The Palgrave Handbook of Methodological Individualism: Volume I [ONE] 9783031415111, 9783031415128

Citation preview

The Palgrave Handbook of Methodological Individualism Volume I Edited by Nathalie Bulle · Francesco Di Iorio

The Palgrave Handbook of Methodological Individualism

Nathalie Bulle · Francesco Di Iorio Editors

The Palgrave Handbook of Methodological Individualism Volume I

Editors Nathalie Bulle French National Center for Scientific Research Paris, France

Francesco Di Iorio Nankai University Tianjin, China

ISBN 978-3-031-41511-1 ISBN 978-3-031-41512-8 https://doi.org/10.1007/978-3-031-41512-8

(eBook)

© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Cover illustration: Gianluca Cavallo, “Pantheon”, 2021, oil on canvas 130x130, source: https://gianlucac avallo.com This Palgrave Macmillan imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Acknowledgments

We wish to express our gratitude to Presses Universitaires de France for free permission to reproduce the translation of Chapters 13 (vol. 1), 14 (vol. 1), 30 (vol. 1), 2 (vol. 2), 18 (vol. 2), 19 (vol. 2), and 21 (vol. 2) that were originally published in French in a special issue on methodological individualism of the journal L’Année Sociologique (Vol. 70(1), 2020). We equally thank other copyright holders for granting us free permission to reuse printed material, namely The European Journal of Sociology for Chapter 23 (vol. 1), Les Éditions de Minuit for Chapter 3 (vol. 1), and Springer for Chapter 25 (vol. 1). We are indebted to the French National Center for Scientific Research (CNRS), the Concepts et Langages doctoral school of Sorbonne University, the Groupe d’Etude des Méthodes de l’Analyse Sociologique de la Sorbonne (GEMASS), Nankai University and Sorbonne University that have generously provided financial support for this project. We sincerely thank Bruce Caldwell, Rosolino Candela, Pierre Demeulenaere, and Leslie Marsh for their guidance in the early stages of its development. Our gratitude extends to all the authors for their contributions, as well as to the editorial staff and production team at Palgrave Macmillan. We also wish to thank Kim Martin and Stephen Procter, who have meticulously edited many chapters of this handbook. Nathalie Bulle Francesco Di Iorio

v

Introduction

Methodological individualism (henceforth MI) is an often criticized and misunderstood approach within the social sciences. A number of authors, who appear to be unfamiliar with MI’s history and theoretical foundations, regard it as a form of naïve and useless reductionism that cannot account for the cultural and structural features of the social world and the influence they exert on its actors. This interpretation does not do justice either to the way in which MI’s logic of explanation was conceived by the classical social scientists who implemented it, nor to the scholars who have followed in their footsteps. This, as well as other common misunderstandings—such as, for example, the tendency to confuse it with a merely utilitarian theory or a political ideology—highlights the need for a comprehensive and upto-date reference work analyzing and presenting MI’s classic arguments and assumptions in the light of contemporary debates in sociology, economics, and philosophy. As a scientific tradition, MI is rooted in the work of Enlightenment precursors such as Adam Smith and was fully developed in the late nineteenth century thanks to the seminal contributions of Carl Menger, Max Weber, and Georg Simmel1 . Its core concerns revolved around fundamental ontological, epistemological, and methodological issues including the explanation of historical and social phenomena, actors’ rationality, and the relationship 1 For more details about the historical emergence of Mi as a scientific paradigm see Bulle (forthcoming).

vii

viii

Introduction

between micro and macro dynamics. As it is well known, the term MI is younger than the scientific practice to which it refers. The expression “individualist method” appears already in the nineteenth century during the Methodenstreit between Carl Menger and the German Historical School. The term MI was used in 1904 by the French philosopher and historian Élie Halévy.2 However, it was Joseph Schumpeter who popularized it in the scientific community through a book chapter that he published in 1908. It is no exaggeration to maintain with Max Weber ([1922] 1978) that by focusing on understandable “social action,” which often produces unintended consequences, MI addresses the central, “constitutive” problem of the social sciences. On this basis, MI aims to explain social enigmas. MI does not overlook the socio-cultural factors and the structural constraints that influence action. Instead, it is concerned with both, taking into consideration the subjective meaning they hold for individuals—a meaning that can be abstractly reconstructed by assuming that human action is based on situated rationality. As already clarified at the beginning of the twentieth century by Schumpeter (1909), MI is a strictly explanatory approach and is neutral to both ethical and political views. Unfortunately, this is a point that is still not always understood in the literature. This is the first handbook devoted to MI and it aspires to give fresh impetus to debates on this perspective. The goal is not only to clarify the aforementioned common misunderstandings about it but also to demonstrate its utility as an explanatory tool in the light of the current debates within the social sciences. Given the historical and scientific importance of MI, the absence of a handbook dedicated to this important and complex tradition of theorizing was a glaring lacuna that needed to be filled. By bringing together contributions from the world’s leading experts on the topic, the present work is a big step in offering a reliable but nonetheless critical discussion of MI. This collective endeavor provides teachers, researchers, and students in sociology, economics, and philosophy with a useful source of reference on this central approach of the social sciences. The historical and contemporary debates about MI are discussed and analyzed, as well as cutting-edge developments within the individualist tradition with philosophical and scientific import.

2 Halévy was not a theorist of MI. The term MI appears in his review of the talk on “the individual and the social” given by Vilfredo Pareto on September 7th, 1904 at The Second World Congress of Philosophy that was held in Geneva, Switzerland (see Halévy, 1904, p. 1108). We thank Massimo Borlandi (2020) for drawing our attention to Halévy’s review.

Introduction

ix

The handbook is divided into two volumes and organized into eight parts. Each volume includes four parts. Part I, The Founding Principles of Methodological Individualism and its Historical Emergence (volume 1). This part provides commentary on and analyses of classical reference texts written by the most eminent economists, sociologists, and philosophers who have been proponents of MI, including Carl Menger, Georg Simmel, Max Weber, Ludwig von Mises, Friedrich Hayek, and Karl Popper. These texts lay the foundations for this approach and their study explains its rationale. Part I also focuses on Adam Smith’s precursory works on the principles of MI, Raymond Boudon’s distinction between holism and individualism, and Jon Elster’s approach. Part II, Methodological Individualism and the Rationality Principle (volume 1). Individualist scholars, from Menger to Boudon, have consistently maintained that human rationality is fundamental to the understanding of social phenomena. However, the status of rationality is an area of considerable debate. Moreover, intense methodological battles have been fought over the very meaning of the concept of rationality itself. The purpose of this second part of the handbook is thus to analyze the most recent debates on this concept from the standpoint of MI. Part III, Methodological Individualism and the Micro-Macro Link (volume 1). This part is devoted to an exploration of the different levels of complexity involved in the explanation of social phenomena. Essays discuss, in particular, their ontological and epistemological status, as well as their relationships and the methodological conditions for moving from one to the other within the general framework of MI. Other topics such as the ties between MI and the theory of complex systems, the role of social institutions within the individualist explanatory model, and the micro-macro link in economic theory are also discussed and analyzed. Part IV, Methodological Individualism and Other Major Intellectual and Scientific Traditions (volume 1). Part IV aims at clarifying the historical, conceptual, and epistemological relationships between MI and other major traditions of theorizing such as the Durkheimian approach, structuralfunctionalism, organizational theory, hermeneutics, analytic philosophy of action, sociology of knowledge, behavioral economics, and agent-based computational simulation. The analysis of the relationships between MI and these other intellectual and scientific traditions is a useful way to appreciate the proper nature of the individualist tradition and its complexity. Part V, Methodological Individualism and Key Research Topics in the Social Science and Humanities (Volume 2). This part highlights the role of MI in the development of the understanding of fundamental social phenomena, such

x

Introduction

as social movements, adherence to social norms, collective beliefs, religions, and ideologies, that have traditionally been accounted for by using holistic approaches. The analysis of the explanation of social change in terms of MI, as well as the relationship between MI and collective intentionality, are also explored and investigated. Part VI, Analytical Tools and Exemplary Case Studies (volume 2). Part VI examines the explanatory power of MI from the standpoint of empirical research. The chapters of this part offer a series of case studies that are inspired by MI and highlight, among other things, the heuristic value of basic mathematical models in MI. Part VI deals with the individualist explanation of concrete phenomena related to history, organizational theory, traditional societies, collective action, religion, the unintended consequences of human action, collective identities, and radicalization. Part VII , Controversial Issues about Methodological Individualism (volume 2). This part focuses on contemporary philosophical and methodological debates about MI, its meaning, and its explanatory power. Particular attention is paid to the interpretation of MI in terms of naïve reductionism, which today dominates Anglo-American philosophy. Part VII focuses, among other things, on holistic bias in contemporary sociology, the limits of the utilitarian individualism used by mainstream orthodox economists, the relationship between MI and cognitive science, and the relationship between MI and symbolic interactionism. Part VIII, Methodological Individualism and its Critics: A Roundtable Discussion (volume 2). The last part of the handbook holds a discussion between its coeditors and some of the most eminent critics of MI, namely Stephen Lukes, Joseph Agassi, and Daniel Little. The focus is on the following relationships: (i) MI, naïve reductionism, and social facts; (ii) MI and institutional individualism; and (iii) MI and methodological localism. This part also includes a list of questions for Margaret Archer, who was originally supposed to take part in this stimulating roundtable to debate the interpretation of MI within critical realism, but finally could not due to personal reasons.

References Borlandi, M. (2020). Raymond Boudon’s methodological individualism. European Journal of Social Sciences, 58(1), 239-266. Bulle N. (forthcoming). Methodological Individualism: Introduction and Founding Texts. London: Routledge (Open Access). Halévy, E. (1904). (Report on the) General Session (of the 11th Congress of Philosophy - Geneva). Revue de Métaphysique et de Morale, 12 (6), 1103-1113.

Introduction

xi

Schumpeter, J. A. ([1908] 2009). The Nature and Essence of Economic Theory. New Brunswick, New Jersey: Transaction. [Transaction. Das Wesen und der Hauptinhalt der theoretischen Nationalökonomie. Leipzig: Verlag von Duncker]. Weber, M. ([1922] 1978) Economy and Society. University of California Press: Berkeley.

Contents

The Founding Principles of Methodological Individualism and its Historical Emergence Methodological Individualism and the Austrian School of Economics Peter Boettke

3

Adam Smith and Sympathetic Individualism Maria Pia Paganelli

17

Max Weber and Understanding Explanation Wolf Feuerhahn

31

Viennese Methodological Individualism Dario Antiseri

57

Popper’s Methodological Individualism and Situational Analysis Makoto Kogawara, Yoji Matsuo, and Han-Seob Yeo

77

Individualism and Holism in the Social Sciences Raymond Boudon

103

What’s the Alternative? Jon Elster

117

xiii

xiv

Contents

Methodological Individualism and the Rationality Principle Emotions and Rationality Paul Dumouchel Rationality, Praxeology, and History: The Contributions of Ludwig von Mises to the Theory of Rationality in the Social Sciences Christian Robitaille and Robert Leroux Methodological Individualism, Scientific Explanation, and Hermeneutics Enzo Di Nuoscio Rationality in Anthropological Explanation Ian Jarvie

141

161

181 203

Ordinary Rationality Theory (ORT) According to Raymond Boudon Jean-Michel Morin

227

Dignity and Axiological Rationality, the Legacy of Raymond Boudon Sylvie Mesure

251

Why Does Comprehensive Sociology Need to Be Enlarged? Gerald Bronner

271

Methodological Individualism and the Micro-Macro Link Methodological Individualism as Holism of the Parts: From Epistemology to Ontology Nathalie Bulle Max Weber’s Methodological Individualism Thomas Schwinn Micro–Macro Transitions and the Paradox of Unintended Consequences Mohamed Cherkaoui

293 321

347

Methodological Individualism and Micro–Macro Modeling Karl-Dieter Opp

377

Complex Methodological Individualism Jean Petitot

407

Contents

The Methodological Foundations of Macroeconomics: A View from Say’s Law Rosolino A. Candela Clarifying Social Institutions in Institutional Individualism Kei Yoshida The Analytical Micro–Macro Relationship in Social Science and Its Implications for the Individualism-Holism Debate Gustav Ramström

xv

433 457

485

Methodological Individualism and Other Major Intellectual and Scientific Traditions The ‘Individual’ in the Writings of Émile Durkheim Anthony Giddens Methodological Individualism (MI) and Methodological Collectivism (MC) in the Era of Postwar American Structural-Functionalism (SF) Charles Crothers

513

535

The Phenomenological Life-World Analysis and the Methodology of the Social Sciences Thomas S. Eberle

565

From Bureaucracy to Organized Action: Methodological Individualism and the Study of Organizations Erhard Friedberg

587

Methodological Individualism and Analytic Philosophy of Action Antonio Rainone Methodological Individualism/Sociology of Knowledge/ Cognitive Sociology Renaud Fillieule Methodological Individualism in Behavioral Economics Malte Dold Agent-Based Models and Methodological Individualism: Are They Fundamentally Linked? Gianluca Manzo

609

631 655

681

xvi

Contents

A Reply to Manzo: The Role of Methodological Individualism for Analytical Sociology Nathalie Bulle

711

Methodological Individualism and Agent-Based Models: A Reply to Manzo Francesco Di Iorio

717

Author Index

727

Index

741

Contributors

Dario Antiseri Department of Political Science, Luiss Guido Carli University, Rome, Italy Peter Boettke Department of Economics and Philosophy, George Mason University, Fairfax, VA, USA Raymond Boudon Department of Sociology, Sorbonne University and Academy of Moral and Political Sciences, Paris, France Gerald Bronner Sociology, Sorbonne University and French Academy of Technologies, Paris, France Nathalie Bulle National Center for Scientific Research and Sorbonne University, Paris, France Rosolino A. Candela F.A. Hayek Program, Mercatus Center, George Mason University, Fairfax, VA, USA Mohamed Cherkaoui Sociology, National Center for Scientific Research, Sorbonne University, Paris, France Charles Crothers Sociology, Auckland University of Technology, Auckland, New Zealand Francesco Di Iorio Philosophy, Nankai University, Tianjin, China

xvii

xviii

Contributors

Enzo Di Nuoscio University of Molise, Campobasso, Italy; Luiss Guido Carli University, Rome, Italy Malte Dold Economics Department, Pomona College, Claremont, USA Paul Dumouchel Department of Philosophy, University of Quebec, Montreal, Qc, Canada Thomas S. Eberle Research Institute of Sociology, University of St. Gallen, St. Gallen, Switzerland Jon Elster Departments of Philosophy and Political Science, Columbia University, New York, NY, USA; Collège de France, Paris, France Wolf Feuerhahn History of Sciences, Centre Alexandre-Koyré, National Centre for Scientific Research and École des Hautes Études en Sciences Sociales, Paris, France Renaud Fillieule Institute of Social Sciences, University of Lille, Lille, France; CLERSE UMR 8019, Villeneuve d’Ascq, France Erhard Friedberg Emeritus of Sociology, Paris, France; Sciences Po, Paris, France Anthony Giddens Sociology, Cambridge University and London School of Economics, London, UK Ian Jarvie Philosophy, York University, Toronto, ON, Canada Makoto Kogawara Kagoshima University, Kagoshima, Japan Robert Leroux Department of Sociology, University of Ottawa, Ottawa, ON, Canada Gianluca Manzo Department of Sociology, Sorbonne University, Paris, France Yoji Matsuo Faculty of Commercial Sciences, Hiroshima Shudo University, Hiroshima, Japan Sylvie Mesure National Center for Scientific Research, Paris, France Jean-Michel Morin Sociology, University of Paris Cité, Ecole Professorale de Paris, Collège des Bernardins and Institut de Philosophie Comparée, Paris, France

Contributors

xix

Karl-Dieter Opp Sociology, University of Leipzig, Leipzig, Germany; University of Washington, Seattle, WA, USA Maria Pia Paganelli Department of Economics, Trinity University, San Antonio, TX, USA Jean Petitot Mathematics, École des Hautes Études en Sciences Sociales, Paris, France Antonio Rainone Department of Human and Social Sciences, University of Naples “L’Orientale”, Naples, Italy Gustav Ramström Political Science, Stockholm University, Stockholm, Sweden Christian Robitaille Department of Sociology, Liverpool Hope University, Liverpool, UK Thomas Schwinn Sociology, Max-Weber-Institute of Sociology, University of Heidelberg, Heidelberg, Germany Han-Seob Yeo School of Commerce, Meiji University, Tokyo, Japan Kei Yoshida Faculty of Social Sciences, Waseda University, Tokyo, Japan

List of Figures

Popper’s Methodological Individualism and Situational Analysis Fig. 1 Fig. 2

The interaction of three worlds D-N explanation schema and SL explanation schema (Source Nadeau 1993, p. 450)

81 95

Micro–Macro Transitions and the Paradox of Unintended Consequences Fig. 1

Relations between macro-phenomena (religious ethics i.e. set of prescriptions and proscriptions in relation to the world and secular institutions such as economy and politics; modern democracy i.e. ultimate political values) and micro-phenomena (qualification i.e. disposition of gifts or abilities that attest to being chosen by God; social relationships e.g. no allegiance to a man, membership in a sect such as voluntary association subject to vote, administration of community affairs)

367

Methodological Individualism and Micro–Macro Modeling Fig. 1 Fig. 2 Fig. 3 Fig. 4

An example for a micro–macro model Kinds of statements in micro–macro explanations Example of a micro–macro model with empirical bridge assumptions Example for a macro explanation of macro propositions

380 380 387 390

xxi

List of Tables

Popper’s Methodological Individualism and Situational Analysis Table 1

Three worlds and their entities

84

Max Weber’s Methodological Individualism Table 1

Combination of Orders and Orientations

337

Methodological Individualism (MI) and Methodological Collectivism (MC) in the Era of Postwar American Structural-Functionalism (SF) Table 1

Individual and collective properties (Lazarsfeld and Menzel 1956)

548

Methodological Individualism/Sociology of Knowledge/ Cognitive Sociology Table 1

Synthetic view of the field of sociology of knowledge: types of explanation × types of knowledge

650

xxiii

The Founding Principles of Methodological Individualism and its Historical Emergence

Methodological Individualism and the Austrian School of Economics Peter Boettke

The concept of methodological individualism finds its origins in the writings of Joseph Schumpeter (1908). But the practice of tracing economic phenomena back to intelligible human action can be found in economists since at least the time of Adam Smith (see Smith 1776, 15, 18). The dinner we enjoy comes from the individual behavior of the baker, brewer, and butcher, and the common woolen coat that ends up on our backs in Smith’s analysis despite the incalculable number of exchange relations required to produce it nevertheless emerges from the purposive behavior of those actors at each nod in the process. The object of our study may well be the “invisible hand” of the market, but how we explain its emergence is through rigorous study of the processes of interactions among various individuals striving to do the best that they can given their circumstances, and weighing their decisions on multiple margins as they cope with their ignorance and wrestle with time. Schumpeter was right to focus his analysis on the implications of the economic theory of methodological individualism, rather than the history of doctrine. Economics begins with the recognition of scarcity. No scarcity, no economics to talk about. Once we recognize scarcity, we acknowledge the

P. Boettke (B) Department of Economics and Philosophy, George Mason University, Fairfax, VA, USA e-mail: [email protected]

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 N. Bulle and F. Di Iorio (eds.), The Palgrave Handbook of Methodological Individualism, https://doi.org/10.1007/978-3-031-41512-8_1

3

4

P. Boettke

necessity of trade-offs. It is individuals that make those trade-offs, and decide to pursue path A and forego path B. Let me be very explicit, economics is not limited to science of choice. It cannot end with the pure logic of individual choice. Economic theory is trying to explain the complex and intricate pattern of social relationships that constitute social cooperation under the division of labor. The explaining is done by tracing back to the purposive behavior of individuals weighing their trade-offs and forming exchange relationships with other purposeful actors. These exchange relationships are formed within institutions that are shaped by legal, political, and social “rules of the game”. David Hume (1739), long ago, argued that without the institutions of property, contract, and consent, society as we know it would not be possible. Individuals would be unable to pursue productive specialization and realize peaceful social cooperation through exchange. Instead, violence and unchecked power and relationships of domination and subordination result whenever there is a lack of stability in possession, abandonment of transference by consent, and disregard for the keeping of promises. Society ceases, and a war of all against all emerges. On the other hand, when the institutions of property, contract, and consent are in operation and reinforced by social mores, norms, and conventions, then individuals will be able to negotiate their trade-offs through the “aids to the human mind” provided by relative prices, profit-and-loss statements, and the pattern of resource ownership. The resulting complex coordination of economic activity through time reflects the intricate matrix of human interaction that constitutes social cooperation. Social phenomena, rather than denied by methodological individualists, is actually explained by their recognition of the subjectivism of value and cost, and their analysis of the dynamic and entrepreneurial market process that follows from the adoption of a methodological individualist stance to the study of emergent social phenomena. That has been the defining characteristic of the Austrian School of Economics from Carl Menger onward. In this essay, I would like to clarify the concept of methodological individualism and its importance by discussing what it is not, and how the Austrian economists—namely Menger, Mises, and Hayek—countered these charges in their work.1 Section 1 will explain that methodological individualism is not methodological atomism. This is one of the first criticisms usually leveled at the concept and to homo economicus. Section 2 will explain that methodological individualism is not asocial. That is another significant criticism that has been leveled against the concept and its usefulness in the social sciences. 1 For an overview of the contributions to the Austrian School of Economics see Boettke and Coyne, ed. (2015), Boettke, ed. (1994).

Methodological Individualism and the Austrian …

5

Section 3 will discuss the intellectual discipline that I believe methodological individualism provides social science. Section 4 will conclude.

1

Not Atomistic

Critics of methodological individualism from the German critics of Adam Smith to Thorstein Veblen’s criticisms of neoclassical economics have tended to dismiss as unrealistic and unproductive depictions of isolated men maximizing their utility by pursuing their self-interest. The German’s debated the das Adam Smith problem which basically argued that the substance of The Theory of Moral Sentiments (1759), which focused on other-regarding behavior, could not be squared with the emphasis on self-interested behavior in The Wealth of Nations (1776). The solution through looking at the relevant span of moral sympathy was largely rejected by the Germans to the detriment of progress in the social sciences. And, this doesn’t even raise the more subtle and tension-filled puzzling observation of F. A. Hayek, that modernity requires us to live in two worlds at once—our intimate order of family and friends and the extended order of the social relations of exchange and production in the “Great Society.” Veblen’s snide remarks aside about lightening calculators of pleasure and pain, his criticisms largely miss their mark because early neoclassical writers such as Alfred Marshall did not postulate an isolated individual maximizing regardless of the social world they occupied. Menger did argue that “Man with his needs and the command of the means to satisfy them, is himself the point at which human economic life both begins and ends” (1871, 108). Menger’s presentation is straightforward economics. Begin with scarcity. As a consequence of scarcity, individual decision-makers must economize in their choices. The individual is striving to do the best that they can given the circumstances they face. Note I said striving because it is to be contrasted with achieving. Menger’s man coping with scarcity is also coping with ignorance as well as uncertainty, and all within a process that takes place in time. Human actors are caught between alluring hopes and haunting fears, and they must rely on various guideposts and aids to the human mind to prod them, direct them, and discipline them so they may stumble their way to pursuing productive specialization and realizing peaceful social cooperation through mutually beneficial exchange relationships with others in the marketplace. The participants in the market process are emersed in a continuous evolution toward a solution. What is critical for the point I am trying to emphasize is that Menger’s contribution to the marginalist revolution—his subjective theory of value and

6

P. Boettke

costs—led him to start his analysis with a focus on the purposive human actor and the causal relationship between means and ends. The actor may have a confused understanding of that relationship or be unaware of that relationship with regard to specific goods and services, or they may be poor at executing their plan due to various human foibles, but they are purposive and as such are forever striving to achieve their goals in the most efficacious way they perceive. Error is inevitable, but it is also correctable. Both at the actor level and the system level, there is learning going on. Erich Streissler (1972, 433, emphasis in original) described this aspect of Menger’s analysis of the market process in the following way: “Mengerian man thus knows little about the present, though, interesting enough, he is constantly trying to increase his knowledge, creating social institutions to gather information, empowering growing droves of middlemen to act on his behalf.” Not only did Menger not work with an atomistic notion of man, but neither did Mises nor Hayek. Society is a composite of the actions of individuals, and not some disembodied entity that acts with some mysterious collective purpose. Society is thus, neither reducible to the intentions of individuals, nor some collective entity separate from the individuals who reside within its boundaries. “Not only man’s action towards external objects,” Hayek (1952, 57–58) tells his readers, “but also all the relations between men and all the social institutions can be understood only by what men think about them. Society as we know it is, as it were, built up from the concepts and ideas held by the people; and social phenomena can be recognized by us and have meaning to us only as they are reflected in the minds of men.” From the Scottish Enlightenment philosopher to the modern practitioners of Austrian economics, social order was understood to be a result of human action, but not of human design. In the spontaneous order of society, the sum is greater than the whole of its parts. As Mises argues in Human Action (1949, 42): “The life of a collective is lived in the actions of the individuals constituting its body. There is no social collective conceivable which is not operative in the actions of some individuals. The reality of a social integer consists in its directing and releasing definite actions on the part of individuals. Thus the way to a cognition of collective wholes is through an analysis of the individuals’ actions.” Hayek in The Counter-Revolution of Science (1952, 69) made this observation about our task as social theorists. “If social phenomena,” Hayek states, “showed no order except insofar as they were consciously designed, there would indeed be no room for theoretical sciences of society and there would be, as is often argued, only problems of psychology. It is only insofar as some sort of order arises as a result of individual action but without being designed

Methodological Individualism and the Austrian …

7

by any individual that a problem is raised which demands a theoretical explanation.” Simply put, if we pursued either atomistic individualism or holistic collectivism there would be no need for social scientific theory. There would just be one-to-one correspondence between the intentions of individuals and the pattern of outcomes we observe in the world, or the general will of society would be perfectly reflected in the social order obtained and maintained. Our quest for the theory comes from our puzzlement and our curiosity being peaked by the disjoint between intentions and results, in either a good or bad way. The “invisible hand” of the market, one must remember, is juxtaposed with the proposition that “the path to hell is paved with good intentions.” Be that as it may, the simple point I am making is the Austrians, while being methodological individualists, are not methodological atomists.

2

Not Asocial

“As a thinking and acting being,” Mises states, “man emerges from his prehuman existence already as a social being” (1949, 43). The evolution of our reason, the languages we communicate in, and the social cooperation under the division of labor we benefit from are social phenomena that come into being through the actions of individuals. We are not isolated— not Crusoe on an island as in the crude thought experiment often use to caricature economics. The Crusoe in Daniel Defoe’s actual book, it should be stressed, while stranded on the island, lives with the voice of his father in his head, and with the social approbation and disapprobation of his lived past before his fateful voyage. This social context which Crusoe carries with him is what prepares him for the relationship formation that follows after he encounters Friday. It is an interesting question to ask if economists, rather than using Robinson Crusoe (1719) to depict the economizing decisions that individuals are compelled to make in choosing against the constraint of the scarcity of nature, chose instead to borrow from the tales Defoe told in Moll Flanders (1722), with its social positioning and posturing, and the games people play with others in their desire to win favor and better their condition. Strategic behavior is crucial to understanding the story, but so is the background of social relations and status in society. Economics should not be seen as a science of choice. Instead, it is more productive, I would argue, to view economics as a science of exchange and the institutions within which exchanges take place. But we can only analyze the

8

P. Boettke

situational logic of organizations and social order if we begin, as Menger put it, with individuals and their needs and their command over the resources to satisfy them. The logic of choice is a necessary component of economic science, but it is not sufficient.2 The logic of choice must be put to work within alternative institutional arrangements in order to tease out the situational logic that specifies the patterns that emerge in our interactions with nature and with others. In his classic essay “Individualism: True and False,” Hayek (1940, 12– 13) argued that the classical political economists from Smith onward focused analytical attention on the situational logic of an institutional configuration to explicate the mechanism behind the “invisible hand.” As he put it: “The chief concern of the great individualist writers was indeed to find a set of institutions by which man could be induced, by his own choice and from the motives which determined his ordinary conduct, to contribute as much as possible to the need of all others; and their discovery was that the system of private property did provide such inducements to a much greater extent than had yet been understood.” As he says elsewhere in the essay: “What then are the essential characteristics of true individualism? The first thing that should be said is that it is primarily a theory of society …” (1940, 32). Mises in Human Action (1949, 42) could not be clearer on this point. As he states: “Methodological individualism, far from contesting the significance of such collective wholes, considers it one of its main tasks to describe and to analyze their becoming and their disappearing, their changing structures and their operation.” Menger famously asked in his Investigations (1883) how it was possible that these social institutions emerge without conscious design. In fact, he argued that this question was the single most important question for the theoretical social sciences. Understanding the spontaneous order that produces for us not only money, prices, and markets, but also norms of property rights, language, and social mores, is critical for progress in social theory. Mises (1949, 405) would go as far as to argue that Menger’s theory of the evolution of money, was not just a correct theory of the origins of money, but an exemplar of how we are to engage in theoretical social science with the aim to produce a theory of society. Methodological individualism and the compositive method are the only tools we economists have to solve 2

This is, I believe, the most productive way to read Hayek’s (1937) paper “Economics and Knowledge”. Economists must, Hayek argued, move from the pure logic of choice to the social learning that is made possible by certain institutional configurations and thwarted by other institutional arrangements in order to give economic science empirical content. See Boettke (2018) for an extended discussion of Hayek’s epistemic institutionalism in comparison with the incentive institutionalism that was developed with New Institutionalism in economics. Hayek’s contribution is not to be seen as in conflict with the incentive-based arguments, but as complementary to them.

Methodological Individualism and the Austrian …

9

the theoretical problem of cooperation without command in a satisfactory manner. Menger’s theory of the price formation process on the market, just as his later development of his theory of property rights and his even more famous theory of the emergence of money, all reflect this invisible hand style of reasoning. Menger was a financial journalist before he became a theoretical economist. His original goal in economic studies and research was to develop a realistic theory of the prices he witnessed in the market. Yet, to do so he had to trace economic phenomena back to the actions of individuals. Menger develops an economic theory that to his mind is both realistic and abstract. His actor is a human actor, not merely a utility machine that can calculate with lightning-speed the marginal benefits and marginal costs of any decision. The central characters in Menger’s theoretical system are fallible but capable human choosers, who are coping with ignorance and uncertainty, but striving to better their condition as best as they humanly can. Menger does not postulate a robotic version of mankind. There are no automatons in his theorizing. Menger instead argues that we study the social world as a process embedded in particular institutional settings and that through the interaction of human agency and social structures manifestations of patterns of human behavior and interaction emerge that we can explicate in our theorizing. The Austrian school of economics since its founding was built to study emergent order and not exclusively on the optimality conditions of a static state of rest.3 Menger’s theory of man is not a decision-maker under the assumption of full and complete information, but instead one of individuals having to choose in an uncertain and ever-changing world, and form exchange relationships with others who are also in the process of deciding in deep uncertainty and coping with their ignorance. In short, Menger’s economics is grounded in human action and human social interaction. Abstract yet realistic theory emerges in this narrative, and it forces us to see the playing out of economic life from the inside-out, rather than from the outside-in. We are who we study, and while it is always a good test of our reasoning to “solve for the equilibrium,” it is critical to our enterprise to acknowledge that we are part of the equilibrium pattern that emerges in our social interactions. It is the clever and creative human actors that populate the economy under investigation that give rise to the patterns, not the theorist who imposes them on the situation in order to comfort their imagination. We must always remember 3 Arthur (2023, 643) writes that economic science must shift its focus from an “economics of nouns” to an ‘economics of verbs’ if it is to make progress as a scientific discipline. But he notes that “a body of theory with verbs already exists in economics – Austrian economics.”

10

P. Boettke

that the baker, the brewer, and the butcher, were already providing dinner to unknown multitudes before Adam Smith theorized about it, and the prices that were formed in those exchanges and the profits or losses realized by the enterprise existed prior to the development of his theory of price. Economic theory renders the patterns of social life intelligible by tracing the phenomena back to the purposes and plans of human actors. “It is the contention,” Hayek (1940, 6-7) states, “that by tracing the combined effects of individual actions we discover that many of the institutions on which human achievements rest have arisen and are functioning without a designing and directing mind; that, as Adam Ferguson expressed it, “nations stumble upon establishments which are indeed the result of human action but not of human design”; and that the spontaneous collaboration of free men often creates things which are greater than their individual minds can ever fully comprehend.”

3

Explanandum and Explanation

As we have seen, the explanandum of the theoretical social sciences is the emergence of society itself. The foundation of society is social cooperation under the division of labor. Mises in Human Action (1949, 143) provides a succinct statement to this effect: “The actions which have brought about social cooperation and daily bring it about anew do not aim at anything else than cooperation…. The total complex of the mutual relations created by such concerted actions is called society. …Society is division of labor and combination of labor. In his capacity as an acting animal man becomes a social animal.” The thing to be explained is social cooperation under the division of labor. As Mises (1949, 144) put it: “In a hypothetical world in which the division of labor would not increase productivity, there would not be any society.” There would not be the opportunity for fellow feelings and goodwill; there would be a biological struggle for survival. On the other hand, when the increased productivity of social cooperation under the division of labor is realized, Mises (ibid.) states: “Within the frame of social cooperation there can emerge between members of society feelings of sympathy and friendship and a sense of belonging together. These feelings are the source of man’s most delightful and most sublime experiences. They are the most precious adornment of life; they lift the animal species man to the heights of a really human existence. However, they are not, as some have asserted, the agents that have brought about social relationships. They are fruits of social cooperation, they

Methodological Individualism and the Austrian …

11

thrive only within its frame; they did not precede the establishment of social relations and are not the seed from which they spring.” Adam Smith (1776, 7) informed his readers in the opening sentence of Chapter 1 of The Wealth of Nations that the source of the greatest increase in the productivity of mankind is due to the division of labor. This empirical fact of the world, as a consequence, brought about the very possibility of society and transformed human behavior from viewing one another as mortal enemies trapped in violence as they wrestle with the implications of the Malthusian trap into trading partners with sentiments of benevolence and goodwill. How we explain the emergence of social cooperation under the division of labor is through economic theory based on methodological individualism. That is our explanation—the logic of choice and the situational logic within alternative institutional arrangements. The intellectual discipline that methodological individualism imposes on social scientists is that they cannot categorize some actors under investigation as dumb and incapable, and others as brilliant, saintly, and uber-component. We cannot analytically cheat in our depiction of the world. We must do social science without recourse to assumptions of omniscience and benevolence, let alone postulate frictionless environments with smooth and continuous functions that are twice differentiable. Such depictions are surely nonsense rather than social science. The world that we occupy and which we study is messy and lumpy. Instead, everyone in the universe we are seeking to understand is pursuing their own interest as efficaciously as they can and responding to the incentives of the alternative institutional arrangements they find themselves operating within, and are busy planning and revising their plans based on the signals and feedback mechanisms of the system that produces the requisite knowledge they utilize. Social order is characterized by the constant need for individuals to adjust and adapt their behavior to one another as circumstances change. In commercial society, the necessary guideposts in this process of adaptation and readaptation to ceaseless change come in the form of property rights, prices, and profit-and-loss. Outside the realm of commercial society, the social system must find a substitute that will serve that function of cajoling us, prodding us, guiding us, luring us, and disciplining us. It turns out that commercial society has advantages that are absent in alternative social systems. The market price system guides individual decisionmakers to pursue productive specialization and realize peaceful social cooperation through an exchange. It is individuals and their purposes and plans that animate the processes of the market that bring about the exploitation

12

P. Boettke

of mutually beneficial exchanges and the spur for entrepreneurial innovation that leads to unimagined progress. As Hayek (1940, 14) put it: “The real question, therefore, is not whether man is, or ought to be, guided by selfish motives but whether we can allow him to be guided in his actions by those immediate consequences which he can know and care for or whether he ought to be made to do what seems appropriate to somebody else who is supposed to possess a fuller comprehension of the significance of these actions to society as a whole.” We live in a world of scarcity, scarcity necessitates trade-offs, the trade-offs must be carefully negotiated, and the tools provided by commercial society in the form of prices and profit-and-loss accounting aid human actors to pursue their plans effectively and cooperate and coordinate their actions with those of others to generate a complex and intricate matrix that constitutes human society. The social system of exchange and production must be so arranged to produce more with less, it will not be an effective promoter of human betterment if it consistently produces less with more. Human beings exhibit two natural propensities that can be identified as Smithian and Hobbesian. Our Smithian propensity is to truck, barter, and exchange. Our Hobbesian propensity is to pillage and plunder. Which propensity is on display in any historical epoch is a function of the social ecology that human actors find themselves operating within. Configurations that steer us on the Smithian path lead to the wealth of nations, whereas those that cater to our Hobbesian nature lead to violence traps and the reason why nations fail and populations suffer. We know this from history because we have the body of economic theory that was built upon the foundation of methodological individualism that can explain the emergence of social cooperation and the determinants that can undermine its attainment.

4

Conclusion

The structure of economic theory that has been built up from the purposive action of individuals through the mediation of alternative institutions that produces a patterned outcome of social order is one of our greatest tools for observation and understanding of the human condition. We have that theory because of the classical political economists and because of the Austrian School of Economics and the work of Menger, Mises, and Hayek in particular. They were the modern theorist that took ideas from Adam Smith and developed the invisible hand style of reasoning in modern economics.

Methodological Individualism and the Austrian …

13

I have in various writings referred to this as “mainline economics,” to reflect the broader intellectual tradition than just the Austrian School of Economics (see Boettke 2012; Boettke et al., ed., 2016). In my rendering, the mainline of economic theory squares the invisible hand proposition with the rational choice or self-interest postulate via institutional analysis. Thinkers who deny that these two can be squared, due either to weaknesses of will on the part of individual actors or the inability of prices to work to guide exchange and production decisions or both, as Keynes argued in “The End of Laissez Faire” (1926), are not mainline economists. But neither are those theorists who for the sake of mathematical tractability, or simply for intellectual comfort and convenience, collapse the two into one another so that there is no social mystery to be solved. For example, thinkers who insist on an equilibrium solution, like the Chicago School of Economics, are also not in the mainline of economic thinking. This emphasis on efficient equilibrium solutions is not found in mainline economic thinkers like David Hume and Adam Smith, nor Menger, Mises, and Hayek, nor is this view to be found in Alfred Marshall or Frank Knight (see Buchanan 2009). Neither of the hegemonic paradigms of the neoclassical synthesis of Paul Samuelson nor the Chicago School of George Stigler and Gary Becker contribute to the further development of mainline economics, though of course, both are solidly within the mainstream of economic science since the mid-twentieth century. Instead, within the contemporary intellectual landscape of economics, mainline economics emerges from reconciling the contributions of property rights economics, law and economics, public choice economics, and market process economics. Economics is about exchange and the institutions within which exchanges take place. It studies economic forces at work, and eschews the preoccupation among many economic thinkers to only focus on the state of affairs after the economic forces have completed their work. And what unites these different views of the science of economics and provides the intellectual discipline to their analysis is the shared commitment to the practice of methodological individualism, properly understood. All social phenomena are traced back to the purposive behavior of acting individuals, and that is how they are rendered intelligible. It is this exercise of tracing back to purposive actors and tracing forward the intended and unintended consequences of those actions—both desirable and undesirable— that constitutes the economist’s task. As we have argued the economist is aided in this mental exercise by their nuanced understanding of the pure logic of choice, and the situational logic that follows from institutional analysis. This approach enables the theorist to articulate the processes that generate the emergent order of social cooperation under the division of labor. The

14

P. Boettke

coordination of complex economic activity through time is the mystery, it is the alert, creative, and clever entrepreneurial actions of individuals that provides the answer. Acknowledgements I would like to acknowledge the research assistance of Michael Wroblewski, and the helpful and encouraging comments on an earlier draft from Rosolino Candela and the editors of this volume. The usual caveat applies.

References Arthur, B. (2023). “Economics in Nouns and Verbs,” Journal of Economic Behavior and Organization 205 (January), 638–647. Boettke, P., Ed. (1994). The Elgar Companion to Austrian Economics. Cheltenham, UK: Edward Elgar Publishing. Boettke, P. (2012). Living Economics: Yesterday, Today and Tomorrow. Oakland, CA: The Independent Institute. Boettke, P. (2018). F. A. Hayek: Economics, Political Economy and Social Philosophy. New York: Palgrave Macmillan. Boettke, P. and C. Coyne, Eds. (2015). The Oxford Handbook of Austrian Economics. New York, NY: Oxford University Press. Boettke, P., S. Haeffele and V. Storr, Eds. (2016). Mainline Economics. Arlington, VA: Mercatus Center. Buchanan, J. (2009). “Economists Have No Clothes,” RMM Journal (January), 151–156. Defoe, D. (1719). The Life and Surprising Adventures of Robinson Crusoe. New York, NY: Penguin Classics, 2003. Defoe, D. (1722). The Fortunes and Misfortunes of the Famous Moll Flanders. New York, NY: Norton, 2003. Hayek, F. A. (1937). “Economics and Knowledge,” reprinted in Hayek 1948. Hayek, F. A. (1940). “Individualism: True and False,” reprinted in Hayek 1948. Hayek, F. A. (1952). The Counter-Revolution of Science. Indianapolis, IN: Liberty Fund, 1979. Hume, D. (1739). A Treatise of Human Nature. Mineola, NY: Dover Publications, 2003. Keynes, J. M. (1926). The End of Laissez Faire. London, UK: Hogarth Press. Menger, C. (1871). Principles of Economics. New York, NY: New York University Press, 1981. Menger, C. (1883). Investigations into the Methods of the Social Sciences. New York, NY: New York University Press, 1983. Mises, L. (1949). Human Action: A Treatise on Economics. Chicago, IL: Henry Regnery, 1966.

Methodological Individualism and the Austrian …

15

Schumpeter, J. (1908). Das Wesen und der Hauptinhalt der theoretischen Nationalökonomie. Leipzig: Duncker & Humbolt. Smith, A. (1759). The Theory of Moral Sentiments. Indianapolis, IN: Liberty Fund, 1994. Smith, A. (1776). An Inquiry into the Nature and Causes of the Wealth of Nations. Chicago, IL: University of Chicago Press, 1976. Streissler, E. (1972). To What Extent Was the Austrian School Marginalist? History of Political Economy, 4 (2), 427–441.

Adam Smith and Sympathetic Individualism Maria Pia Paganelli

James Buchanan (1919–2013), the 1986 recipient of the Nobel Prize in Economics, would complain about economic textbooks because they start with indifference curves and budget constraints and, only later, do they introduce exchange with the Edgeworth box. Indifference curves are a way to model an individual’s preferences, and when a budget constraint is introduced, they are a way to model an individual’s choices. An individual can choose the combination of how many avocados and how much broccoli to produce and consume, given the resources available so that the individual’s utility is maximized. This individual is a Robinson Crusoe, usually explicitly defined as such. He lives alone, in isolation, and in autarky. Because he is alone, he consumes what he produces. It is only when Friday shows up, that exchange is possible. Exchange allows Robinson Crusoe to consume more of what he likes, so his utility increases. Given an endowment, Robinson Crusoe’s and Friday’s indifference curves are represented in an Edgeworth Box, named after the economist who modeled it first, to show that exchange will increase their utility, until a point in which

Thanks to two anonymous referees, to the participants of NYU 2022 colloquium, to Craig Smith and Jimena Hurtado for comments. All mistakes are mine.

M. P. Paganelli (B) Department of Economics, Trinity University, San Antonio, TX, USA e-mail: [email protected]

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 N. Bulle and F. Di Iorio (eds.), The Palgrave Handbook of Methodological Individualism, https://doi.org/10.1007/978-3-031-41512-8_2

17

18

M. P. Paganelli

utility is maximized for both individuals and increasing the utility of one would decrease the utility of the other. Exchange is thus like a technology that allows Robinson Crusoe to consume more. Buchanan would say that starting with utility maximization is an incorrect start. We should start instead as Adam Smith started. Adam Smith (1723– 1790) is the eighteenth-century-thinker usually considered the father of economics. Endorsing his approach, we should start directly with exchange. For Buchanan, the difference is fundamental. With utility maximization, the starting point is one. We need just one individual—Robinson Crusoe—to maximize utility. Actually, we may not even need an individual human being, just an individual. We can model utility maximization of an individual rat and an individual plant too. The logic is the same (Friedman 1953). But when we start with exchange, we must start with two. There cannot be any exchange if there is just one individual. “Give me that which I want, and you shall have this which you want” (WN I.ii.2, emphasis added) tells us Adam Smith at the beginning of the Wealth of Nations ([1776] 1981). To exchange I cannot be by myself, you need to be present as well. To trade, we need to have at least two people. And these two people must be human beings, since only human beings are capable of engaging in a “fair and deliberate exchange,” as Smith believes (WN I.ii.2). Other animals may look like they are exchanging, but they are acting instinctually, not deliberately. Only humans can make contracts and exchange property (WN I.ii.2; see also Wilson 2020). A dog, Smith tells us, can fawn to get its dinner from its master. A person occasionally can do the same, but they do not have to. They can go to the butcher, brewer, and baker and engage with them as peers as they buy their dinner instead (WN I.ii.2). Exchange is a much more effective and dignifying way to obtain what one wants than the one used by dogs, and it is a process that only humans are able to use (Paganelli 2023b). What Buchanan understood is the intrinsic sociality of human beings as manifested in exchange and thus in economics, that same sociality that Adam Smith describes. Human beings are intrinsically social beings, they cannot be understood if they are thought of in isolation. Individualism, as in atomism, is thus an odd label to have become associated with Adam Smith (see for example Menger [1891] 2016), since Smith radically and explicitly rejects the idea of an individual as an atomistic and independent agent (on the differences between Menger and Smith see Hurtado and Paganelli 2023). The only one time he mentions the possibility of an individual growing up isolated from society, he does it to show the absurdity of the claim (TMS III.i.3). And when presented with social contract theories, which presume isolated individuals who eventually join themselves

Adam Smith and Sympathetic Individualism

19

in society because they see the advantages of society, he dismisses the theories as equally absurd (A. Smith, [1762–3; 1766] 1978; LJ (A), March 22, 1763). Friedrich A. Hayek in his Individualism True and False (1948) identified the problem with the label individualism, since individualism can be used to indicate both an atomistic individualism (false individualism) and a social individualism (true individualism). C.B. Macpherson (1989) also distinguishes between different kinds of individualism—developmental individualism where the flourishing of an individual is linked to the flourishing of society; and possessive individualism where the individual is a sort of aggressive and greedy homo economicus. Macpherson identifies Adam Smith as a mild developmental individualist, without any significant explanation of what he means by this. Hayek also does not indulge in explaining the relation between the social and the individual in Smith, being more concerned with explaining social phenomena as the unintended actions of a multitude of individuals, rather than as a conscious plan of one or of a small group of them. I will try to explain what Hayek took for granted, and what Buchanan claimed to be the starting point of economics: how, for Adam Smith, human beings are intrinsically social. Even if one accepts individualism as the “principle [that] states that social processes should be explained by being deducted from principles governing the behaviour of participating individuals and from analyses of their situations, and not from super-individual, ‘holistic’, sociological laws” (Watkins 1952, p. 186), or as “a social theory or ideology which assign a higher moral value to the individual than to the community or society, and which consequently advocates leaving individuals free to act as they think most conductive to their self-interest” (Macpherson 1989, p. 149) like Hayek, I am uncomfortable with using the word individualism because of the association it has with atomism, possessivism, or in more recent times even with rational choice (Heath 2020). Rather than characterize the concept as true or false, like Hayek does, I would prefer thinking instead in terms of “sympathetic individualism” as Maria Alejandra Carrasco has recently done (Carrasco 2022). My true preference would actually be to eliminate such an ambiguous and controversial term. Vernon Smith and Bart Wilson are also trying to discard the ambiguous “individual” for a more human and humane person, who cares about the social context in which they conduct themselves (V. L. Smith and Wilson 2019). Their project of Humanomics is indeed to reintroduce the humane into economics, which among other things means to recognize, like Adam Smith does, the intrinsic sociability of human beings.

20

M. P. Paganelli

This does not imply that social phenomena can be explained through anything but an “individual”—Smith’s methodology is strictly individualistic—but the recognition that a person acts, and is defined, exclusively through social interactions and thoughts in the presence of others. Social phenomena and institutions, from morals, to law, and markets, for Smith are adamantly not the product of any human wisdom but the unintended and unpredicted consequences of individual actions. In this sense, Smith adopts a true methodological individualism. But those individuals are like overcooked noodles left in a dish too long: you can’t pick up just one. If you try to pick up just one, you end up picking up a whole bunch, as each noddle is attached to another and cannot be separated. I will thus stick with Carrasco’s term which I borrowed from her to title this essay (for an earlier use of the term in a completely different context, and unrelated to Adam Smith, see Long 2012). The reason for this choice is because Smith does put the “individual” at the center of all his analysis, be it moral, economical, historical, political, or legal. The main concern and driving force of society is indeed the individual. Social outcomes are unintended consequences of individual actions. But at the same time this “individual” is never alone, in the sense that the individual cannot exist both in reality and analytically as an individual, but the individual is always and without exception a social being, defined by the presence of others. Sympathy, as we will see in a moment, is the mechanism through which we relate with each other and thus through which we can identify as ourselves. This is why Carrasco’s term, “sympathetic,” is very much appropriate for characterizing Smith’s “individualism” (for the idea of a sympathetic liberalism in Smith (and Rawls) see Hurtado 2006, 2008).

1

So, What Is Sympathy?

Smith describes sympathy in his Theory of Moral Sentiments ([1759] 1982), his first published book which made him famous during his lifetime. I say describe rather than define because Smith uses sympathy in different ways. So it is difficult, if not impossible, to offer a quotable definite definition of it (see among others Sayre-McCord 2013). From his work, we can build a working definition of sympathy, though. Smith opens the Theory of Moral Sentiments saying that our body limits our ability to relate with others. I cannot feel what you feel because my body is disconnected from your body. But I can imagine myself in your position and imagine what I would feel if I were in your position (TMS I.i.1.1–4). This imaginative process of placing

Adam Smith and Sympathetic Individualism

21

oneself in the shoes of another and imagining how one would feel if one were in that situation can be thought of as sympathy for Smith. Sympathy implies that when I see you laugh or cry, I will laugh or cry with you, if (and only if ), when I understand the circumstances of your laughing or crying and imagine myself in those same circumstances, I imagine that I also would laugh or cry. But if in your circumstances I imagine that I would cry while you are laughing, or that I would laugh while you are crying, I will not laugh or cry with you. And actually, I will not even approve of your laughing or crying. And if you are able to see yourself with my eyes, you will know that I will not approve of your conduct. And since you want to receive my approbation, you will adjust the pitch of your passion so that I would approve of you (TMS I.i.5). So, for Smith, your conduct is determined by my approval of it. You adjust the expression of your emotions, and you adjust your conduct, in a way that you can gain my approbation because your conduct is deserving of my approbation in my eyes. This is something natural, for Smith. We are born with and driven by our desire to be lovely and approvable, and by our desire to avoid being hateful and disapprovable (TMS III.i.2). Note that for Smith the desire to be lovely or approvable is not the same as the desire to be loved or approved. You can approve of me, but it has to be for good reasons. I need to deserve your approval, not just receive your approval. This process of seeing ourselves with the eyes of another person eventually becomes a sort of second nature so that the other physical person does not need to be physically there to approve or disapprove of our conduct. We see ourselves through “the mirror of society,” developing the ability to see ourselves also, and especially, through the eyes of an imaginary person (TMS III.i.3). This imaginary person lives “within our breast” and, ideally, is able to see us as impartially as someone who is unrelated to us would (TMS III.2.32). This imaginary person, whom Smith calls the impartial spectator, is the one who tells us whether we deserve or not their approbation, whether we deserve or not the approbation of society, that is, whether our conduct is indeed approvable or not. So, Smith explains to us how our conduct is determined, yes, by ourselves, but through the presence of others, through this “mirror of society” without which we cannot see ourselves, judge ourselves, approve or disapprove of ourselves, and thus grow morally or even be ourselves. “I judge of your sight by my sight, of your ear by my ear, of your reason by my reason, of your resentment by my resentment, of your love by my love” (TMS I.3.10). Without the other, there is nothing of myself.

22

2

M. P. Paganelli

Moral Order

Rules of conduct emerge from these constant interactions with one another, by a constantly placing oneself in the shoes of another and imagining how they see us, how they would feel in our place, and whether they would approve of our conduct or not. So, for Smith, I look at you and approve of what you do. I admire your conduct. I also note that everyone else praises you for what you do. And so I make a note to myself that if ever in those circumstances, I should conduct myself as you do, because that conduct is worthy of praise, and I also want to be praiseworthy like you are. Similarly, if your conduct generates my disapprobation and the disapprobation of the people around me, I will know to avoid conducting myself like you because that kind of conduct is disapprovable and I do not want to be the proper object of disapprobation. Thus, for Smith, the recognition that I, like everyone else around me, approve or disapprove certain conduct is what generates general rules of conduct (TMS III.4). Rules of conduct are therefore the product of individuals, yes, but of sympathetic individuals, not seeking to maximize their utility or achieve some socially beneficial outcome, but seeking the approbation of others. And given that the rules of just conduct are what is socially approvable under specific circumstances, they can change with the circumstances, but not by a single individual. They are a sort of “relative absolute absolute” as James Buchanan would say, following his teacher Frank Knight (1944). They are absolute for an individual, but relatively so over time and place. I have been trying to avoid the use of the word behavior, following Smith and Wilson’s advice (V. L. Smith and Wilson 2019). Behavior tends to have the connotation of mechanical action. One individual reacts to a stimulus, without thinking, irrespective of the context, irrespective of the motivation for action. This is generally the use linked to behaviorism. It can imply some sort of Pavlovian response. Conduct, instead, is a word that implies an understanding of the circumstances and judgment of the motivation. It seems thus more appropriate to identify the sympathetic process that Smith describes when he describes and explains human actions and institutions. And in fact, the “as if ” assumption of an atomistic isolated individual has become increasingly inappropriate with the progress of inconsistent results from behavioral economic experiments. The more powerful explanatory models try to incorporate the possible interpretations of contexts and meanings that affect the actions of economic agents. Models that allow for an agent to put themselves in the shoes of the other players, using a sort of theory of mind, and then using approbation or disapprobation to explain their actions, seem to

Adam Smith and Sympathetic Individualism

23

be more effective than a simple rational choice agent in traditional economic models. So, for example, in an experimental game, someone may receive exactly the same payoff from another person or from a computer. They are given the opportunity to express their gratitude or their resentment, given the payoff received. Actions that imply gratitude or resentment are significantly more common if another person is responsible for the payoff received than if a computer is. The payoff is exactly the same. Robinson Crusoe is unable to distinguish between a human and a computer. He would act in the same way with both, as the rational choice model predicts. But a sympathetic individual, and individual who is embedded in society and social norms, would most likely recognize the difference in intentions between a human and a computer and react differently, as observed in economic experiments, many of which Smith and Wilson (2019) report and describe using a Smithian sympathetic apparatus.

3

Legal Order

Adam Smith uses the same “sympathetic” approach that he uses to explain conducts and its rules to also explain the legal order (Paganelli and Simon 2022). Laws do bring and keep order in society but are not meant to do that. The stability of society is a welcomed side effect of the action of individuals who seek the approbation of others, of social beings, of individuals who sympathize with each other, and who understand the intention of each other. For Smith, without this sympathetic process, there would be no legal system, if that is even possible in Smith’s mind. Laws are always present in society, even if the complexity of the legal system will vary with the complexity of the society. Smith thinks that our actions may generate gratitude or resentment in others. When something I do causes resentment in another person, the other person wants to revenge his or her harm against me. The problem is that this revengeful action may or may not be proportionate to the harm done. For Smith, you perceive the loss of your finger as a much bigger tragedy than the death of a multitude of people far away from and unknown to you (TMS III.3.4). Most likely, then, your reaction to my harm to you would be deserved but disproportionate. The punishment I may receive from you, being too much and being perceived by me as even larger than it is, would call for more revengeful action from me, which would ignite a vicious spiral of violence. Laws and the legal system thus emerge, for Smith, as a way to bring

24

M. P. Paganelli

appease to the individual injured, but in a way that allows their punishment to be appropriate, to be proportion to the harm. Laws do not originate from utility, but from resentment. An impartial observer looks at the harm that I did to you and sympathizes with you because you correctly resent my action. But this impartial spectator is able to evaluate the gravity of the harm and find the appropriate punishment for me, the one that will appease your resentment without causing resentment in me. This impartial spectator reflects the view and sentiments of all other members of our society. So again, if one person, as well as many others in the same group, resents an action that caused harm to someone else, they will sympathetically approve and also approve of an appropriate punishment that will appease the resentment of the victim. We see Smith’s sympathetic individualism again. I did something that generated resentment in another because of some harm associated with my action. The other person’s resentment calls for my punishment. Other people see the situation, through the sympathetic process, approve of the resentment of the victim and thus of my punishment. And so I am punished, and my punishment is perceived as well-deserved and thus approved of. In Smith’s account, all revolves around the individual resentment and appeasement of it. It is not a rational calculation of how to maximize utility either of the individual or of society. Punishment is not about social order. Punishment is not about deterrence of future harm. It is about the resentment of a person and the approval of that resentment that another person would feel if they with their imagination would put themselves in the shoes of the victim (and when putting themselves in the shoes of the person who caused harm they would not approve of their action). For Smith, there is a sympathetic individual at the base of the explanation of a legal system as well as the base of the development of rules of just conduct. And it is the same sympathetic individual at the base of the economic order.

4

Economic Order

As early as 1927, Glenn R. Morrow recognized that Smith was consistent in his account of human beings, that both his books are about individualism, and that thus there was no Das Adam Smith Problem. The so-called Das Adam Smith Problem is an alleged problem of consistency in Smith’s thought, created by scholars in the German Historical School at the end of the nineteenth century (Montes 2003). The claim was that the individual that Smith described in the Theory of Moral Sentiments was an other-regarding

Adam Smith and Sympathetic Individualism

25

individual, moved by sympathy (which they read as a form of benevolence), while the one in the Wealth of Nations was a selfish individualistic one, closer to the mythical homo economicus. But the individualism that Morrow describes, even in Smith’s economic work, is one based on sympathy (Glenn R. Morrow 1923), and not on selfinterest as Lorenzo Infantino claimed much later (1998), when he stresses the similarities between Smith and Mandeville, from whom Smith actively wanted to distance himself. And this individualism is even more than the individualism mediated by society that Kenneth Arrow (1994) claims to be typical of Smith. Prices are indeed the results of individual actions mediated by society since no individual can determine a market price. But society plays a much larger role in Smith than in price formation, at least in the sense of the presence of other individuals. In fact, we find the same sympathetic individual at the base of the economic order in Smith. The intrinsic sociability of human beings in Smith’s account is slowly making its way back also in the economic literature (Hurtado and Paganelli 2023; Levy 2002; Peart and Levy 2005), but there are taller walls to climb here because of the association between Adam Smith’s “individual” with the economic actor that is homo economicus, the Robinson Crusoe rational utility maximizer. The two could not be farther apart. For Robinson Crusoe, meeting Friday or discovering a more potent fertilizer have the same effect: they both can increase Robinson Crusoe’s consumption. For Smith exchange is not just a way to increase individual consumption. For Smith, it is a fundamental expression of our humanity. We have an innate propensity to truck, barter, and exchange. We enjoy bargaining. We feel a mutual pleasure in exchanging (Bee 2021) and yes, with it, we increase our consumption. But for Smith, we consume not just because we are hungry, but because eating together is a way of giving and receiving the approbation of others. Big banquets are a way to show off our wealth, which is in its turn a way to attract other people’s approbation. Westminster Hall was a dining room of William Rufus, often not big enough to entertain all his guests (WN III.iv.5). We clothe ourselves to protect ourselves from the elements, yes. But we clothe ourselves also, and because, we do not want to be the object of the disapprobation of others. A working poor person in Smith’s Britain would be ashamed to go without a linen shirt (WN V.ii.k.3). And when we can afford it, we clothe ourselves because the magnificence of our dresses will attract the attention and the admiration of others (WN V.i.g.12). I thus cannot share Andy Denis’s (1999) view that there is no individualism in Smith’s thought because, as he claims, Providence takes care of

26

M. P. Paganelli

everything. Providence is the real source of social order so that social order is not dependent on individuals’ choices (cf. C. Smith 2023). This view, in my opinion, limits Smith’s analysis by eliminating the role of sympathy in human relations. Furthermore, if there is a providential order, it is limited to a very small part of the world. It may be true for Britain or France, that despite human defects, things turned out for the better as people in Britain or France are better off in Smith’s time than many centuries earlier. But it would be a stretch to claim that the indigenous populations of the Americas would be protected by the same Providence, given that they have been brutally massacred by their European conquerors (for example, WN IV.vii.c.63 and WN IV.vii.b.59). Is Providence also generating a beneficent order in Bengal, where the East India Company is generating a famine of such dimension that in a fertile land hundreds of thousand people die of starvation each year (WN I.viii.26)? Or what about the people of China? Are they also benefitting from the same Providence, given that they face the same economic conditions as at the time of Marco Polo and that they are still drowning children like puppies because they can’t afford to raise them to adulthood (WN I.viii.24)? But this paradoxical providential order doesn’t have to be there if we seriously look at the sympathetic individual that Smith presents consistently in all of his works. Our desire to receive the approbation of others is, for Smith, what drives us to gain and accumulate wealth. It is our desire to receive the approved looks of others that drives us to save and accumulate capital, so fundamental for economic growth. But it is the same desire to attract the attention and approbation of others that induces the rich landlords of feudal Europe to sell their birthrights and become “insignificant” to buy a pair of diamond buckles that they alone can have (Paganelli 2009). It is the same desire to receive the approbation of others, through the parade of wealth, which drove Alexander the Great to unthinkable conquests and unthinkable crimes (TMS VI.iii.28). It is the same desire for glory that led Christopher Columbus to the absurd pursuit of Eldorado (WN IV.vii.a). And it is the same desire to accumulate and parade wealth so that we can receive the looks and admiration of others that incentivize merchants and manufacturers to ask and gain monopolies to enrich themselves at the expense of everyone else (Paganelli 2023a). What else could explain why the Chinese mandarins oppress and steal from their subjects so that they have no incentives to innovate or produce more than subsistence (WN I.ix.15)? It is only when the Smithian “individualism” is understood at its sympathetic root, that it can offer a complete explanation, not only of the functioning social orders but also of the dysfunctional ones. Smith is very well

Adam Smith and Sympathetic Individualism

27

aware indeed that when we place ourselves in the shoes of another person to judge their conduct and approve or disapprove of it, we are biased. We generally admire the rich and despise the poor, for no better reason than that the glamor of wealth is more appealing than the lack of it. The glittering of wealth attracts our attention (TMS I.iii.3.2). The dirt of poverty is invisible. There is little to admire in poverty. There is a lot to admire in a golden toothpick or a nail clipper, or a diamond buckle, which nobody else has (TMS IV.i.8). And so we have the poor man’s son who works hard all his life, but honestly so, to ride in a carriage (TMS IV.i.8). We have a prudent man who wisely saves and gradually accumulates wealth (TMS VI.i.7–19). We have our butcher, brewer, and baker, who gives us dinner in exchange for money (WN I.ii.2). And unintentionally we have markets, we have wealth, we have order. But we also have Alexander and Columbus, we have the East India Company and the Chinese mandarins, with all their atrocities and all their glory. We have empires driven by dreams of individuals who put themselves in a bystander’s shoes and admire the rich, so they also seek the approbation of others through gaining as much wealth as possible. We see ourselves only through the mirror of society. That mirror may be a distorting one, but nevertheless, for Smith, there is no other way to be an individual if not through the eyes of another.

References Arrow, K. J. (1994). Methodological Individualism and Social Knowledge. The American Economic Review, 84 (2), 1–9. Bee, M. (2021). The Pleasure of Exchange: Adam Smith’s Third Kind of Self-Love. Journal of the History of Economic Thought, 43(1), 118–140. Carrasco, M. A. (2022). Adam Smith and the Ethics of Care. Presented at the International Adam Smith Society meeting in Bogota. Danis, A. (1999). Was Adam Smith an Individualist? History of the Human Sciences, 12(3), 71–86. Friedman, M. (1953). The Methodology of Positive Economics. In Essays in Positive Economics (pp. 3–45). Chicago: The University of Chicago Press. Hayek, F. A. (1948). Individualism True and False. In Individualism and Economic Order (pp. 1–32). Chicago: The University of Chicago Press. Heath, J. (2020). Methodological Individualism. Summer 2020 Edition. Retrieved from https://plato.stanford.edu/archives/sum2020/entries/methodological-indivi dualism/. Hurtado, J. (2006). Rawls Y Smith: de la utilitad de la ‘simpatia’ para una conception liberal de la justicia. Estudios Publicos, 104, 89–111.

28

M. P. Paganelli

Hurtado, J. (2008). Smith, Rawls y el liberalismo simpático. In J. I. Gonzalez & M. Perez (Eds.), Pluralismo, legitimidad y economia politica. Ensayos criticos sobre la obra de John Rawls (pp. 57–76). Bogota: Universidad National de Colombia. Hurtado, J., & Paganelli, M. P. (2023). Diamonds Are Not Forever: Adam Smith and Carl Menger on Value and Relative Status. Review of Austrian Economics. Infantino, L. (1998). Individualism in Moderrn Thought. London: Routledge. Knight, F. H. (1944). Realism and Relevance in the Theory of Demand. Journal of Political Economy, 52(4), 289–318. Levy, D. M. (2002). How the Dismal Science Got Its Name: Classical Economics and the Ur-Text of Racial Politics. Ann Arbor: University of Michigan Press. Long, N. (2012). Utopian Sociality. Online. The Cambridge Journal of Anthropology, 30 (1), 80–94. Macpherson, C. B. (1989). Individualism. In J. Eatwell, M. Milgate, & P. Newman (Eds.), The Invisible Hand (pp. 149–155): Palgrave. Menger, C. ([1891] 2016). The Social Theories of Classical Political Economy and Modern Economic Policy, E. Dekker & S. Kolev (Trans.). Econ Journal Watch, 13(3), 473–489. Montes, L. (2003). Das Adam Smith Problem: Its Origins, the Stages of the Current Debate, and One Implication for Our Understanding of Sympathy. Journal of the History of Economic Thought, 25 (1), 63–90. Morrow, G. R. (1923). The Significance of the Doctrine of Sympathy in Hume and Adam Smith. The Philosophical Review, 32(1), 60–78. Morrow, G. R. (1927). Adam Smith: Moralist and Philosopher. Journal of Political Economy, 35 (3), 321–342. Paganelli, M. P. (2009). Approbation and the Desire to Better One’s Condition in Adam Smith: When the Desire to Better One’s Condition Does Not Better One’s Condition and Society’s Condition. Journal of the History of Economic Thought, 31(1), 79–92. Paganelli, M. P. (2023a). Adam Smith and the Morality of Political Economy. In P. Sagar (Ed.), Interprting Adam Smith: Critical Essays (pp. forthcoming): Cambridge University Press. Paganelli, M. P. (2023b). Smith at 300: The Dignity of Trade. Journal of the History of Economic Thought, 45 (2), 193–194. Paganelli, M. P., & Simon, F. (2022). Crime and Punishment: Adam Smith’s Theory of Sentimental Law and Economics Journal of the History of Economic Thought, 44 (2), 268–287. Peart, S., & Levy, D. M. (2005). The “Vanity of the Philosopher” from Equality to Hierarchy in Postclassical Economics. Ann Arbor: University of Michigan Press. Sayre-McCord, G. (2013). Hume and Smith on Sympathy, Approbation, and Moral Judgment. Social Philosophy and Policy, 30 (1–2), 208–236. Smith, A. ([1759] 1982). The Theory of Moral Sentiments. Indianapolis: Liberty Classics.

Adam Smith and Sympathetic Individualism

29

Smith, A. ([1762–3; 1766] 1978). Lectures on Jurisprudence (The Glasgow Edition of the Works and Correspondence of Adam Smith ed.). Indianapolis: Liberty Fund. Smith, A. ([1776] 1981). An Inquiry into the Nature and Causes of the Wealth of Nations. Indianapolis: Liberty Classics. Smith, C. (2023). Adam Smith and Spontaneous Order. In Burgio (Ed.), forthcoming (pp. 3–17). Routledge. Smith, V. L., & Wilson, B. J. (2019). Humanomics: Moral Sentiments and Wealth of Nations for the Twenty-First Century. New York: Cambridge University Press. Watkins, J. W. N. (1952). The Principle of Methodological Individualism. The British Journal for the Philosophy of Science, 3(10), 186–189. Wilson, B. J. (2020). The Property Species: Mine, Yours, and the Human Mind . New York: Oxford University Press.

Max Weber and Understanding Explanation Wolf Feuerhahn

Dedicated to Catherine Colliot-Thélène (1950–2022)

In the methodology of the human sciences, to speak of “understanding explanation”1 could be seen as confusion. Traditionally, in these disciplines, the 1 Here we mobilize the term “understanding explanation” to point out that there is no radical opposition between “explanation” and “understanding” in Weber, but the adjective “understanding” does not in any way mean “empathetic.” Weber later coined this expression (verständliche Erklärung or verstehendes Erklären). These expressions can be found in the “Über einige Kategorien der verstehenden Soziologie” articles (On some categories of interpretive sociology, 1913) and “Der Sinn der ‘Wertfreiheit’ der soziologischen und ökonomischen Wissenschaften” (“The meaning of ‘value freedom’ in the sociological and economic sciences”), which was first published in the journal Logos, co-directed by Rickert in 1918 (Band 7): cf . Max Weber, Gesammelte Aufsätze zur Wissenschaftslehre, Tübingen, Verlag von J. C. B. Mohr (Paul Siebeck), 1922/1988 (from now on abbreviated: WL). WL 428, 533 and 503: ihn verstehend zu erklären. Hans-Henrik Bruun translates “intelligible explanation” (Max Weber, Collected Methodological Writings, translated by Hans Henrik Bruun, edited by Hans Henrik Bruun and Sam Whimster, London; New York: Routledge, 2012, 274 (from now on abbreviated: CMW ); CMW , 329: “explanation by means of understanding.”

This work was first published in French in 2005: Feuerhahn, Wolf (2005). Max Weber et l’explication compréhensive. Philosophie, 85, pp. 19–41. Catherine Colliot-Thélène and JeanPierre Grossein contributed their advice to the work.

W. Feuerhahn (B) History of Sciences, Centre Alexandre-Koyré, National Centre for Scientific Research and École des Hautes Études en Sciences Sociales, Paris, France e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 N. Bulle and F. Di Iorio (eds.), The Palgrave Handbook of Methodological Individualism, https://doi.org/10.1007/978-3-031-41512-8_3

31

32

W. Feuerhahn

distinction between “explaining” and “understanding” constitutes a methodological dividing line; to defend the possibility of an “explanation” of human actions is to accentuate the proximity between “human sciences” and “natural sciences,” while to maintain that they must be the object of “understanding” is, on the contrary, to mark the gap between these two poles.2 The thesis of the “understanding” specificity of the human sciences is defended by many researchers who wish, for example, to insist on the irreducibility of the mind to inert nature. In this case, they define “understanding” as an attitude that is either “reflective” or “empathic,” capable of grasping the nature of man. Conversely, those who defend the unity of scientific practices often deny that human phenomena are of a different order from the phenomena of nature. Beyond their disagreements, both share the conviction that the nature of the object analyzed determines the method to be used. Questions of philosophy of science thus seem to be inseparable from the traditional questions of moral philosophy, particularly those relating to the free or determined character of man.3 These debates—whether conducted by philosophers, historians, or specialists in the various sciences of man and society—have always been very lively. Max Weber, considered by the official historiography as one of the founding fathers of sociology,4 is in turn claimed by each of the protagonists of these polemics, who present him respectively as a guarantor capable of legitimizing the position adopted. In order to deepen our understanding of this question, we want to clarify Max Weber’s position in the debate that took shape in his time. In order to do so, it seemed necessary to scrutinize5 one of his articles, in which he explains his position: “Roscher and Knies and the logical problems of historical political economy” published in the Jahrbuch für Gesetzgebung, Verhandlung und Volkswirtschaft im Deutschen Reich directed by G. Schmoller (1838–1917)

2

For a philosophical reconstruction of these debates, see Apel (1979, 33–125). Cf. Colliot-Thélène (2001, 133–168). 4 Les étapes de la pensée sociologique by Raymond Aron (Gallimard, 1967, here 4th edition: 1992) is a paradigmatic example in France. The structure of this book should be analyzed closely, starting by identifying the order chosen: a first part groups together the “founders,” Montesquieu, Comte, Marx and Tocqueville, and a second, “the generation of the turn of the century,” of which the three representatives studied, “Durkheim, Pareto and Weber have in common the will to be scholars” (Aron 1992, 309). Weber, analyzed last, has the privilege of being “our contemporary” (Aron 1992, 564– 571). R. Aron copied his title from Les étapes de la philosophie mathématique (Alcan 1912) published by the director of his main thesis, Léon Brunschvicg, for whom, according to G. Canguilhem, “history is, in reality, for reason, the conquest of its essence that it pursues through its exercise” (Canguilhem 1999, 11). This perhaps explains why “having set out to find the origins of modern sociology, [Raymond Aron notes that he] ended up with a gallery of intellectual portraits” (Aron 1992, 17), these portraits constituting the stages through which sociological reason becomes aware of itself. 5 Weber (2023). 3

Max Weber and Understanding Explanation

33

between 1903 and 1906, before being republished in the Gesammelte Aufsätze zur Wissenschaftslehre 6 in 1922. In it, he tackled head-on the highly controversial question of the criterion for distinguishing between the sciences and questioned the idea that the nature of the object studied was the decisive criterion for classifying the sciences. In this way, he returns to the most commonly shared presupposition in the philosophy of science. Weber’s thought progresses in the opposite direction to the dominant methodological reflections, which in our opinion are wrongly based on his work. Our task is to elucidate these misreadings, which are essentially based on a misinterpretation of Weber’s relationship with philosophical enquiry, and more particularly with the neo-Kantian school. This is reflected not only in the tendency to portray Weber as a defender of the philosophy of values but also in the presentation of his work on the theory of science as the foundations of sociological practice. The Gesammelte Aufsätze zur Wissenschaftslehre and even more so the first of his methodological articles, on Roscher and Knies, are then read as the foundations of a philosophy of social science. This tendency is reinforced by the fact that Max Weber is often referred to as a philosopher. While it is true that Max Weber was very close to Heinrich Rickert and some of his students such as Emil Lask, the historiographical habit of simply making him a member of the Neo-Kantian school in Baden7 must be reconsidered. We shall see how much more complex the situation turns out to be on examination, which will help us to better understand the specificity that Weber attributes to cultural sciences (Kulturwissenschaften).

1

The Reflections on Science in Weber’s Work

Weber usually refused the title of philosopher. According to him, philosophical questions are indeed irreconcilable with the scientific objective he has set himself, as he considers them to be “world views” [Weltanschauungen]. Catherine Colliot-Thélène recalls how “he frequently emphasised that the strictly empirical character of his project of knowledge excluded consideration of the traditional questions of philosophy: those questions which concern the meaning of life and the world, as well as the ultimate justification of the norms of action. It left all this to ‘speculative thinking’ (WL 152), to the ‘metaphysical’ interpretation of the world (WL 156), to the ‘philosophy of 6

Henceforth abbreviated WL. The German secondary literature is immense, but let’s quote the following: Tenbruck (1959), Baier (1969), Wagner (1987), Schluchter (1988), Wagner and Zipprian (ed.) (1994); in French: Aron (1987).

7

34

W. Feuerhahn

values’ (WL 507; ETS 427). An empirical science must confine its ambition to the reasoned ordering of the empirical data: this is one of the leitmotivs of Weberian methodological texts.8 In this way, he acknowledged his debt to the neo-Kantian distinction between being and ought-to-be, between empirical science and philosophy of values.”9 With regard to questions of method, which Max Weber distinguishes from philosophical questions, an examination of how the Gesammelte Aufsätze zur Wissenschaftslehre came into being leads to a reconsideration of the claim that Weber was a “theoretician of science.” Although Weber had decided to compile his logical and methodological essays,10 this project was not completed during his lifetime and was carried out by his wife after Weber’s death. Weber did not want to be a philosopher of science, unlike G. Simmel (1858–1918), who was first and foremost trained as a philosopher, and who, from a neo-Kantian perspective, wanted to determine the conditions of possibility of historical sciences.11 Weber never formulated such an ambition. His articles on scientific practice are above all situational writings and do not aim at a philosophical foundation of the sciences of culture.12 Most of them are reviews: “Critical studies to serve the logic of the sciences of culture” (1906),13 “R. Stammler and the ‘overcoming’ of the materialist conception of history” (1907),14 “the theory of marginal utility and the ‘fundamental psychophysical law’” (1908), “‘energetic’ theories of culture” (1909). Some of his writings also remained unfinished.15 It is necessary here to trace the context in which the essays on Roscher and Knies take place. In 1897 Weber succeeded one of his teachers, Karl Knies (1821–1898),16 who died the following year, as Professor of Political Economy at the University of Heidelberg. Although he had not taught since 1900 due to a nervous breakdown, he was asked by his colleagues in the philosophy faculty of the University of Heidelberg to contribute to a

8

Colliot-Thélène (1990, 5–6). Colliot-Thélène (1990, 6–7), Colliot-Thélène (1992, 125–192). 10 See in particular a letter to his publisher Paul Siebeck dated 8 November 1919 (Max Weber Gesamtausgabe (MWG II/10), pp. 833–834). 11 For Simmel’s neo-Kantian turn: cf. Simmel (1905). 12 This does not mean that Weber should not be interested in Simmel’s texts. On the contrary, he quotes them many times in his own articles. 13 Translated in ETS, 203–299. 14 Weber (2001). 15 “Roscher and Knies,” “Critical studies in the logic of the cultural sciences,” “Rudolf Stammler’s ‘overcoming’ of the materialist conception of history.” 16 Max Weber studied with Karl Knies during his three semesters at the University of Heidelberg (1882–1883). 9

Max Weber and Understanding Explanation

35

volume to celebrate the anniversary of the university reform of 1803 (Marianne Weber 1950, 296, 347). This text was thus primarily the result of a purchase, and, according to Marianne Weber, it represented such a considerable strain on Weber that she calls it a “sighing article” [“Seufzeraufsatz”].17 Max Weber, unable to hand in his manuscript in time, published it in three parts between 1903 and 1906 in the journal edited by Gustav Schmoller. However, it remained unfinished; indeed, a fourth part, entirely devoted to Knies, should have followed. Moreover, Weber repeatedly emphasized that in matters of scientific methodology he was only an “amateur,” which is of importance for a scholar committed to grasping the processes of specialization. As a student in Heidelberg for three semesters between 1882 and 1883, his main subject was law. However, in accordance with the university practice of the time, this meant that he also took courses in history and economics.18 Letters from his youth show that he also attended the logic classes of Kuno Fischer.19 He was thus not only a knowledgeable amateur, but also fully aware of the degree of specialization required to deal with logical and gnoseological questions20 ; “the most important works of specialists in the theory of knowledge […] are therefore hovering so high above the heads of practitioners, that they find it difficult to recognise themselves with the naked eye in the midst of these discussions” (WL 216). But this does not invalidate the reflections of historians or non-specialist economists, on the contrary: “methodological discussions from their own milieu can sometimes help them to become more easily aware of their own problems, despite and, in a certain sense, because of the imperfect formulation of the theory of knowledge point of view” (ibid.)21 In his “Critical Studies in the Logic of the Cultural Sciences,” a critical account of the methodological work Zur Theorie und Methodik der Geschichte 17

Weber, Marianne (1950, 291). See Hennis (1996, 135). 19 Letter to his parents (2 May 1882, 1st semester): “I attend almost as assiduously […] Kuno Fischer’s logic class in the morning from seven to eight” (MWG II/1, p. 255). Kuno Fischer is famous for having delivered three lectures on Kant in Jena in 1860, which are considered to be the foundation of the “return to Kant” in German philosophy (Fischer 1860). On K. Fischer, cf. Köhnke (1993, 195–211) and Ferrari (2001, 16–18). 20 “At its present stage of development, logic is in fact a specialized discipline like any other” (“Critical Studies for the Logic of Cultural Sciences,” CMW ; WL 215). 21 A very similar quotation can be found on the first page of the article on Roscher and Knies: “In so doing, we shall only be concerned with elementary forms of these [logical and methodological] problems. Only this circumstance allows me to discuss them here, as I naturally do not have a professional command of the rapidly growing body of literature in the field of logic. But even scholars in specialized disciplines ought not to ignore those problems; and above all: although they are quite elementary, even their very existence is not widely recognized – as will also become apparent in the context of this study” (“Roscher and Knies” 1903, CMW , 3; WL 1). 18

36

W. Feuerhahn

[On the Theory and Method of History] published in Halle in 1902 by the ancient historian Eduard Meyer (1855–1930),22 Weber agrees with Meyer that the significance of methodological work for the practice of history should not be overestimated: “Methodology can never be more than a self-reflection on the means that have proved useful [bewährt ] in [scientific] practice; and one does not need to be made explicitly aware of those means in order to produce useful work, just as one does not need to have knowledge of anatomy in order to walk ‘correctly’” (CMW 140; WL 217). In no case should methodological reflections order the work of investigation. They do not constitute a prerequisite for research but are a tool for intellectual vigilance since they enable the historian “to avoid being impressed by the sayings of dilettantes posing as philosophers” (CMW 140; WL 217) (ibid.). More importantly, they allow one to face scientific crises: As a rule, such [methodological] considerations only acquire importance for the way in which science is pursued at times when a major shift occurs in the ‘viewpoints’ from which [the] material becomes an object of description, and this [shift] leads to the idea that the new ‘viewpoints’ also require a revision of the logical forms that have hitherto provided the framework for that ‘pursuit’, with the result that one becomes uncertain about the ‘nature’ of one’s work. But this is undeniably the situation in which history finds itself at present; and E[duard] Meyer’s view that, in principle, methodology is of no importance for the ‘practice’ [of history] has therefore, quite justifiably, not prevented him from now letting himself become engaged in methodological reflections]. (CMW 140–141; WL 217–218)

These kinds of reflections are only legitimate insofar as they are called for by the discipline concerned in order to denounce and elucidate a misuse of concepts. Weber’s gnoseological articles do not aim at founding a new science, but at clarifying the use of logical concepts by economists and historians.23 This is why most of his methodological essays are in the form of reviews. They only make sense when placed in their polemical context. The “Essays on Roscher and Knies” contain many indications of this context and thus provide a better understanding of Weber’s positions.

22

Cf. Geschichte des Altertums, 5 vols. 1884–1902, translated into French: Histoire de l’Antiquité 3 vols., Paris, 1912–1926. 23 In Feuerhahn (2023) I show the crucial role of the so called “Lamprecht controversy”. Between 1891 and 1908, the historian Karl Lamprecht published an ambitious History of Germany (Deutsche Geschichte 13 vols) which develops the idea of the existence of psychological laws of history. It sparked a huge methodological dispute within Germany’s academic history establishment, especially Max Weber, who habitually referred to Lamprecht as a mere dilettante.

Max Weber and Understanding Explanation

2

37

The Study of K. Knies Reveals Logical Confusion

Political economy has been in an open crisis since the beginning of the dispute over methods in 188324 between the leader of the historical school of economics, G. Schmoller (1838–1917), and the marginalist C. Menger (1840–1921). Weber denounced attempts to turn political economy into an “ethical science.”25 In this, he does not only criticize Schmoller’s work but also an older tendency, already present in Roscher and Knies, to confuse science and philosophy of values.26 Weber is therefore not a hagiographer,27 he does not provide “a literary portrait of [his] great masters” (CMW 3; WL 1), but wants to “show how certain elementary problems of logic and method, which during the last generation have been discussed in the science of history and in our own specialized discipline, manifested themselves in the early stages of historical economics, and how the first great achievements of the historical method have tried to come to terms with those problems” (CMW 3; WL 1). In this way, he inaugurates the kind of thinking that is used in almost all of his methodological texts. Weber does not analyze the work of Karl Knies exhaustively,28 but rather focuses on the problem of irrationality, showing that it is found in many gnoseological works at the time29 ; indeed, Weber acknowledges that Knies may even appear to the reader as a mere “pretext” for what is said (CMW 29; WL 43 note).

24

Cf. “The ‘Objectivity’ of Knowledge in Social Science and Social Policy” (1904): “We shall begin by stating our position on this question, and then go on to consider the further question: in what sense do ‘objectively valid truths’ exist at all in the domain of the sciences of cultural life? This question must be addressed, in view of the fact that even the seemingly most elementary problems of our discipline, its methods of work, the way in which its concepts are formed and the validity of those concepts, are subject to constant change and give rise to bitter strife” (CMW 101; WL 147–148). 25 “With the awakening of the historical sense, a combination of ethical evolutionism and historical relativism gained the upper hand within our discipline and attempted to divest ethical norms of their formal character, to give substantive content to the realm of ‘morals’ by incorporating into it the totality of cultural values, and in this fashion to raise economics to the dignity of an empirically based ‘ethical science’” (“The ‘objectivity’ of knowledge in social science and social policy” (1904) in CMW 101; WL 148). This passage targets Gustav Schmoller. 26 Apart from the constraints of the commission, the fact that Weber did not directly criticize Schmoller is probably explained by Weber’s complex position, which, while being a member of the Verein für Sozialpolitik led by Schmoller, held Menger’s logical rigor in high regard. 27 “It is only natural that this discussion will in many instances to a large degree also demonstrate the weaknesses inherent to those achievements” (“Roscher and Knies” 1903, CMW 3; WL 1). 28 Cf. WL 43 note: “Here, since it is only a matter of developing certain logical problems, we do not aim at such an exhaustive restitution” of Knies’ work. 29 See WL 46 note.

38

W. Feuerhahn

The general question posed by Knies’ Die politische Oekonomie vom geschichtlichen Standpuncte is the possibility of a science of human actions. According to Knies, two facts are likely to hinder the latter, namely the historical and national relativity of human behavior on the one hand, and its fundamental irrationality on the other. The historical relativity was highlighted by the historical school of economics’ questioning of the cosmopolitanism defended by classical economists.30 These economists, heirs of the natural law theorists,31 based the possibility of economic science on the universality of human reason. The universality of the object was the guarantee of the possible universality of the laws of economic behavior. Adam Müller and, following him, Friedrich List, were the first to criticize such a postulate from two different points of view.32 They emphasized the historical, even ontological, specificity of peoples and rejected the abstract universality proclaimed by classical economics. As for the notion of the irrationality of human behavior, it directly involved philosophical questioning and placed a heavy burden on the possibility of developing a science of human actions. It was considered to be the consequence of human freedom and risked preventing the identification of laws of economics. However, Knies does not give up on making political economy a science. He, therefore, defends a plurality of scientific regimes as a consequence of the ontological differences between their objects.33 The sciences of the mind are therefore opposed to the sciences of nature as the mind is opposed to nature, the mind being the realm of freedom, nature, that of necessity. The freedom of the mind is understood as irrational and therefore unpredictable. But Knies manages to preserve the idea of the regularity of economic behavior. According to him, the object of political economy is both irrational, because it is free, and necessary, because the conditions of human actions are determined by a historical context and by the necessity of nature. This leads to the specificity of the “historical sciences,” of which political economy is a part, and which constitute a third group of sciences between the “sciences of the mind” and the “sciences of nature.” Knies identifies a property of the “mind” that escapes radical irrationality. Even if the “personal (intellectual) element” (478) is not constant in the same way as a physical cause, there must be, according to Knies, “something eternal 30

See Hennis (1996, 145–154). Adam Smith combines his interest in the universality of natural law with his interest in Newtonian physical science. 32 Adam Müller in the context of counter-revolutionary thinking, and Friedrich List for a policy of defending the economy of the German states against the domination of the British economy. 33 Indeed, Knies emphasizes that “objective difference” is decisive for the classification of sciences (Knies 1964, 5–6). 31

Max Weber and Understanding Explanation

39

and identical in all human life and activity, without which no man could belong to the human race” (ibid.). This element must have its “home in the human soul, which determines economic activity like any other activity” (ibid.). The soul, as a characteristic of man, thus makes it possible, in the last instance, to affirm the constancy of the human spirit beyond the diversity of individual behavior. The human soul is not, however, anhistorical. It is part of the constant development that affects human life, both individual and collective. To account for this development, Knies defends an organicist conception of human societies. The economy has to identify the laws of development of peoples, which vary according to “national characteristics” (479). The specificity of the “historical sciences” is thus saved: the comparative work between economies of different times or nationalities cannot lead to identical laws, but only to “analogous laws” (479) of development.34 This explains why Knies was very interested in the statistical research also developed by his teacher Bruno Hildebrand,35 which makes it possible to discover regularities in human actions. However, Weber considered that in doing so he was inscribing them in a metaphysics described,36 following Emil Lask, as “emanatist”37 and for which reality is not opposed to the concept, but comes from it.38 In order to be grasped, the economic activity of a people has to be brought back to its essential “personality.” 34 This brief examination of Knies’ epistemology proves that Weber’s text cannot be reduced to “a real form of murder of the father” as Wilhelm Hennis (1996, 175) states. In this we agree with the criticisms of Hennis made by C. Colliot-Thélène (1992, 81) and F. Chazel (2000, 41–43). 35 Bruno Hildebrand published the Jahrbücher für Nationalökonomie und Statistik from 1863. Knies had already published in 1850 Statistik als selbständige Wissenschaft. Ein Beitrag zu einer kritischen Geschichte der Statistik seit Achenwall , Kassel, Luckhardt. 36 Cf. WL 138–145. 37 Emil Lask (1875–1915), author of a doctoral thesis (Promotion) praised by Weber (WL 16 note 1) and devoted to Fichtes Idealismus und die Geschichte (1902) under the direction of Heinrich Rickert, distinguishes between two theories of the concept, analytic and emanatist: the first “considers what is empirical as the unique and whole reality, and the concept as an artificially separated part, incapable of an existence of its own, […] and which turns out to be only a product of thought. Conceptualization is achieved here by analyzing the immediate given; we can call analytical logic that which defends this point of view. [In the case [of ] [emanatist] theories, the concept must always necessarily appear endowed with a content richer than empirical reality, it must not be thought of as a part of empirical reality, but on the contrary be represented as encompassing the latter insofar as it is a part of itself, an emanation of its supra-real essence [überwirklich]. […] Since the concept delivers each of its realizations, so to speak, from its extreme richness, a conception that promotes such results can be called emanatist logic ” (Lask Gesammelte Schriften, Tübingen, J.C.B. Mohr, 1923, Band 1, 29–30, we translate). Hegel is, according to Lask, the best representative of an emanatist logic. But the historical schools of law and economics would also advocate the insertion of axiological individuality into an axiological totality (Lask op. cit., 21). 38 Weber emphasizes the extent to which Knies inherits from the historical school of jurisprudence: “It soon becomes apparent that – and in what sense - Knies, too, was completely under the sway of that historical version of the “organic” theory of natural law which - in Germany, mainly under the influence of the historical school of jurisprudence - permeated all the areas where human cultural activity was investigated” (CMW 89; WL 138).

40

W. Feuerhahn

The complex gnoseological elaboration proposed by Knies would thus be determined by a metaphysical thesis concerning human freedom. The criticism is harsh from Weber and from historians who, since Leopold Ranke, have been hostile to the mixture of science and metaphysics. Worse still: for Weber, this metaphysics would be very widespread in historical studies in Germany at the time, and particularly among the defenders of specificity in relation to the natural sciences. In general, the introduction of metaphysical considerations on human freedom would lead to recurrent confusion among many historical researchers.39 Human freedom is said to be reflected in the unpredictability of human action, which is itself often seen as an indication of a particular “dignity” of the sciences of human action.40 Weber denounces the idea that the dignity of science depends on the unknowable character of its object.41 He rejects the implicit value judgment that only unpredictable action can be considered, and that only its “understanding,” defined as an empathic relationship, allows us to approach human free will. Finally, he underlines the fact that the different conceptions of the sciences of the mind are based on a double postulate: the first presupposes a difference in nature between human actions and natural phenomena, the second that only two exclusive methodological possibilities would be conceivable, nomological explanation or understanding.

3

Ontology and Gnoseology

By criticizing Knies and many representatives of the historical sciences for determining the methodological characteristics of their sciences according to the nature of their objects, Weber contests the relevance of an ontological foundation of the sciences. This type of foundation had already been denounced by Wilhelm Windelband (1848–1915) and his pupil Heinrich Rickert (1863–1936) who had targeted Dilthey’s Einleitung in die Geisteswissenschaften.42 In his Inaugural Rectoral Address in Strasbourg, entitled “Geschichte und Naturwissenschaft” (History and Natural Science), delivered

39 Weber is not content to denounce such a tendency en bloc, but analyses the positions of Knies, Wundt and Münsterberg separately and in their specificity. 40 See WL 46 and 67. 41 See WL 46 note by Weber. 42 As Windelband with Dilthey, Weber accuses Knies of “taking for granted that the scientific ‘division of labour’ consists in apportioning the objectively given factual material, and moreover, that it is this material, assigned on objective grounds to each science, which dictates the method of that science” (CMW 30; WL 44).

Max Weber and Understanding Explanation

41

in 1894, Windelband opposed a distinction between natural and spiritual sciences based on the ontological difference between nature and spirit: In the division of these disciplines turned towards the knowledge of reality, the separation between the sciences of Nature and the sciences of Spirit is common at present; I do not consider that in this form it is happy. Nature and Spirit is an objective antithesis that was asserted towards the end of ancient thought and at the dawn of medieval thought, and which has been preserved, in all its rigidity, in modern metaphysics, from Descartes and Spinoza to Schelling and Hegel. (Windelband 1894, 22–23)

For Windelband, faithful to transcendental thought, phenomenal nature cannot be considered as external to the mind; on the contrary, he conceives it as the correlate of the mind’s activity. Thus, it is according to the cognitive purposes of the mind that we can distinguish the different sciences: We are faced here with a purely methodological division of the sciences of experience [...]. The principle of division is the formal character of their goals of knowledge. Some seek universal laws, others particular historical facts. To use the language of formal logic: the aim of some is the general apodictic judgement, that of the others the singular and assertoric proposition. [...] Some are the nomological sciences [Gesetzeswissenschaften], others the sciences of the event [Ereigniswissenschaften]; the former teach what is always, the latter what was only once. If I may coin a new expression, I would say that scientific thought is nomothetic in the first case and idiographic in the second. (Windelband 1894, 25–26)43

A student of Windelband and a friend of Weber, Rickert takes up and deepens this critique of the ontological criterion of distinction between the sciences. He refutes the opposition between “physical world” and “psychic world,” which he reads in Dilthey44 and which, according to him, cannot be a valid criterion of distinction between the sciences of nature and the sciences 43

Windelband takes up the Kantian classification of judgements according to their modalities: “problematic” judgements (in which affirmation and negation are accepted as merely possible),“assertoric” judgements (those in which affirmation and negation are regarded as actually real) and “apodictic” judgements (those in which affirmation or negation are regarded as necessary). 44 In fact, Dilthey distinguishes between the sciences according to the mode of donation of their objects: “the sciences of the spirit are distinguished first of all from the sciences of nature in that the latter have as their object facts which present themselves to consciousness as phenomena given in isolation and from the outside, whereas they present themselves to the former from the inside, as a reality and an original living whole” (Dilthey 1947, 148). For a very precise analysis of the debate between Dilthey, Windelband and Rickert see: Gens (2002, 31–53) and on this precise point 50–51 as well as Mesure (1990, 141–167). It seems to me, however, that Rickert is aware that Dilthey thinks in terms of “modes of donation” and not of mere ontological characteristics since he retorts to

42

W. Feuerhahn

of culture. Weber inherits these analyses and even sees it as one of the aims of the article on Roscher and Knies to test the fruitfulness of Rickert’s thoughts for the historical sciences in general.45 He thus borrows the new, formal way of characterizing regimes of scientificity.46 Thus, in the manifesto article of 1904, he specifies: “The fields of inquiry of scientific disciplines are based not on ‘concrete’ relations between ‘things’, but on ‘theoretical ’ relations between ‘problems’” (CMW 111; WL 166). Like Rickert,47 Weber emphasizes that scientific knowledge involves the construction of concepts, which elaborate and transform the empirical given to form specific questions. In order to determine the specificity of science, it is, therefore, necessary to start not from an ontological characterization of its field of investigation, but from its point of view, from its own aim of knowledge.48 Reality does not make sense in itself, it is the scientific interrogation that gives it meaning. The historian or economist does not study pre-existing realities but tries to explain problems, such as the appearance of a new type of man with a rational way of life. This is why all scientific discourse is relative to the point of view adopted. The use of expressions such as “causal need” (kausales Bedürfnis),49 “cognitive purpose” (Erkenntnisziel ), or “interest” (Interesse) is an indication of this perspectival determination of scientific questions. Since each science is defined by the point of view from which it seeks to grasp reality, Weber rejects any quest for fundamental science. There can him that all scientists start “from given immediate facts of consciousness and gradually rise to their systems of concepts” (Rickert 1929, 145). 45 “In the preceding account, I believe that I have more or less faithfully reflected the main points of view – to the extent that they are relevant for us – in Rickert [’s Limits of Concept Formation]. It is one of the purposes of this study to test [the possibility of using [Rickert’s] ideas in the methodology of our discipline. I therefore do not refer to him in every single case where I would normally have been required to do so” (CMW 6 footnote 7; WL 7 footnote 1). 46 “In the further course of our discussion, we shall frequently refer to this distinction, whose significance for the methodology of economics was to a certain extent already (as will be noted later) recognized by Menger, although his conclusions were in part incorrect. After early attempts, to be found in Dilthey (Sciences of the Human Spirit ) and Simmel (Philosophy in History), the precise logical formulation [of this distinction] was first, on important points, briefly sketched out in Windelband ’s Inaugural Rectoral Address (History and Natural Science ), and then comprehensively elaborated in Heinrich Rickert’s fundamental work (Limits of Concept Formation)” (CMW 4–5 footnote 3; WL 3 footnote 2). 47 Rickert (1921, 163 quoted in Ferrari 2001, 128). See also Dufour (2003, 128): “For the Heidelbergians, the distinction between two types of sciences, the sciences of nature on the one hand and the sciences of culture on the other, is therefore not based on the object, nor even properly on the method, but on the aim of each science (its “point of view”), on the goal it pursues (Erkenntnisziel) and which determines its method.” 48 “when new methods are used to investigate a new problem, and this leads to the uncovering of truths that open up new, significant perspectives, then a new ‘science’ comes into being” (CMW 111; WL 166). 49 Cf.: WL 67, 68, 69 in particular (CMW 43–44). This point has already been stressed by Catherine Colliot-Thélène (2001, 158).

Max Weber and Understanding Explanation

43

be no absolute point, no totalizing perspective. Each science investigates specific phenomena, thus allowing the perspective of the other sciences to be completed, complexified, and nuanced, but none of them can be found in the others. While many scientists at the time perceived psychology as a possible equivalent of Newtonian physics for the sciences of the mind, Weber contested that it could be the science of the foundations of human action. He insists on taking into account the diversity of scientific practices defined at the time as psychological50 and underlines the erroneous character of this thinking, which stems from the fact that these specialists in the sciences of the mind characterize the sciences according to their ontological foundation: psychology, the science of the human mind, would have as its object the essence of man and would therefore be first in the hierarchy of the sciences of the mind. According to this conception, social actions would derive from psychic phenomena conceived as elementary, social science being, as it were, a psychology applied to phenomena of greater scope. Weber rejects any ontological starting point that would consist in determining an elementary reality. He starts, on the contrary, from science as it is done, from its work of questioning and elaboration of concepts. For this reason, Weber denounced the idea, common at the time, that the specificity of the sciences of the mind lay in the irrationality of human actions, an idea that ultimately came back to ontological considerations: the irrationality of human actions would be the consequence of man’s own freedom. Weber, for his part, certainly continues to use the notions of “rationality” or “irrationality,” but never in an ontological sense. He redefines them in a perspectivist way: the points of view adopted can lead to judging a phenomenon as either rational or irrational; “what is ‘rational’ may appear ‘irrational’ according to the point of view adopted.”51

4

Cause and Law

By subordinating ontological questions to scientific ones, Weber certainly provides an initial response to the question of the plurality of scientific regimes, which was strongly claimed at the time; but this is not enough to exhaust the question. Having rejected any ontological criterion, Weber had to look for a formal criterion that would allow him to distinguish between

50

Weber underlines the equivocality of the term “psychology” itself, which at the time refers to very different practices and objectives. Experimental psychology, for example, does not invoke the “irrationality of human actions.” One of the aims of our thesis is to elucidate the specificity of Weber’s criticism of each of these psychologies. 51 Max Weber (1996, 502–503).

44

W. Feuerhahn

different types of science. However, he reproaches Knies, in particular, for having confused cause and law and for having thus prevented any discovery of a formal criterion.52 This frequent confusion leads many specialists in the sciences of human action to think that scientific explanation is necessarily nomological and must therefore be opposed to understanding, defined as an empathetic revival of acausal phenomena.53 Weber insists, on the contrary, on the difference between the principle of determinism on the one hand, according to which all phenomena are linked to each other by invariable and necessary relations, and the principle of causality on the other hand, for which, less strictly, all phenomena proceed at least from a cause. While the first applies primarily to the procedures of the sciences that seek to formulate universal and necessary laws, it cannot account for the specificity of scientific practices that seek to explain phenomena produced by several causes or to explain that certain causes can produce various effects. This is the characteristic of the “sciences of reality” [Wirklichkeitswissenschaften], a regime of scientificity that Weber seeks to distinguish from the “nomological sciences” [Gesetzeswissenschaften], the characteristics of which are summarized in the following table.54 Features

Types of science Nomological sciences

Sciences of reality

Task

Ordering infinite, extensive and intensive diversity with a system of concepts and laws that can be universally valid in the most absolute way possible

Logical ideal

Establish a system of universal laws and concepts

Specific logical means

Use of concepts whose extension is ever increasing and whose understanding is ever more restricted

Knowing reality in its singularity and uniqueness = knowing the elements of reality which, in their individual specificity and because of it, appear to us as essential Separating what we consider “characteristic” from what is contingent in individual phenomena Formation of relationship concepts with ever greater understanding and ever smaller extension (continued)

52

“For Knies, as for Roscher, causality [Kausalität ] is identical to law-like regularity [Gesetzmässigkeit ], and he therefore, as a matter of course, regards the influence of natural and ‘general’” (CMW 30; WL 44–45). 53 In Knies’ case, in order to avoid giving up explaining economic phenomena, as we have seen, it leads to the invention of analogous laws of development. 54 This table is based on pages 4–6 of the German edition of Roscher and Knies.

Max Weber and Understanding Explanation

45

(continued) Features

Types of science Nomological sciences

Sciences of reality

Specific logical product

Concepts about relationships with general value (laws)

Scope of practice

Wherever what we consider essential in phenomena can be subsumed under a generic category

Individual concepts about things with historical significance Wherever concrete reality interests us as such

The term “sciences of reality” deserves an explanation. Rickert recalls that, although he inherited the formal distinction introduced by Windelband between “nomological sciences” and “idiographic sciences,” Weber most often avoided the term “idiographic,” which seemed to him to be too aestheticizing and which risked reintroducing the opposition between “explanation” and “understanding.”55 However, by speaking of “sciences of reality,” was he not reintroducing a “material” criterion for distinguishing the sciences? If this were the case, the logical parallelism with the sciences that formulate laws would be lost. But “reality” is understood here as a logical category to be distinguished from “necessity,” which is derived from the nomological sciences.56 The task of the sciences of reality is defined from the limit of the nomological sciences: And on the other hand, [we have] sciences which set themselves the task which the sciences of laws described above are necessarily prevented from fulfilling, owing to the logical nature of their approach: [the task of ] attaining knowledge of reality, with its constant and universal character of qualitative differentiation and uniqueness. However, since it is in principle impossible to reproduce exhaustively even the most limited part of reality with its (at least always intensively) infinite differentiation from all the other [parts of reality], [the object

55

“The ‘idiographic’ practice, he thought, leads to an aestheticism” (Rickert 1929, XXIII). “The expression ‘science of reality’ does correspond completely to the logical nature of history. Rickert and Simmel have already dealt sufficiently firmly with the mistaken view reflected in the popular interpretation of that expression as indicating a pure ‘description’ without preconceptions” (CMW 152; WL 237). Simmel was the first to use the term “Wirklichkeitswissenschaft” to point out the specificity of the historical sciences in relation to the nomological sciences, the “science of history” being the “science of reality par excellence” (Simmel 1892, 43). On the other hand, he insists on the logical character of the distinction between “nomological sciences” and “sciences of reality”: “From a logical-conceptual point of view, there is the greatest difference between these two [types of sciences] that can exist in the field of knowledge” (42), “this solely logical divergence between the intentions [Absichten] of the historical and nomological sciences” (43). 56

46

W. Feuerhahn

of this second group of sciences must be] knowledge of those constituent parts of reality which, with their individual distinctive character and because of that character, are in our eyes the essential ones. (CMW 5; WL 5)

The interest in real singularity cannot be satisfied by establishing universal laws. This does not imply, however, that one should renounce any scientific explanation of real singular phenomena. Simply, the criterion for an “explanation” is first and foremost the principle of causality, not determinism. The scientific explanation does not primarily involve the formulation of “laws.” Rather, in the explanation of concrete processes, “the possibility of formulating judgments of causal necessity is by no means the rule, but [rather] the exception” (CMW 43; WL 66). Weber certainly does not claim that there are lawless causalities, but that it is necessary to distinguish between cognitive purposes: is it a question of identifying and formulating universal laws or of explaining a singular phenomenon? The science of reality must above all bring to light singular causalities and not universal laws, as these do not account for the singular complexity of real phenomena. If Weber criticizes historical materialism, it is not because it wants to make history a science, nor because it has analyzed social and cultural phenomena from the point of view of their economic determination, but simply because some of its defenders have used it “as a ‘universal’ method, in the sense of a deduction of all cultural phenomena […] as being, in the final analysis, economically determined” (CMW 112; WL 168). This causal monism acquires a nomological modality and loses sight of the complexity of singular phenomena. Criticizing the confusion between cause and law thus allows Weber to make the explanation of real phenomena more complex without having to give up their explanation.57 He thus opens the way to a causal pluralism58 —this causal pluralism does not imply an exhaustive reproduction of the complexity of real singular phenomena. Causality is, according to Weber, “a category of ‘our’ thinking” (WL 290) which allows us to isolate certain phenomena as “adequate causes” within an infinite number of conditions with the help of the category of “objective possibility” (cf. WL 266–290; CMW 169–184). Weber thus emphasizes the exceptional character of empirical sciences capable of formulating laws. Most empirical sciences deal with complex singular phenomena. This is not a peculiarity of the sciences of human action. It is not human freedom that is resistant to the laws of nature, but the 57

On Weber’s debate with Marxism, see Colliot-Thélène (1990, 26–51). In Weber’s methodological work, see in particular WL 166–170 and 203–204; CMW 111–113; 131–132. 58 In his presentation to Weber (1996, 56 ff.), J-P Grossein insisted on the Weberian work of causal imputation.

Max Weber and Understanding Explanation

47

logical characteristics of the phenomenon or phenomena that the scientist seeks to explain, in other words, their singularity cannot be accounted for by simply formulating universal laws. In this respect, the explanation of a falling stone or a meteorological phenomenon is of the same order as that of human action. In all these cases, the existence of laws of nature is postulated (the gravitational force applies as much to the falling stone as to the human body), but they do not explain the singularity of these phenomena. The laws, therefore, remain mostly implicit. In the case of human actions, regularities of human action are also assumed, without being formulated. Thus the pedagogue who, most of the time, is content with “common sense” and his “practical experience” (CMW 51; WL 79), with “popular psychology” (CMW 52; WL 81) without having recourse to the formulation of psychological laws to explain the behavior of a child. It is, therefore, necessary to look first and foremost for the singular causes that explain such and such a rock fall or human behavior. Thus, from the point of view of predictability, there is no reason to speak of a specificity of human actions according to Weber.59 The physiological effects of the work of breeding on a head of cattle or the phenomena of weather are no more or less predictable than the reactions of an individual to any order or statement. All singular empirical phenomena are equally predictable. Only a process of abstraction can lead to increased predictability, as in the case of laboratory experiments. The historical sciences share with other sciences that have natural phenomena as their object an interest in singularity; the ontological characteristic of their objects being subordinated to their logical particularity. Human actions are not unpredictable on the grounds that man is a free being. Weber shows that the singularity of these actions does not prevent their predictability, even if this cannot be complete given the complexity of the causal factors involved.60 Against those who emphasize the specificity of human behavior, Weber insists first on what is common to the explanation of singular complex inert or animate phenomena.61 Both are primarily based on the principle 59 “it is certainly not possible to find traces in ‘experienced’ reality of a specific ‘incalculability’ of human conduct” (CMW 41; WL 64). 60 Rickert also rejects the inclusion of such metaphysical considerations for a theory of science: cf. Rickert (1902, 415), Rickert (1997), Science de la culture et science de la nature, transl. A-H Nicolas, Paris: Gallimard (trans. of 7th ed.: 1926), 84 [henceforth abbreviated SCSN ]. 61 Weber also insists on the conceptual character of these sciences of reality, stating: “the term ‘concept’, which is nowadays as contentious as ever, is used here, and will be used in what follows, to indicate any theoretical construct, however individual, that is formed by the logical processing of a perceptual multiplicity of [phenomena] with the aim of acquiring knowledge of what is important ”

48

W. Feuerhahn

of causality. Does this mean that nothing distinguishes the sciences of inert reality from the sciences of human reality? Does Weber renounce the defense of specificity of the sciences, which he called “cultural sciences” at the time?

5

“Understanding Explanation,” a Specificity of the Cultural Sciences

“Science of reality” does not correspond to a particular science, but is one of two possible methodological orientations, typical examples of which are, according to Weber, the exact sciences of nature and political history respectively (WL 3). When referring to the science he practices, Weber more frequently speaks of “social sciences” [Sozialwissenschaften] or “cultural sciences” [Kulturwissenschaften] and frequently defines himself as a cultural historian. Insofar as they encompass all “disciplines that view the events of human life from the perspective of their cultural significance and importance” (CMW 111; WL 165), the cultural sciences include the social sciences.62 Weber borrows the term “science of culture” from Rickert, for whom the formal distinction, emphasized by Windelband, between nomological and idiographic sciences is primary (SCSN 88),63 while any other distinction is secondary. Moreover, when Rickert speaks of “natural sciences,” “the term nature […] is used in a logical sense and in accordance with the terminology defined by Kant” (SCSN 91–92). The same applies to the term “historical sciences” (SCSN 92). History refers to the individual and unique course of a thing. However, this formal principle of distinction between the sciences— generalizing versus individualizing—does not allow us to grasp the principle of concept formation in the cultural sciences. History, aiming to explain singular reality, must have a principle of selection: “In order to form its concepts and provide knowledge, history too must draw boundaries in the continuous flow of what actually happens, and transform its heterogeneity, which is beyond us, into a discretion that we can master ” (SCSN 105–106). The mere distinction between nomothetic and idiographic sciences, though (CMW 6; WL 6). Compare with Rickert’s definition of “concept”: “any grouping of the elements of a reality that are scientifically essential ” (SCSN , 115). 62 “If one wishes to use the term ‘cultural sciences’ to refer to those disciplines that view the events of human life from the perspective of their cultural significance and importance, then social science (in our sense) belongs to that category” (CMW 111; WL 165). 63 “Logic has first of all the task of understanding, starting from the scientific goal proper to history, which consists in the representation of the unique and individual unfolding of a reality, the individualising forms of thought used here as necessary means to achieve this end” (SCSN 93).

Max Weber and Understanding Explanation

49

primary, is therefore not enough: “we must combine material distinctions with formal distinctions” (SCSN 105). This combination is reflected in the name chosen by Rickert, and taken up by Weber, of “cultural historical sciences,” the term “culture” indicating the material criterion that directs our interest differently: We will find that we are usually only interested in realities that have no value or meaning that we can understand, and which we therefore regard as mere ‘nature’ in the sense already mentioned, in the sense of the sciences of nature in the logical sense; their singular configuration is not considered in its individuality, but commonly as a mere example of a more or less general concept. In contrast, the situation is quite different for cultural processes that are charged with value and meaning, and for events that we associate with culture because they represent its preliminary stages; our interest here is directed at the particular and the individual, as well as at their unique course. This is why we want to know them historically in an individualizing way. (SCSN 116)

The characterization of culture as the space of values and meaning provides the distinctive material criterion of the sciences of culture and thus allows them to be differentiated from the sciences of natural reality, which seek to explain singular natural phenomena. Rickert in fact borrows this criterion of differentiation from the theories of Rudolf Hermann Lotze (1817–1881) whom he knew from the work of Windelband.64 According to Lotze, all science, including history, psychology and anthropology, proceeds by causal explanation. But Lotze distinguishes between what is indifferent and what is endowed with values. He replaces the opposition between “being” and “ought to be” with the opposition between “the being of things” and “the worth of meaning contents and values.”65 Values and meaning have an ideal objectivity, but they do not exist in the same way as empirical reality, they are “worth.” This principle, taken up and deepened by Windelband and Rickert, explains why the latter systematically associates meaning and values. Rickert reappropriates these questions by contrasting “what exists” with “what is worth.” “Culture” corresponds to what is beyond pure and simple facticity.66 It is part of an ahistorical and universal transcendental, and therefore cannot be confused with the term “spirit,” which Rickert rejects for its ambiguity.

64

On Lotze see Schnädelbach (1983, 206–218). Lotze, Logik (1874) quoted by Gens (2002, 42). 66 This explains why Rickert does not choose an ontological criterion for classifying the sciences. Culture is not a “reality” in the strict sense. 65

50

W. Feuerhahn

What distinguishes culture from nature in these circumstances is not its psychic content, but its objectively ‘spiritual’ [geistig] content, i.e. that which is not perceptible by sensation, but only intelligible in a non-sentient way, and gives meaning and significance to life. (SCSN 16)

Rickert criticizes Dilthey for defining “spirit” in an almost empiricist way. In reality, their debate is about the definition of values, since, while Dilthey defends an immanentist philosophy of values, Rickert emphasizes their transcendence.67 The historical researcher is not interested in what a particular reality has in common with other realities, but in what distinguishes it from them, i.e., the individual value it carries. The concept of culture plays the role of a principle of selection of the essential within reality because it is thanks to it that the historian or cultural scientist determines what has had a significant causal effect. Without this “value relationship” [Wertbeziehung ] events remain “devoid of any intelligible meaning” (SCSN 123). Not every causal relationship is retained by the historian, he is interested in what matters with regard to the guiding values of his work. However, the fact that Weber’s thought was in contact with the philosophy of values that was developing in his time does not invalidate the fact that Weber cannot and should not be read as a philosopher of values himself. In this, we agree with R. Aron’s interpretation that Weber’s work cannot be seen as the simple sociological application of Rickert’s philosophy.68 Too much attention has probably been paid to the strong impression made on Weber by his reading of Grenzen der naturwissenschaftlichen Begriffsbildung, which is reflected in a letter to his wife.69 Weber does not defend the idea that values are transcendent but analyzes them through the immanent actions of men.

67

Cf. Gens (2002, 42–44). It is this opposition on the definition of values that leads Rickert and Dilthey to propose different definitions of what the work of “understanding” should be: to Dilthey who affirms that “we call ‘understanding’ the process, by which we know something inner [Inneres] from signs that are given from the outside in a sensible way” (“Die Entstehung der Hermeneutik” (1900), Gesammelte Schriften V, 318), Rickert retorts that “signs that are given from the outside must make known more than ‘something inside’ that would be real [ein reales “Inneres”]. Otherwise they remain incomprehensible. […] What is real [Reales…] can only be understood insofar as it ‘carries’ an unreal meaning” (Rickert 1929, 560 and 562). 68 R. Aron has pointed out that Weber cannot be considered a disciple of the Rickertian philosophy of values: R. Aron La philosophie critique de l’histoire (op. cit.), 251–253. 69 Weber, Marianne, op. cit. 296. In this work, Marianne Weber returns to the context of the writing of “Roscher and Knies”: “Heinrich Rickert’s work on the limits of conceptualisation in the natural sciences, the second volume of which had appeared at that time, seems to have occupied and stimulated him. After reading it six months earlier in Florence, he wrote to his wife: ‘I have finished the Rickert. It is very good, I find in it much of what I thought myself, though in an unfinished logical form. I have some reservations about the terminology.’”

Max Weber and Understanding Explanation

51

His aim is not to develop a hermeneutic of the meaning of action but to explain the cause or causes of the actions he studies. This does not mean, however, that Rickert’s role should be overlooked. He was a fundamental interlocutor in the logic of cultural sciences. In this respect, it is interesting to distinguish between what Weber inherits from Rickert and what he does not. Weber certainly distinguishes, following Rickert, between “value judgements” and “value relationships” in that the former are evaluative, while the latter consider the values conveyed by actions as objective facts. However, whereas for Rickert the validity of the guiding values of historical investigations “is at least demanded of all members of a cultural community” (SCSN 135), Weber does not place scientific questioning in an anhistorical transcendentalism,70 but instead emphasizes its historical genesis.71 These questions are not linked to values that tend towards universality.72 The “relation to values” itself, therefore, does not have the same meaning for Weber. For him, “value” is more synonymous with “interest.”73 It is therefore essential to keep Rickert’s work in mind in order to better understand Weber’s definition of understanding explanation. Indeed, Weber agrees with Rickert in that he considers it to be a modality of causal explanation that can only be activated if the phenomena under consideration are the product of motivations that give them meaning. Understanding explanation reveals the motivations of actions that are endowed with meaning.74 When we wish to interpret [human behaviour], we can, at least in principle, set ourselves the goal not make it ‘comprehensible’ [in the sense of ] being ‘possible’ (that is to say: consistent with our nomological knowledge), but also

70 “There is no absolutely ‘objective’ scientific analysis of cultural life - or (to use a term which is perhaps somewhat narrower but which, for our purposes, does not have an essentially different meaning) of ‘social phenomena’ – independent of special and ‘one-sided’ points of view, according to which [those phenomena] are – explicitly or implicitly, deliberately or unconsciously – selected as an object of inquiry, analysed and presented in an orderly fashion” (CMW 113; WL 170). 71 Cf. CMW 101; WL 148: “It is common knowledge that our discipline, like any other scientific discipline dealing with the institutions and processes of human culture […] had its historic roots in practical considerations.” 72 However, this should not lead, in our opinion, to reading Weber’s work as the archetype of a philosophy of axiological polytheism generating relativistic thinking. Weber does not defend one philosophy of values against another, he does not define the essence of values, but merely observes and tries to explain as a historian the causes of certain conflicts between certain values inscribed in concrete realities. 73 Depending on the objective—to study political or fashion history, for example—the researcher is not interested in the same facts. 74 Cf. Weber’s definition of action at the beginning of Economy and Society (Weber 1922/1978, 4): “We shall speak of ‘action’ insofar as the acting individual attaches a subjective meaning to his behavior – be it overt or covert, omission or acquiescence”.

52

W. Feuerhahn

to ‘understand’ it; that is to say, to uncover a concrete ‘motive’ that we can ‘reexperience’ ‘within ourselves’, or a complex of such motives, to which we can impute [that behaviour] (with a degree of precision which will vary according to the nature of the source material). (CMW 43; WL 67)

However, Weber restricted his focus and did not include all motivated human behavior. When, in 1904, he decided to take over the Archiv für soziale Gesetzgebung und Statistik, founded by Heinrich Braun in 1888, together with Sombart and Jaffé, he renamed it Archiv für Sozialwissenschaft und Sozialpolitik and justified the name “social science”: “We now designate the scientific investigation of the general cultural significance and importance of the social-economic structure of human communities, and of their historical forms of organisation” (CMW 110; WL 165). Weber specifies that the limitation of the field of research to the economic and social structure of human collective life is voluntary [“that ‘one-sidedness’ is intentional” (CMW 111; WL, 165)]. Indeed, he refuses to build a “general social science” (CMW 111; WL 165), not only because it would make it impossible to distinguish, within the sciences of culture, between philology, church history, or even the science of the state, but also because any science always depends on a guiding point of view. However, if understanding explanation is to be applied to economic and social phenomena, Weber emphasizes the complexity of the latter, which includes phenomena that are strictly economic, but also phenomena that the economy merely conditions, or phenomena with both economic and social significance. Weber also takes into account the complexity of motivations. To this end, he proposes his famous typology of the determinations of social action classified according to their intelligibility, i.e., according to the possibility of imputing meaning to them; and his typology ranges from rational action with an end in view to traditional action, the latter being a borderline case of actions endowed with an intentional meaning. It is difficult to identify the motivations of ritual or traditional actions; for example, the mystical behavior of pietists, which he focuses on in The Protestant Ethics and the “Spirit” of Capitalism, appears to be of little rationality. By contrast, for the turn-of-the-century European scholar Max Weber, a merchant’s decisions can be “understood” without lengthy analysis. Even the complex strategies of Frederick II in the Seven Years’ War appear to be ultimately rational (see CMW 44; WL 68). The limits of understanding explanation are thus defined differently by Weber and by the historical theorists he opposes. For the latter, understanding stops where the human stops, i.e., where there is an ontological leap, at the

Max Weber and Understanding Explanation

53

edge of the realm of freedom. The boundary thus corresponds to the separation between human and non-human, between freedom and necessity. Weber challenges this boundary. We can only speak of an “understanding explanation” if we discover the motivations for action. However, where these are not known, the cultural scientist will be satisfied with a simple explanation, as in the case of the sudden bouts of dementia suffered by Prussian King Frederick William IV, which psychopathology can at best only explain by the evidence of anatomical lesions (see CMW 44; WL 69). The boundary between “understanding explanation” and simple explanation is thus independent of the boundary that separates human actions from natural phenomena; it separates motivated actions from behavior that the historian cannot link to an intention. This is why, conversely, there are inanimate phenomena that can be understood: To be devoid of meaning is not identical with being lifeless or non-human; every artifact, such as for example a machine, can be understood only in terms of the meaning which its production and use have had or were intended to have; a meaning which may derive from a relation to exceedingly various purposes. Without reference to this meaning such an object remains wholly unintelligible. That which is intelligible or understandable about it is thus its relation to human action in the role either of means or of end; a relation of which the actor or actors can be said to have been aware and to which their action has been oriented. Only in terms of such categories is it possible to ‘understand’ objects of this kind. (Weber 1922/1978, 7)

For Weber, understanding explanation is therefore possible from the moment when we can impute meaning to a phenomenon, even if it is, as in this case, an inanimate tool. On the other hand, all phenomena that are not the effect of a motivation, whose intention cannot be uncovered, such as the mortality rate, or any demographic index, cannot be understood. They can only be explained, that is, their causes can be sought, but not those peculiar types of “causes” that are “motivations.” * ** Weber rejects the alternative that, according to him, dominates the discussions of his time: he is opposed both to a reduction of the specificity of the sciences of culture to the sciences of nature and to the idea of total irreducibility between these two types of sciences. According to him, this false alternative is based on the amalgam between metaphysical theses and scientific procedures. Weber points out that every scientific practice has its own cognitive purpose, and that the purpose of the sciences of reality is not

54

W. Feuerhahn

fully satisfied by the formulation of universal laws but requires the search for singular causes. In this sense, the historical sciences can go further than the other sciences of reality. Through the construction of ideal types of determinations of social action, they can reconstruct the motives behind the events they study, something that a meteorologist, for example, could not do. But Weber is careful not to replace the opposition between “explanation” and “understanding” with an opposition between “causes” and “motivations.” For Weber, “motivations” are “causes” of a peculiar type.

References Apel, K.-O. (1979). Die Erklären-Verstehen Kontroverse in transzendentalpragmatischer Sicht. Frankfurt am Main: Suhrkamp. Aron, R. (1987). La philosophie critique de l’histoire. New ed. revised and annotated by S. Mesure. Paris: Julliard. 1st ed.: Vrin, 1938. Aron, R. (1992). Les étapes de la pensée sociologique. Paris: Gallimard. Baier, H. (1969). Soziologie als Handlungswissenschaft. Münster, ungedruckte Habilitationsschrift. Brunschvicg, L. (1912). Les étapes de la philosophie mathématique. Paris: Alcan. Canguilhem, G. (1999). La problématique de la philosophie de l’histoire au début des années 30. In J.-C. Chamboredon (ed.), Raymond Aron, la philosophie de l’histoire et les sciences sociales (pp. 9–23). Paris: Éditions Rue d’Ulm. Chazel, F. (2000). Aux fondements de la sociologie. Paris: Presses universitaires de France. Colliot-Thélène, C. (1990). Max Weber et l’histoire. Paris: Presses universitaires de France. Colliot-Thélène, C. (1992). Le désenchantement de l’Etat. De Hegel à Max Weber. Paris: Éditions de Minuit. Colliot-Thélène, C. (2001). Études wébériennes. Paris: Presses universitaires de France. Dilthey, W. (1947). Idées concernant une psychologie descriptive et analytique. In W. Dilthey Le monde de l’esprit Volume I (pp. 145–245). Trans. M. Rémy. Paris: Aubier-Montaigne. Dufour, É. (2003). Les Néokantiens. Paris: Vrin. Ferrari, M. (2001). Retour à Kant. Introduction au néokantisme, translation from Italian by Thierry Loisel. Paris: Editions du Cerf. Feuerhahn, W. (2005). Max Weber et l’explication compréhensive. Philosophie, 85, 3–41. Feuerhahn, W. (2023). L’émergence polémique des réflexions wébériennes sur les sciences. In Max Weber, Qu’est-ce que les sciences de la culture? (p. 11–226). Paris: CNRS éditions.

Max Weber and Understanding Explanation

55

Fischer, K. (1860). Kant’s Leben und die Grundlagen seiner Lehre. Drei Vorträge. Mannheim: Bassermann. Gens, J.-Cl. (2002). La pensée herméneutique de Dilthey. Villeneuve d’Ascq: Presses Universitaires du Septentrion. Hennis, W. (1996). La problématique de Max Weber. Translation by Lilyane Deroche-Gurcel. Paris: Presses universitaires de France. Knies, K. (1850). Statistik als selbständige Wissenschaft. Ein Beitrag zu einer kritischen Geschichte der Statistik seit Achenwall . Kassel: Luckhardt. Knies, K. (1964). Die politische Oekonomie vom geschichtlichen Standpuncte. Osnabrück: Otto Zeller (reproduction of the 2nd edition: 1883). Köhnke, K.-C. (1993). Entstehung und Aufstieg des Neukantianismus. Frankfurt am Main: Suhrkamp. Lask, E. (1923). Gesammelte Schriften. Volume I. Tübingen: Mohr (Siebeck). Mesure, S. (1990). Dilthey et la fondation des sciences historiques. Paris, Puf. Rickert, H. (1902). Die Grenzen der naturwissenschaftlichen Begriffsbildung. Eine logische Einleitung in die historischen Wissenschaften. Tübingen: Mohr (Siebeck). Rickert, H. (1921). Die Grenzen der naturwissenschaftlichen Begriffsbildung. Eine logische Einleitung in die historischen Wissenschaften. Tübingen: Mohr (Siebeck). 3rd and 4th Edition. Rickert, H. (1929). Die Grenzen der naturwissenschaftlichen Begriffsbildung. Tübingen: Mohr (Siebeck). 5th expanded edition. Rickert, H. (1997). Science de la culture et science de la nature. Transl. by A-H Nicolas. Paris: Gallimard. (trans. of 7th ed.: 1926). Schluchter, W. (1988). Religion und Lebensführung. Frankfurt am Main: Suhrkamp. Schnädelbach, H. (1983). Philosophie in Deutschland 1831–1933. Frankfurt am Main: Suhrkamp. 1983 Simmel, G. (1892). Probleme der Geschichtsphilosophie. Leipzig: Duncker & Humblot. Simmel, G. (1905). Probleme der Geschichtsphilosophie. Leipzig: Duncker & Humblot. Tenbruck, F. (1959). Die Genesis der Methodologie Max Webers. Kölner Zeitschrift für Soziologie und Sozialpsychologie, 11, 573–630. Wagner, G. (1987). Geltung und normativer Zwang. Freiburg (Breisgau), Munich: Alber. Wagner, G. & Zipprian, H. (ed.). (1994). Max Webers Wissenschaftslehre. Interpretation und Kritik. Frankfurt am Main: Suhrkamp. Weber, M. (1950). Max Weber. Ein Lebensbild . Heidelberg: Verlag Lambert Schneider. 2nd ed. 1950 [1st ed.: 1926]. Weber, M. (1922/1978). Economy and Society. An Outline of Interpretive Sociology. Edited by Guenther Roth and Claus Wittich. Berkeley, Los Angeles, London: University of California Press. Weber, M. (1922/1988). Gesammelte Aufsätze zur Wissenschaftslehre. Tübingen: Mohr (Siebeck).

56

W. Feuerhahn

Weber, M. (1996). Sociologie des religions. Volume translated and presented by JeanPierre Grossein. Paris: Gallimard. Weber, M. (2001). Rudolf Stammler et le matérialisme historique. Translated by Michel Coutu and Dominique Leydet. Québec/Paris: Presses Universitaires de Laval/ Editions du Cerf. Weber, M. (2012). Collected Methodological Writings. Translated by Hans Henrik Bruun, edited by Hans Henrik Bruun and Sam Whimster. London; New York: Routledge. Windelband, W. (1894). Geschichte und Naturwissenschaft. In Das Stiftungsfest der Kaisers Wilhelms Universität Strassburg am 1. Mai 1894 (pp. 15–41). Strasbourg: Heitz & Mündel. Weber, M. (2023). Qu’est-ce que les sciences de la culture? traduit, présenté et édité par W. Feuerhahn, Paris: CNRS éditions.

Viennese Methodological Individualism Dario Antiseri

1

Popper as a Methodological Individualist

Social scientists must continually deal with collective concepts (Kollktivbegriffe) such as “society,” “nation,” “class,” “state,” “revolution,” “people,” “party,” and so on.1 Historically, there have been two opposite currents of thought on the interpretation of these concepts: methodological individualism and methodological collectivism. Their opposition involves three problems: ontological, methodological, and political. The ontological problem can be summarized as follows: What do collective concepts really refer to? Methodological individualists (e.g., Mandeville, Hume, Ferguson, Smith, Menger, Mises, and Hayek) reply that they refer only to individuals. Only individuals exist and only individuals reason and act. The methodological collectivists (e.g., Saint-Simon, Comte, Hegel, Marx, the neo-Marxists, and structuralists), on the contrary, believe that collective concepts refer to

1

This chapter directly draws from Chapter 8 of my book La Vienna di Popper published in 2000 by Rubettino. The English edition of this book was published in 2007 by The Davies Group Publishers under the title Popper’s Vienna.

D. Antiseri (B) Department of Political Science, Luiss Guido Carli University, Rome, Italy e-mail: [email protected]

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 N. Bulle and F. Di Iorio (eds.), The Palgrave Handbook of Methodological Individualism, https://doi.org/10.1007/978-3-031-41512-8_4

57

58

D. Antiseri

substantial realities, i.e., autonomous entities that are independent of individuals and, like “church” or “native land,” act on the individuals and make them conform. The methodological problem can be summarized as follows: Where does social research, which aims at explaining social events and institutions, begin? Since methodological individualists believe that only individuals exist, they maintain that any inquiry into the origins and changes in social events and institutions necessarily must start from the actions of individuals (these actions must be studied in order to explore their unintentional consequences). Since methodological collectivists believe in the reality of collectives, they try, on the other hand, to discover the laws (of progress and decline, dialectics, and so on) that govern the origins and development of such collective entities. The political problem can be summarized as follows: Is the goal of political action a collective entity like a society or nation, or is it the individual with her freedom and responsibility? If a collective entity like a State is real, it is obvious that individuals are at the service of this entity and its collective goals. On the contrary, methodological individualists assume that the end is the individual, not the State, the class, or the party. In their opinion, if an individualistic concept of society is eliminated, the justification for democracy is eliminated too. The reader can find the following prospect in the first volume of Karl Popper’s The Open Society and Its Enemies (Popper 1966a, p. 100): a) Individualism is opposed to b) Egoism is opposed to

a' ) Collectivism b' ) Altruism

According to Popper (Idem), “collectivism is not opposed to egoism nor is it identical with altruism or unselfishness. Collective or group egoism, for instance class egoism, is a very common thing […] and this shows clearly enough that collectivism as such is not opposed to selfishness. On the other hand, an anti-collectivist, i.e. an individualist, can, at the same time, be an altruist?”.2 This counteracts the “ethical” objection of those who are hostile to methodological individualism because it would be the same as an egotistical position. Individualism is, on the contrary, in contrast to collectivism but not to altruism. According to individualism, only individuals exist, and they may

2 For a detailed analysis of the differences between individualistic and collectivist concepts in social philosophy in the period that goes from the Middle Ages to Adam Smith, see K. Pribram (1912, pp. 1–18). On Pribram, see W.M. Johnston (2006, pp. 90–92). See also R. Kerschagl (1925, pp. 66– 114).

Viennese Methodological Individualism

59

be either egoists or altruists. What does not exist is society if it is understood as an independent entity that exists prior to and above the individuals. In 1990, in an interview with Guido Ferrari, Karl Popper clarified this point: Moreover, I‘d like to add, if I may, that it is extremely misleading to talk about society. Naturally one can use a concept like society or the social order, but we shouldn’t forget that they are only auxiliary concepts. What really exists are people, good ones and bad ones — let’s hope there aren’t too many of the latter, in any case human beings, some of whom are dogmatic, critical, lazy, diligent or something else. This is what really exists. (Popper 1990, pp. 24–25)

There are human beings who have ideas and act because of them, and their actions have intended and unintended consequences. Human beings exist, what doesn’t exist is society. However, people believe it exists and so blame everything on society or the social order (ibid., p. 25) So, Popper contends, “one of the worst mistakes is to think that something abstract is concrete. This is the worst kind of ideology” (Idem; see also Green 1968, p. 656; Rothbard, 1979, pp. 57–61). The error of collectivists is to replace a theoretical abstract construct with concrete things. In Popper’s opinion, society does not exist. Not even the police exist: “[…] the laws regarding the police can be changed but the police as such cannot. The police as such does not exist. Laws can change because they are written down in codes and therefore exist” (Popper 1990, p. 25). There are police agents, but not the police. The army does not exist either. In The Poverty of Historicism, he wrote the following: Most of the objects of social science, if not all of them, are abstract objects; they are theoretical constructions. (Even ‘The war’ or ‘The army’ are abstract concepts, strange as this may sound to some. What is concrete is the many who are killed, or the men and women in uniform, etc.) These objects, these theoretical constructions used to interpret our experience, are the result of constructing certain models (especially of institutions), in order to explain certain experiences [...]. Very often we are unaware of the fact that we are operating with hypotheses or theories, and we therefore mistake our theoretical models for concrete things. This is a kind of mistake which is only too common. (1989, pp. 135–136)

According to Popper (ibid., p. 136), “the task of social theory is to construct and to analyse our sociological models carefully in descriptive

60

D. Antiseri

or nominalist terms, that is to say, in terms of individuals, of their attitudes, expectations, relations, etc. — a postulate which may be called ‘methodological individualism’”.3 The word society refers to individuals with some specific ideas who act according to these ideas. There are really only human beings—“good ones and bad ones” (Popper 1990, pp. 24–25). Only they can think and act. As a consequence, “institutions (and traditions) must be analysed in individualistic terms — that is to say, in terms of the relations of individuals acting in certain situations, and of the unintended consequences of their actions” (Popper 1989, p. 324). Only individuals act: “institutions don’t act, only individuals act in or for the institutions” (Popper 2015, p. 12), so that what is needed is “to construct a theory of the intentional and unintentional institutional consequences of purposeful actions. This might lead to a theory of the origin and development of institutions” (ibid.; see also Panebianco 1989).4 In Popper’s view, once these premises have been established, social events can be explained by situational analysis, a research strategy in which a human action—interrelating with the actions of other individuals—is regarded as an attempt to solve some problem.

2

Menger, Simmel, and Weber

“If in the end I have become a sociologist […], it is because I wanted to rule out those exercises based on collective concepts. Sociology itself cannot proceed from anything but actions of single individuals. For this reason, sociology must adopt strictly ‘individualist’ methods” (Weber 1920). Max Weber wrote this in a letter to R. Liefmannin in 1920. His individualistic approach is clear; so is his rejection of “organic sociology,” which attempts “to explain social action by starting from the ‘totality’”—for example, from the totality of the “economy of peoples”—and then, within it, to interpret the individual and his behavior, similar to how physiology considers the position of a bodily “organ” in the functioning of the organism from the point of view of ‘preserving’ it” (Weber 1920). Obviously, Weber contends, the analysis of the functional aspects of the “parts” of a “totality” may play a useful role “in practical explanation and temporary orientation” (Idem) but “it also may

3 For more details on the logic of the situation and methodological individualism in Popper, see the critical remarks by A.M. Petroni (1981, pp. 64–76). According to Petroni (1989, p. 144) “some of Popper’s weakest pages are devoted to the principle of rationality.” 4 On the differences between Popper and Hayek see Scott (1973, pp. 215–220) and Barry (1979, pp. 39–41).

Viennese Methodological Individualism

61

happen that its cognitive value is overestimated because very detrimental false conceptual realism has been accepted” (Idem). All this means that collective concepts should be interpreted individualistically, which means that the social sciences must proceed in an individualistic way, from the actions of “a single individual, of a few individuals, or of many individuals” (Idem, p. 16). Therefore, “even a socialist economy should be included in sociology in virtue of a procedure interpreted ‘individualistically’ — that is, based in the actions of individuals” (Ibidem). Simmel’s view is the same as Weber’s. In 1897 the former was already critical of the misleading illusion whose view of the evolution of the law and other institutions would lead to thinking of society as an “impersonal being” independent of individuals (Simmel 1987). According to Simmel (Idem), society cannot be regarded as something “conceived for itself.” “It is certain,” he contends, “that only individuals exist, that human products have a reality apart from men only if they are of natural materials, and that since the creations we are referring to are spiritual, they live only in personal brains” (Ibidem). In Vienna, Carl Menger, one of the originators of the marginalist revolution in economics, had become an advocate of methodological individualism even before Weber and Simmel. Menger’s basic assumption is that resources are scarce and individuals seek the greatest satisfaction of their needs and desires. In his opinion, it is precisely individuals and their actions that economic theory needs to take into consideration and adopt as a point of departure. According to Menger (1883, pp. 86–87) since “the collectivity as such is not like a great subject, who has his needs, who works, competes; what we call ‘social economy’ is not the economic activity of a society, strictly speaking: it is not analogous to individual economies and does not exist as opposite to individual economies. In its phenomenic form it is a particular multiplicity of particular economies.” In other words, there are no specific realities independent of the individuals that correspond to the collective concepts used in economics and in the social sciences more generally. The facts of social economics are not the immediate manifestations of the life of a collectivity as such, the direct products of its economic activity, but rather the result of countless individual efforts. As a consequence, to understand these complex facts it is necessary to appeal to the real elements of the individual economies and explain how social economy derives from individual economies (ibid., p. 87). For Menger, anyone who insists on following the opposite route, namely the holist approach, does not understand the real nature of social economy and proceeds through a misleading method. As we shall see later on, Menger (ibid., pp. 84–85) shows how the individualistic

62

D. Antiseri

method is crucial to explain the origins of social rules and institutions, which, in his opinion, can arise in two ways: either in a pragmatic way (as the result of the will of individuals directed toward a goal) or as unintentional results of individual actions that had other goals (i.e., through an invisible hand mechanism that can account for the origins and emergence of many social events, even very important ones).

3

Ludwig Von Mises on Human Action and Social Phenomena

Ludwig von Mises was among those who followed and developed Menger’s approach, and, in turn, exerted a significant influence on Friedrich Hayek. Popper knew about Mises’s work in Vienna in 1935, and several months later he met Hayek in London (see Popper 1992, p. 10) Mises (1981a) famously expressed the fundamental principle of methodological individualism as follows: it is only the individual who thinks. It is only the individual who reasons. It is only the individual who acts. As a consequence, the course of history is determined by individual actions and by the consequences of these actions (Mises 1949). From a methodological perspective, all this means that in the social sciences we must always move from individual actions. The idea of a society that works independently of individual actions is absurd. The fact that one is a member of a capitalist society or of a particular association cannot be shown but through individual actions (Mises 1981b, p. 43). This is how methodological individualism makes sense of observable experience. Experience does not show us things like “society” or “nation.” Experience always introduces us to individuals, it makes us listen to their words, read what they write, and see their actions. It is merely an illusion to assume it is possible to visualize collective unities. Everything we know about these unities is the result of the understanding of the individual actions and the meaning that the individuals attach to their actions (Mises 1949). One must continually fight against the temptation to hypostatize our mental constructs, i.e., to attribute substance and real existence to them (Mises 1978, p. 78). The most striking case of this common fallacy is the use of the term “society” in pseudoscientific research traditions. Society is not itself a substance, a power, an agent. Society does not exist independently of the thoughts and actions of individuals. This is because like all the other collective unities, it has no proper interests or goals. The naïve tendency to hypostatize collective concepts is the most dangerous enemy for the social sciences and helps those liberticidal political projects that ascribe more dignity to collectivities than

Viennese Methodological Individualism

63

to individuals, or that consider individuals to be mere abstractions. Instead, everything we can know about the social world can be achieved only by studying the actions of individuals. Methodological collectivism is a mythological view based on the idea that human actions are led by mysterious forces (ibid., pp. 78–82).

4

Friedrich Hayek: Constitutive vs Speculative Opinions

Friedrich von Hayek (1952, p. 94), like Mises, thought that the typical element of methodological collectivism is that it treats social phenomena not as something of which the human mind is a part and the principles of whose organization we can reconstruct from the familiar parts, but as if they were objects directly perceived by us as wholes”. This widespread philosophical position stems from the fact that “the existence, in popular usage of terms like society or economy is naively taken as evidence that there must be definite ‘objects’ corresponding to them. The fact that people all talk about the nation or capitalism leads to the belief that the first step in the study of these phenomena must be to go and see what they are like, just as we should if we heard about a particular stone or a particular animal. (ibid., 94–95)

That is why, Hayek contents, “the error involved in this collectivist approach is that it mistakes for facts what are no more than provisional theories, models constructed by the popular mind to explain the connection between some of the individual phenomena which we observe” (ibid., p. 95). It is a huge mistake “that of treating as facts what are no more than vague popular theories” (Idem). This mistake—naive realism—“is so deeply embedded in current thought about social phenomena that it requires a deliberate effort of will to free ourselves from it” (ibid., p. 96). According to Hayek (ibid., p. 98), it is necessary, once the error of naive realism has been exposed, to confirm that “the social sciences do not deal with ‘given’ wholes but their task is to constitute these wholes by constructing models from the familiar elements.” “The mistake of treating as definite objects wholes that are no more than constructions, and that can have no properties except those which follow from the way in which we have constructed them from the elements, has probably appeared most frequently in the form of the various theories about a “social” or “collective’ mind and has in this connection raised all sorts

64

D. Antiseri

of pseudo-problems” (ibid., pp. 100–101) And, at the same time, it fosters the most ruthless social atrocities. An escape from this trap of naive realism, which hypostatizes collective concepts and makes them become concrete things, is to distinguish between motivating or constitutive opinions on the one hand and speculative or explicative concepts on the other. The “ideas” that the “popular mind” has formed about collective concepts such as, for example, society, the economic system, capitalism, or imperialism—the speculative or explicative ideas—must be regarded by the social scientist as no more than “provisional theories, popular abstractions” rather than as real facts (ibid., p. 64). The “characteristic feature” of methodological individualism, which is strictly connected with the subjectivism of the social sciences, is that the social scientist must consistently refrain “from treating these pseudo-entities as facts,” and systematically start from the concepts guiding individuals in their actions “and not from the results of their theorizing about their actions” (Idem). These concepts that lead people to action are precisely motivating or constitutive opinions: for example, we can say that the cause of changing opinions toward certain goods is in price changes of those goods. Constitutive opinions are beliefs or opinions that “lead a number of people regularly to repeat certain acts, for example, to produce, sell, or buy certain quantities of commodities, are entirely different from the ideas they may have formed about the whole of the ‘society’, or the ‘economic system’, to which they belong and which the aggregate of all their actions constitutes” (ibid., p. 63). Like the natural sciences, the human sciences require contact with experience. Motivating or constitutive opinions are, for Hayek, the empirical base of social research. “It is the so-called wholes, the groups of elements which are structurally connected, which we learn to single out from the totality of observed phenomena only as a result of our systematic fitting together of the elements with familiar properties, and which we build up or reconstruct from the know properties of the elements” (ibid., p. 67). It should be emphasized that, “in all this the various types of individual beliefs or attitudes are not themselves the object of our explanation, but merely the elements from which we build up the structure of possible relationships between individuals” (ibid., p. 68). According to Hayek (Idem) individual beliefs and behavior are the “data” of the social sciences. Moreover, he assumes that the task of the social sciences cannot be equated with the explanation of conscious action: “this, if it can be done at all, is a different task, the task of psychology” (Ibidem). For Hayek (ibid., p. 67), the method of the social sciences is a “compositive” or synthetic method. What are given to the researcher are “elements from

Viennese Methodological Individualism

65

which those more complex phenomena are composed that he cannot observe as a whole” (ibid., p. 66). In other words, “for the social sciences the types of conscious action are data and all they have to do with regard to these data is to arrange them in such orderly fashion that they can be effectively used for their task” (ibid., p. 68). For Hayek (idem), this task—which is the fundamental task of the social sciences—“is done, as we shall soon see, by analyzing the unintentional consequences of intentional human actions.”

5

Popper’s Criticism of Psychologism and the Conspiracy Theories

Only individuals think and take action. This is the basic principle of methodological individualism. It is precisely from the actions of individuals, Popper contends, that the researcher must start in order to explain the origins of and changes in institutions and social phenomena with the awareness that intentional human actions produce unintended consequences. In Secttion 21 of his book The Poverty of Historicism Popper (1989, p. 66) writes the following: “the piecemeal technologist or engineer recognizes that only a minority of social institutions are consciously designed while the vast majority have just ‘grown’, as the undesigned results of human actions”.5 This view, which was only quickly mentioned in The Poverty of Historicism, later became increasingly important in Popper’s reflections on the social world and the role of the human sciences. Thus, in the first volume of The Open Society and Its Enemies, Popper—concerned with the historical analysis of the nature and role of institutions, which have been seen as willed by God or destined by fate or as obedient to powerful historical trends, etc.—counterpoises to the historicist thinker the gradualist social technologist, who is aware that institutions grow similarly to natural organisms. In Chapter XIV of the second volume of The Open Society and Its Enemies, Popper develops on this topic and provides an extensive attack against psychologism, arguing in favor of the autonomy of sociology. Of course, Popper (1966b, p. 93) contends, “it must be admitted that the structure of our social environment is man-made in a certain sense; that its institutions and traditions are neither the work of God nor of nature, but the results of human actions and decisions.” Nevertheless, “this does not mean that they are all consciously designed, and explicable in terms of needs, hopes, or motives. On the contrary, even those which arise as 5 In Popper’s opinion, the two paradigms that respectively assert that institutions are either ‘designed’ or grow spontaneously are typical, on the one hand, of the theorist of the Social Contract and, on the other, of their critics, such as, for example, David Hume (see Popper 1989, p. 65 footnote 1).

66

D. Antiseri

the result of conscious and intentional human actions are, as a rule, the indirect, the unintended and often the unwanted by-products of such actions” (Idem).6 As a consequence, only a small number of social institutions are or have been intentionally designed; the majority of them simply grew, or sprung up, as the unintended result of intentional actions (1966b, p. 91). This is not all for Popper (1966b, pp. 93–94): we can add that even most of the few institutions which were consciously and successfully designed (say, a newly founded University, or a Trade Union) do not turn out according to a plan - again because of the unintended social repercussions resulting from their intentional creation. For their creation affects not only many other social institutions but also ‘human nature’ - hopes, fears, and ambitions, first of those more immediately involved, and later often of all members of the society.

In Popper’s opinion, given the above, i.e., given that most social institutions are not planned and that even institutions that are planned do not implement the plan as it was conceived, there are two inevitable consequences: the first is that psychologism is impossible and the second is that the conspiracy theory of society is untenable. According to the doctrine of psychologism, social research should be reduced to psychology. This is because the origins of and changes in all social phenomena and institutions can be explained by the intentional actions and projects of individuals (Popper ibid., p. 90). However, psychologism does not hold up because “it fails to understand the main task of the explanatory social sciences,” which is to explain the unintended, perhaps even unexpected and undesired, consequences of intentional actions (ibid., pp. 93–94). There are innumerable crucially important social institutions and effects of institutions that are not due at all to intentions, hopes, fears, and conscious projects. Psychologism does not know what to make of them.7 The same is true for the conspiracy theory of society. This theory maintains that a social phenomenon is explained only if the researcher manages to discover the people or social 6

The easiest way to understand the concept of unintended consequences is to use an example: “If a man wishes urgently to buy a house in a certain district, we can safely assume that he does not wish to raise the market price of houses in that district. But the very fact that he appears on the market as a buyer will tend to raise market prices. And analogous remarks hold for the seller” (Op. cit., p. 96). Menger mentions this example in passing, but Popper returns to it several times. For example, see Conjectures and Refutations, cit., p. 452. 7 According to Popper, the inability of psychologism to offer social explanations was well understood by Marx: “To have questioned psychologism is perhaps the greatest achievement of Marx as a sociologist. By doing so he opened the way to the more penetrating conception of a specific realm of sociological laws, and of a sociology which was at least partly autonomous” (op. cit., p. 88; see also Popper 1949, p. 452).

Viennese Methodological Individualism

67

groups that have planned or conspired to promote it. (ibid., p. 95). In other words, according to the conspiracy theory, every social phenomenon, particularly those that generally people abhor like war, unemployment, poverty, and famine, is always the result of successful direct intervention of powerful and malevolent individuals or groups, of conspirators who act behind the scenes. In Popper’s view, this theory is a typical result of the secularization of a religious superstition. The belief in the Homeric gods whose conspiracies explain the history of the Trojan War is gone. The gods are abandoned. But their place is filled by powerful men or groups — sinister pressure groups whose wickedness is responsible for all the evils we suffer from — such as the Learned Elders of Zion, or the monopolists, or the capitalists, or the imperialists. (ibid.; see also Popper 2004, pp. 460–461)

Of course, conspiracies do exist, but this does not mean that all social events and institutions are the result of conspiracies. Even if we admit there are conspiracies, we must also acknowledge that “few of these conspiracies are ultimately successful. Conspiratorsrarely consummatetheirconspiracy” (ibid., p. 95). That the results achieved differ greatly from those sought is so “because this is usually the case in social life, conspiracy or no conspiracy. Social life is not only a trial of strength between opposing groups: it is action within a more or less resilient or brittle framework of institutions and traditions, and it creates — apart from any conscious counter-actions — many unforeseen reactions in this framework, some of them perhaps even unforeseeable” (idem). Consequently, it is obvious that “to try to analyse these reactions and to foresee them as far as possible is, I believe, the main task of the social sciences. It is the task of analyzing the unintended social repercussions of intentional human actions — those repercussions whose significance is neglected both by the conspiracy theory and by psychologism” (ibid., p. 61).8

8

According to Popper (1966b, p. 96), we can understand that this task is essential for the social sciences if we think that “an action which proceeds precisely according to intention does not create a problem for social science (except that there may be a need to explain why in this particular case no unintended repercussions occurred).”

68

6

D. Antiseri

The Task of the Social Sciences According to Popper

In his essay Prediction and Prophecy in the Social Sciences, Popper (2004, p. 452), referring primarily to Marxism, attacks “the doctrine that it is the task of the social sciences to propound historical prophecies, and that historical prophecies are needed if we wish to con duct politics in a rational way.” This doctrine, which Popper calls historicism—and of which Marxism is one of the implementations—is, in Popper’s opinion, untenable. One of the reasons for this is that the historicist thinker is incapable of distinguishing between scientific prediction and unconditional historical prophecy (ibid., p. 454). The predictions of science are always conditional. “They assert that certain changes (say, of the temperature of water in a kettle) will be accompanied by other changes (say, the boiling of the water)” (ibid., p. 456). However, the historicist thinker cannot make these conditional predictions because society does not belong to the category of closed systems such as, for example, the solar system (a closed system is a well-isolated, stationary, and recurrent system). “These systems are very rare in nature; and modern society is surely not one of them” (ibid., p. 454). The solar system and cyclic biological systems, where conditional predictions can be made, are exceptions. Society “is changing, developing. This development, in the main, is not repetitive” (ibid., p. 457).9 As a consequence, it is impossible to make predictions, especially long-range ones. Actually, “the most striking aspects of historical development are non-repetitive. Conditions are changing and situations arise (for example, in consequence of new scientific discoveries) which are very different from anything that ever happened before. The fact that we can predict eclipses does not, therefore, provide a valid reason for expecting that we can predict revolutions” (ibid., pp. 457–458). Political prophecies are not scientific predictions. Moreover, studying “the behaviour of social wholes, such as groups, nations, classes, societies, civilizations, etc.” is not the task of the social sciences (ibid., p. 459). Many sociologists treat these social wholes as empirical objects that can be studied in the same way a biologist studies animals and plants. For Popper, this approach “must be rejected as naive. It completely 9 Popper (idem) argues that in so far as the development of society is repetitive “we may perhaps make certain prophecies. For example, there is undoubtedly some repetitiveness in the manner in which new religions arise, or new tyrannies; and a student of history may find that he can foresee such developments to a limited degree by comparing them with earlier instances, i.e. by studying the conditions under which they arise. But this application of the method of conditional prediction does not take us very far” (idem) because in the history of society situations and conditions change. Since the changing of conditions and their combinations are not predictable, neither are their effects.

Viennese Methodological Individualism

69

overlooks the fact that these so-called social wholes are very largely postulates of popular social theories rather than empirical objects” (ibid.).10 As we have already pointed out, for Popper, another widespread mistaken view is that social phenomena are the results of conspiracies and that the researcher’s task is to discover and explain these conspiracies. This view is untenable because “not all consequences of our actions are intended consequences” (Idem). According to Popper (ibid., p. 460) “the conspiracy theory of society cannot be true because it amounts to the assertion that all events, even those which at first sight do not seem to be intended by anybody, are the intended results of the actions of people who are interested in these results’” (ibid., p. 460). It is surprising that many Marxist thinkers have completely accepted the conspiracy theory, which “objectively” sees a capitalist conspirator behind every event, especially if it is negative. This is surprising because “Marx himself was one of the first to emphasize the importance, for the social sciences, of these unintended consequences. In his more mature utterances, he says that we are all caught in the net of the social system. The capitalist is not a demoniac conspirator, but a man who is forced by circumstances to act as he does; he is no more responsible for the state of affairs than is the proletarian” (Idem).11 However, “perhaps for propagandist reasons, perhaps because people did not understand it” this idea of Marx has been abandoned and a large number of people have taken up “a Vulgar Marxist Conspiracy theory” (ibid., pp. 460–461). For Popper, this is “a come-down — the come-down from Marx to Goebbels. But it is clear that the adoption of the conspiracy theory can hardly be avoided by those who believe that they know how to create heaven on earth. The only explanation for their failure to produce this heaven is the malevolence of the devil who has a vested interest in hell” (ibid., p. 461). What is the task of the social sciences if it is not to make historical predictions, if the theory of sociological holism is mistaken, and if the conspiracy theory is a naïve belief? Popper (ibid., p. 460) answers this question as follows: “the main task of the theoretical social sciences … is to trace the unintended social repercussions of intentional human actions.” Understanding the unintentional, and possibly undesirable, consequences of human actions allows us to detect a similarity between the social and the 10

According to Popper (2004, p. 459), “while there are, admittedly, such empirical objects as the crowd of people here assembled, it is quite untrue that names like ‘the middle-class’ stand for any such empirical groups.” 11 In the second volume of The Open Society and Its Enemies Popper (1966b, p. 323) points out that, as to the view that Marx conceived of social theory as the study of the unintentional repercussions of human actions, he is indebted to Karl Polanyi “who emphasized this aspect of Marxism in private discussions (1924).”

70

D. Antiseri

natural sciences. The laws of nature can be defined as practical technological rules stating what cannot be done. The same is true for the human sciences. Examples of rules of this kind in the social sciences are: “You cannot, without increasing productivity, raise the real income of the working population” and “You cannot equalize real incomes and, at the same time, raise productivity” (Idem). This demonstrates that, although historicism is untenable, science and rationality can benefit us in practical life by “helping us choose our actions more wisely” (Idem). In works published later than those examined above, Popper insists on his criticism of the conspiracy theory and on the relevance of the view that the basic task of the social sciences is to recognize the unintended, especially the unwanted, consequences of individual actions. For example, in Towards a Rational Theory of Tradition, once again he highlights that “we wish to foresee not only the direct consequences but also these unwanted indirect consequences” of human actions (1949, p. 167); “either because of our scientific curiosity, or because we want to be prepared for them; we may wish, if possible, to meet them and prevent them from becoming too important. (This means, again, action, and with it the creation of further unwanted consequences.)” (Idem). Later on, Popper developed his theory of “World 3.” This theory is a generalization for the entire world of culture of ideas such as the autonomy of sociology and the unintended consequences of human actions. In Popper’s opinion, there is an objective World 3 of culture, which is made up, among other things, of problems, theories and argumentations, and a great part of it is an unintended product. This world is a sub-product of human language and language itself is an unintended sub-product of actions that were aimed at other goals (see Popper 1992).12

7

Menger on the Spontaneous Emergence of Institutions

While Popper regarded the analysis of the unintended consequences of human actions as the principal task of the social sciences, Hayek believed that the task of these sciences is completely fulfilled in that analysis. In his 12

In Unended Quest Popper (1976, p. 26) says, “there is an infinity of unforeseeable nontrivial statements belong ing to the informative content of any theory, and an exactly corresponding infinity of statements belonging to its logical content. We can therefore never know or understand all the implications of any theory, or its full significance”. In a certain sense, we never know what we are talking about. A similar view is presented in the Gadamer’s hermeneutics at the beginning of the history of effects (Wirkungsgeschichte). See Gadamer (1975).

Viennese Methodological Individualism

71

view, the problems that the social sciences “try to answer arise only insofar as the conscious action of many men produce undersigned results, insofar as regularities are observed which are not the result of anybody’s design” (Hayek 1952, pp. 68–69). In other words, the task of the social sciences is to account for the unintended effects of intentional actions. This explains the autonomy of sociology. In fact, [...] if social phenomena showed no order except insofar as they were consciously designed, there would indeed be no room for theoretical sciences of society and there would be. as is often argued, only problems of psychology. It is only insofar as some sort of order arises as a result of individual action but without being designed by any individual that a problem is raised which demands a theoretical explanation. (ibid., p. 69)

Well before Hayek and Popper, Carl Menger, another Viennese scholar, had argued very clearly that the solution to the most important problems of economics and the social sciences more generally largely depends on the understanding of social institutions and the fact that many of them have grown not intentionally, but spontaneously. Obviously, nobody can deny that there are social institutions and events that arise from explicit agreements, pacts, or conventions between people. There are social events and institutions that come into being because certain groups of humans wish to achieve particular objectives: think, for example, of insurance companies, urban development plans, press agencies, and price controls. This fact has given credence to the belief in a general theory that all social events can be explained as consequences of deliberate collective activity. Menger (1883, pp. 164–165) calls this belief, “the pragmatic theory of the origin of social phenomena; it is the explanation of the nature and the origin of phenomena based on the aims, opinions and means of human associations and their representatives.” We give a pragmatist interpretation of social phenomena that originate in agreements among people: we consider the aims that have guided some people to create certain institutions as well as the means that were available to them. According to Menger, these phenomena are subjected to pragmatistichistorical examination in so far as in each concrete case we evaluate the scope of the associations in the light of the needs of its representatives. However—Menger (ibid., p. 183) contends—a pragmatistic interpretation of social phenomena cannot account for “a long series” of social phenomena such as language, religion, law, state, market and currency. These institutions are not the result of any agreement or legislation, and this presents an odd, interesting problem, “a curious problem”—perhaps the most curious

72

D. Antiseri

of all the problems in the social sciences” (Idem). The problem is: “How is it possible that institutions which are so important for collective well-being grow without a collective will or stipulation?” (Idem). To make clear the significance of the topic, Menger considers one of the many possible examples: the prices of commodities. Normally, they develop without any regulative influence of the State or collective agreement, as a spontaneous result of social interaction (ibid., pp. 172–178). There is a whole series of very important social phenomena that arise spontaneously or unintentionally. These phenomena belie the universalizing claims of the pragmatist theory given that not all social phenomena arise from explicit agreements or positive legislation. On this ground, Menger argues that the solution of the most important problems of the social sciences and, in particular, economics, is strictly linked to the analysis of the growth of the institutions which developed in an “organic” way, namely in an unintentional and spontaneous way. As pointed out by Menger, explaining the nature and development of social institutions as the result of an agreement between individuals or legislation is an old idea. While this pragmatist explanation was unrealistic, it had the advantage of reducing the understanding of every institution to the same logic of interpretation. However, hypotheses are not true merely because they are simple. This also applies to the pragmatistic explanation of the origin of social phenomena. For example, while it is true that new cities can be formed because a certain number of people of different abilities and professions join together with the purpose of founding a city, most of the time new cities have arisen without a precise general intention. A common will can develop only at advanced stages of the social life: usually, it does not produce the birth of an institution, but only its perfecting. Take the state. The view that the origin of the state can only be explained in an organic way, i.e., through an agreement of powerful people, is mistaken. For Menger, the state is, in its original form, the undesigned result of actions led by private interests. In the same way, Menger maintains that other social institutions such as, for example, language, law, commercial customs, the division of work, developed without an explicit agreement, without legislation, without concern for a collective or public interest, but only as the unintended result of private interests. Of course, in the course of social evolution public powers established new institutions and changed or developed those which arose spontaneously, but at the beginning of the social world, the emergence of institutions was unplanned. Sometimes in the later stages of

Viennese Methodological Individualism

73

social evolution, institutions are the result of combined action of both unintentional and teleological forces, of both organic and positive factors (Menger 1871, pp. 250 ff.) Menger’s analysis of the origin of money is a classic example of explanation in terms of methodological individualism. As he points out, while the fact that one gives to another person a part of his property and receives back good or service is clearly understandable, the fact that in modern societies people exchange useful properties with pieces of paper or metal is quite curious. Is money a desired intentional creation attributable either to an agreement or a legislative act, or is this also a case of an institution that “grew up” spontaneously? (Menger 1882, p. 172). For sure, the idea that certain goods become money because of a legislative act is not false since history offers examples that support this idea. However, usually, the legislative act did not give to a particular good the status of money, but simply the official acknowledgment of this status, while the good was already used as money. By the way, the “pragmatist explanation” of the origin of money does not work in all those relevant cases where money is not the result of legislative acts but was born organically. Now, how is it possible that, with the development of economic civilization, certain goods acquired the status of money without explicit agreement or legislative act? Menger explains this phenomenon in the following way. As long as simple commerce of exchange—i.e., the barter—is in place, every individual tries to exchange excess goods for those he/she needs immediately and refuses all the other goods. In order for an individual to be able to exchange surplus goods with other goods, this person must not only find someone who asks for his/her goods but also who can offer what he/she needs. This is precisely what makes a pure barter system quite difficult and problematic. How is it possible to improve the economic life? Everyone could notice that some goods were requested more than others. For example, cattle could be exchanged more easily than other goods, and this made it easier to find in return the desired goods, among the many people who were interested in acquiring cattle. Those who had to offer goods that were less in demand tended to acquire goods that they did not need but were more exchangeable. In this way, even if they did not get the good that they needed, they gradually approached to get it. This happened without any agreement or legislative act or concern for the public interest. People tended to prefer goods more useful for exchange. The goods which were safer, more lasting, and more easily transferable began to be accepted by everyone as “money.” Money did not originate by a legislative act: it was rather the unintended result of individual actions motivated by strictly personal interests. This explanation shows

74

D. Antiseri

how, with time, it is possible to make sense in a scientific way of phenomena that at other times were attributed to divine entities (see also Merton 1936).

References Barry N.P. (1979). Hayek’s Social and Economic Philosophy. London: Macmillan. Gadamer H.-G. (1960). Wahrheit und Methode. Tubingen: Mohr. English trans, by G. Barden and J. Cumming, Truth and Method . London: Sheed (1975). Green, A.W. (1968). The Reified Villain. Social Research, 35, 1968, p. 656. Hayek F.A. (1952). The Counter-Revolution of Science. Studies on the Abuse of Reason. Indianapolis: Liberty Press. Johnston W.M. (2006). Oesterreichische Kultur-und Geistesgeschichte. Böhlau Verlag Wien, 4., erg. Auflage. Kerschagl R. (1925). Einfiihrung in die Methodenlehre der Nationalokonomie. WienLeipzig: Holder-Pichler-Tempsky. Liefmann R. (1920). From a letter by Max Weber to R. Liefmann. In R. Boudon and F. Bourricaud (Eds.), Critical Dictionary of Sociology (p. 11). London: Routledge. Menger C. (1871). Grundsätze der Volkswirtschaftslehre. Wien: Braumuller. Menger C. (1883). Untersuchungen über die Methode der Sozialwissenschaften, und der politischen Oekonomie insbesondere. Leipzig: Dunker und Humblot Merton R.K. (1936). The Unanticipated Consequences of Purposive Social Action. “American Sociological Review,” 1936, 1 (6): 894–904. Mises von L. (1949). Human Action. A Treatise on Economics. London: Hodge. Mises von L. (1978). The Ultimate Foundation of Economic Science. Kansas City: Sheed Andrews and McMeel, Inc. Mises von L. (1981a). Socialism. Indianapolis Liberty Classics. Mises von L. (1981b). The Task and Scope of the Science of Human Action, in Epistemological Problems of Economics. New York-London: New York University Press. Panebianco A. (1989). L’individualismo metodologico, L’analisi della politica. Bologna: il Mulino. Petroni A.M. (1981). Introduzione to the Anthology K.R. Popper. II pensiero politico. Firenze, Le Monnier. Petroni A.M. (1989). L’Individualisme Methodologique. “Journal des Economistes et des Etudes Humaines,” 1989, 2 (1): 25–62. Popper K.R. (1949). Prediction and Prophecy in the Social Sciences. Proceedings of the Tenth International Congress of Philosophy 1:82–91. Popper K.R. (1966a). The Open Society and Its Enemies, Vol. 1: The Spell of Plato. Princeton, NJ: Princeton University Press. Popper K.R. (1966b). The Open Society and Its Enemies, Vol. 2: Hegel and Marx. Princeton, NJ: Princeton University Press.

Viennese Methodological Individualism

75

Popper K.R. (1976). Unended Quest. London: Routledge. Popper K.R. (1989). The Poverty of Historicism. London: Routledge. Popper K.R. (1990). La scienza e la storia sul filo dei ricordi, Interview by Giulio Ferrari. Bellinzonza: Jaca Book-Edizioni Casagrande. Popper K.R. (1992). The Communist Road to Self-Enslavement. Cato Policy Report, May–June 1992, vol. XIV, n. 3, p. 10. Popper K.R. (2004). Conjectures and Refutations. London: Routledge. Popper K.R. (2015). Die Logik der Sozialwissenschaften. Erkenntnis und Evolution Tübingen: Mohr Siebeck. Pribram K. (1912). Die Entsiehung der individualistischen Sozialphilosophic. Leipzig: C. Hirschfeld. Rothbard M.N. (1979). Individualism and die Philosophy of the Social Sciences. San Francisco: Cato Institute. Scott K.J. (1973). Methodological and Epistemological Individualism. In J. O’Neill (ed.), Modes of Individualism and Collectivism. London: Heinemann. Simmel G. (1987). Comment les formes sociales se maintiennent. L’année sociologique. Vol. 1 (Jan 1, 1896): 71. Weber M. (1920). Economy and Society, volume 1, Berkeley: University of California Press.

Popper’s Methodological Individualism and Situational Analysis Makoto Kogawara, Yoji Matsuo, and Han-Seob Yeo

1

Introduction

An encyclopedia defines methodological individualism (MI) as “a doctrine in the social sciences according to which a proper explanation of a social regularity or phenomenon is one that is grounded in individual motivations and behaviour” (Basu 2018, p. 8715). This seems to suggest that MI involves explanations supported by individual psychology. However, Popper’s own concept of MI is utterly different from this description. According to his conception, it is capable of providing explanations without the need for them to be complemented by individual psychology.1

1

Hans-Joachim Niemann (2004) denies the reduction of sociological laws to psychology.

M. Kogawara (B) Kagoshima University, Kagoshima, Japan e-mail: [email protected] Y. Matsuo Faculty of Commercial Sciences, Hiroshima Shudo University, Hiroshima, Japan e-mail: [email protected] H.-S. Yeo School of Commerce, Meiji University, Tokyo, Japan e-mail: [email protected]

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 N. Bulle and F. Di Iorio (eds.), The Palgrave Handbook of Methodological Individualism, https://doi.org/10.1007/978-3-031-41512-8_5

77

78

M. Kogawara et al.

Popper emphatically rejects holistic or collectivist explanations of collective phenomena based on mass psychology and appeals instead to holistic or collective entities such as Volksgeist (the collective mind of a nation) or Zeitgeist (people’s collective mind in a historical period). He requires that we explain social regularities or phenomena without supposing that collectives or wholes have purposes or goals in the same way as individuals do. As a consequence, many students of the social sciences find the Popperian concept of methodology very ambiguous and not easy to comprehend. Their dissatisfaction forces us to explicate Popper’s MI from the perspective of the methodology or the philosophy of science. It is clear that Popperian methodology leads us to the denial of explanations rooted in individual psychology and at the same time of ontological or methodological assumptions that the collective or the whole is an independent and conscious subject of social phenomena. In this context, it is worth referring to Agassi’s elucidation (1960) of Popper’s ideas about MI because it throws light on our problem.2 According to Agassi, Popper’s methodology of the social sciences, while rejecting methodological holism, makes it possible to reduce psychological factors such as individual motives and other subjective factors to the minimum. Therefore, in this respect, Popper decisively spurns psychological individualism. Furthermore, according to Agassi, while Popper rejects methodological holism, which asserts that social groups and institutions have purposes, desires, and consciousness, he maintains institutionalistic individualism. It is true that Popper’s ontology clearly admits the existence of social groups and institutions and affirms the institutionalist view that the collective elements have the power to put individual behaviors under control. While Popper continues to hold institutionalism, he criticizes the assertion of psychologism that social phenomena should be explained exclusively on the basis of individual psychology. If Popper succeeds in presenting a new type of explanation, it seems that he successfully and effectively connects institutionalism and the refusal of individual psychology-based explanations. He calls it the ‘logic of the situation’ or the ‘situational analysis’,3 which is peculiar to the social sciences. However, his idea of situational analysis has been often misunderstood and underestimated in the history or context of apprehending Popper’s methodology. There are many reasons for this, but the most important one is that

2

See also Di Iorio’s (2016) commentaries on Popper’s MI. Although the ‘logic of the situation’ or the ‘situational analysis’ are used almost interchangeably in Popper’s writings, we will follow Popper (1994a) and give preference to the latter. 3

Popper’s Methodological Individualism …

79

Popper’s later arguments,4 developed in the 1960s, have been overlooked. During this period Popper developed his own ontology consisting of three worlds, called the three-worlds-theory (Popper 1972; Popper and Eccles 1977). One of the most important features of this theory is its assertion of the objective existence of the contents of ideas or sentences outside of individual minds. Relying on this background and using the rationality principle instead of universal laws (Popper 1967, 1983b, 1985, 1994a), Popper began to emphasize the significance of situational models. To be precise, he began to revise his earlier version of situational analysis presented in The Poverty of Historicism (1944–45, 1957) and The Open Society and Its Enemies (1945 [1966]), so it is necessary for us to realize that we can better understand this model only by taking into consideration Popper’s later arguments. For the purpose of introducing and understanding the concept of situational analysis in the context of the studies and controversies concerning Popper’s ideas from the 1960s onwards, we want to make clear that his MI is widely applicable in the social sciences in general, and is a more fruitful method than is usually supposed.5 We have organized our argument as follows: Sect. 2 outlines Popper’s three-worlds-theory, and tries to show that the content of this theory has strong interconnections with Popper’s metaphysical assertions concerning evolutionary theory, indeterminism, mindbody interactionism, and so on. Comprehending the relations of the three worlds will contribute to a better understanding of the reasons why Popper defended institutionalism and committed to individualism simultaneously. Section 3 discusses the epistemological dimension of his MI and tries to elucidate the role that situational analysis is supposed to perform in explaining social phenomena. At the same time, this section explains the reason why Popper places more emphasis on situational analysis than the covering law model, as a type of explanation provided by the theoretical social sciences. Section 4 discusses the function that the rationality principle performs in constituting situational analysis. In this section, we will examine some typical 4 In general, Popper’s thought was enlarged in the 1960s, from scientific methodology to metaphysics, and from physics to biology, the theory of evolution and probability theory. As will be discussed in Sect. 3, the impetus for the expansion was The Post-Script to the Logic of Scientific Discovery (1982a, 1982b, 1983a), which was written between 1951 and 1956. It should be remarked that two articles, ‘The Rationality Principle’ (1967) and ‘Models, Instruments, and Truth’ (1994a), which represent Popper’s later methodology of the social sciences, are also significantly different from The Poverty of Historicism (1944–45, 1957) and The Open Society and Its Enemies (1945 [1966]), written during his exile in New Zealand. 5 In the philosophy of science, methodology is often regarded as the theory of testing methods. At the same time, MI is generally interpreted as a method of explaining social phenomena. If, as Popper (1972, Appendix I) says, testing is another aspect of explanation in terms of the philosophy of science, then the terms MI and methodology should to some extent be regarded as synonymous and will be treated as such in this chapter.

80

M. Kogawara et al.

issues raised particularly by economists in order to show the significance and potential explanatory power of situational analysis for the social sciences.

2

Popper’s MI and the Three-Worlds-Theory

During the 1960s Popper developed the three-worlds-theory. This is a metaphysical and ontological discourse and if we do not have sufficient comprehension of this theory, it will be very difficult to have a deep understanding of his basic arguments about MI. The three-worlds-theory is closely intertwined with Popper’s theories of evolution, evolutionary epistemology (Popper 1972, Chapters 2 and 7), indeterminism (Popper 1966, 1982b), mind-body interactionism (Popper and Eccles 1977; Popper 1994b), and so on. The difficulties of treating this theory in its whole scope force us to restrict our perspective exclusively to MI, in particular to the problem of the interaction of the three worlds. This three-worlds-theory gives us a conceptual framework that Popper depicts in the following way: …, without taking the words “world” or “universe” too seriously, we may distinguish the following three worlds or universes: first, the world of physical objects or of physical states; secondly, the world of states of consciousness, or mental states, or perhaps of behavioral dispositions to act; and thirdly, the world of objective contents of thought, especially of scientific and poetic thoughts and of works of art. (Popper 1972, p. 106, italics in original)

Roughly speaking, as regards human beings, World 1 corresponds to the body, World 2 to the mind in mind-body dualism, and World 3 to the content of the cultural products of the mind. It is necessary to notice that World 3 is an independent, objective, and exosomatic entity produced by the activities of World 2, i.e., stories, myths, tools, scientific theories, scientific problems, social institutions, works of art, and so on, and has no place in mind-body dualism (Popper and Eccles 1977, p. 38). It should be noted that World 3 includes social institutions as the products or results of human actions. We will later explicate this important aspect. The three worlds interact and affect one another (see Fig. 1). Dualism takes within its scope only the interaction between Worlds 1 and 2 in the framework of the mind-body problem, and therefore inevitably faces the insurmountable troubles brought about by Descartes’ dualistic interactionism in the seventeenth century. Descartes’ dualism assumes that the mind and physical entities are mutually independent and therefore closed to the

Popper’s Methodological Individualism …

81

World 3 Compton’s problem World 2

Descartes’s problem World 1

Fig. 1 The interaction of three worlds

other’s influences and affections. It is clear that bodies have extensions and can move through their contacts. However, the mind is shapeless and has no extension. So from the perspective of dualism, there is no possibility that the mind can move the body. However, nobody can deny the fact that there are interactions between them. This gives rise to the problem of how the nonextensive mind can push extensional bodies (and vice versa). Popper called this riddle ‘Descartes’ problem’ (Popper 1966, p. 230). Descartes himself tried to solve it by believing that the pineal gland in the brain makes possible the linkages between mind and body. However, modern developments in cerebral physiology have ruled out Descartes’ theory of the pineal gland and give weight to a materialist views.6 It is impossible today to find views that do not deny interactionism. Popper, in his co-authored book, The Self and Its Brain, criticizes materialism and proposes his own theory of mind-body interactionism by asserting the openness of World 2 to Worlds 1 and 3, and by emphasizing the difference from Descartes’ dualism. As shown in Fig. 1, it is a pluralistic interactionism based on the three-worlds-theory. The interaction between Worlds 2 and 3 is called ‘Compton’s problem’, because Popper pays great honor to Arthur Compton, a famous experimental physicist who asked how we humans are able to be influenced and regulated by World 3 entities such 6 Popper distinguishes four positions in materialism. Of these positions, radical materialism does not accept the existence of mental processes, whereas the rests do. However, what all four positions have in common is that none of them acknowledge the interaction between Worlds 1 and 2 and consider World 1, i.e., the physical world, to be self-contained and closed in its own right. It is against this background that panpsychism, epiphenomenalism, and identity theory are also categorized as materialism.

82

M. Kogawara et al.

as rules and promises (Popper 1966, p. 230). Popper asserts that the discussion of the mind-body problem has been traditionally confined to Descartes’ problem, and says that this is the main reason why the mind-body problem has not been understood or solved on a full scale. Popper suggests that after having succeeded in solving Compton’s problem, we will get a perspective for solving Descartes’s problem as a whole. In other words, Popper thinks that by taking into examination World 3 and its functions, we will be able to understand better the relations between Worlds 2 and 1.7 It is important to note that the three-worlds-theory constitutes another major difference between Popper and Descartes. Popper’s interactionism is built on the foundations of indeterminism and the theory of evolution. The subscription number of each world indicates the order of creation that the universe underwent in the evolutionary process. This means that the procession of the evolutionary process of the universe (World 1) first produced the Earth, and then Life, and thereafter brought about Consciousness (World 2), and lastly gave rise to the World of language and knowledge (World 3). Popper calls the creation or appearance of each World the ‘emergence’ and describes it as a ‘miracle’ in the history of the universe (Popper 1982b, pp. 122–123; Popper and Eccles 1977, p. 22). On the other hand, the concept of emergence throws light not only on Popper’s divergence from Descartes as a result of his perspective on evolution but also on his hostile attitude toward reductionism, i.e., a form of materialism that Popper denies in The Self and Its Brain. Reductionism tries to explain all phenomena by reducing them to physical processes (World 1). These processes are shown by the upward arrows in Fig. 1. As far as the relation between Worlds 2 and 1 is concerned (Descartes’ problem), it is possible to regard all our conscious phenomena and mental 7

The interaction between Worlds 3 and 2 is discussed in detail by Eccles, co-author of The Self and Its Brain. According to Eccles and Robinson (1984), a person is considered to be a human being even at the fetal stage, but the process of personality development, in which a person acquires self-consciousness as a single personality, can be described as precisely the process of interaction between Worlds 2 and 3. They illustrate the process by which each individual consciousness (World 2) interacts with language-based culture (World 3) such as science, literature and art, expanding and elevating each together (Eccles and Robinson 1984, p. 31, see Fig. 3-4). It then states that. “And so each of us has developed progressively in self-creation, and this can go on throughout our whole lifetime. The more the World 3 resources of the human person, the more does it gain in the selfconsciousness of World 2 by reciprocal enrichment. What we are is dependent on the World 3 that we have been immersed in and how effectively we have utilized our opportunities to make the most of our brain potentialities” (Eccles and Robinson 1984, p. 32). Furthermore, immediately after this sentence, they refer to the tragic case of Genie, a young girl whose personality development stopped after she was cut off from contact with World 3 for more than a decade (Eccles and Robinson 1984, p. 32). This is a vivid impression of the reality of World 3 and its influence on World 2, and may serve as a corollary to show how important it is to understand the interaction between Worlds 3 and 2 in order to understand the mind.

Popper’s Methodological Individualism …

83

processes as physical phenomena (e.g., behaviors of molecules, atoms, and even elementary particles). This way of explaining the phenomena of a higher level by using those at a lower level, and not vice versa, is called explanation by ‘upward causation’ (Popper and Eccles 1977, p. 19). Popper held in esteem the various successes of reductionist explanations, but he regretfully pointed out the negative fact that reductionism allows only upward causation.8 Furthermore, Popper believes that higher level phenomena have an ‘enrichment’ or ‘richness’ that cannot be reduced to lowerlevel phenomena, and attempts to introduce the idea of ‘downward causation’ which works from the higher level to the lower level in contraposition to the idea of reductionist upwards causation. We will explain this idea step by step. Popper thinks that the emergence in the evolutionary process produced irreducible ‘enrichment’ or ‘richness’. These are unpredictable and inexplicable from the viewpoint of the lower level. Medawar, the Nobel laureate biologist, pointed out that there is a hierarchical system of nine levels in our universe. The bigger number of levels indicates the evolutionary later and higher stage and every higher level contains the lower level (Medawar and Medawar 1983, p. 230).9 (9) (8) (7) (6) (5) (4) (3) (2) (1)

Society/community Organisms Organs Cells Cellular organelles Macromolecules Molecules Atoms Subatomic particles

At each of these levels of the hierarchy, there are a lot of concepts that are unique and peculiar to that level and cannot be explained by the concepts of the lower levels. For example, the concept of ‘trade deficit’ at the level

8 Popper describes reductionism as follows. “As a research programme, reductionalism is not only important, but it is part of the programme of science whose aim is to explain and to understand” (Popper and Eccles 1977, p. 18, italics in original). The biologist Medawar also stated that “Reductive analysis is the most successful research stratagem ever devised: it has been the making of science and technology” (Medawar and Medawar 1983, p. 227). 9 See Medawar for discussion here (Medawar 1969, pp. 15–19; Medawar and Medawar 1983, pp. 225–232).

84

M. Kogawara et al.

of political economy cannot be found in biology, and to add further examples no one can understand clearly how ideas such as ‘love’ and ‘fear’ arise at the level of organisms can be reduced to physics or chemistry (Medawar 1969, p. 15; Medawar and Medawar 1983, p. 231). These examples show the ‘enrichment’ of the higher level phenomena that Popper mentioned, and at the same time suggest the advantage of his postulation of the reality of three worlds. Furthermore, Popper believes that this kind of ‘enrichment’ controls flexibly lower phenomena through what he calls ‘downward causation’ (downward arrows in Fig. 1).10 To use a familiar expression, this is the idea that the ‘whole’ act on the ‘part’. But here we must be careful not to misunderstand Popper’s idea. He does not repeat the old idea of holism. Here he clearly distinguishes two steps. The first step is the influence from World 3 to World 2. The second step is the influence from World 2 to World 1. There is no direct influence from World 3 to World 1. It will be useful to mention below the framework of his thought (Table 1). Downward causation works from the upper to the lower as ‘plastic control’ because the upper cannot entirely determine the activities of the lower and/ or always cause the definite patterns of reactions of the lower (e.g., reflections or the predetermined reactions of a subordinate to a superior’s order as, for example, in military activities). Downward causation works as a kind of gradually regulating process and does not violate the autonomy of the lower, Table 1 Three worlds and their entities World 3 (the products of the human mind)

World 2 (the world of subjective experiences) World 1 (the world of physical objects)

(6) Works of Art and of Science (including Technology) (5) Human Language. Theories of Self and Death (4) Consciousness of Self and of Death (3) Sentience (Animal Consciousness) (2) Living Organisms (1) The Heavier Elements; Liquids and Crystals (0) Hydrogen and Helium

Source Popper and Eccles (1977, p. 16)

10

Kim (1999), while relying on the hypothesis of causal closure of the physical domain, questions downward causality (action from Worlds 2 to 1). In contemporary philosophy of mind, the physicalist view of the human mind, such as Kim’s, is more dominant, while those who defend the pluralistic interaction theory, such as Popper, belong to rather a minority group. Popper himself criticizes closedness and self-containment in The Self and Its Brain (Popper and Eccles 1977, chapter P3). The questions raised by physicalists seem to be based on mere ontological hypotheses, but surely they suggest a cluster of arguments in defense of Popper’s three-world-theory.

Popper’s Methodological Individualism …

85

where the varieties of reactions are not entirely excluded. And occasionally, the upper itself has a possibility to receive feedback from the lower’s reactions that may lead to the transformation of the upper itself. This is the content of Popper’s interactionism between each hierarchical level. He calls it ‘plastic control’ (Popper 1966, pp. 232, 241; 1994b, p. 112). We have already pointed out that social institutions are inhabitants of World 3. How does this fact connect to plastic control as described above? We may mention traffic as an example. Although pedestrian behavior is neither completely governed nor controlled by traffic rules, and although there is always room for individual choices and decisions, we can observe a certain degree of regularity and stability in traffic. Even if, strictly speaking, traffic rules (entities of World 3) govern traffic, it is impossible to exclude all traffic accidents, caused, for example, by accidentally falling asleep or deliberately speeding. While traffic rules might be insufficient to regulate our behavior, sometimes we can also amend and revise them. It is clear that there is a kind of interactive feedback process between individual behavior (World 1) and social institutions (World 3) via World 2. The above discussion of the interactionism and the plastic control of the three worlds will help us understand Popper’s MI or ‘institutionalistic individualism’ (Agassi 1960). To sum up, we can say that Popper does not rely upon psychological individualism, i.e., he rejects explanations that reduce social phenomena to the actions between individuals as pure atoms and individual psychology. At the same time, Popper’s account of explanation relies upon social institutions, insofar as they are understandable as entities of World 3 and do not have collective aims and interests as methodological holism assumes.11 Social phenomena and processes in our world are indeterministic. They are neither determined entirely by the whole nor by a part, but rather are accidental and always open to emergences.12 In order to comprehend this interactionism of three worlds properly it is necessary to have a unique

11 The institutions that Popper refers to are, so to speak, ‘entities without a knowing subject’, and cannot be reduced to World 2, such as collective consciousness. Social institutions are precisely what Hayek calls ‘the results of human action but not of human design’ (Hayek 1967), which is something completely different from what methodological holism assumes. 12 Nevertheless, a plastically controlled world is neither completely random nor chaotic, where what is to come is completely unpredictable. As mentioned above, we can find there a certain degree of order and stability in our society, created by norms and institutions. In his article “Of Clouds and Clocks” (1966), Popper makes use of the metaphors ‘clouds’, representing utter randomness, and ‘clocks’, representing deterministic phenomena, to argue that many of the physical events in our world (clocks, animals, plants, puppies, clusters of small flies, solar systems, etc.) are neither ‘clocks’ nor ‘clouds’, but are rather there between them. He argues that the world is positioned somewhere between probability 1 (clock) and 0 (cloud). There is little mention of social events in his article, but given their nature, they would likewise be placed somewhere in between 1 and 0.

86

M. Kogawara et al.

position sui generis, i.e., a position which is far away from traditional reductionism, which tries to reduce all social phenomena to the individual (atoms) or to the whole (holon), and simultaneously focus on the interaction between them.

3

The Situational Model and Situational Analysis

In the previous section, we discussed Popper’s ontological (and especially indeterministic) arguments and reached the view that he accordingly had to change his own epistemological or methodological arguments. He moved on to a type of explanation aimed at constructing models of social phenomena. Generally speaking, it is true that both the natural and social sciences use models, but in the field of the social sciences, the model is usually called the ‘situational model’, and has deep connections with the situational analysis that the method of Popper’s MI intends to provide for researchers (Popper 1994a, p. 168).13 Before describing situational analysis, it is useful to first examine the characteristics of this model. Whether natural or social, there are three types of explanation in sciences: (1) causal explanations, which combine causal laws and initial conditions; (2) probabilistic explanations, which combine laws related to probability and initial conditions; and (3) explanations based on models. According to the economist C. G. F. Simkin,14 Popper’s lifelong intimate friend, Popper did not restrict his theoretical explorations only to (1) during his later period. He had come to consider types (2) and (3) as equivalent to the types using theories and had consequently allotted important roles to them (Simkin 1993, p. 65). These attitudes were influenced by the rise of quantum mechanics, which prevailed among the natural sciences in the early 1920s, and also by the successful applications of the theory relying on probabilities. In The Post Script to the Logic of Scientific Discovery (Popper 1982a, 1982b, 1983a) written from

13

This chapter does not treat Popper’s methods of history. For that, see in particular Popper (1945 chapter 25; 1957; 1972 chapter 4). 14 Simkin, born in New Zealand, was an economist, colleague and lifelong friend of Popper’s who he met while at Canterbury University College in Christchurch, where he was posted from March 1937 (Popper 1974, p. 89). In his acknowledgements to The Open Society and Its Enemies, Popper expresses his gratitude to Simkin stating “Professor C. G. F. Simkin has not only helped me with an earlier version, but has given me the opportunity of clarifying many problems in detailed discussions over a period of nearly four years” (Popper 1945 [1966], p. xxix). For more information on Simkin, see Hacohen (2000) and Hogan (1999).

Popper’s Methodological Individualism …

87

1951 to 1956, Popper criticized the subjective interpretation of quantum mechanics (the so-called Copenhagen interpretation) and showed a very critical attitude to this interpretation. In this work, he proposed a propensity interpretation of probability as an objective. Through writing this book, Popper’s view of the theory in general underwent an important change. [Popper] has, indeed, moved more and more to the view that strict causal laws are limiting, and rather rare, cases of scientific explanation which have the general form of probability statements. An exact causal law asserts that predictions made from it, together with all relevant initial conditions, hold with certainly; i.e. with probabilities equal to 1. For probabilistic laws the stated probabilities of predictions are obviously less than 1. Exact causal laws are a limiting case because it is comparatively rare for a variable to be predictable in terms of a manageable number of other variables and initial conditions. (Simkin 1993, pp. 65–66, addition in parentheses)

Popper came gradually closer to the view that the explanations based on strict causal or universal laws are rare cases of the form of probability explanation, and arrived at the conclusion that this type of explanation is the general type of scientific explanation. Moreover, even in the case where neither causal nor probability-based explanations were available, he thought that scientific explanations and theoretical inquiries are possible by model-building. Popper discussed in detail the model-constructing method in his articles: ‘The Rationality Principle’ (1967, 1983b, 1985) and ‘Models, Instruments, and Truth’ (1994a), where he contrasts them with causal explanations and tries to make clear the differences between them. To explain this point, let’s take an example of presenting an explanation or prediction of the next lunar eclipse. Usually, this involves using theories and initial conditions. To acquire rigorous explanations and/or predictions of lunar eclipses, it is necessary to have the following two assumptions satisfied: the laws of nature and the initial conditions which depict the particular situation. On the other hand, in the case of explaining and predicting a certain type of event, such as the repeated occurrence of a lunar eclipse only during a full moon, we need not use the exact laws and initial conditions but only construct the eclipse model to easily solve this problem. A task of the model is here to depict the initial conditions as a type, or, more strictly, to present “a representation—physical, mathematical, graphic or even verbal—of typical initial conditions which bring the relevant phenomena into typical relationships with one another” (Simkin 1993, p. 67, italics in original). It may be freely constructed as a kind of mechanical model or it may be represented by drawing a perspective, for example. Or,

88

M. Kogawara et al.

for our limited purposes, it can be as ‘rough’ as a wooden model that represents the positions of Sun, Earth, and Moon (Popper 1994a, p. 163). The word ‘rough’ means that we for the time being put into parentheses the principle which accounts for the mechanism of how Earth and Moon move and obey. However, if we want to actually move the model, we need an activating (animating—in Popper’s word) principle by which the objects can be put in motion. In the case of the astronomical model, we can mention, for example, something like Newton’s laws of motion. In the problematics of social sciences, we need a principle that drives the actors in the model. This is where the so-called ‘rationality principle’ comes into play, and we will treat this theme in detail in Sect. 4. Before discussing it, it would be appropriate to point out that Popper emphasized the importance of model construction in social sciences, and admitted and emphasized its important roles and functions in theoretical researches. As for the social sciences, I have elsewhere [in the previous selection] proposed that we can construct our models by means of situational analysis, which provides us with models (rough and ready models to be sure) of typical social situations. And my thesis is that only in this way can we explain and understand what happens in society: social events. (Popper 1983b, p. 358, italics in original)

As Simkin says, it is evident that these statements show Popper’s change in his thought. In The Poverty of Historicism (1944–45, 1957) and The Open Society and Its Enemies (1945 [1966]), Popper clearly defends the thesis of the ‘unity of method’, i.e., the thesis that the sciences, whether they are natural or social, apply the same method, and asserts that theoretical social sciences should explore sociological laws.15 It was one of the most important assertions of Popper that theoretical or generalizing social sciences have no other aim than to explore sociological laws. He also argued that “there seems no reason why we should be unable to frame sociological theories which are important for all social periods” (Popper 1957, p. 101). However, when he was writing the articles ‘The Rationality Principle’ (1967, 1983b, 1985) and

15

In support of his argument, Popper shows that there are sociological laws which state as follows: “You cannot introduce agricultural tariffs and at the same time reduce the cost of living”. “You cannot, in an industrial society, organize consumers’ pressure groups as efficiently as you can organize certain producers’ pressure groups”. “You cannot have a centrally planned society with a price system that fulfills the main functions of competitive prices”. “You cannot have full employment without inflation” (Popper 1957, p. 62).

Popper’s Methodological Individualism …

89

‘Models, Instruments, and Truth’ (1994a), he stopped referring to these sociological laws and in his later life did not show any interest in them. Instead, he tried exclusively to replace them with social situation models. Simkin’s interpretation of the change in Popper’s argumentation is that the latter’s growing suspicion of the existence of universal laws in economics and the social sciences exceeded his interest in them (Simkin 1993, p. 112). Indeed, as Popper himself quite clearly affirmed, “In fact, models are even more important here because the Newtonian method of explaining and predicting singular events by universal laws and initial conditions is hardly ever applicable in the theoretical social sciences” (Popper 1994a, pp. 165– 166). The object of analysis of both sciences is never a simple phenomenon. As we have seen in the previous section, our world is indeterministic, accidental, and subject to unexpected emergence. In such a world, deterministic explanations and predictions have hardly any validity.16 Instead, a task of theoretical studies is to discover and depict tendencies and patterns in phenomena. In Hayek’s words, the social sciences try to analyze ‘complex phenomena’, and all we can do is to ‘explain in principle’ or ‘predict patterns’ (Hayek 1955, 1964).17 A paucity of universal laws in the social sciences does not necessarily lead to the abandonment of theoretical research and scientific inquiry. We can say that the situational model shows sufficient closeness to the true reality of the situation. Needless to say, it contains simplifications of reality and does not succeed in representing the complexities of the social world in its completeness. Nevertheless, we can say that it successfully provides approximate explanations. Although the rigorous testing of models (as we shall see below) may be fraught with practical difficulties, it is not logically impossible to improve or modify them through refutations. For Popper, it is clear that the social sciences always maintain a unity of method, in other words, the method of conjectures and refutations. It is true that theoretical social sciences cannot construct models of a situation, unless we can succeed in describing the concrete social situation to which a model is to be applied. We can say that this description corresponds

16

Deterministic explanations and predictions may be possible under the conditions that the effects from surplus variables can be completely controlled, as in the two-body problem in the solar system or in laboratory experiments in the natural sciences. However, it would not be easy to create such ideal experimental conditions in the social sciences. 17 It is possible to say that Popper’s construction of the situational model and Hayek’s explanation of the principle seem to be almost identical in terms of explaining phenomena. Hayek (1955) takes up the construction of the model in the context of examining the explanation of the principle. Although we cannot here take Hayek’s view into consideration, we think that Popper was influenced by Hayek and came to share his view.

90

M. Kogawara et al.

to the initial conditions in the natural sciences. In order to construct a situational model, it is necessary to depict or reconstruct the social situation as a type. Such explanation, being peculiar to the social sciences, is called ‘situational analysis’ or ‘the logic of the situation’. The Poverty of Historicism (1944–45, 1957) and The Open Society and Its Enemies (1945 [1966]) sketched the main features of the method of situational analysis and its applications. Before presenting them, Popper criticized J. S. Mill’s psychological individualism in order to defend situational analysis. By doing so, he tried to show the merit of applying situational analysis. This method attempts to explain in terms of the external circumstances that give rise to an action, while as far as possible avoiding description of the psychological aspects of the actor. An example will illustrate the key points concretely. Popper applies this method to the case of Richard, a pedestrian who hurries to cross the road (Popper 1967, 1983b, 1985, 1994a). To analyze and explain Richard’s behavior, it is necessary to refer to the following main conditions: (a) the physical situation (World 1), such as the traffic jam, the coming and going of many cars on the road, and people’s movements and (b) traffic rules, signals, and other social institutions (World 3) restricting the pedestrian movements.18 In addition to these, there are (c) the beliefs and purposes (World 2) of the pedestrians. The descriptions of Richard’s mental states (World 2) are usually interpreted as the ones treated by psychology, but Popper proposes to treat them not as psychological facts that can be inquired and checked by psychology, but as the one piece of the configurations of “elements of the objective social situation” (Popper 1994a, p. 167, italics in original). For example, Richard’s purpose will be transformed and arranged as an element of the typical problematic situation (World 3) with which the actor is faced. Consequently, his purpose will be interpreted not as his purpose, i.e., ‘I had to catch the train at a certain time to see the opera’, but rather as the purpose in general, i.e., ‘I was about to cross the street’— needless to say, this is a typical problem situation. In this way, we are able to construct a typical impersonal model, which is interpretable by more than a mere description of a singular example of Richard. Thus, situational analysis is an attempt to specify a situation as of a repeatable type (or pattern) by analyzing particular actions or events from the viewpoint that pays attention mostly to the objective situation surrounding the actor, and/or the viewpoint that transfers them to the outside objective situation where the actor performs various movements. In relation to Popper’s three-worlds-theory, it can be said

18

It would be more accurate to say that pedestrian crossings and traffic signals belong to Worlds 1 and 3, because they are both physical objects and social rules.

Popper’s Methodological Individualism …

91

that situational analysis is to analyze and explain social phenomena by using the entities of Worlds 3 and 1, with lesser and lesser reference to World 2. It is possible to illustrate this point by contrasting it to an argument from the social psychologist Lewin. He comprehends human behavior according to the following formula: behavior B is a function of the acting person’s P (predisposition or individual characteristic of the person) and his/her environment E (Lewin 1935). B = f (P, E) Relying on Lewin’s formula, we can observe that the situational analysis tries to minimize the element P as much as possible. The minimization of P implies a thesis that researchers should in principle use only the ‘rationality principle’ to understand P, and everything else should be explained as the objective elements E in the situation as much as possible. In other words, it assumes that a person behaves only adequately to the situation. This assumption is also called the rationality principle on which one can perform one’s own actions or behavior. Under this principle the person P can make one’s actions. It seems to be quite clear that it is a principle or a postulate void of empirical testability because everyone does act according to what he or she sees and understands. In other words, all of one’s movements go into making one’s behavior adjust to a situation from one’s perspective. Concerning the rationality principle, which works as an animating principle of the model, we can say safely that the most important methodological implication of constructing the situational model is to ‘analyze thoroughly the typical relations between E and B’.19

4

The Role of the Rationality Principle in Situational Analysis

An attempt to defend Popper’s MI and his concept of situational analysis inevitably leads us to the question of its methodological status and at the same time the functions of the rationality principle (RP) contained in the situational analysis. In fact, situational analysis and the RP have caused many debates, not only in the field of philosophy of science but also especially in economics, a branch of social science. In this section, we will pay attention exclusively to the critical debates among scholars who are interested 19

Popper’s arguments for situational analysis differ from those of Lewin. The reason to have mentioned Lewin’s formula was only to conveniently illustrate Popper’s claims.

92

M. Kogawara et al.

in Popper’s methodology, such as economists Hands and Caldwell, and the historian of science Nadeau, and examine some of the typical objections raised against Popper. We hope this chapter will help us to proceed to a better understanding of Popper’s situational model, its functions, and the RP itself. Before doing so, we would like to mention a positive evaluation of the situational analysis. While Hands and Caldwell have been critical of RP, they have positively and highly applauded the applications of situational analysis in economics. They believe that explanations in economics have traditionally relied on situational analysis and assert that its applications have resulted in the theoretical achievements of this discipline. According to Hands, the standard microeconomic theory is a special case of situational analysis, through which we can arrive at the foundations to explain not only the behavior of micro-individuals and firms but also aggregate phenomena such as the formation of equilibrium prices in markets (Hands 1992, pp. 28–29; 1996, p. 319). According to Caldwell, situational analysis is also used in research fields such as the economics of information, transaction cost economics, and game theory (Caldwell 2004, pp. 388–391). These claims provide evidence that situational analysis is a very useful and powerful tool in economics. However, they and their colleagues have cast doubt on RP itself. The main points of their doubts will be formulated as follows: Is Popper’s methodology of social sciences, including situational analysis and the RP, compatible with the falsificationism he developed mainly in The Logic of Scientific Discovery, which discusses for the most part the natural sciences and demands the methodological rule that science should have high falsifiability? This problem has been discussed primarily by Koertge (1972, 1979), a philosopher of science, but also by many economists who followed her in the field of methodology of economics and have emphasized that this point is problematic. For example, Hands (1991) describes this problem as a ‘tension’ between Poppers (as a philosopher of social sciences or situational analysis) and Poppern (as a falsificationist philosopher of the natural sciences). Caldwell (1991) calls it a ‘dilemma’ and Nadeau (1993) treats it as an ‘asymmetry’. Various researchers have labeled it in different ways, but we can safely say that they have discussed nearly the same kind of problem. First, we can consider one of the most typical arguments in Hands. He distinguishes two Poppers (1985): Poppern and Poppers . Poppern was the author of The Logic of Scientific Discovery (1959) and Conjectures and Refutations (1965) and was interested in formulating falsifiability as a demarcation criterion to distinguish between so-called scientific and metaphysical statements. Popper could be portrayed in this period as a rigorous falsificationist encouraging bold conjectures and rigorous testing of them in the natural

Popper’s Methodological Individualism …

93

sciences. This is the reason why Hands calls the Popper of this period Poppern . In addition to this, he asserts that there is Poppers , whose claims about economics and the social sciences contradict those of Poppern . Hands refer to a number of disputable points, the most important of which concerns the falsifiability of the RP. Since the rationality principle serves as a general law in the explicans of explanations in social science it must satisfy the necessary conditions for general law status, the most important of these being that it be falsifiable. If it is not (at least potentially) falsifiable then nothing demarcates explanations which employ the rationality principle from merely metaphysical explanations. (Hands 1985, p. 87, italics in original)

However, in referring to Popper’s article, ‘The Rationality Principle’ (1967), he asserts that ‘the RP is neither falsifiable nor testable’, and if the RP is not falsifiable then the social scientific explanations based on this principle will get extremely near to metaphysical explanations. As a result, Popper’s claim that the social sciences are very similar to the natural sciences seems to contradict his defense of the approaches of the social sciences. Hands (1991) repeated similar arguments and again pointed out that the RP will be accused of inconsistencies arising from the conflicts between Poppers and Poppern . If we interpret the RP from the perspective of the actor, then this theory becomes unfalsifiable: if we interpret it from the perspective of the observer, then it is false. Thus the explanations of the social sciences become either irrefutable metaphysical statements or a false cluster of explanatory statements based on false general laws. Furthermore, Hands points out in his review of The Myth of the Framework that even if the RP is falsifiable it is possible to evade falsifications, for example, by using the following maneuvers. When an explanation involving the RP is tested and found faulty, Popper asserts that it is ‘sound methodological policy to decide not to make the rationality principle, but the rest of the theory—that is, the model—accountable’ (Popper 1994a, p. 177). Such a defensive/ad hoc strategy hardly seems consistent with the falsificationist policy of sticking one’s neck out: the method of bold conjecture and severe test. (Hands 1996, p. 319)

Secondly, we should take up Caldwell (1991). He refers to the same point, though in a different way from Hands. Caldwell formulates Popper’s ‘dilemma’ as a conflict between the following five statements.

94

M. Kogawara et al.

(a) Popper maintains that the structure of scientific explanation in both the natural and the social sciences follows the same pattern. In both cases, an explanandum is deduced from an explanans, which contains sentences describing initial conditions and at least one universal law. (b) Popper asserts that “… the explanans ought to be true although it will not, in general, be known to be true; in any case, it must not be known to be false …” (Popper 1983a, p. 132) (c) When theory revision is called for, the prime directive for Popper the falsificationist is to avoid the use of immunizing stratagems, which are ad hoc adjustments of a theory undertaken to protect it from refutation. (d) The universal law used in the social sciences, the rationality principle, is false. (This violates condition b.) (e) Yet Popper the situational analyst insists that the universal law of the social sciences, though false, should never be rejected. Instead, the theory in which it is used should be adjusted until the agent’s actions can be shown to follow from the logic of the situation. (This violates condition c.) (Caldwell 1991, p. 20)

In other words, Caldwell formulates Popper’s dilemma in such a way that there are at least two conflicts, that is, between statements b and d, and statements c and e. (We can formulate some of Hands’ arguments in the same way as Caldwell formulated them, i.e., as conflicts between statements.) Recently, Caldwell again took up the conflict between statements c and e and discussed it (2004).20 The problem that Hands and Caldwell treat has been discussed again and again recently. Although we can partly admit the validity of these criticisms during the days when Popper was writing The Poverty of Historicism and The Open Society and Its Enemies, we would have to say that they had lost their validity in the days of the later Popper. We believe that a good understanding 20 Caldwell (2004) states as follows: “Popper insisted that, whenever a theory employing the rationality principle is falsified, the appropriate thing to do is rethink one’s model of the situation. Crucially, one should never reject the rationality principle. By following this methodological principle, one gets ‘far more interesting and informative’ models (Popper 1985, 362). But it also means that, as a matter of methodological principle, the falsification of theories based on the method of situational analysis is never taken as grounds for rejecting the theory. Instead, any falsification immediately leads to an ad hoc theory adjustment, a redescription of the problem situation, thereby immunizing the theory from falsification. That Popper’s description of how explanation takes place in the social sciences appears to be inconsistent with his prescriptions about the importance of falsifiability and the avoidance of immunizing stratagems has often been remarked on by methodologists (e.g., Hands 1985; Caldwell 1991)” (Caldwell 2004, pp. 395–396, italics in original).

Popper’s Methodological Individualism …

95

of the later Popper’s arguments will easily show how these assertions are misguided and based on false preconceptions and assumptions. Both Hands and Caldwell seem to have equally miscomprehended Popper’s concept of explanations in the social sciences. They have an incorrect view that Popper’s conception of sociological explanation exclusively relies on causal-type explanations based on the so-called Deductive-Nomological (D-N) model. It is easy to ascertain the deficiencies in their view. They assert wrongly that what makes possible both the explanation and testing of hypotheses in the social sciences is the same procedure as is assumed in the D-N model. We have seen in Sect. 3 that this is not the case. The later Popper supported not only causal type of explanations supported by the D-N model but also explanations based on the situational model. Other researchers (Koertge, for example) have completely overlooked this important fact. Also, it may be useful here to mention Nadeau’s argument (1993). (He also points out the inconsistencies in Popper’s alleged assertion that the methodology is different in the natural and social sciences, although we do not here discuss it in detail.) He contrasts Popper’s theoretical accounts of the natural sciences with those of the social sciences, as shown in Fig. 2. On the lefthand side of the diagram we can find the explanatory schema used mainly in the natural sciences, i.e., the so-called D-N model formulated by Hempel and Oppenheim.21 This D-N explanatory schema is presupposed implicitly in Popper’s discussion of the RP, and Popper developed a new type of explanation in the social sciences, i.e., the SL explanatory schema (right side of Fig. 2), which is based on ‘Situational Logic’ (Nadeau 1993, pp. 448–449). It is necessary to add some comments to these schemata. According to Nadeau, the SL explanatory schema shows the structural identity with the D-N explanatory schema, and the RP has the same functional position as D-N Explanation Schema Explanans:

SL Explanation Schema Explanans: or Theory

Initial conditions

Model

Theory or Law

Rationality principle

Explanandum: Natural Event

Explanandum: Social Event-Type

Fig. 2 D-N explanation schema and SL explanation schema (Source Nadeau 1993, p. 450)

21

Nadeau calls it the D-N diagram rather than the D-N model. We will use the names interchangeably.

96

M. Kogawara et al.

the universal law in the D-N explanatory schema.22 In the SL explanatory schema, the explanandum is deduced from the explanans via RP which performs the same role as the covering law in the natural sciences. Thus the RP is ascribed falsifiability like universal laws. This leads Nadeau to ask whether the RP is falsifiable or not. The answer ‘No’ led him to the inconsistency thesis and inevitably to dissatisfaction, and consequently forced him to mistakenly determine that falsificationism had lost its function. Likewise, it is easy to envisage that Hands’ and Caldwell’s arguments presuppose wrongly the structural identity of D-N and SL explanatory schema, so that they are led to an incorrect conclusion.23 We have already shown in Sect. 3 that this presupposition itself is false. The later Popper clearly had explicated the complicated problematic situation of SL explanatory schema. We have already said that the later Popper allows and recommends the application of social situational models instead of using the D-N model in the theoretical social sciences. The RP supports the situational models. They relate primarily to the relations between events as types (results) and initial conditions as types (causes), when strict universal laws and precise initial conditions are not available. Explanations by using models are different from causal explanations based on the D-N explanatory schema. Neither does the RP itself have the same function as the universal laws which work in the D-N explanatory schema nor is it an object at which criticism should be aimed if unsatisfactory cases against the model are discovered. As Popper himself states, the RP is simply the activating (animating) principle of the model and does not play any role as an ‘empirical explanatory theory’. When we discover the defects in the models, it is far better to question the constructed situational models (or initial conditions as a type), instead of the RP. It is better to understand this principle as a ‘sound methodological policy’ which directs us to the explication and discussion of the model (Popper 1994a, pp. 169–170, 177). As is well known, a falsification neither directly/always challenges a universal law nor forces us to discard it. Though it has often been asserted that an immunization of the RP is a bad policy,

22

Nadeau states as follows: “if the universal statement (the theoretical law) constitutes the nomological part of the determinist D-N schema, it is apparently the RP that plays this explanatory role in the SL schema” (Nadeau 1993, p. 449). 23 Hands (1991) calls the tension between Popper and Popper an ‘essential tension’ and argues that n s “it stems from Popper’s unfailing commitment to the covering-law model of scientific explanation” (p. 108). Caldwell (1991) also states that “Popper maintains that the structure of scientific explanation in both the natural and the social sciences follows the same pattern. In both cases, an explanandum is deduced from an explanans, which contains sentences describing initial conditions and at least one universal law” (p. 20).

Popper’s Methodological Individualism …

97

Popper intended exactly the opposite: he recommends treating this principle as a methodological policy to make the model more falsifiable.24 Finally, we would like to make a conjecture concerning the reason why Popper defends the RP. He admits that the RP itself is empirically ‘false’. However, this principle is very useful and fruitful as a methodological council. Without saying it is evident that people do not necessarily behave rationally, sometimes their behaviors look very irrational. However, if we adopt the RP then we must assume that there is rationality in their irrational appearance. Therefore, the RP stimulates us to inquire into the hidden elements in their behavior and to reexamine the conditions of the actors thoroughly. In other words, the RP demands us to attempt a more precise situational analysis; for example, by assuming the counterfactual facts. This shows clearly that the RP is an adequate methodological policy. The RP is a heuristic device or tool for facilitating and advancing situational analysis (Kogawara 2010, p. 140).

5

Conclusion

From a historical perspective, it is very interesting to notice that several economists have raised nearly the same kinds of questions concerning Popper’s methodology of the social sciences, as if they speak in one voice. How can this interesting situation be explained? Applying the situational analysis to these interesting facts, we are perhaps able to constitute hypothetically a type of event, and we may be able to ascertain some of the common situational factors that compelled them to simultaneously misinterpret Popper’s MI. First of all, it should be remarked that Popper’s description of the status of the RP and its functions is vague and not at all easy to comprehend. As Koertge pointed out, it is true that Popper did not always formulate the RP in a clear way, and Popper’s views seem to have changed slightly over time (Koertge 1979, p. 86). It is also true that this has produced various interpretations of the RP, and consequently brought about a number of misunderstandings. Popper himself may deserve at least some of the blame for this misfortune. 24

Although we cannot take it up in detail, Caldwell (1982) seems to think that it is impossible to expose a model to empirical testing (falsification) (p. 241). However, Popper does not think so. Since initial conditions can be falsified, the procedure of empirical testing is essential in constructing the situational model. In fact, in The Logic of Scientific Discovery, Popper describes the procedure of falsification as follows: “By means of this mode of inference we falsify the whole system (the theory as well as the initial conditions) which was required for the deduction of the statement p, i.e. of the falsified statement” (Popper 1959, p. 56, italics in original).

98

M. Kogawara et al.

However, it is not difficult to present a conjecture to explain this situation. It is the conditions of the publications at that time that hindered the understanding of Popper. We can identify these conditions as the worst factor. His most important article, ‘Models, Instruments, and Truth’, was not published until 1994. This was the most comprehensive article concerning the functions of situational analysis. This article describes how to construct the situational model, and how to understand both the situational analysis and the RP in the theoretical social sciences. This article was originally a revised version of a lecture that Popper gave under the same title to the Harvard economics department in 1963, excerpts of which were first published in French in 1967 as the article ‘La Rationalité et le Statut du Principe de Rationalité’, and in English in 1983 and 1985 as ‘The Rationality Principle’.25 It was the English version of these excerpts that almost all the concerned economists referred to. It is for this reason that most of their criticism centered on this article during the period from 1985 to the early 1990s. However, the excerpts amount to less than a third of the original article ‘Models, Instruments, and Truth’, and in terms of the content they do not sufficiently provide a comprehension of the relations between situational models, situational analysis, and the RP, as we have shown in this chapter. It is possible to suppose that Popper’s insufficient account itself brought about the lamentable situation of the misunderstandings and distortions of Popper’s MI (contrary to his intentions). Finally, in 1994, the article ‘Models, Instruments, and Truth’ was published as Chapter 8 of The Myth of the Framework, and it became possible to approach the assertions of Popper’s MI in more detail. We believe that readers of this chapter will recognize that our purpose lies in facilitating the detailed and full comprehension of this article. This chapter has also discussed Popper’s MI from the perspective of his later ideas (indeterminism, evolutionary theory of knowledge, three-worlds-theory, mind-body interactionism, etc.). Some may wonder why such a comprehensive discussion is necessary. In order to understand Popper’s methodology of the social sciences more accurately, it is necessary to locate it in a broader textual and historical context. As Simkin writes, “these [Popper’s] ideas and their developments are spread over a range of writings separated by wide time intervals, and are sometimes partly obscured by distracting polemical or historical material. It has seemed worthwhile, therefore, to attempt bringing them together in a systematic way and as simply as accuracy permits” (Simkin 1993, p. 5). This is very good advice for us to learn the important and necessary conditions for understanding 25

The English versions were published as Pocket Popper (1983) and Popper Selections (1985), edited by David Miller.

Popper’s Methodological Individualism …

99

Popper’s methodology in a better way. Of course, we do not have full confidence in having carried out Simkin’s advice, but it is not a vital point. Even if our interpretation is not adequate, it is evident that somebody else will try to do a good job of improving our interpretations, because the research materials concerning Popper are increasingly abundant and more accessible than ever before.26

References Agassi, J. (1960). The methodological individualism. The British Journal of Sociology, 11(3), 244–270. Basu, K. (2018). Methodological individualism. In S. N. Durlauf and L. E. Blume (eds.), The new Palgrave dictionary of economics. Third ed., Vol. 13. London: Palgrave Macmillan. Caldwell, B. J. (1982). Beyond positivism: Economic methodology in the twentieth century. Revised ed. London: Routledge, 1994. Caldwell, B. J. (1991). Clarifying Popper. Journal of Economic Literature, 29 (1), 1–33. Caldwell, B. J. (2004). Hayek’s challenge: An intellectual biography of F. A. Hayek. Chicago: University of Chicago Press. Di Iorio, F. (2016). World 3 and methodological individualism in Popper’s thought. Philosophy of the Social Sciences, 46 (4), 352–374. Eccles, J. C., & Robinson, D. N. (1984). The wonder of being human: Our brain and our mind . New York: The Free Press. Gorton, W. A. (2006). Karl Popper and the social sciences. New York: State University of New York Press. Hacohen, M. H. (2000). Karl Popper: The formative years, 1902–1945. Cambridge: Cambridge University Press. Haller, M. (2003). Soziologische theorie im systematisch-kritischen vergleich. VS Verlag fuer Sozialwissenschaften, GWV Fachverlage Gmbh, Wiesbaden 2006. Hands, D. W. (1985). Karl Popper and economic methodology: A new look. Economics and Philosophy, 1(1), 83–99. Hands, D. W. (1991). Popper, the rationality principle and economic explanation. In G. K. Shaw (ed.), Economics, culture and education: Essays in honour of Mark Blaug (pp. 108–119). Aldershot: Edward Elgar. Hands, D. W. (1992). Falsification, situational analysis and scientific research programs: The Popperian tradition in economic methodology. In N. De Marchi (ed.), Post-Popperian methodology of economics: Recovering practice (pp. 20–63). Boston/Dordrecht/London: Kluwer Academic Publishers. 26

We may mention a very critical book, William A. Gorton (2006). Also we may refer to the chapter 10.12 of Herbert Keuth (2000).

100

M. Kogawara et al.

Hands, D. W. (1996). Karl Popper on the myth of framework: lukewarm Popperians +1, unrepentant Popperians −1: A Review of Karl R. Popper’s The Myth of the Framework: In defense of science and rationality. Journal of Economic Methodology, 3(2), 317–322. Hands, D. W. (2001). Reflection without rules: Economic methodology and contemporary science theory. Cambridge: Cambridge University Press. Hayek, F. A. (1955). Degree of explanation. In Hayek, F. A., Studies in philosophy, politics and economics (pp. 3–21). London: Routledge & Kegan Paul, 1967. Hayek, F. A. (1964). The theory of complex phenomena. In Hayek, F. A., Studies in philosophy, politics and economics (pp. 22–42). London: Routledge & Kegan Paul, 1967. Hayek, F. A. (1967). The results of human action but not of human design. In Hayek, F. A., Studies in philosophy, politics and economics (pp. 96–105). London: Routledge & Kegan Paul. Hogan, W. P. (1999). Colin George Frederick Simkin, 1915–1998. The Economic Record , 75 (230), 313–322. Keuth, H. (2000). Die Philosophie Karl Poppers. J.C.B. Mohr (English translation: The Philosophy of Karl Popper, Cambridge UP, 2005). Kim, J. (1999). Making Sense of Emergence. Philosophical Studies, 95, 3–36. Koertge (1972). On Popper’s Philosophy of Social Science. PSA: Proceedings of the Biennial Meeting of the Philosophy of Science Association, 1972, 195–207. Koertge N. (1979). The methodological status of Popper’s rationality principle. Theory and Decision, 10, 83–95. Kogawara, M. (1997). Gendaisisou no boukensha tachi dai14kan: Popper Hihantekigourisyugi, Kodansha. Kogawara, M. (2010). Hanshousyugi. Tohoku University Press. Lewin, K. (1935). A dynamic theory of personality. New York and London: McGrawHill Book Company, Inc. Medawar, P. B. (1969). Induction and intuition in scientific thought. London: Methuen & Co. Ltd. Medawar, P. B., & Medawar, J. S. (1983). Aristotle to zoo: A philosophical dictionary of biology. Oxford University Press Paperback. Oxford: Oxford University Press, 1985. Miller D. (1983). Pocket Popper. London: Fontana. Miller D. (1985). Popper selections. Princeton: Princeton University Press. Nadeau, R. (1993). Confuting Popper on the rationality principle. Philosophy of the Social Sciences, 23(4), 446–467. Niemann, H.-J. (2004). Lexikon des kritischen Rationalismus. Mohr Siebeck. Popper, K. R. (1944–45). The poverty of historicism I, II, & III. Economica, 11(42), 86–103, 11(43), 119–137, 12(46), 69–89. Popper, K. R. (1945 [1966]). The open society and it’s enemies. Fifth ed. London: Routledge, 2011. Popper, K. R. (1957). The poverty of historicism. ARK ed. London: Routledge, 1994. Popper, K. R. (1959). The logic of scientific discovery. London: Routledge, 2002.

Popper’s Methodological Individualism …

101

Popper, K. R. (1966). Of clouds and clocks: An approach to the problem of rationality and the freedom of man. St Lous, MO: Washington University Press. Reprinted in Popper 1972, chapter 6. Popper, K. R. (1967). La Rationalité et le Statut du Principe de Rationalité. In E.M. Claassen (ed.), Les fondements philosophiques des systemès économiques (pp. 142– 150), Paris: Payot. Popper, K. R. (1972). Objective knowledge: An evolutionary approach. Revised ed. Oxford: Oxford University Press, 1979. Popper, K. R. (1974). “Intellectual Autobiography,” in P. A. Schilpp (ed.), The philosophy of Karl Popper: Book I (pp. 3–181). Illinois: Open Court. This part is independently published by Fontana, 1976, as a book: Unended Quest; An Intellectual Autobiography. Popper, K. R. (1982a). Quantum theory and the schism in physics. Paperback ed. London: Routledge, 1992. Popper, K. R. (1982b). The open universe: An argument for indeterminism. Paperback ed. London: Routledge, 1988. Popper, K. R. (1983a). Realism and the aim of science. Paperback ed. London: Routledge, 1985. Popper, K. R. (1983b). The rationality principle. In D. Miller (ed.), Pocket Popper (pp. 357–365). London: Fontana. Popper, K. R. (1985). The rationality principle. In D. Miller (ed.), Popper selections (pp. 357–365). Princeton: Princeton University Press. Popper, K. R. (1994a). The myth of the framework: In defence of science and rationality. Paperback ed. London: Routledge, 1996. Popper, K. R. (1994b). Knowledge and the body-mind problems: In defence of interaction. Paperback ed. London: Routledge, 1996. Popper, K. R., & Eccles, J.C. (1977). The self and it’s brain: An argument for interactionism. London: Routledge. Simkin, C. G. F. (1993). Popper’s view on natural and social science. Leiden: Brill.

Individualism and Holism in the Social Sciences Raymond Boudon

In a letter that is frequently cited,1 Max Weber writes: “Sociology, too, can only proceed based on the actions of one, a few or many separate individuals. That is why it must adopt strictly ‘individualistic’ methods.” Firstly, it is important to note that Max Weber places the word “individualistic” in quotation marks, doubtless to indicate that it should not be taken in its usual sense, which is related to ethics or sociology. In the ethical meaning, individualism is a doctrine that makes the person—the individual—a reference point that one cannot go beyond. In this sense, “individualism” is opposed, in particular, to “collectivism.” In the sociological sense, a society is said to be individualistic when the autonomy granted to individuals through laws and moral or social constraints is very broad. To name these types of society, Durkheim uses the related but by no means synonymous terms of “egotism” and “individualism.” Likewise, 1

See Birnbaum (1984, p. 192), Elridge (1970), Mommsen (1965, p. 25).

This chapter is a translation of Boudon, R. (1991). Individualisme et holisme dans les sciences sociales. In Pierre Birnbaum & Jean Leca, Sur l’individualisme. Theories et méthodes (pp. 45–59). Paris: Presses de la Fondation Nationale des Sciences Politiques: https://www.cairn.info/sur-l-ind ividualisme--9782724605945-page-45.htm

R. Boudon (B) Department of Sociology, Sorbonne University and Academy of Moral and Political Sciences, Paris, France © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 N. Bulle and F. Di Iorio (eds.), The Palgrave Handbook of Methodological Individualism, https://doi.org/10.1007/978-3-031-41512-8_6

103

104

R. Boudon

Tocqueville claims to be struck by the development of “individualism” in mid-nineteenth-century American society and uses this term to conclude that American citizens had appeared to him as primarily concerned with their private life and had little interest in public life. In its methodological sense, individualism has quite a different meaning. And it is to indicate that he is dealing here with individualism that will later be described as “methodological,” that Weber places the term in quotation marks in the text cited above. In the methodological sense, the notion of “individualism” is opposed to that of “holism” or, in Piaget’s vocabulary, to that of “totalitarian realism.” In short, there is the same relationship between individualism in the methodological sense and individualism in the ethical or sociological sense as there is between the word “bad” in Persian and the word “bad” in English. According to André Martinet, both words are pronounced in exactly the same way. But because they belong to two distinct languages, their meanings are different. Similarly, the notion of individualism has a very different meaning depending on whether it appears in the context of sociology or ethics, or in the context of knowledge theory. The principle of “methodological individualism” states that, in order to explain any social phenomenon—whether it falls within demography, political science, sociology, or any other particular social science—it is essential to reconstruct the motivations of the individuals concerned by the phenomenon in question, and view that phenomenon as the result of the aggregation of individual behaviors dictated by those motivations. And this principle is valid regardless of the form of the phenomenon to be explained, whether it is unique or statistically consistent, or whether it is reflected in a set of quantitative or qualitative data, etc. Let us consider, for example, a simple relationship between two economic phenomena: Under certain conditions, if a product’s price rises, demand for that product declines. One can, of course, simply record this macroscopic relationship. But it is much more interesting to ask what the causes are, which in this case are not hard to identify. One need only imagine what might go through the mind of a consumer who finds that the price of that product has risen: Except in the case where this product corresponds to an irreducible need and it is in a monopoly situation or in competition with other products with prices that are increasing in the same proportion, the consumers, obeying an easily understandable motivation, will strive to preserve their purchasing power by focusing on other products. Since many, if not all, consumers are likely to respond in the same way—because they will all have the same reasons to react in that way—their aggregated

Individualism and Holism in the Social Sciences

105

behavior will have an overall effect, namely a reduction in the overall demand for the product in question. Let us designate as M the phenomenon we are proposing to explain, which here is the macroscopic relationship between price and demand. The structure of the explanation consists of observing that M results from the aggregation of microscopic, i.e., individual, behaviors. Which can be written as: M = M (m), where M () has the meaning of “function of ” and m is the behavior of any individual. As for this behavior, it is itself a function of a set of data that are not microscopic, but macroscopic, i.e., defined at the system level. These macroscopic data can be referred to as P: m = m (P ). In the example mentioned above, P refers to the fact that the price of the product concerned has risen on the market, but also to the fact that the product is not a vital necessity and is not in a monopoly situation. The structure of the preceding explanation can therefore be represented by the expression M = M [m(P )]. Verbally, the global phenomenon M results from a set of individual behaviors m resulting from motivations that are themselves affected by global data P. When Weber writes that “sociology, too, must proceed from the actions of separate individuals,” he means, since the letter from which the quote is taken is addressed to an economist, that, like economics, the discipline in which this methodology is traditional, sociology must use the principle now known as “methodological individualism.” Is this wishful thinking? Certainly not. Because, while the principle of methodological individualism is often challenged by social sciences other than economics for reasons that will be examined, it is also widely used by them. As for classical sociology, as I have tried to show elsewhere (Boudon 1979/ 1981), many analyses of Marx, Tocqueville, Max Weber, Sombart, or Simmel reflect close adherence to this principle and the conception of explanation it summarizes. And when we look at modern sociology, we can easily see that, while the individualistic paradigm is far from dominant, it is nevertheless present in many studies, and perhaps the most interesting ones. To support this point, I shall simply give a few quick examples. I shall first address the case of the analysis of social movements, mobilization, and collective action. There are, of course, many studies in this field of a holistic type. Le Bon’s old theory that social movements are understandable once one assumes that the individual is, as it were, dissolved in the group is still sometimes taken seriously (Le Bon 1895/1963; Moscovici 1981). In reality, I am not sure that a single example of a collective movement has ever been explained by taking this point of view. And it is likely that Le Bon’s theory explains, at best, very marginal phenomena.

106

R. Boudon

On the other hand, some sociologists continue to analyze, as Hegel did, social movements as carriers of the meaning of history and as the seat of historicity, adopting a holistic and teleological point of view (Touraine 1973, 1984). In contrast, a significant number of studies of social movements use the paradigm of methodological individualism. In this respect, I will only mention the analysis proposed by Anthony Oberschall (1973) of the black Americans’ civil rights movement during the 1960s. It views this movement and interprets the different forms it has taken as the effect of the strategies deployed by the protagonists depending on the context in which they found themselves. To limit myself to one aspect of his work, I will mention the part of his book that explains why the black Americans’ civil rights movement was violent in the north and non-violent in the south of the United States. Often, the non-violence of the movement in the south was viewed as being due to the more oppressive nature of white domination in the deep south. In fact, Oberschall has shown that this holistic explanation is very inadequate and that the non-violence used by the black leaders of the south should be viewed as a strategy that was well adapted to the context of the south. In the first place, it is true that violence posed greater risks of retaliation than in the north. But it is also true that, in 1960, the black movement in the south had the sympathy of the northern white elites, who saw the segregation of the south as an unbearable archaism. Moreover, the movement could rely on the sympathy of the administration, which at that time was trying to promote racial equality. Secondly, the black Americans’ civil rights movement in the south had the support of the Protestant clergy. The black Protestant churches are the only organizations that are not dominated by whites. Furthermore, they play a considerable role because, apart from their religious functions, they take on all kinds of social, cultural, and political functions. For this reason, in the 1960s, black leaders were in a favorable strategic position, able to count on the support of enlightened opinion in the north, the administration, and the Protestant clergy of the south. It is clear that by adopting a non-violent strategy, not only could they hope to minimize retaliation from whites in the south but they were also more likely to retain that support. By resorting to violence, they would have risked alienating the Protestant clergy; they would have placed the administration in a difficult situation; brought water to the mill of conservative parliamentarians in the south; and limited their audience on campuses in northern universities. In the north, the situation was quite different. First, the black community was segmented: the middle class was separated from its old ties, and

Individualism and Holism in the Social Sciences

107

a completely fragmented black proletariat, which did not participate in any form of association, was crammed into ghettos. The black middle class was not interested in the black problem, and the fragmentation of blacks in the ghettos meant that mobilization was very weak. Here, the churches did not play the supporting role they had in the south. This is why the black movement took the form of explosions here, often triggered by blunders by the police. And later, when the attention of the northern elites turned away from the black problem to focus on the Vietnam problem, the black Americans’ civil rights movement in the north took on a radical and sectarian form, because it sought to reclaim the attention of the political and intellectual elites, either by returning to spontaneous violence or by inciting violence. An individualistic methodology is being followed here: The context of the north (P ) is different from that of the south (P ´). Moreover, the situation was not the same when the strategy of non-violence was deployed in the south and when the radical forms of the movement appeared in the north. Let us designate this situational variable as N /N ' . The structure of the analysis consists of showing that the system (P, N ) led to strategies on the part of black leaders and to attitudes m on the part of the black population that differed from the strategies and attitudes m’ that were to develop later in the north, due to the P ' context and the N ' situation. This example allows us to raise an important point. Individualistic methodology does not imply a fragmented vision of societies. It does not preclude, and it even requires that individuals should be considered as inserted into a social context. Furthermore, it invites us to treat as identical those individuals that are in the same situation and therefore makes the analysis of collective phenomena possible. But it differs from holistic methodology in that it is always compelled to highlight the individual reasons for those collective phenomena, and refuses, in principle, to treat a group as an actor that, like an individual, has an identity, a consciousness, and a will. And when it treats a group like an individual, it is because it has good sociological reasons to do so. Therefore, the economic sociology of individualistic aspiration will have no difficulty considering the family group as a decision-making unit. Finally, and I will return to this point, individualistic methodology treats the individual as fundamentally “rational.” In order to show that the individualistic paradigm holds an important place in the most diverse sectors of modern sociology, I will now move on to quite a different chapter, that of developmental sociology. Here, too, one can easily detect the presence of a holistic way of thinking. Authors as different as Margaret Mead (1953) and Hoselitz (1952) argued

108

R. Boudon

that what made it so difficult to lift traditional societies out of underdevelopment, was essentially the result of the fact that in such societies traditions are so binding, and family, religious, and economic institutions are so interdependent, that economic progress involves a complete disruption of their social structures. In the same way, authors such as Lerner (1958) argue that the introduction of technical change or the development of education or means of communication are bound, given the close interdependence of the institutions of traditional societies, to cause a general disruption of their social structures. Other authors, such as Dumont (1966, 1976, 1983), want societies to be dominated by an ideology—or a culture—that permeates social life as a whole. India would be a society organized around the principle of homo hierarchicus, while Western societies would be homo-equals societies. If I refer to Dumont’s work, it is because in the language he uses, he makes a very classical distinction, that of the opposition between Gemeinschaft (community) and Gesellsdiaft (society). Based on Tönnies’ famous distinction, we find with Dumont the holistic premise that societies are coherent wholes, that we can determine the fundamental principles of this coherence, and that there are perfectly discontinuous types of societies. These holistic visions have been widely questioned and tempered by sociological studies following an individualistic type of methodology. To illustrate this point, I will mention, from among the many studies of the effects of technical change in traditional societies, a study of the effects of irrigation in India. This study, which we owe to Epstein (1962), is enough to show that holistic theories of development, like theories that seek coherent and discontinuous systems in societies, are often, to put it in Feyerabend’s words, “fairy tales.” In this study, we find neither the timeless and changeless Indian village dear to Margaret Mead nor Lerner’s automatic chain reaction processes. And we have the impression that, while Indian society is dominated by the homo hierarchicus principle, we can identify processes of considerable social importance that are not unlike those that sociologists have brought to light in homo-aequalis societies. In particular, Mendras’ (1967) work on the modernization of French agriculture is to be considered here. Irrigation allowed rural farmers in villages in southern India to replace traditional subsistence crops such as ragi with sugar cane. These villages switched from a subsistence economy to a trading economy. But the important point is that irrigation also caused opposing social effects, some moving towards reinforcement, and others towards disruption of the traditional structures. Irrigation therefore had a corrosive effect on the paternalistic structure of the family through the following indirect mechanism: as the sugar cane could not be processed by the rural farmer, it was sold to State-owned plants.

Individualism and Holism in the Social Sciences

109

Of course, the administration, by its role, is subject to two constraints in this regard: It must avoid the overproduction of sugar cane and treat farmers on an egalitarian basis. This is why quotas were set, with the State committing to buy a certain amount of sugar cane from each household. But this encouraged farmers to behave strategically, because they understood that, in order to get round the limits imposed on them by the quotas, they simply had to transfer plots to their descendants as quickly as possible. Consequently, whereas before irrigation the actual transfer preceded the legal transfer, after irrigation the legal transfer tended to precede the actual transfer. Of course, the result of this behavior was an effect of self-empowerment of a son in relation to his father, since the former was now, economically and legally, independent of the latter earlier. Similarly, irrigation led to an empowerment effect where husband-and-wife relationships were concerned. Indeed, the appearance of a surplus from sugar cane farming meant that the farmer’s wife could create a reserved area inside the farm: She was responsible for investing part of the surplus, in poultry farming, for example, and marketing its products. Whereas before irrigation, she was simply a part of the division of labor system, after irrigation she became an autonomous economic actor. These first two effects tend towards the corrosion of traditional structures. But other effects go in the opposite direction. Therefore, clientelism between rural farmers and Untouchables was indirectly reinforced by the effect of irrigation. Prior to irrigation, underemployment had reached a very high level. As a result, the Untouchables only worked for rural farmers for a small part of their available time. After irrigation, underemployment decreased. The Untouchables benefited from the boom caused by sugar cane farming, but at the same time, they became more closely tied to the farmers and more dependent on them. It should be noted that this effect left a different mark in the non-irrigated villages neighboring the irrigated ones. Here, farmers, who could not start growing sugar cane, benefited from the economic boom their neighbors were experiencing. Responding to the growing demand for services emanating from the irrigated villages, they became, for example, vendors or repairers of agricultural equipment. In general, they supplemented their traditional agricultural activities with service activities or artisanal activities. But these new activities tended to break the farmers’ traditional ties with their Untouchables, as the former began recruiting staff from outside their traditional customer networks. The study I have just presented very quickly is, again, only one example among many. I could mention many studies of the same type, where development processes are analyzed as the aggregate effect of social actors’ reaction to conditions that are changed, by the introduction of an innovation, for

110

R. Boudon

example. Taken together, these studies tend to reveal the oversimplification of development theories with a general ambition, which seeks to directly analyze the consequences of structural or cultural data by avoiding any analysis of individual motivations and behaviors. Finally, I will refer to a more recent chapter in classical sociology, that of social stratification and mobility, which I had the opportunity to work on directly (Boudon 1973). This is an interesting case because the holistic way of thinking has dominated it for a long time and continues to do so. In general, a sociologist who is interested in this field is content to study, for example, the effects of certain independent variables, such as the development of education, on social mobility, without seeking to analyze the individual behaviors responsible for the structure of mobility flows. And we know that it has long been seen as self-evident that the development of education must lead to an automatic increase in social mobility. And many holistic types of analyses conducted in the field of social mobility assume that, by the effect of “subtle” mechanisms, individuals are forced to adopt behaviors that allow social structures, some argue, to reproduce (Bourdieu & Passeron 1970) or, others argue, to move in a direction laid down by a sense of history. In this type of analysis, it is the structures of society that are supposed to be active elements, while individuals are described as behaving passively, with no freedom other than that of fulfilling a pre-determined destiny. After all, it is important to identify a particular form often taken by holistic methodology, which can be called minimal individualism or shameful individualism: This consists of describing the individual as the exclusive seat or crossing point of collective forces or ideas. The individuals’ expectations or goals would be determined entirely by their social environment. Here, the beliefs of others and, generally, the characteristics of the actors’ environment are not treated as data that the actors would take into account, but as almost mechanistic forces that would determine both their objectives and the means they use to achieve them. It is this type of methodology that is behind, for example, explanations that make economic stagnation the product of the actors’ traditionalism. In this case, the macroscopic phenomenon to be explained is clearly analyzed as the product of microscopic behaviors. But, because the actor is viewed as having no autonomy, the microscopic part of the analysis is pure fiction: The actor is described as obeying invisible mechanisms, the existence of which can only be demonstrated by the behaviors they are supposed to produce… and explain. Of course, this criticism does not indicate that a notion like traditionalism has no reality or utility. Only that it can easily be used tautologically, and

Individualism and Holism in the Social Sciences

111

often is. Evoking the traditionalism of the actor then has the same explanatory force as the so-called “dormitive virtue” of opium. Likewise, it is clear that it is often legitimate to bring in the notion of attitude, the Aristotelian notion of ξ´ις, or the Thomist notion of habitus to account for the behavior of social actors. But when the existence of habitus is used to explain behaviors that are the only evidence of the existence of habitus, then we are in a vicious circle. To return to the problem of social mobility, for my part, I tried to show that if the development of education did not appear to have the expected effects on mobility, this could be interpreted as an aggregation effect resulting from the composition of a multitude of individuals’ rational behaviors in relation first to the education market, and then to the labor market. It is no more necessary to use the tautological hypothesis of traditionalism, or resistance to change, to explain why individually and collectively favorable innovation can only be adopted after a long period of time, than it is to assume that individuals subconsciously obey a desire to reproduce social structures to explain that the structure of mobility flows varies less over time than might have been hoped, even when the education system is undergoing “democratization.” This relative stability can be explained as an aggregation effect resulting from individual strategies, which are not irrational but, on the contrary, understandable if they are related to the actors’ situation. So, it seems indisputable to me that the individualism paradigm, when appropriately applied, can lead to much more acceptable theories than the holistic paradigm. It remains to be explained why—if this diagnosis is justified—the holistic way of thinking remains dominant in the social sciences. I can see several reasons for this state of affairs. First, it must be recognized that the paradigm of methodological individualism is not always easily applicable. In a text that goes into great depth, Georg Simmel (1908, 1981) states: For perfect knowledge, it has to be admitted that there are only individuals. To get to the bottom of things, any phenomenon that seems to constitute a new and independent unit above individuals would be resolved in the reciprocal actions exchanged by individuals. Unfortunately, this perfect knowledge is forbidden to us.

It is important to understand the meaning of this text: Simmel first argues that any social phenomenon can only be viewed as an aggregation effect when it is the result of individual actions, attitudes, or behaviors. And, when he says that this can only be a chimera, he means that there is no way of claiming to analyze, in an objective and controllable way, the behavior of every single

112

R. Boudon

social actor. On the contrary, we must group these actors into types and propose an explanation for the behavior of these ideal–typical actors that will always be a very simplified representation. This representation will always bring into play, as he says elsewhere (Simmel 1908/1981, p. 174) an “abstract psychology” or a “conventional psychology.” In short, individualistic explanations are always made via the construction of models. Just as the economist gives himself the right to describe the behavior of a consumer or a producer under certain circumstances, so the sociologist will choose to create ideal– typical actors, and will give them infinitely simplified “logics” of behavior with regard to the reality. But Simmel immediately adds: That said, just as the biologist has already been able to substitute for the vital force, which appeared to hover over the various organs, the reciprocal action of the latter, the sociologist, in turn, must increasingly seek to reach the particular processes that actually produce social things, while having to remain a certain distance from the ideal.

This passage means that the construction of individualistic models is the goal that knowledge must move towards and that advancing knowledge in the social sciences involves shifting from holistic explanations to individualistic explanations. But, at the same time, the text (and the context in which it appears) suggests that progress is often difficult and that, in any case, it is unequally accessible depending on the problems to be studied (Simmel 1892– 1905/1984). This difficulty in applying the individualistic paradigm is the main explanation for the influence of the competing paradigm. It is true that when considering economic behaviors, their logic can often be reconstructed quite easily. As a result, macroscopic economic phenomena such as those I mentioned above can be analyzed by individualistic models in a convincing way. This is also true for stratification or mobility phenomena, for example. These can be described without too much difficulty as aggregation effects resulting from behaviors that follow relatively simple logics (Boudon 1973). But it is clear that such an analysis is much more difficult with regard to suicide rates, for example. These rates result from behaviors that follow logics that are complex and infinitely variable from one individual to another. That is why, in this case, it is much more difficult to analyze the macroscopic data by bringing them back to their microscopic components. Of course, the example of suicide is not unique. Opinion data also frequently show resistance to individualistic analysis. It is for this reason that one often has to be content, for lack of a better option, with observing that the frequency of a certain opinion varies with a certain socio-demographic characteristic without being able to explain why, that is, with making this

Individualism and Holism in the Social Sciences

113

correlation the result of a set of behaviors that follow a clear logic. Likewise, it is a well-known fact that demographers cannot always bring observed aggregate outcomes back to their individual components: Why was World War I followed in France by a decrease in the birth rate and World War II by an increase? These banal remarks are important because of their consequences, which are often not well perceived. In particular, they explain, at least in part, that individualistic methodology is more commonly used in economics than in sociology. But, on the other hand, we must remember another crucial point developed by Simmel, and which I believe is absolutely right, which is that the progress of knowledge is usually reflected in a shift from the holistic paradigm to the individualistic paradigm. And many examples of such ruptures could be cited; I mentioned a few earlier. These considerations, however, do not sufficiently explain the low popularity of the individualistic paradigm in sociology. Resistance to it also comes from what is often observed in the social sciences, and what Wrong called an “over-socialized conception of Man” (Wrong 1961): the social actors are often viewed as pieces of pliable dough on which data from their environment can be inscribed, to later dictate their behavior in a certain situation. It is for this reason that developmental sociologists often admit that social actors strictly obey the rules laid down for them by tradition and culture. Some have used this to explain, for example, why Indian farmers continue to have large families, when it is against their own interests and those of the community. This is said to be the result of automatic adherence to secular traditions. This “explanation” is not only empty but also overestimates the social actor’s passivity with regard to tradition. In a case like this, the individualistic perspective consists of asking oneself about the actors’ situation and trying to show that if tradition is followed, it is not just because it engenders unconditional and mechanical respect, but mainly because it does not contradict the facts of the situation—in short, it is “rational” for the actors to follow it, and it makes sense to them. Indeed, it has been shown that, in India, the economic context is often such that families benefit from having a large number of children, for reasons of economic and social security. This “over-socialized” conception of the human being—and, more specifically, of the social actor—is very common and is found in the most diverse chapters of sociology. Firstly, it stems, in my view, from the difficulty of escaping what Piaget calls “sociocentrism”: it is easier to interpret behavior you do not understand as unconditionally irrational than to demonstrate that it is irrational only in relation to the observer’s situation, and rational

114

R. Boudon

in relation to the actor’s situation. One of the advantages of the individualistic paradigm and the rationality assumption it includes (i.e., the Weberian assumption in Weber 1922/1968) that any behavior is, in principle, understandable by the observer once the latter is sufficiently informed of the context in which the actor is moving) is that it allows the sociologist to avoid the dreaded trap of sociocentrism. But the “over-socialized conception of Man” also has frankly ideological roots: It derives from a widespread and easily felt sense of the despotism of social structures. But, while this feeling is legitimate and well-founded, it does not allow the ideological derivation that views the social actor as a puppet whose strings are pulled by those structures. To complete the analysis, one would also have to add that the feeling I am talking about is not enough to produce a holistic ideology: This appears above all when the observer also feels that structures are reprehensible and that those who submit to them should, therefore, be held to be irrational or alienated. These two reasons, the despotism of structures and the temptation of sociocentrism, explain why the holistic perspective is so widespread: When the actors are perceived as the plaything of the structures, they can easily be forgotten. In this case, the individualistic program loses all its meaning, and is easily perceived as irrelevant. Societies can therefore be viewed as simple systems that can be described through the use of a number of conceptual oppositions, for example: Gemeinschaft/Gesellschaft, modern society/ traditional society, industrial society/post-industrial society. As for social processes, they can be reduced to basic patterns: class struggle, domination, and dependence.

References Birnbaum, P. (1984). Dimensions du pouvoir. Paris: PUF. Boudon, R. (1973). L’inégalité des chances. La mobilité sociale dans les sociétés industrielles. Paris: Colin. Transl. (1974). Education, Opportunity, and Social Inequality: Changing Prospects in Western Society. New York: Wiley. Boudon, R. (1979). La logique du social . Paris: Hachette. Transl. (1981). The Logic of Social Action. An Introduction to Sociological Analysis. Boston: Routledge & Kegan. Bourdieu, P., Passeron, J.-C. (1970). La Reproduction. Paris : Editions de Minuit. Transl. (1978). Reproduction in Education, Society and Culture. London & Beverly Hills: Sage.

Individualism and Holism in the Social Sciences

115

Dumont, L. (1966). Homo hierarchicus. Essai sur le système des castes. Paris: Gallimard. Dumont, L. (1976). Homo equalis: Genèse et épanouissement de l’idéologie économique. Paris: Gallimard. Dumont, L. (1983). Essais sur l’individualisme. Une perspective anthropologique sur l’idéologie moderne. Paris: Le Seuil. Transl. (1992). Essays on Individualism: Modern Ideology in Anthropological Perspective. Chicago: University of Chicago Press. Elridge, J.E.T. (1970). Max Weber: The interpretation of social reality. London: Nelson. Epstein, T.S. (1962). Economic Development and Social Change in South India. Manchester: Manchester University Press. Hoselitz, B.F. (1952). The Progress of Underdeveloped Areas. Chicago: University of Chicago Press. Le Bon, G. (1895/1963). Psychologie des foules. Paris: PUF. Transl. (1896). The Crowd: A Study of the Popular Mind . New York: Macmillan. Lerner, D. (1958). The Passing of Traditional Society: Modernizing the Middle East. Glencoe: The Free Press. Mead, M. (1953). Cultural Patterns and Technological Change. Paris: UNESCO. Mendras, H. (1967). La fin des paysans. Paris: SEDEIS. Mommsen, W. (1965). Max Weber’s Political Sociology and His Philosophy of World History. International Social Science Journal , 17(1), 23–45. Moscovici, S. (1981). L’âge des foules. Paris: Fayard. Oberschall, A. (1973). Social Conflicts and Social Movements. Englewood Cliffs: Prentice-Hall. Simmel, G. (1892–1905/1984) Les problèmes de la philosophie de l’histoire. Paris: PUF. Transl. (1977). The Problems of the Philosophy of History: An Epistemological Essay. New York: The Free Press. Simmel, G. (1908/1981). Sociologie et épistémologie. Paris: PUF. Transl. (2009). Sociology: Inquiries into the Construction of Social Forms. Boston: Brill. Touraine, A. (1973). Production de la société. Paris: Le Seuil. Transl. (1977). The Self-Production of Society. Chicago: The University of Chicago Press. Touraine, A. (1984). Le retour de l’acteur. Paris: Fayard. Transl. (1988). Return of the Actor. Minneapolis: University of Minnesota Press. Weber, M. (1922/1968) Economy and Society: An Outline of Interpretive Sociology. Berkeley: University of California Press. Wrong, D. (1961). The Oversocialized Conception of Man in Modem Sociology. American Sociological Review, 26 (2), 183–193.

What’s the Alternative? Jon Elster

1

Introduction

I believe that methodological individualism (MI) is trivially true. Yet eminent scholars, including Kenneth Arrow, have put it in question. I believe most of their doubts and criticisms can be traced back to the following fact. In modern societies, and probably in all societies, individuals meet institutions, belief systems, and value systems that they have to take as given. These are without exception effects or residuals of the actions and interactions of other individuals, but their history is not written on their faces. Denials of MI are largely due, in my opinion, to neglect of this fact. Although I shall criticize Marx for his violations of MI, I subscribe to his view that in modern societies “the products of the human brain appear as autonomous figures endowed with a life of their own” (Marx 1867, p. 165).

Unless otherwise noted, all italics in quoted passages are mine.

J. Elster (B) Departments of Philosophy and Political Science, Columbia University, New York, NY, USA e-mail: [email protected] Collège de France, Paris, France

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 N. Bulle and F. Di Iorio (eds.), The Palgrave Handbook of Methodological Individualism, https://doi.org/10.1007/978-3-031-41512-8_7

117

118

J. Elster

Surprisingly, this view was also stated by Émile Durkheim, who is usually seen as a main opponent of the idea that sociology could, at least in principle, be reduced to psychology (1919, pp. 124–5). He argued that what he called “social facts” are “things that have their own existence. The individual encounters them when they are already completely fashioned and he cannot cause them to cease to be or to be different from what they are. Willy-nilly he is therefore obliged to take them into account; it is all the more difficult (although we do not say it is impossible) for him to modify them because in varying degrees they partake of the material and moral supremacy that society exerts over its members. No doubt the individual plays a part in their creation. But in order for a social fact to exist, several individuals at the very least must have interacted together and the resulting combination must have given rise to some new production. As this synthesis occurs outside each one of us (since a plurality of consciousnesses are involved) it has necessarily the effect of crystallizing, of instituting outside ourselves, certain modes of action and certain ways of judging which are independent of the particular individual will considered separately” (ibid ., p. XXII; my italics). I now proceed as follows. In Sect. 2, I propose a definition of MI and distinguish it from other doctrines to which it bears a more or less superficial resemblance. In Sect. 3, I cite social scientists whose work violates MI. In Sect. 4, I discuss some examples that have been proposed to show the inadequacy of MI. In Sect. 5, I consider the question raised in the title of the chapter. Section 6 has a brief conclusion.

2

Definitions and Misunderstandings

“By [MI] I mean the doctrine that all social phenomena—their structure and their change—are in principle explicable in ways that only involve individuals—their properties, their goals, their beliefs, and their actions” (Elster 1985, p. 5). Today, I would characterize MI as an injunction rather than as a doctrine. MI is a form of reductionism, which is the engine of progress in science. It enjoins us to seek for microfoundations. In many cases, a literal attempt to carry out the individualist program will be doomed to failure. The number of agents who contribute to an economic or political decision may be prohibitively large, and the relevant information may be unavailable or unreliable. In such cases, many scholars simply assume that a firm, a political party, or even a nation-state can be considered a single unitary agent. Often, however, that assumption may be so unrealistic as to make the explanatory enterprise a non-starter. Some “big” problems are

What’s the Alternative?

119

too big. Yet whenever it seems at least partly feasible, one should dig in. Ivan Ermakoff (2008, 2015) provides exemplary micro-historical analyses of decisions by large assemblies that were only loosely organized by shifting alliances and factions. At the Federal Convention of 1787, the delegations voted as states, each state delegation aggregating the votes of its individual delegates. The recorded fact that a state voted in favor of a proposal does not allow us to determine whether its vote was close or not. Yet by painstaking triangulations, historians have been able to open these black boxes and make that determination. Economists tend to respect MI, at least in models of general equilibrium. In their econometric applications, they rarely do so, since the units tend to be households or firms rather than individuals. In an article on “intra-household allocations”, Michael Browning and Pierre Chiappori try to extend MI to individual consumption. As they write, “the neoclassical theory of demand applies to individuals, yet in empirical work it is usually taken as valid for households with many members. This paper explores what the theory of individuals implies for households that have more than one member. We make minimal assumptions about how the individual members of the household resolve conflicts. All we assume is that however decisions are made, outcomes are efficient” (Browning and Chiappori 1998, p. 1241). Efficiency (Pareto-optimality) here means that when household members bargain over the allocation of household income, the bargaining never breaks down. In the real world, the situation may not always be so idyllic. To respect MI, efficiency would have to be demonstrated, not stipulated. This example shows that MI can be extremely demanding. Browning and Chiappori dug in, but not all the way down. As I defined it, methodological individualism does not make any substantive psychological assumptions about human behavior. It does not assume that individuals are selfish, or that they are rational. Nor does it treat them as “atoms”, if we understand that metaphor to mean that individuals behave like particles in Brownian motion, bumping into each other randomly. In the belief-desire framework that is standardly used to understand individual behavior, both beliefs and desires may be oriented towards other people, for instance, through empathy and sympathy. The preceding paragraph addressed some tedious and trivial issues, mentioned only because they are occasionally raised by critics of MI. Two more interesting misunderstandings arise from the confusion of MI with political individualism and ethical individualism. The former, which we may also refer as “liberalism”, amounts, roughly speaking, to a claim, or a set of claims, that state interference in the economy and in civil society should

120

J. Elster

be minimized. There is no logical connection between MI and this normative claim, but there may well be a loose sociological connection. In recent French sociology, for instance, Raymond Boudon stood for both methodological and political individualism whereas Pierre Bourdieu was an opponent of both. The present writer and no doubt many others embrace methodological but not political individualism, whereas it seems harder to find the converse combination. These issues belong to the sociology of science, not to the philosophy of science, which is my subject here. Ethical individualism is a less well-known idea. It can be defined, or approximated, as the principle that when assessing the welfare of a society, one should count only the properties of individuals, not properties of supraindividual or non-individual entities (Elster 1993). Marx, I believe, endorsed ethical individualism (Elster 1985, p. 8). He always took it for granted that the goal of communism was the self-realization of human beings, not of humanity (as embodied, for instance, in Gothic cathedrals). Louis XIV, Napoleon, and de Gaulle denied ethical individualism. For them, the greatness of France dominated the welfare of the French.

3

Some Violations of Methodological Individualism

I shall now consider violations of MI. In some cases, these overlap with another violation of the scientific method, namely functional explanations. Generally speaking, these explain actions by their actual (not merely intended) beneficial consequences for someone (e.g., the capitalist class) or something (e.g., social cohesion). Usually, these explanations are metaphysically impossible, as they explain the present by the future rather than by the past. There are exceptions, however. B. F. Skinner (1981) argued for the importance of three ways in which behavior can be explained by its consequences: by natural selection operating on individuals, by reinforcement, and (although Skinner does not use that term) by group selection. I only state without argument that none of these mechanisms fit the examples I shall discuss. I shall first discuss some historical examples and then some contemporary ones. Marx. An example that embodies both violations cited above can be taken from Marx:

What’s the Alternative?

121

The circumstance that a man without fortune, but possessing energy, solidity and business acumen may become a capitalist […] is greatly admired by apologists of the capitalist system. Although this circumstance continually brings an unwelcome number of new soldiers of fortune into the field and into competition with the already existing individual capitalists, it also reinforces the supremacy of capital itself, expands its base and enables it to recruit ever new forces for itself out of the substratum of society. In a similar way, the circumstance that the Catholic Church in the Middle Ages formed its hierarchy out of the best brains in the land, regardless of their estate, birth and fortune, was one of the principal means [Mittel] of consolidating ecclesiastical rule and suppressing the laity. (Marx 1894, pp. 735–36)

The two explananda—upwards mobility into the capitalist class and into the church—have plausible explanations: the absence of a caste system and clerical celibacy. In both cases, the infusion of fresh blood may, arguably, have strengthened the grip of the elites on the people. The agency belongs to the individual upstarts, not to “capital” or the Church. Marx, however, imputed agency to these supraindividual entities. The use of the word “Mittel’ is especially revealing. A means presupposes an end to which it is a means, and an end presupposes an agent for whom it is an end. Thus capitalists and church leaders may benefit from the upward mobility, but if so, only as an accidental byproduct. As in many other passages, notably in the Grundrisse, Marx here treats “capital” as an agent. His most explicit denunciation of MI is the following: The predominance of capital is the presupposition of free competition, just as the despotism of the Roman Caesars was the presupposition of the free Roman “private law”. As long as capital is weak, it still itself relies on the crutches of past modes of production, or of those which will pass with its rise. As soon as it feels strong, it throws away the crutches, and moves in accordance with its own laws. As soon as it begins to sense itself and become conscious of itself as a barrier to development, it seeks refuge in forms which, by restricting free competition, seem to make the rule of capital more perfect, but are at the same time the heralds of its dissolution and of the dissolution of the mode of production resting on it. Competition merely expresses as real, posits as an external necessity, that which lies within the nature of capital; competition is nothing more than the way in which the many capitals force the inherent determinants of capital upon one another and upon themselves. (Marx 1857, p. 651; his italics)

Marx’s economic theories, notably the labor theory of value, rely on these ideas. Consider a firm that produces shirts. The desire for profit will induce

122

J. Elster

the capitalist to innovate, in order to undersell his rivals. In doing so, he will also reduce the value of labor-power, to the extent that shirts form part of the basket of necessary consumption goods, the labor-value of which determines the labor-value of labor-power in all industries, not only in his own. When the value of labor-power is reduced, profit is increased. This collateral benefit—a kind of gift from the shirt producer to all other producers—is, according to Marx, the real aim of the process of capital accumulation. He writes that “the shortening of the working day […], is by no means what is aimed at [bezweckt ] in capitalist production, when labour is economized by increasing its productivity. It is only the shortening of the labour-time necessary for the production of a definite quantity of commodities that is aimed at” (Marx 1867, pp. 437–8). Note, however, that the shirt producer will not “aim” at shortening the necessary labor time, except in the marginal case that his workers consume a large proportion of the shirts. (Henry Ford is reputed, implausibly, to have said that he made his cars cheap so that his workers could afford to buy them.) The benefit to the capitalists of the fall in the value of labor-power occurs only as a collateral byproduct when all producers of consumption goods innovate to cut prices. Marx’s argument is an instance of the fallacy of cui-bono combined with post hoc ergo propter hoc. Marx’s theory of history was teleological, explaining the present by the future rather than by the past. As he stated in the Introduction to the Grundrisse, “human anatomy contains the key to the anatomy of the ape” (Marx 1857, p. 105). He consistently viewed the vicissitudes of nineteenth-century European politics as stepping-stones to the Communist revolution; a defeat of the workers was always seen as reculer pour mieux sauter. The only exception is found in the Preface to the 1859 Critique of Political Economy: In the social production of their existence, men inevitably enter into definite relations, which are independent of their will, namely relations of production appropriate to a given stage in the development of their material forces of production. The totality of these relations of production constitutes the economic structure of society, the real foundation, on which arises a legal and political superstructure and to which correspond definite forms of social consciousness. The mode of production of material life conditions the general process of social, political and intellectual life. It is not the consciousness of men that determines their existence, but their social existence that determines their consciousness. At a certain stage of development, the material productive forces of society come into conflict with the existing relations of production or—this merely expresses the same thing in legal terms—with the property relations within the framework of which they have operated hitherto.

What’s the Alternative?

123

From forms of development of the productive forces these relations turn into their fetters. Then begins an era of social revolution. The changes in the economic foundation lead sooner or later to the transformation of the whole immense superstructure.

Remarkably, coming from Marx, there is no reference, as there is in the Communist Manifesto, to the role of the class struggle in social revolutions. In fact, there is no mention of any kind of agency. The textually most plausible reconstruction of what according to Marx motivates revolutions is that in a given set of property relations, there comes a point when the rate of technical progress becomes inferior to what it would be under another set of property relations. At this point, the class that would gain from a revolution brings about the new property relations (see Cohen 1978; Elster 1985, Ch. 5. 1. 3). Substantively, however, this torch-relay theory of history is far from plausible. Although G. A. Cohen’s reconstruction of Marx’s theory brings out its logical structure admirably well, by showing that it depends on a certain kind of functional explanation, it does not provide microfoundations for the theory. Religion. Methodological individualism is also denied, at least implicitly, when social phenomena are explained by the fact that they satisfy a need . Consider two statements: “Individuals need religion” and “Societies need religion”. With respect to democratic societies, Tocqueville endorsed both statements, but only the first need as explanatory. On the one hand, he wrote that “for me, I doubt that man can ever bear complete religious independence and full political liberty at the same time; and I am led to think, that, if he does not have faith, he must serve, and, if he is free, he must believe” (Tocqueville 2010, volume 3, p. 745). This first claim is explanatory: individuals in democracies adopt religion to alleviate the psychological burden of freedom. On the other hand, he stated that “at the same time that the law allows the American people to do everything, religion prevents them from conceiving of everything and forbids them to dare everything” (ibid ., volume 2, p. 475). This second claim merely states a happy coincidence: because the first claim is true, individuals are not tempted to exploit the opportunities for licentious behavior that democracies create. Tocqueville never says that religion exists because society needs it, that is, because it has good consequences for society. In The Elementary Forms of Social Life, Durkheim (1915, p. 17) wrote that. Society could not abandon the categories to the free choice of the individual without abandoning itself. If it is to live, there is not merely need of a sufficient moral conformity, but also there is a minimum of logical conformity that it

124

J. Elster

cannot do without. For this reason it uses all its authority upon its members to forestall such dissidences. Does a mind ostensibly free itself from these forms of thought? It is no longer considered a human mind in the full sense of the word, and is treated accordingly.

Elsewhere he comments more specifically on religion: Howsoever complex the outward manifestations of the religious life may be, at bottom it is one and simple. It responds everywhere to one and the same need , and is everywhere derived from one and the same mental state. In all its forms, its object is to raise man above himself and to make him lead a life superior to that which he would lead, if he followed only his own individual whims: beliefs express this life in representations; rites organize it and regulate its working. (ibid ., p. 414)

These and many other needs-based explanations ignore the fact that social needs do not generate their own satisfaction. When a need is satisfied, it may be because its satisfaction eases the mind, as in Tocqueville’s first claim, or, more simply, that the members of the society become aware of their need and act consciously and in concert to satisfy it. Kenneth Arrow. As noted, and as further discussed below, Arrow expressed doubts about MI. Consistently with these doubts he suggested, “as one possible interpretation that [norms of social behavior] are reactions of society to compensate for market failures. […] Society may proceed by internalization of these norms to the achievement of the desired agreement on an unconscious level . There is a whole set of customs and norms which might be similarly interpreted as agreements to improve the efficiency of the economic system (in the broad sense of satisfaction of individual values) by providing commodities to which the price system is inapplicable” (Arrow 1971, p. 20). As stated, these ideas are blatant violations of MI. “Society” does not react or act; only people do. People do not agree to establish social norms. In any case, Arrows analysis cannot serve as a complete theory of social norms, as many of them are harmful rather than beneficial (Elster 2015, pp. 357–8). In a restatement of his views, Arrow (2001, p. 1200) seems to refute the view that “social norms arise to meet social needs”. He now associates this view with the work of Talcott Parsons, which he sees as “mostly quite empty of serious content”. At the same time, he found himself “compelled by [his] observations to argue that social norms play a major role in understanding the departures of the medical care market from the standards of markets in general”. This is a weaker and less controversial view than his earlier one. Yet he still maintains the view that “social norms are based on at least perceived

What’s the Alternative?

125

mutual gains” (ibid ., p. 1201), contrary to the many cases of norms that are perceived as harmful and yet persist over time. Political science. This discipline deals partly with the behavior of individuals, notably voters, and partly with the decisions by higher-level actors (institutions), such as the decision by a political party to adopt a certain platform, or the decision by a state to declare war. As everybody knows, in virtually all cases the latter decisions are the result of struggles within the institution. Genuine unanimity is extremely rare, although institutions often present their decisions as unanimous once they are arrived at. Despite these obvious facts, political scientists routinely present institutions, including states, as unitary actors, who act on the basis of coherent preferences and rational beliefs—as rational individuals write large. Scholars rarely, or virtually never, take the trouble of defending the unitary-action assumption, but one can imagine two procedures. First, one could identify the preferences and beliefs of the individuals, and, using an appropriate aggregation mechanism, construct the preferences and beliefs of the collective actor. In assembly politics, aggregation by majority voting seems the most natural procedure. Second, one could impute preferences to the collective actor that coincide with its objective interests and beliefs that correspond to the facts. Regarding the first procedure, political scientists surely all know that preference-aggregation is vulnerable to the Condorcet paradox , which can make the outcome by majority voting essentially arbitrary (Elster 2015, pp. 410– 12). Some of them may also know that belief-aggregation is vulnerable to what is variously called the discursive dilemma or the doctrinal paradox, but which should really be called the Poisson paradox after its first expositor (ibid .). This paradox, too, can leave belief formation by majority voting in a multi-member body essentially indeterminate. Put crudely, a group may not know what it wants or what it believes. The second procedure, which completely disregards subjective elements and the possibility or relevance of disagreement among group members, characterizes the Realist school of international relations. In this approach, which is properly called “methodological statism” (Megiddo 2019), states “are assumed to be unitary actors, their internal dynamics of little importance for studying their behavior internationally” (ibid ., pp. 225–26). I am not a scholar of international politics, but this assumption strikes me as highly implausible. It is violated, for instance, when political leaders to go war to distract from domestic unrest. It is also violated when different branches of government have different interests and different beliefs concerning the conduct of a war, as illustrated by the conflict between the Pentagon and the State Department during the Vietnam War.

126

J. Elster

A special case that undermines the assumption of the unitary actor is the struggle for resources among the branches of the military.

4

Some Alleged Obstacles to MI

As noted at the outset of this chapter, the practical obstacles to MI may be unsurmountable. Because I view MI as an injunction to dig deeper rather than as a research program, these obstacles merely reflect the fact that “ought implies can”. A more forceful objection to MI would arise if one could show that there exist social phenomena that in principle cannot be addressed within the framework of MI, even when all relevant individual-level information is available. I do not think there are any such cases. As it is hard to prove a negative, I can only address the most prominent obstacles and objections. Social norms. A very common objection is that MI cannot account for social norms. As noted earlier, it is certainly true that individuals mostly grow up in an environment where they face strict social norms that are enforced by the avoidance or punishment of norm-violators. It is also true (i) that social norms are shaped by the interaction of individuals and (ii) that even when violations are punished, individuals often violate norms when unseen by others (Elster 2017). Unlike moral norms, social norms are not internalized. (These are part of my stipulative definitions.) It is also true, finally, that no one has proposed a clear and convincing account of how the interaction of individuals causes new norms to appear. If, as Arrow (1971) suggested, social norms arise because of a need to remedy a market failure, this need must be embodied in the perceptions and actions of individuals, including their propensity to avoid or punish norm-violators. But why would anyone punish norm-violators before the norm exists? A possible answer could be found in the simultaneous provision of new information to many individuals about previously unknown harm caused by activities that now become the target of negative norms. Spitting on the floor before the understanding of bacterial dissemination of illness could be an example. The invention of social norms has also been imputed to “norm entrepreneurs”, although it is not always clear exactly what the diffusion mechanism(s) would be. For the time being, feature (i) of norms is stipulated rather than demonstrated. What is certain, however, is that Arrow’s 1971 proposal doesn’t work. “Society” does not create norms. Feature (ii) of norms also counts against Durkheim’s claim that immaterial norms have a compelling and irresistible character, comparable to the circle in the sand that prevents the chicken from going outside it. However,

What’s the Alternative?

127

norms can disappear almost overnight if the victims of harmful norms organize themselves to discontinue the compliance (Mackie 1996; Elster 2023, Ch. 6). Even if the norm persists in the population, some individuals may secretly persist in the proscribed behavior. When they are exposed, consequences can be terrible, as shown by the six Frenchmen who killed themselves in 1997 when they were exposed as consumers of pedophiliac materials. In what sense are social norms social? Not in the sense that they are independent of individuals. They are shared individual expectations about how other people would react to certain actions. The expectations are not necessarily correct. As Tocqueville (2010, volume 4, p. 1149) wrote, the belief in the existence of a norm maybe just “the empty phantom of a public opinion [which] is enough to chill innovators” until someone shows its non-existence by violating it without suffering avoidance or punishment. For more than a century, American Presidents acted on the belief that there existed a political norm requiring them to address Congress in writing rather than orally. When Woodrow Wilson showed up in person in 1920, nobody objected. Until that time, the expectations had causal efficacy. Institutions. Another common objection is that MI cannot account for institutions. These are of course created by individuals and staffed by individuals. They operate by rules such as the principle that an official cannot give orders to other officials more than one step below him or her in the hierarchy or the principle that institutional decisions are subject to appeal. These rules exist on paper and in the minds of the members and the clients of the institution. If the latter do not know (and cannot learn) the rules, as when sheriffs in eighteenth century North Carolina sometimes “prevented publication of the tax rate and collected more than the legal amount” (Douglass 1983, p. 77), we are dealing with a practice rather than with an institution. Institutions exist as networks of mutual expectations among officials and between officials and clients. To individuals subject to their decisions, they can appear as supraindividual, monolithic entities, but to those who operate them, they appear as what they are. To explain their decisions and their changes over time, the interaction of officials with each other and with their clients provides the necessity and sufficient data. A term such as “institutional sclerosis” is merely an umbrella shorthand for a variety of individual-level processes. Arrow (once more). Kenneth Arrow (1994) states a number of objections to MI. The paper is brief, programmatic, enigmatic, and frustrating since it leaves the impression that he could have said much more. He writes notably that “limitations on individualistic methodology appear very strongly when considering the role of information” (p. 2), adding that “the individualistic viewpoint is not to be completely neglected”. He returns to this issue towards

128

J. Elster

end of the paper, where he states that it may be easier to think of information breeding information and to suppress the role of individuals. This is very much like the role of genetic information. As the late nineteenth century writer, Samuel Butler, said: “A chicken is an egg’s way of making an egg" (p. 7) Unfortunately, Arrow does not spell out the argument. Arrow also writes that “the failure to give an individualistic explanation of price formation has proved to be surprisingly hard to cure, though there are some ingenious stories” (p. 4). A powerful example concerns the prices under duopoly: What [..] are the strategies of the two players? Cournot assumed that firms choose quantities. Is it not equally reasonable that firms choose prices? If we assume rational consumers, they will all buy from the sellers with the lowest price. The outcome of this game is entirely different from Cournot’s; under simple assumptions it is in fact the competitive equilibrium even if there are only two firms. What this example shows is that the rules of the game are social. The theory of games gets its name and much of its force from an analogy with social games. But these have definite rules which are constructed, indeed, by a partly social process. Who sets rules for real-life games? More generally, individual behavior is always mediated by social relations. These are as much a part of the description of reality as is individual behavior. (p. 5)

I have little competence to assess this issue. However, a perusal of the literature confirmed my intuition that the form of competition—by prices or quantities?—may be endogenous. “In Bertrand [price] competition (with homogeneous goods) undercutting firms acquire the whole market. Firms can only serve this market if unconstrained by capacity. By contrast, if capacity is constrained, firms produce as much as they can leaving the price to be determined in the marketplace; competition of the Cournot [quantity] type” (Haskel and Martin 1994, p. 23). This simple and attractive idea is perfectly consistent with MI. Sally Haslanger discusses a series of alleged failures of MI. A representative example is the following: Why did the Native Americans in what is now the northeastern United States rely on a diet of root vegetables in the winter months, rather than leafy vegetables and fruits? Answer: Because the climate and storage techniques available wouldn’t support the preservation of leafy vegetables and fruits through the winter. Note that the explanation of the Native American diet is in terms of the climate, technology, and properties of leafy vegetables and fruits, not in

What’s the Alternative?

129

terms of individuals and their interactions. But surely the diet of Native Americans is a social phenomenon to be explained. So explanatory individualism is unwarranted. (Haslanger 2020, p. 3)

This passage shows the need for a clarification of explananda in science. Haslanger asks for the answer to a “Why-question” (Hansson 2006) rather than for an explanation. Since all explanation is causal and since an absence, such as the absence of a diet, cannot be an explanandum (Elster 2015, p. 20), the objection is misplaced.

5

What’s the Alternative?

Critics of MI feel or should feel an obligation to present an alternative. Durkheim’s proposal amounts merely to a stipulation, namely that there exist irreducibly social facts. Marx’s statement that “Capital” is prior to the many individual capitals is not a causal claim, but a confused conceptual one. Arrow’s ideas are not sufficiently spelled out. Here I shall discuss four attempts to provide an alternative to MI: Pierre Bourdieu’s notion of a “field” (champ), Robert Putnam’s concept of “social capital”, various attempts to formulate an idea of “group agency”, and Christian List’s and Kai Spiekermanns proposal to “reconcile” individualism and its apparent opposite, holism. In the vast literature on the subject, there are many other proposals, often couched in impermeable terminology that makes assessment difficult. Although this characterization is also to varying degrees appropriate for the three first alternatives mentioned above, their prominence in the literature justifies the discussion. Bourdieu. The notion of a field is taken from physics, but has been applied by many authors to account for social phenomena (Martin 2003). Here, I shall consider only Pierre Bourdieu. The analogy with physics is only partial. “The magnet may cause the field, but it is the field that has the effects on the iron filings” (ibid ., p. 49). In the social applications, there is no magnet. The field of forces is somehow self-organizing. In an early statement (Bourdieu 1969, p. 89), nevertheless appealed to the analogy: “The intellectual field , which cannot be reduced to a simple aggregate of isolated agents or to the sum of elements merely juxtaposed, is, like a magnetic field, made up of a system of power lines” (his italics). Twenty years later, he had changed his mind: We need to escape the mechanist vision which would reduce the agents to simple particles swept up in a magnetic field, by reintroducing not rational subjects

130

J. Elster

working to fulfil their preferences as far as circumstances permit, but socialized agents who, although biologically individuated, are endowed with transindividual dispositions, and therefore tend to generate practices which are objectively orchestrated and more or less adapted to objective requirements, that is irreducible either to the structural forces of the field or to individual dispositions. (Bourdieu 1988, p. 149)

Presumably, the last sentence means that “practices” (actions?) must be understood (explained?) in terms of both “structural forces” and “individual (or transindividual?) dispositions”. In addition, individuals possess, acquire, or lose various forms of non-economic assets—symbolic, cultural, and, more generally, social capital. These are the objects of strategic competition, where “strategy” is to be understood in an objective sense that does not require strategic intentions (ibid ., p. 62). Let me give an example. Bourdieu’s magnum opus, La Distinction, deals at some length with intellectuals and their “strategies”. Linguistic ease may be manifested either in the tours de force of going beyond what is required by strictly grammatical or pragmatic rules, making optional liaisons, for example, or using rare words and tropes in place of common words and phrases, or in the freedom from the demands of language or situation that is asserted in the liberties taken by those who are known to know better. These opposing strategies, which place one above the rules and properties imposed on ordinary speakers, are in no way mutually exclusive. The two forms of conspicuous freedom, unconventional constraint and deliberate transgression, can coexist at different moments or different levels of the same discourse; lexical “relaxation” may, for example, be counterbalanced by increased tension in syntax or diction, or the reverse (this is clearly seen in condescension strategies, in which the gap thus maintained between the levels of language is the symbolic equivalent of the double game of asserting distance by appearing to negate it). Such strategies—which may be perfectly unconscious, and thereby even more effective—are the ultimate riposte to the hyper-correction strategies of pretentious outsiders, who are thrown into self-doubt about the rule and the right way to conform to it, paralyzed by a reflexiveness which is the opposite of ease, and left “without a leg to stand on”. (Bourdieu 1984, p. 255)

The idea of an unconscious strategy is as absurd as that of a square circle. It is true that writers sometimes experiment with and violate grammar, as when Sartre (1943, p. 53) writes that “le néant est été” (the nothing is been). It may be true upstarts are upset by the discovery that being an intellectual is not a matter of following the rules, but of knowing when to break them. There is no reason, however, to think that this fact, if it is one, can be explained as the

What’s the Alternative?

131

“riposte” of intellectuals to protect themselves from invasions from below. It is hard to tell whether this idea is a violation of MI, by imputing a collective strategy to intellectuals as a group, an instance of functional explanation, or both. These comments apply to most of the analyses in La Distinction (see also Elster 1981). Robert Putnam. As many commentators have noted, Bourdieu’s idea of social capital is very different from Putnam’s; only the (deeply misleading) word is the same. For Putnam, social capital is a force for the good, increasing trust and facilitating cooperation. For Bourdieu, it is a tool of competition and manipulation. For Putnam (2001, p. 45), clubs such as “the Lions club, the Moose club, and the Elks club” are among the best predictors of philanthropy. For Bourdieu (1984, p. 162), elite clubs “aim above all to protect the group against outsiders”. Whereas Putnam tries to measure the amount of social capital, for Bourdieu there is a constant struggle over what counts as social capital. Putnam (2001, p. 1) commits a major methodological error when he writes that “like physical capital, social capital is far from homogenous. Some forms of social capital are good for some things and not for others. Accepting that there is no single form of social capital, we need to think about its multiple dimensions”. However, there is no “it”. Because the error is extremely common, I shall discuss it at some length. All readers of this chapter will have participated in discussions about “what is democracy?”, “what is populism?”, “what is war”, etc., as if there were a fact of the matter that could be ascertained by conceptual analysis. If we ask “what does the moon look like on the side we cannot see from earth?”, we know that there is a fact of the matter, which has been ascertained by sending out a rocket to take pictures. The idea of sending out a conceptual rocket to discover the “real” nature of social capital, democracy, etc. is meaningless and reflects a kind of essentialism that ought to be banished from the social sciences. The proper response is nominalism. If there is a problem that interests us, the first thing to do is to define it as explicitly as possible and stick to that definition throughout the analysis. The only (pragmatic) constraint is that the stipulated meaning should not deviate too much from everyday language, since otherwise, the reader may get confused. Definitions cannot be true or false, but they may be fruitful or not. If, as often happens, we get nowhere, we can redefine the problem or abandon it. The problem can be analytical (explanatory) or normative. If analytical, we can ask about the causes and/or about the consequences of the object we have defined. If we study the consequences, we can also ask whether they are desirable from a normative point of view. In my view, MI and nominalism go hand in hand,

132

J. Elster

not because there is any logical connection between them, but because both serve the cause of clarity. As I see things, these statements are trivially true, but many scholars whom I respect think otherwise. To say that the issue is “essentially contested” is, again in my view, a cop-out. In Putnam’s work, the lack of definitional clarity shows itself in a longstanding controversy over trust: is it a form of social capital or a consequence of it? Sometimes it seems as if social-capital scholars want to have it both ways. Sobel (2002, p. 140 writes that “treating decreases in trusting behavior (as measured by survey responses) as direct evidence for decreases in social capital comes close to equating social capital with good outcomes”. Similarly, Woolcock (2001, p. 13) writes that “it is important that any definition of social capital focus on its sources rather than consequences, i.e., on what social capital is rather than what it does. This approach eliminates an entity such as ‘trust’ from the definition of social capital. Trust is doubtless vitally important in its own right but for our present purposes it is more accurately understood as an outcome” (his italics). Are (properly sanitized) social-capital theories compatible with MI? Might they even require MI? Woolcock (2001, p. 12) writes that “The most compelling empirical evidence in support of the social-capital thesis comes from household and community level (i.e., micro) studies, drawing on sophisticate measures of community networks, the nature and extent of civic participation and exchanges among neighbors”. Interestingly, Arrow (1999, p. 3) suggests that “the social networks guard against market failure that is caused by asymmetric information; they are supplementary activities that exploit monitoring devices not otherwise available”. Unlike his first venture into remedies to market failure (Arrow 1971), this observation refers to microfoundations, namely face-to-face interactions among the same persons over time. Whether these interactions have uniformly benign consequences is an empirical question, which should not be prejudged by the term “capital”. Arrow (1999, p. 4) states that Putnam’s “claim is that membership in associations strengthens political and economic efficiency even though the associations themselves play no role in either the polity or the economy”. In his analyses of the jury, Tocqueville may have been the first to put forward this argument: “The jury serves unbelievably to form the judgment and to augment the natural enlightenment of the people. That, in my opinion, is its greatest advantage. […] I do not know if the jury is useful to those who have legal proceedings, but I am sure that it is very useful to those who judge them” (Tocqueville 2010, volume 2, p. 448). He also emphasized the collateral benefits of local democracy: “This constantly recurring agitation that the

What’s the Alternative?

133

government of democracy has introduced into the political world passes afterward into civil society. Everything considered, I do not know if that is not the greatest advantage of democratic governments, and I praise it much more for what it causes to be done than for what it does. Incontestably, the people often direct public affairs very badly; but the people cannot get involved in public affairs without having the circle of their ideas expand” (ibid ., p. 398). I do not think he would have said that “bowling together” would have the same effects. Group agency. A number of scholars have asserted, more or less explicitly denying MI, that groups can act in ways that are not reducible to individual actions. Perhaps more appropriate terms would be that groups can decide to act or have joint intentions. While shared intentions are consistent with MI, joint intentions are not. Scholars who argue along these lines include Michael Bacharach, Michael Bratman, Robert Sugden, Margaret Gilbert, John Searle, Philip Pettit, and Christian List. They do not argue for exactly the same conclusion, but their arguments have a family resemblance. A recurring phrase is “I intend that we…”. The idea has gained some prominence, as shown by the publication of a Handbook of Collective Intentionality (Jankovic and Ludwig, eds., 2017). Since I claim that MI is trivially true, I may seem to be committed to the claim that these contributions are trivially false. Matters are more complicated, however, since advocates of this view engage only in conceptual analysis. I do not know of a single example of a group-agency scholar trying to (1) explain an empirical phenomenon and (2) argue that individual-level explanations are in principle unable to do the job. I confess to finding group agency and joint intentions unintelligible ideas, at least in their more extreme forms which include the stipulation of irreducibly collective emotions (Gilbert 2002, 2020). I shall not pursue these issues further. Instead, I shall ask whether something like a group agency might explain cooperation in a one-shot Prisoner’s Dilemma. I have in mind “magical thinking” and the closely related “Kantian optimization” (Elster 2017). These occur when the agents—not the scholars who study them —ignore or violate MI. Magical thinking is a well-documented tendency, notably in problems of cooperation, to think that “If I cooperate, others, being like me, will cooperate: therefore I shall cooperate”. The greater the similarity with other members of the relevant group, the stronger the tendency (Quattrone and Tversky 1984, p. 244). Kantian thinking is the imperative: “Do what would be best if everybody did the same”. These hypothetical forms of “group agency” can induce and explain the action.

134

J. Elster

List and Spiekermann. In the abstract of their article, Christian List and Kai Spiekermann (2013, p. 629) write that “Political science is divided between methodological individualists, who seek to explain political phenomena by reference to individuals and their interactions, and holists (or nonreductionists), who consider some higher-level social entities or properties such as states, institutions, or cultures ontologically or causally significant. We propose a reconciliation between these two perspectives”. I shall only consider what they write about the Copenhagen climate summit in 2010. As they write, As a stylized real-world example, consider the failed Copenhagen climate summit in 2010. Leaked audio recordings give a detailed picture of how the negotiations faltered in the final hours. An offended Wen Jiabao had withdrawn to his hotel room, directing his chief negotiator by mobile phone. Nicolas Sarkozy lost his temper. Angela Merkel was sidelined. Finally, Mammohan Singh, Lula da Silva, and others struck a minimal deal with Barack Obama […] Fascinating as this individual-level account is, it is of limited use as a causal explanation. As many analysts have observed, the summit failed for structural reasons, such as the large number of parties, the lack of a common interest, the arguably unhelpful legal framework of the UNFCCC process, and a US president without a domestic mandate for any substantial concessions. (List and Spiekermann 2013, p. 637)

I submit that the individual-level account is a complete causal explanation. What List and Spiekermann, propose is an argument about necessitation (Elster 2015, p. 16). To illustrate, consider a house of cards exposed to occasional wind gusts. That the house will collapse is a certainty. Close observation will tell us which gust made it collapse and the order in which the cards fell. Only the latter provides an explanation of the collapse of the house of cards as it happened. Tocqueville’s writings on the ancient régime provide an illustration. The last chapter of the published work is titled “How the revolution emerged naturally from the foregoing [chapters]”, which I take to be a statement about necessitation. The book is not a narrative and is compatible with many narratives, which have in common, roughly speaking, that feudal institutions break down and the absolute monarchy is transformed into a constitutional one. The posthumously published notes for a second volume show that he also intended to provide a narrative. In these notes, he also provides counterfactual narratives, such as what would have happened if all electoral districts had chosen their delegates to the Constituante by “crossvoting” (Elster 2013, Ch. 5), and concludes that it would not have made much of a difference (Tocqueville 2001, p. 62).

What’s the Alternative?

135

Assuming —a strong assumption—that one can identify “structural reasons” that necessitate the outcome, would they provide better insight into the explanandum than a well-supported narrative would do? My inclination is to say No, but do I have an argument to support it? My preference is partly esthetic: it is more satisfying to follow the actions and interactions of specific and idiosyncratic human beings than to subsume them under abstractions such as “the lack of a common interest”. I have also intellectual grounds, however, for my preference. Only the study of individuals close up can suggest psychological mechanisms for the scholar’s toolbox. Perhaps the victory of the American colonies in the War of Independence can be explained by “structural reasons”, such as the tendency of all large empires to break down because of their very size. In France, Turgot (2018, volume 1, p. 201) observed in 1750 that “colonies are like fruits that remain attached to the tree only until they are ripe: once they are self-sufficient they do what Carthage did [when gaining its independence from the Phoenicians in 650 B.C.] and what America will do one day”. He was right, but I find it more interesting to open that particular black box and find that the Americans were moved by the enthusiasm and not simply by anger (Elster 2023, Ch. 2). Psychologists do not study enthusiasm (Elster 2021), nor do structuralist political scientists, but some historians (Royster 1979; Lovejoy 1985) do.

6

Conclusion

I still believe that MI is trivially true, but I appreciate the fact that others disagree. One person’s trivial conclusion may be another’s deep mistake. Even the interpretive norm of respecting “the plain meaning of the text” does not always give a clear-cut answer. In twenty-one Supreme Court cases decided with an opinion during the spring of the 1993 term, a conflict between majorities and dissents derived at least in part from disagreements over the plain meaning of the statute at issue (Taylor 1995, p. 356, n. 162). In rethinking the issue four decades after my first engagement with it, a major source of reflection has been the four articles by Kenneth Arrow in which he briefly touches on MI. Arrow was arguably the greatest economist since Adam Smith (Keynes and Thomas Schelling are other contenders). He was not only a superb mathematical economist but also extremely wellendowed with intellectual honesty, common sense, and good judgment. From the pieces I have cited, one can conclude that he was both attracted to and worried about MI. He was attracted to it, because as an economist he built on the principle that individual agents are the building blocks of analysis. He

136

J. Elster

was worried about it because as a citizen he deplored that in modern societies the actions of these agents often lead to market failures. Since they did not always do so, he tried to identify the non-market mechanisms that compensate for or prevent such failures. The article on social capital is perhaps the one in which he comes closest to doing so without violating MI. In my opinion, history and psychology are, or ought to be, the main pillars of the social sciences. If one accepts that view, MI follows almost automatically, at least as an injunction. In all of this, I follow Mark Bloch (1964, p. 26), when he wrote that “the good historian resembles the giant of the fairy tale. He knows that wherever he catches the scent of human flesh, there lies his quarry (son gibier)”.

References Arrow, K. (1971). Political and economic evaluation of social effects and externalities. In M. Intriligator (ed.), Frontiers of Quantitative Economics (pp. 3-25). Amsterdam: North-Holland. Arrow, K. (1994). Methodological individualism and social knowledge. American Economic Review 84 (2). Papers and Proceedings of the hundred and sixth annual meeting of the American Economic Association, 1–9. Arrow. K. (1999). Observations on social capital In Dasgupta, P. and Serageldin, I. (Eds.), Social Capital. Washington D.C: The World Bank. Arrow, K. (2001). Reflections on the reflections. Journal of Health Politics, Policy and Law 26, pp. 1197–203. Bloch, M. (1964). The Historian’s Craft. New York: Vintage Books. Bourdieu, P. (1969). Intellectual field and creative project. Social Science Information 8 (2), pp. 89–119. Bourdieu, P. (1984). Distinction. Cambridge MA: Harvard University Press. Bourdieu, P. (1988). The Logic of Practice. Palo Alto CA: Stanford University Press. Browning, M. and Chiappori, P. (1998). Efficient intra-household allocations. Econometrica 66, pp. 1241–78. Cohen, G. A. (1978). Karl Marx’s Interpretation of History: A Defense. Oxford: Oxford University Press. Douglass, E. (1983). Rebels and Democrats. Chicago IL: Elephant Paperbacks. Durkheim, É. (1915). The Elementary Forms of Religious Life. London: Allen and Unwin. Durkheim, É. (1919). Les règles de la méthode sociologique. Paris: F. Alcan. Elster, J. (1981). Review of the French edition of Bourdieu 1984. London Review of Books 3 (20), 10–12. Elster, J. (1985). Making Sense of Marx. Cambridge: Cambridge University Press. Elster, J. (1993). Ethical individualism and presentism. The Monist 76, pp. 333–48.

What’s the Alternative?

137

Elster, J. (2013). Securities Against Misrule. Cambridge: Cambridge University Press. Elster, J. (2015). Explaining Social Behavior. Cambridge: Cambridge University Press. Elster, J. (2017). On seeing and being seen. Social Choice and Welfare 49, pp. 721– 34. Elster. J. (2021). Enthusiasm and anger in history. Inquiry 64, pp. 249–307. Elster, J. (2023). America Before 1787: The Unraveling of a Colonial Regime. Princeton: Princeton University Press. Ermakoff, I. (2008). Ruling Oneself Out. Durham NC: Duke University Press. Ermakoff, I. (2015). The structure of contingency. American Journal of Sociology 121, pp. 64–125. Gilbert, M. (2002). Collective guilt and collective guilt feelings. Journal of Ethics 6, pp. 115–43. Gilbert, M. (2020). Shared Intentionality, joint commitment, and directed obligation. Behavioral and Brain Sciences 43. Hansson, B. (2006). Why-explanations. Theoria 72, pp. 23–59. Haskel, J. and Martin, C. (1994). Capacity and competition. Journal of Industrial Economics 42, pp. 23–44. Haslanger, S. (2020). Failures of methodological individualism: the materiality of social systems. Journal of Social Philosophy, open access, pp. 1–23. Jankovic, M. and Ludwig, K. Eds. (2017). The Routledge Handbook of Collective Intentionality. New York NY: Routledge. List, C. and Spiekermann. K. (2013). Methodological individualism and holism in political science. American Political Science Review 107, pp. 629–43. Lovejoy, D. (1985). Religious Enthusiasm in the New World . Cambridge MA: Harvard University Press. Mackie, G. (1996). Ending footbinding and infibulation. American Sociological Review 61, pp. 999–1017. Martin, J. (2003). What is field theory? American Journal of Sociology 109, pp. 1–49. Marx, K. (1857). Grundrisse, cited after the translation in Penguin Books. London 1973. Marx, K. (1867). Capital I , cited after the translation in Penguin Books. London 1992. Marx, K. (1894). Capital III , cited after the translation in Penguin Books. London 1992. Megiddo, T. (2019). Methodological individualism. Harvard International Law Journal 60, pp. 219–65. Putnam, R. (2001). Social capital: measurement and consequences. Isuma: Canadian Journal of Policy Research 2, pp. 41–51. Quattrone, G. and Tversky, A. (1984). Causal versus diagnostic contingencies. Journal of Personality and Social Psychology 46, pp. 237–48. Royster, C. (1979). A Revolutionary People at War. Chapel Hill NC: University of Carolina Press. Sartre, J.-P. (1943). L’être et le néant. Paris: Gallimard.

138

J. Elster

Sobel, J. (2002). Can we trust social capital. Journal of Economic Literature XL, pp. 139-54. Skinner, B. (1981). Selection by consequences. Science 213, pp. 501–4. Taylor, G. (1995). Structural textualism. Boston University Law Review 75, pp. 321– 86. Tocqueville, A. de. (2001). The Old Regime and the Revolution, volume 2. Chicago: University of Chicago Press. Tocqueville, A. de. (2010). Democracy in America (bilingual edition). Indianapolis IN: Liberty Press. Turgot, A.-R. (2018). Oeuvres, volume 1. Paris: Institut Coppet. Woolcock, M. (2001). The place of social capital in understating social and economic outcomes. Isuma: Canadian Journal of Policy Research 2, pp. 1–17.

Methodological Individualism and the Rationality Principle

Emotions and Rationality Paul Dumouchel

Many researchers, for example Ben-Ze’ev (2000), Cairns (2003), Colombetti (2014), Damasio (2018), DeLancey (2002), De Sousa (1987), Dixon (2009), Dumouchel (1998), Ekman (1992), Elster (2000), Frank (1989), Fridja (1986), Harris (2004), Lebois et al. (2018), LeDoux (1998), Lutz (1988), Oatley (2004), Panksepp (1998), Russell (2003), Solomon (1976), or Wolheim (1999), to name but a few, believe and argue that they have the correct theory of emotions, that they know what emotions are, how many (basic) emotions there are, how they should be construed, and that they understand (at least partially) the role that emotions play in human behaviour. However, a central issue when it comes to writing about emotions in relation to explanation in social sciences, as in this handbook, is that these researchers disagree with each other. Among psychologists, philosophers, political scientists, sociologists, historians, anthropologists, social psychologists, neuroscientists, and even roboticists, there is no consensus as to what P. Dumouchel (B) Department of Philosophy, University of Quebec, Montreal, Qc, Canada e-mail: [email protected]

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 N. Bulle and F. Di Iorio (eds.), The Palgrave Handbook of Methodological Individualism, https://doi.org/10.1007/978-3-031-41512-8_8

141

142

P. Dumouchel

emotions are. How they should be construed, how many there are?1 Or what type of behaviour counts as emotional? These remain open questions. In fact, there is a lack of agreement on most issues concerning emotions, not only among these different disciplines, but also within each one taken individually. Russell (2003) reports that in psychology emotion is a confused and confusing field of study. If consensus concerning the basic elements of a discipline constitutes a sign of scientific maturity, then the science of emotions has clearly not yet reached its maturity. It may be wondered, however, if the social scientist, say the economist or the historian, needs to be committed to a particular theory of emotions. That is to say, unless the question is directly relevant to what she or he is working on does the criminologist need to decide whether the expression of basic emotions is universal or culturally relative, does he or she need to determine how many basic emotions there are? Can the social scientist not simply function using everyday concepts of emotions, words like anger, happiness, or sadness? One of the reasons why a researcher may not need to have a clear and definite concept of emotion to work with is because in everyday life, in everyone’s experience and interactions with others, emotions are evident. Everyone knows, or thinks he or she knows, what emotions are. All have experienced them. All can recognize and feel them. All are aware of how much emotions can influence or motivate behaviour. The common concept and vocabulary of emotions may be all that is necessary to explain behaviour, perhaps not from the point of view of fundamental psychology or in neurological theories, but at least in the context of methodological individualism, which gives central importance to how the agent understands her or his action. For it is precisely the everyday common-sense concept of emotions that agents use to interpret the world in which they act. Furthermore, using that common vocabulary has the advantage of making one’s text and explanation easily accessible to average readers. Terms of emotions, like anger, hope, or fear, also seems indispensable to appropriately describe the behaviours and situations one is analyzing. Do we need more? I think that we do. As Russell and Feldman Barrett (1999: 805) argue “(e)motions is too broad a class of events to be a single scientific category. As psychologist use the term, it includes the euphoria of winning an Olympic gold medal, a brief startle at an unexpected noise, unrelenting profound grief, 1

Elster (1999) reports that out of 17 different classifications of basic emotions there is no emotion that is common to all of them. In other words, there is no consensus among psychologists as to which emotions are basic. While De Sousa (1987) suggests that whenever an emotion has a different target it is a different emotion so that there would be an infinity of emotions, all of them different individual particular mental states. An idea that was already proposed by both Malebranche (1675) and Spinoza (1677).

Emotions and Rationality

143

the fleeting pleasant sensations from a warm breeze, cardiovascular changes in response to viewing a film, the stalking and murder of an innocent victim, lifelong love of an offspring, feeling chipper for no known reason, and interest in a news bulletin. The boundaries to the domain of emotion are so blurry that it sometimes seems that everything is an emotion.” This description also applies, even more so, to how the term is used in everyday life. It is not clear what type of event does or does not belong to the category of emotion, not clear which kinds of behaviour cannot be explained or justified by adverting to a term of emotion, like anger, fear, joy, remorse, resentment, being upset or surprised. Elster in “the place of emotions in the explanation of behavior” an appendix to Alchemies of the Mind (1999: 328–331) gives a list of twenty different ways in which emotions can enter into the explanation of behaviour. He adds that the hodgepodge nature of this list is obvious. However, he does not seem excessively troubled by it, as it reflects, he claims, the complexity of the facts. The problem remains nonetheless that there is no way of knowing if the list, this mixed bag of roles for emotions, is complete or not and merely reflects the particular interests of Jon Elster. It is clear that when we do research in social sciences, we cannot do otherwise than to use a particular conception of emotion in the explanation of behaviour. Whether we are aware of it or not we inevitably tend to pick and chose within that ill-defined category what best serves our purposes. When we are not aware of it, given that the notion is so vast and vague, the danger is that we use not only one conception of emotion, but different ones, surreptitiously sliding from one to the other in view of our explanatory goals. It is therefore necessary for us to discipline our use of terms of emotions, and to be clear about what “emotions” as we propose to understand the term are. As the role they can play in the explanation of behaviour depends on how we conceive emotions and affect in general. In what follows I will look at three conceptions of emotions and their different relations to reason. First, following Elster (2000) emotions as passions that are contrary to reason. Then, in line with recent developments in psychology which have known great success in economics and decisions theory, emotions as rational, as consonant with reason, and indispensable to rational decision taking. Finally, emotions as part of a mechanism of social coordination.

144

1

P. Dumouchel

Passions as the Opposite of Reason

Jon Elster, towards the beginning of Ulysses Unbound (2000: 7) proposes an interesting example of disciplining the use of an emotion concept. “I” use “passions”, he writes, “in an extended sense that covers not only emotions proper such as anger, fear, love, shame and the like, but also states such as drunkenness, sexual desire, craving for addictive drugs, pain, and other “visceral” feelings.” Clearly this is not the way in which we usually use either the word “passion” or “emotion”. We do not usually group together joy and craving for addictive drugs. Yet Elster has a good rationale for this amalgamation of “emotions proper” (or perhaps better of some emotions proper) with these rather different states. In spite of the heterogeneous bunch to which “passions” refers, Elster use of the term in a relatively clear and well-defined way. “Passions” he tells us should be understood as all forms of motivations that are the opposite of rational or reasonable motivations, which Elster characterizes “as any impartial – dispassionate or disinterested – motivation” (Ibid). In spite of this definition’s circularity—passions are all forms of motivation that are not dispassionate!—the term “passions” in Elster’s vocabulary has a specific meaning. It corresponds to all forms of motivation which may bring an agent to act irrationally and against which the agent may precommit or bind him or herself to ensure that, when the time comes, he or she will act rationally. To act irrationally, in this context, is for an agent to act against his or her enlightened or best self-interest, to act against what she or he thinks reasonable, or against what Rawls (1971) views as an agent’s well considered judgement. Passions understood in this way are necessarily conscious—though not necessarily at all time—and these conscious states are recognized by the agent as passions—though not necessarily when they occur. That is to say, in his or her more reasonable moments, the agent views them as motivations against which he or she should protect him or herself because they may bring the agent to act irrationally. Passions are thus characterized by a particular relation not only to reason, but also to time. For if the agent can precommit or bind herself, and finds it rational to do so, it is because reason allows her to foresee that in the future, when in the grip of a strong passion, she will be unable to escape its hold and to evaluate properly either the longer-term consequences of her action or the sacrifice of a cherished objective which giving in to the present motivation may imply. While reason brings us to shape the future with various projects which give sense to our life, passions imprison us in the present and impose a meaning or interpretation which we cannot control.

Emotions and Rationality

145

“Passions” have been understood in this way for a very long time, and this may explain Elster’s choice of that ancient term, rather than the more modern “emotions” (Dixon 2009), to designate the phenomena that interest him. Ancient authors—say Plato, Aristotle, or the Stoics—already presented passions as motivations that impair rational judgment, that distract us from and in many ways make impossible rational decision taking. Passions were conceived as opposed to rational behaviour and the type of states that Elster includes in his version of the category, i.e., addiction, drunkenness, or sexual desire, were typically viewed by ancient and many classical authors if not as passions themselves, at least as the result of unchecked passions. This way of construing passions (and emotions) thus has a very long and prestigious ancestry. Further, Elster (2000) argues that resistance to passions understood as what makes us momentarily lose the ability to act rationally, played an important role in shaping modern political constitutions. However, whether in Elster or ancient authors, passions (or emotions) are essentially viewed as what explains agents’ failures and departures from rationality.2 They provide false interpretations of the world and misrepresents the stakes that reason reveals. They are motivations against which the agent must take means to protect himself.3 Even if Elster recognizes that one passion can be used to protect oneself from another and, following Frank (1989), that passions can sometimes serve as self-binding devices, in all these cases the goal of the agent is to preserve his or her ability to satisfy the requirements of reason.4 Passions are construed as direct motivators, which as such drastically reduce or skew an agent’s choice of options. Self-binding or precommitment paradoxically free the agent from the domination of passions and restore the possibility of choice by reducing the number of available options. Either device functions as a means to take distance from the action or action tendency to which the passion—emotion or “visceral state”—directly leads, by limiting today the options which tomorrow will be available to the agent. 2

Elster (1999) clearly indicates—see especially chapter IV—that he is well aware that there are very different conceptions of emotions and of their relation to rationality, but in Ulysses Unbound he focuses on passions as motivations against which we need to be protected in order to act rationally. 3 On that issue, the major difference between Elster and the ancients is that they usually recommended (direct) self-discipline as a way of resisting to passions, while he essentially analyzes indirect ways of controlling them, methods that make it impossible for the agent to give in to her passions when it irrupts instead of asking her to control herself when the time comes. 4 In his analysis of Frank (1989) who argues that passions can be used as devices for self-binding which allow agents to resolve some paradoxes of rationality, for example the prisoner’s dilemma, Elster (2000: 45–53) proposes that in Frank’s argument it is not such much the passion itself, for example anger or friendship, that plays a fundamental role as the reputation for being vindictive or loyal. Therefore, the emotion is not what allows the agents to act rationality but constitutes information about the other player which allows a rational agent to reach an optimal solution.

146

P. Dumouchel

Passions so construed essentially concern an agent’s relations to him or herself. They transform the agent’s goals, preferences, cognitive states, or even motivations. For example, when in pain an agent may not want anything, except perhaps for that pain to stop, that is to say for his or her state to change and when depressed simply not want anything at all. Sometimes the cause of a passion is in the outside world, but not always, at other times a passion may be due to hormonal unbalance in the agent’s body. However, even when the cause of the passion is a change in external circumstances, my passion may be said to reflect that change, but it is not about that change.5 My passion is primarily about me, my goals, preferences or cognitive state rather than about the world. My passions or emotions have consequences on others through the influence they have on my behaviour. Passions therefore are private rather than social. They primarily concern agents’ characteristics, their behaviour, and desires which they modify. In all cases, passions are not rational because they invade the agent against his or her will and limit his or her ability to decide rationally. That is why Elster suggests that the problems they create should be addressed through devices such as self-binding and precommitment, which do not attempt to regulate the passions directly, but to limit the influence which they have on an agent’s choices.

2

Rational Emotions

In Elster’s list of the twenty ways in which emotions can enter into the explanation of behaviour, emotions are also construed as primarily concerning the agent’s relation to him or herself. The main difference is that in Alchemies of the Mind (1999) emotions, unlike passions, are not always opposed to reason and may at time directly help and support rational decision taking. Even in those circumstances, emotions are not in themselves rational. The subtitle of the book “Rationality and the Emotions” reflects the fact that beginning in the late 1980s and early 1990s the issue of the relation between emotions and rationality was renewed in both philosophy and psychology.6 Breaking with 5

Many authors, for example De Sousa (1987) argue that emotions are intentional states, for example my fear is about the snake which I see. In consequence, just like the belief that John is tall is a different belief than that John is short, then the fear of a snake would be different fear that the fear, say of a viper or of a charging bull. There would then be an infinity of different emotions, just as there is an unlimited number of objects of belief. However, while we explain the consequences of different beliefs in relation to their different content, we tend to explain the effects of different emotions in relation to the emotions that they are—joy, sadness, resentment—rather than in relation to what these authors conceive as their content. 6 De Sousa (1987) and Damasio (1994) respectively are early examples of this transformation in the two disciplines.

Emotions and Rationality

147

what had been until then the dominant position, more and more authors began to argue that emotions are not irrational. That they do not constitute an obstacle to rational decision taking. According to Damasio Descartes’ Error: Emotion, Reason and the Human Brain (1994) emotions, more precisely healthy emotional dispositions, are a necessary condition in order for an agent to act rationally. Though this change of outlook was seen by many as ground-breaking, in fact there is in philosophy a long tradition (Hobbes 1651; Hume 1739) according to which it is only emotions (or passions) that can motivate agents to action.7 According to that tradition, emotions in themselves are not opposed to reason, but necessary for agents to be able to act rationally. As Hobbes (1651: 139) for example wrote “a man who has no great passion… cannot possibly have either great fancy or much judgment”. Passions are not obstacles that prevent rational behaviour, but necessary for reason to function. Only when emotions become too powerful and escape the agent’s control will they lead the agent to act irrationally. Among many of those classical authors affective dispositions were also divided into two categories. One the hand, passions, like pride and vain-glory that were responsible for irrational and immoral behaviours. On the other hand, moral sentiments like sympathy (Smith 1759) or affection and a sense of partnership (Shaftesbury 1711) which were considered consonant to reason and conducive to peaceful living. For various reason, this tradition which was once dominant progressively came to be seen as a historical curiosity which had little to teach us about the true nature and social function of emotions. Hirschman (1977) provides a remarkable analysis of some aspects of how this transformation took place. Another factor that helps to explain the retreat of this tradition is the replacement of the category of passion by that of emotion. As Dixon (2009) shows, the notion of emotion became dominant in nineteenth-century medicine and psychology as an alternative to the idea of passions, because that ancient term carried at least an implicit, and frequently an explicit, moral judgment on the behaviour which it described. The category of passion in consequence came to be seen as neither scientific, nor objective and the concept of emotion was introduced as heralding a new science of the affective dimension of the human mind. In the wake of this rejection of passions, moral sentiments came to be essentially viewed as moral dispositions rather than as sentiments. That is to say, as virtues or the conclusions of moral reasoning, rather than as a particular form of affect.

7

The most famous statement to that effect his Hume’s claim that “Reason is, and ought only to be the slave of passions” (1739: 415).

148

P. Dumouchel

Though Damasio’s Looking for Spinoza (2003) suggests that he is partially aware of this century-old tradition, three important differences separate it from the contemporary reassessment of the relations of emotions to reason. First, as progressively became evident over the year, the new conception of these relations coincides with a transformation of the concept of reason. More precisely, it goes together with a trend at naturalizing reason. Throughout the philosophical tradition reason, rational behaviour always had a normative dimension. In the sense, not only that rational behaviour is distinct from and opposed to irrational behaviour, but also in that reason was thought to provide a criterion that allows us to distinguish one type of behaviour from the other. Reason is prescriptive and not merely descriptive of certain behaviour. It tells us what we should do, for example, if we want to reach our goal or to maximize our utility function. In the naturalizing approach, reason came to be seen as one in a family of strategies, among which natural selection is the most prominent, which augment the probability that an agent will reach his or her desired goal (Gigerenzer and Selten 2002). Simultaneously rational behaviour came to be equated with purposive action. Any action that aimed towards a goal or purpose was construed as being at least minimally rational (De Sousa 2007; Ross 2005). Any strategy that augmented the agent’s probability of success in reaching his or her objective was construed as more robustly rational. Within this context nothing intrinsically distinguishes rationality from other strategies that also augment the probability that the agent will reach his or her objective. In other words, nothing allows us to distinguish a successful behaviour that is rational from one that is motivated by an emotion, or a passion. If that second behaviour also augments the agent’s probability of success it can legitimately be described as “rational”. In consequence it became extremely difficult to distinguish one from the others, reason from emotions. Therefore, if the relation between reason and emotions has recently changed and been renewed, it is not so much because emotions are now evaluated differently, as because the meaning of reason has changed. Reason may still be seen as normative inasmuch as it allows us to distinguish between rational and irrational behaviour, but it is not as prescriptive anymore. The reason why this is so, is because the conclusion that a strategy is rational (i. e. that it augments the probability of the agent’s success) is the result of that strategy’s past rate of success, rather than a consequence of the analysis of the projected action. To put it otherwise, it is not the reason anymore that provides the criteria of which action are not rational, but the statistical

Emotions and Rationality

149

summation of the past success or failure of the strategy that recommends the action which constitutes the criteria of its rationality. As indicated by such books as Gut Feelings: The Intelligence of the Unconscious (Gigerenzer 2008) or Thinking Fast and Slow (Kahneman 2011), within the naturalized approach to human thinking there is little room for reference to either reason or emotions. Both concepts have been subsumed under (or perhaps absorbed into) the categories of “methods of thinking” or of “adaptive tool box”. In a way the two concepts have been abandoned.8 Casting them aside or overlooking their relevance does not, however, have the same consequence for each one of them. Given that emotion is a vague, ill-defined concept and that it is not clear which behaviours do and which do not fall under that heading, the fact that in certain circumstances some researchers avoid the notion does not in anyway threaten its existence and legitimacy. More precisely, whatever transformation is imposed on the idea of emotion in the context of decision taking is of little importance. However, when the claims of reason are dissolved in a pool of related strategies, the outcome is different. It is its ability to guide action that is challenged and its existence as a prescriptive notion that is rejected. The second important difference between the contemporary reassessment of the relation between reason and emotion and the philosophical tradition that construed moral sentiments as consonant with reason, concerns their cause and the object that is the goal or function of these affective dispositions. In the naturalist approach the cause of the “gut feeling” leading to effective decision taking is of little interest. What is important is if and to what extent it helps the agent to reach his or her objective. This indifference to the cause of the affect is indicated, among other things, by the fact that the feeling’s relative success as a rule of decision is usually explained in reference to natural selection. If a fast and sloppy algorithm turn out to be as good, or because faster even better than more rigorous calculations in bringing about the choice of the appropriate behaviour, it is because, the claim goes, it has been selected over a very long period of time and has become part of our “adaptive tool box” (Gigerenzer and Selten 2002). Selective explanations however, as Sober (1984) argues, are not causal explanations. They are distal explanations. Unlike proximate causal explanations they do not exhibit the mechanism that produces the selected phenomena, characteristic,

8

A quick look at the index indicates that the term “reason” only happens once in Kahneman (2011).

150

P. Dumouchel

or behaviour. Selective explanations simply advert to the historical sequence that led to its fixation.9 This indifference to the cause of the behaviour is not accidental, it reflects the fact that only the result is important or relevant. The goal of the emotion, intuition, insight, or gut feelings is to help the agent reach his or her preferred objective or more generally perhaps to enhance the agent’s fitness. Affective dispositions that help agents in successful decision making do not need to be perfect. They can often lead to mistakes and cause bad decisions. This is not surprising given that perfection is never what natural selection produces. Unlike what has often been the claim of reason, those affective dispositions do not promise optimal results.10 All that is necessary is for them to give results that are “good enough”.11 To summarize, when comparing the contemporary “rapprochement” between emotions and rationality with the traditional conception of emotions that are not opposed to reason, we find three major differences. First, because it is now the relative success of the decision rule that provides the criteria of the “rationality” of both affective dispositions and of “rational decisions”, in the contemporary development it is not so much that some emotions are viewed as rational, as the fact that the distinction between emotions and rationality tends to be erased which questions the classical opposition between them. Second, the cause or origin of these emotions has no relevance in their relation to reason. This is closely related to the third point, that these dispositions in their relation to reason do not have any goal in particular, that is to say, they have no goal that is distinct from the agent’s own objectives. In that sense, they are similar to passions as Elster conceives them. They correspond

9 In answer to the question: “why do all the children in this kindergarten speak both French and English?” a selective explanation consists, for example, in the following response: because only bilingual children are accepted in this classroom. Such an answer provides a rule of selection, but it is indifferent to the particular chain of events responsible for the fact that this or that child speaks both languages. For example, because one child was born from French parents in England or because one parent is French and the other English, and so on for every child. Selective explanations, to the extent that they aim at explaining the fixation of a characteristic within a given population, are indifferent to the mechanism that causes the disposition or behaviour. 10 Note however, that rational decision theory or the theory of game only promise optimal choice within the domain of expected utility. In the world of what effectively happens the results can be less (or more) successful, this however has no direct incidence on whether or not the decision was rational. To the opposite, what matters (or should matter—see the next note) for a fast and rough algorithm that was selected are effective results, rather that paybacks in term of expected utility. 11 As I argued elsewhere (Dumouchel 2018) there is at work here a meta-criteria. Over and above the disposition’s relative success, or rather in order to evaluate that success, the analyst uses a form of cost/benefit analysis. The level of success of the rough and ready algorithm is determined by comparing its suggestions with the recommendations of rational decision taking. The comparison takes into account not only the outcomes themselves, but also the cognitive investment involved., what is sometimes referred to as the “accuracy – effort trade-off ” (Stigler 1961).

Emotions and Rationality

151

to changes in the agent’s relation to him or herself, changes that are evaluated in view of their relevance to the agent’s objectives. Positively in the case of “gut feelings” or “insight”, negatively in the case of passions.

3

Moral Sentiments and Interactive Emotions

Moral sentiments theories and interactive approaches to emotions which we will soon consider are different on all three points. Contrary to the current attempts at renewal of the relations between emotions and reason that were presented in the previous section, here the distinction between emotions and rationality is maintained. Moreover, it is a prescriptive view of rationality and rational action that provides the criterion to evaluate if and when affective dispositions are rational. Second, the cause or origin of moral sentiments, or of interactive emotions, is central to these approaches. Their social origin is part of what defines these affects as moral sentiments or interactive emotions. Third, unlike the “gut feelings” that lead to acceptable decisions, or the “passions” that recommend outcomes that go against the agent’s best interest, moral sentiments, and interactive emotions have objectives, or a function, of their own that cannot be reduced to (or identified with) those of the agent who experiences the emotions. To put it otherwise, unlike these, moral sentiments and interactive emotions are not to be primarily construed as a transformation of the agent’s relation to him or herself. Rather, what is characteristic of moral sentiments and interactive emotions is that they lead to a transformation of the agent’s relations with other agents. This function or objective is not accidental and external to what constitutes it as an emotion, but intrinsic to the type of affect that they are and it is in relation to their social function that their rationality, or lack thereof is evaluated. Interactive emotions differ from the conceptions of emotions and of passions previously presented in that they are construed as being primarily about an agent’s relations to other agents, rather than as a transformation of the agent’s relation to him or herself. They are not private, individual events, but social and in a sense public. In psychology and in philosophy the social dimension of affect is recognized as important often enough, but it is not generally considered as central to the concept of emotions (Tiedens & Leach 2004). Either as in Hess and Philippot (2007) the interactive, communicative dimension of emotions is construed as added to and relatively external to the emotion itself. In other cases, some emotions, for example admiration, empathy, or shame (Lewis 1992) are considered to be social, while others, like fear or surprise, are viewed as being individual and “non-social” because the

152

P. Dumouchel

latter are thought to only require an individual’s awareness of his or her physical environment. In this conception of social emotions, being social is not a characteristic of emotions as such, though it is considered to be fundamental of certain emotions. Finally, emotions are thought of as social in the context of emotional contagion (Herrando & Constantinides 2021). For example, a crowd that runs away in panic, laughter that spreads to all members of the assembly, or the mechanism underlying the dissemination of rumours and conspiracy theories. In all these cases, being social is an extrinsic rather than intrinsic characteristic of emotions. It is something that happens to them, or at least to some of them, rather than part of what they are. To the opposite, in what follows I will consider that human affect (and emotions) is primarily or intrinsically interactive and social. This claim has two interrelated aspects. It can be seen both as a proposal concerning the use of the vocabulary of emotions, and as a theoretical hypothesis concerning the nature of human affect. My goal here however is not to propose a new theory of emotions,12 but to illustrate how different conceptions of emotions have different relations to reason. Interactive emotions present a different image of the relation between emotions and rationality than what is the case when emotions are considered to be essentially private individual mental states. First, before we come to the analysis of the relation of reason to interactive emotions, we need to return to my double proposal. On the one hand, it can be understood as a decision to limit the use of the vocabulary of emotion, restricting its use to social phenomena. More precisely, I claim that affect is the central piece of the process of intraspecific coordination among humans, and that the use of that term, and that of emotion, should be restricted to that context.13 This decision to better discipline our vocabulary of emotion can easily be justified from a methodological point of view. Given that the vocabulary of emotion is used in so many different ways, or as Russell and Fledman Barrett (1999) argue, given that “emotions is too broad a class of events to be a single scientific category” it seems that scientific analysis of human affect will only be possible either if we abandon the term “emotion” or, at least, if we limits its use to a subset of the numerous and heterogeneous phenomena that it can indicate in

12

Something which I have done elsewhere. See Dumouchel (1998, 2002, 2008, 2011, 2017a) and Ross and Dumouchel (2004). 13 Actually this form of coordination is not strictly intraspecific as it also extends to our interactions with some animals, especially members of domesticated species.

Emotions and Rationality

153

everyday language.14 This disciplining is more likely to have positive effects if the chosen subset of phenomena has a sufficient coherence. That is to say, if it is possible to provide a criterion that justifies and makes clear the proposed limits of the use of the vocabulary of affect and emotions. This is precisely the goal or function of the second aspect of my proposal. The claim that intraspecific coordination is the central function of human affect suggests guidelines to decide which phenomena, or at least what type of phenomena fall under the category of emotion. One consequence of this disciplining of our vocabulary is that it will exclude from the category of emotions some, perhaps many phenomena that we tend to consider as such in everyday language. This is inevitable though it may give rise to some criticism. Yet, as argued long ago by Bachelard (1938) and by Wolpert (1994) somewhat more recently, we build science by rejecting the common-sense image of the word. It requires us to correct our everyday vocabulary which both reflects and is structured by that common-sense image. There is no reason why what is true of physics and of biology should not also apply to the study of emotions. This last claim, the attentive reader may already see, suggests, at least in its more general form, the nature of a difficulty we will need to address. As long as a theory of emotions respects our every day vocabulary, more precisely as long as it is in sufficient agreement with the way we spontaneously understand them, its findings can easily be integrated into the approach of methodological individualism. Elster (2000) study of self-binding and precommitment is a good example of this. Even though his use of the term “passions” does not correspond to the everyday use of that term, or of “emotions”, it is sufficiently consistent with how agents spontaneously understand the action of diverse motivations which are beyond their control, for his analysis to be built on how they understand and interpret the world and the forces it contains. Agents bind or precommit themselves because they already know that tomorrow they will want what they presently do not want and think that they should not want. They seek to protect themselves from this undesirable transformation. Their reasons for acting as they do are transparent to us as they are to them. This does not require us to share their appreciation of the difficulty, nor does it prevent us from preferring a different strategy, but we understand what they do and why they do what they do, because we understand how they understand the world.

14 See Russell (2003) for a somewhat similar argument. Russell proposes to distinguish between “core affect” which is biologically grounded, while “emotions” correspond to affective episodes distinguished by characteristics of the situation where they appear, including its cultural dimension.

154

P. Dumouchel

As long as we consider emotions or passions as transformations of an agent’s relation to herself—her goals, preferences, or cognitive access to the world—and as long as the agent has access to the result of this process of transformation, both the agent and the observer can be completely blind to the particular mechanism that caused the emotion, because the agent’s emotion, anger or fear, corresponds to her present understanding of the world that explains her action. Of course, one may want to psychoanalyze the agent or give a neuro-physiological explanation of the irruption of the anger or fear, but such explanations determine limits that constrain the agent’s actions and choices, in the same way as do a wall or a river. They enter as part of the explanation, but they are not part of the way a rational agent determines his or her action. This is exactly how in Elster (2000) both the author and the agents consider the passions against which they resort to self-binding. They view them as external limits, events like a flood against which a community protects itself by building a dam. Does the problem situation change in the case where, not only the emotion moves them in ways that they do not understand, but if it is also their misunderstanding of the world that explains how they acted? In other words, do we need to resort to a different explanatory model if making sense of their actions requires that we have access to an understanding of the world to which they do not have and to which they cannot have access? When that happens are we necessarily adopting a different explanatory model, shifting to a causal explanation that acts “behind the back” of the agents and determines their action? Not necessarily, since resorting to an understanding of the world to which the agents did not and could not have access is precisely what we do when we explain a social situation as the unintended consequence of intentional actions, or as what Boudon (1979) calls “perverse effect”, or what Hirschman (1977) described as the unrealized effects of intentional action. In all these cases, the actions of the agents have consequences which they did not and could not foresee, and our explanation rests on an understanding of the world which they did not have and to which generally they could not have access. However, these consequences proceed from (usually from the aggregation of ) their actions as they intended them, rather than as the result of an unexpected accident. It is the agents’ understanding of their action, of its goal, of what they saw as the occasion that prompted it, that is fundamental to the explanation of these unintended effects.

Emotions and Rationality

4

155

Affect as Social Coordination

A central aspect of affect, as indicated above, is that it refers to essentially social or if you prefer, inter-subjective events. Rather than individual characteristics or private events that concern an agent’s relation to him or herself, affect and emotions (as I will use the term from now on) relate agents to one another. Humans are a highly social species in the sense that most of the advantages and disadvantages that befall us take place in the context of our relations with other co-specifics. One is rarely, if ever in the twenty-first century, solely in contact with a natural—in the sense of a non-social—environment. The point is not that all affective events require a social context, rather it is that none take place outside of that horizon which shapes their meaning. All our interactions with the world are mediated by our interactions with others. As the recent pandemic clearly illustrates, even our immune reaction to pathogens is a highly social affair. Affect is what makes us attuned to each other. It is the central piece of the mechanism that makes us social, that allows us to coordinate with one another. By attuned I do not simply mean to be in agreement, for to be opposed to someone is also to be attuned to them in a way. Affective coordination precedes both competition and cooperation, both collaboration and conflict (Dumouchel 2017b; Jackson-Jacobs 2013). The theory of games usually distinguishes between coordination or cooperation on the one hand, and competition on the other, depending on whether the interests of the agents converge (games of coordination) or if they diverge (competition). It considers those interests and the possible moves in the game as given. Competition and coordination are determined in relation to particular actions or goals, and a game is defined by its payoffs matrix. Affective coordination to the opposite is defined in relation to persons. It relates agents to each other directly, rather than in relation to specific goals that are (primarily) attached to each one individually. It precedes both cooperation or competition, because rather than offering an answer to the question “what is the best strategy in this or that game?” it answers the questions: “who do I want to play with?” and “what game do I want to play?”. These two last questions are not unrelated, for no matter what the stakes are, it is not the case that I am ready or want to play any game with just anyone. Moreover, it is affective coordination rather than the payoffs which explains why today I intend to play with you a game, which yesterday, not only I did not want to play, but never thought that I would at some time desire to play. The agents’ interests and the stakes that the theory of games or rational decision theory takes as given are only so momentarily and contextually. One

156

P. Dumouchel

reason why they are subject to change is because of the transformation of our intentions towards each other, and affective coordination is the fundamental process through which our reciprocal intentions are determined and modified. The biological mechanism underlying this process is continuous, in the sense that it is always active. Even when we are alone and not doing anything in particular, we subtly signal possible dispositions towards others that will be picked up by any observer and determined at the moment of an interaction. Emotions can be understood as salient moments in this process of coordination, and because there are many reasons why moments may be salient, including the history of the relation, and the type of relation that already exists (or not) between the agents, emotions will not form a homogeneous class of events. At its most fundamental level at least, this process is not rational for many reasons or in many senses. It does not aim to maximize the agent’s utility functions either taken individually or as partners in a game of coordination. Mostly, it is not a purposeful action. Coordinating with others is not a goal that the agents pursue.15 Further, there is not opposite to affective coordination, only failure or absence of coordination. Unlike what is the case between cooperation and competition, there is no particular behaviour that is the opposite of coordinating with the other, for example agreeing rather than disagreeing. Just like what is arational, rather than irrational, is not the opposite of being rational, but something else,16 in the same way the absence of affective coordination is not its opposite, but simply what it is not. There is no particular goal that affective coordination seeks to realize. The process of affective coordination is not (to be understood as) the result of a teleological mechanism. Coordination is achieved whenever the agent’s uncertainty concerning each other’s intention towards each other is sufficiently reduced. Affective coordination is a phenomenon that takes place spontaneously. It is not any agent’s action. Yet it takes place as an interrelation between agents. Emotions are classically (Frijda 1986) seen as mental states that incline or prepare people to certain actions, for example, to fight, to flee, or to scan the environment attentively. In the context of affective coordination my affective expression may be seen as both an expectation about another agent’s intention towards me and a signal concerning my intention towards him or her. The expression of affect may, but it need not, come together with a conscious 15

At least not at the elementary level. In a second time one may seek to cultivate a particular person’s friendship, but that conscious decision requires that one’s disposition towards that person has already been to some extent determined, and in the continuous on-going process of doing that one may come to abandon the project of making a friend of that person, for reasons which are related to the other, but not necessarily to his or her willingness to accept that friendship. 16 What is arational we could say indicates the complementary set of the rational.

Emotions and Rationality

157

mental state, a mental state that many authors identify with the emotion itself. Most of the time we are not aware of our affective expression, nor that others react to it and that we react to theirs (Trevarthen 2009). At time in this interactive process a conscious mental state appears which may (or may not) be accompanied by a physical transformation (Miller 2004). When this happens, the mental state will also correspond to a salient moment in the process of coordination, I propose to call such events an “emotion”. That mental state however, neither has the propositional content of being an expectation, nor does it correspond with the intention to signal. In consequence, what the agents’ affective expressions do—allow them to coordinate with one another—does not correspond to the agent’s understanding of the world nor to what he or she intends. She intends to flee, or fight, or to spend more time with that person. However, if affective coordination is not rational because it is not an action and does not aim at any behaviour in particular, it determines the basic conditions that allow agents to act rationally. It determines their intention towards each other and colours how they perceive the world. Methodologically, the advantage of understanding affect and emotions as part of a mechanism of social coordination is that it rejects the conception of emotions as private events that are essentially hidden in the person’s mind or body. Emotions are not private events hidden in the agent’s mind or body. Nonetheless, because intraspecific coordination to a large extent takes place at a sub-personal level—mostly we are not aware of it, and aware only of its result—it is not public in the common sense of that term. It is however social. In fact, it is at the core of human sociality, of the interest that we take in each other.

References Bachelard, G. (1938). La formation de l’esprit scientifique. Paris: Vrin. Ben-Ze’ev, A. (2000). The Subtlety of Emotions. Boston: MIT Press. Boudon, R. (1979). Effets pervers et ordre social . Paris: Presses Universitaires de France. Cairns, D. L. (2003). The Politics of Envy: Envy and Equality in Ancient Greece. In Konstan D. and Rutter, K. (Eds.), Envy, Spite and Jealousy: The Rivalrous Emotions in Ancient Greece. Edinburgh: Edingburgh University Press. Colombetti, G. (2014). The Feeling Body. Affective Science Meets the Enactive Mind . Boston: MIT Press. Damasio, A. (1994). Descartes’ Error: Emotion, Reason and the Human Brain. New York. Putnam.

158

P. Dumouchel

Damasio, A. (2003). Looking for Spinoza: Joy, Sorrow and the Feeling Brain. New York. Harcourt. Damasio, A. (2018). The Strange Order of Things: Life, Feeling and the Making of Cultures. New York: Pantheon. DeLancey, C. (2002). Passionate Engines: What Emotions Reveal about Mind and Artificial Intelligence. Oxford: Oxford University Press. De Sousa, R. (1987). The Rationality of Emotions. Boston: MIT Press. De Sousa, R. (2007). Why Think? Evolution and the Rational Mind . Oxford: Oxford University Press. Dixon, Th. (2009). From Passions to Emotions. Cambridge: Cambridge University Press. Dumouchel, P. (1998). Emotions: Essai sur le corps et le social . Paris: Seuil. Dumouchel, P. (2002). Agency, Affiliation and Emotions. In Trappl, R. (Ed.), Cybernetics and System 2002. (pp. 727–733). Vienna: Austrian Society for Cybernetics Studies & University of Vienna. Dumouchel, P. (2008). Social Emotions. In Canamero L. and Aylett R. (Eds.), Animating Expressive Characters for Social Interaction. (pp. 1–19). Amsterdam: John Benjamin Publishing Company. Dumouchel, P. (2011). Emotions and Mimesis. In Garrels, S. (Ed.), Mimesis and Science Empirical Research on Imitation and the Mimetic Theory of Culture and Religion. (pp. 75–86). East Lansing: Michigan State University Press. Dumouchel, P. (2017a). Of Objects and Affect: Artificial Empathy, Pure Sociality, and Affective Coordination. Japanese Review of Cultural Anthropology 18(1), 99– 113. Dumouchel, P. (2017b). Acting Together in Dis-Harmony. Cooperating to Conflict and Cooperation in Conflict. Studi di Sociologia 4, 303–318. Dumouchel, P. (2018). Rationality, Irrationality, Realism and the Good. In Bronner, G. and Di Iorio F. (Eds.), The Mystery of Rationality. (pp. 55–66). Cham: Springer. Ekman, P. (1992). An Argument for Basic Emotions. Cognition and Emotion 6, 169–200. Elster, J. (1999). Alchemies of the Mind . Cambridge: Cambridge University Press. Elster, J. (2000). Ulysses Unbound . Cambridge: Cambridge University Press. Frank, R. H. (1989). Passions within Reason. New York: W.W. Norton & Company. Frijda, N. (1986). The Emotions. Cambridge: Cambridge University Press. Gigerenzer, G. (2008). Gut Feelings: The Intelligence of the Unconscious. New York: Penguin Books. Gigerenzer, G. and Selten, R. (2002). Bounded Rationality the Adaptive Tool Box. Boston: MIT Press. Harris, W. V. (2004). Restraining Rage: The Ideology of Anger Control . Cambridge, Mass.:Harvard University Press. Herrando, C. and Constantinides, E. (2021). Emotional Contagion: A Brief Overview and Future Directions. Frontier in Psychology 12, 712606. https://doi. org/10.3389/fpsyg.2021.712606.

Emotions and Rationality

159

Hess, U. and Philippot, P. (2007). Group Dynamics and Emotional Expression. Cambridge: Cambridge University Press. Hirschman, A. O. (1977). The Passions and the Interests. Princeton: Princeton University Press. Hobbes, Th. (1651). Leviathan. In Macpherson, C. B. (Ed.), Harmondsworth. Penguin Books, 1976. Hume, D. (1739). A Treatise of Human Nature. Oxford: Clarendon Press, 2014. Jackson-Jacobs, C. (2013). Constructing Physical Fights: An Interactionist Analysis of Violence among Affluent Suburban Youth. Qualitative Sociology 36, 23–52. Kahneman, D. (2011). Thinking Fast and Slow. New York: Random House. Lebois, A. M., Wilson-Mendenhal, C. D., Simmons, W. K., Feldman Barret, L. and Barsalou, L. W. (2018). Learning Situated Emotions. Neuropsychologica. https:// doi.org/10.1016/j.neuropsychologia.2018.01.008. LeDoux, J. E. (1998). The Emotional Brain. New York. Simon and Schuster. Lewis, M. (1992). Shame: The Exposed Self . New York. Free Press. Lutz, C. (1988). Unnatural Emotions. Chicago: The University of Chicago Press. Malebranche, N. (1675). De la Recherche de la Vérité. Paris: Galerie de la Sorbonne, 1991. Miller, R. (2004). Emotion as Adaptive Interpersonal Communication. The Case of Embarrassment. In Tiedens, L. Z. and Leach, C. W. (Eds.), The Social Life of Emotions. (pp. 87–104). Cambridge: Cambridge University Press. Oatley, K. (2004). Emotions: A Brief History. Hoboken: Wiley-Blackwell. Panksepp, J. (1998). Affective Neuroscience: The Foundations of Human and Animal Emotions. Oxford: Oxford University Press. Rawls, J. (1971). A Theory of Justice. Cambridge, Mass.: Harvard University Press. Ross, D. (2005). Economic Theory and Cognitive Science. Boston: MIT Press. Ross, D. and Dumouchel, P. (2004). Emotions as Strategic Signals. Rationality and Society 16(3), 251–286. Russell, J. A. (2003). Core Affect and the Psychological Construction of Emotions. Psychological Review 110(1), 145–172. Russell, J. A. and Feldman Barrett, L. (1999). Core Affect, Prototypical Emotional Episodes and Other Things called Emotion: Dissecting the Elephant. Journal of Personality and Social Psychology 76(5), 805–819. Shaftesbury, A. (1711). Characteristics of Men, Manners, Opinions, Times. 3 vol. Indianapolis: Liberty Fund, 2001. Smith, A. (1759). A Theory of Moral Sentiments. Cambridge: Cambridge University Press, 2014. Sober, E. (1984). The Nature of Selection. Boston: MIT Press. Solomon, R. C. (1976). The Passions: Emotions and the Meaning of Life. New York: Doubleday Books. Spinoza, B. (1677). Ethics in The Collected Writings of Spinoza. Vol. 1. Curley, E. (trans.), Princeton: Princeton University Press, 1985. Stigler, G. J. (1961). The Economics of Information. Journal of Political Economy 69(3), 213–225. https://doi.org/10.1086/258464.

160

P. Dumouchel

Tiedens, L. Z. and Leach C. W. (2004). The Social Life of Emotions. Cambridge: Cambridge University Press. Trevarthen, C. (2009). Embodied Human Intersubjectivity: Imaginative Agency, To Share Meaning Cognitive Semiotics 4, 6–56. https://doi.org/10.1515/Cogsem. 2009.4.1.6. Wolheim, R. (1999). On the Emotions, New Haven. Yale University Press. Wolpert, L. (1994). The Unnatural Nature of Science. Cambridge, Mass.: Harvard University Press.

Rationality, Praxeology, and History: The Contributions of Ludwig von Mises to the Theory of Rationality in the Social Sciences Christian Robitaille and Robert Leroux

Ludwig von Mises is known as an economist who, in addition to his contributions to the elucidation of various topics of economic analysis, has also contributed to the establishment of rigorous methodological prescriptions for the proper study of the social sciences in general. Indeed, Mises offered a vigorous defense of methodological individualism at the center of which stands a singular conception of the actor’s rationality. In particular, his conception of rationality as a praxeological category constitutes an epistemological insight which can be quite useful for social analysis. Mises’s methodological rationalism provides a framework for social investigations that are rooted in the desire to understand individual motivations and actions which, in the last analysis, constitute the ultimate causes of all aggregate social enigmas. This chapter hence seeks to clarify Mises’s theory of rationality. In this perspective, the following will be divided into three distinct but interrelated sections. The first section will discuss Mises’s analysis of rationality as a praxeological category, i.e., as a logical implication of the general concept of action. The second section will discuss his understanding of rationality as a tool for the understanding of specific or general historical events, in particular C. Robitaille (B) Department of Sociology, Liverpool Hope University, Liverpool, UK e-mail: [email protected] R. Leroux Department of Sociology, University of Ottawa, Ottawa, ON, Canada

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 N. Bulle and F. Di Iorio (eds.), The Palgrave Handbook of Methodological Individualism, https://doi.org/10.1007/978-3-031-41512-8_9

161

162

C. Robitaille and R. Leroux

in relation to Max Weber’s conception of rationality. Last, this chapter will present two examples of how Mises’s clarifications on the concept of rationality can be useful to the study of specific topics, namely the problem of economic calculation under socialism and the problem of polylogism.

1

Rationality as a Praxeological Category

Classical theories of rationality in the social sciences all share in common a few features stemming from Enlightenment thinking. Indeed, David Hume’s famous dictum—“Reason is, and ought only to be the slave of the passions, and can never pretend to any other office than to serve and obey them” (Hume [1739] 1978, p. 415)—has been taken seriously by political economists and other social scientists from the time Adam Smith began his study of moral sentiments up to the time Ludwig von Mises himself started to write on economics. Reason, in this perspective, is to be understood as instrumental , i.e., as a tool used by individuals to attain their goals which stem from desires determined by various passions. These passions are unanalyzable by reason and are, hence, taken as given by political economists and other social scientists in their analyses. This notion of instrumental rationality is also strongly linked with the conception of actors as self-interested beings. Indeed, the function of reason is precisely to attempt to reach a goal determined, in the last analysis, by the actor’s passions. Hence, reaching the goal is expected to alleviate the actor’s suffering or to increase his pleasure. In other words, utilitarianism is also an important component of rational action in classical economics. Indeed, rationalism, in its classical conception, goes hand in hand with both instrumentalism and utilitarianism (see Demeulenaere 1997, pp. 12–17). These aspects of rationality are all present in Mises’s work, as will become obvious in a moment. There are, however, other elements in Mises’s theory of rationality which are more original than what can be found in the work of his predecessors. These original insights stem from Mises’s acceptance of Carl Menger’s paradigm for the study of economics and from his desire to systematize it (Menger [1871] 2004; [1883] 1985; Robitaille 2022). Indeed, Mises is building a system for the study of the social sciences in general, and of economics in particular. As such, even the elements he borrows from his eminent predecessors are re-evaluated from the perspective of the systematic praxeological foundations he uses. Mises argues that some of these elements are no longer to be considered as mere assumptions, but rather as rigorously demonstrated truths that are presupposed by any proper social

Rationality, Praxeology, and History: The …

163

scientific investigation. In particular, rationality defined in an instrumental and utilitarian manner is conceived as inextricably linked to action and, as a consequence, constitutes a necessary feature of any social analysis. Let us see how Mises proceeded to establish this. To Mises, the epistemological foundations on which the social sciences rest stem from the recognition of human action. Indeed, social epistemology “does not deal with thinking, perceiving and knowing in general, but with human thinking, perceiving and knowing” (Mises 1962, p. 2). Human thinking is not free floating; it is constrained by a few conditions. One of those conditions is that there is no thinking without acting. And there is also no acting without thinking. Mises writes: “Reason and action are congeneric and homogeneous, two aspects of the same phenomenon” (Ibid., p. 42). If it is quite clear that action cannot be detached from thoughts, it is important to stress that, to Mises, thoughts cannot be detached from action either. It is an action that makes us able to distinguish ourselves from the external world and, as such, that makes our ability to think as autonomous beings susceptible to affect the world possible: “In acting, the mind of the individual sees itself as different from its environment, the external world, and tries to study this environment in order to influence the course of events happening in it” (Ibid., pp. 35–36). This is why the category of action must be understood as the ultimate foundation of our understanding and, consequently, as the most basic category of social epistemology (Ibid., p. 8). Human knowledge is constrained by human action; epistemology by praxeology (Hoppe 1995, pp. 20, 81). In other words, action is an a priori of knowledge. If one’s goal is to obtain some understanding of the social world, then one must necessarily take a closer look at the concept of action and at what it means to act. Once the praxeological foundation of human knowledge is established, it is easier to understand Mises’s conception of rationality. Indeed, this concept can be directly deduced from the a priori of human action (Huusen 1989, p. 127). As a consequence, in his work, Mises always conceives of reason in terms of its links to action. In Human Action, Mises ([1949] 1998, pp. 18–19) states: Human action is necessarily always rational. The term “rational action” is therefore pleonastic and must be rejected as such. When applied to the ultimate ends of action, the terms rational and irrational are inappropriate and meaningless. The ultimate end of action is always the satisfaction of some desires of the acting man. Since nobody is in a position to substitute his own value judgments for those of the acting individual, it is vain to pass judgment on other people’s aims and volitions. No man is qualified to declare what would make another man happier or less discontented. The critic either tells us what he

164

C. Robitaille and R. Leroux

believes he would aim at if he were in the place of his fellow; or, in dictatorial arrogance blithely disposing of his fellow’s will and aspirations, declares what condition of this other man would better suit himself, the critic.

It is worth unpacking this definition. Rationality is once again inextricably linked to the category of action here. This means that Mises gives a special meaning to rationality. Indeed, contrary to, for instance, Vilfredo Pareto’s narrow and exclusive definition which can be deduced from his distinction between “logical” and “non-logical” actions, the former being any action which can be objectively linked to the attainment of its goal and the latter being any other action (Pareto 1917, p. 76), Mises’s rationality must be understood as being broader and more inclusive. Even so-called “nonlogical” actions are considered rational so long as they are indeed actions. No proper social investigation can describe an action “non-logical” because the ends sought are considered irrational or even because the means are deemed improper to reach a given end. To understand why Mises conceives of rationality in such a broad manner, it is important to address another insight that can be deduced from the a priori of action, i.e., Menger’s subjective theory of value. Indeed, as the definition presented above indicates, the only ultimate end that can be deduced from the fact of human action is “the satisfaction of some desires.” Adding any further specification would immediately make us leave the realm of transhistorical knowledge about action and enter either that of historical knowledge or that of normative evaluation. The former would require, as a consequence, further specifications about historical actors and their empirical context to assess what goals were indeed pursued and the latter would, according to Mises, make us leave the realm of science altogether to enter that of taste and personal ethics. But when treating of action as such, the truth is that social scientists can only consider the chosen end to be something which the actor considers valuable in order to “remove uneasiness” (Mises 1962, p. 35). What specific ends are perceived by some actors to remove uneasiness in their own lives cannot be judged scientifically. Mises hence deduces the most basic postulate of the Mengerian theory of value from human action: to act is to employ means toward ends in order to make one’s life better, and the fact that an actor does value such-and-such ends is proof that this actor thought these ends were worth pursuing in order to remove some uneasiness in their lives. There are hence no “non-logical” or “irrational” end to pursue from the point of view of a value free social scientific investigation. Ends are “non-rational” (i.e., cannot be analyzed by reason), but the pursuit of any end is always rational in the sense that an actor will choose to pursue goals perceived as removing pain or increasing pleasure.

Rationality, Praxeology, and History: The …

165

This, as discussed earlier, is often accepted by classical political economists; we cannot judge an actor’s ends as they are merely given by the human passions and hence escape the realm of rational discussion. As the old adage stipulates, de gustibus non est disputandum. However, and this is where his definition of rationality becomes more singular when contrasted to classical political economy, Mises argues that there are also no “non-logical” or “irrational” means used to achieve these ends. Mises ([1949] 1998, p. 20) writes: When applied to the means chosen for the attainment of ends, the terms rational and irrational imply a judgment about the expediency and adequacy of the procedure employed. The critic approves or disapproves of the method from the point of view of whether or not it is best suited to attain the end in question. It is a fact that human reason is not infallible and that man very often errs in selecting and applying means. An action unsuited to the end sought falls short of expectation. It is contrary to purpose, but it is rational, i.e., the outcome of a reasonable—although faulty—deliberation and an attempt— although an ineffectual attempt—to attain a definite goal. The doctors who a hundred years ago employed certain methods for the treatment of cancer which our contemporary doctors reject were—from the point of view of present-day pathology—badly instructed and therefore inefficient. But they did not act irrationally; they did their best. It is probable that in a hundred years more doctors will have more efficient methods at hand for the treatment of this disease. They will be more efficient but not more rational than our physicians.

Hence, contrary to the conception of rationality of most social scientists before him, every choice of means is considered to be rational. Indeed, the ex post failure of a procedure employed does not mean that an actor did not believe, based on his knowledge and circumstances, that it would succeed ex ante. On the contrary, the very fact that the actor employed this method (and not another) indicates that he believed it would likely succeed in attaining the goal sought. Otherwise, another method would have been used. As a consequence, every choice of means is rational. Even the choice of denying the rationality of action is rational: However one twists things, one will never succeed in formulating the notion of “irrational” action whose “irrationality” is not founded upon an arbitrary judgment of value. Let us suppose that somebody has chosen to act inconstantly for no other purpose than for the sake of refuting the praxeological assertion that there is no irrational action. What happens here is that a man aims at a peculiar goal, viz., the refutation of a praxeological theorem, and that

166

C. Robitaille and R. Leroux

he accordingly acts differently from what he would have done otherwise. He has chosen an unsuitable means for the refutation of praxeology, that is all. (Mises [1949] 1998, p. 104)

Based on the above, we see once again that Mises conceived of rationality as something which can simply never be detached from human action. Rationality is instrumental (i.e., can only be considered as a reflection about means and not about ends) (Mises [1949] 1998, p. 173; Fillieule 2014, p. 99), utilitarian (i.e., it always seeks to remove uneasiness by satisfying subjectively valued ends) (Mises [1949] 1998, p. 878), and subjective (i.e., based on the actor’s very own information and social and cognitive contexts) (Ibid., p. 21). The first two aspects are taken from classical economics, but Mises attempts to ground them more solidly by changing their status from that of mere “assumptions” to that of “praxeological deductions.” The last one broadens considerably the concept when compared to what had been done by his predecessors. There are some advantages from the social scientific point of view to use such a broad definition of rationality. Indeed, instead of dismissing as irrational actions or beliefs encountered in our study of social phenomena which do not conform to our very own cognitive or valuational framework, approaching them as enigmas that can eventually be understood by investigating the cognitive and valuational contexts of the actors involved may allow us to provide a better explanation of the intriguing phenomena in question. Mises’s example above of the methods used by physicians to treat cancer over time illustrates quite clearly the scope of this conception of rationality; it can allow us to obtain a better understanding of initially perplexing phenomena such as “why were physicians so ineffective a hundred years ago?” without simply dismissing their methods as the result of unfathomable irrational forces. To sum up, Mises’s conception of rationality as a praxeological deduction is thus instrumental, utilitarian, and subjective. We have seen that its “subjective” aspect renders it broader than usual classical conceptions as it leads social scientists to consider the instrumentality of the actor’s thinking, not from their points of view or against objective criteria on cause and effect, but from the point of view of the actor’s singular social and cognitive context. Likewise, it is broader in the sense that the expected “utility” gained from the performance of the action is thought, again, not based on the points of view of social scientists or on any objective, quantifiable, or material conception of self-interest, but based on the chosen goals of the actor. In this sense, Mises writes: “Economics does not assume or postulate that men aim only or first of all at what is called material well-being. Economics, as a branch of the more

Rationality, Praxeology, and History: The …

167

general theory of human action, deals with all human action, i.e., with man’s purposive aiming at the attainment of ends chosen, whatever these ends may be. To apply the concept rational or irrational to the ultimate ends chosen is nonsensical” (Mises [1949] 1998, p. 880). Hence, whatever the end chosen (whether it is linked to material well-being or to goals that logically imply making the actor materially poorer, such as self-sacrifice), the individual acts in accordance with his own self-interest (broadly conceived) because choosing to pursue that goal demonstrates an ex ante preference for it being achieved over any other competing attainable goal.

2

Rationality in the Analysis of Historical Problems

These remarks about rationality as a praxeological category are meant to constrain theoretical analyses in the social sciences. But Mises argued that there are two complementary manners to study the social world: theory (praxeology) and history: “The fullness of reality can be mentally mastered only by a mind resorting both to the conception of praxeology and to the understanding of history” (Mises [1949] 1998, pp. 642–643). Obviously, according to him, the theoretical remarks about rationality made above must constrain any historical analysis, as they constitute necessary transhistorical deductions from the a priori of human action (Mises 1942, pp. 249–250). However, to paraphrase Mises in Kantian terms, if history is indeed blind without theory, praxeological concepts are empty without historical specifications. This means that the concept of rationality as a praxeological category constrains any historical analysis but requires further empirical specifications when used to understand specific social phenomena. Hence, interpretations of historical data merely using the constraints provided by praxeological laws and the laws of the natural sciences remain insufficient. In the realm of historical understanding (and this includes sociology in Mises’s mind, see Robitaille 2019), then, we must attempt to understand what Mises called “thymological” aspects to specific actions. To Mises, “thymology” is the study of human motivations and their historical origins. It is the specific study of the emergence of “ends” in the “means-ends” relationship that constitutes action. Indeed, if ends must be taken as given to praxeological analysis because it looks at the transhistorical structure of action as such (Rothbard 2011, p. 35), it is, on the contrary, the task of thymology to build a “catalogue of human traits” (Mises [1957] 2007, p. 274) aimed at delineating “what we know about human value judgments,

168

C. Robitaille and R. Leroux

the actions determined by them, and the responses these actions arouse in other people” using experience “either from introspection or from intercourse with other men, from our acting in various interhuman relations” (Ibid., p. 312). Thymology thus constitutes a set of additional tools of investigation, in addition to praxeology and descriptive history, for social scientists to understand historical social phenomena. But what is this “catalogue of human traits” and how can we use it for the study of concrete social phenomena? Here, Mises is less explicit than when discussing economic theory or praxeology. A good indication, however, of what Mises had in mind can be inferred from the structure of his Theory and History (1957). Indeed, right after introducing the discipline of thymology, Mises discusses the importance of ideal types in the historical understanding of empirical actions and motivations (Mises [1957] 2007, pp. 315–320). It seems that, according to Mises, the “catalogue of human traits” resulting from the introspective and interactionist bases of thymological knowledge is a set of ideal types which can be used in order for social scientists to gather a better understanding of concrete historical social phenomena. For the study of history, then, in addition to using praxeological constraints for their analyses, social scientists can use profitably the comprehensive methods of Verstehende sociology. In particular, the sociology of Max Weber was perceived by Mises as being of relevance for the study of historical issues. Mises ([1933] 2003, p. 79) writes indeed: Weber was one of the most brilliant figures of German science of the twentieth century. He was a pioneer and trail blazer, and coming generations will have enough to do to make his heritage intellectually their own and to digest and elaborate it. That he was an historian and an investigator into the logical character of history does not mean that he failed with regard to the problems which the period presented and which he undertook to treat. His field was just that of history, and in this field he did far more than his share. And finally, if it is possible today to approach the logical problems of sociology with better conceptual tools, this is primarily due to the work that Max Weber devoted to the logical problems of history.

In spite of these uncharacteristic high praises and their implications for Mises’s theory of rationality applied to the study of historical problems, Mises had some sharp criticisms to address to the famous German sociologist. Indeed, if Mises thought of Weber as a great general historian, he also thought

Rationality, Praxeology, and History: The …

169

that his understanding of praxeological sciences such as economics was unsatisfactory. In an essay called “Sociology and History,1 ” dedicated entirely to addressing the problems with Weber’s method, Mises deals with two main issues: the application of ideal types to the study of theoretical (praxeological) economics and Weber’s conception of rationality. On economics being based on ideal types, Weber (2012, p. 124) mentions that: The abstract economic theory offers us an example of those syntheses that are usually called “ideas” of historical phenomena. It presents us with an ideal image of what goes on in a market for goods when society is organized as an exchange economy, competition is free, and action is strictly rational. This mental image brings together certain relationships and events of historical life to form an internally consistent cosmos of imagined interrelations. The substance of this construct has the character of a utopia obtained by the theoretical accentuation of certain elements of reality.

This “utopia” is created in the researcher’s mind by proceeding to “a one-sided accentuation of one or a number of viewpoints and through the synthesis of a great many diffuse and discrete individual phenomena (more present in one place, fewer in another, and occasionally completely absent), which are in conformity with those one-sided, accentuated viewpoints, into an internally consistent mental image” (Weber 2012, p. 125). Hence, economics is, according to Weber, a science which derives its laws and concepts from the construction of ideal types. It abstracts from historical data aspects of phenomena and accentuates them in order to create “pure” types which can then serve as tools to understand other phenomena. In particular, it abstracts the aspects of phenomena that are in accordance with how omniscient actors would act in given situations (Weber 2012, p. 369) and which are then rebuilt in a systematic fashion. As such they are the result of an “intellectual construct” which “is heuristically useful for the analysis of actual action” because “[e]xperience shows that actual action has a certain tendency to approximate such purely rational action”. In other words, much like Pareto’s understanding of economics, Weber believes that economic theory builds concepts by “making its rational construction of ‘economic man’” 1 Let us note that this essay was written back in the 1920s when Mises called the theoretical study of human action “sociology” rather than “praxeology.” This fact is important, as later writings by Mises will term “sociology” the general historical investigations conducted in light of Weber’s method. As history, then, sociology no longer can be seen as a theory of human action (in the same sense that, for instance, the science of economics is considered theoretical, i.e., as providing transhistorical laws of human actions rather than historically contingent understandings of phenomena). See Mises (2009, pp. 106–107).

170

C. Robitaille and R. Leroux

(Weber 2012, p. 274). This conception of economic theory as the creation of ideal typical structures for historical understanding can be seen as a compromise between Carl Menger’s epistemological insights and those of his German historicist rivals in the Methodenstreit. It follows Menger in the attempt to use an abstract and deductive method for the understanding of social action but it also contests Menger’s claim that economic theory is composed of categories revealing the real general essence of phenomena (Hennis 1991; Yagi 1997, 2011; Robitaille 2022). This is precisely what Mises contests in Weber’s analysis. Indeed, as we have seen above, Mises argued that theoretical economics is a set of praxeological laws that condition the possibility of any particular historical economic phenomena. As such, these laws are real as we cannot conceive of any empirical economic phenomenon without these economic laws being true. True enough, Mises ([1933] 2003, p. 84) says, “[c]oncepts are never and nowhere to be found in reality”, but “they are obtained through reflections having in view the comprehension of what is contained in each of the individual phenomena taken into consideration.” Hence, although a social scientist cannot find a “value” or an “action” as categories in empirical reality (as they are, indeed, general categories), he or she knows for sure that any empirical actor acts based on values. To Mises, it is not an “utopia” or an “ideal image” or a “one-sided accentuation” to talk about categories and laws of action; those are necessary categories or relations that apply to any empirical social phenomenon. The above discussion of Mises’s theory of rationality hence stands as applicable to every empirical analysis of action and, as such, contradicts Weber’s conception of economic theory as using ideal typical constructions such as an accentuated form of rational action. The second criticism that Mises addresses to Weber relates to his typology of social action. Indeed, in his Economy and Society, Weber ([1921] 2019, pp. 101–103) famously divided social action in four types, respectively guided by what he calls “purposive rationality,” “value rationality,” “affect” and “tradition.” This is all rooted in Weber’s quest for meaning . It excludes all behaviors that are merely the result of biological or physiological unconscious reactions, as those reactions do not have meaning and can only be causally described. The meaning of an action is thus rooted either in the utilitarian quest for satisfaction, the desire to conform to higher values, the emotions that impel us, or the social habits that we take for granted and do not put into question. Those motivations to act (purpose, value, emotion, tradition) are, according to Mises, problematic in two ways. First, from a praxeological point of view, the distinction between purposive and value rationality is deemed confused by Mises ([1933] 2003, p. 90). Mises thought indeed that

Rationality, Praxeology, and History: The …

171

social theory (as opposed to social history) aimed at providing transhistorical laws and categories of action. As values are always determined by a complex process that can only be understood historically, we must, as the marginalists have argued before him, consider value as subjective in order to elaborate a transhistorical theory of action. If value is subjective from the point of view of a social theorist, then we cannot, as social scientists, make a clear-cut distinction between “value” rationality and “purposive” rationality. There are no laws of valuation other than that action requires valuing: “Adherence to received views of life or to political convictions is an end like any other, and like any other it enters into the rank order of values. Weber here falls into the old misunderstanding which the basic idea of utilitarianism repeatedly encounters, namely, that of regarding as an ‘end’ only values that are material and can be expressed in money” (Ibid.) Hence, valuing an increase in one’s standards of living is no different from valuing a religious, deontological, or political view from the perspective of social theory. Actions based on either type of motivation preserve the same common logic (Ibid., p. 91). Second, the other two social actions (affective and traditional), not considered as “rational” in Weber’s typology, are nevertheless the result of thought processes (provided they are not merely unconscious reactions) and, hence, cannot be considered otherwise than as rational according to Mises (Ibid., p. 92). There is nevertheless a way to conciliate Mises and Weber on this topic, although the following analysis must be slightly more speculative here. The previously mentioned fact that Mises discusses ideal types right after his description of thymology suggests that ideal types are considered to be the tools of this historical discipline. As such, considering Weber’s typology of social action as historical specifications about motivations to act can be partially consistent with Mises’s own conception of rationality. Mises thought that rationality as a praxeological category constrains any historical understanding of specific actions. As such, even affective or traditional actions must be considered rational. But it requires little work to see that there are often good reasons to pursue affective passions or to resort to traditional ways of doing things. As rational types of social actions, then, they can be considered as historical specifications of motivations to act. It can indeed often be useful for the historian (or thymologist) to classify specific actions as traditional or affective. Likewise, distinguishing between actions aiming at improving one’s wealth and actions aiming at conforming to one’s conception of morality can be useful for the historian who aims at gathering a specific understanding of specific events. Interpreting Weber’s typology as part of Mises’s thymology has the potential to conciliate both authors and to better understand how

172

C. Robitaille and R. Leroux

the praxeological concept of rationality can be operationalized in the study of history. To sum up, Mises’s conception of human rationality has a few implications for social scientists. First, it requires that the social scientist who is confronted with an intriguing phenomenon does not ask himself or herself whether or not the phenomenon is the result of rational action, but rather how it is (Di Iorio & Di Nuoscio 2014, p. 133). Otherwise, the social scientist is not objectively analyzing society but is merely imposing his or her own value judgements on the situation he or she aims to explain. Second, it requires the use of ideal types constrained by praxeological types in order to obtain a specific understanding of historical events and of historical evolution. In brief, theoretical rationality constrains historical rationality, but historical rationality as a set of ideal types is what allows social scientists to obtain a specific understanding of social phenomena. Mises’s conception of value makes it impossible for us to know exactly why actors came to value what they valued without some amount of speculation about which human trait was at play. It is here that Weberian ideal types, as accentuation of such specific human traits, can become helpful.

3

Applications

Let us now discuss two applications of Mises’s theory of rationality in his own work. The first one, Mises’s argument for the impossibility of economic calculation under socialism, shows how the theory of rationality can be used in order to make objective claims about how actors can achieve various goals or not depending on the characteristics of the economic system in which they act. The second one, Mises’s discussion of polylogism, is a more straightforward application of his theory of rationality which demonstrates how various claims made by social scientists can be objectively debunked on its basis.

Economic Calculation Under Socialism and Rationality One of the most famous arguments Mises presented us with in his work relates to the impossibility of economic calculation under the context of socialism. The argument is quite simple to understand. The market economy is characterized by a price system which allows entrepreneurs to make decisions concerning the production process. By comparing anticipated monetary revenues and anticipated monetary costs associated to the use of any alternative factor of production and to the use of any alternative method of

Rationality, Praxeology, and History: The …

173

production for a consumer good, the entrepreneur can make an educated economic decision with respect to which factors to buy and which methods to use (Mises [1922] 1951, p. 115). The entrepreneur can then judge whether he has succeeded in choosing a good method and set of factors once he receives monetary profit or loss. A monetary profit means he was able to allocate factors of production in such a way as to serve some of the most desired needs of consumers. A loss means he failed. Economic calculation thus allows scarce productive resources to be put in the hands of the entrepreneurs who expect to be best able to satisfy consumer needs (Ibid., p. 120). Obviously, errors are possible (Ibid., p. 131), but economic calculation is the condition of possibility for the process which tends to bring scarce factors of production into the hands of those who are best at evaluating the consumers’ desires. Now, to be able to make these decisions (buying or selling factors of production), entrepreneurs need to own these factors. Indeed, if individual entrepreneurs do not own them, they cannot decide to exchange them in a market. If entrepreneur A thinks he can make a higher profit using factor X than entrepreneur B, under market conditions, he could offer B a good amount of money (which will be compensated by the anticipated revenues) to acquire it. If B does not expect to make as much money by using X himself in a process of production, he will simply sell it to A. But if B does not own factor X, he cannot sell it to A. Indeed, public property makes it impossible to have a market for factors of production (Piano & Boettke 2018, pp. 222–223). Hence, economic calculation (anticipated revenues compared to anticipated costs from using factors of production) is simply impossible and, as a result, the process which allows factors to move into the best hands to satisfy consumer needs simply does not occur. In a monetary economy, it makes rational production impossible (Mises [1922] 1951, p. 119). The argument is simply that, if our end is to allocate factors of production effectively, then there is no way to do so under socialism. Capitalism is the only system which allows such a goal to be attained . It seems that Mises uses the term “rational” in a more specific sense here. In particular, this critique of socialism rests on the fact that rational economic calculation is impossible when the means of production are publicly owned: “[A] society based on public property is not viable because it does not permit monetary calculation and thus rational economic action” (Mises [1929] 2011, p. 55). But we have seen in the first section of this chapter that Mises thought every action had to be conceived as rational. Surely, this includes actions under the conditions of socialism. The contradiction is only apparent here. Of course, any historical action performed under socialism remains rational in the sense that an actor would still try to find the best means to attain

174

C. Robitaille and R. Leroux

subjectively chosen goals. The point is that, conceptually, economic calculation is the rational tool of capitalism; it is this intellectual device which allows entrepreneurs to make decisions to achieve their end (making profit) and, consequently, the ends of others in society (obtaining valued goods). Under socialism, this rational tool is inexistent (Salerno 1990, p. 45), and decisions will be based on other, ineffective, rational tools. If the goal of production is to make profit by satisfying the consumers most urgent needs, and if the only means to do so is economic calculation, then socialism, by making this means impossible, also makes rational production (i.e., production based on a reflection on how to best use factors of production) impossible. A central planner who has as a goal to distribute factors of production where they are most likely to be used in the production of the most highly desired consumer goods does not have any rational tool at his disposition to do so in the same way that one has no rational tool at one’s disposition to successfully inform a cat about Plato’s cave. Under the specified conditions, their goals are impossible, which means that their reflection on how to achieve them is doomed to fail. Subjectively, then, if one were to try to distribute factors of production according to their best use under socialism or if one were to try to teach a cat about Plato’s cave, their action could still be understood as rational . Perhaps they are not aware that allocating factors of production effectively necessitates private ownership or that cats cannot understand abstract philosophical arguments. However, comparing reasoning processes under the conditions of capitalism with those under the conditions of socialism reveals this first goal to be an impossible goal; comparing reasoning in a world where cats have the ability to understand abstract arguments with reasoning in a world in which this is not the case also reveals that, in our world, the second goal is an impossible goal. Hence, those being impossible goals under the specified conditions (i.e., socialism and a world where cats are not equipped with the ability for abstract thinking), any rational reflection on how to best achieve them will fail. We can understand the thought process behind attacking windmills; we can also know for a fact that, in a world in which giants do not exist, attacking windmills is a vain attempt at vanquishing them. To conclude, Mises’s broad understanding of rationality can of course help us in understanding why people seek to achieve impossible goals. But, as we have just seen, it can also allow us to make objective claims about the possibility of such goals. It is because of the nature of reflection processes (the rational thinking) of an entrepreneur under market conditions (and its constraints) as compared to that of a social planner under socialism (and its own constraints) that we can determine how decision-making under socialism

Rationality, Praxeology, and History: The …

175

will never be able to reach the same desired outcome as decision-making under capitalism. By looking at how an entrepreneur can reason and by comparing this with the possible reasoning of a social planner, we discover that they simply cannot reason in the same way. To Mises, rational production defined as the reasoning process allowing society to allocate scarce factors of production where they are most likely to be transformed into highly desired consumer goods is not possible under socialism.

Polylogism and Rationality Another important aspect of Mises’s work is his attack on polylogism. Three types of polylogism were attacked by Mises: racial polylogism (Mises [1944] 2011, pp. 163–166), Marxian polylogism (Mises [1949] 1998, pp. 75ff ), and cultural polylogism. These three types of polylogism are rooted on the same mistake: the belief that the very process of thinking is strictly determined by group affiliations. It is quite simple to see, in light of our previous discussion, that if rationality is a universal category, then multiple logics of action are obviously inconceivable. This is different from looking at the logic of action given specified social contexts (e.g., our previous discussion of the logic of action under capitalism as opposed to that under socialism). That different cultures, for instance, provide actors with different social contexts which orient action differently is not contested. The claims which Mises opposes are that the logic of action itself depends on culture, race, or class and that scientific discoveries are dismissed on the basis that they are “bourgeois” or “Jewish.” Although this is not our topic, there are certainly commonalities between these lines of thinking and contemporary polemics on whether or not we should dismiss without consideration theories that are considered as being “white,” “patriarchal,” or “heteronormative.” All of this, Mises claims, is based on fundamental misunderstandings of reason and science. Claims can be true or false, but the ultimate criterion to determine this does not rest in any way on the particular group(s) the person making the claim is affiliated with. Marxian polylogism is the form of polylogism most discussed in Mises’s work. Indeed, Marx, Engels, and their followers argued that the way we think is shaped by the material interests of our social class (Marx and Engels [1846] 1998, p. 42). There is bourgeois thinking (rooted in the justification for capitalism) and proletarian thinking. But capitalism, as a bourgeois ideology, is “erroneous”; it contains fundamental contradictions which, ultimately, will lead to its own destruction. Correct thought is hence only proletarian thought, as once capitalism has been overthrown by the working

176

C. Robitaille and R. Leroux

class, contradictions will disappear and the classless society will emerge. The proletarian logic is correct whereas the bourgeois logic is confused. Mises claims there are many problems with this view. Against this, he indeed argues: “A consistent supporter of polylogism would have to maintain that ideas are correct because their author is a member of the right class, nation, or race” (Mises [1949] 1998, p. 76). Surely, however, not all proletarians agree with Marx. This problem has been addressed to some extent by Marxists. Indeed, noting that proletarians did not always follow their class interest, Engels ([1893] 1968, pp. 689–693) came up with the idea of “false consciousness,” attributing this behavior to the fact that the bourgeoisie created a superstructure which aimed at hiding behind false justifications the exploitation of proletarians. Proletarians who do not adhere to Marxism are hence simply confused by various attempts by the bourgeois to justify capitalism through pseudo-scientific ideology. Classical political economy was thus considered an “ideology,” a superstructural tool of domination. The problem Mises ([1949] 1998, p. 77) notes with this idea is that acting on the basis of a false theory will always lead to less successful action than a true one: “The fact that the practical application of a theory results in the outcome predicted on the basis of this theory is universally considered a confirmation of its correctness. It is paradoxical to assert that a vicious theory is from any point of view more useful than a correct one.” Hence, it is problematic to think that capitalists would choose to act on the basis of knowledge which they fully know to be incorrect; that they would organize production and society on the basis of a theory they know to be false. Hence, the only remaining salvation for polylogism, from the Marxist point of view, would be to argue that capitalists are not aware of the incorrectness of their doctrine. But then, argues Mises, they should be able to point out which aspects of the doctrine are invalid. Mises argues that Marxists failed to do so, as they merely accuse their adversaries of deceiving their readers to gain economic or political advantages (Ibid., p. 79; Mises 1990, p. 206, 257; [1944] 2011, p. 164). Here, Mises uses his broad definition of rationality to argue that all human beings, whatever their social class, share the same mental structure. Action and its categories logically imply that our reasoning processes are structured in the same way. How individuals use this common logic will of course vary depending on circumstances, but its general structure remains the same for all. The mere fact that some bourgeois (like Marx or Engels themselves) can claim to determine what the proletarian logic is better than most proletarians themselves seems to indicate that Marxian polylogism is indeed incorrect (Mises [1949] 1998, p. 84). Moreover, as Mises mentions, the bourgeoisie

Rationality, Praxeology, and History: The …

177

and the proletariat are not homogeneous classes; specific thought processes are unique and individual. They do share, however, a common logic: every individual thinks in terms of means and ends. As a consequence, Mises argues that the laws of economics which are based on this idea are not “ideological” tools of the bourgeoisie to justify their exploitation of the working class, but social scientific truths. Mises addresses racial polylogism slightly differently. Indeed, as the argument is based on biology (rather than on history), claims about racial differences in terms of the structure of thinking must be addressed using different arguments. Against the idea that the logical structure of the human mind differs by race, Mises ([1949] 1998, p. 85) indicates: “The Asiatics and the Africans no less than the peoples of European descent have been eager to struggle successfully for survival and to use reason as the foremost weapon in these endeavors. They have sought to get rid of the beasts of prey and of disease, to prevent famines and to raise the productivity of labor.” Once again, it is merely different motivations or different specific ends that determined perceived differences in ethnic groups: “[T]hese considerations refer only to the motives determining concrete actions, not to the only relevant problem of whether or not there exists between various races a difference in the logical structure of mind” (Ibid., pp. 85–86). Consequently, people of all ethnic groups share the same praxeological structure of the mind: they think in terms of means and ends. Their divergences are merely in terms of the specific ends sought, the specific motivations they have, and the varying cultural and social influences on the choice of means to achieve their ends. This explains, for Mises (Ibid.), the group differences in terms of relative success in increasing standards of living. Similarly to Marxian polylogism, Mises ([1944] 2011, pp. 163–164) says, racial polylogism emerged in order to discredit a group’s achievements without properly analyzing it. The Nazis, for instance, discredited anything deemed to be “Jewish science” (Ibid., p. 165). The problem is that there is no such thing: the correctness of any truth-claim is not determined by a person’s race but by whether or not it stands rational consideration. Finally, there is also cultural polylogism. This is the idea that individuals from “primitive societies” or from more “primitive times” have a different mind structure than individuals living in more civilized societies or civilized times. This idea was famously promulgated by anthropologist Lucien Lévy-Bruhl ([1923] 2018). In particular, after considering a vast amount of ethnographic data, Lévy-Bruhl mentions: “To the primitive mind his preconnections, which are no less imperative than our need to trace every phenomenon back to its causes, establish without any possibility of doubt,

178

C. Robitaille and R. Leroux

the direct transition from such-and-such a sense-impression to such-and-such an invisible force. Or rather, it is not even a transition, for that term, though suited to our discursive operations, does not exactly express the primitive’s mental functioning, which seems more like direct apprehension or intuition” (Ibid., pp. 59–60). This makes it quite clear that Lévy-Bruhl sees the very structure of the “primitive’s” mind as so distinct from that of the civilized mind that it would be impossible for the latter to even fathom the operations of the former other than by analogy. To Mises, although he thought Lévy-Bruhl’s work to be very valuable, the same problem as with other forms of polylogism is present in this line of thinking. He indicates, directly confronting similar arguments, that: “Ethnology shows only that the conclusions arrived at by the reasoning of primitive peoples are different from those which we arrive at and that the range of things primitive peoples are accustomed to think about is different from the circle of our intellectual interests. When primitive man assumes magical and mystical connections where we assume connections of a different kind, or where we find no connection at all, or when he sees no connection where we do see one, this shows only that the content of his reasoning differs from that of our own, but not that his reasoning is of a different logical structure from ours” (Mises [1933] 2003, p. 112). This stance, once again, is rooted in Mises’s view that all of humanity shares a common way to think; in other words, a common rationality. It is how it is used by particular individuals integrated in particular social contexts which matters to the social scientist. And discovering how rationality is used in particular instances is impossible without recognizing that we indeed all share a common structure of thought.

4

Conclusion

In brief, this chapter aimed at presenting Mises’s theory of rationality. It has first shown how rationality was conceived as a praxeological category which, in Mises’s mind, reinforced classical economics’ instrumental and utilitarian conception of rationality by grounding it in a rigorous proof stemming from the nature of human action as such. We have also seen that subjective rationality, deduced from Menger’s theory of value, added generality to the concept. Second, this chapter discussed how Mises’s theory of rationality could be applied in historical analyses by comparing it to Weber’s theory. This chapter then suggested a possible way to conciliate Mises’s and Weber’s respective conceptions by arguing that the latter could be seen as a specification of the former for the study of concrete historical problems. In other

Rationality, Praxeology, and History: The …

179

words, one could see Weber’s typology of social action as part of this “catalogue of human traits” characterizing Mises’s thymology when applied to the study of how historical actors used reason. Finally, this chapter provided two examples from Mises’s work in which his theory of rationality allowed to make objective claims about how actors can achieve various goals or not and to debunk various myths about the structure of the human mind. This theory of rationality proves to be useful for the study of a large range of theoretical and historical social problems and hence stands as an important contribution to the methodological individualist tradition in social research.

References Demeulenaere, P. (1997). Histoire de la théorie sociologique. Paris: Hachette. Di Iorio, F., & Di Nuoscio, E. (2014). Rethinking Boudon’s cognitive rationality in light of Mises’s apriorism and Gadamer’s hermeneutics. Journal des économistes et des études humaines 20 (2), 129–142. Engels, F. ([1893] 1968). Engels to F. Mehring in Berlin. In Karl Marx and Frederick Engels selected works (pp. 689–693). London: Lawrence & Wishart. Fillieule, R. (2014). Raymond Boudon as social theorist: A comparison with Ludwig von Mises. Journal des économistes et des études humaines 20 (2), 91–128. Hennis, W. (1991). The pitiless ‘sobriety of judgement’: Max Weber between Carl Menger and Gustav von Schmoller – The academic politics of value freedom. History of the human sciences 4 (1), 27–59. Hoppe, H.-H. (1995). Economic science and the Austrian method . Auburn: Ludwig von Mises Institute. Hume, D. ([1739] 1978). A treatise of human nature. Oxford: Clarendon Press. Huussen, G. M. (1989). Mises and the praxeological point of view. Journal of economic studies 16 (2), 121-133. Lévy-Bruhl, L. ([1923] 2018). Primitive mentality, L. A. Clare (Trans.). London: Routledge. Marx, K., & Engels, F. ([1846] 1998). The German ideology. Amherst: Prometheus Books. Menger, C. ([1871] 2004). Principles of economics, J. Dingwall & B. F. Hoselitz (Trans.). Auburn: Ludwig von Mises Institute. Menger, C. ([1883] 1985). Investigations into the method of the social sciences with special reference to economics, F. J. Nock (Trans.). New York and London: New York University Press. Mises, L. v. ([1922] 1951). Socialism: An economic and sociological analysis, J. Kahane (Trans.). New Haven: Yale University Press. Mises, L. v. ([1929] 2011). A critique of interventionism. Auburn: Ludwig von Mises Institute.

180

C. Robitaille and R. Leroux

Mises, L. v. ([1933] 2003). Epistemological problems of economics, 3rd Ed., G. Reisman (Trans.). Auburn: Ludwig von Mises Institute. Mises, L. v. (1942). Social science and natural science. Journal of Social Philosophy & Jurisprudence 7 (3), 240–253. Mises, L. v. ([1944] 2011). Omnipotent government: The rise of the total state and total war, B. Bien Greaves (Ed.). Indianapolis: Liberty Fund. Mises, L. v. ([1949] 1998). Human action: A treatise on economics. Auburn: Ludwig von Mises Institute. Mises, L. v. ([1957] 2007). Theory and history: An interpretation of social and economic evolution. Auburn: Ludwig von Mises Institute. Mises, L. v. (1962). The ultimate foundation of economic science: An essay on method . New York: D. van Nostrand. Mises, L. v. (1990). Economic freedom and interventionism: An anthology of articles and essays, B. Bien Greaves (Ed.). Indianapolis: Liberty Fund. Mises, L. v. (2009). Memoirs, A. Oost-Zinner (Trans.). Auburn: Ludwig von Mises Institute. Pareto, V. (1917). Traité de sociologie générale, vol. I. Paris: Payot & Cie. Piano, E. E., & Boettke, P. J. (2018). Rationality of the individual and rationality of the system: A critical examination of the economic calculation problem over socialism. In G. Bronner & F. Di Iorio (Eds.), The mystery of rationality: Mind, beliefs and the social sciences (pp. 219–237). Cham: Springer. Robitaille, C. (2019). Ludwig von Mises, sociology, and metatheory. Quarterly journal of Austrian economics 22(2), 242–270. Robitaille, C. (2022). Carl Menger on theory and history. Cosmos + Taxis: Studies in Emergent Order and Organization 10 (5–6), 61–74. Rothbard, M. N. (2011). Praxeology as the method of the social sciences. In Economic controversies (pp. 29–58). Auburn: Ludwig von Mises Institute. Salerno, J. T. (1990). Ludwig von Mises as social rationalist. Review of Austrian Economics 4, 26–54. Weber, M. ([1921] 2019). Economy and society: A new translation, K. Tribe (Trans.). Cambridge: Harvard University Press. Weber, M. (2012). Collected methodological writings, H. Henrik Bruun (Trans.). Abingdon: Routledge. Yagi, K. (1997). Carl Menger and the historicism in economics. In P. Koslowski (Ed.), Methodology of the social sciences, ethics, and economics in the newer historical school. From Max Weber and Rickert to Sombart and Rothacker (pp. 231–258). Heidelberg: Springer. Yagi, K. (2011). Austrian and German economic thought. From subjectivism to social evolution. London: Routledge.

Methodological Individualism, Scientific Explanation, and Hermeneutics Enzo Di Nuoscio

Émile Durkheim liked to repeat that it is not worth devoting a single hour to the social sciences if they are not useful to society. And the use of the sciences, even those that study social phenomena, cannot be other than promoting the advancement of empirical knowledge, by proposing theories that explain prima facie incomprehensible phenomena, on the basis of rigorous logical and empirical controls. From Weber to Boudon, from Mill to the “Austrian School of Economics”, the tradition of methodological individualism (MI) has been united precisely in this objective: to ensure the scientific character of social theories, avoiding any form of ideology or interpretation that subordinates the search for truth to other ends, such as trying to condition public opinion. The thesis I would like to propose in this chapter is that this goal can be better achieved if MI uses L. von Mises’ theory of praxeology and if is linked to two research traditions often considered incompatible: the nomologicodeductive model and hermeneutics. More specifically, I believe that the notion of rationality of action and its unintended consequences can find in these two streams of study formidable heuristic tools for the definition of a E. Di Nuoscio (B) University of Molise, Campobasso, Italy e-mail: [email protected] Luiss Guido Carli University, Rome, Italy

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 N. Bulle and F. Di Iorio (eds.), The Palgrave Handbook of Methodological Individualism, https://doi.org/10.1007/978-3-031-41512-8_10

181

182

E. Di Nuoscio

general model of MI that ensures a rigorous scientific explanation of social phenomena.

1

Explaining Human Action: Praxeology and “Good Reasons”

The explanation of human action and macro-social phenomena are the two main objectives of MI. Raymond Boudon’s theory of rationality, which aims to explain an action by reconstructing the “good reasons” that led to it, represents one of the best contributions to methodological individualism. The thesis defended in this essay is that this methodological proposal strengthens its explanatory capacity through Mises’ praxeology, which proposes an a priori investigation of human action. More precisely, I believe that there is an epistemological complementarity between the context of action studied by Mises and the context of reasons studied by Boudon, which allows us to define a more efficient model to explain action. Referring to a purely logical plan, Mises proposes an a priori analysis of the notion of action, a conceptual and not an empirical analysis, which aims at establishing the universal and invariable characteristics of an action as such, leaving aside its historical manifestations. It wants to identify truths of reason, not truths of fact, i.e. truths which are not derived from experience, but which are true ex definitione. This is the deductive reasoning that Mises (1949: 10 ff ) proposes in making a conceptual analysis of the notion of action: a. Homo sapiens is necessarily homo agens: he acts to eliminate a state of dissatisfaction. Consequently, he is not omnipotent and infallible; his rationality is necessarily limited; b. The means by which he eliminates his dissatisfaction is action, which can be considered as a problem-solving hypothesis. It is intentional by definition because it is intentionally oriented towards the solution of a problem. Action is therefore the product of a limited rationality: it is always a subjectively satisfying and not objectively optimal solution; c. Each individual aims to eliminate a state of dissatisfaction in the way he or she considers best under the circumstances. From the actor’s point of view, action is always rational . An individual cannot act other than by considering his reasons as the best. For Mises, praxeology is a “science of means” and not of ends, and once he acts, the actor has chosen precisely that action as the best means from his point of view in that situation;

Methodological Individualism, Scientific Explanation …

183

d. In choosing the best solution, the actor always chooses the means he considers the most efficient and appropriate in view of his goal; from this point of view, the action is always, lato sensu “economic”: it is a question of saving the means by considering the end; e. Since the action aims at a goal, that of eliminating a state of dissatisfaction, it has an intrinsically teleological nature. But choosing a means to an end means determining a cause by considering the effects it implies. This means that the teleological means-ends relation presupposes a nomological cause-effect relation. There is no teleology without causality (Mises 1949: 22 ff ). The action can be considered the cause of the expected effects that constitute its goal.1 These necessary and invariable characteristics of action are very valuable information for empirical explanation: a. If action is always intentionally aimed at the elimination of a state of dissatisfaction, then the researcher, in order to explain the action, must reconstruct how the actor perceives this dissatisfaction; in other words, he must formulate hypotheticals about how the individual interprets the situation of uncertainty he finds himself in; b. If action is always rational, then the social scientist knows that at the basis of all action there is a rational strategy (implicit or explicit), which must be conjecturally reconstructed in order to explain an action. This is only possible through an analysis of the situation in which the action takes place: by trying to understand how this situation is perceived by the actor; c. If the action is always lato sensu ‘economic’, the researcher must reconstruct the (non-monetary) cost–benefit calculation which led the actor, once he had chosen the aim of the action, to prefer precisely this means (the action) rather than the other alternatives available to him; d. If the action is teleologically oriented on the basis of the principle of causality, the researcher must establish the nomological knowledge that the individual used in order to choose his action strategy. For obvious logical reasons, only one or more “laws” can in fact “link” cause and effect, i.e. means (action) and end (goal).

1

This thesis had already been advanced by M. Weber (2012 [1903–1906]: 101–2), for whom “the categories of ‘aim’ and ‘means’, without which there is in general no teleological ‘thinking’, contain, as soon as one proceeds scientifically with their aid, conceptually formed nomological knowledge, i.e. concepts and rules developed in relation to the category of causality. Indeed, there is indeed a causal relationship without teleology, but there can be no teleological concepts without causal rules”.

184

E. Di Nuoscio

As can be seen, praxeology determines the universal logical form of action that allows the social sciences to reconstruct the accidental empirical content that changes for each particular action. One could say that empirical research in the social sciences begins where praxeology ends. In fact, every action is characterised by an order, by invariable logical characteristics that allow it to be explained. If we have the essential empirical information, we can reconstruct even the reasons for an action that seems the most incomprehensible, such as, for example, the actions of the Callipids, of whom Herodotus speaks, who would eat the bodies of their dead because they considered them sacred. This is made possible precisely because these actions have logical characteristics identical to those of any other action. Mises’ praxeology thus allows us to give a valid answer to the old question: what do different actions have in common that allows us to understand them regardless of cultural distance? There is thus a complementarity between the aprioristic perspective of Mises and the empirical perspective of Boudon. One could speak of a Mises-Boudon Model : a. Mises identifies the invariable logical structure of action as such, Boudon the empirically variable content of each action; b. Mises’ analysis identifies the a priori categories of action, Boudon’s identifies the a posteriori content of an action, i.e. the reasons; c. Mises’ praxeology refers to the logical context of action, whereas Boudon’s rationality of “good reasons” refers to the sociological context of reasons; d. Praxeology allows the identification of the permanent characteristics of action, which are crucial for reconstructing, by conjecture and refutation, the accidental empirical contenue, namely the reasons, which change from action to action. The sociologist can therefore empirically search for the different typologies of reasons precisely because he knows that these reasons exist and that they can be of any nature (Di Nuoscio 2018: 123 ff ). We can say with Boudon that the individual is homo oeconomicus, homo sociologicus, homo ethicus, etc., precisely because he is above all homo agens, acting on the basis of invariable logical categories that necessarily guide his actions. It is very interesting to note that Mises and Boudon are fighting the same epistemological battle: highlighting the limits of the Rational Choice Model and demonstrating that self-interest is only a type of motivation that inspires action. Like Boudon (2007: 37ff ), and contrary to what is often misunderstood, Mises is a harsh critic of those who equate rationality with

Methodological Individualism, Scientific Explanation …

185

“economic rationality” or “instrumental rationality”. A “human being determined exclusively by economic motives”, he writes, is the “fantoche of the homo oeconomicus” of the utilitarians who has never had a “correspondent in reality” (Mises 1949: 64). Everything explained by economic reasons, in the manner of the classical economists and utilitarians, is a matter of the content of action according to Mises; whereas the permanent attributes of behaviour, starting with its “economic” nature, are to be seen in their “formal sense” and stripped of all “material content” (Mises 1949: 64).

2

Methodological Individualism and the Explanation of Action and of Its Consequences

One of the most interesting aspects of the contemporary epistemological debate on IM concerns causal explanation. To the question of whether individualistic explanation is related to causal explanation we have had opposing answers, which can be summarised by the theses of Raymond Boudon, who gave a positive answer to this question, and those of the leading analytical sociologists, who gave a negative answer. For Boudon, the logic of individualist explanation consists in the reconstruction of the causal link between reasons-actions and action-social phenomena (Di Nuoscio 2004: 163–191). “The causes of action, beliefs, attitudes of social actors must be sought in the meaning they have for them, or more precisely in the reasons why they have adopted them” (Boudon 2005: 29).2 This means that in order to explain an action, the sociologist must use the Nomological Deductive Model, and therefore consider reasons as explanans and action as explandum. He will therefore have to reconstruct these “initial conditions”, represented by knowledge, objectives, resources, representations of the consequences, perception of the situation, without which the actor, ceteris paribus, would not have acted. And among the “initial conditions”, particularly relevant from the causal point of view, will be the reconstruction of the “rational strategy” which leads the individual to choose the goal of the action.3 2 “Motivations and determining beliefs, Hempel and Oppenheim argued, must be classified among the preceding conditions of a motivational explanation, and in this respect there is no difference between a motivational and a causal explanation” (Hempel, Oppehneim 1948: 143). 3 As Weber writes, “the goal is the representation of an effect which becomes the cause of the action” (Weber 2005 [1903–1906]: 11). Equally necessary is the use of NDM to explain the intentional and, above all, unintentional genesis of macro-social phenomena. In this case, the actions and the programmed and unplanned ways in which they combine must be considered as explanans (initial

186

E. Di Nuoscio

Elster, Hedström, Ylikoski, and other analytical sociologists, on the other hand, believed that the individualistic explanation is not also a causal explanation,4 but an explanation by “mechanisms”. Their thesis is that because in the field of social phenomena the same cause is compatible with different effects, and the same effects with different causes, it is impossible to use laws like “If A, then always B”, while it is necessary to refer to mechanisms such as “If A, then sometimes B” or “If A, then sometimes C, D, and B” (Elster 1998: 49), and that explanation is not based “by evoking universal laws, or by identifying statistically relevant factors, but by specifying mechanisms that show how phenomena are brought about” (Hedström 2005: 22). The laws of covering-law explanations, Hedström (2005: 5) explains, “are typically seen as perfectly general and without exceptions while the mechanisms of mechanism-based explanations are not”. The inadequacy of the nomological-deductive model to explain social phenomena is thus related to the nature of covering laws, which are in fact nonexistent in the social sphere, where we are always dealing with nondeterministic explanations (Hedström, Ylikoski 2010: 12). These objections of analytical sociologists are based on a misconception of the Popper-Hempel Model, since they believe that this model assumes only deterministic laws and is therefore only suitable for explaining deterministictype phenomena, which are inexistent in the social world. Such assumption is not acceptable and ignores the “indeterministic turn” in causal explanation of which the main theorists of the application of the nomological-deductive model in the social sciences have been protagonists (Popper, Hempel, Nagel) (Di Iorio, Di Nuoscio 2022: 57–60).

conditions), while the systemic effects, from the most banal to the most complex, as explanda. A very clear example of an individualistic and nomological explanation is that proposed by Weber (2005 [1904–1905]: 101 ff ) of the genesis of capitalism. Weber answers the three key questions for the methodological individualist: (i) who caused the phenomenon: the Calvinist entrepreneurs who avoided accumulating profits and reinvested them; (ii) why they acted in this way: because, on the basis of a religious belief, they regarded professional success as a signum salutis, etc.; (iii) how one moves from the micro level of individual actions to the macro level of macro-social events: the simple addition of a very large number of actions of Calvinist entrepreneurs who behaved in this way gave rise to what Weber called the “spirit of capitalism”. Weber’s theory is built around a dual causal explanation: (i) when he states that Calvinist entrepreneurs reinvested their profits because they believed that professional success was an indication of their salvation, he proposes a causal relationship between these actions and the reasons behind them, based on an implicit law, according to which “anyone who wants to expand his econo-nomic activities tends to reinvest his profits”; (ii) when he makes explicit the link between the individual actions of entrepreneurs and the “spirit of capitalism”, he uses a law that can therefore be made explicit: “The large scale diffusion of economic activities by individual entrepreneurs gives rise to the system of economic relations called ‘capitalism’”. 4 They, beginning as Elster (1998), have held that this model of explanation is to be rejected because it is committed to determinism. “The deductive-nomological model is not applicable because the deterministic social laws that it presupposes do not exist” (Hedström 2005: 22).

Methodological Individualism, Scientific Explanation …

187

Indeed, if we examine in detail the model of causal explanation gradually developed by its theorists, it is not difficult to realise that it assumes that— both in the natural and social sciences—explanations do not necessarily require deterministic laws (see Di Iorio 2015: 149 ff.). And it is certainly no coincidence that the leading proponents of this model, such as J. S. Mill, Weber, Popper, and Hempel, have been theoreticians of MI and staunch critics of any form of historicist determinism even in the social–historical field (see Di Nuoscio, Di Iorio 2019). For defenders of causal explanation, the laws used in the social sciences are “law-like sentences” (Hempel 1965: 458), “law less-than-universal-laws” (Scriven 1959: 464), i.e. expressions that simulate laws but that do not explain all cases, which have also been defined as “common sense maxims” (Nagel 1961: 389) and “judgement[s] of possibilities” (Boudon 1984: 239). These are statements that explain, barring exceptional conditions, what individuals do tendentially (normally, without exceptional conditions), when they find themselves in certain situations (Di Nuoscio 2018: 43–48).5 The author who, perhaps, first and best of all understood how the explanation of social phenomena can be causal and based on nondeterministic laws, was Max Weber himself. In the “Roscher and Knies and the Logical Problems of Political Economy” (Weber 2012 [1903–1906]: 113 ff.), he argued that in the social sciences causal relationships are established on the basis of “rules of experience”, which are only probabilistic and have been acquired through the trial-and-error learning process that has produced our common sense. They are common sense maxims about everyday life, that are not universal because they are often, but not always, valid. They refer to a “nomological knowledge” “describing the concrete causal links” between typical actions and typical situations (Weber 2012 [1903–1906]: 5).6 The Weberian “nomological knowledge” should not be confused either with accidental generalizations (such as e.g. “all my students are for democracy”) which, unlike laws, are mere descriptions of facts, mere accounts of experience, describing definite sequences of facts after they have been verified. Nor should they be confused with trends (such as “with industrial revolution the number of proletarians tends to increase”), which are equally not laws, but a finite sequence of causally related events, that can only be explained through laws. Since they are only a finite number of singular predicates,

5 Examples of laws used implicitly or explicitly are: “Everyone at the beginning (of a war) is more determined” (Thucydides); “Individuals tend not to change what they hold sacred” (Dilthey); “You are more certain to win if your enemies are separated” (Mommsen). 6 Regarding the nature of laws in both the natural and social sciences, see S. Berry, “On the Problem of Laws in Nature and History: A Comparison”, in History and Theory, n. 4, 1999: 121 ff.

188

E. Di Nuoscio

historical trends, and empirical generalisations cannot be projected beyond the cases they describe and therefore have no explanatory or predictive power; consequently, they can be used to support counterfactual hypotheses and thus explanations in terms of the D-N model (Di Nuoscio 2004: 227–231). In History of Italy, Benedetto Croce proposes an historical explanation that is very useful for understanding the explanatory function of nomological common sense knowledge (Di Iorio, Di Nuoscio 2022: 59–60). Here is how the Italian historian and philosopher explains the defeat of the Republican Party in Italy in the 1880 elections 1880s: Republicans included in their election manifestos – see, for instance, their 1882 manifesto – universal suffrage, an allvolunteer army, the abolition of the Law of Papal Guarantees, confiscation of ecclesiastical property, regional autonomies, convocation of the Constitutional Assembly and other things that they should have known – and surely many of them knew – were impossible. As a con-sequence, the Republican Party became increasingly small. (Croce 1966 [1927], translated by the author: 70)

Croce explains the explanandum, the collapse of the Republican Party’s vote, by causally linking it back to “initial conditions” through a nomological rule of common sense, implicit in his theory, which can be explicated as follows: “voters tend not to give their consent to political parties that advance a program they consider unfeasible”.7 As pointed out by Hempel (1965: 39 ff ), because of their trivial nature, probabilistic common sense laws about typical human behaviours are almost always used in an implicit manner. So, only if there is some form of nomological coverage can any mechanism have explanatory power. Consequently, explanation by mechanisms, which analytical sociologists insist, makes sense only if it is based on causal relations, necessarily involve probabilistic laws and “nomological knowledge”.

3

The Primacy of “Situational Analysis”

The epistemological nature of the laws used in social theories led Popper, methodological individualist and theorist of the unified method, to develop the notion of “situational analysis”. Since many laws in the social sciences are “aproblematic”, “trivial”, “common sense”, “unimportant” (Popper 1982 [1974]: 201–204), the nomological component of sociological explanations

7

For more examples of this kind, see Opp (2013).

Methodological Individualism, Scientific Explanation …

189

is both logically necessary and methodologically marginal , i.e. for the construction of explanatory theories. These are “deductively incomplete” explanations (Hempel 1965: 252) in which, due to the absence of universal falsifiable and not falsified laws, there is a lack of what for Hempel and Oppenheimer is one of the characteristics of scientific explanation: the logical deduction of the explanandum from the explanans (Hempel, Oppenheimer 1948: 137). This gives rise to the possibility of cases of “explanatory undetermination” (Hempel 1965: 287), i.e. different explananda compatible with the same initial conditions: explicanda or, viceversa, different explananda compatible with the same fact to be explained. Faced with the possibility of having a plurality of explanations compatible with the event to be explained, the sociologist can only rely on what Popper calls “situational analysis”, i.e. on the most precise reconstruction possible of that unique interweaving of typical aspects which, in his opinion, constitutes the causal context that generated the explanandum. The more detailed the situational investigation, the more the number of alternative explanations that are incompatible with each other and compatible with the explanandum is reduced (Di Nuoscio 2018: 59). An example of this can be found in Thucydides when, in The Peloponnesian War, he proposes to explain the following fact: why, despite the invasion of Athens by the Lacedemonians, and despite the fact that the Athenians “accused Pericles of having convinced them to wage war”, Pericles was reelected. Here is the explanation offered by Thucydides: Not long afterwards, however, according to the way of the multitude, they again elected him general and committed all their affairs to his hands, having now become less sensitive to their private and domestic afflictions, and understanding that he was the best man of all for the public necessities. For as long as he was at the head of the state during the peace, he pursued a moderate and conservative policy; and in his time its greatness was at its height. When the war broke out, here also he seems to have rightly gauged the power of his country. He outlived its commencement two years and six months, and the correctness of his previsions respecting it became better known by his death. (Thucydides 1950 [IV sec. a.C.]: 153)

In this passage, and in the following pages of The History of Peloponnesian War, the Greek historian offers a detailed reconstruction of the causal context that produced Pericles’ reelection. The fact to be explained, the reelection of Pericles, is linked to this context by an implicit law that can be formulated as follows: “when citizens are convinced of the qualities of a ruler and the goodness of his policy, they are ready to change their minds

190

E. Di Nuoscio

and give their consent”. However, it is clear that such a nomological cover, although logically necessary, is hardly sufficient for the explanation. Instead, the identification of “initial conditions” through situational analysis is fundamental. And the more detailed this investigation is, the less incomplete the “explanatory project” is. Clearly, this explanandum is compatible with a plurality of alternative explanations: the Athenians may have been forced to reelect Pericles, or they may have been systematically deceived in some way. It is on the basis of a very detailed situational investigation that Thucydides rules out these other explanations: he points out that Pericles was highly regarded by the Athenians for his policies in the past (even reporting Pericles’ full speech to the Athenians). He shows how Pericles managed to convince the Athenians to reelect him, despite the difficult time he was having in Athens. It is thus an explanation that does not indicate a “necessity”, but a “possibility”, an “objective possibility” as Weber would have said. We are thus confronted with what Popper defined as a “primacy” of situational analysis, which gives “special importance” in the social sciences to the “principle of rationality”, without which it would be impossible to reconstruct the situation (Popper 2002 [1957]: 89–93). It should be pointed out that in “situational analysis”, and more generally in the explanation of action, the “principle of rationality” and the “principle of causality” are complementary. The “principle of rationality” which, in its Popperian formulation, states that “the various individuals or agents act in a suitable or appropriate manner, that is, according to the situation as they have seen it” (Popper 1994 [1964]: 105), can only act as a “principle of order” in human affairs insofar as the actor uses the principle of causality to choose the means he deems appropriate to solve his problem. In turn, the principle of causality makes sense in human affairs, because there are individuals rationally inclined to solve problems. And these attempts at a solution would be unthinkable without the use of the category of causality. It is therefore precisely the use of the principle of causality that enables humans to be rational, i.e. to carry out behaviours that have a logic that can be reconstructed by a well-informed observer. Thus, situational analysis is nothing more than a particular application of the unified method in the social sciences, which, as Boudon has repeatedly stated, has “the scientific ambition […] to follow the procedures obeyed by all the sciences” (Boudon 2011), because “scientific explanation applies indistinctly to natural and social phenomena” (Boudon 2010: 112).

Methodological Individualism, Scientific Explanation …

4

191

Explaining Human Action: Hermeneutics and Praxeology

Gadamer’s “hermeneutic circle” can be considered a complementary methodological perspective to the Mises’ praxeology, Boudon’s explanation by “good reasons” and the nomological-deductive model. Each of these methodological perspectives can make a decisive contribution to developing an effective logic of individualistic explanation.8 If one understands it on the basis of its real epistemological nature, it is not difficult to realise that Gadamerian hermeneutics is not anti-scientific, but on the contrary allows for an improved scientific explanation of human action. If human action is the response to a problematic situation through a “rational strategy”, then it can be considered as the result of an interpretative circuit that is established between the actor’s tabula plena mind and the reference context; between the social actor who becomes an “interpreter” and the situation in which the action develops, which for the subject is nothing more than a “text” to interpret. The action can therefore be considered as a hypothesis of interpretation, proposed on the basis of a “pre-understanding”, and therefore of a horizon of expectation, which changes for each individual. More precisely, just as Gadamer’s interpreter tries to solve the problem of the meaning of a text by resorting to the solution hypothesis that is the “interpretation project”, the social actor tries to solve the problem—to get out of a situation of uncertainty—by this hypothesis that is the action. Just as the interpreter proposes a new interpretation in the case where the first one is falsified by the text or by the context, so the social actor proposes a new action in the case where the first one does not achieve its objective and is therefore falsified by the situation in which it is introduced. Like any interpreter, the social actor faces the problem to be solved with his pre-understanding, which comes from tradition (values, social norms, nomological knowledge, etc.) (Lachmann 1991: 135 ff ).9

8

The interpreter, Gadamer explains, approaches the text with a tabula plena mindset, rich in “preunderstanding”, which derives from tradition. On the basis of this knowledge, he formulates a first “interpretive sketch” about the meaning of the text, which checks on the “text” and the “context”, where “context” means all the information that is “around the text” and that can be useful for checking this interpretive hypothesis. In case of conflict between a part of the text or the context and the “interpretive project”, the interpreter is forced to formulate a new hypothesis, to be submitted to the same control. This cognitive procedure is the “hermeneutic circle”, which for Gadamer is a “virtuous circuit”, in principle infinite, because it advances knowledge of a text by eliminating interpretations that conflict with the text itself and with the context (Gadamer 1976 [1960]: 180ff ). 9 I have developed this thesis in Di Nuoscio (2014).

192

E. Di Nuoscio

Following Popper’s scientific problem-theory-critical method , it can be argued that hermeneutics promotes a scientific explanation of human action because it helps the methodological individualist to: a. Reconstruct how the problem arises, i.e. how the problematic situation at the origin of the action is perceived, and to identify which of the actor’s preunderstandings came into play and which of his unfulfilled expectations led to the situation of uncertainty; b. Understand why this action was chosen as the best solution. This choice is linked to the interpretation of the traces, elements, information and, more generally, of the reference situation of the action; c. Reconstructing how the actor controls the results of the action and, possibly, reorients his strategy. This phase of the action is linked to the individual’s capacity to interpret the new problematic situation created by the action and its consequences, in order to control the achievement of the objective. Hermeneutics thus makes a decisive contribution to the explanation of human action, complementary to the perspective of praxeology. Praxeology identifies the culturally invariable logical structure of human action, establishing that action is ex definitione, intentional, rational, “economic”, causal, and teleologically oriented. Hermeneutics, on the other hand, from a historical-philosophical perspective, insists on the culturally and subjectively variable content of human action (expectations, objectives, perception of the situation, subjective knowledge), trying to reconstruct the interpretative strategy by which each individual uses his or her own pre-understanding to formulate the hypothesis of a solution to a problem represented by the action (Di Nuoscio 2014: 77ff ). Gadamer’s hermeneutics, but also Ricoeur’s (1986: 190ff ), tells us that ideas and actions must be seen as a “text” that can be understood through the cultural categories of the interpreter. In this hermeneutic process, the social scientist can ensure that this particular text represented by actions can assert its “otherness”, can be understood by a sociological explanation without being “violated” by an interpretation that is not open to truth, precisely because he or she does not doubt that there is an order in action. It is a logic common to all actions (known a priori thanks to praxeology) which makes the greatest variety of behaviour possible and understandable, and which pushes the interpreter to look for empirical information which, if found, allows the sociologist to explain what the problem was to be solved and why, according to the actor, this behaviour is the best possible solution.

Methodological Individualism, Scientific Explanation …

193

Praxeology thus refers to the context of action, identifying a universal, culturally invariant logic of behaviour, while hermeneutics refers to the context of reasons, contributing to the reconstruction of the “rational strategy” which is the basis of all action. The theorems of praxeology are thus a decisive component of that “pre-understanding” which allows the interpreter to identify an order even in apparently incomprehensible actions and then to attribute to them, by conjecture and refutation, a meaning. An interpreter who ignores the theorems of praxeology would risk explaining everything in terms of tradition, absolutizing the social and cultural-historical conditioning of human action, even going so far as to decree the irrationality of a behaviour. A praxeologist who ignores, on the contrary, the historical-interpretive dimension of action and its empirically variable content, risks making false generalisations and would not be able to explain action because of the impossibility of reconstructing the subjective “good reasons” that dictated it (Di Nuoscio 2009: 175). Thus, praxeology and the “hermeneutic circle” are complementary: the logical and a priori categories of the former are a prerequisite for the empirical knowledge dealt with by the latter. These two theoretical perspectives can contribute to defining an effective epistemological model for the explanation of human action (Di Iorio 2015: 151).

5

Causal Explanation Presupposes Hermeneutics

Causal explanation presupposes hermeneutics for two reasons: a. Because, as we have seen, the formulation of a theory is linked to the scholar’s ability to interpret, with his or her own “scholarly mind”, a certain problematic context of which the theory itself is intended to be the solution; b. Because the decision to search for a causal order in physical or historical and social truth is a rigorous interpretation linked to the knowledge (laws, models, and rules of common sense) that form a fundamental element of the researcher’s Vorverständnis. In formulating the explanatory hypothesis, the researcher chooses an interpretive perspective that allows him to reconstruct the causal links in order to find the solution to the problem. After all, linking cause and effect, or as Charles S. Peirce would say, antecedent and effect, is a very important task.

194

E. Di Nuoscio

Peirce would say antecedent and consequent is a way of realising an ‘interpretative project’ of the consequent itself. This project is realised by formulating “interpretive sketches”, which are gradually abandoned and replaced by more appropriate hypotheses in case they contradict the facts (Gadamer 1976 [1960]: 189). This strong link between hermeneutics and causal explanation is all the more evident if we consider that the vast majority of explanations in the social sciences, and most of those in the natural sciences, are the result of “knowledge by traces”, linked to the identification of “clues”, allowing us to return to the formulation of the hypothesis. Faced with a problem to be solved, the ability of doctors, detectives, historians, sociologists, economists and, in general, of all those who—in the sciences, as in everyday life—do not have the possibility of experimentally reproducing the phenomenon to be explained, lies in knowing how to recognise the traces that the facts have left over time and to be able to transform them into useful clues for formulating the hypotheses of explanation. This “in-court paradigm” is an explanatory strategy that follows what Peirce defined as the logic of abduction: We observe a surprising fact C; But if A were true, C would be explained as a normal fact; Therefore, there is reason to suspect that A is true. “Therefore—explains Peirce—A cannot be deduced abductively […], if all its content is not already present in the premise. If this were true, C would be explained as a normal fact” (Peirce 1935–1966 [1903], vol. VI, § 5.189). This procedure resembles in every way the Popperian problem-theorycritical method and emphasises that “circumstantial explanation” is a “reverse reasoning” (Copi, Cohen 2014 [1964]: 395), “from experience to hypothesis” (Peirce 1935–1966 [1903]: § 2.755). As Sherlock Holmes was fond of saying, “we begin at the end and arrive at the beginning”. The intrinsically interpretive nature of explanation is obvious: it is only possible to trace causally from known effects to unknown causes through an adequate interpretation of a problematic context of “clues”, made possible by the scientist’s ability to recognise the traces that an event has left in time and to transform them into information important for solving the problem. It is on the basis of their own knowledge—in which the “particular knowledge of time and place” (Hayek 2019 [1945]: 279), “tacit” or “unexpressed” (Polanyi 1958: 104), often plays a decisive role—that a sociologist, like a detective or a doctor, interprets an event or a fact as a clue that can put him or her on the right track to finding a solution. If it is true, as Gadamer

Methodological Individualism, Scientific Explanation …

195

argues, that “whoever begins to interpret a text, always implements a project” and that “even the most immediate meaning of the text manifests itself only insofar as it is read with certain determined expectations” (Gadamer 1976 [1960]: 208), then it can be argued that the identification of traces must be seen as the implementation of a “project of meaning”, carried out under the pressure of a problem, which will determine the choice of the interpretive perspective.10 Moreover, just as the “interpretive sketch” is not reduced to a mechanical deduction of the meaning of the text from the text itself, nor to a mechanical projection of the interpreter’s anticipations of the text, but presupposes a content of meaning from a pre-understanding and through a process of conjecture and refutation which must be considered as a “virtuous circle”, so the abductive one is an ampliative inference. In Peirce’s scheme, it is not reduced (like induction) to a mere statement that “something is”, nor (like deduction) that “something must be”, but assumes that “something can be”, introducing a “new idea” and subjecting it to falsifying control. It can therefore be concluded that even this version of causal explanation described by the abductive method, so widespread in scientific research, cannot be separated from the hermeneutical approach (Di Nuoscio 2014: 93) (Di Iorio, Di Nuoscio 2014).

6

Hermeneutics as “Objective Knowledge”

Contrary to those who have generally associated the hermeneutical perspective with “renunciation of truth”, if not with nihilism, this notion of Gadamer’s “hermeneutical circle” is thus an important contribution to scientific explanation (Albert, Antiseri 2006: 60). From his reflections on aesthetics, in fact, Gadamer’s entire theory of interpretation rests on the idea of the “otherness” of the text, i.e. of a possible objectivity of interpretation, in reaction to what the German philosopher himself defines as the “hermeneutic nihilism” of arbitrary interpretations.11 The meaning of the text, for Gadamer, is a problem to be solved, and the “interpretive sketches” are hypotheses of solution that have to be revised in

10

From a hermeneutical point of view, for traces, whatever they may be, what Ricoeur writes about human action is valid: they are “an open work” to “whoever knows how to read”, whose meaning “is waiting for new interpretations which decide their meaning” (Ricoeur 1986: 189). 11 It is precisely the possibility of producing an objective interpretation (admitted by the former and denied by the latter), which represented one of the most important points of contrast between Gadamer and Derrida (in Michelfelden, Palmer 1989).

196

E. Di Nuoscio

case of a “clash” with the “text” and/or the “context”. “Any correct interpretation - writes Gadamer (1976 [1960]: 288) in his masterpiece, not by accident entitled Truth and Method - must free itself from the arbitrariness and limitations deriving from unconscious mental habits, by looking at the things themselves”. Submitting to one’s object in this way is not a decision that the interpreter makes once and for all, but: the permanent and ultimate task. […] He who wants to understand always makes a projection. He projects a meaning of the whole in advance as soon as a first meaning of the text appears. But such a meaning in turn only manifests itself if one already reads the text with certain expectations as to its precise meaning. The understanding of what is there always takes place through the elaboration of such a prior projection, which must of course be constantly revised as one progresses in the penetration of the meaning. (Gadamer 1976 [1960]: 288)

On the objectivity of this interpretation, Gadamer’s text is even more explicit: He who tries to understand [a text] is exposed to errors deriving from presuppositions that are not confirmed by the object. The permanent task of com-understanding is the elaboration and articulation of correct and adequate projects, which, as projects, are anticipations that can only be validated in relation to the object. There is no other ‘objectivity’ here than that of the confirmation that a pre-opinion can receive in the course of its elaboration. What distinguishes inadequate hypotheses from the fact that, as they develop, they are in-sufficient? Understanding only arrives at authentic possibility if the presuppositions from which it starts are not arbitrary. There is thus a positive sense in saying that the interpreter does not access the text simply by remaining within the framework of the presuppositions already present in him, but rather, in the relationship with the text, tests the legitimacy, i.e. the originality and validity, of these presuppositions. (Gadamer 1976 [1960]: 288)

Basically, Gadamer explains, a correct interpreter from an epistomological point of view: must be prepared to be told something by the text […]. An understanding regulated by a methodical consciousness must apply itself not merely to give free rein to its own anticipations, but to go so far as to become aware of them in order to control them and to start from the things themselves in order to arrive at the understanding. […] A consciousness trained in hermeneutics must therefore be open from the outset to the otherness of the text. But such receptivity does not presuppose ‘neutrality’ in terms of content, nor, above all, the erasure of oneself, but includes the appropriation that brings out the

Methodological Individualism, Scientific Explanation …

197

reader’s preconceptions and personal prejudices. It is a matter of realizing that one has been warned, so that the text itself presents itself in its otherness and thus acquires the possibility of opposing its truth, which is of substance, to the reader’s pre-opinion. (Gadamer 1976 [1960]: 290)

Hermeneutics is thus a theoretically infinite “virtuous circle” that proceeds through historiographical conjectures and textual falsifications. Such a cognitive process differs in almost no way from Popper’s “problem-theorycriticism” method, which describes a cognitive process that evolves towards “conjecture and refutation. Popper’s method of trial-and-error elimination and Gadamer’s Zirkel des Verstehens, explains Dario Antiseri12 : are the same thing described in two different languages: both describe the same event and prescribe the same procedure. Gadamer is the philosopher of prejudice: the interpreter does not approach the text with a Baconian mind void of memory; and Popper has always insisted that we live constantly in the centre of what he calls ‘a horizon of expectations’. Gadamer argues that there is a ‘collision’ that occurs before a text […]; and Popper speaks of problems as disappointed expectations. For Gadamer, it is the ‘collisions’ between our prejudices and the text that make hermeneutics aware of their own prejudices and trigger the chain of increasingly appropriate interpretations; for Popper, it is precisely the problems that make us realise the expectations we had (perhaps unconsciously) and which now no longer hold, and thus trigger the search, the search for new and stronger hypotheses to challenge. Popper argues that the objectivity of theories is one and the same with their falsifiability, i.e. with the fact that they are subject to intersubjective and repeatable empirical control of their consequences; Gadamer asserts that anyone who wants to understand a text must be prepared to be told something. (Antiseri 2001 [1981]: 130)13

12 This thesis finds significant confirmation in the very words of Gadamer, who said: “One element that connects me to Popper—not so much to Albert—is that he sees research as a process of discovering truth and, ultimately, the criterion of truth: the very process of research is revealed as the only possible way to find truth”; (in C. Grossner 1980 [1971], 335). And Popper in turn observed: “in truth, I am as far from positivism as Gadamer: in fact, I have discovered—and my critique of positivism is based on this—that the natural sciences do not operate positivistically, but essentially use a method that operates with ‘prejudices’, only that they eventually use new prejudices and prejudices that are subject to criticism, and expose them to rigorous criticism […]. What distinguishes me from Gadamer is a better understanding of the ‘method’ of natural science, a logical theory of truth and a critical attitude. But my theory is just as anti-positivist as his, and I have shown that the interpretation of texts (hermeneutics) functions with the same methods as the natural sciences”; ivi. 13 It is significant that Ricoeur himself has argued that Gadamer’s “hermeneutic circle” is not a “virtuous circle”, since it is subject to “validation procedures”, which include “invalidation procedures comparable to the criterion of falsifiability defined by K. Popper in his logic of scientific discovery” (Ricoeur 1986: 195).

198

E. Di Nuoscio

On the other hand, Gadamer himself, has made important exceptions to the “unified method”,14 stating that: The theory of attempt and error that Popper elaborates is not limited to the logic of specialized enquiry. For all the shortening and stylization of this scheme, it highlights a notion of logical rationality that goes far beyond the field of scientific inquiry and describes the basic structures of all rationality, even that of practical reason. […]. In the sciences themselves, a hermeneutical dimension as a factor in the maintenance and rooting of ownership - in the natural sciences as a dimension of paradigms and the relevance of one’s own frameworks of investigation. In the social sciences, a similar structure could be described as the self-transformation of social engineering into a social partner. In the historical sciences, finally, it is at work as the outgoing mediation of what once was, what is today, and what will be tomorrow’s row. (Gadamer 1985 [1967]: 165–6)

7

Science as Hermeneutic Knowledge

Therefore, interpretation is based on methodologically relevant rules and not on generic “meta-regulations of behavior”. Objectivity, understood not as certainty, but as the public controllability of interpretative hypotheses that are always revisable, is the principle on which Gadamer’s “hermeneutic circle” is based. Objectivity is possible precisely because the interpreter has a “horizon of expectations”, a web of “prejudices”, which represent the logical presupposition for the formulation of hypotheses. If it is true, as Nelson Goodman (1969: 168) wrote, readjusting a famous phrase of Kant’s, that “the innocent eye is blind and the virgin mind is empty”, and, as Popper (2002 [1963]: 101) said, that “there is no uninterpreted empirical basis”, then all forms of knowledge, even scientific knowledge, are based on interpretation linked to the interpreter’s preunderstanding. The scientist is above all an interpreter and his theories are only interpretative hypotheses of that text represented by natural phenomena or by actions or consequences of human actions. Hypotheses that must always be revised in case of falsifications. As Gadamer himself observed, Even in the natural sciences, the rootedness of scientific knowledge cannot avoid the hermeneutical consequences of the fact that what is called ‘given’

14

“The scheme of ‘formulation of the hypothesis’ and its ‘proof ’ - writes Gadamer - is valid for any kind of research, even in the humanities, even in philology” (Gadamer 1976 [1960]: 701).

Methodological Individualism, Scientific Explanation …

199

cannot be separated from interpretation. It is only in the light of interpretation that something becomes a fact, and it is only in the processes of interpretation that an observation can be expressed. […]. Thus, the hermeneutical understanding of something as something has been discovered even in what one calls oneself (in Michelfelden, Palmer 1989: 31). Scientific knowledge is thus possible thanks to the tabula plena mindset of the researcher-interpreter, who draws on the reserve of what Gadamer calls “prejudices” (Vorurteilen), to delimit an aspect of reality and try to explain it. On the basis of his own values and interests, the scientist chooses the object of investigation, and the perspective from which he will investigate it, in order to draw from his basic knowledge the indispensable tools (laws, models, ideotypes, even metaphysical hypotheses that favour explanation), to explain in a causal way the selected phenomenon. The scientist’s “pre-understanding”, which makes scientific investigation possible, is therefore composed above all of values and “nomological knowledge” which make possible, respectively, the Weberian “relation to values” and the use of problem-oriented “nomological knowledge”, without which it would not be possible to formulate any scientific theory (Di Nuoscio 2014: 86). “Science itself is a form of hermeneutics” (Eger 1993: 2)15 and thus the sociologist is also an interpreter whose research work can only benefit from the analysis of the interpretive procedure offered by hermeneutics. As Gadamer (1976 [1960]: 701) explained, The problem of the relevance of hermeneutics cannot be limited to the humanities. What are considered facts in the natural sciences are not all arbitrary quantities, but only the results of measurements that represent the answer to a question, a confirmation or refutation of a hypothetical. Hermeneutic reflection is tested everywhere in scientific practice, which is based on falsifiability. Approaching Popper’s thesis, Gadamer concludes that “all science has a hermeneutical component” (Gadamer 1996 [1985]: 427) and that the schema of “hypothesis formulation” and “proof ” “applies to all kinds of research” (Gadamer1976 [1960]: 701). Thus, “it is certainly right” to conceive, “with Popper”, scientific research “as a solution to problems by trial and error” (Gadamer 1996 [1985]: 427). Moreover, hermeneutics can be of additional help to “all sciences” because, Gadamer explains, “by making transparent the respective thematic pre-understandings of each science, it can uncover new problematic dimensions, and thus indirectly serve the methodological work” (Gadamer 1985 [1967]: 281).

15

The same thesis has been expressed by R.J. Bernstein (1983: 30).

200

E. Di Nuoscio

References Albert, H., Antiseri, D. 2006. L’ermeneutica è scienza? [Is Hermeneutics a Science?]. Soveria Mannelli: Rubbettino. Antiseri, D. 2001 [1981]. Teoria unificata del metodo [Unified Theory of Method]. Torino: Utet. Bernstein, R. J. 1983. Beyond Objectivism and Relativism. Science, Hermeneutics and Praxis. Philadelphia: University of Philadelphia Press. Boudon, R. 1984. La place du désordre [Transl. 1991. Theories of Social Change: A Critical Appraisal. New York: Polity Press]. Paris: PUF. Boudon, R. 2005. Teoria della scelta razionale e individualismo metodologico: sono la stessa cosa? [Rational Choice Theory and Methodological Individualism: Are They the Same Thing?] In Borlandi, M., Sciolla L. (a cura di) 2005. La spiegazione sociologica. Bologna: il Mulino. Boudon, R. 2007. Essais sur la théorie générale de la rationalité [Essays on the General Theory of Rationality]. Paris: PUF. Boudon, R. 2010. La sociologie comme science [Sociology as Science]. Paris: La Découverte. Boudon, R. 2011. Les deux grandes orientations de la sociologie [The Two Main Orientations of Sociology]. Paris: Académie des Sciences Morales et Politiques. https://academiesciencesmoralesetpolitiques.fr/2011/09/26/reflexions-surla-sociologie/. Copi, I. M., Cohen, C. 2014 [1964]. Introduction to Logic. Edinburgh: Pearson Education Limited. Croce, B. 1966 [1927]. Storia d’italia dal 1871 al 1915 [History of Italy from 1871 to 1915]. Bari: Laterza. Di Iorio, F., Di Nuoscio, E. 2014. Rethinking Boudon’s Cognitive Rationality in the Light of Mises’ Apriorism and Gadamer’s Hermeneutics. Journal des Économistes et des Études Humanines, 2, 129–142. Di Iorio, F., Di Nuoscio, E. 2022. On Situational Analysis and the Explanatory Power of Mechanisms: Analytical Sociology and the Deductive Nomological Model. Social Sciences Information, 61(1), 54–76. Di Iorio, F. 2015. Cognitive Autonomy and Methodological Individualism. New York: Springer. Di Nuoscio, E., Di Iorio, F. 2019. Filosofia della scienza e scienze sociali [Philosophy of Science and Social Sciences]. In P. Barrotta & E. Montuschi (eds.) La Filosofia della Scienza in Italia. Roma: Armando. Di Nuoscio, E. 2004. Tucidide come Einstein? La spiegazione scientifica in storiografia [Thucydides as Einstein? The Scientific Explanation in Historiography]. Soveria Mannelli: Rubbettino. Di Nuoscio, E. 2009. The Rationality of Human Action: Toward a Model MisesPopper-Boudon? In M. Cherkaoui & P. Hamilton (eds.) Raymond Boudon. A Life in Sociology. Oxford: The Bardwell Press.

Methodological Individualism, Scientific Explanation …

201

Di Nuoscio, E. 2014. Ermeneutica ed economia [Hermeneutics and Economics]. Soveria Mannelli: Rubbettino. Di Nuoscio, E. 2018. The Logic of Explanation in the Social Sciences. Oxford: Bardwell Press. Eger, M. 1993. Hermeneutics as an Approach to Science: Part I. Science & Education, 2. Elster, J. 1998. A Plea for Mechanisms. In P. Hedström & R. Sweldberg (eds.) Social Mechanisms. Cambridge: Cambridge University Press. Hedström, P., Ylikoski, P. 2010. Causal Mechanisms in the Social Sciences. Annual Review of Sociology, 36, 49–67. Hedström, P. 2005. Dissecting the Social: On the Principles of Analytical Sociology. Cambridge: Cambridge University Press. Gadamer, H.-G. 1976 [1960]. Verité et méthode [Truth and Method]. Paris : Seuil. Gadamer, H.-G. 1985 [1967]. Rhetoric, Hermeneutics and Critique of Ideology. In The Hermeneutics Readers. New York: Kurt Mueller-Vollmer. Gadamer, H.-G. 1996 [1985]. Natura e mondo. Ermeneutica e scienza della natura [Nature and the World. Hermeneutics and the Science of Nature]. In Verità e metodo [Truth and Method] 2, tr. it. Milano: Bompiani. Goodman, N. 1969. Languages of Art. Oxford: Oxford University Press. Grossner, C. 1980 [1971]. I filosofi tedeschi contemporanei tra neomarxismo, ermeneutica e razionalismo critico [Contemporary German Philosophers between NeoMarxism, Hermeneutics and Critical Rationalism] tr. it. Roma: Città Nuova. Hayek, von F. A. 1945. The Use of Knowledge in Society. In 2009 [1949]. Individualism and Economic Order. Auburn: The Mises Institute. Hempel, C. G. 1965. Aspects of Scientific Explanation an Others Essays. New York: McMillan. Hempel, C. G., Oppenheim, P. 1948. Studies in the Logic of Explanation. Philosophy of Science, 15(2), 135–175. Lachmann, L. 1991. Austrian Economics. A Hermeneutic Approach. In D. Lavoie (ed.) Economics and Hermeneutics. London: Routledge. Michelfelden, D. P., Palmer, R. E. 1989. Dialogue and Deconstruction. The GadamerDerrida Encounter. New York: State University of New York Press. Mises, L. von. 1949. Human Action a Tratise on Economics. New Haven: Yale University Press. Nagel, E. 1961. The Structure of Science. New York: Harcourt Brace & World. Opp, K-R. 2013. What is Analytical Sociology? Strengths and Weaknesses of a New Sociological Research Program. Social Science Information, 52(3), 329–360. Peirce, C. S. 1935–1966 [1903]. Pragmatism and Abduction. In Collected Paper, Cambridge MA: Harvard University Press. Polanyi, M. 1958. Personal Knowledge: Towards a Post-Critical Philosophy. Chicago: University of Chicago Press. Popper K. 1982 [1974]. Undended Quest. An Intellectual Biography. La Sale (Ill.): Open Court.

202

E. Di Nuoscio

Popper, K. 1994 [1964]. Models, Instruments and Truth. In The Myth of the Framework. In Defense of the Science and Rationality. London: Routledge. Popper, K. 2002 [1957]. The Poverty of Historicism. London: Routledge. Popper, K. 2002 [1963]. Conjectures and Refutations: The Growth of Scientific Knowledge. London: Routledge. Ricoeur, P. 1986. DutTexte à l’action [From Text to Action]. Paris: Seuil. Scriven, M. 1959. Truism as the Ground for Historical Explanation. In P. Gardiner (ed.) Theories of History. Glencoe: Free Press. Thucydides. 1950 [iv sec. a.C.]. Thucydides The History of Peloponnesian War. London: E.P. Dutton and Company. Weber, M. 2005 [1904–1905]. The Protestant Ethic and the Spirit of Capitalism. London: Routledge. Weber, M. 2012 [1903–1906]. Roscher and Knies and the Logical Problems of Historical Economics. In H. Henrik Bruun & S. Whimster (eds.) Collected Methodological Writings. London: Routledge.

Rationality in Anthropological Explanation Ian Jarvie

1

Introduction

The Revolution in Anthropology that goes to the credit of Bronislaw Malinowski and A. R. Radcliffe-Brown sought to replace the historical and evolutionary explanations of human society offered by the previous generation of anthropologists (especially Sir Edward Tylor and Sir James Frazer) with explanations inspired by the sociology of Durkheim. Durkheim was a functionalist whose explanations sought out how a social phenomenon worked in and contributed to the working of society right now, not what possible historical origins it may have had and which have now atrophied. An immediate objection to functionalism was that it saw social life in holistic terms, the various social institutions and practices in some way contributing, propping up, the whole edifice. Human agency seemed to be diminished. A sharp version of the objection to this came from the tradition of Weber and Schumpeter via Hayek and Popper, namely the demand that as social scientists we adopt methodological individualism. the quite unassailable doctrine that we must try to understand all collective phenomena as due to the actions, interactions, aims, hope, and thoughts of individual men, and as due to traditions created and preserved by individual I. Jarvie (B) Philosophy, York University, Toronto, ON, Canada e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 N. Bulle and F. Di Iorio (eds.), The Palgrave Handbook of Methodological Individualism, https://doi.org/10.1007/978-3-031-41512-8_11

203

204

I. Jarvie

men. But we can be individualists without accepting psychologism.1 The ‘zero method’ of constructing rational models is not a psychological but rather a logical method. (Popper 1957, pp. 157–158)

(‘Man’ and ‘men’ in the quotation is old-fashioned sexist language for ‘people’ and ‘persons’.) Functionalism, however, proved very fruitful as a method and the question arose of trying to reconcile these two perhaps at odds methodologies. Most, but not all, participants in the discussions aimed at a scientific, explanatory anthropology. Were functionalist explanations at odds with methodological individualism, was there no reconciliation possible? One bright spot was Popper’s concession that (following closely in this respect Hayek) “the social sciences are largely concerned with the unintended consequences, or repercussions, of human action” (Popper 1957, p. 158). Not unintended in the sense of consciously, but rather unintended in the sense that repercussions may violate all interests of the actor, conscious or unconscious. Think of those articles in the mass media that promise a getaway to the quiet, the unknown, the unspoiled. How long can those qualities survive their being widely broadcast? In the present chapter I shall discuss functionalist explanation in anthropology and I shall try to show that the hope of reconciling it with methodological individualism, a hope I once shared, was forlorn; the triumph of Durkheimian sociology requires a different emphasis in methodology, one that might be better called ‘Institutional Individualism’ (Agassi 1975) or ‘Situational Individualism’ (Wisdom 1970). My title links two tangled fields: rationality and explanation in anthropology. There is no escape from the tangle other than by stipulating and simplifying. Rationality: my previous publications on rationality take it as part and parcel of social science explanation; settle for a minimalist account of what rationality involves; and argue moreover that rationality is a matter of degree, not of either/or (Jarvie 1964, 1965, 1970, 1984; Jarvie and Agassi 1967). Explanation in anthropology is also contested ground. Some authors claim that anthropology does not explain at all, it is purely descriptive (Leach 1989; Geertz 1988; Clifford and Marcus 1986). Others argue about the terms in which the descriptions are couched. They argue that such explanations as are offered by the people being studied should be taken at face value 1

This sentence is of the greatest importance. The Wikipedia entry on methodological individualism, to take an example, states it to be “the principle that subjective individual motivation explains social phenomena, rather than class or group dynamics”. In this way, methodological individualism gets mixed up with methodological psychologism and collapses towards the views of J. S. Mill, decisively criticized and rejected by Popper (Popper 1945, Chap. 14).

Rationality in Anthropological Explanation

205

(Holbraad 2020). Some argue that the search for explanation (and of description too) is in and of itself a scientific or scientistic presumption and one that anthropology should eschew. Each of these positions can be and often are formulated as prescriptions rather than simply descriptions; prescriptions for the health of anthropology, so to speak, so that it can avoid various moral and political transgressions. To by-pass these sorts of controversies for purposes of this chapter, I shall stipulate that my interest is anthropology as explanatory science.2 The aim of explanatory anthropology is to explain repeatable descriptions that are puzzling, that are problematic. What is problematic is relative to point of view. Much of social life is not problematic to the people living it. Problems arise when comparisons are made, by the anthropologist, to earlier states, to expectations, to how they do things in other places. If problems are relative to point of view then explanations will be relative to point of view as well. Historically, anthropology has gone through two phases of explanation: explanation from history and explanation from function. Some anthropologists write as though we are now at a third stage, called “postmodernism” perhaps, where new norms have arisen (Gellner 1992). What these are is hard to grasp. Accepting the functionalist contention that conjectural reconstructions of human history and development do not explain the present state of things, I shall confine myself to functional and structural–functional explanations and discuss their rationality. Malinowski, for example, offered to explain the regular and ritualized trading pattern of the kula in the Trobriand Islands (Malinowski 1922). He found that ceremonial exchange of decorative items was sustaining reciprocal acknowledgement and status respect. Organized and dominated by hereditary chiefs, these visits to exchange necklaces and bracelets bonded the Islanders to one another, enhanced of diminished status, and were occasions for a certain amount of gifting and bartering trade. Radcliffe-Brown, for another, explained by means of structure and function features of certain systems of kinship reckoning. Descent, succession, and inheritance were all regulated by kinship (Radcliffe-Brown 1952). Evans-Pritchard, pupil of both of the previous scholars, explained how the segmented kinship system of the Nuer made possible a society that functioned without institutions of government, how the peace was kept or restored, and how quarrels and feuds did not become unlimited in scope (Evans-Pritchard 1940). By and large this functionalist work has held up rather well. Attempts to falsify the claims

2

Thus leaving aside anthropology as simple description, salvage anthropology, anthropology as literary exercise (fiction or nonfiction), anthropology as untestable history of the development of humankind, postmodern anthropology, and any other characterization.

206

I. Jarvie

serve to illustrate their fruitfulness. It has been said that science offers solutions to problems only for the solutions in turn to spawn countless further problems. Functional explanation must be used with caution, but it is a fruitful approach. The main limitation to a functionalist approach is that some problems that interest anthropologists involve social change and thus cannot readily be shown to serve standing social functions. In this paper I shall illustrate, amplify and maybe clarify arguments with two case studies: an old chestnut, Melanesian cargo cults, and an up-to-the-minute-of-writing case: various sociological problems raised by social behaviour during the COVID-19 pandemic of 2020–2021. Cargo cults were socially disruptive events occurring across vast areas of Oceania. Social reactions to COVID-19 are part of social adjustments to an unprecedented public health emergency on a world-wide scale. To an extent, my focus on scientific anthropology and its scientific explanations begs the question of bringing together anthropological explanation and rationality, if only because, with Popper, I think that the enterprise of science is our paradigm of rationality. We shall see, however, by looking closely at our case studies, that the role rationality plays in such scientific explanation is far from simple and straightforward. It seems to me that we have to concede that rationality comes in degrees, or on a sliding scale, and that the scale is an open interval: it does not have endpoints. That is, there is no explanation that attributes zero rationality; and there is no perfect or 100% rationality either. In all situations rationality could be improved or could be weakened. The three tribal ancestors I have invoked—Malinowski, Radcliffe-Brown, and Evans-Pritchard—all insisted on one or other kind of functional explanation. Institutions were explained when the anthropologist managed to show how they contributed to a stable social system that maintained and reproduced itself. Malinowski was given to stressing the system’s ability to satisfy all the basic human needs. Radcliffe-Brown was especially drawn to the ways in which normal human affections became extended beyond their titular recipients. And Evans-Pritchard is particularly compelling when he explains how the Nuer coordinate in solidarity despite lack of any central direction. All three anthropologists take swipes that their predecessors for engaging in what Radcliffe-Brown referred to as “pseudo-history” and their acceptance of the implicit premise that present practices can be explained by tracing them back to earlier practices. A favourite example was that some kinship customs might be a “survival” of an earlier stage, for example a survival of earlier matriliny in a presently patrilineal society. There were, the functionalists maintained, no such “survivals”. If a social practice was identified now

Rationality in Anthropological Explanation

207

then it must have its proper place now, and specifying how it worked now was its best explanation. Much of the work of these classic functionalists is illuminating. They saw themselves as students of the immensely influential work of Emile Durkheim and his students. It was commonplace in the mid-twentieth century to argue that, despite this impressive body of work, there was an incoherence at the heart of these endeavours: functionalism when formulated as a general doctrine about society was not remotely plausible. Consider some such version of functionalism as: societies are integrated social wholes, each part contributing and none “surviving” from earlier periods like the appendix in human anatomy. Any such claim runs against the obviously ramshackle character of much human sociality, the fuzzy boundaries, and the fact that social institutions can be abolished, modified, replaced, recycled and that new ones can be invented and introduced. In other words, society can be changed and sometimes of its own volition it just changes. It is false that all societies are fully integrated, fully functioning units. For one thing, the boundary between one functioning society and another may be manifest on Pacific Islands, but in societies that dwell on the major continents it is likely to be conventional. The institutions of language and of money, for example, certainly are functional, yet their contribution extended well beyond any single society, however defined. Functionalism also clashed with methodological individualism because it looked at societies as functioning wholes, with contributions from parts, and seemingly left out of account the role of individual, aim-guided actions. Individuals were not likely to have the aim of contributing to social functioning. In my original critique of functionalism (Jarvie 1964) I stressed that its successes could be attributed to taking it as a method, as advice on how to do research. It enjoyed thus: Do not treat the puzzling or the quaint as without social value. Look for how institutions work with other institutions. And so on. There was something very valuable in the functional approach which had enabled the leading social anthropologists to make their contributions. Given that strong statements of the functional integration of societies were too strong, in fact false, what made it valuable as a method? The answer is that functionalism was a case of looking for the unintended consequences of rational action. The rubber stamp example is the way many tramping feet in a meadow or a woodland create a path, the better for later feet and for guiding those uncertain of the way. The path is an unintended consequence of intentional human action. The rise in the post-world war II world of international air travel was a boon to all kinds of communication and tourism. It also provided a new vector for the spread of disease including, as

208

I. Jarvie

we know, COVID-19. Unintended consequences can be desirable—a path through unmarked territory—and undesirable—the spread of disease. Acting adequately to a situation is a minimal assumption of the ability to think things through. Once the situation is assessed, the route to an aim or even if there is a route to the aim points the way to the action to take. If the aim is to cross the road and the actor turns on her heel and enters a store then only a few explanatory options present themselves: the actor has changed her aim; the actor has decided to go the long way round; or the actor’s behaviour is erratic. If it is erratic then by definition it is not repeatable and if it is not repeatable it is not testable and so the question arises whether it is explicable at all. That for the most part actors act adequately to realize their aim, given their appreciation of the situation, has been called “the rationality principle”. It is quite minimal in that it says “mostly”: that is, it concedes from the start that it is at best a partial truth. It indicates a degree of idealization or simplification in our modelling of action. Functionalism had, it goes without saying, to distinguish itself from teleology, the idea that there was design or intent behind successful functioning; rather, the functional outcome was mostly an unintended consequence of actions towards other ends or due to the social arrangements. Human societies are not in general designed, yet they function. Their actors pursue mundane aims yet the overall system works by coordinating across both small and large scale. Some of these outcomes we can call unintended consequences of goal-directed human actions. Take a mass vaccination program such as was launched in 2021. The goal was to give protection to all of humanity from COVID-19, and to reach a herd immunity that would tame the virus. Among the outcomes would be a certain degree of failure of the vaccines, and some unwanted side effects. These failures and side effects are unintended consequences. The consequences are not unexpected but they are nonetheless unintended. Were a vaccine without side effects available, it would be deployed. One can logically see that there might also be unintended but desirable side effects. Some scientists think that public health measures helped suppress the flu season of winter 2020–2021. F. A. Hayek went so far as to argue that unintended consequences are the central problematic of the social sciences (Hayek 1955, p. 39). Intended consequences, being intended, are their own explanation. The actors assessed the situation correctly and acted appropriately to achieve their aims. Unintended consequences provide most of the interesting cases in the social sciences. No Nuer, as far as we know, dwelling within their kinship system, has the intent of making an acephalous political system work. Yet, according to Evans-Pritchard, that is what results as unintended consequences from their actions.

Rationality in Anthropological Explanation

209

What I want to do in this chapter is to retrace my steps. My original monograph on explanation in anthropology used a case study (Jarvie 1963, 1964). I want to return to that case study, to Melanesian cargo cults, and look at the explanation of them in order to display the role of rationality. Rationality being such a contested idea, looking at explanation in action will focus us on just what role it plays and just what sort of thing it comprises in such an explanation. To broaden and update the claim I shall add a case study attuned to current events: the pandemic of COVID-19. This world-wide phenomenon offers many problems of an anthropological kind alongside hosts of other problems in natural science, government, and administration, and, perhaps most complicated of all, communication.

2

First Case Study: Rationality in Cargo Cults

An example of a repeatable social phenomenon in need of explanation would be the “cargo cults” that appeared in Melanesia in the first quarter of the twentieth century. These were popular movements that were remarkably similar, although appearing on islands scattered across vast ocean distances. To Western observers they resembled a cult or religious movement. They usually involved a prophet, the prophet’s prophecy, a prophecy that portended something like the end of the world, an ending in which injustice and exploitation would cease, and quantities of material wealth, which the prophet’s people lacked, would be delivered by ship or by aeroplane. Some prophesied that the souls of the dead would rise, and people would be reunited with their loved ones. Less common but hardly insignificant were prophecies that power would shift from the colonial authorities to the natives and that white skin would turn black and vice versa. Various syncretic rituals were performed, different in different places, sometimes involving trance-like states or “possession”. What needed explanation was the rather sudden appearance of these cults in one cultural area, Melanesia; their structural parallels, and their rapid spread, despite their involving considerable sacrifice of time and resources by their followers. When first encountered by administrators and expatriates they were classified as a kind of mass madness. One was named “The Vailala Madness” (Williams 1923). Because the cults promised material rewards in this world they were for long regarded as a new twist in the catalogue of cult sociology. On closer examination, and better comparison sets, they were found to be less outliers than at first thought. Once an esoteric phenomenon

210

I. Jarvie

known mainly to anthropologists, the phrase “cargo cult” is now in general use.3 Cargo cults presented a number of explanatory problems. They arose in a vast area of the Pacific Ocean when inter-island communications were, to say the least, attenuated. What means of transmission and synchrony could achieve this? They were in effect millenarian or end of the world cults. F. E. Williams, a colonial official, was alarmed that cultists might squander or destroy valuable assets, including animals and crops, and thus put their own welfare in jeopardy (Williams 1923). What could explain such self-harm on a large scale? Why was it, also, that cultists clung to magical thinking in the face of empirical evidence? In a striking experiment, the Australian government (as the colonial power) flew a cult leader to Australia to show him how “cargo” goods were manufactured, in the hope that he would realize that ships full of such goods would not just appear one day out of the blue, so to speak. The leader, Yali, was unconvinced (Lawrence 1964). Cargo cults were something new in Melanesia, it seemed, although cults were not new in human society or the study of sociology.4 Cults are conventionally contrasted with “organized religion”, although one would be hard-pressed to characterize the supposed differences. There might be some explanatory traction there. The resemblance between cargo cults across vast waters of the Pacific Ocean offered to bring in both structural similarities and long-distance communication previously thought impossible. And the cults were clearly going to change things in ways that would challenge functionalism. Early observers such as Williams feared they would bring about social disintegration. After reviewing the extant literature on cargo cults, the explanation that I plumped for was an application of the logic of the situation. This modelbuilding exercise postulates that actors have aims which they act to realize in a situation not of their own making. They need to assess that situation and its possibilities for achieving their aims. The situation permits various ways to achieve aims, these ways are its logic. A situation has physical or material aspects, as well as social aspects, including ideas generally held about the role if any of supernatural forces. The situation provides both tools to realize aims and obstacles to their use. The force of gravity, for example, prevents us from flying by flapping our arms, but also ensures that goods stored on a jetty do not float off into the air. The capacity to organize multiplies human 3

There is a Wikipedia entry that offers a good survey and which informs us that some anthropologists are now nervous about the term and its use to stigmatize the Melanesians. 4 Subsequent research has claimed the cults’ ancestry goes back to Fiji in the last quarter of the nineteenth century. Wikipedia says a number of named cults are still active.

Rationality in Anthropological Explanation

211

effort but it also requires discipline and the provision of basic needs. So the situation combines physical enablers and obstacles to realizing aims, as well as social enablers and obstacles. Williams’ report describes how the Papuans had built platforms, an example of their using physical materials to hand, and old and new kinds of social organization to accomplish the work. The platforms were to receive the cargo and perhaps also the returning ancestors. Not to be forgotten is the nuance that the situation is what it is, both naturally and socially, but there is also the question of whether the actor makes a correct assessment of just how it is. If the aim is to obtain material wealth (of which the cultists are said to feel deprived) then the natural and the social enablers and obstacles are obvious to the anthropologist. The goods desired are trade goods (cargo) manufactured industrially a long way away. The traders, colonial officials, missionaries, and military that have and control the trade goods import them by sea and air transport. Virtually none is made locally in Melanesia. Intriguingly, the first three classes of outsiders pay cash for the cargo; the military, however, do not. They just have a lot of stuff and they also leave a lot of bits and pieces behind when they move on. The situational model, then, comprises an attempt to attribute the rationality of means/ends action to the Melanesians and rests the explanatory particulars on the interplay and the gap between their situation as it actually is, naturally and socially, and their appreciation or understanding of it. We are dealing with a cargo cult, one that has some concrete aims (the acquisition of material goods) and has some less concrete, such as the return of the ancestors or the inversion of social roles between those ruled and dominated becoming the rulers and dominators. The model postulates the Melanesians as trying to figure out how they can control and hence get a bigger share of the material goods but at the same time an alleviation of their subordinate condition. After all, the traders demand money and money comes to those who take directions. The missionaries promise a better world, but how does the Melanesians know that they do not withhold important information? The governing colonial officials are concerned with social order and servicing the economic contacts of the outside world: they exercise both economic and physical power. Given the mix of material and supernatural matters some would argue that this is not a route to a rational explanation. Since there is no alternative to a rational explanation, or, to put it another way, the alternative is no explanation at all, I think logic of the situation copes just fine. If Melanesian actors conjecture that both the making and the delivery of cargo goods involve supernatural forces, then attempts to identify and manipulate them make completely rational sense. In this they are no different to the millions

212

I. Jarvie

in the society from which the anthropologist comes who, in the face of a catastrophic storm, or a pandemic disease, or a personal loss, turn to fervent prayer to what they think of as the supernatural governor(s) of all that befalls us. Their actions are rational given their aims and their appreciation of the situation they are in. They are hardly as rational as they could be, it goes without saying. If the problem is a catastrophe or a pandemic or a loss of life it makes little sense to think that a spirit that initiated it can be persuaded to mitigate it; otherwise what sense does its initiation make in the first place? But here I reveal my secularist bias where I obviously think that the supernatural is an unnecessary surplus hypothetical. For a social actor to act rationally in seeking to achieve their aims there are a number of aspects to assess. First, are the actors quite clear about their aims and their feasibility? Aiming to become filthy rich, or the leader of the free world, or dictator of your own country is rather vague and distant and there is no clear path for achieving them in any situation. Hence setting one’s hat at such aims is not as rational as it would be to focus on some others. Second, in choosing means, there are similarly those that take account of the natural and social situation and those that get it wrong. For example, most aims could be realized if the social actor could find a magic wand that would accomplish their aims. The problem is that neither the wand nor magic exist in the natural and social world, hence the quest for those means is futile. There are less flagrantly irrational actions. A social actor whose means involve marshalling the efforts of many other social actors is giving hostages to fortune. The actions of visionaries and initiators and generals may look at first high risk and low rationality only to have to be reconsidered when the vision is realized, the initiative bears fruit, the battle is won. So, pennywise pound foolish action may be deemed irrational, but not a lot would be achieved if no one ignored the maxim. These and other examples show that rationality is a judgement call: the observer assesses the natural and the social situation and then reckons whether the actors have a correct or near correct grasp of it. If they have an incorrect grasp their plans are bound to go awry since they are acting on a poor appreciation of what they actually face, the true enablers of, and constraints on, their action towards their aims. The observer can also note whether the social actors in practice use such correct knowledge of the situation as they have. Actors may revert, waive aside, or even just forget what they know and hence act less than optimally even by their own standards. Come back now to ordinary situations in anthropology. Melanesian actors faced with a natural and social situation where they aim to acquire more material goods and they also aim to change the power dynamics between themselves and the

Rationality in Anthropological Explanation

213

various groups of outsiders who come to their shores, not to mention their universal aim of reunion with their departed loved ones. They resort to a mixture of natural and supernatural means. They build airstrips and jetties ready to receive cargo. Except that the cargo is not on its way. They engage in rituals, including rituals with high social costs and this too invokes false premisses, since rituals cannot conjure up the cargo. Their grasp of their own social situation in the big picture, for example during and in the aftermath of the second world war in the Pacific, is faulty in some respects. The armies of Japan and of the United States and Australia were certainly well equipped and supplied with material goods but there was in fact no way, natural or supernatural, that the cultists could divert some of this materièl to themselves. All such anthropological problems can be formulated only against a background. They are not, so to speak, problems to those involved; they operate against a different background. Hence these cults contradict the idea that anthropologists ought to confine themselves to the point of view of those they study. Important though that point of view is, it is not the scientific point of view. Durkheim would have got nowhere if he had confined himself to the outlook and motivations of each individual suicide (Durkheim 1897). On the contrary: the suicides that interested him were a statistical artefact and he was interested in the rates, and the fluctuations of these rates from place to place, time to time, social group to social group. His point of view was not available to each and every suicide. Similarly, the comparative questions about the cargo cults, their overall similarity, their separation by enormous expanses of ocean, and their likelihood of self-harm were hardly available to the cultists or even to the cult leaders. In setting out to find explanations of these cults the move that I advocated and that, quite independently, the anthropologist Peter Lawrence (Lawrence 1964) advocated, was to reconstruct the general situation of people in these scattered island communities faced with only partly explicable life changes. Foreigners arrived first to establish commerce in societies without it; they were followed by others who assumed roles of governance among peoples who heretofore governed themselves in their own way; and, of course, when the second world war supervened on all this, heavily armed men accompanied by mountains of “cargo” came, fought each other, and went. Not to be overlooked is the fact that almost every one of these island communities had also experienced the ministrations of Christian missionaries, both foreignborn and locally recruited, who had offered interesting new ideas about the nature of the world, its origins and workings, and the important power of ritual, including prayer. The missionaries brought with them sacred books which contained, they said, god’s message to the world. They also directly

214

I. Jarvie

suggested that their ways offered better control of the supernatural than did traditional ways. Cults, from the Vailala Madness through to that around Donald J. Trump, arise when a charismatic prophet offers a syncretic message that seems to those who join to offer a way out of their present distress, lack, yearning, and suchlike sentiments. It is essential, however, that the anthropologist not endorse magical thinking. Or, to put it another way, those who take magical thinking seriously are not part of the project of scientific anthropology. Anthropology is a secular child of the Enlightenment. Anthropology aspires to be a general and comparative science of society, that is, it seeks to solve problems that can only be formulated by those who stand back from the local point of view and seek the overview. Cult joining has a degree of rationality in a reconstructed logic of the situation. It contains magical thinking, which means it is hardly optimal. The cult also usually depends on the charisma of its prophet and that is also an element of lesser rationality. And of course any apocalyptic element flies in the face of commonsense and the fact that previous or other prophecies of The End have signally failed to be fulfilled. What is the status of rationality in this situational model? Traditional views of rationality were all or nothing, black and white. Science was rational, superstition and magic were irrational. This hardly will work here. Cargo cults are not a form of madness. Neither the prophet nor the followers need be classified that way on any scale of sanity. The prophets need not be thought disingenuous either. He or she is convinced that they have solved some local problem concerning the missionaries, merchants, colonial administrators and, eventually, fighting military that have intruded into their space, and the huge discrepancies between their own material condition and that brought with them by the outsiders. We might see those joining the cult as having a similar yearning to make sense of events and conditions and hence to eagerly receive the prophetic message. The language of magic and the supernatural is not out of the ordinary and it would hardly behoove the missionaries to cast any doubt on it. It is very important to take the cults at face value: they explain and rectify the situation of peoples whose lives are being disrupted and who are being offered confusing and confused explanations of what is going on. By the Enlightenment view of rationality, error is not rational. This will not do on general grounds, so even a science descended from the Enlightenment, as anthropology is, must do without it. If we invoke a pragmatic standard of rationality and ask: will the cults achieve their goals, the answer has to be, “Alas, no”. That is, if the Melanesians feel excluded and exploited by the commercial set up, if they feel demeaned and oppressed by the colonial set up, if they feel deprived and impoverished when they see the materièl that

Rationality in Anthropological Explanation

215

the armed forces deploy, it is clear to the anthropological observer that the cult injunctions to build jettys or airstrips, to sacrifice animals and crops, to conduct rituals that will bring on the arrival of the cargo, will all fail. The commercial intrusions, the colonial intrusions, and the military intrusions will not be stopped or rectified by the means the cults deploy. This is a plain fact evident to the anthropological observer. Marxists, seeking rationality in history as it unfolds, argue that perhaps the cults are the beginning of a sort of class-consciousness that shows promise, sooner or later, of being a way to overthrow exploitation and colonial oppression (Worsley 1957). Hence, again, the rationality found is not evident to the natives but is evident to the loftily positioned surveyor of history unfolding. It is also crucial to factor in the role of Christian missionaries to the whole mix. These are outreach groups, they are interested on the face of it not so much in resources, trade, and real estate as in conveying messages of hope and salvation to the local peoples. They claim to have knowledge of the way things really are and they offer to share it. They bring new rituals and a sacred text that contains the gist of their message. They are emphatic that what is important is the supernatural, not so much the natural. But the message is mixed. As well as services and preaching, Christian missionaries set up clinics and schools. Clinics deal with natural ills naturalistically. Schools teach about the natural world as well as the supernatural. When anthropologists describe the cults as syncretic they have in mind that some elements of Christian teaching find their way into practices. One of the most striking is the idea that although the Christians share their knowledge, they have removed from Bibles crucial pages that contain the secrets of the power of the white men.5 What can we discern of rationality here? The answer is that the cultists have worked out a good solution to realizing their aims in the situation as they see it. These aims are nonetheless not going to be achieved: no cargo, no return of the ancestors, no inversion of governed and government, no exchange of skin pigment. To the anthropologist, who brings a different perspective, the cults have an element of poignancy, even tragedy. If the world were the way the cultists have it, then their rituals and gestures might have succeeded. They did not; hence something is amiss in the native appreciation of the situation they are in, and only an improved appreciation has any hope of their grasping a better reconstruction of that situation, and hence of achieving at least some of their aims. Two reappreciations might be particularly difficult to assimilate. One is that some of the aims can be achieved only by natural and not

5 Missionary activities are here taken at face value. It goes without saying that they also offer opportunities for “unmasking”.

216

I. Jarvie

by supernatural means. The second is that both the natural and the supernatural are very hard to manipulate successfully. It follows that even better appreciation of the true situation does not immediately produce success at acting in it to bring about one’s aims. This is simultaneously a practical and a metaphysical point. The natural/supernatural distinction itself may be very hard to grasp. The supernatural and the magical permeate the materialist and consumerist societies from which the anthropologist comes, just as much as they do Melanesian society. The anthropologists may have their own particular combination of views on it. Any correction the Melanesians may need to make will have to draw a different boundary where no boundary at all was known before. The anthropologist may engage in supernatural obeisance on Fridays, Saturdays, or Sundays depending upon their background. But usually, the anthropologists will not extend supernatural thinking to ordering supplies, collecting mail, awaiting transport, and suchlike. These will match entirely naturalistic means to their ends and transmitting that knowledge to those not yet comfortable with the natural/supernatural distinction may be, to say the least, difficult and liable to misunderstanding. The cargo cults arose in areas that were under colonial control; that had been on the end of the world’s trading networks for many years; and which were penetrated by missions about the same time. Contact with these intruders had not created a secular realm that matched that of the intruders so it is a ticklish point whether further contact with the delicate sensibilities of anthropologists will do the trick. Come back again to the overall view of rationality. If we treat rationality as a quality of actions to realize aims based on best available appreciation of the situation, then we will conclude that cargo cults are quite rational. If a physician thinks disease is caused by the system of humours, then treating it with the means best understood of dealing with humours is rational. The cargo cultists have a mistaken appreciation of their situation, as mistaken as the system of humours in medicine. The colonialist, traders, missionaries, and soldiers operate in a situation they construe differently, and their construal is a lot closer to the way things are. Similarly, disease is not caused by the humours and treatment prescribed by the humorous diagnosis is mostly ineffective. Disease operates in a different situation than the one the humoral doctors and patients think they are operating in. So the situational model allows there to be rationality when the actors are utilizing false theories. This is not a surprising result. Most science in the history of science, including most present science, is false. Yet humans have been thought to be acting rationally a lot of the time. The falsity of a world view or of what is thought to be science cannot be the measure of rationality.

Rationality in Anthropological Explanation

217

Most modern rationalists are uncomfortable with this: they think science is rational and any appreciation of the situation that does not incorporate or acknowledge science cannot be called rational. That is their mistake. They too are misappreciating the general human situation. Most people in all societies, including the society from which the anthropologist comes, are ignorant of science and they do not respect a natural/supernatural distinction. A very common attitude towards science in our society is to treat it as a form of magic. People have attitudes to disease in their own bodies, to the behaviour of prices, to the impacts of social change, that are hardly consonant with known science on these matters. Yet their actions can be explained as acting rationally to a degree given their appreciation of the situation. For example, given that one understands neither one’s own body scientifically, nor the arcana of present-day scientific medicine, is it rational to submit oneself to the doctors and their treatment? One may be ignorant of why they do what they do but one does know that they know a great deal one does not know. And since the natural/supernatural boundary is a matter of dispute this means that some people gladly accept vaccination whilst silently offering a prayer. They view both actions as rational and given their appreciation of the situation, they are not mistaken nor are they irrational. Those who equated rationality with science wanted to argue that being mistaken about the way the world is can be treated as a failure of rationality. This is in itself a mistake and a mistake we should see as partly rational, or rational to some degree. Obviously, facing a situation, the more accurate one’s assessment of it the better one will be able to cope with it. Facing an abyss but mistaking it for solid ground makes taking a step forward a disaster. But if science is conjectural then we are always in the position of not knowing whether we face an abyss or solid ground. So we have to act on our best guess. Equating rationality with having perfect knowledge is to equate it with utopian conditions that nowhere obtain.

3

Second Case Study: Rationality and COVID-19

To try to correct for ethnocentrism let us turn to our present condition, as it displays itself in medical and social reactions to the ravages of a novel virus (SARS 2 COV) which causes the disease known as COVID-19, treating our own experience as ethnography. The aim is again to treat anthropology and anthropological explanation as a comparative science of society, not an exercise in ideological dominance of the other. The WHO declared COVID-19

218

I. Jarvie

a pandemic on March 11, 2020, “pandemic” being a term used to indicate that the disease is likely to spread to every society on earth. Every society on earth tries to fathom the situation they are being put in, and tries to cope with the disease and, of course, to the virus all humans are equal in being potential hosts.6 All kinds of supernatural, conspiracy, and clearly nonsensical accounts of the situation are circulating as I write (in the course of 2021), especially in the unfiltered megaphone of “social media”. Among those who adopt the scientific outlook their appreciation of the situation is without comfort. The virus is a tricky, actively reproducing, mutating, semi-living thing.7 It kills some of the hosts it invades and spares far more. It is invisible to all humans and most machines. Only electron microscopes can visualize it. Yet it is insidious: it distributes itself via aerosolized particles, by clinging to surfaces, in the shedding of the asymptomatic as well as the symptomatic. At least in outline, this is the situation as science, natural and social together, describe it. Unlike with the cargo cults, where the anthropologist was able to take an outsider view and to consider problems that may not have been problems for the cultists, this pandemic situation makes all humanity, including the scientist-anthropologist, a participant as well as an observer. For the social scientist the following are some of the problems that present themselves from the overview. Why is there organized resistance to public health measures? Why is there organized resistance to the medical intervention of vaccination? Why do scientists and physicians face intense scepticism and lack of trust? Why do so many resort to magical thinking? The answers lie, I argue, in the logic of the situation. So first we have to get clearer what the situation is, and then try to grasp how it looks to those it affects and whose actions puzzle us. This virus has created a situation that many people could not grasp and so they substituted imaginary constructions of the situation. In this they resemble the cargo cultists. The first situational problem was this: the enemy, the virus which caused the pandemic, was invisible. It was invisible in two ways. Viruses are extremely small and can only be visualized indirectly. So the enemy was literally invisible. The second way it was invisible was that

6 Some human groupings are more susceptible to illness and death than others. All the usual variables have been explored. 7 ‘Viruses teeter on the boundaries of what is considered life. On one hand, they contain the key elements that make up all living organisms: the nucleic acids, DNA or RNA (any given virus can only have one or the other). On the other hand, viruses lack the capacity to independently read and act upon the information contained within these nucleic acids. “A minimal virus is a parasite that requires replication (making more copies of itself ) in a host cell,” said Jaquelin Dudley, a professor of molecular biosciences at the University of Texas at Austin. “The virus cannot reproduce itself outside the host because it lacks the complicated machinery that a [host] cell possesses.”’ (Vidyasagar 2016).

Rationality in Anthropological Explanation

219

some infected people showed no symptoms; that is, tests detected viral antibodies in people who presented no clinical symptoms. They appeared to be well yet the scientific view was that they were incubating and shedding the virus. Of course, science classes in school should have rendered these two features unproblematic: lots of science invokes a world of invisibles. The invisibility meant that a commonsense approach to the viral danger is inadequate. The danger was hidden and deceptive. So, people aiming simply to get on with their lives were confronted with various kinds of advice, controls, guidelines, and such like which seemed quite at odds with the visible world, which looked much the same as it always had. Except that hospitals were filling up with sick people and alarming numbers were dying quickly. Much of this toll took place in segregated institutions, such as care homes, hospices, prisons, and hospital intensive care units, so the general public was shielded from any visual indication of what was happening. As with all highly infectious diseases, the public was kept away from such sites. Again, there was concentration of cases in minority groups, often poor and living in segregated neighbourhoods. Thus the conditions were perfect for the oblivious to deny there was a disease outbreak at all, or to claim its size was much exaggerated, or even to venture that it was a hoax devised for some nefarious purpose by some invisible conspirators. Once such claims about the situation were made a logical impasse was reached. Claims such as that the disease does not exist, or is only a minor threat, or is a downright hoax, are irrefutable. You cannot show that something does not exist because the claim can always be made that you did not search hard enough or far enough afield for it. Once such ideas take hold, it is easy to explain away even confrontation with the disease, such as someone close dying from it, or getting it yourself. An invisible disease agent is matched by invisible actors hoaxing the rest of us. The stealthy invisibility of the disease means that scientists have to reconstruct what it is caused by and how it works. In short, they have to use imagination to interpret the data on what seems to be going on in the invisible world. It is tempting, in a democratic ethos, to think that perhaps anyone is entitled to use their imagination, including to imagine the disease away, or to postulate a conspiracy or hoax. And it is true, freedom to use our imagination is fundamental and universal. The difference between scientific imagining and general imagining is the critical attitude. The scientists operate by testing their imagination against the data and changing their ideas when the data do not fit. Sometimes they reverse that and challenge the data, insisting on a re-do. That is how they maintain contact with world as it is, rather than simply as they imagine it to be. They encourage nature, or

220

I. Jarvie

whatever one wants to call it, to adjudicate on the products of their imagination. The scientific objection to imagining a conspiracy or a hoax is that it is done uncritically, in a revelatory manner. Conspiracies and hoaxes are certainly abroad. If they are postulated, however, it is rational to expect some suggestion as to how the claims can be independently tested. Why was there a turn away from science, or, a failure to comply with public health measures recommended by scientists? To begin with, scientific consensus was not total, as is to be expected, and some hesitation appeared about spelling out (marginal) details. Moreover, the political world is subdivided into countries, and some countries into smaller units united into federations. This choir did not sing one song. On the contrary, politicians saw advantage in varying, pooh-poohing, and even mocking public health basics. Public health recommendations are a mixed bag. Some are directed by the scientific view of the virus: get vaccinated, increase hand hygiene, but all such measures have a social dimension: wear masks; stay at a distance; avoid crowds and social gatherings; do not go in to workplaces. They are directed by the view that many social activities facilitate the spread and hence the ravages of the virus. Perhaps more important, at least by some reckonings, the scientific recommendations changed over time because the science changed. For example, there was a period when it was thought that the virus could spread through droplets on surfaces and so rigorous cleaning of surfaces was enjoined, and those that could not be cleaned were closed. This was never shown to be false, but its fading from attention indicated it was no longer seen as a primary vector of the disease. Later, the emphasis shifted to the dangers of aerosols, i.e. microdroplets so small that they float in the air rather than settle on surfaces. This raised a new concern that was not stressed initially, namely ventilation and air circulation in buildings, vehicles, and so on. Similarly, various drugs and nostrums were claimed to have protective or beneficial effects; careful investigation showed this not to be the case. The authority that scientists wanted to claim was undermined in the mind of the public (some of them, anyway) by these changes. Unfortunately quite a lot of the public have utterly wrong expectations about science, expecting it to be monovocal and certain. No doubt the powers that be in science are partly responsible for proliferating such expectations. And no doubt the pretence to certainty stems from the false equation between error and irrationality. A major research initiative was attempted, in several countries, using the initial genome sequence of the virus, to devise formulas for vaccination. After many trials and some failures a handful of successful vaccines were developed. But the testing and genomic sequencing soon revealed that variations of the original sequence were abroad and they were also found to have slightly

Rationality in Anthropological Explanation

221

different properties: more or less transmissible, more or less virulent, more or less lethal. Would the new vaccines give protection against these Variants of Concern? Each solution spawned more problems and each reassurance that became controversial changed the situation. The public were given a very hard lesson in the realities of the situation in science: science is conjectural; scientific data need interpretation; there is sometimes consensus but just as often there is controversy. Such nostrums as “follow the evidence”, or “science-led” offer false promises of certainty. There is no certainty. Even the best vaccine fails in some subjects, poisons rather than protects some subjects, and is limited to a specified effectiveness discovered empirically. This situation is what explains the phenomenon of anti-vaxxers, which antedates COVID-19. Similarly, a jab plus a prayer strikes many people as rational, and a cheerful “keep your fingers crossed” goes unremarked for its invocation of superstition. We read of countries where the authorities recommend traditional herbal remedies, but we are not told whether those are countries where modern medical treatment is in short supply. We read of countries where the leaders pretend to know more than the scientists and then substitute their commonsense for medical counsel. Unknown people declare themselves in possession of the true secret of the disease and offer their wisdom for sale. We read that some people hunt the web because they feel that the authorities are not to be trusted; yet somehow web sources are to be trusted. And so on. In such a bewildering and rapidly changing situation we cannot take seriously that the generality of humankind is behaving irrationally. We can only go by the obvious idea that people usually act on what they take to be their best appreciation of the situation. No doubt many appreciations of the situation are faulty and very far from the situation as it is. But no one is in a position to say just exactly what the truth of it is right now; perhaps there never will be such a time. There are viral epidemiologists who argue that the question, where exactly did the virus come from, will never be given a definitive answer. They say that with the authority of science. Science advises setting aside certain questions because they cannot be answered. What we can assume is that most people, most of the time, aim to survive, even to thrive. They do not want to get sick with a viral illness. It is their misappreciation of the situation created by the virus that explains those actions that deviate in some ways from what the observer considers to be optimal. In taking this lofty overview of the situation I hope to show how much rationality there is in the general chaos and disorder. That there is also much irrationality, including not just fear but panic, perhaps does not need to be stressed.

222

4

I. Jarvie

Discussion and Assessment

Treating rationality in the way outlined here, as a judgement of actors’ appreciation of the situation, we can be humane, we can be empathetic, and we can also do social science that is in line with the ideals of the Enlightenment. However, it is not hard to imagine certain lines of criticism that might be directed at this situational model of anthropological explanation. It gives a pass to flagrantly irrational behaviour, such as that by politicians or such as that displayed by people defiantly gathering together for drunken parties. In response I would note that it is not the business of the anthropologist to give or to deny passes. The assertion that something is explained by something else should be a testable claim, decidedly not a moral imperative. The criticism equivocates on the word “irrational”, attacking its role in the explanation whilst using it not for explanation but as an epithet. This confuses. We began by trying to show that cargo cults, also abused as “madness”, turn out to be nothing of the sort, the trance-like behaviour sometimes noticed being no more a sign of madness than possession behaviour in Pentecostal sects, trances in Voodoo, and mystical enlightenment as Plato envisaged it. The anthropologist explains. The moralizer judges. All too many anthropologists today want to fuse both roles. The rational part of their position is that they believe it is their responsibility to counter decades of moral indifference or even moral evil that was embedded in the colonial encounter. A second line of criticism could argue that attempts to explain actions situationally are condescending, since they substitute the view of things of the observer for those of the subject. An effort was made to deflect this criticism by noting that the very formulation of the problem to be explained involved the point of view of the observer who might be noting patterns, repeatable events, not known to or available to the subjects. Cargo cults are an example. The brief sketch of the problems created by the COVID-19 pandemic is another in that the observer has to switch back and forth between being a member of human society and being an observer of its patterns and repetitions. Durkheim, I noted, focused on problems around suicide that were not visible to the suicide in the sense that the suicide’s point of view could not be expected to throw any light on them (Durkheim 1897). A third line of objection might be that my account substitutes a notion of rationality as action adequate to the situation as assessed for a better and more traditional view that rationality is a capacity for reasoning and for behaviour based on reasoning that does not vary from situation to situation. I readily concede, as I did at the beginning, that the model of rationality I invoke,

Rationality in Anthropological Explanation

223

which is not all or nothing and which recognizes degrees, is not the traditional model. The traditional model is, however, defunct. There is no rational faculty. There is no cut and dried correct reasoning that can be used to measure human thinking. There is deductive logic, of course, but quite what its relation to thinking and to reasoning is, is highly controversial. At best we can compile lists of fallacies, that is, examples for invalid reasoning but then we can also compile lists of psychological experiments that show how, if anything, humans are prone to fall into these fallacies. So what exactly is going on? A final line of objection might be that my assessments of rationality depend on assessments of the way the actors face the situation, social and physical. This postulate, it might be said, is presumptuous. The anthropologist is dealing with people who see the world differently, and may even not accept the idea that there is anything like the way things are. A good point. It is not to be avoided. Science and scientific explanation do assume, one way or the other, that there is explanation and that the world is the way it is and not another way. I would claim, however, that it is the critic who is misdescribing matters. There is no inference from the obvious claim that one’s anthropological subjects may see the world, or aspects of the world, very differently from the way the anthropologist does, to the suggestion that they would deny that there is anything resembling the way things are. This view is relativism. As Ernest Gellner so emphatically insisted, as it is an anthropological reality that relativism comes from the anthropologists not from the tribespeople, relativism is yet another cargo cult that does not deliver the good it promises to deliver (Gellner 1995). The COVID-19 case is useful here because all we humans, anthropologists and their other, are in the affected group together. It is true that there are a few relativists out there, however disingenuous one may suspect their relativism is. The vast majority, however, do think there is a way things are, ranging from acceptance of the scientific picture put out by the WHO, through to the conspiracy theorists’ view that it is all a giant hoax. They think one or the other is true, is actual, is the way things are. Some have argued that this very claim undergirds the siblinghood of humanity.

References Agassi, J. (1975) Institutional Individualism. British Journal of Sociology, 26:144– 155. Clifford, J. & G. Marcus, Eds. (1986) Writing Culture: The Poetics and Politics of Ethnography. Berkeley and Los Angeles, CA: University of California Press.

224

I. Jarvie

Durkheim, E. (1897). Le Suicide. Etude de sociologie. Paris: Felix Alcan. English edition: 1951. Suicide: A Study in Sociology, translated by John A. Spaulding and George Simpson, edited with an introduction by George Simpson. New York: The Free Press. ISBN 0-684-83632-7. Evans-Pritchard, E. E. (1940). The Nuer. Oxford: Oxford University Press. Geertz, C. (1988). Works and Lives. The Anthropologist as Author. Cambridge: Polity. Gellner, E. (1992). Postmodernism, Reason and Religion. London: Routledge. Gellner, E. (1995). The Uniqueness of Truth. In Anthropology and Politics. Revolutions in the Sacred Grove. Oxford: Blackwell. Hayek, F. A., (1955). The Counter-Revolution of Science. Studies on the Abuse of Reason. Glencoe, IL: The Free Press. Holbraad, M. (2020). The Shapes of Relations: Anthropology as Conceptual Morphology. Philosophy of the Social Sciences, 50(6):495–522. Jarvie. I. C., (1963). Theories of Cargo Cults: A Critical Analysis. Oceania XXXIV (1, September):1–31 & XXXIV:(2, December)108–136. Jarvie, I. C. (1964). The Revolution in Anthropology. London: Routledge. Jarvie, I. C. (1965). Limits to Functionalism and Alternatives to it in Anthropology. In D. Martindale (Ed.) Annals of the American Academy of Political and Social Science, Monograph 5, Functionalism in the Social Sciences. Philadelphia: February, pp. 17–34. Jarvie, I. C. (1970) ‘Understanding and Explanation in Sociology and Social Anthropology’, and ‘Reply’ (to Peter Winch). In Borger R and Cioffi F (Eds.), Explanation in the Behavioural Sciences. Cambridge, Cambridge University Press, pp. 231–45 and 260–9. Jarvie, I. C. (1984). Rationality and Relativism. In Search of a Philosophy and History of Anthropology. London: Routledge. Jarvie, I. C. and J. Agassi (1967). The Problem of the Rationality of Magic. British Journal of Sociology, 18:55–74. Lawrence, P. (1964). Road Belong Cargo. Manchester, UK: Manchester University Press. Leach, E. (1989). Writing Anthropology. American Ethnologist, 16(1;Feb.)137–141. Malinowksi, B. (1922). Argonauts of the Western Pacific. London: Routledge & Kegan Paul, Ltd. Popper, K. R. (1945). The Open Society and Its Enemies. London: George Routledge & Sons, Ltd. Popper, K. R. (1957). The Poverty of Historicism. London: Routledge. (Originally published in Economica, 1944, vol. 11(42):86–103; vol. 11(43):119–137;1945, vol. 12(46):69–89.) Radcliffe-Brown, A. R. (1952). Structure and Function in Primitive Society, London: Cohen & West. Vidyasagar, A. (2016). What Are Viruses? Live Science, www.livescience.com/53272what-is-a-virus.html. Accessed 23 August 2021.

Rationality in Anthropological Explanation

225

Williams, F. E., (1923). The Vailala Madness and the Destruction of Native Ceremonies in the Gulf Division. Anthropology Report No. 4, Territory of Papua, Port Moresby: Edward George Baker, Government Printer. Wisdom, J. O. (1970). Situational Individualism and the Emergent Group Properties. In R. Borger and F. Cioffi (Eds.), Explanation in the Behavioural Sciences (pp. 271–296). Cambridge: Cambridge University Press. Worsley, P. (1957). The Trumpet Shall Sound: A Study of “Cargo” Cults in Melanesia. London: Macgibbon & Kee.

Ordinary Rationality Theory (ORT) According to Raymond Boudon Jean-Michel Morin

Science is Nothing more than a Refinement of our Everyday Thinking. Einstein, 1936, quoted by Boudon (2009: 60)

Raymond Boudon (1934–2013) wrote dozens of books and hundreds of articles. Yet, the entire latter part of his work is devoted to refining the analysis of individuals’ reasons for acting and believing. He sees in this the cornerstone consolidating the theoretical edifice destined to rebuild sociology. This central theme can be found early in his writing. Throughout his works, in his last ten years, he explicitly confronts it. – In Reason. Good reasons (Boudon 2003), he adapts Weber’s model and positions his own proposal. – In Towards a General Theory of Rationality: A Defence of Common Sense (Boudon 2007), he explains the contrast between his proposal and that of rational choice theoreticians (RCT), and he offers applications of his theory, in particular in the field of religious beliefs. – In his synthesis of rationality (Boudon 2009), he establishes his vocabulary, in calling his proposal: ordinary rationality theory (ORT).

J.-M. Morin (B) Sociology, University of Paris Cité, Ecole Professorale de Paris, Collège des Bernardins and Institut de Philosophie Comparée, Paris, France

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 N. Bulle and F. Di Iorio (eds.), The Palgrave Handbook of Methodological Individualism, https://doi.org/10.1007/978-3-031-41512-8_12

227

228

J.-M. Morin

– In Belief and knowledge. Thinking the political, the moral and the religious (Boudon 2012), he stresses scientificity but also the utility of such a model, in particular, in understanding normative, that is, moral and political, beliefs. Even in his writings published posthumously (2013), he also endeavours to refute all the theoreticians who refuse to take “reasons” into account in explaining human behaviour, in short, those who oppose methodological individualism. He especially focuses on: materialists, positivists, culturalists, and “all neuronal” theoreticians. It is not a question of engaging in an exercise in the history of thought here. I have presented the complete work of Boudon elsewhere (Morin 2020). Instead, it is here a matter of presenting the complete model of ordinary rationality theory, in its final version, according to the author. – First, the ordinary rationality theory is presented internally, with the dimensions of reasons which are included: instrumental, axiological, and cognitive (1); but also with the dimensions which are eliminated: affective, traditional, and biased dimensions (2). – Next, the ordinary rationality theory is presented externally, along with the inextricable links which are established between: individual reasons, collective effects, and social contexts, according to the equation S = f [a (r, C)] (3); but also with the rupture which this creates with theories which either deny reasons or only see reasons (4). – Finally, a comprehensive positioning of the model within social science is proposed: on one hand, with the cartography of six postulates which Boudon refines between 2003 and 2013 (5); on the other hand, with the adjustments which some of those close to him have suggested since 2020 (6).

1

The ORT Internally: Instrumental, Axiological, and Cognitive

The theory of ordinary rationality, which Boudon opposes to the dominant instrumentalist model of rationality, is based on a broad conception of rationality that he identifies as also cognitive and axiological. X may be a belief or even an action. “We will say that X is explained by ordinary rationality if X is, in the eyes of the individual who adheres to it, the consequence of a system S of reasons, including all the elements acceptable to him or her,

Ordinary Rationality Theory (ORT) According …

229

and if there is not within sight a system S’ of reasons which would lead to subscribing to X’ rather than to X” (Boudon 2009: 51). X could be a positive belief relating to what is “true.” For example, I believe that the earth is flat, right up until the moment when new elements lead me to think that it is round. This could also be a normative belief related to what is “just.” For example, I believe that the work week should be reduced to 35 hours to better share the jobs, up until new elements lead me to think otherwise. The important thing is to reconstruct the S system of reasons which lead each person to believe such things, until there is evidence to the contrary. Now we easily understand our ancestors, lacking telescopes, who, when contemplating the endless horizon, thought they were inhabiting a vast platform. It is more difficult to grasp the precise reasons for controversies concerning the governmental reduction of working hours. A new reason could be to note that, if everyone worked 35 hours to do the work which took 40 hours before this, no jobs would be created. Another reason may cross one’s mind: the work of one is not necessarily transferable to another. Again: jobs are constantly created or destroyed and this affects everyone’s tasks and the time required to accomplish them. Also: generations succeed one another at work, ultimately affecting the distribution of jobs. Thus, there are a multitude of reasons to believe in something, to the point of being convinced to take action. Yet, how then can we comprehend a system of reasons? To reconstitute a system of reasons, before Weber’s synthesis, there were two opposing schools. – On one hand, the utilitarians, often Anglo-Saxons since Adam Smith, claim that the principal reasons can be reduced to the often selfish interests of individuals who hope for maximum utility. – On the other hand, the idealists, often German since Kant, claim that pure reason leads individuals to follow universal principles, reflecting this duty to respect everyone’s rights. Fortunately, Weber points out that these two dimensions come together in a system of instrumental and axiological reasons (Weber 1922). We are moved simultaneously by both interests and values. For example, if a merchant scrupulously returns the change to a young child who is doing errands without knowing how to count, his or her system of reasons is probably composed of two dimensions. On one hand, it is instrumental to say that the consequences would be disastrous if one were suspected of dishonesty. Here, the fear of losing one’s reputation would override the temptation to earn more money. On the other hand, it is axiological to tell oneself that it is wrong to

230

J.-M. Morin

abuse the confidence of one’s clients, especially the most vulnerable. Here, a moral sense, professional ethics, determines the decision. The problem is that a system of reasons surely consists of other dimensions. Weber mentions an affective dimension; in our example, the child is sweet; and a traditional dimension, still with this example, for this merchant, the practice becomes habitus and has become a routine to the point where there is not even a question of behaving otherwise. We will see that Boudon does not return to these dimensions. Instead, he forcefully introduces a cognitive dimension. In the example, the merchant must know how to count. To summarize, ordinary rationality frequently has us combine three dimensions in our system of reasons: instrumental, axiological, and cognitive reasons. The latter dimension bears on our knowledge, our capacity to learn.1 To test the necessary and sufficient aspect of this way of modelling a system of reasons, Boudon multiplies the examples. One of his favourite examples is first presented during his inaugural lecture at the Royal Society of Canada, in Ottawa, on November 15, 2001. He then regularly returns to this (Boudon 2007, p. 285, 2012, pp. 139–141). It is a matter of reprocessing, with his decoding model, the data provided by the World Value Survey, a forty-year investigation, in five successive waves, in most countries of the world. Let us just examine one table concerning France, in 2000, an index of tolerance measured by a question on the rejection of a neighbour. The issue behind this is to detect racism, antisemitism, islamophobia, xenophobia, homophobia, etc. Yet the question of nonacceptance of the neighbour is subtle. A priori, respondents to the questionnaire admit the other more easily as a neighbour than as a colleague at work or as a future son-in-law or daughter-in-law. Consequently, their possible rejection is that much more revealing.

1 From this point of view, axiological rationality can be considered as the application of cognitive rationality to normative problems (see Boudon 2007, 2008). To further refine the three dimensions of ordinary rationality, instrumental rationality is doubtless also derived from some of our cognitions.

Ordinary Rationality Theory (ORT) According …

231

Investigation of values: Tolerance (France, 2000) Rejection of the neighbour

% of yes

– – – – – – – – – –

−8% −9% −13% −16% −17% −18% −24% −29% −45% −50%

Large family Other race Foreigner Sick with AIDS Emotionally unstable Criminal Homosexual Extremist Drug addict Alcoholic

These X opinions are a result of respondents’ S systems of reasons. According to the ordinary rationality theory, one must then reconstitute the dimensions: instrumental, axiological, and cognitive, which form the basis for the reasons for these choices. Indeed, respondents combined these elements. (1) I’m familiar with this, and this neighbour seems to me to be responsible for his or her behaviour: cognitive (2) This behaviour is reprehensible with regard to a principle: axiological (3) This has negative consequences for me: instrumental It must be noted that these reasons come up. Furthermore, it would seem that these are in descending order. – The large family is viewed positively because: (1) the situation is familiar; (2) and, moreover, it is approved; (3) even if it can prove noisy, so disturbing. Therefore, respondents are more cognitive and axiological than instrumental. – Alcoholics and drug addicts are poorly seen because: (1) we believe we understand the situation while holding them responsible; (2) we disapprove of these excesses; and (3) this could be disturbing. Here, all the reasons weigh against this type of neighbour. – At the top of the table, racism and xenophobia are weak, with respect to this degree of involvement with a neighbour because: (1) the other does not choose his or her skin colour or country of origin; (2) cohabitation seems accepted; and (3) there is no insurmountable disturbance. – The emotionally unstable individual is considered to not be responsible, therefore, is quite well tolerated, even if at times annoying. Here again, we find this hierarchy of reasons, where the cognitive and axiological weigh more heavily than the instrumental.

232

J.-M. Morin

– The individual with a criminal record is perhaps not more dangerous, but is condemned axiologically. The instrumental dimension also plays a role, if this person appears threatening. – The extremists on the right or left are not disturbing any one since they rarely lobby their neighbours, but they are condemned for axiological reasons, above all. – The homosexual neighbour is not rejected for instrumental reasons. Fortunately, this level of homophobia has fallen from 24% in 2000 to less than 10% in 2020. Thus, one discerns an evolution, above all axiological and cognitive. Incidentally, this indicates that the educational campaigns over twenty years have borne fruit. – The case of the neighbour with AIDS, who was worrisome, especially for cognitive reasons of ignorance of the exact means of transmission before 2000, no longer comes up in recent questionnaires. Respondents are now tolerant on this theme, as a result of acquired knowledge. Ultimately, the test is convincing. The three dimensions cover the essential aspects to understand each individual’s ordinary rationality. Thus, a social phenomenon as important as the tolerance of another is explained, using this example of the rejection or acceptance of the neighbour. Furthermore, we note that the instrumental dimension weighs much less heavily than anticipated. Finally, individuals are not always utilitarian, even selfish. They may also have moral principles, mixed with bounded scientific knowledge. In short, they have axiological and cognitive reasons which often outweigh purely instrumental reasons. Clearly, for the individual, here the just and the true take precedence over the useful.

2

The ORT Internally: Neither Affective, Traditional, nor Biased

While Boudon enriches Weber’s model, in adding a cognitive dimension to instrumental and axiological dimensions, he eliminates elements which he considers less decisive. He does not pick up on the affective and traditional dimensions highlighted by the great German scholar (Weber 1922).2 Indeed, it seems to him that these dimensions generally boil down to the three already mentioned which constitute the whole of ordinary rationality and which thus seem sufficient. 2 Weber already states that emotional or traditional behaviours (habits) are borderline cases. These are more reactive forms of behavior. As a result, they are located at the limit of rationality.

Ordinary Rationality Theory (ORT) According …

233

For the affective dimension, Boudon frequently explains his omissions. Indeed, he explicitly swims against the tide in considering the feelings which permeate human action, indeed which underly all relations, secondary. He is even ironic about this “romantic” thought where the EQ [Emotional Quotient] has a tendency to replace the IQ [Intellectual Quotient] (Boudon 2009: 151). His argument is simple. In his view, the affective aspect is merely a consequence of reasons already mentioned. Therefore, these reasons are sufficient to understand the individual and to explain the seeds of choices and actions, and the emotions or attitudes which develop subsequently. His favourite example is that of the indignation we feel when we observe a strong individual taking advantage of someone weaker. This feeling that overcomes us emotionally is very clear. It is not to be confused with any other: fear, joy, etc. It reinforces the determination to then pass judgement, make a decision, or act. For all that, this is assuming that we have already done the preliminary rational analysis. This indignation arises because we have observed the scene, noted a power imbalance, and then an abuse of power, and have sided with the weakest. In short, we have drawn upon the usual system of reasons: cognitive, axiological, and instrumental. In summary, if I see a thief steal an older woman’s purse, I will first identify the situation, and note that it is abnormal and that it violates a principle which is dear to me. If necessary, I will imagine that I would not like to be in the place of the older woman, even think that society cannot function on such a basis. Only then is the time for indignation. Others will experience fear if they have reason to feel threatened themselves. Yet others will feel a form of approval if they have cause to identify more with the aggressor or to agree with such a transgression of the usual established social order. The example is powerful. The analysis appears generalizable. With emotions following reasons, reasons are enough to reconstitute the process at the source. Consequently, it is not necessary to include emotions in the ordinary rationality theory. The traditional dimension proposed by Weber will undergo the same editing by Boudon. Certainly, some collective actions or beliefs are transmitted from one age to another. De facto, the reasons we invoke are often: “we’ve always done it like this,” through force of habit, even a respect for the legacy of our elders. Nonetheless, it also seems there that these elements could be reclassified as of secondary importance, like emotions. On one hand, we are completely capable of changing our behaviour or institutions which no longer appear suitable. This only indicates that we evaluate them regularly and that we have reasons to either change them or allow them to continue as they are. On the other hand, it is most often possible to reconstruct the reasons which led the elders to establish these norms, rules, and rites in their

234

J.-M. Morin

particular context. Consequently, all this could become routine and transform into new habits. For all that, there is meaning behind the rite; behind the established rule, the reasons for having installed it remain; behind the norm, there remains arbitration which had led to its adoption. To rediscover all these reasons behind the measures enacted, one may simply take the example of traffic lights. It is more complex to take examples such as that of money or that of language. In all cases, one thing is certain: behind the traditions, costumes, usages, and rules, there are always reasons. Once again, it is these that the ordinary rationality theory is happy to reconstruct. In other words, analysis of reasons for tolerance of the neighbour today may also apply effectively to the explanation of phenomena arising from a past which seems further away: traffic lights, money, and language. Moreover, there is nothing more vibrant than tradition. In general, it evolves all the time. Weber recognizes this when he describes a “disenchantment of the world,” characterized by “rationalization.” As with neglecting emotions, not taking into account tradition, indeed habits, in the forefront, remains controversial among sociologists. In France, Pierre Bourdieu proposes a model which highlights habitus. This includes these behaviours, born of will, which become second nature, routine. For example, by riding a bicycle, I end up no longer falling, to the point where then, even if I haven’t been on a bicycle for a long time, “I never forget it.” In fact, I cannot even unlearn it. If I would like to lose my balance, I couldn’t. However, in the beginning, I had trouble managing it. Yet, a sort of conditioned response has developed. This example shows that habitus exists. Perhaps they may often be found in cultural heritage phenomena among social classes, selected by Bourdieu: someone newly rich will maintain the habits of one who used to be poor. However, it is risky to present habitus as the key to all behaviour. Participation in social life being more complex than bicycling, it can happen that my past is not a reliable predictor of my present. I may even have reasons to change and the possibility of managing to do so. Again, it is then the ordinary rationality theory which is proposed to reconstitute our reasons, ahead of emotions and habits, affections and traditions. Additional dimensions, such as that of cognitive biases, have been suggested more recently, beginning in the 1970s, by such authors as Tversky and Kahneman (1982). There again, Boudon proposes considering these refinements as of secondary importance. Constructing a model is not simply a matter of adding the maximum number of factors. On the contrary, it is a question of bringing out the strict minimum of salient features, here of human rationality. This sufficient minimum can be found in the ordinary

Ordinary Rationality Theory (ORT) According …

235

rationality theory. Thus, introducing cognitive biases here seems superfluous. Boudon already concludes this in The Origin of Values (1995, p. 173). He subsequently reiterates this (2009, p. 13). Of course, our thinking may evolve and our cognition may be biased by: availability, anchoring, and representativeness. For example, a traveller arriving in England who comes across a redheaded woman will perhaps think that: “all Englishwomen are redheaded.” However, he will doubtless subsequently encounter others with different coloured hair, which will lead him then to qualify his judgement. Of course, at times, one can’t see the forest for the trees. Naturally, a straight stick partially submerged in water may appear curved. Yet, cognitive reason ultimately corrects the biases which induce errors in our thinking. In short, it is not because we sometimes persuade ourselves of unsubstantiated ideas that our entire reason is intrinsically biased. Ultimately, reason is a little like this vehicle which often delivers us to the right destination but which sometimes leads us into the ditch. In the latter case, the guilty party is the driver rather than the vehicle. Therefore, it is appropriate to incriminate the individual, without assuming that his or her reason was flawed. Perhaps there is an art to driving one’s reason effectively? Regardless, there is a sociological method to correctly reconstitute the ordinary reasons of each individual. This is what the ordinary rationality theory formalizes. So, it is superfluous to include in this the few biases which will eventually be overridden.

3

The ORT Externally: Between Collective Effects and Social Contexts

In the ordinary rationality theory, Boudon proposes formalizing the internal system of reasons of every individual in his or her daily life. Nonetheless, one must link this internal system of reasons to two inseparable external elements to provide a thorough sociological explanation. – On one hand, every social phenomenon must be analysed as being the collective effect of actions, certainly motivated by individual reasons; – On the other hand, these reasons must be understood, with the meaning they had for individuals in their particular context. In other words, individuals’ reasons fall between the collective effects that they provoke and their social contexts. Incidentally, we realize the misinterpretation of detractors of methodological individualism, when they claim that this method reduces everything to the individual. Certainly, reasons are

236

J.-M. Morin

individual. Yet, on one hand, the method succeeds in explaining collective effects and, on the other hand, it starts from social contexts to understand the reasons. In short, we go from macroscopic to macroscopic, via microscopic, referring to reasons and actions. Boudon formalizes this twofold link in an equation S = f [a (r, C)], formulated in Theories of social change: A Critical Appraisal (1984, p. 40) and refined in Methods in sociology (Boudon and Fillieule 2002, p. 75). The meaning of this equation is that: each S social phenomenon is considered the collective effect f, of actions a, which are due to reasons r, in context C. One will have noted that the macroscopic variables S and C are in capital letters. The microscopic variables a and r, as well as f, which formalizes the emergence of the collective from the individual, are not capitalized. Above all, one observes here the central place of reasons. They are at the heart of the collective contextual action. In summary: Explaining a social phenomenon S Entails understanding the reasons r Between collective effects of actions f(a) And social contexts C S = f [a (r, C)]

The collective effects f(a) come last for the actor but first for the sociologist. The actor, situated in a given context, develops reasons which trigger his or her actions or beliefs, ultimately, with collective effects. The sociologist, observing at the end of the story, starts from collective effects to then goes back to the actions and reasons behind them. We are aware of the difficulty of this reconstitution. Everything would be simple if the effects obtained were always consistent with the reasons expressed. In fact, nothing is less certain. Sometimes, individuals obtain results different from their initial goals. Even more, the collective composition of their actions often triggers unforeseen outcomes, even undesired effects. Pollution, inflation, and viral circulation of a rumour are examples of these largely undesired consequences of intentional action. Boudon (1977) endeavours to index these effects in The Unintended Consequences of Social Action, following in the path of Merton (1949), his professor and friend from Columbia. We observe that there are often effects produced which are far from the initial intentions. A prime example of that is the traffic jam. While everyone hoped to move along the autoroute smoothly, they all find themselves unable to move, because there are too many in the same place at the same time, without prior arrangement. This example reveals

Ordinary Rationality Theory (ORT) According …

237

a perverse effect and neutralization: its result is negative and it leads to immobility. There are also sometimes positive effects, nice surprises. There are still amplification effects, where “the beating of a butterfly’s wing awakens a sleeping volcano.” In short, everyone’s ordinary reasons lead to actions, then numerous and often unexpected collective effects. When we reach the end of the chain, as with the sociologist observing the results, only an in-depth investigation allows us to trace the observed effects back to the reasons behind them. The social contexts C are in the forefront for the actor but come last for the sociologist. When the latter comes to the reasons behind this, through the investigation which begins with the effects obtained, the first difficulty has been resolved. Then, it remains for the sociologist to understand the social actor’s reasons more precisely. Yet, these reasons are only meaningful in their context. This is the second difficulty for the sociologist. After having explained the social phenomenon as an emerging effect, he or she must comprehend the reasons leading to the actions. This entails reconstructing the context. For example, it was determined that an influx of migrants is due to individuals’ desire to flee the overwhelming poverty of their countries of origin: one moves from effects to reasons and contexts. Yet, nonetheless, a French public servant who is a sociologist has difficulty understanding the reasons of someone who comes from far away and who, until now, was living on less than two dollars a day: how then is this sociologist able to move from reasons to such contexts, in order to fully grasp the meaning or issues behind these actions? In The Logic of Social Action, Boudon (1979) distinguishes different types of contexts to effectively understand the reasons at play. – In the contexts of functional interactions, each expects the other to play a role: client and supplier, leader and subordinate, student and professor, doctor and patient. – In contexts of interdependence, this is no longer the case: the decisions of individuals whom I do not know and who overlook me still have an impact on me. – There are also static or dynamic contexts. Boudon multiplies the examples of systems of roles or contexts of interdependency, whether static or dynamic, to show to what degree reasons are always contextual. He often returns to a famous example of Coleman’s. The latter shows that doctors who have reasons to want to prescribe a new medication will not act in the same way, depending on whether they are in the context of a hospital or in that of a private office. In the former

238

J.-M. Morin

case, doctors wait for instructions from their head of department. In the latter, they must decide on their own, without even being able to rely on the opinion of doctors in other offices. These are certainly colleagues but, above all, rivals. This creates an interdependence among them rather than the system of roles observed in the hospital. Consequently, medication will not at all be distributed in the same way in these two contexts. Prescriptions shift from zero to a hundred per cent in the hospital, once the head of department has had the opportunity to decide. The distribution curve is much more progressive in city offices. Certain doctors start early. Others will never prescribe the new particular medication. It is as if there were a vanguard, followers, and then those who resist novelty right until the end. Yet, it is the same reason which remains key for everyone: wanting to care for and heal patients. However, with contrasting contexts, the actions and their effects prove to be very different.

4

The ORT Externally: Neither Naturalism nor Constructivism

With the ordinary rationality theory, Boudon establishes an inextricable link between individual reasons, social contexts and the collective effects produced. Consequently, he dismisses two nebulous theories in Montaigne’s spinning wheel: A theory of belief (2013, Chapter 1): – On one hand, the theories which underestimate reasons, as if the initial context erased any move forward. First, one finds sensualism, and then positivism and, above all contemporary naturalism, in various forms. – On the other hand, the theories which overestimate reasons, as if the contexts or effects meant nothing, the real returning to the purely rational. Above all, one finds contemporary constructivism in multiple forms. The theories which underestimate reasons follow one another, in three waves. – The first wave arrives in the XVIIIth century. Sensualism claims that our reason is negligible since we have no innate ideas. Everything which dwells within us is acquired, and depends entirely on the social context. Our senses, sight, hearing, touch, smell, and taste, reflect this context. This sensualism has political consequences. Humans are blank pages on which

Ordinary Rationality Theory (ORT) According …

239

everything has yet to be written. No individual reason may come to contradict this. Thus, many revolutionaries rely on this purported collective malleability. They want to change society by stating that no individual rationality will thwart this plan. – The second wave arrives in the XIXth century. Positivism claims that our reason is perhaps there but unobservable. Yet, there would only be science of what is observable. Physicians are taken here as models. They observe external phenomena, from the largest to the smallest scale. From the object plunged in liquid to the falling of an apple, observation allows for the reconstitution of immutable laws, even the prediction of what is to come. Going further, positivists carefully eliminate any psychological component in the explanation. In short, they avoid reasons. Obviously, the apple falls, not driven by reasons (of a spirit of the forest?) but due to the law of gravity. This is perfectly based on physics. The problem in sociology is that a human being is not an apple or a liquid. Certainly, I stay home “because I’m sick,” the positivists will say, referring to an objective cause. But I also stay home “to heal,” one could add, invoking a very ordinary reason. These are the reasons which the positivists do not wish to consider. In physics, they are within their rights. In social science, they are wrong. Moreover, even in physics, recourse to unobservable concepts, for example the concept of strength, is acceptable. Then, why refuse to allow sociologists the certainly unobservable concept of reason? – The third wave becomes a tidal wave in the XXth century. Before this, for the sensualists, human beings had no innate ideas or, according to the positivists, their reasons were unobservable. From then on, human is moved by forces completely beyond its control: economic, structural, cultural, and neuronal. There are so many variants of this programme. It is said to be “naturalist,” because these forces are in external nature and never in reasons inside the human being. For a supporter of the Marxist theory, the relationships of production rule the world. In short, everything is in the material causes of the economy. For a structuralist, without us being aware of this, underlying structures determine all our behaviours. In short, everything circulates like words in language or a code in an algorithm. For a culturalist, everything is enshrined in a culture of belonging. For a fan of “neuronal man” theories, one only has to observe the connection of neurons in the brain, using medical imagery, to know what constitutes the person; for example, one has merely to photograph a “math whiz.” The point is that there is no need to go seeking the reasons that the individual is learning mathematics. It all comes down to connections.

240

J.-M. Morin

Boudon is very disturbed by the success of all these doctrines which underestimate reasons. They include those of: Condillac the sensualist,3 Comte the positivist, and then Marx, Lévi-Strauss, Foucault and Bourdieu, as well as Changeux, among the naturalists, in some versions, including: material, structural, cultural, and neuronal factors. Unfortunately, these incorrect perceptions often reinforce one another: “The neuronal man and the Zombie, unwittingly manipulated by the habits inflicted on them by social structures, would thus constitute two sides of the same coin” (2013, pp. 47–48, note 7). The theories which overestimate reasons perhaps emerged in reaction to those which underestimate them. It is always the case that the pendulum swings too far in the other direction. The movement is driven by a claim: the real is rational. This leads to an omission: of the weight of contexts or the strength of emerging effects. The constructivist asserts that reality is entirely a “social construction.” We shift from: nothing is due to reasons to: everything is only due to reasons. Faced with the success of the naturalist programme, we may certainly applaud the beneficial reaction of the constructivist programme. Nonetheless, excessive versions—according to which, if everything is constructed, nothing is true—emerge which may even lead to stalemates. One precursor of this trend is doubtless Kant (alongside Marx and Nietzsche, who may have inspired the radical version of constructivism), when he indicates that a phenomenon can only be grasped as such but that it is always represented in us, through the categories of our understanding. In short, one never has direct access to reality but only to images of this reality, forged by our reason. Applied to sociology, this outline is intended to also clarify scientific knowledge, religion, politics, or morality. Finally, everything is constructed by reason and nothing is given. Nothing is discovered but everything is found. For example, science utilizes laboratory instruments. For all that, is the resulting knowledge a pure construction? This is what some new sociologists of science, from Kuhn to Feyerabend or Latour, contend. We even find the same process claiming that values are pure human constructs. Whether it be in the search for what is true or in the quest for justice, above all, such an approach establishes a strong relativism. Finally, when all is constructed, everything is possible but nothing is certain, to such a degree that what is built asks only to be “deconstructed.” This applies right up to categories which seem the most “natural:” gender, age, ethnicity, and humanity. Even sex differences, of lives or of species, are then to be rebuilt. It is enough to break some representations which are presented as pure stereotypes. 3 Here, Condillac is used as a symbol of sensualism. His work is more complex, as Nathalie Bulle would rightly have me mention.

Ordinary Rationality Theory (ORT) According …

241

Boudon is worried about this other overwhelming wave, even if it is opposed to a pernicious naturalism. In fact, constructivism is also pernicious. Whether reason is nothing or everything, we are missing the essential. – On one hand, reasons are not nothing. They are something. They have a precise, ordinary content: instrumental, axiological, and cognitive. – On the other hand, reasons are not everything. They are linked: to social contexts to which they belong and to collective effects they trigger, sometimes without the individual’s intention. Therefore, it is a priority to propose a third path between naturalism and constructivism. This path is that of ordinary rationality theory. Reasons there have their place.

5

Boudon’s Positioning of the Model: Between Three and Six Postulates

Boudon is not the only one to want to trace a path for the social sciences to enjoy a genuine theory of rational action. John Goldthorpe (1998) inventoried the victors of this combat. He listed four finalists: Simon (1945), Popper (1957), Gary Becker, and Boudon. At the end, it seems to him that the most appropriate model would be that of Boudon. This has all the advantages: procedural like that of Simon; supple like Popper’s; and general like Gary Becker’s. Since the 2000s, Boudon himself made an effort to identify the comparative advantages of his ordinary rationality theory, over the closest alternatives. Six postulates allow him to classify closely related theories (2003, 2007, 2009, 2012). Six postulates to classify closely related theories – P 1: individualism – P 2: comprehension – P 3: rationality – P 4: consequentialism – P 5: egoism – P 6: maximization

242

J.-M. Morin

Certain theories deny recourse to methodological individualism to explain social phenomena. – Already, the first postulate consists of taking this step, consisting of considering that collective effects are due to individuals. – The second postulate is based on the fact that the observer is capable of understanding the other. Otherwise, if the individual is at the origin of the collective but no one can understand how, then the attempt at explanation comes to a halt. – If I can understand the other, the third postulate indicates that behaviour of the social actor is due to reasons. (The most ordinary are: instrumental but also axiological and cognitive, without needing to invoke other elements: affective, traditional, and biased.) – The fourth postulate is retained by the authors who claim that reasons are always oriented towards consequences, if necessary, collective ones, and to action. This disqualifies axiological reasons motivated by principles and not consequences. – With the fifth postulate, some add that, furthermore, the individual retains no more than the consequences affecting him or her. In short, each person always behaves as an egoist, even if he or she sometimes claims the contrary. This disqualifies some altruistic reasons. – Finally, bolstering the model with a sixth and final postulate, some theoreticians affirm that the consequentialist and egotistical individual constantly maximizes the difference between the costs and the advantages of action. This disqualifies some reasons tainted with ignorance, progressivity or a more limited sense of satisfaction. We see that the postulates accumulate. We cannot adopt one without having already admitted its precedents. Thus, Boudon advises adopting the three first postulates systematically but only adding the following three in certain, less frequent, cases. In doing so, he is deliberately quarrelling with the detractors of methodological individualism who accept none of the postulates. Yet he also risks falling out with important allies. They are those who form a common front within the extended intellectual family of methodological individualism but who wish to systematically adopt the six postulates, or at least five of them. It is interesting to position, though in a perfunctory fashion, the major authors with whom he is engaging in debate throughout his work. With the cartography of accumulated postulates, this provides the following indication.

Ordinary Rationality Theory (ORT) According …

243

Positioning of the ORT According to Boudon Based on six accumulated postulates – – – –

None: Marx, Comte, Freud, Lévi-Strauss, Foucault, Bourdieu, Changeux P.1: Durkheim (Sometimes) P 1 + P 2: Weber P 1 + P 2 + P 3: Boudon following Tocqueville, ORT = ordinary rationality theory – P 1 + P 2 + P 3 + P 4: Merton – P 1 + P 2 + P 3 + P 4 + P 5: Simon, Crozier, BRT = bounded rationality theory – P 1 + P 2 + P 3 + P 4 + P 5 + P 6: Coleman, Becker, RCT = rational choice theory

Between a connection with the comprehensive sociology of Weber and a closeness to the more flexible functionalism of Merton, Boudon thus finds himself in keeping with Tocqueville, with the ordinary rationality theory. He is again close to certain writings of Durkheim when the latter was no longer claiming a youthful positivism. We find that he is virulently opposed to the horde of naturalists, those who deny even the first postulate. Yet, in a reversal, he is also opposed, but less bitterly, to the utilitarians who claim that postulates P5 or P6 are generally applicable and must be adopted systematically. Thus, he finds himself confronting friends such as Michel Crozier (1977) in France or James Coleman (1990) in the United States. For Boudon, utilitarian, even egotistical and maximizing behaviour merely occurs in particular cases. Certainly, individuals have interests but they also have values and knowledge. In short, one must integrate all the reasons. They are instrumental sometimes. However, one should never neglect axiological or cognitive reasons. The bounded rationality theory (BRT) neglects axiology. Furthermore, the rational choice theory (RCT) denies basic cognitive acquisitions. Individuals are often satisfied with a minimum. Even supposing that they sometimes want the maximum, they are rarely capable of this, lacking the required knowledge. Certainly, we can establish some nuances between the BRT which stops at P5 and the RCT which goes up to P6, in particular, in comparing Simon and Coleman (Maclouf 2022). Yet here, definitely, one must adopt Boudon’s ordinary rationality theory (ORT). For the author, this is the one which covers all the possible reasons, in a given context and with precise collective effects.

244

J.-M. Morin

General Model of Behaviour, From Ordinary Rationality Theory

Ordinary Rationality

Bounded Rationality, Rational Choice

Boudon takes great care to distinguish himself, on one hand from theoreticians who fail to take account of individuals and their reasons but, on the other hand, also from theoreticians who attribute exclusively consequentialist, egotistical, even maximizing reasons to individuals. It is because he is refusing to be caught between Charybdis and Scylla. – On one side, the detractors of methodological individualism claim that individuals are nothing, that their behaviours are incomprehensible, and that their reasons are unobservable. – On the other side, too often, the defenders of methodological individualism claim to be reducing individuals to purely self-interested, egotistical, and maximizing persons. This is usually such a caricature and so unrealistic that the detractors have no difficulty discrediting this. For Boudon, the ordinary rationality theory constitutes the right compromise in presenting an efficient explanatory model. The issue for him is to rebuild sociology as a science, while incorporating results useful for our life in common.

Ordinary Rationality Theory (ORT) According …

6

245

The Adjustment of the Model by Scholars Close to Boudon: Reduced or Enhanced Versions

We consider that a model is adjusted, not when we can add nothing further, but when we can take nothing more away. On one side, it is necessary to develop it enough, so that it is not too reductive. On the other side, it is sufficient to circumscribe it in this way without going further; otherwise, it becomes too complex. From this perspective, we may observe that some sociologists, a priori close to Boudon, may go in three possible directions. – Some want to reduce the model, which is considered already too complex. They are content with a version of homo economicus, driven solely by his or her own interests. Therefore, they simply adhere to the RCT. – Others wish to present an “enhanced” version of the model. They judge that one or two variables, which have been too overlooked, should be added. – The third group considers that the version proposed by Boudon has been adjusted. However, they suggest sometimes incorporating a few nuances. The position of being satisfied with the rational choice theory (RCT) is widespread. Boudon himself was approached by Coleman to take this approach. Gianluca Manzo (2012, 2020) is committed to this direction. Specializing in multi-agent simulations, he considers, following Coleman (1990), that rationality reduced to agents’ interests is sufficient. Implicitly, there is no need to refine the modelling in developing a complete ordinary rationality. In the situations studied by Manzo, it is justified. But, moreover, this common position generally leads to an amalgam of methodological individualism and utilitarianism going as far as the RCT. This is what Francesco Di Iorio (2020) and Alban Bouvier (2020) are lamenting. In a 2020 issue of L’Année Sociologique devoted to this theme, they vigorously denounce all these forms of reductionism: first, psychological and semantic; then that imported from analytical philosophy. The desire to present an “enhanced” version of the model of ordinary rationality is rarer. In France, it is supported by two sociologists, who were initially close to Boudon. Gérald Bronner (2020) considers that it is necessary to even more resolutely draw nearer to cognitive psychology, indeed neuroscience. In particular, he proposes incorporating cognitive biases, following Tversky and Kahneman (1982). He references, doubtless ironically, this management review which lists as many as 150 cognitive biases. As human cognition

246

J.-M. Morin

handles approximately 7 concepts at a time, managers will have difficulty mastering such a list to improve their performances. More seriously, Bronner thinks that Boudon was too severe in his criticism of this dimension of biases. In turn, Pierre Demeulenaere (2020) admits that beliefs and actions are based on ordinary reasons, beyond the narrow rationality of homo economicus. Nonetheless, he proposes enhancing the ORT model with at least two additional dimensions. On one hand, he insists again on the greater importance to be accorded the emotions. On the other hand, he introduces more forcefully the influence of social norms. In Italy, Enzo Di Nuoscio (2020) proposes seeking further additions, in particular with Gadamer’s hermeneutics. Finally, there are those who judge that the ordinary rationality theory proposed by Boudon has been adjusted with a few slight variations. Sylvie Mesure (2020) fully accepts the proposed ORT. She insists on the axiological dimension of rationality, reformulated based on ideal–typical dichotomy of Weber. This dimension of values is important since it constitutes its principal distinction from all the utilitarian theories, within the intellectual family of methodological individualism. Nathalie Bulle (2020) also fully agrees with the method of ordinary rationality. Nonetheless, she stresses that the author meticulously restricts himself to methodology, without ever contributing to ontology. In other words, Boudon consistently refuses to speak out on the human being, the anthropology of the person, for fear of contributing in a metaphysical form. Nathalie Bulle proposes exploring the nature of the individual further, beyond a methodological assumption, all while sharing the refusal to contribute to metaphysics. Thus, she proposes establishing a bridge between Aristotelian and Kantian references. Personally (Morin forthcoming), I share the project of pursuing this ontological path, less prudently with respect to the risk incurred. The quest for the human being, behind the individual, may effectively result in pursuing anthropology. This can only serve to further solidify the process. Certain authors investigating rationality even end up studying theology (Hirschfeld 2018), in this case, Thomist. Always strongly supportive, others close to Boudon encourage the possibilities of connections with Hayek (Nemo 1988), indeed with the entire Austrian school of economics (Fillieule 2010). In sociology, specifically, Bernard Valade (1996) indicates that Boudon’s primary admiration for the “TWD” trio, Tocqueville-Weber-Durkheim, could be tempered by an intensified reverence for Simmel, Tarde, even Pareto. Clearly all these nuances are expressed against a background of support for the ordinary rationality theory. Boudon arrives at the ordinary rationality theory which he places at the heart of any explanation of a social phenomenon. To understand the individual, one must necessarily reconstitute his or her “good” reasons, even if

Ordinary Rationality Theory (ORT) According …

247

the observer is surprised at first sight by certain behaviours and beliefs of the other. Three elements are crucial. – First, these reasons may be: instrumental, axiological, and cognitive. – Next, these reasons can only be understood in the individual’s social context. – Finally, these reasons then trigger actions which have collective effects, sometimes unintended, even by those behind them. Those who wish to reduce or enhance this model all seek to offer the best explanatory efficacy. As refined by Boudon, it already provides an excellent method for grasping social mechanisms. This simple and effective manner of taking social reality into account is intended to face a dual challenge in keeping with the author’s spirit: re-establishing scientific solidity in sociology, and offering useful and reliable results to those who wish to improve our ways of living together in our most ordinary daily lives, as well as during major crises.

References Boudon, R. (1977). Effets pervers et ordre social . Paris: PUF. Transl. (1982). The Unintended Consequences of Social Action. London/Basingstoke: Macmillan. Boudon, R. (1979). La logique du social . Paris: Hachette. Transl. (1981). The Logic of Social Action. London/Boston: Routledge & Kegan Paul. Boudon, R. (1984). La place du désordre. Paris: PUF. Transl. (1986). Theories of social change: A Critical Appraisal . London: Basil Blackwell/Polity Press. Boudon R. (1995). Le juste et le vrai: études sur l’objectivité des valeurs et de la connaissance. Paris: Fayard. Transl. (2001). The Origin of Values. New Brunswick/ London: Transaction. Boudon, R. (2003). Raison. Bonnes raisons [Reason. Good reasons]. Paris: PUF. Boudon, R. (2007). Essais sur la théorie générale de la rationalité: action sociale et sens commun. Paris: PUF. Transl. (2009). Towards a General Theory of Rationality: A Defence of Common Sense. Oxford: Bardwell Press. Boudon, R. (2008). How can axiological feelings be explained? International Review of Sociology, 18, 3, 349–364. Boudon, R. (2009). La rationalité [Rationality]. Paris: PUF, Que sais-je? Boudon, R. (2012). Croire et savoir. Penser le politique, le moral et le religieux [Belief and knowledge. Thinking the political, the moral and the religious]. Paris: PUF. Boudon, R. (2013). Le Rouet de Montaigne: une théorie du croire [Montaigne’s Wheel: a theory of belief ]. Paris: Hermann.

248

J.-M. Morin

Boudon, R. & Fillieule, R. (2002). Les méthodes en sociologie [The methods in sociology]. Paris: PUF, Que sais-je? Bouvier, A. (2020). L’individualisme méthodologique au défi des critiques de la philosophie analytique récente. L’Année sociologique, 70 (1), 45–67. Paris: PUF. Transl. Methodological Individualism facing recent criticisms from Analytical Philosophy. Artificial reconstructions and genuine controversies. This volume. Bronner, G. (2020). Pourquoi une sociologie compréhensive augmentée? L’Année sociologique, 70 (1), 153–174. Paris: PUF. Transl. “Why does comprehensive sociology need to be enlarged?” This volume. Bulle, N. (2020). Trois versions de l’individualisme méthodologique à l’aune de l’épistémologie [Three versions of methodological individualism in the light of epistemology]. L’Année sociologique, 70 (1), 97–128. Coleman, J. S. (1990). Foundations of Social Theory. Cambridge/London: The Belknap Press of Harvard University Press. Crozier, M. & Friedberg E. (1977). L’acteur et le système. Paris: Le Seuil. Transl. 1980. Actors and Systems. The Politics of Collective Action. Chicago: The University of Chicago Press. Demeulenaere, P. (2020). L’individualisme méthodologique et les représentations collectives. L’Année sociologique, 70 (1), 69–95. Transl. Methodological Individualism and Collective Representations. This volume. Di Iorio, F. (2020). Individualisme méthodologique et réductionnisme. L’Année sociologique, 70 (1), 19–44. Transl. Methodological Individualism and Reductionism. This volume. Di Nuoscio, E. (2020). Individualisme méthodologique comme méthode scientifique: théorie de la rationalité, explication causale, herméneutique. [Methodological individualism as a scientific method: theory of rationality, causal explanation, hermeneutics] L’Année sociologique, 70 (1), 129–151. Fillieule, R. (2010). L’école autrichienne d’économie. Une autre hétérodoxie [The Austrian school of economics. Another heterodoxy]. Lille: Presses Universitaires du Septentrion. Goldthorpe, J. H. (1998). Rational Action Theory for Sociology. British Journal of Sociology, 49 (2), 167–192. Hirschfeld, M. L. (2018). Aquinas and the Market. Cambridge: Harvard University Press. Transl. (2021). Thomas d’Aquin et le marché. Paris: Cerf. Maclouf, P. (2022). De Herbert Simon à James Coleman [From Herbert Simon to James Coleman]. Actes de la journée “Fonder l’action organisée” en l’honneur des trente ans du livre de Coleman, le jeudi 15 octobre 2020. Paris: Université de Dauphine. Forthcoming. Manzo, G. (2012). Reason-Based Explanations and Analytical Sociology: a rejoinder to Boudon. Revue européenne des sciences sociales, 50 (2), 35–65. Manzo, G. 2020. Agent-Based Models and Methodological Individualism: are they fundamentally linked? L’Année sociologique, 70 (1), 197–229. Repr. This volume. Merton, R. K. (1949). Social Theory and Social Structure. Glencoe: The Free Press. Transl. (1997). Eléments de théorie et de méthode sociologique. Paris: Colin.

Ordinary Rationality Theory (ORT) According …

249

Mesure S. (2020). Dignité et rationalité axiologique. L’héritage de Raymond Boudon. L’Année sociologique, 70 (1), 175–195. Transl. Dignity and Axiological Rationality, The Legacy of Raymond Boudon This volume. Morin, J.-M. (2020). Boudon. Un sociologue classique [Boudon. A classical sociologist ]. Paris: L’Harmattan. Nemo P. (1988). La société de droit selon F. A. Hayek [The law society according to F. A. Hayek]. Paris: PUF. Popper, K. R. (1957). The Poverty of Historicism. London: Routledge & Kegan Paul. Simon H. (1945). Administrative Behavior. A Study of Decision-Making Processes in Administrative Organization. New York: The Free Press. Transl. (1983). Administration et processus de décision. Paris: Economica. Tversky A, Slovic P. & Kahneman D. (Eds). (1982). Judgment under Uncertainty. Cambridge: Cambridge University Press. Valade B. (1996). Introduction aux sciences sociales [Introduction to the social sciences]. Paris: PUF. Weber M. (1922). Wirtschaft und Gesellschaft. Tübingen: Mohr. Transl. (1968). Economy and Society: An Outline of Interpretive Sociology. Berkeley: University of California Press.

Dignity and Axiological Rationality, the Legacy of Raymond Boudon Sylvie Mesure

It may seem surprising at first to want to reread Raymond Boudon’s work on axiological rationality based on the guiding principle of equal dignity for all, through which the moral equality of all human beings is asserted. Yet, he saw dignity as the central value of our western modernity, dominating our entire moral hierarchy. The normative concept of dignity is indeed omnipresent in the public arena. It is just as omnipresent in discussions on bioethics or the end of life as it is in various claims related to identity, culture or religion, which express requests for recognition. It is mobilized in both feminist struggles for equality and gay and lesbian movements for new rights (“marriage for all”). More generally, the requirement of dignity is what appears to constitute the unity of all contemporary challenges, as noted by Manuel Castells (2013, p. 140): from New York and the Occupy Wall Street movement to the Spanish Indignados, and the watchwords that were heard on Tahrir Square in Egypt or on Bourguiba Avenue in Tunisia (Béshir Ayari 2011) during what was rightly or wrongly labeled the “Arab Spring” (“bread, work, democracy ,

This article was originally published in French: «Dignité et rationalité axiologique. L’héritage ee Raymond Boudon», L’Année Sociologique, 70(1): "L’Individualisme méthodologique", 2020, pp.175–195.

S. Mesure (B) National Center for Scientific Research, Paris, France e-mail: [email protected]

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 N. Bulle and F. Di Iorio (eds.), The Palgrave Handbook of Methodological Individualism, https://doi.org/10.1007/978-3-031-41512-8_13

251

252

S. Mesure

and dignity”),1 it was the same call for the normative concept of dignity that was expressed, despite the diversity of the contexts. Moreover, this notion seems to have become an integral part of our contemporary consciousness: Beyond the demands related to gender, we now hear those concerning the dignity of the sick, the disabled, the dying, the elderly, the very young, etc., which correspond to so many demands for recognition as part of the democratic logic toward greater equality. This massive recourse in the public sphere to a notion that seemed to have become obsolete after the many questions regarding the figure in the subject or the human being, notably in the French Theory of the 1960s–1970s, has led to a reassessment and revalorization of this concept in many works, particularly in the Anglo-Saxon world.2 And, while few books in France are devoted directly to the concept of human dignity, some authors, following up on the recognition theory developed by Axel Honneth (1992, 2006), refer to it, however, by interpreting that dignity as the opposite of humanity and contempt, to develop reflection on what should be a “non-humiliating” or “decent” society, within the meaning that Avishai Margalit gives it (1996), that is, capable of treating everyone with due respect.3 It was with these two considerations in mind that I reread Boudon’s work on axiological rationality, discovering that he also gave a central place to the principle of human dignity in his thought, although he did not feel the need to place clear emphasis on it, he was so busy, in his defense of the idea of moral objectivity, sparring with his relativist adversaries. The purpose of this article is to highlight the central place of the question of dignity in the Boudonian argument, which, to my knowledge, has not yet been done in the literature. At the same time, the interpretation of the notion of axiological rationality supported by Boudon will also be brought into question. We know that the latter has always defended the idea that axiological rationality has a coherence of its own, regardless of the instrumental rationality dear to rational-choice theorists, who, in their explanation of social action, take into account only the simple interest of the social actors. His methodological individualism was based on a “broad rationality theory” that could also regulate the field of normativity, which he also referred to as an “ordinary rationality theory” or as a “cognitive rationality theory” (Boudon 2011, p. 51).

1 Indeed, the events that took place in Tunisia between 2010 and 2011 were called by Tunisians themselves the “Dignity Revolution.” See on this point, Rainer Grote and Tilmann J. Röder (2018, p. 5). 2 Recent works by George Kateb (2011), Michael Rosen (2012), Jeremy Waldron (2012), Christopher McCrudden (2013) and the Cambridge Handbook of Human Dignity (2014) may also be cited. 3 In particular, see the work of Emmanuel Renault (especially 2017a, 2017b).

Dignity and Axiological Rationality, the Legacy …

253

He considered therefore that axiological rationality was “a variant of cognitive rationality,” asserting in this way that, far from being considered mere reflections of our social conditioning, our axiological judgments were based on reasons that had to be understood and evaluated. In an article published in 2010, significantly entitled: “The cognitive approach to morality” (Boudon 2010a),4 he included axiological rationality in the field of methodological individualism, affirming that the same explanatory source is used to explain both social actions and our adherence to certain value judgments: in both cases, it is a matter of seeing things from the actor’s point of view and understanding the reasons behind them. By arguing that in the domain of morality a distinction had to be made between the more or less powerful reasons we have for subscribing to a particular normative statement, however, he went beyond the simple explanation level to address the level of justification for our normative statements.5 I would also like to question this move at the end of this article, taking as the guiding thread the principle of dignity and its central place in his argument.

1

The Principle of Equal Dignity in the Boudonian Argument

Dignity as the Central Value of Modernity There are many texts where Boudon recognizes dignity as a central value of modernity, and it is not my ambition here to identify them all, especially as those who are familiar with his work know just how committed he was to tirelessly updating his texts to clarify them and improve their presentation. We can simply note that in his 2012 work, Croire et savoir: penser le politique, le moral et le religieux (Believing and knowing; reflecting on politics, morals and religion), he writes, in a strongly Durkheimian tone: “The overriding sacred value today is that of human dignity” (Boudon 2012, p. 268). We recall that Durkheim had, in “Individualism and intellectuals” (written at the 4

I thank Nathalie Bulle for bringing this text to my attention. I thank Gianluca Manzo for submitting to me the hypothesis that, by making axiological rationality a mere variant of cognitive rationality, Boudon seemed to have gradually abandoned the problem of justifying our normative statements in favor of explaining axiological beliefs. In fact, it seems to me that the two perspectives have always been simultaneously present in Boudon’s thought, including in his last posthumous article, which I also thank Gianluca Manzo for pointing out to me (Boudon 2014b), where the question of why and for what reasons people endorse certain values (explanation) always coexists with the question of whether they are right to adopt those values, in the sense that they would have solid and objective reasons for doing so (justification). This second perspective is, of course, discussed in this article. 5

254

S. Mesure

time of the Dreyfus Affair), emphasized the extent to which modern societies think of an individual as someone who “needs reasons” so that his/her “reason bows down before that of others”—an individual with cult status whose “first dogma is the autonomy of reason” and “the first rite,” “free inquiry” (Durkheim 1898, p. 9). It was therefore put forward that, with the retreat of tradition, it was the individual and his/her autonomy that took center stage, an individual whose dignity had to be respected through the creation of fair institutions (Mesure 2017). Boudon certainly remembered this when he wrote the same sentence in Croire et savoir,6 recognizing Durkheim as a true liberal (Boudon 2004a, pp. 121–122) and justifying his own adherence to political liberalism by the fact that he “places mankind at the center of the universe” (Boudon 2003, p. 782). One might also recall other passages in which he claims that “Respect for the dignity of others, regardless of their skills, practices or choices, increasingly appears to be the value that dominates the whole hierarchy of moral values” (Boudon 2006a, pp. 889–890) and that: “Modern morality tends to be reduced to a single principle: respect for human dignity” (Boudon 2012, p. 124). The notion of the dignity of every human being, as a human being, is therefore considered to be eminently modern. It is the result of a long historical path that led to the universalization and democratization of a principle that was once thought of as an aristocratic principle of hierarchy separating an elite of equals from the rest of humankind and was more readily referred to as “honor.” On this point, it is worth reviewing Peter Berger’s now very old text, “On the obsolescence of the concept of honor” (1970), which, because it accurately places the logic of modernity in the universalization of the concept of dignity, still constitutes a compulsory reference point on the question.7 According to Berger, the shift from traditional societies and the Ancien Régime to modern societies would lead from aristocratic honor to democratic dignity. Honor is an aristocratic concept that prevailed in pre-modern, highly hierarchical societies; it is a statutory honor that is a source of solidarity between the socially equal individuals of a group, caste or order, and at the same time a strong demarcation line with respect to those who are considered to be “inferior.” In such societies, people are not apprehended regardless of their status, which is linked to sociolegal institutions that closely circumscribe their lives and identities; this is in contrast to modern societies, in which rightsholders assert themselves in the universality of their shared moral status.

6

Indeed, Boudon explicitly cites Durkheim’s text in The Poverty of Relativism (2004b, p. 84). For a more recent discussion on this point, see also Jeremy Waldron (2012) and Jack Donnelly (2015).

7

Dignity and Axiological Rationality, the Legacy …

255

According to Berger, and we follow him on this point, the logic of modernization has therefore led to a real obsolescence of the concept of honor, which has become alien to our democratic Weltanschauung.8 The idea of dignity, such as it belongs to us all as members of a common humanity, is therefore a product of democratic modernity. This is what Boudon, too, recognizes, not only referring to Durkheim, but also, significantly, to Tocqueville, when he writes: “Equality has become a fundamental value, a ‘general and dominant passion,’ simply because it affirms, in the coded manner that characterizes religious thought, the dignity of a human being as a human being, irrespective of his individual skills, characteristics and achievements” (Boudon 1999, p. 303). It would be unnecessary and tedious to provide more citations, as the notion of dignity is so recurrent in Boudon’s work, right up to and including his last book, which was published posthumously: Le Rouet de Montaigne: une théorie du croire [Montaigne’s wheel: a theory of belief] (Boudon 2014a, pp. 98–100). Instead, let us consider how the normative concept of dignity, in addition to expressing a moral conviction that is deeply rooted in Boudon and which he considers to be a universal value (Boudon 2006a, p. 887), came for him to be at the very heart of axiological reasoning.

Dignity at the Heart of Axiological Reasoning If it is fashionable today to say that we must “have done with the normative approaches to values” (Heinich 2018)9 and give up questioning their objectivity, this is not, as we know, Boudon’s position. He vigorously and tirelessly insisted on the power and role of rationality in the domain of normativity. As from Le Juste et le Vrai: Études sur l’objectivité des valeurs et sur la connaissance (Transl. The Origin of Values), published in 1995 and which constitutes one of his first great forays in the normative domain, he defends, based on reading

8 The distinction between the honor of aristocratic societies and the dignity of the democratic age is also noted by Boudon based on his reading of Tocqueville (Boudon 2007, pp. 124–125). 9 We certainly can admit that the question of the objectivity of values is not of interest to Nathalie Heinich, or perhaps even to sociology as a discipline, at least according to a certain conception of the latter. But is it not important to know whether we are right or wrong to say, for example, that democracy is a better regime than a theocratic regime based on Sharia law? That equality between men and women in all fields, both private and public, is a good thing? That “racializing” human beings by introducing a difference in “nature” between them is reprehensible? There is little understanding of how trying to answer these fundamental questions as best as possible can be the object of such disdain.

256

S. Mesure

Max Weber, the idea of the consistency of axiological rationality. Returning to this idea in a collection of articles published in 1999, he then wrote: My own reading of Weber’s theory of normative beliefs as outlined in his notion of ‘axiological rationality,’ or, if you prefer, the model inspired by Weber’s analyses, can be finally summed up by the following proposals:

1. Adherence to normative beliefs is the result of processes analogous to those leading to adherence to descriptive beliefs. 2. We believe that X is true if there are strong reasons to believe it to be true. 3. Similarly, we believe that X is good, fair, etc. if we have strong reasons to judge it to be so. 4. These strong reasons must be perceived by the subject as having objective value” (Boudon 1999, p. 105). With the intention of stressing the idea that a moral argument can be as strong as scientific arguments, Boudon went on to provide an even more formal or formalized definition of this in his Essais sur la théorie générale de la rationalité (Transl.Towards a General Theory of Rationality: A Defence of Common Sense): “Axiological rationality can be formally defined as follows: an argument system {Q} → N containing at least one axiological proposal and concluding that a norm N is valid, all components of {Q} being acceptable and mutually compatible. It is axiologically rational to accept N if no argument system {Q}’ leading to a norm N’ that is incompatible with N is available and preferable to {Q}” (Boudon 2007, pp. 98, 99). With this model of axiological rationality, which he will continue to deepen right up to his final texts, Boudon intended to defend against both naturalist or culturalist theories and consequentialist and functionalist types of models—the idea that value judgments (and not values themselves)10 can be objectively founded, or, in any case, that some value judgments can be stronger than others because they are based on reasons that are stronger than others. He also invited us to consider, against positivism (Boudon 1997), that normative beliefs, contrary to descriptive beliefs, could not be the subject of

10 Talking about a “values theory” is inappropriate, as Boudon stresses when he refers above all, like Weber, to a valorization theory (Boudon 1998, p. 17). But the distinction between the two is not necessarily so clear. It is indeed difficult to clearly separate the question of the objectivity of values from that of the objectivity of the judgments based on them. What Boudon categorically dismisses is above all an essentialist conception of values, the objectivity of which would be conceived of independently of any judgment (Boudon 1999, p. 11).

Dignity and Axiological Rationality, the Legacy …

257

a scientific demonstration, in the sense that their validity could be demonstrated by a factual verification, but that it was possible to found it through an argument based on a system of reasons that we perceive as objective and shared, or in any case shareable.11 This is why Boudon could invoke the idea of an “argumentative basis of beliefs” (Boudon 1995a, p. 200), insofar as, for him, understanding why X is perceived as good, legitimate, illegitimate, etc., implied reconstructing an argument that seems stronger than the arguments leading to the opposite conclusion.12 There is no need to place any more emphasis on this model of axiological rationality defended by Boudon.13 Let us simply recall that he conceived of his theory of axiological rationality as a particularization of his theory of cognitive rationality (Boudon 2007, p. 63), which he also referred to as a “theory of ordinary rationality” and viewed as the “backbone of social sciences” (Boudon 2010b). Let us also remember that, convinced that one could explain prescriptive beliefs through the same process as that used for descriptive or positive beliefs, he concluded with great optimism, and perhaps a little boldly, “that we can have as much certainty in morality as we can in geometry” (Boudon 1995b, p. 40). As a result, Boudon, as we know, has always relativized the Weberian theme of the “war of the gods,” the polytheism of values and the conflict between the ultimate values, which he described as “a Shakespearean interpretation” of the axiological conflict (Boudon 2006a, p. 890). This was not the Weber that he was so fond of, and he acknowledged that he was not claiming to reconstitute the latter’s thought “as it really was” (Wie es eigentlich geen wesist ) (Boudon 1998, p. 56). While he recognized that there could be undecidable situations between two arguments on the basis of a value judgment, he refused to dramatize the conflict and the axiological dissension (Boudon 2007, p. 65). It seemed obvious to him that some value judgments are stronger than others, and therefore are preferable to others, because they rely on reasons or arguments that are themselves stronger than others, since they are touted as being universally shared—in law, if not in fact. In the domain of moral truths, he considered that arguments invoking the principle of dignity were among them. Because for him the principle of dignity was universal, he considered that when brought into an argument, whether 11

For Boudon: “It is in the nature of axiological judgment that it is perceived by the subject as shared, at least as shareable” (Boudon 1995a, p. 39). Indeed, a value judgment is always accompanied by “potential universalization” (Boudon 1995a, p. 194). 12 We have to be able to justify our value judgments through axiological argumentation and reasoning, whether for ourselves or with regard to others, because any value judgment implies a presumption of objectivity. 13 Here we limit the normative scope to ethics or morality, but Boudon also extended his model of axiological rationality to esthetics (Boudon 1995c).

258

S. Mesure

implicit or not, justifying a value judgment, it could lead us to believe in the soundness of that judgment. Because he saw respect for human dignity as the central feature of modern morality, Boudon thought that this principle could serve as a legitimate criterion for distinguishing between moral theories. More generally, for him, it is through consideration of this criterion of respect for human dignity that the legitimacy of any norm and institution should be assessed, as he explicitly states in his book on the decline of morality, published in 2002: “The dignity of the individual is the ultimate criterion for the legitimacy of any standard of any level whatsoever, whether microscopic or societal” (Boudon 2002, p. 79). The notion of dignity is therefore not marginal in Boudon’s theory of values. It is at the very heart of axiological reasoning. Moreover, it is based on the principle of equal dignity that he legitimizes a number of his personal positions, for example in favor of universal suffrage, (Boudon 2008a, p. 1148), human rights (Boudon 2006b, p. 878), gender equality (Boudon 2004a, p. 96), against slavery (Boudon 2002, p. 77), against the death penalty (Boudon 2007, p. 323), against polygamy (Boudon 2012, p. 46), and against excision (Boudon 2006a, p. 883). With regard to the latter in particular, he stands against the culturalist positions that would justify it in the name of incommensurability14 and the equal dignity of cultures and not of human beings. Despite accusations of ethnocentrism, he argues that the notion of dignity, far from being a purely Western creation, is valid for all. Everywhere, people are aware of their dignity and the respect that is due to them, through rights15 in particular, even if the basic structure of their society, to use a Rawlsian term, does not allow them to express this requirement and to enforce such rights. Convinced of the objectivity of the notion of human dignity, he argues, against the relativistic mood prevalent at the time, that it is possible to judge and condemn cultural practices that appear to be contrary to it (Boudon 2000). We see this with regard to polygamy, excision and any other form of oppression of women, and the point is stressed that while we have

14

Samuel P. Huntington and his book The Clash of Civilizations and the Remaking of World Order (1996) is also regularly targeted by Boudon. 15 Let us recall that it is according to the normative principle of dignity that the Universal Declaration of Human Rights based all human rights, and that this basis has never been called into question; it was used again in the Covenants signed in 1966 and in all subsequent major legal instruments, including the Vienna Declaration and Program of Action of 1993, which recognizes and affirms in its preamble “that all human rights derive from the dignity and worth inherent in the human person, and that the human person is the central subject of human rights and fundamental freedoms, and consequently should be the principal beneficiary and should participate actively in the realization of these rights and freedoms.” See on this point: Morsink (1999), Hugues (2011) and Mesure (2019).

Dignity and Axiological Rationality, the Legacy …

259

had to wait a long time in history for gender equality, “we are still waiting for it” (Boudon 2012, p. 186).16 Beyond the sphere of ethics, it is also possible to show that it is by logically invoking the principle of equal dignity that Boudon justifies his political liberalism. In an article published in 2003: “Why intellectuals don’t like liberalism,” which gave rise to an eponymous book in 2004, he argues that it is “paradoxical that intellectuals are not more attracted by a world view that treats the dignity and autonomy of individuals as core values” (Boudon, 2003, p. 775). And it is always by referring to the principle of human dignity that he seeks to justify his advocacy of a representative democracy, which seems to him to be, from the point of view of axiological rationality, the best possible regime, since it derives its legitimacy from a principle that is constitutive of it. that is, respect for the dignity of all and consideration of every person as a citizen, that is, as someone capable of autonomy, rationality and judgment, that is, as someone capable of claiming rights. Above all, it is against this same normative principle of dignity that it would be possible to assess the various axiological positions. Boudon is confident on this point: in fact, he thinks “that from the moment an axiological innovation emerges, if it seems to strengthen human dignity, it tends to be selected socially” (Boudon 1999, p. 230). He believes in moral progress, signs of which he sees, for example, in the deepening of the idea of equality through the demand for new rights, and which he interprets as so many steps toward the realization of the idea of dignity.17 He also believes in the irreversibility of certain moral ideas which, once they have emerged, and because they seem to tend toward deepening and realizing human dignity, will no longer leave people’s minds, even though historical or social forces may sometimes seem to be going against them and causing them to recede. Thus, he sees moral evolution as a process of “diffuse rationalization” (Durchrationalisierung ), according to an expression he borrows from Weber, during which axiological innovations, such as the emergence of new ideas or institutions, 16

But refuting the charge of ethnocentrism is not so easy, because, while every human being has a sense of dignity, as Boudon rightly argues, this does not mean that that its interpretation is unequivocal and capable of transcending cultural differences. Women in some African societies where the practice of excision is accepted, and even valued, certainly have a very different conception of their dignity from that which should be allowed to every human being as a human being, and through which value should be placed on rationality, autonomy and equality in law for all of humankind. But acknowledging does not entail yielding to relativism. While this interpretation of dignity is not demonstrable, in the sense that it could be proved, it is possible, however, to make arguments in its favor. For a discussion of Boudon’s theses concerning the practice of excision, see the article by Ernest-Marie Mbonda (2012). 17 Boudon (2010c, p. 80) further states that: “Concepts of table-companionship, citizenship, human rights, democracy, separation of powers of the Church and the State, etc. are the result of variations of the program that summarizes the concept of human dignity” (the italics are Boudon’s).

260

S. Mesure

are then rationally selected by the human mind. This is a moral evolution that he also describes as a “program,” the implementation of which is identical to that of the idea of dignity in history: “To clarify the concept of program as I take it here, we can reconcile the history of morality, law and political philosophy with that of science. Science is born of a vague program: to describe reality as it is. The value of this program is indemonstrable because the ultimate values are indemonstrable. Once put in place, this program […] has continued to come to fruition. Like the history of science, the history of morality, law and political philosophy is that of the realization of a program: to develop rules and institutions to ensure the best respect for the dignity of the individual and for his or her vital interests (Boudon 2011, p. 90).

If one was not convinced of the centrality of the idea of dignity in Boudon’s axiological reasoning, the latter text clearly testifies to the contrary interpretation. To be sure, the validity of the program mentioned—the realization of the idea of dignity in the world—is not demonstrable, no more than is the program representing science, the purpose of which is to determine reality: “ubiquitous” and “still incomplete,” these programs “directly guide human activity in all its facets” (Boudon 1999, p. 82) and are to be understood as regulating ideas, the meaning of which is clarified and deepened with the gradual emergence of what we believe to be a partial realization. In fact, on the principle of dignity, Boudon states: “There are no general criteria for ‘respect for dignity.’ The principle of dignity is a guiding idea. Just as the content of the principle of reality is only discovered through the history of knowledge, the content of the principle of dignity is only discovered through the history of morality” (Boudon 1999, p. 247). It is this status of dignity as the regulating idea of moral life that we must now emphasize and clarify further, as it may appear to undermine the thesis that Boudon has always defended, which is that it is possible to achieve the same certainty in the normative field as in the scientific field.

Dignity as the Regulating Idea of Moral Life The principle of equality of dignity for all is indemonstrable in that it is not empirically verifiable (Boudon 1999, p. 75). This is what Boudon concedes, from his early texts on axiological rationality to his final work, Le Rouet de Montaigne (2014a). Referring to the Münchhausen trilemma popularized by Hans Albert, the problem of the famous baron who claimed to have pulled himself out of a swamp by his own hair, Boudon wrote: “This is indeed a

Dignity and Axiological Rationality, the Legacy …

261

trilemma, because a person adhering to any principle is confronted with three equally unsatisfactory possibilities: 1. seek to demonstrate the principle in question based on other principles and ad infinitum, 2. seek to demonstrate it in a circular way based on its consequences, 3. take the principle in question at face value and give up on demonstrating it” (Boudon 2014a, p. 27). Boudon again alludes to this Münchhausen trilemma by reformulating when alluding to Montaigne’s wheel: “To judge appearances that we receive from subjects, we would need a judicatory instrument: to verify that instrument, we would need a demonstration; to verify the demonstration, an instrument, now we are on the wheel” (Boudon 2014a, p. 26). Faced with the impossibility of demonstrating the first principles on which science and morality are based, Boudon excludes a “loyalty” position, for which adherence to the first principles is a matter of faith, as much as he does a skeptical conclusion, arguing that any knowledge would be founded on sand and wind (Boudon 1999, pp. 31–40). In his view, the robustness and fertility of such principles are found to be proven though the progressive realization of the program, the principles of which are the regulatory ideas according to the “diffuse rationalization” process mentioned above. For him, “[…] the end of apartheid, the emergence of the “right of interference” or of international criminal law, the multiplication and increasing role of non-governmental organizations,” and “the detention of Augusto Pinochet in the United Kingdom or the appearance of Slobodan Milosevic before the International Criminal Tribunal” (Boudon 2006a, p. 890) are to be considered as moments in the deepening of a program that consists of “defining the institutions, rules, etc. intended to best respect the dignity and vital interests of every individual” (Boudon 2006a, p. 885). The undetermined and empirically indemonstrable character of the principle of human dignity is therefore not, according to him, an obstacle to the idea of axiological rationality. While circularity is inevitable here, insofar as it is through consideration of moral evolution that the fertility of the principles that seem to guide it is experienced, and as this evolution can in turn be assessed on the basis of such principles (Boudon 1999, pp. 75–79), rationality is not devoid of resources in the field of normativity and it seems possible for him to conclude that there are “axiological truths” just as there are scientific truths. This is a position that was described as “hyperrationalist” (Heinich 2017, p. 373), and one we would like to question by confronting it with the thought of another great liberal, Ronald Dworkin who, in his last book: Justice for Hedgehogs (2015 [2011]) also saw human dignity as one of the uppermost values of our democratic modernity (Khurshid, Malik & Rodriguez-Blanco 2018).

262

S. Mesure

Like Boudon, Ronald Dworkin believes in the objectivity of values. Like him, he refuses to give axiological dissensus an incommensurable dimension and a tragic character. Also like him, he believes that it is through a rational confrontation of moral theories, and through argued reasoning, that it is possible to achieve any kind of moral truth. Finally, like him, he considers that “for an axiological judgment to be true, there must be a reason that makes it true” (Dworkin 2015 [2011], p. 131). By contrast, however, he takes for granted Hume’s principle that states that value judgments can never be based on judgments of fact (Dworkin 2015 [2011], p. 58) and which, when “properly understood,” does not seem to him to strengthen normative skepticism, but rather to underline “the independence of morality as an autonomous province of knowledge with its own criteria for analysis and justification” (Dworkin 2015 [2011], p. 30). Far from any scientific verificationism, he thinks that moral argumentation fundamentally falls within the interpretative register and that the normative concept of dignity must be understood, like all normative concepts and unlike other types of concepts, such as the descriptive concept of a lion, or the geometric concept of a triangle, for example, as an interpretative concept insofar as it contains a value and already contains a value judgment, and therefore an interpretation of its meaning (Dworkin 2015 [2011], p. 227). But the meaning of the notion of dignity is far from unequivocal. In fact, it is a concept that is “essentially contested,” in the meaning of Walter B. Gallie, that is, as a concept the very nature of which evokes a multitude of interpretations and controversies (Gallie 1956; Rodriguez 2015). We saw in this notion not only an expression of hidden ethnocentrism and Occidentalism (Chibundu 2012) but also an affirmation of an unfair speciesism by virtue of which the superiority of the human being over the animal was asserted (Rossello 2016). We were able to detect in it a useless and superfluous (Macklin 2003), even dangerous and freedom-destroying18 concept. It is said that it is often used as a knock-out argument to make a final decision in ethical debates and to silence divergent opinions. We were even able to speak of the “stupidity of dignity” (Pinker 2009). However, reason here is not entirely powerless. Refusing to give in to skepticism, Dworkin considers that it is possible to produce a reasoned argument in favor of the principle of dignity. This is an argument, however, that, by virtue of the Hume principle, cannot borrow the rationality regime of scientific thought and obeys its own justification criteria. About the Humean 18 This type of argument was mainly used in France during the famous “dwarf tossing” case, during which a large part of the doctrine, rising against its prohibition by the State (decision of the Council of State of October 27, 1995), placed freedom (here, freedom to work) in opposition to dignity.

Dignity and Axiological Rationality, the Legacy …

263

prohibition, he writes: “Hume’s principle […] is a good principle: Any argument in favor of moral argumentation, or opposed to another moral affirmation – must itself include or presuppose other moral assertions or other moral hypotheses” (Dworkin 2015 [2011], p. 115). Falling within the register of interpretation, moral reasoning links values together and ensures their consistency so as to produce the best possible interpretation. The argued defense of the value constituted by human dignity will therefore consist in unfolding an interpretative chain that shows its consistency with all the other values we uphold by placing it within a tight axiological network, the elements of which reinforce one another. Indeed, an argument in favor of the human dignity value cannot mobilize the idea of equality, especially since it assumes the idea of equal dignity, just as, in turn, a normative interpretation of the idea of equality can only be conducted by calling upon other values that justify and strengthen its meaning. For Dworkin, moral argumentation therefore has a holistic and circular character that is not without repercussions on the objectivity it may claim. Considering that “truth in morality is none other than the best plea in its favor” (Dworkin 2015 [2011], p. 139), he considers in fact that it is not necessary to speak of the truth or falsity of normative interpretations but of their greater or lesser soundness and consistency (Brown 2018; Wood 2018). Here we measure the deviation from Boudon’s position, who, for his part, rejects the “Humean guillotine” (Boudon 2008b, p. 10), or at least the skeptical consequences that can be drawn from it. For Boudon, values are nested in the facts, just as, but distinct from it, axiological rationality is closely hinged on instrumental rationality, so that it is fallacious to conclude there is an insurmountable gap between facts and values: “A system of reasons leading to the conclusion and consequent belief that ‘X is good , fair, legitimate,’ etc.” always includes factual proposals next to normative proposals. However, factual proposals are testable […],” and Boudon concludes, a little further on in his reasoning, and directly against Hume this time: “We can very well draw a conclusion in the imperative mood from a set of reasons that are all in the indicative mood, only one being in the imperative mood” (Boudon 2004c, p. 353). That is why we could speak of “objectivism” with regard to Boudon’s position, suspecting him of wanting to reduce values to facts (Heinich 2017, p. 372). However, this does not do him justice, if we consider his repeated refusal to reduce axiological rationality to instrumental rationality and to the fundamental role he gives to the idea of dignity in moral reasoning. However, if the idea of dignity cannot be factually verified, like the value judgments that call upon it, and therefore if it is not possible to demonstrate “that it is good to try to ensure equal dignity for all” (Boudon

264

S. Mesure

1999, p. 75), the status of axiological reason—which is distinct from instrumental rationality, but one wonders in what sense, ultimately, in Boudon’s thought—seems to be weakened and to lose consistency. Take, for example, the condemnation of excision, an example that Boudon often uses. It can be condemned from a consequentialist point of view (instrumental rationality)19 by evoking the unnecessary suffering it inflicts, the reduction of female pleasure it involves, or the health disasters it can cause; it can also be condemned from the point of view of strictly axiological rationality by mobilizing the value of equal dignity for all human beings. However, if we can test the consequences for women of a practice such as excision, how can we establish the objectivity of the statement that such a practice is contrary to their dignity? The answer to this is not so easy. In any case, it pushes us to renew our questioning of the status of axiological rationality, and of the limits of rationality in the normative sphere. Here, I do not wish to bring into question the substantialist positions of Boudon on the centrality of the idea of dignity in our moral reasoning, and on its importance in the justification of many of our value judgments, or his thesis of the moral advancement of our societies considered in light of this major normative concept. Of course, there is also no question of giving in to the siren calls of relativism by denying the role of reasoning and the power of the best argument that serves our moral convictions. Not all value judgments are equal. We should simply question the power, and also the weakness, of rationality when it occupies the prescriptive or normative field. Raymond Boudon, while admitting that there are many undecidable cases that are difficult to resolve, had no doubt that, in other cases, it is possible to justify our value judgments as objectively as we do our scientific judgments. And especially when these value judgments involve the principle of equal dignity. Slavery or apartheid, for example, seemed to him to be unfair and condemnable because they are the very negation of the idea of equal dignity, which is why we have good reasons, that is, sound and universally valid reasons, to condemn them. We can only agree with him on this point. But can we follow him right to the end when he asserts that we can have certainties in the axiological field that are as sound as those in geometry and convince ourselves, and others in particular, that this is indeed so? We can remain doubtful on this point. Convinced of the power of moral reasoning in the justification of our value judgments, he underestimated its 19 There is, of course, no question here of denying the legitimacy and fertility of a consequentialist approach to values, one that makes it possible to sort them, favoring some over others, in order to guide an action or a policy, but simply of stressing that adopting such a point of view does not resolve the issue of their objectivity.

Dignity and Axiological Rationality, the Legacy …

265

interpretative nature. This is why he had no doubt that we could achieve, for some of our value judgments, the same objectivity as for scientific judgments. While we can put forward reasons in favor of the principle of dignity, or justify by reasoned argument value judgments that are supported by a certain principle, the interpretative nature of any axiological reasoning constrains us to more humility than Boudon would like to admit. Here, reason is certainly not without its advantages, as both Raymond Boudon and Ronald Dworkin vigorously maintained, but if we do not share the skeptic and pessimistic position of Hans Joas concerning the rational justification of ultimate values (Joas 2013), let us admit, nevertheless, that it also has its limits.20 Limits on the type of objectivity it can claim in the field of ethical values and choices. Limits also on the power it would itself have, by virtue of its sole and unique strengths, to elicit the moral assent of others. For we can only hope to convince an individual who has already engaged in a moral conversation and who is ready to listen and to produce moral reasons. We can only hope to convince those who believe in the value of argumentation and rationality.21 Is it possible to convince the initiator of a genocide not to implement his lethal plan? Is it possible to convince a determined racist of the moral equality of all human beings? Could we have convinced the Bataclan terrorists of the value of every human life independent of a single faith community? We have to recognize that in this field, while we are not without a compass—because we have reasons to believe in reason in the field of normativity—we also unfortunately, and often, find ourselves “on the wheel.”

References Bashir Ayari, M. (2011). Des maux de la misère aux mots de la “dignité”. La révolution tunisienne de janvier 2011 [From the woes of poverty to the words of “dignity”. The Tunisian revolution of January 2011]. Revue Tiers Monde, 5, 209–217. Berger, P. (1970). On the obsolescence of the concept of honor. Archives européennes de sociologie/European Sociology Archives] /European Journal of Sociology, 11(2), 339–347.

20

Let us specify here that this is not, however, Ronald Dworkin’s position, which we put to one side on this point, as he was convinced of the strength of normative argumentation and its power to elicit moral adherence. 21 At this level, we find the Weberian theme of the “war of the gods” that Boudon had so much wanted to relativize, if not eliminate. See on this point: Sylvie Mesure and Alain Renaut (1996).

266

S. Mesure

Boudon, R. (1995a). Sens et raisons : théorie de l’argumentation et sciences humaines [Meaningfulness and its reasons: argumentation theories and the human sciences]. In Boudon R. (1995a). Le juste et le vrai: études sur l’objectivité des valeurs et de la connaissance (pp. 161–201). Paris: Fayard. Transl. (2001). The Origin of Values. New Brunswick/London: Transaction. Boudon, R. (1995b). Le juste et le vrai: études sur l’objectivité des valeurs et de la connaissance. Paris: Fayard. Transl. (2001). The Origin of Values. New Brunswick/ London: Transaction. Boudon, R. (1995c). De l’objectivité des valeurs artistiques ou Les valeurs artistiques entre le platonisme et le conventionnalisme [On the objectivity of artistic values or Artistic values between Platonism and conventionalism]. Archives de philosophie du droit, t. 40, 76–95. Boudon, R. (1997). Qu’est-ce qu’une bonne théorie ? Peut-on être positiviste aujourd’hui ? [What is a good theory? Can we be positivist today?] In Ch.-H. Cuin (Ed.) Durkheim d’un siècle à l’autre: Lectures actuelles des “Règles de la méthode sociologique” (pp. 265–287). Paris: PUF. Boudon, R. (1998). La rationalité axiologique [Axiological rationality] In S. Mesure (Ed.) La Rationalité des valeurs (pp. 13–58). Paris: PUF. Boudon, R. (1999). Le Sens des valeurs [The Meaning of values]. Paris: PUF. Boudon, R. (2000). Pluralité culturelle et relativisme [Cultural diversity and relativism]. Comprendre, 1: Les identités culturelles, 311–339. Boudon, R. (2002). Déclin de la morale ? Déclin des valeurs ? [Decline in morality? Decline in values?] Paris: PUF. Boudon, R. (2003). Pourquoi les intellectuels n’aiment pas le libéralisme [Why Intellectuals Dislike Liberalism]. Commentaire, 104, 773-784. Boudon, R. (2004a). Pourquoi les intellectuels n’aiment pas le libéralisme [Why intellectuals dislike liberalism]. Paris: Odile Jacob. Boudon, R. (2004b). The poverty of relativism. Oxford/Cambridge: The Bardwell Press. Boudon, R. (2004c). Une théorie judicatoire des jugements moraux [A judicatory theory of moral judgments]. L’Année sociologique, 54 (2), 327–357. Boudon, R. (2006a). À propos du relativisme des valeurs : retour sur quelques intuitions majeures de Tocqueville, Durkheim et Weber [About the relativism of values: a look back on some major intuitions of Tocqueville, Durkheim and Weber]. Revue française de sociologie, 47 (4), 877–897. Boudon, R. (2006b). Misère du relativisme [The poverty of relativism]. Commentaire, 116, 772–784. Boudon, R. (2007). Essais sur la théorie générale de la rationalité: action sociale et sens commun. Paris: PUF. Transl. (2009). Towards a General Theory of Rationality: A Defence of Common Sense. Oxford: Bardwell Press. Boudon, R. (2008a). La démocratie éclairée [Enlightened Democracy]. Commentaire, 124, 1147–1150. Boudon, R. (2008b). Le Relativisme [Relativism]. Paris : PUF.

Dignity and Axiological Rationality, the Legacy …

267

Boudon, R. (2010a). The cognitive approach to morality. In S. Hitlin & S. Vaisey. Handbook of the Sociology of morality (pp. 15–33). New York: Springer. Boudon, R. (2010b). La rationalité ordinaire : colonne vertébrale des sciences sociales. [Ordinary rationality: the backbone of the social sciences]. L’Année sociologique, 60 (1), 19–40. Boudon, R. (2010c). La sociologie comme science [Sociology as science]. Paris: La Découverte. Boudon, R. (2011), La rationalité [Rationality]. Paris: PUF. Boudon, R. (2012). Croire et savoir : penser le politique, le moral et le religieux [Believing and knowing; reflecting on politics, morals and religion]. Paris: PUF. Boudon, R. (2014a). Le Rouet de Montaigne : une théorie du croire [Montaigne’s wheel: a theory of belief ]. Paris: Éditions Hermann. Boudon, R. (2014b). What is context? Kölner Zeitschrift für Soziologie and Sozialpsychology, 66 suppl., 17–45. Brown, A. (2018). Interpretation in normative domains. In S. Kurshid, L. Malik & V. Rodriguez-Blanco (Eds). Dignity in the legal and political philosophy of Ronald Dworkin (pp. 60–80). New Delhi: Oxford University Press. Castells, M., Khosrokhavar F. & Touraine A. (2013). L’unité des grandes contestations contemporaines [The unity of great contemporary challenges]. Socio. 2: Révolutions, contestations, indignations, 139–167. Chibundu, M. O. (2012). The parochial foundations of cosmopolitan rights. In M. N. S. Sellers (Ed.), Parochialism, cosmopolitanism and the foundations of international laws (pp. 172-211). Cambridge: Cambridge University Press. Donnelly, J. (2015). Normative versus taxonomic humanity: varieties of human dignity in the western tradition. Journal of Human Rights, 14 (1), 1–22. Durkheim, E. (1898). L’individualisme et les intellectuels [Individualism and intellectuals]. Revue bleue, Fourth series, t. X, 7–13. Düwell, M., Braarvig, J., Brownsword, R. & Mieth, D. (Eds.). (2014). The Cambridge Handbook of Human dignity: Interdisciplinary Perspectives. Cambridge: Cambridge University Press. Dworkin, R. (2011). Justice for Hedgehogs. Boston: Harvard University press. Transl. (2015) Justice pour les hérissons. Genève : Labor et Fides. Gallie, W. B. (1956). Essentially contested concepts. Proceedings of Aristotelian Society, 56 (1), 167–198. Grote, R., & Röder, T. J. (2018). Constitutionalism, human rights and Islam after The Arab Spring. New York: Oxford University Press. Heinich, N. (2017). Des valeurs. Une approche sociologique [Values. A sociological approach]. Paris: Gallimard. Heinich, N. (2018). Pour en finir avec les approches normatives des valeurs [To have done with normative approaches to values]. Revue du Mauss, 51, 177–184. Honneth, A. (2006). La Société du mépris. Vers une nouvelle théorie critique [The society of contempt. Toward a new critical theory]. Paris: La Découverte. Honneth, A. (1992). The struggle for recognition. The Moral Grammar of Social Conflicts. Cambridge: Polity Press.

268

S. Mesure

Hugues, G. (2011). The concept of dignity in the Universal Declaration of Human Rights. Journal of Religious Ethics, 39, 1–24. Huntington, S. P. (1996) The Clash of Civilizations and the Remaking of World Order. New York: Simon & Schuster. Huntington, S. P. (1997 [1996]). Le Choc des civilisations. (J.-L. Fidel, Trans.). Paris: Odile Jacob. Joas, H. (2013). The Sacredness of the Person: A New Genealogy of Human Rights. Washington: Georgetown University Press. Kateb, G. (2011). Human dignity. Cambridge MA/London: Harvard University Press. Khurshid, S., Malik, L. & Rodriguez-Blanco, V. (2018). Dignity in the legal and political philosophy of Ronald Dworkin. New Delhi: Oxford University Press. Macklin, R. (2003). Dignity is a useless concept. It means no more than respect for persons or their autonomy. British Medical Journal , 327(7429), 1419–1420. Margalit, A. (1996). The Decent Society. Boston: Harvard University Press. Mbonda, E.-M. (2012). Rationalité, objectivité et universalité des valeurs morales : à propos du débat sur l’excision [Rationality, objectivity and universality of moral values: concerning the debate on excision]. Ethica, 17 (2), 97-120. McCrudden, C. (Ed.). (2013). Understanding Human dignity. Oxford: Oxford University Press. Mesure, S. (2017). Le lien social à l’épreuve de l’individualisme. Le culte de l’individu chez Émile Durkheim [The social bond put to the test by individualism. The cult of the individual in Émile Durkheim’s work]. Revue Internationale de Philosophie, 280, 157–180. Mesure, S. (2019). Cosmopolitisme, droits humains et dignité [Cosmopolitanism, human rights and dignity]. In A. Policar. Le Cosmopolitisme sauvera-t-il la démocratie ? (pp. 47–63). Paris: Classiques Garnier. Mesure, S. & Renaut A. (1996). La Guerre des dieux. Essai sur la querelle des valeurs [The war of the gods. Essay on the quarrel over values]. Paris: Grasset. Morsink, J. (1999). The Universal Declaration of Human Rights. Origins, Drafting and Intents. Philadelphia: University of Pennsylvania Press. Pinker, S. (2009). The stupidity of dignity. The New Republic, May 28, 28–41. Renault, E. (2017a). L’Expérience de l’injustice. Essai sur la théorie de la reconnaissance [The experience of injustice. Essay on the theory of recognition]. Paris: La Découverte. Renault, E. (2017b). Reconnaissance, conflit, domination [Recognition, conflict, domination]. Paris: CNRS Éditions. Rodriguez, Ph.-A. (2015). Human dignity as an essentially contested concept. Cambridge Review of International Affairs, 28(4), 743-756. Rosen, M. (2012) Dignity: its history and meaning. Cambridge: Harvard University Press. Rossello, D. (2016). All in the (human) family? Species aristocratism in the return of human dignity. Political Theory, 45 (6), 749-771.

Dignity and Axiological Rationality, the Legacy …

269

Waldron, J. (2012), Dignity, rank and rights. New York: Oxford University Press. Wood, A. W. (2018). Interpreting human dignity. In S. Khurshid, L. Malik & V. Rodriguez-Blanco. Dignity in the legal and political philosophy of Ronald Dworkin (pp. 161-199). New Delhi: Oxford University Press.

Why Does Comprehensive Sociology Need to Be Enlarged? Gerald Bronner

1

Introduction: A Sociology Founded on Acceptance of Mental Invariants

The tradition of comprehensive sociology as it can be understood if we take Max Weber’s work as a starting point considers the individual as the bearer of a subjectively considered meaning while her/his interactions represent the appropriate level of analytical reduction for those wishing to look at social phenomena as an object for science. The German sociologist placed human rationality’s ideal–typical way of functioning at the core of his definition of sociology. In an oft-cited passage (1971, p. 28), he describes this discipline as “a science that in construing and understanding social action seeks causal explanation of the course and effects of such action. By ’action’ is meant human behavior linked to a subjective meaning on the part of the actor or actors concerned.” The reconstruction of this subjectively aimed at meaning presupposes the existence of mental invariants in the human species. This is a fundamental point because it is the smallest common denominator for

This article was originally published in French : “Pourquoi une sociologie compréhensive augmentée ?” , L’Année sociologique, 70(1): “L’Individualisme méthodologique aujourd’hui: Fondements, problèmes et controverses”, 2020, pp.153–174.

G. Bronner (B) Sociology, Sorbonne University and French Academy of Technologies, Paris, France e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 N. Bulle and F. Di Iorio (eds.), The Palgrave Handbook of Methodological Individualism, https://doi.org/10.1007/978-3-031-41512-8_14

271

272

G. Bronner

any comprehensive method. Boudon (2003) described this smallest denominator as being founded on three postulates. The first postulate (P1) is that of individualism which states that individuals are the only possible substrata of action and supports all actionist approaches. The second postulate (P2) is that of understanding which states that all mental activity can be understood whether it translates into action or not, i.e., the meaning it conveys is never beyond an observer’s understanding. The third postulate (P3) is that of rationality which states that individual behavior can generally be described by the reasons that underpin it even when the actors involved themselves are not clearly aware of these reasons. This conception of rationality does not need to be axiomatized (we know that Boudon condemned this possibility on several occasions as did Weber (1922/1949, p. 125) in Collected Essays on Scientific Theory 1 to suggest a form of universality of human thought. It is easy to find many extracts from Boudon or Weber’s texts in which this universality is claimed to exist beyond time and the culture of Man given that we all know you don’t need to be Caesar to understand Caesar. This universality derives from at least two aspects. The first aspect is that of instrumental rationality, the conventional conception of which since Aristotle depicts the rational individual as a person who uses the appropriate means in a given context to achieve her/his aims. The second is that of cognitive rationality which, as Boudon (1999, pp. 148–149) stated, implies “the validity and compatibility between themselves of the propositions that make up a theory or of the reasons a belief is based on as well as their compatibility with reality.” These two aspects of human rationality can be considered among our species’ cognitive invariants and enable us to understand —in the Weberian sense—social activities in cultures that are very different from the observer’s own. The intention of this article is to show how this sociological tradition can be enlarged somehow by the contemporary contributions of cognitive and brain sciences. Such sciences are far from contradictory to the epistemological intentions of the comprehensive tradition and in fact, enable this tradition to be extended.

1 The criticism of the marginalist school that he offers in Collected Essays on Scientific Theory needs to be understood in this spirit: he criticizes Menger for defending the project of an axiomatized, normative and universal model of rationality and, in doing so, for being influenced by the natural sciences’ almost exclusive interest in the reason to know.

Why Does Comprehensive Sociology Need to Be …

2

273

A Universal Disposition for Understanding

We are generally able to understand the meaning that others give to their actions without having to resort to sociological methods. We thus understand , for example, why a given individual may have become angry in certain circumstances. We could even presume that such anger has its own meaning, without knowing the exact reasons behind it. Neuroscience states that this ability to understand other people’s actions is one of the human species’ specific brain characteristics. As Decety and Jackson (2008, p. 68) pointed out: “In the five to eight million years since the hominid lineage diverged from that of our closest relatives, chimpanzees, one of the important evolutionary changes characterizing humans is the large increase in brain size. This increase is especially the case for those regions of the brain that play an important role in the mechanisms that enable us to understand that our behavior and that of other people is caused by our mental states.” We are therefore said to be able to simulate another person’s point of view to help understand and predict their behavior. Certain areas of the brain are involved in this comprehension skill2 —the temporoparietal region, the anterior temporal lobe and, particularly, the medial prefrontal cortex which is systematically involved in attributing intentions to other people.3 These dispositions are well-established from a very early age—a theory of mind is perfectly established by the age of four and there are serious indications that it is already very solidly in place by the age of eighteen months.4 For example, it has been shown5 that children as young as a year old naturally attributed intentions to objects that seemed self-directing and even made positive or negative evaluations of such intentions depending on how various objects interacted among themselves. Therefore, a kind of system of spontaneous understanding actually exists although this does not mean that we spontaneously practice comprehensive sociology. Nevertheless, this initial observation takes us away from the idea that ordinary knowledge of the social sphere is totally illusory and that the task for sociologists should be to make an inaugural “epistemological break,” as suggested by Bourdieu, Chamboredon and Passeron (1968, p. 37): “Sociology can only represent a science which is truly cut off from common sense as long as it opposes the systematic pretensions of spontaneous sociology with the organized resistance of a theory of

2

See Brunet, Sarfati, Hardy-Bayle and Decety (2000). See Frith (2001). 4 Leslie and Kebble (1987). 5 Premack and Premack (1997). 3

274

G. Bronner

knowledge of the social dimension, the principles of which contradict every presupposition of initial social philosophy.” It would be wrong to think that the sociologist’s sociology is formed in reaction to “popular” sociology in the same way as scientific physics constructs itself in reaction to spontaneous physics. This would be the case if there were the same (epistemological) distance between learned sociology and commonsense sociology as the distance that exists between our natural physics and the physics of physicists. The difference is that more often than not we can use what we know about the reasons for our own actions to identify the reasons behind others’ actions with a fair chance of being right. Thus, we can go on to explain the social facts they cause or have caused. In other words, because the causes of social actions are also reasons, they are not always concealed from sociologists interested in such causes in the same way as the causes of the phenomena physicists aim to explain. This is why Karl Popper was able to state that formulating sociological hypotheses is easier than formulating their counterparts in physics (1986, pp. 172–178). However, he also pointed out that social science is only comparatively easier, that it also has to construct models to explain the phenomena it studies. Furthermore, it is not spared from submitting its hypotheses and models to the test of refutation without which we shall always find what we want: “we shall look for, and find, confirmation, and we shall look away from, and not see, whatever might be dangerous to our pet theories.” (ibid., p. 168). It is therefore an unfounded accusation against comprehensive sociology to profess that it is not scientific because those who represent it do not believe themselves compelled to “break” with the ordinary rationality of other individuals to explain social facts produced by their actions. It is neither more nor less scientific than all the other sciences even though it is based on a method that cannot be exactly the same as the method used in the natural sciences. This is impossible because the social actors at the origin of a fact studied by sociology are neither things nor creatures having absolutely nothing to do with their own actions. However, sociology can also establish itself as a science in the Weberian sense of the term because this kind of spontaneous comprehension inevitably comes up against several obstacles. We propose to define these in the following section with a view to clarifying the definition of the scope of comprehensive sociology and seeing how this scope could be enlarged. This, then, is the idea of an enlarged comprehensive sociology.

Why Does Comprehensive Sociology Need to Be …

3

275

Comprehensive Sociology Is Not Spontaneous Sociology

This mental disposition is very necessary to life in society but it is also sometimes ineffective. In some ways, it is inhibited by certain situations which means we are no longer able to understand others and also presume them to be easily driven by a form of irrationality. In other words, we interpret their actions as the consequence of causes and not reasons. This is particularly the case when we wish to report on actions and beliefs inspired by systems of representation to which we do not adhere to and/or which lead individuals to act in ways we do not approve of. When we observe our contemporaries, their attitudes can sometimes seem as baffling as hieroglyphs whose meaning troubled people so much in previous centuries. Such attitudes are indecipherable and we do not necessarily have as much time and motivation to try and understand our fellow humans on a daily basis as Champollion had when deciphering the Rosetta stone. Why then do others’ forms of logic not seem clear to us in certain circumstances? Here is another way of asking the question which also gives us an idea of an answer—why are we generally less easy on others than ourselves when trying to explain erroneous beliefs or counterproductive practices? We would all readily admit to having endorsed beliefs several times in our lives that we now know to be false (didn’t everyone believe in Father Christmas?). Nonetheless, we would find it unfair to be considered stupid or insane just because of these errors and would have thought so even at the time we believed such things. We know that our approximate and erroneous decisions, our political or religious beliefs, our unfeasible New Year’s resolutions and so forth are or were the consequences of a logical capacity that is partially weakened by a situational effect rather than being the symptoms or effects of an actual temporary form of madness. We know this because no effort is required to reconstruct our own logic at the time as the reasons that led us to this error are often readily available to us. This then is quite simply the information we lack when observing the manifestations of others’ unfamiliar beliefs. The reasons behind them are not beyond our grasp but we do not have immediate access to them and finding out such reasons requires an effort we are not always ready to make. We opt instead to react more rapidly and may decide the proponents of such beliefs to be stupid, insane and/or evil, thus giving in to what Kant (1986, p. 526) called “lazy reason.” This is why comprehensive sociology cannot

276

G. Bronner

only be based on ordinary psychology when attempting to establish robust knowledge.6 Three main realities hinder our spontaneous comprehension and methodological effort is required if we are to reconstruct the subjective meaning that social actors believe in.

4

The Three Obstacles That Hinder Our Spontaneous Perception of Others’ Rationality

Three difficulties face those using the comprehensive method of reconstructing the mental universe of others. The first is that individuals exist in an informational universe which is limited in time and space (the dimensional conditions of human rationality). The spatial limit on our judgment is the easiest to conceive of. This condition is common to all humanity because the information processed by our brains is limited by the capacities of our senses. Also, most often we only have access via our experience to a limited sample of reality which we use as a basis for conjecture. The sociologist of knowledge De Gré (1941) offers a didactic illustration of this. He imagines four individuals placed in front of a pyramid. Each person claims the object is a different color—one says it is blue, another red and so on. The first idea that will come to the mind of the observer asked to comment on this will perhaps be that some of them must be lying or have got it wrong. In this case, this observer would not be taking the spatial limits of the four individuals’ rationality into account. Actually, the individuals placed in front of a different face of the pyramid can only see the color of that side of the pyramid which is in fact multicolored. We can draw two conclusions from this example. First, the four individuals would need to go all round the pyramid to obtain a more objective idea of its color or, in other words, to use a method to free themselves of the limits weighing on their rationality. Second, an observer wishing to understand the reasons behind their errors needs to work in a comprehensive manner to be able to put them down to a form of judgment limited by spatial conditions. Despite the simplicity of this example, it reveals the first pitfall of “spontaneous understanding.” This is because there is an immense scope of social phenomena that can best be explained by taking this limit of our rationality alone into account. The social worlds we live in, whether in 6 Another essential reason which will not be discussed at length in this article is that social actors have absolutely no awareness of the social effects which emerge from their individual actions and which are phenomena of prime importance for sociology.

Why Does Comprehensive Sociology Need to Be …

277

physical or digital reality, preferentially expose us to a certain type of information or argument. It is not impossible to make the required effort to extract oneself from such echo chambers7 but this requires a certain degree of motivation. All comprehensive sociologists know how to integrate this element into their approach to social phenomena. There are many relevant examples in this context and we need only recall the way Boudon8 insisted on the importance of these situational effects several times. We also need to add temporal limits which have an effect on our judgment to the list of dimensional limits to our rationality. Here too there are many illustrations of how our judgment is affected by our temporal condition. Economists have particularly concentrated on this,9 starting with Samuelson (1937) and Shackle (1967). Piatelli Palmarini (1994) gave an experimental illustration of this. Three distinct groups of individuals are asked to plan their lunches. The first group have to plan their lunches for the coming month (on day D, they have to choose their menus for all the other days of the month). The second group plan lunch on a daily basis while the third group also has to plan daily menus but does so immediately after quite a large breakfast. In fact, the idea to an extent is to look at what these individuals anticipate their desires regarding lunch to be in different temporal decision-making contexts. In the first group, whose subjects were asked to plan their meals for a whole month, the results of the experiment show that a varied and balanced program was chosen on average. Conversely, people in the second group opted for a rather monotonous program (for each individual) as these individuals preferred ad-hoc menus featuring food they actually like rather than basing choices on dietary concerns. The third group opted for frugal lunches which is not surprising as they made their decisions on a full stomach. It is hardly a new discovery that we do not know ourselves well or that at the least we see ourselves differently according to the position we put ourselves in to observe ourselves. Also, the future of our desires appears to vary a great deal according to the kind of present we find ourselves in when anticipating those desires. Our consciousness remains definitively limited in terms of time and space and thus our minds never stop trying to make up for these limits by calling on beliefs. Quite often, individuals adhere to all sorts of false ideas that may seem irrational to observers. These observers would think differently about this if they saw that such false ideas come from minds that

7

Regarding the digital sphere, see for example Bessi et al. (2015). For example, Boudon (1986, 1990). 9 Sociologists have focused less on this but we may nonetheless cite the work of Dupuy (1997) or Elster (1986) for example. 8

278

G. Bronner

are limited by their dimensional situation as this prevents them from being able to access all the necessary information. The second variable we need to reconstruct to understand other people involves the cultural limits that influence our rationality. The systems of representations we implement through socialization help us to understand the world but quite often these representations are also an obstacle between us and the world around us. We process information through this interpretative filter and the result does not always correspond to the norms of objective rationality. Exotic societies with very different cultural systems from our own were long thought to be peopled by naïve individuals with childish or even illogical or prelogical mentalities.10 However, when anthropologists worked on understanding these individuals, they of course found their beliefs were the result of culturally situated logics. These observers thus rejected the stereotypical images of slightly stupid “primitives” and instead only saw individuals behaving reasonably given their own representations of the world. For example, people have long been surprised by the stability of the caste system in India. This situation seemed paradoxical as the “untouchables” who, a priori, had the most reason to contest this order (as they made up the lowest level of this stringently hierarchical society) were also the people who were the most likely to respect it according to Max Weber (1971, p. 461). To shed light on this paradox, the German sociologist explained that this system ordered individuals to respect the professional activity prescribed to their caste and carry out the corresponding duties. However, as Max Weber explained, this rule was only followed so well because it was linked to the idea of the transmigration of souls by which people believe they improve their chances of positive reincarnation by respecting the precepts of their caste. Weber highlighted this paradox but showed how it was only a paradox from the standpoint of observers outside the social system concerned. Indeed, given their beliefs, the lowest castes have the most to gain from conforming to the duties linked to their social status. Certain beliefs, therefore, seem absurd or counterproductive to us because we do not make the necessary effort to reconstruct the cultural universe which inspires them. This is such a commonplace element in the comprehensive method that it is not the subject of a great deal of work and few illustrations of the idea can thus be found. The third major limit which affects our rationality and requires methodical reconstruction work involves our cognition and the failings inherent to certain of our inferences. Some of our false ideas are actually a consequence

10

See for example Frazer (1890/1981)or Lévy-Bruhl (1960).

Why Does Comprehensive Sociology Need to Be …

279

of our minds’ “normal” way of functioning. Particular effort is required to perceive this way of functioning which is based on mental procedures that we are so used to they become habits and almost routines. We use them so frequently because they are of valuable service to us, often providing acceptable solutions to our problems. Many of our mistakes come from the excessive trust we have in these mental habits which is why they do not seem totally unreasonable even when their consequences can be either comical or dramatic. Often daily situations are of a complexity that exceeds our cognitive capacities in the short term so we sometimes give in to specious reasoning that is intuitively satisfying but produces false ideas. Comprehensive sociological work requires observers of such occurrences to ask themselves if the beliefs of their contemporaries may result from these predictable ways of functioning of the mind. It is this third point that seems to me to make an enlarged comprehensive sociology a possibility or even a necessity. The two first limits of rationality have been explored by many sociologists or anthropologists, whether they claim to be using the comprehensive method or not.

5

Cognitive Biases: Key Elements of an Enlarged Comprehensive Sociology

Countless thinkers could be mentioned, including Aristotle, Cicero, Bacon, Malebranche, Descartes or Condorcet, who grappled with the task of formalizing the right ways of thinking while distrusting the criterion of evidence and the seductive traps of sophistry. A special place should be reserved in this list for John Stuart Mill (1843/1948). We should also probably add the contributions of Vilfredo Pareto in his Treatise on General Sociology, Daniel Bernouilli’s resolution of the St. Petersburg paradox in 1738, the ideas of Allais (1953) about the model of rational choice and so forth. However, in reality, all these contributions only foreshadow the research carried out at the end of the twentieth century by Amos Tversky and Daniel Kahneman who acknowledge their predecessors. These were Paul Meehl (1954), and his research on the comparison of clinical and statistical predictions; Ward Edwards (1968) who introduced subjective probability studies into psychology in the framework of the Bayesian paradigm; Herbert Simon (1957) and his program for thought about reasoning strategies; Jerome Bruner (1957), who was one of the first to give an empirical illustration of this program; and Fritz Heider (1944) and his pioneering work on the ordinary perception of causality.

280

G. Bronner

In the 1970s, Amos Tversky and Daniel Kahneman inaugurated a program founded on a set of experiments that marked history and led to Kahneman being awarded the 2002 Bank of Sweden Prize in Economic Sciences in Memory of Alfred Nobel. These experiments were carried out in the framework of an ambitious scientific project aiming to shed light on how human thought functions. Their work was followed by hundreds of other projects which all aimed to draw up the cartography of cognitive errors and the intuitions guilty of leading us into them. In the few decades between today and these psychologists’ experiments, all these researchers wrote what seems to me to be one of the most important pages in twentieth-century human sciences. Their proposals have given rise to intense internal discussions11 and yet, curiously, although these works often profoundly transformed several disciplines (psychology, economics, management, etc.), they have yet to make themselves fully heard in sociology.12 Boudon did mention these works as early as 1986 in L’idéologie ou l’origine des idées recues whereas the first collective summary report of cognitive biases was published in 1982.13 However, Boudon only made reference to these in this book and subsequently14 to challenge Kahneman and Tversky’s irrationalist interpretations of their results rather than to attempt to increase the comprehensive model’s descriptive power. This discussion was worth having15 but it probably took the French sociologist’s attention away from the extent of the analytical services all this work on human reasoning could offer comprehensive sociology. Raymond Boudon contested Tversky and Kahneman’s interpretations but did not doubt the validity of their results. Comprehensive sociology was therefore just a step way from enriching itself through the results of a study of cognitive biases, those great invariants of human thought. This article proposes to take this step forward.

11

Some authors, notably Gerd Gigerenzer (1991) or (1995), have shown that changing the presentation framework of conventional cognitive bias exercises—particularly the base rate neglect bias—means the expected error rate can be varied. This change of framework makes the existence of several resolving strategies clearer and enables individuals to resist certain forms of specious reasoning easier. These authors often work in evolutionary psychology and generally contest the idea that these biases express a form of fundamental irrationality inherent to the human species given that they can be related to evolutionary advantages. This discussion would be fascinating to pursue but it is outside the scope of this article and may only be a category of confusion between a form of instrumental and phylogenetic rationality and another form considered in the light of objective standards of reasoning. 12 Part of the explanation for this is simply that a part of this discipline has little scientific consideration for cognitive phenomena. See Bronner and Géhin (2017). 13 Kahneman, Tversky, Slovic (1982). 14 For example, Boudon (1993a, 1993b, 1994). 15 I have carried this on myself in Bronner (2007).

Why Does Comprehensive Sociology Need to Be …

6

281

Cognitive Biases Are Necessary Elements for Comprehensive Sociology

The sociological tradition has shown that the “meaning subjectively believed in” by social actors requires observers to go further than the framework of spontaneous sociology’s skill set, especially when this meaning relates to a particular dimensional or cultural situation. Sociologists may also go beyond this framework when the motives of actors concerned are tangled up in cognitive processes which lead them to make errors of appreciation. As I wrote earlier—not only can sociology take account of the results of the cognitive psychology of error which has referenced around one hundred and fifty different cognitive biases,16 it must do so. This is the case because reference to the cognitive limits of rationality is often highly necessary even if they rarely suffice alone to develop an acceptable explanatory model for a social phenomenon. Their usage is often complementary, notably to references to the cultural context in which the power of such limits asserts itself. In other words, many social phenomena can only be effectively explained by the fact that they are also caused by implicit, erroneous and statistically predictable modes of reasoning. To take just one example, the many researchers who have studied questions of risk perception by the general public have found that collective representations are often far from an objective evaluation of the phenomena concerned. Paul Slovic and his team carried out a set of studies17 that led to the conclusion that certain risks can be greatly overestimated or underestimated. For example, one of their surveys found that the number of deaths by drowning is thought to be far higher than in reality and, conversely, that the number of deaths because of cardiovascular disease is greatly underestimated. Several studies have found this type of result18 but how can it be accounted for? The individuals who are asked such questions are generally not competent to answer them as they cannot be reasonably imagined to be able to refer to public health data when replying. However, they do not answer by chance either, as this would show up in the results of such surveys. We can always conjecture that they have access to a sample of facts via their social lives which are linked to matters they are asked questions about and hyperbolically deduce results that do not actually comply with reality. In fact, this sampling technique often gives good results in everyday life but it works much less well when it is affected by the availability bias. What is this? 16

Halvorson and Rock (2015). Slovicet al. (1982). 18 Flanquart (2016). 17

282

G. Bronner

Tversky and Kahneman proposed the definition of availability bias as the tendency we have to estimate the probability or frequency based on the ease with which we find examples that, by their type illustrate the event we wish to estimate. The problems caused by availability bias in fact represent a breakdown in inductive logic. This involves estimating the validity of a general statement on the basis of the particular examples we are able to recall. This breakdown can occur in two ways. First it can be due to the complexity of a problem and the concomitant ease of finding a certain number of examples that seem likely solutions. Alternatively, it can be the consequence of the fact that the examples found are not emotionally neutral. This means that certain illustrations of a phenomenon will be remembered better than others. One of Tversky and Kahneman’s (1973) experiments illustrates one aspect of this bias. The problem given to subjects was as follows: “Let’s consider the letter r. In English words of three or more letters, do you think the letter r appears more frequently in first or third position?” A large majority of individuals said there were more words in the English language beginning with the letter r than words with r as their third letter. In fact, the opposite is the case. The answer given by subjects to the first question is understandable. As all cruciverbalists know, finding a word by trying to think of its third letter is a more complex mental operation, than finding words when you know the first letter. Therefore, in the same given and limited time, it is quite likely people will find a word beginning with the letter r more frequently than with the more complex operation of finding a word whose third letter is r. This leads to the subjectively rational but false impression that there are more words in the English language which have r as their first letter than as their third letter. Availability bias is enveloped in a broader cognitive process namely the inductive logic which gives the illusion that general statements are valid on the basis of specific cases being valid. We know full well that this is not valid according to the ideals of pure logic but still naturally trust in induction simply because it has endlessly been proven to be effective in our daily lives as social actors. Availability bias is however a specific breakdown in inductive logic because it involves a sampling failure in the evaluation of a phenomenon. In this way, the error highlighted in the problem of the letter r only occurs because the easiest examples to find are in reality less numerous than competing examples. However, there is no reason to believe this to be the most frequent or even the only case in reality. We may even suppose that Tversky and Kahneman had to use a little imagination and experiment to find an example that could illustrate their theory. Staying with the problem of the letter r, it is

Why Does Comprehensive Sociology Need to Be …

283

also possible to suggest this involves the existence of an inferential temptation more than mechanical recourse to a heuristic. Indeed, if instead we set the following problem: “In your view, are there more words beginning with the letter z or more words containing the letters r, s and e in the French language?” it is likely that most people would opt for the second solution and they would be right. And yet, it is easier to find examples of French words beginning with the letter z in a limited amount of time (say, thirty seconds). This means the temptation to resort to the logic of availability is outweighed here by the knowledge that words in the French language beginning with the letter z are rarer than those containing the letters r, s, and e. Finally, we should also add that this inferential temptation of availability can be significantly reduced by protocols that facilitate retro-judgment19 which shows that availability bias’s power over our mind is not mechanical in nature. In other words, the observed effects of availability bias can easily be included in a comprehensive sociology program as long as they are linked to the idea of a cognitive limit to rationality. Slovic’s team noticed that availability bias had played a significant role in the results they obtained and thus tried to find out why certain examples of fatal risks were more likely to come to mind easier than others, ceteris paribus. Their investigation led them to the following result which is both intuitive and fascinating— the media which often provide examples of these kinds of subjects cover certain risks more than others in quantitative terms. The intensity of coverage has more to do with the logic of the attention economy20 than statistical rationality. Thus, the media tend to give more visibility to certain tragic events (death by drowning during a flood for example) than others (death by heart attack for example). The paradox here is that the rareness of a risk is also what gives it a competitive advantage in the cognitive markets which means that certain forms of rareness receive too much media coverage and are therefore overestimated thanks to availability bias. This is not without consequence particularly for public health matters because a certain current of opinion may call for—and sometimes obtain— costly measures thus causing political decisions to be taken which are contrary to a certain form of rationality.21 Unsurprisingly this mechanism can be found in other forms of spontaneous sociological beliefs. An international

19

Schwarz, Bless, Strack, Klumpp, Rittenauer-Schatka and Simons (1991). On this subject, see Citton (2014). 21 Cf. Bronner and Géhin (2010). 20

284

G. Bronner

survey carried out by the Ipsos Mori institute22 showed that in many countries people have a tendency to overestimate the number of Muslims living there. For example, the Australians are the champions for this. They think that Muslims represent 18% of the national population when in fact there are only 2%. Similarly, Americans, Canadians and French people significantly overestimate the number of Muslims living in their countries. In general, the actual proportion of Muslims in the 14 countries surveyed is around 3% but the average response given is 16%. And yet if you think about it, none of us can answer the question as to what proportion Muslims represent in the national community unless we are aware of the available statistics. We only have access to a limited sample of the population through the neighborhood we live in or information the media supply us with. This means that whenever an issue is the subject of intense coverage in the public space, individuals will tend to overestimate its expressions as an almost mechanical consequence. In these conditions, even answering the question with certitude might require as serious an investigation as the one conducted by Slovic and his team. However, we can reasonably conjecture that the socio-cognitive process at work in the overestimation of the number of Muslims is similar to that underpinning the underestimation of certain risks. We are fairly well aware of all the reasons that have led to Muslim populations receiving intense media coverage giving them a level of social visibility which goes far beyond the statistical reality they actually represent in terms of the population. As we have seen, it is easy for an unprepared mind to mix this kind of social visibility up with a form of representativeness. I chose to particularly focus on availability bias because it is a classic of the psychology of error, because it has varied and yet commonplace consequences in social life and because it can be related to the register of implicit reasoning and thus to comprehensive sociology. It is an example of a mental invariant that needs to be taken into account for certain social phenomena. This means it is necessary, but not sufficient, for sociological study, as we have seen. This is because its expression only makes sense when the configuration of the cognitive market is also taken into account and, as in the second example, when the existence of social stereotypes is considered. It can thus be observed at work in objects that are legitimately studied by sociologists and not just by psychologists. It also reveals the full extent of the perimeter of a discipline that studies the hybridization between mental invariants and social variables.

22

https://www.ipsos-mori.com/researchpublications/researcharchive/3466/Perceptions-are-not-reality10-things-the-world-gets-wrong.aspx.

Why Does Comprehensive Sociology Need to Be …

7

285

In Conclusion: A Cognitive Ecology Which Makes Enlarging Comprehensive Sociology an Urgent Matter

This deregulation can be explained first by the cyclopean mass of information available which is unprecedented in the history of humanity and second by the fact that everyone can “pour” their own representation of the world into this “ocean of information.”23 This situation has weakened the role of the traditional gatekeepers like journalists, academic experts or indeed anyone previously considered socially legitimate to take part in public debate. These people all used to play a regulatory role in the information market. Public debate is a major sociological event that has had all sorts of consequences including that it favors the spread of certain forms of credulity, as many authors have noted.24 Credulity can only be satisfactorily explained by using multi-factorial models but clearly certain false ideas dominate, persist and are sometimes more successful than more reasonable and balanced ideas. Indeed, they take advantage of dubious intellectual processes which are nonetheless attractive for the mind. In other words, cognitive biases are one of the variables which give a competitive advantage to credulity in this deregulated market.25 A study published in the journal Science showed that disinformation spread six times faster on average than more demanding information26 and that people remembered it better.27 One possible explanation28 for the phenomenon of false information that goes viral is that it flatters our minds’ most intuitive ways of functioning, but not always the most honorable ways. This unprecedented situation brings up an ecological point that serves as an argument in favor of an enlarged comprehensive sociology rather than an epistemological argument. Certain phenomena that were already part of social reality even before the Internet have been amplified since the beginning of the 2000s. For example, this is the case of mistrust of vaccines, conspiracy theories or the multiplication of all sorts of health or environmental alerts that are not always well-founded in reality. Some of these phenomena can

23

See for example Patino (2019). For example, Quattrociocchi (2018). 25 Qiu, Oliveira, Shirazi, Flammini, and Menczer (2017). 26 Vosoughi, Roy and Aral (2018). 27 Pantazi, Kissine and Klein (2018). 28 The success of contemporary credulity is a complex question which is not only caused by the existence of cognitive biases. I attempted to explore certain aspects of this in particularly in Bronner (2013). 24

286

G. Bronner

only be satisfactorily explained by realizing that they also result from how our brain overestimates low probabilities,29 focuses on the costs of a proposal rather than its potential benefits30 and gives greater consideration to the consequences of our actions than to those of doing nothing.31 All these commonplace elements of our psychic life have been experimentally tested and often duplicated.32 As discussed earlier in this article, they also hybridize with social variables that can be explained by the other limits that have an effect on rationality. Sociology needs to take seriously the idea of increasing its comprehensive methods if it aims to fully respond to the challenge of our contemporary informational context. This situation of deregulation of the cognitive market is unprecedented in the history of humanity and brings up the following question. Is the competitive advantage which certain proposals appealing to people’s credulity benefit from, likely to be long-lasting? Or will the competition between intellectual proposals favor the best argued and the closest to the ideals of rationality in the medium or long term? It is hard to give anything but a speculative answer to this question but it does provide a conclusion for this article in the form of dialogue with Raymond Boudon. Keen readers of the French sociologist will remember that he proposed a progressive theory of ideas in his later work.33 Obviously, he recognized that collective opinions can be led astray (this was indeed one of the main subjects of his work) but he claimed, taking de Tocqueville as his inspiration, that ideas which benefit the common good34 win out in the long term in history. The idea at the heart of this “evolutionist” theory works equally well for matters opposing what is true and what is false as for those involving good and evil. It is based on the idea of reasons which he calls “trans-subjective” because they have the “capacity to be adopted by a group of people even if they cannot be said to have objective validity.” (1995, p. 67). This notion would no doubt have merited the more powerful analysis because it led the author to formulate the following very powerful intellectual proposal—the trans-subjectivity of certain ideas gives them longevity and a “natural” form of dissemination when social conditions facilitate this with the result that, when an idea of this kind has taken hold, public opinion does not go back on it. On many occasions, Boudon defended the idea that 29

Prelec (1998). Tversky and Kahneman (1986). 31 Tversky and Kahneman (1982). 32 Kahneman, D. (2012) Système 1 Système 2. Les deux vitesses de la pensée, Paris, Flammarion. 33 He first introduced this theory in Boudon (1995). 34 Boudon (2000, p. 26). 30

Why Does Comprehensive Sociology Need to Be …

287

there is a kind of rational “safety lock” in both the descriptive and normative domains. He liked to illustrate this with the example of the death penalty— once abolished, its principle is no longer discussed. This really is therefore a form of optimistic evolutionism that characterized the way Raymond Boudon conceived of the selection of ideas and which we find in his later work in Le sens de valeurs and also in his short book Déclin de la morale? Déclin des valeurs? The trans-subjectivity of a belief can therefore be gauged according to the belief ’s capacity to transfer to other minds, all things being equal . This last condition is particularly important because it assumes there is a cognitive market in which competition between ideas is pure and perfect. This situation is very rare in reality if only because cognitive markets have a history and because certain ideas benefit from an oligopolistic or even monopolistic status and from dissemination effects that ensure their longevity and not because of their trans-subjective character.35 However, can we really make the intellectual bet that a free market of ideas would favor the emergence of the cognitive products that are the best argued from the standpoint of the norm of rationality? This idea can be discussed with reference to the notion of trans-subjectivity which helped Boudon conceive of his evolutionist theory. This notion does enable us to understand how cognitive competition can free individual judgments from their dimensional and cultural determinants over the long term in history. However, it also serves to explain how certain inferential temptations can converge in a deregulated cognitive market and lend false or dubious ideas a veneer of plausibility. The question is whether competition favors the best or just the most satisfying product. The two are sometimes synonymous in many markets but the situation in the cognitive market clearly portrays the distance between methodical thought and credulity. These questions remain speculative for the time being but should interest everyone concerned by the renewal of methodological individualism.

References Allais, M. (1953). Le comportement de l’homme rationnel devant le risque: critique des postulates et axiomes de l’école américaine. Econometrica, 21(4), 503–546. Bessi, A., Zollo, F., Del Vicario, M., Scala A., Cardarelli, G., Quattrociocchi, W. (2015). Trend of Narratives in the Age of Misinformation. PloS one 10 (8), e0134641. 35

Bronner (2002).

288

G. Bronner

Boudon R. (1986). L’Idéologie. Paris, Fayard. Transl. (2013). The Analysis of Ideology. New York: Polity Press. Boudon R. (1990). L’Art de se persuader: Des idées douteuses, fragiles ou fausses. Paris, Fayard. Transl. (1997). The Art of Self-Persuasion: The Social Explanation of False Beliefs. New York: Polity Press. Boudon R. (1993a). L’explication cognitiviste des croyances collectives. Cahiers de recherche sociologique, 21, 143–162. Boudon R. (1993b). Toward a Synthetic Theory of Rationality. International studies in the philosophy of science, 7 (1), 5–19. Boudon R. (1994). Reconstructing Complex Systems Needs a Complex Theory of Rationality. In H.-U. Derlien, U. Gerhardt, & F. W. Scharpf (Eds.) Systemrationalität und Partialinteresse (pp. 167–188). Baden-Baden: Nomos-Verl.-Ges. Boudon R. (1995). Le Juste et le vrai. Paris: Fayard. Transl. (2001). The Origin of Values. New Brunswick/London: Transaction. Boudon, R. (1999). Le Sens des valeurs. Paris: PUF. Boudon, R. (2000). Etudes sur les sociologues classiques II . Paris: PUF. Boudon, R. (2002). Déclin de la morale? Déclin des valeurs ? Paris : PUF. Boudon, R. (2003). Raison, Bonnes Raisons. Paris: PUF. Bourdieu, P., Chamboredon, J-C., Passeron, J-C. (1968). Le Métier de sociologue. Paris: Mouton. Bronner, G. (2007). L’empire de l’erreur – Eléments de sociologie cognitive. Paris: PUF. Bronner, G. (2013). La démocratie des crédules. Paris : PUF. Bronner, G., Géhin, E. (2010). L’inquiétant principe de précaution. Paris: PUF. Bronner, G., Géhin, E. (2017). Le danger sociologique. Paris: PUF. Brunet, E., Sarfati, Y., Hardy-Bayle, M. C., Decety, J. (2000). A PET Investigation of the Attribution of Intentions with Nonverbal Task. Neuroimage, 11, 157–166. Bruner, J. (1957). Going Beyond the Information Given. In Jerôme Bruner et al. (Eds) Contemporary approaches to cognition (pp. 143–175). Cambridge: Harvard University Press. Citton, Y. (Ed) (2014). L’économie de l’attention. Paris, La découverte. Decety, J., Jackson, P. L. (2008). Le sens des autres, La Recherche, special issue 30, 64–68. Dupuy, J. P. (1997). Temps et rationalité: les paradoxes du raisonnement rétrograde. In Jean-Pierre Dupuy & Pierre Livet (Eds) Les Limites de la rationalité (pp. 30– 58). Paris: La Découverte. Edwards, W. (1968). Conservatism in Human Information Processing. In B. Kleinmutz (Ed.) Formal Representation in Human Judgement (pp. 17–52). New York: Wiley. Elster, J. (1986). Le Laboureur et ses enfants (The Ploughman and his children) Paris: Les éditions de Minuit. Flanquart, H. (2016). Des risques et des hommes. Paris: PUF. Frith, U. (2001). Mind Blindness and the Brain in Autism. Neuron, 32, 969–979. Heider, F. (1944). Social Perception and Phenomenal Causality. Psychological Review, 51, 358–373.

Why Does Comprehensive Sociology Need to Be …

289

Frazer J. G. (1890/1981). Le Rameau d’or (The Golden Bough). Paris, Robert Laffont. Gigerenzer, G. (1991). How to Make Cognitive Illusions Disappear: Beyond Heuristics and Biases. European Review of Social Psychology, 2, 83–115. Gigerenzer, G., Hoffrage, U. (1995). How to Improve Bayesian Reasonig Without Instruction: Frequency Format. Psychological Review, 102, 684–704. De Gré, G. (1941). The Social Compulsions of Ideas, toward a sociological analysis of knowledge. New Brunswick. Halvorson, H. G., Rock, D. (2015). Beyond Bias. Strategy + Business, 13(7), https://www.strategy-business.com/article/00345?gko=ed7d4. Kant, E. (1781/1986). Critique de la raison pure (Critique of Pure Reason). Paris: PUF. Leslie, A. M., Kebble, S. (1987). Do Six-Month-Old Infants Perceive Causality? Cognition, 25, 265–288. Lévy-Bruhl, L. (1960). La mentalité primitive. PUF : Paris. Meehl, P. (1954). Clinical versus Statistical Prediction: A Theorical Analysis and a Review of the Evidence. Minneapolis: University of Minnesota Press. Mill, J. S. (1843/1988). Système de logique (System of Logic ). Bruxelles : Mardaga. Pantazi, M., Kissine, M., Klein, O. (2018). The Power of the Truth Bias: False Information Affects Memory and Judgment Even in the Absence of Distraction. Social Cognition, 36 (2),167–198. Patino, B. (2019). La civilisation du poisson rouge: Petit traité sur le marché de l’attention. Paris: Grasset. Quattrociocchi, W. (2018). Liberi di crederci: Informazione, internet e post-verità. Rome: Codice Edizioni. Qiu, X., Oliveira, D., Shirazi, A. S., Flammini, A. & Menczer, F. (2017). Lack of Quality Discrimination in Online Information Markets. Preprint arXiv:1701. 02694. Pareto, V. (1912/1968). Traité de sociologie générale (Treatise on General Sociology). Genève-Paris: Droz. Piatelli Palmarini, M. (1994). « Leurres du jugement » dans Manger magique (dirigé par Fichler C.). Paris, Autrement, n°149. Popper, K. (1956/1986). Misère de l’historicisme (The Poverty of Historicism). Paris: Plon. Prelec, D. (1998). The Probability Weighting Function. Econometrica, 47 , 313–327. Premack, D., Premack, A. J. (1997). Infants Attribute Value to the Goal Directed Actions of Self-Propelled Objects. Journal of Cognitive Neuroscience, 9 (6), 848– 856. Samuelson, P. (1937). A Note on Measurement of Utility. Review Economics Studies, 4, 155–161. Schwarz, N., Bless, H., Strack, F., Klumpp, G., Rittenauer-Schatka, H., Simons, A. (1991). Ease of Retrieval as Information: Another Look at the Availability Heuristic. Journal of Personality and Social Psychology, 61(2), 195–202.

290

G. Bronner

Shackle, G. L. S. (1967). Décision, déterminisme et temps (Decision-making, Determinism and Time). Paris: Dunod. Simon, H. (1957). Models of Man: Social and rational . New York: Wiley. Slovic, P., Fischhoff, B., Lichtenstein, S. (1982). Facts versus Fears: Understanding Perceived Risk. In Amos Tversky, Daniel Kahneman et Paul Slovic (Eds.). Judgment Under Uncertainty: Heuristics and Biaises (pp. 463–490). Cambridge: Cambridge University Press. Tversky, K., Slovic, P. (Eds) (1982). Judgment Under Uncertainty: Heuristics and Biaises. Cambridge: Cambridge University Press. Tversky, A., Kahneman, D. (1982). The Simulation Heuristic. In Amos Tversky, Daniel Kahneman et Paul Slovic (Eds.). Judgment Under Uncertainty: Heuristics and Biaises (pp. 160–173). Cambridge: Cambridge University Press. Tversky, A. & Kahneman, D. (1986). Framing of Decisions and the Psychology of Choice. In Jon Elster (Ed.) Rational choice (pp. 123–141). Oxford, Basil Blackwell. Vosoughi, S., Roy, D., Aral, S. (2018). The spread of true and false news online. Science, 359-6380, 1146-1151. Weber, M. (1922/1971). Économie et société (Economy and Society). Paris : Plon. Weber, M. (1922/1949). The Methodology of the Social Sciences (Transl. E. A. Shils et H. A. Finch) Glencoe: The Free Press.

Methodological Individualism and the Micro-Macro Link

Methodological Individualism as Holism of the Parts: From Epistemology to Ontology Nathalie Bulle

1

The Holism of Parts

Jan Smuts, South African Statesman, scientist, and philosopher, is referred to by his biographer (Crafford 1943, p. 11) as “the world’s leading holist.” Smuts introduced the term holism (from ×λoς = whole) in his work Holism and Evolution (1926/1927) in a different sense from that which later dominated the social sciences, and which will be identified in the following with causal holism.1 Smuts’ holism is particularly interesting to detail somewhat, even though he tends, in a questionable way, to make it an a priori principle of his metaphysics of evolution, underlying the dynamics of evolutionary creation 1

Two broad forms of causal holism can be distinguished here. The first form considers the causal capacity of wholes to act on their parts in such a way that the parts respond to logics defined at the whole level (e.g., through the notion of spirit of the people in certain nineteenth century historicist approaches, or the notion of function in various forms of functionalism in the social sciences). The second form, a more sophisticated version of the first, attributes causal power to the relational or organizational structures characterizing the arrangement of the parts. These structures may determine the specific causal properties of the parts (e.g., through the neo-Marxist notion of class habitus).

I would like to thank Francesco Di Iorio for his constructive comments on this chapter. However, any remaining errors or oversights are solely my responsibility.

N. Bulle (B) National Center for Scientific and Sorbonne University, Paris, France e-mail: [email protected]

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 N. Bulle and F. Di Iorio (eds.), The Palgrave Handbook of Methodological Individualism, https://doi.org/10.1007/978-3-031-41512-8_15

293

294

N. Bulle

based on an inherent tendency of natural entities to unite into wholes: “Holism is a real operative factor, a vera causa” (Smuts 1926/1927, p. 88). Smuts refers to the existence of entities or parts that exhibit certain properties and activities that are inherent to the whole they form and, therefore, cannot be described in terms of, entities and properties in an independent state: “The whole is in the parts and the parts are in the whole, and this synthesis of whole and parts is reflected in the holistic character of the functions of the parts as well as of the whole.” To this notion of ontological dependence of the parts with respect to the wholes they form, is added the idea, which is especially important, that the whole is not added to the parts but only characterizes their unity: “A whole is not some tertium quid over and above the parts which compose it; it is these parts in their intimate union and the new reactions and functions which result from that union” (Smuts 1926/1927, p. 130). For the same reasons that the whole is not something additional over and above its parts, the whole, Smuts (1926/1927, pp. 125, 134) insists “does not act as a separate cause, distinct from its parts (…) holism is of the parts and acts through the parts (…) the causality of the whole (is) exhausted by the causal operation of its parts [so that there is nothing] over and above the influence of the parts which must be attributed to the whole as such.” Let us note with Smuts that this does not contradict the possibility of a central control, as is the case in organisms. Such control does not represent the action of the whole on the parts, but results from a differentiation of the functions of the parts. According to Smuts, the creativeness of matter is limited to the transformation of new structures out of preexisting material units, while genuine creativity in the evolution of organisms involves qualities or properties that “are not mere reshufflings of characters which were there” but instead constitute irreducible entities and properties that combine and create specific wholes. In this regard, Smuts invokes the idea of “creative evolution” referring to the appearance of radically new entities and properties within the evolutionary process: The process of creative Evolution is not a mechanical rearrangement of old material; it involves the qualitatively new at every stage, from the most minute elements to the most complex structure. It is not merely the structure which is new and different from what has gone before, some of the materials are also new (…) There is the creation of the new variety or species (the new structure) ; there is the creation of new unit characters (parts of the structure) which justify the new species or variety ; and there is behind the new unit characters not

Methodological Individualism as Holism of the Parts …

295

mere re-arrangements of elements of old character units, but an integration of new materials or quality elements with the old elements in the formation of the new unit characters (…) If there is this evolution or making not only of new wholes or structures, but also of new quality elements therein, the whole fabric of Mechanism as ordinarily understood is shaken to its foundations. (Smuts 1926/1927, pp. 137–9)

Although Smuts holds holism as the fundamental creative factor of evolution, he also takes into account the role played by mutational processes in explaining the development of specific phenomena related to life and consciousness. He details various types of structures that he orders according to grading levels of binding intensity between their constituents, from the first mechanistic structures to different types of holistic structures. Consequently, the holistic character of wholes is defined by the activity of their parts and increases with their inner cooperation: natural, instinctual, and then, at the highest stage, conscious. Five stages are then essentially distinguished (Smuts 1926, pp. 88, 109). The first stage is related to the physical mechanism or is reduced to it: The parts of the structure have causal properties independent of the whole they form. The second stage is paradigmatically characterized by chemical structures, which represent creative syntheses with new qualitative elements. The third stage is that of functional structures in living organisms, whose unity intensifies in animals, with the appearance of a system of central control, regulation, and coordination of the parts. The fourth stage is that of the minds “or psychical organs where the Central Control acquires consciousness, freedom and creative power.” Finally, the sixth stage, the highest and most evolved structure in the universe, is that of the human personality, which as such constitutes the “part” of the society. Society does not act as a separate and distinct cause from its parts any more than holistic wholes do. Furthermore, it is not a unit as is a biological organism: The society or group is “holistic without being a whole” (Smuts 1926, p. 348). But as is the case in holistic wholes, the human personality is inseparable from society. The capacities that characterize it are inherently linked to the formation of human groups: “The individual becomes conscious of himself only in society and from knowing others like himself ” (Smuts 1926, p. 234). Through their activities, which are inherent to their essential social nature, individuals develop social structures in the form of institutions of all sorts intended to provide intellectual and moral education, as well as the coordination of their actions and interactions. Holism and Evolution enjoyed a brief period of popularity, but its real contribution has been largely overlooked (see Poynton 1987; Jàros 2002). The vague notion associated with Smuts’ concept of holism has its roots in

296

N. Bulle

Aristotle’s reflections in the Metaphysics, according to which the whole, if is not a mere heap, is something “more than the sum of its parts.” This understanding retains nothing of Smuts’ original contribution, for it presupposes that the parts are to be considered as independent entities, transformed by their belonging to the whole, which leads to a dubious separation of the parts from the whole. Once we acknowledge with Smuts the intrinsic interdependence of the parts, there is no plausible reduction that would allow us to theorize the parts with properties independent of the whole. Arthur Koestler (1978, p. 26), who considers the work as remarkable, blames its neglect by academic circles on the fact that it went against the spirit of the times. However, it should be noted that Smuts’ emergentist contemporaries, Samuel Alexander or Lloyd Morgan, whose philosophies of evolution were also opposed to prevailing reductionism, are better known and often cited in this regard. But Smuts’ conceptions differ from those of the emergentists. According to him, the creative force of evolution, the holistic “operative factor,” is the inherent tendency toward the union of parts, and not the emergence as such of new wholes from lower levels of organization. This difference is not minor, as Smuts’ holism focuses on the activity of parts, while emergentists focus on the organizational structures of wholes, by invoking a pyramidal scheme (Morgan 1923, p. 9) involving the levels of composition of “complexes” emerging at each level according to “progressively ascending grades.”2 In this regard, Smuts contends that emergentism is an inverted vision of preformism.3 Consequently, the omission of Smuts’ philosophical contribution may be explained not so much by his opposition to reductionism, but more precisely by the fact that his holism is based on the parts (those related to life and to consciousness implying intrinsic evolutionary creations). It broke more fundamentally with the spirit of the times than did emergentism and its pyramidal compositional scheme, ultimately built from elementary physical events. It must be acknowledged that Smuts’ insistence on “holism” as such as the dynamic of evolution probably did not allow him to differentiate from other forms of holism. This may be confirmed by Karl Popper’s assertion attributing the popularity of historicism to the attraction of evolutionism, from which the great systems of evolutionary philosophy issued by “Bergson, Whitehead, Smuts and others” are presented as an emanation (Popper 1936/

2 “Each higher entity in the ascending series is an emergent ‘complex’ of many entities of lower grades, within which a new kind of relatedness gives integral unity” (Morgan 1923, p. 10). 3 See his commentary on Lloyd Morgan (1923) in the second edition (Smuts 1926/1927, p. 330) published one year after the first.

Methodological Individualism as Holism of the Parts …

297

1957, p. 97). However, it is worth noting the change in Popper’s conceptions on this subject, as he will adopt from the 1960s onward notions such as those of holistic structure and, above all, creative evolution.4 Lastly, it is worth questioning whether the term holism itself is more appropriate for the general qualification of systems that grant causal power to wholes, and whether it alone could justify the misunderstanding and neglect of the specificity of Smuts’ holism, which is “of the parts” and rejects the idea that the action of the whole can be added to that of the parts. On the epistemological level, the activity of parts in Smuts marks from the outset an opposition to the Humean forms of explanation that involve only correlations between discrete particulars. Moreover, his opposition to reductionism5 derives from the fact that wholes are not explained by their parts with characteristics in an independent state, any more than the parts are explained by the wholes speculatively separated from them. In the following, I will argue that the compatibility of methodological individualism (MI) with the holism of parts, and correlatively, its impossible integration into the dominant physicalist frames of thought, can explain its assimilation to a form of reductionism which represents the alternative and opposite option to the forms of causal holism with which Smuts’ holism has been conflated. In summary, holism in its original and forgotten sense is a “holism of the parts” which implies a non-positivist and non-reductionist epistemology. Holism of the parts refers to the appearance of new entities and properties in the course of evolution that tend to unite into new wholes. It evokes properties and activities of the parts that are inherent to the whole they form and prohibits the whole from acting as an added cause to the activity of the parts. The most evolved structure of the universe is, in this framework, represented by the human personality.

4 Creative evolution refers to the chance-like character of mutations and natural selection, underlying the creation of intrinsically new entities and properties within nature (Popper 1978a, p. 352). 5 The microreduction of one theory by another mainly refers to the idea that the entities, properties and explanations of one theory can be accounted for by a more fundamental theory. See the section below 5.1.

298

2

N. Bulle

Theoretical Decomposition of a Complex Whole into Basic Units

The classical scientific approach to decomposition of a complex whole into basic units has been applied in various fields, including psychology, where specific approaches have been based on the explicit identification of analytically constituted units.6 For example, Boris Sidis (1908), a now-forgotten psychologist, defines the percept as the “basic unit” in the perceptual process. In doing so, he challenges the legacy of classical empiricism, which does not differentiate qualitatively between sensations and ideas. Sidis likens the percept to a cell, as it “forms one organic whole” with its “constituent elements firmly integrated into a single living organization.“ This involves a specific form of perceptual consciousness, making the percept irreducible to the sensory elements that compose it, and implying characteristics inherent to the whole perception process (Sidis 1908, p. 53). Similarly, psychologist Kurt Koffka (1935), in his Principles of Gestalt Psychology, contrasts between approaches to behavior conceived as physiological (molecular) processes composed of discrete or separate units, such as sensory contents in associationism or stimulus–response sequences in behaviorism, and (molar) approaches that focus on “interconnected groups or units,” where the phenomena of meaning and value can contribute to the psychological approach. On his part, the Russian developmental psychologist Lev Vygotsky (1934/1986) utilizes the paradigmatic example of the water molecule to illustrate his method of studying verbal thought. He defines this approach as breaking down a complex whole into basic units, which represent “living parts” that can no longer be broken down without losing their properties (such as H2 O), and exhibit all the fundamental properties inherent in the whole. According to Vygotsky, the basic unit that encapsulates the simplest aspect of verbal thought as a whole is the internal aspect of the word, that is, its (subjective) meaning, which involves an entire act of thought. In sociology and economics, the analytical decomposition of a social whole into basic units7 was made explicit in the works of Carl Menger (1883/1986), the first theorist of methodological individualism. According 6 It should be noted here that a whole literature associated with holism in various, often “strong” or “causal”, forms (i.e., hegelianism, historicism, organicism, functionalism, etc.) uses the concept of analysis critically in a reductionist sense to express something quite different—a mere decomposition or “dissection” of a whole into elementary components, endowed with characteristics in a state independent of the whole. See in particular D. C. Phillips (1976) who refers to this literature and maintains the confusion regarding the meaning of “analysis” in science. 7 Gustav Ramström (2018/2023) discusses the fact that in the social sciences the micro–macro link is an analytical type of relationship.

Methodological Individualism as Holism of the Parts …

299

to Menger, empirical realism, which considers that phenomenal reality exists autonomously, cannot lead to exact laws. Theoretical analysis goes beyond phenomenal reality by isolating the subsystems studied and by bringing into play “the simplest and strictly typically conceived constitutive factors (susceptible of exact inquiry) of phenomena” (Menger 1883/1986, p. 113). These theoretical units are not supposed to be independent of the whole: “[the whole] cannot be imagined in its normal appearance and function without some essential part or other. Nor, conversely, can such a part be imagined in its normal nature and function when separated from the unit [the whole]. It is obvious that we have here a certain analogy between the nature and the function of natural organisms on the one hand and social structures on the other” (Menger 1883/1986, p. 130). In the same way, Ludwig von Mises (1949/1966, p. 51) explains that there is no precedence of the whole over its parts or of the parts over the whole: “Logically the notions of a whole and its parts are correlative.” In this theoretical framework, social wholes and their parts do not refer to two interacting realities that can be separated by a speculative path but to relative constructs, as illustrated in the example of water and water molecules. Therefore, the reference to the parts of a whole does not involve any reduction, but rather characterizes the proper approach of theoretical sciences: The charge of atomism in the above indicated sense of the word is [thus] a misunderstanding (…) Every theory (…) strives first and foremost to make us understand the complicated phenomena of the research field peculiar to it as the result of the coworking of the factors responsible for its origin. This genetic element is inseparable from the idea of theoretical sciences. (Menger 1883/1986, pp. 93–94)

Max Weber (1922/1958, p. 58) essentially conveys the same idea when he states that understanding sociology regards “the individual and their actions as the basic unit, or its ‘atom’. Weber acknowledges that this analogy, which evokes the parallel between physical and social sciences, is debatable, a connection also drawn by Menger, albeit with a hint of provocation concerning the accusation of atomism.8 This can be made clearer if the 8 Nevertheless, Weber criticizes the conclusions drawn by Menger from the theoretical approach to economics: “In the further course of our discussion, we shall frequently refer to this distinction [between the theoretical (exact) and the empirical-realistic], whose significance for the methodology of economics was to a certain extent already (as will be noted later) recognized by Menger, although his conclusions were in part incorrect.” (Weber 1903/1906/1975, note p. 4). Especially, Weber disputes the idea that social phenomena could be explained from abstract theories developed within social sciences, as if they would “contain the true reality of the object.” Weber emphasizes that they merely

300

N. Bulle

abstract scientific construct represented by the atom is recognized, which does not involve reductionism insofar as the atom properties are not assumed to be independent from the physical wholes under study. In this respect, Weber even criticizes the notion of interaction in Georg Simmel’s work, arguing that it assumes independent entities, which is not the case, even in the physical realm: “Gravitation is always reciprocal gravitation; not only the collision of two bodies moving in different directions but also the impact of a moving body upon a resting one affects both (by transmission of motion, alteration of speed and direction of kinetic energy, and generation of heat). Indeed, one can in general say that within the realm of physical reality an influence that is not somehow “reciprocal” in the strictest sense of the word and as a general phenomenon is scarcely conceivable (…) only with the greatest artificiality will one be able to conceptualize a pure “one-way” influence, i.e., an instance of one man being influenced by another where there is not some element of “interaction” (Weber 1908/1972, pp. 162–163). Popper (1957, p. 76) also asserts that the opposition between the ‘atomistic’ and the ‘Gestalt’ approaches is baseless, at least in atomic physics, because the study of particles systems is concerned with organized structures or wholes. Popper adds in note that “to the social scientist, such ideas as competition or division of labour should make it abundantly clear that an ‘atomistic’ or ‘individualistic’ approach in no way prevents us from recognizing that every individual interacts with all others.” Correlatively, Weber defines human behavior as “action” only when individuals attach subjective meaning to it, and action as “social” when the meaning that orients it takes account of the behavior of others. The basic units of Weber’s sociological approach, the individuals and their actions, cannot be broken down without losing the property of the individuals that is central to the approach—the capacity of understanding, which involves conscious processes.9 As we will see, this capacity is inherently tied to individuals’ membership in social groups and relies on their handling of culturally formed tools of thought. In this regard, Max Weber (1922) highlights the importance of functional questioning in sociology, which starts from the “whole,” characterizing typical orientations of actions of social actors in

provide heuristic devices for understanding them: “This procedure gives rise to no methodological doubts so long as we clearly keep in mind that ideal-typical developmental constructs and history are to be sharply distinguished from each other, and that the construct here is no more than the means for explicitly and validly imputing an historical event to its real causes while eliminating those which on the basis of our present knowledge seem possible” (Weber 1922/1949, pp. 101–102). 9 Let us note the proximity of the respective approaches of Weber and Vygotsky here (cf. Bulle 2018, 2021): Both refer to the role of consciousness and the notion of meaning, the subjective meaning of words in Vygotsky, and the subjective meaning of actions in Weber.

Methodological Individualism as Holism of the Parts …

301

relation to a given institutional context. He also points out the significant misunderstanding in confusing individualistic method with individualistic valuation “in all possible senses.” In summary, the analytical decomposition of a complex whole into basic units has been favored by social scientists who support non-positivist and non-reductionist approaches in epistemology. In these theoretical perspectives which include methodological individualism, the units of analysis have properties inherent to their belonging to the whole, and they account for the phenomena that develop at the level of the whole they form. Therefore, these units do not exist in an independent or separate state, and the whole does not act as a cause superadded to the activity of the parts. When these conditions are met, the theoretical decomposition of a complex whole into basic units ontologically refers to a “holism of the parts.”

3

Basic Units and Scientific Realism: from Epistemology to Ontology

Rom Harré (2006) explains that to understand the activities of the basic units distinguished in explanatory science, one must move beyond observable phenomena. This requires a commitment to scientific realism and exposes one to the difficulty of not transgressing Kant’s prohibition against using empirical concepts “constitutively” beyond the limits of the observable. However, our conviction in the existence of an order in nature, which underlies the very idea of causation, tends to require the persistence of things beyond change, as the anthropological epistemology of Emile Meyerson (1908) maintains. This explains our belief in the action of persistent entities and powers whose variable arrangements can account for observable changes. Scientific causality implements these causal powers on theoretical grounds since, from the ontological point of view, we have no way of understanding how things ultimately act on each other, and how “efficient” causality operates. Therefore, to understand the processes in question, it is necessary to break them down into theoretical components and properties, and to show how one can produce a given effect, in a given situation, by reassembling the components involved. It is assumed that these components will behave in their new arrangements in the same way as in the previous ones, in accordance with their “nature” (Cartwright 1999). Causal powers or capacities are dispositional properties of the entities involved, which scientists thus tend to interpret as real. In this respect they are not subjected to ceteris paribus conditions and continue to produce their effects in a trans-situational manner,

302

N. Bulle

interfering with the action of other factors or processes. In this sense, a form of causal realism underlies the “active”10 approaches in contemporary philosophy of science (which, nevertheless, represent a great diversity of ontological points of view), whose common claim is their opposition to Humean metaphysics where the interpretation of causal bonds is essentially descriptive, and where things have no inherent tendency to behave as they do, so that causal powers do not exist. Certain active approaches to causality thus legitimize the idea of generative mechanisms (Harré 1970; Harré and Madden 1973, and for applications in sociology, e.g. Cherkaoui 2005), supposed to unearth the factors that fundamentally generate the phenomena to be explained. As we have seen, Menger maintains that, to explain a given phenomenon, the goal of a theory is to account for “the coworking of the factors responsible for its origin,” which places the real at the horizon of the theoretical. Many commentators attribute this perspective to the ontological landscape of the Austrian school.11 Again, this can be clarified by the idea that, even if ontological reality is not directly accessible, theoretical science tends to rely on the notion that an inherent order in nature guides its development, and this is also reflected, for instance, in Friedrich Hayek’s (1952/1976, p. 173) statement: “We cannot regard the phenomenal world in any sense as more ‘real’ than the constructions of science: we must assume the existence of an objective world (or better, of an objective order of the events which we experience in their phenomenal order) towards the recognition of which the phenomenal order is merely a first approximation.”12 In short, as science tends to approach genuine causal factors, it moves away from the phenomenal world and naive realism, and develops new models of interrelationships, where the characteristics of theoretical entities, defined as the basic units of analysis, imply trans-situational causal properties.

10

Active epistemologies involving “activities” (Machamer et al. 2000), “capacities” (Cartwright 1989), or “causal powers”, and their opposition to “passive” approaches involving laws or counterfactual considerations (see for example: Bird, et al. 2012; Ellis 2009; Marmodoro 2010). 11 For further references on this subject see Bostaph (1978), Campagnolo (2010), Cowan and Rizzo (1996), Cubeddu (1993), Grassl and Smith (1986), and Smith (1990). 12 It is not sufficient, as a form of pragmatic neo-Kantianism would have it, to merely assign theoretical concepts an “as if ” regulatory role over empirical data. These theoretical concepts should aim to model real properties going beyond observable data. Although Weber’s idealizations may have had a regulatory function in the study of observable phenomena, it is likely that his central reference to the rational capacity of individuals is intended to serve explanation insofar as it tends to represent a genuine human capacity.

Methodological Individualism as Holism of the Parts …

4

303

MI’s Principle of Rationality and Holism of Parts

Based on the preceding discussion, I propose to demonstrate that the principle of rationality, which broadly refers to the ability of social actors to give meaning to their actions, underpins two core stances of MI. The first stance is non-positivist, wherein MI pursues an explanatory objective in the sense of causal realism. The second is non-reductionist, positing that rationality is a causal power inherent to the social nature of individuals. This suggests that MI advocates for a form of “holism of the parts.” Let us take a closer look at these two positions. Rational capacity characterizes humanity since the beginnings of Western philosophy: Humanity is gifted with logos, that is, with speech, and with reason. In particular, human beings make use of an articulated language thanks to the recursive capacities of their thought, which underlies their faculties of abstraction and reasoning. This human capacity does not imply any normative conception of rationality, and it is not essentially of a logical or instrumental nature; rather it is a general capacity for understanding, that is, attributing meaning (Bulle 2022). The driving role of meaning in human reason is involved in the idea that motivations, or “good reasons”13 (abstractly reconstructed by the observer), explain actions. Understood in a broad way, rationality constitutes a causal power of a trans-situational nature, so the more the analysis tends to refer to such a trans-situational role freed from “ceteris paribus” conditions, the more it tends to causal realism.14 The “understanding” approach of methodological individualism refers to this trans-situational nature of rationality, which accounts for the possibility of a relationship of understanding between the observer and the observed, beyond the situational differences that separate them. From this perspective, it assumes the possibility of understanding the influence of social actors’ situations on their actions by examining the meaning they attribute to them, and it identifies the causes of their actions with the reasons that drive them. This relation of understanding distinguishes the social sciences from the natural sciences, precisely when they both refer to generating causes. In the case of

13

On “good reasons” in Boudon’s work, see especially by Jean-Michel Morin (2023, this volume) and Di Nuoscio (2018). 14 In terms of human rationality, progress has been made in moving away from instrumentalcomputational models and toward the concept of axiological rationality (Boudon 1999/2001; see also the chapter by Sylvie Mesure 2023, in this volume).

304

N. Bulle

the social sciences, the causal power of the acting units at stake is not apprehended from the outside, and theoretically reconstructed from observational bases, but it is grasped from the inside. This is what Weber (1922/1968, p. 15) expresses: We can accomplish something which is never accomplishable in the natural sciences, namely the subjective understanding of the action of the component individuals. The natural sciences on the other hand cannot do this, being limited to the formulation of causal uniformities in objects and events and the explanation of individual facts by applying them. We do not “understand” the behavior of cells, but can only observe the relevant functional relationships and generalize on the basis of these observations. This additional achievement of explanation by interpretive understanding as distinguished from external observation, is of course attained only at a price, the more hypothetical and fragmentary character of its results. Nevertheless, subjective understanding is the specific characteristic of sociological knowledge.

This explains why interpretations invoking processes that, in principle, cannot serve as conscious motivations, leading to the failure of reference to the intertemporal continuity of individuals’ rational capacities, have less explanatory power according to the understanding approach of MI.15 It is the case for approaches that rely on internalized dispositions, mentalities, or the “class habitus” previously mentioned. But it is also the case for approaches that link human rational capacity to a given normative model, such as those developed by rational choice theories.16 These more or less general patterns of action are confined to a single kind of manifestation and are closer to empirical laws limited by ceteris paribus conditions.17 Moreover, the rationality principle underpins a non-reductionist stance because it refers from the outset to a capacity of individuals inherent to their social being. The rational capacities of individuals are rooted in their social learning. They become effective through the internalization of meaning structures related to Popper’s world 3. This world, as we know, refers to all the productions of the human mind in the form of thought contents (as distinct from thought processes), such as language, numerical systems, conceptual systems, etc., which are transmitted and developed socially.18 The 15

On MI and understanding sociology see Bulle (2023, forthcoming) and Feuerhahn (2023, this volume). 16 But these may be enlarged (see Opp 2023, this volume). 17 This is why, for instance, Cartwright’s epistemology is based on capacities, not dispositions which tend to refer to definite schemes of action. 18 Popper’s world 3 comprises intersubjective, objectifiable structures of meaning that have their own existence but no inherent causal action. This action is always mediated by individuals’ understanding:

Methodological Individualism as Holism of the Parts …

305

reference to the reasons for individual actions implies taking into account this cultural and institutional learning that underlies the meaning they give to these actions. Even the self-controlled nature of individuals’ actions is inherent in their social participation, by virtue of the structured tools of thought socially transmitted which allow for this self-control (Vygotsky 1934/1986). Therefore, the social participation of individuals is an essential condition of their rational capacity and their causal power. In contrast, it is when the principle of rationality is abandoned, because no situation seems to be able to justify the behavior of social actors on the basis of this principle, that the analysis resorts to reductionist explanations, referring to blind habits or automatisms, or other forms of physical state dominated by unconscious mechanisms of an organic nature, such as illness, nervousness, or madness. This can be seen, for example, in Weber (1903–1906/ 1975, p. 125), where he explains that if human action cannot be meaningfully understood, then it tends to become indistinguishable from actions carried out by inanimate objects, thereby referring to mechanical processes beyond the scope of understanding sociology: “[I]f human conduct cannot be interpreted in this way [meaningfully], it is no different from the fall of the boulder from the cliff. In other words, ‘incalculability’—in the sense of the non-satisfaction of the conditions for this type of interpretation—is the principle of the ‘madman’.”19 It should be noted that even though Weber contrasts meaningful action with purely reactive, and thus, unconscious action, such meaningful action is usually not completely conscious. As regards imitation, for example, only mechanical behaviors have no meaningful orientation. But a meaning appears as soon as the very attachment to habits assumes a certain degree of selfconsciousness, so that rationality in value is, for instance, involved. The same is true of traditional actions, which are not confined to thoughtless habits when they involve a subjectively perceived meaning, which, for example, refers to collective entities with normative authority.

“This influence is, to the best of my knowledge, always indirect. World 3 theories and world 3 plans and programs of action must always be grasped or understood by a mind before they lead to human actions, and to changes in our physical environment” (Popper 1978b, p. 164). For more on world 3 and MI, see Di Iorio (2016) and the discussion between Bulle, Di Iorio, and Joseph Agassi (2023, this volume), especially Popper (1965/1972; 1978b). 19 To complete Weber’s argument: “Whenever our search for historical knowledge leads us to focus on behaviour that is “irrational” in the sense of being inaccessible to interpretation, our need for causal [explanation] will indeed, as a rule, have to be content with having a “conception” of [that behaviour] on the basis of the nomological knowledge of, say, psychopathology or similar sciences, just like [having a “conception” of ] the pattern of those rock fragments” Weber (1903–1906/1975, p. 44).

306

N. Bulle

In summary, the rational capacity of individuals is a general, transsituational capacity inherent to their social being. As a trans-situational causal power, rationality involves a non-positivist stance, and as inherent to individuals’ social being, it involves a non-reductionist stance. Rationality is the fundamental property of the basic units defined by MI for the social sciences, the social actors, that makes MI consistent with the notion of holism of parts.

5

The Role of Epistemology and Ontology in the Misunderstanding of MI

The regularly stratified picture of reality based on compositional criteria that define successive autonomous “parts” and “wholes,” constitutes the ontological background of a large part of the work in the Anglo-Saxon physicalist tradition, and represents, as I will defend here, a major source of the misunderstandings developed about methodological individualism. The characteristics of this layered representation can be described as follows. 1. Levels are hierarchized in nature from most elementary to most complex structures. 2. Wholes or entities at a given level Xn are structures composed of parts or entities of level Xn-1 whose properties are thus independent of their belonging to Xn and so on.20 3. Entities and properties in this layered representation of nature correspond to the entities and properties studied by the different sciences.21 The non-reductionist physicalist perspectives assume that: 4. Some structural systems have identical functional consequences while involving very different entities and properties at lower levels. Such an assumption is at the basis of the notion of multiple-realization developed by contemporary non-reductionist physicalism. Emergentists add the following condition which, we will see, tends to abandon the physicalist framework: 20

“Anything of a given level except the lowest must possess a decomposition into things belonging to the next lower level” (Oppenheim and Putnam 1958, p. 9). 21 “The idea of reductive levels employed in our discussion suggests what may plausibly be regarded as a natural order of sciences.” (Oppenheim and Putnam 1958, p. 28).

Methodological Individualism as Holism of the Parts …

307

5. Under certain conditions, phenomena of organization and complexity specific to wholes at level Xn exert a causal power on parts or entities at level Xn-1. It is on such ontological bases, that a debate on reductionism developed in the twentieth century within analytic philosophy. At the heart of the opposition between the notions of micro (reduction) and emergence lies the issue of reducing entities and properties at Xn in terms of entities and properties at Xn-1.22 The relevance of the layered model, referring to the physical constitution of things, is today questioned because it conflates issues of constitution with causal power issues.23 Such conflations are fostered by the mainstream stance in analytic philosophy that associates a physicalist ontology with a principle of continuity established as a naturalistic quasi-dogma. This continuity principle involves the continuous constitution of things by physical components representing the building blocks of natural phenomena at inferior levels, united in wholes by organizational structures at superior levels. This representation originally opposed extra-natural factors introduced by metaphysical assumptions such as those derived from substance dualism, idealistic philosophies, or else vitalism.24 But its metaphysical limitations are reflected in Smuts (1926)’s statement about the role of naturalism in preventing any idea of radical novelty within natural processes: “Naturalism is wrong where it fails to recognize that there is creative evolution, and that real new entities have arisen in the universe, in addition to the physical conditions of the beginning” (Smuts 1926, p. 327).

The Bias of Reductive Physicalism on MI: The Case of Logical Empiricism The layered representation of the world and science was promoted in the course of the twentieth century by the logical empiricists who influenced the developments of analytic philosophy. To understand this development of scientific thought, let us recall that logical empiricists distinguish only two categories of meaningful propositions: analytic propositions—whose truth is exclusively based on their form and which are true or tautological, or false or

22

For an account and a critical discussion of emergentists’ claims in the first half of the twentieth century, see for instance Nagel (1961, pp. 336–397). 23 See especially, for instance, Craver (2007) and Potochnik and McGill (2012). 24 See in particular Kim (2003).

308

N. Bulle

contradictory—and synthetic propositions—whose truth is based on experience, and which are supposed to be validated by an empirical test. Any other kind of proposition is considered metaphysical. On these grounds, the role attributed to the entire philosophical enterprise is closely dependent on logic through strictly analytical work applied to language. Apart from the logical issues, the meaning of a proposition is entirely invested in the empirical field. A consequence is that all sciences share the same observational basis, which offers the possibility in principle of ultimately translating the different scientific languages into physical language (see Ayer 1959, Kistler 2007, pp. 3–4). In this framework, the possibility of reductionism involved building bridges between sciences by semantically linking their worlds of discourse term by term, given their unity in principle. Hence the project of the unity of science that animated the philosophy of logical empiricism, which was engaged in a general program of inter-theoretical reduction, level by level, of all sciences to physics. This program was implicitly associated with a stratified ontology by levels of physical constitution. Contrary to their rejection of metaphysics, certain neopositivists tended to guarantee their reductionist project by referring to the ontological evolution of the physical world and of life: If we find that there was a time when a certain whole did not exist, and that things on a lower level came together to form that whole, it is very natural to suppose that the characteristics of the whole can be causally explained by reference to these earlier events and parts; and that the theory of these characteristics can be micro-reduced by a theory involving only characteristics of the parts. (Oppenheim and Putnam 1958, p. 15)25

Even when not referring to a global ontology, neopositivists considered the possibility, in principle, of systematically linking properties of entities at higher levels to those of their constituents in an independent state, thus pertaining to lower ontological levels (see for example Ernst Nagel 1961/ 1979, p. 395). It is interesting to note that the reductionist interpretations of MI in the above sense appeared in the period of influence of logical empiricism, while the issue of MI was lively debated in the social science scene in the context of Hayek’s and Popper’s critiques of historicism.26 The interpretive bias involved by the neopositivist representation manifests itself, for example, 25

See also Kemeny and Oppenheim (1956) and Nagel (1961). The “atomistic” interpretation, in the reductionist sense, is of course ancient (see Menger above who refers to it), but with related epistemological foundations, derived from Humeanism. 26

Methodological Individualism as Holism of the Parts …

309

in May Brodbeck (1954, p. 141)’s feeling of inconsistency of MI when, in reaction to Hayek (1952) The Counter-Revolution of Science, she presents MI as an approach that is “reductionist in one sense, antireductionist in another,” because “the behavior of groups must be explained in terms of the behavior of individuals; but the psychology of individuals cannot be reduced to anything else.” This confirms the misunderstanding of the path followed by MI: that of the holism of parts. Indeed, MI opposes both causal holism and (micro)reductionism. For his part, Maurice Mandelbaum (1955, 1957), although a critical realist27 arguing against logical empiricism, interprets MI as an enterprise of “reduction of societal facts to facts concerning individual behavior,” which is also a form of inadequate translation of MI into the very language of logical empiricism. In the same period, logical empiricists themselves tended to interpret MI from the perspective of their reductionist stance and, on this ground, opposed interpretive sociology in the Weberian sense of Verstehen.28 Unity of science implied the translation of mental data into observable behavioral terms by behaviorist psychology largely supported by neopositivists (Carnap et al. 1929/2010; Hempel 1952/1965; Oppenheim and Putnam 1958). On this subject, Carl Hempel (1966, p. 110) refers the reader to Ernest Nagel (1961)’s “lucid” discussion of methodological individualism in his influential text on intertheoretical reduction, The Structure of Science. This text can shed light on MI’s interpretative drift in the reductionist direction. Ernest Nagel correctly identifies methodological individualism as the approach in social science applied to the understanding of social phenomena from the intentional, meaningful, or subjective categories of human experience, which he opposes to “methodological collectivism” or “holism” (Nagel 1961/1979, pp. 540–542), according to which (contrary to the holism of parts defined above) the whole acts as a separate cause, through the hypostasis of collective terms, or else, through the idea that social systems constitute “wholes” governed by sui generis macro-laws, etc. Nevertheless, Nagel reinterprets MI according to the logical-empiricist layered conception of the world and of science, where the parts of the wholes involved refer to entities with independent characteristics, by asserting that in MI: All the descriptive terms occurring in satisfactory explanations of social phenomena must belong to a special subclass of individual terms, namely, terms denoting ‘subjective’ or ‘psychological’ states of individual human beings. Methodological individualism thus subscribes to what is often advanced as a

27 28

On critical realism, see below. See in particular Abel (1948), Hempel (1952/1965) and, for a general perspective, Uebel (2010).

310

N. Bulle

factual thesis (although it is perhaps better regarded as a program of research) concerning the reducibility of all statements about social phenomena to a special class of (‘psychological’) statements about individual human conduct. (Nagel 1961/1979, pp. 541–542; see also Oppenheim and Putnam 1958, pp. 17–18)

This reductionist description of MI, essentially involving terms describing individual psychological states, is inappropriate, but it follows from a logical empiricist view of the structure of science. Since the general project of MI cannot satisfy the neopositivist ambitions for unity of science, involving the strictly empirical approach of the then in favor of behaviorist psychology, Nagel deduces the limited interest of the associated research program. This is an important point, which may help us to better understand the difficulties of reception of MI in circles dominated by the premises of analytic philosophy rooted in these empirical-physicalist frameworks. Nagel (1961/ 1979, pp. 473–475) defends in substance that only individuals’ overt behaviors really satisfy the requirements of scientific explanation, so that without denying the existence of subjective mental states, behaviorism represents a research program that “hopes to achieve a comprehensive system of explanation for human behavior through the ‘reduction’ of psychology to other sciences.“ Posterity has typically adhered to Nagel’s reductionist definition of MI, but often overlooks his adequate comprehension of the organic link between MI and “understanding” interpretation, which, in reality, stands in opposition to reductionism. This can be explained by the lack of interest in conscious phenomena of the still dominant empirico-physicalist trends of the analytic philosophy of sciences. As Alfred Schutz points out, Nagel’s reductionist framework of interpretation does not allow him to understand the specificity of the explanation in MI which refers to the causal role of reason, a role that the overt behavior approach cannot substitute for: Our authors are prevented from grasping the point of vital concern to social scientists by their basic philosophy of sensationalistic empiricism or logical positivism, which identifies experience with sensory observation (…) The identification of experience with sensory observation in general and of the experience of overt action in particular (and that is what Nagel proposes) excludes several dimensions of social reality from all possible inquiry. (Schutz 1954, pp. 261–262)

Methodological Individualism as Holism of the Parts …

311

Schutz argues that social scientists cannot interpret social processes simply as chains of events connected by external links, with the sole aim of establishing regularities. Besides, the individuals of MI who are apprehended in the perspective of the Weberian Verstehen have an essentially social character: From the outset, we, the actors on the social scene, experience the world we live in as a world both of nature and of culture, not as a private but as an intersubjective one, that is, as a world common to all of us, either actually given or potentially accessible to everyone; and this involves intercommunication and language. (Schutz 1954, p. 261)

However, the reductionist interpretation of MI derived from neopositivism has persisted. It is found in Steven Lukes (1968, p. 120)’s influential article where MI is defined as the doctrine according to which “facts about society and social phenomena are to be explained solely in terms of facts about individuals” and in many other critical texts of MI in philosophy of social sciences. In Zahle and Collins’ (2014, p. 8) more recent book supposedly about the individualism-holism debate, MI is explicitly defined as a model of inter-theoretic reduction where “individualist explanations involve individualist theories which contain descriptions of individuals only.”29 No reference is made in such definition to the criterion of subjectivity constitutive of MI, since its emergence with marginalist economics, but especially in the methodological developments of its major representatives in the social sciences.

The Bias of Non-Reductive Physicalism on MI: The Multiple-Realizability Argument Non-reductionist approaches to physicalism have been developed largely with the argument of multiple-realizability. This argument has been put forward particularly in the philosophy of mind. It is inspired by the computational metaphor and the analogy with the functional relationship between hardware and software: As long as certain computing principles are respected, the same program can be realized with different hardware supports, just as the same function (cognitive for example) can be realized in different physical systems (that is “multiple-realizability”) prohibiting the explanation of the functional properties of the wholes by those of their physical constituents, and thus posing a challenge to physicalist reductionism. Putnam (1967)’s 29

For a discussion of contemporary fallacies regarding methodological individualism within analytic philosophy, see Alban Bouvier (2023, this volume) and Di Iorio (2023, this volume).

312

N. Bulle

multiple realizability argument is notoriously based on the observation that mental states, such as pain, can be achieved by different physical states of the nervous system in different types of organisms, so that the (mental) function involved has to be distinguished from the (physical) basis that realizes it, even if it is dependent on it, because this dependency is not strict. Organizational structures at higher levels are assumed to have explanatory power as long as the elements at lower levels are interchangeable with other components that meet the conditions required for the structures to play their role. This is in line with the physicalist conception of the natural and human world made of physical components structured at different levels. This representation explains the opposition between reductionist approaches, which defend that the components at a given level explain in principle the phenomena resulting from their interactions in complex wholes, and non-reductionist approaches which defend the relative autonomy of these wholes, and correlatively, the autonomy of special sciences with respect to physics. I have discussed (with Denis Phan) elsewhere the multiple-realizability argument in relation to debates about the possible independence of analytical sociology from methodological individualism (see Bulle and Phan 2017). In essence, our argument is that there is explanatory complementarity between the structural generic properties (implemented by the multi-agent models favored by analytical sociology) and the semantic conditions of interpretation involved by the thematic conceptual research framework (MI for example). The structures have explanatory power in this respect which is of the order of partial causes, but the explanation, in order to be complete and to account for the generating mechanisms at work, must be based on an interpretative framework defined by the research, bringing into play the way in which the activities or causal powers of the entities involved (the individuals endowed with rational capacity in MI) underlie, and possibly condition, the structural logics highlighted.

The Emergentist Bias on MI: The Case of Critical Realism Contemporary emergentist conceptions are closely related to emergentist doctrines developed in the early twentieth century (see Mclaughlin 1992), which raise similar problems. They assume that entities at level Xn-1 are endowed with new causal properties when participating in wholes at the next level Xn of the natural hierarchy, due to the constitutive role of structures that link them. Critical realists30 defend a form of emergentism, based on an 30

Critical realism has the interest of discussing the necessity of a form of realism to justify the idea that the human mind can build knowledge of an independent reality. It was originally developed

Methodological Individualism as Holism of the Parts …

313

ontology of causal powers, where autonomous and causally efficient wholes are supposed to emerge at different levels of organization and complexity. Social wholes are no exception. According to the contemporary critical realist perspective, individuals are situated at an ontological level inferior to that of social wholes (i.e., considered in an independent state), and emerge as social agents, formed and empowered by social structures, by virtue of the “relationally emergent properties of things” (Elder-Vass 2010). This attribution of proper causal power to social structures which acts on individual parts tends to reify them and commit “the fallacy of collectivism” as noted by Rom Harré and Charles Varela (1996): The thesis that social structures are not causal factors but (efficient) causal powers is a violation of the principle that causal power “is the activity of powerful particulars at work.” Critical realists’ holism may be qualified as causal insofar as the whole is supposed to act as a separate cause on its parts. It opposes individuals and social structures as two interacting ontological levels. Hence, the reductionist interpretation of MI which denies that social wholes act as separate causes.31 We have seen that this is a misconception because MI, consistent with holism of the parts, from the outset holds individuals to be constituent parts of social wholes whose action or causal power, referring to their rational capacity, is inherent to their social being. In short, the physicalist ontology tends to support a layered conception of the world and science that confers an efficient causal power only to physical parts and, in the mainstream non-reductive approaches, an irreducible explanatory power to functional structures. Emergentist approaches that defend the existence of non-physical causal powers tend to depart from the physicalist framework, bringing into play the constitutive role that relations between entities have on their properties, as is the case with critical realism. MI has been misunderstood by both the reductionist and non-reductionist approaches developed within the framework of mainstream physicalism supported by analytical philosophy, as well as by emergentist approaches of critical realism. When it comes to the social world, all these approaches oppose individuals (assumed to have independent properties) and social structures as two separate levels of reality. Such metaphysics is incompatible with MI.

by Roy Wood Sellars (father of Wilfrid Sellars) whose work is largely neglected today, so much so that contemporary proponents of critical realism generally make no mention of it (Groff 2008 is an exception). 31 See Bhaskar (1979, p. 26), and the discussion of critical realism of Di Iorio and I (2023, this volume).

314

6

N. Bulle

From Ontology to Epistemology: The Issue of Creative Evolution

I have argued here that MI represents a social science method based on an analytical decomposition of wholes into basic units comparable to methods in other sciences, as is the case in various approaches of psychology. Such decomposition defines theoretical units that are active units involved in scientific explanation and whose properties or causal powers are most often hold as inherent to their belonging to the whole they form. According to MI, basic units of social sciences are social actors endowed with a general rational capacity which represents a trans-situational capacity to attribute meaning or, in other words, to understand. The rationality principle involved in MI assumes that the relevant subjective factors of individuals’ situations (including internal structures of meaning) can be objectified, alongside external social features, to account for the reasons underlying their thoughts and actions. This principle justifies the understanding perspective of MI, that is, the ability of the interpreter to understand the motives of social actors by trying to grasp their situation, as they perceive it, albeit in an abstract, ideal-typical way. This is possible because the interpreter is presumed to share the same fundamental rational capacity as the social actors. Even if reasons often recede into the unconscious, the observer can account for them on the basis of the implicit meaning they hold for the social actor. Moreover, the MI method is “individualist” in that the causality of social wholes is exhausted by the causal operation of their parts, the social actors. Consequently, MI represents a fundamentally non-positivist and non-reductionist methodological approach for the social sciences, which is consistent with an ontology referring to a holism of parts originally defined by Smuts (1926) according to which (1) the properties and activities of the parts are inherent to their belonging to the whole they form, and (2) the whole does not act as a separate cause distinct from its parts. Additionaly, I have argued that misunderstandings of MI in the philosophy of social sciences literature are rooted in epistemological and ontological issues. They can be illuminated by considering three alternatives concerning the relations of parts to wholes among theoretical constructs. These are the reductionist and non-reductionist physicalist approaches or emergentist alternatives on the one hand, and the holism of parts on the other. In explaining phenomena at level Xn, reductionist physicalists involve physical entities at level Xn-1, whereas non-reductionist physicalists and emergentists consider the causal power of wholes at level Xn. Proponents of these perspectives incorrectly assume that individuals in MI are endowed with independent

Methodological Individualism as Holism of the Parts …

315

characteristics from their belonging to social wholes. The holism of parts, which underpins the genuine ontological stance of MI, engages a fundamentally different approach that can be supported by the notion of creative evolution. Based on this notion, throughout evolution, the development of new causal capacities underlies the appearance of intrinsically new entities and properties forming new wholes. This is exemplified by the human rational capacity, which is one with human social being, opening a third path between reductionist and non-reductionist varieties of contemporary physicalism or emergentist conceptions. According to this third path, the parts involved in MI have no separate existence from the social wholes (reductionism rejected) while the social wholes do not act as separate causes (causal holism rejected).

References Abel, T. (1948). The Operation Called Verstehen. American Journal of Sociology, 54 (3), 211–218. Ayer, A. J. (ed.) (1959). Logical Positivism. New York: Free Press. Bhaskar, R. (1979). The Possibility of Naturalism. Brighton: Humanities Press. Bird, A., Ellis, B., Sankey, H. (2012). Properties, Powers and Structures: Issues in the Metaphysics of Realism. London: Routledge Bostaph, S. (1978). The Methodological Debate Between Carl Menger and the German Historicists. Atlantic Economic Journal , 6 (3), 3–16. Boudon, R. (1999/2001). The Origin of Values: Sociology and Philosophy of Beliefs. London: Routledge. Bouvier, A. (2023). Methodological Individualism Facing Recent Criticisms from Analytic Philosophy Artificial Reconstructions and Genuine Controversies. This volume. Brodbeck, M. (1954). On the Philosophy of the Social Sciences. Philosophy of Science, 21(2), 140–156. Bulle, N., Phan, D. (2017). Can Analytical Sociology Do Without Methodological Individualism? Philosophy of the Social Sciences, 47 (6), 379–409. Bulle, N. (2018). Methodological Individualism as Anti-Reductionism. Journal of Classical Sociology, 19 (2), 161–184. Bulle, N. (2021). Vygotsky versus Dewey on Mental Causation: The Core of Two Divergent Conceptions of Human Thought. New Ideas in Psychology, 63(4): 100898. Bulle, N. (2022). Rationality as a Meta-analytical Capacity of the Human Mind: From the Social Sciences to Gödel. Philosophy of the Social Sciences. https://doi. org/10.1177/00483931221115345

316

N. Bulle

Bulle, N. (2023). Methodological Individualism: Introduction and Founding Texts. London: Routledge (open access, Forthcoming). Bulle, N., Di Iorio, F. (2023a). Methodological Individualism and Institutional Individualism: A Discussion with Joseph Agassi. This volume. Bulle, N., Di Iorio, F. (2023b). Methodological Individualism and Critical Realism: Questions for Margaret Archer. This volume. Campagnolo, G. (2010). Criticism of Classical Political Economy. Menger, Austrian Economics and the German Historical School . London: Routledge. Carnap, R., Hahn, H., Neurath, O. (1929/2012). Wissenschaftliche Weltauffassung – Der Wiener Kreis. In Friedrich Stadler and Thomas Uebel (Eds.) Wissenschaftliche Weltauffassung. Der Wiener Kreis. Reprint of original Edition with Translations (pp. 75–116). Vienna: Springer. Cartwright, N. (1989). Nature’s Capacities and Their Measurement. Oxford: Clarendon Press. Cartwright, N. (1999). The Dappled World. A Study of the Boundaries of Science. Cambridge: Cambridge University Press. Cherkaoui, M. (2005). Invisible Codes: Essays on Generative Mechanisms. Oxford: Bardwell Press. Cowan, R., Rizzo, M. (1996). The Genetic-Causal Tradition and Modern Economic Theory. Kyklos, 49, 273–317. Crafford, F. S. (1943). Jan Smuts: A Biography. Kessinger Publishing. Craver, C. F., Bechtel, W. (2007). Top-Down Causality Without Top-Down Causes. Biology and Philosophy, 22, 547–63. Cubeddu, R. (1993). The Philosophy of the Austrian School . New York: Taylor and Francis. Di Iorio, F. (2016). World 3 and Methodological Individualism in Popper’s Thought. Philosophy of the Social Sciences, 46 (4), 352–374. Di Iorio, F. (2023). Methodological Individualism and Reductionism. This volume. Di Nuoscio, E. (2018). On the Explanation of Human Action: ‘Good Reasons’, Critical Rationalism and Argumentation Theory. In G. Bronner and F. Di Iorio (Eds.) The Mystery of Rationality: Mind, Beliefs and the Social Sciences (pp. 37–52). New York: Springer. Elder-Vass, D. (2010). The Causal Power of Social Structures. Emergence, Structure and Agency. Cambridge: Cambridge University Press. Ellis, B. (2009). The Metaphysics of Scientific Realism. New York: Routledge. Feuerhahn, W. (2023). Max Weber and Understanding Explanation. This volume. Grassl, W., Smith, B. (1986). Austrian Economics: Historical and Philosophical Background . London: Croom Helm/Routledge. Groff, R. (Ed.) (2008). Revitalizing Causality: Realism About Causality in Philosophy and Social Science. London: Routledge. Harré, R. (1970). The Principles of Scientific Thinking. Chicago: The University of Chicago Press. Harré, R., Madden, E. H. (1973). Natural Powers and Powerful Natures. Philosophy, 48(185), 209–230.

Methodological Individualism as Holism of the Parts …

317

Harré, H. R., Varela, C. R. (1996). Conflicting Varieties of Realism: Causal Powers and the Problems of Social Structure. Journal for the Theory of Social Behaviour, 26 , 313–25. Harré R. (2006). Un champ sémantique étendu pour les dispositions et le rôle fondationnel des pouvoirs causaux. In B. Gnassounou, M. Kistler (Eds.) Causes, pouvoirs, dispositions en philosophie [Causes, Powers, Dispositions in Philosophy] (pp. 119–133). Paris: Éditions de la rue d’Ulm/PUF. Hayek, F. (1952) The Counter-Revolution of Science: studies on the abuse of reason. New York: The Free Press. Hayek, F. (1952/1976). The Sensory Order. An Inquiry into the Foundations of Theoretical Psychology. Chicago: The University of Chicago press. Hempel, C. G. (1952/1965). Typological Methods in the Natural and the Social Sciences. In Carl Hempel (Ed.) Aspects of Scientific Explanation and other Essays in the Philosophy of Science (pp. 155–171). New York: The Free Press/CollierMacMillan. Hempel, C. G. (1966). Philosophy of Natural Science. Englewood Cliffs: PrenticeHall. Jàros, G. (2002). Holism Revisited: Its Principles 75 Years On. World Futures: The Journal of General Evolution, 58(1), 13–32. Kemeny, J., Oppenheim, P. (1956). On Reduction. Philosophical Studies, 7 (1-2), 6-19. Kim, J. (2003). The American Origins of Philosophical Naturalism. Journal of Philosophical Research 28(9999), 83–98. Kistler, M. (2007). La réduction, l’émergence, l’unité de la science et les niveaux de réalité. Matière première, 67–97. Koestler, A. (1978). Janus: The summing up. London: Picador. Koffka, K. (1935). Principles of Gestalt psychology. London: Routledge & Kegan Paul. Lukes, S. (1968). Methodological Individualism Reconsidered. The British Journal of Sociology, 19 (2), 119–129. Machamer, P., Darden, L., Craver, C. F. (2000). Thinking About Mechanisms. Philosophy of Science, 67 (1), 1–25. Mandelbaum, M. (1955). Societal Facts. British Journal of Sociology, 6 , 305–317. Mandelbaum, M. (1957). Societal Laws. British Journal for the Philosophy of Science, 8, 211–224. Marmodoro, A. (ed.) (2010). The Metaphysics of Powers: Their Grounding and Their Manifestations. London: Routledge. Mclaughlin, B. P. (1992). The Rise and Fall of British Emergentism. In A. Beckermann, H. Flohr, J. Kim (Eds.) Emergence or Reduction? Essays on the Prospects of Nonreductive Physicalism (pp. 49–93). Berlin: Walter de Gruyter. Menger, C. (1883/1986). Investigations into the Methods of the Social Sciences. New York: New York University Press. Mesure, S. (2023). Dignity and Axiological Rationality, The Legacy of Raymond Boudon. (This volume) Meyerson, É. (1908). Identité et Réalité. Paris: Félix Alcan.

318

N. Bulle

von Mises, L. (1949/1966). Human Action. A Treatise on Economics. Chicago: Contemporary Books. Morgan, C. L. (1923). Emergent Evolution. Londres: Williams & Norgate. Morin, J.-M. (2023). Ordinary Rationality Theory (ORT). According to Raymond Boudon. This volume. Nagel, E. (1961/1979). The Structure of Science. Problems in the Logic of Scientific Explanation. Cambridge: Hackett Publishing Company. Opp, K-D. (2023). Methodological Individualism and Micro-Macro Modeling. This volume. Oppenheim, P., Putnam, H. (1958). The Unity of Science as a Working Hypothesis. In H. Feigl, M. Scriven, G. Maxwell (Eds.) Concepts, Theories, and the Mind-Body Problem (pp. 3–36). Minneapolis: University of Minnesota Press. Phillips, D. C. (1976). Holistic Thought in Social Science. Stanford: Standford University Press. Popper, K. (1957). The Poverty of Historicism. London: Routledge & Kegan Paul Popper, K. (1965/1972). Objective Knowledge. Oxford: Clarendon Press. Popper, K. (1978a). Natural Selection and the Emergence of Mind. Dialectica, 32(3/ 4), 339–355. Popper, K. (1978b). Three World . The Tanner Lecture on Human Values Delivered at The University of Michigan April 7, 1978b. Potochnik A., McGill, B. (2012). The Limitations of Hierarchical Organization. Philosophy of Science, 79 (1), 120–140. Poynton, J. C. (1987). Smuts’s Holism and Evolution Sixty Years On. Transactions of the Royal Society of South Africa, 46 (3), 181–189. Putnam, H. (1967). Psychological Predicates. In W. H. Capitan, D. D. Merrill (Eds.) Art, Mind, and Religion (pp. 37–48). Pittsburgh: University of Pittsburgh Press. Ramström, G. (2018/2023). The Analytical Micro–Macro Relationship in Social Science and Its Implications for the Individualism-Holism Debate. Philosophy of the Social Sciences, 48(5), 474–500, reprinted in this volume. Schutz, A. (1954). Concept and Theory Formation in the Social Sciences. The Journal of Philosophy, 51(9), 257–273. Sidis, B. (1908). Psychological Review, 15 (1), 44–68. Smith, B. (1990). Aristotle, Menger, Mises: An Essay in the Metaphysics of Economics. In B. Caldwell (Ed.) Carl Menger and His Legacy in Economics (pp. 263–288). Durham/London: Duke University Press. Smuts, J. C. (1926). Holism and Evolution. New York: The Macmillan Company. Uebel, T. (2010). Opposition to Verstehen in Orthodox Logical Empiricism. In U. Feest (Ed.) Historical Perspectives on Erklären and Verstehen (pp. 291–310). New York: Springer. Vygotsky, L. S. (1934/1986). Thought and Language. Cambridge: MIT Press. Weber, M. (1903–1906/1975). Roscher and Knies and the Logical Problems of Historical Economics. New York: The Free Press.

Methodological Individualism as Holism of the Parts …

319

Weber, M. (1908/1972). Georg Simmel as Sociologist. Social Research, 39 (1), 155– 163. Weber, M. (1922/1981). Some Categories of Interpretive Sociology. The Sociological Quarterly, 22, 151–180 Weber, M. (1922/1968). Economy and Society: An Outline of Interpretive Sociology. Berkeley: University of California Press. Weber, M. (1922/1958). From Max Weber: Essays in Sociology. In H. H. Gerth, C. Wright Mills. Oxford: Oxford University Press. Weber, M. (1922/1949). The Methodology of the Social Sciences. Translated and edited by Edward A. Shils and Henry A. Finch. Glencoe: The Free Press. Zahle, J., Collin, F. (eds.) (2014). Rethinking the Individualism-Holism Debate: Essays in the Philosophy of Science. Dordrecht: Springer.

Max Weber’s Methodological Individualism Thomas Schwinn

1

Introduction

Max Weber is undisputedly an important progenitor of methodological individualism.1 Wherever social scientists define and develop their own theoretical position, they do so in relation to him as a reference author and a benchmark. This is true for the alternative enterprise of methodological collectivism in its most elaborated system-theoretical variants in Talcott Parsons and Niklas Luhmann (Schwinn 2013). Structuration theories also refer to him (Giddens 1984, p. xxxvi; Archer 1995), as well as exponents of rationality theories (Boudon 1987; Coleman 1987; Esser 1993; Hedström and Ylikoski 2019). However, these discussion fronts have not led to clearly outlined research programs. Concerning the methodological individualism discussed in this article, some researchers complain about “a bewildering number of different versions” (Udehn 2001, p. 346), like weak and strong versions. Additionally to this, there is no consensus on the question of what constitutes the methodological core and the status of the Weber paradigm. Some interpreters, such as his wife Marianne Weber (1988 [1926], p. 265, T. Schwinn (B) Sociology, Max-Weber-Institute of Sociology, University of Heidelberg, Heidelberg, Germany e-mail: [email protected] 1 Carl Menger, who was born 24 years before Weber, influenced his thinking as well as that of the entire Austrian branch of Methodological Individualism.

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 N. Bulle and F. Di Iorio (eds.), The Palgrave Handbook of Methodological Individualism, https://doi.org/10.1007/978-3-031-41512-8_16

321

322

T. Schwinn

308f.) or Hennis (1988, p. 157f.) regarded his methodological writings as unsystematic occasional work, without any significant importance for his work (similar Lichtblau 2020, p. 218). Tenbruck, who initially followed this judgment (Tenbruck 1959), completely revised it some decades later and now discovered in Weber’s methodological work a unity that could be understood as the key to his entire work (Tenbruck 1989). Albert even perceives a “moderate methodological holism” in Weber’s work, a term that he subsequently corrects: “holistic individualism” was the appropriate term (Albert 2009, p. 518). Even an excellent expert like Schluchter, who has contributed like no other to the systematization of Weber’s sociology, is undecided and unsure: from “methodological individualism” (Schluchter 2003, p. 59) to “moderate methodological holism” (Schluchter 2007, p. 307) up to “methodological relationism” (Schluchter 2018, pp. 10, 22) in the most recent works, he uses changing terms for Weber’s methodology. Following Albert, Schluchter (2018, p. 229) speaks of a “third sociology” beyond the two methodological families of holism and individualism.2 These assessments and controversies also move on an epistemological and ontological level.3 Methodological questions are certainly not entirely independent of these, but not identical with them. They can be treated separately (Henrich 1952, p. 15). Epistemological and ontological categories constitute our concept of reality, methodological forms make them perceptible and provide methodological guidance on how to attain true scientific knowledge. The focus of this article lies in the latter mentioned. Methodological questions shape the basic sociological problem of how action and order, micro and macro level, can be related to one another. “Of all the classical sociologists, only Weber seemed to see a clear way out of this traditional dilemma. Although in the end Weber’s formulations are still not entirely satisfactory … his contributions have remained central to every subsequent effort at establishing a micro–macro link” (Alexander and Giesen 1987, p. 15). This assessment contrasts in a remarkable way with an insufficient reconstruction of Weber’s version of the micro–macro link in the secondary literature. Punctual references dominate, without giving an idea of his general conception. I would like to remedy this deficit by revealing the differentiated methodological terms that Weber offers for the problem of action and order. In his model

2 For an excellent critique of these hybrid formulations, which create more confusion than clarity, see Greve (2008). 3 The relationship between epistemological and ontological positions in Weber is controversial. Whereas Henrich (1952, p. 11, 35) can detect no explicit epistemological theory in Weber, for Schluchter (2006, p. 218ff.) the Neo-Kantian epistemology of Heinrich Rickert and Emil Lask is valid for Weber. It is not possible to follow this complicated discussion here.

Max Weber’s Methodological Individualism

323

he still integrates what disintegrates in the respective schools later, microsociology on the one hand and macro-sociology on the other hand.4 This makes him interesting for the current theoretical discussion.

2

Ontological Premises of Weber’s Methodology

First, the ontological premises of Weber’s methodology, his basic understanding of the individual and the social will be briefly presented. For him, the single individual and his actions are the only bearer of a meaningful behaving, “its most basic unit, its ‘atom’” (CMW (Weber 2012), p. 280; WL (Weber 1982), p. 439; ES (Weber 1978a), p. 13f.; WG (Weber 1980), p. 6f.). The human subject is also a complex of genetic, neurological, and psychological processes. But they do not form the basis of his interpretive sociology. At best, they may be a condition for the constitution of meaning, but it cannot be derived from them. Hence, Weber explicitly distances himself from psychologism (CMW, p. 90, 276; WL, p. 140, 432; ES, p. 13, 19; WG, p. 6, 9, cf. Weiß 2018, p. 10f., 14, 72). The market actions of an actor, for instance, cannot be made intelligible through a psychological analysis, but through an analysis of the situation, such as the supply-demand relationships. Popper (1966, p. 89ff.) later speaks of the “logic of the situation” which cannot be derived from psycho-logic. The meanings that orientate and motivate human action, which can be found in religion, politics, economy, science, etc., have their own kind of rationality. Of course, human action has psychological preconditions, but the meanings that appear in the motives must not be derived from them. Thus, psychology is not a basic science of sociology and Weber’s individualism is not a psychological one (Udehn 2002, p. 482). The individual is also the “upper limit and the only agent of meaningrelated behaviour” (CMW, p. 280; WL, p. 439). To collective entities such as “the state” or “the economy” or “society” must not be imputed any subjectanalogous qualities of action. According to Weber, collectivism is particularly 4

This can also be justified in terms of the history of theories. The classics Alfred Schütz and Talcott Parsons had a quite significant influence on the methodological debates. Both build on Weber’s work and regard their own work as a further development of his sociology (cf. the correspondence between Schütz and Parsons in 1940/41, Schütz and Parsons 1977). However, they come to completely different theoretical concepts, which significantly shaped the sociological micro–macro topic through the respective school-forming influences. If Schütz’s and Parsons’ selective connections to Weber are an essential cause of this split, the reconstruction of Weber’s sociological program promises a retrospective correction of this development (Schwinn 1993).

324

T. Schwinn

susceptible to ideologies. “The use of the undifferentiated, collective concepts belonging to everyday language is always a cover for woolly thinking or unclear aims of action; quite often, it is a tool that leads to questionable half-truths; and it invariably serves to inhibit the correct formulation of the problem” (CMW, p. 137; WL, p. 212; see also Burger 1994, p. 83ff.; Müller 2020, p. 50). All social phenomena are to be explained in terms of the mutual orientation of the actors and cannot be separated from them. “The ontological peculiarity of the social lies in the fact that it ceases to exist without the orienting contribution of individuals” (Greve 2008, p. 72). The causal potential of social phenomena and characteristics always has to be manifest in individuals, too. The influence of the social is an influence of individuals on individuals. According to Weber, there is therefore no emergence beyond certain levels of complexity, as if the quality of the social would change beyond a certain threshold. A relationship, such as a friendship has the same ontological status and requires the same methodology as a state. They are to be analyzed from the orientations and actions of individuals, whose meaning orientations are reciprocal or unilateral and whose meaning contents can coincide in different degrees (ES 27f.; WG, p. 13f.; CMW, p. 130; WL, p. 200f.). It was wrongly assumed a “monological, pre-communicative model of action” in Weber’s work (Habermas 1984, I, p. 279f.). This is a misjudgment, since Weber does not pursue any “methodological atomism” (Albert 2009, p. 518; Greve 2008, p. 72; cf. also Udehn 2002, p. 500). Subjects don’t find their orientation within or starting from closed monads, as “isolated subjects,” but they are directed toward others and their action has to be clarified from this reciprocal relationship. “An individualism that follows Weber does not presuppose that what is individual can be clearly defined in terms of its content beyond social orientations; rather, there is only a minimal anthropological conception in the background, which regards people as needy and interpretive beings, but this is a thin concept that by no means deduces social structures from a fixed human nature” (Greve 2008, p. 67; emphasis by TS). Subjects do not start in a social zero state, as if the actors were spinning the social relationships out of themselves without any precondition. A basic premise of Weber’s thinking is that all action is historically embedded (CMW, p. 114f.; WL, p. 172). As a result, the actors always relate to each other via predefined and pre-structured conditions that have to be considered. Weber’s basic terms of the social via the reciprocal orientation of actors avoid two recurrent reifications that are reflecting each other: a collectivism that breaks away from the human subjects and a pre- or non-social psychologism or solipsism. Following Giddens and Archer, Schluchter (2018,

Max Weber’s Methodological Individualism

325

p. 3f., 11, 22) speaks of a duality of structure and action in Weber’s work. Duality and not dualism that ontologically reifies the two sides. Weber’s methodological instruction based on these premises is presented now.

3

The Concept of Chance, Ideal Types, and Types of Social Action

Weber’s understanding of social relationship and order is a very broad one. “Within the realm of social action certain empirical uniformities can be observed, that is, courses of action that are repeated by the actor or (simultaneously) occur among numerous actors since the subjective meaning is meant to be the same. Sociological investigation is concerned with these typical modes of action” (ES, p. 29; WG, p. 14). How are these “typical modes of action” explained? Weber operates with three concepts to grasp the problem of action and social order: the types of social action, the concept of chance, and the ideal type. The types of social action in § 2 of the Basic Sociological Terms (ES, p. 24f.; WG, p. 12f.) are based on different motives for social action. Purposively rational, value-rational, affectual, and traditional action not only address whether an actor bases his actions on subjective meanings and in which way. They also give different possibilities of how actors form relationships with each other and on what basis these relationships can be made permanent. Accordingly, the four types of action correspond to four types of order: coordination by self-interest, legitimate order, manners and customs, and affective communal relationship.5 The correspondence of types of action and order follows from Weber’s thesis that sociality only exists through references to the behavior of others and that principally it has to be explained from this perspective. The fact that there is no micro-sociological reductionism in this becomes clear by further developing the basic definitions of the action-order relationship through the concept of chance. Being reproduced and thus the existence of a social order is based on the chance of the course of a certain pattern of human action. “In order to avoid fallacious impressions, let it be repeated that it is only the existence of the probability that, corresponding to a given subjective meaning, a certain type of action will take place which constitutes the ‘existence’ of the social relationship” (ES, p. 28; WG, p. 14). The concept of chance opens up analytical insight into the way social orders are reproduced. 5 Affective communal relationship is missing in section 4 and 5 of the Basic Sociological Terms. However, it has to be adopted as an independent type of order.

326

T. Schwinn

In rational choice approaches, the micro-macro linking is referred to as the problem of transformation (Lindenberg 1977; Coleman 1987; Wippler and Lindenberg 1987; Esser 1991, p. 40ff.). How do you get from the actions of the actors and their effects to the collective level and processes and vice versa from the macro to the micro level? How do these two levels interact? Weber tries to answer this question using the concept of chance.6 This term bundles a number of questions that are central to the micro-macro topic. Weber distinguishes between the subjective and the objective chance. The empirical validity of an order can mean two things: “(1) that individuals will in fact (subjectively) on average – like the cheat and the thief – expect that the other societized persons will, on average, behave ‚as if‘ they were guided by the wish to comply with the instituted order; or (2) that individuals could objectively have that expectation – based on an (on average) applicable estimation of probable human conduct (a particular variety of the category of ‘adequate causation’). Logically, (1) and (2) should in principle be strictly distinguished from each other: one is an expectation that is subjectively held – that is to say: the researcher assumes that it is ‘on average’ held – by the acting person who is the object of investigation; the other is a probability to be calculated objectively by the subject seeking knowledge (the researcher), taking into account the probable knowledge and ways of thinking of the participants” (CMW, p. 283, 291; WL, p. 443f., cf. 456; Winckelmann 1976, p. 21f.; Hopf 1991, p. 132ff.).7 Here Weber addresses two different expectations and thus terms of chance. In the first case, it is about the subjective expectations of the actors that the others will behave as if they were guided by the order, i.e. the subjectively assessed chance. In the second case, an analyzer estimates the objective chance based on his methodically controlled knowledge using the category of objective possibility. The importance of this distinction can be demonstrated by two examples. In Nazi Germany, people believed in a “thousand-year Reich.” This widespread subjective chance calculation of the reproduction of the political order gave it a high legitimacy. For a scientific observer, this subjective calculation of chances could not correspond to the objective calculation 6

The concept of chance has found few attention in the secondary literature. Dahrendorf (1979, p. 93ff.) is one of the few authors who takes it up and even sees it as a “key” to Weber’s methodology. However, there is no systematic reconstruction in Dahrendorfs work. In Weber’s work he can only discover a “chance salad” (Dahrendorf 1979, p. 104). Dahrendorf (1979, p. 93) discovers in the first volume of Economy and Society alone over a hundred passages in which Weber uses the word and the term “chance”. In the edition published in English by Roth and Wittich (ES = Weber 1978a), the term is missing from the index; it is listed in the German edition (WG = Weber 1980, p. 882) and in CMW, Weber (2012, p. 535). 7 In this quotation, Weber refers to purposively rational orders. However, these reflections can be transferred to the other types of order; see CMW, p. 281f., 291; WL, p. 441, 456.

Max Weber’s Methodological Individualism

327

of chances. At the latest with the beginning of the war, it was foreseeable that the numerous opponents and fronts could neither be managed militarily by the strength of the German Wehrmacht nor economically by the potential of the German economy. Before the financial crisis of 2008, borrowers and lenders subjectively assessed the possibilities of financing a property as particularly easy. To an experienced observer of the financial market though there were few who warned—this opened up fictitious and unrealistic repayment possibilities. In both examples, it was precisely the difference between subjective and objective calculations of chance that led to the reproduction of a political and an economic order for a certain period of time and then to its collapse or to the crisis. The distinction between subjective and objective chance calculation only makes sense for a scientific observer. For a participant, the objective conditions he perceives are always already integrated into his subjectively calculated chance, otherwise he would not act.8 A scientific analysis, on the other hand, must clarify the extent to which the participants’ subjectively assumed conditions of action stand up to objective scrutiny. These are the non-intentional effects of action. Such analyses are one of the primary tasks of the social sciences. With these specifications, one can conceive the micro-macro link in a differentiated manner. The concept of chance helps on the one hand to avoid the hypostatization of collective concepts and on the other hand it avoids their reduction to the micro level. It is unimportant at which level one starts the analysis, as long as feedback processes are considered. Contexts initially set “objective” conditions in the sense of opportunities and limitations for action. These conditioning factors have an influence on the actors, they are included in the subjective chance calculations. Therefore, this subjective side must not be neglected. It has considerable consequences how actors perceive the situation, how they evaluate it, and what expectations they have. However, one must not remain at this level. Actors seldom fully consider and anticipate the consequences of their actions. This is the task of the scientific observer. The objective calculation of chance comprises the subjectively expected consequences of action as well as the consequences that are independent of intentions. In the course of action actors consider the shifts of conditions, do they remain the same or change from situation t1 to t2 , in variable degrees. In this perspective, orders can only be thought dynamically. They only exist as 8 In rational choice theories, an actor’s action is the product of imagined utility and the probability of achieving the goal of action. However, both aspects are subjective constructs of the actor, including the chance of achieving a goal under the given conditions.

328

T. Schwinn

a process. The chances must be used and activated in actions. As a result, the scope of possibilities is either reproduced or modified. Orders live on the continuous interaction of their regulatory specifications for action and the maintenance or change of their patterns through action.9 Therefore, the term “order” always means two things: instructing action in accordance with their standards and the fact that action is generating these standards. Here, the micro and the macro level can just be distinguished analytically (cf. Alexander and Giesen 1987, p. 1). “Macro-laws” only work as long as the implicit assumptions about the actors’ chance calculations and the resulting courses of action, as well as the conditioning factors remain stable within certain limits. Inversely, the “micro-logic” or logic of the situation is only the same as long as the assumed contextual conditions remain constant. Another central concept of Weber’s methodology is the “ideal type.” In addition to the concept of chance, it opens up an insight into the way orders are reproduced. The way how the chances are perceived by the actors is crucial for the existence of an order. How are the opportunities perceived and used? Using the concept of the state, he illustrates the variety of people’s perceptions of and their reactions to an order (CMW, p. 130; WL, p. 200f.). This variety of modes of perception and reaction cannot be dealt with by mere aggregation, i.e., the mere summation of all individual real ideas and acts in which something of this order is actualized, but only by means of an ideal-typical analysis. In order to grasp the empirical consequences of the often only fragmentary aspects of meaning and action, they must be mentally “extended” to form the unity of the unreal (i.e., ideal-typical) structure. The fragments of meaning in the motives of the actors are inserted into a more comprehensive picture in which those motives are ideally “at home” (von Schelting 1934, p. 381ff.; Schwinn 2013, p. 1ff.).10 But in order to know on what basis the actions may be put together, how to measure them, one has to know what is the vanishing point of this construct. Weber distinguishes between ideal types of a general character and those of an individual character. The ideal types of a general character are “theoretical constructions in which empirical material is used for illustrative purposes” (CMW, p. 133; WL, p. 205). The duty of sociology is the construction of such general concepts in order to facilitate the causal attribution and analysis (ES, p. 19f.; WG, p. 9f.). These general terms are formed under idealized

9

Giddens (1979, p. 69ff.) calls this the “duality of structures”: these are both the medium and the result of action. 10 For example, the legal constitution of a legitimate order is an important framework. However, the sociological, in contrast to the legal analysis of a legitimate order, must not concentrate solely on these formal elements, but must also take into account the factual ones.

Max Weber’s Methodological Individualism

329

conditions. The tenets of national economics serve as a model for Weber. Assuming a homo oeconomicus therein, fictitious teleological action schemes and processes are formed (ES, p. 8f., 20f.; WG, p. 4, 10; CMW, p. 84, 247ff.; WL, p. 130, 394ff.). Accordingly, Weber constructs the general type of order by interests. The most important example for this is the economic market. The participants want to realize their interests. In order to do this, they have to consider the interests of each other under supply-demand conditions. How could you comprehend such an order? In accordance with neoKantianism Weber assumes that the full complexity of reality couldn’t be grasped. Scientific analysis always need theoretical models of reality. “Model” means reduction of this complexity by concentration on certain features. That’s why a theoretical model is not reality, but an idealized picture. Weber calls this “ideal-type.” In the case of a social coordination by interests, the idealization or theoretical construction consists in isolating the purposively rational action orientation and in seeing how such action would take place, given the hypothetical assumption that the actors have a complete insight into the situational circumstances and that without exception they take the most suitable means for the pursued purposes according to efficiency criteria. By acting strictly purposive rational, through error, affects or value considerations uninhibited, and by reacting in the same way to given situations, actors create similarities, regularities, and continuities of action (ES, p. 30; WG, p. 15). Order does not arise here through the intervention of any values or norms, but through the criterion of purposive rationality as the determining factor of social action. According to Weber, in contrast to a one-sided, normativist treatment of the problem of order in the tradition of Durkheim and Parsons, the regularities of action resulting from the mutual intertwining of interests represent an independent type of order that is not necessarily less stable than the one resulting from norms. In Parsons’ interpretation, the term interest does not function without addressing the “ultimate goals” in which the interests are grounded (Parsons 1968, p. 653ff.). Weber, in contrast, is of the opinion that “many of the especially notable uniformities in the course of social action are not determined by orientation to any sort of norm which is held to be valid, nor do they rest on custom, but entirely on the fact that the corresponding type of social action is in the nature of the case best adapted to the normal interests of the actors as they themselves are aware of them” (ES, p. 30; WG, p. 15). One could get the impression here that Weber is assuming that ordering by interest implies a return to rational choice and neoclassic economics. For these theories, the interest model is the only one, for Weber it’s one under several in his analytical tool box. And he distinguishes between

330

T. Schwinn

general ideal type and individual ideal type. As will be shown below, the latter is always a mixture between several types of action. The general ideal type of the order of interests is constructed in such a way that the subjective and objective calculation of chances coincide. The course of an order or macro-processes are cognitively present to the actors, and they continuously adjust their actions accordingly. “However, when constructing general concepts, sociology also attributes to the acting persons the average subjective intellectual ‘capacity’ required for estimating that probability. That is to say: [sociology], once and for all, ideal-typically assumes that the objectively existing average probabilities will also, on average, be taken approximately into account by persons acting with purposive rationality” (CMW, p. 283; WL, p. 444). Why does Weber suppose this unrealistic premise that subjective and objective calculation of chances coincide? Comparing the fictional general ideal type with the real conditions “help us to identify precisely those elements of actual economic action that are teleologically non-rational and, consequently, to make the actual course of that action understandable” (CMW, p. 84; WL, p. 130f.). Only by constructing the model of a perfect homo oeconomicus, i.e., ideal type, it is possible to understand the real reproduction of an order by measuring their deviation from the model. The motives of social action are not exhausted with the purposive rational orientation. Weber’s transition from the economic to the sociological frame of reference is guided by the insight that the former rests on a too narrow model of action (Schluchter 1988, I, p. 23ff.). The types of action in § 2 of the Basic Sociological Terms (ES, p. 24; WG, p. 12f.) outline this sociological frame of reference. Social action can also be determined and made permanent by traditional orientations. The general ideal type of order manners/customs (ES, p. 29; WG, p. 15)11 is based on a different idealization than in the case of interest. The traditional attitude is isolated and it is assumed that the behavior runs without any purpose or value-rational, or affective moments of orientation. “Strictly traditional behavior, like the reactive type of imitation discussed above, lies very close to the borderline of what can justifiably be called meaningfully oriented action, and indeed often on the other side. For it is very often a matter of almost automatic reaction to habitual stimuli which guide behavior in a course which has been repeatedly followed. The great bulk of all everyday action to which people have become habitually accustomed approaches this type” (ES, p. 25; WG, p. 12) One can speak here of a de facto order. The habitual consolidations are no longer guided by any

11

Today one would rather speak of habitually solidified courses of action.

Max Weber’s Methodological Individualism

331

subjective calculations of opportunity. Here, the processes of order take place independently of the intentions of the actors. The chance calculations made by the scientific observer do not correspond to any subjective ones on the side of the actors. Objective chance calculation means the determination of those conditions under which such habitual regularities are reproduced: e.g., stable environmental and contextual conditions, but also possibly value-rational or purposively rational intentions at the origin of custom and manner, which would then be lost. Regularities of social action that are neither purposively rational, traditional, or affectual, but dominantly value-rational are a central object of Weber’s sociology. He calls them legitimate orders (ES, p. 31, 212ff.; WG, p. 16f., 123). For their general ideal type, an idealized actor is assumed, similar to the homo oeconomicus of economic theorems. Under the fictitious assumption that the actors are guided only by value-rational motives and that they have a full awareness of the legitimacy content of the order, regularities of action arise which are stabilized by the binding forces of values. Thus, a principally different coordination mechanism of actions is given than in the case of interests. In the latter, actions are based on motives of utility or success: the purpose, as the idea of success, is the cause of the action (CMW, p. 120f.; WL, p. 183). In the legitimate order, actions are based on motives of obligation: value, as the idea of an obligation, is the cause of an action. Here, action is taken for the sake of value, regardless of success. In both cases, however, the interactions and thus the order are stabilized by the actors’ well-founded expectation, i.e., the general ideal type is conceived in such a way that the order is not merely made of the fragile as-if expectations of the interactants, but by justified expectations (Hopf 1991, p. 132). The network of relations and patterns of processes that constitute a social order is held together and reproduced by the well-founded expectations of the actors. Therefore, the fact that the objectively existing opportunities are subjectively conscious and taken into account also applies to the general concept of legitimate order. In the case of interests and legitimate order, however, something different is expected in each case. In the case of coordination by interests, the subjective calculation of chances consists in the justified expectation that the others also seek to realize their interests and purposes and that ego must take this into account for its own. In the case of non-observance of the interests of the others, there is no sanctioning instance, as in the case of normative breach, but the “punishment” occurs implicitly, similar to technical rules: ego’s own interest suffers damage. The stability of interests is based on the fact that the interests of others must be taken into account as conditions for the realization

332

T. Schwinn

of one’s own interests (CMW, p. 289; WL, p. 453; ES, p. 30f.; WG, p. 15f.). In the coordination mechanism of legitimacy, the subjective calculation of chances consists in the justified expectation that the order is also binding for the other(s). “It should at once be emphasized that there is more to the empirical ‘validity’ of an [instituted] order than the fact that the ‘expectations’ of the societized persons in relation to their actual behavior are, on average, well founded. This is only the most rational sense [of the term], and at the same time – sociologically speaking – the most directly comprehensible one. But a situation where the behavior of each and every participant would exclusively be oriented only according to ‘expectations’ concerning the behavior of others would constitute the absolute limiting case in relation to ‘communal action’ and would imply the absolute instability of those expectations in themselves. On the contrary: the average probability that those expectations are ‘wellfounded’ will be the greater, the more one may, on average, count on the action of the other participants being not only oriented according to the[ir] expectations concerning the actions of others, and the more widespread (to a relevant degree) the subjective view among the[se other participants] that the ‘legality’ (comprehended in terms of its subjective meaning-relatedness) of the [instituted] order is ‘binding’ upon them” (CMW, p. 284f.; WL, p. 446; see CMW, p. 292f.; WL, p. 459). The coordination mechanism of “legitimacy” explained by Weber here on the basis of the legal order results in a plus of stability compared to a pure coordination of interests. In the latter, the factual action of the other(s) is the condition of the possibility to realize my own interest. I only produce something if others will buy it. In contrast to this factual coordination, the normative one is counterfactually stabilized. Interest-based actions on an economic market are factually coordinated because every change of a competitor has to be answered by a change or adaption of the other participants, for example a price challenge: If a competitor deviates from the normal price of a product you will be kicked out of the market if you hold on the old price. Value-based action in the context of a legitimate order is counterfactually stabilized, because the deviation from a norm or their violation is not answered by giving up the norm. The norm is defended against the deviator (Luhmann 1987, p. 43f.). Here, the action of the other is not a condition for me to act as well, but the reason for my action per se is the obligatory nature of a norm or what is obligatory, what ought to be valid in an order. Insofar as “the subjective view that an order is binding is widespread to a relevant degree,” ego can rely on it in interactions and well-founded expect the other to behave in accordance with the order. If ego can no longer well-founded expect the general diffusion of the obligatory character of the order, the value

Max Weber’s Methodological Individualism

333

content can still be something that ought to be valid; but in interactions, ego will no longer rely on this being true for the others as well. Indeed, legitimacy still applies to ego, but it will hardly be able to function as a coordination mechanism. Thus, it is of central importance in the construction of the general ideal types to grasp the different action orientations and types of order in a conceptual purity and to assign them to each other in an exclusive way under the fictitious assumptions. Compared to the concrete historical reality they are relatively empty of content and far from reality, but they offer an optimum of unambiguity. The result is a tableau of theoretical constructions, each with a principally different conception of the relationship between action and order, between micro- and macro-level.

4

Models of Order and Empirical Reality

What role do such general ideal types play in the context of a concrete empirical investigation? In order to clarify this, some general remarks on the function of the general and individual ideal types are necessary. First, the general ideal types fulfill a classificatory or constitutive task in delimiting the object under investigation. Weber illustrates this with the example of skat, a card game. “The intellectual content of the ‘norm’ [the rules of skat, T.S.] is therefore the criterion according to which what is ‘conceptually important’ is selected from that multiplicity of cigar smoke, beer drinking, banging on the table and comments of every kind that furnishes the customary context of a good old German game of skat, as well as from the incidental ‘milieu’ of the concrete game. We ‘classify’ a complex of phenomena as ‘skat’ when it contains phenomena that are deemed to be relevant for the application of the norm. […] What defines the object of the investigation is [its] relevance from the point of view of the norm. […] The sole – [but] important – function of the norm [in this respect] is that its content constitutes the point of reference from which we [can] identify those facts and processes whose causal explanation a hypothetical ‘historical interest’ would focus on. That is to say: [those facts and processes] constitute those points within the given multiplicity from which the causal analysis goes backwards or forwards” (CMW, p. 214; WL, p. 340f.). Analogously, the theoretical construction of a legitimate order, for example, serves to read out the essentials relating to the concept of the state from the multiplicity of diffuse and discrete actions and tolerations, partly of a unique, partly of a regularly recurring character, as well as from the ideas

334

T. Schwinn

and thoughts in the heads of concrete people, which are partly developed intellectually, partly felt darkly, partly accepted passively, and shaded in the most diverse ways (CMW, p. 130; WL, p. 200f.). This means, that we classify a complex of actions and ideas as ‘state’ if such processes, which are considered relevant for the application of the general ideal type, can be found in it. The next function of the general concepts is a heuristic one in the causal cognition (CMW, p. 215, 131; WL, p. 342, 201). The possibility of classifying a complex of actions and ideas as a state presupposes that the content of the general ideal type, in however modified or attenuated form, is or was a causal determinant in the actions of the concrete actors, just as knowledge of and adherence to the formal rule of skat is one of the empirical maxims of skat players, thus causally influencing their actions. The observer uses the general ideal type as a heuristic means to fathom the true intentions. Thereby, the factual course of action is compared with the ideal typically expected course of action in order to facilitate the knowledge of the real motives and causes by reference to the distance between both. The teleologically rationally constructed schemes of action make it possible to recognize the teleologically non-rational elements of practical action and thus to make the latter understandable in its actual course (CMW, p. 84f.; WL, p. 130f.). The more complex and multifaceted the phenomenons are, the representation and causal attribution of which are about, the less it is possible to get by with only one such concept, and the more the general ideal type merges into the individual ideal type (CMW, p. 129, 131; WL, p. 198, 201f.). As the complexity of the object of investigation increases, circumstances and motives that are not contained in the presuppositions of a single theoretical construction take effect (CMW, p. 332; WL, p. 537). The individual ideal type is therefore composed of various conceptual elements (CMW, p. 131; WL, p. 201), that is, of various general ideal types. “For example, the same historical phenomenon may be in one aspect feudal, in another patrimonial, in another bureaucratic, and in still another charismatic. In order to give a precise meaning to these terms, it is necessary for the sociologist to formulate pure ideal types of the corresponding forms of action which in each case involve the highest possible degree of logical integration by virtue of their complete adequacy on the level of meaning. But precisely because this is true, it is probably seldom if ever that a real phenomenon can be found which corresponds exactly to one of these ideally constructed pure types. The case is similar to a physical reaction which has been calculated on the assumption of an absolute vacuum. Theoretical differentiation (Kasuistik) is possible in sociology only in terms of ideal or pure types” (ES, p. 20; WG, p. 10).

Max Weber’s Methodological Individualism

335

What does this mean for the problem of action and order that is of interest here? With the transition from the general to the individual ideal type in the context of an empirical investigation for the purpose of causal attribution, the exclusive assignment of types of action and types of order is not possible. The individual ideal type of an order is composed of several types of action. Thus, for example, a legitimate order of a particular country is characterized by several action orientations: value-rational, purposively rational, affectual, and traditional. In the empirical order process, a multiplicity of orientations and action regularities, heterogeneous with regard to the general type of order, is constantly reproduced. The individual ideal type must take this into account, the assignment of types of action to types of order can vary here but not arbitrarily. In order to be able to speak of certain regularities of action as of a specific type of order, it is necessary that among the multitude of actions oriented toward this order, a certain core set of action orientations constitutive for this type of order is found. It is incompatible with the logic and consistency of Weber’s methodological individualism to assign an identical type of order to arbitrarily heterogeneous types of orientation. Externally identical rule processes can have a completely different meaning for the actors and thus have to be assigned to different types. Types of order and types of action can stand in certain, but not arbitrary mixing ratios to each other. A definitional condition for a type of order is therefore the possibility that for a certain core set of actions there is an exclusive relationship between a type of actions and a type of order. In other words: In order to be able to speak of the same type of order in the duality of order as a medium and a result of actions at time t2 , there must have been a certain number of type-constituting action orientations. Orders are never static, they exist only as a process. In a biological organism, the DNA instructs the billions of cells to always have the same reproduction cycles and thus keeps the structure of the organism constant. If copying errors occur to a certain extent, we speak of disease. This term makes no sense for social phenomena because here actions, which are not in accordance with the order, are constantly taking place. The relationship of the actor to the orders cannot be understood according to the biological model of cells and organisms. Social phenomena have no DNA in the sense of firmly fixed information on how to act. Only certain social rules are written down in legal texts. But even here, most people know the content of the law only superficially or not at all and some transgress it, in all degrees of severity. In the case of phenomena that are not formally framed, this possibility of control is completely absent. For the social scientist, this creates the problem of having to determine the identity and change of a social order: At

336

T. Schwinn

what point does a marriage enter a crisis, when is it a failure? What degree of nonconformist action, for example, by social movements, can a legitimate state order tolerate before it collapses? What degree is unproblematic or even conducive to stability? When do economic markets lose their function through monopolization processes? Weber addresses this issue by means of his consideration of a sliding transition of the existence of one order into another (CMW, p. 283f.; 286f.; WL, p. 444f., 448f.). The empirical existence of an order can be the resultant of entirely different motives, purposes, and orientations (CMW, p. 293; WL, p. 460). However, this heterogeneity of action orientations has certain limits, that have to be empirically determined in given cases, beyond which they affect the maintenance of an order. Weber explains this heterogeneity using the example of the “state.” “When we ask what, in empirical reality, corresponds to the idea … of the ‘state’, we find an infinite number of diffuse and discrete human actions … This idea is found in manifold nuances in the mind of each individual, partly intellectually well thought-out, partly vaguely felt, (and) partly passively accepted. … The concrete (historical) substance of the ‘state’, (as it is present) in those syntheses formed by human beings living in history, can in its turn only be expressed in terms of ideal-typical concepts” (CMW, p. 130; WL, p. 200f.). The scientific concept constitutes the features that allow to grasp a particular order from a complex of heterogeneous actions. For example, a core set of value-rational action orientations must be present in a multitude of actions in order to form the ideal type of a legitimate order. Consequently, it is only possible to determine when one type of order ceases to exist or when it merges into another through the criteria provided by the general ideal types. That’s the reason why Weber emphasizes that sociological casuistry is possible only in terms of pure types. In the face of the heterogeneity of action orientations and different motives, the chance for the reproduction of an order will be improved, if elites interpret the legitimacy content and actively carry it out via organs. Their intention is above all to secure the objective chance of an order being reproduced by specifying expectations. For this purpose, in addition to the internal guarantees, external guarantees in terms of legislation and a corresponding coercive staff are usually necessary (CMW, p. 31f., 212ff., 946f.; WG, p. 16, 123, 544f.). These guarantee the order by “the expectation of specific external effects, that is, by interest” (ES, p. 33; WG, p. 17). For a large mass of actors, action is based on mere docility to what is familiar, acquired, without explicit value orientation. This “audience” knows the order of contents and the chances resulting from it just as far as they are necessary for the attainment of advantages or the avoidance of the most drastic

Max Weber’s Methodological Individualism

337

inconveniences (CMW, p. 300; WL, p. 472). This is, of course, an empirical question. Particularly in times of change, the indifferent attitude gives way to a more conscious one that addresses questions of legitimacy, especially when doubts about legitimacy spread among the elites of the order itself. If they no longer attempt to specify and secure the chance structure in the sense of a reproduction according to the order, the chance structure shifts, and the order dissolves, like in many socialist countries after 1989. Since orders always have to fulfill interpretive functions as well as supply functions, the chances of legitimation also depend on the structure of social inequality. Depending on these empirical mixing ratios of action orientations, the objective chance of the existence and duration of an order varies.

5

Combinations of Orders and Orientations

For a social theory that follows the premises of methodological individualism, certain types of action orientations must correspond to a certain type of order. As already pointed out, actors do not orient themselves accordingly to orders either. Many people do not observe norms or even consciously act against them. This raises the question of how many improper actions an order can stand in order to remain constant, and at what point it dissolves or merges into another. To do this, one must distinguish the types of action coordination or order from the types of action orientation and then indicate the possible combinations. It is possible to list the conceivably possible combinations and transitions thematized by Weber (Bader 1989, p. 310) (Table 1): Here the action-order combinations will be displayed for legitimate orders, which are central for Weber’s sociology of domination. Weber distinguishes four types of action and three types of domination. To explain the connection Table 1

Combination of Orders and Orientations

Types of Action Orientation

Types of Coordination/Order Custom/ Manner

Affective Communalization

Coordination by Interests

Legitimate Order

Traditional Affectual Purposive rational Value rational

+ – –

– + –

– – +

– – –

–

–

–

+

+type constituting action orientation –not type constituting action orientation

338

T. Schwinn

between the typology of action orientations and the typology of legitimate orders, one must not directly assign the respective identically worded types. That’s why traditional action is not basic for traditional domination. Weber makes a difference between power and domination (ES, p. 53; WG, p. 16). In the first one people can be forced to do what a dictator wants to be done, even despite resistance. In contrast domination works only when people voluntarily obey the rules. This kind of orientation is anchored in values, which are considered as desirable. Therefore, the constituting type of action that is the basis of all types of legitimate order is the value-rational one. Legitimation needs the existence of maxims that are regarded as binding or exemplary. The vocabulary of being valid and ought in distinction from benefit-calculating motives goes back to Kant’s differentiation of the categorical and the hypothetical imperative (CMW, p. 223; WL, p. 355; Schluchter 1988, I, p. 206ff.; 2006, p. 272).12 This Kantian distinction transforms Weber into corresponding conceptions for historical and sociological research: purposively rational and value-rational action, interests, and legitimate order. The normative character of all three types of legitimate order can be constituted and guaranteed by only one type of action: the value-rational action.13 The traditional legitimacy type is therefore not constituted by the traditional action type, but by the value-rational belief in tradition, “valid is that which has always been” (ES, p. 36; WG, p. 19), which is something different (Bader 1989, p. 313).14 In the more or less vaguely proceeding traditional action, the actors have no conception at all of something that ought to be valid. This is only possible when the attachment to the habitual can be consciously maintained. Nor is the core of the legal or rational type of domination formed by purposively rational action. The latter has nothing to do with normative validity. What exactly is the value-rational character of this type of order? It is not enough here to see it merely, as Bader (1989, p. 324) does, in the “belief in specific formal procedures as a value in itself.” Such legitimation through procedures does not go far enough. Bader seems to notice this, too, and he

12

Imperatives according to Kant set basic premises for the orientation and action of subjects. Categorical Imperatives we respect and obey for themselves. We feel the duty to act in accordance with a value or a norm. In contrast, hypothetical imperatives are rules of cleverness and skillfulness. We obey them not for themselves, but to be successful, to reach a purpose. 13 Weber is not always consistent in this, e.g. ES, p. 212; WG, p. 122. 14 Weber also speaks of “tradition formation”: “It is by way of conventional rules that merely factual regularities of action, i.e., usages, are frequently transformed into binding norms, guaranteed primarily by psychological coercion. Convention thus makes tradition. The mere fact of the regular recurrence of certain events somehow confers on them the dignity of oughtness” (ES, p. 326; WG, p. 191).

Max Weber’s Methodological Individualism

339

recommends questioning the usual opposition of procedures and substantial values. Formal procedures cannot exist without values that justify them and are institutionalized in them. Weber’s remarks on human and civil rights indicate that they represent a certain material-ethical value minimum as a condition of validity of the formal procedures (Weber 1988, p. 42f., 62f.; 1994, p. 45f., 68ff.; Schluchter 1979, p. 154f.). The value-rational character of the legal type of domination, what ought to be valid, therefore consists of a specific combination of value-contents and formal procedures. Charismatic legitimacy cannot be constituted either, nor guaranteed by purely affectual orientations (Döbert 1989, p. 239; Bader 1989, p. 314).15 Therefore, it is important to distinguish it from affectual communal relationship. How can it be justified that here, too, the value-rational is the decisive type of action? For this, it is necessary to recall the conditions under which the charisma arose and the course it took. Charismatic relevant situations are situations of inner or outer distress, in which traditional forms and interpretations of life fail. The collectively perceived discrepancy becomes affectively charged. People jump into this gap and reinterpret the situation of existential concern. The decisive factor is therefore not purely affectual action, but interpreted affects, i.e., the fluid affective potential is bound by new interpretative patterns and engaged. This is done through the principle of mission. Legitimate validity is ascribed to the charismatic leader by the follower not only because of his person as such, but also on the basis of faith in the validity of his mission (ES, p. 36, 215; WG, p. 19, 124). This value-rational component sets internal limits and restrictions to any charismatic domination. The obedience and allegiance demanded by the leader by virtue of his mission must be proven by success (ES, p. 1112f. WG, p. 655). Similar to traditional domination, charismatic domination also involves an inner tension. “While on the one hand there is the power to rule by virtue of a personal gift of grace, which is not inhibited by any tradition or statute, on the other hand there is the inescapable compulsion to proof the ruler in the constitutive idea of mission, which carries the ‘vocation’. From here, the very most significant legitimacy check against arbitrary exercise of power can take place and lead to the abrupt fall of the charismatic ruler after blatant failure of his mission” (Winckelmann 1966, p. 239). In terms of action theory, charismatic domination is not based on affectual “uncontrolled reaction” but affective action “is … usually well on the road to rationalization” (ES, p. 25; WG, p. 12). 15

Cf. WL, p. 475, where Weber explicitly states that legitimacy cannot be established by a purely affectual orientation. At this point, Weber confirms the thesis adopted here that only the valuerational type of action can constitute a legitimate order. Contradictory to this: ES, p. 212; WG, p. 122.

340

T. Schwinn

From the latter alone can be deduced what is characteristic ought to be valid for every legitimate order. Now that the type-constituting value-rational orientation for the three legitimate order types has been stated, the combinatorial possibilities can be expounded. Traditional action cannot constitute and guarantee the legitimacy of an order, but to a considerable extent it can stabilize its empirical existence (Bader 1989, p. 313f.; ES, p. 31f., 212ff.; WG, p. 16, 122; WL, p. 475). According to Weber, the traditional and strategic orientations are the most common type of inner attitude in everyday life. The actions of the mass of the population usually proceed without any knowledge of purpose and meaning, even of the existence of a legitimate order (CMW, p. 300f.; WL, p. 473).16 The motiveless docility into the familiar, familiarized, acquired, always repeating is an essential stabilizer of legitimate orders, whose meaning does not have to constantly take the “detour” via a conscious meaningrelatedness in order to become relevant for action. To the extent that this traditional orientation becomes dominant, no more instances or groups maintain the claim to legitimacy, and an order is only habitually anchored, legitimacy turns into custom. Affective orientations cannot constitute or guarantee legitimacy either, but they can very well stabilize it. As a type of legitimacy, charismatic domination belongs to a different category of order than affective communal relationship. The latter is not a relationship of domination and therefore does not necessarily require value-rational moments. As a probationary relationship, charismatic domination cannot exist without these moments. Indeed, any affective communal relationship also normally has some value reference, but the constitutive and defining moment is based on the affects as such. The integrative moment of this relationship lies precisely in a certain diffuseness of the emotional states, which is lost when it is overloaded with meaning contents by which action is to be measured. An actor adopts a purposive rational orientation to a legitimate order if he, out of benefit-calculating motives, takes its existence into account for his actions (ES, p. 31f.; WG, p. 16; CMW, p. 300; WL, p. 472). It is likely that in many cases there is an attitude of sanction avoidance. One does not violate the order, but also does not consider it as something that ought to be valid. Moments of coercion and fear have to be considered in the subjective as well 16

“The broad mass of the participants act in a way corresponding to legal norms, not out of obedience regarded as a legal obligation, but either because the environment approves of the conduct and disapproves of its opposite, or merely as a result of unreflective habituation to a regularity of life that has engraved itself as a custom. If the latter attitude were universal, the law would no longer ‘subjectively’ be regarded as such, but would be observed as a custom” (ES, p. 312; WG, p. 182).

Max Weber’s Methodological Individualism

341

as objective calculation of chances. A thief orients himself to the validity of order by concealing his violation. Between both cases there are fluid transitions of more or less “partial deviations” (ES, p. 32; WG, p. 16). Order-related action is to be distinguished from these cases (CMW, p. 285; WL, p. 447). Here, one also orients oneself purposively rationally to an existing order, but not necessarily from utility motives, but also from normative and valuerational motives. A distinction must be made here between actions that are directed toward a planned change of a legitimate order within a legal framework and those that also accept illegal means for changes. To the extent that the purposively rational orientation toward an order that merely claims legitimacy becomes dominant, it changes into a coordination by interests or dissolves completely.

6

Final Remarks

In the methodological debates, the sociological schools approach the micro– macro problem from the premises of their own program. This often leads to an overgeneralization of single variables and analytical dimensions. “Indeed, it is this very loyalty to initial starting points, … which limits the success of most of these linkage proposals in fundamental ways” (Alexander and Giesen 1987, p. 37). It is not to be expected, as Parsons (1968) still assumed, that theoretical efforts converge in a single model. Nevertheless, one may wonder whether more can be achieved than that each theory develops its own micromacro solution and that one is now confronted with as many connections as there are number of schools. In this situation, Weber’s methodology is an interesting conceptual offer that can help to mediate the paradigms working on a common problem from different directions and with different premises. Implicitly, it already accomplishes this task. Unlike many classics and more recent theorists, many approaches refer to him. Although Weber unmistakably pursues a methodological individualism, the conceptual genealogy in the Basic Sociological Terms, action—social action—social relation—order, must not be misunderstood in the way that the actors spin the relations and orders out of themselves without any context. A basic premise of Weber’s thinking is the historical embeddedness of action. No action begins in a social zero or isolated state, but is always already interwoven with macro presuppositions. Therefore, one can also rearrange the order in the Basic Sociological Terms and start with the paragraphs on social structures and orders. Actions take place within the framework of certain structural presuppositions and reproduce them or lead to modified or new ones. The concept of chance

342

T. Schwinn

and the ideal types offer analytical instruments that provide insights into the internal mode of reproduction of social orders. In which way order processes are present to the actors; when do they vanish from the intentions; where do the objective chance calculations of the observer no longer correspond to the subjective ones of the participants? Which mixed ratios of action orientations reproduce an order? But it is not only the fact that he avoids approaching the micro-macro problem unilaterally from the level of action or order that makes Weber an interesting interlocutor in current discussions, but also the fact that a differentiated understanding of action and order can be found in his work. Certain analytical aspects, which are overgeneralized in some schools as a representative model, rational-benefit-calculating or symbolic-interpreting actor, penetrative or restrictive conceptions of order (Münch and Smelser 1987, p. 376ff.), are present in Weber’s work as variants. These types of action and order allow to sort the respective aspects of the empirical circumstances and to combine them in corresponding individual ideal types. Thus, the micromacro problem cannot be solved in a theoretical model in a general way, but rather, depending on the concrete mixing ratios between types of action and order, the linking of these levels takes different forms. In this regard, Weber’s conceptual offer is still richer than much of what is offered on today’s theory market.

References Albert, G. (2009). Weber-Paradigma. In G. Kneer & M. Schroer (Eds.), Handbuch Soziologische Theorien (pp. 517–554). Wiesbaden: VS Verlag. Alexander, J. C. (1987). Action and Its Environments. In J. C. Alexander, B. Giesen, R. Münch & N. J. Smelser (Eds.), The Micro-Macro Link (pp. 289–318). Berkeley, Los Angeles, London: University of California Press. Alexander, J. C., & Giesen, B. (1987). From Reduction to Linkage: The Long View of the Micro-Macro Link. In J. C. Alexander, B. Giesen, R. Münch & N. J. Smelser (Eds.), The Micro-Macro Link (pp. 1–42). Berkeley, Los Angeles, London: University of California Press. Alexander, J. C., Giesen, B., Münch, R., & Smelser, N. J. (Eds.). (1987). The MicroMacro Link. Berkeley, Los Angeles, London: University of California Press. Archer, M. S. (1995). Realist Social Theory: The Morphogenetic Approach. Cambridge: Cambridge University Press. Bader, V.-M. (1989). Max Webers Begriff der Legitimität. Versuch einer systematisch-kritischen Rekonstruktion. In J. Weiß (Ed.), Max Weber heute. Erträge und Probleme der Forschung (pp. 296–334). Frankfurt a.M.: Suhrkamp.

Max Weber’s Methodological Individualism

343

Blau, P. M. (1987). Contrasting Theoretical Perspectives. In J. C. Alexander, B. Giesen, R. Münch & N. J. Smelser (Eds.), The Micro-Macro Link (pp. 71–85). Berkeley, Los Angeles, London: University of California Press. Boudon, R. (1987). The Individualistic Tradition in Sociology. In J. C. Alexander, B. Giesen, R. Münch & N. J. Smelser (Eds.), The Micro-Macro Link (pp. 45–70). Berkeley, Los Angeles, London: University of California Press. Burger, T. (1994). Deutsche Geschichtstheorie und Webersche Soziologie. In G. Wagner & H. Zipprian (Eds.), Max Webers Wissenschaftslehre. Interpretation und Kritik (pp. 29–104). Frankfurt a. M.: Suhrkamp. Coleman, J. S. (1987). Microfoundations and Macrosocial Behavior. In J. C. Alexander, B. Giesen, R. Münch & N. J. Smelser (Eds.), The Micro-Macro Link (pp. 153–173). Berkeley, Los Angeles, London: University of California Press. Colomy, P. (Ed.). (1990). Neofunctionalist Sociology. Aldershot, Brookfield: Elgar Reference Collection. Dahrendorf, R. (1979). Lebenschancen. Frankfurt a. M.: Suhrkamp. Döbert, R. (1989). Max Webers Handlungstheorie und die Ebenen des Rationalisierungskomplexes. In J. Weiß (Ed.), Max Weber heute. Erträge und Probleme der Forschung (pp. 210–249). Frankfurt a.M.: Suhrkamp. Esser, H. (1991). Alltagshandeln und Verstehen. Zum Verhältnis von erklärender und verstehender Soziologie am Beispiel von Alfred Schütz und „Rational Choice“. Tübingen: Mohr (Siebeck). Esser, H. (1993). Soziologie. Allgemeine Grundlagen. Frankfurt/New York: Campus. Gerhards, J. (1988). Soziologie der Emotionen. Weinheim, München: Juventa. Giddens, A. (1979). Central Problems in Social Theory: Action, Structure and Contradiction in Social Analysis. Berkeley und Los Angeles: University of California Press. Giddens, A. (1984). The Constitution of Society: Outline of the Theory of Structuration. Cambridge: Polity Press. Greve, J. (2008). Das Makro-Mikro-Makro-Modell: From Reduction to Linkage and Back Again. In J. Greve, A. Schnabel & R. Schützeichel (Eds.), Das MikroMakro-Modell der soziologischen Erklärung (pp. 49–78). Wiesbaden: VS Verlag. Habermas, J. (1984). The Theory of Communicative Action, Volume 1. Reason and the Rationalization of Society. Transl. by Thomas McCarthy. London: Heinemann. Hedström, P., & Ylikoski, P. (2019). Sociological Individualism. In S. Randeria & B. Wittrock (Eds.), Social Science at the Crossroads (pp. 55–66). Leiden/Boston: Brill. Hennis, W. (1988). Max Weber: Essays in Reconstruction. London: Allen & Unwin. Henrich, D. (1952). Die Einheit der Wissenschaftslehre Max Webers. Tübingen: Mohr (Siebeck). Hopf, W. (1991). Regelmäßigkeiten und Typen. Das Durchschnittshandeln in Max Webers Methodologie. Zeitschrift für Soziologie, 20, 124–137. Knorr-Cetina, K., & Cicourel, A. V. (Ed.). (1981). Advances in Social Theory and Methodology: Toward an Integration of Micro- and Macro-sociologies. Boston, London und Henley: Routledge & Kegan Paul.

344

T. Schwinn

Lichtblau, K. (2020). Zur Aktualität von Max Weber. Einführung in sein Werk. Wiesbaden: Springer. Lindenberg, S. (1977). Individuelle Effekte, kollektive Phänomene und das Problem der Transformation. In K. Eichner & W. Habermehl (Eds.), Probleme der Erklärung sozialen Handelns (pp. 46–84). Meisenheim am Glan: Anton Hain. Luhmann, N. (1987). Rechtssoziologie. 3rd ed. Opladen: Westdeutscher Verlag. Müller, H.-P. (2020). Max Weber. Eine Spurensuche. Berlin: Suhrkamp. Münch, R., & Smelser, N. J. (1987). Relating the Micro and Macro. In J. C. Alexander, B. Giesen, R. Münch & N. J. Smelser (Eds.), The Micro-Macro Link (pp. 356–387). Berkeley, Los Angeles, London: University of California Press. Parsons, T. (1968) [1937]. The Structure of Social Action. New York: The Free Press. Popper, K. R. 1966 [1945]. The Open Society and Its Enemies, Vol. 2: Hegel and Marx. London: Routledge & Keagan Paul. Schelting, A. von (1934). Max Webers Wissenschaftslehre. Tübingen: Mohr (Siebeck). Schluchter, W. (1979). Die Entwicklung des okzidentalen Rationalismus. Tübingen: Mohr (Siebeck). Schluchter, W. (1988). Religion und Lebensführung, 2 Bde. Frankfurt a.M.: Suhrkamp. Schluchter, W. (2003). Handlung, Ordnung und Kultur. Grundzüge eines weberianischen Forschungsprogramms. In G. Albert, A. Bienfait, S. Sigmund, & C. Wendt (Eds.), Das Weber-Paradigma. Studien zur Weiterentwicklung von Max Webers Forschungsprogramm (pp. 42–74). Tübingen: Mohr Siebeck. Schluchter, W. (2006/2007). Grundlegungen der Soziologie. 2 Bände. Tübingen: Mohr Siebeck. Schluchter, W. (2018). Handeln im Kontext. Neue Studien zu einem Forschungsprogramm im Anschluß an Max Weber. Tübingen: Mohr Siebeck. Schütz, A., & Parsons, T. (1977). Zur Theorie sozialen Handelns. Ein Briefwechsel . Frankfurt a.M.: Suhrkamp. Schwinn, T. (1993). Jenseits von Subjektivismus und Objektivismus. Max Weber, Alfred Schütz und Talcott Parsons. Duncker & Humblot, Berlin. Schwinn, T. (2013). Max Weber und die Systemtheorie. Studien zu einer handlungstheoretischen Makrosoziologie. Tübingen: Mohr Siebeck. Tenbruck, F. H. (1959). Die Genesis der Methodologie Max Webers. Kölner Zeitschrift für Soziologie und Sozialpsychologie, 11, 573–630. Tenbruck, F. H. (1989). Abschied von der Wissenschaftslehre. In J. Weiß (Ed.), Max Weber heute. Erträge und Probleme der Forschung (pp. 90–115). Frankfurt a. M.: Suhrkamp. Udehn, L. (2001). Methodological Individualism: Background, History and Meaning. London and New York: Routledge. Udehn, L. (2002). The Changing Face of Methodological Individualism. Annual Review of Sociology, 28, 479–507. Weber, M. (1988) [1926]. Max Weber: A Biography. New Brunswick: Transaction Books.

Max Weber’s Methodological Individualism

345

Weber, M. (1978a) (ES). Economy and Society: An Outline of Interpretive Sociology. 2 Vol., edited by G. Roth and C. Wittich. Berkeley: University of California Press. Weber, M. (1978b). Gesammelte Aufsätze zur Religionssoziologie I, 7. Aufl. Tübingen: Mohr (Siebeck). Weber, M. (1980) (WG). Wirtschaft und Gesellschaft, 5. Aufl. Tübingen: Mohr (Siebeck). Weber, M. (1982) (WL). Gesammelte Aufsätze zur Wissenschaftslehre, 5. Aufl. Tübingen: Mohr (Siebeck). Weber, M. (1988). Gesammelte Politische Schriften, 5. Aufl. Tübingen: Mohr (Siebeck). Weber, M. (1994). Political Writings, edited by P. Lassman & R. Speirs. Cambridge: Cambridge University Press. Weber, M. (2012) (CMW). Collected Methodological Writings, edited by H. H. Bruun & S. Whimster. London and New York: Routledge. Weiß, J. (2018). Einleitung. In Max Weber Gesamtausgabe. MWG1/12: Verstehende Soziologie und Werturteilsfreiheit (pp. 1–92). Tübingen: Mohr Siebeck. Winckelmann, J. (1966). Max Webers Verständnis von Mensch und Gesellschaft. In K. Engisch, B. Pfister & J. Winckelmann (Eds.), Max Weber (pp. 195–243). Berlin: Duncker & Humblot. Winckelmann, J. (1976). Erläuterungsband zu ‘Wirtschaft und Gesellschaft’ . Tübingen: Mohr (Siebeck). Wippler, R., & Lindenberg, S. (1987). Collective Phenomena and Rational Choice. In J. C. Alexander, B. Giesen, R. Münch, & N. J. Smelser (Eds.), The MicroMacro Link (pp. 135–152). Berkeley, Los Angeles, London: University of California Press.

Micro–Macro Transitions and the Paradox of Unintended Consequences Mohamed Cherkaoui

Like other sciences, sociology investigates macrophenomena as well as microphenomena. These two classes of theoretical constructs are considered to conform two scales of organization to which two types of explanation are associated, the basic structures of which are for the moment neither entirely homogeneous nor perfectly unifiable thanks to a particular point of view, although methodological individualism ideally aims at making macrophenomena intelligible through individual actions. Recognizing the existence of levels of reality should not, however, lead us to believe in the existence of ontologically distinct entities. I will leave aside what some people understand by the “mesological” level, which is identified either with the analysis of local contexts from which individuals act or with that of networks for which the observation unit is not the individual but the relationships and which would combine both the individual and the institutional. The term “mesology” was coined by the demographer Adolphe Bertillon (1897), who gave it the meaning of the study of the environment. Auguste Comte borrowed the concept from Henri de Blainville and used it as the key-word in the argument of the forty-third lesson of his Cours de philosophie positive, where he sketched out “a theory of M. Cherkaoui (B) Sociology, National Center for Scientific Research, Sorbonne University, Paris, France e-mail: [email protected]

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 N. Bulle and F. Di Iorio (eds.), The Palgrave Handbook of Methodological Individualism, https://doi.org/10.1007/978-3-031-41512-8_17

347

348

M. Cherkaoui

environments.” It will have a considerable and diffuse posterity going from botany or zoology to sociology; present therefore in all the researches which estimate that the environment is an imperative entry to any intelligibility of reality. The purpose of this contribution is twofold. Firstly, I would like to present and clarify the debate on the links between the two levels, the transition from one to the other, and the problems posed by the foundation of macrology on micrology or, if one prefers the traditional terminology, of the social on the individual. Methodological individualism aims to propose solutions. Secondly, I would like to analyze the transitions between classes of phenomena and highlight the unexpected consequences of individual actions or collective decisions. Four cases result from the combination of the two levels: we can describe and account for the individual through the individual, the collective through the individual, the collective through the collective, the individual through the collective. In each of these four figures, the problem of expected and unexpected consequences arises. In this presentation, I will limit myself to the two levels of sociological phenomena, while recalling, along the way, the identity or differences between the problems of economics and sociology.

1

Micro–Macro: Definitions, Tensions, and Foundations of the Two Orders

Though the problems raised by the relations between macro- and microsociology are highly relevant today, few of them are new. They first arose with the institutionalization of sociology as an autonomous scientific discipline positioned at the center of the debates opposing utilitarianism-inspired theorists like Spencer to sociologists who, in search of a foundation for the specificity of their object, method of analysis, and explanations of social phenomena, rejected the assumptions of classical economics. This controversy was not limited to the Spencer–Durkheim opposition but was present in all debates on method (Cherkaoui 2005).

Research Programs and Intelligibility of Macro-Sociological Phenomena: The Place of Methodological Individualism Far from being rhetorical, these questions of the relationships between levels in fact determine researchers’ choice of units of observation, methodological

Micro–Macro Transitions and the Paradox …

349

instruments, and explanatory procedures. A hypothesis tested at the level of the individual may turn out to be false at the systemic level; another, verified at the global or contextual level, may prove invalid for the individual level. The error of inferring the social from the individual corresponds to that of inferring the individual from the social (“contextual fallacy”): though it is sometimes a mistake to deduce (without any supplementary hypothesis) the voting behavior of individuals from information on how ecological units such as precincts voted, it is just as wrongheaded to affirm, on the basis of a relation of positive dependence between an individual’s degree of education and his or her salary, that when the collective level of education rises, the average salary does too. Let us assume, following the founding fathers, that the major objective of social sciences consists in explaining the genesis, structure, and dynamics of macro phenomena. Tocqueville (1856) sought to account for political change, mainly revolutionary phenomena; Durkheim (1893) for the increasing differentiation of functions and its consequences; Weber (1905) for the birth of a new type of socio-economic organization, capitalism; Sombart (1906) for the absence of socialism in the United States of America; and Michels (1911) for the inevitable formation of the oligarchic structure within democratic organizations. All these issues pertain to the systemic level. The size of the system may vary, but whatever its components and their number, it constitutes a unit that must not be dissolved and should be studied sui generis. We have three competing approaches for conducting an explanation of macro-sociological facts. The first and most current is also the least satisfactory. It consists in analyzing the relations between the social phenomenon under consideration, explicandum, and independent social factors or explicans. Explaining the social by means of the social is the leitmotiv of the Durkheimian sociological program. But in order for that first macrological procedure to be legitimate, one’s observations must pertain to the same level. This condition would seem to be met by the previously mentioned studies, though the observations reached are not explained in terms of specific indicators. Paradoxically, it can be said that that revolution is the result of improvement in economic, political, and social conditions; that an increase in the volume and density of work leads to a division of labor; that religious ethics facilitates the birth of capitalism; that a high degree of social mobility makes the development of strong leftwing parties less likely. It is worth noting in passing that these consequences are all unexpected. It seems, however, that this condition of consistency between the content of a sociological proposition and the unit of observation and analysis is not always satisfied. Indeed, in most macrosociological studies conducted by

350

M. Cherkaoui

interview or questionnaire, the unit remains the individual, and a hiatus is created between the objective of the theory, which is to formulate macrosociological propositions, and the empirical analysis, which remains focused on explaining individual behavior. Without consistency we are caught in an impasse: many hypotheses deduced from macrosociological theories cannot be adequately tested. The second procedure consists in collecting observations at an infrasystemic level and developing hypotheses on the behavior of such units (individuals, groups, institutions), with the purpose of explaining systemic relations through an appropriate synthesis of these observations. In this case micrological variables or so-called transition mechanisms positioned in an intermediary position between dependent and independent macroscopic factors are introduced. The above-mentioned sociologists were perfectly aware of this, even if they did not formalize it. It is argued that improvement in conditions (macro) generates an inflationary spiral of expectations that cannot be met, thus a rate of frustration (micro), such that it increases the probability of a contestation of the socio-political order and collective action that leads to revolution (macro); that an increase in volume linked to a growing densification of the population aggravates competition between system units confronting each other on the same market, and that this competition is resolved by division of labor (but also by suicide, and the migration of people and capital); and that social mobility, whether real or perceived as real by individuals, reduces the search for a collective solution by making people believe in the utility of personal effort, thus making the emergence and institutionalization of strong socialist parties less likely. It is often argued that an explanation operating at this infrasystemic level is sometimes considered more satisfying, stable, and general than one expressed entirely in macroscopic terms. This is Popper’s point of view (1934, 1963), as well as Harsany’s (1968), Simon’s (1977), Boudon’s (1973, 1977), and his school (e.g. Bulle 1999; Cherkaoui 1982, 2018; Manzo 2009). It should be noted, however, that this modus procedendi of methodological individualism can be traced back at least to Menger (1883, 1892) and Weber (1921), to whom Schumpeter (1908, 1954) acknowledges his intellectual debt, although some have mistakenly believed that the author of the History of Economic Analysis was its father. Since, it is argued, the behaviors and dynamics of a social system are in fact the result of the actions of its components (individuals, sub-groups, institutions), it is appropriate to analyze these in order to better account for and predict systemic characteristics. Without getting involved in a detailed discussion of this crucial problem, I would like to point out, however, that sometimes such a principle is not adhered to. Some

Micro–Macro Transitions and the Paradox …

351

macrosociological or macroeconomic predictions seem, if not better, at least more effective and practical than those based on micrological indicators. Macroeconomic propositions are the real basis for many decisions in political economics. The economic management of major industrial countries like the United States or European Union by their central banks is largely based on analysis of developments in a few simple monetary indicators (e.g., the price index, the inflation rate, the unemployment rate, the price of a barrel of oil, the money supply M1, M2, or M3). I also add that public policy (on demographics, the city, or the family, for instance), like predictions about social behavior, is based on macrosociological propositions. The third approach to conducting an explanation is to analyze effects due to the nature of the positions and distributions of certain variables, on the behavior of the system’s component units without formulating hypotheses about individuals. Here we are no longer interested, for example, in the influence of religion (or any other variable) on social relations, but rather in the distribution of religious affiliation and the structural parameters of that distribution, understood to determine the dependent variables. This solution is advocated by some contemporary macro-sociologists, but its origin lies in the work of Simmel (1908) on the quantitative determinations of groups and the crossing of social circles. Whereas microsociology seeks to account for social relations in terms of psychological traits or processes such as exchange, communication, competition, and cooperation, which shape individual relations, structural macrosociology, that of Peter Blau (1977; Blau and Schwartz 1984) in particular, accounts for the relations between the different constituent parts of a given society in terms of differentiation and integration. For the latter approach, explaining certain aspects of relations between religious or ethnic groups, such as marriage or friendship, is to deduce them logically from the structural properties of these groups (notably their respective size) without having recourse to additional assumptions about individuals, such as their degree of liberalism. Coleman (1964) had sketched a first formalization of a Simmelian property; Fararo (1981), Fararo and Skvoretz (1984, 1986), and Skvoretz (1983) developed it. Let us take a simple example. Suppose we have identified a correlation between religion and marriage. To account for mixed marriage, macrostructuralists do not have recourse to religious orientations or to the degree of tolerance of religious groups. Let us consider two groups, Protestants and Catholics, of different sizes A and B (A > B), whose individuals can come into contact with each other and whose members are likely to enter into mixed marriages. The first theorem deduced from these assumptions is that

352

M. Cherkaoui

the rate of intergroup association of group B must exceed that of group A. If A and B consist of 10,000 and 100 individuals respectively and if there are 10 intermarriages (10 pairs of spouses), the intermarriage rates for A and for B are equal to, respectively, 10/10,000 = 1/1000 and 10/100 = 1/10. The previous theorem allows us to state that the mixed marriage rate of Catholics is higher than that of Protestants. The correlation is thus explained without making any additional assumptions about the followers of the two churches. But however important analysis of social systems may be, we cannot say that it is the one and only aim of the social sciences. Sociological or economic theories also aim at explaining individual behavior. It is indeed legitimate to study voting behavior, household consumption, and more generally the attitudes and choices of individuals. Apart from the explanations proposed by psychological theories, sociologists and economists explain these behaviors by analyzing characteristics of individuals in their relation to their environment.

A Multidimensional Definition of Micro and Macro Could we reduce the definition of micro and macro to the size of the unit of observation? Micrology is understood to be limited to small units, whereas macrology analyses groups, collectivities, communities, organizations, institutions, and even whole societies. This first, unidimensional definition, is, in fact, valid for neither sociology nor economics. On the other hand, the observation units used with certain theories do not fall into either of the two categories. The ultimate observation unit in symbolic interactionism or ethnomethodology, for example, is not the individual but the situation in which individuals interact. Nor is size a more relevant criterion, because the unit of observation for certain macrosociological theories, for example, is a limited number of small groups. Similar problems of definition are encountered in economics. The classic distinction between micro- and macroeconomics also refers to several levels. While microeconomics is the study of the behavior of individual decisionmakers (households, firms) or, more exactly, choice as limited by different conditions, macroeconomics seeks to analyze the economy as a whole by structuring its hypotheses around aggregate variables (e.g., national income or consumption). But what is true of sociology is also true of economics: it is not so much the unit of observation that matters as the way in which problems are simplified. The second definition, and the one we will adopt here, is multidimensional. It has at least three dimensions: (1) the nature of the patterns to be explained; (2) the type of hypotheses—rationalist or a-rationalist; (3) the unit

Micro–Macro Transitions and the Paradox …

353

of observation and analysis used in empirical testing of propositions. These units may be individuals, situations, or structures; the point is that they must be consistent with the hypothesis demonstrated, as discussed above. The first dimension pertains to the particular problem under study, which may involve either individual behavior regularities or social and group regularities. Voting for a particular political party, consuming this product rather than that one, adopting a particular attitude, making this economic or social choice rather than that one—these are all, of course, examples of regularities in individual behavior. On the other hand, accounting for the degree of heterogeneity in a given society (the frequency of intermarriage between ethnic or religious groups, for example); for the existence and intensity of cooperation or conflict within an organization; for integration and differentiation; for macroscopic balances and imbalances, involves studying social patterns. Let us consider the following problem taken from social stratification studies, understandably one of the main fields for macrosociology. When we carry on empirical research on the effects of a father’s socioprofessional category or an individual’s level of education on the social status attained by his son or daughter, we are in fact positioning ourselves at the microsociological level (though we may not be aware of it)—despite the fact that our intellectual project is macrosociological. As long as the propositions demonstrated are micrological, we are not violating logical rules; we can legitimately make assertions of the type “When, for individuals, the level of education increases, so does socioprofessional status.” Such affirmations derive from empirical research carried on in accordance with known, standardized rules (e.g. construction of a random sample, use of analytical statistical methods in which the individual is the relevant unit). But in proceeding thus, we are working from several implicit hypotheses, the simplest of which affirms that individuals’ decisions are entirely independent of each other. Nothing prohibits us from applying such simplifications as long as our hypotheses or empirically tested propositions do not go beyond the microsociological level. However, it is clear to any sociologist of stratification that (1) individual choices are in fact interdependent; (2) there are not an infinite number of socioprofessional positions; (3) the set of such positions constitutes a structure whose parameters function as constraints on individual choices. Obviously, when I decide to apply for a research post in sociology in a given institution, (1) I enter into competition with other applicants (interdependence of individuals); (2) I know that the number of positions in my discipline for which applicants are competing is limited; (3) the job openings in sociology that applicants are competing to fill are the direct result of what other

354

M. Cherkaoui

jobs are available both in other sections of the same department and other departments of the same institution; and (4) when making its decision, the institution that has activated competition for the jobs will take into account the interest other institutions of the same type may have in sociology. The second dimension of the definition of micro- and macrosociology proposed here refers to theoretical simplifications and the objectives of such simplifications. I will summarily distinguish between two types of hypothesis about the behavior of units of observation and analysis: a theory requires or implies either hypotheses that are rationalist or hypotheses that are not rationalist (this last type may be divided into a-rational and non-rational). “Rational” here is being used in Simon’s (1982, p. 405) broad sense: “a style of behavior (A) that is appropriate to the achievement of given goals, (B) within the limits imposed by given conditions or constraints.” The rational individual, then, cannot be reduced to Homo oeconomicus, whose main goal is to maximize the expected value of his utility function—such as monetary profit, for example. In other words, economic rationality is merely a particular species of a genus spanning numerous sociological theories, as is discussed for instance, in some of Boudon (1992, 1995, 1998a) and Demeulenaere (1996). But these theoretical simplifications of course only concern individual actors or units, not structures. If we preclude Hegelian-type visions, which reify universals and endow them with intentionality or rationality, it is obvious that under certain conditions, groups, organizations, and institutions may be assimilated to actors. The third and last dimension is the level of observation and analysis deemed most relevant to a given theory. That level may be individuals, households, organizations, a context-situation, or population distributions according to quantitative or qualitative criteria. The individual, the town, or any administrative division, are the units most commonly used in electoral sociology. Whereas for most economic and sociological research, the household is considered the most appropriate unit for studying consumption, in sociology of organizations and microeconomics theory of the firms the preferred basic unit is the single firm. For some correlations analysis— that between literacy and the proportion of the state budget allocated for education, for example—rates are required. In the example borrowed from social stratification research, we saw why observations made at the level of the individual do not suffice to give us answers to macrosociological questions; they may even be an inappropriate place to look for such answers. The empirical tradition at the beginning of the 1950s and popularized by Blau and Duncan’s American Occupational Structure (1967) and Featherman and Hauser’s research (1978), among others, does not allow for a move

Micro–Macro Transitions and the Paradox …

355

from micro- to macrosociological. Is it possible to identify these transition mechanisms from one level to the other?

2

The Four Forms of Micro–Macro Transition and the Paradox of Consequences

Previously, I distinguished four types of relationships between phenomena that belong to the two micro and macro levels: it happens that the researcher wishes to describe and account for the individual by the individual, the collective by the collective, the collective by the individual, and the individual through the collective. Let us now examine them more closely.

Micro-Micro For this trivial case, there is no need to postulate the existence of a transition mechanism. In general, it is sufficient to argue that the action we are seeking to make intelligible conforms to the principle of rationality in the double Weberian sense of the concept and to draw out the logical consequences. Moreover, I am inclined to think that the form of social action that we describe as traditional—the fourth type according to Weber—is often in conformity with what I call “second-order rationality” (Cherkaoui 2006, 2007). I leave aside the purely emotional act, whose status is not the clearest despite the many studies in recent decades on the importance of emotions in decision-making or in behavior more generally. Firstly, action conforms to the logic of instrumental rationality when the actor takes into consideration the means at his disposal, the ends he pursues, and the subsidiary consequences of his action: this is the case of a household that decides to buy a consumer good or of the general staff of an army that develops a military tactic. Daily life offers us many examples of such things. I limit myself to the one that I borrow from Max Weber (1922) because it allows us not only to illustrate the rational conduct of an individual but also to distinguish two types of social phenomena. “I observe,” he notes at the beginning of Economy and Society, “that people in the street simultaneously open their umbrellas when it starts to rain.” The action of each of them is easily explained: not only are we able to answer the question of why they act in this way, but furthermore we do not need to look for causes other than their determination to protect themselves from the rain. The result of all these actions is a social fact that we qualify as resulting from it insofar as it is nothing other than the sum of individual actions

356

M. Cherkaoui

that are independent of each other. If the actions of some were influenced by those of others—as in the dissemination of a fashion, a rumor, or news, by word of mouth, by social networks, or by any other means—they would produce a social phenomenon that we would describe as emergent according to Hempel’s terminology (1965) derived in fact from John Stuart Mill (1843). Secondly, the action conforms to the principle of axiological rationality if the actor makes a decision based on an ultimate value, duty, or honor, regardless of the costs and consequences of his action. This last type of behavior is therefore based on a non-consequentialist prescriptive belief. Antigone’s decision to give a burial to one of her two brothers, Polynices, defying the prohibition of the holder of power, her uncle Creon, results from her conformity to a divine norm which is to bury the dead. The same is true of the suicide attacks by members of the Ismaili brotherhood of the Hashshashin, founded and led by Hasan As-Sabah, the Old Man of the mountain of Alamut, or the ultimate sacrifice of those resisting the invasion of their country by a foreign army. No consequentialist explanation makes these acts understandable to us, the outcome of which is often death, which the actors knowingly expected, and that Durkheim (1897) calls altruistic suicide. Only the theory of axiological rationality, which assumes that the actor acts according to strong beliefs perceived as legitimate and anchored in a community context, makes these actions comprehensible to us. In either case, we understand individual behavior to the extent that we can identify the reasons responsible for it and also to the extent that we need not ask any further questions about the causes of such conduct since the explanation is final and thus definitive. We do not need to assume the existence of any “black box,” of any of these “reasoning beings,” which most often exist only in the heads of those who use them (Boudon 1998b). It happens that we appeal to black boxes, such as the primitive mentality referred to by LévyBruhl or the habitus of Bourdieu, which are for us only the asylum of our ignorance, according to the formula that Spinoza applies to those who take refuge in the will of God. Let us now consider the unexpected consequences of the actor’s action when we are at the strictly micrological level.

Limits of Instrumental Rationality and Their Effects The first source of the emergence of unexpected consequences that Weber describes as Paradoxie der Folgen or the “paradox of consequences” stems from the relationship between the means used and the ends pursued by an

Micro–Macro Transitions and the Paradox …

357

actor (for the Weberian typology of generative matrices of unexpected consequence of those paradoxa, see Cherkaoui 2006; Merton 1968). Ideally, in a definitively rational action, one for which the objective is well defined, the most adequate means are implemented, for which therefore the logical relations between the antecedent and the consequent are known and are under control, no unexpected consequences will appear. But it often happens that the information is incomplete or that the relations between means and ends are inappropriate. If therefore there is ignorance or error, the predictions being mistaken and the consequences of the action unexpected, and they will be all the more distant from the agent’s objectives as cumulative processes may arise and locate themselves between this inadequacy and other unexpected effects which are due to other sources, such as the interdependence between actions or the conflicts between the rationalities of logically autonomous spheres, such as, for example, between religion and the economy or between religion and politics. The problem posed by the incomplete information on the world held by the actor and the difficulties inherent in the calculations that the latter must carry out in a very short time impose on him costs that are difficult to bear, and which persuade him to use rational procedures that I describe as being of a second order, such as imitation or tradition (Cherkaoui 2006, 2007). Moreover, when there is a feedback loop between the information, never pure and perfect, which the actor has about the world, and his decisions, we must expect the appearance of identical effects insofar as human behaviors and social systems are all open. Forecasting is, in this case, strictly speaking, impossible. We are able to correctly predict the future states of systems as long as we have access to the laws and theories that describe and explain the logical relations between phenomena, and provided also that the system is really closed or that the hypothesis of its closure is a reasonable one, that is to say that it is closed to any influence or external interference which risks disturbing its operation. Certainly, the predictions of the position of the planets of the solar system are based on the laws and theory of motion and gravitation as well as on previous empirical data; but they also turn out to be correct because the system in question is not disturbed by external interference. However, this closure condition is rarely satisfied when we are concerned with individual behavior and social systems.

358

M. Cherkaoui

The Paradox of the Consequences of Axiological Rationality While it is true that unintended consequences affect all types of action, they are more of a problem for actions that have rational purposes than for those that fall under the rubric of axiological rationality. Since, indeed, axiologically rational actions relate only to ultimate values and that they can often be executed without regard to the consequences they produce for the actor and for others, their sometimes unexpected character does not represent, in fact, an enigma either for the sociologist or for the actor. It does not matter that the action turns against the person who made the decision to act and destroys him as long as he agrees to conform to the ethics of the Sermon on the Mount which he holds for an ultimate value that gives meaning to his life. It even happens quite often to those who act axiologically that they have to face consequences that are sometimes contrary to the values that govern their interpretation of the world, their decision-making, and their action. After all, everyone knows that evil can result from good but also that it can, conversely, engender good. In one of the Letters to the Romans (7,19), Saint Paul expressed this idea of heterotely by affirming: “For I do not do the good I want to do, but the evil I do not want to do—this I keep on doing.” Similarly, in Faust, Goethe has Mephistopheles say: “I am part of this force which always wants evil and yet produces good.” In a famous passage from Politics as a Vocation, Weber (1919) reminds the politician of the paradoxical ethical consequences that result from the action he undertakes to achieve his goal. One who acts under the maxims of the ethics of conviction, without explicit or implicit reference to the consequences of his action, must expect to face unexpected outcomes. But even someone who acts according to the rules of the ethics of responsibility and who, in principle, chooses the most appropriate means to achieve his end, who considers and accepts responsibility for the consequences of his decisions, is not always free from paradoxical side effects. It is also conceivable that such phenomena are encountered in all actions that fall under the heading of instrumental rationality as well as axiological rationality, and formal rationality as well as substantive rationality.

Micro–Macro The problems posed by the transition from micro to macro are both complex and varied due in particular to the nature of macro-phenomena. It is indeed not certain that the same explanation is, strictly speaking, equally valid for a norm, a sphere of norms, an institution, or morphology. The structure of the

Micro–Macro Transitions and the Paradox …

359

prisoner’s dilemma, a jack of all trades of rational choice theorists, makes it possible to understand the production of a single norm one at a time, but it cannot account for a cosmos of norms, i.e., let us say, a relatively coherent set of prescriptive rules. The most eloquent example is the emergence of a tribe, a state, or an empire as morphology. Each is the consequence of a long process of crystallization of interactions between several social units. The prisoner’s dilemma model finds its cognitive limits here (see Cherkaoui 2023). Moreover, norms also differ in terms of their nature: a social norm is different from a legal norm as defined by Kelsen (1932, 1934, 1979) for example. The former is diffuse, the latter is institutionalized. The transposition of the first to the second is often the responsibility of the legislator and is supposed to conform to the logic of positive law according to a hierarchy of norms, according to Kelsen, which does not necessarily exist in the universe of social norms.

Paradigmatic Examples Let us illustrate the nature of the transition between the two orders with two classic and simple examples. I will raise later the problems posed by the differences in the nature of the macrophenomena. The first, and the most famous, is the mechanism of the “invisible hand” which transforms purely selfish individual behavior into an unexpected macrophenomenon that is the general interest. The purpose of the actions of economic actors is not the production of this good, which is moreover unpredictable when we analyze the motives that incite them to act. Without the existence of this transitional device that is market competition, we would remain at the individual level and we will not understand what is happening at the collective level. Competition reflects a form of interdependence between individual decisions, i.e., the fact that the action of one agent influences those of others. It is this that is the cause of this unexpected consequence. It has been said that Adam Smith, like so many others, was influenced by Mandeville (1705). It should be noted, however, that the analyses of the Scottish economist go far beyond the insightful and surprising findings of the author of The Fable of the Bees. The Wealth of Nations provides us with two more occurrences of unexpected effects. The first is the division of labor “from which flow so many advantages, that it must not be regarded in its origin as the effect of a human wisdom which has foreseen and which has had as its goal that general opulence which is the result of it.” warns Adam Smith. He adds, however, that the division of labor also has unfortunate consequences. By dint of carrying

360

M. Cherkaoui

out simple operations in a very limited number, the mass of the people sees its intelligence weakened: the people “thus naturally lose the habit of deploying or exercising these faculties, and generally become as stupid and as ignorant as it is possible for a human creature to become…As for the great interests, the great affairs of his country, he is totally out of the position to judge, and unless some very special pains have been taken for them to prepare for it, he is equally unfit to defend his country in war: the uniformity of his sedentary life naturally corrupts and breaks down his courage, and makes him contemplate with an aversion mingled with dread the varied, uncertain and hazardous life of a soldier….Thus, his dexterity in his particular trade is a quality which he seems to have acquired at the expense of his intellectual qualities, his social virtues and his warrior dispositions.” The second occurrence rarely noticed and yet of capital importance is the theory of the decline of feudal society to which chapter four of Book III of The Wealth of Nations is devoted. Smith explains it in terms of methodological individualism. This social change is the unintended and unforeseen consequence by social actors. Indeed, the ruin of the wealthy landowners was not intended and could not be predicted either by them or by the merchant class who pursued no other goal than their own private interests. “The motive of the great proprietors was to satisfy a ridiculous vanity. The merchants and manufacturers, much less ridiculous, acted purely in view of their own interest, and according to that principle familiar to every merchant class, that a small profit should not be neglected as soon as it is possible to realize it. Not one of them sensed or foresaw the great revolution which the extravagance of some and the industry of others were gradually bringing to an end.” Smith evokes several mechanisms that explain this transition, among which I mention two main ones. Firstly, the wealth acquired by the inhabitants of cities was frequently employed in purchasing such lands as were to be sold, of which a great part would frequently be uncultivated. Merchants are commonly ambitious of becoming country gentlemen, and when they do, they are generally the best of all improvers. A merchant is accustomed to employ his money chiefly in profitable projects, whereas a mere country gentleman is accustomed to employ it chiefly on expenses. Secondly, commerce and manufactures gradually introduced order and good government, and with them, the liberty and security of individuals, among the inhabitants of the country, who had before lived almost in a continual state of war with their neighbors and of servile dependency upon their superiors. The effects of interdependence are therefore sometimes unexpected and happy ones, and sometimes unexpected and unfortunate ones as in the classic examples of the tragedy of the Commons (Hardin 1968), the psychology of

Micro–Macro Transitions and the Paradox …

361

crowds (Le Bon 1895) or panic of whatever kind it is, whether one affecting a stock market or a social disaster as in the case of a fire in a theatre (Brown 1965). The stock market panic or “run” is not explained by the assumption of irrationality on the part of the actors involved in it, but on the contrary as being the consequence of rational behavior. It is assumed that the individual who withdraws his assets bases his action on information, true or false, which makes him doubt the credibility of his bank. If a large number of individual depositors share the same belief and ask to be reimbursed and if the amount of withdrawals exceeds that of the reserves, the bank ends up insolvent. Everything happens as if there was an agreement between the individuals whose behavior becomes significant to others. In fact, there is neither agreement nor connivance. Panic is, indeed, an example where two processes combine. The first is the simple sum of the actions of isolated individuals. The second is no longer a sum but the aggregation of behaviors that generates a sui generis phenomenon that individuals did not expect. It assumes the existence of a structure of interdependence between the actors, who are no longer isolated. This structure conditions their expectations, and leads them to make the same decision. As soon as we reach a critical mass, we see a singular macrophenomenon appear. We translate this mechanism by the proposition that the behavior of an individual depends on the number of those others who have also adopted it. In fact, behind this sibylline phrase lie complex phenomena, the detailed study of which can be found in Schelling (1978). If, therefore, individuals make decisions based on information, the effect is by no means an intended outcome. It does not matter whether the information they hold is objectively true or false, the important thing is that they believe it.

Externalities and Production of a Norm Methodological individualism is also able to account for the production of a norm (Boudon et al. 2001a, b eds.). Social norms are not the product of the simple addition of frequent behaviors but the crystallization of responses that members of a society have given to problems they have encountered. According to this meaning, the norm is not a statistical concept, defined by the average or by the majority of individual behaviors as in the case of a Laplace-Gauss curve usually described as a normal curve. Rather, it is a prescriptive rule, a regulating principle. By embracing the heritage of the sociological tradition of the founding fathers, the protagonists of methodological individualism reject the point of

362

M. Cherkaoui

view of certain rational choice theorists who consider that the norm is superfluous and that the principle of utility maximization is fully sufficient to account for individual actions. For many sociologists, norms are macrophenomena which are certainly based on individual actions but which are only produced under certain conditions. They emerge if the actions of some individuals, Ai, cause for others, Bj, negative or positive externalities and if they cannot be reduced through transactions that give the Bj the right to control the actions of the Ai. Those who are affected by negative externalities will have an interest in establishing norms to eliminate them, while those who cause positive externalities will seek to institute others to compensate. Norms are second-order public goods that individuals will seek to produce, i.e. means to achieve objectives common to these individuals but not yet achieved. This of course assumes that (1) we have good information on the effects of the norm and on the type of actions to be taken to produce it, (2) the problem of the free rider is solved as in the case of any collective action (Olson 1965). The same logic applies to the emergence of unintentional norms, taking as a starting point not the production of a public good but that of a private good. Finally, one or more mechanisms are necessary for the production of norms. They can be generated either intentionally by institutions or organizations such as the parliament or the judiciary or unintentionally by the aggregation of individual behaviors. The second example relates to the hypothesis that one norm can perfectly well be generated by another. Attempts to solve this problem within the framework of instrumental rationality do not win approval either. Coleman (1990) recognizes that rational choice theory is only able to explain the genesis of one norm at a time (Cherkaoui 2003a). However, norms are generally not independent of each other. How then can we account for the relationships they have with each other and the fact that some can produce others? Although rational choice theorists remain silent on this subject, Weber on the other hand provides a model for solving this problem when he studies in particular the links between several macro-phenomena. How is it possible to explain the genealogy of a constellation of norms understood in the sense of a coherent whole? How are they structured? What are the influences on them? A first answer to these questions is partly dealt with in the third section of the second part of this text where I show how Weber highlights the strong relations between the two macrophenomena that are religious ethics and democracy. It is also easy to show how legal norms, rational law, are produced by processes similar to those that Hobbes identifies in Leviathan.

Micro–Macro Transitions and the Paradox …

363

Methodological individualism is also able to explain the genesis of an institution. Let us refer to Carl Menger’s theory of money (1892). The origin of money is neither an invention of the state nor the result of a social convention. It is, according to Menger, the spontaneous result, the unforeseen consequence of the actions of agents who have gradually succeeded in discriminating and climbing up the different levels of market exchange of goods. Menger constructs a simulation to account for money’s emergence. Let us summarize it very briefly. Let us imagine a utopian society where the economy is based on barter. If an individual wants to exchange his good A for another good B, he must find somebody who at the same time has B and wants A. At the beginning of the process, the two individuals exchange their goods. It is easy to imagine the difficulties encountered by barterers who possess different goods with different values to them and for whom supply and demand do not correspond. Very rapidly, more and more exchangers see that certain goods C are more in demand or more easily sold than others. To possess goods C seems more advantageous than to have any other goods if exchange is what is wanted. An ever growing number of individuals demand C for all the other goods they exchange even if they do not have an immediate need for them. The goods that are most appropriate to this indirect exchange have all the attributes of money. In the end, indirect exchange becomes general and attains a threshold of universality such that we can say that money as a means of exchange is created. Menger believes that the most important task of the social sciences is to show how institutions that serve the common good come into existence without a common will directed toward their production. It also requires that the explanation of unintentional phenomena be based on methodological individualism. Better still: he is the only representative of the Austrian school who is fully aware of the need for a theory of the genesis of political authority. He offers some brilliant insights into this—akin to his theory of money—even if they are not as developed as one might have wished. But is methodological individualism able to make intelligible the emergence of a social morphology like the band or the tribe? I do not know. On the one hand, the birth and development of such phenomena take place over a long period of time, which is measured in millennia; on the other hand, we only have hypotheses on their formation because the archaeological data are tenuous and rare. All that can be advanced are only fragile conjectures or the results of simulation models such as those Menger devises for money. Take the case of the tribe. It is both a social organization and a political order. It is present in virtually all of the contexts present at one time or

364

M. Cherkaoui

another in human history. It is not the creation of any power. In all likelihood, it was born out of demographic and economic issues facing the band that needed to be resolved (Cherkaoui 2023 forthcoming). It is not a Taxis, an order wanted and produced by the actors, but a Kosmos, an endogenous, self-generated order, not wanted by the actors themselves but which is nevertheless the result of the crystallization of collective decisions and solutions. It will be understood by this that I am taking up Hayek’s dichotomy here (Hayek 1973).

Macro-Macro One of the major objectives of sociologists is to make such macrophenomena as revolution, the division of labor, a system of social norms, the birth of capitalism, etc., intelligible. I will take two classic examples to illustrate the methodological approach and the transition from macro to macro. The first relates to Tocqueville’s explanation of the Revolution, which I have dealt with elsewhere and to which I will not return (Cherkaoui 2003b, 2005). I borrow the second example from Weber who identifies the relationship between religious ethics and democracy, two macrophenomena. Why, asks the German sociologist, has Puritanism, which is an eminently elitist religious doctrine, strongly contributed to the institutionalization of modern democracy? How is it possible to make such a paradox intelligible? For him, democracy does not consist solely or mainly of a set of institutions intended to select political elites. It is above all a value or an ideal insofar as we believe in it. It is therefore not possible for our faith or our belief in the excellence of democracy to be demonstrated or subjected to an empirical test. For its part, religious ethics is a coherent set of prescriptions and proscriptions. Let us avoid any misunderstanding and specify from the outset that Weber does not support the thesis according to which Puritanism is the essential cause of democracy. He is aware of the existence of several types of democracy, ancient and medieval, which have left their mark on modern democracy. However, only the specificity of the latter interests him. Why have those societies influenced by Protestant asceticism historically been the cradle of modern democracy? Why was the new Calvinist-inspired Church so difficult for monarchies to accept? Why did Louis XIV and Richelieu, the great promoters of absolute monarchy, clearly perceive the danger that the Huguenots posed both to the monarchical order and to the hierarchy of the Catholic Church because of the democratic tendencies of the

Micro–Macro Transitions and the Paradox …

365

communities of the Reformed and their synods? Why were they generally hostile to all Caesarism, to all authority? Weber’s answer is complex. Let us present it in a simplified form. We should firstly recall the personal contribution of Calvin, whose elitist doctrine rejects democracy as a source of disorder. We will then examine the decisive role of the Protestant sects. God distributed his gifts unequally, argues Calvin. It is therefore up to the best to govern. It is moreover with religious power as with political power which is at the service of the former since both must be at the service of God. The basis of authority is the absolute sovereignty of God, not a natural human right. Calvin’s political ideal is therefore neither monarchy, which leads to personal abuse and arbitrariness, nor democracy, which presupposes equality of gifts between citizens and which, according to him, leads to confusion and disorder, taking up thus on his account the Aristotelian doctrine. His ideal is indeed an aristocracy. It stems from the religious qualification of each individual. Either the latter is qualified or elected or he is not. However, paradoxically, we are going to see that Calvin’s doctrine generates an unexpected and undesired effect which is modern democracy. In this discussion of the best form of government, what must in his eyes prevail over all other considerations is the necessity of preserving in society what he calls “a well-tempered liberty,” which maintains both rulers and the ruled in the performance of their duties to each other. But, he points out, however great the gifts and however vast the knowledge of the pastor, he can only exercise his magisterium under the guarantee of constitutional laws and as long as he is accepted by the faithful. In addition, the procedures for electing church leaders contain many democratic elements. Whether they are pastors, elders, or deacons, and although they are chosen by the Governing Council, they must be presented to the General Council of the city for an annual election. The Church is the result of cooperation between all the faithful. The pact of alliance which binds the pastor to his faithful should not be interpreted as an expression of popular sovereignty but as the recognition of the right that the community possesses to submit to discipline only if it is freely consented to, which is indeed the germ of a conception of the political contract. Secondly, we can deduce from the dogma a certain number of properties relating to social relations which are characteristic of modern democracy. The immunity of Puritan societies to Caesarism and providential men was emphasized by Weber (1905–1906). Among the Quakers, the principle is pushed to its extreme limit: they refuse even manifestations of politeness such as uncovering the head, bowing or using the plural when addressing

366

M. Cherkaoui

a person of high social status. “We know,” wrote Weber (1905–1906), “how this principle [of the absolute rejection of the idolatry of creatures] of ascetic Protestantism historically founds the specific character of contemporary democracy among those peoples influenced by Puritanism in contrast to that of peoples of the ‘Latin spirit’. It also constitutes, in part, the historical background of the ‘disrespectful’ attitudes of Americans, so irritating or so comforting according to one or the other.” There is no doubt that the organization and mode of operation—particularly the recruitment—of Protestant sects has shaped the structure and specificity of American society and democracy. It will be recalled that recruitment in the many clubs and associations was done by means of the vote. No one could access it without first having proven themselves. However, it is true of the confirmed social qualification of the individual as of his religious qualification which is the only orthodox way: neither the magical nor hereditary sacraments nor money alone are likely to give the individual access to social honor, to prestige, or to make it possible to buy them. The sect functions as a selection body that separates the qualified person or the chosen one from the one who is not. Only qualification, consequently social success which is expressed above all by acceptance by ballot in associations— clubs or sects—constitutes the foundation of community association; and it is indeed according to the model of the sect that all the social organizations that characterize American society have been constituted. It is not difficult to show the elective affinity between the sect and modern political democracy by examining a structural element common to both organizations, that of the administration of the affairs of the community by its own members. Cults insist on direct and democratic administration by the community and on treating religious “employees” as servants of the community. Finally, let us add that the organization of the Calvinist Church is democratic since it is the lower levels that in principle elect their delegates to the higher level and not as in the Catholic Church where the hierarchy is descending since it starts from the Pope to reach down to the parish priest. It would be neither wrong nor historically uninteresting to recognize in the ascending hierarchy of the Reformed Church (mainly among the American Presbyterians whom Weber knew from having observed them) a model of the structure of modern democracy. Let us summarize this discussion with Fig. 1, which visualizes the relationships between the macro and micro levels.

Micro–Macro Transitions and the Paradox …

Religious ethics

Religious qualification

367

Modern democracy

Social relations

Fig. 1 Relations between macro-phenomena (religious ethics i.e. set of prescriptions and proscriptions in relation to the world and secular institutions such as economy and politics; modern democracy i.e. ultimate political values) and micro-phenomena (qualification i.e. disposition of gifts or abilities that attest to being chosen by God; social relationships e.g. no allegiance to a man, membership in a sect such as voluntary association subject to vote, administration of community affairs)

Macro–Micro We can observe on a daily basis that many individual (micro) behaviors conform either to institutional rules, which are explicit or formal, or to social norms, which are generally implicit or informal (macro). How and why does the individual agree to direct his conduct according to a law, a maxim, a moral rule? How is it possible to make intelligible the constraining power of an institution, the fact that it exerts an influence on the representations, projects, action, and models of action of the individual? Why, when it is not written anywhere and is not sanctioned by any legal authority, does the norm also have coercive power? This is indeed one of the leading problems with which the sociologist has been confronted since the birth of the science of societies. But does not account for the multiple effects of a macrophenomenon, institutional order, or normative order, on a micro phenomenon lead us to reify abstract concepts and to account for particular facts by universals? The passage from the macro to the micro, from the social rule to the conformity of individual behavior, is explained by the sanction in the event of a violation of a law or a norm (Durkheim 1893; Parsons 1951), or by the hopes of gain that the agent believes he can derive by respecting them instead

368

M. Cherkaoui

of violating them (costs-benefits). This point of view of rational choice theorists (e.g. North 1992, 2005) seems to me to be redundant in relation to normative theories. The difference between the two conceptions is purely semantic; because the sanction is not limited to penalties but also includes benefits. The range of punishments and rewards is very wide, ranging from fines to imprisonment or ostracism, from pecuniary compensation, and social prestige, to the compensation promised by religion, such as the examples I previously mentioned of Antigone or the perpetrators of suicide attacks. The norm has, let us recall, two meanings. The first is purely statistical. It refers to the average or the majority of behaviors or facts described by mathematical laws, that of Laplace-Gauss or the so-called normal law in particular, without reference to any prescriptive rule. The distributions of several demographic characteristics, such as the size of a population at a given age, are examples. According to the second meaning, the one that concerns us here, it is a prescriptive statement or formula that specifies the way in which the individual must or should behave in given circumstances. The social norm is informal, more or less diffuse, and sanctioned by public opinion. The legal norm is, for its part, explicit, made enforceable, and repressed by judicial institutions if it is violated. There are also cognitive norms or “norms of thought” according to the expression that Durkheim (1912) uses in the Elementary Forms of Religious Life, such as the logical principles which determine our modes of thought.

The Norm and Suicide The influence of the normative sphere on voluntary death is undoubtedly the most famous and disconcerting example. It offers us the model of a disruption of the normative sphere, when it becomes either insignificant and without control over actions, or excessive and tyrannical. In the first case, it leads to anomic suicide, in the second to fatalistic suicide. Durkheim (1897) opens the chapter on anomic suicide with the following affirmation: “Society is not only an object which attracts to itself with unequal intensity the feelings and activity of individuals. It is also a power that regulates them.” In other words, society is not only a social morphology that encloses the individual in social networks such as family, friends, and professional associations; but it is also a set of norms whose function is to regulate individual behavior (Besnard 1987). It integrates as much as it regulates. It can produce conformism or deviance with its varieties such as innovation or rebellion. When social norms are relatively clear, stable, and firm, any transgression calls for a sanction from the group to which the individual belongs or from

Micro–Macro Transitions and the Paradox …

369

the entire community. When, on the other hand, they become doubtful, uncertain, vague, or, worse, when they are so strongly weakened that they are violated without reprobation, individuals feel free from the weight of the constraint which they usually exert on them. This situation characterizes anomie, a key concept of the Durkheimian theoretical heritage that will be taken up by the sociological tradition. The individual thus freed from all social pressure “cannot attach himself to ends that are superior to him and submit to a rule.” He “no longer knows where legitimate needs stop.” Anomie engenders the “vanity of boundless ambitions” in Durkheim’s words, excessive expectations, and projects that have a strong chance of failing. It is therefore to be expected that the resulting frustration will aggravate the suicide. Anomie spares no sector of social activity. Suicide particularly analyzes the consequences of both economic anomie and conjugal anomie. The first follows rapid recessions or expansions. The rapidity of economic progress determines a new way of thinking and behaving. In a society where the economy is experiencing strong and sustained growth, it is to be expected that individuals tend to formulate very high expectations which are not necessarily met. This frustration is causing suicide rates to worsen. Rather than subscribing to the Durkheimian explanation by anomie, Boudon (1977) offers an interpretation of the influence of strong economic growth on suicide that consists in simulating the winnings of competing actors according to their participation in a lottery. But as original and subtle as it is, this exegesis applies only to strong economic expansions and not to economic recessions. But these depressions have the same influence on suicide: they increase it not because they impoverish but because they “strongly disturb the social order” in the same way as impetuous prosperity, asserts Durkheim. That economic crises increase the suicide rate, current data, more precise than those available at the end of the nineteenth century, demonstrate beyond doubt. However, is the explanation offered by the French sociologist convincing? How should we follow Durkheim when he affirms that “misery protects against suicide” being a brake on desires? The hypothesis is so ambiguous that it is undecidable: it can be held to be both true and false. It is true that suicide rates are relatively lower in poor countries than in rich countries. However, we must immediately add that poverty does not have the same meaning in the former societies where solidarity is strong as it does in the latter, which are more individualistic, where it stigmatizes more. But economic recession, which is not poverty, definitely influences mental health and suicide for other reasons. On this point, Durkheim betrays his own theory relating to the effects of discrepancies between expectations and reality. The brutal recession, which follows a period of growth perceived

370

M. Cherkaoui

as normal to which the actors are accustomed, is indeed an example of the impact not of misery but of this sudden trend reversal which produces social downgrading. This explains why men are sensitive to the increase in unemployment in countries which hardly experienced it before the crisis, as well as the distress of peasants today, their isolation, and the strong constraints that are exerted on them to survive (debt and lower prices imposed by central purchasing), that account for their excess mortality by suicide. Indeed, deciphering this recession-related increase in suicide rates requires taking both integration and regulation into account. For its part, marital anomie is due to “the weakening of marital discipline that divorce implies.” It is observed that the suicide rate increases with the increase in divorces. Recent American longitudinal studies show that divorced people are 2.5 times more likely to commit suicide than married people. All other things being equal, divorce increases the probability of suicide for the divorced man more than the divorced woman: in the United States, the former is ten times more likely to commit voluntary death than the latter. Similar results have been established for other countries. The same reasoning applies to conjugal anomie as it does to economic anomie. Marriage, singularly monogamous, obliges the man to attach himself to only one woman. He limits his desires; in a way, it closes the horizon of possibilities. The dissolution of the marriage frees him; he finds himself in the situation of the bachelor who can attach himself to whoever pleases him and aspires to everything. Nothing pleases him. It nourishes new hopes which are often disappointed. Such frustration is accompanied by this “sickness of infinity that anomie brings everywhere with it.” This explains why the suicide rate is higher among single and divorced people than among married people. The norm exerts an influence on individual behavior according to the following process: deregulation (macro), due to rapid reversals of the economic trend or of divorce, generates a substantial gap between individual expectations and reality (micro) which determines in its in turn the frustration or social downgrading that produces suicide (micro). While it is moderate, the regulation is enough to contain the individual within limits without hindering his freedom. When it is weak, it breeds anomie. And when it is excessive it produces fatalism which increases the probability of committing suicide. ******* At the end of the combinatorial analysis of the two levels of reality, we are able to conclude that methodological individualism remains the most promising and satisfactory research program insofar as it makes it possible to define a methodological procedure authorizing the founding of the macro on the

Micro–Macro Transitions and the Paradox …

371

micro and to make macro-macro, micro-micro, and macro–micro relationships intelligible. If the difficulties encountered by the supporters of this modus procedendi mainly concern the explanation of the emergence of social morphologies such as the band or the tribe, as well as other phenomena such as the prohibition of incest or myths, whose formation and crystallization only occur over a very long period of time, which is not the short time period commensurate with the individual, they can hope to overcome them when the long series of data will be available to researchers or, failing that, when they are able to build simulations to generate these macrophenomena from hypotheses on the behavior of actors.

References Agénor, P.-R. (2012). Public Capital, Growth and Welfare: Analytical Foundations for Public Policy. Princeton, NJ: Princeton University Press Aghion, P., & Howitt, P. (1998). Endogenous Growth Theory. Boston: MIT Press. Alexander, J. et al. (1987). The Micro-Macro Link. Berkeley: University of California Press. Bertillon, A. (1897). Mésologie [Mesology]. In Amédée Dechambre (dir.), Dictionnaire encyclopédique des sciences médicales [Encyclopedic Dictionary of Medical Sciences] (pp. 211–212). Paris: Masson. Besnard, P. (1987). L’anomie [anomia]. Paris: Presses Universitaires de France. Blau, P. (1977). Inequality and Heterogeneity. New York: The Free Press. Blau, P., & Duncan, O. (1967). The American Occupational Structure. New York: John Wiley and Sons. Blau, P., & Schwartz, J. (1984). Crosscutting Social Circles: Testing a Macrostructural Theory of Intergroup Relations. New York: Academic Press. Boudon, R. (1973). L’inégalité des chances. Paris: Armand Colin [Transl. 1974. Education, Opportunity and Social Inequality. New York: Wiley]. Boudon, R. (1977). Effets pervers et ordre social . Paris: PUF [Transl. 1982. The Unintended Consequences of Social Action. London/Basingstoke: Macmillan]. Boudon, R. et al. (ed.). (1992). Traité de sociologie [Treatise of Sociology]. Paris: Presses Universitaires de France. Boudon, R. (1992). Action. In R. Boudon et al. (ed.), Traité de sociologie [Treatise of sociology] (pp. 21–55). Paris: Presses Universitaires de France. Boudon R. (1995). Le juste et le vrai: études sur l’objectivité des valeurs et de la connaissance. Paris: Fayard [Transl. 2001. The Origin of Values. New Brunswick/London: Transaction]. Boudon, R. (1998a). La rationalité axiologique [Axiological rationality]. In S. Mesure (ed.), La rationalité des valeurs [The rationality of values]. Paris: Presses Universitaires de France.

372

M. Cherkaoui

Boudon, R. (1998b). Social Mechanisms Without Black Boxes. In P. Hedström & R. Swedberg (eds.), Social Mechanisms. Cambridge: Cambridge University Press. Boudon, R. (2003). Raison. Bonnes raisons [Reasons. Good Reasons]. Paris: Presses Universitaires de France. Boudon, R. (2006). Bonne et mauvaise abstraction [Right and wrong abstraction. L’Année sociologique, 56 (2), 263–284. Boudon, R., & Cherkaoui, M. (2000). Central Currents in Social Theory. London: Sage. Boudon, R., Bulle, N., & Cherkaoui, M. (eds.). (2001a). Ecole et société. Les paradoxes de la démocratie [School and Society. The Paradoxes of Democracy]. Paris: Presses Universitaires de France. Boudon, R., Demeulenaere, P., & Viale, R. (eds.). (2001b). L’explication des normes sociales [The Explanation of Social Norms]. Paris: Presses Universitaires de France Brown, R. (1965). Social Psychology. New York: Free Press. Bulle, N. (1999). La rationalité des décisions scolaires. Analyse comparée de l’évolution des systèmes d’enseignement secondaire français et américain au cours du xxe siècle [The rationality of school decisions. A comparative analysis of the evolution of the French and American secondary education systems during the 20th century]. Paris: Presses Universitaires de France. Cherkaoui, M. (1982). Les changements du système éducatif en France 1950-1980 [Changes in the French Educational System 1950-1980 ]. Paris: Presses Universitaires de France. Cherkaoui, M. (1997). Le réel et ses niveaux. Peut-on toujours fonder la macrologie sur la micrologie? [The Real and Its Levels. Can We Still Base Macrology on Micrology?] Revue Française de Sociologie, 28, 497–524. Cherkaoui, M. (2003a). Les transitions micro-macro. Limites de la théorie du choix rationnel dans les Foundations of Social Theory [Micro-Macro Transitions. Limitations of Rational Choice Theory in Foundations of Social Theory]. Revue Française de Sociologie, 44, 231–254. Cherkaoui, M. (2003b). L’État et la Révolution. Logique du pouvoir monopoliste et mécanismes sociaux dans L’Ancien régime de Tocqueville [The State and the Revolution. Logic of monopolistic Power and Social Mechanisms in Tocqueville’s The Old Regime]. Revue européenne des sciences sociales, 41(126), 61–79. Cherkaoui, M. (2005). Invisible Codes. Essays on Generative Mechanisms. Oxford: Bardwell-Press. Cherkaoui, M. (2006). Le paradoxe des conséquences [The Paradox of Consequences]. Genève: Droz. Cherkaoui, M. (2007). Good Intentions: Max Weber and the Paradox of Unintended Consequences. Oxford: Bardwell Press. Cherkaoui, M. (2018). Essai sur l’islamisation. Changements des pratiques religieuses dans les sociétés musulmanes [Essay on Islamization. Changes in Religious Practices in Muslim Societies]. Paris: Presses Universitaires de la Sorbonne.

Micro–Macro Transitions and the Paradox …

373

Cherkaoui M. (2023). L’empire et la tribu. Essais sur les morphologies sociales et les ordres politiques [The Empire and the Tribe. Essays on Social Morphologies and Political Orders]. Forthcoming. Coleman, J. (1964). Introduction to Mathematical Sociology. New York: Free Press. Coleman, J. (1990). Foundations of Social Theory. Cambridge: Belknap Press of Harvard University Press Comte, A. (1830–1842). 1975. Cours de philosophie positive [Course of Positive Philosophy]. Paris: Hermann. Demeulenaere, P. (1996). Homo oeconomicus. Enquête sur la constitution d’un paradigme [Homo oeconomicus. Investigation into the Constitution of a Paradigm]. Paris: Presses Universitaires de France. Durkheim, E. (1893). 1967. De la division du travail social [The Division of Labor in Society]. Paris: Presses Universitaires de France. Durkheim, E. (1897). 1912. Le suicide [Suicide]. Paris: Librairie Félix Alcan. Durkheim, E. (1912). 1968. Les formes élémentaires de la vie religieuse [The Elementary Forms of Religious Life ]. Paris: Presses Universitaires de France. Fararo, T. J. (1981). Biased Networks and the Social Structure Theorems. Social Networks, 3, 137–159. Fararo, T., & Skvoretz, J. (1986). E-State Structuralism: A Theoretical Method. American Sociological Review, 51(5), 591–602. Fararo, T., & Skvoretz, J. (1984). Biased Networks and Social Structure Theorems II. Social Networks, 6 , 223–258. Featherman, D., & Hauser, R. (1978). Opportunity and Change. New York: Free Press. Goethe, J. (1809). 1973. Die Wahlverwandtschaften [The Elective Affinities]. In Goethes Werke, vol. 6. Hamburger Ausgabe: München Oscar Becke. Hardin, G. (1968). The Tragedy of the Commons. Science, New Series, 162(3859), 1243–1248. Harsany J. (1968). Individualistic and Functionalistic Explanations in the Light of Game Theory: The Example of Social Status. In I. Lakatos & A. Musgrave (eds.), Problems in the Philosophy of Science (pp. 305–348). Amsterdam: North-Holland publishers. Hayek, F. (1967). Studies in Philosophy, Politics and Economics. Chicago: University of Chicago Press. Hayek, F. (1973). Law, Legislation and Liberty. London: Routledge and Kegan Paul. Hayek, F. (1978). New Studies in Philosophy, Politics, Economics and History of Ideas. London: Routledge and Kegan Paul. Hedström, P. & Swedberg, R. (eds.). (1998). Social Mechanisms. Cambridge: Cambridge University Press. Hempel, C. 1965. Aspects of Scientific Explanation and Other Essays in the Philosophy of Science. New York: The Free Press. Kelsen, H. (1932). 2004. Vom Wesen und Wert der Demokratie, La démocratie, sa nature, sa valeur [On the Nature and Value of Democracy]. Paris: Dalloz.

374

M. Cherkaoui

Kelsen, H. (1934). 1962. Reine Rechtslehre Mit einem Anhang: Das Problem der Gerechtigkeit. Théorie pure du droit [Pure theory of law]. Dalloz: Paris. Kelsen, H. (1979). 1996. Allgemeine Theorie der Normen, Manz Verlag Wien. Théorie générale des normes [General Theory of Norms]. Paris: Presses Universitaires de France. Le Bon, G. (1895). 1975. Psychologie des foules [Psychology of Crowds]. Paris: Retz. Mandeville, B. (1714). 1988. The Fable of the Bees. Oxford: Clarendon Press. Manzo, G. (2009). La spirale des inégalités. Choix scolaires en France et en Italie au XXe siècle [The spiral of inequalities. School choices in France and Italy in the 20th century]. Paris: Presses de l’Université Paris-Sorbonne. Menger, C. (1883). Untersuchungen über die Methode der Socialwissenschaften und der politischen Oekonomie insbesondere. Transl. 1985. Investigation into the Methodology of the Social Sciences with Special Reference to Economics. New York: New York University Press. Menger, C. (1892). On the Origin of Money. Economic Journal, 2, 239–255. Merton, R. (1968). Social Theory and Social Structure. New York: Free Press of Glencoe. Mesure, S. (ed.). (1998). La rationalité des valeurs [The Rationality of Values]. Paris: Presses Universitaires de France. Michels, R. (1911). 1914. Zur Soziologie des Parteiwesens in der modernen Demokratie. Untersuchungen über die oligarchischen Tendenzen des Gruppenlebens. Les partis politiques [Political parties]. Paris: Flammarion. Mill, J. S. (1843). 1973. A System of Logic, Ratiocinative and Inductive (Books I-III). In J. S. Mill. The Collected Works, vol. 7. Toronto: The University of Toronto Press. North, D. (1992). Transaction Costs, Institutions, and Economic Performance. San Francisco: ICS Press. North, D. (2005). Understanding the Process of Economic Change. Princeton: Princeton University Press. Olson, M. (1965). The Logic of Collective Action. Cambridge: Harvard University Press. Parsons, T. (1951). The Social System. London: Routledge and Kegan Paul. Paul. Saint. Circa 54–58. Epitre aux Romains [Letter to the Romans] (7,19). Popper, K. (1934). 1959. Die Logik der Forschung, Wien, Julius Springer, The Logic of Scientific Discovery. London: Routledge. Popper, K. (1945). The Open Society and Its Enemies. London: Routledge and Kegan Paul. Popper, K. (1963). Conjectures and Refutations. London: Routledge and Kegan Paul. Popper, K. (1972). Objective Knowledge: An Evolutionary Approach. Oxford: The Clarendon Press. Schelling, T. (1978). Micromotives and Macrobehaviors. New York: Northon and Co. Schumpeter J. (1908). Das Wesen und der Hauptinhalt der theoretischen Nationalökonomie [The Nature and Essence of Economic Theory]. Berlin: Duncker und Humblot.

Micro–Macro Transitions and the Paradox …

375

Schumpeter, J. (1942). 1976. Capitalism, Socialism and Democracy. London: George Allen & Unwin. Schumpeter. J. (1954). History of Economic Analysis. London: G. Allen & Unwin. Simmel, G. (1908). Soziologie. Untersuchung über die Former der Vergesellschaftung [Sociology. Investigation into the Forms of Socialization]. Berlin: Duncker & Humblot. Simon, H. (1957). Models of Man, Social and Rational . New York: John Wiley. Simon, H. (1977). Models of Discovery. Boston: D. Reidel Pub. Co. Simon, H. (1982). Models of Bounded Rationality. Cambridge: Harvard University Press. Skvoretz, J. (1983). Salience, Heterogeneity and Consolidation of Parameters: Civilizing Blau’s Primitive Theory. American Sociological Review, 4 (3), 360–375. Smith, A. (1776). 1975. An Inquiry into the Nature and Causes of the Wealth of Nations. Oxford: Clarendon Press. Sombart, W. (1906). Warum gibt es in den Vereiningen Staaten keinen Socialismus? Tübingen: Mohr (Paul Siebeck). Tocqueville, A. de. (1856). 1953. L’Ancien régime et la Révolution [The Old Regime and the Revolution]. In Œuvres Complètes. Paris: Gallimard. Weber, M. (1905). 1984–2013. Die protestantische Ethik und der “Geist” des Kapitalismus [The Protestant Ethic and the “Spirit” of Capitalism]. In Max Weber Gesamtausgabe [Max Weber Complete Edition]. Tübingen: Mohr. Weber, M. (1919). 1984–2013. Wissenschaft als Beruf. Politik als Beruf [Science as a Vocation. Politics as a Vocation] In Max Weber Gesamtausgabe [Max Weber Complete Edition]. Tübingen: Mohr. Weber, M. (1921). 1984–2013. Gesammelte Aufsätze zur Religionssoziologie, Max Weber Gesamtausgabe [Collected Essays on the Sociology of Religion]. In Max Weber Gesamtausgabe [Max Weber Complete Edition]. Tübingen: Mohr. Weber, M. (1922). 1984–2013. Wirtschaft und Gesellschaft [Economy and Society]. In Max Weber Gesamtausgabe [Max Weber Complete Edition]. Tübingen: Mohr.

Methodological Individualism and Micro–Macro Modeling Karl-Dieter Opp

1

Introduction

The following analysis addresses the question of how macro propositions can and should be explained. To illustrate, assume that societies with market institutions have higher economic growth than societies with central planning. This is a relation between macro properties (market institutions, economic growth), i.e., properties of collectives (ranging from dyads to entire societies). Social scientists usually want to explain such macro propositions. The previous hypothesis could be explained by analyzing how the macro properties are related to properties of individual actors, i.e., by considering the micro level. For example, market institutions (a macro property) might have a causal impact on the micro level: the institutions influence incentives of individual actors to realize their goals. These incentives lead to individual economic activities (such as founding enterprises) which then aggregate to economic growth. This procedure to explain macro propositions by considering the micro level is pursued by the research program of Methodological Individualism (MI), also called structural individualism or the structuralindividualistic research program. The question whether such explanations are K.-D. Opp (B) Sociology, University of Leipzig, Leipzig, Germany e-mail: [email protected] University of Washington, Seattle, WA, USA

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 N. Bulle and F. Di Iorio (eds.), The Palgrave Handbook of Methodological Individualism, https://doi.org/10.1007/978-3-031-41512-8_18

377

378

K.-D. Opp

possible is also called the structure-agency problem (“structure” is similar to the macro level, “agency” is similar to the micro level). MI assumes that such individualistic explanations are possible and scientifically fruitful.1 There are different versions of MI. In what follows a version of MI is presented and discussed that is based on the work of practicing social scientists. This will become particularly clear when we discuss examples for micro–macro modeling.

2

Basic Ideas of Methodological Individualism, Illustrated with the Example of Income Inequality

As the introductory example illustrates the basic idea of MI in the social sciences: macro propositions can be explained by recourse to individual actors and their properties. This is the explanation thesis. It is a factual assertion that can be falsified: one might find macro propositions that can not be explained by considering the micro level. MI further consists of an explanation postulate. The claim is that it is scientifically fruitful to explain macro propositions by drawing on properties of individual actors. How exactly can macro propositions be explained individualistically? The procedure is micro–macro modeling. It can be depicted with a graphical representation that is based on the so-called Coleman boat (see in particular Coleman 1986; 1990: 1–26), sometimes also called the Coleman bathtub. The scheme is based on a figure from McClelland (1961: 47) who is not cited by Coleman. A similar model is proposed already by Diekmann and Opp (1979), Hummell and Opp (1968: 209; 1971: 15), Hernes (1976), and von Wright (1971: 137). I use this scheme to analyze in detail the procedure and problems of micro–macro modeling. I start with an example: the effects of social inequality. In order not to complicate matters unnecessarily, I deal with income inequality. For the sake of simplicity, I simply speak of “inequality.” A high inequality in a society has numerous effects (e.g., Osborne et al. 2019). One finding is:

1

Collections of essays about MI are Alexander et al. 1987; Buskens et al. 2012; Wagner 2017; Zahle and Finn 2014. For reviews and discussions of MI see Hodgson 2007; Raub 2021b; Udéhn 2002. For discussions of MI (and reductionism) by philosophers see Brodbeck 1958; di Iorio 2015, 2016; Kincaid 1986; Zahle and Kincaid 2019. A textbook on explanation in the social sciences, based on MI, is Parri 2014. See further the special issues 2 and 3 of Cosmos + Taxis, 2016.

Methodological Individualism and Micro–Macro …

379

Macro proposition: The higher the inequality of a society, the higher is the likelihood of mass political violence.

One would like to know why inequality is (or is under certain conditions) related to collective violence. An intuitive explanation could be the following. Mass political violence is composed of actions of individuals like blocking streets and seizing buildings at a certain time and place. The macro variable “mass political violence” is an (arithmetic) aggregation of individual actions such as the number of people who engage in those actions. The next step of explaining the macro proposition is to explain the individual actions (and, thus, the aggregated dependent macro variable). Individual behavior could be explained by individual goals (such as the wish to change government decisions) and beliefs (such as the perceived impact of one’s behavior on the realization of one’s goals). What is missing so far is how inequality is related to mass-political violence. If individual goals and beliefs lead to macro-political violence—via individual violent actions, then inequality could explain mass-political action by changing goals and beliefs of people (that indirectly cause mass-political violence). For example, relatively high inequality may lead to high general individual dissatisfaction (i.e. goals such as having a higher income are not achieved) and a belief that violence may help to realize the goals. Note that there is not a direct but only an indirect causal effect of the independent on the dependent macro variable. Between the macro variables there is a correlation. This example is an illustration of the basic structure of a micro–macro explanation, i.e., of the explanation of the relation between inequality and mass political violence by considering the micro level. Figure 1 summarizes the explanation. Note the different kinds of relations between the variables. The relation between the macro variables is a correlation. The independent macro variable has a causal effect on the individual motivating factors. They have a causal effect on the individual behavior. This is analytically (i.e. logically) related to the dependent macro variable. As will be seen below, there might be other relations as well. The macro-to-micro and the micro-to-macro relations are called bridge assumptions: they are links between the macro and micro level. One important implication of the model should be noted already at this point: the micro–macro explanation assumes certain macro-to-micro and micro-to-macro relations. Only if they are valid, the explanation of the macro proposition is correct (assuming that the micro theory is valid as well). This implies that micro–macro modeling specifies the conditions for the validity of the macro proposition.

380

K.-D. Opp Macro level

Inequality Causal effect

Micro level

Goals and beliefs

Mass political violence

Correlation

Aggregation Causal effect

Individual behavior

Fig. 1 An example for a micro–macro model

The model in Fig. 1 can be extended in various ways (see below). For example, there may be more independent macro variables. The model explains a macro proposition, i.e., a relation between at least one independent and one dependent variable. It is also possible to explain macro phenomena. A question could be why there is mass political violence. One might omit the macro-to-micro part and look at the causes for individual violence and its aggregation. In what follows we will only deal with macro propositions. The micro–macro model can best be described as a set of methodological rules that can be applied to explain macro propositions. Given a macro proposition, the scheme suggests finding a micro theory (or micro propositions) that could explain the macro proposition. Then the micro proposition needs to be connected with the macro variables by bridge assumptions. The micro–macro model thus suggests how to proceed to realize the program of MI . We will now discuss each component of the micro–macro model: we analyze the kind of statements of the macro, micro propositions, and the bridge assumptions. Figure 2 summarizes the questions to be discussed. The previous example of explaining the impact of inequality is again chosen for illustrative purposes. As will be seen below, other explananda are possible as well. Income (3) Singular statements? Laws? Collective inequality action (2) Bridge assumptions: (a) Analytical – if so: kind of aggregation? (b) Empirical – if so: singular statements or laws?

Individual Goals and (1) Singular statements? Laws? behavior beliefs Lines symbolize any relationship. Based on Opp 2011: 213. Fig. 2 Kinds of statements in micro–macro explanations

Methodological Individualism and Micro–Macro …

3

381

The Micro Foundation of Micro–Macro Explanations

Micro–macro modeling requires hypotheses about explaining individual behavior (or other individual properties such as attitudes). These hypotheses are often called the “micro foundation” of MI. “Foundation” does not mean that there is something “foundational” in the sense that it is exempted from critique. The term simply refers to the lower part (the “foundation” in a physical sense) of the micro–macro model. In the inequality example (Fig. 1) the micro foundation was a behavioral theory asserting that goals and beliefs determine behavior. The micro foundation might be any valid proposition that explains individual action. Examples are the attitude-behavior theory by Fishbein and Ajzen (2010) or the social psychological value expectancy theory (which is a version rational choice theory – see below). One might also apply a theory of the middle range such a theory of political violence. Any valid micro propositions are admissible. It is preferable to apply a clearly formulated and valid theory, i.e., a general conditional statement that has been empirically confirmed. A theory specifies what factors in general determine an explanandum. It can thus be applied to identify the factors in the specific situation. Many advocates of micro–macro modeling use rational choice theory (RCT) as the micro foundation. Micro– macro modeling that applies RCT is often called the rational choice approach. There are different versions of RCT.2 One version that is increasingly accepted can be called a wide version. It is applied in sociology, behavioral economics (e.g., Thaler 2015) and social psychology (e.g., Tyler 2013). It is consistent with a “bounded rationality” approach (H. Simon). There is a large literature that does not explicitly mention RCT but implicitly applies it: authors use a wide range of goals and beliefs (i.e. rational choice variables) to explain social phenomena. It is further a common assumption in the social science literature that actors do what they think is best for them in the present situation. This is the assumption of subjective (and not objective) utility maximization. The assumptions of the wide version can be summarized as follows. (1) All kinds of goals may determine behavior (be they goals to increase one’s

2 For expositions and discussions of this theory see Kirchgässner 2008; Opp 2011, 2020a; Raub 2021a. For the different versions see Diekmann 2022; Opp 1999, 2011, 2020a, 2020b: 28–75. See also Knauff and Spohn 2021. A version of this theory that is often applied by sociologists and social psychologists is value expectancy theory. See Feather 1990; Nagengast et al. 2011; Wigfield et al. 2016.

382

K.-D. Opp

own or others’ wellbeing or goals to observe norms). (2) All kinds of beliefs are relevant. They may refer to one’s abilities or to the reactions of the social environment to one’s behavior. (3) Beliefs may be biased . (4) Goals and beliefs may change. (5) There is subjective utility maximization: people do what they think best in the present situation. So far the concept of rationality is not used—except when we refer to a theory called “rational choice theory.” “Rationality” has numerous meanings and is often not clearly defined (Opp 2018a). Applying the wide version of “rational” choice theory does not need the rationality concept. One might even change the name of the theory into goals-beliefs theory. This refers to the variables of the theory: preferences (i.e., goals) and beliefs. Other authors avoid “rational” as well. Gintis (2006) uses the label “beliefs, preferences, and constraints” or BPC model. Hedström’s DBO (“desires, beliefs, opportunities”) model is a version of RCT as well (see Hedström 2005 and the critique by Opp 2013). We will take up the relation between DBO und rational choice theory below. RCT is controversial. The critique usually addresses “the” theory of rational action and then describes the narrow version without even mentioning the wide version. A recent example is the discussion of MI and RCT by Ylikoski (2016: 142) when he asserts that “rational choice theory … is not a very rich or empirical account of individual agents. The strong assumptions about rationality and knowledge are justified by their analytical convenience.” All this does not apply to a wide version. RCT “is not a very rich or empirical account” apparently means that the empirically relevant motivations (including the motivation to follow norms) and beliefs (which can be wrong) are not included. All this is clearly wrong for a wide version. This is “empirical” in the sense that what the specific goals and beliefs are needs to be measured empirically. There are further no “strong assumptions about rationality and knowledge”—neither objective utility maximization, transitivity of preferences, or full information are required. Another widely held misunderstanding is that RCT assumes deliberative purposive agents. People may act spontaneously, and RCT explains when a conscious or spontaneous mode of action is chosen (see already Becker 1976: 7). The most extreme misunderstandings one can think of come from Bhaskar (1978: 7–8). He asserts that “relations play no part in this model.” Not even in a narrow version there is an assumption of atomized actors. The micro foundation often consists theories of the middle range (TMR) explaining specific kinds of phenomena such as crime or migration. A major objection against applying only TMR is that general behavioral theories specify the conditions under which middle-range theories hold. Evidence for

Methodological Individualism and Micro–Macro …

383

this assertion has been provided already by Homans (1974) and Malewski (1964, 1967). These and other authors show the limitations of numerous middle-range propositions and the conditions for their validity (see also Opp 1970, 2013). Hummell and Opp (1969) provided a comparative empirical test of hypotheses of learning theory and hypotheses from “Union Democracy” (Lipset et al. 1956). The results show that the latter had to be modified considering the former. Hummell and Opp (1968, 1971) also demonstrated for several sociological hypotheses that they need to be modified when they are compared with general behavioral propositions. Other analyses indicate how a wide version of RCT modifies macro and micro propositions from the social movements and deviant behavior literature (e.g. Opp 2009, 2020b). One argument to justify focusing on TMR is the belief that RCT is highly deficient (e.g. Hedström and Ylikoski 2014: 58–61, see recently Hedström 2021). Again, this critique is based on the narrow version of RCT. This cannot be shown here for space restrictions. Nonetheless, the following reasons speak against focusing only on TMR. Hedström himself (2005) proposed a general theory: DBO theory (with the independent variables desires, beliefs, and opportunities). This is a deficient variant of the wide version of RCT (see Opp 2013: 341–344). Preferences in RCT are identical with desires in DBO theory. “Beliefs” have the same meaning in both theories as well. However, in DBO theory it is not specified which kinds of beliefs are relevant for explaining a specific action. The “O” refers to behavioral options open to an agent. These are the explananda of RCT. A difference between the theories is that DBO theory does not include utility maximization (see the description of DBO in Hedström 2005: 38–42). This makes DBO theory unfalsifiable. If, for example, an individual has conflicting goals or conflicting opportunities: it is not specified which action is chosen. The conclusion thus is that DBO theory is a deficient version of RCT. DBO theory is not even mentioned in the 2021 article (Hedström 2021). One might therefore assume that the wide version of RCT is acceptable. This is confirmed when Hedström and Swedberg (1996: 127–128) write about RCT ten years earlier: Rational choice theory provides an action theory that is useful in many branches of sociology and, perhaps even more importantly, rational choice theory represents a type of theorizing that deserves to be emulated more widely in sociology. This type of theorizing is analytical; it is founded upon the principle of methodological individualism; and it seeks to provide causal cum intentional explanations of observed phenomena. (Hedström and Swedberg 1996: 127–128, emphasis added).

384

K.-D. Opp

This argument seems to be based on DBO theory. I therefore conclude that this version—and, thus, the wide version of RCT—is still acceptable. It is striking that advocates of Analytical Sociology who denounce RCT are blind to the major deficiencies of TMR. In their contribution (e.g., Hedström and Ylikoski 2014: 63) TMR are praised in about half a page and are characterized as “a clear, precise, and simple type of theory.” One would have expected a detailed analysis of at least some TMR (e.g., anomie theory or the theory of status inconsistency). Are their major concepts (e.g., anomie) really clear? What about their empirical confirmations (which are not mentioned)? But assume one regards TMR as fruitful for many specific phenomena. An exclusive focus on TMR is dogmatic and an immunization of TMR if one ignores general theories. What is missing is extensive empirical theory comparison, i.e., comparative research of the relationships between RCT and TMR. The explananda on the micro level need not be behavior. Individual cognitions, and goals (including the goal to follow norms or to intrinsically benefit others) may be explananda as well. There are general theories that explain these phenomena. For example, there is a growing literature on dual-process theories that state conditions for activating and explaining preferences and beliefs (e.g., Gawronski and Creighton 2013). The cognitive learning theory by Bandura (e.g., 1999) or the attitude theory by Fishbein and Ajzen (2010) could be applied. Balance theory (e.g., van de Rijt 2011) or dissonance theory (Harmon-Jones and Mills 2019) are further candidates for a micro foundation. The conclusion is that a micro–macro explanation may have quite diverse micro foundations. Instead of a factor approach (simply introducing ad hoc some variables that are deemed causal determinants of the dependent micro variables) the application of a confirmed theory is preferable.

4

Bridge Assumptions I: Macro-To-Micro Relations

The relation between inequality and the independent variables of the micro theory (i.e. the macro-to-micro relation) is an empirical causal effect. This effect will only exist if inequality is perceived . People often have biased beliefs. They might overestimate or underestimate existing inequality. Let us assume that inequality is perceived correctly. Osborne et al. (2019) specify a “macro–micro model of inequality.” They argue that individuals compare themselves with those who are higher in

Methodological Individualism and Micro–Macro …

385

the income hierarchy. This leads to relative deprivation. This is an empirical effect: the perception of relatively high inequality causes relatively high relative deprivation. Ferreira et al. (2022) compile a table that lists numerous other effects. Could this causal effect of inequality on relative deprivation be a law? It is plausible that inequality leads to relative deprivation only under certain conditions. If individuals perceive inequalities as just, i.e. if unequal income is regarded as being based on merits so that those higher in the income hierarchy deserve what they get, there is no deprivation. The hypothesis thus is: If income inequality is high and is regarded as unjust, there is relative deprivation.

The “and” refers to an interaction effect. The macro-to-micro proposition (income inequality leads to deprivation) is thus not a law. It is only a singular causal statement (i.e., a statement referring to some time and place). One might call it an empirical regularity, i.e., it holds sometimes, if certain conditions are given. If macro-to-micro relations are singular statements, the question is how their validity can be examined. In the example, a general hypothesis was applied stating that norm violation leads to deprivation of those who hold the norm. The complete argument is: Theory: If a norm is violated, this leads to deprivation of those who accept a norm. Initial conditions: There is income inequality in a society, and this violates an accepted norm.

––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Conclusion: Income inequality in this society leads to deprivation.

The general procedure to validate singular macro-to-micro relation is to apply a theory. There may also be analytical macro-to-micro relations. Let a macro proposition be: the larger a group, the less likely is the production of a public good (e.g. Olson 1965). Assume the micro theory that explains individual contributions to a public good is RCT. Let the major explanatory factor be the influence of an individual on the production. This is negligible in a large group. The macro-to-micro relation is thus from group size to individual influence. This is empirical if the independent micro variable is perceived influence: the larger a group, the less influence on the provision of a public good is perceived. However, influence can be defined as 1/group size (1

386

K.-D. Opp

divided by group size). In a group with 100 members the influence is 1/ 100. This is an analytical (i.e. logically true) macro-to-micro relation. Macro-to-micro relations may thus be quite different. What the relation is depends on the research problem.

5

Bridge Assumptions II: Micro-To-Macro Relations

In the inequality proposition the relation between the dependent micro variable (individual political violence) and the dependent macro variable (mass political violence) is an analytical relation, i.e. an aggregation: the number of individual actions is counted (and perhaps divided by the size of the group). There are other micro-to-macro relations. The first detailed analysis of the relation between individual and collective properties is an article by Lazarsfeld and Menzel (1961). The work by Hummell (1973) and Jasso (2010) is in this tradition. Lazarsfeld and Menzel distinguish three kinds of collective properties. Analytical properties are “properties of collectives which are obtained by performing some mathematical operation upon some properties of each single member” (ibid.: 427). The standard deviation of individual incomes is an example. Structural properties are “properties of collectives which are obtained by performing some operation on data about the relations of each member to some or all of the others” (ibid.: 428). Assume friendship relationships between members of a group are measured. Cohesion as a group property could then be defined as the number of relationships between members of a group. These are analytical relations as well. Although “friendship” is a relational property (a is a friend of b, formally: Fab) it is also a property of each member of the relationship. Cohesion is thus a macro property constructed based on individual properties. Global properties are “not based on information about the properties of individual members” (ibid.: 426). An example is a written language of a society. Although these properties are not constructed from individual properties as analytical and structural properties, they can be explained by micro–macro modeling (see the analyses in Hummell and Opp 1971: 35– 38; 1968: 211–212). For example, let the dependent macro variable be the existence of a written language in a society. This is measured by the existence of written documents (laws, newspapers, schoolbooks). The dependent variable is thus measured without explicit reference to individuals. Nonetheless,

Methodological Individualism and Micro–Macro …

Government decision

Correlation

Causal effect

Goals and beliefs

387

Change of the governement decision Causal effect

Causal effect

Protest against the government decision

Fig. 3 Example of a micro–macro model with empirical bridge assumptions

a language is a spontaneous outcome of the actions of numerous individuals. To explain the existence of a written language thus means to explain why a language or a certain kind of language is created by individuals. Explaining the origin of a language in a micro–macro model one thus does not explain an object (language) but individual actions that create an object. There is thus an analytical relation of the language-creating individuals and the macro property of a language. Micro-to-macro relations may also be empirical (for a detailed analysis see Opp 2014: 170–171). Let us illustrate this with an example in which all bridge assumptions are empirical (see Fig. 3; for other variants of the micro– macro model see Opp 2011). Assume a city administration wants to build a new highway through a residential area. This is the independent macro variable. This decision activates incentives to individual protests. They bring about individual protests which prompt the government to give up the plan to build a highway. There are thus causal bridge assumptions, a causal micro theory, but a correlation on the macro level. Is the empirical micro-to-macro statement a law? There are numerous protests that do not change government decisions. The micro-to-macro relation in the figure is a singular causal statement. As in the case of the previous empirical macro-to-micro relation one could try to explain why the government followed the goal of the protesters. In this case we can again apply RCT. The government consists of individuals who respond to incentives: members of governments have the goal not to lose support. Giving in to the demands of the protesters may fulfill this goal.

6

The Macro Propositions

Although macro propositions are correlations it is often assumed that they are causal statements. This seems to be the implicit assumptions when macro propositions are formulated in a general way such as: “inequality leads to

388

K.-D. Opp

mass political violence.” Sometimes empirical research confirms these propositions because they are tested in specific situations that are in line with the propositions. For example, assume research is largely restricted to countries in which high inequality is regarded as unjust and in which strong incentives to protest prevail. The confirmation of such propositions seemingly suggests the existence of a law. But a severe test including omitted relevant variables will show that the macro proposition is at best an empirical regularity. However, there are macro laws (for details see Opp 2012, 2017). For example, there is no society that is democratic and has central planning. Those laws have a low informative content. In addition, they do not specify the conditions for their validity. For example, a democracy is not related to central planning because the institutional provisions of a democracy provide incentives to citizens not to vote for parties proposing central planning. The law is thus due to stable relations in a micro–macro model.

7

The Logical Structure of Micro–Macro Models

In micro–macro modeling macro phenomena or macro propositions are explained . To explain a proposition means to deduce it from other propositions. If a macro proposition is explained by referring to the micro level, it is deduced from the bridge assumption and the micro theory (see already the analyses in Hummell and Opp 1968, 1971). Let us illustrate this with the slightly modified inequality example which can be formalized in the following way: Law: If there are incentives, there is the relevant behavior (INC ⊃ BEH). Bridge assumption 1: If there is inequality, there are incentives (INE ⊃ INC). Bridge assumption 2: Individual behavior aggregates to collective action (BEH ⊃ CA). ––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Macro proposition (inference): If there is inequality, there is collective action (INE ⊃ CA) Symbols: ⊃ is an implication sign (if..., then...). In this example it is a material (empirical) implication. ∩ means a conjunction (and). The line symbolizes that the statement under the line can be deduced from the statements above the line.

Methodological Individualism and Micro–Macro …

389

If we write this axiomatic system in a different way, the logical structure becomes clearer: [(INE ⊃ INC) ∩ (INC ⊃ BEH) ∩ (BEH ⊃ CA)] → (INE ⊃ CA). The → is a logical implication (the statement after this sign is deduced from the statements before the sign). Micro–macro models from the research literature are much more complicated, as will be seen below. Nonetheless, the logical structure is the same: it is a logical “chain” such as A leads to B leads to C, etc.

8

Why Micro–Macro Modeling?

Assume the explanation thesis (macro propositions can be explained by considering the individual level) holds true. Why should macro propositions be explained in this way? This is the explanation postulate. The argument is that micro–macro modeling is scientifically fruitful. This means the following. 1. Micro–macro modeling leads to an increase of our knowledge about causal social processes. The previous example illustrates this. Just stating that inequality is related to mass political violence is unsatisfactory. Why are inequality and mass violence associated? Answering this question extends our knowledge: we will get information about why a certain proposition holds. In this case, we will discover intervening processes between inequality and mass political violence. It is an undisputed goal of scientists to increase our knowledge. 2. Micro–macro modeling leads to the discovery of conditions for the validity of macro propositions. As has been shown, the inequality proposition holds only if the bridge assumptions and the micro theory are true. However, the bridge assumptions are most of the time subject to change. Even if they remain constant for some time, their stability is a condition for the validity of the macro proposition. Problems of the validity of macro propositions become particularly clear if there are conflicting macro propositions. For example, research shows that increasing state repression for protest sometimes has a deterrence effect (protests decrease) and sometimes a radicalization effect (protests increase). What happens depends on the existing incentives of those who

390

K.-D. Opp

are exposed to repression (for details see Opp 2022). Again, micro–macro modeling provides new information about social processes. 3. Rejecting micro–macro modeling results in ad hoc theorizing. Those who formulate macro propositions without micro–macro modeling argue ad hoc—just following their unfounded intuitions. The reason is that there are no informative macro theories (see before). Macro propositions are at best empirical generalizations (holding in some situations without knowing the conditions for their validity). 4. Even if macro propositions can be explained by other macro propositions, these explanations are incomplete. One could suggest a macro explanation for the inequality proposition (see Fig. 4). Assume that we find high-income inequality in societies with extractive economic institutions—“extractive because such institutions are designed to extract incomes and wealth from one subset of society to benefit a different subset” (Acemoglu and Robinson 2012: 76). Assume that these institutions do not only lead to inequality but to collective action aimed at changing these institutions and the inequality as well (see the upper panel of Fig. 4). Again, one would ask why these macro relations exist. The lower panel of Fig. 4 sketches a possible explanation. The extractive institutions provide the opportunities and incentives to individuals of a certain social class which aggregate to inequality. This inequality activates individual incentives (discontent etc.) which leads to individual collective action that aggregates to collective action. The three macro variables of the upper panel may thus be related to the micro level. All this must be tested empirically. An important conclusion from the micro–macro model is that the causal relations run via the micro level and that there are only correlations at the macro level. The figure suggests how complicated micro–macro models can be. 5. Explaining macro propositions individualistically leads to a clarification of macro concepts. Any comparison of theories consists of a comparison of the meaning of concepts. In the proposition “the higher cohesion of a group, the lower is the crime rate” the concepts “cohesion” and “crime” Inequality

Institutions

Goals and beliefs

Individual behavior

Goals and beliefs

Fig. 4 Example for a macro explanation of macro propositions

Collective behavior

Individual behavior

Methodological Individualism and Micro–Macro …

391

need to be clarified. The necessity to explicate the often ambiguous macro concepts is a desirable side effect of micro–macro modeling. 6. Micro–macro modeling is a test of the applied micro theories. An empirical theory comparison implies that each of the propositions could be falsified. A wrong micro theory will not lead to a valid explanation of the macro proposition. If the micro theory is “deprivation leads to collective action” we will find that inequalities are often not related to collective action because deprivation is not sufficient to bring about collective action (as research clearly shows). Micro–macro modeling thus informs also about the validity of the applied micro theory.

9

Complications

The previous examples were simplified in several respects. The following complications are particularly important. 1. Number of levels. There may be several macro levels. For example, groups may be hierarchically ordered – such as a city (e.g. Leipzig) that is a member of a state (Saxony), which is part of a country (Germany). 2. Macro factors may have different effects. If there are different collectives and their properties at the macro level, there may be different effects on micro variables. For example, individuals may be members of different groups who provide oppositive incentives for a given behavior. 3. Macro effects depend on pre-existing individual incentives. There are normally different individual motivations that influence individual behavior. Often only some of them are influenced by macro variables. For example, an increase in repression (macro variable) may activate a norm to do something about an unjustified regime action (micro variable). The pre-existing discontent with the regime and a strong belief, that now one’s own protest participation would make a difference, might immediately lead to an explosion of protest actions. If pre-existing incentives are weak, the impact of increasing repression is weak as well. 4. Not only behavior may be a dependent micro variable. As has been said before, explananda may be attitudes, norms, goals, and beliefs. For their explanation existing social psychological theories can be applied. 5. There may be non-linear and interaction effects between macro variables and between micro and macro variables. In one of the previous examples an interaction effect was assumed. An example of a non-linear relation on the macro level is the inverted u-curve for repression: if it is low increasing

392

K.-D. Opp

repression raises collective protests up to a certain point, then protest decreases (for a summary see Opp 2022). We will not go into details. We only want to allude to the possible complications of micro–macro modeling.

10

Examples for Complex Micro–Macro Models in the Social Sciences

Micro–macro explanations are practiced in all social science disciplines. To illustrate this, I will provide a short review of some fields where micro–macro modeling is common. I hope that the following is helpful for philosophers who criticize MI without referring in detail to the work of social scientists (see the section on objections to MI).

Collective Action “Collective” action—a macro variable—means that at least two individuals jointly act to realize their goals. There are numerous kinds of collective action for which micro–macro models have been formulated (Francisco 2010; Marwell and Oliver 1993; Olson 1965; Sandler 1992). One example is the explanation of revolutions—a macro variable. Tiruneh (2014) writes that the three most important factors that trigger a revolution are economic development, regime type, and state ineffectiveness. Whatever the exact meaning of these factors is, a question is why those factors lead to revolutions. An intuitive answer is that revolutions are the outcome of individual actions (a micro variable) that have a causal effect on regime changes such as deposing a government. “Regime change” also refers to the behavior of individual actors: there are persons who resign and persons who take office. Such micro–macro modeling is common in many explanations of revolutions (e.g., Salert 1976; Taylor 1988). Revolutions are “complex emergent processes” (Goldstone 2014: 12) that involve numerous micro–macro relations. It is surprising that authors who are not suspicious of adhering to MI suggest micro–macro explanations. Engels (1886) writes that everyone consciously pursues his goals, and the outcome is history. Revolutions are such an outcome of history. Elster (1985) and Analytical Marxism (e.g. Roemer 1986) reconstruct Marx as a methodological individualist. The diffusion of innovations is another example in which existing macro factors are an important set of conditions that influence individual behavior.

Methodological Individualism and Micro–Macro …

393

For example, Becker et al. (2020) explain the diffusion of the Protestant Reformation (a kind of collective action) by macro factors such as multiplex networks that influence Luther’s students and merchants who spread the ideas of the Reformation. Other macro factors were the existing technology (such as the printing press) as well as existing shipping routes for trade.

Institutional Analysis The origin and impact of social institutions (i.e., sets of rules such as private property) are a major theme since the beginning of the social sciences. An early example is Adam Smith’s classic analysis of the impact of market institutions on individual economic behavior. This is complemented by analyses showing why centrally planned economies fail (see the summary by Brus and Laski 1989). These analyses consist of micro–macro modeling of complex social processes. The macro variables are, for example, the state as a guard for enforcing the laws such as private property rights. These provide incentives for individual action such as founding new enterprises or inventing new products. Invisible hand explanations (e.g., Ullmann-Margalit 1978) are micro–macro explanations as well. Explaining why certain rules originate spontaneously and why state actors intervene is a complex interplay of micro–macro levels (e.g., Opp 2018b). Another example of institutional analyses is the explanation of the origin, development, and impact of private property (and other macro factors) on economic growth. North and Thomas (1973) and more recently Acemoglu and Robinson (2012) explain the historical development of human societies. Max Weber’s (2005, first 1930) analysis of Protestantism and the “spirit” of capitalism can be reconstructed as a complex micro–macro model. Institutional economics (e.g., Eggertsson 2013) in general is another illustration of how complex micro–macro modeling proceeds. One of the most impressive examples of explaining the use of common pool resources is Ostrom’s work (e.g., Ostrom 1990). These analyses do not explicitly identify the structure of the micro-models. We don’t find graphs like Fig. 1. However, the complex modeling can easily be reconstructed as micro–macro models.

11

Agent-Based Modeling

To formulate models of complex micro–macro processes often requires agentbased simulations (ABS, e.g., Macy and Flache 2009; Manzo 2020). Such simulations can deal with heterogeneous individuals, interactions between

394

K.-D. Opp

different agents, and different bottom-up and top-down processes (i.e., bridge assumptions). ABS thus are ideal to set up complicated micro–macro models (e.g., Mäs and Opp 2016). There is a debate about how MI and ABS are related (Marchionni and Ylikoski 2013; Bulle and Phan 2017; di Iorio and Chen 2019; Linares 2018; Manzo 2020). I strongly agree with di Iorio and Chen’s critique of Marchionni and Ylikoski. In contrast to the latter authors ABS does not only allow a bottom-up research strategy but top-down modeling as well. Even if there are “structural assumptions” that “play an irreducible explanatory role” (Marchionni and Ylikoski 2013: 337), micro–macro explanations are possible and can be modeled with ABS. An example is the impact of an “irreducible” fact such as an earthquake on processes of collective action or the invention of the printing press and its impact on the diffusion of the Protestant Reformation. An early example that illustrates the fruitfulness of ABS is Thomas Schelling’s (1978) micro–macro segregation model. Another impressive example is simulating “growing artificial societies” (Epstein and Axtell 1996; Sawyer 2003).

12

Problems of Micro–Macro Modeling

In this section I discuss possible problems of MI that seem most serious for the version of MI advocated in this article.

Limits of the Individualistic Reconstruction of Collective Properties It is controversial whether collectives and their properties can be defined by individuals and their properties (e.g., Brodbeck 1958; Nagel 1961: 336–397). The individual properties may refer to one-place predicates (such as Ca—a has a cold) or relational predicates (such as Iab—a interacts with b). It is often difficult to judge definability because many collective concepts are ambiguous. An example is “society.” In this case the question is whether collective concepts can be reconstructed (or explicated, see Hanna 1968) as individuals and their properties. “Reconstruction” means that one proposes a more precise definition (that does not contradict the existing vague definition) before a meaning analysis is performed. The issue thus is whether collective concepts can be reconstructed by concepts about individuals. The reconstruction thesis asserts this.

Methodological Individualism and Micro–Macro …

395

This thesis can be tested: one could analyze the meanings of concepts referring to collectives and examine whether they can be defined or reconstructed as properties of individuals. For example, can “cohesion of a group” be reconstructed as properties of individuals? So far there is only one such systematic test (Hummell and Opp 1968, 1971: 35–50). The authors analyzed several central sociological concepts. They found that each concept can be reconstructed by concepts referring to individuals and their properties. Why is this relevant for micro–macro explanations? Several cases can be distinguished. (1) If bridge assumptions are analytical, this means that the collective concepts can be reconstructed individualistically at least in part. This is sufficient for an individualistic explanation. To illustrate, let a macro factor be Protestantism (as a system of beliefs and values), and let the micro factor be certain individual beliefs (such as being chosen for salvation). The relevant belief thus is, by definition, included in the Protestant belief system. There is thus an analytical macro-to-micro relation. Another example is the micro-to-macro aggregation of individual crimes to the crime rate. There are thus definitional (i.e., analytical) macro-to-macro and micro-to-macro relations so that a micro–macro explanation is possible. (2) Now let the bridge assumptions be empirical. Assume regime repressiveness (a macro property) has an empirical effect on the activation of protest norms of individuals. It is irrelevant whether the macro factor can be reconstructed individualistically. Only the causal effect of the macro factor on individual properties is relevant. Nonetheless, “regime repressiveness” is a behavior of a set of individuals and can thus be reconstructed individualistically. But this is not required for a micro–macro explanation. Another example is the collision of an asteroid with the earth as a macro event. Let this have a causal effect on incentives of individual actors to take measures to protect themselves against future collisions. An individualistic reconstruction of “collision” is not necessary for a micro–macro explanation and not possible either. The same holds for microto-macro effects. Let many individuals commit crimes that lead to an increase of the state budget for combatting crimes. Individualistic reconstruction of “state budget” is again irrelevant, although “increase of the state budget” can be individualistically reconstructed: it consists of actions of individuals. The conclusion is that the validity of the reconstruction thesis is not required for micro–macro explanations. Thus, claims that sometimes individualistic reconstructions of “wholes” are not possible and that this impedes micro–macro explanations are untenable.

396

K.-D. Opp

The Emergence of Collective Properties as an Obstacle to Micro–Macro Explanations The concept of emergence has several meanings.3 A common definition is that a whole (such as a collective) has properties that its parts do not possess. These new collective properties are denoted as emergent. For example, individuals have an age, but only collectives have a variance (or standard deviation) of age. The age variance is thus an emergent property. In statistics the variance of a property x is defined as the sum of the squared differences of individual ages from the mean, divided by the number of cases. The standard deviation is the square root of the variance. There is thus, as it is sometimes put, supervenience between the emergent and the lower-level properties. The question is to what extent the existence of emergent properties is a problem for micro–macro modeling. In the previous example a high average age (emergent from age as an individual property) might have effects on individual beliefs that it is easy to get jobs (because a high number of people are retired). In this example the emergence of the age structure means that there is an analytical bridge assumption from individual age to the age structure. But for explaining the impact of the emergent property on individual properties it is irrelevant whether the macro property is emergent. Now look at micro-to-macro relations. Income inequality is an emergent property from individual incomes. Here the emergent property is a dependent variable: differential individual efforts for earning income cause inequality of incomes. Nonetheless, as has been shown, it can be explained when individual actions bring about unequal incomes. Emergence of properties does thus not mean, as is often claimed, that collective phenomena are “unexplainable, or unpredictable, on the basis of information concerning … constituents” (Hempel and Oppenheim 1948: 147). The previous analysis suggests that MI is capable of modeling emergence. This is explicitly the goal of what Sawyer (e.g., 2005: 2) calls the third wave of systems theory which is concerned with emergence: “the processes whereby the global behavior of a system results from the actions and interactions of agents.”

3 About emergence in general and in the social sciences see in particular Bedau and Humphreys 2008; Hempel and Oppenheim 1948: 146–152; Hummell and Opp 1971; Kincaid 2016; Nagel 1961: 366–397; Sawyer 2005: 63–99.

Methodological Individualism and Micro–Macro …

13

397

Reductionism and Methodological Individualism

Labeling a theory or a theoretical approach as reductionist is often considered a devastating critique. This is not surprising if one looks at different meanings of this term.4 For example, for Ylikoski (2012: 24) reductionist explanations refer to the micro level and replace or eliminate the higher-level facts. Kincaid (2016: 113) writes that reductionism is the “idea that theories about social entities can be translated into theories that are only about individual behavior.” None of these theses holds for MI as it is outlined in this article. Nothing is “replaced” or “eliminated” (Ylikoski)—the goal is to explain macro propositions (as has been emphasized before). Kincaid’s characterization is flawed because the question is not about “translation” but about explanation. Such misrepresentations of MI explain the widely shared anti-reductionist positions in sociology. The claim that “sociology” should be “reduced” to “psychology” (e.g., Hummell and Opp 1968, 1971) meant that sociological macro propositions could and should be explained individualistically. The issue is not to eliminate a discipline but explaining its propositions. “Reduction” should therefore be replaced by “explanation”—which has been repeatedly emphasized already by Homans (e.g., 1967), a pioneer of MI in sociology.

14

The Program of Methodological Individualism: A Proposal

It seems that every philosopher of science creates his or her own version of MI. For example, according to Bhaskar (1978: 5) MI asserts that “facts about society and social phenomena are to be explained solely in terms of facts about individuals.” Bridge assumptions apparently do not exist in this version. Elder-Vass (2015: 85) asserts—without giving any references: “Methodological individualists question whether groups and other social entities can have causal significance at all.” Obviously, this is wrong: there are causal micro-to-macro and macro-to-micro relations, as has been shown before.

4 Some meanings are listed in Kincaid 1986: 493. About reductionism in general and in the social sciences see Hummell and Opp 1968, 1971; Kincaid 1986, 2016; Malewski 1964; Nagel 1961: 336–366; Nowak 1970; Sawyer 2005; Wagner 2017.

398

K.-D. Opp

MI is a research program that is pursued by numerous social scientists. Most of the time micro–macro models are not spelled out in detail (e.g., whether bridge assumptions are empirical or analytical). But the arguments can easily be reconstructed as micro–macro modeling. From my knowledge of the social science literature, I believe that the following theses are shared by practicing social scientists who are interested in rigorous individualistic explanations of macro propositions. (1) Macro propositions (or single macro phenomena) can be explained by theories about individual behavior and bridge assumptions. The latter can be empirical or analytical. The empirical bridge assumptions can further be validated by applying theories or by the methods of empirical research (causal analysis). This is the explanation thesis of MI. (2) On the micro level confirmed theories about individuals (referring also to relations of individuals) should be applied. This is preferable to applying unconfirmed ad hoc causal statements. (3) On the micro level, general behavioral theories should be applied and empirically compared with theories of the middle range. (4) Micro-macro explanations lead to scientific progress: they enhance our knowledge about social processes and are therefore desirable. This is the explanation postulate.

I hope that this formulation of MI is helpful to social scientists as a guideline for outlining in a clearer way what their bridge assumptions and micro theories are.

15

Conclusion and Discussion

Our general conclusion is that MI and micro–macro modeling, as described before, are a fruitful explanatory strategy. The social science examples presented before support this claim. Reading the vast philosophical literature about MI, emergence, and reductionism is extremely frustrating for a practicing empirical-theoretical social scientist (as the author of this article). It is surprising that there is most of the time no detailed reference to the existing social science literature. This can be exemplified by Epstein’s (2015) critique of MI. Sugden (2016: 1378) criticizes correctly his “abstract philosophical reasoning that takes no notice of existing theory or evidence” and that “expresses a breathtaking disdain for 250 years of work in the discipline [of economics – KDO]” (see also di Iorio and Herfeld 2017: 17 for a similar critique). Another extreme example for

Methodological Individualism and Micro–Macro …

399

ignoring existing social science is Bhaskar’s (1978: 6) statement that “sociology is not concerned, as such, with large-scale, mass or group behaviour.” Mass-protests, crime rates, riots, revolutions, etc., are some counter examples sociologists are dealing with. This situation suggests two methodological rules for further discussions of MI: (1) If you criticize MI, precisely specify the version your critique is directed at. (2) Show in detail the extent to which the version attacked is advanced by practicing social scientists. For limitations of space, it was not possible to address philosophical reflections referring, for example, to ontology, constructivism, postmodernism, critical rationalism, pragmatism, and critical realism that address MI. The position taken here is that of an analytical social science that is committed to rigor, to explanations of social phenomena by applying confirmed general theoretical hypotheses, to severe empirical tests by applying the methods of empirical research and multivariate analyses, and to empirical theory comparison. My position concurs with a rigorous modern philosophy of science, based on analytic philosophy (similar to Schurz 2014) and critical rationalism (Popper 1968; Albert 1985).

References Acemoglu, D., & Robinson, J. A. (2012). Why Nations Fail. The Origins of Power, Prosperity, and Poverty. New York: Random House/Crown Publishers. Albert, H. (1985). Treatise on Critical Reason. Princeton: Princeton University Press. Alexander, J. C., Giesen, B., Münch, R., & Smelser, N. J. (Eds.). (1987). The MicroMacro Link. Berkeley: University of California Press. Bandura, A. (1999). Social Cognitive Theory: An Agentic Perspective. Asian Journal of Social Psychology, 2(1), 21–41. Becker, G. S. (1976). The Economic Approach to Human Behavior. Chicago and London: Chicago University Press. Becker, S. O., Hsiao, Y., Pfaff, S., & Rubin, J. (2020). Multiplex Network Ties and the Spatial Diffusion of Radical Innovations: Martin Luther’s Leadership in the Early Reformation. American Sociological Review, 85 (5), 857–894. Bedau, M. A., & Humphreys, P. (Eds.). (2008). Emergence: Contemporary Readings in Philosophy and Science. Cambridge, MA, and London, UK: MIT Press. Bhaskar, R. (1978). On the Possibility of Social Scientific Knowledge and the Limits of Naturalism. Journal for the Theory of Social Behaviour, 8(1), 1–28. Brodbeck, M. (1958). Methodological Individualisms: Definition and Reduction. Philosophy of Science, 25 (1), 1–22. Brus, W., & Laski, K. (1989). From Marx to the Market. Socialism in Search of an Economic System. Oxford: Clarendon Press.

400

K.-D. Opp

Bulle, N., & Phan, D. (2017). Can Analytical Sociology Do without Methodological Individualism? Philosophy of the Social Sciences, 47 (6), 379–409. Buskens, V., Raub, W., & Van Assen, M. A. L. M. (Eds.). (2012). Micro-Macro Links and Microfoundations in Sociology. London and New York: Routledge. Coleman, J. S. (1986). Social Theory, Social Research, and a Theory of Action. American Journal of Sociology, 91(6), 1309–1335. Coleman, J. S. (1990). Foundations of Social Theory. Cambridge, MA and London: Belknap Press of Harvard University Press. Di Iorio, F. (2015). Cognitive Autonomy and Methodological Individualism. Heidelberg: Springer. Di Iorio, F. (2016). Introduction: Methodological Individualism, Structural Constraints, and Social Complexity. Cosmos + Taxis, 3(2 and 3), 1–8. Di Iorio, F., & Chen, S.-H. (2019). On the Connection Between Agent-Based Simulation and Methodological Individualism. Social Science Information, 58(2), 354-376. Di Iorio, F., & Herfeld, C. (2017). Book Review: The Ant Trap: Rebuilding the Foundations of the Social Sciences, by Brian Epstein. Philosophy of the Social Sciences, 48(1), 105–128. Diekmann, A. (2022). Rational Choice Theory. Heuristic Potential, Applications and Limitations. In K. Gërxhani, N. D. de Graaf & W. Raub (Eds.), Handbook of Sociological Science: Contributions to Rigorous Sociology (pp. 100–119). Cheltenham: Edward Elgar. Diekmann, A., & Opp, K.-D. (1979). Anomie und Prozesse der Kriminalitätsentwicklung im sozialen Kontext. Vorschläge für die Weiterentwicklung und Formalisierung der Anomietheorie. Zeitschrift für Soziologie, 30 (4), 330–343. Eggertsson, T. (2013). Quick Guide to New Institutional Economics. Journal of Comparative Economics, 41(1), 1–5. Elder-Vass, D. (2015). Developing Social Theory Using Critical Realism. Journal of Critical Realism, 14 (1), 80–92. Elster, J. (1985). Making Sense of Marx. Cambridge: Cambridge University Press. Epstein, B. (2015). The Ant Trap: Rebuilding the Foundations of the Social Sciences. New York: Oxford University Press. Epstein, J. M., & Axtell, R. (1996). Growing Artificial Societies: Social Science from the Bottom Up. Cambridge, MA: MIT Press. Feather, N. T. (1990). Bridging the Gap between Values and Actions: Recent Applications of the Expectancy-Value Model. In E. T. Higgins & R. M. Sorrentino (Eds.), Handbook of Motivation and Cognition: Foundations of Social Behavior (Vol. 2, pp. 151–192). New York: Guilford Press. Ferreira, I. A., Gisselquist, R. M., & Tarp, F. (2022). On the Impact of Inequality on Growth, Human Development, and Governance. International Studies Review, 24 (1), 1–28. Fishbein, M., & Ajzen, I. (2010). Predicting and Changing Behavior: The Reasoned Action Approach. New York and Hove: Psychology Press.

Methodological Individualism and Micro–Macro …

401

Francisco, R. A. (2010). Collective Action Theory and Empirical Evidence. New York: Springer. Gawronski, B., & Creighton, L. A. (2013). Dual Process Theories. In D. E. Carlston (Ed.), The Oxford Handbook of Social Cognition (pp. 282–312). Oxford and New York: Oxford University Press. Gintis, H. (2006). The Foundations of Behavior: The Beliefs, Preferences, and Constraints Model. Biological Theory, 1(2), 123–127. Goldstone, J. (2014). Revolutions: A Very Short Introduction. Oxford: Oxford University Press. Hanna, J. F. (1968). An Explication of ‘Explication’. Philosophy of Science, 35 (1), 28-44. Harmon-Jones, E., & Mills, J. (2019). An Introduction to Cognitive Dissonance Theory and an Overview of Current Perspectives on the Theory. In E. HarmonJones (Ed.), Cognitive Dissonance: Reexamining a Pivotal Theory in Psychology (2nd ed., pp. 3–24). Washington, DC, US: American Psychological Association. Hedström, P. (2005). Dissecting the Social: On the Principles of Analytical Sociology. Cambridge: Cambridge University Press. Hedström, P. (2021). Coda—The Past and Future of Analytical Sociology. In G. Manzo (Ed.), Research Handbook on Analytical Sociology (pp. 490–505). Cheltenham: Edward Elgar. Hedström, P., & Swedberg, R. (1996). Rational Choice, Empirical Research, and the Sociological Tradition. European Sociological Review, 12(2), 127–146. Hedström, P., & Ylikoski, P. (2014). Analytical Sociology and Rational-Choice Theory. In G. Manzo (Ed.), Analytical Sociology: Actions and Networks (pp. 57– 70). Chichester, UK: Wiley. Hempel, C. G., & Oppenheim, P. (1948). Studies in the Logic of Explanation. Philosophy of Science, 15 (2), 135–175. Hernes, G. (1976). Structural Change in Social Processes. American Journal of Sociology, 82(3), 513–547. Hodgson, G. M. (2007). Meanings of Methodological Individualism. Journal of Economic Methodology, 14 (2), 211–226. Homans, G. C. (1967). The Nature of Social Science. New York: Harcourt, Brace & World. Homans, G. C. (1974). Social Behavior: Its Elementary Forms (2nd ed.). New York: Harcourt, Brace & World. Hummell, H. J. (1973). Methodologischer Individualismus, Struktureffekte und Systemkonsequenzen. In K.-D. Opp & H. J. Hummell (Eds.), Soziales Verhalten und soziale Systeme (pp. 61–134). Frankfurt am Main: Athenäum. Hummell, H. J., & Opp, K.-D. (1968). Sociology Without Sociology—The Reduction of Sociology to Psychology: A Program, a Test, and the Theoretical Relevance. Inquiry, 11(2), 205–226. Hummell, H. J., & Opp, K.-D. (1969). On the Relation Between Theory and Research in Sociology. Quality & Quantity, 3(1–2), 23–61.

402

K.-D. Opp

Hummell, H. J., & Opp, K.-D. (1971). Die Reduzierbarkeit von Soziologie auf Psychologie. Eine These, ihr Test und ihre theoretische Bedeutung. Braunschweig: Vieweg. Jasso, G. (2010). Linking Individuals and Societies. The Journal of Mathematical Sociology, 34 (1), 1–51. Kincaid, H. (1986). Reduction, Explanation, and Individualism. Philosophy of Science, 53(4), 492–513. Kincaid, H. (2016). Reductionism. In L. McIntyre & A. Rosenberg (Eds.), The Routledge Companion to Philosophy of Social Science (pp. 133–143). Routledge. Kirchgässner, G. (2008). Homo Oeconomicus: The Economic Model of Behaviour and Its Applications in Economics and Other Social Sciences. New York: Springer. Knauff, M., & Spohn, W. (Eds.). (2021). The Handbook of Rationality. Cambridge, MA: The MIT Press. Lazarsfeld, P. F., & Menzel, H. (1961). On the Relation Between Individual and Collective Properties. In A. Etzioni (Ed.), Complex Organizations: A Sociological Reader (pp. 422–440). New York: Holt, Rinehart & Winston. Linares, F. (2018). Agent Based Models and the Science of Unintended Consequences of Social Action. Revista Española de Investigaciones Sociológicas, 162, 21–38. Lipset, S. M., Martin, T., & Coleman, J. S. (1956). Union Democracy. The Internal Politics of the International Typographical Union. Glencoe, IL: Free Press. Macy, M., & Flache, A. (2009). Social Dynamics from the Bottom Up: Agent-Based Models of Social Interaction. In P. Hedström & P. Bearman (Eds.), The Oxford Handbook of Analytical Sociology (pp. 245–268). Oxford: Oxford University Press. Malewski, A. (1964). Two Models of Sociology. In H. Albert (Ed.), Theorie und Realität. Ausgewählte Aufsätze zur Wissenschaftslehre der Sozialwissenschaften (pp. 103–115). Tübingen: Mohr. Malewski, A. (1967). Verhalten und Interaktion. Die Theorie des Verhaltens und das Problem der sozialwissenschaftlichen Integration. Tübingen: Mohr. Manzo, G. (2020). Agent-Based Models and Methodological Individualism: Are They Fundamentally Linked? L’Année sociologique, 70 (1), 197–229. Marchionni, C., & Ylikoski, P. (2013). Generative Explanation and Individualism in Agent-Based Simulation. Philosophy of the Social Sciences, 43(3), 323–340. Marwell, G., & Oliver, P. (1993). The Critical Mass in Collective Action: A MicroSocial Theory. New York: Cambridge University Press. Mäs, M., & Opp, K.-D. (2016). When Is Ignorance Bliss? Disclosing True Information and Cascades of Norm Violation in Networks. Social Networks, 47 (1), 116–129. Mathieu, J. E., Aguinis, H., Culpepper, S. A., & Chen, G. (2012). Understanding and Estimating the Power to Detect Cross-Level Interaction Effects in Multilevel Modeling. Journal of applied psychology, 97 (5), 951–967. McClelland, D. C. (1961). The Achieving Society. New York: The Free Press.

Methodological Individualism and Micro–Macro …

403

Nagel, E. (1961). The Structure of Science: Problems in the Logic of Scientific Explanation. London: Routledge & Kegan Paul. Nagengast, B., Marsh, H. W., Scalas, L. F., Xu, M. K., Hau, K.-T., & Trautwein, U. (2011). Who Took the “×” Out of Expectancy-Value Theory? A Psychological Mystery, a Substantive-Methodological Synergy, and a Cross-National Generalization. Psychological Science, 22(8), 1058–1066. North, D. C., & Thomas, R. P. (1973). The Rise of the Western World: A New Economic History. Cambridge: Cambridge University Press. Nowak, S. (1970). The Logic of Reductive Systematizations of Social and Behavioural Theories. The Polish Sociological Bulletin, 22, 79–102. Olson, M. (1965). The Logic of Collective Action. Cambridge, MA: Harvard University Press. Opp, K.-D. (1970). Theories of the Middle Range as a Strategy for the Construction of a General Sociological Theory: A Critique of a Sociological Dogma. Quality and Quantity, 4 (2), 243–253. Opp, K.-D. (1999). Contending Conceptions of the Theory of Rational Action. Journal of Theoretical Politics, 11(2), 171–202. Opp, K.-D. (2009). Theories of Political Protest and Social Movements: A Multidisciplinary Introduction, Critique and Synthesis. London and New York: Routledge Opp, K.-D. (2011). Modeling Micro-Macro Relationships: Problems and Solutions. Journal of Mathematical Sociology, 35 (1–3), 209–234. Opp, K.-D. (2012). Can There Be Causal Effects on the Macro Level? Sociologica, 1, 1–7. https://www.rivisteweb.it/issn/1971-8853/issue/100. Opp, K.-D. (2013). What Is Analytical Sociology? Strengths and Weaknesses of a New Sociological Research Program. Social Science Information, 52(3), 329–360. Opp, K.-D. (2014). Das Aggregationsproblem bei Mikro-Makro-Erklärungen. In J. Friedrichs & A. Nonnenmacher (Eds.), Soziale Kontexte und Soziale Mechanismen. Special Issue 54 of the Kölner Zeitschrift für Soziologie und Sozialpsychologie (pp. 155–188). Wiesbaden: Springer VS. Opp, K.-D. (2017). Warum gibt es keine brauchbaren Makrogesetze in den Sozialwissenschaften? In M. Wagner (Ed.), Die Provokation der Reduktion. Beiträge zur Wissenschaftstheorie der Soziologie (pp. 69–96). Wiesbaden: Otto Harrassowitz. Opp, K.-D. (2018a). Do the Social Sciences Need the Concept of “Rationality”? Notes on the Obsession with a Concept. In F. Di Iorio & G. Bronner (Eds.), The Mystery of Rationality: Mind, Beliefs, and the Social Sciences (pp. 191–217). Wiesbaden: VS Springer. Opp, K.-D. (2018b). The Interdependence of Spontaneous Order and Institutional Design. Table Manners, Language, Daylight Saving Time, and the Erosion of Institutional Design under Communist Rule. In A. Mica, K. M. Wyrzykowska, R. Wi´sniewski & I. Zieli´nska (Eds.), Sociology and the Invisible Hand (pp. 197– 228). Frankfurt: Lang. Opp, K.-D. (2020a). Rational Choice Theory and Methodological Individualism. In P. Kivisto (Ed.), The Cambridge Handbook of Social Theory (pp. 1–23). Cambridge: Cambridge University Press.

404

K.-D. Opp

Opp, K.-D. (2020b). Analytical Criminology: Integrating Explanations of Crime and Deviant Behavior. London and New York: Routledge. Opp, K.-D. (2022). Advanced Introduction to Social Movements and Political Protests. Cheltenham Edward Elgar. Osborne, D., García-Sánchez, E., & Sibley, C. G. (2019). Identifying the Psychological Mechanisms Underlying the Effects of Inequality on Society: The MacroMicro Model of Inequality and Relative Deprivation (MIRED). In J. Jetten & K. Peters (Eds.), The Social Psychology of Inequality (pp. 275–293). Wiesbaden: Springer. Ostrom, E. (1990). Governing the Commons: The Evolution of Institutions for Collective Action. Cambridge: Cambridge University Press. Parri, L. (2014). Explanation in the Social Sciences: A Theoretical and Empirical Introduction. Soveria Mannelli: Rubbettino. Popper, K. R. (1968 (3rd ed.)). Conjectures and Refutations: The Growth of Scientific Knowledge. New York: Routledge. Raub, W. (2021a). Rational Choice Theory in the Social Sciences. In M. Knauff & W. Spohn (Eds.), The Handbook of Rationality (pp. 611–623). Cambridge, MA: MIT Press. Raub, W. (2021b). Analytical Sociology and Its Lesser-Known Antecedents: Structural Individualism in European Sociology. In G. Manzo (Ed.), Research Handbook on Analytical Sociology (pp. 80–99). Cheltenham: Edward Elgar. Roemer, J. (Ed.). (1986). Analytical Marxism. Cambridge: Cambridge University Press. Salert, B. (1976). Revolutions and Revolutionaries: Four Theories. New York: Elsevier. Sandler, T. (1992). Collective Action: Theory and Applications. Ann Arbor: Michigan University Press. Sawyer, R. K. (2003). Artificial Societies: Multi Agent Systems and the Micro-Macro Link in Sociological Theories. Sociological Methods and Research, 31, 37–75. Sawyer, R. K. (2005). Social Emergence: Societies as Complex Systems. Cambridge: Cambridge University Press. Schelling, T. C. (1978). Micromotives and Macrobehavior. New York & London: W.W. Morton and Company. Schurz, G. (2014). Philosophy of Science: A Unified Approach New York: Routledge. Sugden, R. (2016). Ontology, Methodological Individualism, and the Foundations of the Social Sciences. Journal of Economic Literature, 54 (4), 1377–1389. Taylor, M. (Ed.). (1988). Rationality and Revolution. Cambridge: Cambridge University Press. Thaler, R. (2015). Misbehaving. The Making of Behavioral Economics. New York: Norton. Tiruneh, G. (2014). Social Revolutions: Their Causes, Patterns, and Phases. SAGE Open, 4 (3), 1–12. Tyler, T. R. (2013). Why People Cooperate: The Role of Social Motivations. Princeton: Princeton University Press.

Methodological Individualism and Micro–Macro …

405

Udéhn, L. (2002). The Changing Face of Methodological Individualism. Annual Review of Sociology, 28, 479–507. Ullmann-Margalit, E. (1978). Invisible-Hand Explanations. Synthese, 39, 263–291. van de Rijt, A. (2011). The Micro-Macro Link for the Theory of Structural Balance. Journal of Mathematical Sociology, 35 (1 to 3), 94–113. von Wright, G. H. (1971). Explanation and Understanding. London: Routledge. Wagner, G. (Ed.). (2017). Die Provokation der Reduktion. Beiträge zur Wissenschaftstheorie der Soziologie. Wiesbaden: Harrassowitz. Weber, M. (2005, first 1930). The Protestant Ethic and the Spirit of Capitalism (translated by Talcott Parsons). London and New York: Routledge. Wigfield, A., Tonks, S., & Klauda, S. L. (2016). Expectancy-Value Theory. In K. R. Wentzel & D. B. Miele (Eds.), Handbook of Motivation at School (pp. 55–75). London and New York: Routledge. Ylikoski, P. (2012). Micro, Macro, and Mechanisms. In H. Kincaid (Ed.), The Oxford Handbook of Philosophy of the Social Sciences (pp. 21–45). Oxford: Oxford University Press. Ylikoski, P. (2016). Methodological Individualism. In L. McIntyre & A. Rosenberg (Eds.), The Routledge Companion to Philosophy of Social Science (pp. 135–146). New York: Routledge. Zahle, J., & Kincaid, H. (2019). Why Be a Methodological Individualist? Synthese, 196 (2), 655–675. Zahle, J., & Collin, F. (Eds.). (2014). Rethinking the Individualism-Holism Debate. Essays in the Philosophy of Social Science. New York: Springer.

Complex Methodological Individualism Jean Petitot

1

Introduction

This chapter contextualizes the question of Methodological Individualism (MI) in the social sciences into a broader and more general scientific context including physical, biological, and ethological sciences. This approach closely links MI with the sciences of complexity, hence the denomination of complex methodological individualism (CMI). Many other chapters of the Handbook deal with the strictly social, political, economic, cultural, and institutional aspects of MI, as well as with the difficult epistemological problems they raise in the humanities. The focus here is on the contrary on the intrinsically transdisciplinary and transversal character of CMI. CMI concerns here the emergence of global collective properties (structures, organizations, processes) at the macroscopic level in populations composed at the microscopic level of a very large number of elementary individuals interacting with each other.

J. Petitot (B) Mathematics, École des Hautes Études en Sciences Sociales, Paris, France e-mail: [email protected]

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 N. Bulle and F. Di Iorio (eds.), The Palgrave Handbook of Methodological Individualism, https://doi.org/10.1007/978-3-031-41512-8_19

407

408

2

J. Petitot

Levels of Reality

In many domains, we come across several levels of description, and at least a micro-level and a macro-level (and often also an intermediary meso-level). At the micro-level, there are individual entities endowed with elementary individual states and rules of behavior and interacting through short range very local interactions. In contrast, at the macro-level we observe global structures which can be adequately described using concepts whose content has nothing to do with micro entities. One of the main challenges is to understand conceptually, causally, and moreover mathematically, the relations between the two levels. It is a true challenge because any sort of semantic reductionism is inapplicable since the concepts used for describing the two levels don’t share any content. There is therefore an antinomy between the thesis of independence and the antithesis of the non-independence between levels. The problem is age-old.1 It is extremely difficult and transversal to the classical disciplinary fields and in particular to the traditional division between the natural and human sciences. Let us begin with two examples from physics where it has been solved. The first, and most celebrated, example is provided by thermodynamics. To describe macroscopically a substance like water, one uses concepts such as temperature T , pressure P, and volume V , which do not refer directly to any atomic or molecular concept (no semantic reductionism).2 It took a long time to find the equations correlating these thermodynamical variables and to describe accurately such striking phenomena, called critical phenomena, as phase transitions (water boils at the critical value of 100° Celsius under a pressure of 1 atmosphere).3 But after the molecular structure of water (H2 O) was discovered in the late eighteenth century,4 it was natural to ask for a microlevel interpretation of thermodynamical variables in terms of local behavior (vibrations) and interactions (impacts, collisions) between molecules. Already Lavoisier had introduced the hypothesis that heat was due to some molecular agitation. But it required the genius of Ludwig Boltzmann (1872) and Willard Gibbs (1902) to achieve this revolutionary reinterpretation of thermodynamics in terms of statistical physics. The interpretation of T as the mean kinetic energy of molecules is one of the major achievements in the 1

It dates back to Antiquity. T and P are intensive magnitudes, while V is an extensive magnitude. 3 At the molecular level, a liquid to gas phase transition is an incredibly dramatic revolution: all the links giving to the liquid its cohesiveness are suddenly broken and the collective state becomes a dispersed state of “atomic” independent units. 4 Remember the chemical revolution resulting from the celebrated controversy on Oxygen and the atomic composition of water between Joseph Priestley and Antoine Laurent de Lavoisier (1779). 2

Complex Methodological Individualism

409

history of science. The micro-dynamics is incredibly complex, but some simple macro “mean field” “order parameters” emerge. Another example, more recent, is provided by the geometry and the morphological dynamics of sand dunes and sandpiles. Here, the micro elementary entities are grains of sand locally interacting by rolling against each other. The micro-dynamics is very complicated and the physics of dry cohesionless granular materials is a very active discipline. The macro geometric shape emerges from a fractal entanglement of myriads of micro avalanches of different scales, and geometric macro parameters such as the slope of the dune or pile (called the “angle of repose”) can be explicitly computed using non-linear equations. It is an example of a complex spontaneous order . The remarkable fact is that this geometric slope is a critical value and the complex multiscale system of sand grains has the extraordinary property of stabilizing on its critical state, while criticality seems to be the opposite of stability. This phenomenon, called “self-organized criticality,” was discovered in 1987 by Per Bak, Kurt Wiesenfeld, and Chao Tang. Once again, there is no semantic reductionism: the geometrical concepts used to adequately describe the morphology of dunes have nothing to do with fractal micro-avalanches of rolling sand grains. These two purely physical examples share some epistemological features: 1. The two micro/macro levels are phenomenologically heterogeneous and conceptually autonomous. Their relation is a relation of emergence. 2. But they are not completely causally autonomous. The macro-level “supervenes” on the micro-level. Moreover, causality operates bottom-up from the micro to the macro level. There is no direct top-down physical causality of the slope on the grains but when the wind is added as an “external field,” as is the case for dunes in the desert, then the morphology of the dune produces on the air flow all the well-known phenomena of overpressure-front and rear depression and turbulence. It therefore acts top-down on the granular micro-dynamics and generates complicated morphologies. In life sciences or social sciences top-down causalities become omnipresent and essential. 3. The heterogeneity of the two levels results from the immense number of elementary micro units. The local interactions between these units must be iterated (in a sequential or parallel mode) and the iteration of elementary rules can reach any possible complexity (it is a theorem on networks of automata). 4. The mathematical modeling of the relation of emergence between the two levels overcomes their conceptual incommensurability.

410

J. Petitot

5. An emerging macro structure can be rather simple (a temperature, a slope). But its emerging simplicity encompasses an immeasurable micro complexity. 6. In physical situations, the system is frequently modeled by a large system of differential equations and the emerging properties are represented by order parameters which have often a statistical content (as in “mean field” theories). 7. The conceptual independence of the emerging structures is mathematically justified by proving that the order parameters are essentially independent of the fine-grained underlying microphysics. For instance, in statistical physics, Kenneth Wilson won the Nobel Prize in 1982 for proving that critical behaviors in magnetic materials can be classified, via what is called the renormalization group, in universal classes largely independent of the atomic-molecular structures of the substrates. Without such a proof of invariance, emergence would have no scientific content. Thus, CMI is justified as a computational synthesis which posits that highlevel phenomena, structures, and processes can be deduced, as far as their scientific explanation is concerned, from underlying lower-level phenomena, structures, and processes. We have just seen that it is by no means an “eliminativist” reductionism and is perfectly compatible with emergence, “supervenience” or “functionalism.” Functionalism means that macro structures having a functional role can exist only if they are materially implemented in an underlying material substrate, but are at the same time, as functionally meaningful structures, largely independent of the fine grained material structure they are implemented in.5 Beside physical examples, another paradigmatic example is the opposition software/hardware in computer sciences (see philosophers like Hilary Putnam, Jerry Fodor, Zenon Pylyshyn, etc.). Complex systems having different levels of reality whose emergence at different scales can be proved are countless in nature: critical phenomena, percolation, self-organized criticality, reaction–diffusion systems, far from equilibrium dissipative structures, turbulence, cellular automata, neural networks, ant colonies, swarms, stock markets, etc. The proven emerging levels are not epiphenomenal. But they are not interpretable according to a holist realism positing their irreducible reality. They are causally reducible but not semantically reducible.

5

See e.g. Petitot, 2010.

Complex Methodological Individualism

411

This key point has been also emphasized by Viktor Vanberg in the social sciences concerning the “invisible-hand” explanations we will comment on later6 : conceptual descriptive adequacy is by no means sufficient to justify any emergence thesis. Emergence must be proved using mathematical models, that is a computational synthesis. Vanberg criticized for instance Hayek, because the latter recognized correctly the “synthetic” and “compositive” character of emergence, but did not insist on the obligation that synthesis must be carried out. It is the most difficult challenge. Remark. In this chapter we will make many references to Hayek because we consider him an exemplary representative of the CMI. Indeed, Hayek has since his first reflections on the Sensory Order constantly developed MI within the general framework of the theory of complex systems.

3

“From Physics to Politics”

The scientific study of multiscale self-organized complex systems is a wellestablished interdisciplinary field which goes far beyond physics. In 1981, Paul Dumouchel and Jean-Pierre Dupuy organized an important Conference entitled “Self-organization: from Physics to Politics.” So, let us progress toward politics through biology, psychology, and ethology.

Neural Functional Architectures A fascinating micro/macro example in life sciences is the relation between neurology and psychology. Activities of billions of elementary micro-neurons locally connected through specific systems of connections called “functional architectures,” implement a macro-level of emerging psychological states and processes. Results depend today upon revolutionary new techniques of in vivo optical imaging. Emergence is proved looking at huge systems of non-linear differential equations expressing how neurons fire and emit spikes when they are (i) activated by external stimuli and (ii) connected through inhibitory and excitatory connections having specific weights and functional architectures. This specificity results from experience and learning. It encodes in a radically distributed and non-conceptual way the knowledge and the cognitive resources of the system. Such equations, due to Jack Cowan, Hugh Wilson, and David Hopfield, are very similar to those found in statistical physics in 6

See Vanberg, 1986.

412

J. Petitot

spin glasses theories. Their numerical study is necessary to prove emergence by a computational synthesis. It is today the issue of some of major international research programs.7 Emerging properties are in particular properties of synchronization of pools of neurons. They explain many macro-psychological facts because the common phase of a synchronized population can act as a label for further processing (what is called “binding” and “labeling hypothesis”). The idea of interpreting mathematically psychological structures in terms of complex neural networks goes back to 1940–1950 with John von Neumann, Norbert Wiener, Warren McCulloch, Walter Pitts, and the famous Macy Conferences (1942–1953) at the origin of cybernetics, cognitive sciences, and information sciences.8 Since then, incredible progress has been made, e.g. concerning the explanation of categorization or learning. In fact, as acknowledged by such great neuroscientists as Gerald Edelman (Nobel 1972) and Joaquin Fuster, the great precursor was Friedrich von Hayek in The Sensory Order, a unique masterpiece analyzed by Bruce Caldwell, Barry Smith, and Francesco Di Iorio.9 It goes back to the 1920s and was taken up in the 1952 book. This critique of behaviorism is essential in Hayek’s intellectual journey because it is a key example of a complex spontaneous order : neural cells are connected into complex, organized, and specific networks which act as communicational infrastructures for complex fluxes of neural spikes (“impulses” said Hayek) and encode learning and memory. We have to take this micro neural organization as the basis for explaining our macro psychological sensory experiences. The specificity of what are now called “neural functional architectures” introduces a gap between, on the one hand, the sensorial micro transduction of the stimuli by peripheral receptor organs such as the retina; and, on the other hand, the perceived scenes endowed with their Gestalt patterns processed by the central cortical brain areas. In The Sensory Order After 25 Years, Hayek’s summary is very similar to the contemporary concept of functional architecture: “Mind thus becomes to me a continuous stream of impulses, the significance of each and every contribution of which is determined by the place in the pattern of channels through which they flow within the pattern of all available channels.” (p. 291)

7 For example, the billion euros Human Brain Project, which aims to simulate cortical modules of the visual cortex. It uses a computational power up to a million teraflops (a teraflop is 1000 billions operations per second). 8 See Jean-Pierre Dupuy’s book The Mechanization of Mind . 9 They all emphasize the importance of this early contribution of Hayek to the neurophysiology of psychology.

Complex Methodological Individualism

413

These initial reflections were a model for the further works of Hayek in economics and, after he had come into contact in the 1950s with the transdisciplinary and polymathematical sciences of complex self-organized systems (Norbert Wiener, Ludwig von Bertalanffy, John von Neumann, Warren Weaver, see above), he wrote his 1952 book. In the Hayek archives managed by Bruce Caldwell at the Hoover Institution of Stanford, there is interesting correspondence with James Gibson, the future author of The Ecological Approach to Visual Perception (1979), who was “much impressed” by The Sensory Order (24 February 1954) and wanted to invite Hayek to Cornell for a Conference on the “Fundamental Problems of Perception.” In the letters there are interesting remarks concerning Heinrich Klüver (1897–1979), the great neuropsychologist at Chicago and a privileged interlocutor of Hayek in psychology and neurology. Klüver was a student of Max Wertheimer and introduced Gestalt theory to the US. He arrived in Chicago in 1928 and joined the “Neurology Club” (Karl Lashley, Percival Bailey, A. Earl Walker, Ralph Gerard, Stephen Polyak, Charles Judson Herrick, and Roy Grinker). His works on vision are now well modeled using Wilson-Cowan equations and tools of Neurogeometry.10

Ethology Other fundamental examples in life sciences are found in ethology. In the last decades, a great deal of research and modeling has been dedicated to the collective behavior of large communities of animals: hordes, flocks of birds, fish banks, insect swarms, etc. The case of starling flocks is spectacular and well known. Very complex global geometric patterns can result from very simple local rules, as simple as (i) move in the direction of the neighbors, (ii) move to the center of the group, and (iii) if another animal is coming too close, move away from it. What is particularly interesting in these collective movements is that individuals possess internal states controlling their external motions. They are agents endowed with sensory-motor faculties and cognitive resources (even if limited). Since they are active and dissipate energy to act, they constitute thermodynamically far from equilibrium systems and can therefore structure themselves by violating the second law of thermodynamics. The emergent global geometry of their collective motions is totally unplanned and unpredictable. Transcending the cognitive resources of individuals, it is an unintentional global macro consequence of intentional, local, 10

See Petitot, 2008.

414

J. Petitot

and micro actions. It therefore already represents a social phenomenon which is a spontaneous order completely transcending individual capacities. The three rules of (i) local alignment, (ii) group attraction, and (iii) short distance repulsion were implemented in particular by Tamás Wicsek and colleagues in 1995 in a simplified statistical model which shares many features with the spin glasses models sketched above. But the models went far beyond the modeling of mere collective motions. They progressively led to the key idea of multi-agent collective, distributed, and decentralized intelligence as a new paradigm for complex problem solving, more precisely for solving problems that individual agents are unable to solve—what is now called “swarm intelligence,” “distributed artificial intelligence,” and “ant colony optimization algorithms”. The idea is that an algorithm solving a complex problem can be implemented in a distributed way into the collective intelligence of a network of micro-elementary agents sharing very limited cognitive resources and interacting through very elementary rules. The collective intelligence is incommensurable with individual intelligence and emerges from global cooperation. In nature, the most striking examples of collective intelligence are provided by social insects: bee hives, wasp nests, ant nests, and termite mounds. This leads us toward politics because social insects have been considered, since Aristotle, as “political animals” (zoôn politikon) because they cooperate to produce public goods. This is the key point. In his Politics, Aristotle considers that there are essentially two kinds of political species: humans and social insects. A lot of empirical observations and comparisons led him to a behavioral taxonomy. Animals are either solitary (felines, spiders) or social. Social species can live in small groups (primates, wolf packs) or in large groups. In the latter case, they can have gregarious (herds, flocks) or “political” behavior. Aristotle claimed “man is by nature a political animal,” but he is not the only political species. What is true is that man is the only rational political species. What is especially interesting in social insects is that: 1. The collective intelligence is incommensurable with the individual intelligence of the agents. 2. The “political” performances (i.e. collective production of public goods) result from simple but extremely efficient ethological rules selected by evolution. 3. The collectivity constructs global architectures, called “architectures without architects” as honeycombs with miraculous hexagonal tesselations, or immense mounds (they would extend up to ten kms at human

Complex Methodological Individualism

415

scale) with pillars, external walls, galleries, cellars, channels, ridges, spiral conducts for ventilation and cooling, valves, brood chambers, etc.

Stigmergy Highly sophisticated buildings such as termite mounds are essential for the species to survive. They result from elementary spatial moves marked by pheromones. The process was called stigmergy in 1959 by Pierre-Paul Grassé. Stigmergy is the key concept of the theories of swarm intelligence. In constructing a mound, initially, each individual termite rolls a mudball, invests it with pheromones, and deposits it randomly in some place. But the pheromones are attractive and act as signs (hence the term of “stigmergy”: stigma = sign, ergon = action). So, many mudballs become deposited in the same place. This positive feedback produces high pillars that collapse beyond a certain height, a catastrophe which triggers the iterated building of new pillars, etc. This spontaneous emergence of coherent activity builds gradually, without any centralized planning and control, seemingly intelligent structures. Their mathematical models are highly non-trivial.11 It must be emphasized that, in contrast with the physical examples, the emerging macro architectures apply here a strong top-down “imerging” causality upon the micro individuals.

Eusociality This very particular paradigm of sociality “invented” by evolution is called eusociality in ethology, a neologism coined in 1966 by Suzanne Batra. Eusociality (from Greek “eu” = “good” and “social”), is defined (see e.g. Wikipedia) as the highest level of organization of animal sociality and characterized by cooperative brood care, overlapping generations within a colony of adults, and a division of labor into reproductive and non-reproductive groups. The “queen” and reproductive males monopolize reproduction. Division of labor creates specialized behavioral groups (castes) within the society. Soldiers and workers are sterile and specialize in brood, foraging, defense, or maintaining food and resources. At the evolutionary level, in a strict neo-Darwinian gene-centered view based on the selfish gene principle (individuals maximize the fitness of their 11

See, e.g., works by Jean-Louis Denebourg, Guy Theraulaz, Eric Bonabeau, or Bernard Manderick.

416

J. Petitot

genes), eusociality seems rather paradoxical. How can organisms increase the transmission of their genes if they become sterile and work for their close relatives? How can we imagine genetic bases for cooperation and altruistic behavior? The sociobiologist William Hamilton explained (1964) that it might be possible for a non-reproductive (sterile) individual to share more genes with a close relative than with his or her offspring (“inclusive fitness” principle). It is precisely the case with many species of social insects which are haplodiploid : males are haploid (i.e. have a single set of chromosomes from the mother) while females are diploid (i.e. have two sets of chromosomes from the mother and the father). As a result sisters can share up to 75% of their genomes: more than the 50% they can share with their offspring.12

Cognitive VS Eusocial Complexification The eusocial paradigm is the opposite of the dominant paradigm found in superior species of mammals, that of small “communities” which are like extended family groups. In that later case, the communal links are accessible to the experience of the agents. Social organization is “cognitively commensurable” and we can speak of a sort of social reflexivity. For the eusocial paradigm it is not the case. Social organization is not community-based and is cognitively inaccessible. The key point is that there exist (at least) two evolutionary ways for complexifying intelligence. On the one hand a “vertical” cognitive complexification of the individual intelligence, and, on the other hand, a “horizontal” eusocial complexification leading to a “swarm intelligence” with two levels micro–macro. Either individual intelligence increases but the social groups remain small communities, and it is the case for evolution leading toward primates and Homo Sapiens; or individual intelligence remains very limited but the groups increase drastically and become eusocial, which enables a distributed collective intelligence to emerge.

Mandeville’s Fable of the Bees The “political” character of social insects and its analogy of structure with modern human open societies has a long history in Occidental thought and recurred in the modern period. Its best-known occurrence is The Fable of the Bees of Bernard Mandeville (1670–1733) which, according to Hayek, “asked

12

For a critique of Hamilton’s inclusive fitness, see Nowak, Tarnita, Wilson 2010.

Complex Methodological Individualism

417

the right question.”13 His apologue The Grumbling Hive: or, Knaves Turn’d Honest (1705), later extended and called The Fable of the Bees; or, Private Vices, Publick Benefits (1714, 1723, 1729) had a major impact. The expression “private vices” meant at that time the prevalence of self-interest; what we now call choice rationality and maximization of utility. The thesis is that a hive functions properly only when each individual bee, each with its very limited representational resources, does what it has to do in the framework of division of labor and follows strict innate rules without bothering about collective advantages or disadvantages. They are the interactions of the bees according to efficient selected rules, and not “moral” virtues, which produce the collective wealth, benefits, and public goods of the hive. Individual bees don’t have the cognitive resources to enable them to have and share any representation of the hive. In other words, Mandeville’s fable on bees as Aristotelian “political animals” concerns the social value of selfish behavior in complex societies where myriad individuals cooperate through labor division. Mandeville made explicit in a provocative way the conflict arising at his time between the new-born economic liberalism and traditional Christian ethics. Since the beginning of the seventeenth century, some philosophers had already tried to explain that “enlightened” self-interest could be socially positive. This political, and in fact theological , thesis can be found in Blaise Pascal and Pierre Nicole (1625–1695, a Jansenist of Port-Royal14 ): society should be based upon “enlightened” self-interest rather than upon charity.15 Mandeville inspired Adam Smith and his concept of the “invisible hand” as a mechanism ensuring the “Wealth of Nations.” This first formulation of selforganized spontaneous order operates at the metaphysical, theological, and political levels as a “ruse of reason.” As Pierre Nicole claimed, to make selfish interests cooperate in an unintended way to the benefit of public welfare is “the secret plan of God,” “the hidden order of God.” “There is no need of virtuous individuals to get a virtuous society.” Selfish individuals are able, without knowing it and willing it, to do “an admirable thing”: the more the persons aim at their own interests, the more they become interdependent, and the more they compose “a superior reality able to transcend each bet.” 16 Thus, God’s “ruse of reason” is a “ruse of passion” for men.

13

Mandeville was from a family of liberal-progressive physicians of Rotterdam who emigrated to England following conflicts with the Orange Party and Calvinists. 14 See his Essais de morale, 1671. 15 See, e.g., Faccarello, 2006. 16 Here, “transcendence” means incommensurability and emergence.

418

J. Petitot

Later, we find the same kind of “ruse of reason” in Kant (1784, Idea for a Universal History with a Cosmopolitan Purpose) to solve the political antinomy of the “asocial sociality” (ungesellige Geselligkeit ), which is the key concept of Kant’s anthropology and philosophy of history. The question is no longer to speculate on the original pre-political nature of humans as hostile wolves (Hobbes) or kind lambs (Rousseau). The problem is that the human is a social species while his individual nature (ambition, domination, cupidity) is anti-social. But, by a “ruse of reason,” Nature uses this Hobbesian pathology to compel humans to accept collective rules of law. The Fable of the bees triggered a tremendous controversy (in particular with Bishop Berkeley) and was condemned in 1724 for its “diabolic attempts against religion.” Even today many people judge it outrageous to introduce a principle of moral inversion between micro-social individual intentions and global emerging macro-social properties, to posit that intentionally selfish individuals (“rational” in the sense of the theories of rational choice) governed by their private and local self-interest can, by means of their interactions, generate, in an unintended way, a global social order propitious to public interest.

Hayek on Mandeville The moral evaluation of Mandeville’s thesis is irrelevant. Mandeville was concerned by the economy of large cities and societies, by wealth rather than by virtue. He explained that a “sympathy fusion” between individuals is not necessary to reach a spontaneous “harmony” of interests. This said, it is not yet clear if, in Mandeville, the “invisible-hand” paradox was solved by a “hidden Providence” as it was previously the case in Pierre Nicole, or already by a natural self-organizing mechanism as it will be in Adam Smith, or by an artificial legislation as it will be in a physiocrat such as Helvétius or a utilitarian like Bentham. In his Lecture on a master mind : Dr Bernard Mandeville, given at the British Academy on March 23, 1966,17 Hayek explains that Mandeville was not a moralist but an excellent physician and psychologist-psychiatrist, and that the question is not to know if the “harmony” is natural (phusis and cosmos) or artificial (nomos and taxis), but to understand that it is an emerging unintended spontaneous order. Mandeville had already raised the issue of the incommensurability of social structures with respect to reason, and of the selection of “good” social rules by cultural evolution. 17

In News Studies in Philosophy, Politics, Economics and the History of Ideas.

Complex Methodological Individualism

4

419

Eusociality in Humans

The evaluation of Mandeville’s thought must be contextualized with respect to the different conceptions of social order trying to solve the “asocial sociality” of humans in modern open societies.

The Paradigms of Social Order Three conceptions of social order have long dominated political philosophy.18 They are analyzed in other articles of the Handbook. But none of them set out to explain the emerging phenomena of self-organization and spontaneous orders in modern societies. 1. The paradigm of hierarchical order and absolute power theorized from the Renaissance by Machiavelli (1469–1527), then Bodin (1529–1596) and Hobbes (1588–1679). In Hobbes’ Leviathan, the coordination problem is solved by a centralized hierarchical “vertical” power (king’s scepter and bishop’s crook) which imposes a top-down coordination by coercion. This is possible because the wealth surplus produced by global cooperation is monopolized by narrow elites enjoying exclusive privileges and able to pay for the police and the army protecting their power. It is in reaction to this form of absolutism that many demands arose for tolerance and human rights, from Grotius (1583–1645), Bayle (1647–1706) and Locke (1632–1704) to Kant (1724–1804), Humboldt (1767–1835) and Constant (1767–1830). The “vertical” orders were challenged by more “horizontal” and “democratic” conceptions advocating, like Mandeville, a sort of eusociality. 2. The rational “constructivist” (in Hayek’s sense) paradigm posits that an efficient order can be computed, planned, and applied. 3. The conservative paradigm of natural order, which champions a form of organicist holism and accuses individualism for “atomizing” society and destroying “natural communities” (family, corporations, churches, etc.). For example, for Saint-Simon (1760–1825, De la physiologie appliquée à l’amélioration des institutions sociales, 1813) and Auguste Comte (1798– 1857, Système de politique positive, 1851) holism was a sort of “organicism,” a “physiological” conception of the social reality opposing “mechanistic atomism.”

18

See e.g. Nemo, 2002.

420

J. Petitot

An Eusocial Primate But none of these three paradigms takes into account a major evolutionary characteristic of societies specific to urban civilizations. Homo Sapiens is a broadminded primate situated on the “vertical” axis of cognitive evolution. Our ancestors began to live in families, small groups, clans, hordes, and tribes where there existed a cognitive commensurability between individuals and collectivity. A kind of social reflexivity was then possible and there was no eusocial organization. A decentralized non-coercive coordination was possible because each member could control the behavior of the others. But after the prehistoric sedentarization of “hunter-gatherers,” the advent of agriculture and farming, with the apparition of great urban civilizations in Mesopotamia, Egypt, Carthage, Rome, China, large cities, new techniques (writing, accounting arithmetic, surveying geometry), currency, land and sea lines of communication, trade exchanges, etc., a fundamental break occurred and a “horizontal” eusocial-like complexity emerged, complexity which was completely alien to the evolutive line of Homo Sapiens. This dimension is well explained by many authors, e.g., Daniel Andler in his La silhouette de l’humain (2016). The challenge is to articulate cognitive science and social theories from an evolutionary perspective; to understand the very fast hyper sociality of Homo Sapiens with its division of labor, its cooperation between agents allowing group performance, its “altruism” and its other behaviors closer to those of social insects than those of primates. All the classic debates between MI and MH reappear when we try to understand the evolutionary dialectic between the cognitive and the social. From a phylogenetic perspective, the problem is very difficult. Some hypotheses proposed the concept of a “social brain” resulting from a mind/sociality co-evolution having led to a double complexity: a cognitive complexity improving that of primates coupled with a social complexity of the social insect type. Homo Sapiens would have become “pro-social” and “eusocial.” We can cite in particular the works of Richard Byrne, Andrew Whiten, Nicholas Humphrey, and Robin Dunbar, all of whom have been studied in depth by Daniel Andler. The expression “eusocial-like” uses the term “eusocial” in a cultural sense and not in a biological one. When Edward Wilson, the sociobiologist of Harvard, ant specialist (see The Ants 1990, written with Bert Hölldobler) and founding father of Sociobiology (1975) wrote The Social Conquest of Earth: humans are eusocial apes (2012), after having published with Martin Nowak and Corina Tarnita The evolution of eusociality (2010), he triggered a live debate with sociobiologists specialist in eusocial altruism such as Herbert Gintis and Bill Hamilton, or biologists and ethologists like Richard Dawkins

Complex Methodological Individualism

421

(the author of The Selfish Gene, 1976), and also cognitivist psychologists such as Steven Pinker. But, in much the same way as Hayek explained that the moral debate about Mandeville was irrelevant, the biological debate about Wilson is irrelevant. Of course, Homo Sapiens is not a eusocial species. Human eusociality results from a cultural (and not biological) evolution. The key problem is the mismatch of this eusocial cultural evolution with our primate brain. Our biological inheritance is not adapted to global social coordination. The later, requires the introduction and the acceptation of eusocial impersonal, “objective,” and external rules. But what type of rules?

Spontaneous Orders and Complex Methodological Individualism Through all his life, Hayek defended the concept of spontaneous self-organized orders, which posits that pluralism and individual freedom are not sources of disorder, anarchy, and social struggle but, on the contrary, a factor conducive to higher forms of organization. His CMI stands in sharp contrast with the other paradigms of social order and conceives social order as neither natural (permanent and universal) nor artificial (rationally construed), but pluralist and self-organized, non-hierarchical and polycentric.19 As the masters of the Scottish Enlightenment David Hume (1711–1776) and Adam Ferguson (1723–1816) emphasized,20 they are the results of human actions but not of human intentions. This conception—that individuals are the basic social entities but interact in a contractual society protected by the rule of law—is in general attributed to the tradition stemming from John Locke (1632–1704) and Adam Smith (1723–1790) and the “invisible hand ” (Theory of Moral Sentiments, 1759, The Wealth of Nations, 1776). The essential feature of the invisible hand is that it drives subjects to collective ends that do not proceed from their intentions. But as we have seen, it was already present in Pierre Nicole. And in the volume I edited with Philippe Nemo on The History of Liberalism in Europe, we find a chapter by Gilbert Faccarello on one of the main precursors of Adam Smith with Mandeville, namely Pierre de Boisguilbert (works between 1695: Le Détail de la France, and 1705: Factum de la France).21

19

Evident examples of such orders are language, law and morals: they are not natural in the strict sense of the term, but neither are they artificial since nobody has ever made them. 20 Ferguson. An Essay on the History of Civil Society. 21 Faccarello, 2006.

422

J. Petitot

For CMI, global macro rules must be rules of a law-abiding state. The function of the state is neither to uphold an “innate” order nor to impose a “rational” one, but only to secure the institutions enabling the emergence of an open and evolutive spontaneous order. CMI concerns mechanisms of self-organization, in which the agents do not have in general the cognitive resources for rationally computing all the consequences of their actions. So many collective effects of these actions are unintented. It is evident in the case of the evolution of language. Social cohesion, cooperation, and prosperity are non-intentional effects emerging from an aggregation of selfish interests. But, of course, there is a major problem. It is not sufficient to refer to empirical examples, remarkable as they may be, as language, religion, law, money, market, etc., and to claim that they are unintended results of human actions and not the intentional outcome of a general collective will. As we have already emphasized with Viktor Vanberg, the possibility of emergence must be proved using mathematical modeling and computational synthesis. This is particularly non-trivial because, as we have seen, the rationality (in the sense of the rational choice theories since William Stanley Jevons and Léon Walras) of selfish agents able to compute the maximization of their utility seems to be incompatible with global coordination. The rational economical calculus for maximizing pleasure and minimizing pain seems to be anti-social.

Evolutionary Game Theory Let us now give an idea of how we can mathematically model phenomena of social coordination. We will take the example of the emergence of cooperation in evolutionary game theory. See, e.g., the celebrated works of Robert Axelrod. The best-known example is the so-called “prisoner’s dilemma.” It is a game intended to model the dynamics of cooperation. In the simplest case we consider two players, A and B, each having two possible behaviors, namely d = defection (betraying) and c = cooperation. The game is defined by a matrix giving the payoffs of the players for each of the four possibilities (b A , b B ): A plays b A and B plays b B : T = (d , c ) = Temptation (A betrays B who cooperates), S = (c, d ) = Sucker (A cooperates and is betrayed by B), R = (c, c ) = Reward (A and B cooperate and are rewarded), P = (d , d ) = Punishment (A and B betray and are penalized for it). For the game to be relevant, betrayal T must be more profitable than cooperation R (which explains the “temptation” of betrayal), cooperation R more

Complex Methodological Individualism

423

profitable than generalized defection P and generalized defection P more profitable than unconditional cooperation S. In other words, we must assume that the payoffs satisfy the conditions: T > R > P > S. This very simple game represents a situation where individual rationality comes into conflict with collective rationality. Indeed, in the case of a one shot game between rational players. i. If player A plays c, then player B wins R if he plays c and T if he plays d . As T > R, player B has an interest in playing d . ii. If player A plays d , then player B wins S if he plays c and P if he plays d . As P > S, player B has an interest in playing d . iii. If player B is rational , he will therefore play d whatever the behavior of A. We say that the betrayal strategy d strictly dominates the cooperation strategy c: for the player B, d does better than c whatever the behavior of the other player. iv. The same is true for A by symmetry. v. The result of the game is therefore (d , d ) = (loose, loose), generalized defection which leads to the bad collective gain (P, P ). vi. But clearly, the cooperation (c, c ) = (win, win) leading to the collective gain (R, R ) would have been a much superior strategy since R > P. With such a payoff matrix, the general defection strategy (d , d ) is the only Nash equilibrium of the game, i.e., the strategy such that each player does worse if he changes strategy unilaterally. Countless examples of (loose, loose) failures of cooperation are observable in all the domains of action, from personal relationships to international geopolitics. We therefore have an elementary micro behavior of pairs of agents with its very precise rules leading systematically to a very precise result, generalized defection. We can then consider societies of such agents, introduce interactions, and wonder if new macro collective behaviors, in particular cooperation, which is the very opposite of defection, can emerge. This is typically a CMI issue. And that is indeed the case. The situation changes drastically when the game becomes more complex. First, one can iterate the game, which implies that defection can then be punished and cooperation rewarded. In this case, one can introduce more elaborate individual strategies such as unconditional cooperation, unconditional defection, “tit for tat” (TFT : start with cooperation, then play what the other player played at the previous move), “vindictive” (start with cooperation and defect forever as soon as the other player defects, defection being punished as an irreversible betrayal), etc. Moreover, games can be evolutionary games, where polymorphic populations

424

J. Petitot

of individuals use different strategies and generate new generations using the scores in a generalized competition: strategies with good scores increase their number of representatives while those with bad scores progressively vanish. In these models, agents are considered as “phenotypes” expressing “genotypes” identified with strategies, and simple “micro” strategies influence complex “macro” population dynamics. Evolutionary game theory is more realistic than the classical one based on individual choice rationality. It substitutes a collective selective scheme to an untractable variational calculus. Moreover, it enables us to understand the dynamics that drive agents toward global equilibria. Simulations and computational synthesis prove then that anti-cooperative strategies can be eliminated, and that cooperation can win and become stable. The best strategies are nicely cooperative, rapidly reacting to defections (“retaliatory”), rapidly forgiving, and simple (“clear,” without wiles). The best known is the “tit for tat” (TFT ) strategy. But such cooperative strategies are fragile with respect to mutations. Indeed, as far as unconditionally cooperative mutants exhibit exactly the same cooperative behavior as TFT agents in a TFT environment, they can therefore substitute themselves progressively and “silently” for TFT , without any observable effect. But then “bad” (unconditionally defective) mutants can destabilize, invade and destroy the system. So, to be retaliatory is a condition for being collectively stable. Moreover, one can, as did Karl Sigmund, Martin Nowak and Robert May, spatialize this evolutionary game by introducing local neighborhood relations between the agents. Simulations show that the transition between non-cooperative global states toward emerging cooperative global states has the status of a critical phenomenon, exactly as phase transitions in physics.22

Hayek and the Complexity Problem Hayek was at the same time anti-holist and anti-rationalist. For him, man is not a rational chooser. His cognitive resources are very limited and drastically imperfect, he is unable to perform the economical calculus and has always a dramatically incomplete knowledge of economical situations. Hypotheses of perfect information are wrong. He has to choose and to act according to 22

See the (Modèles formels de la ‘main invisible’: de Hayek à la théorie des jeux évolutionniste) “Formal models of the ‘invisible hand’. From Hayek to evolutionary game theory”, (Histoire du libéralisme en Europe) The History of Liberalism in Europe. See also Petitot, 2016, where fractal structures characteristic of phase transitions are computed for models of an iterated spatialized prisoner dilemma.

Complex Methodological Individualism

425

fallible expectations and very local and limited knowledge. It is for that reason that he needs social institutions, as prices in a market, which, as emphasized by Jack Birner in Cosmos and Taxis, provide a “communication structure that transmits price information efficiently and rapidly.”23 The source of complexity has to be found in the fact that, in an open society, knowledge, competencies, and information are distributed, scattered over a great number of cognitively limited and interacting agents. The systemic properties of such systems cannot be conceptually controlled. The political control of social and economic orders rests on a methodological error. Many consequences derive from this fundamental fact. 1. First, complexity prohibits at the same time a centralized hierarchical organization and a communal link of reciprocity characteristic of small, closed communities. In modern open societies the interaction between agents is no longer ensured by consensus on shared values but by exchange of signals such as prices in a market. The market is a way of circulating information in a multi-agent system whose very complexity makes it opaque to its own agents. In a Hayekian “catallaxy,” everyone cooperates with everyone else but without any shared ends. The individual aims are incommensurable with each other but mechanisms such as free trade and markets guarantee, nevertheless, a viable cooperation. 2. Complexity is an evolutionary process resulting from a selection of historico-cultural rules of behavior, practices, and institutions that are impossible to master conceptually. 3. A third consequence of complexity is that rules that govern social exchanges and communication are abstract and formal. Social selforganized complex systems are governed by civil rights guaranteed by public laws.

Cultural Evolution and Emerging Ethical Maxims At the cognitive level, be it individual or social, according to Hayek, the rules governing perception and action, as well as of conventions and norms, are products of an evolutionary process. They result from a cultural selection— a collective learning—which is a competitive/cooperative process having favored the individuals and groups that applied them. They are like cultural

23

Birner, 2016.

426

J. Petitot

shortcuts enabling people to behave rapidly and adaptively without having to recapitulate every time all the experiences and beliefs necessary to action. For Hayek, common-sense is a library of tacit knowledge routines and practical schemes patterning our experience according to generic default schemes. It is necessary to act without being overwhelmed by the overflow of irrelevant information coming from the environment. For Hayek (as for Mandeville, Hume, or Ferguson), common sense norms are not repressive constraints but, on the contrary, cognitive achievements deeply adapted to the contingencies of life. Traditions express an “embodied knowledge” which is “phylogenetic” in the sense of cultural evolution, and it is therefore rational to comply with them “ontogenetically.” In much the same way as in evolutionary biology, phylogenetic a posteriori operates as ontogenetic a priori, common sense rules operate for the subjects as a priori frames. In this sense, we find in Hayek an evolutionary theory of the self-transcendence of behavioral rules. Like linguistic rules, they proceed from symbolic institutions whose origin is neither a rational omniscient intelligence nor a deliberative social contract. We see how Hayek articulates cognitive psychology (the “sensory order”) with the sociology of complex spontaneous orders (the “catallaxy”). Of course, the very concept of cultural evolution is quite problematic. Defended by a zoologist such as Vero Copner Wynne-Edwards and many sociobiologists, but also strongly criticized and rejected by other biologists such as John Maynard Smith or George Williams, it was philosophically analyzed (and defended) by Elliott Sober and David Sloan Wilson as a case of multi-level selection. The latter posits that evolution can operate at many levels, the three main levels being gene selection, individual selection, and group selection. There can exist complex cooperative/competitive relations between the levels. Group selection is quite well accepted by specialists of the eusociality in social insects—including Nowak, Tamita, and Wilson—beyond Hamilton’s inclusive fitness (see above) and also by anthropologists who consider that social norms are the result of some group selection. For Hayek, as for Popper, cultural evolution selects groups and not individuals, subjects having to comply with rules that maximize the collective performances of their group. Among others, Robert Nadeau and Paul Dumouchel have deeply investigated this point. We can emphasize again the fact that, in the case of social insects, biological Darwinian evolution has selected “good” eusocial rules of collective organization; that is, efficient group rules.

Complex Methodological Individualism

427

However, regardless of the answer to this question, what is sure is that the subjects themselves cannot understand in what operational sense norms of just conduct can be socially fruitful because they encode a “phylogenetic” historical evolution. These norms are not “moral” in the traditional sense. That’s why subjects interpret them as duties. We must emphasize the originality of this conception: 1. As individuals cannot understand the pragmatic efficacy of norms, they accept them for deontic reasons. We recognize here a thesis we find also in Kantian ethics in the Critique of Practical Reason. 2. However, norms being socially useful, we recognize also a utilitarian conception of ethics (Jeremy Bentham, John Stuart Mill). The main difference is that the “computation” of moral maxims and actions is cognitively inaccessible to individuals. Therefore, according to Hayek, cultural evolution implies that maxims of action can act for individuals as transcendent “categorical” imperatives24 while they are at the same time immanent “hypothetical” (pragmatic) imperatives for cultures.25 For cultures, maxims are caused by the viability of a social order from which individuals gain much. As was emphasized by John Gray, Hayekian utilitarianism is indirect and exemplifies the general evolutionary principle (Haeckel’s law) according to which phylogenetic a posteriori operates ontogenetically as a priori. Hayek was able to reconcile, from within a sort of CMI loop (i) individual actions, axiological rationality, values, and (ii) social norms resulting from group selection and cultural evolution and prescribing rules and maxims to individuals. It is interesting to highlight how Hayek succeeded in renewing the notion of the categorical imperative as a deontological (non-consequentialist) conception of actions. According to deontological theses, actions must be evaluated in a principled way, independently of their consequences, while according to consequentialist theses they must be evaluated on the basis of a computation of the costs and benefits of their consequences. But as that kind of computation is intractable for a finite and limited rational mind, it is performed by cultural evolution. As was emphasized by Jean-Pierre Dupuy, cultural evolution is “utilitarian” but bears on “deontological” maxims that can be interpreted in accordance with a test of “categoricity.” 24

For Kant, a normative judgement is “categorical” when it is independent of any end. Categorical prescriptions are purely “procedural”. 25 For Kant, a normative judgement is “hypothetical” when it is conditioned by an end and prescribes means to achieve the end (consequentialism).

428

J. Petitot

References Amit, D. (1989). Modeling Brain Function: The World of Attractor Neural Networks. Cambridge: Cambridge University Press. Andler, D. (2016). La silhouette de l’humain. Quelle place pour le naturalisme dans le monde d’aujourd’hui ? Paris: Essai NRF-Gallimard. Axelrod, R., Cohen, M., Riolo, R. (1998). The Emergence of Social Organization in the Prisoner’s Dilemma: How Context Preservation and Other Factors Promote Cooperation. Santa Fe Institute Working Paper, 99-01-002. Bak, P., Tang, C., Wiesenfeld, K. (1987). Self-Organized Criticality: An Explanation of the 1/f Noise. Physical Review Letters, 59, pp. 381–384. Barabási, A.L., Albert, R. (1999). Emergence of Scaling in Random Networks. Science, 286/5439, pp. 509–512. arXiv:cond-mat/9910332 Binmore, K. (1994). Playing Fair. Cambridge: MIT Press. Birner, J. (2016). Metaphysical Models of Man in Economics. Cosmos and Taxis, 3/ 2–3, pp. 78–93. Boltzmann, L. (1872). Weitere Studien über das Wärmegleichgewicht unter Gasmolekülen. Wiener Berichte, 66, pp. 275–370. Bonabeau, E., Theraulaz, G., Deneubourg, J.-L., et al. (1998). A Model for the Emergence of Pillars, Walls and Royal Chambers in Termite Nests. Philosophical Transactions of the Royal Society of London, B, 353, pp. 1561–1576. Boudon R. (1986). Theories of Social Change: A Critical Appraisal . Cambridge: Polity Press. Bressloff, P., Cowan, J., Golubitsky, M., Thomas, P., Wiener, M. (2001). Geometric Visual Hallucinations, Euclidean Symmetry and the Functional Architecture of Striate Cortex. Philosophical Transactions of the Royal Society of London B, 356, pp. 299–330. Bulle, N., Phan, D. (2017). Can Analytical Sociology Do Without Methodological Individualism. Philosophy of the Social Sciences, 47/6, pp. 379–409. Chaté, H., Ginelli, F., Grégoire, G., Raynaud, F. (2008). Collective Motion of Self-Propelled Particles Interacting Without Cohesion. Physical Review, E 77, p. 046113. Chavalarias, D. (2020). From Inert Matter to the Global Society—Life as MultiLevel Networks of Processes. Philosophical Transactions of the Royal Society B, 375/ 1796. Coleman, J.S. (1987). Microfoundations and Macrosocial Behavior. In J. Alexander, B. Giesen, R. Münch, and N. Smelser (Eds.), The Micro-Macro Link (pp. 153– 177). Berkeley/Los Angeles: University of California Press. Di Iorio, F. (2015). Cognitive Autonomy and Methodological Individualism: The Interpretative Foundations of Social Life. Cham: Springer. Di Iorio, F. (2016). Introduction: Methodological Individualism, Structural Constraints, and Social Complexity. Studies in Emergent Order and Organization, Cosmos and Taxis, 3/2–3, pp. 1–8.

Complex Methodological Individualism

429

Di Iorio F., Chen S.-H. (2019). On the Connection Between Agent-Based Simulation and Methodological Individualism. Social Science Information, 58/2, pp. 354–376. Douady, S., Hersen, P. (2011). Dunes, the Collective Behavior of Wind and Sand. In P. Bourgine and A. Lesne (Eds.), Morphogenesis. Berlin-Heidelberg: Springer. Dupuy, C., Torre, A. (1999). The Morphogenesis of Spatialized Cooperation Relations. European Journal of Economic and Social Systems, 13/1, pp. 59–70. Dupuy, J.-P. (2000). The Mechanization of Mind: On the Origins of Cognitive Science. Princeton: Princeton University Press. Dupuy, J.-P. (2004). Vers l’unité des sciences sociales autour de l’individualisme méthodologique complexe. Revue du MAUSS, 24/2, pp. 310–328. Dupuy, J.-P., Dumouchel, P. (Eds.). (1983). L’auto-organisation: de la physique au politique. Colloque du Centre Culturel International de Cerisy-la-Salle 1981. Paris: Éditions du Seuil (reprinted by Hermann, Paris, 2022). Faccarello, G. (2006). La liberté du commerce et la naissance de l’idée de marché comme lien social. In P. Nemo and J. Petitot (Eds.). Histoire du libéralisme en Europe (pp. 205–253). Paris: Presses Universitaires de France. Ferguson, A. (1767/1996). An Essay on the History of Civil Society. Cambridge: Cambridge University Press. Gibbs, J.W. (1902). Elementary Principles in Statistical Mechanics. New York: Charles Scribner’s Sons/London: Edward Arnold. Gordon, D.M. (2010). Ant Encounters. Interaction Networks and Colony Behavior. Princeton: Princeton University Press. Grassé, P.P. (1959). La Reconstruction du nid et les coordinations interindividuelles chez Bellicositermes natalensis et Cubitermes Sp.; la théorie de la stigmergie. Insectes Sociaux, VI/1. Paris: Masson. Hamilton, W. (1964). The Genetical Evolution of Social Behaviour. I&II. Journal of Theoretical Biology, 7/1, pp. 1–16 and 17–52. Hayek, F. (1952). The Sensory Order. London: Routledge & Kegan Paul. Hayek, F. (1974). The Sensory Order After 25 Years. In W.B. Weimer and D.S. Palermo (Eds.), Cognition and the Symbolic Processes, Vol. 2, pp. 287–293. Hillsdale, NJ: Erlbaum. Hayek, F. (1978) New Studies in Philosophy, Politics, Economics and the History of Ideas. Chicago: University of Chicago Press. Heider, F., Simmel, M. (1944). An Experimental Study of Apparent Behavior. American Journal of Psychology, 57, pp. 243–259. Hodgson, G. (1986). Behind Methodological Individualism. Cambridge Journal of Economics, 10, pp. 211–224. Hölldobler, B., Wilson, E. (1990). The Ants. Cambridge, MA: Harvard University Press. Jaynes, E.T. (1957). Information Theory and Statistical Mechanics. Physical Review, 106/4, pp. 620–630.

430

J. Petitot

Kant, E. (1784). Idee zu einer allgemeinen Geschichte in weltbürgerlicher Absicht. Engl. Transl. (1991) Idea for a Universal History with a Cosmopolitan Purpose. Cambridge: Cambridge University Press. Landau, L., Lifshitz, E. (1967). Physique statistique. Moscow: Mir. Engloish Edition: (1958) Pergamon Press. Laurent, A. (2006). L’individualisme méthodologique. Paris: Presses Universitaires de France. Leroux, A., Nadeau, R. (Eds.). (2000). Friedrich Hayek et la philosophie économique (special issue). Revue de Philosophie économique, 2. Mandeville, B. (1705/1714 [1957]). The Fable of the Bees: Or, Private Vices, Public Benefits. F. B. Kaye, ed. Oxford: Clarendon. Manzo, G. (2020). Agent-Based Models and Methodological Individualism: Are They Fundamentally Linked? L’Année sociologique, 70/1, pp. 197–229. Nadeau, R. (1998). L’évolutionnisme économique de Friedrich Hayek. Philosophiques, XXV/2, pp. 257–279. Nemo P. (1998). La société de droit selon F.A. Hayek. Paris: Presses Universitaires de France. Nemo, P., Petitot, J. (Eds.). (2000). Histoire du libéralisme en Europe. Paris: Presses Universitaires de France, 2006. Italian translation by F. Di Iorio, Rubbettino, Soveria Mannelli, 2012. Nicole, P. (1671/1999/2016). Essais de Morale, contenus en divers traités sur plusieurs devoirs importans, 1671. Paris: Presses Universitaires de France (1999). Paris: Les Belles Lettres (2016). Nowak, M., Tarnita, C., Wilson, E. (2010). The Evolution of Eusociality. Nature, 466, pp. 1057–1062. Petitot, J. (1982). Introduction aux phénomènes critiques. http://jeanpetitot.com/Art iclesPDF/Petitot_CritPh.pdf. Petitot, J. (2006). Modèles formels de la ‘main invisible’: de Hayek à la théorie des jeux évolutionniste. In P. Nemo and J. Petitot (Eds.), Histoire du libéralisme en Europe (pp. 1095–1114). Paris: Presses Universitaires de France. Petitot, J. (2008). Neurogéométrie de la vision. Modèles mathématiques et physiques des architectures fonctionnelles. Paris: Les Editions de l’Ecole Polytechnique. Petitot, J. (2010). Reduction and Emergence in Complex Systems. In R.E. Lee (Ed.), Questioning Nineteenth-Century Assumptions about Knowledge, II Reductionism. NY: SUNY Series. Petitot, J. (2014). Landmarks for Neurogeometry. In G. Citti and A. Sarti (Eds.), Neuromathematics of Vision (pp. 1–85). Berlin, Heidelberg: Springer. Petitot, J. (2016). Complex Methodological Individualism. Cosmos and Taxis, 3/ 2–3, pp. 27–37. Poundstone, W. (1993). Prisoner’s Dilemma. Oxford: Oxford University Press. Rumelhart, D.E., McClelland, J.L. (1986). Parallel Distributed Processing: Exploration in the MicroStructure of Cognition. Cambridge: MIT Press. Saint-Raymond, L. (2009). Hydrodynamic Limits of the Boltzmann Equation. Lecture Notes in Mathematics. Berlin: Springer-Verlag.

Complex Methodological Individualism

431

Samuelson, L. (1997). Evolutionnary Games and Equilibrium Selection. Cambridge: MIT Press. Sawyer, R. K. (2002). Nonreductive Individualism, I. Philosophy of the Social Sciences, 32/4, pp. 537–559. Sawyer, R. K. (2002/2003). Nonreductive individualism, II. Philosophy of the Social Sciences, 33/2, pp. 203–224. Searle, J.R. (1995). The Construction of Social Reality. London: Penguin. Toulouse, G., Pfeuty, P. (1975). Introduction au groupe de renormalisation et à ses applications. Grenoble: Presses Universitaires de Grenoble. Udehn, L. (2001). Methodological Individualism: Background, History and Meaning. London: Routledge. Uffink, J., Valente, G. (2010). Time’s Arrow and Lanford’s Theorem. Séminaire Poincaré (Institut Henri Poincaré, Paris), XV, 141–173. Van der Waals, J.D. (1873). Over de continuiteit van den gas en vloeistoftoestand . Leinden: Sijthoff. Vanberg, V. (1986). Spontaneous Market Order and Social Rules: A Critical Examination of F. A. Hayek’s Theory of Cultural Evolution. Economics and Philosophy, 2, pp. 75–100. Vicsek, T., Czirók, A., Ben-Jacob, E., Cohen, I., Shochet, O. (1995). Novel Type of Phase Transition in a System of Self-Driven Particles. Physical Review Letters, 75, p. 1226. Watkins, J.W.N. (1952). The Principle of Methodological Individualism. The British Journal for the Philosophy of Science, 3/10, pp. 186–189. Weibull, J. (1996). Evolutionary Game Theory. Cambridge: MIT Press. Wilson, E. (2012). The Social Conquest of Earth: Humans Are Eusocial Apes. New York: Liveright. Wilson, K., Kogut, I. (1974). The Renormalization Group and the ε Expansion. Physics Reports, 12/2, pp. 75–199.

The Methodological Foundations of Macroeconomics: A View from Say’s Law Rosolino A. Candela

1

Introduction

Macroeconomic theories are often understood in terms of the policy conclusions with which they are identified, particularly with regard to the role of government in mitigating macroeconomic fluctuations in unemployment, output, inflation, and other macroeconomic phenomena. An overemphasis on the “interventionist” or “laissez-faire” policy implications of different macroeconomic theories can generate the presumption that what we call “macroeconomic theory” is little more than a policy conclusion in search of a set of assumptions that justify a preconceived policy conclusion, rather than theory referring to the reasoning out a set of policy conclusions from a particular set of assumptions. Moreover, such an understanding of macroeconomic theory directs attention away from the fact that theories that are

I wish to thank Francesco Di Iorio and Nathalie Bulie for their invitation to make this contribution, as well as their feedback, which improved an earlier draft of this book chapter. I also greatly benefited from Peter Boettke and Michael Wroblewski from comments through the process of writing this book chapter, for which I express much gratitude. Any remaining errors are entirely my own.

R. A. Candela (B) F.A. Hayek Program, Mercatus Center, George Mason University, Fairfax, VA 22030, USA e-mail: [email protected]

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 N. Bulle and F. Di Iorio (eds.), The Palgrave Handbook of Methodological Individualism, https://doi.org/10.1007/978-3-031-41512-8_20

433

434

R. A. Candela

defined by different policy recommendations often share common theoretical foundations, or that theories that share similar policy recommendations have different theoretical foundations. Therefore, methodological investigation of the underlying assumptions of macroeconomic theory can provide a fruitful approach to understand the commonalities and differences between different macroeconomic theories that might not otherwise be apparent, and that identifying macroeconomic theory in terms of its policy conclusions can be misleading. The purpose of this chapter is to explain the evolution of macroeconomic theory from a methodological standpoint. In doing so, my argument will be twofold. First, any theoretical explanation regarding the causes and consequences of macroeconomic phenomena cannot be understood without an underlying theory of microeconomics, whether explicitly or implicitly stated. To state this in modern parlance, all macroeconomic theory, both past and present, has entailed an underlying set of “microfoundations” of one form or another, entailing a theory of individual action (or inaction for that matter). Secondly, if the purpose of theory for the macroeconomist is to reason out a set of macroeconomic consequences from a set of assumptions, then the purpose of methodology is to understand how the selection of particular assumptions drives not only the questions that motivate different theories, but also the answers that flow from different macroeconomic theories. Thus, from a methodological standpoint, different microeconomic foundations will yield different macroeconomic explanations. To be clear, my goal will not be to provide a comprehensive survey of the variety of macroeconomic theories in the history of economic thought, per se. Such an exercise would entail an overview of how different economists across various schools of thought have applied theories to explain macroeconomic phenomena, a task beyond the scope of a single chapter. However, understanding macroeconomic theory in terms of its methodological foundations is a distinct exercise that is not mutually exclusive from understanding the history of macroeconomic thought. This being said, the central theme that will undergird my entire argument focuses on the methodological foundations of what is known as “Say’s Law” (or the fundamental law of markets), its various interpretations, and the policy implications that flow from its microfoundations. Stated briefly, Say’s Law postulates that the production of particular goods or services by one group of individuals (or firms) is the source of demand for other goods produced by other individuals (or firms). Therefore, there can be no inherent tendency in the market process toward a general “glut” or an economy-wide overproduction of all goods and services.

The Methodological Foundations …

435

The central relevance of Say’s Law to my argument is twofold. The first reason has been best stated by Axel Leijonhufvud in the following way: Say’s Law, “though simply and outwardly trivial, is crucial for clear understanding of macro-theory. Indeed, there is hardly a single problem in macro-theory (or, for that matter, micro-theory) that can be consistently analyzed without it ” (emphasis added, 1981: 80).1 Therefore, what are regarded as “problems” in macroeconomic theory are fundamentally driven by microeconomic explanations. Secondly, as noted by Tyler Cowen, “Say’s law of markets was not conceived as an isolated economic principle, but was rather part of the base of the entire edifice of Say’s liberal philosophy” (1982: 167–168). Such an observation explains why the interventionist policy recommendations made by John Maynard Keynes are a by-product of what was his first and foremost priority: a denial of the validity of Say’s Law. Steven Kates has gone so far as to state that the “Keynesian Revolution was about Say’s Law and nothing else” (Kates 2003: 8). However, understanding Keynes’s attack on Say’s Law, as well as macroeconomic theory and policy in general, ultimately turns on a methodological investigation into the nature of individual choice and the role of the price mechanism, particularly intertemporal prices (i.e. interest rates), in coordinating production and consumption decisions across time. A large body of literature has already poured over the various interpretations (and misinterpretations) of Say’s Law (Lange 1942; Mises [1950] 1960; Becker and Baumol 1952; Hazlitt 1959; Sowell 1972; Hutt 1974; Leijonhufvud 1981; Cowen 1982; Blaug 1997; Kates 1998; Baumol 1999; White 2012). Therefore, I claim no originality in reinterpreting Say’s Law, but to use this existing literature in order to distinguish between what I will refer to as “price-theoretic” and “choice-theoretic” microfoundations of macroeconomic theory.2 I use this distinction to render explicit and illustrate the methodological foundations that have underpinned the coevolution of both macroeconomic theory and Say’s law as a by-product of differences in microfoundations.

1 As a note of detail, Leijonhufvud refers to Say’s Law as “Say’s Principle” in his argument. I will elaborate further on the distinctions below. 2 Elsewhere, Peter Boettke and I have made this distinction to draw distinctions in the evolution of microeconomic theory (see Boettke and Candela 2017, 2020). See also Evans and Dragos Aligica (2016).

436

2

R. A. Candela

Which Microfoundations of Macroeconomic Theory: Price-Theoretic or Choice-Theoretic?

The central puzzle of economic theory, from the days of Adam Smith onwards, has been primarily focused on the possibility of social cooperation without command. Accordingly, it has been motivated by the following question: how can it ever be possible for individuals, pursuing their own individual goals, to unintendedly generate social order without design? If economic theory, or “microeconomics” as it is known today, explains social order as the coordination of individual plans about consumption and production across time and place, resulting in an economy-wide clearing of markets, then the obverse of this puzzle can be stated in the following way: how can it ever be the case that the pursuit of one’s self-interest unintendedly results in a systematic cluster of errors in decision, resulting in periods of economic discoordination in which markets fail to clear as well as a corresponding general glut of resources? From this standpoint, what is known today as modern macroeconomic theory can be understood as addressing the obverse of the central puzzle motivating microeconomic theory. Framed in this way, Say’s Law is the underlying economic principle that addresses both of these questions. In the next section, I will provide an overview of how Say’s Law has been understood, and how such interpretations have paralleled trends in macroeconomic theorizing, but before doing so, it is important to outline the methodological underpinnings upon which such trends have been founded. Economic theorizing, whether we are referring to “microeconomics” or “macroeconomics,” can distinguished between two paradigms: (1) a choicetheoretic paradigm; and (2) a price-theoretic paradigm. This distinction also illustrates different meanings of methodological individualism, a term first used in 1904 by the French philosopher and historian Élie Halévy, but later popularized it in the scientific community by Joseph Schumpeter (1909).3 Methodological individualism is an analytic principle that treats individual action as basic unit of analysis for explaining social phenomena (Boettke and Candela 2015; Di Iorio 2020). In drawing this distinction, I will highlight how these paradigms parallel a difference between two conceptualizations of methodological individualism: (1) an atomistic form of methodological individualism; and (2) a compositive form of methodological individualism.4 3

See also Hodgson (2007). I will mention a point of clarification regarding the distinction I am drawing here. The term “compositive” is borrowed from Hayek ([1952] 1979), but Hayek himself borrowed the term from a 4

The Methodological Foundations …

437

Given that both approaches are methodologically individualist, they are both reductionist in the sense that theoretical explanations of economic phenomena must be grounded in individual human action, and therefore economic phenomena are the aggregation of individual plans (Nozick 1977: 353). Therefore, to the extent that critics of methodological individualism use the term reductionist in a pejorative sense, they misdirect attention away from the relationship between the micro-level of human action to the macro-level of the economic coordination. The key to understanding the micro–macro relationship in economics can be understood in terms of the relationship between what economists refer to as “optimization” and “equilibrium.” Optimization refers to a course of action in which an individual has chosen, from an available set of means, the most effective for the purpose of achieving a particular end. An equilibrium refers to an economic outcome in which the interaction of optimizing individuals has exhausted all the potential gains from trade, and therefore a situation in which the plans of individuals are perfectly coordinated. An atomistic form of methodological individualism conflates the distinction between the two, such that an equilibrium represents the mere aggregation of all optimizing activity. In effect, the sum equals its parts, negating the relevance of time as a medium of human action, as well as negating the relevance of institutions, such as money, as a medium of exchange. However, a compositive form of methodological individualism implies that an equilibrium outcome is greater than the mere aggregation of optimizing activity, whereby institutions, such as market prices denominated in terms of money, emerge as an outcome of exchange, which in turn guides individuals toward an optimal course of action in terms of their decisions to buy and sell goods and services as well as their decisions to hold money. Therefore, the key distinction between these two approaches to methodological individualism, from a macroeconomic standpoint, is the relevance of time and money, which Roger Garrison (1984) refers to as the two universal concepts of macroeconomic theorizing. Whereas a choice-theoretic approach renders time and money irrelevant for macroeconomic analysis, time and money is central to a price-theoretic approach of macroeconomic analysis. Mainstream macroeconomic theorizing follows a choice-theoretic approach to economic theorizing, which collapses the optimizing activity of an agent onto the conditions of general competitive equilibrium as the ideal welfare standard of analysis. According to this paradigm, the logic of

manuscript note of Carl Menger, who used the term in his personal annotated copy of his review of Gustav Schmoller’s Methoden der Socialwissenschaften (see Hayek [1952] 1979: 212, fn. 33). However, Menger ([1883] 1985) himself identified methodological individualism as atomistic, and therefore did not imply a pejorative meaning with which it is often understood in contemporary debates.

438

R. A. Candela

human choice is not a subset of macroeconomic theory, but its defining characteristic. Such a conflation implies a direct link between the presumed rationality (or irrationality) of an economic agent and the extent to which market outcomes are consistent with a coordination (or discoordination) of resources according to consumer demands. To be clear, to argue that there is a direct link between the two levels of analysis does not imply a direct relationship between the two. In other words, macroeconomic theories that model suboptimal outcomes are not necessarily based on irrational behavior. As I will explain below, macroeconomic theories based on the same microfoundations of rational, maximizing agents can result in optimal as well as suboptimal macroeconomic outcomes. What all choice-theoretic models have in common is that they are consistent with an atomistic form of methodological individualism, since economic phenomena are simply an aggregation that can be directly reduced to the individual “choice” of individuals. James Buchanan best explains the implications of macroeconomic modeling in choice-theoretic terms as follows: “Precisely because it has divorced itself from the central proposition relating to human behavior, modern macroeconomic theory is really no theory at all. It has evolved, and remains, a set of models for the workings of economic aggregates, models that have little predictive value. Lord Keynes, of course, recognized this, and it was for this reason that he tried to tie his theoretical structure to basic psychological propensities” (Buchanan [1966] 1979: 121), what Keynes referred to as “animal spirits” (Keynes [1936] 1964: 161) to explain declines in aggregate demand, and therefore why Say’s Law fails to hold. While this may seem to be an unfair conclusion given that Keynes argued that “[w]e should not conclude from this that everything depends on waves of irrational psychology” ([1936] 1964: 162), the rise of New Keynesian macroeconomics has been predicated on “the absence of a theoretical underpinning for this Keynesian price stickiness” (Mankiw 1985: 529). Although New Keynesian economics has been characterized as a reconstruction of Keynesian macroeconomics on neoclassical microfoundations, it has only done so by reducing the macroeconomy to a representative agent model that is continuously optimizing. “Representative agent models in macro”, according to Richard Wagner, “are no different than treating a beehive as a gigantic bee or a traffic jam as a gigantic car moving in reverse. More generally, macro theorizing is filled with formulations where one macro variable acts directly upon another macro variable” (2010: 141). If the “economy” is modeled as an individual, and the “government” as another individual acting over the economy, then it easily follows that macroeconomic outcomes can

The Methodological Foundations …

439

be changed by manipulating macroeconomic variables as objects of choice, such as aggregate demand, using fiscal or monetary policy. The policy implications that follow from a choice-theoretic conceptualization of macroeconomic theory can be drawn from three different macroeconomic models. First, according to the New Classical vision of macroeconomic theorizing, prices and wages instantaneously adjust so that markets are continuously clearing, and therefore labor and other resources do not remain idle. The economy is always and everywhere in an optimal state of affairs as an approximate fact of reality, given the underlying preferences, resource availability, and technology of the economy. The policy conclusion that follows is that government intervention is redundant and unnecessary. Secondly, according to the Old Keynesian5 vision, macroeconomic outcomes are hopelessly stuck in depression, characterized by a general glut of goods and services, perpetual unemployment, and idle resources. The policy conclusion that follows requires government intervention to continuously boost aggregate demand. Third, the New Keynesian vision of macroeconomic theorizing, which shares the same neoclassical microfoundations of New Classical economics, postulates “all agents are optimizing and all prices result from that optimization” (Mankiw 1985: 535), but aggregate demand failures at the macro level of analysis can be directly reduced to suboptimal price adjustments at the individual level of analysis. The resulting conclusion is a policy of active macroeconomic stabilization due to “too much price adjustment following an expansion of aggregate demand and too little price adjustment following a contraction in aggregate demand” (Mankiw 1985: 536). This being said, if Mankiw is correct in that the “conflict between modern neoclassical and traditional Keynesian theories of the business cycle centers upon the pricing mechanism” (1985: 529), then the mutually exclusivity of policy choices between these theories is a by-product of their choice-theoretic microfoundations. Macroeconomic policy is regarded as either passive or otherwise redundant, based on fully flexible, perfect prices. Or, an active role of macroeconomic policy is necessary, based on the prevalence of imperfect or “sticky” prices. This choice-theoretic paradigm, however, precludes the theoretical possibility of passive macroeconomic policy, based on general rules rather than arbitrary discretion, being predicated on imperfect pricing and imperfect market-clearing. Why is this the case? 5 I use the term “Old Keynesian” to refer to Keynes’s theory, as discussed in his General Theory, in order to distinguish it from: (1) “Post Keynesian” economics, which evolved “unalloyed with neoclassical microeconomics” (White 2012: 153); and (2) “New Keynesian” economics, which is foreshadowed by the “neoclassical synthesis” developed by Paul Samuelson, but explicitly builds on the neoclassical microeconomic foundation of consumer optimization to explain aggregate demand failures (see Mankiw 1985, 1990; Gordon 1990).

440

R. A. Candela

Different meanings of the word “imperfect” frame our understanding of macroeconomic phenomena and the policy implications that are derived from understanding how macroeconomic forces work. From a choice-theoretic standpoint, to say that markets are “imperfect” in terms of pricing and market-clearing implies that they are flawed, suboptimal, or otherwise not ideal compared to the ideal of general competitive equilibrium. The policy implication here is either Say’s law always holds, meaning that markets are always clear based on fully flexible pricing, rendering macroeconomic policy unnecessary or redundant. Or, if prices are regarded as “sticky” and therefore markets are prone to aggregate demand failures, then government intervention is the deus ex machina that saves the macroeconomy from its own “imperfections” either through activist monetary policy or activist fiscal policy. Why? The answer is twofold. First, to use a quote from Frank Knight often used by James Buchanan, “to call a situation hopeless is to call it ideal.” The narrative that is constructed is one in which, outside the conditions of the ideal of general competitive equilibrium, there is no hope but for government intervention to save the market from itself. This conclusion, Hayek (emphasis added; 1989: 3) states, is based on “a mistaken conception of the proper scientific procedure” that “consists in the assertion that there exists a simple positive correlation between total employment and the size of the aggregate demand for goods and services; it leads to the belief that we can permanently assure full employment by maintaining total money expenditure at an appropriate level.” Moreover, as Tranditis and Boettke (2022: 215) argue: “Reducing economic phenomena to the input of a sum of aggregates or basic units that can be modelled for analysis fosters the idea that policy interventions can rearrange these parts following this analysis. Reductionist analysis helps policymakers build confidence in their capacity to control and direct economic activity. Informed by the right model or data analysis, the government supposedly knows what it should do.” Macroeconomic theorizing according to a price-theoretic approach, however, is primarily about the study of how individuals pursue their separate goals through exchange, which in turn create exchange ratios (i.e. market prices) as by-products of their purposive behavior. Such market prices, which include interest rates, in turn guide individuals in their consumption and production decision-making. Human choice is not absent according to this rendering of macroeconomic theorizing; rather, it is a necessary subset of macroeconomic theorizing, though not sufficient, for understanding the invisible hand processes that generate social order according to Say’s Law. Nor does it imply that markets allocate resources instantaneously clear based on fully flexible pricing. Rather, a price-theoretic approach to macroeconomic

The Methodological Foundations …

441

theorizing is based on a compositive form of methodological individualism, one in which there is an indirect link between a human agent and the tendency toward equilibrium, and therefore market outcomes are not directly reducible to the individuals that constitute the macroeconomy. As Cachanosky (emphasis original; 2021: 281) elaborates, “the market process is the outcome of a multiplicity of interactions among heterogeneous, not representative, economic agents that do not know the model [in which] they live.” That is, prices emerge from the act of exchange between individuals engaging in open-ended choice under a world of uncertainty, but not of human design. However, once emerged, prices then become guides for future action. Thus, whether markets “work” or “fail” do not depend on the behavioral characteristics of individuals, but whether institutions secure and enforce the ability for individuals to exchange (i.e. private property). From this standpoint, a “market failure” associated with aggregate demand failures or macroeconomic instability is not a failure of markets to “work” but a failure to establish the conditions for a market to exist, namely due to a lack of stable enforcement of general rules that enforce residual claimancy of decisionmaking and constrain the arbitrary discretion of policymakers. Two examples illustrate this point, one from the Great Depression and another from the more recent Great Recession. Robert Higgs (1997) has argued that “regime uncertainty” explains why the Great Depression lasted as long as it did. Higgs’s argument is a variant of Keynes’s claim that a lack of “animal spirits” or confidence among businessmen to invest their capital resulted in the Great Depression. However, for Higgs, the lack of “animal spirits” is not an exogenous cause, but a consequence of “a pervasive uncertainty among investors about the security of their property rights in their capital and its prospective returns” (1997: 564). In the context of the Great Recession, Lawrence White has argued that an absence in the rule of law regarding bankruptcy is what precipitated the uncertainty and subsequent financial shock following the collapse of Lehman Brothers in 2008. Rather than consistently applying the rule of law, in which banks would have had the expectation to bear both the profits and losses resulting from their decision-making, policymakers exercised discretion in terms of choosing to allow particular banks to fail or bailing them out. The collapse of Lehman Brothers was preceded by the bailout of Bear Sterns, which had set the expectation that Lehman Brothers would also be bailed out. Given these expectations, discretionary policy generated an undesirable consequence, in which “Lehman Brothers increased its leverage and its exposure to risky mortgage assets” (emphasis in original; White 2010: 458). The financial crisis that resulted cannot be

442

R. A. Candela

attributed to a market failure, but a failure of policy to enforce the conditions for market discipline in the banking industry to exist in the first place. From a price-theoretic standpoint, by using the same word “imperfect”, the narrative about macroeconomic forces and the policy implications that follow are radically different. Tracing the etymology of the word “imperfect” back to its Latin origins, and breaking it down from its constituent parts (“im-perfect”) you will learn that “im” expresses the negation, “per” comes from the Latin word meaning “thorough” or “thoroughly done” and “fect” comes from the Latin verb “facere,” meaning “to do.” Thus, rather than saying that something, or some state of affairs, is flawed, suboptimal, or nonideal, another way to interpret the meaning of “imperfect” is an act or process that is not thoroughly done, or incomplete (Candela 2020). An appropriate understanding of an “imperfect market” reveals that the market is a process of continuous tendency toward perfection, or completion, where all the gains from trade are exhausted and all plans between buyers and sellers are perfectly coordinated. As Ludwig von Mises states, the “market process is the adjustment of the individual actions of the various members of the market society to the requirements of mutual cooperation” ([1949] 2007: 258). Thus, markets will always be imperfect, but that is precisely why markets exist in the first place. Market processes exist precisely because they generate the information necessary to better coordinate the plans and purposes of individuals in a peaceful and productive manner consistent with marketclearing tendencies. The entrepreneurial lure for profit and the discipline of loss is what guide such imperfect processes in a tendency toward the creation of more complete information between buyers and sellers. Money pricing is what guides this perfecting process of “completing” the plans of buyers and sellers. None of this implies that money prices are sufficient for market clearing. Thus, although not immediately obvious by its epithet, a price-theoretic approach to macroeconomic theorizing implies that market clearing is not entirely based on the price mechanism, but also on a corresponding set of institutions that emerge to facilitate the intermediation of goods and services through time, including banking, clearinghouses, and insurance markets. Moreover, in a world in which all exchanges could be perfectly coordinated through the price mechanism, we would not observe inventories of goods or resources, including unemployment of labor. Because macroeconomic phenomena are in a process of completion that cannot be directly reduced to the choice of individuals, then this implies that interventionist macroeconomic policy is faced with both an “epistemic problem” (Trantidis and Boettke 2022). Given that the knowledge necessary to generate macroeconomic coordination is an emergent phenomenon

The Methodological Foundations …

443

that is context-specific to the process of exchange itself, macroeconomic theory generates different implications for macroeconomic policy that differs from a choice-theoretic approach. If policymakers do not have the knowledge necessary to generate equilibrium outcomes in the macroeconomy, then they will be incentivized to respond to the knowledge that is available to them. That is, democracies under the guise of “independent” central banking will use macroeconomic policy to manipulate relative prices for the approval of elected authorities seeking political survivability by generating conditions consistent with reelection.6 Rather than assuming that policymakers acting in the public interest can manipulate macroeconomic variables, in accordance with a positive relationship between total employment and the size of the aggregate demand for goods and services in the economy, analogous to a high tide lifting all boats proportionally, a focus on relative price distortions paints a different picture regarding the implications of discretionary macroeconomic policy. Given the logic of political processes is to concentrate short-term benefits of higher employment and incomes on well-organized interest groups and diffuse the long-term costs of inflation and macroeconomic instability on the ill-informed masses of the population, only a price-theoretic approach focused on the structure of relative prices in the economy, rather than focusing on the level of absolute prices, makes it possible to explain how political discretion can be used to generate politically-motivated business cycles, and the undesirable consequences of macroeconomic instability that followed from it (Wagner 1977). Thus, the implications of discretionary macroeconomic policy led Milton Friedman to argue the following: “Any system which gives so much power and so much discretion to a few men that mistakes – excusable or not – can have such far reaching effects is a bad system…Mistakes, excusable or not, cannot be avoided in a system which disperses responsibility yet gives a few men great power, and which thereby makes important policy actions highly dependent on accidents of personality,” concluding that “money is much too serious a matter to be left to the Central Bankers” (emphasis added; Friedman [1962] 2002: 50–51; see also White 2010; Boettke et al. 2021).

6

As an example, see Abrams (2006), which presents a case study of U.S. President Richard Nixon pressuring Federal Reserve Chairman Arthur Burns to use expansionary monetary policy, in spite of rising inflation, in order to increase his likelihood of reelection in 1972.

444

3

R. A. Candela

The Evolution of Macroeconomic Theory: A View from Say’s Law

The intellectual trajectory of economic theory from Adam Smith onward can be roughly divided with changes in the manner in which Say’s Law has been interpreted. From 1776 to 1871, which marks the publication of Adam Smith’s An Inquiry into the Nature and Causes of the Wealth of Nations, and the Marginal Revolution, respectively, is the period of classical political economy. During this period and among early neoclassical economists up to the 1930s, the unifying substantive proposition of social order that economists held in common was the study of invisible hand theorizing, in which private property and freedom of contract under the rule law were the governing institutional prerequisites for social cooperation without command. Economists following this line of inquiry has been dubbed by economist Peter Boettke as “mainline economics,” which is the study of the uniting substantive propositions held by economists going back to Adam Smith. This is distinguished from “mainstream economics,” which refers to what is perceived to be scientific and fashionable among leading economists at a particular time (Boettke 2012: xvii). The primary thrust of the mainline of macroeconomic theorizing during this period was, as the title of Adam Smith’s book suggests, about economic growth. Such luminary figures of this period included Jean Baptiste-Say, David Ricardo, James Mill, and John Stuart Mill, and discussions regarding the nature and causes of the wealth of nations were inextricably tied to Say’s Law. As its namesake suggests, the fundamental law of markets, or Say’s Law, is credited to Jean Baptiste Say ([1803] 1971). However, Say’s Law can trace its roots back to Adam Smith,7 and would later take its more mature form in James Mill’s Commerce Defended ([1808] 1965), through Say’s debate with Thomas Malthus ([1821] 1967), as well as John Stuart Mill’s classic essay “Of the Influence of Consumption on Production” ([1844] 1960). Two reasons can explain why debate and controversy continues to ensue over the meaning of Say’s Law. One reason is that “Say’s Law, as it emerged in the classical period,” Sowell writes, “was a cluster of related propositions contributed and refined by a number of individuals” (2006: 26). Had it not been for the piecemeal nature in which Say’s Law had developed, Steven Kates argues that “the controversy that has surrounded the meaning of Say’s Law during the past two centuries would have never occurred” (2003: 8). 7 Smith states the essence of Say’s Law in the following manner: “though a particular merchant, with abundance of goods in his warehouse, may sometimes be ruined by not being able to sell them in time, a nation or country is not liable to the same accident” ([1776] 1981: 439).

The Methodological Foundations …

445

But as Kates admits, Say’s Law “was accepted by virtually every economist of the classic period and was accepted just as strongly following the marginal revolution of the 1870s as it had been before” (2003: 7). If there was a general consensus over the meaning of Say’s Law that was generally accepted in the mainline of classical and early neoclassical economics, then how did its rejection become mainstream? This second reason, upon which my argument is based, is a paradigmatic shift from price-theoretic “microfoundations” to choice-theoretic “microfoundations” without there might have been no controversy. As Cowen states, the “extent to which a new paradigm can color and influence succeeding economic thought for decades (sometimes even centuries) to come should not be underestimated” (1982: 160). Understanding the evolution of Say’s Law in the history of economic thought requires that we draw a distinction between its various interpretations by its proponents and critics alike. Bernice Shoul (1957: 615) provides four distinct meanings of Say’s Law as follows: (1) Supply creates its own demand; hence, aggregate overproduction or a “general glut” is impossible. (2) Since goods exchange against goods, money is but a “veil” and plays no independent role. (3) In the case of partial overproduction, which necessarily implies a balancing underproduction elsewhere, equilibrium is restored by competition, that is, by the price mechanism and the mobility of capital. (4) Because aggregate demand and supply are necessarily equal, and because of the equilibrating mechanisms, output can be increased indefinitely and the accumulation of capital proceed without limit.

The first interpretation of Say’s Law, which is the most well-known among mainstream economists today, was provided by Keynes in his General Theory ([1936] 1964: 25). Although we will elaborate on the microfoundations of Keynes’s argument later, it is important to raise here that Keynes is attacking a strawman in order to undermine Say’s Law. It is therefore more appropriate to dub this meaning of Say’s Law as “Keynes’s Law” (Hazlitt 1959: 32; Cowen 1982: 178). In its simplest rendition, Say and his followers were not arguing that supply creates its own demand in the sense that “supply of shoes creates demand for shoes. It means that supply of shoes creates demand for everything other than shoes” (emphasis original; White 2012: 144–145). As Say argued (emphasis in original; Say [1803] 1971: 134): It is worth while to remark, that a product is no sooner created, than it, from that instant, affords a market for other products to the full extent of its own

446

R. A. Candela

value. When the producer has put the finishing hand to his product, he is most anxious to sell it immediately, lest its value should diminish in his hands. Nor is he less anxious to dispose of the money he may get for it; for the value of money is also perishable. But the only way of getting rid of money is in the purchase of some product or other.

Even the most important critic of Say’s Law, namely Thomas Malthus, with whom Say debated through correspondence (see Say [1821] 1967), would not object to this specific point. The main point of contention was over whether production would lead to an overall “general glut” of goods and services, specifically over the claim that increasing productivity due to capital investment would generate an abundance of supply of goods and services that would be insufficient to maintain demand due to the erosion of profits and corresponding fall in prices. Capital investment would bring about underconsumption due to prices being driven down to unprofitable levels. The error of reasoning is an omission of the fact that falling prices reflect decreasing costs of production. More importantly, increasing profitability and cost minimization are two ways of expressing the same tendency of the market process. The fact that producers are able to realize greater profits through increasing productivity is what frees up resources to be used in additional lines of production, hence “the mere circumstance of the creation of one product immediately opens a vent for other products” (emphasis added; Say [1803] 1971: 134–135) in an ever-widening scope of productive specialization and exchange that drives economic growth. While indeed a “general overproduction of commodities is impossible because the production of one commodity creates a market for another” (Cowen 1982: 171), a reassessment of the first meaning of Say’s Law does not imply that we should interpret Say’s Law in a vulgar sense of an automatic market-clearing outcome based upon fully flexible and instantaneous price adjustment, but as a process of discovery consistent with (3) and (4) of the list above. As Leijonhufvud elaborates, Say’s Law “refers only to purchase and sale intentions; it asserts absolutely nothing about the possibility of their realization” (emphasis added; 1981: 89). The fact that Say’s Law, appropriately understood, refers only to a general tendency in the market process means that it is impossible to assert that all trades are executed as planned; rather, the market serves as an error-correcting mechanism in which “prices and trading plans must be revised ” (emphasis added; Leijonhufvud 1981: 92). These points raised by Leijonhufvud are consistent with the price-theoretic approach outlined in Sect. 2, implying that Say’s Law is not a static concept, but an “imperfect” dynamic concept that reflects a process by which the plans

The Methodological Foundations …

447

of buyers and sellers become coordinated overtime, not guaranteed at every moment in time. The implication here is that Say’s Law, according to the second interpretation of the term, does not suggest that all cash balances are immediately spent, and therefore exerting no relative price effects, as would be the case if money were a “veil” as suggested by the second meaning of Say’s Law, or better known as “Say’s Identity” (Cowen 1982: 162). In this respect, institutions such as money serve as a lubricant of commerce “to minimize the likelihood of entrepreneurial error, and helps to ensure that input and output flows shall be continuous even under the complexities of the highly advanced division of labor which the monetary mechanism has rendered possible” (emphasis original; Hutt 1974: 60). Given that goods are generally not exchanged directly for goods, but indirectly via money, with “trivial exceptions every market is a market for money” (emphasis original; Garrison 1984: 199). Once we take into account that money is a good for which, like all other goods and services, there is a demand, this “keeps Say’s Law from being true in the vulgar sense” in which monetary equilibrium always holds, as implied by Lange (1942: 52). “The play in the system associated with the use of money allows for” disequilibrium in an economic system, whereby “deviations between the quantities of nonmonetary goods supplied and quantities demanded” exist (Garrison 1984: 202). Thus, if an excess supply of some goods implies an excess demand for others, then those other goods must also include money (Yeager 1956: 439). During the period between 1920 and 1950, a sharp divide between the mainline and mainstream of macroeconomic theorizing would become apparent. This was particularly evidenced by the publication of John Maynard Keynes’s The General Theory of Employment, Interest and Money in 1936. Keynes’s attack of classical macroeconomics was predicated on a rejection of Say’s Law, but such a rejection was not a sui generis argument. Its roots can be traced back to the choice-theoretic microfoundations that had developed in mainstream economics, which in turn would serve as the basis for the development of “market failure” theory during the second half of the twentieth century. Based on its choice-theoretic microfoundations, Say’s Law would become a static concept, consistent with an identity principle (i.e. “Say’s Identity”) understood in terms of an outcome or a state of affairs, rather than as a dynamic process as it was understood among economists of the classical and early neoclassical period. The result of this choice-theoretic approach would be an interpretation of Say’s Law in terms of Say’s Identity, the implications of which would render

448

R. A. Candela

macroeconomic policy as either necessary and active, or passive and redundant. However, such policy implications are by-product directly reducing macroeconomic outcomes from the choices of individuals, whereby general equilibrium (or a lack thereof ) analysis moves to the foreground of analysis. One rendition related to Say’s Identity is what is known as mainstream economics is known as “Walras Law” developed by Oskar Lange (1942). This particular interpretation of Say’s Law is one in which the impossibility of a general glut follows directly from its underlying assumptions. In this Walrasian model, disruptions to an equilibrium generate instantaneous adjustment to a new equilibrium, given the assumption of zero transaction costs. Moreover, commodities are strictly homogenous, and individuals have perfect information regarding what they are exchanging as well as the terms of trade (i.e. prices), rendering trade instantaneous. Since all resources are allocated through a “Walrasian auctioneer,” individuals passively respond to equilibrium prices, which serve as a sufficient statistic by which to allocate resources to their highest-valued uses. Mainstream macroeconomic theorizing in the vein of New Classical Economics (and DSGE8 modeling) has proceeded through this equilibrium lens. By ruling out the process by which prices emerge to generate a tendency whereby excess demands and excess supplies of resources become eliminated, all that is required to illustrate the “imperfection” of markets, according to this choice-theoretic paradigm, is to point to the existence of unemployed resources, as Keynes did. According to Keynes ([1936] 1964: 28), it is only by “accident or design” that a market economy achieves its potential of full employment. The basis for Keynes’s argument is explained by the fact that money is not a “tight joint” (Garrison 1984), whereby cash balances are immediately spent to reflect a match between aggregate supply and aggregate demand. Rather, money is a “broken joint” whereby the possibility of intertemporal coordination is precluded. Moreover, by eliminating the communicative role that interest rates play in channeling available savings into investment, money that is not spent is not saved (i.e. deferred for future consumption) but hoarded, resulting in a decline in both consumption and investment that manifests in fall in aggregate demand, a fall in output, and increasing unemployment. For Keynes, it is not supply that creates demand; rather, it is demand that creates supply. The problem is not one of a lack of production, but an overabundance of idle resources. The policy recommendation that flows from these assumptions is one in which active fiscal policy can rescue the macroeconomy from market imperfections. However,

8

“DSGE” is the acronym for Dynamic Stochastic General Equilibrium.

The Methodological Foundations …

449

as Leijonhufvud argues, Keynes’s rejection of Say’s Law was not a removal of the foundations of classical economics, but the removal of an assumption in classical economics falsely interpreted in terms of choice-theoretic microfoundations of Walrasian general equilibrium: The only thing which Keynes ‘removed’ from the foundations of classical theory was the deus ex machina – the auctioneer which is assumed to furnish, without charge, all the information needed to obtain the perfect coordination of the activities of all traders in the present and through the future (Emphasis original; Leijonhufvud 1981: 15).

In effect, government intervention replaced the role played by the Walrasian auctioneer as the deus ex machina, predicated on the ad hoc assumption of wage and price stickiness, the rationale for which New Keynesian economics would later attempt to reconstruct based on microfoundations. However, as macroeconomists Roger Garrison (2001), Lawrence White (2012), and Leland Yeager (1997) have pointed out, this label “new” is a misnomer since the connection is premised on the fact that both Old Keynesian and New Keynesian models are uniquely premised on “sticky” or downward rigid prices. Moreover, such a premise is not unique to Keynesian economics, and misleadingly implies that all other macroeconomic models are premised on fully flexible pricing and instantaneous market clearing (Garrison 2001: 232). Mainstream economic theorizing has evolved in such a way that a defense or rejection of Say’s Law on the basis of flexible or “sticky” pricing, respectively, misdirects attention away from the fact that such conclusions are based on choice-theoretic microfoundations. Moreover, it also overlooks the fact that the divergence between mainstream economics and mainline economics has fueled attempts to recapture, relearn, and rearticulate the fundamental roots of mainline macroeconomic theorizing on price-theoretic microfoundations. This continuation of the mainline has corresponded not only with a reevaluation and rehabilitation of Say’s Law, but also to correct for the misinterpretations that Keynes had introduced and remain predominant in the mainstream of macroeconomic theorizing today. What choice-theoretic microfoundations rule out is the possibility that Say’s Law, “by itself, could not possibly pose a mental block to the development of unemployment theory. On the contrary, correct and systematic application of it is necessary for the construction of a consistent theory of any disequilibrium (or equilibrium) phenomenon” (emphasis added; Leijonhufvud 1981: 93). Understood according to price-theoretic foundations, Say’s Law is not inconsistent with an excess withholding of “idle” resources by suppliers

450

R. A. Candela

of good and services (including one’s own labor) to account for time required to discover the expected demand from consumers, thereby providing reliable availability of goods and services until buyers have formed their plans. Alchian addressed the puzzle traditionally confined to macroeconomics using price-theoretic microfoundations, rather than appealing to ad hoc theorizing of “sticky wages”: what explains the continued unemployment of the Great Depression in the face of non-decreasing aggregate demand? Perhaps the best way to understand the answer he provided, and the manner in which he approached it, is to reframe the question in the following way: in the event of a fall in the demand of a particular firm’s output, or a general economic downturn, why do we not observe individuals voluntarily accepting pay cuts to avoid unemployment? The answer rests on the fact that, given the heterogeneity of labor and human capital in alternative productive uses, an individual’s information about the demand for his or her own labor and human capital in alternative productive activities is imperfect. When the wages currently paid to workers fall, the opportunity costs of discovering the values of their labor and human capital elsewhere also decline. The gap between being employed in one job, before being rehired by a different employer in another job, requires “time for discovery” (emphasis original; Alchian 1969: 122) not in the form of unemployment, as it’s usually understood, but “self-employment,” that is, seeking job information (Alchian 1969: 111). For Alchian, unemployed labor and human capital can be understood within a broader theory of resource allocation, marshaled to illustrate how the holding of inventories of resources of all kinds economizes on the costs of discovering market-clearing prices across time and place (Alchian 1969: 113). The implications of Alchian’s price-theoretic approach are that discretionary macroeconomic policy intended to mitigate the undesirable consequences of economic downturn, such as a declines in output and unemployment, may have counterproductive effects. “A sequence of measures by the government (NIRA, Guffey Coal Act, agricultural price support, and the Labor Relations Act, minimum wages) arbitrarily and successively raised prices and wages over the period – not once and for all in 1932. In the absence of these autonomous factors pushing up permissible (though not the equilibrating) wages and prices, 1933–37 would have shown greater employment and output” (emphasis in original; Alchian 1969: 126–127). In the context of the more recent Great Recession, and consistent with Alchian’s conclusion, Casey Mulligan argued that “parts of the 2009 ‘stimulus law’” implemented in the United States increased “subsidies such as unemployment insurance for people who did not find a job” (2012: 4). Therefore, delayed economic recovery manifested in the form prolonged unemployment, as was the case

The Methodological Foundations …

451

throughout the Great Depression and in the more recent Great Recession was neither a result of idle resources, nor due to a fall in aggregate demand, but a result of discretionary macroeconomic policies that keep relative prices from adjusting to reflect underlying supply and demand conditions. The resulting uncertainty only decreases the opportunity cost of being “employed” seeking alternative job opportunities. The implications of such an analysis do not imply that macroeconomic policy is unnecessary, per se; rather, counterintuitively, it implies that macroeconomic policy, rather than being active in stimulating aggregate demand, must be “active” in maintaining a nondiscretionary role. Analogous to a referee in a game, the purpose is not to manipulate the game to achieve a particular outcome, but to enforce a set of rules that allow for the fundamental law of markets (i.e. Say’s Law) to emerge indirectly from human action, though not directly from human design.

4

Conclusion

The importance of approaching macroeconomic theorizing from a methodological standpoint is to illustrate why it is misleading to conclude that macroeconomic theorizing is simply a set of public policy conclusions in search of a set of premises. This applies no less when we are discussing a theory’s policy implications. Viewing mainstream macroeconomic theorizing purely in ideological terms would misleadingly suggest that “laissez-faire” or “interventionist” macroeconomic theorists have chosen different sets of methodological assumptions to justify their policy conclusions. Such an interpretation of macroeconomics misdirects attention away from the fact that mainstream macroeconomists of a laissez-faire or of an interventionist stripe are based on a common methodological denominator: they have adopted “microfoundations” that are choice-theoretic in nature. The conclusion that macroeconomic outcomes can be directly reduced to the choices of individuals yields a binary choice in macroeconomic policy as either necessary and active or redundant and passive. Rendering explicit the methodological foundations of macroeconomic theorizing, however, directs our attention to a price-theoretic approach to macroeconomic theorizing, which transcends the false dichotomy implied by a choice-theoretic approach. Thus, the point here has not been to argue that there is no theoretical role for government in terms of macroeconomic policy, per se. As Hayek writes, “a government that is comparatively inactive but does the wrong things may do much more to cripple the forces of a market economy than one that is more concerned with economic affairs but

452

R. A. Candela

confines itself to actions which assist the spontaneous forces of the economy” (1960: 222). Rather, if there is a necessary place for macroeconomic policy, a price-theoretic approach implies that macroeconomic policy must remain passive, namely by being “active” in enforcing general rules, which is necessary for Say’s Law to emerge, rather than be directed . The key to macroeconomic stability is not the “independence” of central bankers or macroeconomic policymakers in general, but restraint in their ability to exercise discretion based on the presumed ability to manipulate macroeconomic variables as objects of their choice in accordance with the goal of achieving macroeconomic stability, whether that includes discretion over the money supply, tax policy, or other policy variables to affect unemployment and output (White 2010: 460).

References Abrams, B. A. (2006). How Richard Nixon Pressured Arthur Burns: Evidence from the Nixon Tapes. Journal of Economic Perspectives, vol. 20, no. 4, pp. 177–188. Alchian, A. A. (1969). Information Costs, Pricing, and Resource Unemployment. Western Economic Journal , vol. 7, no. 2, pp. 109–128. Becker, G. S., and W. J. Baumol. (1952). The Classical Monetary Theory: The Outcome of the Discussion. Economica, vol. 19, no. 76, pp. 355–376. Baumol, W. J. (1999). Retrospectives: Say’s Law. The Journal of Economic Perspectives, vol. 13, no. 1, pp. 195–204. Blaug, M. (1997). Say’s Law of Markets: What Did It Mean and Why Should We Care? Eastern Economic Journal , vol. 23, no. 2, pp. 231–235. Boettke, P. J. (2012). Living Economics: Yesterday, Today, and Tomorrow. Oakland: The Independent Institute. Boettke, P. J., and R. A. Candela. (2015). What Is Old Should Be New Again: Methodological Individualism, Institutional Analysis and Spontaneous Order. Sociologia: Rivista Quadrimestrale di Scienze Storiche e Sociali, vol. 49, no. 2, pp. 5–14. Boettke, P. J., and R. A. Candela. (2017). Price Theory as Prophylactic Against Popular Fallacies. Journal of Institutional Economics, vol. 13, no. 3, pp. 725–752. Boettke, P. J., and R. A. Candela. (2020). The Applied Theory of the Bourgeois Era: A Price-Theoretic Perspective. Journal of Contextual Economics—Schmollers Jahrbuch, vol. 140, nos. 3–4, pp. 355–366. Boettke, P. J., A. W. Salter, and D. J. Smith. (2021). Money and the Rule of Law: Generality and Predictability in Monetary Institutions. New York: Cambridge University Press. Buchanan, J. M. ([1966] 1979). Economics and Its Scientific Neighbours. In What Should Economists Do? (pp. 115–142). Indianapolis: Liberty Fund.

The Methodological Foundations …

453

Cachanosky, N. (2021). Microfoundations and Macroeconomics: 20 Years. The Review of Austrian Economics, vol. 34, no. 2, pp. 279–288. Candela, R. A. (2020). Are Markets Imperfect? Of Course, But That’s The Point! The Library of Economics and Liberty, May 18. https://www.econlib.org/are-mar kets-imperfect-of-course-but-thats-the-point/. Cowen, T. (1982). Say’s Law and Keynesian Economics. In R. H. Fink (Ed.), Supply Side Economics: A Critical Appraisal (pp. 160–184). Frederick: University Publications of America. Di Iorio, F. (2020). Individualisme Méthodologique et Réductionnisme. L’Année Sociologique, vol. 70, no. 1, pp. 97–128. Evans, A. J., and P. Dragos Aligica. (2016). The Microfoundations of Austrian Economics Through a New Classical Theoretical Lens. Review of Political Economy, vol. 28, no. 1, pp. 113–133. Friedman, M. ([1962] 2002). Capitalism and Freedom, 40th Anniversary Edition. Chicago: University of Chicago Press. Garrison, R. W. (1984). Time and Money: The Universals of Macroeconomic Theorizing. Journal of Macroeconomics, vol. 6, no. 2, pp. 197–213. Garrison, R. W. (2001). Time and Money: The Macroeconomics of Capital Structure. New York: Routledge. Gordon, R. J. (1990). What Is New-Keynesian Economics? Journal of Economic Literature, vol. 28, no. 3, pp. 1115–1171. Hayek, F. A. ([1952] 1979). The Counter-Revolution of Science: Studies on the Abuse of Reason. Glencoe: The Free Press. Hayek, F. A. (1960). The Constitution of Liberty. Chicago: University of Chicago Press. Hayek, F. A. (1989). The Pretence of Knowledge. The American Economic Review, vol. 79, no. 6, pp. 3–7. Hazlitt, H. (1959). The Failure of the “New Economics.” Princeton: D. Van Nostrand. Higgs, R. (1997). Regime Uncertainty: Why the Great Depression Lasted So Long and Why Prosperity Resumed After the War. The Independent Review, vol. 1, no. 4, pp. 561–590. Hodgson, G. M. (2007). Meanings of Methodological Individualism. Journal of Economic Methodology, vol. 14, no. 2, pp. 211–226. Hutt, W. H. (1974). A Rehabilitation of Say’s Law. Athens: Ohio University Press. Kates, S. (1998). Say’s Law and the Keynesian Revolution: How Macroeconomic Theory Lost Its Way. Northampton: Edward Elgar. Kates, S. (2003). Introduction. In S. Kates (Ed.), Two Hundred Years of Say’s Law: Essays on Economic Theory’s Most Controversial Principle (pp. 1–33). Northampton: Edward Elgar. Keynes, J. M. ([1936] 1964). The General Theory of Employment, Interest and Money. New York: Harcourt. Lange, O. (1942). Say’s Law: A Restatement and Criticism. In O. Lange, F. McIntyre, and T. Yntema (Ed.), Studies in Mathematical Economics and Econometrics: In Memory of Henry Schultz (pp. 49–68). Chicago: University of Chicago Press.

454

R. A. Candela

Leijonhufvud, A. (1981). Information and Coordination: Essays in Macroeconomic Theory. New York: Oxford University Press. Mankiw, N. (1985). Small Menu Costs and Large Business Cycles: A Macroeconomic Model of Monopoly. The Quarterly Journal of Economics, vol. 100, no. 2, pp. 529–537. Mankiw, N. G. (1990). A Quick Refresher Course in Macroeconomics. Journal of Economic Literature, vol. 28, no. 4, pp. 1645–1660. Menger, C. ([1883] 1985). Investigations into the Method of the Social Sciences with Special Reference to Economics. New York: New York University Press. Mill, J. ([1808] 1965). Commerce Defended . New York: Augustus M. Kelley. Mill, J. S. ([1844] 1960). Of the Influence of Consumption on Production. In H. Hazlitt (Ed.), The Critics of Keynesian Economics (pp. 24–45). Princeton: D. Van Nostrand. Mises, L. von. ([1949] 2007). Human Action: A Treatise on Economics (4th Ed.). Indianapolis: Liberty Fund. Mises, L. von. ([1950] 1960). Lord Keynes and Say’s Law. In H. Hazlitt (Ed.), The Critics of Keynesian Economics (pp. 315–321). Princeton: D. Van Nostrand. Mulligan, C. B. (2012). The Redistributionist Recession: How Labor Market Distortions Contract the Economy. New York: Oxford University Press. Nozick, R. (1977). On Austrian Methodology. Synthese, vol. 36, no. 3, pp. 353–392. Say, J-B. ([1803] 1971). A Treatise on Political Economy; or the Production, Distribution, and Consumption of Wealth. New York: Augustus M. Kelley. Say, J-B. ([1821] 1967). Letters to Mr. Malthus. New York: Augustus M. Kelley. Schumpeter, J. (1909). On the Concept of Social Value. The Quarterly Journal of Economics, vol. 23, no. 2, pp. 213–232. Shoul, B. (1957). Karl Marx and Say’s Law. The Quarterly Journal of Economics, vol. 71, no. 4, pp. 611–629. Smith, A. ([1776] 1981). An Inquiry into the Nature and Causes of the Wealth of Nations. Indianapolis: Liberty Fund. Sowell, T. (1972). Say’s Law. New York: Oxford University Press. Sowell, T. (2006). On Classical Economics. New Haven: Yale University Press. Trantidis, A., and P. J. Boettke. (2022). Macroeconomic Policy as an Epistemic Problem. Journal of Public Finance and Public Choice, vol. 37, no. 2, pp. 211– 231. Yeager, L. B. (1956). A Cash-Balance Interpretation of Depression. Southern Economic Journal , vol. 22, no. 4, pp. 438–447. Yeager, L. B. (1997). The Fluttering Veil: Essays on Monetary Disequilibrium. Indianapolis: Liberty Fund. Wagner, R. E. (1977). Economic Manipulation for Political Profit: Macroeconomic Consequences and Constitutional Implications. Kyklos, vol. 30, no. 3, pp. 395– 410. Wagner, R. E. (2010). Mind, Society and Human Action: Time and Knowledge in a Theory of Social Economy. New York: Routledge.

The Methodological Foundations …

455

White, L. H. (2010). The Rule of Law or the Rule of Central Bankers? Cato Journal , vol. 30, no. 3, pp. 451–463. White, L. H. (2012). The Clash of Economic Ideas: The Great Policy Debates and Experiments of the Last Hundred Years. New York: Cambridge University Press.

Clarifying Social Institutions in Institutional Individualism Kei Yoshida

1

Introduction

The aim of this paper is to clarify social institutions in institutional individualism, a critical rationalist version of methodological individualism.1 The conflict between methodological individualism and methodological collectivism has been a perennial in philosophy of the social sciences. Many philosophers and social scientists have been debating what the social sciences need as the unit of explanation. Methodological individualists claim that social phenomena should be explained not in terms of wholes, but in terms of individuals and their actions, interactions, etc. (Popper 1945b, 88).2 Methodological collectivists counter that individuals and their actions, interactions, etc., are not sufficient for explaining social phenomena, and thus wholes that are independent of individuals and their actions, interactions, etc., must be included in social scientific explanations. From these claims, some

1 John O. Wisdom uses the term “situational individualism” to indicate a critical rationalist version of methodological individualism (Wisdom 1970). In this paper, I shall use the term “institutional individualism” to cover recent discussions about it. 2 For the history of methodological individualism, see Udehn (2001).

K. Yoshida (B) Faculty of Social Sciences, Waseda University, Tokyo, Japan e-mail: [email protected]

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 N. Bulle and F. Di Iorio (eds.), The Palgrave Handbook of Methodological Individualism, https://doi.org/10.1007/978-3-031-41512-8_21

457

458

K. Yoshida

might think that this debate is simply methodological. This is not the case, however. Methodological positions are closely intertwined with ontological views. This does not necessarily mean that methodology implies ontology. In other words, methodological individualism is not equal to ontological individualism. Ontological individualism is the position that only individuals exist. A typical example of ontological individualism would be Margaret Thatcher’s following claim: “[W]ho is society? There is no such thing! There are individual men and women and there are families…” (Thatcher 1987, 29–30). Methodological collectivism also does not mean ontological collectivism. Contrary to Thatcher’s claim, ontological collectivism assumes that not only individuals, but also wholes such as society exist. Nevertheless, the difference between methodological individualism and ontological individualism or between methodological collectivism and ontological collectivism was not carefully considered (Yoshida 2021, Chapter 1). This is not a general claim that all methodological individualists and methodological collectivists do not distinguish between methodology and ontology. My claim here is more nuanced, namely, that some if not all of them do not sufficiently pay attention to the difference between methodology and ontology. Let us look at a few examples to illustrate my point. The first one is Thomas Hobbes. From the framework of this handbook, it would not be appropriate to regard him as a methodological individualist. But he is often regarded as such, at least, in the English-speaking world. For instance, game theorists and economists who follow the tradition of social contract tend to claim that they are methodological individualists. Steven Lukes’s paper “Methodological individualism reconsidered” illustrates this situation (Lukes 1968). On its first page, he quotes from Hobbes and John Stuart Mill to show what he means by “methodological individualism” (Lukes 1968, 119). When Lukes’s paper was published in 1968, it had already been the received view to understand methodological individualism in this way, at least, in the English-speaking world. In this received view, not only Hobbes, but also Mill is regarded as a methodological individualist. True, the received view is problematic from the perspective of this handbook. As suggested below, they would be better called psychologistic individualists. Here I discuss only Hobbes, because Mill’s view will be explained in relation to Karl Popper’s version of methodological individualism. Hobbes proposed the theory of social contract. According to him, human beings are equally created. There may be some minor differences between them; however, they are physically and intellectually similar to each other. They pursue their self-preservation. But if resources that are available to them are

Clarifying Social Institutions in Institutional …

459

limited, they fight against each other (Hobbes 1651, 60–61). If there is no common power that governs them, human beings are in the state of nature where there is a “[w]ar of every one against every one” (Hobbes 1651, 62). To avoid such a state of war, they make a social contract with each other and establish a commonwealth that has the power to govern them. Hobbes does not argue that the state of nature historically existed (Watkins 1965, 72). But his argument presupposes that only individuals exist prior to the establishment of the commonwealth. Thus, he does not carefully differentiate between methodology and ontology. Hobbes may not be an appropriate example of methodological individualism. But unfortunately, a similar problem can be seen even in Max Weber, a methodological individualist proper. According to Weber, the most basic unit of his interpretive sociology is an individual and his/her action. Although he admits that this is a questionable analogy, he calls the unit an “atom” (Weber [1913] 2012, 280). Given his admission of its being questionable, I do not intend to claim that Weber does not differentiate between methodology and ontology. He might have just made a slip of the pen; however, having made a slip of the pen suggests that he was not sufficiently careful about the distinction between methodology and ontology. If he had been, he would have avoided using the term “atom,” first and foremost.3 Let us look at one more example from methodological collectivism: Émile Durkheim. Durkheim is also not sufficiently careful about the distinction between methodology and ontology. Durkheim argues that sociology needs to explain social facts that are constituted by “the beliefs, tendencies and practices of the group taken collectively,” and their reality is independent of individual facts (Durkheim [1895] 2013, 23–24). Furthermore, he contends that society as a combination of individuals is not a mere sum of them, but an independent entity that has its own characteristics. Society thinks, feels, and acts differently from individuals (Durkheim [1895] 2013, 86). We have so far looked over some scholars who are not sufficiently careful about the distinction between methodology and ontology. I cannot generalize this point, because that is an inductivist error that I do not want to commit. This also does not necessarily mean that they do not distinguish between methodology and ontology. My point here is a matter of degree. Some might find it subtle, but it is an important point. 3 One of the referees suggests that Weber’s atom can be interpreted as a scientific construct. This may be possible. But this analogy seems to have another problem. Popper argues that thinking about the social sciences in terms of social atoms and their direct knowledge often leads to a conclusion that the method of the social sciences is subjective or psychological (Popper 1945a, 2:308, n. 14). Given Weber’s admission and such a problematic usage, it would be better to interpret Weber without this analogy. See also Note 5 and my discussion about Popper’s situational analysis at the end of Sect. 2.

460

K. Yoshida

Joseph Agassi is one of the contemporary philosophers who try to differentiate between methodology and ontology and seek for a third view.4 He calls his view institutional individualism (Agassi 1960, 1975). Although Popper’s view in philosophy of the social sciences is usually known as a version of methodological individualism, Agassi employs Popper’s criticism of psychologism and recasts Popper’s view as institutional individualism to distinguish between it and psychologistic individualism.5 Furthermore, Agassi argues that wholes such as social institutions exist, although they do not exist in the same way as individuals exist (Agassi 1975, 152).6 This is an important point, which suggests that we need to avoid a problematic confusion between methodology and ontology. There is still a problem, however. Although Popper and his followers emphasize that social institutions are important, they do not precisely explain what they are. For instance, Popper argues as follows: The term “social institution” is used here in a very wide sense, covering bodies of a private as well as of a public character. Thus we shall call a business, whether it is a small shop or an insurance company, an institution, and likewise a school, or an “educational system,” or a police force, or a Church, or a law court. … [O]nly a minority of social institutions are consciously designed while the vast majority have just “grown,” as the undesigned results of human actions. (Popper 1944b, 122)

Two points deserve our attention. First, Popper considers social institutions in a very broad way. This point will be discussed shortly. Second, according to Popper, most of social institutions are “the undesigned results of human actions.” This is what we usually call the unintended consequences of human actions. If we look at the history of the social sciences, this idea is emphasized by many scholars from Niccolò Machiavelli, Adam Smith, and Adam Ferguson to Friedrich von Hayek, Popper, and Robert K. Merton (Yoshida 2014, 110–111). It also takes an important role in evaluating several views of social institutions that will be discussed in the following sections. 4

As other examples, see Kincaid (1997), Sawyer (2002, 2003), Little (2007, 2020), and Epstein (2015). Although they seek for a third view, they do not necessarily agree with each other. They respectively propose their own candidates, and thus there are many and different third views. It is beyond the scope of this paper to examine the difference between their views and mine. But at the end of this paper, I shall suggest where my discussion will lead. 5 Francesco Di Iorio argues that Carl Menger and Weber anticipate institutional individualism (Di Iorio 2016, 366–367, n. 3). It is beyond my ability to investigate whether this is really the case. But I wonder how Di Iorio deals with Weber’s problematic analogy. 6 Agassi initially ascribes this claim to Popper (Agassi 1960, 247). But he later treats it as his own claim.

Clarifying Social Institutions in Institutional …

461

As to the first point, Popper elsewhere states that social institutions are “all those things which set limits or create obstacles to our movements and actions almost as if they were physical bodies or obstacles. Social institutions are experienced by us as almost literally forming part of the furniture of our habitat” (Popper 1994, 167). Furthermore, he gives language, market, price, etc., as examples of social institutions. From these remarks, we can discern that Popper’s definition is so broad that almost everything can be included in it (Yoshida 2019, 154). Popper is critical of essentialism, according to which examining terms and figuring out their essential meanings are necessary for scientific research (Popper 1944a, 95). Hence he would insist that trying to precisely define the term “social institutions” commits an essentialist error that he has severely criticized. I have no problem with Popper’s anti-essentialism. If we must find the essence of social institutions at the beginning, we can go nowhere. I do not think that there is such essence. Still, it would be helpful to clarify social institutions in that they play an important role in the social sciences and they are integral in institutional individualism. This paper aims at such a clarification to take one step further. The problem of how we should consider social institutions is one of the hotly debated issues in philosophy of the social sciences. In this paper, I shall first examine Popper’s criticism of psychologism. Then I shall discuss how Agassi recasts Popper’s view as institutional individualism. Next, John Searle’s and Francesco Guala’s views of social institutions will be examined. Searle argues for a rule-based approach to social institutions, and Guala proposes a unified view by integrating the rule-based approach into an equilibriumbased one. Then I shall scrutinize how Agassi and his fellow traveler, Ian Jarvie, respond to these works. Agassi and Jarvie’s view looks like one of the equilibrium-based approaches; however, they claim that social institutions cannot be reduced to rules or games. Instead, they appeal to the institution of the scholarly symposium. This comes from their institutional view of rationality, which regards social institutions as workshops. In a workshop, people can exchange and discuss ideas and solve problems in a collaborative way, and a Socratic dialogue based on workshop rationality plays an important role. I shall finally argue that Avner Greif ’s view of institutions can be interpreted in terms of Popper’s theory of World 3 and has the potential to supplement Agassi and Jarvie’s view.

462

2

K. Yoshida

Popper’s Criticism of Psychologism

Popper was originally trained as an educational psychologist. But he had a strong interest in philosophy of science, and he published the first book Logik der Forschung, which was later translated into English as The logic of scientific discovery (Popper [1935] 1959). Soon after the publication of this book, Popper, a son of converted Jews in Vienna, had to immigrate New Zealand because of Nazi Germany’s persecution of Jews. In New Zealand, he worked on political philosophy and philosophy of the social sciences to fight against totalitarianism. His works during this period are known as a series of papers “The poverty of historicism” (Popper 1944a, b, 1945b; later published as a book in 1957) and a two-volume book The Open Society and Its Enemies (Popper 1945a; hereafter The Open Society). In The Open Society, Popper develops his view into a general framework and calls it critical rationalism. Its core is the following attitude: “I may be wrong and you may be right, and by an effort, we may get nearer to the truth” (Popper 1945a, 2:213; italics in the original). Just after the publication of The Open Society, Popper arrived in London to teach at the London School of Economics and Political Science (LSE) (Simkin 1993, 2). At LSE he taught graduate students including Agassi and Jarvie who will be discussed in this paper. To understand Popper’s version of methodological individualism, his criticism of psychologism is a good starting point. Popper discusses it in Chapter 14 “The autonomy of sociology” of The Open Society. Popper’s main target is John Stuart Mill. As indicated in the Introduction, Mill is sometimes regarded as a methodological individualist. In 1843, he published a book A System of Logic. In its book VI “On the logic of the moral sciences,” Mill takes a reductionist position and argues that all laws of social phenomena can be derived from the laws of human nature (Mill 1843, 2:537). This means that we do not need social laws which are independent of the laws of human nature. Popper calls such a view psychologism. According to him, it is “the doctrine that, society being the product of interacting minds, social laws must ultimately be reducible to psychological laws” (Popper 1945a, 2:86). Hence, it would be better to call Mill a psychologistic individualist. Against psychologism, Popper argues for institutionalism. According to it, we must refer to social institutions and social environments to explain human actions. Because of that, we cannot reduce sociology to psychology, and thus the autonomy of sociology must be maintained. In Popper’s opinion, defenders of psychologism would riposte that since social institutions and social environments are created and introduced by human beings, they can and must be explained in psychological terms (Popper 1945a, 2:86–87).

Clarifying Social Institutions in Institutional …

463

According to Mill, “[m]en are not, when brought together, converted into another kind of substance, with different properties” (Mill 1843, 2:537). Since society as a group of individuals does not differ from them, it is not problematic to explain social institutions and social environments in terms of human nature. Popper finds Mill’s view problematic in that it would lead to the myth of the social contract. True, Mill does not necessarily agree with Hobbes (Mill 1843, 2:552–553). But according to Popper, emphasizing the psychological origin of social institutions means that we must consider the first steps of how they develop. For this reason, Popper argues: “Psychologism is thus forced, whether it likes it or not, to operate with the idea of a beginning of society, and with the idea of human nature and human psychology as they existed prior to society” (Popper 1945a, 2:88–89; italics in the original). In Popper’s view, the social contract is a historical and methodological myth in that human ancestors were social beings before becoming human beings. This means that social institutions exist prior to what they call human nature and human psychology. Hence if any reduction is possible, it must be not from sociology to psychology, but from psychology to sociology (Popper 1945a, 2:89). Popper’s criticism of psychologism is closely related to his view of methodological individualism. He values Mill’s psychologism only because it espouses methodological individualism. Popper writes that “it [psychologism] rightly insists that the ‘behaviour’ and the ‘actions’ of collectives, such as states or social groups, must be reduced to the behaviour and to the actions of human individuals. But the belief that the choice of such an individualistic method implies the choice of a psychological method is mistaken..., even though it may appear very convincing at first sight” (Popper 1945a, 2:87). The former part of this remark might give the impression that Popper intends to reduce wholes to individuals, as does Mill. Since Popper uses the phrase “must be reduced to,” such an impression is understandable. But in the latter part, he clearly opposes psychologism. Thus it is difficult to understand why a follower of Popper’s such as John W. N. Watkins initially argues for psychologistic individualism (Watkins 1952, 28–29).7 At the very end of Chapter 14 of The Open Society, Popper argues as follows: [W]e must not overlook the great merits of psychologism in propounding a methodological individualism and in opposing a methodological collectivism; for it lends support to the important doctrine that all social phenomena, and

7

Watkins (1957) later retracts his psychologism. For an institutionalist criticism of Watkins, see Gellner (1956). Jarvie suggests that Agassi starts taking a distance from Popper under the influence of Gellner’s argument for social institutions (Jarvie 2022, 339).

464

K. Yoshida

especially the functioning of all social institutions, should always be understood as resulting from the decisions, actions, attitudes, etc., of human individuals, and that we should never be satisfied by an explanation in terms of so-called ‘collectives’ (states, nations, races, etc.). The mistake of psychologism is its prejudice that this methodological individualism in the field of social science implies the programme of reducing all social phenomena and all social regularities to psychological laws. (Popper 1945a, 2:91)

From these remarks, it would be appropriate for us to interpret Popper as arguing for the methodological individualist position that social phenomena should be explained not in terms of wholes, but in terms of individuals and their actions, interactions, etc. Since Popper defends institutionalism, social institutions will not necessarily be reduced to individuals. Nor must all social laws be reduced to psychological laws. Thus Popper differentiates his view from psychologistic individualism. Popper writes that we should not be satisfied with explanations in terms of wholes such as social institutions. Then how does he explain social phenomena? To answer this question, his idea of situational analysis would be helpful.8 To illustrate situational analysis, Popper raises an example of Richard who intends to cross a road to catch a train. Popper states that we need to consider several things to explain Richard’s movement under the assumption that he wants to avoid a collision and to safely cross the road. First, there are physical objects to limit Richard’s movements. They include vehicles, other pedestrians, etc. Next, Richard’s movements are restricted by other factors such as “the rule of the road, police regulations, traffic signals, zebra crossings, and other such social institutions” (Popper 1994, 166–167; italics in the original). Some of them are closely related with physical objects or human bodies; others are not. For instance, traffic signals, zebra crossings, and police regulations presuppose that physical objects or human bodies exist. The rule of the road is, however, more abstract than them. Nevertheless, it would restrict Richard’s movements. As quoted in the Introduction, all things that limit our movements and actions as if they were physical objects or obstacles can be regarded as social institutions. Furthermore, Popper argues that we need to ascribe several elements to Richard in explaining his movements. They are the aim of crossing the road and knowledge and information about social

8 Popper originally uses the terms such as “situational logic,” “logic of the situation,” and “logic of situations” (Popper 1945a, 2:90, 1945b, 85). But he later proposes to use the term “situational analysis” to avoid an unnecessary misunderstanding that the theory of human action is deterministic (Popper 1968, 39). Here I follow Popper’s modification.

Clarifying Social Institutions in Institutional …

465

institutions such as traffic regulations. But these elements need not be psychologically understood. Rather, they are “elements of the objective social situation” (Popper, 1994, 167; italics in the original). In this regard, psychologically revealing what Richard is in fact thinking about is irrelevant in explaining his movements. Nor is relevant what he generally knows. Popper intends to use this situational analysis to explain social phenomena. Some might think that people’s personal opinions about and subjective interpretations of situations should play an important role in Popper’s situational analysis. But in The Open Society, Popper emphasizes that these opinions and interpretations play only a limited role. According to him, there is no reason against using our direct knowledge. But only if we can generalize it in such a way that it holds true for others, our direct knowledge is useful. Furthermore, such a generalization is hypothetical, and thus it must be subject to empirical examinations (Popper 1945a, 2:308, n. 14). This means that people’s personal opinions and subjective interpretations must be objectively treated and reconstructed, namely, as “elements of the objective social situation.” Popper argues that situational analysis deals not with a singular case, but with a typical case or a model (Popper 1994, 168). Hence it does not use the opinions and interpretations, as they are.9 This may be one of the reasons Popper argues that the so-called hermeneutics will be superfluous or simplified, when he presents his theory of World 3 that will be discussed below (Popper [1970] 1994, 170, n. 5).10

3

Agassi’s Institutional Individualism

We have so far discussed how Popper criticizes psychologism and intends to explain social phenomena from an institutional perspective. But even a follower of Popper’s such as Watkins does not properly understand Popper’s view. By recasting Popper’s view as institutional individualism, Agassi tries to resolve such a misunderstanding. Agassi was a student of Popper’s at LSE. He worked on philosophy and history of physics at the beginning of his career. But he has also written many books and articles on philosophy of the social sciences. As to methodological individualism, he published a paper “Methodological individualism” in 1960 (Agassi 1960). Then fifteen years later, Agassi published a revised 9

I am grateful to two referees who draw my attention to this point. This Note 5 was added in 1984 when Popper published a collection of his papers Auf der Suche nach einer besseren Welt in Germany (this was later translated into English as In search of a better world ).

10

466

K. Yoshida

version of it, “Institutional individualism” (Agassi 1975). Agassi’s view in the papers, particularly, in the latter paper is not a simple interpretation of Popper. Rather, we can discern that Agassi takes some distance from Popper and presents his own view. Agassi examines the debate between individualism and collectivism (or holism) and argues that many scholars assume the following two propositions. The first proposition is that “[i]f ‘wholes’ exist then they have distinct aims and interests of their own” (Agassi 1960, 245, 1975, 147; italics in the original). The second one is that “[e]ither society is primary, or the individual is primary, but not both” (Agassi 1975, 149).11 According to Agassi, these two propositions are regarded as equivalent. As a result, individualism is identified with psychologism, and collectivism is regarded as the same as institutionalism. By rejecting the two propositions, Agassi tries to differentiate between individualism and psychologism and between collectivism and institutionalism. Hence the following four options are available to us, although Agassi admits that they do not cover all possibilities: (1) psychologistic individualism, (2) institutional collectivism, (3) psychologistic collectivism, and (4) institutional individualism. Options 1 and 2 are familiar to us. Option 3 is rare; however, its examples are Plato’s view of the state and Carl G. Jung’s view of subconsciousness (Agassi 1960, 246; 1975, 151).12 Agassi argues for option 4: institutional individualism. Agassi rejects the first proposition, or more precisely, its consequent. In other words, he accepts its antecedent, and thus admits that wholes exist— although not in the same way as people exist. By “wholes,” he means a wide variety of social groups and social institutions such as customs, constitutions, neighborhoods, states, etc. In this regard, his definition is broad, as is Popper’s.13 By rejecting the consequent, Agassi claims that wholes do not

11

Agassi (1960) refers only to the first proposition. The second proposition is added in Agassi (1975). Also, Agassi uses the term “holism” as a counterpart of individualism. But for the sake of consistency, I shall use the term “collectivism,” and thus my terminology differs from his. 12 It may be tempting to identify psychologistic individualism with ontological individualism or atomism. In fact, Agassi takes psychologistic individualism as “the view that only individuals exist and have interests” (Agassi 1960, 245; 1975, 147–148; italics in the original). But following Popper, Agassi argues that “[p]sychologism is the programme of explaining all social phenomena solely with the aid of psychological theory” (Agassi 1960, 246; see also Agassi 1975, 144). Both individualism and collectivism can be psychologistic, and thus it is not an ontological thesis. He also defines individualism as the doctrine that ascribes the power to decide, and then to act only to individuals (Agassi 1975, 144; see also Agassi 1960, 244). Since both psychologism and individualism are not ontologically defined, Agassi has made a slip of the pen. Given his definitions of them, it would be appropriate not to identify psychologistic individualism with ontological individualism or atomism. 13 Even John Wettersten, who is strongly influenced by Agassi, does not seem to explain the term “social institutions.” For details, see Wettersten (2006). This may be one of the defects that are shared by critical rationalists.

Clarifying Social Institutions in Institutional …

467

have their distinct aims and interests. Rather, if they have aims and interests, this is because we give aims to them or because we act following what we think must be their interests (Agassi 1960, 247; 1975, 152). Some might wonder why Agassi needs to reject the consequent of the first proposition. This is because if wholes have their distinct aims and interests, then we will need to ask for the help of psychological theories to explain them. This would lead us to accept psychologism, according to which the aims and interests must be ultimately explained by psychological theories. This is what Agassi wants to avoid. That is why he cannot accept the consequent of the first proposition. By rejecting the second proposition, Agassi admits that both the individual and society can be primary. The individual is affected by society, and vice versa. Hence the sense of primacy is blurred. In relation to this, Agassi argues that Popper’s moral philosophy tends to lean toward psychologistic individualism, according to which individual conscience in criticizing society is regarded as important. By contrast, Agassi claims that from the perspective of institutional individualism, not only individual conscience, but also the sense of duty to one’s society is significant. True, this does not mean that we must accept tribalism or collectivist ethics (Agassi 1975, 154). But this is the implication of rejecting the second proposition. We have so far examined Popper’s and Agassi’s views; however, they do not necessarily clarify what they mean by “social institutions.” In the following sections, I shall discuss several attempts to explain what they are.

4

Searle’s Rule-Based Approach to Social Institutions

In this section, I shall examine Searle’s view of social institutions. Searle has written many books and articles about philosophy of mind and language. He has tried to extend his views of mind and language to society, and thus he has proposed his own philosophy of society (Searle 1995, 2010). The term “philosophy of society” might suggest that he is interested in the social sciences. But this is not necessarily the case. In proposing his philosophy of society, he mostly rely on his works on philosophy of mind and language. Hence, Searle’s work in this field is sometimes called social ontology. Searle proposes his view of social institutions to be consistent with his philosophy of mind and language. In so doing, he bases his argument on what he calls brute or basic facts. The brute or basic facts are ones provided by the natural sciences such as physics, biology, chemistry, etc. Searle uses

468

K. Yoshida

these brute or basic facts in discussing society and social institutions. His ideas would be like this. Neurobiological processes create mental phenomena of human beings, and these phenomena develop social institutions. Hence his argument lineally proceeds from brain to mind to society; however, he avoids using terms such as reduction and supervenience (Searle 2010, 4–5). Searle considers social institutions in terms of rules, and thus his approach is rule-based. Searle regards an institution as “a system of constitutive rules” (Searle 2010, 10). To explain constitutive rules, Searle compares them with regulative rules. The difference between the two rules is that regulative rules are supposed to control people’s actions, but constitutive rules create the very possibility of these actions. Searle argues that “regulative rules have the form ‘Do X,’” but “constitutive rules have the form ‘X counts as Y in context C’” (Searle 2010, 10). For instance, Joe Biden (X) counts as the 46th President (Y) of the United States of America (C). Such a constitutive rule gives Biden the status of being the 46th President, and this is collectively accepted or recognized by people. In Searle’s view, social institutions as constitutive rules are created by declaration. He argues: “[A]ll of human institutional reality is created and maintained in existence by (representations that have the same logical form) as SF [Status Function] Declarations, including the cases that are not speech acts in the explicit form of Declarations” (Searle 2010, 13; italics in the original). This shows that Searle seriously takes the role of language in creating social institutions. According to Searle, status functions are functions that can be performed by having the collectively recognized status (Searle 2010, 7). By having the collectively recognized status of the 46th President of the United States of America, Biden can perform some functions that are specific to the American president. Searle identifies status functions with institutional facts, and he argues that they have deontic powers that provide “desire-independent reasons for action” (Searle 2010, 23). He claims that to make institutions work properly, we need to accept or recognize deontic powers. Desireindependent reasons for action include obligations, rights, duties, etc. To clarify desire-independent reasons, Searle raises an example of an obligation that he must give a lecture tomorrow. This obligation stands regardless of whether he wants to give a lecture. Even if he does not want to do so, he has the obligation. This is why he regards this obligation as desire-independent. According to him, human rationality lies in the fact that human beings can create desire-independent reasons and act based on them (Searle 2010, 127–128). Some might have the impression that Searle takes institutions as restricting human lives; however, he claims that the former also provide the

Clarifying Social Institutions in Institutional …

469

latter with many possibilities. In his opinion, without institutions such as money, marriage, or the American presidency, we would have no desire of earning much money, getting married, or becoming the American president (Searle 2010, 123–124). This means that institutions give human beings not only restrictions, but also possibilities. Searle’s view of social institutions is interesting in many respects; however, there are a few problems with it. First, as already suggested, his philosophy of society is based on disregard of social scientific research. But many social scientists have spent enormous time and energy in investigating social phenomena or society. It is not appropriate to ignore such efforts. Second, when talking about the creation of social institutions, Searle gives an important role to declaration. True, some of the existing institutions were created by declarations. But not all of them were created in such a way. Searle admits that some institutions may unconsciously evolve. He states that “money may simply have evolved without anyone ever thinking, ‘We are now imposing a new function on these objects’; and once money has evolved, people may use money to buy and sell without thinking about the logical structure of imposed function” (Searle 1995, 22). Nevertheless, he adds that in order for money to have an agentive function, which is intentionally assigned by an agent, some people in the exchange system must be able to understand that money is for buying and selling. Otherwise, its function will be regarded as nonagentive. An example of a nonagentive function is pumping blood as the function of heart. According to Searle’s hierarchical taxonomy of facts, only agentive functions that are performed by collective acceptance or recognition are regarded as status functions, namely, institutional facts (Searle 1995, 121– 125).14 Because of that, Searle’s view pays attention only to an intentional aspect of social institutions, and thus does not necessarily capture another important aspect of them: the unintended consequences of human actions.

5

Guala’s Unified View of Social Institutions

Guala is a philosopher of economics, and he worked on the methodology of experimental economics. But he has recently published a book Understanding institutions (Guala 2016). As a result of publication of this book and other articles on institutions, those who study institutions are interested in his works.

14

I would like to thank one of the referees who draws my attention to this point.

470

K. Yoshida

Guala examines rule-based approaches such as Searle’s and claims that they are not satisfactory in explaining social institutions. In claiming so, Guala appeals to an equilibrium-based approach and tries to integrate it with a rulebased one. This is what he calls the rules-in-equilibrium approach. Guala takes institutions as “rules that people are motivated to follow” (Guala 2016, xxv). To understand Guala’s unified view, we need to know the equilibriumbased approach that he has in mind. According to him, it was defended by scholars such as Thomas Schelling and David Lewis. Guala states that “[t]heories within the equilibria approach view institutions as behavioral patterns that tend to persist because individuals have no incentive to deviate from the pattern unilaterally (unless everyone else does the same)” (Guala 2016, xxiii). Guala is not satisfied with a rule-based approach to social institutions in that there are many ineffective rules. He refers to traffic lights in Italy as an example of such rules. According to him, people in Milan, Rome, and Naples respond differently to them—although the traffic rule is the same (regulation in Milan; suggestion in Rome; decoration in Naples). Since the rule is the same in these cities, there should be something else that makes the rule properly work in some cities, but not in other cities. Guala argues that the equilibrium-based approach is helpful in explaining what it is, because “effective institutions are backed up by a system of incentives and expectations that motivate people to follow the rules” (Guala 2016, xxiv; italics in the original). Guala claims that the equilibrium-based approach is compatible with the rule-based approach. From an external point of view, institutions show behavioral patterns that correspond to equilibria of coordination games. “But each equilibrium strategy also takes the form of a rule that dictates each player what to do in the given circumstances” (Guala 2016, xxv). In a sense, the equilibrium-based approach and the rule-based approach examine the same phenomena from different perspectives. By “equilibria,” what does Guala mean? He raises an example of traffic lights. The traffic lights on streets suggest what we should do, and thus they work as signals for coordination. If all of us follow the signals, then we will have no accident. But if any of us ignores one of the signals, then we might have an accident. Guala calls such signals “correlation devices” and argues that “the actions of people who use these signals constitute correlated equilibria” (Guala 2016, xxv–xxvi; italics in the original). In such a correlated equilibrium, an institutional rule takes the form of a conditional: “if X then do Y” (Guala 2016, xxv). In a sense, this form can be regarded as one of the regulative rules.

Clarifying Social Institutions in Institutional …

471

According to Guala, not only physical things such as the traffic lights, but also language can work as a correlation device. Guala’s example is a message in preparing for a dinner party: “I shop and you cook.” This gives you an expectation that I will go shopping while you cook, and thus it creates an equilibrium. If I say to you “I cook and you shop,” then another equilibrium will be created (Guala 2016, xxvi). Understanding the role of language as a correlation device, Guala focuses on Searle’s view of social institutions that places language in its center. Guala considers Searle’s distinction between regulative rules and constitutive rules and finds it unacceptable. According to him, the role of constitutive rules is limited, and they can be eliminated by translating into regulative rules. For instance, think about money. Searle refers to the constitutive rule of money as follows: “Bills issued by the Bureau of Engraving and Printing (X) count as money (Y) in the United States (C)” (Searle 1995, 28). Guala argues that constitutive rules refer to conditions when and how theoretical terms are employed to name institutions. This means that if conditions of a constitutive rule are specified, then it can be safely translated into a regulative rule with no loss. Thus following Guala’s idea, we can translate the constitutive rule of money into a regulative rule as follows: “‘[I]f a bill has been issued by the Bureau of Engraving and Printing, then use it to purchase commodities or save it for the future’” or something like that (Guala 2016, xxvii). The form of this rule is a conditional, namely, a regulative rule: “if X then do Y.” This is how Guala integrates Searle’s rule-based approach into the equilibrium-based one. In my view, Guala’s view is better than Searle’s in that although Searle disregards social scientific research, Guala seriously takes it. But it also has a problem, because he does not discuss the unintended consequences of human actions. In a joint paper with Frank Hindriks, Guala argues that according to their view, “institutions can be the unintended consequences of individual actions” (Hindriks and Guala 2021, 2038). Unfortunately, they do not develop this point further. I do not know why Guala does not discuss the unintended consequences of human actions. But explaining them is one of the main aims of the social sciences, and thus he should have stated what he thinks about them.

472

6

K. Yoshida

Agassi and Jarvie’s Institutional View of Rationality and Responses to Searle and Guala

We have so far looked over Searle’s and Guala’s approaches to social institutions. In this section, I shall examine how Agassi and Jarvie propose a different view of social institutions. While arguing for institutional individualism, Agassi develops Popper’s critical rationalism with Jarvie. Jarvie was also a student of Popper’s and a colleague of Agassi’s for many years. Jarvie works on philosophy of the social sciences. Agassi and Jarvie have participated in the socalled “rationality” debate and have coauthored many papers. This debate was triggered by Peter Winch’s criticism of Edward E. Evans-Pritchard’s anthropology, and its main question was whether different cultures, or institutions, are rational or not.15 In this debate, Agassi and Jarvie have developed an institutional view of rationality. To understand it, it would be helpful to look over what they have argued in this debate. In one of their papers, they argue that rationality can be seen as a matter of degree, and thus they distinguish rationality into three levels. Rationality 1 is that of “the goal-directed action of an agent with given aims and circumstances,” inclusive of his/her knowledge and opinions (Jarvie and Agassi 1979, 353–354). Rationality 2 contains “the element of rational thinking or thinking which obeys some set of explicit rules” (Jarvie and Agassi 1979, 354). Rationality 3 is “that of goal-directed action (rationality 1) subject to thinking that conforms to not just some rule or standard of rational thinking (rationality 2), but the best available rules for, the highest standards of, rational thinking” (Jarvie and Agassi 1979, 354; see also Jarvie 2005, 475– 476). From this perspective, the problem lies not in the difference between rational and irrational, but in the difference between more rational or less rational. To satisfy the highest standards of rational thinking, we must be self-critical. But that is not enough. We sometimes try to protect our pet views. To avoid this, we need mutual criticism, and this cannot be done alone. Through their close collaboration, Agassi and Jarvie came to view rationality not as a matter of individual mentality, but as that of action and contexts (Jarvie 2005, 472–473). As a result, they emphasize the role of social institutions and contexts, and thus their view can be regarded as an institutional view of rationality. But their view is ambiguous in that they do not precisely explain the term “social institutions.” What kind of social institution do they have in 15

For my view of this debate, see Yoshida (2014).

Clarifying Social Institutions in Institutional …

473

mind? Is their view rule-based, equilibrium-based, unified, or none of them? At first sight, it appears to be equilibrium-based. There are several passages that support this interpretation. For instance, Agassi once claims that “society is the conventional means of co-ordination between individual actions” (Agassi 1960, 264; italics in the original). Jarvie also states that “[f ]or Popper, institutions are intermediate between persons: they serve as means of co-ordination between the diverse aims of diverse people” (Jarvie 1972, 32). Furthermore, Jarvie criticizes Searle’s disregard for unintended consequences and argues that Agassi’s view of institutions as coordinating mechanisms anticipated Lewis’s version (Jarvie 1998, 378–379, n. 7). These remarks look like game-theoretic expressions, and thus some might be tempted to conclude that Agassi and Jarvie’s view is one of the equilibrium-based approaches. This is what I thought when I compared Jarvie with Guala (Yoshida 2019, 155). That was, however, a hasty conclusion. Referring to Malcolm Rutherford, Hindriks states that in economic theories, the rule-based approach is part of old institutionalism, and new institutionalism represents the equilibrium-based approach (Hindriks 2022, 355). Furthermore, Rutherford and Fernando Toboso show that institutional individualism is used both in old institutionalism and in new institutionalism (Rutherford 1994, Chapter 3; Toboso 2001, 2008). Thus it would be appropriate to regard Agassi and Jarvie’s view as a mixture of the rule-based and the equilibrium-based approaches. If so, then what is the difference between them and Guala? When I compared Jarvie with Guala, I have argued that the differences between them would be their views of the unintended consequences and of functionalism (Yoshida 2019, 155–156). Jarvie emphasizes the importance of the unintended consequences; however, Guala does not necessarily do so. Furthermore, Jarvie is ambivalent about functionalism. He claims that functionalism is false as a theory, but fruitful as a method (Jarvie 1973, 35). He elsewhere states that “asking functionalist questions” is good sociology, but “using functionalist explanations” is bad sociology (Jarvie 2003, 268). By contrast, Guala takes a functionalist view of institutions (Guala 2016, 30; Hindriks and Guala 2021). These are the differences that I could discern. In an article on the review symposium on Guala’s book, Agassi and Jarvie argue for a different approach that relies on the scholarly symposium as a social institution (Agassi and Jarvie 2019). Interestingly, they take Searle’s and Guala’s approaches as reducing institutions to either rules or games. Furthermore, they argue that most discussions in the review symposium take individuals as being prior to institutions and focus on explaining how the former create the latter (Agassi and Jarvie 2019, 58). This comes

474

K. Yoshida

from Popper’s criticism of psychologism, and it is a basis of Agassi’s institutional individualism. Furthermore, Agassi and Jarvie lament that classics in the social sciences are rarely mentioned in the current debate about social institutions. While criticizing Searle, Guala, and discussions about them, Agassi and Jarvie propose their own view of social institutions as follows: The scholarly symposium has elements of a game (of arguing): it follows rules (mostly unwritten). For example, symposiasts know they should try to be clear and to use good arguments; what in local contexts this amounts to is often folded into the discussion of the point at hand. Still, to say that a symposium is a game is scarcely an adequate description because, among other things, it aims at the truth of the matter rather than at winning, and to say it is a set of rule-following activities likewise falls short because there is no set of rules that leads to the truth systematically. A symposium is an institution that serves a goal, or multiple goals—in the present case, communications directed at trying to understand social institutions. Not all institutions serve goals. (Agassi and Jarvie 2019, 57)

This passage suggests that the scholarly symposium has aspects of the equilibrium-based approach and the rule-based approach. But both approaches are not satisfactory in explaining social institutions such as the scholarly symposium. Some might think that Agassi and Jarvie suddenly present such a view; however, that is not necessarily the case. They have already presented a similar view under the name of workshop rationality. They propose the idea of workshop rationality in relation to their argument for rationality as a matter of degree. In their view, rationality is not a matter of individual mentality, and thus individual efforts are not sufficient to meet the highest standards of rational thinking, which are represented by rationality 3. Doing so requires mutual criticism of each other’s views, and a workshop provides an arena for it. In the workshop, people discuss ideas in a collaborative way. The relation between people is not rigid such as that of master and apprentice or of teacher and student. Rather, they can change their roles—if necessary. Their model of workshop is a Socratic dialogue (Jarvie and Agassi 1979, 360; Jarvie 2005, 474–475). It is not a coincidence that Agassi and Jarvie refer to the scholarly symposium, because they take the term “symposium” from one of Plato’s dialogues (Agassi and Jarvie 2019, 56–57). Hence we can discern a continuity between workshop rationality and the scholarly symposium, because both assume a Socratic dialogue as their model. In this regard, their view of the scholarly symposium can be regarded as an extension of workshop rationality.

Clarifying Social Institutions in Institutional …

475

We have so far examined Agassi and Jarvie’s institutional view of rationality and their responses to Searle and Guala. As mentioned above, Agassi and Jarvie have emphasized game-theoretic aspects in their works; however, their view can be seen as a mixture of the rule-based and the equilibrium-based approaches. They reject a simple view that reduces social institutions to rule or games. But if social institutions cannot be reduced to rules or games, then what are they after all? True, an easy solution should be avoided; however, Agassi and Jarvie seem to make the problem more complicated than ever.

7

Greif’s Another Unified View of Institutions

In the previous section, we have seen Agassi and Jarvie’s view that social institutions cannot be reduced to rules or games. To supplement this view, I shall discuss Greif ’s view of institutions. Greif is an economic historian who had graduate training both in economics and in history. In his book Institutions and the Path to the Modern Economy, he tries to unify historical analysis and game theory (Greif 2006). He examines historical records about how traders in the middle age solved their problems, and then analyzes the records using game theory. Greif is also trying to unify the rule-based approach with the equilibrium-based one, as is Guala. Nevertheless, his view does not simply reduce institutions to rules or games, and thus it is compatible with institutional individualism. Emrah Aydinonat and Petri Ylikoski argue that Greif ’s view lacks philosophical ambition in that he does not try to construct a philosophical view of institutions (Aydinonat and Ylikoski 2018, 562). But as I shall show below, Greif ’s view is philosophically, or more precisely, ontologically interesting. In this regard, his view deserves a close examination. Greif calls his approach comparative and historical institutional analysis. According to him, it departs from major institutional analyses in two respects. First, Greif claims that his approach does not take institutions as monolithic entities. Rather, they are constituted by distinct, but interrelated elements such as rules, beliefs, norms, and organizations. These elements are exogenous to individuals, and thus the former influence the latter (Greif 2006, 14). This point is fundamental in Greif ’s view of institutions, which will be discussed later. Second, his approach is a combination of two perspectives: a structural/cultural and an agent/functionalist. The former can be seen in sociology, and the latter is common in economics. Furthermore, Greif argues that his approach studies institutions as equilibria where they work as structures affecting individuals, but the latter’s responses reproduce the former (Greif

476

K. Yoshida

2006, 14). But if his view takes institutions as equilibria, then what is the difference between him and Guala? Greif ’s answer is that he does not view games and institutions as the basic unit of analysis (Greif 2006, 14–15). This point will also be explained in due course. First, let us look at Greif ’s view of institutions. He defines an institution as follows: An institution is a system of social factors that conjointly generate a regularity of behavior. Each component of this system is social in being a man-made, nonphysical factor that is exogenous to each individual whose behavior it influences. Together these components motivate, enable, and guide individuals to follow one behavior among the many that are technologically feasible in social situations. I often refer to such social factors as institutional elements. The institutional elements that this work focuses on are rules, beliefs, and norms as well as their manifestation as organizations. An institution is a system of rules, beliefs, norms, and organizations that together generate a regularity of (social) behavior. Each of these elements satisfies the conditions stated previously. (Greif 2006, 30; italics in the original)

To understand Greif ’s unified view, I shall focus on the following two questions. First, what does it mean that institutional elements are social, and thus artificial and non-physical? Second, how does Greif consider the institutional elements? Let us think about the first question. When reading the above passage, some might wonder what these institutional elements are. If they are not physical, then do they exist in the minds of people? But Greif clearly states that they are exogenous to individuals. If so, then they do not exist in the minds of people. Greif does not explain the ontological status of the institutional elements; however, we can use Popper’s theory of World 3 to clarify it. As to the so-called body-mind problem, Popper argues against monism and dualism. Instead, Popper claims that “the world consists of at least three ontological categories” (Popper 1968, 26). Popper calls these categories the first, the second, and the third worlds; however, he later renames them Worlds 1, 2, and 3. World 1 is the world of the physical, World 2 is that of the mental, and World 3 is that of objective ideas. The objective ideas include many things such as theories, propositions, arguments, etc. According to Popper, most of social scientific objects such as social institutions are abstract and theoretical constructions (Popper 1945b, 80). Thus social institutions are also included in World 3 (Simkin 1993, 145; Di Iorio 2016, 356; Jarvie 2016, 286). These objects in World 3 are constructed by human beings, but do not exist in the

Clarifying Social Institutions in Institutional …

477

minds of people, namely, in World 2. For this reason, they are not subjective, but objective and autonomous. According to Popper, there is no direct interaction between World 1 and World 3. The human mind in World 2 works as a mediator between them. As the human mind sees or grasps objects in World 1, it sees or grasps those in World 3 (Popper 1968, 26). As I have already written, Popper argues that the theory of human actions is not deterministic, but that social institutions can limit or restrict human actions in one way or another. This means that social institutions as objects in World 3 do not completely control human actions. Rather, if social institutions such as traffic regulations need to be modified, then we may be able to change them after the careful consideration. Hence the relation between social institutions and human beings is not unidirectional, but bidirectional.16 By comparing Greif ’s view and Popper’s, some may be surprised to discern how they are similar to each other. As far as I know, there is no indication that Greif has read Popper’s works. Their similarity may be a coincidence. But even if this is the case, it is sufficient for us to show that Greif ’s view of institutions can be interpreted in terms of Popper’s theory of World 3. This should be enough for the first question. Let us focus on the second question about Greif ’s view of the institutional elements. Greif regards an institution as a system of the institutional elements such as rules, beliefs, norms, and organizations. Some might think that if so, then an institution as a system must have all institutional elements. But Greif denies this. According to him, he uses the term “system” to emphasize that the elements are interrelated, and thus all elements are not necessarily included in an institution (Greif 2006, 30, n. 1). Furthermore, Greif argues that all rules, beliefs, and norms are not necessarily regarded as institutional elements. He states that “[a] legal rule, a constitutional provision, a moral code, or beliefs that do not influence behavior are not components of an institution” (Greif 2006, 30). In Greif ’s view, influencing behavior and generating a behavioral regularity are important conditions of being part of an institution. But if a rule does not motivate people to follow it, it does not influence behavior. Because of that, Greif claims that people’s motivation to follow a rule makes it part of an institution. To explain his view, Greif refers to the infringement of property rights. Think about robbery as a concrete example of it. It is commonly known that robbery is a crime, and if it is discovered, a robber will get arrested and punished. Even a robber usually knows this. Such a common knowledge is based on beliefs in legal sanctions, and these beliefs presuppose the

16

I am grateful to one of the referees who draws my attention to this point.

478

K. Yoshida

existence of legal organizations such as the police and the court. In Greif ’s opinion, one important point about an organization is its dual character. It is part of an institution, but at the same time, it also constitutes an institution (Greif 2006, 31). If we think about organizations such as the police or the court, this point is not difficult to understand. The police or the court is part of a larger institution; however, it is also an institution. As we can see, a rule that influences and regulates behavior is important. According to Greif, such a rule needs to be socially articulated and disseminated. Through the social rule, people learn what is socially and morally acceptable behavior and what is not. In this sense, the social rule provides people with a cognitive and normative guide about how to behave. People cognitively and normatively understand the situation and coordinate behavior. But they must be motivated to follow a rule. Beliefs and norms are supposed to give people such motivation. Greif refers to driving on the right as an example of it. In his view, people drive on the right not because of the rule about it, but because of the belief that since others drive on the right, it would be the best for them to do so. According to him, two kinds of beliefs motivate behavior: “internalized beliefs and behavioral beliefs (expectations).” The former are about “the structure and details of the world we experience (and potentially other worlds) and the implied relationship between actions and outcomes.” Internalized beliefs directly or indirectly influence behavior. The latter are about “the behavior of others in various contingencies, whether or not the behavior actually occurs” (Greif 2006, 36). The belief about driving on the right is an example of behavioral beliefs. From these remarks, we can discern that each of the institutional elements has its own role. Rules teach people what is socially and morally acceptable and what is not. Beliefs and norms motivate them to follow the rules. Then what is the role of organizations? Greif argues that organizations play the following three roles: “to produce and disseminate rules, to perpetuate beliefs and norms, and to influence the set of feasible behavioral beliefs” (Greif 2006, 37). Since an organization has a dual character, it is related to other institutional elements in an interesting way. This is how Greif presents his view of institutions as systems of the institutional elements such as rules, beliefs, norms, and organizations. Since we have examined the first part of how Greif ’s approach departs from major institutional analyses, let us discuss the second part. As to this part, the relation between Greif ’s approach and game theory is crucial. Although he analytically and empirically uses game theory, he does not regard it as a theory of institutions (Greif 2006, 18). Such a view can also be discerned from his treatment of rules. Game theory assumes that people are playing against the

Clarifying Social Institutions in Institutional …

479

rules of the game. Against this, Greif argues that people are playing against rules (Greif 2006, 16, 367). Some might wonder what he means. In his view, rules that people play are not ones revealed and described by game theory. Rather, the rules that Greif has in mind are social, and thus they are exogenous to people. In other words, Greif ’s point is that people play against these social rules that are not necessarily the same as what game theory reveals and describes. Think about Greif ’s example of driving on the right. Let us assume that this works well in a city adopting it. Game theorists would call such a situation an equilibrium. But most drivers are not familiar with game theory, and thus they do not know whether it is called an equilibrium. They simply accept a social rule that it would be best for them to follow what others drive on the right. This clearly suggests that Greif does not intend to reduce institutions to games. He uses game theory as an analytic tool “in situations in which institutionalized rules prevail.” In such situations, rules are followed, behavior is reproduced, and belief and norms that motivate it are kept intact. Thus, the sets of institutionalized rules and of admissible beliefs will be restricted. In short, when institutionalized rules prevail, game theory limits the scope of analysis (Greif 2006, 151). We have so far examined Greif ’s view of institutions. It has a strong similarity to Popper’s theory of World 3. Both Popper and Greif do not take institutions or institutional elements as physical, and do not assume that they exist in the minds of people. Greif ’s view does not intend to reduce institutions to rules or games. True, it does not provide a general view of them. But since his approach is a mixture of different perspectives, this is a price that Greif must pay. If our aim is to have a general view of institutions, then this may be problematic. But if we want to explain them as social phenomena, such a problem is inevitable. Institutions in the real world are too complex to be explained by the general view that reduces them to rules or games. True, if the general view can explain all institutions, then that would be excellent. I cannot examine all available approaches; however, we do not seem to have such a general view. At least, there is no agreement about it. Then we will have to be satisfied with a mixture of different approaches. This does not mean that we must give up finding a better view. What we can do is to improve what we have now, even if that may not lead to a better view. In this regard, Greif ’s approach tries to explain institutions without simplifying them. This does not mean that it has no problem. For instance, Greif does not seem to be interested in the unintended consequences of human actions. This is probably because he argues for an endogenous theory of institutional change. Greif tries to fill in a gap between two perspectives of institutions: the intentionally created perspective and the evolutionary perspective. According

480

K. Yoshida

to the former, institutions are intentionally created by forward-looking individuals, and thus their changes are endogenous. This can be seen in new institutional economics. By contrast, the latter emphasizes that institutions emerge as the unintended consequences of human interactions. This can be represented by old institutional economics and the Austrian economics. But in explaining institutional change, Greif adopts the former. As a result, he does not attach an importance to the latter, according to which institutions can develop evolutionarily and unintentionally (Greif 2006, 153–156). There may be other problems that I could not find. Still, Greif ’s view seems to have the potential to supplement Agassi and Jarvie’s view of social institutions.

8

Conclusion

In this paper, I have tried to clarify social institutions in institutional individualism. First, I have explained Popper’s criticism of psychologism. Then I have examined how Agassi recasts Popper’s view as institutional individualism. Next, I discussed several views of social institutions. Searle proposes a rule-based approach; however, it is not based on a serious examination of social scientific research. In this sense, it is not clear whether it is helpful in developing social scientific theories of social institutions. Guala tries to integrate the rule-based approach into the equilibrium-based one. But it is also problematic, because Guala does not discuss the unintended consequences of human actions. Arguing against reducing social institutions to rules or games, Agassi and Jarvie appeal to the institution of the scholarly symposium. To supplement their view, I have scrutinized Greif ’s view of institutions and argued that it can be interpreted in terms of Popper’s theory of World 3. Although I cannot start writing a new paper here, the current discussions about methodological individualism often turn on the idea of supervenience that comes from physicalism in philosophy of mind. Physicalism is generally believed as “the thesis that everything is physical” (Stoljar [2001] 2021). Because of that, it can be seen as a version of monism. Supervenience is the idea that the higher-level entities depend (or supervene) on the lower-level ones, but statements at the higher level are not necessarily reduced to those at the lower level (Little 2007, 345). From this perspective, higher-level entities such as the mental do not exist independently of lower-level ones such as the physical, because the former depend (or supervene) on the latter. But this does not mean that there are no statements or regularities at the higher level such as the mental. Such a view is usually called non-reductive physicalism in philosophy of mind. Some scholars apply this idea to the relation

Clarifying Social Institutions in Institutional …

481

between the social and the individual. As the mental depends (or supervenes) on the physical, the social depends (or supervenes) on the individual. In this regard, it is not surprising that those who rely on supervenience argue for ontological individualism regardless of their theoretical differences (Sawyer 2002, 2003; Little 2007, 2020; see also Epstein 2015, 74). They admit that there are social laws or regularities, but do not accept that social entities exist independently of the individual. In terms of the distinction in the Introduction, they are ontological individualists, but not methodological individualists (Udehn 2001, 2).17 Popper’s theory of World 3 will shed a new light on the current discussions in that it does not endorse physicalism as a version of monism. It can be used to refute ontological individualists who rely on the idea of supervenience and its underlying physicalism. This point needs to be investigated further. Anyway, the starting point of this paper was that Popper’s definition of social institutions was too broad. I have attempted to narrow it, and I hope that my attempt does not make it too narrow. Acknowledgements An early version of this paper was presented at Rationality in the Transnational World: New Perspectives (10th Birthday Conference of the Collegium, Institute of Advanced Studies of Lyon). I am grateful to the audience. This paper was completed during my sabbatical at York University in Toronto. I would like to express my gratitude to people at Waseda University and York University who have made my sabbatical possible. A usual disclaimer applies here.

References Agassi, J. (1960). Methodological individualism. British Journal of Sociology, 11(3), 244–270. Agassi, J. (1975). Institutional individualism. British Journal of Sociology, 26 (2), 144–155. Agassi, J., & I. C. Jarvie. (2019). Institutions as a philosophical problem: A critical rationalist perspective on Guala’s “Understanding institutions” and his critics. Philosophy of the Social Sciences, 49 (1), 42–63. Aydinonat, N. E., & P. Ylikoski. (2018). Three conceptions of a theory of institutions. Philosophy of the Social Sciences, 48(6), 550–568. Di Iorio, F. (2016). World 3 and methodological individualism in Popper’s thought. Philosophy of the Social Sciences, 46 (4), 352–374. Durkheim, É. ([1895] 2013). The rules of sociological method. In S. Lukes (Ed.), The rules of sociological method and selected texts on sociology and its method

17

Lars Udehn has already pointed out the relation between the popularity of this position and the idea of supervenience (Udehn 2001, 357, n. 5). I did not understand it until quite recently.

482

K. Yoshida

(pp. 1–120). 2nd ed. Translated by W. D. Halls. Houndmills, Basingstoke, Hampshire, UK: Palgrave Macmillan. Originally published as Les règles de la méthode sociologique. Paris: F. Alcan. Epstein, B. (2015). The ant trap: Rebuilding the foundations of the social sciences. Oxford: Oxford University Press. Gellner, E. A. (1956). Explanations in history. Proceedings of the Aristotelian Society, Supplementary Volume 30, 157–176. Greif, A. (2006). Institutions and the path to the modern economy: Lessons from medieval trade. New York: Cambridge University Press. Guala, F. (2016). Understanding institutions: The science and philosophy of living together. Princeton, NJ: Princeton University Press. Hindriks, F. (2022). Institutions and their strength. Economics & Philosophy, 38(3), 354–371. Hindriks, F., & F. Guala. (2021). The functions of institutions: Etiology and teleology. Synthese, 198(3), 2027–2043. Hobbes, T. (1651). Leviathan, or the matter, forme and power of a commonwealth ecclesiasticall and civil . London: Printed for Andrew Crooke. Jarvie, I. C. (1972). Concepts and society. London: Routledge & Kegan Paul. Jarvie, I. C. (1973). Functionalism. Minneapolis, MN: Burgess. Jarvie, I. C. (1998). Situational logic and its reception. Philosophy of the Social Sciences, 28(3), 365–380. Jarvie, I. C. (2003). Fuller on science. Philosophy of the Social Sciences, 33(2), 261– 285. Jarvie, I. C. (2005). Workshop rationality, dogmatism, and models of the mind. In C. E. Erneling & D. M. Johnson (Eds.), The mind as a scientific object: Between brain and culture (pp. 471–486). New York: Oxford University Press. Jarvie, I. C. (2016). Popper’s philosophy and the methodology of social science. In J. Shearmur & G. Stokes (Eds.), The Cambridge companion to Popper (pp. 284– 317). New York: Cambridge University Press. Jarvie, I. C. (2022). Popper and Agassi at odds. Philosophy of the Social Sciences, 52(6), 329–340. Jarvie, I. C., & J. Agassi. (1979). The rationality of dogmatism. In T. F. Geraets (Ed.), Rationality today (pp. 353–362). Ottawa: University of Ottawa Press. Kincaid, H. (1997). Individualism and the unity of science: Essays on reduction, explanation, and the special sciences. Lanham, MD: Rowman & Littlefield. Little, D. (2007). Levels of the social. In S. P. Turner & M. W. Risjord (Eds.), Philosophy of anthropology and sociology (pp. 343–371). Amsterdam: Elsevier. Little, D. (2020). A new social ontology of government: Consent, coordination, and authority. Cham, Switzerland: Palgrave Macmillan. Lukes, S. (1968). Methodological individualism reconsidered. British Journal of Sociology, 19 (2), 119–129. Mill, J. S. (1843). A system of logic, ratiocinative and inductive: Being a connected view of the principles of evidence, and the methods of scientific investigation. 2 vols. London: John W. Parker.

Clarifying Social Institutions in Institutional …

483

Popper, K. R. ([1935] 1959). The logic of scientific discovery. London: Hutchinson. Originally published as Logik der Forschung. Vienna: Springer. Popper, K. R. (1944a). The poverty of historicism, I. Economica, 11, 86–103. Popper, K. R. (1944b). The poverty of historicism, II: A criticism of historicist methods. Economica, 11, 119–137. Popper, K. R. (1945a). The open society and its enemies. 2 vols. London: George Routledge & Sons. Popper, K. R. (1945b). The poverty of historicism, III. Economica, 12, 69–89. Popper, K. R. (1968). On the theory of the objective mind. In Proceedings of the XIVth international congress of philosophy, vol. 1 (pp. 25–53). Vienna: Herder. Popper, K. R. ([1970] 1994). An objective theory of historical understanding. In In search of a better world: Lectures and essays from thirty years (pp. 161–170). Translated by L. J. Bennett. London: Routledge. Originally published as Eine objective Theorie des historischen Verstehens. Schweizer Monatshefte, 50 (3), 207– 215. Popper, K. R. (1994). Models, instruments, and truth: The status of the rationality principle in the social sciences. In M. A. Notturno (Ed.), The myth of the framework: In defence of science and rationality (pp. 154–184). London: Routledge. Rutherford, M. (1994). Institutions in economics: The old and the new institutionalism. Cambridge, UK: Cambridge University Press. Sawyer, R. K. (2002). Nonreductive individualism. Part I—supervenience and wild disjunction. Philosophy of the Social Sciences, 32(4), 537–559. Sawyer, R. K. (2003). Nonreductive individualism. Part II—social causation. Philosophy of the Social Sciences, 33(2), 203–224. Searle, J. R. (1995). The construction of social reality. New York: Free Press. Searle, J. R. (2010). Making the social world: The structure of human civilization. New York: Oxford University Press. Simkin, C. (1993). Popper’s view on natural and social science. Leiden, the Netherlands: E. J. Brill. Stoljar, D. ([2001] 2021). Physicalism. In E. N. Zalta (Ed.), Stanford encyclopedia of philosophy. https://plato.stanford.edu/entries/physicalism/. Accessed on September 30, 2022. Thatcher, M. (1987). Interview for Woman’s Own (‘no such thing as society’). September 23. Margaret Thatcher Foundation. https://www.margaretthatcher. org/document/106689. Accessed on September 30, 2022. Toboso, F. (2001). Institutional individualism and institutional change: The search for a middle way mode of explanation. Cambridge Journal of Economics, 25 (6), 765–783. Toboso, F. (2008). On institutional individualism as a middle-way mode of explanation for approaching organizational issues. In S. S. Batie & N. Mercuro (Eds.), Alternative institutional structures: Evolution and impact (pp. 203–243). London: Routledge.

484

K. Yoshida

Udehn, L. (2001). Methodological individualism: Background, history, and meaning. London: Routledge. Watkins, J. W. N. (1952). Ideal types and historical explanation. British Journal for the Philosophy of Science, 3, 22–43. Watkins, J. W. N. (1957). Historical explanation in the social sciences. British Journal for the Philosophy of Science, 8, 104–117. Watkins, J. W. N. (1965). Hobbes’s system of ideas: A study in the political significance of philosophical theories. London: Hutchinson. Weber, M. ([1913] 2012). On some categories of interpretive sociology. In H. H. Bruun & S. Whimster (Eds.), Max Weber: Collected methodological writings (pp. 273–301). Translated by H. H. Bruun. London: Routledge. Originally published as Über einige Kategorien der verstehenden Soziologie. Logos, 4, 253–294. Wettersten, J. (2006). How do institutions steer events? An inquiry into the limits and possibilities of rational thought and action. Aldershot, UK: Ashgate. Wisdom, J. O. (1970). Situational individualism and the emergent groupproperties. In R. Borger & F. Cioffi (Eds.), Explanation in the behavioural sciences (pp. 271–296). London: Cambridge University Press. Yoshida, K. (2014). Rationality and cultural interpretivism: A critical assessment of failed solutions. Lanham, MD: Lexington Books. Yoshida, K. (2019). Jarvie on rationality and cultural relativism. In R. Sassower & N. Laor (Eds.), The impact of critical rationalism: Expanding the Popperian legacy through the works of Ian C. Jarvie (pp. 145–158). Cham, Switzerland: Palgrave Macmillan. Yoshida, K. (2021). Shakaikagaku no tetsugaku ny¯umon [Philosophy of the social sciences: An introduction]. Tokyo: Keiso Shobo.

The Analytical Micro–Macro Relationship in Social Science and Its Implications for the Individualism-Holism Debate Gustav Ramström

1

Introduction

There is a longstanding discussion within philosophy of science on the relationship between higher level facts and lower level facts (see, for example, Churchland 1985; Craver and Bechtel 2007; Fodor 1974; Kemeny and Oppenheim 1956; Kim 1999; Nagel 1961; Wimsatt 2000). This debate is often framed in terms of how “micro” facts relate to “macro” facts. At the core of this debate stands the question of how the properties, activities, and interrelations of microelements relate to higher level/systemic properties and events. How do atoms relate to molecules, cells to organs, neural activity to psychological experience, and individuals to social entities and events?1 1 Some social scientists define the micro and macro levels in terms of the size of social units, see, for example, Munch and Smelser (1987, pp. 356–57) and Ritzer (2000, pp. 499–505). However, the majority of both individualists and collectivists agree that the micro/macro debate must be couched in terms of relations between properties at different levels of analysis and not in terms of group size (Archer 1995, pp. 8–9; Sawyer 2001, p. 552; Wimsatt 1985, p. 260).

This article was previously published: Ramström, G. (2018). “The Analytical Micro–Macro Relationship in Social Science and Its Implications for the Individualism-Holism Debate.” Philosophy of the Social Sciences, 48(5), 474–500. https://doi.org/10.1177/0048393118789793.

G. Ramström (B) Political Science, Stockholm University, Stockholm, Sweden e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 N. Bulle and F. Di Iorio (eds.), The Palgrave Handbook of Methodological Individualism, https://doi.org/10.1007/978-3-031-41512-8_22

485

486

G. Ramström

Although each scientific discipline, of course, provides a different answer to what particular entities or phenomena actually constitute higher (macro) level versus lower (micro) level ones, the discussion of micro–macro relationships within the philosophy of science has often had a cross-disciplinary and general character. That is, the issue of how higher level facts relate to lower level ones has often been treated as a general philosophical problem, and examples relating to a wide range of scientific disciplines have been invoked to support a certain argument or illustrate a certain perspective on this issue. Micro–macro relationships as seemingly diverse as atoms and molecules, computer memory and the characteristics of memory chips, and individuals and social collectives have been discussed as a piece and used as examples in a unifying or universalizing part-whole debate concerned with general issues such as inter-level causality, emergence, or multiple realization (see, for example, Craver and Bechtel 2007; Fodor 1974; Humphreys 1997; List and Menzies 2009; Wimsatt 2000). Discipline-specific micro–macro debates have, of course, also emerged, but the distinction between them is often blurred, and arguments from one discipline are quickly generalized and adapted to other disciplines. This tendency is perhaps especially prevalent in philosophy of social science, where, as (Bouvier 2011, p. 4) points out, for quite some time the discussion has been taking place at several removes from what social scientists actually do. It has been particularly popular to draw inferences and conclusions about the micro–macro relationship in the social sciences based on the micro–macro discussion taking place in philosophy of mind and philosophy of biology (e.g., Kincaid 1986, 1996; List and Spiekermann 2013; Sawyer 2001, 2002, 2003; Vromen 2010). In an article discussing how developments in philosophy of mind can be applied to the issue of “emergence” in sociology, Sawyer (2001, p. 65)—in reasoning typical of contemporary philosophy of science—proclaims that “although philosophical arguments about emergence and reducibility have focused on the mind-brain relation, they can be generalized to apply to any hierarchically ordered sets of properties.” In this article, I argue that while this cross-fertilization and partial unification of the micro–macro debate in many instances may have been fruitful and advantageous, it has had a quite detrimental effect on our understanding of the particular characteristics of the social scientific micro–macro relationship. This is so, I will argue, because the micro–macro relationship in the social sciences differs from that of the disciplines from where its participants typically borrow or import their arguments (such as philosophy of mind and biology) in a very significant respect: the social scientific micro–macro relationship is analytical rather than physical/empirical . The failure to pay due

The Analytical Micro–Macro Relationship in Social …

487

attention to this fact, I argue, has significantly decreased clarity of thought within the social scientific micro–macro discussion and led to both ontological ambiguities and methodological misunderstandings. Here I attempt to address some of these issues.2 More specifically, this article has three interconnected purposes: (a) to clarify the distinction between analytical and empirical micro–macro relationships and how it marks a clear disanalogy between the social sciences and most other scientific disciplines; (b) to illustrate how this disanalogy implies a phenomenally monist social ontology (as opposed to the phenomenally dualist micro–macro relationship found in most other parts of science); and (c) to show how this fact invalidates the current mainstream argument for social macro causation (the so-called “multiple realization argument” which has been imported from the phenomenally dualist discipline philosophy of mind). The remainder of the article is divided into four parts; the content of the three first parts mirrors the three purposes outlined above, and the forth part contains some concluding remarks.

2

Analytical Versus Empirical Micro–Macro Relationships

In the often general and cross-disciplinary micro–macro debate found in philosophy of science, scholars typically (as discussed above) assume that arguments about the relationship between micro and macro properties in one discipline can also be applied to hierarchically ordered sets of properties in other disciplines. However, this intuition ignores the fact that there are different kinds of macro properties which relate to the micro level in very different ways. That is, a property of X that we label as a “macro property” may refer to X taken as a unified whole (type A) or to X taken as a composite system (type B). While type A macro properties refer to properties that emerge at a higher level of manifestation given lower level circumstances, type B macro properties rather label, summarize, and describe the lower level circumstances that characterize the disposition and structure of the system. If, for instance, we identify a diamond as “hard,” this would be a type A macro property, while if we characterize the crystallized structure formed by the diamond’s carbon atoms as “concentrated,” we are talking about a type B macro property. Both types of properties are macro properties in the sense that they are ascribed to the system as a whole rather than to any of its parts in 2

I address further problems which this lack of clarity has generated in Ramström (2018).

488

G. Ramström

particular. However, they are also different in a significant respect; type A macro properties are identified at a distinct “macro level of manifestation” (or macro level of zoom/grain), while type B macro properties are inferred from micro-manifest circumstances (i.e., circumstances identified at a micro level of zoom/grain). We may thus call type A properties macro-manifest macro properties, while type B macro properties are often called microstructural , micro-based , or micro-manifest macro properties (see also Ramström 2018). In many scientific disciplines, micro examination and the knowledge about the micro base (the properties, behavior, and interrelations of micro entities) acquired through such examination are used to explain or understand macromanifest properties or, if you will, “higher level” properties. To return to our geological example above, a diamond is hard because it has a concentrated atomic structure. In chemistry, likewise, the boiling point of water (type A) depends on the behavior and interrelations of water molecules (type B), and in physiology the properties and interrelations of muscle fibers (type B) determine the amount of weight a muscle is able to lift (type A), and so on. In fact, this form of higher level/lower level analysis—in which micro-based macro properties are simply treated as a part of the micro-level facts that “gives rise to” or “constitutively explains” macro-manifest properties (or simply macro properties)—is the kind of analysis that the micro–macro discussion in philosophy of science traditionally has undertaken. Phrased differently, the discussion has concerned the relationship between the micro base (including the interrelatedness of microelements) and the macro properties to which this micro base gives rise (Kim 1999, p. 6).3 The micro–macro discussion in philosophy of science has thus traditionally concerned the relationship between higher versus lower levels of phenomenal expression or, if you will, the relationship between phenomena at different levels of zoom or grain within a given system. Intuitively, it would seem that the same perspective could be applied to social science as well. Just as states/events at the neurological level of phenomenal expression give rise to psychological experiences at the mental level of phenomenal expression or atomic-level states/events give rise to phenomena at the molecular level of phenomenal expression, it may seem reasonable to suggest that the activities and interactions of individuals give rise to phenomena at the social level. However, this is not the case. Social phenomena do not have macro-phenomenal expressions.

3

The “micro structure” or “micro-base,” following Kim, includes the microelements that constitute the system, all of the intrinsic properties of these microelements, and the relations that configure these microelements into a structure (Kim 1999, p. 6).

The Analytical Micro–Macro Relationship in Social …

489

Let us take a look at the often invoked (but poorly understood) relationship between social science and philosophy of mind to demonstrate this lack of macro manifestations in the social sciences. Compare the relationship between psychological states (macro) and neural activity (micro) in the philosophy of mind/psychology to the relationship between societal phenomena (macro) and the activity of individuals (micro) in social science. While a mental experience such as “pain” refers to a macro psychological state rather than a certain type of micro neural activity, there is no parallel macro manifestation of a social phenomenon such as a “riot.” That is, no events at a level higher than the activities of human beings can be said to constitute an empirical manifestation of a “riot,” whereas the sensation “pain” is manifested both at the macro level (in terms of a psychological experience) and at the micro level (in terms of neural activity). Unlike a riot, we can thus identify pain at the macro level and then (afterward ) start investigating how it is manifested or realized at the micro (neurological) level. The experience “pain” can thus be identified without resorting to neurological-level observation/information, whereas a riot cannot be identified from a vantage point from which we do not have information about the activities of individuals. This is not to say that a whole consisting of individuals could never produce any type of macro-manifest properties. If we, for instance, were sitting in a helicopter at high altitude and made out a group of Buddhist monks engaged in a political protest as a square-shaped, slow-moving orange entity on the ground, all these properties (square shape, slow moving, and orange) would be macro manifested. That is to say, we could identify these properties from a vantage point where we did not have any information about the microelements or their interrelations. However, without micro-level information, it would not be possible to identify what we saw as a phenomenon of interest to us as social scientists. It would not be possible to identify the phenomenon as a political protest, or even as a social phenomenon of any kind. Indeed, social phenomena are recognized as social phenomena by the fact that they involve the activities and interrelations of human beings. The kinds of systems, and the kinds of properties of these systems, that we as social scientists are interested in thus cannot be identified without micro-level information. That is, neither families, companies, nor nations manifest themselves in any other form than in terms of the activities, interrelations, mental properties, and so on, of individuals in physical settings, no matter how far we zoom out. We could, for the same reason, not identify a society as democratic, a company as hierarchical, a nation as torn by civil war, or a family as dysfunctional without information about the activities, interrelations, and properties of individuals.

490

G. Ramström

The level of individuals who act, interact, and experience mental properties is thus the phenomenal level of expression that (both macro and micro) social science is concerned with. The social scientific macro level does not constitute a separate level of manifestation. Individual-level circumstances do not “give rise to” any higher level phenomena that preoccupy the macro social sciences. While the activities of neurons give rise to mental experiences (which psychology is about) and the activities/interactions of atoms give rise to molecular level states and events (which molecular biology is about), no part of social science ever leaves the individual level of phenomenal expression to take an interest in phenomena taking place at another level of phenomenal expression. Social properties are thus micro-based macro properties, rather than macro-manifest macro properties. That is, they are “type B macro” and not “type A macro.” This implies that social macro properties, like all type B macro properties, label, summarize, and describe the lower level circumstances that characterize the disposition and structure of the system. Thus, unlike the macro-manifest macro properties typically discussed in biology, chemistry, or philosophy of mind, social macro properties do not arise from micro-level circumstances. They are not distinguishable phenomena at their own level of manifestation; rather, they are inferred from and refer to sets of micro-level circumstances. While the micro–macro discussion in philosophy of science traditionally has been concerned with the relationship between the micro base and macro-manifest properties—about the physical “gives rise to” type of relationships—the social scientific micro–macro distinction implies an analytical “qualifies as” type of relationship. Philosophy of social science has failed to fully grasp and clearly distinguish between these two types of micro–macro relationships. Indeed, the fact that philosophers use the same word supervenience to depict both kinds of relationships is symptomatic of this lack of clarity. When we say that the social level “supervenes on” (i.e., is determined by) the individual level in social science, we must necessarily mean something quite different than what we mean when we say that facts about molecules “supervene on” facts about atoms or that the mental “supervenes on” the physical. In social science, “supervenes on” implies “is analytically inferred from.” For example, a player on the team scoring a goal “counts as” the team scoring a goal, or the actions of negotiation carried out by a representative of a steel union “qualifies as” the steel union engaging in negotiations (see Zahle 2003, p. 91). As Zahle puts it, “The fact that actions ascribed to collectives, let’s call them group actions,

The Analytical Micro–Macro Relationship in Social …

491

are attributed on the basis of individual action may in more modern terminology be formulated as the point that group actions supervene on individual actions” (Zahle 2003, p. 91). In physics, biology, and the other natural sciences, however—where we typically are talking about type A micro–macro relationships—“supervenes on” instead implies “is physically determined by.” For example, the hardness of a rock “supervenes on” the crystallized atomic structure that makes up its micro level. While this latter use of the term implies a physical constitutive relationship in the actual empirical world (the lower level circumstances synchronically “gives rise to” the higher level circumstances), the alternative social scientific use of the term implies a logical or conceptual relationship—an analytical relationship. The lack of a clear distinction between physical and analytical micro– macro relationships did not start with the popularization of the (fuzzy) term “supervenience” during the last few decades, however. In fact, with Oppenheim and Putnam’s influential categorization of the world into a hierarchy of six levels—(1) elementary particles, (2) atoms, (3) molecules, (4) cells, (5) multicellular living things, and (6) social groups (Oppenheim and Putnam 1958, p. 9)—each connected to distinct scientific disciplines, the stage was already set for this conflation to occur. The movement between individual organisms and groups in Oppenheim and Putnam’s hierarchy is different from the previous four shifts, in that the higher level properties of groups are inferred from observations of, and information about, the organism level. The supervenience relationship (the relationship of determination of higher level properties by lower level properties) between levels 1, 2, 3, 4, and 5 is thus physical, but the supervenience relationship between levels 5 and 6 is analytical. Another way to phrase it is to say that the first four shifts are mechanistically mediated, while the fifth is not. Since the interrelations and behavior of the parts at the individual level do not produce any higher level effects of interest to social scientists, what we are left to explain or understand in social science is why the parts behave or interrelate in the manner in which they do. In disciplines which display macro-manifest higher level properties a second question is actualized: “how” the behavior and interactions between the parts give rise to the macro-manifest property. While there is an answer to how the properties and organization of water molecules give rise to the transparency of water (there is a mechanism we can account for which makes the macro-manifest circumstances intelligible given micro-level circumstances), no such synchronous “how” can be accounted for with regard

492

G. Ramström

to the connection between the behavior and interactions of sets of individuals and the properties of social groups.4 The difference between the physical dependence relations regarding levels 1 through 5 and the analytical relationship between 5 and 6 also shows us why the infinite regress argument against methodological individualism does not hold. Ever since Watkins infamous “rock bottom” claim about individuallevel explanation (Watkins 1957, p. 106) holists have argued there is no reason to stop at the individual level: that is, who is to say that the individual level is the rock-bottom level? Why not the cellular, molecular, or atomic level? The difference between physical and analytical micro–macro relationships shows us the logic of Watkins’s intuition. The individual level is the rock-bottom level in social science because it is the highest manifest level for the discipline. In this light, Watkins’s plea is merely a plea for deabstraction to the highest level of manifestation in social science, rather than a claim that the level of acting and interacting individuals somehow would have a privileged position in relation to other levels of manifestation (only to nonmanifest levels of analysis, that is, privileging manifest states and events to more abstract descriptions of such states and events).

3

A Phenomenally Monist Social Ontology

Regarding the ontological part of the individualism-holism debate in philosophy of social science, it is widely held that the discipline has reached a “consensus view.” The vast majority of philosophers of social science today explicitly adhere to ontological individualism5 (e.g., Epstein 2009, p. 187; List and Spiekermann 2013, p. 630; Udehn 2001, p. 2; Zahle 2007, p. 315; Zahle and Collin 2014, p. 4). Although ontological individualism may be defined in different ways, the two claims most commonly associated with this position are (a) the composition claim and (b) the supervenience (or determination) claim.6

4

See Ramström (2018) for a detailed discussion on so-called micro–macro mechanisms and why they are not found in social science. 5 Proponents of critical realism are the notable exception to this rule. See below. 6 As Zahle and Collin (2014, p. 4) put it, “the majority of theorists tend to be concerned with the composition and determination criteria. More specifically, most participants in the debate regard themselves as ontological individualists by reference to one or both of these criteria.” See, for example, Kincaid (1986, pp. 495–96) for an example of when both criteria are equally emphasized, and List and Spiekermann (2013, p. 629) for an example of when the supervenience criterion is put in focus.

The Analytical Micro–Macro Relationship in Social …

493

The composition claim: The social is composed of constellations of individuals: that is, solely made up of individuals (and their material artifacts). The determination claim: The social is fully determined by the individual level: that is, there can be no change at the social level without a change at the individual level. The problem with formulating ontological individualism in this sense, however, is that it is compatible with both micro–macro phenomenal dualism and micro–macro phenomenal monism. That is, the composition claim and the supervenience claim are both compatible with the view that social macro concepts refer to macro-level phenomenal expressions (i.e., states, events, phenomena, entities, or properties which are independently identifiable at the macro level), which in turn are composed of and determined by individuals (phenomenal dualism). Both claims are, however, also compatible with the view that there is only one phenomenal level at which all states, events, phenomena, entities, or properties are identified/inferred from (phenomenal monism). The analytical nature of the social scientific micro–macro link discussed above clearly implies that the monist perspective is the accurate one. Social phenomena do not take on identifying macro-phenomenal expressions regardless of how far we zoom out. The social scientific micro–macro distinction is analytical, not a reflection of a naturally occurring phenomenal dualism between levels of grain in the empirical world. The micro-manifest character of social states and events thus implies phenomenal monism: social entities, properties, phenomena, states, and events all appear at and all refer to/ are inferred from states and events at the level of individuals who act, interact, and experience mental properties in physical settings. However, as discussed above, most other disciplines (including philosophy of mind and biology) deal with macro-manifest phenomena and thus display phenomenal dualism between micro and macro.7 This difference is important, as ontological individualism is often compared and likened to ontological physicalism in philosophy of mind (e.g., Kincaid 1986, p. 495; List and Spiekermann 2013, pp. 632ff.; Sawyer 2002, p. 543) and to the belief in the ontological reducibility of higher levels to basic particles in biology (Kincaid 1996, p. 188; List and Spiekermann 2013, p. 629). Both these positions are, however, typically characterized by a phenomenally dualist perspective

7 While, as discussed above, the level of individuals who act, interact, and experience mental properties in physical settings is the phenomenal level of expression that both micro and macro social science is concerned with (phenomenal monism), the micro level in philosophy of mind refers to states and events at the neurological level of phenomenal expression and the macro level to states and events at the psychological/mental level of phenomenal expression (phenomenal dualism).

494

G. Ramström

on the micro–macro relationship. In philosophy of mind, ontological physicalism is—just like ontological individualism—based on the composition and the supervenience criteria (that the world is ultimately composed of physical matter and mental states supervenes on the physical states in the brain) and is therefore frequently referred to as the counterpart of ontological individualism. However, very few ontological physicalists are phenomenal monists. Claiming that mental phenomena lack their own level of phenomenal expression in philosophy of mind is considered a radical position, as it goes against our day-to-day experiences of the existence of mental states.8 Ontological physicalists are thus typically not phenomenal monists, just like those who believe in the ontological reducibility to basic particles in biology/ chemistry do not deny that collections of atoms phenomenally appear—and are identifiable as—molecules at a zoomed-out level of grain. Ontological physicalists in philosophy of mind and those who believe in ontological reducibility to basic particles in biology/chemistry are thus typically phenomenal dualists but ontological monists, which means that higher level phenomena are perceived as ontologically reducible to lower level states and events. That is, rather than concluding that phenomenal dualism implies an ontological dualism between mind and body, ontological physicalists argue that the physical brain is responsible for both neurological events and the mental experiences to which they give rise. In social science, however, due to the phenomenal monism implied by the micro-manifest character of social macro phenomena, the aspiring reductionist runs into the problem that there is nothing to reduce. As discussed above, social-level states and events are not empirically determined by individual-level states and events, nor does the former level represent a higher level expression of the latter. The activities, interactions, and mental properties of individuals in physical settings do not represent some lower level base to which social phenomena potentially could be ontologically (or explanatorily, as I will discuss below) reduced; they instead constitute the only (or, if you will, highest) empirical/phenomenal level of the social world.9 8

The claim that mental states/phenomena do not exist is typically discussed as “eliminative materialism.” See, for example, Ramsey (2016) for an overview of this position. See also Footnote 15. 9 The alternative would seemingly be to argue that our interest in the activities of human beings (i.e., in the social world as we experience it) merely is due to their role as indicators of experientially inaccessible higher level phenomena. From such a metaphysical standpoint, one could potentially make the case that social science is individualistic only in an epistemological sense: that we identify social phenomena via micro-level observation in the same way (only reversed) that particle physicists identify the properties of quarks by observing the composite particles quarks combine to form (i.e., hadrons). However, since social scientists, as discussed in the last section, are not interested in the macro-manifest properties which collections of individuals may give rise to from a zoomed-out perspective (such as color, shape, or volume), such a position would seem to have to be based on the

The Analytical Micro–Macro Relationship in Social …

495

Phenomenal monism, of course, also implies property monism. If there is only one phenomenal level (the micro level) at which social phenomena and properties can be identified, then all properties regardless if we characterize them as “micro” or “macro” must be inferred from this level. That is, rather than there being micro-based properties that characterize the microlevel and macro-manifest properties that characterize the macro level, there are only micro-based properties described in micro language, and the same type of properties described using summarizing macro concepts. What makes it possible to talk about property dualism in philosophy of mind, for example, is the phenomenal dualism that implies that one and the same “brain-event” may have mental properties at the mental level of phenomenal expression and neurological properties at the neurological level of phenomenal expression. This type of property dualism is obviously not possible in the phenomenally monist social scientific disciplines. When philosophers of social science talk about “property dualism” in the social realm (e.g., Sawyer 2002, pp. 542ff.) they are thus necessarily referring to something quite different, whether they realize it or not: namely, to the trivial fact that sets of individuals logically have different properties than single individuals (a single individual’s activities can never have the property of being “a riot,” for example). This kind of trivial “one vs. many” property dualism can be equated to the fact that one single neuron cannot display the property “abnormal neural connectivity,” since this property describes a state involving many neurons. Of course, the interesting type of property dualism with regard to the micro– macro discussion concerns the fact that micro-based properties (such as abnormal neural connectivity) in phenomenally dual systems can give rise to distinct macro-manifest properties (such as schizophrenia) at the higher level of phenomenal expression (in this case, the mental level; Ruiz et al. 2013). Not even the most hardline reductionist would question that sets of entities can have properties that single entities cannot display. What separates social property concepts from mental property concepts, however, is that all social property concepts refer to or describe micro-level circumstances (types

type of mystical ontological holism we saw during the nineteenth and early twentieth centuries (e.g., something akin to Hegel’s ideas about “objective spirit” or Durkheim’s “collective consciousness”). Of course, no party in the contemporary individualism-holism debate officially supports such a radical ontological holism or has made any attempts toward addressing any of the many questions such a position would raise, such as what the inaccessible higher level of social facts is and what it consists of, how it relates to our experiences, and through what type of mechanisms individual-level circumstances would give rise to these experientially inaccessible macro phenomena. I discuss the many problems such a hypothetical position would entail in more detail elsewhere (Ramström 2018).

496

G. Ramström

of activities and interactions of individuals), while mental property concepts refer to or describe macro-level circumstances (types of mental experiences).10

4

The Impossibility of Explanatory Holism

Within the individualism-holism debate, it is often argued that the ontological question about whether social phenomena have a sue generis existence should be addressed separately from the methodological question of whether social phenomena could be considered casually efficacious in their own right (e.g., Jepperson and Meyer 2011; Kincaid 1986, 1996; List and Spiekermann 2013; Little 2012; Sawyer 2002, 2003). Ontological individualism, it is argued, does not necessarily imply methodological individualism. As List and Spiekermann formulates it, Facts about molecules or chemical compounds, for example, supervene on (are determined by) facts about the atoms and particles of which they consist. Similarly, facts about Social phenomena supervene on facts about individuals and their interactions. From this, however, it does not follow that all social phenomena can be explained in individualistic terms alone. This mirrors the observation that, despite the supervenience of chemical and biological phenomena on microphysical ones, chemical and biological explanations cannot necessarily be reduced to purely physical ones. In a slogan, “supervenience” does not imply “explanatory reducibility” (as observed by Kincaid [1986] and Sawyer [2002, 2003]). (List and Spiekermann 2013, p. 629)

Such arguments are used to warrant the combining of ontological individualism with explanatory holism.11 This type of contemporary mainstream methodological holism, sometimes called “nonreductive individualism” (NRI),12 accepts ontological individualism but uses the so-called multiple realization argument to maintain that under certain circumstances macro-level phenomena may nevertheless qualify as casually efficacious from a correlational (or counterfactual) view of causality. NRI, which most notably is represented by Kincaid (1986, 1996), Sawyer (2001, 2002, 2003), and List and Spiekermann (2013), has emerged as the main alternative to methodological individualism in the contemporary 10

When Sawyer (2002, pp. 542ff.) compares social science to philosophy of mind and argues that both disciplines are characterized by “property dualism” he seemingly misses this crucial distinction. 11 The same point is sometimes formulated using the composition criteria, as in Little’s statement “composition and explanation can be separated” (Little 2012, p. 146). 12 This is the terminology that Sawyer suggests (Sawyer 2002, 2003), which I will apply henceforth.

The Analytical Micro–Macro Relationship in Social …

497

individualism-holism debate. As explanatory holism based on sociological realism, so-called “causal powers holism” is widely regarded today as either unsustainable (if based on a strong conception of emergence) or simply as methodological individualism rephrased in a collectivist vocabulary (if based on a weak conception of emergence), NRI, to most philosophers of social science, stands out as the only viable contender to methodological individualism.13 The NRI-position (in accordance with the cross-disciplinary character of the micro–macro debate discussed above) builds on an argument that was first introduced in the micro–macro discussion in philosophy of mind, a discipline which saw a development toward consensus on a nondualist ontology (of the mind-brain distinction) similar to that philosophy of social science reached with regard to social ontology during the latter part of the last century. As so-called “ontological physicalism” (or “ontological materialism”)—the idea that everything is ultimately composed of the same elementary particles—grew stronger in philosophy of mind and marginalized perspectives based on substance dualism between mind and brain, those who argued in favor of mental (macro) causation had to find other ways to legitimize their claims. The way forward for most came in the form of the notion of “multiple realization.” In philosophy of mind, the multiple realization argument against reduction (for nonidentity between types of mental states and types of neurological states) states that one and the same kind of mental state (such as “pain”) may be “brought about” or “be realized” in many different (multiple) and potentially unpredictable and unrelated (wildly disjunctive) brain states or patterns of neural activity. That is, although ontological physicalism entails that any token mental state or event is “supervenient on” (determined by) a particular neural state or event, different token instances of neural states or events can give rise to the same kind of mental state or event, thereby preventing any claim of identity between a particular type of mental event/state and a particular neural event/state (see, for example, Baker 2009; Davidson 2001; Fodor 1974; List and Menzies 2009; Putnam 1975; Stoljar 2017).

13

See Zahle (2014) and Wahlberg (2014) for the point that causal-powers holism based on emergence in the weak sense (i.e., relational emergence, as represented by, for example, Elder-Vass 2010) is fully compatible with individualism. Causal-powers holism based on a strong sense of emergence, which would ascribe a causal impact to a social entity at t over and above “the sum (or net effect) of the impacts of its proper parts interrelated as they in fact are interrelated at t ” (Wahlberg 2014, p. 783) would amount to a defense of unexplainable higher level casual powers and is not a position officially represented by any party in the contemporary individualism-holism debate. See also Kaidesoja (2007) and Sawyer (2001, pp. 565ff.) for critical overviews of the causal-powers tradition in social science.

498

G. Ramström

Nonreductive physicalism (or nonreductive materialism), as this position is often called, thus acknowledges that all mental states/events are made possible and are determined by some neural state/event. Yet at the same time it allows for nonidentity between types of mental states and types of neural states in instances where there are wildly disjunctive multiple realizations of mental states.14 Whether or not a particular mental state may have multiple brain state realizations in a wildly disjunctive way is regarded as an empirical question. The potential nonidentity between types of mental states and types of neural states due to multiple realizations, it is argued, implies that a mental state (such as pain) could display a more systematic (lawful) relationship to some outcome variable Y (grimacing, for example) at the macro level than in terms of any of its particular/specific micro-level realizations. If the experience “pain” has numerous potential realizations in terms of neural states, and “pain” always is followed by the pain-experiencing person grimacing, we have a situation where grimacing is more systematically (lawfully) correlated with pain than with any particular neural state (i.e., if pain is absent, grimacing does not follow, but if a particular “pain-realizing” neural state is absent, grimacing could still follow if a different “pain-realizing” neural state is present). This, the nonreductive physicalists contend, implies that it would be more correct to pose “pain” as the cause of grimacing than any specific neural state or a limited set of them. In such instances, it is argued, higher level (psychological) causal laws are thus irreducible to lower level (neurological) laws.15 Nonreductive individualists assume that the same argument can be applied to social science. They argue that a social outcome variable Y may be more systematically (lawfully) correlated with a social macro phenomenon X than with any particular micro-level realization of X , rendering social causal laws irreducible and social phenomena causally efficacious (nonepiphenomenal; Kincaid 1986, 1996; List and Spiekermann 2013; Sawyer 2001, 2002, 2003). However, as discussed in the previous sections, the social scientific micro– macro relationship is different from that found in philosophy of mind. Social macro phenomena are micro-based macro phenomena rather than macro-manifest macro phenomena. That is, as opposed to the macro level in philosophy of mind, the macro level in social science lacks its own level of 14 The “wild disjunction” criterion is necessary, since it blocks reducibility to a finite list of predictable realizations or to some common denominator of “different” realizations. Sawyer writes, “Multiple realizability alone does not necessarily imply irreducibility; if there are only a few realizing states, or if those states display some common features” (Sawyer 2001, p. 557). 15 List and Menzies (2009) arguably present the best-developed version of this argument to date. See, for example, Sawyer (2003, pp. 214ff.) for an overview of earlier versions.

The Analytical Micro–Macro Relationship in Social …

499

phenomenal expression. The social macro level is merely analytical: it consists of summarizing concepts and mathematical computations (statistics) which describe the micro level. The argument thus runs into a fundamental problem when applied to social science: there are no social macro-manifest phenomena to save from epiphenomenalism. In philosophy of mind, both explanatory reductionists and nonreductionists agree that macro concepts refer to mental experiences (appearing empirically as experiences) and not to neural-level states or events. In biology and chemistry, the same argument also has been used by opponents of explanatory reduction and all parties similarly agree that molecular concepts refer to molecular-level states/events that appear at the molecular level of zoom/grain and not to atomic-level states or events. In these disciplines, the discussion about macro causation versus explanatory reduction thus starts from the premise that we are dealing with two separate levels of phenomenal expression, where the higher level phenomena could either be causally efficacious or epiphenomenal.16 In philosophy of social science, however, this premise is not present. As discussed in the previous section, the social world is phenomenally monist. In applying the multiple-realization argument to social science, nonreductive individualists thus make the error of treating micro-based social properties that relate to the micro level analytically as if they were macro-manifest properties that relate to the micro level empirically. Indeed, NRI-proponents (just like nonreductive physicalists in philosophy of mind) explicitly claim that the micro realization of a social macro phenomenon is an empirical question (e.g., Keith Sawyer 2012, p. 272; Kincaid 2014, pp. 142, 147; List and Spiekermann 2013, p. 641). As discussed above, however, this is not the case. In social science, macro properties refer to/are inferred from micro properties based on an analytical “qualifies as” or “equates to” move rather than an empirical “gives rise to” move. Since social concepts, as discussed above, do not refer to macro-manifest properties of collective entities (such as shape, color, or volume), the empirical content of social variables that such concepts depict necessarily must be located at the micro level. This fact implies that NRI cannot be sustained, and that we must let go of the idea that we can distinguish between micro and macro causal relationships in social science. Nothing arises at the social scientific macro level; nothing takes place or 16

As Kim (1989, p. 43) points out, “nonreductive physicalism is not to be a form of eliminativism; that is, it acknowledges the mental as a legitimate domain of entities... as a noneliminativist the nonreductive physicalist believes that there are events in her ontology that have mental properties (e.g., being in pain, being a belief that snow is cold, etc.).” To be a nonreductive physicalist, according to Kim, implies that one believes that an event causes another event “in virtue of its mental property” (Kim 1989, p. 43) rather than in virtue of its physical property.

500

G. Ramström

happens at this level. To try to save the social scientific macro level from epiphenomenalism thus equates to trying to save analytical constructs and/or mathematical computations (statistics) from epiphenomenalism. While the multiple realization argument from philosophy of mind may be used to support macro causation in a domain characterized by the ontological reducibility of higher level phenomena (e.g., the body-mind domain), it makes no sense to apply it to a domain where there is no higher level of phenomenal expression. We could never have a macro-manifest social phenomenon (X-macro) which was more systematically or “lawfully” correlated to a particular outcome (Y ) than its micro realizations (X-micro) were, simply because there are no macro-manifest social phenomena (Xmacro). As discussed above, the micro (individual level) realization of a social phenomenon is its highest realization. Phrased differently, the idea that supervenience does not necessarily imply explanatory reducibility may very well be correct when we are dealing with empirical supervenience relationships, but it does not apply to analytical ones. One could thus apply the multiple-realization argument to claim that causal relationships between social phenomena (i.e., causal relationships between sets of activities, interactions, and mental properties of human beings in physical settings) could not be explanatorily reduced to causal relationships at lower empirical levels (e.g., to descriptions of sets of causally related events at the cellular or neurological level). However, it makes no sense to apply this argument to the analytical/conceptual distinction between “social facts” and facts about interacting human beings. Given that the micro–macro discussion in philosophy of science traditionally has concerned the relationship between the micro base and macromanifest macro properties (as discussed in part 1), the use of the term multiple realization within this discipline has typically referred to a situation where multiple different lower level circumstances can “give rise to” one and the same type of empirical phenomena at the higher level. In the social scientific context, however, if macro concepts are to have any empirical content at all, they necessarily must refer to micro-level circumstances, either directly or indirectly. The most intuitive conclusion would seem to be that social macro concepts refer directly to micro circumstances, that is, that the micro realization of a macro phenomenon is stipulated by the empirical definition of the concept that is used to depict the phenomena in question (I will address less intuitive alternatives later on).17 To translate a macro causal law into micro language 17

It should be noted that the empirical content that we assign to a variable, that is, its empirical definition is something very different from its operational definition. An empirical definition gives

The Analytical Micro–Macro Relationship in Social …

501

would thus merely mean using the empirical definition of the concept rather than the abstract summarizing macro label. If, for example, we concluded that the type of independent variable “riot” was associated with some type of outcome variable Y , then a “reduction” of this casual statement would merely mean that we were explicit about which kinds of empirical circumstances we had observed to correlate with Y : that is, that we were explicit about what we referred to in using the label “riot” (e.g., “any instance when at least one hundred people engage in violent public disturbance resulting in some form of property damage or bodily harm”). Multiple realization and wild disjunction would obviously be out of the question, since the whole reason for using a common concept X to label A, B, and C is because A, B, and C have something in common that warrants the use of a common label. It is the thing that is common to A, B, and C that would make them instances of X , and it is to this commonality we would refer when using the concept X . In fact, this is the position taken by Mitrovic in his critique of the social scientific use of the multiple-realization argument. He argues that, for the individualist, multiple realization will not occur, since for such scholars, social entities and phenomena are the sets of individuals and their interactions that have certain, specific properties that define them. If the properties of the actions and interactions of a set of individuals satisfy the definition of a bank, then these individuals and their interactions are a bank; if the properties of the interactions of a set of individuals satisfy the definition of a revolution, then these interactions are a revolution. (Mitrovic 2017, p. 35)

While Mitrovic’s argument—that this perspective follows logically from ontological individualism—is unsustainable (as discussed above, ontological individualism is fully compatible with phenomenal dualism and thus with the notion that macro concepts have macro referents), it would undoubtedly be the most natural way to perceive the relationship between macro concepts and their referents if we accepted the phenomenally monist ontology suggested in the current article.18 From such a perspective, it would be logical to perceive empirical content to a variable (X ) by explicating to what states or events in the world it refers. An operational definition, on the other hand, constitutes a specification of which indicator-values results in variable X being identified/registered as present if observed in a particular case (e.g., the presence of police or media reports may be used to determine if a “riot” has occurred in a particular case, or survey-answers may be used to learn about people’s beliefs and desires). 18 Mitrovic interprets ontological individualism (OI) as implying that social properties are properties of sets of individuals and their interactions (i.e., micro-based/microstructural properties; Mitrovic 2017, p. 40). However, this is seemingly not how nonreductive individualists tend to interpret OI, all of whom, as discussed above, explicitly claim that the micro-realization of a macro phenomenon is an empirical question rather than an analytical or definitional one, which would be the case if they believed that social predicates simply referred to properties of individuals and their interactions.

502

G. Ramström

the social scientific micro–macro link as identical to a purely conceptual “concept-referent” link. Such a micro–macro link would be incompatible with multiple realizations. When the micro–macro link is empirical, it is the structure of the world that determines how a macro phenomenon is realized at the micro level and multiple realization may very well occur, however, when the micro–macro link is conceptual, it is the way we use a particular word that determines how a phenomenon is “realized,” and the logic of language implies that we do not use one and the same word to label a multitude of unrelated referents.19 To illustrate this fact, we may take a look at List and Spiekermann’s argument that the variable “democracy” in democratic-peace theory is a multiply realizable variable.20 List and Spiekermann argue that it would be practically impossible to list all the ways the higher level property “being a democracy” could be instantiated at the individual level, and that it thus would not be feasible to phrase the democratic-peace causal law in micro-level terms: However, listing all possible individual-level constellations that can instantiate “being a democracy” is practically impossible. Furthermore, the point of the democratic peace hypothesis is that there are structural features of democracies that lead them to avoid wars with one another, independently of the individual political details of each case (e.g., the personalities of the relevant politicians, the specifics of the public debate on the issue, the voting pattern in the legislature, and even the details of the democratic constitution itself ). For this reason, if there is a causal relation between “being a democracy” and

Indeed, Kincaid, as the original translator of the nonreductionist position into social science, and as one of the nonreductive individualism (NRI) proponents that Mitrovic argues against, clearly distinguishes between social properties and microstructural properties. Far from equating the two, Kincaid (1986, pp. 510–11) claims that just as in other scientific disciplines, facts about microstructural circumstances (i.e., information about “individuals and their interrelations,” Kincaid 1986, p. 511) at best could provide an “explanatory extension” (Kincaid 1986, p. 511) to our understanding of macro social phenomena. A successful rebuttal of NRI must thus show these underlying assumptions to be incorrect, that is, it must show that social phenomena are in-fact micro manifest, and that the social scientific micro–macro relationship consequently is purely analytical. 19 Indeed, even if we defied the logic of language and used the same word X to refer to a multitude of unrelated states, events, properties, or entities, in a causal X –Y theory it is hard to see how anyone would accept these unrelated states/events as instances of the same type of cause. 20 This particular example was chosen because it represents the best-developed example of a macro causal relationship that has been presented by the nonreductive individualist camp to date. Indeed, it is rare to find examples of such alleged relationships at all in the NRI-literature. Sawyer, for example, tends to return to the example of “church” as a multiply realizable social phenomenon in his writings (Sawyer 2001, p. 557; 2002, p. 542), but he never presents any examples of macro causal relationships in which this phenomena potentially could be involved. Regardless, the counterargument against the independent variable “being a democracy” in the democracy–peace relationship being multiply realizable applies equally well to the hypothetical variable “being a church.”

The Analytical Micro–Macro Relationship in Social …

503

“avoiding wars with other democracies,” it must be micro realization robust. (List and Spiekermann 2013, p. 640)

However, in making this argument List and Spiekermann seem to be forgetting that democracy (like all social properties/phenomena) is a microbased macro fact rather than a macro-manifest macro fact. While, for example, the organization of water molecules is a part of the lower level realization of the transparency of water in an empirical sense,—that is, it is part of what physically/empirically (synchronically) “gives rise to” the macro-level phenomena that the concept “transparency” refers to—the same is not true with regard to the relationship between the interactional patterns of individuals in a society and a society being democratic. While the former property depends on the micro structure, the latter refers to it. What kinds of properties and organization of microelements “gives rise to” the transparency of water is an empirical question, while what kinds of properties and organization of microelements “qualify as” democracy is determined analytically (by definition). The transparency of water could thus theoretically be produced by several different and potentially unexpected sets of micro circumstances. When it comes to democracy, on the other hand, it is a certain set of micro circumstances that “makes” the property. In other words, if societies A, B, and C are all democratic, it is because they all display the micro-level circumstances which qualify as “democracy” according to our empirical definition of this concept. It is, of course, true in a logical sense that democracy would not be present in a particular case if the interactional patterns of individuals were different (e.g., more hierarchical). However, this is not due to the fact that the phenomenon we label democracy suddenly would not emerge at some higher level if individuals interrelated differently; rather, it is because we would not label such an interactional (microstructural) pattern between individuals “democracy.” The very existence of democratic-peace theory itself must thus necessarily be based on the observation of micro circumstances. This, of course, implies that it is not only practically possible to define what a democracy is in micro-level terms but that such a definition logically must have been used by those who discovered the correlation that List and Spiekermann are discussing. Democratic peace theory is a theory about the empirical relationship between two kinds of phenomena, and since social phenomena are micro manifest, it could only have been observed at the micro level (i.e., the level of individuals that are acting, interacting, and experiencing in physical settings). If we were unable to summarize what type of individual-level activities makes a country “democratic,” we would not be able to determine

504

G. Ramström

whether countries that wage wars are democratic or not. In short, there would be no democratic-peace hypotheses to begin with.21 List and Spiekermann’s argument is not only fueled by a conflation between physical and analytical supervenience relationships, however; the authors also seem to confuse the fact that a type of phenomena can vary in its nonidentifying characteristics, with the fact that a higher level phenomenon/ cause can vary in its lower level realization. The fact that the personalities of relevant politicians, the specifics of the public debate, or voting pattern can vary between different cases that qualify as democracies in no way implies that different and unrelated circumstances may “give rise to” or “qualify as” democracy (i.e., multiple realizations). Rather, it means that the criteria that stipulate what qualifies as a democracy do not include these variables. If these circumstances can vary, it is because they do not define or realize what makes a democracy a democracy. With List and Spiekermann’s logic, it would be equally impossible to specify, for example, what a “tennis player” is, since tennis players may be of any number of different ages or ethnicities and display a host of different physical and psychological attributes. Of course, the very reason why tennis players can display such differences and are still all considered tennis players is because the thing that realizes a tennis player as a tennis player is stipulated by the definition, and such definitions typically do not refer to age, ethnicity, physical, or psychological characteristics but rather to the fact that someone regularly plays tennis. The fact that what makes phenomena A and B be of a certain type (X ) is some exclusive set of criteria, while other circumstances which have nothing to do with A and B qualifying as X may vary between A and B, is true with regard to all “type” identifications for all phenomena at all levels. It has nothing to do with the micro–macro issue or with the notion of multiple realization. Just like age or weight may vary between tennis players, the personalities of relevant politicians and the specifics of the public debate may vary between democracies because the definition of “democracy” used by political scientists and international relations scholars does not refer to these circumstances. Just as the fact that the climate may be different in different democracies does not prevent us from stating what counts as “democracy” in micro/empirical terms (e.g., citizens are allowed to vote in reoccurring elections and to express their opinions freely), the fact that the personalities of politicians may be different in different democracies does not prevent us from arriving at a definition. 21

See Mitrovic (2017), as well as Greve (2012), for similar arguments regarding the unidentifiability of social phenomena when individual-level information is lacking.

The Analytical Micro–Macro Relationship in Social …

505

Indeed, the argument cannot be applied to social science even if we go out of our way and try to define social macro concepts in ways that would allow multiple realizations. That is, even if we could break the direct/explicit conceptual relationship between social predicates and particular types of micro-level circumstances by using abstract or nonempirical definitions, particular micro circumstances would nevertheless always be implied, provided the identifying characteristics of the variable were manifested at the individual level (which is true for all social phenomena/properties, as discussed above). Nonreductive individualist Harold Kincaid, for example, has argued that “multiple realizations are particularly likely in the case of high-level causal claims that abstract from individual detail or when we define terms by their role vis-a-vis other social factors” (Kincaid 1996, p. 154). More specifically, Kincaid argues that social macro phenomena/entities can be defined by reference to their causal/functional relations to other social macro phenomena, rather than by reference to their empirical manifestations. Kincaid (1996, 164) uses the example of the social entity “state” to demonstrate what such a functional definition could look like: “The state, for example, is a social entity, one which is surely defined by its social role and properties. It is, at least in part, the entity which has more control over organized violence than any other institution.” However, the problem with Kincaid’s argument is that regardless of definitions, micro and macro share a common empirical level of phenomenal expression in social science (as discussed above) and macro phenomena thus necessarily must be inferred from micro circumstances. Before we start explaining a function of a social entity, we obviously have to identify its existence, and since the micro-level realization of a function of a social entity is its only realization, the micro realization of the function of a social entity must be determined before it can be identified, not to mention explained. That is, explaining a function (X ) of a social entity equates to explaining the (micro)manifest states and events from which X is inferred. Phrased differently, the micro-manifest nature of the social world implies that the actions of social macro entities upon which their functional roles are identified are assigned to these entities based on the actions of individuals (a fact Kincaid himself acknowledges 1996, pp. 155–56). If we take a look at Kincaid’s definition of “state” above, we can illustrate this problem. Kincaid’s definition of state not only presupposes the existence of a state in all societies (which would make political scientists’ claim that there are stateless societies untrue per definition), but more importantly it raises the question of what, in the empirical world (i.e., in the world of the manifest) we refer to with the concept “entity”? We obviously need an answer to this

506

G. Ramström

question before we can start investigating and comparing different “entities” with regard to their relative control of violence. We similarly need to know what “organized violence” refers to in terms of empirical events (in terms of the activities of human beings). Once we have answers to all such questions, we can identify “states,” but we would also have defined “state” in empirical terms: that is, in micro-level terms. Thus, defining social phenomena in functional or other abstract terms in no way makes unpredictable realizations or “wild disjunction” possible in social science (as opposed to in philosophy of mind, for example, where the identification of functions can be carried out at a “separate” macro level of phenomenal expression, that is, at the level of mental experiences and bodily responses). It merely makes the stipulation of what qualifies as a particular macro phenomenon/property in micro-level terms indirect or implicit, rather than direct and explicit. The fact that there are no macro-manifest social phenomena thus not only blocks the possibility of using states and events at this level of phenomenal expression as direct referents of social concepts (as discussed above) but it also, naturally, blocks the possibility of producing functional definitions based on states and events at this level of phenomenal expression. Whereas a mental experience is an event that can be ascribed a functional role without any knowledge or investigation of its (potentially multiple) neural-level realizations, if we wish to ascribe a type of social state, event, or entity a functional role, it would necessarily have to be a micro-manifest state/event/entity. In social science, a functional role can thus not be found to be fulfilled in a surprising or unexpected way. Rather, such a role is identified from the bottom up; the criteria for fulfilling the functional role must be formulated in terms of micro-level states and events. How a social functional role is realized at the micro level must thus be known before the functional role can be identified. Which is to say, its realization is a definitional rather than an empirical question.

5

Conclusion

Social macro phenomena are micro-based macro phenomena rather than macro-manifest macro phenomena. The relationship between micro and macro in social science is thus analytical; that is, the social macro level merely consists of concepts and statistics which summarize and describe the empirical micro level. The argument for macro causation is borrowed from philosophy of mind, which builds on the presence of an empirical micro–macro relationship, thus cannot be applied to social science. Phrased differently: the

The Analytical Micro–Macro Relationship in Social …

507

nonreductionist argument NRI is based on is modeled to fit a phenomenally dualist ontology and not a phenomenally monist ontology. The multiple realization argument is logically incompatible with social science, but more fundamentally, the monist social ontology implies that it has no potential purpose to serve in the social realm. NRI applies an antireductionist argument to a macro level that harbors nothing to reduce. While the ultimate purpose of arguing for multiple realization in philosophy of mind always has been to defend free will by saving mental phenomena (such as impulses, ideas, desires, beliefs, and so on) from being deemed causally inefficacious byproducts (epiphenomena) of neurological processes, the social scientific use of the multiple realization argument fills no comparable purpose. Given phenomenal monism and the fact that methodological individualists do not oppose the use of macro concepts as convenient shorthand, in practice all NRI amounts to is a defense of the right not to empirically define variables in social scientific theorizing.22 As discussed above, however, to conduct empirical scientific inquiry, we need empirical definitions, either explicit or implicit. I find it hard to believe that NRI proponents would consciously question this fact, or the prevailing paradigm in explanatory theory that we should be as explicit as possible about what (in the world) we theorize to cause or correlate with what.23 Rather, I believe that NRI is simply based on a conflation between macro-manifest macro properties and micro-based macro properties (and thus between empirical and analytical supervenience relationships). This is a mistake they certainly are not alone in making, as the conflation is more or less built into the cross-disciplinary character and vocabulary of the micro–macro discussion in philosophy of science.24 Hopefully, however, we will be able to prevent similar metatheoretical slip-ups from occurring in the future by clearly distinguishing between the analytical character of the social scientific micro–macro relationship and the 22

If a concept does not refer to anything in particular at the micro level and there is no empirical macro level to which it could refer, there is no empirical definition we could give the concept. That is, it effectively does not refer to any particular phenomenon at all. 23 I would thus be surprised if the conscious objective of NRI-proponents were to encourage explanatory theorists to cloak or conceal what empirical relationships their theories are actually about (e.g., by rewriting their definitions in functional or other abstract terms). 24 Indeed, the idea of “micro–macro mechanisms,” which figures prominently in Coleman’s metatheory, as well as in the contemporary meta-theory of the so-called mechanistic movement (e.g., analytical sociology and process tracing) is based on the same conflation. As scholars in this camp typically do not adhere to NRI or rely on the notion of macro causation, I will not address this issue further here, however, I discuss it at length elsewhere (Ramström 2018).

508

G. Ramström

empirical micro–macro relationships found in most other fields of science and by explicitly including phenomenal monism as a part of our conception of ontological individualism.

References Archer, M. (1995). Realist Social Theory: The Morphogenetic Approach. Cambridge: Cambridge University Press. Baker, L. R. (2009). “Non-Reductive Materialism.” In A. Beckermann, B. P. McLaughlin, & S. Walter (Eds) The Oxford Handbook of Philosophy of Mind (pp. 109–27). Oxford: Oxford University Press. Bouvier, A. (2011). “Individualism, Collective Agency and the Micro-Macro Relation.” In I. C. Jarvie & J. Zamora-Bonilla (Eds) The Sage Handbook of the Philosophy of Social Sciences (pp. 199–216). London: Sage. Churchland, P. M. (1985). “Reduction, Qualia, and the Direct Introspection of Brain States.” The Journal of Philosophy, 82(1), 8–28. Craver, C. F. & Bechtel, W. (2007). “Top-Down Causation without Top-Down Causes.” Biology & Philosophy, 22(4), 547–63. Davidson, D. (2001). Essays on Actions and Events. Oxford: Oxford University Press. Elder-Vass, D. (2010). The Causal Power of Social Structures: Emergence, Structure and Agency. Cambridge: Cambridge University Press. Epstein, B. (2009). “Ontological Individualism Reconsidered.” Synthese, 166 (1), 187–213. Fodor, J. A. (1974). “Special Sciences (Or: The Disunity of Science as a Working Hypothesis).” Synthese, 28(2), 97–115. Greve, J. (2012). “Emergence in Sociology: A Critique of Nonreductive Individualism.” Philosophy of the Social Sciences, 42(2), 188–223. Humphreys, P. (1997). “How Properties Emerge.” Philosophy of Science, 64 (1), 1– 17. Jepperson, R. & Meyer, J. W. (2011). “Multiple Levels of Analysis and the Limitations of Methodological Individualisms.” Sociological Theory, 29 (1), 54–73. Kaidesoja, T. (2007). “Exploring the Concept of Causal Power in a Critical Realist Tradition.” Journal for the Theory of Social Behaviour, 37 (1), 63–87. Kemeny, J. G. & P. Oppenheim. (1956). “On Reduction.” Philosophical Studies: An International Journal for Philosophy in the Analytic Tradition, 7 (1–2), 6–19. Kim, J. (1989). “The Myth of Nonreductive Materialism.” Proceedings and Addresses of the American Philosophical Association, 63(3), 31–47. Kim, J. (1999). “Making Sense of Emergence.” Philosophical Studies: An International Journal for Philosophy in the Analytic Tradition, 95 (1), 3–36. Kincaid, H. (1986). “Reduction, Explanation, and Individualism.” Philosophy of Science, 53(4), 492–513.

The Analytical Micro–Macro Relationship in Social …

509

Kincaid, H. (1996). Philosophical Foundations of the Social Sciences: Analyzing Controversies in Social Research. Cambridge: Cambridge University Press. Kincaid, H. (2014). “Dead Ends and Live Issues in the Individualism-Holism Debate.” In J. Zahle & F. Collin (Eds) Rethinking the Individualism-Holism Debate: Essays in the Philosophy of Social Science (pp. 139–52). Cham: Springer. List, C. & Menzies, P. (2009). “Nonreductive Physicalism and the Limits of the Exclusion Principle.” The Journal of Philosophy, 106 (9), 475–502. List, C. & Spiekermann, K. (2013). “Methodological Individualism and Holism in Political Science: A Reconciliation.” American Political Science Review, 107 (4), 629–43. Little, D. (2012). “Explanatory Autonomy and Coleman’s Boat.” Theoria, 74, 137– 51. Mitrovic, B. (2017). “Is Multiple Realizability a Valid Argument Against Methodological Individualism?” Philosophy of the Social Sciences, 47 (1), 28–43. Munch, R. & Smelser, N. J. (1987). “Relating the Micro and Macro.” In J. C. Alexander, B. Giesen, R. Munch, & N. J. Smelser (Eds) The Micro-Macro Link (pp. 356–87). Berkeley: University of California Press. Nagel, E. (1961). The Structure of Science: Problems in the Logic of Scientific Explanation. New York: Harcourt, Brace & World. Oppenheim, P. & Putnam, H. (1958). “Unity of Science as a Working Hypothesis.” Minnesota Studies in the Philosophy of Science, 2, 3–36. Putnam, H. (1975). “The Nature of Mental States.” In H. Putnam (Ed.) Philosophical Papers: Mind, Language and Reality, Vol. 2 (pp. 429–40). Cambridge: Cambridge University Press. Ramsey, W. (2016). “Eliminative Materialism.” In N. Edward (Ed.) The Stanford Encyclopedia of Philosophy, Winter 2016 edition. https://plato.stanford.edu/arc hives/win2016/entries/materialism-eliminative/. Ramström, G. (2018). “Coleman’s Boat Revisited—Causal Sequences and the Micro-Macro Link.” Sociological Theory, 36 (4), 368–91. Ritzer, G. (2000). Modern Sociological Theory. Boston: McGraw-Hill. Ruiz, S., Birbaumer, N. & Sitaram, R. (2013). “Abnormal Neural Connectivity in Schizophrenia and fMRI-Brain Computer Interface as a Potential Therapeutic Approach.” Frontiers in Psychiatry, 4, 17. Sawyer, R. K. (2001). “Emergence in Sociology: Contemporary Philosophy of Mind and Some Implications for Sociological Theory.” American Journal of Sociology, 107 (3), 551–85. Sawyer, R. K. (2002). “Nonreductive Individualism.” Philosophy of the Social Sciences, 32(4), 537–59. Sawyer, R. K. (2003). “Nonreductive Individualism Part II—Social Causation.” Philosophy of the Social Sciences, 33(2), 203–24. Sawyer, R. K. (2012). “Response to ‘Emergence in Sociology.’” Philosophy of the Social Sciences, 42(2), 270–75.

510

G. Ramström

Stoljar, D. (2017). “Physicalism.” In Edward N. Zalta (Ed.) The Stanford Encyclopedia of Philosophy, Winter 2017 edition. https://plato.stanford.edu/archives/win 2017/entries/physicalism/. Udehn, L. (2001). Methodological Individualism: Background, History and Meaning. London: Routledge. Vromen, J. (2010). “MICRO-Foundations in Strategic Management: Squaring Coleman’s Diagram.” Erkenntnis, 73(3), 365–83. Wahlberg, T. H. (2014). “Elder-Vass on the Causal Power of Social Structures.” Philosophy of the Social Sciences, 44 (6), 774–91. Watkins, J. W. N. (1957). “Historical Explanation in the Social Sciences.” The British Journal for the Philosophy of Science, 8(30), 104–17. Wimsatt, W. C. (1985). “Forms of Aggregativity.” In A. Donagan, N. Perovich, & M. Wedin (Eds) Human Nature and Natural Knowledge (pp. 259–93). Dordrecht: D. Reidel. Wimsatt, W. C. (2000). “Emergence as Non-Aggregativity and the Biases of Reductionisms.” Foundations of Science, 5 (3), 269–97. Zahle, J. (2003). “The Individualism-Holism Debate on Intertheoretic Reduction and the Argument from Multiple Realization.” Philosophy of the Social Sciences, 33(1), 77–99. Zahle, J. (2007). “Holism and Supervenience.” In S. P. Turner & M. W. Risjord (Eds) Philosophy of Anthropology and Sociology (pp. 311–41). Amsterdam: Elsevier. Zahle, J. (2014). “Holism, Emergence, and the Crucial Distinction.” In J. Zahle & F. Collin (Eds) Rethinking the Individualism-Holism Debate: Essays in the Philosophy of Social Science (pp. 177–96). Cham: Springer. Zahle, J. & Collin, F. (2014). “Introduction.” In J. Zahle & F. Collin (Eds) Rethinking the Individualism-Holism Debate: Essays in the Philosophy of Social Science (pp. 1–14). Cham: Springer.

Methodological Individualism and Other Major Intellectual and Scientific Traditions

The ‘Individual’ in the Writings of Émile Durkheim Anthony Giddens

During his life-time, Durkheim’s methodological writings were notoriously subject to controversy, and his ‘sociologism’ was widely condemned. These early critiques, often totally misinformed as to the real character of Durkheim’s views, have long since ceded place to critical interpretations of Durkheim’s writings which are founded upon a more adequate understanding of the themes and the dilemmas inherent in his sociology.1 Nonetheless, it is arguable that we still await a treatment, which fully explores the strengths and weaknesses of Durkheim’s method. One main reason for this is that most secondary interpreters of Durkheim have failed to connect his analytical discussion (and rejection) of individualism as a methodological approach to social theory with his developmental conception of the emergence of individualism as a morality brought into being by the growth of the differentiated division of labour. It is commonly accepted—and, indeed, he himself stressed this very strongly—that Durkheim’s methodological ideas must be 1

Perhaps the best of the earlier studies of Durkheim is Roger Lacombe (1927).

Giddens, A. (1971). The ‘Individual’ in the Writings of Émile Durkheim. European Journal of Sociology, 12(2), 210–228. © Archives Européennes de Sociologie.

A. Giddens (B) Sociology, Cambridge University and London School of Economics, London, UK e-mail: [email protected]

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 N. Bulle and F. Di Iorio (eds.), The Palgrave Handbook of Methodological Individualism, https://doi.org/10.1007/978-3-031-41512-8_23

513

514

A. Giddens

evaluated in relation to their concrete implementation in his more empirical works. But this is generally taken to mean showing how successfully or otherwise he ‘applied’ his methodological views in his other studies. The point I wish to make in this paper is that there is a reciprocal relationship between Durkheim’s substantive discussion of the development of individualism and his abstract formulations of sociological methods. Durkheim is often regarded as being fervently “anti-individualist”. But, in fact, his works contain a vigorous defence of individualism—understood in a specific way. In other words, Durkheim’s writings represent an attempt to detach “liberal individualism”, regarded as a conception of the characteristics of the modern social order, from “methodological individualism”.2 It is necessary, of course, to specify the main senses in which the term ‘individualism’ appears in Durkheim’s writings. In his very earliest works, such as his long review article devoted to a study of certain leading contemporary German thinkers (Durkheim 1887a), he uses the term indiscriminately to refer to any branch of social philosophy which accords the ‘individual’ some sort of definite primacy over ‘society’—whether in ethical terms (as in Kant) or in methodological terms (as in utilitarian philosophy). But, he soon came to perceive more clearly that there is an essential difference between these two types of ‘individualist’ philosophy, and the elaboration of the precise nature of this difference became a dominant theme in his own sociological perspective. Whereas utilitarian individualism must be rejected as a methodology—sociology cannot be based upon a theory, which treats the individual as the starting point of the analysis—ethical or “moral individualism” refers to a social process, which is of the greatest significance in modern society. The latter form, which Durkheim also frequently refers to as the “cult of the individual”, is created by society: it is this very fact, which demonstrates the inadequacy of utilitarianism as a social theory, because what it takes as its premise is actually the outcome of a long-term process of social development. Durkheim’s discussion of the origins and nature of moral individualism thus constitutes one major dimension of his attempt to meet the age-old question of the relationship between the individual and society. The general framework of this position is set out in De la division du travail social , but it is substantially clarified in later writings.3 The primary question which Durkheim sets himself in the former work is as follows:

2

This point is mentioned by Steven Lukes (1969) in introducing his translation, Durkheim’s “Individualism and the Intellectuals”. 3 My article Giddens (1971a), forthcoming.

The ‘Individual’ in the Writings of Émile Durkheim

515

Comment se fait-il que, tout en devenant plus autonome, l’individu dépende plus étroitement de la société? Comment peut-il être à la fois plus personnel et plus solidaire ? Car il est incontestable que ces deux mouvements, si contradictoires qu’ils paraissent, se poursuivent parallèlement [How is it that, while becoming more autonomous, the individuals become more closely dependent on society? How can they be both more personal and more supportive? For it is indisputable that these two movements, however contradictory they may seem, go on in parallel]. (Durkheim 1960 [1893], p. xliii)

The formulation of the problem has become famous; its resolution, however, has frequently been represented as ambiguous by those who have sought to render the substance of Durkheim’s argument. For it has seemed as though, while mechanical solidarity refers to a moral attachment between the individual and society, organic solidarity refers to a purely economic relation—functional interdependence within the division of labour—binding individuals in a social unity. Thus it appears as though it is only somewhere near the middle of the work that Durkheim perceives that organic solidarity must have a moral underpinning in the same way as mechanical solidarity does: and hence that there is a crucial shift in his argument from the first to the concluding half of the book.4 But, in Durkheim’s own eyes at least, there is clearly no such transition in his thinking in the work. Some years prior to the publication of De la division, Durkheim made just this point in criticising Tönnies’ differentiation of Gemeinschaft und Gesellschaft: that, while Tönnies recognises the moral basis of solidarity in traditional societies, his portrayal of Gesellschaft treats modern society in a way comparable to that of the utilitarians, as lacking the moral character of the traditional type. Where, then, is the main line of Durkheim’s argument? It lies, in fact, in the proposition that there is a duality in the relationship between the individual and society, a duality whose balance becomes radically altered with the movement from mechanical to organic solidarity. In mechanical solidarity each individual remains largely unconscious of his ‘separateness’ as an individual since, dominated by the conscience collective, he shares similar traits with the other members of the society; the limits of his autonomy are strictly bounded. The paradox in this, an apparent paradox only, is that the strength of the moral integration of the conscience collective is inversely related to the strength of the ties which attach the individual to the group; like a simple organism, such a society can shed individuals, and even whole segments of itself, without difficulty. The characteristic of organic solidarity, on the other hand, is that the attachment of the individual to the conscience collective is mediated by 4

See, for example, Nisbet (1965), p. 37 and passim.

516

A. Giddens

his ties to other groups: especially, of course, those created by occupational specialisation in the division of labour. Durkheim is never in any doubt that these are moral ties—that is, that there is a “non-contractual element” which governs the negotiation of contracts. The point is that this presupposes moral diversity rather than uniformity: the “universal man” of traditional society is replaced by modern “specialist man”. The greater self-autonomy which is not only allowed for, but thereby demanded, of the individual, however, does not detach him from society, but actually increases the strength of the reciprocal tie between them. The main issue which is not fully clarified in De la division is not the nature of organic solidarity itself, but the relationship between the “morality of specialisation” generated by organic solidarity, and the morality of the conscience collective, which becomes focussed upon the “cult of the individual”; because Durkheim (1960 [1893], p. 396) makes it clear that “il restera toujours, tout au moins, ce culte de la personne, de la dignité individuelle […] qui, dès aujourd’hui, est l’unique centre de ralliement de tant d’esprits [there will always remain, at least, this cult of the individual, of individual dignity […] which, from today, is the only rallying center of so many minds]”. What does this “cult of the individual” refer to, and how is it compatible, as a general, shared morality, with the moral specialisation enjoined by organic solidarity? While all the main elements of Durkheim’s answer are present in De la division, he did not elaborate upon them fully until later, in the context of his subsequent appreciation of the intimate connections between religion and moral authority. In these later writings, three important characteristics of moral individualism are worked out in some detail: what the ‘individual’ refers to in the “cult of the individual”; why it is a ‘cult’; and what is the proximate ideological source from which it derives. ‘The individual’ who is the subject of the ideals embodied in moral individualism is not the concrete individual, the particular personality, but ‘man’ in general. The morality of the “cult of the individual” is composed of those values given intellectual expression by the eighteenth-century philosophers and inspiring the French Revolution. These are values which emphasise the dignity and worth of ‘man’ in the abstract; as such, not only do they not derive from the ‘egoism’ of the utilitarians, but they are its direct opposite. Egoism is the pursuit of self-interest. But these values imply sentiments of sympathy for others and for human suffering.5 Precisely because they are created by society, they have a religious quality—although, if Durkheim’s definition of religion given in Les formes élémentaires be strictly applied, it is by no means clear what

5

L’individualisme et les intellectuels (1898), Revue Bleue, 7–13, translated in Lukes (1969).

The ‘Individual’ in the Writings of Émile Durkheim

517

corresponds to the ‘church’ in relation to the ‘sacred’ beliefs of moral individualism. Finally, it can be shown, as Durkheim indicates in L’évolution pédagogique en France, that the ideals of moral individualism have their immediate origin in Protestantism, and are more generally founded upon conceptions common to all forms of Christianity.6 Understood in these terms, moral individualism is not merely compatible with the moral diversification of organic solidarity, but directly stimulates its development. Respect for the individual, and the concomitant demand for equality, become moral imperatives: as such, they entail that the welfare and self-fulfilment of every member of society should be sought after. Human life can no longer be contained within the narrow limits enforced in traditional society. Specialisation of occupational function according to talent and capacity is the principal mode in which the (concrete) individual can realise himself. Hence the emergence and strengthening of the “cult of the individual” progresses hand in hand with the diversification of the division of labour. Now Durkheim’s theory of moral authority holds that all moral norms have two components: they command respect, and have a positive attraction; but they also have a ‘dutiful’ character, and are supported by sanctions against deviation. It does not follow from this, however, that the form of moral authority is everywhere the same. The moral authority of mechanical solidarity is repressive: the individual is subject to the “tyranny of the group”. Now, human freedom, according to Durkheim, consists of the autonomy of individual action. This is not acquired by the dissolution of moral codes, but by the transformations implicit in the development of moral individualism: Droits et [...] libertés ne sont pas choses inhérentes à la nature de l’individu comme tel. Analysez la constitution empirique de l’homme, et vous n’y trouverez rien de ce caractère sacré dont il est actuellement investi et qui lui confère des droits. Ce caractère lui a été surajouté par la société. C’est elle qui a consacré l’individu; c’est elle qui en fait la chose respectable par excellence. L’émancipation progressive de l’individu n’implique donc pas un affaiblissement, mais une transformation du lien social. [Rights and [...] freedoms are not inherent in the nature of individuals as such. Analyze the empirical constitution of human beings, and you will find none of that sanctity with which they are now invested and which confers rights upon them. This character has been added to them by society. It is society that has consecrated the individuals; it is society that makes them the respectable things par excellence. The progressive emancipation of the individuals therefore does not imply a weakening, but a transformation of the social bond.]. (Durkheim 1924, p. 106)

6

Durkheim (1938). See, for example, pp. 322–323.

518

A. Giddens

Durkheim’s conception of anomie bears directly upon each of the two types of ‘individualism’ distinguished above: in a positive sense, it provides an important reinforcement to his analysis of moral individualism; negatively, it constitutes a major aspect of his critique of utilitarian philosophy. The concept of anomie, as Durkheim employs it in his writings, appears at first sight to be fairly simple in form; further investigations of it, however, discloses various overlapping strands which enter into his formulation. As an element in his critical evaluation of utilitarianism, it rests first of all upon a set of empirical observations, which seem to have constituted the mainspring of Durkheim’s first appreciation of the significance of anomie.7 These observations concern the proposition that there is no direct correlation and, under certain circumstances, an inverse correlation, between the growth of economic prosperity and the advance of human happiness. The evidence from the analysis of suicide rates provides the clearest empirical index of this. In an early empirical study of suicide, Durkheim notes that, if human wants are not simply ‘given’, but are plastic, then the rising capacity of society to provide for the needs of its members may in turn generate new needs, and thereby increase the gap between wants and their satisfaction (Durkheim 1888). This conclusion is affirmed in De la division. The theoretical side of this, however, remained obscure until the publication of Le suicide. In that work, the empirical observations (concerning, for example, the rising rate of suicide with the advance of material civilisation, and the relation between suicide and socioeconomic position with the occupational system) are drawn together in terms of a coherent theoretical analysis. Durkheim’s discussion of anomie in Le suicide is far from unambiguous, but one major point can be gleaned from it. There is an essential distinction between biological needs and those which are engendered by society. The former is fixed in their object and in their limits: these two aspects may vary independently in socially produced needs. That is to say, whereas biological needs are specific in what they demand of motivated action (e.g. hunger: food) and in their point of satiation (there are given organic reactions which reduce appetite in relation to the intake of nourishment), neither of these is necessarily fixed in a determinate fashion in the case of socially created needs or desires. The importance of the differentiation between these two aspects is never clearly brought out in any of the discussions of anomie which appear at various places in Durkheim’s work: as will be shown below, it is a source of certain basic flaws in his writings. For

7

The term ‘anomie’ first appears in Durkheim’s (1887b) writings in his review of Guyau’s L’irréligion de l’avenir. The latter author uses the term, however, in a sense closer to Durkheim’s conception of moral individualism.

The ‘Individual’ in the Writings of Émile Durkheim

519

the present, however, it is sufficient to note the significance of this theoretical position for the critique of utilitarianism. This critique is also twofold. Firstly, human wants cannot be taken as givens, but are socially created and thus historically variable. Secondly, and equally important: that the creation of wants does not automatically produce the circumstances which make possible their realisation. This latter point becomes focal to Durkheim’s critical assessment of socialist doctrines. Socialism is superior to utilitarianism in recognising that human wants are not simply ‘contained’ in the individual, but are socially created; but it shares with utilitarianism the notion that society does not have to intervene in the satisfaction of needs. According to socialist theory, production thus has to be regulated, but consumption must be freed from social control. The positive connections between the conceptions of moral individualism and anomie, apart from the early formulation in De la division were nowhere explicitly stated in any detail by Durkheim, and thus have tended to be among the most frequently misunderstood parts of his writings. As I have tried to show elsewhere, faulty interpretation of Durkheim on this matter has helped to sustain two of the most prevalent misrepresentations of Durkheim’s sociology in the secondary literature: that which sees his work as primarily concerned with an abstract “problem of order”, and the closely related view of his writings as advancing a heavily authoritarian theory of moral discipline.8 If there is a basic opposition in all of Durkheim’s work, it is not that of social integration (normative control) versus social disintegration (lack of normative regulation: anomie), but, as with virtually all leading social thinkers of his time, that of ‘traditional’ versus ‘modern’ society, with all the profound social transformations which this latter distinction implies. It does not seem to have been generally appreciated that there is necessarily a historical dimension to Durkheim’s treatment of anomie: this is integral to the very conception of ‘socially generated need’, but it is also important in regard to the second aspect of anomie, that of provision for wants. In the traditional social order, human faculties and needs are kept to a low level, and therefore are readily provided for. The dominance of the conscience collective plays a role in each of these respects: on the one hand, by restricting the development of ‘individuation’—the liberation of the individual personality—and, on the other, by setting strict limits to what may legitimately be striven for by an individual in a given social position. The process of evolution away from traditionalism both increases the level of individuation and at the same time undermines the

8

See my Capitalism and Modern Social Theory (Giddens 1971b), Chapters viii and xv; and my “Introduction” to Émile Durkheim: Selected writings ( forthcoming ).

520

A. Giddens

fixed moral boundaries characteristic of previous ages. It is these twin developments which create the important theoretical problem—which Durkheim seeks to resolve in terms of his analysis of the emergence of moral individualism. Anomie, therefore, is a phenomenon specific to the modern order (as is indexed by the documentation of the growth of anomie suicide in Le suicide); it is to be understood in relation to individuation and moral individualism. Although Durkheim concedes that “a certain level” of anomie is inevitable in modern society, which is committed to rapid and continuous change, anomie as pathology is to be traced to the temporarily inadequate development of moral individualism. The upsurge of anomie found in the contemporary age is mainly centred in economic life, which has broken away from the confines of tradition, but has not yet been sufficiently penetrated by the new morality of individualism. It might be useful at this juncture to summarise what has been said so far: (1) There are two, connected, aspects of Durkheim’s discussion of the relationship between the ‘individual’ and ‘society’: a defence of the fundamental place of moral individualism in modern society, coupled with a disavowal that this implies acceptance of ‘individualism’ as developed in utilitarian philosophy, as a viable methodological basis for sociology. (2) Moral individualism is consequently quite distinct from ‘egoism’: the increasing range of individual autonomy (in the sense of ‘individuation’) is contingent upon the emergence of this ‘cult of the individual’, in conjunction with the other social changes entailed in the transition from mechanical to organic solidarity. Both the critique of utilitarianism and the analysis of moral individualism receive an important reinforcement from the elaboration of the theory of anomie; the remedy for anomie does not consist in the re-imposition of traditional, repressive moral discipline, but the further advance of the liberal morality of individualism. With these points in mind, it is possible to turn to Durkheim’s critique of “methodological individualism”, as set out in his discussion of “social facts” in Les règles de la méthode sociologique. Like the rest of the work, much of this is bluntly polemical: while its impact upon the reader is correspondingly powerful and convincing, the positive elements in the exposition of the characteristics of social facts are in fact difficult to unearth. As in his analysis of the difference between egoism and moral individualism, there are two conceptions of the ‘individual’ implied in what he has to say: but whereas in the former case, he himself specifies the nature of the distinction, in his treatment of social facts this is not clearly brought out. This time it is a distinction between the concrete, flesh-and-blood individual, on the one hand, and the ‘pre-social’ individual on the other—that is, the ‘individual’ as he is conceived

The ‘Individual’ in the Writings of Émile Durkheim

521

of in utilitarian theory. The concrete individual is necessarily, of course, the ‘carrier’ of society: remove all human organisms, and there is no society. In holding that social facts are ‘external’ to the individual, Durkheim has in mind the “pre-social individual”: there is no role for sociology as an independent discipline if we accept the utilitarian premise. But the manner in which this is expressed is such that it is hardly surprising that Durkheim has been accused of an illegitimate reification of the ‘social’. The same ambiguity arises in relation to the second criterion which he applies: that of ‘obligation’ or ‘constraint’. If Durkheim’s ‘individual’ is the “concrete individual”, then his analysis of the ‘constraining’ character of social facts in Les règles is, to say the least, unsatisfactory; for other phenomena ‘external’ to the concrete individual share the same character—such as those given in the geographical environment. Social facts would be merely residual, placed ‘outside’ the individual and resistant to his will.9 But it is evident that Durkheim does not wish to hold this. Social facts are distinct from those of the physical world, because “ils consistent en représentations et en actions” (Durkheim 1950 [1894], p. 8). In what sense, then, are social facts ‘constraining’? Much hangs on the answer to this question, of course, since it has been commonly accepted, even by those generally sympathetic to Durkheim’s viewpoint, that his stress upon the constraining nature of social facts leaves no place for the social actor as a conscious willing agent. In Les règles, Durkheim offers two sorts of examples to support his argument—without clearly recognising the difference between them: Si j’essaye de violer les règles du droit, elles réagissent contre moi de manière, à empêcher mon acte s’il en est temps, ou à l’annuler et à le rétablir sous sa forme normale s’il est accompli et réparable, ou à me le faire expier s’il ne peut être réparé autrement. S’agit-il de maximes purement morales ? La conscience publique contient tout acte qui les offense par la surveillance qu’elle exerce sur la conduite des citoyens et les peines spéciales dont elle dispose [...] Ailleurs, la contrainte, pour n’être qu’indirecte, n’en est pas moins efficace. Je ne suis pas obligé de parler français avec mes compatriotes, ni d’employer les monnaies légales; mais il est impossible que je fasse autrement. Si j’essayais d’échapper à cette nécessité, ma tentative échouerait misérablement. Industriel, rien ne m’interdit de travailler avec des procédés et des méthodes de l’autre siècle mais, si je le fais, je me ruinerai à coup sûr [If I try to violate the rules of law, they react against me in such a way, to prevent my act if it is time, or to cancel it and restore it to its normal form if it is accomplished and reparable, or to make me atone for it if it cannot be repaired otherwise. Are these purely moral maxims? The public conscience contains any act that offends them through the

9

Parsons, however, makes too much of this. See Parsons (1949, pp. 350–353).

522

A. Giddens

surveillance it exercises over the conduct of citizens and the special penalties available to it [...] Elsewhere, the constraint, for being only indirect, is not no less effective. I am not obliged to speak French with my compatriots or to use legal currencies; but it is impossible for me to do otherwise. If I tried to escape this necessity, my attempt would fail miserably. Industrial, nothing prevents me from working with processes and methods of the other century but, if I do, I will ruin myself for sure]. (Durkheim 1950 [1894], p. 7)

In the examples mentioned in the first part of the quotation, law and custom, the constraint involved is that of moral obligation. In those given in the second part, constraint does not derive from direct moral commitment, but from factors, which comprise a ‘factual’ element in the horizon of the social actor. Thus, the industrialist who ignores certain technical requisites of production fails to prosper. If Durkheim realised the full significance of this distinction at the time at which he wrote Les règles, he gives little sign of it, and it is clear that he tends to identify social facts primarily as moral obligations. Later, he became more clearly aware of the importance of the difference between these two sorts of social ‘constraint’, perceiving that the moral sanction is different in character from the constraining factor which is present in the other type. The feature of the moral sanction is that it is not intrinsic to the act which is sanctioned: it cannot be deduced from the properties of that act itself. The same act which is morally reprehensible in one type of society, for instance, may be tolerated or even actively encouraged in another. This is not the case with the consequences which stem from failure to adhere to prescriptions or practices which are not moral obligations (Durkheim 1924, pp. 60–62). In this case, the ‘constraint’ is an undesired consequence, which follows automatically from the transgression of the rule—as in the case of the industrialist who fails to observe the rules of efficient business practice. Most secondary interpreters of Durkheim have assumed that this later discussion represents a shift from the conceptualisation of “social facts” set out in Les règles, holding either that he was unsure of his standpoint in that work, or that he abandoned the notion that social facts should include nonnormative social constraint. Durkheim nowhere actually says this, however, and he continued to rank within the field of sociology such areas of study as “social morphology”, which is not defined in normative terms. Moreover, it is important to remember that he insisted that his own contributions to sociology were all primarily located in this one restricted area of the discipline: the study of morality. In virtue of this, it is perhaps not surprising that most of his writings should concentrate upon the characteristics of moral obligation. At any rate, this is certainly the case: as has been mentioned, even in the relatively early discussion, in the first edition of Les règles, he places most

The ‘Individual’ in the Writings of Émile Durkheim

523

of the weight, in setting out the nature of social facts, upon constraint as a moral obligation. Thus, in evaluating his position, it is of vital importance to determine the sense of the term ‘obligation’: and it is in this respect that we need to consider the implications of his other writings. Two of these implications are especially significant here. One is that the character of moral obligation varies in different forms of society; the second is that, according to the theory of anomie, moral obligation not only sets limits to human action, but also focusses it and gives it a defined objective. Each of these two propositions indicates, in a differing way, that ‘obligation’ is not to be identified with ‘restriction’ tout court, as might be suggested by Durkheim’s habit of using ‘obligation’, ‘constraint’, and even, occasionally, ‘coercion’ as interchangeable terms. To be sure moral obligations are always constraining in the sense that deviation produces sanctions: but the degree to which acceptance of obligation is ‘restrictive’ upon the actor is contingent upon the moral form in question. The moral obligations involved in the ‘cult of the individual’ confer increased autonomy of action, as compared to the rigid discipline of traditional society. This is possible precisely because moral codes define the content of human motivation, at least with regard to socially generated needs. While Durkheim no doubt did not achieve complete clarity on this latter point until after he had written his account of the nature of social facts in Les règles, he does specifically point out in the book that his understanding of moral obligation diverges from the ‘constraint’ of Hobbes and Rousseau, who neglect the ‘spontaneous’ character of moral life. These thinkers, according to Durkheim, assume that there is a dislocation between society and its component individuals: the human being is treated as naturally hostile to social life. Society, therefore, can only exist if men are compelled, by the action of the state, to adapt themselves to it. Their views contrast with those of the utilitarians, who see social life as a spontaneous phenomenon. In the first conception, ‘constraint’ is seen as fundamental to social life, but is considered as a purely ‘external’ sort of coercion. The second standpoint eliminates constraint, in any form, from the operation of society. His own position, Durkheim says, comprises elements of each of these, and also rejects both: there is no contradiction in holding that social life is both ‘constraining’ and ‘spontaneous’. These considerations are surely enough to suggest that Durkheim may be absolved from the cruder type of criticism which has frequently been directed against his methodological writings: that he accords society’s ‘objective’ reality only at the expense of denying any reality to the active subject. Nor is there much substance in the equally common assertion that Durkheim’s work

524

A. Giddens

should be ranged with that of those thinkers (mainly of a broadly ‘conservative’ persuasion) who hold that there is an inherent antinomy between man and society, such that the continuation of social life depends upon the strict repression of individual desires. He clearly recognises that what the (concrete) individual is, depends upon ‘internalised’ norms which are, in part, the condition of freedom of action. But his treatment of this matter involves definite inconsistencies. This can be seen quite clearly in his various discussions of ‘egoism’. In his earlier writings, ‘egoism’ has reference to the utilitarian model of self-interest. In Durkheim’s view, this presupposes a ‘pre-social’ man, and his critique of this conception in Les règles takes this as its foil. But he evidently soon came to perceive, that according to the position which he had taken in criticising utilitarian individualism, egoism itself must be a product of society. That is to say, there can be socially created self-interest. He even gives a specific illustration of this. Moral individualism, he emphasises, does not derive from egoism; but the growth of moral individualism nonetheless produces, as an offshoot, an expansion in the range of egoistic inclinations: Assurément la pitié pour autrui et la pitié pour nous-mêmes ne sont pas étrangères Tune à l’autre, puisqu’elles progressent ou reculent parallèlement ; mais l’une ne vient pas de l’autre. S’il existe entre elles un lien de parenté, c’est qu’elles dérivent toutes deux d’un même état de la conscience collective dont elles ne sont que des aspects différents. Ce qu’elles expriment, c’est la manière dont l’opinion apprécie la valeur morale de l’individu en général. S’il compte pour beaucoup dans l’estime publique, nous appliquons ce jugement social aux autres en même temps qu’à nous-mêmes ; leur personne, comme la nôtre, prend plus de prix à nos yeux et nous devenons plus sensibles à ce qui touche individuellement chacun d’eux comme à ce qui nous touche en particulier [...] Tant il est vrai que même les sentiments qui semblent le plus tenir à la complexion personnelle de l’individu dépendent de causes qui le dépassent ! Notre égoïsme lui-même est, en grande partie, un produit de la société [Assuredly pity for others and pity for ourselves are not foreign to each other, since they progress or retreat in parallel; but one does not come from the other. If there is a kinship between them, it is because they both derive from the same state of the collective consciousness of which they are only different aspects. What they express is the way in which opinion appreciates the moral value of the individual in general. If it counts for much in public esteem, we apply this social judgment to others as well as to ourselves; their person, like our own, takes on more value in our eyes and we become more sensitive to what affects each of them individually as well as to what affects us in particular [...] It is so true that even the feelings that seem most dependent on the individual’s personal complexion depend on causes that go beyond him!

The ‘Individual’ in the Writings of Émile Durkheim

525

Our selfishness itself is, in large part, a product of society.]. (Durkheim 1930 [1897], p. 411)

Yet, in other writings, Durkheim reverts to a conception of ‘egoism’ which counterposes it in a direct way to social learning as if the two are necessarily mutually exclusive. Man, he argues, everywhere conceives of himself as a homo duplex, as being composed of two beings, which are usually represented in religious thought as the body and the soul.10 This corresponds to a psychological division between sensations, on the one side, and concepts and moral activity on the other. Sensations and sensory needs, according to Durkheim, are necessarily egoistic, because they originate from, and refer to, conditions of the biological organism. Conceptual thought and moral activity, are ‘impersonal’; they are social products, and do not ‘belong’ to any particular person who uses them. These, therefore, are two opposed aspects of personality. They are not merely separate from one another, but are in constant conflict. ‘Egoism’ is thus identified solely with the ‘pre-social’, and is portrayed as wholly foreign to the “penetration of the individual by society”. The implications of this will be further discussed below. But it is worth pointing out that a similar inconsistency appears in various other parts of Durkheim’s writings, which have reference to the connection between the ‘biological’ and the ‘social’ components of personality. An example may be given from his work on suicide.11 The main thesis, which Durkheim develops in Le suicide is that the suicide rate, as compared with particular cases of suicide, “constitue par lui-même un fait nouveau et sui generis, qui a son unité et son individualité [constitutes by itself a new and sui generis fact, which has its unity and its individuality]”, and has to be explained sociologically (Durkheim 1930 [1897], p. 8). To this end, he develops his typology of the social factors underlying observed differences in suicide rates. To hold that the suicide rate can only be explained in sociological terms does not mean, however, he goes on to say, that there is no place for psychological studies of suicide. It is the role of the psychologist to examine the particular characteristics which cause one man, rather than another, to kill himself, when placed in a given set of social circumstances: not everyone who is exposed to a situation of anomie actually commits suicide. While this may be a neat division of tasks between sociology and psychology, it is a clearly inadequate position, because it implicitly assumes that the traits of the ‘suicidal personality’ are formed outside of society. The (social) factors which cause a certain rate of suicide are presented by Durkheim as separate from those relevant to the aetiology 10 11

Durkheim (1914), translated in Wolff (1964). See “The Suicide Problem in French Sociology”, in Giddens (1971c).

526

A. Giddens

of the individual case. But this would only be so if it were true that the characteristics of suicidal personality were ‘pre-social’: i.e. settled in the biological constitution of the organism. In fact, Durkheim accepts that this is not the case, showing that the “individual type” of suicide is strongly influenced by the “social type”. If this be admitted, however, the explanatory model which Durkheim sets up in the book is immediately defective. It follows that there is a complicated interplay between society and personality in the aetiology of rates of suicide: the social conditions which he treats as simply acting ‘directly’ upon “suicide-prone” individuals also must have an influence in producing that “suicide-proneness” as a complex of personality characteristics. Enough has been said this far to indicate that, although Durkheim’s attempt to detach moral from methodological individualism is much more subtle and profound than has been assumed by many of his critics, what results is a brittle synthesis and essentially an unsatisfactory one. The ambiguities, and the very serious deficiencies, which run through his works, however, have to be understood in the light of this attempt. As happens so often with a writer whose works are strongly polemical in tone, ultimately he was unable to abandon certain of the very premises of which he was most critical in the writings of his opponents. In this paper, I have given most prominence to Durkheim’s critique of utilitarian individualism, but this was not his only target of attack. A second tradition of thought that he sought to evaluate critically, while also being strongly influenced by it, was that of ‘idealist holism’. A variety of substantively quite different doctrines, of course, may be subsumed under such a rubric. But the most important element in these doctrines which formed a negative frame of reference for the development of Durkheim’s thought was the notion that a ‘universal’ moral consensus is the fundamental condition of the existence of society, of whatever type As Durkheim attempted to show in De la division, this is a mistaken view. While this is indeed the condition of unity in traditional society, moral diversity is the necessary concomitant of the modern societal type. If modern society is, and must be, a moral order, it is distinct from the traditional form. In developing this stand point, as is plainly evident in his early critical articles and reviews, Durkheim was, in a certain sense, positively influenced by utilitarianism, especially as it was adapted by such writers as Spencer. These authors at least recognise that the character of modern social life is distinctively different from the life of man in traditional society, and their version of ‘individualism’ gives expression to this awareness. In exploring the insufficiencies of utilitarian individualism, however, Durkheim borrowed heavily from the other school of thought, asserting that moral ideals are irreducible to individual motivations or interests.

The ‘Individual’ in the Writings of Émile Durkheim

527

Along the historical dimension, therefore, Durkheim leans partly upon utilitarianism in rejecting the conception of the relationship between the individual and society implicit in holistic idealism—that the individual is merely a ‘microcosm’ of society. This may be, in a sense, an apt characterisation of the situation in traditional society, in mechanical solidarity, but it is quite inappropriate to the modern order, in which “la personnalité individuelle se développe avec la division du travail [the individual personality develops with the division of labor]” (Durkheim 1960 [1893], p. 399). This shows, of course, that idealism is methodologically wanting. The individual in society is not simply a passive imprint of social forms, but an active agent. But, even while he recognises this, he relies upon the holistic standpoint in working out his critique of utilitarian individualism. From each of these two aspects, historical and methodological, this rests upon the proposition that society is not a creation of the (pre-social) individual, but exists ‘prior’ to him and moulds him. How, then, is it possible that the (concrete) individual is an active agent? It is at this point that the two dimensions, the historical and the methodological, in Durkheim’s thought diverge. The answer which he reaches via his study of the evolution of solidarity, and his analysis of moral conduct more generally, is that it is possible because the cognitive and motivational structure of the personality of the individual is shaped by social learning. He is not just moulded by society; the active orientation of his conduct is framed by ‘internalised’ moral norms. But there is a second answer, to which his thought constantly tends to revert, and which is undoubtedly a derivative of his preoccupation with utilitarianism. This is that whatever is actively willed by the individual is a ‘pre-social’ impulsion. In other words, in seeking to reject utilitarianism, Durkheim tends to deal with it in its own terms. Society cannot be conceived as the outcome of pre-formed individual wills— because society makes, and must make, demands upon the individual which are foreign to his own wishes! Hence, we reach the position that there is irremediable conflict between the (‘egoistic’) inclinations of the individual, and the moral commands which society enjoins upon him. Durkheim never managed adequately to reconcile these two strands of his thought, and this is reflected in several of the ambiguities noted previously. Thus, his discussion of social facts in Les règles, as has been mentioned, shifts from one meaning of ‘individual’ to another. The ‘pre-social individual’ is the ‘individual’ of utilitarian theory. Because Durkheim’s polemic here is mainly concerned to rebut this conception, his treatment easily lends itself to the interpretation that social facts are empirically ‘external’ to individual conduct, and exert a constraining force over that conduct which is of the same kind as that produced by geographical or climatic forces. The same kind

528

A. Giddens

of enforced polarisation between the ‘internal’ and the ‘external’ appears in Durkheim’s use of argumentuum per eliminationem in Le suicide.12 Various sorts of ‘external’ or ‘pre-social’ factors (e.g., the influence of climate) are eliminated separately as explaining observed differentials in suicide rates. These include the influence of inherited insanity. Now the assumption that this is an ‘external’ factor is acceptable if it be granted that insanity is wholly inherited, and thus is built into the constitution of the organism. But Durkheim explicitly points out that this is not so: social factors have a definite role in the aetiology of mental disorder. The insight is, however, allowed to remain undeveloped. Yet, as has been indicated above, Durkheim at no time wished to advance what he calls at one juncture the “absurd notion” that society is ‘external’ to the individual in the same sense as the geographical environment is; and most of his argument in Les règles is only understandable if his ‘individual’ is the “concrete individual”. A similar point may be made about the concept of anomie. It has already been shown that, in Durkheim’s formulation, the concept involves two components, which tend to be assimilated in his own use of it. One of these refers to the degree to which human action is provided with definite objectives; the other concerns how far these ends are realisable. The distinction is a fundamental one, yet Durkheim glosses over it. It is easy to see how closely this is tied in to the ambiguity in his treatment of social facts. In so far as one uses the concept of anomie as referring primarily to the first element, lack of a coherent norm, which provides a firm objective for a man to strive for, it presupposes that one is talking of the “concrete individual”. This is certainly the type of case in most of Durkheim’s abstract discussions of anomie. But many of the examples of anomie which he applies in his more empirical analyses (e.g. in relation to class conflict) actually concentrate upon the second element: not that the objective of conduct is not clearly defined, but that it cannot be attained. In such cases, his argument again tends to slip back into a direct opposition to the utilitarian position, and thus relates to the “pre-social individual”: utilitarian theory is mistaken because an external limitation (which can only come from society) must be placed upon individual desires. It is no accident that Durkheim’s illustrations often refer to cases of biological drives: e.g. that ‘an insatiable thirst cannot be quenched’. The fact that Durkheim failed to utilise the distinction between the two aspects of anomie is a source of some of the most basic flaws in his sociology. As the more recent literature employing the notion of

12 This is, of course, reinforced by other aspects of Durkheim’s methodology which I do not discuss in this paper: such as the ‘rule’ that “an effect can only have one cause”.

The ‘Individual’ in the Writings of Émile Durkheim

529

anomie demonstrates, the concept takes on quite divergent theoretical applications according to which aspect is emphasised. If anomie is taken to refer mainly to ‘normlessness’—the first aspect—then it tends to support a standpoint emphasising the dimension of ‘meaning’/‘lack of meaning’ in individual conduct.13 The end result of this is likely to be a position which, implicitly or otherwise, treats social conflict as ‘pathological’—that is to say, which links conflict to ‘deviance’ produced by “imperfect socialisation”. Although, as I have argued elsewhere,14 it is quite mistaken to hold that Durkheim regards all social conflict in this way, there are obviously strong elements of this viewpoint in his writings. If the other aspect is emphasised (Merton 1963), on the other hand, it tends to lead to a conception of ‘normative strain’, rather than ‘normlessness’. Here, the objectives of conduct may be quite clear to the actor, and there is not the strong overtone of irrationality, which appears to characterise the conduct of the individual where the first aspect is stressed. The importance of this conception, then, is that it allows a much greater scope for conceptualising conflicts which derive from divisions of interest in society. Durkheim undoubtedly minimises the significance of this form of conflict in his writings. We are now in a clearer position to see why this is so. There are two respects in which it is possible to claim that a given objective is not ‘realisable’. One is that there exist barriers in the society which prevent its realisation. This may mean that it cannot be achieved—that it asks for something which, in any conceivable form of society, is simply impossible of attainment; or it may mean that its realisation demands some kind of change in the existing organisation of society. This perspective certainly has a substantial place in Durkheim’s writings. Thus, he emphasises that, in order to overcome the class conflict which characterises modern industry, social change (involving principally the removal of blockages upon equality of occupational opportunity) is necessary. But there is another sense in which an objective may be said to be ‘unrealisable’, and Durkheim continually reverts to this—for reasons explained above—in his anxiety to refute the utilitarian position. This is where the objective in question has no limits: the insatiable appetite. We can now also see more clearly the significance of another ambiguity, which has been referred to in a previous part of this paper, concerning Durkheim’s treatment of moral obligation and ‘factual’ constraint. Although he did not seem to attach any importance to this distinction in Les règles, he did later recognise that moral constraint is quite different in character from the ‘factual’ consequences of actions. But, he only elucidates one dimension of this difference—the difference in the nature of the sanction involved. 13 14

This is the aspect given most prominence by Parsons. See especially Parsons (1951, p. 39). “Introduction” to Émile Durkheim: Selected Writings.

530

A. Giddens

He fails to consider the theoretical significance of the possibility that moral obligations themselves may be ‘factual’ elements in the horizon of the acting individual. A person (or a group) may acknowledge the existence of the obligations, and take account of them in orienting his conduct, without feeling any strong commitment to them. Such action is not necessarily ‘criminal’, in the sense of directly flouting the moral prescriptions in question. But it rests neither solely upon fear of the sanctions which would be invoked as punishment for transgression, nor solely upon moral commitment. While Durkheim accepts that there are varying degrees of attachment to moral norms,15 he has no place for this in his theoretical analysis of the nature of moral obligation. Nor is there any recognition of the differential ‘interpretation’ of moral norms. These inadequacies can also be seen at the level of Durkheim’s historical analysis and, indeed, partly derived from that analysis. It is significant that Durkheim has been called both a ‘materialist’ and an ‘idealist’. The first label, so it has been thought, can be readily applied to his standpoint as expressed in De la division. In that work, he argues that [la civilisation] est elle-même une conséquence nécessaire des changements qui se produisent dans le volume et dans la densité des sociétés. Si la science, l’art, et l’activité économique se développent, c’est par suite d’une nécessité qui s’impose aux hommes; c’est qu’il n’y a pas pour eux d’autre manière de vivre dans les conditions nouvelles où ils sont placés. Du moment que le nombre des individus entre lesquels les relations sociales sont établies est plus considérable, ils ne peuvent se maintenir que s’ils se spécialisent davantage, travaillent davantage, surexcitent leurs facultés; et de cette stimulation générale résulte inévitablement un plus haut degré de culture. De ce point de vue, la civilisation apparaît donc, non comme un but qui meut les peuples par l’attrait qu’il exerce sur eux, non comme un bien entrevu et désiré par avance, dont ils cherchent à s’assurer par tous les moyens la part la plus large possible, mais comme l’effet d’une cause, comme la résultante nécessaire d’un état donné [(civilization) is itself a necessary consequence of the changes that occur in the volume and density of societies. If science, art, and economic activity develop, it is as a result of a necessity that imposes itself on men; it is that there is no other way for them to live in the new conditions in which they are placed. As soon as the number of individuals between whom social relations are established is greater, they can only maintain themselves if they specialize more, work more, overexcite their faculties; and from this general stimulation inevitably results a higher degree of culture. Therefore, from this point of view, civilization appears not as a goal that moves peoples by the attraction it exerts on them, not as a good

15

See for instance Durkheim (1924, pp. 56–57).

The ‘Individual’ in the Writings of Émile Durkheim

531

foreseen and desired in advance, of which they seek to secure for themselves by all means the largest possible share, but as the effect of a cause, as the necessary resultant of a given state.]. (Durkheim 1960 [1893], p. 327)

The apparent theme of such statements is that the transformation of society from mechanical to organic solidarity is wholly a matter of changes in the social ‘infrastructure’; that the changes which occur in the character of moral conduct are simply ‘effects’ of these ‘causes’. The perspective developed in Les formes élémentaires seems to contrast markedly with this. In the latter work, so it would appear, Durkheim treats moral ideals as the force dirigeante in social life. It is mainly by reference to this work that he has been regarded as advancing an ‘idealist’ position. Thus, it seems as though, if one follows through the stages of his intellectual career, he begins from a ‘materialistic’ standpoint, and moves later towards a view, which is directly opposed to that which he initially adopted. In fact, however, throughout his writings, Durkheim frequently and specifically denies that he wishes to adopt either of these positions. In De la division, for example, the section which has just been quoted is immediately followed by the assertion that «une conception mécaniste de la société n’exclut pas l’idéal [a mechanistic conception of society does not exclude the ideal]», and that «c’est à tort qu’on lui reproche de réduire l’homme à n’être qu’un témoin inactif de sa propre histoire [he is wrongly accused of reducing man to an inactive witness of his own history]». On the other hand, while his later writings are full of statements such as “la société, c’est l’idéal [society is the ideal]”, he is always careful to insist that such propositions must be interpreted to mean that ideals are créations of human society, not ‘given’ forces which determine social conduct. This is, after all, the main proposition developed in Les formes élémentaires, and Durkheim himself evidently still considered that his critics would regard it as another version of his earlier ‘materialism’. Near the end of the book, he takes pains to make clear that this is not what he is proposing: there is, he emphasises, an active interplay between the conscience collective and its ‘infrastructure’. But he never solved the problems which the analysis of this interplay presents and, where he discusses social change, it is often as if there are two quite independent sets of processes going on: those in the ‘infrastructure’, on the one hand, and those in the sphere of moral ideals on the other. This is undoubtedly one reason which explains the apparent ‘break’ in the argument in De la division which was mentioned earlier. The only direct connections which Durkheim is able to establish between the expansion of the division of labour and the changing nature of the conscience collective are those whereby the influence of the latter is weakened. It is not at all clear why its content

532

A. Giddens

should become changed: what factors promote the emergence of the ‘cult of the individual’? Even in the theory developed in Les formes élémentaires, the origin of the content of religious beliefs remains obscure. These beliefs are created in the fervour of collective ceremonial. But sacred beliefs, while framed in terms of categories (‘space’, ‘time’, etc.), which are formed upon the characteristics of society, are basically ‘random’: any object may become sacred, and there is, applying Durkheim’s theory, a practically infinite range of potential primitive classifications. Durkheim’s sociology lacks any systematic theoretical treatment of the social mechanisms which mediate the relationship between infrastructure and conscience collective. The reasons for this touch upon weaknesses in Durkheim’s thought which have not been explored in this paper. But the analysis given earlier is certainly of basic importance here: for it helps to show how it is that, as a result of the theoretical impasse in which his thought becomes wedged, he is unable to deal in a satisfactory way with socially generated interests, and more especially with conflicts which stem from the opposition of such interests. This applies both to the conduct of the individual actor vis-à-vis others, and to the relationship between subgroups in society.

References Durkheim, E. (1887a). La science positive de la morale en Allemagne [The Positive Science of Morals in Germany]. Revue Philosophique de la France et de l’Étranger, 24, 113–142. Durkheim, E. (1887b). L’irréligion de l’avenir, étude de sociologie Guyau [The Irreligion of the Future, a Study in Guyau Sociology]. Revue Philosophique de la France et de l’Étranger, 23, 299–311. Durkheim, E. (1888). Suicide et natalité: étude de statistique morale [Suicide and Birth Rate: A Study in Moral Statistics]. Revue Philosophique, XXVI, 446–463. Durkheim, E. (1960 [1893]). De la division du travail social . Paris: Presses Universitaires de France. [Transl. (1933) The Division of Labor in Society. New York: The Macmillan]. Durkheim, E. (1950 [1894]). Les règles de la méthode sociologique. Paris: Felix Alcan [Transl. (1982) Rules of Sociological Method . New York: The Free Press]. Durkheim, E. (1930 [1897]). Le Suicide. Paris: Felix Alcan [Transl. (1951) Suicide. New York: The Free Press]. Durkheim, E. (1914). Le dualisme de la nature humaine et ses conditions sociales [The Dualism of Human Nature and Its Social Conditions]. Scientia, XV, pp. 206–221. Durkheim, E. (1924). Sociologie et philosophie. Paris: Félix Alcan [Transl. (1974) Sociology and Philosophy. New York: The Free Press].

The ‘Individual’ in the Writings of Émile Durkheim

533

Durkheim, E. (1938). L’évolution pédagogique en France. Paris: Presses Universitaires de France [Transl. (1977) The Evolution of Educational Thought. London: Routledge and Kegan Paul]. Giddens, A. (1971a). Durkheim’s Political Sociology. Sociological Review, 19(4), 477–519. Giddens, A. (1971b). Capitalism and Modern Social Theory. An Analysis of the Writings of Marx, Durkheim and Max Weber. Cambridge: Cambridge University Press. Giddens, A. (Ed.) (1971c). The Sociology of Suicide. London: Routledge. Lacombe, R. (1927). La méthode sociologique de Durkheim [Durkheim’s Sociological Method]. Revue internationale de l’enseignement, 81(1), 119–123 Lukes, S. (1969). Durkheim’s “Individualism and the Intellectuals.” Political Studies, XVII, 14–30. Merton, R. K. (1963). Social Structure and Anomie. In Social Theory and Social Structure. Glencoe: The Free Press. Nisbet, R. A. (1965). Emile Durkheim. Englewood Cliffs: Praeger Publishers Inc. Parsons, T. (1949 [1937]). The Structure of Social Action. Glencoe: The Free Press. Parsons, T. (1951). The Social System. Glencoe: The Free Press. Wolff K. (Ed.) (1964). Émile Durkheim: Essays on Sociology and Philosophy. New York/London: Harper and Row.

Methodological Individualism (MI) and Methodological Collectivism (MC) in the Era of Postwar American Structural-Functionalism (SF) Charles Crothers

1

Introduction

A major theme in social theory has concerned the respective roles of individualism and collectivism (holism), but the handling of this distinction has not always been played out explicitly and directly. The debate about the explanatory role of each has occurred in social philosophy as well as in social theory, particularly sociology but also economics. The philosophical debates (e.g. Udehn 2002) tend to overshadow the merely theoretical, but this account remains within sociological discourse. In sociology, the emphasis on MI or MC has varied across particular time periods and has taken different forms. This chapter attempts to document the ways in which early versions of these doctrines were instantiated in sociological theory in the second half of the twentieth century. And, in particular, to illuminate the role of the ‘Columba Tradition’1 in working through relatively practical forms of the debate. 1

The ‘Columbia Tradition’ encompasses a stream of theoretical and empirical sociology which emerged from the University of Columbia Department of Sociology from mid-century through into the 1980s and was marked by the wide visibility of the work of its leaders—Robert K Merton and Paul F. Lazarsfeld.

C. Crothers (B) Sociology, Auckland University of Technology, Auckland, New Zealand e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 N. Bulle and F. Di Iorio (eds.), The Palgrave Handbook of Methodological Individualism, https://doi.org/10.1007/978-3-031-41512-8_24

535

536

C. Crothers

Within sociology differences in orientations towards MI/MC are likely to be important, not only between theory and methodology, but in terms of particular subject fields: for example, it seems likely that MI is more prevalent in social psychology, and MC in macro-sociology. But, such propensities are only incidentally examined here. Although often presented as a contrast, more nuanced positions can be specified: however, the extremes of MI and then MC have few advocates. Somewhat related, more empirically based, distinctions involve levels: usually divided into micro-, meso-, and macro-sociologies. In the period being studied in this chapter, sophisticated concepts of MI were yet to develop but developments set the stage for the rise of MI. This can be glimpsed as the sociology of the time began to supplement its earlier resolutely MC stance with moves towards more individualism, albeit of a general type. Such indirect effects came from the development of survey research. But also ‘methodological positivism’. But, it is important to note that both these can be interpreted within MI or MC frames, while in practice tending to be more readily coupled with an individualist interpretation. Survey research can play an important auxiliary role to MI in providing data which can then be fed into more abstract models of social action. Most often survey research is broadly individualistic, given the usual singlerespondent context of interviews. But, there are immediate qualifiers—sometimes group interviews are carried out and others may be present in the interviewing situation. Some surveys collect data directly on corporate entities, usually from authorized ‘spokespeople’, while others trace networks of contacts. Almost universally in data analysis, responses are aggregated into results for each of the various social categories. During the period of American sociology covered, there was a strong pressure towards ‘methodological positivism’ (or scientism), which also tended to direct sociologists towards more individualistic approaches. But again, this is an indirect effect as MI as an explanatory approach completely rejects the ‘covering law’ approach in favour of models of generative processes. A final preliminary point is that social phenomena can simultaneously be both personal and social2 However, this ‘split level’ understanding is not further explored in this chapter.

2

Always personal, but very often also social (Greenwood 2003). We have private ideas (expressed or not), but also (not always) may express the views held by a group. This point was crucial in Robert Merton’s analysis of prejudice (1949) where he argued that people can be prejudiced but not discriminate, while sometimes people might not be prejudiced and yet still discriminate, and yet, others are “active bigots”—both prejudiced and discriminatory.

Methodological Individualism (MI) …

2

537

MI and MC as Doctrines Within Sociology

Sociologists are not always conscious of the doctrines implicit in their analyses. The availability of ‘packaged’, explicitly developed, doctrines of MI, or MC in sociology has varied over time—and space, and this has suppressed formal affiliations with either. MI stretches back to Adam Smith at least, and in more recent times the writing of Max Weber and Joseph Schumpeter. Since then, Friedrich Hayek in the early 1940s wrote some highly pertinent methodological essays, while in the 1960s, JWN Watkins took the lead. More recently, there has been an array of philosophical discussions across a range of social philosophy and sociological journals. But these developments have largely bypassed sociology. Philosophical debate needs to be lifted out of its arenas of circulation and translated into more accessible forms for many sociologists to take any note. Of the various writings, those of Max Weber were undoubtedly the most influential for sociology, but the relevant passages were translated to English by Edward Shils and Henry Finch only in 1949. Moreover, the impact of Max Weber’s views was discounted by the ambiguity that whereas his formal methodology stressed MI this was not a strong feature of some of his actual comparative-historical analyses—while the Protestant Ethic work has a strong MI component, in his work on Chinese villages, ancient civilizations or the Russian Revolutions this is less emphasized (Turner 2022, per comm.). Some flourishes of MI were also developed within American sociology: the early Talcott Parsons and Robert Merton ‘social action’ models in the 1930s and then later (from the 1960s) George Homans’s behaviourism, plus Peter Blau’s and Si Goode’s ‘exchange theory’ micro-analyses. These approaches endeavoured to ‘bring back individuals’3 by showing how people’s behaviour was shaped at an individual level. Another wave of MI theorizing emerged in the early 1980s with theorists such as James Coleman and Gary Becker, together with the development of ‘Analytical Marxism’ (see Elster 1982) that attempted to provide a micro-foundation for Marxist analyses. In this era, there was much debate around ‘rational choice’ theory, which has become increasingly popular since. Alongside these were subtle analyses flowing from the work of Pierre Bourdieu and Anthony Giddens on ‘structuration’, which is an approach attempting to evade being impaled on either horn of the agency/structure and micro/macro dilemmas, but rather melding the two (e.g. Elder-Vass 2012). MC was seldom explicitly argued as a doctrine but has suffused sociology from its founding. Collectivism was deeply implicated in Marxian

3

Merton was not convinced that ‘men’ had ever left sociology.

538

C. Crothers

approaches, but also adopted by most sociological writers. A key and popular plank was the enunciation of Durkheim’s strong collectivity doctrine. As part of its development, and in reaction to the cross-Atlantic migration of ideas, American sociology began to consolidate its conceptual apparatuses from the 1930s. There were many possibilities, but few clear-cut ‘best bets’, as illustrated by several theory textbooks of the time which covered large arrays of ideas. One line of theorizing that had developed from within this clash (or chaos) of many schools was ‘structural-functionalism’ (SF) or Functional Analysis: the terms are broadly interchangeable. SF endeavours to explain features (e.g. social institutions) in terms of the contributions they make to the wider systems of which they are part, the whole is seen as bigger than the sum of the elements, and emphasis is on the system achieving stability. SF builds on a very intuitive approach—form is seen as explained by function. SF also endeavours to disentangle the relationships between the various elements of a system. But, a plausible feedback mechanism must be pointed at to turn a functional into a causal argument: for example, people in a system may sense potential systemic failure and move to staunch the problem, thereby restoring the system’s ‘health’. However, SF often remains at a descriptive level without going on to identify how whole systems operate to achieve some degree of equilibrium. Nevertheless, it can be a very useful ‘mapping’ approach, especially where the social system being investigated appears quite different from more well-known examples, perhaps because of the cultural distance of the observer. This can be a fecund recipe for generating hypotheses for further investigation. ‘Functional requisite’ analysis seeks to identify those key operations necessary for the system to keep operating rather than collapse. For the most part, SF can be seen as resolutely MC, operating ‘behind the backs’ of participants. Although there are many historical accounts of the development of sociological and other structural-functionalisms few are entirely satisfactory, with this chapter attempting a more grounded interpretation. Various early social theorists indulged in functional analysis. In his The Rules of Sociological Method Emile Durkheim (1895/1982), considered sociology was “the science of social facts”, which “consist of representations and actions”—meaning that “they cannot be confused with organic phenomena, nor with psychical phenomena, which have no existence save in and through the individual consciousness”. Durkheim insisted that social facts can be explained only by other social facts. His examples included kinship and marriage, currency, language, religion, political organization, and other societal institutions. A further implication was that deviating from the influence of institutions renders individuals subject to being stigmatized as deviants.

Methodological Individualism (MI) …

539

In the 1930s, functional analysis was renewed through the promising development of British anthropological functionalism available in two versions: the more sociological and structural version proposed by Alfred RadcliffeBrown, which examined the relationship between institutions, and the more psychologically-based version of Bronislaw Malinowski which derived societal requirements in part from the needs of its members. Malinowski argued that individuals have physiological needs (reproduction, food, shelter) and that culturally-specific social institutions existed to meet these needs. He also argued that there were four basic “instrumental needs” (economics, social control, education, and political organization), that also require institutional performances. However, Malinowski’s psychological functionalism received much less attention than the more structural version, in large part as it could not be readily shown how institutions arose to meet the postulated psychological needs. Both schemas provided ‘mapping’ capability, helping analysts locate how the institutional framework of a society/culture fitted together, as opposed to earlier approaches, which had attempted to trace the evolution of particular cultural elements over time, and their diffusion between different social contexts. The methodological complement of this approach required extensive ethnography in each selected tribe. Anthropologists were expected to learn the language and obtain in-depth information on all aspects of the culture, thereby allowing, inter alia, subsequent analysis of functional connections. By the mid-1930s, sociologists working in the ambit of Talcott Parsons at Harvard (Nichols 2010) were focussed on the development of more formal sociological theory. In his Structure of Social Action (1937), Parsons claimed a convergence across several important classical sociological authors of an individual-level ‘theory of action’ in which subjective elements were crucial (see below for an elaboration). But, this approach did not provide much analytical leverage for examining larger social and cultural structures, and Parsons then moved beyond this approach. Anthropological functionalism began to percolate into sociology as several sociological writers explored the possibilities of adapting it to fructify the analyses needed to understand modem societies, adding to Sociology’s growing concern with theory. Davis and Merton (both former students of Parsons) had a lengthy correspondence over several years in which they attempted to develop a joint functionalist manifesto (Merton 1976, p. 245). The joint work did not materialize but Davis produced several functional analyses and Merton weighed in with a high-impact methodological analysis published in his 1949 collection of essays: Social Theory and Social Structure.

540

C. Crothers

SF clearly operated at the MC level, while sometimes including more individually orientated aspects. Several examples of such analyses in the late 1930s can be readily adduced: such as Willard Waller (dating, 1937) or Kingsley Davis (prostitution, Inequality, 1937, 1945). A standard account is that from the 1940s or so through the 1970s, a broad, albeit loosely organized, ‘School’ of functional analysis or structuralfunctionalism4 could be formally identified. It is usually seen as emerging from the Harvard circle of Parsons (Nichols 2010) and was vigorously pushed in the 1950s and 1960s by a stable of writers among his students: Robin Williams, Kingsley Davis, Robert Merton, and Marion Levy. However, this ‘School’ is generally reported as faltering from the 1970s under the mounting concerns that it was unable to explain change or conflict, and it was aggressively outflanked by a revival of Marxian approaches on the one hand and by an array of promising micro-analyses (symbolic interactionism, phenomenology, and ethnomethodology) on the other. However, it was not completely eviscerated and structural-functionalism has continued well beyond this period as a broad sociological doctrine, albeit while further loosening its cognitive and social cohesion. A pivotal contribution was Robert Merton’s (1949) lengthy essay on functionalism which was concerned to identify theoretical and methodological difficulties in SF, and then present a framework to facilitate SF analyses. Merton defines SF as: “the practice of interpreting data by establishing their consequences for the larger structures in which they are implicated” (1968, p. 101). Consisting of several key passages in his highly popular collection (STSS) this was a very visible and influential statement, designed to shape SF as a viable analytical approach. Merton attempts to clear away various obstacles to and difficulties of a SF programme. These include terminological issues and a defence that functional analysis is not ideologically biased, since analysts of a variety of value positions have converged on the same truths. He differentiates between two models of SF, separating focussed functional analyses from full-scale functional prerequisite analysis which he does not recommend pursuing. Subsuming the spirit of his doctrine of middle-range theorizing Merton has a preference for adding to knowledge through partial studies of particular social entities rather than attempting the far more ambitious task of trying to pin down the operation of the whole system (which a Parsonian approach is more likely to propel).

4 Merton disparaged ‘-isms’ but the suffix often lets the text flow more easily, used here without assuming essentialism. SF is variously referred to as a theory, method or mode of analysis.

Methodological Individualism (MI) …

541

Adapting functional analysis to the study of more complex societies is seen to require a rejection of three postulates usually held to be ‘essential’ to classical functional analysis (1968, pp. 79–91): these postulates hold first, that standardized social activities or cultural items are functional for the entire social or cultural system; second, that all such social and cultural items fulfill sociological functions, and third, that these items are consequently indispensable.

This rejection of the trinity of classical functional postulates has several direct implications for a reformulated functional analysis: that the consequences of social units for other areas of social life be recognized as multiple, specified as either functional or dysfunctional, and not inherently tied to a particular form (but rather, units may have alternatives). But perhaps, the leading methodological argument in this crucial essay on functional analysis concerns the ‘heuristic purposes’ of the identification of latent functions (i.e. the functions not intended or recognized by participants: see above) for theoretical work in sociology. The several examples of identification of Latent Functions identified by Merton have been subject to much subsequent criticism (e.g. see below for treatment of Campbell’s appraisal). The Hopi rain dance seems a relatively uncontroversial example. In drought conditions, it was a custom of Hopi Indians to perform rain dances, which (at least from a Western scientific viewpoint) are unlikely to bring rain. However, it can be hypothesized that such ceremonies fulfil the latent function of increasing social cohesion at a time of heightened social stress. In sum, Merton boosted SF (and more generally sociology) by codifying a range of analytical tools for analysing social structures. Parsons, though, was ‘missing in action’. His first extended writing on functionalism wasn’t produced until the mid-1940s, although after his broaching of the subject, an avalanche of writing ensued. This led a wider group of sociologists to attempt to pin down a formal structural–functional (pre-requisites) model of society, and its major institutions. An even wider group of writers contributed methodological reflections or analyses of particular institutions. Thus, in the immediate post-WW2 period structural– functional sociology became arguably the discipline’s ‘first’ widely shared explanatory framework, albeit one represented more by its labelling, than by its actual analytical practice. It is important to try to conjure up a picture of what an actual SF can look like. After reviewing much of the relevant literature it is possible to argue that some of the features usually associated with SF include methodological and theoretical tendencies to:

542

– – – – – – – – – –

C. Crothers

be couched at a Macro sociological level be comparative in scope, open to scanning a wide array of societies stress interrelationships amongst aspects of collectivities focus on the recent past (the ‘ethnographic present’) rather than longer historical spans see as sociologically significant values, norms, roles, etc. attempt to separately identify Manifest vs. Latent layers of functions expect citations from appropriate scholarly authorities include many textbook publications5 often claim value neutrality explore micro-foundations, but often in ad hoc ways.

More broadly, SF in the several decades after WW2 can in turn be seen as being set within ‘Standard American Sociology’ (Mullins 1973) as a loosely connected assemblage of related approaches, which covered much of the sociology in the early post-war years, although a much smaller proportion more recently: – – – – –

Functional systems analysis Structural analyses (especially of role structures) Cultural analyses of values Social action theory Survey (and other systematic) social research.

These various components travelled on different trajectories, and had various interpenetrations (e.g. see below for links between functional analysis and survey research). My view is that SF had morphed into a general-purpose language for describing and explaining institutional patterns that has been deployed as a broad vision by almost all sociologists. This is rooted, too, in the ‘textbook sociology’ needed to provide introductory surveys to generations of sociology students (see below). Such treatments in part take the form of ‘long march through the institutions’ with the number of separate institutional areas (economy, religions, etc.) covered depending on the number of weeks available for instruction. The SF language allows sociologists to describe institutions, noting their geographical and other variations and essential forms (if any), to provide evocative structural descriptions and illustrations,

5

Some tightly organized others less so, and varying in sophistication.

Methodological Individualism (MI) …

543

and to provide historical and functional analyses. As well as guiding textbook sociology, such a residual SF provides an all-important springboard for social research: for example, certain functional consequences of an institutional pattern might be hypothesized which can then be checked out using systematic social research. Similarly, the general-purpose language facilitates the development of explanations for raw social research findings, since the structural relevance of social facts needs to be interpreted.

3

MI in Sociology

Despite the general MC thrust of Sociology, MI doctrines have broken through at various points. Parsons’s The Structure of Social Action (1937) was a manifesto for the development of formal theory, with his focus on the unit act specifying a MI intention. The components of a unit act are that it has an active, conscious agent, aiming at particular ends, who must choose amongst a range of choices or means available to them, an environment exercising various constraints, and a normative orientation, which provides the central subjective viewpoint. Action is more-or-less rational, but not only shaped by objective factors, but more by a subjective frame of reference. In order to analyse any action, the point of view of the actor whose actions are being considered has to be ascertained. Talcott Parsons clearly locates his unit act within a MI frame: “There are no group properties that are not reducible to properties of systems of action and there is no analytical theory of groups which is not translatable into terms of the theory of action” (Parsons 1937, p. 747). However, more interestingly, in later writing, as he turned to more systemic concerns, Talcott Parsons extends the scope of his action framework. “In Toward a General Theory, the scheme was substantially revised and its relevance extended from role-analysis in the social system to the analysis of all types of systems of action” (1960, p. 467). “In carrying out analyses at a level of the total action system, the concept ‘actor’ is extended to define not only individual personalities in roles but other types of acting units – collectivities, behavioral organisms, and cultural systems”. (Parsons 1960, p. 468)

As a result of this extension, he resiles from using psychological references such as motivation. Merton provided a more concrete analytical framework of the unit act at much the same time in his study of the unanticipated consequences of purposive social action (1936). (This was confined to isolated purposive acts

544

C. Crothers

compared the coherent system of action.) In summary, “The consequences result from the interplay of the action and the objective situation, the conditions of action”. This is directed towards intended and anticipated outcomes, which are relatively desirable to the actor, although outcomes were often bedeviled by other factors throwing off intentions. However, neither strongly further pursued this path of analysis (e.g. see Turner and Maryanski 1988). Parsons built conceptions of collectivities from the crystallization of the interaction patterns pointed to by his action theory in his Social Systems (1951). And Merton didn’t even include his social action essay in his definitive 1949 Social Theory and Social Structure compilation, and although individual-level analyses are threaded through his complete oeuvre, formal attention to their MI nature was seldom made. For Merton, individuals were occupants of multiple status/role positions, with this very multiplicity providing some freedom of choice from the constraints imposed by any one position. Moreover, people are portrayed as making choices amongst socially structured available alternatives relating to their values, knowledge and the means at their disposal (which are circumscribed by their social position and this often broadly predictable). The outcomes of their choices, and the actions ensuing, could have structural consequences, often unanticipated. Merton’s more important theoretical models each had such a structure, and in some the operation was illustrated by biographical material—although of course many were descriptive MC studies. And throughout, his focus was on outcomes, especially at an emergent level. (Merton seldom broached macro-sociological level questions and more often the collectivities of interest were at a meso level.) Nor was it Merton who drew attention to this aspect of his theoretical work, which appeared first in a brilliant commentary essay by Arthur Stinchcombe (1975: for an extension see Crothers 2020, pp. 127ff.). In his final paradigm for structural analyses (1975), Merton cleaved explicitly more to Durkheim and Marx than Weber (although his approach is very Weberian) with the listing of stipulations including both collective and individual level analytical points. Merton has retrospectively been honored6 as providing early intimations of Analytical Sociology (Hedstrom and Udehn 2009), and also as having influenced this approach through his student James Coleman. He defines social mechanisms as “social processes having designated consequences for designated parts of the social structure” (1968, p. 42). The most visible instantiation of his attention to mechanisms was an item in his paradigm 6

He attended the conference from which the collection Social Mechanisms was constructed. Analytical Sociology is usually seen as an ally of MI.

Methodological Individualism (MI) …

545

of functional analysis. He notes that the study of mechanisms should be accounted, which are ‘concrete and detailed’ and specified that they are social not psychological. He then raises the issues of what is the presently available inventory of such social mechanisms and what methodological problems are entailed in discerning their operation. Broadly, his concepts focus on the mechanisms through which functions are fulfilled. The term/ concept is invoked variously through his writings: such as the mechanisms through which role-sets function. However, his use of the term is rather unself-conscious. There is no particular criticism of the covering law model of theorizing. Hedstrom and Udehn more broadly suggest that developing mechanism explanations is a good instance of middle-range theorizing. From within the citadels of SF, rebelling MI ‘social exchange’ approaches emerged in the 1960s (Cook and Rice 2001) to explicitly and directly challenge earlier collectivist approaches. George Homans (based at Harvard alongside Parsons) had earlier carried out historical analysis on medieval villages, but by changing his approach his intervention demanding ‘the return of men’ was clearly impactful. Indeed, he “…replaced Parsons as the most cited in the US by 1964” (Steinmetz 2005, p. 119). His analysis of ‘elementary forms of behaviour’ was based on reductionist psychological behaviourism. “Homans’ theoretical formulation portrayed human behaviour as a function of its payoffs, the consequences of rewards and punishments” (Cook and Rice 2001, p. 700). Homans then developed explanations for phenomena such as “…cooperation, conformity and competition, structures of sentiment and interaction, status and influence, satisfaction and productivity, leadership, the exercise of power and authority, distributive justice and the emergence of stratification” (Cook and Rice 2001, p. 701). Peter Blau had carried out an analysis of interactions amongst social workers in formal organizations for his Columbia Ph.D. research and (while at the University of Chicago) extended this into a theoretical approach: “…Blau moved beyond the micro level to the institutional level, dealing with authority and power, conflict and change in the content of institutionalised systems of exchange” (Cook and Rice 2001, p. 701) seeing these as emergent properties. Their basis was an assumption that rational actors were acting to maximize their utility in exchanges with others. People were seen as having differential stocks of scarce social resources and were involved in exchanges of these, with social subordination to the resource-rich, a likely cost of having limited social and/or economic resources. Blau then embroidered a complex multi-level theory based on this. Somewhat ironically, after his return to Columbia, Blau later became an extreme MC proponent, dismissing any need for micro-foundations (e.g. 1990).

546

C. Crothers

A wide range of micro–macro studies were further extensions, e.g. Robert Emerson’s power dependence relations or Randall Collins’ interaction ritual chains, which became well-established, although in many ways, a more technical niche sociology that failed to catch much mainstream attention. Nevertheless, it is clear the supposed hegemony of SF had not precluded the development of MI approaches.

4

Implementing (Tooling Up for) Analytical Complexities in Sociology

Over the immediate post-war period, the Columbia University Department of Sociology (together with its research unit the Bureau of Applied Social Research) hosted a remarkable programme developing sociology, fueled by a stream of postgraduate students and a range of funding sources. Although there was a broad alliance between the Columbia Tradition sociological enterprise and that of the Harvard Department, the Columbians were far less committed to SF (and therefore, MC) and often critical of it (in part for its lack of explanatory heft). Merton and Lazarsfeld were a dueling ‘couple’ who spent much time in discussion. The dynamic between them partly involved theory methods specializations, but largely revolved around what constituted legitimate sociology above and beyond psychology. In their conversations, Lazarsfeld tended to push back7 against the idea of collective properties, seeking clear justifications, and procedures for identifying such a level of analysis. Several methodological developments ensued: – Elaboration/causal survey analysis through the use of control variables – Scaling and classifications – Panel studies: where causality is better tested by studying respondents at several time periods (often using ‘turnover’ tables) 7 Reflecting on their relationship Merton later reported on Lazarsfeld being pulled, over time, towards more collectivist emphasis:

The further convergence between us during our years of collaboration did involve transitions in the foci of our respective interests. For Paul, it chiefly involved a shift from a prime interest in determining the place of psychological attributes of individuals and processes in social action to such quintessential sociological matters as ‘the measurement of collective properties, contextual propositions, the analysis of macro-sociological variables’. (1998, p. 176)

Methodological Individualism (MI) …

547

– Relational analysis: where the interactions and influences of people on others are asked about and even added to the study (e.g. through snowball sampling) – Diffusion studies through networks (following chains of adoption of an innovation: e.g. study of the inter-physician spread of drug prescriptions) – Network studies examining various dimensions of cliques and their effects – Contextual studies where differences between research sites were traced of the effects on members (e.g. Merton’s housing studies in which some of the effects of communities with different characteristics were examined). Lazarsfeld and his students experimented with many of these and worked out how to analyse each, with many other sociologists (e.g. Sam Stouffer, of Harvard) also contributing. These methodological innovations bring in both collective and individual features. Elaboration analysis involves data analysis procedures, which facilitate the identification of causality. The original empirical relationship to be ‘elaborated’ was a cross-tabulation (usually presented in percentages) of the distribution of dependent variable results (e.g. political party voting intention) by each category of an independent variable (e.g. social class of respondent). Often, a relationship was found, and then the analysis would further examine whether this was stable once other variables had been brought into the analysis: for example, it might be shown that a third variable better explained the dependent variable and that the first analysed independent variable had only a ‘spurious’ apparent effect. This data analysis approach was later overtaken by regression and especially log-linear models. In an attempt to map out the various units and their different properties, that can be deployed in sociological analyses, Paul Lazarsfeld and Herbert Menzel (1956/1993, p. 172) developed a very useful typology. They say: This paper attempts to clarify some of the operations involved in the construction and use of such [i.e. collective] variables in empirical research and provides a nomenclature for the different ways in which information about individuals and about collectivities may be interwoven in these properties. The properties will be classified according to the measurement operations involved in their construction (Table 1).8

8 More widespread utilisation of this superb typology has been hampered by its low-visibility publication history.

548

C. Crothers

Table 1 Individual and collective properties (Lazarsfeld and Menzel 1956) Relation Collectivities Analytical Structural Global Members Absolute Relational Comparative

Contextual

Operation

Example

Mathematical operation on each member Relational, comparative Unique property of collectivity

Average

Characteristics of individuals Relationships between members Comparison of characteristics of individuals against distribution of all members of collectivity Describes members by property of collective

Age Rating of other members Sibling order

‘Best-liked’ member Degree of democracy

Being a member of a large group

Within the Columbia Tradition energy for developing new methods (e.g. network analysis) was exerted by the group of ‘young Turks’,9 which included Charles Kadushin who was to become a specialist expert on networks and James Coleman who was in many ways a culmination of the several strands of the Columbia Tradition, incorporating both Merton’s interest in theory development (although Coleman was far more formal in his theorizing) and Lazarsfeld’s interest in methods and mathematics. In his comments on his vision of sociology, Coleman illustrates several of the points made in this chapter. My years in graduate school, the early 1950s, constituted a kind of watershed for sociology, best exemplified by the contrast between community studies which were dominant until then, and survey research, which was dominant after. One might think of the Lynds’ Middletown as the typical community study, but there were many others …. The typical survey project could be exemplified by the Michigan voting studies, the first of which, titled The Voter Decides, studied the 1952 election, or earlier, Stouffer’s American Soldier volumes. (Coleman 1994, p. 29)

Some differences between the two types of study were their different methods and units of analysis. The focus in survey research is on variation in terms of a dependent variable (e.g. individual’s actions or attitudes) and then

9

This is a common term for any reform group, referring to the movement for Turkish political reforms in the early twentieth century.

Methodological Individualism (MI) …

549

flows of causation from independent variables—which can include collective properties—which sought to measure variation in the dependent variable ‘caused’ by the independent variables. The two figures who symbolized the two sides of the watershed were Robert S. Lynd and Paul F. Lazarsfeld. … Yet what was most striking, was that this watershed was being crossed … most of the work having elements from the new survey research side also contained strong elements of the community study side. (Coleman 1994, p. 30)

Indeed, the post-war Columbia studies were a halfway house: they were sited in communities rather than having a national-level framing and they included multi-methods. These attributes were clearly strengths, but they were also limitations as society had changed such that single stand-alone communities no longer so closely bound the fate of residents: labor markets, media circulations, and geographical mobility all now transcended single communities (Coleman 1986). This shift renders smaller-scale studies less pertinent and encourages nation-level surveys, which can be covered relatively cheaply by reasonably small samples. Coleman adduces other forces behind the move to survey research. In part, the decisive shift was due to the increased prominence of government agencies in funding of applied social research. Governments are less interested in analysis of social system functioning than in descriptive statistics for well-defined populations, such as the population for which they have some responsibility. Thus they will favor research which gathers data on a representative sample, from which inferences about the population can be rigorously drawn. (Coleman 1994, p. 30)

And they are little interested in studies of social systems which are more central to sociological analyses. Taken all together, this then is my vision for sociology. One part of that vision is to combine the substantive focus on the social system that characterized the pre-watershed period with quantitative methods and models developed and used since the watershed. This implies the use of a theory of action, because it involves a micro to macro transition. In some cases, it may involve formal modelling as well, since the linkage between individual action and system behavior is a complex one. A second part of that vision is increased utility of social theory for the ongoing functioning of society. (Coleman 1994, p. 34)

550

C. Crothers

Coleman is usually lauded for his (MI) rational choice approach, which indeed he was strongly committed to. However, the interest in collectivities (corporate entities), which was so prominent in his earlier career continued and much of his later analysis involved corporate entities. As with other theorists, he was able to ‘scale up’ frameworks appropriate at a more individual level to also examine collective entities (see Marsden 2005). In sum, the Columbia Tradition (and allies) stood out from the mainstream of US sociology to develop a multi-level approach, which continues to be relevant to contemporary sociology.

5

Controversies Illuminating the Relationships Between MI and MC in Sociology

In support of this argument for the continuance of a modified sociological visions including an MC emphasis, this chapter now reviews arguments relevant to the untangling of MI and MC positions variously advanced by: – – – – – – – –

The Ecological Fallacy Functionalism and Mainstream Sociology The Manifest/Latent Distinction Functionalism and textbook sociology Functionalism and Survey Research Methodological Positivism Empirical data on MI and MC in Sociology National Traditions.

The Ecological Fallacy MC was central in many social research studies, especially up to the postwar period. A particular temptation was to deploy data on characteristics of aggregates and to provide explanations of these in terms of the correlated distributions of other collective characteristics. Given the limited supply of published census and other data and the primitive data analysis technologies then available, such analyses seemed to be useful first steps: Moreover, such data was often mapable, which can be vividly illuminating. In his famous analysis of what became termed ‘the aggregate (or ecological or population) fallacy’, William Robinson (1950) found that several patterns that appeared

Methodological Individualism (MI) …

551

highly correlated at a collective level, failed to show such patterns in individual data. For example, he found a high negative correlation between the 1930 state-level illiteracy rate and the proportion of the population born outside the US which became a small positive correlation at the individual level which was explained by immigrants tending to settle in states where the non-immigrant population was more literate. As noted by Firebaugh: It would be difficult to overstate the impact Robinson’s article has had on social science research during the second half of the twentieth century. The use of ecological correlations to study individual-level relationships had been commonplace before Robinson’s article, and the article sharply curtailed that practice. The article also served to motivate the development of survey research. (cited in Subramanian et al. 2009, p. 4024)

However, Subramanian et al. (2009) argue that an assumption was made by Robinson that the individual level was the default correct level and as a result collective-level relationships were often incorrectly neglected in subsequent studies. They go on to argue that the 1950s development of McCarthyism and Cold was ideological individualism meant proponents of collectivism sometimes found themselves under fire (or even prosecution). It is possible that these ideological currents affected the take-up of particular theoretical approaches, stunting the development of the full range of empirical research. The study team is able to make important analytical gains by including then-available but unused data (e.g. on the ‘Jim Crow’ context of different States) to provide explanations of the differences between patterns at the two levels. And they consequently advocate multi-level research.

Functionalism and Sociology Both Kingsley Davis (1959) and later Si Goode (1973) argued that SF had become merged with sociology as a whole (or large portions of it). They observe that in much sociological writing, the terms seemed to have become interchangeable. Davis seems to have made a U-turn concerning the promise of functionalism as he had been a particularly active early architect of the approach. Davis argues that functionalism and comparative sociology in general share the same questions and notes the commonality of those questions: …most closely identified with functionalism—for example: What features of social organization or behavior appear in all or nearly all societies? Why are

552

C. Crothers

these features so nearly universal while others are more variable? What particular features characterize each type of society, and how do they mesh together in the operation of that type? How in a concrete community are the parts of the social structure mutually congruent or incongruent, as exhibited in attitudes, roles, and conduct? (1959, p. 762)

Actual functional analysis can be characterized as primitive theory. Davis suggests that the ease of analysing smaller societies as totalities led to the popularity of functionalism in anthropology. He also pointed out that the analysis of functional prerequisites is particularly methodologically difficult as experimental variation is not possible and the universality of some social phenomena just has to be posited. Goode (1973) renews Davis’s argument maintaining that it is best now to look at functionalism as an “empty castle”, since there are few defenders against the attackers of functionalism. He pleads that the functionalist specter might be exorcised once and for all by translating into causal language the various propositions advanced in functionalist terminology.10 While functional arguments can be subsumed into Sociology, they need not (and do not) constitute a separately identified approach.

The Manifest/Latent Distinction A central concept of revised SF was Merton’s distinction between manifest and latent functions. Colin Campbell (1982) launched a vigorous critique of this distinction which he suggested was seldom used. More than thirty years after he first introduced the terms in the book Social Theory and Social Structure (1949), Merton’s concepts of manifest and latent function are still cited in the literature of the social sciences, especially in sociology and anthropology textbooks, where they are commonly defined and illustrated … None of these references is critical, the majority being openly complimentary, variously defining the categories as “important” … “valid” …, “useful” and “helpful”. By contrast, the judgments of theorists are more mixed, with observations on fundamental ambiguities and weaknesses typically associated with a qualified approval …. These discussions are, however, noticeable for their brevity… Finally, this picture of textbook enthusiasm and critical nearneglect is rounded out by apparently total indifference at the level of empirical inquiry. For although one can find the words manifest and latent employed in

10

Functional ‘explanations’ explain an element by its contribution to the system, so this can be reversed by showing that the state of the system is explained by the operation of its elements.

Methodological Individualism (MI) …

553

sociological discourse, it is rare to find the concepts themselves employed in the research context. (1982, p. 29)

Lengthy critique follows Merton’s definitions and examples. Campbell sees that the distinction causes difficulties as it is (according to him) a joint product of functionalism and action theory, which are not coterminous. Another major slippage Campbell identifies is that the method to measure the distinction replaces Merton’s definition of the concept of manifest function in terms of consciously intended and recognized consequences as ‘commonsense’, whereas latent consequences are those illuminated as a result of sociological analyses. Campbell’s critique is useful in reviewing a distinction where there is fuzziness, if not dubiousness. However, his criticisms seem to overshoot. The latent-manifest distinction belongs in the general purpose vision level of sociological discourse, generating hypotheses for systematic research to test.11 Consequently, it is possible, but unlikely, that claims about latent functions will be reported in the research literature which Sociological Abstracts records and which Campbell reports. And functions are indeed collective (MC) level characteristics, not to be confused (although they can be empirically related) with the comparable concepts (unanticipated consequences) at the ‘action’ (MI) level. The conflation between manifest and latent functions and sociological and commonsense knowledges also seems stretched since manifest functions are likely to be sufficiently well-known that they are experienced as ‘common sense’ by many; although perhaps, a more rigorous methodology could be deployed. Campbell’s issues seem to almost vanish once correct interpretations are provided.

Functionalism and Textbook Sociology While SF has long been abandoned as a living tradition of sociological research it seemingly is a lingering presence within the genre of textbook sociology, as a generalized marker for various forms of a MC approach. This section reviews the trajectory of SF content in English-language sociology textbooks. In his study of UK textbooks, John Scott suggests that a mainstream approach had developed through the 1930s and 1940s. “This was not fundamentally altered until after the Second World War, when US structural– functional sociology began its major influence and dominance of the field 11

Indeed, latent functions might be indirectly posited from reflection on empirical findings.

554

C. Crothers

until the proliferation of new and radical perspectives in the 1960s” (2014, p. 211). A mark of the SF texts was their use of ‘social systems’ terminology. The voluptuous US undergraduate market has produced not only large numbers of textbooks but also many studies of them: I draw mainly on Jeff Manza et al. (2010). Undergraduate texts are ‘big business’ and publishers take a controlling interest in the broad organization and content (but not detailed content) and especially the presentation of textbook material. Textbook sociology may well have a poor structural ‘fit’ with research sociology since, as Manza et al. (2010) point out, the undergraduate sociology market is largely located in lower level tertiary education institutions, that almost all texts have been written by authors from such institutions, and that many textbook authors see themselves as detached from the research front. Textbook sociology has attracted much criticism in published reviews for its outdated inclusion of material and many other deficiencies. Manza et al. (2010) depict mainstream sociology textbooks as locked into providing a standardized sequence, including an early placed presentation of theory, together with methods, and then followed by chapters on culture, social structure, socialization, and inequality, followed by coverage of various institutional areas (Manza et al. 2010, p. 277). Theory chapters are almost always presented in terms of an SF, conflict, symbolic interaction triad with very few other positions or authors included. They go on to explicate how SF is usually described: The structural-functional paradigm is a framework for building theory that sees society as a complex system whose parts work together to promote solidarity and stability. As its name suggests, this paradigm points to social structure, meaning any relatively stable pattern of social behavior. Social structure gives our lives shape, whether it be in families, the workplace, or the classroom. This paradigm looks for a structure’s social functions, or consequences for the operation of society as a whole. All social structure - from simple handshake to complex religious ritual - functions to keep society going, at least in its present form .... (Manza et al. 2010, p. 281)

Moreover, Manza et al. (2010, p. 282) suggest that it is a particular brand within the ‘broad church’ of Functional analysis that is presented: Textbook presentations of “structural functionalism” not only postulate a vision of society that few contemporary American sociologists hold, but also typically reproduces a version of the most grandly ambitious of Parsons’ (1951) midcentury work, where whole social systems are explained by their underlying

Methodological Individualism (MI) …

555

functions …. It is, in other words, a particular and narrow version of functionalism that is routinely invoked…, largely devoid of any role for social action that Parsons’ modern defenders would insist upon …..

Manza et al. (2010) compare this triad with sociology’s current traditions, pointing out that there is no extant grouping of functionalists, while conflict theory is also no longer a living tradition but a pastiche of disparate lines of analysis. Although Symbolic Interactionism retains some vitality (e.g. there are still at least two active symbolic interaction journals), it doesn’t exhaust the range of available micro-sociologies. Moreover, in as much as SF and Conflict Theory are mirror images of each (both MC) including both is somewhat redundant.12

Functionalism and the Survey In the post-war period, there seemed to some observers to be a conjunction between MC and survey research methods with its MI tendency. This apparent link is addressed by Jennifer Platt in an article (1986), which examined the extent to which the development of particular research methods is shaped by, or linked to, theoretical developments. She instances the widely bruited purported nesting of survey research with functionalism. After lengthy considerations, Platt reaches the conclusion that although there was a definite historical coincidence, at the departmental level, of joint emphases on functionalism and of survey method, each arose out of “separate chains of causation which converged in time”. In sum, Platt argues that in the postwar period, SF was hegemonic in the theory sphere while survey research was hegemonic in the methodology sphere, with the pair constituting ‘standard American sociology’ (Mullins 1973). 12 As well as simple labels, theoretical positions are often marked by the names of key theorists. Thus, for example, Mallory and Cormack (2018, p. 1) make a similar argument about overuse of simplistic markers of theoretical positions in their complaint about the presentation of Durkheim’s work in popular English-language Canadian sociology textbooks which show two distinct Durkheim.

First, they characterize him as a founder of the discipline and the sociological project of challenging common-sense explanations of social life. Second, Durkheim appears as the father of structural functionalism who advocates a conservative, integrating vision of society. We argue that to understand why these two versions of Durkheim persist in sociology textbooks, we must appreciate the symbolic place of classical authors in the discipline. The two “textbook Durkheim’s” endure because they operate as symbols for both the coherence and divisions of the discipline.

556

C. Crothers

Platt is indubitably correct, and indeed no obvious virtuoso sociologist spanning both can be pointed to.13 Moreover, disciplinary (or subject matter) differences in the relationship between theory and methods are instructive: whereas Anthropological functionalism is paired with in-depth ethnography, sociological functionalism is not so connected. Platt is overrestrictive in limiting her discussion to survey research when a wider term such as systematic social research would have been more encompassing. Platt’s generalization that theory and methods are not linked is too quickly made. The procedures of any method are underlaid by a theoretical framework, although this is often not made explicit. For example, survey research is based on theories concerning appropriate sampling frames, the effect of social interaction in interview situations on the information yielded and theories about the subject content asked about in interviews. SF indeed does relate to survey research through a shared understanding of social subject matter. SF emphasizes values and related items such as knowledge, attitudes, emotions, norms, behaviours, etc., and these are the very content of much survey research, and are usually deployed as ‘dependent’ variables. Moreover, functional analysis often referred to the status/role structures of societies, which are built into the routinely collected information on sociodemographic variables at the heart of survey analysis—these form the ‘independent variables’ on which variation of dependent variables is plotted. As Merton pointed out: …such categories as sex, age, education and income happen also to correspond to some of the chief statuses in the social structure, the procedures evolved for audience measurement by the students of mass communication are of direct interest to the sociologist.... (Merton 1949/1968, p. 505)

The lesson is that while methodologies clearly do have theoretical foundations, the relationships may be complex and need to be carefully considered.

Methodological Positivism (MP) Methodological Positivism (MP) tends to push towards individualism, but is based on a radically different approach to explanation compared to MI, which is resolutely non-positivist in terms particularly of covering laws, preferring instead the modelling of generative mechanisms. George Steinmetz (2007) argues that, MP was a powerful doctrine such that in contrast to an interwar ‘intellectually splintered discipline’, post-war sociology was quite 13

The pairing of Merton and Lazarsfeld requires two sociologists to bridge both aspects.

Methodological Individualism (MI) …

557

unified (whether or not particular writers were supporters) around a doctrine of ‘methodological positivism’ or ‘natural science’. MP has three main dimensions: – epistemological commitment to covering laws, albeit probabilistic – empiricist ontology: i.e. science about empirically observable events – scientism: natural and social science use the same approach social facts.14 Steinmetz draws a link with survey research: “Although we cannot equate quantitative or statistical methods with MP, they clearly certainly had elective affinities” (2007, p. 361). He goes on to provide a broad and highly speculative sociological linkage pointing to the development of Fordism as a basic social condition: especially. Stable predictable social life involving not just industrial work, but also consumption and ideas, was highly compatible with the development of MP. After the war social reality became more orderly and was presented using tropes of stability, repetition and ‘the end of history’. All of this corresponded more closely to the positivist expectation that social practices can be subsumed under universal covering laws. Social actors now seemed increasingly atomized and interchangeable, losing any distinctive cultural peculiarities and thus lending themselves to general models of subjectivity…. (Steinmetz 2007, p. 364)

He suggests that the massive federal funding of social science during and after WW2, on top of substantial private funding, helped shape this approach, with its emphasis on individualism.

Units of Analysis in Sociology: Empirical Data To what extent are MI/MC emphases present in the sociological literature, or held by particular communities of sociologists, at various times/places? There are some relevant data on the theoretical approaches supported by sociologists, especially North American sociologists, and some relevant data on the empirical units deployed in journal articles. Apparently, none, however, directly measure attitudes to or use of MI compared to MC interpretations. Some surveys of sociologists’ theoretical affiliations have shown a precipitate decline in support for functionalism. Whereas, the figure was 82% in 14

“Scientism’s specifically methodological implication has been that the social sciences should strive to become quantitative and experimental like the natural sciences and should eschew normative evaluations” (Steinmetz 2007, 317).

558

C. Crothers

1967, it had dropped to 17% by 1992 and more recently (amongst Canadian sociologists) was negligible. Sanderson and Ellis’s small 1992 survey of American sociology theorists recorded (reported in Crothers 2005, p. 97) showed major concentrations lying amongst Conflict, Eclectism, and Symbolic Interaction (about a quarter each); Functionalism and Structuralism (with about a fifth each) with smaller proportions cleaving to Marxism, Weberianism, Phenomenology/ethnomethodology and Exchange theory. Michalski (2016) shows that the contemporary array of theoretical affiliations stresses conflict theory,15 feminism, etc. There is some older data concerning the entities that populate the main US sociology journals in the late 1970s. Four levels were identified (Crothers 2005, p. 98 drawing on the study from Wells and Picou): – L1 individual/role was the unit in about half the articles (amongst Symbolic interactionist articles, not surprisingly, up to 73.4%; – L2 population/aggregations/classes 5%; – L3 groups/families/communities/associations 20–30%; – L4 institutional/societies/confederations 15–25%. Coding for levels is problematic, and without examples being provided, it is not easy to envisage what the differences might be. The data suggest that at that point US sociology seemed broadly split between individuals as units and collectivities. Amongst the latter there again seemed to be a fairly even split between the meso- and the macro-levels. Given the prevalence of survey research, which might be expected to be coded at level 2, the tiny proportion reported seems unexpected. And of course, the contemporaneous mix may be quite different.

National Traditions This chapter has focussed on fairly recent decades in the US because of the importance and availability of relevant source material. But some historical treatments may also identify long-seated National Traditions of sociology that may favour MI or MC. Donald Levine (1995) argues that while UK sociology tends to the former, French is more staunchly MC as are Hellenic, Marxian, and Italian sociologies. German sociology stresses the power of language and human action. He then maps a listing of current orientations back to these 15

It is interesting given Manza et al.’s (2010) strictures about the lived reality of conflict theory that it is fact so popular. And that a few still cleave to SF!.

Methodological Individualism (MI) …

559

traditions. He argues that the predilection for MI/MC mix partly arises out of the personal life experiences of sociologists but also national political circumstances. While his account is admittedly highly speculative it is incredibly insightful (Levine 1995, pp. 273–275): – Britain: “The British preoccupation with the properties, rights, and utilities of individual actors fits well with the long British struggle for personal liberties and with Calvinist strains… that focused attention on the individual’s personal conduct, while the ability to take the nation-state for granted relieved British thinkers from attending to issues of collective identity and organisation…”. – The French belief in collectivities and that these are natural phenomena arises from “The persisting French political need to integrate its disparate provinces into a national whole and the communal orientation of organised Catholicism”. – Germany: “Certain features of German culture made their thinkers especially averse to the project of grounding an ethic on foundations supplied by the lawful properties of natural phenomena. These included the inwardlooking cast of Lutheranism, with its doctrines of salvation through faith and its celebration of subjective freedom. In addition, the highly repressive character of German political life has often been associated with a German tendency to find freedom in an ideal realm since it was unavailable in political reality”. – Italy was a “late-blooming nation-sate that looked anxiously for a stratum of competent rulers” leading to an emphasis on elites and their role in collectivities. – US culture favoured both scepticism of rigid categories and established authorities, and promotes a tendency towards social activism. Levine’s listing is organized historically so that earlier national traditions can influence later: indeed he sees complex dialogic interactions over time. Further, substantiation of Levine’s glimpses would be difficult but doubtless most rewarding.

6

Conclusions: Sociological Visions

A broad distinction made in science studies lies between the ‘research front’ and the consolidated ‘core’ of a discipline. Their literature behave according to different dynamics. Thomas Kuhn (1962) argues that textbooks produce

560

C. Crothers

stable and formulaic presentations of the then-dominant paradigm. Cole (1992) suggests that only a trickle of well-confirmed findings are slowly added to the core from the research front. And there can be fractured slippage between the two. My conception is that, while hard-packed consolidated knowledge is indeed part of the core, the core also functions in an expansive way, it involves a vision (and sometimes, a mandate for achieving this). A core vision of sociology casts a broad purview of the discipline’s subject matter. There can be gaps between the discipline’s demands for knowledge and the supply through research: E.g. just because there is no research on a topic does not mean that authors drawing broadly on disciplinary knowledge (and perhaps non-research-based sources of information) won’t address that topic. The core’s vision is also drawn on in constructing textbooks and teaching curriculums. A major source of the current sociological vision is a continuation of the SF (and Conflict Theory) of the early post-war period, with many accretions and some losses. This vision remains particularly weighted towards an MC stance. The research front in sociology, though, is far more sophisticated. Multi-level studies are frequently carried out, although there seem to be few manifesto statements extolling this approach.16 It is possible to develop intimations of a sociology of how the mix of MI/ MC in sociology has varied over time. Unfortunately, this necessarily remains quite speculative. Exploration should be possible by measuring the MI/MC mix through content analyses of texts, while tables reporting data could be coded for the type of unit appearing in them. It just is that such studies (with one exception—see above) have not been done. There does seem prima facie evidence. Individualistic approaches were boosted from the 1940s on by the rapid growth of survey research, and by expansive funding support with academic and policy interests in the attitudes/behaviours of broad population samples. It is possible, too, that they were further propelled by the increasing Fordism and broader geographical scale of societal configurations and by the sharp ideological drive of McCarthyism and individually orientated political ideologies. Nevertheless, individualistic approaches were far from hegemonic, and MC approaches continued, especially in the theory portions of sociology textbooks. In particular, the Columbia Tradition seemed to hold itself somewhat apart from wider trends in Sociology, although its approaches later diffused widely. Although individualism was placed more firmly on the sociological

16

One textbook which more than most is organized to draw attention to different levels is Johnson (2008).

Methodological Individualism (MI) …

561

map, a wider set of tools were developed which have been very considerably continued since. The inadequacies of the current sociological visions presented in textbook theory chapters suggest a need for more powerful and adequate alternative visions of Sociology. The consensus/conflict dichotomy needs to be replaced by a consensus, competition, conflict continuum, applied as appropriate and relevant without prior sorting into essentialist categories. Levels (and perhaps various levels separately treating at least meso- and macro-) continue to be important and an appropriate terminology for discussing these would be a useful component of a vision. Degree of social cohesion is an important dimension and maybe sociological approaches pertinent to particular levels of social cohesion might be a useful addition in portrayals of a sociological vision. But, in addition, the vision sociology espouses must involve much more of the range of both MI and MC traditions which are so important in the sociological research arena.

References Blau, P. (1990). Structural Constrains and Opportunities: Merton’s Contribution to General Theory. In Clark, J., et al. (Eds.) Robert K. Merton: Consensus and Controversy (pp. 141–158). London: Falmer Press Campbell, C. (1982). A Dubious Distinction? An Inquiry into the Value and Use of Merton’s Concepts of Manifest and Latent Function. American Sociological Review, 47(1), 29–44. Cole, S. (1992). Making Science: Between Nature and Society. Cambridge: Harvard University Press. Coleman, J. S. (1986). Social Theory, Social Research, and a Theory of Action. American Journal of Sociology, 91(6), 1309–1335. Coleman, J. S. (1994). A Vision for Sociology. Society, 32(1), 29–34. Cook, K. S. & Rice, E. (2001). Exchange and Power: Issues of Structure and Agency. In Turner, J. H. (Ed.) Handbook of Sociological Theory (pp. 699–719). New York: Springer. Crothers, C. (2005). The Diversity and Insularity of Sociological Traditions. In Calhoun, C., Rojek, C. & Turner, B. (Eds.) The Sage Handbook of Sociology (pp. 79–102). London: Sage Publications. Crothers, C. (2020). Reintroducing Robert K. Merton. London: Routledge. Davis, K. (1937). The Sociology of Prostitution. American Sociological Review, 2(5), 744–755. Davis, K. & Moore, W. E. (1945). Some Principles of Stratification. American Sociological Review, 10(2), 242–249.

562

C. Crothers

Davis, K. (1959). The Myth of Functional Analysis as a Special Method in Sociology and Anthropology. American Sociological Review, 24(6), 757–772. Durkheim, E. (1895/1982). Rules of Sociological Method . New York: Simon and Schuster. Elder-Vass, D. (2012). The Reality of Social Construction. Cambridge: Cambridge University Press. Elster, J. (1982). Marxism, Functionalism, and Game Theory. Theory and Society, 11(4), 453–482. Goode, W. J. (1973). Explorations in Social Theory. Oxford: Oxford University Press. Greenwood J. (2003) Social Facts, Social Groups and Social Explanation. Nous, 37(1), 93–112. Hedstrom, P. & Udehn, L. (2009). Analytical Sociology and Theories of the Middle Range. In Bearman, P. & Hedstrom, P. (Eds.) Oxford Handbook of Analytical Sociology (pp. 25–50). Oxford: Oxford University Press. Johnson, D. P. (2008). Contemporary Sociological Theory: An Integrated Multi-Level Approach. New York, NY: Springer. Kuhn, T. (1962). The Structure of Scientific Revolutions. Chicago: Chicago University Press. Lazarsfeld, P. & Menzel, H. (1956/1993). Analyzing the Relations Between Variables: On the Relation Between Individual and Collective Properties. In Boudon, R. (Ed.) Paul F. Lazarsfeld. On Social Research and Its Language (pp. 172–188). Chicago: Chicago University Press. Levine, D. (1995). Visions of the Sociological Tradition. Chicago: University of Chicago Press. Mallory, P. & Cormack, P. (2018). The Two Durkheims: Founders and Classics in Canadian Introductory Sociology Textbooks. Canadian Journal of Sociology, 43(1), 1–24. Manza, J., Sauder, M. & Wright, N. (2010). Producing Textbook Sociology. Archives Europeennes de Sociologie, 51(2), 271–304. Marsden, P. V. (2005). The Sociology of James S. Coleman. Annual Review of Sociology, 31, 1–24. Merton, R. K. (1936). The Unanticipated Consequences of Purposive Social Action. American Sociological Review, 1(6), 894–904. Merton, R. K. (1949). Discrimination and the American Creed. In Maclver, R. (Ed.) Discrimination and National Welfare (pp. 99–126). New York: Harper & Brothers. Merton, R. K. (1949/1968). Social Theory and Social Structure. New York: The Free Press. Merton, R. K. (1975). Structural Analysis in Sociology. In Blau, P. (Ed.) Approaches to the Study of Social Structure (pp. 21–52). New York: The Free Press. Merton, R. K. (1976). Sociological Ambivalence. New York: The Free Press. Merton, R. K. (1998). Working with Lazarsfeld: Notes and Contexts. In Lautman, J. & Lecuyer, B.-P. (Eds.) Paul Lazarsfeld 1901–1976. La Sociologie de Vienne à New York (pp. 163–211). Paris: L’Harmattan.

Methodological Individualism (MI) …

563

Michalski, J. H. (2016). The Epistemological Diversity of Canadian Sociology. Canadian Journal of Sociology, 41(4), 525–556. Mullins, N. (1973). Theories and Theory Groups in Contemporary American Sociology. New York: Harper & Row. Nichols, L. (2010). Merton as Harvard Sociologist: Engagement, Thematic Continuities, and Institutional Linkages. Journal of the History of the Behavioral Sciences, 46(1), 72–95. Parsons, T. (1937). The Structure of Social Action. New York: McGraw-Hill. Parsons, T. (1951). The Social System. New York: The Free Press. Parsons, T. (1960). Pattern Variables Revisited: A Response to Robert Durbin. American Sociological Review, 25(4), 467–483. Platt, J. (1986). Functionalism and the Survey: The Relation of Theory and Method. The Sociological Review, 34(3), 501–536. Robinson, W. S. (1950). Ecological Correlations and the Behavior of Individuals. American Sociological Review, 15, 351–357. Scott, J. (2014). Building a Textbook Tradition: Sociology in Britain, 1900–68. In Holmwood, J. & Scott, J. (Eds.) The Palgrave Handbook of Sociology in Britain (pp. 211–235). Hampshire: Palgrave Macmillan. Shils, E. & Finch, H. (1949). Max Weber on the Methodology of the Social Sciences. New York: Free Press. Steinmetz, G. (2005). The Genealogy of a Positivist Haunting: Comparing Prewar and Postwar U.S. Sociology. Boundary 2, 32(2), 107–133. Steinmetz, G. (2007). American Sociology Before and After World War 2: The (Temporary) Settling of a Disciplinary Field. In Calhoun, C. (Ed.) Sociology in America: A History (pp. 314–366). Chicago: University of Chicago Press. Subramanian, S. V., Jones, K., Kaddour, A. & Kreiger, N. (2009). Revisiting Robinson: The Perils of Individualistic and Ecologic Fallacy. International Journal of Epidemiology, 38(2), 342–360. Turner, J. & Maryanski, A. (1988). Is ‘Neofunctionalism’ Really Functional? Sociological Theory, 6(1), 110–121. Udehn, L. (2002). The Changing Face of Methodological Individualism. Annual Review of Sociology, 28, 479–507. Waller, W. (1937). The Rating and Dating Complex. American Sociological Review, 2(5), 727–734.

The Phenomenological Life-World Analysis and the Methodology of the Social Sciences Thomas S. Eberle

The work of Alfred Schutz’s changed my life and I was in a continuous dialogue with him most of my life. I have published two books (Eberle 1984, 2000) and many essays on Alfred Schutz in German, in particular, in regard to phenomenology and the methodology of the human sciences and I have co-edited the volume with his methodological writings in the new German complete edition of Alfred Schutz’s works (Schutz 2010). I guess many of us would be more than happy if our own work was still published and discussed in as lively a manner as his, 50 years after our death. Alfred Schutz’s work is multilayered and can thus be viewed from a variety of perspectives. His central motive, however, was without a doubt the contribution of life-world analysis to the methodology of the social sciences. It was the declared goal of his major work The Phenomenology of the Social World to develop a “philosophically founded theory of method” for social science research (Schutz 1967, p. xxxi). Despite the numerous topical ramifications of his work, Schutz never lost sight of this goal. Accordingly, his draft of the

This article was presented as the Alfred Schutz Memorial Lecture at the meetings of the Society for Phenomenology and the Human Sciences in Arlington, VA, October 30, 2009. And was previously published in Human Studies (2010) 33, 123–139. https://doi.org/10.1007/sl0746010-9146-9.

T. S. Eberle (B) Research Institute of Sociology, University of St. Gallen, St. Gallen, Switzerland e-mail: [email protected]

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 N. Bulle and F. Di Iorio (eds.), The Palgrave Handbook of Methodological Individualism, https://doi.org/10.1007/978-3-031-41512-8_25

565

566

T. S. Eberle

structure for the planned opus “The Structures of the Life-World” features a chapter titled “Sciences of the Life-World” (Schutz and Luckmann 1973, p. xvi) as the crowning conclusion.1 In this concluding work, the diverse facets of the phenomenological analysis of the lifeworld were therefore arranged to culminate in a philosophical grounding of the sciences of the life-world. Since Luckmann decided to cut this chapter without substitution when editing the Structures posthumously, some scholars have lost sight of this basic intention. Therefore, this memorial lecture intends to discuss the interrelation between phenomenological life-world analysis and the methodology of the social sciences. Of course, the topic is much too broad for a single lecture and has to be focused . I will pursue two major goals. First, I will concentrate on the methodological postulate of adequacy, exploring how Max Weber devised it, what Alfred Schutz made of it, and how it could in my view be interpreted in a much more radical way and made much more relevant for modern qualitative research. Secondly, as I am a German-Swiss, I will report on some developments in German Europe—which comprises Germany, Austria, and parts of Switzerland—developments which are not well known over here as only very few publications were translated into other languages so far.

1

The Life-World as a Foundation of the Social Sciences

The Phenomenology of the Social World is—in my view—the key work of Alfred Schutz. It lays the foundation for all further streams of reflection in his later works. Compared to the analytical depth and systematic approach of the analyses found here, some of his later examinations seem much less elaborated—especially his methodological considerations. On the other hand, he expanded the spectrum of his analysis with important additions, particularly through his contributions on the lifeworld as the unquestioned ground of science (subsequent to the Crisis-book by Husserl [1936] 1970), on multiple realities (referring to William James 1907), on the systems of relevance as well as through the distinction between the everyday world and the world of science and the elaboration of their interrelation, in terms of different levels 1 In the quoted English edition it says “sciences in the life-world” but this is a mistake. In Schutz’s German notebooks, it is clearly written “sciences of the life-world,” which was translated correctly in the appendix of the second volume of the Structures (Schutz and Luckmann 1989: Chapter 6, p. 177).

The Phenomenological Life-World Analysis …

567

of construction and different rationalities. Because of this, two and a half decades later a more differentiated view on the methodology of the social sciences had emerged. According to the outline of chapters for The Structures of the Life-World preserved in his index cards, Schutz planned a chapter on the Sciences of the Life-World as the culmination or conclusion of his analyzes. Luckmann, however, refrained from publishing this final chapter. He stated two main reasons for this: firstly, Schutz’s drafts did not seem to add anything to his paper “Common-Sense and Scientific Interpretation of Human Action” (Schutz 1962a, b, c), which constitutes the most substantial methodological contribution after The Phenomenology of the Social World . Secondly, Luckmann made clear that his own convictions differed from those of Alfred Schutz and that due to the lack of directions, he could not have written this chapter true to Schutz’s intent (Luckmann 1974, p. xvii). Both reasons are comprehensible, but it had the effect that the immanent interrelation of Schutz’s analysis of the life-world and the methodology of the social sciences is completely absent in the Structures of the Life-World—, which should have represented the summary volume of his work. The fact that the social world is always meaningfully pre-interpreted and that this constitutes a difference between the natural and the social sciences, usually finds a consensus nowadays. Which consequences are to be drawn from this, remains, however, disputed. Which consequences has Schutz drawn? Schutz maintains that the distinctive feature of the social sciences lies in their understanding of the world of contemporaries. This follows from the distinct attitude of the scientist as an “uninterested observer,” whose interpretation of meaning is not bound to pragmatic motives but who is striving for truth. The boundaries of this quest are set on the one hand by the scientific system of relevance, especially by the present problem, and on the other hand by the corpus of knowledge of the respective discipline. Schutz subsumes the principles by which theoretical models should be construed under the following methodological postulates (Schutz, 1962a, b, 1964a, b): (1) The principle of relevance: as already mentioned. (2) The postulate of logical consistency: which is undisputed. (3) The postulate of subjective interpretation: which means that explanations in the social sciences have to refer ultimately to the subjective meaning of action. (4) The postulate of adequacy: which means that the constructs of the social scientist have to be consistent with the commonsense constructs of actors.

568

T. S. Eberle

(5) In addition, he sometimes invokes the postulate of rationality: which means that models of rational actions are preferred since they are especially evident and form a point of reference for the characterization of deviant types. This postulate is not mandatory, although economics especially continues to adhere to it. And it may be added that Schutz says in his correspondence to Adolphe Lowe, that Pareto was right when he said that rational actions constitute the realm of economics and irrational actions the realm of sociology (Schutz 1955a).

2

The Postulate of Adequacy

In my view, the postulate of adequacy is crucial, and I would like to discuss it here. When are scientific constructs adequate?

Adequacy in Max Weber’s Work Let us briefly recapitulate Max Weber, from whom Schutz draws in this matter: explanatory understanding in Weber’s sense, which captures the actual as well as the motivational meaning, has to conform to two kinds of adequacy: to adequacy of meaning and to causal adequacy: The interpretation of a coherent course of conduct is “subjectively adequate ” (or “adequate on the level of meaning ”), insofar as, according to our habitual modes of thought and feeling, its component parts taken in their mutual relation are recognized to constitute a “typical” complex of meaning. It is more common to say “correct.” The interpretation of a sequence of events will, on the other hand, be called causally adequate insofar as, according to established generalizations from experience, there is a probability that it will always actually occur in the same way. … Thus causal explanation depends on being able to determine that there is a probability, which in the rare ideal case can be numerically stated, but is always in some sense calculable, that a given observable event (overt or subjective) will be followed or accompanied by another event. (Weber [1922] 1978, pp. 11–12—emphasis added by T.S.E.)

In other words, adequacy of meaning is only reached if explanatory understanding is evident. But even an evident interpretation remains only a hypothesis as long as the criterion of causal adequacy is not fulfilled as well. Conversely, a statement that is causally adequate remains only a nonunderstandable statistical probability as long as the criterion of the adequacy

The Phenomenological Life-World Analysis …

569

of meaning has not been fulfilled (Weber [1922] 1978). In Weber’s words: “Statistical uniformities constitute understandable types of action and thus constitute ‘sociological generalizations’ only when they can be regarded as manifestations of the understandable subjective meaning of a course of social action.” Such generalizations always exhibit a “correspondence between the theoretical interpretation of motivation and its empirical verification” (Weber [1922] 1978, p. 11). Weber tried to build a bridge here between the theory of objective probability as the physiological psychologist Johannes von Kries had developed, and the interpretive tradition since Dilthey. Von Kries saw that general causal relationships between actions and results are relevant for penologic attributions, as an action must be suited to promote a certain result according to common (statistical) experience. Only if this is the case, an “adequate causation” can be stated. For example: if a coachman falls asleep and his horses take the wrong road at an intersection, and his passenger is subsequently killed by lightning, there is no adequate causation between falling asleep and the death of the passenger, “since the sleeping of coachmen in general does not heighten the possibility of being killed by a lightning, (that is) it is generally not capable of inducing it” (von Kries 1889, p. 532—transl. T.S.E.). Von Kries referred to jurisprudence, where causal adequacy has been an ongoing issue until today. Weber saw analogical issues in history and sociology. Scientific analysis has to rely on “nomological knowledge” which is composed of “certain known empirical rules, particularly those relating to the ways in which human beings are prone to react under given situations” (Weber 1949, p. 174). Since human beings tend to react differently each time and therefore divert from “empirical rules,” only probabilistic statements can be made about their actions. For this reason, Weber takes up von Kries’s notion of “adequate causation”: In relation to human action, the opposite of “coincidence” cannot be “necessity”, but only “adequacy.” Therefore, the construction of a causal relationship refers to the relative frequency of occurrence of a type of action, in other words: to its objective likelihood. If such an interrelation cannot be “adequately” established, one needs to speak of “coincidental” causation (Weber 1949, p. 185). Weber coined the term “adequacy of meaning ” in analogy to the notion of causal adequacy, and he suggests assessing the degree of adequacy of meaning “according to our habitual modes of thought and feeling” (Weber 1994, p. 7). In the neo-Kantian framework of Rickert’s work, an interpretive sociology had to deliver explanations, which fulfilled both types of adequacy: causal adequacy as well as adequacy of meaning.

570

T. S. Eberle

Schutz’s Dismissal of Causal Adequacy As is generally known, Schutz dismissed the notion of causal adequacy. Already in his early notes written in Vienna which have been recently published in German (Schutz 2007), one can read: (…) the social sciences have to reject the question of causality within the interrelations of their objects as inadequate and refer it to the realm of mechanistic world explanations, which will indeed (what can be shown a-priori) never be able to solve a single social-scientific problem, be it with the help of neurosciences, theories of psychophysical parallelisms or any similar theories. (Schutz 2007, p. 227—transl. by T. S. E.)

What Weber means by the postulate of causal adequacy, says Schutz in The Phenomenology of the Social World , is nothing else than “the postulate of the coherence of experience” (Schutz 1967, p. 232). For every typical construct, there must exist a chance that “according to the rules of experience an act will be performed in a manner corresponding to the construct” (Schutz 1967, p. 232). If causal adequacy means that the typical construct of a human act has to be consistent with the entirety of our experience and if every experience of human action implies its embeddedness in a meaning-context, “all causal adequacy which pertains to human action is based on principles of meaningadequacy of some kind or other” (Schutz 1967, p. 233). For Schutz, causal adequacy is only a special case of meaning adequacy. Subsequently, Schutz integrated both postulates into the postulate of adequacy. This move had two major implications: First, in contrast to Weber, Schutz abandons talk of causality completely, be it causal adequacy in von Kries’s sense on the level of objective probabilities, or be it on the level of subjective probabilities. Crucial for Schutz are the in-order-to-motives, which may be linked to because-motives but not in a causal sense. In The Phenomenology of the Social World , he is also quite explicit that causality in the social sciences can be substituted by rationality— in accordance with the rational choice theories developed in economics. I will come back to this. The second—and more important—implication is, that Schutz dropped the requirement of empirical verification from the postulate of adequacy, while it was an integral part of Weber’s concept. Let me repeat Weber’s quotation: Sociological generalizations always exhibit a “correspondence between the theoretical interpretation of motivation and its empirical verification” (Weber 1994, p. 6).

The Phenomenological Life-World Analysis …

571

By contrast, Schutz restricts the postulate of adequacy more and more to the adequacy of meaning in a narrow sense. In 1943, in conjunction with the remarks in The Phenomenology of the Social World , he still wrote: The postulate of adequacy requires that the typical construction be compatible with the totality of both our daily life and our scientific experience (Schutz 1964b, p. 88).

But it is exactly this aspect that he drops later on. The final version of the postulate of adequacy, as found in “Common-sense and scientific interpretation of human action” reads: Each term in a scientific model of human action must be constructed in such a way that a human act performed within the life-world by an individual actor in the way indicated by the typical construct would be understandable for the actor himself as well as for his fellow-men in terms of commonsense interpretation of daily life. Compliance with this postulate warrants the consistency of the constructs of the social scientist with the constructs of commonsense experience of the social reality (Schutz 1962a, p. 44—emphasis by T.S.E.; analogous to Schutz 1962b, p. 64; 1964a, p. 65).

The subjunctive makes us hesitate: Obviously, only the consistency between scientific and common sense constructs is addressed, while causal adequacy, that is, the conformity with experience, has been dropped from the postulate. In this version, the criterion of adequacy is already fulfilled if an action that coincides with the scientific construct would be understandable in the course of common sense thinking—obviously it is no longer important whether this action does in fact occur empirically or whether it remains a sheer model. Schutz reduced the postulate of adequacy to the aspect of meaning -adequacy. At the same time, he does not tie it to his postulate of the coherence of experience, which he said was the substitute for Weber’s causal adequacy. This shift of focus compared to the adequacy concept of Weber can in my view be traced back to his specific view of science which was profoundly influenced by his colleagues from the Mises circle, who were mainly economists. Ludwig Mises, a prominent member of the second generation of the Austrian School of economics, advanced a view of economics which consisted of apriori statements. He drew a sharp distinction between theoretical economics which were aprioristic, and economic history which dealt with empirical data. Max Weber endorsed the Austrian concept of theoretical economics— in opposition to the Historical Economic School around Gustav Schmoller

572

T. S. Eberle

in Germany—but criticized the naturalistic self-misunderstanding of aprioristic theory. In fact, he contended, they are working with ideal types. Mises reproached Weber’s misunderstanding of what aprioristic science means, and went on to develop a praxeology, an aprioristic theory of human action (published in 1940 in German, shortly before his emigration to the United States, and published in English in 1949 in a reworked edition). Schutz was caught in the middle and undertook a careful attempt to reconcile the two: Mises was right, he argued, that on a very abstract level his concepts and propositions were aprioristic, but as soon as they are related to concrete data they would inevitably become ideal types (Schutz 1967). However, Schutz hardly ever came into contact with empirical research. His knowledge of social science was above all influenced by his friends from the Mises-Circle, in other words, from Austrian Economics (cf. Prendergast 1986). The impressive work of Fritz Machlup, with whom Schutz maintained close contact, consists as well mainly of those typical non-empirical economic models which are based on relatively simple assumptions in order to keep the theoretical complexity manageable (see Machlup 1978). If Schutz describes scientific constructs as constructs of homunculi and ties the postulates of logical consistency, subjective interpretation and adequacy to them (Schutz 1962a, p. 43), then even economic models most aloof from empirical data seem to match these principles. Schutz’s methodological postulates read like descriptions of the prevalent self-conception of the economics of that time (see Eberle 1988). It shall be added at this point that the famous economist Milton Friedman from the University of Chicago, published the so-called Friedman-theorem in 1953 which found wide appeal among economists: Truly important and significant hypotheses will be found to have ‘assumptions’ that are wildly inaccurate descriptive representations of reality, and, in general, the more significant the theory, the more unrealistic the assumptions (in this sense) (Friedmann 1953, p. 14). For Friedman, unrealistic assumptions of a theoretical model are not a problem at all—the only thing that counts is the success of the prognoses derived from it. But can a model that is based on unrealistic assumptions comply with the methodological postulate of adequacy? Interestingly enough, Machlup completely takes sides with Friedman on this matter by claiming with regard to another great American economist, Samuel Samuelson, that he wrote his best works whenever he made unrealistic assumptions (Machlup 1964, p. 753). Machlup, however, holds the Friedman-theorem to be in need of additions: the assumptions would not need to be realistic, but adequate (Machlup 1954, p. 17). Needless to say, if mental images that are based

The Phenomenological Life-World Analysis …

573

on unrealistic assumptions are supposed to comply with the postulate of adequacy, it obviously must be broadly defined. Machlup defines it—with reference to Schutz—in the following way: The fundamental assumptions of economic theory are not subject to a requirement of independent empirical verification, but instead to a requirement of understandability in the sense in which man can understand the actions of fellowmen. (Machlup 1954, p. 17)

Schutz’s self-conception as a methodologist has always been characterized by modesty: “Methodology is not the preceptor or the tutor of the scientist. It is always his pupil …” (Schutz 1964b, p. 88). His conception of science was deeply influenced by the “method of mental images” (Mises 1949, pp. 227ff.) of the Austrian School of economics. This holds also true in regard to the conceptualization of rationality, in which he, however, takes sides with Weber against Mises. But the postulate of rationality that social-scientific models have to comply with is formulated subjunctively: The rational course-of-action and personal types have to be constructed in such a way that an actor in the life-world would perform the typified action if he had a perfectly clear and distinct knowledge of all the elements, and only of the elements, assumed by the social scientist as being relevant to his action and the constant tendency to use the most appropriate means assumed to be at his disposal for achieving the ends defined by the construct itself (Schutz 1962a, p. 45—emphasis by T. S. E.) Again, the subjunctive indicates that the scientific models form an (objective) possibility from which acts in the everyday world deviate more or less strongly. Earlier, Schutz had extensively elaborated the “paradox of rationality on the level of common-sense action,” which is to be found in the fact that … the more standardized the pattern is, the less the underlying elements become analyzable for commonsense thought in terms of rational insight. … Only on the level of models of interaction patterns constructed by the social scientist in accordance with certain particular requirements defined by the methods of his science does the concept of rationality obtain its full significance (Schutz 1962a, p. 33). The postulate of rationality formulated in the subjunctive, however, can again be fulfilled by all economic models, as long as they are just based on the homo economicus—even if their assumptions are completely unrealistic. Schutz obviously employs his analyses of the life-world primarily in order to explicate the differences in orientation between homunculi-constructs and actors guided by common sense—and therefore, the inevitable distance between a scientific model and social reality as it is experienced in the life-world .

574

T. S. Eberle

A More Radical Interpretation of the Postulate of Adequacy Before I became a sociologist and a phenomenologist I was trained as an economist. It is probably due to this background, that I see much critical potential in Schutz’s phenomenological life-world analysis. Many economists still uphold the self-description that their discipline is the queen of the social sciences—a designation, which Auguste Comte once reserved for sociology. At the same time, economics usually applies rather simple explanatory models of human behavior. If you would have such homunculi in an organization as Economic Men, you would never see a collaboration happen. Many economists went so far as to abolish the phenomenon of altruistic motives, by reinterpreting them into factually egoistic motives: altruists, too, are actually egoists, they just maximize their personal benefit by helping others. By this procedure, the universe of meanings of the social world is dramatically reduced. This is the so-called KISS procedure, keep the model simple and stupid, so you can handle it well. Although this is said to be a methodological procedure, it finally results in a kind of simple ontology as well. I would like to recommend therefore that we apply a more restrictive interpretation of Schutz’s postulate of adequacy: The postulate of adequacy means that the constructs of the social scientist describe and explain a concrete course of action empirically appropriately in the actor’s perspective. Schutz’s detailed analysis of the lifeworld would not make much sense if we did not require the social sciences to refer to the actor’s actual orientation in everyday life. And it is not just intelligibility of the scientific constructs which is required—economic models of man are usually highly intelligible in a commonsense perspective—it is their empirical appropriateness which is essential. And it is not just the commonsense constructs that must be referred to by a scientist, but the actual orientation of actors in their actions and interactions. As Garfinkel (1967) has shown, commonsense constructs are often just glosses, which are not precise enough to grasp the actors’ orientations. If we employ a concept of adequacy which explicitly refers to life-world analysis and implies empirical “verification” in Weber’s sense, we can reject many simplistic economic explanations as being “inadequate.” Is this in Schutz’s sense? In The Phenomenology of the Social World , Schutz calls the pure economics of Mises a “perfect example of an objective meaning-complex about subjective meaning-complexes” (1967, p. 245), and the law of marginal utility is interpreted as “a stipulation that merely marks out the fixed boundaries of the only area within which economic acts can by definition take place” (1967, p. 245).

The Phenomenological Life-World Analysis …

575

Unlike Kaufmann, whom he invokes, Schutz decouples this economic principle from the economic context and generalizes it formally. This, as I have outlined before, was a heroic attempt to reconcile Weber and Mises. However, formulated in this generalized way, the economic principle can be applied to almost anything: Not only to decisions of consumers or business persons, but also to lovers and their relationships, to employees in an organization, to the behavior of family members, and so on—Gary Becker (1991) won the Nobel Prize for his “Treatise on the Family” for analyses of this sort. The fact that Schutz as a methodologist never elevated himself to be a teacher or tutor of the economists but instead remained their scholar, could explain his reluctant formulation of the postulate of adequacy. However, he was much more critical in his personal correspondence than in his publications. In a letter to Adolphe Lowe, he criticized his teacher Ludwig Mises in no less than three different respects. (1) the decisive problem involved (in the process of choosing) is taken just for granted by Mises, that is the problem how it comes that things stand to choice at all (Schutz 1955b, pp. 5–6); (2) He overlooks also the difference which seems to be vital for me, namely on the one hand choosing between objects equally within my reach and, on the other hand, choosing between projects of actions which have to be carried out by me (Schutz 1955b, p. 6); (3) Mises is trying to develop a general praxeology which he identifies— erroneously, as I think—with the theory of economic action, namely an action according to the assumed scale of preferences of the actor. … If this were the case there would be no human action whatsoever which were not an economic action (Schutz 1955b, p. 3). And if gentlemen prefer blondes, they are economic subjects. The first two arguments point to the process of choosing in the subjective consciousness and the third to the criteria which define the realm and subject matter of economics. So, what did Schutz do about it? Schutz dealt with both problems in a longer manuscript in the middle of the 1940s. In the first part, he examined the act of choosing in everyday situations, and in the second he traces its conception in a scientific model and illustrates this with the example of theoretical economics. After having been caught between two stools with this attempt, “the philosophical part being of no interest to economists and the economic part of no interest to philosophers” (Schutz 1955a, p. 1), he finally published the first part separately in a phenomenological-philosophical journal in 1951 (Schutz 1962c) and held

576

T. S. Eberle

back the second part until the end of his life (it was published posthumously by Lester Embree: Schutz 1972). With his analysis of the acts of choosing in the life-world , Schutz hoped to show that the utilitarian modeling of acts of choosing is inadequate and that Mises’s praxeological model is in need of elaboration in several critical aspects (see Eberle 2009). A phenomenological description of the process of choosing must not be based on reconstructions of past experiences, but has to start right from amidst the stream of consciousness. This prerequisite has been fulfilled by the works of Husserl, Bergson, and Leibniz, whose findings Schutz combines for this reason. From Husserl, he takes the constitution of problematic options as the precondition to every possible choice, from Bergson the time perspectives implicated in the process of choosing, and from Leibniz the concurrence of volitive intentions which leads to the final “fiat” of the decision. On this basis, it becomes immediately clear that the utilitarian model of choosing and decision-making is just an interpretation scheme for explaining because-motives of actions already foregone and is missing the polythetic course of choosing (Schutz 1972, pp. 573f.). These statements by Schutz provide sufficient evidence to me to suggest that Schutz’s understanding of adequacy was actually more restrictive than his definitions make visible. Indeed, his lifelong effort to achieve a theory of the meaning constitution of the social world would not have made much sense if he had not aimed at contributing towards a more adequate methodology of social scientific research. Many interpreters of Schutz have overlooked the radical difference between ideal types in Schutz’s and Weber’s sense. Whenever the two are confused, a fundamental misinterpretation may result. Ideal types in Weber’s sense are constructions within a neoKantian framework, ideal types in Schutz’s sense are constructions within a phenomenological framework. Kantianism was concerned with formal aprioris and phenomenology with material aprioris. For neo-Kantianism, the (scientific) method is crucial, for phenomenology, the experience is crucial. The construction of an ideal type in Weber’s sense is based on methodical decisions by the scientist only. Scientific constructions in Schutz’s sense, however, must relate to the concrete experiences of actors in their everyday world. This does not mean that scientific ideal types must relate to the actors’ commonsense typifications, but, as has been demonstrated above with the example of the act of choosing, scientific constructions must relate to how actors actually make sense in their everyday life. Schutz’s analysis of the meaning constitution in the social world provides the link between the constructions of the first and second order. Whenever Schutz’s approach is

The Phenomenological Life-World Analysis …

577

equated with Weber’s methodology, the crucial point is missed (cf. Eberle and Srubar 2010). Schutz’s life-world analysis also provides the means to make sense of his methodological postulates. As I have argued, this results in a radicalized interpretation of the postulate of adequacy. In this vein, scientific constructions can only be accepted as adequate, if they are on the one hand designed according to the processes of meaning constitution in daily life and if on the other hand they grasp the actual orientation of actors in concrete situations empirically correctly. In this form, the postulate of adequacy would be suited to function as a quality criterion of qualitative social research and to constructively replace the usual suspects like objectivity, validity, or reliability which originated in the context of quantitative research.

3

Phenomenology and Sociology in the German Context

Schutz tied his analysis of the life-world closely to Weber’s theory of action, and his methodological arguments are oriented to Weber’s Outline of interpretive sociology (Weber [1922] 1978) as well as to the Austrian School of Economics. This happened for biographical reasons, not out of necessity. The structures of the lifeworld are not just compatible with a certain kind of sociology. Firstly, as a proto-sociology they form a framework in which in principle any kind of social science can be located. Secondly, as Garfinkel (2002, 2006) has shown with ethnomethodology, the analysis of the life-world can also be redrafted completely. Since the processes of the constitution of meaning form the core of the analysis of the lifeworld, they are, however, only compatible with an interpretive sociology and a hermeneutic approach to the social world. By now, a number of approaches to social-scientific hermeneutics have been developed, among which several explicitly refer to Schutz’s analysis of the life-world, if in different ways. As promised, I am going now to report on some developments in the German part of Europe. First of all—and this may surprise many of you—there is no such thing as a “Phenomenological Sociology.”2 The reason is that Thomas Luckmann (1983) sharply distinguished the two: either you do phenomenology or you do sociology. Phenomenology is a philosophy, its method proceeds egologically, and it analyzes phenomena of subjective consciousness. Sociology is a 2 In the United States, the concept of phenomenologicala sociology had quite a different career and was prominently forwarded by Psathas (1973, 1989), among others.

578

T. S. Eberle

science, its method proceeds cosmologically and it analyzes phenomena of the social world. The phenomenological lifeworld analysis delivers a protosociology and provides a mathesis universalis, a formal framework for any kind of social scientific endeavor. I don’t know if you have noticed, but The Social Construction of Reality by Berger and Luckmann (1966) is built on the same distinction: The first part about the analyses of Schutz is called “pre-sociological ,” and then follow the two sociological parts, society as an objective reality and society as a subjective reality. As all German-speaking sociologists with a phenomenological, notably Schutzian background were directly or indirectly trained by Luckmann, most of them adhere to this distinction between phenomenology and sociology, and many take Berger and Luckmann (1966) as their theoretical orientation. In the last 15 years, most phenomenologically oriented sociologists who do empirical research, used ethnography as a multiple-method procedure to analyze different sorts of data. The formal structures of the life-world are considered anthropological givens and as universal, and all phenomena which are historically or culturally variable are considered objects not of phenomenological but of sociological research. How rich and well-differentiated the empirical research building on Schutz’s analysis of the life-world presents itself today is shown impressively by two recent volumes, one on The Untranslatability of Cultural Differences, edited by Jochen Dreher and Peter Stegmaier (2007), and another one on Phenomenology and Sociology, edited by Jurgen Raab et al. (2008). On the one hand, they mark the enormous difference between today’s social-scientific research and the role models that Schutz oriented to. On the other hand, they both mirror the variety of empirical approaches to the social world that strive for adequacy. Let me pick three examples. Hubert Knoblauch and Bernt Schnettler developed an approach, which they call Ethnophenomenology. In their research on near-death experiences (Knoblauch and Soeffner 1999) as well as in their research on visions (Knoblauch and Schnettler 2001; Schnettler 2004), both researchers realized that the egological analysis of phenomenologists remains tied to their specific biographic situation: Mundane phenomenology can only describe one’s own experiences. Therefore, phenomenologists cannot make any analytic statements regarding the constitution of transcendental experiences that they themselves did not have. This explains, why the “multiple realities” of Schutz remain incomplete. The term ethnophenomenology points to the observation that philosophical laymen

The Phenomenological Life-World Analysis …

579

are quite able to reflect on their own modes of experiences. The attention of these people to the way of their extraordinary experiences is called Ethnophenomenology by us. (Schnettler 2008: 145—transl. by T.S.E.)

The analogy to ethnomethodology is obvious: Ethnophenomenology intends to examine the structures of actions and experiences of members of society empirically, and describes the research approach as well as its subject matter. But in contrast to ethnomethodology, it is not methodologically produced, observable, ordinary communicative acts, but non-observable, extraordinary subjective experiences of actors that are empirically explored. Schnettler (2004) showed in his study of visionary experiences that, within the interview data, passages with ethnophenomenological descriptions of form differed clearly from the descriptions of the content of the experiences— in fact, the content of the witnessed was often of secondary importance in comparison with the extraordinary mode of the experience. Finally, he was able to elicit a number of recurring features of an Ethnophenomenology of visions of the future. Knoblauch and Schnettler carefully differentiate between the different reference levels of mundane phenomenology and Ethnophenomenology: mundane phenomenology aims at establishing a proto-sociological general theory with a universal relevance by describing general forms of human experience. Ethnophenomenology reconstructs sociologically and empirically the communicatively conveyed descriptions of extraordinary experiences (for example of near-death experiences) by everyday people in a certain historical epoch, and it transforms their generalizations into theoretical notions of a “medium range” (Schnettler 2008, p. 142). Another ethnographic approach that builds closely on Schutz is the socalled lifeworld analytic ethnography of Ronald Hitzler, Anne Honer, and Michaela Pfadenhauer. In the course of their research, data is collected on the one hand by participant observation, interviews in the field, analysis of artifacts and documents, and then gets hermeneutically interpreted, much like in other ethnographic approaches. On the other hand—and this is specific about this approach—the subjective experience of the researcher in the field is used explicitly and reflexively as an “instrument ” of data generation and collection. The researchers thus do not only rely on participant observation for their data collection, but also on—what they call, with a different emphasis: observing participation in a field-specific role, and analyze their experiences phenomenologically. For example, a certain experience of wellbeing during a techno rave is not only researched through observation of and interviews with other participants, but also through a systematic phenomenological analysis of their own personal experiences as co-participants. The basic idea is that

580

T. S. Eberle

the genuine form of an experience is lost, once it is brought into an objectified form, for example by transcribing it and subsequently interpreting it hermeneutically. As a researcher, one should therefore use the immediate access to one’s own subjective experience—for example, of a rave—to conduct a methodologically controlled phenomenological analysis of the experienced— that is, of the experiences and their correlates—through systematic reductions (or bracketings) (Hitzler 2005; Hitzler and Eberle 2004). In contrast to other ethnographic approaches, “the native’s point of view” is not understood indirectly, but is complemented by an “existential view from the inside” (Honer 2004). This way they explore “small social life-worlds” in the sense of Benita Luckmann (1970), as raves and other events, fitness studios, and so on. As you may guess, this is quite a different approach than ethnophenomenology. As a third approach, I should mention social scientific hermeneutics as developed by Hans-Georg Soeffner (2004), a sequential hermeneutical analysis. It is well established in Germany but hardly known in other countries. Interview or interaction sequences are recorded and transcribed, and then analyzed word by word, sentence by sentence, turn by turn. The goal is to detect the hidden meaning structures in the text, the structure of the “case.” In contrast to Conversation Analysis they strive for a reconstruction of meaning, not of formal structures. This kind of analysis has more and more disseminated in Germany. If I do not present you an approach of my own, it is due to the fact that I have not developed one. Like Schutz, I am a methodologist. I am interested in the different research approaches, their epistemological and methodological foundations, and their methodical procedures. And, of course, in the relevance of Alfred Schutz’s life-world analysis for the methodology of the social sciences. Let us instead open up the closing, and return to the problem of adequacy. What does it mean to conduct adequate empirical research? Are the approaches, that I briefly described, particularly evident examples of adequate empirical research? It is obviously not that easy. With his lifeworld analysis, Schutz explicated in detail, if unknowingly, how difficult empirical research actually is: he pointed to the manifold, interlaced interrelations of meaning, to the undistinguishable, diffuse layers of meaning and the implicit horizon of the taken-for-granted, to the limits of what we are able to remember and the approximate character of understanding the other, and finally, to the “paradox of rationality” on the level of everyday actions. If we consider Schutz’s analysis of an act of choosing in everyday life, how can we, for example, empirically grasp the petites perceptions of other actors?

The Phenomenological Life-World Analysis …

581

Even if the hermeneutic approaches try to build upon the current subjective meaning structures of everyday actors as closely as possible—complete adequacy remains an unreachable ideal to which only approximations or variants are possible. Therefore, Schutz’s analysis of the life-world rather conveys a fundamental awareness of the complexity of subjective meaning constitution and social meaning construction, than an empirical instruction for adequate research. No wonder that Schutz loved analyzing literature: The thoughts, reflections and experiences of the proponents in Wilhelm Meister’s apprenticeship and journeyman years (Schutz, forthcoming) or in Don Quixote and the problem of reality (Schutz 1964c) are available in detailed description. The reader gains direct access to the actors’ subjective worlds by their creator’s hand and is able to explore them further analytically. The hermeneutical access to the alter ego in social reality, however, proves to be considerably more difficult. Schutz’s Structures of the Life-World represent a proto-hermeneutics and reveal the basic operations of the constitution and interpretation of meaning as well as the fundamental problems of a hermeneutic approach. They do not, however, provide a practical recipe as to how to unlock the subjective meaning of social action empirically. Accordingly, the different approaches in interpretative social research start with quite diverse premises: Some only trust in interview data which then are interpreted in a sequential analysis, and others only trust in audio-visual recordings of action—and interaction— sequences which are subsequently transcribed meticulously. Some reconstruct life courses from biographical interviews, others regard them as pure narrations and solemnly examine the form of this narration or the structure of the discourse. Some intend to research the content of subjective consciousness, others confine themselves exclusively to communicative practices. Some only trust data from focus groups, others only data of social processes in socalled “natural” situations. Some rely on interpretive interviews, others only on observations of courses of action in real time. Although Schutz’s mundane phenomenology did not bracket the ontological assumptions of the natural attitude (unlike Husserl’s transcendental phenomenology), it does not provide criteria on how to assess such different types of data and how to deal with them in doing research. The various approaches of scientific research differ in terms of their ontological , epistemological and value-theoretical premises, as well as in terms of their theoretical presuppositions. And as the examples mentioned above show, they also differ from certain presuppositions of the actors in the everyday world. How empirical social research can be designed adequately, is thus not only determined by the structures of the life-world but also by additional theoretical premises: The “adequacy” of a study is judged along

582

T. S. Eberle

fairly different criteria by, for example, ethnomethodologists as opposed to the representatives of the so-called Objective Hermeneutics (Reichertz 2004). Let me come to a conclusion: I presented a plea to resume talking about adequacy, but in a more radical fashion than is found in the formulations of the late Alfred Schutz. I argued that Schutz stripped the postulate of adequacy of its empirical requirement and made it thereby compatible with the non-empirical, partly even unrealistic models of his economist friends from the Mises-circle. I further provided some evidence that Schutz actually endorsed a more radical version of his postulate. Using adequacy as a criterion to evaluate social scientific descriptions and explanations, he introduces a different notion into qualitative research than the usual suspects, like objectivity, validity, or reliability. I enjoy looking at facial expressions when I tell colleagues that their measurements are doubtlessly valid and reliable—but completely inadequate. However, I am not a dreamer and do not expect an easy solution for all our methodological and methodical problems. Adequacy is also a kind of elusive ideal. But if it helps to explicitly refer to the phenomenological life-world analysis and to make explicit the additional theoretical and pre-theoretical premises of a sociological approach, we are on the right track. After all, it was this that Alfred Schutz was striving for: a reflective methodological analysis, and a reflexive sociology. I guess this is the goal of all of us.

References Becker, G. S. (1991). A treatise on the family. Cambridge, MA: Harvard University Press. Berger, P. L., & Luckmann, T. (1966). The social construction of reality. A treatise in the sociology of knowledge. Garden City, New York: Doubleday. Dreher, J., & Stegmaier, P. (Eds.). (2007). Zur Unüberwindbarkeit kultureller Differenz. Grundlagen- theoretische Reflexionen [On the insuperability of cultural difference. Basic theoretical reflections]. Bielefeld: Transcript. Eberle, T. S. (1984). Sinnkonstitution in Alltag und Wissenschaft. Der Beitrag der Phänomenologie an die Methodologie der Sozialwissenschaften [Constituting meaning in everyday life and science. The contribution of phenomenology to the methodology of the social sciences]. Bern/Stuttgart: Paul Haupt. Eberle, T. S. (1988). Die deskriptive Analyse der Ökonomie durch Alfred Schutz [The descriptive analysis of the economy by Alfred Schutz]. In E. List, & I. Srubar (Eds.), Alfred Schütz. Neue Beitrage zur Rezeption seines Werkes. In Studien zur Österreichischen Philosophic. Bd. XII [Alfred Schutz. New contributions to the reception of his work. In Studies in Austrian philosophy. Vol. XII] (pp. 69–119). Amsterdam: Rodopi.

The Phenomenological Life-World Analysis …

583

Eberle, T. S. (2000). Lebensweltanalyse und Handlungstheorie. Beiträge zur verstehenden Soziologie [Lifeworld analysis and theory of action: Contributions to understanding sociology]. Konstanz: UVK. Eberle, T. S. (2009). In search for aprioris: Schutz’s life-world analysis and Mises’s praxeology. In H. Nasu, L. Embree, G. Psathas, & I. Srubar (Eds.), Alfred Schutz and his intellectual partners (pp. 459–484). Konstanz: UVK. Eberle, T. S., & Srubar, I. (2010). Einleitung. In A. Schutz. Zur Methodologie der Sozialwissenschaften. Alfred Schutz Werkausgabe (ASW). Vol. IV [On the methodology of the social sciences. Alfred Schutz Edition of Works (ASW). Vol. IV] (pp. 9–44). Friedmann, M. (1953). The methodology of positive economics. In Essays in positive economics (pp. 3–43). Chicago: University of Chicago Press. Garfinkel, H. (1967). Studies in ethnomethodology. Englewood Cliffs. Garfinkel, H. (2002). Ethnomethodology’s program: Working Out Durkheim’s aphorism. Lanham, MD: Rowman & Littlefield. Garfinkel, H. (2006). Seeing sociologically: The routine grounds of social action. Lanham, MD: Rowman & Littlefield. Hitzler, R. (2005). Die Beschreibung der Struktur der Korrelate des Erlebens [The description of the structure of the correlates of experience]. In U. Schimank, & R. Reshoff (Eds.), Was erklart Soziologie? [What does sociology explain?] (pp. 230– 240). Berlin: LIT-Verlag. Hitzler, R., & Eberle, T. S. (2004). Phenomenological lifeworld analysis. In U. Flick, E. V. Kardorff, & I. Steinke (Eds.), (B. Jenner, Trans.), A companion to qualitative research (pp. 57–72). Thousand Oaks, CA: Sage. Honer, A. (2004). Life-world analysis in ethnography. In U. Flick et al. (Eds.), A companion to qualitative research (pp. 133–171). Thousand Oaks, CA: Sage. Husserl, E. ([1936] 1970). The crisis of European sciences and transcendental phenomenology (D. Carr, Trans.). Evanston, IL: Northwestern University Press. James, W. (1907). Principles of psychology. New York: Holt. Knoblauch, H., & Schnettler, B. (2001). Die kulturelle Sinnprovinz der Zukunftsvision und die Ethnophanomenologie [The cultural province of meaning in the vision of the future and ethnophanomenology]. Psychotherapie und Sozialwissenschaft: Zeitschrift fur Qualitative Forschung, 3(3), 182–203. Knoblauch, H., & Soeffner, H. G. (Eds.). (1999). Todesnähe. Interdisziplinäre Beiträge zu einem aussergewöhnlich Phänomen [Near death: Interdisciplinary contributions to a common phenomenon] (pp. 271–292). Konstanz: UVK. Luckmann, B. (1970). The small life-worlds of modern man. Social Research, 37, 580–597. Luckmann, T. (1974). Preface. In A. Schutz, & T. Luckmann. The structures of the life-world. Vol. 2 (pp. xvii–xxvi) (R. Zaner, & H Engelhardt, Trans.). Evanston, IL: Northwestern Press. Luckmann, T. (1983). Philosophy, science, and everyday life. In Life-world and social realities (pp. 3–39). London: Heinemann.

584

T. S. Eberle

Machlup, F. (1954). The problem of verification in economics. Southern Economic Journal , 22(6), 1–21. Machlup, F. (1964). Paul Samuelson on theory and realism. American Economic Review, LIV(September), 733–736. Machlup, F. (1978). Methodology of economics and other social sciences. New York: Academic Press. Mises, L. V. (1940). Nationalökonomie. Theorie des Handelns und Wirtschaftens [National economics: Theory of action and economy]. Geneva: Editions Union Genf. Mises, L. V. (1949). Human action: A treatise on economics. New Haven, CT: Yale University Press. Prendergast, C. (1986). Alfred Schutz and the Austrian school of economics. American Journal of Sociology, 92(1), 1–26. Psathas, G. (1973). Phenomenological sociology. New York: Wiley. Psathas, G. (1989). Phenomenology and sociology: Theory and research. University Press of America. Raab, J., Pfadenhauer, M., Stegmaier, P., Dreher, J., & Schnettler, B. (Eds.). (2008). Phänomenologie und Soziologie. Theoretische Positionen, aktuelle Problemfelder und empirische Umsetzungen [Phenomenology and sociology: Theoretical positions, current problem areas and empirical implementations]. Wiesbaden: VS Verlag fur Sozialwissenschaften. Reichertz, J. (2004). Objective hermeneutics and hermeneutic sociology of knowledge. In U. Flick, E. von Kardorff, & I. Steinke (Eds.), A companion to qualitative research (pp. 290–295). Thousand Oaks, CA: Sage. Schnettler, B. (2004). Zukunftsvisionen: Transzendenzerfahrung und Alltagswelt [Visions of the future: Transcendent experience and the everyday world]. Konstanz: UVK. Schnettler, B. (2008). Soziologie als Erfahrungswissenschaft: Überlegungen zum Verhältnis von Mundanphänomenologie und Ethnophänomenologie [Sociology as an experiential science: Reflections on the relationship between oral mundane phenomenology and ethnophenomenology]. In J. Raab, M. Pfadenhauer, P. Stegmaier, J. Dreher, & B. Schnettler (Eds.), Phänomenologie und soziologie: Theoretische Positionen, aktuelle Problemfelder und empirische Umsetzungen [Phenomenology and sociology: Theoretical positions, current problems and empirical realizations] (pp. 141–149). Wiesbaden: VS Verlag fur Sozialwissenschaften. Schutz, A. (1955a). Letter to Adolph Lowe from October 17th (unpublished correspondence from the Sozialwissenschaftliches Archiv der Universitat Konstanz). Schutz, A. (1955b). Letter to Adolph Lowe from December 7th (unpublished correspondence from the Sozialwissenschaftliches Archiv der Universitat Konstanz). Schutz, A. (1962a). Common-sense and scientific interpretation of human action. In Collected papers Vol. I: The problem of social reality (pp. 3^4–7). The Hague: Martinus Nijhoff.

The Phenomenological Life-World Analysis …

585

Schutz, A. (1962b). Concept and theory formation in the social sciences. In Collected papers Vol. I: The problem of social reality (pp. 48–66). The Hague: Martinus Nijhoff. Schutz, A. (1962c). Choosing among projects of action. In Collected papers Vol. I: The problem of social reality (pp. 67–98). The Hague: Martinus Nijhoff. Schutz, A. (1964a). The social world and the theory of social action. In Collected papers Vol. II: Studies in social theory (pp. 3–19). The Hague: Martinus Nijhoff. Schutz, A. (1964b). The problem of rationality in the social world. In Collected papers Vol. II: Studies in social theory (pp. 64–88). The Hague: Martinus Nijhoff. Schutz, A. (1964c). Don Quixote and the problem of reality. In Collected papers Vol. II: Studies in social theory (pp. 135–158). The Hague: Martinus Nijhoff. Schutz, A. (1967). The phenomenology of the social world . Evanston, IL: Northwestern University Press. Schutz, A. (1972). Choice and the social sciences. In L. E. Embree (Ed.), Lifeworld and consciousness: Essays for Aron Gurwitsch (pp. 565–590). Evanston, IL: Northwestern University Press. Schutz, A. (2007). Sinn und Zeit. Frühe Wiener Arbeiten und Entwürfe. Alfred Schutz Werkausgabe (ASW) Vol. I [Sense and time. Early Viennese works and designs. Alfred Schutz work edition (ASW) Vol. I]. Konstanz: UVK. Schutz, A. (2010). Zur Methodologie der Sozialwissenschaften. Alfred Schutz Werkausgabe (ASW) Vol. IV. [On the methodology of the social sciences. Alfred Schutz work edition (ASW) Vol. IV]. Konstanz: UVK. Schutz, A. (forthcoming). Wilhelm Meisters Lehrjahre (1948-11-1); Zu Wilhelm Meisters Wanderjahren (1948-II-2) [Wilhelm Meister’s Apprentice Years (194811-1); On Wilhelm Meister’s Wandering Years (1948-II-2)]. In Schriften zur Literatur. Alfred Schutz Werkausgabe (ASW) Vol. VIII [Writings on literature. Alfred Schutz work edition (ASW) Vol. VIII]. Konstanz: UVK. Schutz, A., & Luckmann, T. (1973). The structures of the life-world Vol. 1. (R. Zaner, & H. T. Engelhardt, Jr., Trans.). Evanston, IL: Northwestern University Press. Schutz, A., & Luckmann, T. (1989). The structures of the life-world Vol. II (R. Zaner, & D. J. Parent, Trans.). Evanston, IL: Northwestern University Press. Soeffner, H.-G. (2004). Social-scientific hermeneutics. In U. Flick, E. von Kardorff, & I. Steinke (Eds.), A Companion to qualitative research (pp. 95–100). Thousand Oaks, CA: Sage. von Kries, J. (1889). Über die Begriffe der Wahrscheinlichkeit und Möglichkeit und ihre Bedeutung im Strafrecht [On the concepts of probability and possibility and their meaning in criminal law]. Zeitschrift fur die Gesamte Strafrechtswissenschaft, 9, 528–537. Weber, M. ([1922] 1978). Economy and society: An outline of interpretive sociology (G. Roth, & C. Wittich, Eds.). Berkeley: University of California Press.

586

T. S. Eberle

Weber, M. (1949). The methodology of the social sciences (E. Shils, & H. Finch, Trans.). New York, NY: The Free Press. Weber, M. (1994). Sociological writings (W. Heydebrand, Ed.). London: Continuum.

From Bureaucracy to Organized Action: Methodological Individualism and the Study of Organizations Erhard Friedberg

Without exaggeration, it can be said that organizational analysis and methodological individualism, if at all, did not come together easily.1 This is all the more paradoxical as Max Weber (1922, 1964) can be considered to be the spiritual father of organizational analysis. In his methodological reflections, he always proposed and defended methodological individualism as the proper perspective for the analysis of social phenomena. However, his main and lasting impact on organizational sociology and organizational analysis came from his focus on bureaucracy as one of the three ideal types of domination or governance (Herrschaft ). He observed the increasing importance of big formal organizations in the society of his time and interpreted it as the sign that a formalized form of domination (bureaucracy) was gradually

1 In a previous article, (Friedberg, 2019), I have tried to document what seemed to me a more general reluctance in sociology to accept methodological individualism as a legitimate framework, a reluctance which appears to be linked to the fact that this perspective is identified as recognizing human agency as central for the analysis of social action, and as accepting all the complexities for social analysis that come with this recognition.

E. Friedberg (B) Emeritus of Sociology, Paris, France e-mail: [email protected] Sciences Po, Paris, France

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 N. Bulle and F. Di Iorio (eds.), The Palgrave Handbook of Methodological Individualism, https://doi.org/10.1007/978-3-031-41512-8_26

587

588

E. Friedberg

replacing the two other ideal types and was becoming the hallmark of modern societies. Retrospectively, one can only pay homage to the farsightedness of this diagnosis, given the degree of bureaucratization of present-day societies worldwide. In the narrower field of organizational analysis, however, it could be argued that the notion of bureaucracy and its interpretation greatly contributed to “harden” the traditional idea of organizations and the three premises it was built on. The first of these premises was the instrumental nature of organization: they were seen as instruments for the realization of predetermined goals on the formulation of which they had no influence. They were understood as nothing more than passive and transparent transmission belts of the missions in service of which they have been created. The rationality and legitimacy of their goals could be questioned from the outside, but they were considered to be non-problematic for their members: indeed, they were seen to provide the yardstick for defining what was to be considered “normal” behavior in organizations. The second premise was that organizations were cohesive, unified, and homogenous entities. As such, they were capable of effectively structuring the behavior of their members and of obtaining their respect of the formal rules and regulations inscribed in their formal structure. Conflicts had no reason to exist, indeed were considered to be abnormal, not to say pathological, precisely because as members of the organization, all shared the same preoccupations and values. And third, organizations were considered to have clear and unambiguous boundaries which unequivocally separated an inside and an outside and were capable of shielding the inside from the repercussions of events taking place outside. From today’s vantage point, this “classical” conception of an organization is so clearly unrealistic that it seems impossible that anybody could have seriously believed it, even though, in comparison to today, it may well be that it actually came close to describing reality in the late years of the nineteenth and the early years of the twentieth century. Since then, however, since at least the early years of the twentieth century, the steady flow of organizational studies has successfully taken aim at one or the other of these three premises, replacing this oversimplistic model with a much more complex conceptualization. Covering the whole range of organizations, from workshops, factories, and administrations to voluntary associations like political parties or labor unions and what now would be called NGOs, these studies were by nature inter-or transdisciplinary: they were located at the cross-section of a great number of disciplines, with sociology far from being the most prominent: from psychology, social psychology and the sociology of work reflecting on the mutual and complex relations between the functioning of organizations

From Bureaucracy to Organized Action …

589

and the psychological development of individuals and their processes of socialization, to economics, the “sciences“ of management and the studies of decision-making highlighting the gap between the normative theories of decision making and the actual observed behavior within organizations, private and public.2 In a previous essay (Friedberg, 1993), I have attempted to document the thrust of what these studies contributed to our understanding of the structures and functioning of formal organizations.3 Summing up their results, it can be said that they document three main “deficits” or gaps in regard to the canonical conception of an organization. First, formal organizations suffer from a “deficit” of legitimacy: their overall goals and missions are incapable of structuring the day-to-day behavior of their members, as they are constantly relativized by the development of partial, “local” and sometimes openly factional interests, goals, and values. The normative integration of their members is never complete, even under the conditions of highly paternalistic (highly “inclusive”) management styles.4 No organization, not even a firm operating on a highly competitive market, is a unified normative space: it is always characterized by the coexistence of different “worlds” representing differentiated value systems and logics of action. Second, organizations suffer from a “deficit” of interdependence. Empirically, this is manifest in the frequency of weak ties within organizations or more generally in the heterogeneity of ties between the various members and spaces (contexts) within an organization: links and ties between them have neither the same intensity nor the same frequency, nor, even more importantly, the same relevance and importance. Action fields within organizations are heterogenous and characterized by all sorts of discontinuities, because in their day-to-day dealings and behavior, participants tend to limit as far as possible and feasible, their dependence on other members (parts) of the organization and will attempt to construct their cooperation around “rules of the game” specific to their contexts.

2 Special mention should be made here of one strand of organizational sociology which is interactionist in its approach. Somewhat marginal in the bulk of organization studies, no reference is made to it in what will be presented in the following pages. Focused on the study of the dysfunctions of bureaucracy, this strand was initiated and developed by Merton and his disciples, in particular Selznick (1949, 1957) and Gouldner (1954, 1955) in the United States. Outside the USA, it was to be particularly influential in France where it can certainly be traced to the beginning of what would later become the French school of organizational sociology. In his seminal work Le Phénomène Bureaucratique, M. Crozier (1963) has acknowledged and documented its influence on his interpretation of his studies of French bureaucracies. Cf The Multimedia Encyclopedia of Organization Theory: from Taylor to today (2011), www.screeningorganization.com . 3 Cf. Friedberg (1993, pp. 37–112). 4 In present day language, such management styles would probably be qualified as “inclusive.”

590

E. Friedberg

Contrary to a persistent myth, organizations are not homogenous, unified action contexts: their degree of fragmentation will of course differ. However, such fragmentation is an ever-present ingredient in organizational reality and a perpetual threat to their unity, all the more so as in their day-to-day interactions with groups of actors located outside the organization, some at least will be drawn to identify with the latter’s interests and import them into their organization, thereby increasing its internal fragmentation.5 In short, organizational unity is never a given. It has to be created and maintained through shrewd management. And, thirdly, organizations suffer from a “deficit of rationality”: formal organizations are not the guarantees for rational action they were thought to be. The bounded rationality characteristic of all human activities infuses everything in organizations: the behavior of organization members as well as the structures, tools, and instruments which are supposed to channel and steer the latter, in order to articulate their behavior with the goals and the mission of the organization as a collective entity. Everything in and about an organization is subject to the same bounds of rationality: formal organizations cannot pretend to represent a form of rationality that would be superior to, or free from, the affective and cognitive limitations that characterize human action in general. They are formalized structures of bounded rationality. Special mention should be made here of a strand of studies linked to the name of James March (1978, 1988) which highlight the theme of ambiguity in organizations, challenging the idea that organizations are different in this respect, because their rationalization had precisely been able to reduce, if not altogether eliminate, ambiguity. In a series of studies, he and his colleagues (Cohen et al., 1972; March 1978/1988; March and Olsen 1989) document that just like the world around us, organizations were characterized by all sorts of ambiguities, the impact of which on the conditions of choice in organizations have been grossly underestimated. Ambiguity in organizations is important because its existence blurs the causality of events in organizations, makes it difficult to unequivocally attribute results to events and/or to individuals, or to arrive at a shared distinction between goals and means. In his article of 1978, March categorizes four dimensions of ambiguity. First, the ambiguity of preferences which he had been highlighting for some time already and referring to the fact, well documented by cognitive dissonance theory (Festinger, 1957), that preferences are not independent of choice, but rather are as much produced by choice as producing them. Second, the ambiguity of relevance which challenges the apparently clear 5

This is the one of the central themes of Ph. Selznick’s classic TVA and the grassroots (Selznick 1949).

From Bureaucracy to Organized Action …

591

distinction between means and ends, between problems and solutions (with solutions needing problems which in turn call for solutions and vice-versa, and with actors in organizations being able to create problems in order to generate an interest in what they can propose as solutions).6 Third, the ambiguity of history and experience, referring to the difficulty of unequivocal learning cycles, which depend on a clear (i.e., unambiguous) understanding of past events and of their relevance for the present or future. And last, March mentions the ambiguity individuals in organizations encounter when they try (as individuals, but also collectively) to make sense and extract the meaning of present events. Meaning, he states, is never unambiguous, but basically subjective, just as the participants’ rationalities. It seems obvious that the overall effect of ambiguity in organizations is to further dilute their normative integration and to accentuate the disconnect between their units, if not their individual members. Bounded legitimacy, bounded interdependence, and bounded rationality: these three bounds, the impacts of which obviously reinforce each other, make complete the deconstruction of the social object “formal organization” which loses all the attributes justifying its existence as a distinct social object. Social life and social action within formal organizations (whether they be bureaucratic or not) are in no way different by nature from what they are in society at large. The apparent simplicity of formal organizations, their seeming transparency, and what could be called their “finalization,” i.e., their self-proclaimed subsumption to formalized goals, are illusions. Having to live with (cope with) these three bounds or deficits, contexts of action characterized by formal organization are not fundamentally different from other, apparently “less organized” contexts of action, where the interactions and dealings between participants seem to be less structured and formally codified. The difference is not one of nature, but of degree. Surprisingly, this conceptual evolution, which from a theoretical perspective might nowadays be considered a trivial point, has not really penetrated the world of action. The governing elites in the private as well as in the public sector continue in their majority to act as though organizations were simple tools to be used to accomplish collective goals, as though the only thing that really mattered, were the invention and definition of more and more complex goals which could then be accomplished by organizations which have been designed to do just that and can be considered to be totally submissive to their design. While paying of course lip service to the complexities of organizations, they overwhelmingly continue to adhere to a mental model

6

This is the central theme of the garbage can model of organizational choice (Cohen et al. 1972).

592

E. Friedberg

in which organizations are simple tools for the accomplishment of collective goals, notwithstanding the overpowering empirical evidence that reality is more complicated, that organizations seem to develop their own dynamic and that they produce results not only different from, but more frequently contrary to the intentions that presided over their creation or design. On the conceptual level, the conclusion was and had to be different. The banalization and complexification of formal organizational as an actioncontext different in degree, but not in nature from other action contexts in the wider society, forced to recognize the basic fact that the collective entity organization does not eliminate the individuals within it. Because they have decided to participate in an organization, does not mean that ipso facto they abdicate their individuality, their identities, and personal interests, or lose their capacity to choose their behavior in order, if possible, to rearrange their situation in a way that fits their interests (whatever they may consider them to be). Therefore, organizational analysis has to start with the recognition of the inescapable limits of an organization’s control of the behavior of its members whose behaviors and logics of action are never completely dictated or predetermined by organizational dispositions and prescriptions. In other words, it has to respond to the question of how organizations come into existence and maintain a minimum of regularity in their functioning with building blocks as volatile as the behavior of individuals (actors) who at any time retain their capacity to stop cooperating by, in the very least, withholding the contribution expected from them by their partners, while remaining physically present. The irreducible unpredictability of human behavior within formal organizations as well as in society at large is the starting point for any serious attempt in understanding organizational reality. Instead of confronting this empirical reality, however, there has been a long tradition in organization theory of denying this basic fact and trying to find a way of going around it by developing theories of motivation that would make human behavior predictable. F. Taylor’s theorization of homo economicus can be seen as the first expression of such an attempt: motivated solely by the perspective of material gain, human behavior, such was the reasoning, becomes predictable and can be exploited by appropriate pay schemes. However, experience has quickly forced to replace or rather, complement this rudimentary motivational theory with more complex ones, from “social man” and his “need to belong” of the human relations perspective (Roethlisberger et al. 1939), to “self-actualizing man” in search of “self-realization” (Argyris 1964). The most elaborate

From Bureaucracy to Organized Action …

593

and for this reason most widely influential of these attempts was probably A. Maslow’s motivational pyramid (Maslow 1954).7 Maslow proposed to consider human needs to be rank ordered in a pyramid, from the most basic needs at the bottom (food and clothing) to ever more complex and elaborate needs, with self-actualization being at the top of the pyramid. And he theorized that in step with the social and economic development of societies, individual needs would rise along this pyramid accordingly. What made this theory attractive for organizational analysis (and HR-management, it should be added), is the additional premise that only the needs not yet satisfied were relevant to predict and explain human behavior. One can easily see the simplistic and quasi-mechanical reasoning that can be developed on this basis: depending on where people (and societies) are located on the developmental pyramid, only one type of motivation is predominant, making individual behavior predictable again. From the 1970s on, organizational psychology had to admit that analyses based on Maslow’s perspective did not produce any tangible empirical results. In recognition of that, it developed the notion of “complex man,” which meant that it had to accept the irreducible multiplicity of human motivation in real situations and recognize that individuals retain a minimum of distance and autonomy in regard to their “needs” and psychological motivations. “Complex man” is the description of individuals who are active in their situations and therefore basically unpredictable. In a way, it can be understood as the admission that the attempts to psychologically model human behavior have in the end turned out to be neither very meaningful nor very helpful in practice.8 Contrary to what it pretends, behavioral economics and its cognitive perspective encounters, the same limitations.9 Apparently, less mechanistic and despite its libertarian dimensions, its mode of reasoning comes close to the perspective of organizational psychology of the 1960s of last century. The cognitive bias producing the errors of judgment that cognitive psychology

7 As one might remember, Maslow proposed to imagine human needs to be rank ordered in a pyramid. At the bottom he put elementary needs (like food and clothing) that dominate all the others as long as they are not satisfied. Next in line would be the needs for security, followed by social needs (the need to belong, the need to be recognized and so forth) and then the needs linked to personal development (dignity, self-realization). It is obvious that historically, this perspective expresses an evolutionist perspective theorizing in psychological terms the rise of the affluent society in the 60s and 70s of last century. 8 Given this evidence, it is amazing and somewhat disquieting to observe the persistent influence of Maslow’s motivational theory in HR and general management circles as well as in their training centers. 9 For a detailed critique of the many problems raised by behavioral economics, cf. Bergeron (2018) and also Pilmis et Castel (2020).

594

E. Friedberg

has highlighted, simply takes the place here of the “needs” prior psychological reasoning was based on. On the basis of an exploration of the bias which can introduce errors in the individuals’ reasoning and have them choose decisions that might be “irrational,” i.e. decisions contrary to what would be their “authentic” interests, behavioral economics proposes à device, the “nudges,” supposed to bring individuals back to rationality and thus to make them if not predictable, but at least easier to steer in the right direction. This perspective has been very positively received and put into practice, more particularly in public management. It has brought success, fame, and the 2017 Nobel Prize to its inventor, Richard Thaler. However, it is based on two premises which can be easily challenged. On the one hand, it seems to assume that individuals, independently of their situation, are always and under any circumstance, the slaves of their cognitive bias, just like mainstream organizational psychology considered (and still seems to consider today) individuals to be the slaves of their psychological needs. On the other hand, this perspective calls for (and implicitly builds on) a normative model which assumes that it is possible to define and to know the “authentic” interest of individuals and thus the “good” rationality to be applied to their decisions. This model can then provide the basis for “enlightened” authorities to justify the use of the appropriate “nudges” to correct the predictable “irrationality” of individual choices and steer them toward their “authentic” interests. The underlying paternalism of such a mode of reasoning goes without saying.10 All conceptual attempts to make human behavior in organizations predictable have failed. The unpredictability of behavior of humans, wherever they are and whatever their activities, is an inescapable empirical reality. Neither the notion of economic rationality and strictly utilitarian calculation, nor the recourse to “psychological needs,” nor the invocation of cultural influences or group identities (like race, gender, sexual orientation), nor the cognitive perspective of behavioral economics or of cognitive psychology offers a way around this basic fact. The reason for this is that despite all their differences, these attempts share two shortcomings. On the one hand, they ascribe the unpredictable behavior to isolated, unrelated individuals11 : they implicitly seem to consider that unpredictability of behavior is an individual 10

Richard Thaler, Nobel prize 2017, calls it “libertarian paternalism” to justify the use of “nudges” in public policy. Even qualified as libertarian, paternalism is still paternalism: it is based on the idea that people claiming this perspective for them, have the right, if not the “moral duty” to correct other persons’ decisions and actions under the pretext that their decisions are in fact irrational, i.e., deviate from what the paternalistic, enlightened person would claim to be a rational decision. Such paternalism is libertarian in words only. It could be added that the advances in the neurosciences have added a new dimension to this perspective, but do not change the problematic side of it. 11 Or individuals considered to be completely absorbed by (without any distance to) their identities which then become easier to parameter.

From Bureaucracy to Organized Action …

595

problem. On the other hand, and more importantly, they seem to cling to the idea, that, if studied long enough and appropriately framed (organized), individual behavior can be made predictable in the end. Cooperation can be established and guaranteed once and for all and can therefore be neglected as a problem in the design of collective projects and outcomes. The rationale behind such reasoning is based on what could be called a functionalist fallacy, which can be described as follows: As one of the functions of organizations is to create cooperation around complex tasks that create interdependence, functioning organizations are the living proof that they have indeed succeeded in fulfilling this function. As a consequence—so the reasoning—it can be concluded that, within their bounds a least, cooperation ceases to be a problem: with the creation of an organization, the margin of freedom of individual members, their capacity to vary in their commitments, even if they continue to participate, disappears behind the organizational roles formal organization creates for them. If this were verified, organizations would indeed be instruments faithfully implementing the complex designs in the service of which they have been created.12 However, the banalization of formal organization and the accumulation of what we know about their actual way of functioning, makes this simplification untenable. Even if it is of course constrained, within a formal organization, the freedom to choose one’s behavior and to adapt in unforeseen ways to changing circumstances, never disappears behind organizational roles, as clearly defined, and codified as they might be. On the level of each separate individual, this margin of freedom may seem negligible. However, in collective action, it is interrelated to the margins of freedom of all the other participants. This interrelation and the resulting interactions amplify its scope and produce the radical and empirically documented unpredictability of collective efforts, this grain of sand that invariably derails the trajectory of all projects which create interdependence and require lasting cooperation. In other words, the behavioral unpredictability of each individual member creates a collective problem because it translates into the unpredictability of organizational outcomes. This is so because the actual use of any participant’s margin of freedom is not a question depending only on their individual motivation: it is not merely an individual problem. No psychological perspective can convincingly account for it. It is intractable to any attempt of modeling 12

This is why in Hirschmann’s triangle of exit, voice and loyalty (Hirschmann, 1970), the third term is the most crucial, but also the most problematic and ambiguous. Loyalty can indeed mean loyally sticking to one’s commitment, continuing to cooperate as before, despite all disappointments and problems. However, it can just as well mean that one continues to act as though one were loyal in the first meaning of the world, when in fact one has ceased to be mentally present in one’s job, even though one continues to be physically present.

596

E. Friedberg

based on individual motivations defined out of context and, more generally, intractable to any psychological (and or cultural) definition of human nature arrived at in the abstract and independent of context. Because of its interaction with the margins of freedom of other participants, it becomes “socialized” and contextualized and is beyond the control of any of them, even the most powerful. Seen from this perspective, it has become a social fact. As such, it is inseparable from the very interaction dynamics which develop in concrete action contexts with their always specific constraints. Action within organizations is always the product of contextualized dynamics of interaction, the outcome of which is radically unpredictable as well as beyond the control of any of the participants. Organizational outcomes (the mode of functioning of any organization) are the result of how these dynamics are aggregated, and that aggregation is in itself an open and undetermined social process the result of which in no way guarantees that outcomes will match the intentions of any of the participants, even the most powerful. As we know at least since Merton (1936), non-intentional or even “perverse” results (i.e. at least partially contradictory with initial intentions), are omnipresent in purposive, collective action, in the public as well as in the private sector, and formal organizations are of course no exception. R. Boudon (1973, 1977, 1984) has demonstrated their existence in a variety of fields, more specifically in his controversy with P. Bourdieu, in relation to policies aimed at rectifying educational inequalities in France. In the narrower field of organizations, Ph. Selznick expressed the same idea this way: “I helped form this institution, but whether it is really fulfilling the ideas we had in the first place yes, partly. If we look at the structure of this situation here, the idea was to create an interdisciplinary program within the Law School, and we did. However, there are dynamics to interdisciplinary programs. They are fragmenting dynamics, and they are very hard to control, and it takes very strong leadership and insistence on conceptual coherence and all of those things, to really have it hang together and it is very hard to do.” 13 At this juncture, organizational analysis encounters methodological individualism. For, this is the theoretical perspective that takes seriously the radical unpredictability of human action, rooted in the individuals’ capacity of behavioral choice, the rationality of which is always at the same time subjective (according to one’s dispositions linked to one’s socialization and past experiences) and contextual, i.e., dependent on, or reflective of, the specific constraints and opportunities of one’s position and interactions in

13

Philip Selznick, interview with the author Cf. Friedberg (2011).

From Bureaucracy to Organized Action …

597

a given context of action.14 Correspondingly, it is also the theoretical perspective, which thinks of collective outcomes as composition effects of interaction dynamics generated by the cooperation between relatively autonomous actors positioned in specific contexts of action. The deconstruction or banalization of formal organization as just one possible form of collective action implies that it is built on the temporary and inherently fragile form of cooperation that exists between actors who have accepted (for the time being, for a certain period of time and under certain conditions) to cooperate with one another while retaining their relative autonomy, i.e., their capacity to modulate (change, interrupt) that cooperation at any time. Like any form of collective action, formal organization can only function and thrive if it has solved this “problem of cooperation.”15 Cooperation is an attempt to manage existing interdependencies—this is the resource side of cooperation. At the same time, it creates the necessity of exchange and the social interactions which make it possible: this is the constraint side of cooperation. The two faces of cooperation are closely linked to one another, in the same way that decisions of one actor in any given context are always at the same time the response (the reaction) to decisions of some other participant, and of course the other way around. In other words, the decisions of any of the participants in any given context cannot be understood outside the reality of interdependence and limited reciprocity characteristic of this context. Such a context characterized by a causality different from the one commonly used, a causality, which could be called complex, circular,16 or “systemic.” Here, each participant’s behavior is simultaneously cause and effect of the behavior of one of the partners whose behavior in turn will be cause and effect of the behavior of the latter. Instead of explaining one fact by its cause (or by a multiplicity of causes), circular causality consists in taking cause and effect as the interdependent condition one for the other, as two facts causing and producing each other. The two represent the effect

14

I have reformulated this distinction based on Boudon (1986). Fligstein et McAdam (2012) have tried to deny this inherent fragility of cooperation with a strange conceptual construction called the “existential function” of the social (the collective) which makes free riding an unattractive alternative because in their words affiliation with groups and other collectives is a highly desired end in and of itself. The argument may seem interesting, but in closer examination turns out not to be very convincing. Indeed, one does not understand why this “existential function of the social” would eliminate an actor’s possibility to modulate at wish their cooperation and to choose to exit, if feasible and possible. Society is the place of multiple collectives, and one can always choose to satisfy “the existential function of the social” in some other collective. Contrary to what Fligstein and McAdam pretend, the “existential function of the social” does not eliminate the unpredictability of individuals engaged in cooperation. 16 The expression is taken from Mary Parker Follett (1924). 15

598

E. Friedberg

of a system (a game) the nature and properties of which have to be reconstructed in order to understand the logic of the behaviors in question. In his book Micromotives and Macrobehavior, Thomas Schelling (1978)17 has presented a simplified model which illustrates this circular causality and its most important implications. The model consists of chips meant to impersonate real people. The chips are of two different colors: blue and red, or black and white, or yellow and blue, any color, as long as there are only two. Chips are then located in a square on a checkerboard, one blue chip next to one yellow chip next to one blue chip again, and so on until the whole checkerboard is filled with chips, except the four corners of the checkerboard. Each chip thus has a location and an environment composed of the chips adjacent to its location. These environments are all similarly mixed. Schelling then proceeds to disturb this uniform distribution in two steps: first, by removing at random, a certain number of chips (32), and then in a second move, by replacing, again at random, a smaller number of these chips (12) back on the checkerboard. The end result is a checkerboard, where there are squares left empty, and where the number of blue and yellow chips not only will be unequal, but also unevenly distributed on the checkerboard. Chips are then hypothetically endowed with two capacities: first, with a capacity to choose their location on the checkerboard, if they dislike the environment in which they find themselves. And second, with preferences: chips are supposed to prefer environments in which they are not in a minority, i.e., in which their own color is at least slightly in the majority.18 Schelling then moves the chips remaining on the checkerboard until each is supposed to be “happy” with the composition of its environment. When all is said and done, the chips end up on a checkerboard which overall is clearly more segregated than it would have to be, were the chips’ spatial distribution faithfully representing their individual preferences. Several outcomes of this operation are of course possible, depending on where it is started and how the chips are moved.

17

Schelling (1978), Chap. 4, pp. 135-167. For the exercise, Schelling provides a quantitative definition of what it means for any given chip “not to be in the minority.” For instance, if a chip has only one neighbor, it would have to be of its color. If a chip has two, three of four neighbors, two of them at least would have to be of its color; And if a chip has five, six, seven, or eight neighbors, three of them would have to be of the same color. Note that this defines not a preference for segregated environments, but for environments where diversity is limited, but not excluded.

18

From Bureaucracy to Organized Action …

599

This little model has provoked a good amount of critical discussion.19 One very common error in those comments was to have interpreted this model as an attempt to explain or may be even to euphemize the segregationist tendencies of the American society. Be that as it may, these comments seem to me to miss the crucial point illustrated by Schelling’s exercise, which is to highlight how collective outcomes can become disconnected from the initial intentions and choices of participants. The reason why Schelling’s little exercise merits to be mentioned here is because it demonstrates the existence of the cumulative nature of the aggregation of participants’ behavioral choices producing what can properly be called a composition effect of individual choices. This effect is of course the product of each participant’s decision to stay in their neighborhood or to leave it. At the same time, however, each participant is positioned in a context in which the abovementioned circular causality is fully operative. The choice of each participant is influenced by the decisions of other participants as well as in turn influences their decisions. By deciding to leave a neighborhood and to choose another one, each participant transforms the diversity of the respective neighborhoods, and in turn, transforms the parameters which enter into the decisions of the participants in either context (the one that has been left as well as the new one that has been chosen) whether to stay or to leave. It thus becomes difficult to distinguish cause and effect. The overall outcome (the composition effect of all “residential choices” on the checkerboard) is produced by all these choices, but it is not caused by anyone in particular. It is the product of their interaction, and as such it is in large part disconnected from the participants’ intentions and preferences. At the same time, however, it amplifies the implications of the latter, because their overall effect is to create a residential structure on the checkerboard which is much less integrated than the one that would correspond to the preferences which Schelling has initially defined in the protocol for his exercise. This effect is not only independent from the participants’ intentions and preferences, but clearly goes beyond what they would have chosen had they been able to do so. Attempting to avoid being in a clear minority, participants end up residing in neighborhoods, which are far more homogenous than what they would initially have preferred.

19

In particular, it has been argued that contrary to Schelling’s conclusion, this model was a witness of the American society’s latent racism, by pointing out that more moderate preferences than those used by Schelling would lead to a less segregated outcome. For an example, cf. Forsé and Parodi (2010). This argument, however, is misleading: a slight preference not to be in a minority in regard to one’s ethnic identity is not proof of latent racism, unless one considers of course that it would be normal for individuals not to have any preference on this matter.

600

E. Friedberg

Before going any further, let us highlight the premises on which Schelling’s demonstration is built. Three of them are of particular relevance. First, the chips/individuals don’t communicate in any way. They do not speak to each other, nor do they attempt to coordinate in any way. They only have the option to stay or to leave. If they dislike their location, their only means of protest is to leave. They cannot choose to stay and use Hirschmann’s (1970) “voice” to protest and/or engage in an attempt to change their situation. Exit is their only option. Second, the chips/individuals in question are insensitive to the monetary and affective consequences of their choices. The degree of diversity of their neighborhood is the only dimension they take into account when they have to decide whether to leave or to stay. Whatever the moving costs, whatever the affective costs of leaving one’s neighborhood: when its diversity goes beyond a certain threshold, leaving is the only option. And lastly, socio-economic inequalities are excluded from the model: the chips/individuals have identical capacities for staying or leaving, and they have the same way of perceiving and judging the diversity of their neighborhood. Highlighting these premises clearly shows that this model is really built around rudimentary individuals, using simplistic binary rationality. It is obviously a very simplified, not to say unrealistic model of social life. It creates a world without structure and without constraints other than the pursuit of one’s satisfaction in reaction to the other participants’ moves. However, this simplism has a virtue which is of great interest in the perspective of organizational analysis. On the one hand, it highlights how the composition (aggregation) of the dynamics emerging in all interactive contexts produces effects of its own. The sheer emergence of these effects is not only probable. It can be predicted with a good chance of success. However, neither the precise nature nor the direction of these “effects” can be anticipated: they remain an open question and can only be understood after the fact. On the other hand, it demonstrates that these effects, also called “counterintuitive” effects, not only are independent and diverge in unforeseen (and unpredictable) ways from the participants’ intentions, but also are very difficult to steer in advance. One can only correct them ex post, and even then only within certain limits, as these attempts will themselves be subject to counterintuitive consequences. To clarify this last point, imagine for one moment that some enlightened actors, having understood the “perverse” results of these interactive dynamics, decide to intervene in order to steer and, as far as possible, correct the effects

From Bureaucracy to Organized Action …

601

of their aggregation (composition). What would be their possible means? They could choose to use behavioral and economic incentives, they could attempt to define a set of rules in regard to the exchange of residential locations, they could even go so far as to establish a platform facilitating such exchanges, but also defining a set of constraints for its use, or they could use other organizational devices. In short, they could and would try to “organize” in some way the free-flowing interactions underpinning the emergence of the undesired results. There is a fair chance that such devices will have some impact on the behavior of the participants in the relevant action context. However, there is no way to predict precisely the way in which this impact will materialize in the end. For their impact will have to go through the filter of the ongoing interactive dynamics which continue to locally structure the relations of participants with one another. In other words, they will only add new constraints for these dynamics, but they will not and cannot do away with them. The result produced by the interactive dynamics within this new “organizational context” will certainly be different from those produced by them in the absence of such devices. But it will still be the result of the composition of interaction dynamics developing within this reorganized action context and could therefore be called an “organizational” effect. And like any other “organizational effect,” the overall result (for example the mode of functioning of a formal organization created for this end), can only be observed ex post, as it is the product of a radically open and thus unpredictable process. Understanding organizations as one form of collective action has two main implications. On the one hand, it implies giving priority to the observation of action in organizations and to the analysis of the concrete action contexts and the interactive dynamics which develop within them. On the other hand, and as a consequence, it forces the observer to relativize the impact of the formal frame-works organizations are based on. Organizational instruments—formal rules and prescriptions, hierarchies of authority, management tools, etc.— all the tools for organizing that have been invented to “better organize,” do not exist outside of, or independently from, the interactive dynamics without which no organization would function and produce any outcome. More importantly, they never eliminate these dynamics, which continue to exist and structure the behavior in the concrete action contexts in organizations. They are nothing more than never fully successful attempts to steer and control the dynamics emerging from the interplay between the participants’ interdependent actions and the effects produced by their composition (aggregation). As a consequence, the analysis of their impact cannot be separated from an understanding of the latter. And the emergence of unanticipated

602

E. Friedberg

and frequently undesired outcomes cannot be excluded. On the contrary, it should be taken for granted and interventions should be tailored to cope with them as far as possible. As Merton has shown, unanticipated consequences are the “normal” result of purposive action and should therefore always be considered an object for careful attention. And as Crozier (1963) has shown in his analysis of French bureaucratic organizations, these unforeseen consequences frequently take on the form of vicious circles, i.e., can be described as self-reinforcing dynamics. Summing up the argument developed so far, leads to conceiving of any organization as a more or less tightly connected juxtaposition of scenes or contexts of interaction. These contexts are similar to the checkerboards in Schelling’s model, but of course also much more complex. One has to imagine them as spaces where participants concretely deal with each other and have to find a way to construct some enduring modalities for their necessary cooperation (their “rules of the game”) despite their possibly diverging interests. These scenes are located in an overall structure formally relating one to the other into a collective entity endowed with a mission or overall goals, and they are subject to a set of formal rules and prescriptions making up the formal structure of the organization. Unlike the chips in Schelling’s model, the individuals acting within these contexts are unequal humans endowed with differentiated resources and capacities, linked to each other not only by their task interdependence, but also by a hierarchical structure with its own constraints. And unlike the chips in Schelling’s model, these individuals all have different intentions and interests in the service of which they reason, communicate, exchange, and strategize, while being simultaneously aware of the constraint imposed by the interdependence in their tasks, at least as long as they decide to “stay in the game.” Each of these contexts has its own power structure and each is characterized by a specific imbrication of the interactive dynamics which emerge from the day-to-day dealings between the participants and the formal organizational prescriptions and which result in a set of “rules of the game” stabilizing the necessary cooperation between participants. The structure of these contexts is obviously much more complex than what characterized the contexts of action in Schelling’s model. But similar interactive dynamics can be observed, which function with a circular logic quite similar to what can be observed in Schelling’s model, and are prone to produce similar composition effects, i.e., a mode of functioning specific to this particular context. And similar to what can be observed in Schelling’s model, this mode of functioning will invariably diverge from the intentions of any of the participants,

From Bureaucracy to Organized Action …

603

even the most powerful. Unintended consequences are at the heart of organizational processes, just as they are at the heart of all concerted efforts, i.e., what Merton (1936) called purposive action.20 In other words, each context is structured by a specific transactional space, which develops its own rhythm and logic, depending on, in particular, the way participants have integrated, used, or misused the organizational prescriptions in their attempt to construct and stabilize their day-to-day cooperation. Each context thus develops its own local and contextualized “field of rationality,” it will have its own culture and “traditions” It does not disappear in the organization; it will tend to have an existence of its own. The important implication being that the same organizational instrument such as for instance a performance indicator, a formal rule, a measurement tools, a modification in the formal structure, etc., will not be understood and implemented the same way in all the contexts an organization is composed of. The impact of such tools will always remain and must be considered an open question. It remains essentially unpredictable and will have to be observed on a case-to-case basis. It is not possible to predict the differences between the organizational contexts which will eventually materialize. Intra-organizational differences and variation, therefore, can be expected to be high rather than low.21 However, documenting their frequent inconsistencies and disconnected way of functioning of formal organizations, demonstrating that their modes of functioning are structured along diversified logics which are never merely instrumental to the accomplishment of the official goals, cannot be an end in itself. Such observations should not discourage organizational analysis to look for the underlying logic behind that empirical reality. This is also the methodological preoccupation of Karl Weick known for his theorizing on organizations as loosely coupled systems (Weick 1969, 1976). Revisiting his own theorizing (Orton and Weick 1990), he argues that referring to organizations as loosely coupled systems, merely amounts to replacing one simplistic model of organization (the cohesive organization) by another simplistic model (the incoherent and unpredictable organization). Adopting such a view, he argues, has to be the beginning of a process of questioning,

20

In a forthcoming essay (Friedberg Erhard (forthcoming), Cette irréductible part de liberté [This irreducible part of freedom] I will provide a more detailed description and analysis of the implications of such a conceptualization for the management (the governance) of organizational and societal change. 21 In a book attempting to synthesize his life-long experience and deep empirical knowledge of development projects in Africa, Jean Pierre Oliver de Sardan (2021) has highlighted the crucial difference made by the social structure of “local contexts” for the implementation of abstract models of development which he with good reason has labeled “modèles voyageurs [traveling models].”

604

E. Friedberg

as it calls for differentiating the perception, for deepening the descriptions on where, in what areas, on what matters the concept of loose coupling really applies, and what therefore its real significance might be. The genuine contribution of this perspective is thus to propose a heuristic which encourages the observer to become attentive to the discontinuities and disconnects behind apparent smooth chains of communication or hierarchies, to become sensitive to the co-existence, within organizations, of tight and loose coupling, and of the always difficult articulation between interdependence and autonomy. By following this perspective, Weick argues, it becomes apparent that organizational structures are not inert properties, but the product of permanent elaboration which can only be understood by rooting them within the interactive dynamics of which they are simultaneously the context and the object. It needs contextual analysis to situate and, if possible, decode what off-hand would only look like anomalies. ********* This stylized evocation of the real processes, which in their (varying) complexity produce the functioning of any organization, might provoke two seemingly contradictory reactions. On the one hand, an amazement about the amount of control retained despite all these odds by big rationalized organizations such as private corporations or huge public administrations, documented by their capacity of (more or less) successfully steering their dayto-day operations. On the other hand, a deep awareness of the irreducible and unsurpassable limits imposed by this very complexity on unbridled managerial constructivism and its illusions of control and direction, as well as the recognition of the radical and obvious inadequacy of all the mechanical metaphors which too frequently underpin the use of the term organization. Organizations are neither mere transmission belts of directives and goals defined from above, nor a set of tightly connected cogs functioning in unison. The machine metaphor for an organization leads to systematically overestimating the importance of goals and the logical arraignment of the means deemed necessary for their implementation, and as a consequence to underestimate, if not utterly neglect the complexities of the process of implementation itself, with its cascade of mediations and filters imposed by the interactive dynamics going on in the multiple transaction contexts any organization consists of. Bureaucracy as a metaphor for a more and more formalized organization may well be a way for making domination (Herrschaft ) acceptable. This was Weber’s basic insight, and this is what might explain its resilience and its worldwide diffusion as a model. It does structure the interdependence of participants and the interactive dynamics which implement it. It does its

From Bureaucracy to Organized Action …

605

part in regulating them by taming their potential for conflict, and thus may very well facilitate cooperation. However, cooperation, even when facilitated, remains a problem, its potential for conflict may be tamed, but does not altogether disappear. In other words, formal organization (bureaucracy) never fully succeeds in what it is purported to do. And more importantly, it does so at a price. The price is that the supposed instrument takes on a life of its own, generates “organizational effects” and the unintended consequences they are associated with. Looking at organizations from the perspective of methodological individualism, the problem of cooperation, or put differently the management by individuals of their interdependence in organized action, is a central and eternal problem at the heart of organization. Cooperation, the response to interdependence, is always both a solution and a permanent problem. It cannot be solved once and for all. It needs to be constructed and maintained locally and contextually, in the day-to-day dealings between participants going on in the diverse contexts of action any organization consists of. The rules of the game which emerge from the resulting interaction dynamics, while contributing to their regulation, need continued monitoring as they depend on trust and on promises, and, as such, are vulnerable to moral hazard. Reasoning from a perspective of methodological individualism implies that the understanding of the way an organization functions, can only be built bottom-up, starting in the various action contexts where the various participants’ rationalities are formed by the construction of the tradeoffs which enable them to durably cooperate. As has been shown above, these rationalities do not reflect pure individual subjectivity. They are always relational and cannot be separated from the interactions which have generated them. In the end, it is these rationalities, and the exchange dynamics they entertain, which will decide whether and to what extent an organizational design, or a reform project, will be implemented. The analysis of organized action thus calls for the reconstruction of the contexts in which participants construct their cooperation on a day-to-day basis, thus successfully managing their task-interdependencies.22 From the perspective of methodological individualism, there is no such thing as an organization understood as a well-defined object driven by some overarching, all-encompassing rationality. Like all structures of and for collective action, organization must be understood as a social process of the 22

On the basis of such an analysis, I have recently highlighted the crucial importance of the managers of these different work-contexts in processes of organizational transformation in private and public organizations. In it I argued that the somewhat condescending term of “middle management” should be replaced by the notion “context managers” the crucial importance of which should be recognized, instead of trying to do away with it, as is too frequently the tendency today. Cf. Friedberg (2021).

606

E. Friedberg

construction and maintenance of a minimum of (localized) order between a set of interaction dynamics set in motion by interdependent and yet relatively autonomous actors. The overall logic of this order is under the discrete control of none of the participants, while all participate in its construction and maintenance. It is a never-ending process, always in progress and its outcome is by nature radically uncertain and undetermined. What methodological individualism teaches us in the field of organization, is that we must accept not only the random, radically uncertain nature of the outcome of collective action by humans, but also the random nature of the results of the instruments we use to steer them. Neither the quality or the enlightened nature of the goals to be pursued, nor the sheer logical and analytical quality of the rules and procedures invented for their implementation can guarantee that the expected outcome will be delivered. Only observation of the processes in real-time can provide an answer to this question. Looked at from this perspective, organizational sociology becomes a heuristic, a way for trying to understand the different forms of collective action in our societies and the complex, never linear relations between their inputs and their outcomes. A heuristic skeptical of overambitious social engineering, a heuristic forcing the enlightened to moderate their ambitions and stay vigilant to the unavoidable and ever-present unforeseen consequences of their efforts to ameliorate society or simply to act.

Bibliography Argyris, C. (1964). Integrating the Individual and the Organization. New York: Wiley. Bergeron, H., Castel, P. et al . (2018). Le Biais Comportementaliste [The Behaviorist Bias]. Paris: Presses de Sciencespo. Boudon, R. (1973). L’inégalité des chances. Paris: Armand Colin. Transl. 1974. Education, Opportunity and Social Inequality. New York: Wiley] Boudon, R. (1977). Effets pervers et ordre social . Paris: PUF. Transl. 1982. The Unintended Consequences of Social Action. London/Basingstoke: Macmillan. Boudon, R. (1984). La place du désordre. Paris: PUF. Transl. 1986. Theories of Social Change: A Critical Appraisal . London: Basil Blackwell/Polity Press. Boudon R. (1986). L’idéologie. Paris: Fayard. Transl. 2013. The Analysis of Ideology. New York: Polity Press. Cohen, M. D., March, J. G. & Olsen, J. P. (1972). A Garbage Can Model of Organizational Choice. Administrative Science Quarterly, 17 (1), 1–25. Crozier, M. (1963). Le Phénomène Bureaucratique. Paris: Ed. du Seuil. Transl. 1964. English Edition: The Bureaucratic Phénomenon. Chicago University Press. Festinger, L. (1957). A Theory of Cognitive Dissonance. New York: Harper.

From Bureaucracy to Organized Action …

607

Fligstein, N. & McAdam, D. (2012): A Theory of Fields. New York: Oxford University Press. Follett, M. P. (1924). Creative Experience. New York: Longmans, Green and Co. Forsé, M. & Parodi, M. (2010). Low Levels of Ethnic Intolerance Do Not Create Large Ghettos, A Discussion About an Interpretation of Schelling’s Model. L’Année sociologique, 60 (2), 445–473. Friedberg, E. (1993). Le pouvoir et la règle. Paris: Editions du Seuil. Transl. 1997. Local Orders: Dynamics of Organized Action. Stamford: JAI Press, 1997). Friedberg, E. (2011) (Ed.). The Multi-Media Encyclopedia of Organization Theory. Paris: R&O Multimedia. Friedberg, E. (2019). Purposive Action Is a Genuinely Sociological Perspective. Advances in Sciences and Humanities, 5 (6), 171–179. Friedberg, E. (2021). Organizations Need Context Managers. https://www.druckerfo rum.org/blog/organizations-need-context-managers-by-erhard-friedberg/ Gouldner, A. W. (1954). Patterns of Bureaucracy. Free Press: Glencoe, Ill. Gouldner, A. W. (1955). Wild Cat Strike? New York: Harper. Hirschmann, A. O. (1970). Exit, Voice and Loyalty. Cambridge: Harvard University Press. March, J. G. (1978/1988). Bounded rationality, Ambiguity and the Engineering of Choice. In J. G. March. Decisions and Organizations. London: Basil Blackwell. March, J. G. & Olsen, J. P. (Eds), (1989). Ambiguity and Choice in Organizations. Bergen: Universitetsforlaget. Maslow, A. H. (1954). Motivation and Personality. New York: Harper. Merton, R. K. (1936). The Unanticipated Consequences of Purposive Social Action. American Sociological Review, 1(6), 894–904. Olivier de Sardan, J.-P. (2021). La Revanche des contextes [The Revenge of Contexts]. Marseille: Karthala. Orton J. D. & Weick, K. E. (1990). Loosely Coupled Systems: A Reconceptualization. The Academy of Management Review, 15 (2), 203–230. Pilmis, O. & Castel P. (2020). Le virus des biais: Ce que la crise du Covid-19 révèle du comportementalisme [The Bias Virus: What the Covid-19 Crisis Reveals About Behaviorism]. AOC , 12 juin 2020. https://aoc.media/analyse/ 2020/06/11/le-virus-des-biais-ce-que-la-crise-du-covid-19-revele-du-Comportem entalisme/ Roethlisberger, F. J., Dickson W. J. et al. (1939). Management and the Worker. Cambridge: Harvard University Press. Schelling, T. C. (1978). Micromotives and Macrobehavior. New York: Norton. Selznick, P. (1949). TVA and the Grass-Roots. Berkeley: University of California Press. Selznick, P. (1957). Leadership in Administration. Berkeley: University of California Press Thaler, R. (2015). Misbehaving. The Making of Behavioral Economics. New York: W.W. Norton

608

E. Friedberg

Weber, M. (1922/1964) Wirtschaft und Gesellschaft [Economy and Society]. Berlin/ Cologne: Kiepenheuer und Witsch. Weick, K. E. (1969). The Social Psychology of Organizing. Reading: Addison-Wesley Weick, K. E. (1976). Educational Organizations as Loosely Coupled Systems. Administrative Science Quarterly. 21(1), 1–19.

Methodological Individualism and Analytic Philosophy of Action Antonio Rainone

The origin of action – its efficient, not its final cause – is choice, and that of choice is desire and reasoning with a view to an end. – Aristotle, Nichomachean Ethics, VI, 2, 1139a, 31–32. (Quoted from Aristotle 2009). But in the night of thick darkness […] there shines the eternal and neverfailing light of a truth beyond all question: that the world of civil society has certainly been made by men, and that its principles are therefore to be found within the modifications of our own human mind. – Giambattista Vico, The New Science, Book I, iii, § 331. (Quoted from Vico [1744] 1948).

A. Rainone (B) Department of Human and Social Sciences, University of Naples “L’Orientale”, Naples, Italy e-mail: [email protected]

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 N. Bulle and F. Di Iorio (eds.), The Palgrave Handbook of Methodological Individualism, https://doi.org/10.1007/978-3-031-41512-8_27

609

610

1

A. Rainone

The Two Fundamental Principles of Methodological Individualism

The issue of methodological individualism (hereinafter MI) was hotly debated in the last fifty years of the twentieth century and is still discussed among philosophers and social scientists. Not infrequently considered obscure and based on constraints and methods to which it is impossible to adhere, the principle of MI (or the individualistic approach to the social sciences) has had a wide circulation thanks to some observations by K.R. Popper in The Open Society and its Enemies (1945) and in The Poverty of Historicism (1957), but it actually originated with Max Weber, who had opposed the “individualistic method” in the explanation of historical-social events to the orientation of the so-called German “historical school” (inaugurated by Friedrich C. von Savigny, continued by Wilhelm G.F. Roscher, K. Knies, and G. von Schmoller, and whose method in economics had been challenged by Carl Menger, the famous founder of the Austrian School of economics). The historical school based its socio-historical explanations of economics and law on holistic-metaphysical notions of romantic, Hegelian and Herderian, derivation, such as that of the Volksgeist. The central thesis of MI, widely defended by Weber (and then by J. Schumpeter 1909, who spread the name), is that explanations of social phenomena must be given in terms of the behaviour (actions) of individuals and not of holistic concepts such as those that can be found in the works of G.W.F. Hegel and other Romantic philosophers and historians. According to an excellent definition, the MI “amounts to the claim that social phenomena must be explained by showing how they result from individual actions, which in turn must be explained through reference to the intentional states that motivate the individual actors” (Heath 2020). In other words, the characteristics of the individualistic method—which has nothing to do with ethics or politics—are essentially two: on the one hand, the explanation of social (and historical) phenomena as a result of individual actions; on the other, the explanation of actions as results of the intentional states of the agents. Therefore, the motivating intentional states (desires, beliefs, intentions) are taken to cause the actions of the individuals, and such actions would cause the historical-social phenomena. There is probably no better illustration of the individualistic method of explanation than that provided by Weber on the origins of capitalism. (This explanation is perhaps a paradigm model of individualistic explanation of a social phenomenon.) As is well known, capitalism originated for Weber from the behaviour of a social group (the Calvinists in particular) whose

Methodological Individualism and Analytic …

611

members had based their actions on certain religious beliefs (the doctrine of predestination above all) which required them to pursue the economic success understood as signum salutis, as proof of being among those predestined for the salvation of the soul. In this interpretation the two highlighted characteristics operate: on the one hand, the explanation of the historicaleconomic phenomenon of capitalism as a consequence or result of individual human actions (the continuous reinvestment of profits in order to obtain even more economic profits); on the other hand, the explanation of the actions of group’s individual members in terms of motivating reasons, namely the desire or intention to obtain an unearthly sign of salvation, and the belief that economic worldly success may be taken as such a sign, a proof of salvation, to be part of the elect (and be in the state of divine grace). It is not easy to say whether the actions of the believers that Weber described are included among those oriented to purposive rationality (Zweckrationalität, means-end rationality) or to value rationality (Wertrationalität, value-oriented action), according to the Weberian distinctions (we leave here aside traditional actions and affective or emotional ones). Probably we have a mixture of the two types of action in the Weberian account: Calvinist believers oriented their actions towards a value (the salvation of their souls and the belief in the doctrine of predestination) and, from this point of view, they were indifferent to any negative (or collateral) consequences of their actions, according to the Weberian model of the value-oriented action; but they were also convinced that their actions were the most effective and optimizing (rational) means to achieve the goal of obtaining the certitudo salutis. Whichever is the more correct hypothesis, what remains unchanged is in any case the macrosocial explanation of the origin of capitalistic behaviour, consisting in its being an unintended consequence or effect of the aggregation of individual actions. Here is how Weber commented on his historical investigation of the origin of capitalism: We thus take as our starting-point in the investigation of the relationship between the old Protestant ethic and the spirit of capitalism the works of Calvin, of Calvinism, and the other Puritan sects. But it is not to be understood that we expect to find any of the founders or representatives of these religious movements considering the promotion of what we have called the spirit of capitalism as in any sense the end of his life-work. We cannot well maintain that the pursuit of worldly goods, conceived as an end in itself, was to any of them of positive ethical value. Once and for all it must be remembered that programmes of ethical reform never were at the centre of interest

612

A. Rainone

for any of the religious reformers […]. They were not the founders of societies for ethical culture nor the proponents of humanitarian projects for social reform or cultural ideals. The salvation of the soul and that alone was the centre of their life and work. Their ethical ideals and the practical results of their doctrines were all based on that alone, and were the consequences of purely religious motives. We shall thus have to admit that the cultural consequences of the Reformation were to a great extent, perhaps in the particular aspects with which we are dealing predominantly, unforeseen and even unwished-for results of the labours of the reformers. They were often far removed from or even in contradiction to all that they themselves thought to attain (Weber [1904–1905] 2005, p. 48).

If macrosocial events are to be explained as the (involuntary) products of aggregated actions—which is one of the two specified principles of MI—there is still something to add about the way in which Weber conceived individual actions. Human actions (both individual and social) were for Weber actions endowed with sense, that is, provided with a subjectively intended meaning, with a subjective motivation that causes a certain behaviour aimed at the achievement of a purpose or to testify the faith in a value (Weber [1922] 2019, pp. 84–85). And it is the interpreter’s job to understand (Verstehen) the sense of actions. This is particularly true of the behaviour of Calvinists: it is not to the “objective social institutions of the older Protestant Churches” that Weber turns his attention, but rather to the “results which subjective adoption of an ascetic faith might have had in the conduct of the individual” (Weber [1904–1905] 2005, p. 99; my italics). In short, according to Weber and his individualism it is the subjective appropriation of religious asceticism that causes the behaviour of the followers of Calvinism, that is, a set of individual motivational psychological states (reasons, beliefs and desires, purposes or intentions), not certain institutional facts that would have had no causal power without the formation of these psychological states. In other words, it was not Calvinist ethics or Protestant religious institutions that caused the ‘spirit’ of capitalism: there is no causal relationship between the former and the latter, even if in ordinary language we tend to summarize this relationship by means of use of terms or concepts that seem to refer to supra-individual entities that constitute their reifications. MI, therefore, defends the principle that human actions, whatever their unforeseen or unintended results, are essentially caused by intentional psychological states, not by institutional or supra-individual facts.

Methodological Individualism and Analytic …

2

613

From Durkheim to Analytic Philosophy of Action

2.1. It shouldn’t be surprising, then, if the name of Émile Durkheim has often been contrasted with that of Max Weber. The French sociologist, indeed, had argued, more or less in the same years in which Weber was formulating his individualistic methodology, radically different and opposite methodological theses. According to Durkheim, human behaviour is not at all caused by intentional mental states, and agents often just believe that they are acting autonomously, while in reality their actions are caused by external “social facts” that have a coercive power on individuals. Here we have Durkheim’s famous theory of “social facts”: When I perform my duties as a brother, a husband or a citizen and carry out the commitments I have entered into, I fulfil obligations which are defined in law and custom and which are external to myself and my actions. Even when they conform to my own sentiments and when I feel their reality within me, that reality does not cease to be objective, for it is not I who have prescribed these duties. I have received them through education.[…] Not only are these types of behaviour and thinking external to the individual, but they are endued with a compelling and coercive power by virtue of which, whether he wishes it or not, they impose themselves upon him (Durkheim [1895] 2013, p. 20).

The same would also apply to religion. In a way that would appear as an ante litteram reply to Weber, Durkheim noted also that “the believer has discovered from birth, ready fashioned, the beliefs and practices of his religious life; if they existed before he did, it follows that they exist outside him” (ibid .: p. 20). There is, in the contrast between Weber and Durkheim, a paradigmatic example of the opposition, in the social sciences, between individualist and anti-individualist theories, the former founded on autonomy and the intentional states of individuals, the others on the dependence of human actions on supra-individual “social facts”—that is, norms and institutions. These facts appear to be the cause of human behaviour and (causal) explanations for these behaviours should be based on them. An immediate objection that can be levelled at this way of conceiving social explanations is that it takes into consideration only standardized behavioural models; moreover, as has been repeatedly claimed against sociological functionalism, it is unable to explain social change, unlike Weber’s individualistic method (as we have seen with the capitalism example). Durkheim’s theory seems to limit itself exclusively

614

A. Rainone

to rule-governed behaviour, neglecting behaviour based on rational deliberation, problem solving and the strategies pursued by individuals to achieve their goals. Nevertheless, it might be of some surprise that Durkheim’s theory had a remarkable influence in the philosophy of the human sciences of the second half of the twentieth century, in some cases directly, in others indirectly and involuntarily—a sign that the question of individualism against anti-individualism, regardless of metaphysical orientations, constitutes an important and inevitable thematic core of the methodology of the social sciences. Durkheim’s position finds an echo first of all in an important essay by the philosopher Maurice Mandelbaum with the significant (and Durkheimian) title Societal Facts (1955). Here the author aims particularly to criticize the individualistic theses of Popper and von Hayek, according to which denying the validity of MI would be tantamount to subscribing to holistic concepts such as those of Hegel, Comte or Marx. For Mandelbaum, by contrast, a critical orientation towards MI does not at all imply an organicistic and holistic conception of society and history, but, more modesty, the impossibility of reducing social events to individual behaviours and to the psychological states of individuals. This is because in order to describe or explain certain human actions it is implicitly necessary to refer to certain ultimate societal facts or concepts that have no equivalent in psychological states and in the behaviour of the individuals: “If […] societal facts”, Mandelbaum noted, “are as ultimate as are psychological facts, then those concepts which are used to refer to the forms of organization of a society cannot be reduced without remainder to concepts which only refer to the thoughts and actions of specific individuals” (Mandelbaum 1955, p. 307). Mandelbaum’s famous example, in this connection, is that of a person who goes to a bank to withdraw money: explaining the behaviours involved here, let’s say to someone who knows nothing of the banking system, implies a necessary reference not only to the purposes of the people involved (according to the means-end model), but also to their roles and the banking system as a whole; and this additional information (for example, opening a bank deposit account, depositing or withdrawing money) cannot be reduced to the observed behaviours or the psychological states, but imply reference to the institutional organization of the society. The same applies to all inter-individual relationships that characterize social behaviours; Mandelbaum wrote: In all cases of this sort, the actual behaviour of specific individuals towards one another is unintelligible unless one views their behaviour in terms of their status and roles, and the concepts of status and role are devoid of meaning

Methodological Individualism and Analytic …

615

unless one interprets them in terms of the organization of the society to which the individuals belong (ibid .: p. 308).

Certainly, Mandelbaum’s argument appears in many ways plausible, especially considering that his anti-individualism is by no means intended as a vindication of the old holistic and organicistic metaphysical conceptions. Moreover, even in his rehabilitation of Durkhemian themes, he does not seem to propose an ontological view of societal facts, but rather a semantic analysis of the meaning of the statements with which human behaviour is described and explained within a social context. However, there is more to say about anti-individualism, especially with regard to certain implications of the semantic theories of the later Wittgenstein. 2.2. The question of MI seems to have been extraneous to analytic philosophy with the exception of a few sporadic cases. However, in various publications inspired by the Wittgenstein’s Philosophical Investigations (1953) we can find arguments that in one way or another respond to it implicitly, in particular in the developments of Wittgensteinian philosophy of language regarding the problem of the explanation of behaviour—that is, in the so-called action theory. As far as we know, Durkheim had no direct influence on Wittgenstein; nevertheless, despite starting from a completely different perspective from Durkheim’s, Wittgenstein seems to support theses very similar to those of the French sociologist. Wittgenstein’s starting point was the analysis of the notion of meaning, extended to the problem of the ontological status of mental states and to the traditional problem of their presumed causal power with respect to actions, which the Viennese philosopher had set out in those notes that later would become the Philosophical Investigations (1953) and the Blue and Brown Books (1958). For Wittgenstein, these themes were closely interconnected. As is well known, by questioning the denotationist conception of meaning— which, under the influence of Bertrand Russell, he had supported in the Tractatus (1921)—Wittgenstein proposed in the two posthumous works a very different view of meaning. In particular, he paid close attention, among other themes, to the meaning of mentalistic terms such as ‘desire’, ‘intention’, ‘expectation’, ‘belief’, ‘will’, etc., highlighting the implausibility of the traditional conception and of common sense, which considers them, in a Cartesian fashion, as referring to mental entities endowed with causal power. Mentalistic terms are misleadingly regarded as names of actually existing entities, the misunderstanding arising from considering a substantive as the name of an object, “one of the great sources of philosophical bewilderment” which induces us to “look for a thing that corresponds to it” (Wittgenstein

616

A. Rainone

1958, p. 1). But these terms should be understood as being devoid of an objective reference, and rather as part of our way of expressing ourselves, in the practice of our language, regarding the explanations or descriptions of certain behaviours or actions. Wittgenstein’s observations did not question our way of explaining human actions by what Russell had called propositional attitudes, linguistically expressed by statements such as ‘to believe that p, to know that p, to desire p’, etc., provided with intentional contents by means of which we explain, give the reasons, for our actions (for example, ‘I did y because I wanted , intended to obtain x ’). Wittgenstein instead had refused the ontological and causalistic thesis that statements of propositional attitudes and mentalistic terms refer to mental states that would cause our actions. According to Wittgenstein mentalistic terms amount instead to only a façon de parler —more precisely, a way of interpreting human behaviour, but essentially devoid of ontological, mentalistic and causalistic implications with respect to behaviour. The most important consequence of these antimentalistic theses is a conception that, while not questioning purposive behaviour, considers it based not on individual-subjective states, but on social rules and conventions. The intention with which (i.e. the end in view of which) one acts—a central concept of Wittgensteinian reflection that formed the core of the important essay Intention by G.E.M. Anscombe (1957)—should not for Wittgenstein be located in a mental nor subjectivistic (nor yet causalistic) ontological dimension, but rather in our language, social practices and conventions or rules. Intentions and other mentalistic terms are just ways of describing behaviours in public or social contexts. (For example, if I see a man with an umbrella, I can describe or interpret his behaviour by saying that he believes it will rain and that he intends not to get wet. But these descriptions do not imply the existence of mental states that cause his behaviour.) Here is Wittgenstein’s famous claim about the concept of intention: “An intention is embedded in its situation, in human customs and institutions. If the technique of the game of chess did not exist, I could not intend to play a game of chess” (Wittgenstein 1953, § 337). This holds good both for language and for action, which according to Wittgenstein, like language, is always rule-governed. Wittgenstein’s antimentalism and his conception that purposive or goaldirected behaviour is always of the rule-following or rule-governed type have had a particular influence in subsequent analytic philosophy. Peter Winch, an authoritative follower and interpreter of Wittgenstein, for example, stressed that we understand what a motive or a reason for acting is only in reference to the rules in force in a social context: “learning what a motive is belongs

Methodological Individualism and Analytic …

617

to learning the standards governing life in the society in which one lives” (Winch 1958, p. 83). What about means-end behaviour which, according to a tradition dating back at least to the Aristotle’s Nichomachean Ethics, constitutes our paradigm of the concept of action? Actions are performed for reasons, that is with certain intentions or to achieve purposes that represent their subjectively intended meaning, as Weber had argued in his individualistic perspective. If reasons for actions can only be understood with reference to the social context in which the actions are performed, then the means-end model is subordinated to certain social norms without which it would not be possible to understand the meaning (or purposes) of the actions. Perhaps there is some truth in this consideration: to take an ethnological example, could we understand the behaviour of a shaman (the ends or the sense of his actions) without a preliminary knowledge of the shared habits and beliefs of the tribe to which he belongs? Maybe yes, maybe not. For we can understand his behaviour starting from the shared habits and beliefs of the tribe, but yet also understand the shared habits and beliefs starting from the shaman’s behaviour and from his interaction with the community members. There is obviously a reciprocity between the two types of understanding. Be that as it may, the thesis of the priority of social norms, and of their constitutive function with respect to action, has been supported by more than one author. Of particular interest appears above all what R.S. Peters argued in a famous book dedicated to the concept of motivation (Peters 1958), in which we can find arguments very similar to those of Wittgenstein’s and his followers. Peters’ starting point was means-end behaviour; as you can read in book’s introduction, The paradigm case of a human action is when something is done in order to bring about an end. So the usual way of explaining an action is to describe it as an action of a certain sort by indicating the end which [an agent] had in mind. We therefore ask the ‘why’ question in a more specific form. We ask what was his reason for doing that or what was the point of it, what end he had in mind (Peters [1958] 1960, p. 4).

As we have already noted, this is one of the two fundamental tenets of the MI, the one concerning the explanation of the action. It is by no means new: it goes back to Aristotle’s Nicomachean Ethics, and it seems difficult to question it. Peters certainly accepts this principle, but with the caveat that “this simple purposive model of a man taking means to bring about an end is further complicated by the fact that norms enter into and often entirely define the end” (ibid .: p. 5). There is a wide range of ends, such as passing an examination, getting married, becoming a professor, which involve social

618

A. Rainone

rules and conventions—indeed they “are defined almost entirely by social convention” (ibid .: p. 4). Even natural or biological ends (eating, drinking, having sex) involve social conventions for men living in a society. Here is Peters’ fundamental thesis: Man is a rule-following animal. His actions are not simply directed towards ends; they also conform to social standards and conventions, and unlike a calculating machine he acts because of his knowledge of rules and objectives. […] Ends are not given as natural terminating points like a chain of oases distributed across a desert. And to a large extent, what counts as falling within a means-to-end explanatory framework is determined by convention. […] Man in society is like a chess-player writ large. Requests for explanation are usually reflections of our ignorance about the particular rule or goal which is relevant to the behaviour in question (ibid .: pp. 5–7).

It is not surprising, therefore, that Peters refers to Durkheim to support his (Wittgensteinian) theses: “Men are born into a social environment; social norms confront them from the start ‘comme les choses’, as Durkheim said” (ibid ., p. 96). Nor is it surprising that Weber is never mentioned in his book. Is Peters’ thesis a form of anti-individualism? Certainly, Peters’ starting point is means-end (or goal-directed) behaviour—which is a pivotal tenet of the MI; and yet it seems that Peters is subordinating the purposive behaviour to social practices, rules and norms, which thus appear to be the authentic causes of human actions. In this perspective individuals act only in a derivative sense, we could say. Perhaps there is some truth in the Peters’ view, even if there seems to be some exaggeration. It is certainly true that I can set myself the goal of becoming a professor or a doctor or of getting married only on condition that I know the norms and rules that define socially both the ends and the means to achieve those ends. After all, even a natural end (or desire) such as that of drinking some water is often mediated by technical and social norms, such as, to take an example of Thomas Nagel ([1970, 1978, p. 33), when we satisfy it after inserting a coin in a drinks vending machine out of which pops a bottle of water. Of course, there is a lot of social information we need to have when we take actions to pursue our goals, but this does not mean that our actions depend on the social context and norms rather than on our desired ends and on our beliefs about the best ways to achieve them. Suppose, for example, that a high school graduate intends to study medicine to become a doctor; becoming a doctor (his goal) and studying medicine (the best means to become a doctor) are certainly goals and means socially defined: as Wittgenstein would have said, I cannot have the intention of becoming a doctor or studying medicine

Methodological Individualism and Analytic …

619

if the profession of doctor did not exist and if there were no medical faculties. But this seems obvious (as it is obvious that a student of the 1950s could not have had the desire or end to study computational medicine). However, our graduate, after having decided to enrol in a medical faculty, will still have to evaluate which, among different medical faculties, is the best for him, listing the features he prefers and deciding accordingly. Even taking into account the constraints imposed by the social norms and rules, what primarily matters in the purposive behaviour is the structure of the practical reasoning by which an agent chooses the action (or the course of actions) he deems best to achieve his desired ends. This is the basic principle of our concept of action and it is the basis of human rationality. If human actions were exclusively determined or conditioned by norms, practices and rules, then there would be no deliberative rationality in agents, but only rule-following behaviours. And, more importantly, if all actions were of the rule-following type (similar to automatic, blind or mechanical routines), then they could not be distinguished from those that Weber called “traditional actions” (Traditionales Handeln), tracing back this behaviour to ingrained habits and patterns devoid of any form of rational deliberation. But is it possible that this should always be the case? There is another and more detailed argument of the same type as Peters’, formulated by a philosopher influenced by Wittgenstein and Anscombe. In his Free Action (1961) A.I. Melden argued a radical form of contextualism relative to the explanation of human behaviour. As we have already noted, Wittgensteinian theories of human action are essentially antimentalistic and are based in particular on social norms and conventions. In this perspective, social norms provide all that is needed to explain (or, better, understand ) actions. Contextualism is perhaps the best example of this view. A famous (and outdated) Melden’s example involves a driver who stretches his arm out of the car window. Why did he raise his arm? is the question that someone unfamiliar with what has happened might ask. ‘The driver has stretched out his arm to indicate he was preparing to make a turn’ is the answer that we might give. This answer appears obvious: his reason (or motive) for raising his arm was that he wanted to signal a turn to other drivers and believed that raising his arm was the way to signal his intention to make a turn. This is a typical (and individualistic) common sense explanation of the driver’s action, but is not the obvious explanation Melden gives us. His point of view is completely different from that of common sense’s. According to Melden (who follows Wittgenstein and especially Anscombe) we can describe his arm raising as a signalling a turn only by reference to the rules governing road traffic; a more detailed explanation involving his intention or reason would

620

A. Rainone

add practically nothing to what appears implied in the previous description. An intentional (and mentalistic) explanation of the ‘in order to’ type is not necessary, given the context of the action; as Melden specifies, “the driver intended to make the turn – this is why he raised his arm – he raised his arm in order to indicate to others that he was preparing to make a turn. But does ‘in order to indicate that…’ mark some mental occurrence?” (Melden 1961, p. 87). Not at all. Here’s Melden’s argument: In short, citing the motive was giving a fuller characterization of the action; it was indeed providing a better understanding of what the driver was doing. […] the statement that declares the agent’s motive refers not to something that cross the agent’s mind at that moment or to any other interior occurrence, but to a matter of public performance. To say, in response to the question ‘Why did you raise your arm?’, ‘I did so, in order to indicate that I was preparing to make a turn’, is to call attention not to some mental occurrence, but to the action that was performed and the circumstances in which this occurred. […] Raising one’s arm and thereby signaling is a highly conventionalized method of communication. This means, first, that there are fixed rules of the road that determine the precise form in which the relevant bodily movement is to be executed. Second, it involves the convention that whenever this bodily movement occurs in the appropriate circumstances, then, whatever the agent’s intentions may be, the movement of one’s arm will be understood as an instance of signalling (ibid .: pp. 88–91).

In cases of this sort, concludes Melden, “the conventions, rather than the agent’s intentions, determine what is being done” (ibid .: p. 92). (This is a very unhappy statement, as we shall see below.) Melden’s anti-mentalism presents a curious consequence: we may signal a turn without even knowing it, or without intending to do so. And this is because, according to his point of view, it is the rules that matter, not our alleged mental states (intentions, desired ends, decisions). It is clear that in a highly conventionalized context, our behaviour can be interpreted according to the rules that characterize it, but that’s not the whole story. In order to signal (a turn or anything else) an agent must have the intention to do so; furthermore, even if there is a rule that establishes how to do it, he must at least have the intention to follow the rule. In other words, actions, as had said Weber, have a subjectively intended meaning . The anti-mentalistic perspective, however, is not Melden’s only assumption; there is, in his argument, another important Wittgensteinian assumption: anti-causalism. If the explanations of our actions are ‘determined’ by conventions, then, we may suppose, they aren’t caused by mental occurrences, contrary to what a long tradition (from Aristotle to Descartes to Hume) had claimed. Anti-mentalism

Methodological Individualism and Analytic …

621

and anti-causalism are closely related to each other in Melden (and in other Wittgensteinian philosophers such as Anscombe, von Wright, Stoutland and Searle). Actions are not caused by mental states but, as Melden said, they are ‘determined’ (a word semantically similar to the word ‘caused’) by social conventions and rules. But if so, nevertheless, then there exists a form of social causation (neither Humean nor scientific) of our behaviour—a somewhat obscure form of causation. From this point of view there emerges a form of Durkheimian anti-individualism. However, Melden follows Wittgenstein and he particularly insists on our understanding of actions as meaningful gestures in a social context; the interpretation of a gesture as an action comes from language learning in appropriate (external) circumstances—in short from the education we have received. Thus social training in a social context is essential to learning not only language, but also to understanding meaningful behaviour, the gestures with which we intend to communicate something (ibid .: pp. 187–190).

3

Causalism, Rationality, Deliberation and Methodological Individualism

3.1. Perhaps we could be left rather perplexed about the theories of action inspired by Wittgenstein’s theses. It seems that these theories deal only with explicit bodily movements within a socio-normative context such as raising an arm or a hand (for example, I can raise my hand during a lecture to request to speak). True, these are intentional actions aimed at an end, but they are very elementary actions whose understanding is based on the social context and its rules—on forms of life, if we want to use this famous Wittgenstein’s expression. But there are also many more complex actions whose understanding is by no means reducible to a normative social context—all those actions that are based on deliberation and decision, on the human rationality. Thus, on the one hand there is the standardized behaviour based on social norms and rules, on the other hand there are individual behaviours based on norms of individual rationality. And we cannot limit or trace all behaviour to those of the former type. This is a consideration that surely both Anscombe (1957) and von Wright (1971) must have had in mind when they proposed the Aristotelian practical syllogism as a model for understanding human action. As is well known, the practical syllogism is a real enigma: it is not difficult to find counterexamples for it, demonstrating its inadequacy. Aristotle himself did not seem very convinced of its adequacy for understanding or explaining human

622

A. Rainone

action, attaching more importance, in the Nicomachean Ethics, to deliberation regarding the (causal) origin of action. According to Aristotle action is a means to attain an end, and it is performed only after deliberation, which is the search for the best means to achieve the end (cf. Aristotle 2009: III, 3, 1112b, 15–20). Moreover, Aristotle had also insisted on causation: “The origin of action – its efficient, not its final cause – is choice (prohairesis), and that of choice is desire and reasoning with a view to an end” (ibid.: VI, 2, 1139a, 31–32). Was Aristotle already a (methodological) individualist? The connection between desire, choice (or intention) and causation is perhaps Donald Davidson’s main contribution to action theory. Davidson was the philosopher who rediscovered the Aristotelian doctrine of action. Since 1963, the year of publication of his fundamental and influential essay “Actions, Reasons, and Causes”, he criticized the theories of action based on the Wittgenstein’s anti-mentalism and anti-causalism, while accepting a central Anscombe’s thesis (the idea that a single action can fall under many different descriptions). According to Davidson actions are caused and explained by reasons, but reasons are not social norms and rules: in other words they are not obscure social entities. Reasons for actions are pairs of intentional mental states (whatever their real nature may be: mental, physical or neurochemical): pro attitudes and beliefs. Pro attitudes are individual dispositions or propensities towards actions of a certain type; beliefs have as their content particular actions chosen—among various available options—by the agents to satisfy their pro attitudes. Here’s Davidson’s famous argument: A reason rationalizes [explains] an action only if it leads us to see something the agent saw, or thought he saw, in his action – some feature, consequence, or aspect of the action the agent want, desired, prized, held dear, thought dutiful, beneficial, obligatory, or agreeable. […]. Whenever someone does something for a reason, therefore, he can be characterized as (a) having some sort of pro attitude toward actions of a certain kind, and (b) believing (or knowing, perceiving, noticing, remembering) that his action is of that kind (Davidson [1963] 1980, pp. 3–4).

Prima facie, this thesis may perhaps appear complicated and difficult to understand. Indeed, despite some difficulties in understanding, what Davidson means is that actions are means to ends—the actions chosen to satisfy the pro attitudes—and ends may be of various types: we may have a pro attitude towards taking a holiday or something more specific such as attending a certain philosophy conference or visiting Naples, and so on. Each pro attitude selects our ends or desires in a general way, while a particular

Methodological Individualism and Analytic …

623

action (our choice) is what we believe is necessary to satisfy the desire or achieve the end. The Davidson’s argument is particularly interesting for various reasons. Firstly, because it takes up Aristotle’s doctrine of action and deliberation (and of desire and choice as the causes of action), which seemed to have been overlooked by the philosophers influenced by later Wittgenstein. Secondly, because it appears akin to the Weberian theory according to which to understand an action it is necessary to identify an agent’s subjectively intended meaning (which can be an end or even a value). This affinity is particularly interesting for our discussion, because, thirdly, Davidson’s argument appears as a defence of MI, at least as regards action explanation (as in Weber’s thesis). Let us try to explore this point. From a careful reading of Davidson’s definition of the concept of a pro attitude, it may be noted that this concept is intended in an essentially antiWittgensteinian and individualistic perspective. Davidson includes in the class of the pro attitudes a wide range of psychological dispositions, such as desires, wanting, urges, promptings, and a great variety of moral views, aesthetic principles, economic prejudices, social conventions, and public and private goals and values in so far as they can be interpreted as attitudes of an agent directed towards actions of a certain kind (ibid .: p. 4).

On Davidson’s account there is no difference between private and public ends, values and social conventions, and so on, so long as they are all considered as psychological and individual attitudes (endowed with causal force) of the agents. In particular, for Davidson we may consider all pro attitudes as forms of desire or wanting, since whatever an agent (intentionally) does, he does it because, all things considered, he wants to do it; for example, if I’m driving my car and I find myself in front of a red traffic light, I am obliged to stop my car, but I stop my car because I want, desire or intend to do so. Thus, as Davidson had pointed out, “it is not unnatural, in fact, to treat wanting as a genus including all pro attitudes as species” (ibid .: p. 6). It should be emphasized even more, moreover, that we usually act for individual reasons even in those cases where the reasons are provided by social norms or rules. In this respect it seems interesting that a single action can be explained by various and conflicting reasons. Let’s go back for a moment to the example of the red traffic light. I may stop my car because I have a pro attitude towards respecting rules and laws (I want to respect them), and I believe that stopping my car is a way of respecting rules and laws; or I may stop the car because I want to avoid the penalties that I would incur in case of non-compliance if I didn’t stop my car, and I believe stopping the car is a way

624

A. Rainone

of avoiding penalties (in this second case my attitude would be to avoid as far as possible the collateral damage of my actions); or, again, I may stop the car to avoid a collision with a car coming from my right or left side. In all three cases, therefore, the action is the same, but the reasons for performing it are obviously different: the difference lies in the desires and beliefs that cause and rationalize the action. Explaining an action, therefore, means identifying the right reason or the right desire-belief pair; to disregard this is to overlook the difference between an authentic explanation and a simple justification (and this would be Melden’s mistake, as Davidson noted). In a nutshell, it is my pro attitudes and beliefs that explain (cause) my actions, not social rules. There have been various objections to this view. For example, John Searle (2001) argued, against the Davidsonian theory, that there are reasons for actions independent of desires: Searle’s claim is that obligations and institutional actions are not so much based on (internal) desires or wants and beliefs, as on our commitment to respect the (external) rules and norms of the social context in which we find ourselves living. This argument is perhaps similar to Melden’s and does not offer a clear analysis of why we follow certain rules (what exactly does a commitment consist of? and what does distinguish it from what I want or intend to do?). In fact, it seems rather reasonable to presume that social rules must be internalized in order to be followed, and it is not clear how they can be effective without first entering our cognitive and conative system. Moreover, Searle also criticizes the means-end model, that is the instrumental model of human action, which Davidson had took up from Aristotle and Hume. His critique appears therefore as a form of anti-individualism. Davidson’s model of action explanation has become known as desire-belief model . According to this model we explain an agent’s action by citing a desire (an end or a value or any other pro attitude) of the agent and a belief about the best way to satisfy it. Belief is particularly important because we assume that the agent has evaluated the various ways or options to achieve his goal, finally choosing the one he believes is the best. (This is the so-called principle of practical rationality.) Desires and beliefs are subjective states and, together, they provide both a reason and a cause for action; as Davidson said, a reason for acting is a rational cause (Davidson [1974] 1980, p. 233). It goes without saying that Davidson’s desire-belief model shows a close relationship with the fundamental principles of decision theory. We can note this in the fundamental Davidsonian theory of practical reasoning and intention (see Davidson 1978). Moreover, here’s a clear argument from Davidson on this point:

Methodological Individualism and Analytic …

625

Beliefs and desires conspire to cause, rationalize, and explain intentional actions. We act intentionally for reasons, and our reasons always include both values and beliefs. We would not act unless there were some value or end we hoped to achieve (or some supposed evil we hoped to avoid), and we believed our course of action was a way of realizing our aim. Decision theory is a way of systematizing the relations among beliefs, desires, and actions. It does this by imposing a complex, but clearly defined, pattern on the way in which people’s beliefs and desires interact. […] the [decision] theory answers to our intuitions about how actual decisions are made; in effect it simply spells out our commonsense apparatus for explaining intentional action. For we all, whether we think about it or not, make our decisions in terms of how we weigh the values of various possible outcomes of our actions, and how likely we think one or another course of action is to attain those values (Davidson [1997] 2001, pp. 125–126).

Of course, if what Davidson argued is correct, our explanations of actions are based on a strictly individualistic method. The desire-belief model and decision theory are in fact the main individualistic approaches to human behaviour. We may here add that Davidson’s causal theory of action is intended as an alternative to the theories of action inspired by Wittgenstein. From this point of view Davidson is very clear: we act on the basis of our desires, beliefs and preferences, and the social context is not sufficient to explain an action: as Davidson had noted in criticizing Melden’s argument, it makes no sense to say the driver signalled a turn if he didn’t have the relevant desire and belief: perhaps, in the given context, he had signalled a turn involuntarily, without realizing it, without to have the (subjective) intention to signal, and if so then he had no reason to signal a turn, whatever the social rules may be. 3.2. Davidson’s theses are important and have been very influential. There has been no philosopher of action who has not discussed them, defending them (such as A. Goldman 1970) or criticizing them (such as J. Searle 2001). Perhaps his theses are not explicitly aimed at defending individualism, and yet they offer a clear model of individualistic explanation of human action, even if not of social phenomena. In any case, psychological explanation is relevant for MI because some forms of anti-individualism aim to deny the legitimacy of intentional explanations of human actions. In analytic philosophy this point has been specifically discussed by Bernard Williams, whose views are in part similar to Davidson’s. In a paper with the significant title “Formal and Substantial Individualism” (1985) Williams recognizes the impossibility of identifying MI with a reductionist thesis; as

626

A. Rainone

Williams claims, we have to consider the issue of individualism essentially as a question of explanation, not of reduction: Questions about individualism – at least as an issue in the social sciences – have often been understood to be questions about the possibility of reduction, but I take the interesting questions in that dimension to concern explanation rather than reduction (Williams 1985, p. 122).

We can certainly agree on this. Remember, in this regard, Mandelbaum’s anti-individualist argument that it is not possible to reduce descriptions of human behaviour involving social rules and institutions to descriptions devoid of social terminology. It is very difficult to dispute this argument. Nevertheless, this would be a destructive objection to MI only if MI were indeed a reductionist conception. But that’s not the case at all. The use of institutional terminology is not in conflict with MI simply because MI is not a reductionist theory but a methodological theory according to which human action is explained by intentional states (plus deliberation), even in cases of institutional actions (as Weber had already pointed out). In this perspective, Williams outlined two types of individualism, one weak and one strong. The weak type he calls formal individualism, while the strong type he calls substantial individualism. Formal individualism is a minimalistic form of individualism: its fundamental principles are consciousness and deliberation. The consciousness of an agent carrying out an intentional action is intimately connected to a previous deliberation. As Williams writes, […] deliberative or practical questions are radically first-personal, where that means that they are individually first-personal. That is one of two truths that together constitute the position I shall call ‘formal individualism’. The other is to the effect that what an individual does is often explained by the individual’s deliberation, and, to the extent that his or her action is intentional, it can be explained in terms of a deliberation that the individual could have conducted. It follows from the two claims together that intentional action can always be explained by reference to a consciousness which the agent at least could have had and in many cases did have, and which refers to the agent (ibid .).

It is worth noticing, then, that there doesn’t seem to be much particularly new about individualism. The principles specified by Williams are in fact those advocated by both Aristotle and Davidson (among other philosophers). They refer to our image of human beings as rational individuals who pursues their own ends and are usually aware of what they are doing and, albeit within certain limits, the way to achieve those ends.

Methodological Individualism and Analytic …

627

Formal individualism does not claim to eliminate any reference to social norms and institutions from the explanations of individual behaviour, as a substantial individualism probably would claim. But what of a substantial individualism understood as a form of eliminativism according to which social and individual explanations should dispense with social concepts altogether? Here we would like to put forward the hypothesis that the strong form of individualism, i.e. a “substantial individualism”, is essentially an equivocal or fictitious and impossible goal constructed by MI’s opponents. But a criticism based on this impossible goal would be too simple a criticism: if MI’s fundamental principle were to give up social concepts in the explanations of the social sciences, then it would be trivially false. “According to formal individualism”, Williams wrote, “it is true (indeed, unproblematically true) that intentional actions can be explained by reference to the intentional states of individuals”; but yet “some opponents of individualism seem impelled to deny even this much” (ibid .: 123–124). In other words, anti-individualists contest the image of human being as a rational individual, or, at least, the idea that behaviour can be adequately explained by reference to the intentional states (plus deliberation) of individuals. We can give various examples of this view: the most important are Structural Marxism of Althusser, Social Structuralism, Sociological Functionalism and Psychoanalysis (in particular in Jung’s version which hypothesized the existence of a supra-individual entity as the Kollektiven Unbewussten, Collective unconscious—a radical re-dimensioning of consciousness and intentionality). On this point there is much to be learnt from Williams’ argument about the depth of the formal (and substantial) individualism: according to some anti-individualists—he observed—individualism is an inadequate model of explanation because “it says nothing about the explanation of […] intentional states themselves”: The problems of depth are raised by theorists who claim for instance that all or most of the things that people do, or the things they do that are socially significant (whatever quite that may turn out to mean), can be explained by their position in the social structure: at the limit, that individuals act just as Träger or bearers of social roles, at (or as) the conjunctions of certain social practices. […] The point is that even this type of theory does not deny, or should not try to deny, that these actions can be explained by reference to the intentional states of the agents; what it offers is an account of those intentional states and their content, in terms of social structure (ibid .: 124).

If Williams’ argument is correct (and we believe it is), then antiindividualism tries to explain intentional actions at a deeper level, that is,

628

A. Rainone

on a sub-intentional and social level: society (a vague notion that needs to be clarified) would be the authentic cause of our intentional states (desires, beliefs, intentions) and deliberations from which our actions would derive only in appearance. This claim appears akin to Wittgensteinian theories of action, based on forms of life and social practices. But what evidence do we have for abandoning the model of the rational agent who acts on the basis of intentional states and which has proved so effective from an epistemological and interpretative point of view? Should we really give up this two-thousandyear-old paradigm and replace it with strongly metaphysical concepts devoid of any adequate confirmation? (On this point see also Pettit 1980).

References Anscombe, G.E.M. (1957). Intention. Ithaca (NY): Cornell University Press. Aristotle. (2009). The Nichomachean Ethics. Translated by D. Ross, revised by L. Brown. Oxford and New York: Oxford University Press. Davidson, D. (1963). Actions, Reasons, and Causes, repr. in Davidson (1980): pp. 3–19. Davidson, D. (1974). Psychology as Philosophy, repr. in Davidson (1980): pp. 229– 244. Davidson, D. (1978). Intending, repr. in Davidson (1980): pp. 83–102. Davidson, D. (1980). Essays on Actions and Events. Oxford: Clarendon Press. Davidson, D. (1997). The Emergence of Thought, repr. in D. Davidson, Subjective, Intersubjective, Objective. Oxford: Clarendon Press, 2001: pp. 123–134. Durkheim, É. (1895). Les Règles de la méthode sociologique. Translated by W. D. Halls, ed. by S. Lukes, The Rules of Sociological Method . London: Palgrave Macmillan, 2013. Goldman, A. I. (1970). A Theory of Human Action. Englewood Cliffs (NJ): PrenticeHall. Heath, J. (2020). Methodological Individualism. The Stanford Encyclopedia of Philosophy (Summer 2020 Edition), Edward N. Zalta (ed.), https://plato.sta nford.edu/archives/sum2020/entries/methodological-individualism/ [consulted in November 2020]. Mandelbaum, M. (1955). Societal Facts. The British Journal of Sociology 6: pp. 305– 317. Melden, A.I. (1961). Free Action. London: Routledge & Kegan Paul. Nagel T. ([1970] 1978). The Possibility of Altruism. Princeton (NJ): Princeton University Press. Peters, R.S. (1958). The Concept of Motivation. London: Routledge & Kegan Paul, 1960.

Methodological Individualism and Analytic …

629

Pettit, P. (1980). Rational Man Theory. In C. Hookway and P. Pettit (Eds.), Action and Interpretation. Studies in the Philosophy of the Social Sciences (pp. 45–63). Cambridge (UK): Cambridge University Press. Popper, K.R. (1945). The Open Society and its Enemies, 2 vol. London: George Routledge & Sons. Popper, K.R. (1957). The Poverty of Historicism. London: Routledge & Kegan Paul (Published for the first time in Economica, New Series, vol. 11, 12, in three issues: nos. 42, 43 and 44, 1944–1945). Schumpeter, J. (1909). On the Concept of Social Value. The Quarterly Journal of Economics 23: pp. 213–232. Searle, J.R. (2001). Rationality in Action. Cambridge (MA): The MIT Press. Vico, G. ([1744] 1948). La scienza nuova giusta l’edizione del 1744. a cura di F. Nicolini, Bari: Laterza, 1928. Engl. transl. from the 1928 ed. by Thomas Goddard Bergin and Max Harold Fisch, The New Science. Ithaca (NY): Cornell University Press, 1948. von Wright, G. H. (1971). Explanation and Understanding. Ithaca (NY): Cornell University Press. Weber, M. ([1922] 2019). Wirtschaft und Gesellschaft. T Mohr. Engl. transl. by K. Tribe, Economy and Society. Cambridge (MA), London: Harvard University Press, 2019. Weber, M. ([1904–1905] 2005). Die protestantische Ethik und der Geist des Kapitalismus. Transl. by Talcott Parsons with an introduction by Anthony Giddens, The Protestant Ethic and the Spirit of Capitalism, London and New York: Routledge, 2005. Williams, B. (1985). Formal and Substantial Individualism. Proceedings of the Aristotelian Society, New Series, 85 (1984–1985): pp. 119–132. Winch, P. (1958). The Idea of a Social Science and its Relation to Philosophy. London: Routledge & Kegan Paul. Wittgenstein, L. (1953). Philosophische Untersuchungen/Philosophical Investigations. Ed. by G.E.M. Anscombe, R. Rhees. Oxford: Basil Blackwell. Wittgenstein, L. (1958). The Blue and Brown Books. Ed. by R. Rhees. Oxford: Basil Blackwell.

Methodological Individualism/Sociology of Knowledge/Cognitive Sociology Renaud Fillieule

For all those who adhere to the principles of an individualistic methodology, a collective belief can only be explained insofar as it makes sense for an idealtypical subject. (Boudon 1997, p. 38)

The sociology of knowledge got its name from Scheler (1926) but already existed long before his work. Classical sociologists such as Marx, Tocqueville, Durkheim, Weber, etc., all developed theories relating knowledge to its social conditions of production. This field has gone through three phases, first classical , then modern (inaugurated by Barnes 1974), and finally cognitive (with the work of Boudon 1986). This chapter retraces this history in the light of methodological individualism (hereafter, MI). Our main thesis is that the sociology of knowledge, just like any other field in the social and economic sciences, greatly benefits from the application of an “individualistic” methodology. In the classical sociology of knowledge, the relationship between thought and social factors was not satisfactorily nor consistently analyzed. The modern sociology of knowledge made the problem worse by refusing to use reasons as causes of beliefs. Finally, the cognitive turn initiated by Boudon R. Fillieule (B) Institute of Social Sciences, University of Lille, Lille, France e-mail: [email protected] CLERSE UMR 8019, Villeneuve d’Ascq, France

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 N. Bulle and F. Di Iorio (eds.), The Palgrave Handbook of Methodological Individualism, https://doi.org/10.1007/978-3-031-41512-8_28

631

632

R. Fillieule

in the mid-1980s provided the proper “individualistic” viewpoint and made it possible to develop satisfactory theories of collective beliefs, and also to offer convincing interpretations of older theories. It is quite astonishing that Boudon is not referenced even once in the recent Oxford handbook of cognitive sociology (2019), when he has written some of the most significant sociological works in this field with his books on ideology (1986), self-persuasion (1990), and social values (2001). While this Oxford handbook covers several topics such as culture, identity, perception, attention, memory, and so on, this paper strictly focuses on the core subject matter of the sociology of knowledge, namely the explanation of collective beliefs–in both a theoretical and a history of thought perspective. “Knowledge” will be understood here mainly in its proper sense of positive beliefs. Normative judgments will only be mentioned incidentally.

1

The Classical Paradigm in Sociology of Knowledge

The sociology of knowledge traditionally studies the relationship between knowledge and social factors. Merton (1968 [1945]) offers a classic presentation of the theories developed by founding figures such as Marx and Engels (1970 [1846]), Durkheim and Mauss (1901, 1902), Scheler (1926), and Mannheim (1936 [1929]). The approaches of these founders vary, and Merton classifies them according to the answers they give to a series of questions. The first question is: “Where is the existential basis of mental production located? ” Marx and Engels find this basis in the relations of production or class position, Scheler in what he calls “real factors” (impulses of sex, hunger, and power), Mannheim in the relevant group to which the individual belongs (generation, status or occupational group, etc.), and Durkheim and Mauss in social rites, group structure and the spatial configuration of group gatherings (in their study of Australian tribes). Here is the second question: “What mental productions are being sociologically analyzed? ” The range of mental constructs studied is very wide since it encompasses ideologies, opinions, moral and religious beliefs, and social norms. The relevance of sociology will rarely be challenged in the study of ideologies, and especially of the wrong belief systems that promote the class or group interest of those who hold them. One of the main issues in this respect is whether scientific knowledge is within the range of the sociology

Methodological Individualism/Sociology …

633

of knowledge, and if so, how a scientific theory can be both socially determined and objectively valid. Marx and Engels granted to the natural sciences some autonomy from social factors and related in this case the social determination more to the focus of scientific attention than to the conceptual apparatus (solving at the same time the issue of combining social determination and objective validity). Mannheim was on a similar line and exempted the “exact sciences” from existential determination: “… natural science, especially in its quantifiable phases, is largely detachable from the historical-social perspective of the investigator…” (1936 [1929], p. 261). He also considered that, whereas the outlooks of the workers and of the entrepreneurs are bound by their class affiliation, the intellectuals are largely free from any social determination. Intellectuals are “unattached” (p. 140) because they come from all strata of society. They are therefore, according to him, in a position where they can search for a dynamic synthesis of the various conflicting political viewpoints in the interest of the whole society. The third question is the most important because it addresses the issue of the sociological explanation of knowledge: “How are mental productions related to the existential basis? ” The authors generally postulate a link of determination between the existential conditions and thought, but Merton observes that this fundamental problem “has often been treated by implication rather than directly” (p. 530). Marx and Engels presuppose a causal relationship between the social basis and ideas. However, the nature of this causal link is complex, and they do not explain satisfactorily how it operates. Some of their texts suggest that ideas are determined by social class and interests, and that this determination is hidden from the thinkers themselves behind a “false consciousness.” The objective determination is therefore subjectively concealed. Also, if the ideology of the ruling class predominates, then the exploited class will believe in an ideology that goes against its objective interest. Merton then turns to Mannheim and recognizes that he “has not appreciably clarified the connectives of thought and society” (p. 533). As far as Durkheim and Mauss’ analysis is concerned, Merton calls it a “naive theory of correspondence” in which “the categories of thought are held to ‘reflect’ certain features of the group organization” (p. 534). In Scheler’s theory, ideas belong to their own realm, and existential factors operate like some kinds of floodgates that let some ideas flow into society while preventing others from pouring through. Scheler also uses the concept of “structural identity” to denote the common ground between knowledge and the socio-economic sphere, for instance, between the mechanistic thought of the modern era and economic individualism.

634

R. Fillieule

The classical contributions are important and delineate a significant field of study, but Merton’s exegesis makes it clear that they are far from being satisfactory. They seem to suffer from a foundational problem that makes them unable to correctly theorize the relationship between knowledge and social factors. This chapter will show that MI provides a satisfactory solution to this problem by clarifying and illustrating what it means to explain a collective belief.

2

MI as a Framework for the Sociology of Knowledge: An Illustration

Although MI was initially conceived to analyze actions and their combination into social phenomena, it can also be used to analyze beliefs and their prevalence in society. MI is a method that seeks to explain social phenomena starting from rational (in a broad sense) actors in their social context. Now, collective beliefs are social phenomena indeed, as they represent shared and common knowledge (“The Earth is round”). If, in order to explain why people reach these beliefs or why they adhere to them, we start with a rational individual in a given social context, then we can properly say that the individualistic methodology has been applied. Boudon (1986) defends and illustrates this idea in his groundbreaking work on ideology.1 Let us take a simple example that applies MI to the explanation of beliefs. Many people in the general population (and even among the intellectuals) believe that machines cause—or will eventually cause—massive unemployment. Here is the typical reasoning that an average person can make on this topic: . . . .

Machines perform many tasks much faster and better than human labor To be more productive, firms tend to replace labor with machines Workers are fired Since the use of machines is more and more prevalent, job opportunities become increasingly scarce . The unemployed workers face less and less job openings . A massive unemployment ensues.

1

This paper is written from the perspective of the “Boudonian paradigm” (so to speak) in sociology of knowledge, but for a critical assessment of Boudon’s cognitive model see for instance Opp (2014). At a more general level, Sepulvado and Lizardo (2017) offer an overview of cognitive sociology in France and go through the Boudonian program, the naturalistic program, and the linguistic program.

Methodological Individualism/Sociology …

635

This kind of reasoning was initially limited to physically unqualified labor but was then applied to physically qualified labor (that can be replaced by robots), and even in recent years to intellectual labor (that can be replaced by artificial intelligence). The important point here is that an explanation of a widespread belief has been provided. Why do many people believe that machines create unemployment? Because when they think about it, their reasoning leads them to conclude that machines indeed have this effect. Their conclusion is incorrect from the viewpoint of a more comprehensive analysis, but the refutation that machines create unemployment requires a complex multistep argumentation.2 We cannot expect ordinary persons to perform complicated reasoning outside of their own field of expertise: they will stop at the first or at most the second step of the much longer and intricate reasoning required for a valid conclusion. When people conclude that “machines cause unemployment,” their belief is explained by the reasons they find to hold it, even though they are mistaken. More generally, people believe in a general proposition because they find satisfactory (to them) reasons to do so. This explanation, furthermore, is sociological in three respects. First, it explains a social phenomenon, namely, a collective belief related to an important question for society: what causes the major social problem of unemployment? Second, people do not ponder the question of technological unemployment out of the blue. This question will likely arise in specific contexts such as soaring unemployment or the appearance of major labor-saving innovations (such as automation or artificial intelligence). The economic or technological situation, therefore, plays a role in these collective beliefs. In the case of a rising and worrisome unemployment, the belief in the adverse effects of “machines” on jobs can even appear as factually corroborated. Finally, the explanation is sociological in a third respect, which is the experience of the social actors according to their place and role in society. If for instance, in their everyday life, people see with their own eyes that workers are replaced by “machines” (waiters replaced by self-ordering kiosks at fast-food restaurants, cashiers replaced by self-checkouts at stores, and so

2 A classic refutation of the claim that mechanization causes unemployment is offered by Hazlitt (2008 [1946]). He explains that even though machines replace labor and as such can destroy jobs, they also require workers for their production and maintenance (jobs creation). Then, since machines lower the price of the consumption goods they help produce, consumers can spend more money on other goods, which leads to the development of the branches producing these other goods (jobs creation). Furthermore, if the demand for the good benefiting from the labor-saving machines is elastic, then employment will paradoxically increase in this very branch. If the demand is inelastic, then employment will increase in branches producing new types of goods. In this way, machines create roughly the same number of jobs that they destroy, and they ultimately increase the productivity of labor and therefore also the average material standard of living.

636

R. Fillieule

on), they may more easily tend to conclude that these labor-saving devices destroy jobs overall. This illustration shows that MI can be extended from the study of combinations of actions to the study of collective beliefs and is therefore highly relevant to the field of the sociology of knowledge. This methodological framework is now going to help us solve the explanatory problem plaguing the classical paradigm.

3

The Classical Paradigm Through the Lens of MI

The previous illustration can now be used to answer the fundamental questions asked by Merton in the field of the classical sociology of knowledge. The first one raises the issue of the location of the “existential basis of mental production.” In the case of the belief that machines cause unemployment, this basis is easy to locate. First and obviously, this belief can only appear in a capitalist society with salaried labor and a sufficiently high technological level. Second, this belief will be more widespread in the general context of rising unemployment. In this situation, people may legitimately wonder where this unemployment comes from and fall back on the technological unemployment argument. Also, the appearance of major labor-saving innovations can make people worry about the future level of employment. Third, there are positions in society that favor this belief, where people see jobs replaced by automata or where they experience this replacement themselves. The extracognitive factors encompass here the type of society, economic outlook, and social position. The second question asks what types of mental productions are “sociologically analyzed.” The belief in the harmful effects of mechanization on employment is interesting in this regard because it is a good example of an ideological belief . First, it relates to the overall functioning of the economic system, and second, it is erroneous. This belief, furthermore, is not trivial or insignificant. An economic policy based upon the rejection of machines would save some jobs, but at the same time prevent the appearance of a similar number of more productive ones. Such a policy would for sure have very detrimental consequences upon the average material standard of living. This kind of belief is therefore especially suited for an explanation in terms of sociology of knowledge: people can be mistaken on topics that are important for the proper functioning of the system in which they themselves participate.

Methodological Individualism/Sociology …

637

This example also illustrates a clear divergence between scientific knowledge and ideology. According to Merton, the main question for a sociology of knowledge is the third one, namely “How are mental productions related to the existential basis? ” In other words, what is the relationship between thought and society? He finds the answers given to this question by the originators in this field unconvincing, which poses a foundational problem. If Merton does not find any satisfactory answer to his question, it may be because it is badly put. From the viewpoint of MI, the question is not so much to determine a relationship (between social factors and thought) as it is to explain beliefs. This kind of explanation, as we have seen, accounts for collective beliefs through human reasoning , or in other words through human cognition. An actor receives information based on his or her social position, processes it, and reaches a reasoned conclusion. This conclusion is believed because it is based upon facts and logic (upon information and inferences) that are accessible to this actor. Of course, these facts can be wrong and this logic can be flawed from the viewpoint of a more knowledgeable individual (an expert, a scientist, a statistician, etc.). The relevant human reasoning or rationality used by MI is therefore subjective (Boudon, 1989). The belief in the negative consequences of machinery is a good example of this “subjective rationality.” Some people reach this conclusion by observing the disappearance of jobs in their daily life, then generalizing this local observation to the whole system: if jobs disappear here, they must for similar reasons disappear elsewhere also, which explains a high or rising unemployment. This induction from the particular to the general is usually justified. In this case, however, the inference is invalid because of the complex interdependence between the numerous branches of the market economy. The reasoning is objectively wrong because it is too simplistic, but it can nevertheless be subjectively convincing as a rationale for people who are not specialized in economics: the cognitive process explains the belief and the existential basis is in some of its aspects represented in the premises of the inference.

4

The Universality of Human Reason: MI Versus “Polylogism”

In the field of sociology of knowledge, MI heavily relies on human reasoning. A preliminary question must therefore be raised: can it be that reason and logic themselves are socially determined? Mises (1998 [1949]) has coined the term “polylogism” to refer to the idea that human reason and logic are

638

R. Fillieule

not universal, but instead specific to their social context. “Polylogism” asserts that the human mind is compartmentalized between societies, nations, social classes, or races.3 If this was indeed the case, then one of the most important applications for a sociology of knowledge would be to explain how the various forms of human reason are causally related to particular social groups or systems. However, the existence of a multiplicity of human reasons and logics would have momentous implications: this heterogeneity would hamper or prevent altogether the mutual understanding or meaningful communication between groups and societies. In the framework of MI, polylogism and its implications are thoroughly rejected . The most famous attempt at theorizing the concept of polylogism was made by Lévy-Bruhl (1910, 1922). His thesis was that in traditional or “primitive” societies, thought proceeds in a different way than in modern societies.4 According to him, the “primitive” mind differs from the modern mind in two respects. First, the “primitive” mentality is “mystical” in that it attributes natural phenomena to supernatural and anthropomorphic forces. Where we perceive the effects of purely natural forces, the “primitive” sees instead the manifestation of invisible spirits moved by feelings of benevolence or hostility. Practical activities such as hunting, fishing, going to war, etc., could not be successful without the favorable intercession of these supernatural forces. Second, and more importantly, the “primitive mentality” is “pre-logical” in the sense, not that it is devoid of logic, but rather that it follows logical principles that are not only different from, but even incompatible with ours. Notably, it does not apply the non-contradiction principle, but instead what Lévy-Bruhl calls the “law of participation,” according to which a being can be two different things or at different places at the same time. This theory of the “primitive mind” could be interpreted to mean that human reason itself is socially determined and that the principle of the unicity of the human mind should be rejected. However, Lévy-Bruhl himself adopts

3 When Mises invented the term “polylogism” in the 1930’s, he used it to describe a tactic of the totalitarian movements: discrediting their opponents by denouncing their social or ethnic background instead of criticizing their ideas. Marxists claimed that “proletarian” economics was true and “bourgeois” economics false, just by virtue of the fact that the proletariat was allegedly the anointed class. National socialists claimed that “German” science was true and “Jewish” science false, just by virtue of the fact that the “German race” was allegedly the superior one. Mises completely rejected all these forms of polylogism. He simply observed that their supporters were never able to expound in any detail the nature of their supposedly superior system of thinking. Polylogism in these contexts was nothing else but what he called a “revolt against reason” by people unable to deal with rational criticism. 4 For a critical analysis of Lévy-Bruhl’s contribution on this topic, see for instance Horton (1973) and Boudon (1990, pp. 41–43). An in-depth theorization of the similarities and differences between traditional thought and modern science is provided by Horton (1967).

Methodological Individualism/Sociology …

639

a more nuanced interpretation of his own theory. In his later work (LévyBruhl, 1931), he clearly states that the “structure of the mind” is the same throughout all societies. It is the “way of thinking” that can differ between “primitive” and “civilized” societies. He writes that the law of participation is “incomprehensible” to us, but the use of this term is questionable. We can in fact understand the members of a traditional society who claim they are both a human and an animal. The “law of participation” is not outside of the scope of the Weberian comprehension principle. From the viewpoint of our modern knowledge, however, we just do not see how this dual nature is possible in the real world and find such claims to be objectively false. Lévy-Bruhl recognizes that there is a universal substratum in human thinking (the “structure of the mind”), and a variation of contents or inference rules in different societies. He also writes that “primitive” people use the law of contradiction in a number of contexts, and that there are “primitive” elements of thought in modern societies. He acknowledges here that human reason is not divided into two separate and irreconcilable spheres between the two types of societies, but he still clings to the distinction between a “logical” and a “pre-logical” thought. Since the 1970s, cognitive psychologists have analyzed and popularized a great number of cognitive biases (Kahneman et al. 1982). When facing rather simple (maybe deceptively simple) cognitive tasks, most people err and reach a wrong result. One could conclude, à la Lévy-Bruhl, that these are instances of “primitive” or “pre-logic” thinking in modern societies. Boudon (1986, 1990) reviews several of these biases and argues against this polylogist interpretation. Even though people tend to reach mistaken conclusions, the various “heuristics” (availability, representativeness, etc.) that they use are not some kind of “pre-logical” inferences. They are reasonable rules that work in many instances, and it is therefore perfectly understandable that people resort to them in these problem-solving contexts. Their inferences do not fundamentally differ from those used in scientific research. These experiments are carefully crafted by the cognitive psychologists to mislead the subjects into using generally valid inferences that turn out to be inadequate to solve these specific problems. However, these mistakes do not support the thesis that ordinary and scientific inferences belong to two separate logical spheres. While an anthropologist such as Shweder (1977) argues that there is a discontinuity between ordinary (he calls it “magical”) and scientific thought, Boudon concludes from his review of cognitive biases that this type of polylogism is unwarranted. Ordinary thought displays less rigor than scientific thought (as expected) but operates within the same set of inferential rules.

640

R. Fillieule

The foremost proponents of MI are on the same line on this subject matter. Weber expresses the universalism of science in a famous quotation: “It has been and remains true that a systematically correct scientific proof in the social sciences, if it is to achieve its purpose, must be acknowledged as correct even by a Chinese–or–more precisely stated–it must constantly strive to attain this goal, which perhaps may not be completely attainable due to faulty data” (1949 [1904], p. 58). In other words, the validity criterion of scientific knowledge transcends societies and cultures. Popper reminds us of the fact that the different mother tongues can be translated into one another and upholds “the unity of human reason” (1966 [1945], p. 239). Mises speaks of the “uniformity and immutability of the logical structure of the human mind” (1998 [1949] p. 4). Boudon defends the “universality of the rules of functioning of thought” (1990, p. 85). They all fully acknowledge the unicity of human reason. In the framework of MI, human thinking and reason are universal and not variable with culture or social organization. Mises (1998 [1949], p. 84) quite unfairly reduces the “sociology of knowledge” to an “offshoot” of “Marxian polylogism.” This chapter shows that sociology of knowledge does not deserve this disparagement, since it can also be interpreted in terms of MI and freed from any remnants of polylogism.

5

The Explanation of Ideology: MI Versus “False Consciousness”

Ideologies are an important phenomenon in modern societies, and a major topic for the sociology of knowledge. The word “ideology” was coined by Destutt de Tracy (1801) to mean a general science of ideas in the sensationalist tradition of Condillac, and it was popularized half a century later by Marx as a key element of his dialectical materialism: ideologies are part of the “superstructure” of society, and as such determined by the “infrastructure” of material productive relations. One of the core tenets of Marx’s theory is that ideologies are mistaken beliefs about the functioning of the social and economic system. Because of their social class situation, people subscribe to a wrong interpretation of social reality. Furthermore, they are unable to realize that their view is distorted and just a cover-up for the selfish class-interest of the ruling class. Boudon (1986) has written a major study on the topic of ideology in the MI framework. He follows Marx in defining “ideology” as a set of positive beliefs, i.e., beliefs that fall under the criterion of true and false. More recent definitions have chosen instead to conceive ideologies as symbolic actions.

Methodological Individualism/Sociology …

641

Boudon argues that these definitions miss the mark because symbolism encompasses too many elements of social discourse. In the Leninist tradition, ideologies are defined as simple tools, useful in the political struggle to rise to power, but this conception also misses the mark. People subscribe to ideologies because they believe, not only in their usefulness to conquer power, but also and above all in their truthfulness and therefore in their capacity to actually fulfill their promises. The definition of ideologies as positive beliefs seems to be contradicted by the fact that they often incorporate value judgments. Boudon argues, however, that in this context, the divide between value judgments and factual statements is not as insurmountable as it seems. A value judgment cannot be deduced from statements of fact, this is not in dispute. Ideological value judgments can nevertheless heavily rely upon statements of fact. In the heyday of the communist movement, the famous motto (“Workers of the world, unite!”) would not have been quite as convincing without the support of the theory of exploitation that claims as a fact that profit is extracted from the value produced by labor. Ideologies should of course not be confused with economic or social science, but there is an undeniable relationship between them. Classical liberalism is supported by the tradition of Classical and (to some extent) Neoclassical economics, socialism by the Marxist tradition, and conservatism by classical sociology with its emphasis on community, authority, status, and the sacred (as shown by Nisbet 1966). The appearance of ideologies in the modern sense of the word has accompanied the rise of the social sciences and the will to organize society according to science instead of religion or tradition. Boudon also follows Marx in defining ideologies as distorted perceptions of reality. They are not only positive beliefs, but false beliefs about the social order. Where he differs from Marx, on the other hand, is in the suggested explanation of ideological phenomena. Boudon puts forward an explanation of ideologies as comprehensible or rational beliefs (in the Weberian tradition), while Marx develops an irrationalist (or a-rationalist) explanation encapsulated in the concept of “false consciousness.” Marx himself never used that expression, but it adequately summarizes his general theory of ideology. It first appeared in a letter written by Engels to Franz Mehring in 1893: “Ideology is a process accomplished by the so-called thinker consciously, it is true, but with a false consciousness. The real motive forces impelling him remain unknown to him; otherwise it simply would not be an ideological process.” In this theory, people think they adhere to an ideology for reasons they are aware of, but their beliefs are in fact determined by an entirely different causal process. The determination process takes place, not in their

642

R. Fillieule

conscious argumentation, but rather in their “material activity.” The social and economic life in which they are embedded shapes the ideological superstructure. Even though the ideology favors the interests of the ruling class, neither the dominant nor the dominated have any conscious control over it. Marx’s general theory of ideology can be called “irrationalist” because the conscious reasoning of the social actors is just a front for a process that takes place unconsciously in the material social life. Now, Boudon remarks that there is a divergence between this general theory and the case studies offered by Marx. The latter do not exemplify the former. In his famous analysis of “commodity fetichism,” for instance, Marx (1867) observes that people believe that the value (price) of commodities is determined by their exchange against other commodities, when in fact (according to his labor theory of value) this value is determined by the quantity of labor needed to produce them. His explanation for this discrepancy is that sellers are mainly interested in the commodities they will get in exchange. Focused on these exchanges, they erroneously believe that the prices originate in the trade between the commodities themselves. In this explanation of a belief characteristic of the capitalist system, Marx does not resort to a hidden causal process originating in the “material activity.” He notices that people deal with a dual process of production and exchange. They simply make the mistake of focusing on the wrong part of this process, namely, on exchange rather than on production and labor time. This mistake is understandable because—as Marx himself points out—the most important part of this process for the traders lies in its finality, i.e., in obtaining the goods they want through exchange. People therefore “look” at the part of the process that is the most relevant to them (the exchanges), find there a plausible explanation of value (the law of supply and demand), and understandably infer that it is the true explanation. Furthermore, buyers and sellers are not economists familiar with the labor theory of value. In this example, Marx gives a rational explanation that does not match his (irrational) theory of false consciousness.5 Boudon offers other illustrations of this divergence, such as Marx’s explanation of the mercantilist illusion, the support by the French bourgeoisie of the parliamentary monarchy, and so on.

5 We are not interested in this section in knowing whether Marx is right or wrong, but in the type of explanation of beliefs that he provides in his specific illustrations. We should however note that the mistake he reports is in fact not a mistake at all. The traders are right, and Marx is wrong. Exchange value is indeed determined by supply and demand, not by the quantity of incorporated labor, as was demonstrated with the marginalist revolution of the 1870’s. Marx’s mistake, in turn, is not due to any hidden process of “material activity” either. It comes from the inability of Classical economics to solve the paradox of value and therefore having to resort to the (erroneous) labor theory of value.

Methodological Individualism/Sociology …

643

In his own approach, Boudon does not come up with a comprehensive theory but rather with different models explaining various sources of ideological beliefs. He presents three models, each one exemplifying a type of misleading effects: the situational effect, the communicational effect, and the epistemological effect. (i) Commodity fetichism illustrates the situational effect. People hold erroneous beliefs because their perception of social reality is biased due to their specific situation or viewpoint. They “look” at a social phenomenon and understandably believe in what they “see,” but they do not realize that their perspective is too narrow and therefore misleading. The belief that machines increase unemployment also falls into this category. (ii) An example of a communicational effect occurs when people trust an authority figure on a social or economic topic. They are not irrational. They just lack the time and expertise to go through the nuts and bolts of a specialized topic, and therefore have to rely on persons or sources they regard as trustworthy. This effect is reinforced and can easily lead to ideological beliefs when the “authoritative” source manages to bypass the evaluation by the relevant scientific community and conveys the work directly to the public. Boudon takes the example of the famous book by Foucault (1975) on prisons. He shows that one of the main theses of this book (prisons increase crime!) is highly questionable and rejected by the specialists in the relevant field. However, the success and fame of this thesis cannot be denied. This thesis was disseminated by influential social groups concerned with state repression, but why was it considered persuasive? The first explanation is that Foucault has written a book based on high-quality historical material, giving the readership the appearance of an overall scientifically solid book. The educated public is not specialized enough to be able to dissect the thesis and spot its weak points (Boudon calls this the “black box” effect). A situational effect may also have played a role: the neutralization and deterrent impacts of prison are invisible (criminal acts that are never committed), while recidivism is all over the news when an ex-convict commits a gruesome crime. As a result, the public could consider, not only that Foucault’s thesis was plausible at face value (situational effect), but that it was scientifically established (communicational effect). (iii) The third model covers the epistemological effects. They originate in the fact that thought always proceeds within a framework. A framework, in turn, necessarily rests upon several hypotheses, some of which remain implicit. If the implicit hypotheses happen to be inadequate to address a particular problem, then reasoning that is perfectly acceptable and convincing within the framework will lead to mistaken conclusions and beliefs. For instance, speaking of “economics of development” implicitly suggests the dubious idea that “development” takes a similar form across countries and continents.

644

R. Fillieule

These three types of effects can be applied, separately or in combination, to explain why reasonable people are led to believe in questionable or wrong ideas about society, politics, or the economy. A methodologically individualist sociology of knowledge thus very usefully contributes to the explanation of ideology. On the other hand, even though the explanation by a “false consciousness” may, at first sight, seem appealing, it becomes useless as soon as one seeks to explain an actual ideological belief.

6

The Sociology of Scientific Knowledge (1): MI Versus the “Strong Program”

Starting in the early 1970s, some sociologists of knowledge attempted to apply a purely causal (arational) explanation to the knowledge developed by the natural sciences and even by mathematics (Barnes 1974; Bloor 1991 [1976]). The classical sociology of knowledge had put the theories of the natural sciences beyond the scope of the “sociological” explanation. Kuhn (1962) laid the groundwork for a sociological conception of science with his analysis of “normal science” as a consensus-based and routine activity carried out within a scientific community. The “modern” sociology of knowledge went all the way and sought to explain scientific theories themselves by social factors. Bloor called this endeavor the “strong program” in the sociology of knowledge. He summarized it with the four concepts of causality, impartiality, symmetry, and reflexivity. According to him, the sociology of scientific knowledge should be “causal” in the sense that it should study “the conditions which bring about beliefs or states of knowledge.” It should be “impartial with respect to truth and falsity, rationality or irrationality, success or failure.” It should also be “symmetrical” in that the “same types of cause would explain, say, true and false beliefs.” Finally, it should be “reflexive,” i.e., “its patterns of explanation would have to be applicable to sociology itself ” (1991 [1976], p. 7). The most contentious principle, and the only one in fact, between MI and the “strong program” is the first one, namely causality. When Bloor speaks of the “conditions that bring about beliefs,” he means the sociological or psychological conditions. He excludes reasons or rationality as causes of knowledge and only takes social (or psychological) factors into account. His rejection of reasons as causes of the adoption of scientific theories by scientists themselves may at first sight appear astonishing. It becomes understandable once we realize that he conceives sociology as a Durkheimian science that seeks to explain social phenomena (here, scientific beliefs) by showing that they are

Methodological Individualism/Sociology …

645

effects caused by social factors (Durkheim 1895). In other words, he upholds a positivist and holistic conception of sociology as a science that empirically shows the cause-and-effect relationships between objective (observable) social facts. If sociology was a science of this kind, then indeed his “strong program” would be the only suitable way to devise a sociology of knowledge. However, the Durkheimian view of sociology is obviously not the only one and is furthermore highly questionable. This holist methodology stands in blatant contradiction with MI. Surprisingly, Bloor never discusses this methodological issue, as if Durkheimian positivism was the only method prevalent in the discipline. The present handbook is a testimony to the importance and relevance of MI for social science. Examples of convincing theories that follow the precepts of MI abound in all fields of sociology (and of course of economics also). It seems to us indisputable that the MI tradition carried forward by Weber (1978 [1921]), Mises (1998 [1949]), Hayek (1952), and Boudon (1984), to mention only these four major authors, has won the methodological debate against Durkheimian positivism. Only MI can provide genuine explanations of social phenomena, and the subjective (unobservable) rationality of the actors plays a key causal role in these explanations. Social phenomena are combinations of rational actions and collective beliefs appear because a lot of people reach a similar conclusion through typical reasoning. Barnes’ and Bloor’s rejection of the causal role of reasons on human beliefs or knowledge is untenable from the methodological standpoint of an explanatory social science. If the “strong program” was as relevant and useful as Bloor claims it is, he would quickly provide an illustration displaying its strengths and value. Instead, he very briefly reviews examples that do not clearly support his thesis and formulates an attack against the teleological model of epistemologists. As he puts it, the “teleological model” (as against the causal model that he defends) rests upon an asymmetry between true (or rational) beliefs and false (or irrational) beliefs: true beliefs are believed because they are true and do not need any further explanation, while false beliefs are to be explained by sociological or psychological causes. As we have seen, Bloor rejects this asymmetry and chooses to follow a symmetry principle instead: whether true or false, scientific or not, beliefs should be explained within the same causal apparatus. It is notable that on this point, there is an agreement between MI and the “strong program.” MI fully endorses the symmetry principle. It applies the same kind of explanation (“good reasons”) to both true and false beliefs, to both scientific knowledge and ideology, as Boudon’s work

646

R. Fillieule

(1986, 1990) clearly shows.6 Bloor implicitly presupposes that wrong beliefs cannot be explained by reasons. This premise must be rejected. The main criticism that Bloor levels at “epistemologists,” namely, asymmetry, does not hold against MI. As a final comment, let us note that Bloor’s approach is inconsistent. How does he try to convince us of the validity and relevance of his “strong program?” Obviously through a reasoned argumentation. He wants our reason to persuade us that he is right against the epistemologists. If he was consistent, he would ask us to get into some specific “social conditions” or to enter a required psychological state that would lead us to believe his theory through an a-rational process. This is of course absurd, as is the denial that reasons are the causes of beliefs.

7

The Sociology of Scientific Knowledge (2): MI Versus “Constructivism”

By the end of the 1970s, the “constructivist” paradigm appeared in the sociology of scientific knowledge, mainly developed by Latour and Woolgar (1986 [1979]).7 Its major claim is that, in the realm of the natural sciences, scientific facts are entirely “constructed” by the social activity of the scientists. According to this radical view, the scientists are not discovering an external reality by way of observation and rational argumentation. Instead, their interaction with one another and with the scientific devices literally fabricates the scientific facts. This idea originated in an ethnographic study in a biology laboratory where Latour observed for several months the scientific activity that eventually led to an important discovery in the field. All he could see were social and localized processes such as negotiations, conflicts, trust and mistrust, traditions, habits, and sheets of paper coming out of machines (or “inscription devices,” as he calls them). He concluded that there was no external reality pre-existing these interactions between scientists and between scientists and machines. The so-called “thought processes,” “logical 6 On the side of the natural sciences, Laudan (1977) shows how commonsense rationality explains why scientists decide to adopt a theory. Basically, a theory aims at solving problems, some of them empirical and some conceptual. A theory that solves more problems will be chosen over one that solves less problems. But there is also a hierarchy of problems, some being considered more important and some less important. So, even if a theory solves less problems, it can nevertheless be chosen because it solves more important problems. Finally, a theory is not an isolated entity but part of a “research tradition.” A theory that solves fewer or less important problems can still be chosen by scientists because it is part of a promising and fast-progressing research tradition. All these criteria are perfectly understandable in the Weberian sense of a comprehensive sociology. 7 For a brief and synthetic statement, see Woolgar (1988).

Methodological Individualism/Sociology …

647

reasoning,” and “intellectual factors” were just post hoc rationalizations by the scientists, when in fact there were only social (a-rational) processes. Latour and Woolgar adopt here an anti-realist metaphysical stance. Is it the role of sociologists to take a stand on this subject matter? The “social construction of reality” paradigm is a venerable tradition in sociology going back to Schutz (1967 [1932]). Two of its main exponents, Berger and Luckmann, explicitly refused to endorse a philosophy of reality. Right from the introduction of their book, they make clear that questions such as “What is reality?” and “What is knowledge?” are intended for the philosopher, not the sociologist: “we immediately disclaim any pretension to the effect that sociology has an answer to these ancient philosophical preoccupations” (1991 [1966] p. 13). Latour and Woolgar, on the other hand, not only claim to know what factual scientific reality is and where it comes from, but also to know it better than the scientists and philosophers themselves. This is peak sociological hubris. Is it justified? The anti-realism of the “constructivists” is in fact highly questionable. Latour and Woolgar tell us that proof that scientific facts are constructed is that they can be deconstructed (1986 [1979], p. 179). But when a scientific fact is deconstructed, it implies that it had not been constructed at all in the first place. If the scientists, through their interactions with their devices, had not constructed the fact, then what had they done? This question remains unanswered. In the realist framework, when a fact is “deconstructed,” it just means that an empirical hypothesis turns out to be false, i.e., not in correspondence with the pre-existing reality. In the constructivist framework, the scientific activity of construction, deconstruction, reconstruction, and so on, appears as a purely arbitrary and senseless social process. Popper, in his classic defense of what he calls “metaphysical realism,” puts it very simply: “Rational discussion, that is, critical argument in the interest of getting nearer to the truth, would be pointless without an objective reality, a world which we make it our task to discover” (1983 [1956], p. 81). Since Latour and Woolgar deny the very existence of an “objective reality” to be uncovered, there is indeed no place for “rational discussion” in their paradigm. Just like Bloor (to whom they repeatedly refer), they want to get rid of rationality that they consider antithetical to social processes. The previous section has already made clear that their rejection of rationality stands in complete opposition to MI. An additional and relevant methodological point can be raised. Latour and Woolgar claim that they apply the anthropological method, which according to them consists in making “the activities of the laboratory seem as strange as possible.” They want to create an artificial unfamiliarity “in order not to take too much for

648

R. Fillieule

granted” (1986 [1979], p. 30). However, anthropologists do not apply this method at all. They do not seek to make strange what is familiar, but on the contrary to make familiar what at first sight appears to them quite strange in traditional societies. A good example is the anthropological debate about the nature of religious rituals in traditional societies. Beattie (1966) and the “symbolist” school claim that these rituals are what art is in our society, i.e., a way of expressing meaning and values. Horton (1962, 1967) and the “intellectualist” school, on the other hand, claim that these rituals aim at understanding and controlling nature, and are therefore similar to what we call technology. In both cases, anthropologists are trying to erase the strangeness of the activity and make it understandable for their Western readers. If Latour and Woolgar applied the anthropological method (uncovering a hidden familiarity), they would compare laboratory activity to a recognizable activity where people seek the truth through observation and critical argumentation, such as investigative journalism for instance. An anthropologist such as Horton applies MI and develops the tradition of a Weberian comprehensive social science, while Latour and Woolgar essentially seek to explain why the lab scientists do not know what they are really doing. MI and constructivism are irreconcilable, with understanding and humility on one side, estrangement and hubris on the other.

8

Conclusion: From Sociology of Knowledge to Cognitive Sociology

The classical sociology of knowledge rests upon a dual system of explanation: explanation by reasons for scientific or otherwise validated knowledge, and explanation by social factors for ideological and wrong or questionable beliefs. Merton thought that all the approaches to the sociology of knowledge shared a “central point of agreement,” namely, that “thought has an existential basis in so far as it is not immanently determined and in so far as one or another of its aspects can be derived from extra-cognitive factors” (1968 [1945], p. 516). An epistemologist such as Laudan is exactly on the same page as Merton, as he writes that “the sociology of knowledge may step in to explain beliefs if and only if those beliefs cannot be explained in terms of their rational merits” (1977, p. 202). But what kinds of beliefs are without “rational merits?” He gives the example of the belief in socialism by nineteenth-century German workers. He claims that it would take “an enormous degree of credulity” to explain their adoption of the socialist doctrine by its “rational well-foundedness” (p. 200), but it is on the contrary completely believable. A typical German worker

Methodological Individualism/Sociology …

649

could observe that he was doing all the hard work at the factory, while the capitalists got huge profits without lifting a finger; socialism promised to put an end to this exploitation by confiscating the means of production from the freeloaders and giving the proceeds back to the genuine producers, namely the workers. Contrary to Laudan’s assertion, believing in socialism would be rational indeed for such a typical worker. It would be all the more rational that by then the refutation of Marx’s exploitation theory by Böhm-Bawerk (1959 [1883]) was only known in very specialized circles, the disastrous socialist experiments of the twentieth century had not yet taken place, and the main scientific criticism leveled against collectivism—the impossibility of economic calculation—had not yet been formulated by Mises (1935 [1920]). This example shows that believing in a wrong doctrine and being located at a certain place or time in society does not shut down rationality, but instead illustrates how subjective rationality proceeds in this specific context. Despite what Merton and Laudan think, the dual (rational vs. social) explanation of beliefs in the classical sociology of knowledge is untenable because in most cases knowledge is the result of a cognitive (immanent) process by an actor who is socially situated (extracognitive factors). The modern sociology of knowledge focused on scientific knowledge and tried to solve this problem by going all in with social factors, rejecting rationality altogether. The “strong program” (Bloor 1991 [1976]), however, was a failure. It rests upon an antiquated conception of social science (holism) and is selfrefuting (it uses reasons to convince us that reasons are not causes of beliefs). “Constructivism” (Latour and Woolgar 1986 [1979]) suffers from the same defects and adds to them an unwarranted metaphysical conceit. In the mid-1980s, Boudon (1986) initiated the cognitive turn in sociology of knowledge, followed by a number of his students (Bouvier 1995, 2007, 2012; Bronner 2003, 2007, 2011; Demeulenaere 2003, 2020). This “cognitive sociology” rejects the asymmetrical explanations of the classical paradigm on the one hand, and the exclusively a-rational explanations of the modern paradigm on the other. Instead, it adopts a broad conception of rationality to account for ideological beliefs as well as scientific knowledge. It focuses on the cognitive processes of the individual and lets social factors play a role through the position and disposition of the actor. At a general level, cognitive sociology changes the way the problem of the sociology of knowledge is framed. The sociology of knowledge traditionally asks what the “relationship” between social factors and knowledge or thought is. Cognitive sociology asks a very different question: starting from the fact that there are collective beliefs, the problem is just to explain them. Cognitive sociology is part of MI, not only because it starts with the individual (in a social context), but

650

R. Fillieule

also because its aim is essentially explanatory. In this framework, the problem cannot be to define or analyze a “relationship” between phenomena, but quite simply to provide satisfactory explanations of interesting social phenomena. Now, the only way to explain why people believe in specific (collective or personal) ideas is to determine their own reasons to adopt them. A cold and narrow economic interest can explain why some people pretend to believe in ideas that suit their instrumental purposes, but it cannot explain a genuine belief in these ideas: a genuine belief requires reasons that are convincing from the viewpoint of the actors themselves. Some ideas are held passionately or even fanatically, but this strong affective element cannot explain a belief either because it is a consequence of a belief system, not its cause. If for instance one’s knowledge system leads to conclude that the current social order is deeply unjust, then it may understandably generate feelings of anger and revolt. It appears on the other hand quite unlikely (but not impossible, of course) that someone angry and revolted will look for an ideological or religious system that will justify ex post these feelings. Table 1 summarizes in broad strokes a good deal of the topics that have been discussed above. Starting with the explanation of ideology by social factors (bottom-right box), the sociology of knowledge moved to the arational explanation of scientific knowledge (bottom-left), then shifted upwards with the cognitive turn (top line, and especially the top-right box that is highly relevant, as Boudon’s contribution has shown). Finally, it can be noted that the field of sociology of knowledge is nowadays somewhat neglected. There is not one scientific journal under this name. The Cambridge handbook of sociology (2017) does not devote a single chapter to Table 1 Synthetic view of the field of sociology of knowledge: types of explanation × types of knowledge Types of knowledge

Types of explanation

Rational: cognition A-rational: social factors

True or validated (ex.: science)

False or questionable (ex.: ideology)

Epistemology Authors: Popper, Laudan, etc Modern sociology of knowledge Authors: Bloor, Latour, etc

Multi-discipline Authors: Boudon, Horton, etc Classical sociology of knowledge Authors: Marx, Mannheim, etc

Methodological Individualism/Sociology …

651

this topic,8 the Cambridge dictionary of sociology (2006) offers no entry for “sociology of knowledge” or even for “beliefs,” and the Oxford handbook of cognitive sociology (2019) is to a large extent off topic as far as sociology of knowledge in the strict sense is concerned.9 The reason for this neglect might quite simply be that MI and rationality turned out to be indispensable in this field and are not very popular among sociologists overall, but this situation is nonetheless unfortunate because the explanation of collective beliefs is an important and perennial issue for social science.

References Barnes, B. (1974). Scientific knowledge and sociological theory. London: Routledge and Kegan Paul. Beattie, J. (1966). Ritual and social change. Man, 1, 60–74. Berger, P. L., and Luckmann, T. (1991 [1966]). The social construction of reality: A treatise in the sociology of knowledge. London: Penguin Books. Bloor, D. (1991 [1976]). Knowledge and social imagery. 2nd ed. Chicago: The University of Chicago Press. Böhm-Bawerk, E. von (1959 [1883]). Capital and interest, vol. 1: History and critique of interest theories. South Holland: Libertarian Press. Boudon, R. (1984). La place du désordre. Critique des théories du changement social . Paris: Presses Universitaires de France. Translated as: Theories of social change: A critical appraisal . Cambridge: Polity Press, 1986. Boudon, R. (1986). L’idéologie ou l’origine des idées reçues. Paris: Fayard. Translated as: The analysis of ideology. Cambridge: Polity Press, 1989. Boudon, R. (1989). Subjective rationality and the explanation of social behavior. Rationality and Society, 1(2), 173–196. Boudon, R. (1990). L’art de se persuader des idées douteuses, fragiles ou fausses. Paris: Fayard. Translated as: The art of self-persuasion: The social explanation of false beliefs. Cambridge: Polity Press, 1994. Boudon, R. (1997). L’explication cognitiviste des croyances collectives. In R. Boudon, A. Bouvier and F. Chazel (Eds.), Cognition et sciences sociales. La dimension cognitive dans l’analyse sociologique (pp. 19–54). Paris: Presses Universitaires de France.

8

There are 42 chapters in the volume on “Specialty and Interdisciplinary Studies,” with for instance, a chapter on “Visual sociology” and one on “Sociology of emotions,” but none on the sociology of knowledge. 9 Mannheim is only quoted in the first chapter. Boudon is not mentioned once in a book that is almost 800 pages long. Also, there is no in-depth discussion of rationality, except for a few scattered references to “economic rationality” and “bounded rationality.”

652

R. Fillieule

Boudon, R. (2001). The origin of values: Sociology and philosophy of beliefs. London: Transaction Publishers. Bouvier, A. (1995). L’argumentation philosophique. Étude de sociologie cognitive. Paris: Presses Universitaires de France. Bouvier, A. (2007). An argumentativist point of view in cognitive sociology. European Journal of Social Theory, 10 (3), 465–480. Bouvier, A. (2012). Pareto, Mill and the explanation of collective beliefs: Unnoticed “middle-range theories” in the Trattato. In J. Femia and A. Marshall (Eds.), Beyond disciplinary boundaries: Essays on Pareto (pp. 131–151). Farnham: Ashgate. Brekhus, W. H. & Ignatow, G. (eds). (2019). The Oxford handbook of cognitive sociology. Oxford: Oxford University Press. Bronner, G. (2003). L’empire des croyances. Paris: Presses Universitaires de France. Bronner, G. (2007). L’empire de l’erreur. Éléments de sociologie cognitive. Paris: Presses Universitaires de France. Bronner, G. (2011). The future of collective beliefs. Oxford: Bardwell Press. Demeulenaere, P. (2003). Les normes sociales. Entre accords et désaccords. Paris: Presses Universitaires de France. Demeulenaere, P. (2020). L’individualisme méthodologique et les représentations collectives. L’Année sociologique, 70 (1), 69–95. Destutt de Tracy, A. (1801). Projet d’éléments d’idéologie. À l’usage des Écoles centrales de la République française. Paris: Didot. Durkheim, É. (1895). Les règles de la méthode sociologique. Paris: Alcan. Translated as: The rules of sociological method . Chicago: University of Chicago Press, 1938. Durkheim, É., and Mauss, M. (1901–1902). De quelques formes primitives de classification. Contribution à l’étude des représentations collectives. L’Année Sociologique, 6 , 1–72. Foucault, M. (1975). Surveiller et punir. Naissance de la prison. Paris: Gallimard. Translated as: Discipline and punish: The birth of the prison. New York: Random House, 1977. Hayek, F. A. (1952). The counter-revolution of science: Studies on the abuse of reason. Glencoe: The Free Press. Hazlitt, H. (2008 [1946]). Economics in one lesson. Auburn: The Ludwig von Mises Institute. Horton, R. (1962). The Kalabari world-view: An outline and interpretation. Africa, 32, 197–220. Horton, R. (1967). African traditional thought and Western science. Africa, 38, 50–71 and 155–185. Horton, R. (1973). Lévy-Bruhl, Durkheim and the scientific revolution. In R. Finnegan and R. Horton (Eds.), Modes of thought: Essays on thinking in Western and non-Western societies (pp. 249–305). London: Faber and Faber. Kahneman, D., Slovic, P., and Tversky, A., Eds. (1982). Judgment under uncertainty: Heuristics and biases. Cambridge: Cambridge University Press. Kuhn, T. (1962). The structure of scientific revolutions. Chicago: University of Chicago Press.

Methodological Individualism/Sociology …

653

Latour, B., and Woolgar, S. (1986 [1979]). Laboratory life: The construction of scientific facts. 2nd ed. Princeton: Princeton University Press. Laudan, L. (1977). Progress and its problems: Towards a theory of scientific growth. Berkeley: University of California Press. Lévi-Strauss, C. (1962). Le totémisme aujourd’hui. Paris: Presses Universitaires de France. Translated as: Totemism. London: Merlin Press, 1964. Lévy-Bruhl, L. (1910). Les fonctions mentales dans les sociétés inférieures. Paris: Félix Alcan. Translated as: How natives think. London: George Allen & Unwin, 1926. Lévy-Bruhl, L. (1922). La mentalité primitive. Paris: Félix Alcan. Translated as: Primitive mentality. London: George Allen & Unwin, 1923. Lévy-Bruhl, L. (1931). La mentalité primitive: The Herbert Spencer lecture. Oxford: Clarendon Press. Mannheim, K. (1936 [1929]). Ideology and utopia: An introduction to the sociology of knowledge. London: Routledge and Kegan Paul. Marx, K. (1867). Das Kapital. Kritik der Politischen Oekonomie. Hamburg: Otto Meissner. Marx, K., and Engels, F. (1970 [1846]). The German ideology. New York: International Publishers. Merton, R. K. (1968 [1945]). The sociology of knowledge. In R. K. Merton, Social theory and social structure (pp. 510–542). New York: The Free Press. Mises, L. von (1935 [1920]). Economic calculation in the socialist commonwealth. In F. A. Hayek (Ed.), Collectivist economic planning: Critical studies on the possibilities of socialism (pp. 87–130). London: George Routledge & Sons. Mises, L. von (1998 [1949]). Human action: A treatise on economics. Auburn: The Ludwig von Mises Institute. Nisbet, R. A. (1966). The sociological tradition. New York: Basic Books. Odell Korgen, K. (ed.) (2017). The Cambridge handbook of sociology, vol. 2: Specialty and interdisciplinary studies. Cambridge: Cambridge University Press. Opp, K.-D. (2014). The explanation of everything. A critical assessment of Raymond Boudon’s theory explaining descriptive and normative beliefs, attitudes, preferences and behavior. Papers. Revista de Sociologia, 99 (4), 481–514. Popper, K. (1966 [1945]). The open society and its enemies, vol. 2: The high tide of prophecy: Hegel, Marx, and the aftermath. London: Routledge & Kegan Paul. Popper, K. (1983 [1956]). Realism and the aim of science. London: Hutchinson. Scheler, M. (1926). Die Wissenformen und die Gesellschaft. Leipzig: Der Neue-Geist Verlag. Schutz, A. (1967 [1932]). The phenomenology of the social world . Evanston: Northwestern University Press. Sepulvado, B., and Lizardo, O. (2017). Cognitive sociology in France. The American Sociologist, 48(3/4), 366–381. Shweder, R. A. (1977). Likeliness and likelihood in everyday thought: Magical thinking in judgments about personality. Current Anthropology, 18(4), 637–659. Turner, B. S., Ed. (2006). The Cambridge dictionary of sociology. Cambridge: Cambridge University Press.

654

R. Fillieule

Weber, M. (1949 [1904]). “Objectivity” in social science and social policy. In M. Weber, On the methodology of the social sciences (pp. 50–112). Glencoe: The Free Press. Weber, M. (1978 [1921]). Economy and society: An outline of interpretive sociology. Berkeley: University of California Press. Woolgar, S. (1988). Science: The very idea. London: Tavistock Publications.

Methodological Individualism in Behavioral Economics Malte Dold

1

Introduction: Methodological Individualism in Traditional Microeconomics

In microeconomics, methodological individualism is typically understood as the claim that ‘macro’ level phenomena—such as market demand or supply—must ultimately be explained on the ‘micro’ level in terms of the characteristics and choices of individual actors involved (Arrow 1994). Put differently, social scientific explanations of group-level phenomena must be built up from the actions of individuals (Bowles 2004, p. 478). Typically, in microeconomics those actions are further specified by an action-theoretic mechanism, i.e., a model of individual choice (Alexander 1987). There is a long tradition in economics that understands the actiontheoretic mechanism as intentional individual action and social phenomena as the unintended consequence of the interaction of intentionally acting individuals (Menger [1871] 1985; Hayek [1942–44] 1979). In this tradition, methodological individualism means that economists “shall not be satisfied with any type of explanation of social phenomena which does not lead us ultimately to a human plan.” (Lachmann 1969, p. 94). In microeconomics of the mid-twentieth century and onwards (‘traditional microeconomics’), the M. Dold (B) Economics Department, Pomona College, Claremont, USA e-mail: [email protected]

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 N. Bulle and F. Di Iorio (eds.), The Palgrave Handbook of Methodological Individualism, https://doi.org/10.1007/978-3-031-41512-8_29

655

656

M. Dold

idea of intentionality is dropped and the action-theoretic mechanism qualified by the assumptions of rationality and self-interest (Sen 1987). Thus, traditional microeconomics asserts that social phenomena can be explained purely in terms of the characteristics of rational, self-interested agents, the actions available to them, and the constraints that they face (Gintis 2009, p. xiv). In doing so, the commitment to methodological individualism entails a restriction to what arguments microeconomists focus on when explaining social phenomena—viz., individual choice explanations—and which actiontheoretic mechanism they employ when modeling individual choice—viz., the homo economicus model. This ‘double commitment’ forms the heart of methodological individualism in traditional microeconomics and leads to the following three principles: (i) Self-interest principle: objectives are assumed to be self-interested and located at the level of the individual economic actor. (ii) Rationality principle: the individual actor behaves rationally given her objectives and beliefs; she always chooses the best feasible option given the constraints she faces. (iii) Social change principle: changes in social phenomena are caused by individual actors’ responses to changes in constraints. Principles (i) and (ii) form the action-theoretic mechanism of traditional microeconomics and principle (iii) ties back discussions of changes in social phenomena to the level of individual choice. Principle (iii) is central to explanations in traditional microeconomics: preferences are assumed to be stable so any change in behavior is explained by changes in the ‘external’ incentive structure, such as income or relative prices (Stigler and Becker 1977, p. 76).1 Admittedly, the ‘double commitment’ to individual choice explanations and the specific action-theoretic mechanism of the homo economicus model is a ‘thick’ description of methodological individualism. Yet, it is justified in the context of traditional microeconomics. While the homo economicus model is often qualified in microeconomics today (Angner 2019), before the ascend of behavioral economics in the last quarter of the twentieth century, the assumptions of self-interest and rationality were considered part and parcel of the methodological ‘hard core’ of the field (Sen 1987). For instance, Stigler

1

Becker (1976, p. 5) defends the assumption of preference stability; it “provides a stable foundation for generating predictions about responses to various changes, and prevents the analyst from succumbing to the temptation of simply postulating the required shift in preferences to ‘explain’ all apparent contradictions to his predictions.”

Methodological Individualism in Behavioral Economics

657

(1981) defends the homo economicus model as the action-theoretic mechanism of microeconomics by arguing that “we live in a world of reasonably well-informed people acting intelligently in pursuit of their self-interests” (p. 190) and when self-interest and ethical values are in conflict “[m]uch of the time, most of the time in fact, the self-interest theory … will win” (p. 176). A description of markets is an illustration of how traditional microeconomists apply the three principles. On both sides of the market, there are a number of individual actors who are assumed to be rational and selfinterested: consumers maximize their utility and producers maximize their profit. Individual consumers choose consumption demands and offer labor, individual producers choose inputs and outputs. These choices depend on many factors, such tastes, attitudes toward risk, beliefs about the future, and constraints. Yet, importantly, these factors are located at the level of the individual consumer and producer (Arrow 1994). They typically exclude any reference to social characteristics and social relationships. Choices on both sides of the market are explained in terms of individual economic actors having their own separate beliefs and objectives (Davis 2009). The sum of individual consumer and producer choices gives then rise to market demand and supply which in turn defines the market price.2 If exogenous shocks occur (e.g., a pandemic or a natural catastrophe), market participants face a new set of incentives (e.g., certain goods and services are now needed more, others less). The market price for these goods and services will adjust as a result of individual responses to those new incentives. Hence, whatever happens in a market on an aggregate-social level is ultimately described as the consequence of a series of rational individual reactions to changes in constraints. Schumpeter (1954, p. 888) sums up this individualist framework: “the selfgoverning individual constitutes the ultimate unit … all social phenomena resolve themselves into decisions and actions of individuals that need not or cannot be further analyzed in terms of superindividual factors.”3 This means that aggregative analysis that makes use of social entities—such as firms or corporations—are regarded as only provisionally legitimate and ultimately be treated as the resultants of rational choices of individuals (Brennan and Tullock 1982, p. 225). Personifications of the firm or the corporation are only interim theoretical constructs, which can be broken down into a series of 2 Strictly speaking, each individual takes the price as given at the moment of choice and the market price that comes to prevail is an outcome of the choices of all individuals (Arrow 1994). 3 Schumpeter used the term ‘sociological individualism’ to describe what many people today describe as ‘methodological individualism’ (Hodgson 2007, p. 213).

658

M. Dold

rational choices of individuals who form those institutions, such as managers, shareholders, workers, etc. There are at least two methodological reasons for adopting this individualist approach in microeconomics (Heath 2020). One, the individualist perspective goes beyond the observation of statistical correlations between economic variables (e.g., between levels of market concentration and market prices) and seeks to deliver a causal explanation for social phenomena (e.g., prices are higher in concentrated markets due to the price-setting power of the monopolist that is in turn a result of the inelastic demand of consumers). And, two, the individualist perspective cautions against the “postulate [of ] a purpose without a purposive actor” (Elster 1982, p. 454). In other words, it cautions against the identification of a group interest without analyzing what sort of objectives and constraints govern its individual constituents, e.g., the notion of ‘class interest’ must ultimately be derived from the analysis of the interests and constraints of individual class members. If this individualist perspective is ignored, theorists might overlook intricate collective action problems and falsely predict the power of groups with dispersed interests.

2

A Very Brief History of Behavioral Economics

In the second half of the twentieth century, an increasing number of scholars were motivated by dissatisfaction with the unrealistic assumptions of microeconomics and a conviction to put forth descriptively more accurate models of individual choice using insights from cognitive psychology (Sent 2004). The most prominent spokesman of this ‘old school’ of behavioral economics was Herbert Simon as illustrated by his prolific work on bounded rationality (Simon 1955, 1956, 1959). According to Simon (1955, p. 99), the economist’s “task is to replace the global rationality of economic man with a kind of rational behavior that is compatible with the access to information and the computational capacities that are actually possessed by organisms […] in the kinds of environments in which such organisms exist.” Simon (1956, p. 29) argues further that this insight leads to the acknowledgment of satisficing behavior: individuals “[fall] far short of the ideal of ‘maximizing’ postulated in economic theory. Evidently, [they] adapt well enough to ‘satisfice’; they do not, in general, ‘optimize’.” However, the ‘old’ behavioral economics of Simon never fully crossed over into the economic mainstream, in part because of its explicit effort to radically depart from basic assumptions of rationality in microeconomics, and

Methodological Individualism in Behavioral Economics

659

in part due to its lack of rigorous experimental evidence (Simon 1991, p. 385). Moreover, Friedman’s (1953) ‘as-if ’ defense of microeconomics’ unrealistic assumptions successfully circumvented the introduction of psychology into economics during the mid-twentieth century. According to Friedman, competition within markets ensures that economic actors on both sides of the market act ‘as-if ’ they followed the maxims of rational, self-interested choice. Therefore, economists would not need to consider the psychological complexity of human decision-making when analyzing markets. Furthermore, Friedman argued that psychological realism is not a methodological value as such. According to Friedman, theories and their underlying assumptions should be judged in terms of their predictive power rather than their proximity to reality. Friedman’s defense strategy came under increasing attack in the second half of the twentieth century. In part due to a lack of predictive power of economic theory and in part because a new group of behavioral economists convincingly argued that competitive pressure on markets do not easily eliminate market actors’ system 1 thinking, i.e., their fast and intuitive way of making decisions (Kahneman 2003, 2011). While being cognitively frugal, system 1 thinking often leads people to make biased choices. Biases persist on the individual level and influence market outcomes because there are often clear limits to individuals’ learning opportunities, e.g., one-off decisions or repeated decisions that lack clear feedback make it difficult for individuals to correct their cognitive biases even in competitive market environments (Mullainathan and Thaler 2000). Also, many market participants—mostly on the supply side—have an incentive not to correct the other market side’s system 1 thinking since it is more cost-effective to make money from people’s biases than to educate them (Russell and Thaler 1985). For instance, a company might profit more from exploiting people’s existing biases that leads them to consume sugary drinks, fatty food, or nicotine products in great quantities than by first educating the consumers about the long-term consequences of those consumption habits and then to market an alternative product. Taken together, proponents of this ‘new’ strand in behavioral economics argue that individual psychology matters for the explanation and prediction of market outcomes. This has been a powerful critique of Friedman’s aforementioned as-if assumption. Compared to Simon et al., proponents of ‘new’ behavioral economics have been far more successful in establishing themselves as a field within economics. In 2002, the Nobel Prize in Economics was awarded jointly to Daniel Kahneman and Vernon Smith; Richard Thaler was awarded the Prize in 2017. The reasons for the success of ‘new’ behavioral economics are

660

M. Dold

manifold and this chapter does not aim to do justice to all of them.4 One important aspect that contributes to the ongoing success of ‘new’ behavioral economics is that its proponents do not try to fundamentally challenge the methodological framework of microeconomics, including its commitment to methodological individualism. As Laibson and List (2015, p. 385) point out “behavioral economics is a series of amendments to, not a rejection of, traditional economics.” The notion of bounded rationality in ‘new’ behavioral economics upholds the focus on individual choice and the idea that people try to optimize, i.e., they try to choose the best feasible option. Yet, they sometimes fail to achieve their intended goals due to systematic and predictable cognitive biases. In doing so, new behavioral economics takes the rationality assumption of traditional microeconomics as a benchmark from which to consider behavioral deviations. Kahneman (2003, p. 1449f ) clarifies this point: “The rational-agent model was our starting point and the main source of our null hypotheses. […] Theories in behavioral economics have generally retained the basic architecture of the rational model, adding assumptions about cognitive limitations designed to account for specific anomalies.” While Simon started from the conviction that traditional microeconomists “were not all that serious” (Sent 2004, p. 750), proponents of ‘new’ behavioral economics do not attempt to fundamentally change microeconomics’ methodological framework, e.g., ‘new’ behavioral economics upholds traditional microeconomics’ reliance on maximizing behavior, albeit under budgetary and cognitive constraints. Another factor that has contributed to the success of ‘new’ behavioral economics is its proponents’ reliance on lab and field experimental evidence that supports their theoretical claims (Angner and Loewenstein 2012). Importantly, both lab and field data focus on individual choice. This illustrates ‘new’ behavioral economics’ commitment to methodological individualism not just on a conceptual-theoretical but also on an empirical-practical level. The idea behind this reliance on individual choice data is that controlled lab conditions enable economists to study individual behavior in situations that, in simplified and pure forms, mimic those encountered in markets and other forms of economic interaction. In a nutshell, ‘new’ behavioral economics by and large follows the individualist methodology of standard microeconomics in that it aims to explain social phenomena by understanding and modeling individual choice.

4

See Heukelom (2014) for a comprehensive discussion of the history and success of ‘new’ behavioral economics.

Methodological Individualism in Behavioral Economics

3

661

Methodological Individualism in Behavioral Economics

Methodological individualism is rarely explicitly defended or precisely defined in behavioral economics. None of the introductory textbooks discusses methodological individualism (e.g., Angner 2020; Cartwright 2018; Wilkinson and Klaes 2017) and only a handful of articles have recently been published on the topic (e.g., Davis 2015; Frerichs 2019; Lecouteux 2022; Ross 2022). Despite this omission, an implicit advocacy of methodological individualism is as widespread in behavioral economics as it is in traditional microeconomics: “Recent developments in behavioral economics have involved some retreat from rationality, but the individualism of classical and neoclassical economics has been retained. The idea that explanations that are based on individual-level assumptions are superior to, or more satisfying than, those that are not, is deeply rooted in the practice of economics.” (Sugden 2016, p. 1379). Whitman (2022, p. 456) puts it succinctly: “behavioral economics … clearly work(s) within the individualist tradition.” Like traditional microeconomists, behavioral economists study and model the manner in which individual economic actors make choices under conditions of scarcity and the results of those choices for social outcomes, particular markets. Hence, like traditional microeconomics, behavioral economics aims to explain social events in terms of the preferences, beliefs, and constraints of individuals. However, based on a plethora of experimental findings about people’s actual behavior, behavioral economists qualify the three principles mentioned in Sect. 1 that lie at the heart of traditional microeconomics’ commitment to methodological individualism: the self-interest principle, the rationality principle, and the social change principle. This section argues that while the relaxation of the self-interest and the rationality principles confirm behavioral economists’ commitment to methodological individualism, the relaxation of the social change principle pushes the boundaries of individualist methodology as it highlights the bidirectional relationship between social institutions and individual choice.

Relaxation of the Self-interest Principle There are a number of experimental findings about people’s actual behavior in game theoretical settings that standard microeconomies cannot easily explain, e.g., why people give a substantial amount of their endowment to others in dictator games (Camerer 2011, Chap. 2). This has led behavioral economists to relax the self-interest principle of traditional microeconomics

662

M. Dold

and assume bounded selfishness: people do not only care about their own welfare but also about those of relevant others. Behavioral economists typically model bounded selfishness by introducing social preferences (such as altruism, inequity aversion, or reciprocal fairness) into individuals’ objective functions. The literature on dictator and ultimatum games initiated the research on social preferences (Güth and Kocher 2014). In dictator games, an anonymous sender and an anonymous recipient are paired. Senders decide how much of their endowment (typically $10) they want to give to the recipients. Recipients are passive in this game. Traditional microeconomics’ self-interest principle predicts that senders keep $10 for themselves. However, in experiments, senders typically propose a division in which the recipient receives between $2 and $4 (Camerer 2011, p. 56). The assumption of social preferences (e.g., inequity aversion or altruism) on behalf of the players helps make sense of the observed findings. A similar pattern can be observed in ultimatum games, in which the recipient is not passive but can either accept or reject the division. In the event of rejection, both players receive nothing. Here, likely driven by a mix of pro-social and strategic concerns (senders anticipate that many receivers would reject offers that are perceived as too low and thus unfair), the recipient receives slightly higher offers of $4–$5 compared to the dictator game (Camerer 2011, p. 49).5 Social preferences also help explain cooperation in n-player public good games in which each player is given an initial endowment (e.g., $10) and each player has the option of transferring a part of their endowment to a public account where it is multiplied by some factor, typically between 1 and 3. Contrary to the prediction of mutual defection based on the self-interest and rationality principles of standard microeconomics, experimental studies find that corporation is remarkably robust and subjects contribute on average 40–60% of their endowment to the public good in a one-shot game, even when varying the monetary stakes or group size (Dawes and Thaler 1988). Social preferences identified on the individual level can help explain and predict market-level phenomena, e.g., social preferences help explain the attenuated effect of competition on wages (Fehr and Fischbacher 2002) and the counterproductive effect of incentive-based regulation on pro-social behavior (Bowles and Polanía-Reyes 2012). Crucially, for the discussion of methodological individualism, social preferences identified in the lab play a 5 Like standard microeconomics, behavioral economics assumes that incentives still matter for individuals’ social preferences. When experimenters exogenously increase the stakes, recipients become more willing to accept unfair offers (Andersen et al. 2011).

Methodological Individualism in Behavioral Economics

663

significant role for explanations of outcomes in natural markets. For instance, sellers who are more pro-social in a lab setting are more successful in natural markets: they achieve higher prices, have better trade relations, and better abilities to signal trustworthiness to buyers (Leibbrandt 2012). Furthermore, in a study with fishermen whose main source of income stems from the use of fishing grounds with open access, those who exhibit social preferences in the laboratory are also more collaborative outside the lab and less likely to exploit the common pool resource (Fehr and Leibbrandt 2011). These studies indicate that social preferences play a crucial role in economic decisions with lasting consequences for the people involved. Moreover, they are an important contributing factor to the solution of real-world social dilemmas. There is a rich empirical literature that shows that people in real-life settings are not helplessly trapped in social dilemma situations since, among other things, individuals hold self-interested and other-regarding preferences (Ostrom 2010, p. 648). A more realistic understanding of the interaction between self-interested and other-regarding concerns allows economists to calibrate agent-based models which yield better predictions of the chances of success of self-organization and bottom-up solutions in social dilemma situations (Poteete et al. 2010). Accordingly, behavioral economists follow the rationale of methodological individualism which stipulates that macro phenomena (e.g., social cooperation) should ultimately be explained by micro models that refer to the properties and predispositions of individuals (e.g., objective functions that incorporate social preferences).

Relaxation of the Rationality Principle Experimental evidence and empirical observation suggest that people sometimes make choices that do not align with their interests (Hausman and McPherson 2009). Behavioral economists identify bounded willpower and decision biases as main sources for this observed divergence between actions and interests (Mullainathan and Thaler 2000). Hence, they relax the rationality principle of traditional microeconomics and model economic actors as boundedly rational. Bounded willpower captures the phenomenon that people often make plans that entail actions with upfront costs (e.g., to diet or to work out) for a later reward (e.g., to be healthy or good-looking). But when it comes to the instantiation of their plans, they are overwhelmed by instant gratification and decide to forgo the larger later reward for a smaller-sooner one (e.g., eat the burger or watch the Netflix show). A famous study in this context is Read and van Leeuwen (1998) who show that snacks chosen one week in advance

664

M. Dold

are typically healthy, but snacks chosen for immediate consumption are not. This systematic pattern, called present bias, leads to time-inconsistent preferences and violates standard rationality axioms. Behavioral economists explain present bias by means of (quasi)hyperbolic discounting which captures the observation that people are relatively patient for options that are further removed in the future, but discount heavily between the current and the next period (Laibson 1997; O’Donoghue and Rabin 1999).6 Besides bounded willpower, decision heuristics can drive another wedge between people’s goals and actions. Heuristics are often useful since they reduce time and cognitive effort by applying simpler judgmental operations, particularly in complex environments. Yet, research in behavioral economics suggests that those heuristics can be biased and lead to systematic errors of judgment (Kahneman 2003). For instance, when making judgments of the likelihood of an event people often give too much attention to salience (availability heuristic), similarity (representativeness heuristic), or arbitrary anchors (anchoring-and-adjustment heuristic). Such behavioral observations of systematic departures from the rationality principle of traditional microeconomics motivated prospect theory (Kahneman and Tversky 1979). A core insight of prospect theory is that people make decisions based on how circumstances compare to a reference point (e.g., their current endowment, a subjective income target, or the income of a peer group). Outcomes are then coded in terms of losses or gains relative to this reference point. Experimental evidence suggests that people are prone to loss aversion: they suffer from losses about twice as much as they benefit from gains of the equal magnitude which can explain the endowment effect (Kahneman et al. 1990).7 This means that people’s value function is steeper for the domain of losses than gains. Also, the gain and loss parts of the value function display diminishing sensitivity: people value the first incremental changes from the reference point more than changes of equal magnitude that are further removed from the reference point. A third feature of prospect theory is the insight that individuals overweight small probabilities and underweight medium to high probabilities.

6

This is different from the standard assumption of exponential discounting in traditional microeconomics in which the discount factor is independent of the time horizon. 7 A famous example in this context is the coffee mug experiment (Kahneman et al. 1990). Following a random allocation, half of the subjects get a mug and half of the subjects a chocolate bar of approximately the same monetary value. When trade is allowed between coffee mug-subjects and chocolate bar-subjects, fewer than one quarter of subjects will take up this offer. Yet, traditional microeconomics would predict that half of them should trade.

Methodological Individualism in Behavioral Economics

665

The literature on hyperbolic discounting and prospect theory are examples of the ways in which behavioral economists uphold the individualist methodology of traditional microeconomics while tweaking its action-theoretic mechanism. Both are descriptive theories that help explain choice data that deviate from the rationality principle of standard expected utility theory. Crucially, these deviations are systematic and can help explain marketlevel phenomena.8 For instance, the model of hyperbolic discounting can explain the connection between innovations in financial products and the ongoing decline in US savings rates by stressing that these innovations increased liquidity and eliminated commitment opportunities for hyperbolic consumers (Laibson 1997). Moreover, hyperbolic discounting can make sense of the finding that many consumers subscribe to long-term contracts (e.g., for a gym or a weekly magazine) but overpay given their actual consumption (DellaVigna and Malmendier 2006). Hyperbolic discounting explains this finding by stipulating that people are overconfident in their future selfcontrol, both in terms of their usage of a good or service and the cancellation probability of automatically renewing contracts. Prospect theory can also explain a series of market-level phenomena.9 For instance, it can make sense of the finding that people buy simultaneously lottery tickets and insurances. Both are small probability events whose relative frequencies of occurring are overestimated (Kahneman and Tversky 1979, p. 263). And loss aversion can clarify the consumer behavior of mental accounting where consumers pool gains and losses differently to maximize their experienced value of a series of events (Thaler 1985). Loss aversion can also help explain why economists observe less trade in many real-world markets than standard theory would predict. For instance, many investors are less willing to sell a stock that has decreased in value than one that has increased in value (Odean 1998). Due to loss aversion, many investors are reluctant to realize capital losses because it would mean that they have to ‘declare’ a loss. Yet, wealth maximizing behavior would require the opposite since the ‘loser stocks’ underperform relative to the ‘gainer stocks.’ These examples illustrate that the main difference between traditional microeconomics and some of the most famous models and theories in behavioral economics does not lie in different understandings of methodological individualism. Instead, the main difference lies in the belief of behavioral economists that the modification of the action-theoretic mechanism of traditional microeconomics leads to a more adequate descriptive 8

For an overview of the applicability of prospect theory to real-world markets, see Barberis (2013). For more comprehensive discussions of ways in which prospect theory explains social phenomena outside the lab, see Camerer (2000) and Barberis (2013). 9

666

M. Dold

theory of individual choice that can explain a series of real-world market-level phenomena that traditional microeconomics couldn’t explain. In doing so, many behavioral economists are committed to an individualist methodology, both in terms of data collection and in terms of model building. Yet, the next section argues that a more recent strand in behavioral economics has started pushing the boundaries of methodological individualism as understood in traditional microeconomics.

Relaxation of the Social Change Principle Traditional microeconomics explains changes in social phenomena by referring to changes in the constraints individuals face. This social change principle of traditional microeconomics is based on the assumption that institutions, norms, and rules do not directly cause changes in individuals’ preferences or beliefs; rather, they affect behavior only indirectly via changes in the set of feasible alternatives (Stigler and Becker 1977). In contrast, the aforementioned heuristics-and-biases program suggests that preferences are often context-dependent. For instance, it matters for people’s risk preferences whether options are framed in terms of potential gains or losses (Kahneman and Tversky 1986) or for the willingness to pay for various consumption items whether people have been exposed to arbitrary anchors or not (Ariely et al. 2003). In doing so, the heuristics-and-biases program expands the social change principle and includes the idea that individuals respond not just to economic incentives but also to non-economic factors, such as informational framing and situational cues. The heuristics-and-biases program is still close to traditional microeconomics in that it models individuals as actors who are constituted of a rational ‘inner self ’ with stable and consistent preferences; yet this actor is assumed to be trapped in a psychological ‘shell’ (Sugden 2018, pp. 53–76). The idea of this approach is to model individuals as ‘faulty econs’: individuals aim to instantiate their well-ordered preferences, but their decision-making process is distorted by psychological biases which cause them to make systematic mistakes (Heukelom 2014, pp. 172–180). Context-dependent preferences are understood as situational deviations from rationality when intuitive system 1 thinking makes people pay too much attention to normatively-irrelevant factors in the decision environment (such as cues or anchors). Yet if people were given enough time to activate their deliberate system 2 thinking,

Methodological Individualism in Behavioral Economics

667

they would likely see through those irrelevant contextual factors and make decisions that realizes their ‘true’ preferences (Hoff and Stiglitz 2016, p. 28).10 In recent years, a strand in behavioral economics has emerged that does not only acknowledge situational framing effects and preference instability but suggests that individual preferences are shaped on a deeper and more durable level by sociocultural factors (Fehr and Hoff 2011; Bowles 2016; Hoff and Stiglitz 2016; Hargreaves Hep 2020). Hoff and Stiglitz (2016) call this strand two behavioral economics to contrast it with the heuristicsand-biases approach. Strand two argues that institutions and the wider sociocultural environment are not only regulative (in the sense that they affect relative prices and constraints) but they are also constitutive (in the sense that they affect people’s preference development): “prolonged (and sometimes even brief ) exposure to a given social context shapes who people are.” (Hoff and Stiglitz 2016, p. 26). Hence, strand two behavioral economics has contributed to a further relaxation of the social change principle in behavioral economics by conceptualizing people’s perception and preferences as endogenous to the sociocultural context they have been exposed to.11 Strand two behavioral economics is motivated by experimental findings that reveal striking cross-cultural differences in outcomes of the games subjects typically play in economics, such as the dictator game, the ultimatum game, or the public goods game (Henrich et al. 2001; Henrich et al. 2004). For instance, in ultimatum games subjects from industrialized societies tend to divide the money more equally and reject low offers, while subjects from non-industrialized societies are neither inclined to make equal offers nor punish those who make low offers (Henrich et al. 2010a). These results suggest that preferences of individuals to be fair in anonymous transactions (and to punish unfairness) increase with the level of the society’s market integration. Of course, economists cannot unambiguously conclude from these observed correlations that a market-based society (with its distinct institutions, rules, and norms) causally shapes people’s social preferences. It is plausible that market integration has shaped preferences for fairness, e.g., since the quid pro quo rationale is salient in market exchanges and has thus supported the development of fairness preferences. Yet, it is also plausible that the casual link runs the other way: stronger fairness preferences have fostered market integration by honoring informal contractual obligations.

10

For critical discussions of the assumption of ‘true’ preferences, see Dold (2018) and Sugden (2018). This sociologically enriched version of behavioral economics is supported by insights from evolutionary psychology (Heyes 2018) and anthropology (Henrich 2016, 2020). A core insight is that the sociocultural environment physically rewires people’s brains and thereby shapes how they think and what they want. 11

668

M. Dold

Strand two behavioral economics has recently begun to disentangle this endogeneity problem and tried to gain a better understanding of when and how the sociocultural environment affects preferences, by studying, among other things, the effects of social conflicts, laws and regulations, as well as cultural and professional identities on people’s preference development. A series of experimental studies suggest that social conflicts between groups and societies can have considerable impact on people’s time, risk, and social preferences. Voors et al. (2012) conducted a field experiment in rural Burundi. They find that subjects who happen to live in regions with higher levels of violence have higher discount rates, are more riskseeking, and display more altruistic behavior toward their neighbors. The study suggests that adverse shocks (such as violent conflicts) can significantly alter people’s saving and investment decisions that persist even beyond the temporary shocks. In another study, Gneezy and Fessler (2012) find experimental evidence that wartime can have a deep effect on social preferences and within-group cooperation. The authors elicited people’s preferences nine months before, during, and a year after the 2006 Israel–Hezbollah war. They find that during wartime, people are more willing to pay costs to sanction greedy behavior and reciprocate generous acts by group members. The study confirms the hypothesis of people’s preference for in-group favoritism and suggests that inter-group conflict increases it. Furthermore, strand two behavioral economics suggests that laws and regulations may not just affect people’s behavior via changes in constraints but more directly by impacting their preference development. In a lab experiment, Falk et al. (2006) studied the effect of minimum wages on subjects’ reservation wages. The study finds that the temporary introduction of a minimum wage leads to a rise in subjects’ reservation wages. Crucially, this effect persists after the minimum wage has been removed. In other words, subjects are less willing to work for a given wage after the experience of a minimum wage. The study suggests that minimum wage laws may affect people’s behavior by shaping their preferences for fair remuneration by creating entitlement effects. In a famous field study, Gneezy and Rustichini (2000) analyze the effects of a change in the regulatory regime of daycare centers in Israel. At six centers in Haifa, a fine was introduced for parents who were late in picking up their children. Contrary to the prediction of traditional microeconomic theory, parents respond to the fine by doubling the fraction of time they arrive late. Similar to the Falk et al. (2006) study, the effect persists after the regulatory intervention (i.e., the fine) was revoked. In other words, the impact of incentives on preferences exceeds situational framing effects; they constitute part of a learning environment in which

Methodological Individualism in Behavioral Economics

669

preference change can become durable.12 Bowles (2016, p. 5) hypothesizes that the introduction of economic incentives negatively affects the parents’ social preferences by reframing lateness as another commodity they could purchase. In signaling what is considered normatively legitimate, changes in laws and regulations can directly affect people’s preferences—a phenomena that is described as the ‘expressive function of law’ (Sunstein 1996).13 Finally, experimental evidence suggests that people internalize norms and values of the social or professional identities they are exposed to and those norms and values can be activated by situational cues. In a study of employees of a large international bank, Cohn et al. (2014) find that subjects depict are higher willingness to cheat when their professional identity as bank employees is made salient; yet, they behave on average honestly in a control condition without the priming of their professional identity. Interestingly, this effect is specific to bank employees: in control experiments with students and employees from other industries, subjects do not become more dishonest when their professional identity or bank-related items are made salient. The results suggest that the prevailing business culture in the banking industry weakens social preferences. In another set of priming experiments with Asian–American subjects, LeBoeuf et al. (2010) and Benjamin et al. (2010) study the hypothesis that people have multiple identities, and that making one identity more salient than others would evoke different, sometimes conflicting preferences. Both studies corroborate the hypothesis. LeBoeuf et al. (2010) find that priming the subjects by making them think of their Asian heritage impacts their social preferences and leads to a significantly lower defection rate in prisoner’s dilemma games. Benjamin et al. (2010) find that the Asian prime impacts subjects’ time preferences (they become more patient). These findings suggest that people learn to associate different professional and cultural identifies with different preference rankings. Consequently, “social institutions do not just impose constraints and shape beliefs about others behavior but are also preference elicitation devices, frames and anchors that may render particular identities, and thus particular values and normative commitments, more salient.” (Fehr and Hoff 2011, p. 404). This brief review of studies illustrates some ways in which strand two behavioral economics highlights ‘deep social determinants’ of preferences. 12

According to Bowles (1998, p. 80), this durable preference change is a core aspect of preference endogeneity and differentiates it from mere preference instability: “preferences learned under one set of circumstances become generalized reasons for behavior. Thus, economic institutions may induce specific behaviors—self-regarding, opportunistic, or cooperative, say—which then become part of the behavioral repertoire of the individual.” 13 The effectiveness of default rules (e.g., in the context of retirement savings, insurance plans, and organ donation) is sometimes explained along similar lines, see Fehr and Hoff (2011).

670

M. Dold

Among those determinants are institutions, norms, rules, and social practices that provide people with cultural mental models (Hoff and Stiglitz 2016, p. 36). Those mental models “shape the way we attend to, interpret, remember, and respond emotionally to the information we encounter and possess” (DiMaggio 1997, 274). Mental models affect people’s preference formation since they are the way people categorize the world around them, including products, people, political parties, lifestyles, etc. (Dold and Lewis 2022). Individuals are not defined by a single mental model. Instead, they hold multiple models that their minds can draw upon to interpret a given situation. The selection of a model in any given situation happens largely unconsciously and “is guided by cultural cues available in the environment” (DiMaggio 1997, 275).14 Also, the repertoire of mental models is not static but develops in the process of individuals’ exposure to different institutions, norms, rules, and social practices (Frerichs 2019, pp. 11–12). Consequently, the wider sociocultural environment and the context at the moment of choice constantly interact. While the wider environment provides a pool of mental models that individuals are exposed to, the situational context primes and activates concrete mental models.

4

Bounded Individualism and Why It Matters

When behavioral economists qualify the action-theoretic mechanism of traditional microeconomics in the heuristics-and-biases approach, they are still committed to methodological individualism. In fact, relaxing the self-interest principle and the rationality principle has been motivated by the idea to ground explanations of social outcomes (e.g., the observation of consumer behavior on markets) in psychologically more realistic theories of individual behavior (e.g., prospect theory or hyperbolic discounting). The assumptions of self-interest and rationality are not necessary conditions for a methodologically individualist perspective. As Arrow (1994, p. 4) points out “[t]he individualist viewpoint is in principle compatible with bounded rationality, with violations of the rationality axioms, and with the biases in judgment characteristic of human beings. The additional step to rational choice is, of course, of the greatest practical importance to theory formation, but it is not in principle necessary for the individualist viewpoint.”

14

For instance, when primed of their professional identify, bankers do not choose the lens through which to analyze the choice options; rather their active mental model is the outcome of an unconscious reaction to environmental cues (Hoff and Stiglitz 2016, p. 39).

Methodological Individualism in Behavioral Economics

671

In contrast, the relaxation of the social change principle and the discussion of endogenous preferences illustrate that a more recent strand in behavioral economics supports a ‘bounded’ understanding of methodological individualism. It is bounded since it does not solely emphasize that social aggregates need to be built up from individual choices but also acknowledges the influence of social aggregates—such as institutions, norms, and rules— on individuals’ mental models and their preferences.15 In contrast to the heuristics-and-biases program which is based on the idea of stable cognitive biases that are shared across individuals and societies, strand two behavioral economics stresses that mental models vary within and between societies and they can change over time. Individuals’ sociocultural environment may alter the process by which people come to acquire mental models and hence influence individuals’ preferences over time. Social institutions, norms, and rules structure interactions both by providing a pool of mental models and by creating situations in which certain mental models are repeatedly primed and activated. The more a certain model is activated—e.g., a model of rivalry or competition—the higher the chance that this model becomes a general trait of the individual across domains. Bowles (2016, p. 117) points out that “[the] way in which we come to have our particular preferences is much like the way in which we come to have our particular accents. The process takes place early in life, is for the most part unwitting, and depends critically on our social interactions with others.”16 These insights of strand two behavioral economics have implications for several issues in economics. The remainder of this chapter highlights two of them: how economists may need to be more nuanced when interpreting experimental results and when doing comparative institutional analysis. Interpretation of experimental results: The heuristics-and-biases program typically interprets behavioral patterns revealed by subjects in lab experiments as outcomes of invariant cognitive properties and ‘biases’ of individual agents (Frerichs 2019). In contrast, strand two behavioral economics provides a different perspective. Behavioral patterns revealed in the lab can inform the theorist not so much about invariant cognitive properties or ‘biases’ but about the structure of social institutions, norms, and rules the individual has been exposed to (Hargreaves Heap 2022; Lecouteux 2022). For 15

In doing so, strand two behavioral economics shifts the boundaries of methodological individualism as understood in traditional microeconomics, but not necessarily the boundaries of methodological individualism in general. For instance, in sociology in the tradition of Max Weber methodological individualism typically assumes that individuals’ preferences and beliefs are context-dependent, i.e., influenced by the material, social, and cultural environment. 16 For a discussion of historical, social survey, and ethnographic data supporting this view, see Bowles (1998).

672

M. Dold

instance, experimental economists have claimed that their results suggest that poor consumers are more present-biased than their richer counterparts (Ashraf et al. 2006; Tanaka et al. 2010) and women are more risk averse than men (Croson and Gneezy 2009). This perspective concentrates economists’ attention on individual psychology, i.e., the cognitive faculties of individuals are the dominant explanation of their behavior. In contrast, strand two behavioral economics emphasizes social factors that have led to the observed differences in time and risk preferences. Impulsive behavior is seen as the result of structural poverty and peer group behavior that cause the internalization of a mental model of short-term ‘scarcity thinking’ (De Bruijn and Antonides 2022). And differences in risk preferences between women and men are not interpreted as the result of ‘natural’ differences in cognition but of repeated exposure to social narratives that propagate risk-seeking behavior for boys and risk aversion for girls (Lecouteux 2022). This shift in explanatory emphasis away from individual cognition to social factors matters both for scientific explanations and for policy discourse. If it is true that many observed preferences are “invariant only over a particular society or a particular era, or even over a particular social or professional group within a society” (Simon 1990, p. 16), economists must be cautious with scientific generalizations of their behavioral models (Henrich et al. 2010b). Moreover, a policy discourse based on the assumption of invariant cognitive properties likely leads to different policies than emphasizing the ‘heterogeneity’ and ‘plasticity’ of people (Frerichs 2019). For instance, if individual cognition is identified as the source of present bias, then policy discussions will likely focus on ‘micro-interventions’ meant to help individual consumers to overcome their ‘biases,’ e.g., by means of default rules in saving contracts. However, if structural factors are identified as the cause of short-term thinking, policy discussions will likely be about institutional reforms, e.g., by directly targeting socioeconomic imbalances through poverty alleviation or educational campaigns. Comparative institutional analysis: Typically, economists take preferences as antecedent to institutional analysis and understand institutions as instrumental, regulative devices that constrain individual behavior (Arrow 1994). Yet, if strand two behavioral economics is correct, institutions are also constitutive and preferences are endogenous to different institutional arrangements. For instance, experimental evidence suggests that social preferences change when the same exchange is organized in a market as compared to a nonmarket setting (Bowles 2016). In this case, economists cannot explain the emergence of a particular set of institutions by referring to the exogenous

Methodological Individualism in Behavioral Economics

673

preferences of the parties to the exchange as it is typically done in neoinstitutionalism (McCloskey 2022). And for prescriptive purposes, economists cannot simply rely on preference satisfaction as the standard of welfare to recommend one institutional arrangement over another because preferences may change across institutions (Hargreaves Heap 2020). Consequently, discussions of the relative merits of different institutions need to go beyond efficiency considerations: a choice of institutions becomes a choice of sets of preferences individuals will likely develop. What is thus required is a discussion of which preferences and values a society wants to propagate, i.e., what constitutes a “good society” (Hoff and Stiglitz 2016, p. 51). For instance, when economists discuss the advantages of competition in many social contexts (e.g., by promoting the idea of meritocracy), they cannot solely refer to positive incentive effects of selection contests (Sen 2000). Instead, they also need to consider how the implementation of the meritocracy principle in various areas of social life would change the actors involved, e.g., by inculcating habits of rivalry and resentment (Sandel 2020; Dold and Gewecke forthcoming). Frank Knight made this point 100 years ago in his famous essay The Ethics of Competition (1923, p. 586): “An examination of the ethics of [an] economic system must consider the question of the kind of wants which it tends to generate or nourish as well as its treatment of wants as they exist at any given time.”17 Strand two behavioral economics can also illuminate a discussion of the quality of the process through which preferences are formed under different institutional arrangements. For instance, it can open a discussion of whether individuals have access to and are exposed to social situations that allow them to reason about which preferences to hold (Dold and Lewis 2023). In this sense, strand two behavioral economics may be “less concerned with forms of preference satisfaction and more concerned with individual autonomy.” (Hargeaves Heap 2013, p. 985). A focus on autonomy can help disentangle dysfunctional preference-norm equilibria that economists traditionally have a hard time analyzing, such as the Indian caste system (Hoff et al. 2011) or racial segregation (Lang and Kahn-Lang Spitzer 2020). These are examples of shared cultural mental models that people are not born with but form during their upbringings (Kinzler and Spelke 2011). An autonomyoriented perspective would for instance ask whether, in forming those cultural mental models, people were exposed to ‘experiments in living’ that allowed 17

Knight (1923, p. 587) further explicates: “the issue as to the influence of the economic system on character … should at least be raised. Emphasis will be placed on the particular phase of competitive emulation as a motive and of success in a contest as an ethical value. The competitive economic order must be partly responsible for making emulation and rivalry the outstanding quality in the character of the Western peoples who have adopted and developed it.”

674

M. Dold

them to reflect on which preferences to hold. In this regard, autonomy can be understood as an individual’s degree of cognitive independence from a specific situational or socio-cultural context (Whitman 2022); it means that “the cause of human behavior is more inside the individual than outside it.” (Di Iorio 2013, p. 153). Crucially, this autonomy-oriented perspective highlights that strand two behavioral economics does not subscribe to the idea of social determinism: environmental influences are not direct and mechanical, but always mediated by the interpretive skills of the individual. In summary, the analytical focus of strand two behavioral economics shifts away from the decisions of isolated individuals with given preferences to the preference-shaping power of the context in which decisions are made. In doing so, strand two behavioral economics acknowledges the bidirectional relationship between social institutions and individual choice. Individual preferences and group-level institutions are conceptualized as coevolving (Bowels 2004, Chaps. 11–13). People’s preferences and actions contribute to the emergence and stability of institutional arrangements and hence must be seen as the product of individual action. For instance, the distribution of mental models (e.g., perceptions of gender or racial differences) in a population at any given point in time, influences the emergence of social equilibria in various areas of social life and stabilizes certain norms (e.g., gender and racial equality). Yet, those social equilibria also exert causal influence on individuals’ processes of preference development by creating priming effects and repeatedly exposing people to certain mental models. In this sense, institutions, norms, and rules emerge from individual choices and individual choices are in turn shaped by institutions, norms, and rules. Ultimately, the concrete phenomenon at hand might inform economists about which direction of causality they need to focus on. The traditional approach that explains behavior change in terms of changes in incentives might be helpful in standard market analyses with shorter time horizons. Here, the assumption that preferences are unaffected by changes in constraints might be well-justified. Yet, upholding this assumption stringently might prevent economists from a deeper understanding of many social issues, including dysfunctional preference-norm equilibria and market phenomena with longer time horizons.

Methodological Individualism in Behavioral Economics

675

References Alexander, J. (1987). The micro-macro link. Berkeley: University of California Press. Andersen, S., Ertaç, S., Gneezy, U., Hoffman, M., & List, J. A. (2011). Stakes matter in ultimatum games. American Economic Review, 101(7), 3427–39. Angner, E. (2019). We’re all behavioral economists now. Journal of Economic Methodology, 26 (3), 195–207. Angner, E. (2020). A course in behavioral economics, 3d . London: Macmillan. Angner, E., & Loewenstein, G. (2012). Behavioral economics. In U. Mäki (Ed.), Philosophy of economics (pp. 641–690). Amsterdam: North-Holland. Ariely, D., Loewenstein, G., & Prelec, D. (2003). “Coherent arbitrariness”: Stable demand curves without stable preferences. The Quarterly Journal of Economics, 118(1), 73–106. Arrow, K. J. (1994). Methodological individualism and social knowledge. The American Economic Review, 84 (2), 1–9. Ashraf, N., Karlan, D., & Yin, W. (2006). Tying Odysseus to the mast: Evidence from a commitment savings product in the Philippines. The Quarterly Journal of Economics, 121(2), 635–672. Barberis, N. C. (2013). Thirty years of prospect theory in economics: A review and assessment. Journal of Economic Perspectives, 27 (1), 173–96. Becker G. S. (1976). The economic approach to human behavior. Chicago: University Press. Benjamin, D. J., Choi, J. J., & Strickland, A. J. (2010). Social identity and preferences. American Economic Review, 100 (4), 1913–28. Bowles, S. (1998). Endogenous preferences: The cultural consequences of markets and other economic institutions. Journal of Economic Literature, 36 (1), 75–111. Bowles, S. (2004). Microeconomics: behavior, institutions, and evolution. Princeton: Princeton University Press. Bowles, S. (2016). The moral economy: Why good incentives are no substitute for good citizens. Yale: Yale University Press. Bowles, S., & Polania-Reyes, S. (2012). Economic incentives and social preferences: substitutes or complements? Journal of Economic Literature, 50(2), 368–425. Brennan, G., & Tullock, G. (1982). An economic theory of military tactics: Methodological individualism at war. Journal of Economic Behavior & Organization, 3(2–3), 225–242. Camerer, C. F. (2000). Prospect theory in the wild: Evidence from the field. In D. Kahneman and A. Tversky (Eds.), Choices, Values and Frames (pp. 288–300). Cambridge: Cambridge University Press. Camerer, C. F. (2011). Behavioral game theory: Experiments in strategic interaction. Princeton: Princeton University Press. Cartwright, E. (2018). Behavioral economics, 3d . Abingdon: Routledge. Cohn, A., Fehr, E., & Maréchal, M. A. (2014). Business culture and dishonesty in the banking industry. Nature, 516 (7529), 86–89.

676

M. Dold

Croson, R., & Gneezy, U. (2009). Gender differences in preferences. Journal of Economic literature, 47 (2), 448–74. Davis, J. B. (2009). Individualism. In J. Peil and I. van Staveren (Eds.), Handbook of economics and ethics (pp. 261–266). Cheltenham, UK: Edward Elgar Publishing. Davis, J. B. (2015). Bounded rationality and bounded individuality. In A Research Annual (Vol. 33, pp. 75–93). Emerald Group Publishing Limited. Dawes, R. M., & Thaler, R. H. (1988). Anomalies: cooperation. Journal of Economic Perspectives, 2(3), 187–197. De Bruijn, E. J., & Antonides, G. (2022). Poverty and economic decision making: a review of scarcity theory. Theory and Decision, 92(1), 5–37. DellaVigna, S., & Malmendier, U. (2006). Paying not to go to the gym. American Economic Review, 96 (3), 694–719. Di Iorio, F. (2013). Cognitive Autonomy and Epistemology of Action in Hayek’s and Merleau-Ponty’s Thought. In R. Frantz and R. Leeson (Eds.), Hayek and behavioral economics (pp. 149–176). New York: Palgrave Macmillan. Dold, M. F. (2018). Back to Buchanan? Explorations of welfare and subjectivism in behavioral economics. Journal of Economic Methodology, 25 (2), 160–178. Dold, M., & Gewecke, A. (forthcoming). The Way we Change: Meritocracy in its Social Context. In: Mathis, K. and Tor, A. (eds.). Law and Economics of Justice: Efficiency, Reciprocity, Meritocracy. Springer. Dold, M., & Lewis, P. (2022). FA Hayek on the political economy of endogenous preferences: An historical overview and contemporary assessment. Journal of Economic Behavior & Organization, 196 , 104–119. Dold, M., & Lewis, P. (2023). A Neglected Topos in Behavioural Normative Economics: The Opportunity and Process Aspect of Freedom. Behavioural Public Policy, 1–11. DiMaggio, P. (1997). Culture and cognition. Annual Review of Sociology, 23(1), 263–287. Elster, J. (1982). Marxism, functionalism and game theory. Theory and Society, 11(4), 453–82. Falk, A., Fehr, E., & Zehnder, C. (2006). Fairness perceptions and reservation wages—the behavioral effects of minimum wage laws. The Quarterly Journal of Economics, 121(4), 1347–1381. Fehr, E., & Fischbacher, U. (2002). Why social preferences matter–the impact of non-selfish motives on competition, cooperation and incentives. The Economic Journal , 112(478), 1–33. Fehr, E., & Leibbrandt, A. (2011). A field study on cooperativeness and impatience in the tragedy of the commons. Journal of Public Economics, 95 (9–10), 1144– 1155. Frerichs, S. (2019). Bounded sociality: behavioral economists’ truncated understanding of the social and its implications for politics. Journal of Economic Methodology, 26 (3), 243–258. Friedman, M. (1953). The methodology of positive economics. In: Essays in positive economics (pp. 3–43). Chicago: University of Chicago Press.

Methodological Individualism in Behavioral Economics

677

Gintis, H. (2009). The bounds of reason: Game theory and the unification of the behavioral sciences. Princeton: Princeton University Press. Gneezy, U., & Rustichini, A. (2000). A fine is a price. The journal of legal studies, 29 (1), 1–17. Gneezy, A., & Fessler, D. M. (2012). Conflict, sticks and carrots: war increases prosocial punishments and rewards. Proceedings of the Royal Society B: Biological Sciences, 279 (1727), 219–223. Güth, W., & Kocher, M. G. (2014). More than thirty years of ultimatum bargaining experiments: Motives, variations, and a survey of the recent literature. Journal of Economic Behavior & Organization, 108, 396–409. Hayek, F. A. ([1942–44] 1979). “Scientism and the Study of Society.” In The counter-revolution of science: studies on the abuse of reason (pp. 17–182). Indianapolis: Liberty. Hargreaves Heap, S. P. (2013). What is the meaning of behavioural economics? Cambridge Journal of Economics, 37 (5), 985–1000. Hargreaves Heap, S. P. (2020). Two accounts of the relation between political economy and economics (and why it matters which account is better). Social Philosophy and Policy, 37 (1), 103–117. Hargreaves Heap, S. P. (2022). Laboratory experiments and Austrian economics. In D.J. D’Amico and A.G. Martin (Eds.), Contemporary methods and Austrian economics (pp. 55–67). Bingley: Emerald Publishing Limited. Hausman, D. M., & McPherson, M. S. (2009). Preference satisfaction and welfare economics. Economics & Philosophy, 25 (1), 1–25. Heath, J. (2020). Methodological individualism, The Stanford Encyclopedia of Philosophy (Summer 2020 Edition), Edward N. Zalta (ed.), https://plato.stanford.edu/ archives/sum2020/entries/methodological-individualism Henrich, J. (2016). The secret of our success: How culture is driving human evolution, domesticating our species, and making us smarter. Princeton: Princeton University Press. Henrich, J. (2020). The WEIRDest people in the world: How the West became psychologically peculiar and particularly prosperous. New York: Macmillan. Henrich, J., Boyd, R., Bowles, S., Camerer, C., Fehr, E., Gintis, H., & McElreath, R. (2001). In search of homo economicus: behavioral experiments in 15 smallscale societies. American Economic Review, 91(2), 73–78. Henrich, J., Boyd, R., Bowles, S., Camerer, C., Fehr, E., & Gintis, H. (Eds.). (2004). Foundations of human sociality: Economic experiments and ethnographic evidence from fifteen small-scale societies. Oxford: Oxford University Press. Henrich, J., Ensminger, J., McElreath, R., Barr, A., Barrett, C., Bolyanatz, A., & Ziker, J. (2010a). Markets, religion, community size, and the evolution of fairness and punishment. Science, 327 (5972), 1480–1484. Henrich, J., Heine, S. J., & Norenzayan, A. (2010b). The weirdest people in the world? Behavioral and Brain Sciences, 33(2–3), 61–83. Heukelom, F. (2014). Behavioral economics: A history. Cambridge: Cambridge University Pres

678

M. Dold

Heyes, C. (2018). Cognitive gadgets: The cultural evolution of thinking. Cambridge: Harvard University Press. Hodgson, G. M. (2007). Meanings of methodological individualism. Journal of Economic Methodology, 14 (2), 211–226. Fehr, E., & Hoff, K. (2011). Introduction: Tastes, castes and culture: The influence of society on preferences. The Economic Journal , 121(556), 396–412. Hoff, K., Kshetramade, M., & Fehr, E. (2011). Caste and punishment: The legacy of caste culture in norm enforcement. The Economic Journal , 121(556), 449–475. Hoff, K., & Stiglitz, J. E. (2016). Striving for balance in economics: Towards a theory of the social determination of behavior. Journal of Economic Behavior & Organization, 126 , 25–57. Kahneman, D. (2003). Maps of bounded rationality: Psychology for behavioral economics. American Economic Review, 93(5), 1449–1475. Kahneman, D. (2011). Thinking, fast and slow. New York: Macmillan. Kahneman, D., & Tversky, A. (1979). Prospect Theory: An analysis of decision under risk. Econometrica: Journal of the Econometric Society, 47 (2), 263–291. Kahneman, D., Knetsch, J. L., & Thaler, R. H. (1990). Experimental tests of the endowment effect and the Coase theorem. Journal of Political Economy, 98(6), 1325–1348. Kinzler, K. D., & Spelke, E. S. (2011). Do infants show social preferences for people differing in race? Cognition, 119 (1), 1–9. Knight, F. H. (1923). The ethics of competition. The Quarterly Journal of Economics, 37(4), 579–624. Lachmann, L. M. (1969). Methodological individualism and the market economy. In E. W. Streissler (Ed.), Roads to freedom: Essays in honor of Friedrich A. von Hayek (pp. 89–103). London: Routledge and Kegan Paul. Laibson, D. (1997). Golden eggs and hyperbolic discounting. The Quarterly Journal of Economics, 112(2), 443–478. Laibson, D., & List, J. A. (2015). Principles of (behavioral) economics. American Economic Review, 105 (5), 385–90. Lang, K., & Kahn-Lang Spitzer, A. (2020). Race discrimination: An economic perspective. Journal of Economic Perspectives, 34 (2), 68–89. LeBoeuf, R. A., Shafir, E., & Bayuk, J. B. (2010). The conflicting choices of alternating selves. Organizational Behavior and Human Decision Processes, 111(1), 48–61. Lecouteux, G. (2022) The Homer economics narrative: from cognitive psychology to individual public policies. GREDEG WP No. 2022-29, https://hal.archivesouvertes.fr/hal-03791951. Leibbrandt, A. (2012). Are social preferences related to market performance? Experimental Economics, 15(4), 589–603. McCloskey, D. N. (2022). Beyond Positivism, Behaviorism, and Neoinstitutionalism in Economics. Chicago: University of Chicago Press. Menger, C. [1963] 1985. Investigations into the Methods of the Social Sciences with Special Reference to Economics. New York: New York University Press

Methodological Individualism in Behavioral Economics

679

Mullainathan, S., & Thaler, R. H. (2000). Behavioral economics. NBER Working Paper 7948. Odean, T. (1998). Are investors reluctant to realize their losses? The Journal of Finance, 53(5), 1775–1798. O’Donoghue, T., & Rabin, M. (1999). Doing it now or later. American Economic Review, 89 (1), 103–124. Ostrom, E. (2010). Beyond markets and states: polycentric governance of complex economic systems. American Economic Review, 100 (3), 641–72. Poteete, A. R., Janssen, M. A., & Ostrom, E. (2010). Working together. Princeton: Princeton University Press. Read, D., & Van Leeuwen, B. (1998). Predicting hunger: The effects of appetite and delay on choice. Organizational Behavior and Human Decision Processes, 76 (2), 189–205. Ross, D. (2022). Economics is converging with sociology but not with psychology. Journal of Economic Methodology, 1–22. https://doi.org/https://doi.org/10.1080/ 1350178X.2022.2049854. Russell, T., & Thaler, R. (1985). The relevance of quasi rationality in competitive markets. The American Economic Review, 75 (5), 1071–1082. Sandel, M. J. (2020). The tyranny of merit: What’s become of the common good? London: Penguin UK. Schumpeter, J. A. (1954). History of economic analysis. New York: Oxford University Press. Sen, A. K. (1987). On ethics and economics. Oxford and New York: Basil Blackwell. Sen A. K. (2000). Merit and justice. In: K. Arrow, S. Bowles, and S. Durlauf (Eds.), Meritocracy and economic inequality. Princeton: Princeton University Press. Sent, E. M. (2004). Behavioral economics: How psychology made its (limited) way back into economics. History of Political Economy, 36 (4), 735–760. Simon, H. A. (1955). A behavioral model of rational choice. Quarterly Journal of Economics, 69 (1), 99–118. Simon, H. A. (1956). Rational choice and the structure of the environment. Psychological Review, 63(2), 129–138. Simon, H. A. (1959). Theories of decision-making in economics and behavioral science. American Economic Review, 49(3), 253–83. Simon, H. A. (1990). Invariants of human behavior. Annual Review of Psychology, 41(1), 1–20. Simon, H. A. (1991). Models of my life. New York: Basic Books. Stigler, G. J. (1981). Economics or ethics? In S. McMuffin (Ed.), Tanner lectures on human values, vol. II. Cambridge: Cambridge University Press. Stigler, G. J., & Becker, G. S. (1977). De gustibus non est disputandum. The American Economic Review, 67 (2), 76–90. Sugden, R. (2016). Ontology, methodological individualism, and the foundations of the social sciences. Journal of Economic Literature, 54 (4), 1377–89. Sugden, R. (2018). The community of advantage: A behavioral economist’s defense of the market. Oxford: Oxford University Press.

680

M. Dold

Sunstein, C. R. (1996). On the expressive function of law. University of Pennsylvania Law Review, 144 (5), 2021–2053. Tanaka, T., C. F. Camerer, & Nguyen, Q. (2010). Risk and time preferences: linking experimental and household survey data from Vietnam. The American Economic Review, 100 (1), 557–571. Thaler, R. (1985). Mental accounting and consumer choice. Marketing Science, 4 (3), 199–214. Thaler, R. (2015). Misbehaving: The Making of behavioral economics. New York, NY: W.W. Norton & Company. Tversky, A., & Kahneman, D. (1986). The framing of decisions and the evaluation of prospects. In R. Barcan Marcus, G. Dorn, and P. Weingartner (Eds.), Studies in logic and the foundations of mathematics, Vol. 114 (pp. 503–520). Amsterdam: North-Holland. Voors, M. J., Nillesen, E. E., Verwimp, P., Bulte, E. H., Lensink, R., & Van Soest, D. P. (2012). Violent conflict and behavior: a field experiment in Burundi. American Economic Review, 102(2), 941–64 Whitman, G. (2022). Austrian behavioral economics. Journal of Institutional Economics, 18(3), 449–466. Wilkinson, N., & Klaes, M. (2017). An introduction to Behavioral Economics, 3d. London: Macmillan.

Agent-Based Models and Methodological Individualism: Are They Fundamentally Linked? Gianluca Manzo

1

Introduction

Agent-based computational models (hereafter, ABMs; or, ABM for agentbased model/modeling) are stylized representations of a set of entities that, given certain contextual constraints, interact over time according to given behavioral rules, thus creating, usually through feedback loops across different levels of analysis, outcomes, and trends that were not present at the outset of the simulated process. ABMs are computational models in the sense that they

This chapter reproduces the following paper: Manzo, G. (2020). Agent-Based Models And Methodological Individualism: Are They Fundamentally Linked? L’Année sociologique, 70(1), 197–229. I am sincerely grateful to Francesco Di Iorio, John Goldthorpe and Petri Ylikoski who took time to provide extensive written comments on an early draft of this paper. Preparatory materials were presented at workshop on “Methodological Individualism Today” held at the École française de Rome in June 2019. I thank the organizers and participants for their feedback. I also benefited from some helpful and detailed remarks from three anonymous reviewers of L’Année sociologique. Last but not least, I am grateful to Peter Hamilton for his careful linguistic revision of the original article as well as to Alexandra Frénod for her highly professional assistance in implementing the required editorial style for the current publication. Obviously, the usual disclaimers apply.

G. Manzo (B) Department of Sociology, Sorbonne University, Paris, France e-mail: [email protected]

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 N. Bulle and F. Di Iorio (eds.), The Palgrave Handbook of Methodological Individualism, https://doi.org/10.1007/978-3-031-41512-8_30

681

682

G. Manzo

are solved by the execution of a computer program that translates algorithmically the hypothesized entities’ rules, interactions, and constraints (for an introduction, see Wilensky and Rand 2015). Over the last twenty years or so, this type of computer simulation has been applied across a variety of disciplines to topics as diverse as cancer growth (Norton et al. 2019), blood circulation (Bora et al. 2019), animal behaviors (Tang and Bennet 2010), disease diffusion (Ajelli et al. 2010), city changes (Hosseinali 2013), traffic jams (Barthélemy and Carletti 2017), pedestrian movements (Kerridge 2001), ancient civilization dynamics (Heckbert 2013), crime (Malleson 2012), fertility decisions (Gonzales-Bailon and Murphy 2013), mating behaviors (Billari et al. 2007), tax evasion (Hokamp et al. 2018), financial trends (Lux and Marchesi 1999), social inequalities (Manzo 2013), the diffusion of innovations (Manzo et al. 2018), or opinion dynamics (Flache et al. 2017), to name only a few topics. Originally dominated by over-simplistic models, the field is becoming aware of the importance of ABMs that seek more systematic connections with existing sociological theories and empirical data (compare, for instance, the studies reviewed in Macy and Willer [2002] with those discussed in Bianchi and Squazzoni [2015]). The proliferation of ABMs was paralleled by meta-analyzes on their potential and limitations. Some debates concerned highly technical issues such as the tools for the implementation of ABMs (Railsback et al. 2017), the ways they can be studied (Thiele et al. 2014), or how an ABM can be calibrated and/or validated empirically (Smith and Burow 2018); the attempt to build shared research guidelines is also a recurrent line of reflection (Müller et al. 2014). Other debates focused on the epistemological status of ABMs. In particular, epidemiologists started to investigate whether an ABM can produce causal knowledge (Marshall and Galea 2014) whereas philosophers of social science questioned its explanatory potential (Elsenbroich 2012). Recently, a new stream of discussion, involving both practitioners and philosophers of social science, appeared and added a new question to the debates on offer: is there an inescapable connection between ABMs and methodological individualism (MI, hereafter) (see, in order of publication: Marchionni and Ylikoski [2013], Bulle and Phan [2017], Di Iorio and Chen [2019], Zahle and Kincaid [2020])? In the present article, I focus on this specific debate. My aim is to propose a new way of framing the question of whether or not ABM can/should be seen as the ideal tool to study formally theoretical explanatory models conceived in MI terms. My argument stems from the observation that existing analyzes of the connection between ABMs and MI are all premised on specific interpretations of what MI is. By “specific,” I refer here to definitions of MI that are

Agent-Based Models and Methodological …

683

given in terms of types of entities that can legitimately enter an explanatory model, in order for the model to remain within the realm of MI. That current analyzes endorse this starting point is certainly not surprising. As documented by Lars Udehn, the entire history of MI across authors and disciplines can indeed be thought of as a slow evolution from “strong” to “weak” versions of MI, where the distinction essentially refers to the proportion of collective and institutional (versus individual) variables, entities, and properties that are accepted within the explanation (see, in particular, Udehn [2001: 347]; see also, Udehn [2002: 499, 502]). Although understandable, the problem with starting from a specific interpretation of MI is that, given the variety of definitions of MI on offer, the analysis of the link between ABMs and MI is transformed into a “battle” over the “right” understanding of MI (this is what I will call the “definition battle”). As right here essentially means deciding what entities can be legitimately introduced within an explanation inspired by MI, this battle in turn triggers a more subtle dispute about why this or that type of entities are legitimate or not and should have a primary explanatory role, a dispute that ultimately shifts to the ontological level (I will refer to this as the “ontological battle”). Thus, within this argumentative framework, the answer that one can offer to the question of whether or not ABM can/should be seen as an implementation of MI will only continue to be controversial. To overcome these problems, I suggest reversing the way existing analyzes approach the connection between ABM and MI. Rather than basing the investigation on a specific definition of MI, thus implicitly going from MI to ABMs, I suggest starting with ABM by identifying the generic properties of this method. By “generic” properties, I refer to properties that are not defined in connection with any particular type of entities that a given ABM may be able to represent. My inspiration here is taken from the concept of “generic instruments,” a concept that sociologists of science created to describe those material devices that are conceived in such a way that their functioning can be reshaped again and again as a function of the specific requirements of a given context of application (see Joerges and Shinn [2001: 9, 2002: 208, 212, 217, 244], Shinn [2007, 2008]). ABM’s generic properties thus tell us only the type of problems the method is best suited for. The main proposal of the paper then is to move from ABM to MI in order to apply this understanding of “genericity” to MI and formulate the working hypothesis that ABM’s generic properties are compatible with principles that partisans of MI may also support. To defend this view, the paper goes through the following steps. I first retrace the way the connection between ABMs and MI was established.

684

G. Manzo

Secondly, I offer an overview of the arguments defended by those authors who have systematically addressed the question of whether or not the connection between ABMs and MI is inevitable. Third, I show that these investigations share two common problems that are likely to make the answer to the focal question continuously controversial. Finally, I formulate my own view on the connection between ABMs and MI, and explain in what sense ABMs and MI share generic properties that make them intrinsically associated.

2

The Origin of the Link Between ABM and MI

In the following comment, Macy and Flache formulate a typical statement on the intimate association that computational sociologists tend nowadays to establish between ABM and MI: We begin, in Section 11.1, by introducing ABC [agent-based computational] modeling as a computational implementation of ‘methodological individualism’, the search for the micro-foundations of social life in the actions of intentional agents [...] the agent-based approach replaces a single integrated model of the population with a population of models, each corresponding to an autonomous decision maker. This reflects the core methodologicalindividualist interest in the emergence of population dynamics out of local interaction. Although methodological individualism is older by many decades, ABM modeling can be characterized as its fullest formal representation. (Macy and Flache 2009: 246, 248; emphasis added)

The link between ABMs and MI here is motivated by identifying MI with the quest for actor-centered explanations of macroscopic patterns and ABM as the methodological tool used to study micro-based explanatory models. The origin of this association between MI, micro-foundation, and ABMs has a specific history. It is related to a well-defined technical problem— i.e., the so-called “transformation problem”—with which MI has struggled for decades. Making this history explicit helps to see that the way current analyzes of the connection between ABMs and MI frames the problems, i.e., with specific definitions of MI in mind, in fact, results from a longstanding, but particular, line of reasoning. Being aware of this peculiar context is a preliminary step to argue, as I will do, that the link between ABM and MI can in fact be established on more general bases (see below, section: A new perspective on the connection between ABM and MI).

Agent-Based Models and Methodological …

685

From the Transformation Problem to ABMs Within the history of MI, Coleman is usually associated with the so-called Coleman’s boat (see, in particular, Coleman [1987]). In Udehn’s terminology, the diagram visually illustrates a “weak” form of MI where macroscopic outcomes are seen as the result of concatenations of dynamic loops between structures and actions—as opposed to “strong” forms of MI where structural conditions, possibly pre-existing and out of the control of actors, are excluded from the focus of the analysis (Udehn 2001: 292–306). While, as clearly suggested by Raub and Voss (2017), Coleman should be credited with popularizing this non-reductionist form of MI more than for having invented it—strictly equivalent formulations can be found in Boudon (1986: 30), for instance, or, even earlier, among Dutch sociologists who in fact coined the term “structural individualism” (see, for instance, Wippler [1978: 143])—, it seems fair to say that Coleman discussed the main problem posed by the rigorous implementation of this form of MI more explicitly than others. Since his highly quoted article on the theory of action, Coleman (1986: 1320–1321) indeed emphasized that the most difficult theoretical problem for action-oriented sociologists is “[…] the means by which purposive actions of individuals combine to produce a social outcome […].” Coleman called this problem the “micro-to-macro problem” but he acknowledged that, in Europe, the expression of “problem of transformation” was often in use (see Wippler and Lindenberg [1987]). When discussing his diagram, Coleman then translated this general statement into a claim about which step of the explanation should require the highest attention (Coleman 1987: 155). For him, this is the third one, “[…] for it is the third which moves back up from the individual level to the societal level.” Coleman’s chapter in fact is entirely devoted to showing how many strands of the sociological literature have developed explanations that fail fully to elaborate on the specific ways individual actions combine to produce a given social outcome. The failure, Coleman notes, consists in assuming that “simple aggregation” is enough to account for the micro-to-macro transition (Coleman 1987: 157). The interdependence between actions is the missing element. Some years later, in the Foundations of Social Theory, he reiterated the point and noted that most economic theory, too, eludes the micro-to-macro transition by relying on the metaphor of the “representative agent” (Coleman 1990: 10). Interestingly, the same critique simultaneously appeared in economics where Kirman forcefully argued that this metaphor “is not simply an analytical convenience […] but is both unjustified and leads to conclusions which are usually misleading and often wrong” (Kirman 1992: 117). The

686

G. Manzo

reason is that the “representative agent” is a “fiction” (ibid., p. 125) that gives economists the impression of having solid micro-foundations while in fact the specific ways actors’ heterogeneity is transformed into complex macro-outcomes—like economic cycles or price dynamics—, through interdependent behaviors, is simply ignored (ibid ., p. 132, 134).1 It should be noted that, although without using the name, methodological individualism in sociology, too, can adopt this simplification. Boudon, for instance, defines “individualistic” statements as “[…] statements explaining why the ideal–typical individuals to relevant categories behave the way they did” (Boudon 1998: 176). Only to consider “typical” individuals of a given category amounts to accepting that heterogeneity between individuals within the category can be ignored without consequences: the behavior of one actor is assumed to be sufficient to deduce the aggregate behavior of the whole. In contemporary advanced game theory, the construction of (a limited number of ) “types” of players is indeed the standard way to account for actors’ potential heterogeneity in strategic behavior (see, for instance, Fehr and Gintis [2007]). To go back to Coleman, the interesting point of his analysis is that he formulated a precise hypothesis to explain why sociologists and economists have been obliged for so long to adopt such simplifying assumptions to handle interdependence and heterogeneity. According to him, the responsible is the “[…] little development of methods for characterizing systemic action resulting from the interdependent actions of members of the system” (Coleman 1986: 1316). The absence of appropriate quantitative tools would have impeded formal analysis of theoretical models that include action interdependence (and heterogeneity), thus allowing sociologists to do only half of the job (Coleman 1987: 157), and, on the other hand, as technique tends to feedback into ways of thinking (on this general mechanism, see Abbott 1988, 1990), not having a method suited for interdependence and heterogeneity, may have slowed down advanced theoretical elaboration along these lines. To my argument, what now matters is that, around 2000, ABMs entered the social sciences precisely as the methodological solution to study macroscopic patterns and trends from the bottom-up through the algorithmic implementation of hypotheses on the heterogeneous behaviors of heterogeneous interdependent actors. 1

This is the complete definition Kirman provides of the “representative agent” approach: “[…] macroeconomic models […] typically [they] assume that the choices of all the diverse agents in one sector—consumers for example—can be considered as the choices of one ‘representative’ standard utility maximizing individual whose choices coincide with the aggregate choices of the heterogeneous individuals” (Kirman 1992: 117).

Agent-Based Models and Methodological …

687

In one of the founding manifestos of the field, Joshua Epstein claimed: “Artificial society modeling allows us to ‘grow’ social structures in silico demonstrating that certain sets of microspecifications are sufficient to generate the macrophenomena of interest” (Epstein 2007: 8). In a call for a stronger link between economics and ABMs, Farmer and Foley, after reiterating the critique of the “representative agent” and the mathematical toolkit requiring such a simplification, stated: “[…] There is a better way: agentbased models. […] Done right, the agent-based method can provide an unprecedented understanding of the emergent properties of interacting parts in complex circumstances where intuition fails” (Farmer and Foley 2009: 685, 686). In sociology, Macy and others explicitly recognize that one of the virtues of ABMs is that they “[…] have repeatedly demonstrated how the emergence of macro-social patterns out of micro-social interactions is not always a matter of simple aggregation” (Macy et al. 2011: 252). The connection between ABMs and MI was established through this path. MI fundamentally requires explaining macroscopic patterns resulting from the combination of individual actions. The various ways in which individuals’ actions may be interdependent, and how this interferes with various forms of actors’ heterogeneity, however, make it difficult to study theoretical models in the absence of methods being able to treat both elements formally. ABMs were seen as the technique that could be used to face this challenge. Thus, almost naturally, a long list of statements explicitly conceiving ABMs as a “formal implementation of methodological individualism” (ibid .) has appeared over the last few decades (see, in chronological order, O’Sullivan and Haklay [2000: 1413–1414], Sawyer [2004: 263], Manzo [2007: 5], Neuman [2008], Macy and Flache [2009: 246], Wan [2011: 188]). These statements are not all based on exactly the same understanding of MI; nor do all authors see the link between ABMs and MI in a positive way. But even those— like O’Sullivan and Haklay (2000: 1413–1414), Sawyer (2004: 263)—who consider that framing ABMs as a tool for MI may restrict ABMs potentialities do not contest that ABM has a privileged connection with MI in that it allows this perspective to implement its quest for the micro-foundations of social regularities.

688

3

G. Manzo

Partisans and Critics of the Link Between ABM and MI

Over the last few decades, the claimed link between ABMs and MI has been investigated in a systematic way by Marchionni and Ylikoski (2013), Bulle and Phan (2017), Di Iorio and Chen (2019), Zahle and Kincaid (2020). These analyzes reached opposing conclusions. While Marchionni and Ylikoski (2013) argued that the link in fact is neither necessary nor desirable (and as did Zahle and Kincaid [2020]), Bulle and Phan (2017) concluded that this connection is necessary and fruitful (as did Di Iorio and Chen [2019]). In this section, I scrutinize the arguments put forward by each contribution; although briefly (and in footnotes), I also point to particular problems raised by each line of reasoning. The problems that, in my view, are common to the four articles, which constitute the focus of my critical analysis, will instead be discussed in the next section. Marchionni and Ylikoski initiated the debate with the aim of demonstrating that ABM “[…] should not be conflated with the doctrine of methodological individualism” (Marchionni and Ylikoski 2013: 324). To defend this thesis, they start from two premises. On the one hand, they conceive of a mechanistic explanation within an ABM as amounting to identifying the model’s assumptions that at the same time are sociologically interpretable and produce a difference in the outcome that is of interest when they are systematically manipulated (ibid ., p. 328). On the other hand, they define MI as the principle according to which “Social phenomena can only be explained (or they are best explained) by accounts that only refer to individuals, their properties and their interactions” (ibid ., p. 332). On this basis, Marchionni and Ylikoski inspect a published ABM on opinion diffusion and argue that the variables that are systematically manipulated in fact refer to “structural” properties—i.e., “properties that are attributed to larger scale entities than individuals or if they are attributed to individuals, they presuppose some larger scale entities” (ibid ., p. 334). Thus, Marchionni and Ylikoski conclude that as some of the variables to which ABM modeler attributes explanatory value are “structural,” the association between ABM and methodological individualism is unsound and should be broken (ibid ., p. 335).2 2 In my view, Marchionni and Ylikoski’s conclusion raises three specific problems. First of all, they start with a definition of MI that excludes by construction the potential explanatory role of contextual and structural factors. While many contemporary philosophers of social sciences may accept this definition as a “philosophical commonplace” (in Ylikoski’s own words, private correspondence), it should also be admitted that MI has received other, and more inclusive, interpretations during its long history, and that premising an argument on this or that definition is not neutral for

Agent-Based Models and Methodological …

689

In response to Marchionni and Ylikoski, Bulle and Phan (2017: 404, proposition I) elaborate a different line of reasoning. They accept that in principle there is no inherent connection between ABMs and MI but argue that, in practice, this connection is needed to maximize ABMs’ capacity to deliver mechanism-based explanations. Bulle and Phan, too, premise their argument on a specific definition of MI. In particular, they see MI as the “generic approach in the social sciences” according to which “[…] forces in action in society are governed by the subjective meaning of the reasons for individual actions,” and claim that “[…] For methodological individualism, social/relational structures have an explanatory or causal role in the representation of generative mechanisms only insofar as they affect the subjective meaning of/the reasons for individual actions by the contextual properties they define” (ibid ., p. 332). As to ABMs, Bulle and Phan also formulate an important premise. They correctly remark that an ABM intrinsically is nothing more than a symbolic system, which, “[…] from the point of view of the phenomenal world […] ‘speaks’ of nothing” (ibid ., p. 387). That is why, Bulle and Phan continue, any ABM needs an interpretative framework that assigns specific meanings to the entities, properties, and connections postulated by the model. The authors regard MI as the required interpretative framework because of its emphasis on actors’ rationality. They consider indeed that actors’ rationality has the property of being “trans-situational,” a property that Bulle and Phan. in turn, see as the necessary feature that the entities represented by a given ABM must possess in order to make the mechanisms postulated by the model generalizable (ibid ., p. 399, 402, 404, proposition E).3

the conclusion. Second, the selected definition of MI regards the interactions among actors as a legitimate explanatory variable. It is then unclear why properties concerning these interactions and their structure are exclusively considered “structural” and lead an ABM including them to go beyond MI. Finally, Marchionni and Ylikoski do not consider the possibility that, when ABM modelers systematically manipulate structural variables, they are in fact exploring the conditions under which the postulated actor-centered mechanism is able to trigger the expected consequences. In this case, structural properties should not be regarded as “crucial explanatory variables” but simply as scope conditions. And structural variables as conditions for the explanation seem a less strong violation of the proposed view of MI than if structural variables were “the” explanation. 3 By stressing the conceptual emptiness of ABM, Bulle & Phan make a potentially highly consequential point (I will come back to it in the section: A new perspective on the connection between ABM and MI). However, their emphasis on actors’ rationality, and in particular on the supposedly “trans-situational” nature of this rationality, makes the entire argument questionable. The concept of “trans-situationality” is implicitly defined in terms of “invariance” but it is unclear in what sense and on which basis the authors claim that “rationality” is “invariant.” Theoretically, Boudon explicitly insisted on the fact that actors’ reasons can be “context-free” as well as “context-dependent” (Boudon 2014); empirically, it has repeatedly shown that actors’ rationality tends to change across decision contexts (Gigerenzer and Gaissmaier 2011) and frames (Della Vigna 2009). It is also unclear why “trans-situationality,” and, consequently, generalizability, should be seen as a necessary condition for

690

G. Manzo

Di Iorio and Chen have recently pursued the debate through a different line of reasoning (Di Iorio and Chen 2019). They directly attack one of the premises of Marchionni and Ylikoski’s argument, namely their understanding of MI. According to Di Iorio and Chen, “Marchionni and Ylikoski agree with the dominant interpretation of MI in terms of reductionism […],” an interpretation that “[…] is misleading and philologically incorrect […]” (ibid ., p. 2; emphasis added). Di Iorio and Chen thus deliver a synthetic study of MI over the long run and argue that a non-atomistic variant of MI exists that “[…] conceives of society in an emergentist and systemic way and the agent as being influenced by many structural and socio-cultural factors that limit his/her freedom” (ibid ., p. 6). Within this understanding of MI, the authors continue, “non-individual agents” are also admitted provided one can imagine shared decision procedures that justify the attribution of choices and actions to the collective actors (ibid ., p. 15). When Di Iorio and Chen conclude, it will be accepted that “the widespread interpretation of MI in terms of reductionism is historically incorrect and must be rejected,” then it will become obvious that the presence and the manipulation of structural properties simply means that ABMs “[…] must be regarded as explanations in terms of non-reductionist MI” (ibid ., p. 17).4 However, what seems to be going without saying to Di Iorio and Chen is again challenged by Zahle and Kincaid (2020). Similarly, to Marchionni and Ylikoski, they want to demonstrate that “[…] ABMs need not, and sometimes should not, be seen as embodying the methodological individualist program” (ibid ., p. 1). Again, the argument is premised on a certain definition of MI. According to Zahle and Kincaid, “the key MI claim” is that “explanations should be in terms of individuals” (ibid ., p. 2). They acknowledge however that this claim has received different interpretations. To some (but they do not give any specific reference), individualistic explanations amount to only considering individuals in isolation. Given this

good mechanistic explanations. One may indeed defend the opposite claim, as Kroneberg did in a recent and provocative article (Kroneberg 2019). 4 To those who are familiar with careful historical analyses of MI—like the monumental summa provided by Udehn (2001), Di Iorio and Chen’s general line of reasoning is perfectly acceptable. However, their conviction that it is possible to demonstrate that one (or a few) of the existing variants of MI is (are) “philologically” and “historically correct,” to take their own words, seems more debatable. Who has the skills to establish the proper variant of MI? What selection of writings should be included within the analysis? Which disciplines should we consider? What periods of time? From a methodological point of view, the idea of the existence of a correct version of MI seems problematic. It is certainly possible to ascertain that this rather than that interpretation of MI is at work within a specific (set of ) piece(s) of research, and that this rather than that version is heuristically more useful for the research purpose at hand. By contrast, to reach a stable consensus, among scholars and across disciplines, once and for all, on what MI really is seems a task doomed to fail.

Agent-Based Models and Methodological …

691

understanding, they argue, many ABMs clearly escape MI because they represent social interactions as well as a non-physical environment (ibid ., p. 5). To others (Zahle and Kincaid explicitly refer here to Watkins), however, “relations between individuals” are legitimate explanatory elements within an individualistic explanation, a view that would then obviously make ABMs compatible with MI. In some accounts of MI, continue the authors (who refer here to Popper, Agassi, and Jarvie as examples of “institutional” individualists), even institutions (like firms, universities, states, or codified laws and social norms, they say) are accepted as legitimate elements defining the context of individuals’ actions. Within this view of MI, ABMs can also be seen as a tool for MI because they are able to model institutions (ibid ., p. 5). If so, is there any element that is out of the acceptable explanatory toolbox of MI while being (possibly) present within ABMs? According to Zahle and Kincaid, these elements are of two types: collective actors (among which they mention “households, firms, nations, banks, central banks, political parties, elites, and governments”) and aggregative properties (among which they consider “statistical properties” and “formal network properties”). In both cases, they claim, these elements are often introduced within an ABM not as simple background features defining the context of actions of individual agents but, more fundamentally, as central explanatory variables. To Zahle and Kincaid, this is “unacceptable even to institutional individualists” (ibid ., p. 6).5 In sum, contrary to computational modelers who suggested that ABM is a “formal implementation of methodological individualism” (Macy et al. 2011: 252), scholars who systematically investigated the claimed link between ABMs and MI reached symmetrically opposed conclusions on the soundness of this link. How can such markedly different answers to the same question be explained? In the next section, I discuss two reasons that seem to me at the origin of the disagreement.

5 Zahle and Kincaid’s analysis has the merit of progressively increasing the inclusiveness of MI definitions and showing how the answer to the question of whether ABMs fall within MI depends on these modifications. However, what they consider as the frontier separating MI from non-MI explanations illustrates the difficulty of this “definitional” approach. On the inclusion of collective actors, indeed, Di Iorio and Chen defend the opposite interpretation of MI when they overtly claim that “the fact that ABS [agent-based system] methodology sometimes refers to non-individual agents is not a sufficient condition to conclude that ABS is incompatible with MI” (Di Iorio and Chen 2019: 15).

692

4

G. Manzo

Common Problems with the Current Analyzes

Existing principled analyzes of the supposedly deep link between ABMs and MI can be questioned on this or that point of their specific line of reasoning.6 However, I would like to suggest that these analyzes in fact also have two fundamental limitations in common. These limitations trap the debate in what I propose to label the “definition” and the “ontological” battle, respectively. These problems require a systematic analysis because they are likely to make the debate on the connection between ABMs and MI endless and unproductive. I discuss them in turn.

The “Definition Battle” The four articles that I have analyzed share a first general, but fundamental, point: they all premise their analysis of the link between ABMs and MI on a specific understanding of MI. The authors have opposite orientations on the status of the selected definition, however. The contrast is well illustrated by Marchionni and Ylikoski (2013: 331)— who stated that “[…] the debate over the proper definition of methodological individualism is a distraction from the real methodological issues”—and, on the other hand, Di Iorio and Chen (2019: 4)—who claim that “[…] within the frame of philosophical analysis, the problem of the compatibility between MI and ABS cannot be satisfactorily analyzed without considering the debates on the proper definition of MI, i.e., without assuming that the substantial aspects of this problem are relevant.” Both perspectives—which I suggest to call respectively “stipulationist” and “realist”—seem problematic. Marchionni and Ylikoski implicitly ask the reader to accept their starting definition of MI as a convention that one must stipulate for the sake of the argument and/or in order to be understood within a certain academic circle. This “stipulationist” stance is problematic because the answer to the question of whether or not ABMs and MI are intrinsically connected in fact crucially depends on the starting definition of MI. Thus, why should one regard convincing the conclusion of an argument that is premised on a definition of MI that sounds acceptable only for a specific audience? Marchionni and Ylikoski’s claim could also be read as meaning that any starting definition of MI should only grasp the essential features of this doctrine. This view would be in line with Zahle and Kincaid’s starting point according to which 6

See footnotes [2–5].

Agent-Based Models and Methodological …

693

“the key MI claim” is that “explanations should be in terms of individuals” (Zahle and Kincaid 2020: 2). However, Zahle and Kincaid’s paper itself shows that several interpretations exist of what an individualistic definition is so that the proposed definitional strategy raises the problems of knowing if one has considered all the possible interpretations and of establishing which one better corresponds to “the key MI claim.” For all these reasons, Di Iorio and Chen’s view seems more transparent in that it overtly raises the issue of the definition of MI rather than trying to bypass it. The shortcoming of their approach, however, is the implicit conviction that some interpretations of MI are more correct than others. Ultimately, the stipulationist view—according to which only the essential features of MI (as seen within a certain audience) matter in discussing whether or not ABMs are individualist—like the “realist” one—according to which it is necessary to establish the proper interpretation of MI—underestimate the difficulty of reaching a consensus on how MI should be interpreted. Detailed historical investigations of MI concluded that “methodological individualism exists in a bewildering number of different versions” (Udehn 2001: 346). Even a single author—like Popper (see Udhen [2001: Chap. 7])— or specific variants of MI—like “rational choice individualism” (ibid ., ch. 10)—can in fact be received differently. This is not to deny that progress in understanding what MI is can be made. But this progress may realistically come precisely from accepting the existence of a variety of versions of MI and the absence of a “true version of this doctrine” (Udehn 2002: 480). This lesson is highly consequential for the debate on the connection between ABMs and MI. The existing analyzes all premise their assessment of whether or not ABM can be seen as an implementation of MI on a specific understanding of MI without considering the intrinsic difficulty of finding an agreement on how MI should be interpreted. This strategy can only lead to conclusions that are likely to remain controversial: it only suffices to disagree on the chosen definition(s) of MI to be tempted to reject the conclusion on the nature of ABM.

The “Ontological Battle” The difficulty of agreeing on the proper definition of MI is reinforced by a second, and deeper, problem. When arguing in favor of or against a given understanding of MI the authors of the articles under examination are in fact discussing what kind of entities can legitimately be inserted within a given theoretical model. This is the defining feature of what I have termed

694

G. Manzo

the “specific” interpretation of MI in the introduction. Now, to defend the legitimacy of this or that entity, the authors inevitably end up developing, more or less explicitly, an argument on the existence of these entities, and, based on this, on their explanatory primacy. Thus, no matter if this is explicitly endorsed or not, the dispute over the proper content of a MI-inspired model, shifts silently and smoothly into an ontological dispute. Similarly, to the definition problem, the authors animating the debate on the link between ABMs and MI display opposite views as to the impact that ontology may have on their conclusions. Marchionni and Ylikoski regard MI as “a thesis about explanation, not about ontology,” and claim that ontological arguments are not directly consequential for theses about explanation (Marchionni and Ylikoski 2013: 332).7 Di Iorio and Chen challenge this interpretation of MI and claim that the proper understanding of MI—i.e., the non-reductionist variant, in their view—is based on explicit “ontological individualism” (Di Iorio and Chen 2019: 11).8 Explicitly admitting the potential relevance of the ontological level seems a more defensible perspective. A commitment to the disjunction between the explanatory and ontological levels is indeed likely to be violated in practice as soon as the argument is pushed farther. Zahle and Kincaid well illustrate this potential inconsistency when, after arguing that ABMs are not limited to individualistic explanations, they ask whether ABMs should be still used only to provide individualistic explanations. To challenge one of the possible arguments from the individualist camp in favor of this thesis—namely, that “social phenomena are causally inert,”—they note: [...] as social properties supervene on individualist properties, the latter supervene on biological properties that, in turn, supervene on chemical properties that supervene on physical properties. [...] As a result, individualist properties, qua being supervenient, are causally inert too and MIs must admit that it isn’t legitimate to offer individualist (causal) explanations either (Zahle and Kincaid 2020: 7).

7 Zahle and Kincaid think along similar lines when they state that “methodological individualism is a claim about explanations” (Zahle and Kincaid 2020: 2). 8 Here is Di Iorio and Chen’s full statement: “Even regarding this point, the dominant interpretation seems to us to be historically inaccurate. […] Many non-reductionist individualists such as Carl Menger, Max Weber, Georg Simmel (1858–1919), Herbert Spencer (1820–1903), Ludwig von Mises, Friedrich Hayek and Karl Popper argued that MI is based on ontological individualism […] According to ontological individualism, while the word ‘individual’ does correspond to a real entity, collective nouns such as ‘capitalism’ or ‘society’ do not, in the sense that they refer to a collection of individuals and the consequences of their interaction (which must be described in systemic terms)” (Di Iorio and Chen 2019: 11).

Agent-Based Models and Methodological …

695

Although the discussion is framed in explanatory terms, the argument ultimately (but implicitly) is based on the assumption that certain entities of the social and physical realm exist and that these entities are organized in a clearly identifiable hierarchy. This is an ontological debate, not an epistemological one. However, when ontology is explicitly taken into account, as in Di Iorio and Chen’s approach to MI, problems that are likely to trigger further debates that are difficult to settle also arise. Given the focus of my discussion, let me illustrate this point by evoking one of these debates, namely, the vexata quaestio of the existence of individuals and of their analytical primacy (for a detailed analysis, see Goldthorpe [2006: Chap. 8, 174–183]). Within MI, the fact that only individual actors exist is usually regarded as something that goes without saying. To express this point, Hedström proposed the following metaphor: “The causal efficacy of actions would be readily seen if we were able to press a pause button that suddenly froze all individuals and prevented them from performing any further actions. All social processes would then come to an immediate halt” (Hedström 2005: 28). Similarly, Abell claimed: “The substantive point is that the motor energy of individual actions is necessary for the creation of any pattern of social causality” (Abell 2003: 259; emphasis in the original). To those authors, it is obviously that the “energy” has its origin in actors’ beliefs and intentions but this does not go at all without saying for scholars of other orientations. Sperber, for instance, formulates this concern as follows: “Individual agency is taken as a primitive in this approach, rather than as a tentative construct that should be unpacked and possibly questioned by psychology and biology” (Sperber 2011: 64). He remarks that, empirically, “no primary reasons to give pride of place to intentions among the cognitive determinants of behavior”; “attention or memory,” he noted, may equally claim a primary role (ibid ., p. 71). From a relational (rather than sub-individual) perspective, Padgett and Powell (2012: 2) formulate a similar objection when they remark that, within MI, actors’ preferences are not further unpacked.9 In sum, as these contrasting statements show, the “definitions” and the “ontological” battles are inevitably related. The defense of interpretations of

9 Here is Padgett and Powell’s full statement: “Most social science proceeds according to the logic of methodological individualism. […] ‘Actors’ are objects imbued with boundaries, purposes, and choices whose teleological behavior is explained thereby. […] To assume axiomatically that real people are actors makes them logically impenetrable to the theories built upon them. No theory can derive its own axioms. The problem is not that the social science concept of actor is not useful. The problem is that the atomic conception of actor precludes investigation into the construction and emergence of the real people and organizations that we refer to by that abstraction” (Padgett and Powell 2012: 2; emphasis in the original).

696

G. Manzo

MI that are specific requires the justification of the legitimacy of the postulated entities, and, to argue in favor of this legitimacy, one is ultimately required to make ontological assumptions explicit, which in turn raises issues that are difficult to ascertain empirically. For these reasons, if the question of the connection between ABMs and MI is addressed by starting with a specific definition of MI, the answer to the question is likely to inherit all the difficulties related to the chosen definition. The assessment of the link between ABM and MI thus remains unstable because it entirely depends on the extent to which an observer agrees on the posited (specific) account of MI.

5

A New Perspective on the Connection Between ABM and MI

To frame differently the question of the link between ABMs and MI, let me start with the following observation. From a computer programming perspective, an ABM is nothing more than a set of “objects,” i.e., computational units defined by certain properties (attributes) and rules of behavior (methods or procedures). According to the bundle of properties and rules defining an object, the latter can be used to model the behavior and interactions of a set of particles, molecules, cells, beliefs, actors, groups (of particles, molecules, etc.), organizations, etc. (see Manzo [2010: 146–147], Manzo [2014: 440– 441]). Within the debate under examination, although using different terms, Marchionni and Ylikoski, Bulle and Phan, and Zahle and Kincaid have all rightly remarked on this aspect when they acknowledged that the interpretation of the entities and the rules on which an ABM is based is not dictated a priori by the method itself (Marchionni and Ylikoski 2013: 335: Bulle and Phan 2017: 387; Zahle and Kincaid 2020: 6). None of them, however, drew out all the consequences of this important technical observation. The intrinsic substantive emptiness of an ABM indeed ultimately implies that including or excluding this or that type of entities (and/or relationship between them) from a given account of MI is inconsequential to knowing whether or not ABMs fall within the realm of MI: no matter what choice one makes, it will be always possible to find (or to design) an ABM where the computational objects and their relations represent the entities posited by the selected definition of MI. This is a fundamental reason for looking at the connection between ABMs and MI from a different perspective. In particular, I propose to take seriously the intimate substantive emptiness of ABM and to look for ABM’s

Agent-Based Models and Methodological …

697

generic properties. As anticipated in the introduction, I take here inspiration from the concept of “generic instruments” that sociologists of science and techniques coined to describe electronic or mechanical devices—like micro-processors or ultra-centrifuges (Joerges and Shinn 2002)—or symbolic tools—like programing languages (see Shinn [2007])—that are based on principles and properties that are malleable, adaptable, and transversal to specific application domains (see Joerges and Shinn [2001: 9], Shinn [2008]). From this perspective, ABM’s generic properties can be defined as those technical features that are independent of the specific type of entities that a given model may want to represent. Once these theoretical lenses are endorsed, the problem of the link between ABM and MI can be addressed, first, by identifying ABM’s generic properties, and, then, by assessing the extent to which these properties can also be seen as defining MI irrespectively from the type of entities that are regarded as legitimate within a given understanding of the doctrine. I discuss these problems in turn in the two following subsections.

ABM’s Generic Properties Behind the endless diversity of ABMs in terms of macroscopic outcomes, entities, connections, dynamic processes, and levels of analysis involved, existing ABMs in fact share fundamental properties that are independent from the specific content of each model.10 Among these properties, five of them constantly recur across applications. I propose to call them “levelchange” propensity, “recursivity,” “fractality,” “generalized interdependence,” and “heterogeneity.” Let me briefly discuss each of them in turn. By “level-change” propensity, I refer to ABM’s flexibility in understanding a given pattern in terms of mechanisms that are conceptualized (by the modeler) as located at a different level of analysis. “Level-change” can concern problems where a macro-pattern needs to be understood in terms of micropatterns (“upward” level-changes) as well as problems where a micro-pattern is conceived of as being impacted by some macro-patterns (“downward” levelchanges). For instance, if one wants to understand the specific location of a pedestrian (low-level) within a given physical space (high-level), an ABM

10 This astonishing variety could be documented by inspecting literature reviews of ABMs in disciplines as diverse as sociology (Bianchi and Squazzoni 2015), political science (De Marchi and Page 2014), demography (Grow and Van Bavel 2016), psychology (Eberlen et al. 2017), criminology (Birks et al. 2012), epidemiology (Tracy et al. 2018), economics (Neugart and Richiardi 2018), geography (O’Sullivan 2008), ecology (Schulze et al. 2017), archaeology (Wurzer et al. 2015), or biology (Metzcar et al. 2019).

698

G. Manzo

can be used to model the way the structure of the environment (macropatterns) impacts on the likelihood of the pedestrian being in a given point in the space at some point in time (micro-pattern) (Zheng et al. 2009). What matters is that the modeler thinks that moving from one level to another can generate some explanatory gain—whatever the content of these levels and/ or the “direction” of the movement are. “Level-change” propensity should be analytically distinguished from the operation of micro-reducing macropatterns. ABMs are indeed often perceived simply as a tool to demonstrate that macroscopic patterns in fact do not have sui generis properties and that it suffices to understand the behavior of the parts to explain the properties of the whole (see, for instance, Little’s distinction between “generativity” and complexity [Little 2016: Chap. 5]). Quite the contrary, an ABM is often built to show how moving up and down across levels, i.e., “level-change” propensity, can help to understand the origin of properties that no low-level entity can possess on its own (for example, in the field of social stratification, see Manzo [2013]). ABM’s “level-change” propensity is not a static feature. Within an ABM, moving up and down across sets of entities conceived of as located at different levels of analysis is an operation that is repeated iteratively. This is what can be called “recursivity,” which I regard as the second generic feature of an ABM. Recursivity is a direct consequence of the fact that ABM is a simulation-based method. There is no simulation without computational time elapsing. Concretely this means to iterate a set of rules a certain number of times. The intrinsic dynamic nature of an ABM—see, on this point, Miller and Page (2007: 80–81, 83–84)—thus implies that the procedures defining the entities’ behaviors trigger a chain of activities, reactions, and updating that constitute the ultimate higher-level outcome of interest through a cascade of loops composed of upward local aggregations and downward effects. As we have seen earlier, the capacity to handle “upward” level changes was the initial motivation for adopting ABM in sociology, in particular within the MI tradition (see section: The origin of the link between ABM and MI). This should not hide however the fact that, within an ABM, “moving up” is never disconnected from “moving down:” iteratively looping levels over time, i.e., recursivity, is the second deepest generic feature of ABM. The third one is its fractal nature. By this, I refer to ABM’s capacity to realize “level-change” propensity every time one wants to decompose a given level of analysis into more elementary mechanisms. This property is another consequence of the basic unit on which an ABM is based, i.e., computational objects. Within the object-oriented paradigm, everything can be “objectified.” This means that, as soon as a given class of objects is created to represent a given set of entities/properties/activities/connections, another class of

Agent-Based Models and Methodological …

699

objects representing another set of entities/properties/activities/connections that is regarded as internal to the first set can be created to better ground the behavior of the first posited class of objects. For instance, if a given class of objects represents the properties and the activities of a population of actors (still the most frequent use of ABMs in social sciences), the brain functioning of each actor may be modeled by means of another class of objects representing neurons, their properties, activities, and connections; and one can let the two classes of objects depend on, and communicate with, each other. This is done, for instance, when an agent-based model at the actor level is coupled with an agent-based model at the infra-actor level (through the use of artificial neural networks—see, for instance, Hayward [2006]). The fourth feature that recurs within ABMs across disciplines is the dependence of one entity’s state(s) and behavior(s) on other entities’ state(s) and behavior(s). This is possible because computational objects can communicate with each other through message passing at the deep level of the computer’s memory addresses (Hummon and Fararo 1995). In fact, interdependence comes under three different forms that must be distinguished analytically. First, entity interdependence can be mediated by a global aggregate, namely, an outcome derived from the behavior of all entities present in the artificial population that feedbacks into the subsequent behaviors of each entity. Second, entity interdependence can be mediated by a local aggregate, namely, an outcome derived from the behavior of all entities to which the focal entity is connected that feedbacks into the subsequent behaviors of the focal entity. Third, when the relevant input for a given entity comes from a single other entity, entity interdependence is purely dyadic. In the latter two cases, the relevant entity’s neighborhood can be defined on a spatial and/or relational basis. In any minimally complex ABM, these forms of interdependence combine in various ways. In this sense, I propose to call “generalized” interdependence the fourth generic feature of ABM. Finally, heterogeneity clearly is a fundamental generic property of this method. ABM’s capacity to handle heterogeneous entities again comes from the underlying object-oriented nature of the technique. Objects can be heterogeneous in a variety of ways (Epstein 2007: Chap. xvi, 7). First, objects within the same class, while by definition sharing the same properties (and activities), are such that these properties can get different values. Second, objects in different classes by construction possess different types of behavior. Third, by playing with the objects’ scheduling, objects can be represented as being heterogeneous in terms of behavioral sequences (i.e., the time at which a given behavior is realized). Finally, as I remarked at the beginning of this section, objects are conceptually “empty,” meaning that, by playing

700

G. Manzo

with variables, vectors, or other data structures, the objects’ states can model any attributes of the entities of interest, and, by creating logical and numerical functions over these states, the objects’ “methods” can be used to model every activity of these entities. As a consequence, heterogeneity can also take the form of multiple classes of objects representing different types of entities at different levels of analysis, such as organizations and actors (see Ferber et al. [2005]). It should now be clear that “level-change” propensity, “recursivity,” “fractality,” “generalized interdependence” and “heterogeneity” have a fundamental feature in common: they are not defined with reference to the specific entities, behaviors, connections, and levels of analysis a given model may want to represent. This explains why these properties travel across ABMs in different fields. These properties are “generic” because they are transversal to models representing an infinite variety of specific entities and relationships among them. These five properties can thus be seen as the ingredients that an explanatory problem should have to justify the use of an ABM irrespective of the specific entities involved in the problem at hand. Could one defensibly argue that MI can be interpreted along similar lines?

ABM’s Generic Properties and MI The way James Coleman comments on MI in the Foundations of Social Theory (and related minor writings) constitutes an interesting starting point to assess this hypothesis. Coleman’s understanding of MI has indeed received radically different interpretations, some scholars seeing it as the best example of micro-reductionism (see, in particular, the harsh critique by Jepperson and Meyer [2011]) while others reading it as the most structuralist variant of MI that admits “social wholes, in the form of structures of interrelated positions, which exist independently of the particular individuals who happen to occupy these positions” (Udhen 2001: 347). In this sense, as Coleman’s MI is seen to span from the “strongest” to the “weakest” forms of MI, it constitutes a useful benchmark to provide a first assessment of the presence of ABM’s generic properties across a variety of possible specific accounts of MI.11 As to “change-level” propensity, which I defined as ABM’s requirement of treating an explanatory problem by moving up and down across levels of analysis, Coleman clearly expresses this idea when he states that his mode of explanation “[…] of the behaviour of social system entails examining 11

Moreover, Coleman’s diagram inspires large portions of contemporary empirical, experimental, and formal sociology (see Hedström and Swedberg [1998], Raub et al. [2011]) and, for this reason, it can be regarded as representative of the way a majority of social scientists nowadays understands MI.

Agent-Based Models and Methodological …

701

processes internal to the system, involving its component parts, or units at a level below that of the system” (Coleman 1990: 2, 5). Some pages later, he even detaches the principle of not remaining at the same level of analysis from a specific type of entities that one should assign by construction of the “lower” level (Coleman 1990: 5).12 To the objection that Coleman would be in fact emphasizing here only one type of “level-change,” namely, “top-down” (whereas, for ABM, I stressed the equal importance of “downward” and “upward” level-changes), a response can be found in another remark where he clearly indicates that system-level regularities should be thought as generated by chains of macro-to-micro constraints and micro-to-macro aggregative events.13 This principle also clearly implies that Coleman sees moving up and down across a set of entities conceived of as located at different levels of analysis as an iterative operation, a feature that I call “recursivity” and I regarded as ABM’s second generic property.14 The third property, i.e., “fractality,” which I defined as the possibility to operate “level-changes” every time a given level of analysis is required to be decomposed into more elementary mechanisms, is also visible in Coleman’s comments on MI when he explicitly acknowledges that the concept of “system behavior” is relative, its content is not defined a priori once and for all, and the principle of looking into the internal component of a given set of entities located at a given level of analysis is independent from the specific types of entities and level at hand.15

12

Here is Coleman’s full quotation: “Furthermore, there is no implication that for a given purpose an explanation must be taken all the way to the individual level to be satisfactory. The criterion is instead pragmatic […]. This criterion will ordinarily require an explanation that goes below the level of the system as a whole, but not necessarily one grounded in individual actions and orientations” (Coleman 1990: 5). 13 Here is Coleman’s full quotation: “[…] structure at one time (macro-level) generates the conditions which together with existing interests shape the actions of actors (micro-level) that jointly produce outcomes which modify the structure of a later time (macro-level) which generates conditions that again (through constraints and incentives) shape actions (micro-level) that jointly produce outcomes (macro-level) and so on” (Coleman 1993: 63). 14 Abell and Engel also acknowledge this property of the Coleman’s boat when they note that “The ‘boat shape’ is used to imply passage of time from left to right. Furthermore, repeated diagrams whereby the exogenous macro cause is the outcome/effect of a previous cycle may be conceived” (Abell and Engel 2018; see Manzo [2007: Fig. 2] for an explicit graphical representation of idea). 15 Here is Coleman’s full quotation: “The system behaviour, the macro-level in the terminology of this book, sometimes is appropriately conceived as being merely the behavior of a system of actors whose actions are interdependent. In some case the system behavior can be regarded as the action of a supraindividual actors, for example, the action of a nation resulting from the interdependent actions of actors internal to the nation. Similar to this is the case in which the macro-level is a formal organization and the micro-level is made up of departments in the organization or person occupying positions in it” (Coleman 1990: 12).

702

G. Manzo

“Generalized interdependence,” the fourth generic property that I proposed for ABM, is the easiest feature to find within MI. ABM was imported within MI as the tool that finally enabled social scientists to handle interdependence structures potentially involved in the “transformation” of the “micro” into the “macro” (see section: The origin of the link between ABM and MI). Coleman explicitly stresses the importance of conceptualizing a variety of forms of interdependence that goes far beyond simple dyadic interactions among actors (Coleman 1990: 20–21). Moreover, he makes the important (but unnoticed) observation that, in some cases, the “macroto-micro transition” is contained within a given interdependence structure, which is in full agreement with the proposed idea that an ABM is designed to model interdependence both as an aggregation mechanism and a source of constraints. The last generic feature that I suggested to be at the heart of ABM is “heterogeneity.” Similarly to interdependence, the quest for a tool able to treat how heterogeneity dynamically travels through interdependent actions, thus affecting macroscopic patterns in not-easy-to-predict ways, was at the origin of the adoption of ABM by MI-oriented economists and sociologists. Despite the absence of an explicit representation of heterogeneity within his acclaimed diagram, Coleman’s overt attack on the metaphor of the “representative agents” (see section: The origin of the link between ABM and MI)—an analytical strategy that ignores actors’ heterogeneity—clearly suggests that he regarded this element as a crucial explanatory ingredient for MI-based theoretical models.16 In sum, Coleman’s meta-reflection on MI is clearly characterized by a number of principles that hold independently of the specific type of entities one assigns a given level of analysis, and even independently from the specific levels of analysis a given explanatory problem may involve. These principles astonishingly resemble the features—namely, “level-change” propensity, “recursivity,” “fractality,” “generalized interdependence,” and “heterogeneity”—that I suggested as the defining generic properties of ABM. As Coleman has been equally well interpreted as the most extreme “micro-chauvinist” (see Jepperson and Meyer [2011: 57]) or the most accomplished “structural individualist” (see Udehn [2001: 292–306]), it seems fair to consider him as speaking for a variety of MI accounts. In this sense, the resemblance that I found between ABM’s and MI’s generic properties within his masterwork 16

On this point, I agree with Ylikoski, who, in an extensive analysis of the Coleman’s boat, notes: “The second unrepresented but crucial idea is the heterogeneity of agents. […] Nothing in the logic of the diagram implies that we should think that the problem of micro-foundations could be resolved by intellectual compromises like representative agents” (Ylikoski 2016).

Agent-Based Models and Methodological …

703

can be regarded as a first test of the working hypothesis that ABM’s generic properties are transversal to “strong” and “weak” forms of MI, and that, consequently, MI can legitimately be conceived in terms of (these) generic properties (i.e., as a generic instrument) rather than in terms of specific entities (and levels of analysis).

6

Concluding Remarks

The rapid diffusion of ABM across disciplines over the last two decades has been scrutinized from various technical, methodological, and epistemological points of view. In this paper, I focused on one of the most recent debates fueled by a set of articles systematically studying the soundness of the link that computational modelers posited between ABM and MI. I argued that the answer to this question is likely to continue to be controversial as long as the analysis starts with a “specific” understanding of MI, i.e., any interpretation of MI that, firstly, gives a specific content the entities that can be legitimately introduced within an explanatory model, and, second, argues in favor of the analytical primacy of these entities. When this is done, a cascade of issues that have historically proved to be difficult, if not impossible, to be settled arises, so that it suffices to change the premise—i.e., the chosen “specific” definition of MI—to challenge the conclusion—i.e., the claimed/denied existence of a connection between ABM and MI. To overcome this pitfall, I took inspiration from the concept of “generic instruments” that sociologists of science and technology forged to describe mechanical and symbolic devices that have historically proved to be able to travel across application domains because of the flexibility of the principles and rules of functioning on which they were based (see Shinn [2008]). I proposed to look into ABM and MI as “generic instruments” potentially sharing similar generic properties, meaning, in my case, properties that are defined without reference to the specific entities (and levels of analysis) a given model may want to represent. Thus, starting from the empirical observation that ABMs across disciplines involve an astonishing variety of specific entities, behaviors, connections, and (relationships between) levels of analysis, I identified “level-change” propensity, “recursivity,” “fractality,” “generalized interdependence,” and “heterogeneity” as those generic properties that recur most frequently across ABMs with very different specific contents. Then, I moved on to MI and considered the possibility that this perspective can be characterized by the same generic properties. In particular, I inspected James Coleman’s MI account and

704

G. Manzo

I have shown that this understanding of MI is in fact explicitly based on the same five generic ingredients that recur within ABMs. As Coleman’s MI has received diametrically opposing interpretations with respect to the entities it is supposed to accept and the hierarchy among levels of analysis it is supposed to admit, thus arguably speaking for “strong” as well as “weak” forms of MI, I finally claimed that it seems realistic to formulate the hypothesis that ABM’s generic properties are transversal to various forms of MI and that, consequently, ABM can (and should be) legitimately seen as a powerful technical support for the analysis of theoretical models conceived in MI terms. I see two potential objections to this line of thought. First, one may object to my approach that in fact it is also based on a specific understanding of MI (and of ABM). The subtle relevant distinction here is between “particular” (or distinctive) and “specific.” I certainly proposed an account of MI that is particular in the sense of being different from the way MI is usually described. However, this particular interpretation is not “specific” in the sense I used the term in this paper: my particular view on MI is not based on any claims concerning the type of entities (and level of analysis) that one should admit to remain within the realm of MI. The “genericity” of my account of MI thus makes it at the same time “particular” (in the sense of being distinctive) and transversal (in the sense of generalizable) to various “specific” interpretations of MI. And here comes the second objection. One may indeed retort that my analysis did not prove the compatibility of ABM’s generic properties and MI across its multiple “strong” and “weak” versions, which weakens the claimed generality of the proposed interpretation of MI in terms of generic properties. I am obviously well aware that, given the variety of accounts MI has received, in order to establish whether ABM’s generic properties are compatible with MI, one should in principle scrutinize every single proposed interpretation of MI. For this reason, as the production of a full empirical proof of this kind was out of my reach, I insisted several times that my modest goal was to formulate a working hypothesis that historically minded philosophers of science and as well as historians of ideas and social theory may test in a more systematic way than I was able to do in this study. For the time being, the reader could at least be ready to follow me in stressing the heuristic value of this working hypothesis. Conceiving MI and ABM as “generic instruments” indeed has the potential of teaching us something new about both ABMs and MI rather than obliging us to rehash old and endlessly controversial issues. As to the ABM, the focus on its generic properties forces us to consider that, while the link between ABMs and MI was historically created within this perspective to solve a specific methodological problem—i.e., the transition from interdependent actions to their

Agent-Based Models and Methodological …

705

unintended consequences—, ABM is a method that can be used to solve a much broader set of explanatory problems, where the entities involved are not necessarily human actors and the transition from the micro to the macro-level is not necessarily the main focus of the analysis. Without these generic lenses, it would not be possible to account for the impressive variety of specific entities, behaviors, connections, and (relationships between) levels of analysis that ABM modelers have been able to treat across fields and disciplines. On the other hand, with respect to MI, the focus on its generic properties makes it possible to see that MI has a larger scope than sociological MI. Although this perspective has historically been forged within the social sciences to frame the explanatory problem of social order (see Udhen [2001: Chap. 2]), many aspects of physical and biological life, too, require thinking in terms of the mechanisms that create complex and dynamic connections between the parts and the whole (see Lieberson and Lynn [2002]). Thus, similar to ABM, to stop identifying MI with a specific set of entities and behaviors helps us to see how MI can travel across fields and disciplines.

References Abbott, A. (1988). Transcending general linear reality. Sociological Theory, 6 (2), 169–186. Abbott, A. (1990). Conceptions of time and events in social science methods: causal and narrative approaches. Historical Methods, 23(4), 140–150. Abell, P. (2003). The role of rational choice and narrative action theories in the sociological theory. The legacy of Coleman’s Foundations. Revue française de sociologie, 44(2), 225–273. Abell, P., Engel O. (2018). The Coleman Diagram, Small N Inquiry and Ethnographic Causality. Sociologica, 12(2), 115–125. Ajelli, M., Gonçalves, B., Balcan, D., Colizza, V., Hu, H., Ramasco, J. J., Merler, S., Vespignani, A. (2010). Comparing large-scale computational approaches to epidemic modeling: Agent-based versus structured metapopulation models. BMC Infectious Diseases, 10, article 190. Barthélemy, J., Carletti, T. (2017). An adaptive agent-based approach to traffic simulation. Transportation Research Procedia, 25, 1238–1248. Bianchi, F., Squazzoni F. (2015). Agent-based models in sociology. Wires Computational Statistics, 7, 284–306. Billari, F., Fürnkranz-Prskawetz, A., Diaz, B. A., Fent, T. (2007). The “weddingring”, an agent-based marriage model based on social interaction? Demographic Research, 17, article 3, 59–82. Birks, D., Townsley M., Stewart A. (2012). Generative explanations of crime: using simulation to test criminological theory. Criminology, 50(1), 221–254.

706

G. Manzo

Bora, S¸ ., Evren, V., Emek, S., Çakırlar, I. (2019). Agent-based modeling and simulation of blood vessels in the cardiovascular system. Simulation, 95(4), 297–312. Boudon, R. (1986). Theories of Social Change: A Critical Appraisal . Cambridge: Polity Press. Boudon, R. (1998). Social mechanisms without black boxes. In P. Hedström, R. Swedberg (Eds.), Social Mechanisms. An Analytical Approach to Social Theory (pp. 172–203). Cambridge: Cambridge University Press. Boudon, R. (2014). What is context? Kölner Zeitschrift für Soziologie und Sozialpsychologie, 66(1) (supp.), 17–45. Bulle, N., Phan D. (2017). Can analytical sociology do without methodological individualism? Philosophy of the Social Sciences, 47(6), 379–409. Coleman, J. S. (1986). Social theory, social research and a theory of action. American Journal of Sociology, 91(6), 1309–1335. Coleman, J. S. (1987). Microfoundations and macrosocial behavior. In J. Alexander, B. Giesen, R. Münch, N. Smelser (Eds.), The Micro- Macro Link (pp. 153–177). Berkeley/Los Angeles : University of California Press. Coleman J. S. (1990). Foundations of Social Theory. Cambridge: Harvard University Press. Coleman, J. S. (1993). Reply to Blau, Tuomela, Diekmann and Baurmann. Analyse und Kritik. De Marchi, S., Page, S. E. (2014). Agent-based models. Annual Review of Political Science, 17, 21–42. DellaVigna, S. (2009). Psychology and economics: evidence from the field. Journal of Economic Literature, 47(2), 315–372. Di Iorio, Fr., Chen S.-H. (2019). On the connection between agent-based simulation and methodological individualism. Social Science Information, 58(2), 354–376. Eberlen, J., Scholz, G., Gagliolo, M. (2017). Simulate this! An introduction to agent-based models and their power to improve your research practice. International Review of Social Psychology, 30(1), 149–160. Elsenbroich, C. (2012). Explanation in agent-based modelling: functions, causality or mechanisms? Journal of Artificial Societies and Social Simulation, 15(3), article 1. Epstein, J. M. (2007). Generative Social Science. Studies in Agent-Based Computational Modelling. Princeton: Princeton University Press. Farmer, J. D., Foley, D. (2009). The economy needs agent-based modelling. Nature, 460(6), 685–686. Fehr, E., Gintis, H. (2007). Human motivation and social cooperation: experimental and analytical foundations. Annual Review of Sociology, 33(1), 43–64. Ferber, J., Michel, F., Baez, J. (2005). AGRE: integrating environments with organizations. In D. Weyns, H. Van Dyke Parunak, F. Michel (Eds.), Environments for Multi-Agent Systems, (pp. 48–56). Berlin, Heidelberg: Springer.

Agent-Based Models and Methodological …

707

Flache, A., Mäs, M., Feliciani, T., Chattoe-Brown, E., Deffuant, G., Huet, S., Lorenz, J. (2017). Models of social influence: towards the next frontiers. Journal of Artificial Societies and Social Simulation, 20(4), article 2. Gigerenzer, G., Gaissmaier, W. (2011). Heuristic decision making. Annual Review of Psychology, 62, 451–482. Goldthorpe, J. H. (2006). On Sociology (Second Edition), vol. 1: Critique and Program, Stanford: Stanford University Press. Gonzales-Bailon, S., Murphy, T. E. (2013). The effects of social interactions on fertility decline in the nineteenth-century France: An agent-based simulation experiment. Population Studies, 67(2), 135–155. Grow, A., Van Bavel, J. (Eds.) (2016). Agent-Based Modelling in Population Studies: Concepts, Methods, and Applications, Berlin: Springer. Hayward, S. (2006). Agent-based modelling with wavelets and an evolutionary artificial neural network: applications to CAC 40 forecasting. In A. Chatterjee, B. K. Chakrabarti (Eds.), Econophysics of Stock and Other Markets (pp. 163–174). Milano: Springer. Heckbert, S. (2013). MayaSim: an agent-based model of the ancient Maya socialecological system. Journal of Artificial Societies and Social Simulation, 16(4), article 11. Hedström, P. (2005). Dissecting the Social: On the Principles of Analytical Sociology, Cambridge: Cambridge University Press. Hedström, P., Swedberg, R. (1998). Social Mechanisms: an introductory essay. Social Mechanisms. An Analytical Approach to Social Theory (pp. 1–31). Cambridge: Cambridge University Press. Hokamp, S., Gulyás, L., Koehler, M., Wijesinghe, S. (2018). Agent-based modeling and tax evasion: theory and application. In S. Hokamp, L. Gulyás, M. Koehler, S. Wijesinghe (Eds.), Agent-Based Modeling of Tax Evasion: Theoretical Aspects and Computational Simulations (pp. 1–36). Chichester: Wiley and Sons. Hosseinali, F., Alesheikh, A. A., Nourian, F. (2013). Agent-based modeling of urban land-use development, case study: Simulating future scenarios of Qazvin city. Cities, 31, 105–113. Hummon, N. P., Fararo, T. J. (1995). Actors and networks as objects. Social Networks, 17(1), 1–26. Jepperson, R., Meyer, J. W. (2011). Multiple levels of analysis and the limitations of methodological individualisms. Sociological Theory, 29(1), 54–73. Joerges, B., Shinn, T. (Eds.) (2001). Instrumentation Between Science, State and Industry, Dordrecht: Kluwer Academic. Joerges, B. & Shinn, T. (2002). The transverse science and technology culture: dynamics and roles of research technology. Social Science Information, 41(2), 207–251. Kerridge, J., Hine, J., Wigan, M. (2001). Agent-based modelling of pedestrian movements: the questions that need to be asked and answered. Environment and Planning B: Planning and Design, 28(3), 327–341.

708

G. Manzo

Kirman, A. P. (1992). Whom or what does the representative individual represent? Journal of Economic Perspectives, 6(2), 117–136. Kroneberg, C. (2019). Theory development in comparative social research. Kölner Zeitschrift für Soziologie und Sozialpsychologie, 71(1), 29–51. Lieberson, S., Lynn, F. B. (2002). Barking up the wrong branch: scientific alternatives to the current model of sociological science. Annual Review of Sociology, 28, 1–19. Little, D. (2016). New Directions in the Philosophy of Social Science, London: Rowman & Littlefield International. Lux, T., Marchesi, M. (1999). Scaling and criticality in a stochastic multi-agent model of a financial market. Nature, 397, 498–500. Macy, M. W., Willer, R. (2002). From factors to actors: computational sociology and agent-based modeling. Annual Review of Sociology, 28, 143–166. Macy, M. W., Flache, A. (2009). Social dynamics from the bottom up: agent-based models of social interaction. In P. Hedström, P. Bearman (Eds.), The Oxford Handbook of Analytical Sociology (pp. 245–268). Oxford: Oxford University Press. Macy, M. W., Centola, D., Flache, A., van de Rijt, A., Willer, R. (2011). Social mechanisms and generative explanations: computational models with double agents. In P. Demeulenaere (Ed.), Analytical Sociology and Social Mechanisms (pp. 250–265). Cambridge: Cambridge University Press. Malleson, N. (2012). Using agent-based models to simulate crime. In N. Malleson, A. J. Heppenstall, A. Crooks, L. M. See, M. Batty (Eds.), Agent-Based Models of Geographical Systems (pp. 411–434). Dordrecht: Springer. Manzo, G. (2007). Comment on Andrew Abbott/2. Sociologica, 1(2), 1–8. Manzo, G. (2010). Analytical sociology and its critics. European Journal of Sociology, 51(1), 129–170. Manzo, G. (2013). Educational choices and social interactions: a formal model and a computational test: Class and Stratification Analysis, Comparative Social Research, 30, 47–100. Manzo, G. (2014). The potential and limitations of agent-based simulation. An introduction. Revue française de sociologie, 55(4), 653–688. Manzo, G., Gabbriellini, S., Roux, V., M’Mbogori, F. N. (2018). Complex contagions and the diffusion of innovations: evidence from a small-N study. Journal of Archaeological Method and Theory, 25(4), 1109–1154. Marchionni, C., Ylikoski, P. (2013). Generative explanation and individualism in agent-based simulation. Philosophy of the Social Sciences, 43(3), 323–340. Marshall, B. D., Galea, S. (2014). Formalizing the role of agent-based modeling in causal inference and epidemiology. American Journal of Epidemiology, 181(2), 92–99. Metzcar, J., Wang, Y., Heiland, R., Macklin, P. (2019). A review of cell-based computational modeling in cancer biology. JCO Clinical Cancer Informatics, 3, 1–13.

Agent-Based Models and Methodological …

709

Miller, J., Page, S. E. (2007). Complex Adaptive Systems: An Introduction to Computational Models of Social Life. Princeton: Princeton University Press. Müller, B. et al. (2014). Standardised and transparent model descriptions for agentbased models: current status and prospects. Environmental Modelling & Software, 55, 156–163. Neugart, M., Richiardi, M. (2018). Agent-based models of the labor market. In S.H. Chen, M. Kaboudan, Y.-R. Du (Eds.), The Oxford Handbook of Computational Economics and Finance (pp. 667–687). Oxford: Oxford University Press. Neumann, M. (2008). Homo socionicus: a case study of simulation models of norms. Journal of Artificial Societies and Social Simulation, 11(4), article 6. Norton, K. A., Gong, C., Jamalian, S., Popel, A. S. (2019). Multiscale agent-based and hybrid modeling of the tumor immune microenvironment. Processes (Basel), 7(1), 37. O’Sullivan, D. (2008). Geographical information science: agent-based models. Progress in Human Geography, 32(4), 541–550. O’Sullivan, D., Haklay, M. (2000). Agent-based models and individualism: is the world agent-based? Environment and Planning A: Economy and Space, 32(8), 1409–1425. Padgett, J. F., Powell, W. W. (2012). The problem of emergence. In J. F. Padgett, W. W. Powell (Eds.), The Emergence of Organizations and Markets (pp. 1–29). Princeton: Princeton University Press. Railsback, S. F. et al . (2017). Improving execution speed of models implemented in NetLogo. Journal of Artificial Societies and Social Simulation, 20(1), article 3. Raub, W., Buskens, V., Van Assen, M. A. L. M. (2011). Micro-macro links and microfoundations in sociology. The Journal of Mathematical Sociology, 35(1–3), 1–25. Raub, W, & Voss, T. (2017). Micro-macro models in sociology: antecedents of Coleman’s diagram. In B. Jann, W. Przepiorka (Eds.), Social Dilemmas, Institutions, and the Evolution of Cooperation (pp. 11–36), Berlin/Boston: De Gruyter. Sawyer, R. K. (2004). The mechanisms of emergence. Philosophy of the Social Sciences, 34(2), 260–282. Schulze, J., Müller, B., Groeneveld, J., Grimm, V. (2017). Agent-based modelling of social-ecological systems: achievements, challenges, and a way forward. Journal of Artificial Societies and Social Simulation, 20(2), article 8. Shinn, T. (2007). When is simulation a research technology? Practices, markets and lingua franca. In J. Lenhard, G. Küppers, T. Shinn (Eds.), Simulation: Pragmatic Constructions of Reality (pp. 187–203). Dordrecht: Springer. Shinn, T. (2008). New sources of radical innovation: research-technologies, transversality and distributed learning in a post-industrial order. Social Science Information, 44(4), 731–764. Smith, J. A., Burow, J. (2018). Using ego network data to inform agent-based models of diffusion. Sociological Methods and Research [online].

710

G. Manzo

Sperber, D. (2011). A naturalistic ontology for mechanistic explanations in the social sciences. In P. Demeulenaere (Ed.), Analytical Sociology and Social Mechanisms (pp. 64–77). Cambridge: Cambridge University Press. Tang, W., Bennett, D. A. (2010). Agent-based modeling of animal movement: a review. Geography Compass, 4(7), 682–700. Thiele, J. C., Kurth, W., Grimm, V. (2014). Facilitating parameter estimation and sensitivity analysis of agent-based models: a cookbook using NetLogo and R. Journal of Artificial Societies and Social Simulation, 17(3), article 11. Tracy, M., Cerdá, M., Keyes K. M. (2018). Agent-based modeling in Public Health: current applications and future directions. Annual Review of Public Health, 39, 77–94. Udehn, L. (2001). Methodological Individualism: Background, History and Meaning, London: Routledge. Udehn, L. (2002). The changing face of methodological individualism. Annual Review of Sociology, 28, 479–507. Wan, P. Y. (2011). Emergence à la systems theory: epistemological totalausschluss or ontological novelty? Philosophy of the Social Sciences, 41(2), 178–210. Wilensky, U., Rand, W. (2015). An Introduction to Agent-based Modeling. Modeling Natural, Social, and Engineered Complex Systems with NetLogo. Cambridge: MIT Press. Wippler, R. (1978). The structural-individualistic approach in Dutch sociology: toward and explanatory social science. The Netherlands Journal of Sociology, 14(2), 135–155. Wippler, R., Lindenberg, S. (1987). Collective phenomena and rational choice. In J. Alexander, B. Giesen, R. Münch, N. Smelser (Eds.), The Micro-Macro Link (pp. 135–153). Berkeley/Los Angeles: University of California Press. Wurzer, G., Kowarik, K., Reschreiter, H. (2015), Agent-based Modeling and Simulation in Archaeology. Dordrecht: Springer. Ylikoski, P. (2016). Thinking with the Coleman Boat, IAS Working Paper, Series, n° 1. Linköping: Linköping University. Zahle, J., Kincaid H. (2020). Agent-based modelling with and without methodological individualism. In H. Verhagen, M. Borit, G. Bravo, N. Wijermans (Eds.), Advances in Social Simulation. Looking in the Mirror. Berlin: Springer. Zheng, X., Zhong, T., Liu, M. (2009). Modeling crowd evacuation of a building based on seven methodological approaches. Building and Environment, 44(3), 437–445.

A Reply to Manzo: The Role of Methodological Individualism for Analytical Sociology Nathalie Bulle

In our article “Can Analytical Sociology Do Without Methodological Individualism?” Denis Phan and I (2017) argue that (a) from an explanatory point of view, the syntactic properties of models require semantic conditions of interpretation drawn from a conceptual research framework; (b) in such a framework, social structures have only partial, explanatory power (counterfactual); and (c) taking the explanation further through generative mechanism modeling necessitates calling upon methodological individualism’s generic framework of interpretation. According to this interpretive framework, forces in action in society are governed by the subjective meaning of/the reasons for individual actions. Gianluca Manzo (2023, 2020, this volume) objects to this, saying that rationality, which is associated to the understanding interpretation of methodological individualism (MI), cannot be taken to be trans-situational. This reply argues that either (1) this is a misunderstanding resulting from a confusion between the general rational capacity of human beings, to which Bulle and Phan refer, and the specific behavioral patterns associated with given situations and models or (2) this objection supports a structural reductionism in which the structural level tends to “reduce” the individual level, so that the behavioral patterns are directly implied by the structures (on this N. Bulle (B) National Center for Scientific Research and Sorbonne University, Paris, France e-mail: [email protected]

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 N. Bulle and F. Di Iorio (eds.), The Palgrave Handbook of Methodological Individualism, https://doi.org/10.1007/978-3-031-41512-8_31

711

712

N. Bulle

subject authors speak, for example, of “embodied structures”) and therefore, these patterns are not rationally anchored: “They [Bulle and Phan] consider indeed that actors’ rationality has the property of being ‘trans-situational’, a property that Bulle and Phan in turn see as the necessary feature that the entities represented by a given ABM (Agent Based Model) must possess in order to make the mechanisms postulated by the model generalizable (ibid., p. 399, 402, 404, proposition E).” In the mentioned article, as the following excerpts show, no reference is made to any normative, “generalizable” model of rationality. By rational capacity, we simply refer to the capacity of human subjects that underlies their ability to understand and reason, the one generally attributed to homo sapiens, based on the human ability to abstract and manipulate conventional signs for attributing meaning to their situation (see Bulle 2022). To make the discussion clear, we will return to the excerpts from our article referenced by Manzo. First, we defend that the developments of the conceptions of human rational capacity in the social sciences were made in the direction of a broadening of the understanding of motivations from instrumental ends toward axiological ends: [p. 399: Advances in the conception of human rationality reveal it to be not only a capacity that is transcultural in its essence, with “reasons” varying according to the auxiliary means for thought that individuals possess, but also a general capacity that is not reserved for utilitarian reasoning and includes in its scope even the deepest individual purposes (Boudon 1999/2001)]

Moreover, we defend that explanatory models (which are of a conceptual nature) refer indirectly to real (trans-situational by definition) causal powers: [p. 398: Causality, conceived as a methodological principle necessitating isolated systems, is not only consistent with the active approach to causation— that is, the idea of “trans-situational” causal power—but strengthens on this basis. Whereas the causality principle applies to the representation/description of things, the idea of causal power applies to the things represented.]

Therefore, the more the explanation aims at causal realism, the more the models’ validity tends to be freed from ceteris paribus conditions. The reference to the causal power of the entities involved allows us to explain in greater depth the phenomenon in play and the trans-situational anchoring of this causal power allows us to apply the causal system to other particular conditions:

A Reply to Manzo: The Role of Methodological …

713

[p. 402: We note that the reference to rationality as a trans-situational capacity allowing understanding of the meaning of/reasons for actions for the methodological individualist not only deepens the explanation by virtue of the range of possible “interventions.” The explanation is deepened by virtue of the deduction allowed of the various regularities/patterns observed through the valid possibility of “closing” the explanatory system, of detaching it from the different contexts of “intervention,” and, also, of combining different models].

Referring to real processes, the model can thus represent generating mechanisms which, contrary to a simple mechanism, tends to uncover the causes that genuinely produce a given phenomenon: [p. 404 (proposition D): A generative mechanism explanation is a causal explanation involving the (trans-situational) capacities/powers of the active entities in play.]

The consequence is that the generating mechanisms represented by multiagent models, indirectly referring to real causes, must use action models involving a rational capacity whose scope is ultimately trans-situational: [p. 404 (proposition E): One necessary condition for agent-based models to model generative mechanisms is that the properties of the active entities in play in the conceptual modeling framework are associated, in the thematic conceptual framework, with (trans-situational) capacities/power].

Contrary to what Manzo assumes, there is no question of invariance and possible generalization of determined patterns of action (Manzo 2020, p. 206, note 3: The concept of “trans-situationality” is implicitly defined in terms of “invariance” but it is unclear in what sense and on which basis the authors claim that “rationality” is “invariant”). In this article, we oppose the claims of analytical sociologists to be able to stick to structural logics. The structural logics are no more explanatory than the nomological machines as Cartwright calls them, their limitation to “ceteris paribus” conditions is precisely what in essence Kroneberg (2019), whom Manzo cites in a note, denounces. In order for the multi-agent models developed to be able to claim an explanation that goes beyond the counterfactual properties of the causal structures involved, the behavioral principles put into play must be able to be modified according to the circumstances in such a way as to be able to always refer (indirectly) in reality to the same rational capacity of individuals (which, for example, is evoked by methodological individualists through the notion of “good reasons,” which is likely to allow an understanding of the contextual variation of reasons). It is only in

714

N. Bulle

this case that the models can move from a partial explanation, of a counterfactual nature referring to structures without any understanding of their impact on the entities’ behavior, to an explanation revealing the generating mechanisms. If a condition of generalization of behavioral models operates with respect to explanation, it is therefore due to a process of abstraction, allowing their adaptation to other circumstances and, in so doing, the constancy of the reference to a rational capacity of a trans-situational nature, and certainly not due to any invariance of the behavioral models used in computation. As Phan and I further explain (2017, pp. 399–400), which may highlight our point: Works referring to mentalities, dispositions toward, and habitus, for instance, that assume there to be no trans-situational, rational common ground, prevent us from combining explanatory models that are concerned with different domains of social action. The conception of rationality as a trans-situational capacity accounts for the very meaning of interpretive/comprehensive sociology, through the relationship of possible understanding between the observer and the observed. This does not mean that the meaning of/reasons for individual actions might be easily accessible to the observer, or even that the social actor might be fully aware of them, but that ideally, contextual information can render this meaning/these reasons sufficiently clear that the observer might grasp them in an internal way, through his or her own means, such as through empathy. On this subject, the concept of rationality does not play a normative role, but an epistemological one: it enables us to judge the relative relevance of alternative explanations based on the allowed understanding of the meaning of/ reasons for individual actions. Hence, Proposition F is suggested: Proposition F: The rational (in the broad sense) tendency that underpins the meaning of/ reasons for individual actions for methodological individualism represents the type of trans-situational capacity that allows the representation of generative mechanisms in the social sciences.

Challenging the (indirect) reference of multi-agent models to the transsituational quality of individuals’ rational capacity seems to us to be either a misunderstanding or a negation of the subjectivity of the social actor, which could feed an important discussion that cannot be developed here.

References Boudon, R. (1999/2001). The Origin of Values: Philosophy and Sociology of Beliefs. Piscataway: Transaction Publishers.

A Reply to Manzo: The Role of Methodological …

715

Bulle, N., Phan, D. (2017). Can Analytical Sociology Do Without Methodological Individualism? Philosophy of the Social Sciences, 47 (6), 379–409. Bulle, N. (2022). Rationality as a meta-analytical capacity of the human mind: From the social sciences to Gödel. Philosophy of the Social Sciences. Kroneberg, C. (2019). Theory development in comparative social research. Kölner Zeitschrift für Soziologie und Sozialpsychologie, 71(1), 29–51. Manzo, G. (2023/2020). Agent-Based Models and Methodological Individualism: are they fundamentally linked? L’Année sociologique, 70 (1), 197–229.

Methodological Individualism and Agent-Based Models: A Reply to Manzo Francesco Di Iorio

1

Introduction

In this chapter, I focus on the stimulating and enriching contribution published by Gianluca Manzo in this handbook under the title “Agent-based models and methodological individualism: are they fundamentally linked?”1 The purpose is to reply to the constructive criticism that he levelled at an article entitled “On the Connection Between Agent-Based Simulation and Methodological Individualism” that I coauthored with Shu-Heng Chen (Di Iorio and Chen 2019). Manzo agrees with the main conclusion of our article, which is that there is a clearly identifiable link between these two approaches. However, he disagrees with the line of reasoning that we have developed to support this conclusion. My objective is showing that Manzo’s criticism is based on a misunderstanding of our argumentative strategy that may stem from the fact that we did not clarify some implicit relevant methodological assumptions of our analysis. In the following pages, I shall attempt to elucidate and make explicit these assumptions to demonstrate that the distance 1 I would like to thank Nathalie Bulle, Gianluca Manzo and Enzo Di Nuoscio for providing valuable remarks on the topic of this chapter. All errors remain mine.

F. Di Iorio (B) Philosophy, Nankai University, Tianjin, China e-mail: [email protected]

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 N. Bulle and F. Di Iorio (eds.), The Palgrave Handbook of Methodological Individualism, https://doi.org/10.1007/978-3-031-41512-8_32

717

718

F. Di Iorio

between Manzo’s and our line of reasoning is less than it may appear at first sight.

2

Manzo’s Criticism of Our Article

In his chapter Agent-based models and methodological individualism: are they fundamentally linked?, which was first published in a special issue of L’Année sociologique edited in 2020 by Nathalie Bulle and then republished in this volume, Gianluca Manzo maintains that methodological individualism and agent-based simulation are linked using innovative and penetrating arguments. The conclusion of his analysis matches well with the central thesis of an article that I coauthored with Shu-Heng Chen entitled “On the Connection Between Agent-Based Simulation and Methodological Individualism” that was published in Social Science Information in 2019. However, while Manzo agrees with our central thesis, he disagrees with the line of argument that Shu-Heng Chen and I have developed to support this thesis. Manzo’s article elaborates a complex and articulated analysis that explores a variety of theoretical dimensions and focuses on different topics and authors related to the debates on the agent-based models. I am not going to discuss here the whole of his article, but only the criticism Manzo levels at Shu-Heng Chen and myself with the purpose of showing that this criticism is based on a misunderstanding of our position. I recognize that this misunderstanding may stem from the fact that Chen and I did not clarify some implicit methodological assumptions of our approach. As a consequence, in the next section, I shall attempt to fix this problem by making these assumptions explicit. However, I would like first to shortly summarize our article and Manzo’s criticism of it. In our article Shu-Heng Chen and I investigate the relationship between methodological individualism and agent-based models using a thesis defended by Caterina Marchionni and Petri Ylikoski (2013) as the starting point of our analysis. This thesis can be summarized as follows: since methodological individualism is often considered to be a form of naïve reductionism, implying that the social world is not characterized by systemic or structural properties, it is confusing and meaningless to assume that agent-based simulation, which is a non-reductionist and emergentist approach, is committed to methodological individualism (ibid.). While Marchionni and Ylikoski agree with the interpretation of methodological individualism in terms of naïve reductionism that is dominant in analytic philosophy of the social sciences,

Methodological Individualism and Agent-Based …

719

in their article they do not clarify their reasons for doing so. They maintain that it is irrelevant to clarify these reasons because, for their argument, what matters is only what is commonly understood by methodological individualism by analytic philosophers and their followers. In other words, they suggest breaking down the relationship between methodological individualism and agent-based models “because of practical and utilitarian reasons” (Di Iorio and Chen 2019, p. 355). Shu-Heng Chen and I (ibid.) criticize this view on the ground that to avoid confusion regarding agent-based simulation, the problem of the relationship between methodological individualism and the former cannot be analyzed by assuming the dominant philosophical interpretation of methodological individualism “to be the only criterion that must be considered.” We contend that this interpretation “is misleading and philologically incorrect” (ibid.). On this basis, we provide an alternative and more charitable interpretation of methodological individualism, which we consider “historically more accurate than the dominant one,” and assess the compatibility between our interpretation of this approach and agent-based simulation (ibid.). In our opinion, the former is compatible with the latter because “reductionism is only the most simplistic variant” of individualist analysis—a variant that none of the main theorists of methodological individualism endorsed and implemented (ibid.). We clarify that agent-based explanations must be regarded as explanations in terms of “non-reductionist methodological individualism” (ibid.). According to Shu-Heng Chen and myself, methodological individualism is often criticized on the basis of strawman arguments. One of the problems is that, since large sectors of the contemporary philosophy of the social sciences are disconnected from the history of the ideas and the practice of science and are thus pointlessly abstract, they do not focus on this explanatory model in the proper way, i.e., on the basis of a careful study of the works of its many representatives. In our article Shu-Heng Chen and I (ibid.) argue that the historical analysis clearly shows that, while approaches like social contract theories and economic equilibrium models are reductionist forms of individualism that are rooted in a mistaken mechanistic philosophy, methodological individualism as understood by the Mengerian tradition (e.g., Mises, Hayek, Popper) and interpretative sociology (e.g., Weber, Simmel, Coleman, Boudon), as well as by their Scottish Enlightenment precursors like Adam Smith, is a non-atomistic and structural orientation. This orientation should not be confused with the abovementioned non-empirical and unrealistic approaches. As pointed out by the representatives of the non-reductionist individualist tradition, the fact that society is more than a mere sum and that it is thus characterized by emergent properties is trivially true. This fact

720

F. Di Iorio

entails that the analysis of the social world and the constraints that this world imposes on individuals can only be implemented in structural or systemic terms (Popper 1957, p. 82; see also, for example, Boudon 1971; Coleman 1990; Hayek 1967, 1973, 1978; Menger 1985 [1883]). Shu-Heng Chen and I (2019, p. 260) recognize that, given that philosophers often regard methodological individualism, in all its variants, as committed to a pointless naïve reductionism, the risk of conceptual confusion that can be caused, in certain intellectual sectors, by arguing that methodological individualism is compatible with the emergentist and structural assumptions of agent-based simulation is real. However, we maintain that this risk depends on a lack of historical knowledge about methodological individualism, whose main theorists never endorsed a simplistic and atomistic view. We also argue that in order to prevent ambiguities, from a practical standpoint, this risk “can be avoided without breaking down the association between agent-based simulation and methodological individualism” by stressing that there are two different variants of the latter “and taking care to define agent-based simulation as a non-reductionist methodological individualism rather than simply as methodological individualism” (ibid.). As pointed out above, Manzo (2020, p. 2006) agrees with us that it is correct to argue that agent-based models and methodological individualism are essentially linked, but he maintains that the line of reasoning developed by Chen and myself is “debatable” because of our conviction that it is possible to demonstrate that one (or a few) of the existing variants of MI [methodological individualism] is (are) “philologically” and “historically correct…Who has the skills to establish the proper variant of MI? What selection of writings should be included within the analysis? Which disciplines should we consider? What periods of time? From a methodological point of view, the idea of the existence of a correct version of MI seems problematic” (id., pp. 206-207).

This is because it is impossible “to reach a stable consensus, among scholars and across disciplines, once and for all, on what MI really is” (p. 207). Following Udehn (2001, 2002), Manzo (2020, p. 209) argues that our view underestimates. the difficulty of reaching a consensus on how MI should be interpreted. Detailed historical investigations of MI concluded that…methodological individualism exists in a bewildering number of different versions…This is not to deny that progresses in understanding what MI is can be made. But this progress may realistically come precisely from accepting the existence of a

Methodological Individualism and Agent-Based …

721

variety of versions of MI and the absence of a…true version of this doctrine (Ibid.).

3

Our Fallibilist Interpretation of Methodological Individualism

Let me clarify now why the distance between Manzo’s and our line or reasoning is less than it may appear at first sight by making more explicit the epistemological assumptions of the latter. Our article did not aim at showing that there is only one variant of methodological individualism that should be regarded as the true implementation of this approach, but, on the contrary, that the dominant definition of methodological individualism in analytic philosophy is untenable precisely because, as mentioned above, there are different historical implementations of methodological individualism, many of which are clearly non-reductionist. It is important to understand that, in our analysis, Chen and I were not inspired by any form of dogmatism and essentialism, but by an approach that can be accounted for in terms of Popper’s fallibilism. According to Popper (1980, p. 353), the science of textual interpretation (namely, classical hermeneutics), is based on the same empirical logic of the natural sciences, i.e. on the method of trial and error advocated by fallibilism (see also Antiseri 2006; Di Nuoscio 2018; Grondin 2002; Mantzavinos 2006; Ricoeur 2007). From this epistemological standpoint, while no absolute and perfect truth about methodological individualism can be grasped and made manifest, an account describing the general features of this intellectual tradition needs to be objective and nonarbitrary in the sense that it must be falsifiable and not falsified (it can be provisionally acceptable). Such an account is a conjecture referring, among other things, to the content of the texts written by the advocates of methodological individualism that may be either compatible or incompatible with this content. This conjecture can be empirically rejected because of this content. From the standpoint of this fallibilist view, to conclude that there is something wrong with a general account of methodological individualism it is unnecessary to consider all the texts written by its advocates: all its different implementations. It is sufficient that the content of at least one or some of their texts is in conflict with the account under investigation. Of course, as Popper (2002, 1959], p. 50) recognized, falsifications are not certain because of the Duhem-Quine problem, but, as long as it is not proven that the falsification depends on the falsity of a background assumption erroneously regarded as correct, the falsification must be accepted.

722

F. Di Iorio

Let’s consider, for example, the following statement by Coleman (1990 p. 5), in which he clarifies what is meant by methodological individualism in his book Foundations of Social Theory: No assumption is made that the explanation of systemic behavior consists of nothing more than individual actions and orientations, taken in aggregate. The interaction among individuals is seen to result in emergent phenomena at the system level, that is, phenomena that were neither intended nor predicted by the individuals. Furthermore, there is no implication that for a given purpose an explanation must be taken all the way to the individual level to be satisfactory.

This statement is clearly a falsification of the widespread view that methodological individualism, in all its variants, is at odds with emergentism and a systemic approach (and, as a consequence, with agent-based models too). Since, as Chen and I show, similar statements can be found in many other theorists of methodological individualism (e.g., Menger, Hayek, Popper, Crozier, Boudon), we have abundant empirical evidence that there is something wrong with this dominant view. Our conclusion is not “problematic,” to take up an expression used by Manzo, because its objective validity can be easily checked through textual analysis. Note that this empirical evidence shows not only that methodological individualism is compatible with a systemic approach, but also that multiple-realization arguments against methodological individualism are invalid because of the ideal-typical feature of the individualist models that do not take the concrete historical features of the agents into account (see Lukes et al. 2023). Chen’s and my goal was much humbler than making manifest the absolute truth about the nature of methodological individualism: it was simply showing the empirical inconsistency of the definition of this explanatory model provided by Marchionni and Ylikoski and largely accepted in analytic philosophy. While the objective falsity of this definition is publicly testable, the problem is that contemporary philosophy of the social sciences is often excessively abstract because disconnected from the history of ideas and the practice of science, whose analysis is necessary to empirically check hermeneutical hypotheses. It must be noted that, while Chen and I criticize the short definition of methodological individualism provided by Marchionni and Ylikoski (2013, p. 10), we avoid offering an alternative short definition of it. This is because short definitions of longstanding, complex, and articulated intellectual traditions are normally poor and fail to foster useful scientific discussions about them (see Popper 2002). Conjectures about the features of intellectual traditions need to be couched in terms of multiple arguments that

Methodological Individualism and Agent-Based …

723

clarify why one thinks such and such about these features with precise references to the sources, i.e., to the protagonists of these traditions, so that these interpretive conjectures may be easily discussed and checked by the rest of the scientific community. Some sectors of contemporary philosophy of the social sciences are obsessed with short definitions because they follow the principle that any line of reasoning must be based on clear and precise assumptions. However, this principle cannot be implemented in all research fields and the validity of the assumptions of an argumentative structure is obviously more important than their simplicity. In many articles methodological individualism is criticized through quite technical arguments (such as, for example, the multiple realization problem) on the basis of a short simplistic and caricatural definition of it provided either by the same author of the article or by another critic of this approach cited by the article usually without any serious and detailed reference to the texts written by the individualist social scientists. Manzo (2020, p. 205) is aware of the problem created by this unfortunate tendency and these comments are obviously not a criticism of his position. They only aim at providing a better understanding of the approach implemented by Chen and myself. There is another important point that needs to be addressed. In his stimulating article, Manzo (ibid., p. 197) maintains that agent-based models and methodological individualism must be regarded as linked because they “share some basic principles irrespective of the entities and levels of analysis involved by the explanatory problem under examination.” In other words, they should be considered connected independently of “what kind of entities can legitimately be inserted within a given theoretical model” (the ontological problem) and of what kind of level of analysis (micro, meso, macro) should be taken into account and regarded as causally relevant in relation to the other levels of analysis (id. 210–214). As to the level of analysis, in our paper, Chen and I explain that the non-reductionist variant of methodological individualism is compatible with any kind of causal link (for example, micro-to-macro bottom-up causality as well as macro-to-micro top-down causality). I am not going to discuss this point here. For more details, the reader is referred to our paper. Manzo (id. p. 210) does not comment on our discussion of this topic, but focuses on our view that methodological individualism is related to a specific ontological claim, i.e. nominalism. He argues that our understanding of methodological individualism as underpinning this particular ontological thesis does not allow us to make a strong case for the compatibility between this explanatory approach and the agent-based models. This is because, if methodological individualism is assumed to incorporate a particular ontological thesis, it is more difficult to find an agreement about its compatibility

724

F. Di Iorio

with agent-based simulation. Manzo stresses that the acceptability and legitimacy of nominalism is often challenged in the social sciences. According to him, to support the compatibility between methodological individualism and agent-based models the former should be disconnected from nominalism understood as a view allowing only “specific entities”—i.e., the individuals— instead of any kind of entities (e.g., collective entities) within a given model (id. p. 219). In his opinion, methodological individualism should be interpreted as not necessarily linked to any specific ontological claim about the nature of the entities allowed by the social explanations. While Manzo’s analysis is astute, it does not properly take into account the way Chen and I understand the relationship between methodological individualism and nominalism. Two remarks need to be made here. First, according to our article, this relationship, which we regard as in need of being analyzed from a historical standpoint, is not absolutely essential and inescapable. Chen and I (2019, p. 364) challenge the widespread view in analytic philosophy that no connection whatsoever exists between ontology and methodological individualism by stressing that historically “many nonreductionist individualists… argued that methodological individualism is based on ontological individualism” (italics added). We recognize that some individualist scholars—like Boudon—are not fond of nominalism. To make it short, our paper denies the total absence of connection between nominalism and methodological individualism, but it is not supportive of the view that only nominalist individualism is the true implementation of methodological individualism. Second, in our article, Chen and I (ibid. p. 367) clarify that ontological individualism as historically understood by authors like Mises, Hayek, and Popper has nothing to do with the necessity to rule out the reference to entities other than individuals. We do not center the discussion of the relationship between methodological individualism and ontological individualism on the admissibility or non-admissibility of some specific entities. This is because we explain that nominalist individualists never conceived this relationship in these terms (ibid.). It seems to me that the distance between Manzo’s understanding of methodological individualism and the position of the nominalist individualists is more nuanced than Manzo allows. His argument in favor of disconnecting nominalism and methodological individualism to foster an agreement on the connection between the latter approach and agent-based simulation is based on a debatable understanding of nominalism. According to nominalist individualists, explanations that refer to non-individual agents are compatible with their approach. “This is so insofar as they do not conceive non-individual agents in holistic terms, i.e.,

Methodological Individualism and Agent-Based …

725

as supra-individual substances (concrete entities) that ontologically exist independently of the individuals and control their thoughts and actions” (ibid.). Think, for example, of Spencer’s and Hayek’s analyzes of cultural evolution in terms of group selection (see Di Iorio 2015; Di Nuoscio 2018; Whitman 1998, 2005).

References Antiseri, D. (2006). Epistemology and hermeneutics. In M. Alai, and G. Tarozzi (Eds.), Karl Popper philosopher of science. Soveria Mannelli: Rubbettino. Boudon, R. (1971). Uses of structuralism. London: Heinemann. Coleman, J. S. (1990). Foundations of Social Theory. Cambridge, MA/London: The Belknap Press of Harvard University Press. Di Iorio, F. (2015). Cognitive autonomy and methodological individualism: The interpretative foundations of social life. Cham: Springer. Di Iorio, F. and Chen, S-H. (2019). On the Connection Between Agent-Based Simulation and Methodological Individualism. Social Science Information, May 2019–OnlineFirst, pp. 1–23. Di Nuoscio, E. (2018). The Logic of Explanation in the Social Sciences. Oxford: The Bardwell Press. Grondin, J. (2002). Gadamer’s basic understanding of understanding. In R.J. Dostal (Ed.), The Cambridge companion to Gadamer. New York: Cambridge University Press. Hayek, F.A. (1967). Studies in philosophy, politics and economics. Chicago, IL: University of Chicago Press. Hayek, F.A. (1973). Law, legislation and liberty, Vol. 1: Rules and order. Chicago, IL: University of Chicago Press. Hayek, F.A. (1978). New studies in philosophy, politics, economics and the history of ideas. London: Routledge & Kegan Paul. Lukes S., Bulle N., & Di Iorio F. (2023). Methodological individualism and social facts: A discussion with Steven Lukes. In N. Bulle & F. Di Iorio (Eds.), Palgrave handbook of methodological individualism, Vol. II, Palgrave MacMillan, London. Mantzavinos, C. (2006). Naturalistic hermeneutics. Cambridge, Uk: Cambridge University Press. Manzo, G. (2020). Agent-based models and methodological individualism: Are they fundamentally linked? L’année Sociologique, 70, 197–229. Marchionni, C., & Ylikoski, P. (2013). Generative explanation and individualism in agent-based simulation. Philosophy of the Social Sciences, 43(3), 323–340. Menger, C. (1985 [1883]). Investigations into the method of the social sciences with special reference to economics. New York/London: New York University Press. Popper, K.R. (2002 [1952]). The Logic of Scientific Discovery. London & New York: Routledge.

726

F. Di Iorio

Popper, K.R. (1957). The Poverty of Historicism. Boston: Beacon Press. Popper, K.R. (1980). Autointerpretazione filosofica e polemica contro i dialettici. In C. Grossner (Ed.), I filosofitedeschi contemporanei (Verfall der philosophie. Politik deutscher Philosophen). Rome: Città nuova. Ricoeur, P. (2007). From text to action: Essays in hermeneutics, II . Evanston, IL: Northwestern University Press. Udehn L. (2001). Methodological Individualism: Background, History and Meaning. London: Routledge. Udehn L. (2002). The changing face of methodological individualism. Annual Review of Sociology, vol. 28, p. 479–507. https://doi.org/https://doi.org/10.1146/ annurev.soc.28.110601.140938. Whitman, D. G. (1998). Hayek contra pangloss on evolutionary systems. Constitutional Political Economy 9: 45–66. Whitman D. G. (2005). Group Selection and Methodological Individualism: Compatible and Complementary. Evolutionary Psychology and Economic Theory ISBN: 978-0-76231-138-5, eISBN: 978-1-84950-294-8. Publication date: 5 January 2005

Author Index

A

Abel, T. 309 Abrams, B.A. 443 Acemoglu, D. 390, 393 Agassi, J. 78, 85, 204, 305, 460–463, 465–467, 472–475, 480, 691 Ajzen, I. 381, 384 Albert, G. 322 Albert, H. 195, 260, 324, 399 Alchian, A.A. 450 Alesheikh, A.A. 682 Alexander, J.C. 26, 27, 322, 328, 341, 378, 655 Allais, M. 279 Althusser, L. 627 Andersen, S. 662 Andler, D. 420 Angner, E. 656, 660, 661 Anscombe, G.E.M. 616, 619, 621, 622 Antiseri, D. 195, 197, 721 Apel, K.-O. 32 Aral, S. 285 Archer, M.S. 321, 324, 485

Argyris, C. 592 Ariely, D. 666 Aristotle 145, 272, 279, 296, 414, 617, 620–624, 626 Aron, R. 32, 50 Arrow, K. 25, 117, 124, 126–129, 132, 135, 655, 657, 670, 672 Arthur, B. 9 Ashraf, N. 672 Axelrod, R. 422 Axtell, R. 394 Aydinonat, E.E. 475 Aydinonat, N.E. 475 Ayer, A.J. 308

B

Bachelard, G. 153 Bader, V.-M. 337–340 Baez, J. 700 Baier, H. 33 Baker, L.R. 497 Bak, P. 409 Balcan, D. 682 Bandura, A. 384

© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 N. Bulle and F. Di Iorio (eds.), The Palgrave Handbook of Methodological Individualism, https://doi.org/10.1007/978-3-031-41512-8

727

728

Author Index

Barberis, N.C. 665 Barnes, B. 631, 644, 645 Barr, A. 667 Barrett, C. 667 Barry, N.P. 60 Barthélemy, J. 682 Basu, K. 77 Baumol, W.J. 435 Bayuk, J.B. 669 Beattie, J. 648 Bechtel, W. 485, 486 Becker, G.S. 13, 241, 382, 435, 537, 575, 656, 666 Becker, S.O. 393 Bedau, M.A. 396 Bee, M. 25 Benjamin, D.J. 669 Bennett, D.A. 682 Ben-Ze’ev, A. 141 Bergeron, H. 593 Berger, P. 254, 255 Berger, P.L. 578, 647 Bernstein R.J. 199 Bertillon, A. 347 Bertrand, E. 128 Besnard, P. 368 Bessi, A. 277 Bhaskar, R. 313, 382, 397, 399 Bianchi, F. 682, 697 Billari, F. 682 Birbaumer, N. 495 Bird, A. 302 Birks, D. 697 Birnbaum, P. 103 Birner, J. 425 Bishop, R. 418 Blaug, M. 435 Blau, P. 351, 354, 537, 545 Bless, H. 283 Bloor, D. 644–647, 649 Boettke, P.J. 4, 8, 13, 173, 433, 435, 436, 440, 442–444 Böhm-Bawerk, E. von 649 Boltzmann, L. 408

Bolyanatz, A. 667 Bonabeau, E. 415 Bora, S¸ . 682 Bostaph, S. 302 Boudon, R. 103, 105, 110, 112, 120, 154, 181, 182, 184, 185, 187, 190, 191, 227–230, 232–238, 240–247, 251–265, 272, 277, 280, 286, 287, 303, 321, 350, 354, 356, 361, 369, 596, 597, 631, 634, 637–643, 645, 649–651, 685, 686, 689, 712, 719, 720, 722, 724 Bourdieu, P. 110, 120, 129–131, 234, 240, 243, 273, 356, 537, 596 Bouvier, A. 245, 311, 486, 649 Bowles, S. 655, 662, 667, 669, 671, 672 Boyd, R. 667 Bratman, M.E. 133 Brennan, G. 657 Brodbeck, M. 309, 378, 394 Bronner, G. 245, 246, 280, 283, 285, 287, 649 Brown, A. 263 Brown, R. 361 Bruner, J. 279 Brunet, E. 273 Brunschvicg, L. 32 Brus, W. 393 Buchanan, J.M. 13, 17–19, 22, 438, 440 Bulle, N. 240, 246, 253, 300, 303–305, 312, 350, 394, 682, 688, 689, 696, 712, 717, 718, 722 Bulte, E.H. 668 Burger, T. 324 Burow, J. 682 Buskens, V. 700

Author Index C

Cachanosky, N. 441 Cairns, D.L. 141 Çakırlar, I. 682 Caldwell, B. 92–97, 412, 413 Camerer, C. 667 Camerer C.F. 661, 662, 665 Campagnolo, G. 302 Campbell, C. 541, 552, 553 Candela, R.A. 14, 435, 436, 442 Canguilhem, G. 32 Cardarelli, G. 277 Carletti, T. 682 Carnap, R. 309 Carrasco, M.A. 19, 20 Cartwright, N. 301, 302, 304 Castells, M. 251 Centola, D. 687, 691 Cerdá, M. 697 Chattoe-Brown, E. 682 Chazel, F. 39 Chen S.-H. 394, 682, 688, 690–695, 717–724 Cherkaoui, M. 348, 350, 355, 357, 359, 362, 364 Chibundu, M.O. 262 Choi, J.J. 669 Churchland, P.M. 485 Citton, Y. 283 Clifford, J. 204 Cohen, G.A. 123 Cohen, M.D. 590, 591 Cohn, A. 669 Coleman, J.S. 237, 243, 245, 321, 326, 351, 362, 378, 507, 537, 544, 548–550, 685, 686, 700–702, 704, 719, 720, 722 Cole, S. 560 Colizza, V. 682 Collin, F. 492 Colliot-Thélène, C. 31–34, 39, 42, 46 Colombetti, G. 141

729

Comte, A. 32, 57, 240, 243, 347, 419, 574, 614 Constantinides, E. 152 Cook, K.S. 545 Copi, I.M. 194 Cormack, P. 555 Cowan, R. 302 Cowen, T. 435, 445–447 Coyne, C.J. 4 Crafford, F.S. 293 Craver, C.F. 302, 307, 485, 486 Creighton, L.A. 384 Croce, B. 188 Croson, R. 672 Crothers, C. 544, 558 Crozier, M. 243, 589, 602, 722 Cubeddu, R. 302

D

Dahrendorf, R. 326 Damasio, A. 141, 146, 148 Darden, L. 302 Davidson, D. 497, 622–626 Davis, J.B. 657, 661 Davis, K. 539, 540, 551, 552 Dawes, R.M. 662 Dawkins, R. 420 Decety, J. 273 Defoe, D. 7 DeLancey, C. 141 DellaVigna, S. 665 Del Vicario, M. 277 De Marchi, S. 697 Demeulenaere, P. 162, 246, 354, 649 De Sousa, R. 141, 142, 146, 148 Destutt de Tracy, A. 640 Diaz, B.A. 682 Diekmann, A. 378, 381 Di Iorio, F. 78, 172, 186–188, 193, 245, 305, 311, 313, 378, 394, 398, 412, 433, 436, 460, 476,

730

Author Index

674, 681, 682, 688, 690–695, 717, 719, 722, 725 Dilthey, W. 40, 41, 50, 187, 569 DiMaggio, P. 670 Di Nuoscio, E. 172, 184–189, 191–193, 195, 199, 246, 303, 717, 721, 725 Dixon, Th. 141, 145, 147 Döbert, R. 339 Dold, M.F. 667, 670, 673 Donnelly, J. 254 Douglass, E. 127 Dreher, J. 578 Dufour, É. 42 Dumont, L. 108 Dumouchel, P. 141, 150, 152, 155, 411, 426 Dunbar, R. 420 Dupuy, J.P. 277, 411, 412, 427 Durkheim, É. 32, 103, 118, 123, 126, 129, 181, 203, 207, 213, 222, 243, 253–255, 329, 349, 356, 367–369, 459, 495, 513–532, 538, 544, 555, 613–615, 618, 631–633, 645 Dworkin, R. 261–263, 265

E

Eberlen, J. 697 Eccles, J.C. 79, 80, 82–84 Edwards, W. 279 Eger, M. 199 Eggertsson, T. 393 Ekman, P. 141 Elder-Vass, D. 313, 397, 497, 537 Ellis, B. 302 Elridge, J.E.T. 103 Elsenbroich, C. 682 Elster, J. 118, 120, 123–127, 129, 131, 133–135, 141–146, 150, 153, 154, 186, 277, 392, 537, 658 Emek, S. 682

Engels, F. 175, 176, 392, 632, 633, 641 Epstein, B. 398, 460, 481, 492 Epstein, J.M. 394, 687, 699 Epstein, T.S. 108 Ermakoff, I. 119 Ertaç, S. 662 Esser, H. 321, 326 Evans, A.J. 435 Evans-Pritchard, E.E. 205, 206, 208, 472

F

Faccarello, G. 417, 421 Falk, A. 668 Fararo, T.J. 351, 699 Farmer, J.D. 687 Featherman, D. 354 Feather, N.T. 381 Fehr, E. 662, 663, 667, 669, 686 Feldman Barrett, L. 142 Ferber, J. 700 Ferguson, A. 10, 57, 421, 426, 460 Ferrari, M. 35, 42 Fessler, D.M. 668 Festinger, L. 590 Feuerhahn, W. 31, 304 Fillieule, R. 166, 236, 246 Fischbacher, U. 662 Fischer, K. 35 Fishbein, M. 381, 384 Flache, A. 393, 682, 684, 687 Flammini, A. 285 Flanquart, H. 281 Fligstein, N. 597 Fodor, J.A. 410, 485, 486, 497 Foley, D. 687 Follett, M.P. 597 Forsé, M. 599 Foucault, M. 240, 243, 643 Francisco, R.A. 392 Frank, R.H. 141, 145 Frazer J.G. 203, 278

Author Index

Frerichs, S. 661, 670–672 Freud, S. 243 Friedberg, E. 587, 589, 596, 605 Friedman, M. 18, 443, 572, 659 Frijda, N. 156 Frith, U. 273

G

Gadamer, H.-G. 70, 191, 192, 194–199, 246 Gaissmaier, W. 689 Galea, S. 682 Gallie, W.B. 262 Garfinkel, H. 574, 577 Garrison, R.W. 437, 447–449 Gawronski, B. 384 Geertz, C. 204 Géhin, E. 280, 283 Gellner, E. 205, 223 Gens, J.-Cl. 41, 49, 50 Gewecke, A. 673 Gibbs, J.W. 408 Giddens, A. 321, 324, 328, 513, 514, 519, 525, 537 Giesen, B. 322, 328, 341 Gigerenzer, G. 148, 149, 280, 689 Gilbert, M. 133 Gintis, H. 382, 420, 656, 686 Gneezy, U. 668, 672 Goethe, J. 358 Goldman, A.I. 625 Goldstone, J. 392 Goldthorpe, J.H. 241, 681, 695 Gonzales-Bailon, S. 682 Goode, W.J. 552 Goodman, N. 198 Gordon, R.J. 439 Gorton, W.A. 99 Gouldner, A.W. 589 Grassé, P.P. 415 Grassl, W. 302 Green, A.W. 59 Greenwood, J. 536

731

Greif, A. 461, 475–480 Greve, J. 322, 324, 504 Groff, R. 313 Grondin, J. 721 Grossner, C. 197 Grote, R. 252 Grow, A. 697 Guala, F. 461, 469–476, 480 Güth, W. 662

H

Habermas, J. 324 Hacohen, M.H. 86 Hahn, H. 309 Haklay, M. 687 Halévy, E. 436 Halvorson, H.G. 281 Hamilton, W. 416 Hands, D.W. 92–96 Hanna, J.F. 394 Hansson, B. 129 Hardin, G. 360 Hardy-Bayle, M.C. 273 Hargreaves Heap, S.P. 667, 671, 673 Harmon-Jones, E. 384 Harré, R. 301, 302, 313 Harris, W.V. 141 Harsany, J. 350 Haskel, J. 128 Haslanger, S. 128, 129 Hauser, R. 354 Hausman, D.M. 663 Hayek, F. 4–6, 8, 10, 12, 13, 19, 57, 60, 62–65, 70, 71, 85, 89, 194, 204, 208, 246, 302, 308, 364, 411–413, 416, 418, 419, 421, 424–427, 436, 437, 440, 451, 460, 537, 645, 655, 694, 719, 722, 724, 725 Hayward, S. 699 Hazlitt, H. 435, 445, 635 Heath, J. 19, 610, 658 Heckbert, S. 682

732

Author Index

Hedström, P. 186, 321, 382–384, 695, 700 Hegel, G.W.F. 39, 41, 57, 106, 495, 610, 614 Heider, F. 279 Hempel, C. 95, 185–189, 309, 356, 396 Hennis, W. 35, 38, 39, 170, 322 Henrich, D. 322 Henrich, J. 667, 672 Herfeld, C. 398 Hernes, G. 378 Herrando, C. 152 Hess, U. 151 Heukelom, F. 660, 666 Heyes, C. 667 Hine, J. 682 Hirschfeld, M.L. 246 Hirschman, A.O. 147, 154 Hitzler, R. 579, 580 Hobbes, T. 147, 362, 418, 419, 458, 459, 463, 523 Hodgson, G.M. 378, 436, 657 Hoff, K. 667, 669, 670, 673 Hogan, W.P. 86 Hokamp, S. 682 Holbraad, M. 205 Hölldobler, B. 420 Homans, G.C. 383, 397, 537, 545 Honer, A. 579, 580 Honneth, A. 252 Hopf, W. 326, 331 Hoppe, H.-H. 163 Horton, R. 638, 648, 650 Hoselitz, B.F. 107 Hosseinali, F. 682 Hume, D. 4, 13, 57, 65, 147, 162, 262, 263, 421, 426, 620, 624 Hummell, H.J. 378, 383, 386, 388, 395–397 Hummon, N.P. 699 Humphreys, P. 396, 486 Huntington, S.P. 258 Hurtado, J. 17, 18, 20, 25

Husserl, E. 566, 576, 581 Hutt, W.H. 435, 447 I

Infantino, L. 25 J

Jackson-Jacobs, C. 155 Jackson, P.L. 273 James, W. 566 Jankovic, M. 133 Jarvie, I.C. 204, 207, 209, 461–463, 472–476, 480, 691 Jasso, G. 386 Jepperson, R. 496, 700, 702 Joas, H. 265 Joerges, B. 683, 697 Johnson, D.P. 560 Johnston, W.M. 58 K

Kahneman, D. 149, 234, 245, 279, 280, 282, 286, 639, 659, 660, 664–666 Kahn-Lang Spitzer, A. 673 Kaidesoja, T. 497 Kant, E. 35, 48, 198, 229, 240, 275, 301, 338, 418, 419, 427, 514 Kateb, G. 252 Kates, S. 435, 444, 445 Kebble, S. 273 Kelsen, H. 359 Kemeny, J. 308, 485 Kerridge, J. 682 Kerschagl, R. 58 Keuth, H. 99 Keynes, J.M. 13, 135, 435, 438, 439, 441, 445, 447–449 Khurshid, S. 261 Kim, J. 84, 307, 485, 488, 499 Kincaid, H. 378, 396, 397, 460, 486, 492, 493, 496, 498, 499,

Author Index

502, 505, 682, 688, 690–694, 696 Kinzler, K.D. 673 Kirchgässner, G. 381 Kirman, A.P. 685, 686 Kissine, M. 285 Kistler, M. 308 Klaes, M. 661 Klein, O. 285 Klumpp, G. 283 Knauff, M. 381 Knies, K. 32–35, 37–40, 42, 44, 50, 187, 610 Knight, F.H. 13, 22, 440, 673 Knoblauch, H. 578, 579 Kocher, M.G. 662 Koertge, N. 92, 95, 97 Koffka, K. 298 Kogawara, M. 97 Köhnke, K.-C. 35 Kries, J.V. 569, 570 Kroneberg, C. 690, 713 Kuhn, T. 240, 559, 644

L

Lachmann, L. 191, 655 Lacombe, R. 513 Laibson, D. 660, 664, 665 Lange, O. 435, 447, 448 Lang, K. 673 Lask, E. 33, 39, 322 Laski, K. 393 Latour, B. 240, 646–650 Laudan, L. 646, 648–650 Lawrence, P. 210, 213 Lazarsfeld, P. 386, 535, 546–549, 556 Leach, C.W. 151 LeBoeuf, R.A. 669 Lebois, A.M. 141 Le Bon, G. 105, 361 Lecouteux, G. 661, 671, 672 Ledoux, J.E. 141

733

Leibbrandt, A. 663 Leijonhufvud, A. 435, 446, 449 Lerner, D. 108 Leslie, A.M. 273 Levine, D. 558, 559 Lévi-Strauss, C. 240, 243 Lévy-Bruhl, L. 177, 178, 278, 356, 638, 639 Levy, D.M. 25 Levy, M.J. 540 Lewin, K. 91 Lewis, M. 151 Lewis, P. 670, 673 Lichtblau, K. 322 Lieberson, S. 705 Liefmann, R. 60 Linares, F. 394 Lindenberg, S. 326, 685 Lipset, S.M. 383 List, C. 129, 133, 134, 486, 492, 493, 496–499, 502–504 List, J.A. 660 Little, D. 480, 481, 496, 698 Lizardo, O. 634 Loewenstein, G. 660 Long, N. 20 Luckmann, T. 566, 567, 577, 578, 647 Ludwig, K. 133 Luhmann, N. 321, 332 Lukes, S. 311, 458, 514, 516, 722 Lutz, C. 141 Lux, T. 682 Lynn, F.B. 705

M

Machamer, P. 302 Machiavelli, N. 419, 460 Machlup, F. 572, 573 Mackie, G. 127 Macklin, P. 697 Macklin, R. 262 Maclouf, P. 243

734

Author Index

Macpherson, C.B. 19 Macy, M.W. 393, 682, 684, 687, 691 Malebranche, N. 142, 279 Malewski, A. 383, 397 Malik, L. 261 Malleson, N. 682 Mallory, P. 555 Malmendier, U. 665 Mandelbaum, M. 309, 614, 615, 626 Mandeville, B. 25, 57, 359, 416–419, 421, 426 Mankiw, N. 438, 439 Mannheim, K. 632, 633, 650, 651 Mantzavinos, C. 721 Manza, J. 554, 555, 558 Manzo, G. 245, 253, 350, 393, 394, 682, 687, 696, 698, 701, 711–713, 717, 718, 720–724 Marchesi, M. 682 Marchionni, C. 394, 682, 688–690, 692, 694, 696, 718, 722 March, J.G. 590, 591 Marcus, G. 204 Maréchal, M.A. 669 Margalit, A. 252 Marmodoro, A. 302 Marsden, P.V. 550 Marshall, B.D. 682 Martin, C. 128 Martin, J. 129 Marwell, G. 392 Marx, K. 32, 57, 66, 69, 105, 117, 120–123, 129, 175, 176, 240, 243, 392, 544, 614, 631–633, 640–642, 649, 650 Maryanski, A. 544 Maslow, A.H. 593 Mäs, M. 394 Mauss, M. 632, 633 Mbonda, E.-M. 259 McAdam, D. 597 McClelland, D.C. 378

McCloskey, D.N. 673 McCrudden, C. 252 McElreath. R. 667 McGill, B. 307 Mclaughlin, B.P. 312 McPherson, M.S. 663 Mead, M. 107, 108 Medawar, J.S. 83, 84 Medawar, P.B. 83, 84 Meehl, P. 279 Megiddo, T. 125 Melden, A.I. 619–621, 624, 625 Menczer, F. 285 Mendras, H. 108 Menger, C. 4–6, 8, 9, 12, 13, 18, 37, 42, 57, 61, 62, 66, 71–73, 162, 164, 170, 178, 272, 298, 299, 302, 308, 321, 350, 363, 437, 460, 610, 655, 694, 720, 722 Menzel, H. 386, 547, 548 Menzies, P. 486, 497, 498 Merler, S. 682 Merton, R.K. 74, 236, 243, 357, 460, 529, 535–537, 539–541, 544, 546–548, 552, 553, 556, 589, 596, 602, 603, 632–634, 636, 637, 648, 649 Mesure, S. 41, 246, 254, 258, 265, 303 Metzcar, J. 697 Meyer, J.W. 496 Meyerson, É. 301 Michalski, J.H. 558 Michel, F. 700 Michelfelden, D.P. 195, 199 Michels, R. 349 Miller, D. 98 Miller, J. 698 Miller, R. 157 Mill, J.S. 90, 187, 204, 279, 356, 427, 444, 458, 462, 463 M’Mbogori, F.N. 682 Mommsen, W. 103, 187

Author Index

Montes, L. 24 Morgan, L.H. 296 Morin, J.-M. 228, 246, 303 Morrow, G.R. 24, 25 Morsink, J. 258 Moscovici, S. 105 Mullainathan, S. 659, 663 Müller, B. 682 Müller, H.-P. 324 Mulligan, C.B. 450 Mullins, N. 542, 555 Münch, R. 342 Murphy, T.E. 682

735

O’Donoghue, T. 664 Oliveira, D. 285 Oliver de Sardan, J.-P. 603 Oliver, P. 392 Olsen, J.P. 590, 591 Olson, M. 362, 385, 392 Oppenheim, P. 95, 185, 306, 308–310, 396, 485, 491 Opp, K.-D. 304, 378, 381–383, 386–388, 390, 392–397, 634 Opp, K.-R. 188 Orton J.D. 603 Osborne, D. 378, 384 O’Sullivan, D. 687, 697 Ostrom, E. 393, 663

N

Nadeau, R. 92, 95, 96, 426 Nagel, E. 186, 187, 307–310, 394, 396, 397, 485 Nagel, T. 618 Nagengast, B. 381 Nemo, P. 246, 419, 421 Neugart, M. 697 Neurath, O. 309 Nguyen, Q. 672 Nichols, L. 539, 540 Nicole, P. 417, 418, 421 Niemann, H.-J. 77 Nillesen, E.E. 668 Nisbet, R.A. 515, 641 Norenzayan, A. 672 North, D. 368 North, D.C. 393 Norton, K.A. 682 Nourian, F. 682 Nowak, M. 416, 420, 424, 426 Nowak, S. 397 Nozick, R. 437

O

Oatley, K. 141 Oberschall, A. 106 Odean, T. 665

P

Padgett, J.F. 695 Paganelli M.P. 18, 23, 25, 26 Page, S.E. 697, 698 Panksepp, J. 141 Pantazi, M. 285 Pareto, V. 32, 164, 169, 246, 279, 568 Parodi, M. 599 Parri, L. 378 Parsons, T. 124, 321, 323, 329, 341, 367, 521, 529, 537, 539, 541, 543–545, 554, 555 Passeron, J.-C. 110, 273 Patino, B. 285 Peart, S. 25 Peirce, C.S. 193–195 Peters, R.S. 617–619 Petitot, J. 410, 413, 424 Petroni, A.M. 60 Pettit, P. 133 Pfaff, S. 393 Phan, D. 312, 394, 682, 688, 689, 696, 711, 712, 714 Philippot, P. 151 Phillips, D.C. 298 Piano, E.E. 173

736

Author Index

Pilmis, O. 593 Pinker, S. 262, 421 Platt, J. 555, 556 Polanía-Reyes, S. 662 Popel, A.S. 682 Popper, K. 58–60, 62, 65–71, 77–99, 186–190, 192, 197, 198, 203, 204, 206, 241, 274, 296, 297, 300, 304, 305, 308, 323, 399, 426, 457–467, 472–474, 476, 477, 479–481, 610, 614, 640, 647, 691, 693, 694, 719–722, 724 Poteete, A.R. 663 Potochnik, A. 307 Powell, W.W. 695 Prelec, D. 286 Premack, A.J. 273 Premack, D. 273 Prendergast, C. 572 Pribram, K. 58 Psathas, G. 577 Putnam, H. 308–311, 410, 491, 497 Putnam, R. 129, 131, 132

Q

Qiu, X. 285 Quattrociocchi, W. 285 Quattrone, G. 133

R

Rabin, M. 664 Radcliffe-Brown, A.R. 203, 205, 206, 539 Railsback, S.F. 682 Ramasco, J.J. 682 Ramsey, W. 494 Ramström, G. 298, 485, 487, 488, 492, 495, 507 Rand, W. 682 Raub, W. 378, 381, 685, 700 Rawls, J. 144

Read, D. 663 Reichertz, J. 582 Renault, E. 252 Reschreiter, H. 697 Richiardi, M. 697 Rickert, H. 31, 33, 39–42, 45, 47–51, 322, 569 Ricoeur, P. 192, 195, 197, 721 Rittenauer-Schatka, H. 283 Ritzer, G. 485 Rizzo, M. 302 Robinson, D.N. 82 Robinson, J.A. 390, 393 Robinson, W.S. 550, 551 Robitaille, C. 162, 167, 170 Rock, D. 281 Röder, T.J. 252 Rodriguez-Blanco, V. 261 Rodriguez, Ph.-A. 262 Roemer, J. 392 Roethlisberger, F.J. 592 Rosen, M. 252 Ross, D. 148, 152, 661 Rossello, D. 262 Rothbard, M.N. 59, 167 Rousseau, J.-J. 418, 523 Roux, V. 682 Royster, C. 135 Rubin, J. 393 Ruiz, S. 495 Russell, J.A. 141, 142, 152, 153 Russell, T. 659 Rustichini, A. 668 Rutherford, M. 473

S

Salerno, J.T. 174 Salert, B. 392 Salter, A.W. 443 Samuelson, P. 13, 277, 439 Sandel, M.J. 673 Sanderson, S.K. 558 Sandler, T. 392

Author Index

Sankey, H. 302 Sarfati, Y. 273 Sartre, J.-P. 130 Sauder, M. 554, 555 Sawyer, R.K. 394, 396, 397, 460, 481, 485, 486, 493, 495–498, 502, 687 Say, J.-B. 444–448 Sayre-McCord, G. 20 Scalas, L.F. 381 Scheler, M. 631–633 Schelling, T.C. 41, 135, 361, 394, 470, 598–600, 602 Schelting, A. von 328 Schluchter, W. 33, 322, 324, 330, 338, 339 Schnädelbach, H. 49 Schnettler, B. 578, 579 Scholz, G. 697 Schumpeter, J.A. 3, 203, 350, 436, 537, 610, 657 Schurz, G. 399 Schutz, A. 310, 311, 566, 567, 571, 576, 577, 580, 581, 647 Schwarz, N. 283 Schwinn, T. 321, 323, 328 Scott, J. 553 Scott, K.J. 60 Scriven, M. 187 Searle, J.R. 133, 461, 467–475, 480, 621, 624, 625 Sellars, W. 313 Selten, R. 148, 149 Selznick, P. 589, 590, 596 Sen, A.K. 656, 673 Sepulvado, B. 634 Shackle, G.L.S. 277 Shafir, E. 669 Shaftesbury, A. 147 Shils, E. 537 Shinn, T. 683, 697, 703 Shirazi, A.S. 285 Shoul, B. 445 Shweder, R.A. 639

737

Sibley, C.G. 378, 384 Sidis, B. 298 Simkin, C.G.F. 86–89, 98, 462, 476 Simmel, G. 34, 45, 61, 105, 111–113, 246, 300, 351, 694, 719 Simmons, W.K. 141 Simon, F. 23 Simon, H. 241, 279, 350, 354, 381, 658, 659, 672 Simons, A. 283 Sitaram, R. 495 Skinner, B. 120 Skvoretz, J. 351 Slovic, P. 280, 281, 283, 284 Smelser, N.J. 342, 485 Smith, A. 3, 5, 10–13, 18–20, 23–25, 38, 58, 135, 147, 162, 229, 302, 359, 360, 393, 417, 418, 421, 436, 444, 460, 537, 719 Smith, B. 302, 412 Smith, C. 17, 26 Smith, J.A. 682 Smith, V. 19, 22, 659 Smuts, J.C. 293–297, 307, 314 Sobel, J. 132 Sober, E. 149, 426 Soeffner, H.-G. 578, 580 Solomon, R.C. 141 Sombart, W. 52, 105, 349 Sowell, T. 435, 444 Spelke, E.S. 673 Spencer, H. 348, 526, 694, 725 Sperber, D. 695 Spiekermann, K. 129, 134, 486, 492, 493, 496, 498, 499, 502–504 Spinoza, B. 41, 142, 356 Spohn, W. 381 Squazzoni, F. 682, 697 Stegmaier, P. 578 Steinmetz, G. 545, 556, 557 Stigler, G.J. 13, 150, 656, 666

738

Author Index

Stiglitz, J.E. 667, 670, 673 Stoljar, D. 480, 497 Stoutland, F. 621 Strack, F. 283 Streissler, E. 6 Strickland, A.J. 669 Subramanian, S.V. 551 Sugden, R. 133, 398, 661, 666, 667 Sunstein, C.R. 669 Swedberg, R. 383, 700

T

Tanaka, T. 672 Tang, C. 409 Tang, W. 682 Tarnita, C. 416, 420 Taylor, G. 135 Taylor, M. 392 Tenbruck, F. 33, 322 Thaler, R. 381, 594, 659, 662, 663, 665 Thatcher, M. 458 Theraulaz, G. 415 Thiele, J.C. 682 Thucydides 187, 189, 190 Tiedens, L.Z. 151 Tiruneh, G. 392 Toboso, F. 473 Tocqueville, A. de 32, 104, 105, 123, 124, 127, 132, 134, 243, 255, 286, 349, 364, 631 Tonks, S. 381 Touraine, A. 106 Tracy, M. 697 Trautwein, U. 381 Trevarthen, C. 157 Tullock, G. 657 Turner, J. 544 Tversky, A. 133, 234, 245, 279, 280, 282, 286, 664–666 Tversky, K. 280, 639 Tyler, T.R. 381 Tylor, E.B. 203

U

Udehn, L. 323, 324, 457, 481, 492, 535, 544, 545, 683, 685, 690, 693, 702, 720 Uebel, T. 309 Ullmann-Margalit, E. 393

V

Valade, B. 246 Van Assen, M.A.L.M. 700 Vanberg, V. 411, 422 van de Rijt, A. 384 van Leeuwen, B. 663 Van Soest, D.P. 668 Verwimp, P. 668 Vespignani, A. 682 Vidyasagar, A. 218 von Mises, L. 4, 6–8, 10, 12, 13, 57, 62, 161–178, 181–185, 191, 299, 435, 442, 572, 573, 575, 637, 638, 640, 649, 694, 719 von Neumann, J. 412, 413 von Wright, G.H. 378, 621 Voors, M.J. 668 Vosoughi, S. 285 Voss, T. 685 Vromen, J. 486 Vygotsky, L.S. 298, 300, 305

W

Wagner, G. 33, 378, 397 Wagner, R.E. 438, 443 Wahlberg, T.H. 497 Waldron, J. 252, 254 Waller, W. 540 Wang, Y. 697 Wan, P.Y. 687 Watkins, J.W.N. 19, 459, 463, 465, 492, 537, 691 Weber, M. 31–37, 39, 40, 42–48, 50–54, 60, 61, 103–105, 114, 162, 168–171, 179, 183, 185,

Author Index

187, 190, 203, 227, 229, 230, 232–234, 243, 246, 256, 257, 259, 271, 272, 278, 299, 300, 302, 304, 305, 321–326, 328–332, 335–342, 349, 350, 355, 358, 364, 366, 393, 459, 537, 544, 566, 568–573, 575–577, 587, 604, 610–613, 617–620, 626, 645, 719 Weiß, J. 323 Weick, K.E. 603, 604 Wettersten, J. 466 White, L.H. 435, 439, 441, 443, 445, 449, 452 Whitman, D.G. 725 Whitman, G. 661, 674 Wiesenfeld, K. 409 Wigan, M. 682 Wigfield, A. 381 Wijesinghe, S. 682 Wilensky, U. 682 Wilkinson, N. 661 Willer, R. 682 Williams, B. 625 Williams, F.E. 209–211, 626, 627 Wilson, B.J. 18, 19, 22, 23 Wilson, E. 416, 420 Wilson, K. 410 Wilson-Mendenhal, C.D. 141 Wimsatt, W.C. 485, 486 Winch, P. 472, 616, 617 Winckelmann, J. 326, 339 Windelband, W. 40, 41, 45, 48, 49 Wippler, R. 326, 685

739

Wisdom, J.O. 204, 457 Wittgenstein, L. 615–619, 621–623, 625 Wolff, K. 525 Wolpert, L. 153 Wood, A.W. 263 Woolcock, M. 132 Woolgar, S. 646–649 Worsley, P. 215 Wrong, D.H. 113 Wurzer, G. 697

X

Xu, M.K 381

Y

Yagi, K. 170 Yeager, L.B. 447, 449 Yin, W. 672 Ylikoski, P. 186, 382–384, 394, 397, 475, 681, 682, 688–690, 692, 694, 696, 702, 718, 722 Yoshida, K. 458, 460, 461, 472, 473

Z

Zahle, J. 311, 378, 490–492, 497, 682, 688, 690–694, 696 Zehnder, C. 668 Zheng, X. 698 Zhong, T. 698 Ziker, J. 667

Index

A

Abstract psychology 112 Adaptive 149 Adequacy 165, 334, 411, 566, 568–578, 581, 582, 621 Affect 22, 80, 143, 147, 149, 151–153, 155–157, 163, 336, 358, 452, 666–670, 689 Affective 147, 149–151, 153, 155–157, 171, 228, 230, 232, 233, 242, 325, 330, 337, 339, 340, 590, 600, 611, 650 Agency 9, 121, 123, 133, 203, 378, 537, 587, 695 Agent-based 663, 684, 687, 691, 699, 713, 717–720, 722–724 Aggregation 104, 105, 111, 112, 125, 154, 328, 361, 362, 380, 386, 395, 422, 437, 438, 558, 596, 599–601, 611, 685, 687, 698, 702 Aggregative 657, 691, 701 Altruism 58, 420, 662 Analytical Marxism 392, 537 Analytical sociology 312, 507, 544

Analytic philosophy 307, 310, 311, 399, 615, 616, 625, 718, 721, 722, 724 Anomie 369, 370, 384, 518–520, 523, 525, 528, 529 Anthropological 206, 209, 213, 215, 217, 222, 223, 301, 324, 539, 556, 578, 647, 648 Anthropology 49, 204–206, 209, 212, 214, 217, 246, 418, 472, 552, 667 Atomism 4, 18, 19, 299, 419, 466 Atomistic 6, 7, 18, 19, 22, 300, 308, 436–438, 690, 720 Austrian School of Economics 4, 9, 12, 13, 246, 571, 573, 577, 610 Axiological 39, 51, 228–233, 241–243, 246, 247, 251–253, 255–265, 303, 356, 358, 427, 712

© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 N. Bulle and F. Di Iorio (eds.), The Palgrave Handbook of Methodological Individualism, https://doi.org/10.1007/978-3-031-41512-8

741

742

Index

B

Behavioral economics 381, 593, 594, 656, 658–662, 664–669, 671–674 Behaviorism 22, 298, 310, 412 Behaviorist 309, 310 Belief 63, 64, 67, 69, 71, 90, 110, 117–119, 124, 125, 127, 146, 166, 175, 185, 186, 227–229, 233, 236, 246, 247, 253, 255–257, 263, 272, 275, 277–279, 283, 287, 301, 338, 356, 361, 364, 379, 381–384, 391, 395, 396, 426, 440, 459, 463, 475–479, 493, 499, 501, 507, 517, 532, 559, 610–613, 615, 617, 618, 622, 624, 625, 628, 631, 632, 634–637, 640–646, 648–651, 656, 657, 661, 665, 666, 669, 671, 695, 696 Biology 79, 84, 153, 177, 411, 426, 467, 486, 490, 491, 493, 494, 499, 646, 695, 697 Bounded rationality 381, 590, 591, 651, 660, 670

C

Capitalism 63, 64, 173–176, 186, 349, 364, 393, 610–613, 694 Causal 6, 42, 46, 47, 49–51, 84, 86, 87, 95, 96, 127, 129, 134, 149, 154, 183, 185–189, 192–195, 199, 271, 293–295, 297, 298, 301–307, 309, 310, 312–315, 324, 328, 333–335, 377, 379, 383–385, 387, 390, 392, 395, 397, 398, 497–500, 502, 505, 538, 546, 552, 568–571, 612, 613, 615, 622, 623, 625, 633, 641, 642, 644, 645, 658, 674, 682, 689, 694, 695, 712, 713, 723

Causal explanation 49, 51, 86, 87, 96, 149, 154, 185–187, 194, 195, 271, 305, 613, 658, 694, 712 Causality 44, 46, 48, 84, 183, 190, 279, 294, 301, 314, 409, 415, 486, 496, 546, 547, 570, 590, 597–599, 644, 674, 695, 712, 723 Causation 83, 84, 301, 302, 326, 487, 497, 499, 500, 506, 507, 549, 569, 621, 622, 712 Cause 38, 44, 46, 51, 64, 84, 108, 118, 129, 132, 146, 149–151, 166, 183, 185, 186, 193, 233, 239, 264, 274, 294, 295, 297, 301, 309, 313, 314, 323, 331, 359, 362, 364, 379, 396, 441, 498, 502, 504, 507, 525, 528, 530, 531, 597, 599, 609, 610, 613, 616, 622, 624, 625, 628, 634–636, 644, 645, 650, 666, 672, 674, 701 Chance 274, 281, 297, 325–328, 331, 336, 337, 341, 342, 369, 570, 600, 601, 671 Class struggle 114, 123 Cognitive 41, 42, 46, 53, 61, 146, 150, 154, 166, 191, 197, 228, 230–235, 241–243, 245, 247, 252, 253, 257, 272, 279–287, 311, 359, 368, 384, 411–414, 417, 420, 422, 424–426, 478, 527, 590, 593, 594, 624, 631, 634, 637, 639, 649, 650, 658–660, 664, 671, 672, 674, 695 Cognitive bias 280, 593, 594 Cognitive psychology 245, 281, 426, 593, 594, 658 Cognitivist 421 Collective action 105, 233, 350, 362, 388, 390–394, 595–597, 601, 605, 606, 658

Index

Collective belief 632, 634–637, 645, 649, 651 Collective concept 57, 60–62, 64, 324, 327, 394, 395 Collective consciousness 85, 495, 524 Collective entities 58, 78, 305, 323, 499, 550, 724 Collective properties 386, 396, 407, 546, 548, 549 Collectivism 7, 57, 58, 63, 103, 309, 313, 321, 323, 324, 457–459, 463, 466, 535, 537, 551, 649 Collectivist 57–59, 63, 78, 457, 458, 467, 485, 497, 545, 546 Communism 120 Communist 122, 641 Community 19, 51, 52, 83, 106, 108, 113, 132, 154, 213, 265, 284, 352, 356, 365–367, 369, 413, 416, 419, 425, 436, 547–549, 552, 557, 558, 617, 641, 643, 644, 723 Complex 4, 10, 12, 14, 33, 37, 40, 46, 47, 52, 61, 65, 112, 124, 171, 186, 234, 240, 245, 282, 285, 294, 296, 298, 301, 306, 312, 323, 333, 334, 336, 358, 361, 365, 392, 393, 409–414, 417, 423–426, 479, 526, 541, 545, 549, 554, 556, 559, 568, 574, 588, 591–593, 595, 597, 602, 606, 621, 625, 633, 635, 637, 664, 686, 687, 699, 705, 718, 722 Complexity 23, 46, 47, 52, 143, 279, 282, 307, 313, 324, 329, 334, 407, 409, 410, 420, 425, 572, 581, 604, 659, 698 Comprehensive sociology 243, 271, 273–275, 279–281, 283–285, 646, 714

743

Computational 303, 311, 410–412, 422, 424, 619, 658, 681, 684, 691, 696, 698, 699, 703 Concept 3, 4, 6, 19, 36, 39, 42–45, 47–50, 57–59, 66, 77–79, 82, 83, 91, 95, 129, 142, 144, 147–149, 151, 161–163, 166, 167, 172, 178, 239, 251, 252, 254, 255, 259, 260, 262, 264, 298, 303, 322, 324–328, 333, 334, 336, 341, 347, 355, 361, 369, 382, 395, 396, 408, 412, 415, 417, 418, 420, 421, 426, 446, 447, 500, 501, 503, 505, 507, 518, 528, 529, 543, 545, 552, 553, 570, 571, 573, 574, 577, 604, 616, 617, 619, 623, 633, 638, 641, 683, 689, 695, 697, 701, 703, 713, 714 Conceptual 47, 61, 80, 114, 129, 131, 168, 182, 304, 312, 333, 334, 341, 342, 409–411, 491, 500, 502, 505, 525, 538, 591, 592, 594, 597, 633, 646, 660, 689, 711–713, 720 Condorcet pradox 125 Conflict 8, 51, 93, 94, 119, 122, 125, 135, 155, 191, 257, 353, 357, 417, 423, 439, 457, 525, 527–529, 532, 540, 545, 554, 555, 558, 561, 588, 605, 626, 646, 657, 668, 721 Consciousness 41, 42, 78, 80, 82, 84, 85, 107, 118, 122, 176, 196, 215, 252, 277, 295, 296, 298, 300, 538, 575–577, 581, 626, 627, 633, 641, 642, 644 Consequentialism 241, 427 Consequentialist 242, 244, 256, 264, 356, 427 Constructivism 238, 240, 241, 399, 604, 648, 649 Cooperation 4, 5, 7, 9–13, 131, 133, 155, 156, 295, 351, 353,

744

Index

365, 416, 419, 420, 422–425, 436, 442, 444, 545, 589, 595, 597, 602, 603, 605, 662, 663, 668 Coordination 4, 14, 143, 152, 153, 155–157, 295, 325, 329, 331–333, 337, 341, 419–422, 436–438, 442, 448, 449, 470 Covid-19 206, 208, 209, 217, 221–223 Critical realism 309, 312, 313, 399, 492 Critical realist 309, 312, 313 Cultural 33, 37, 39, 46, 48–51, 53, 80, 106, 110, 130, 153, 175, 177, 184, 192, 193, 209, 234, 239, 240, 258, 259, 278, 281, 287, 305, 407, 418, 420, 421, 425–427, 475, 538, 539, 541–543, 557, 594, 596, 612, 667–671, 673, 725 Cultural evolution 418, 420, 421, 426, 427, 725 Culturalist 228, 239, 256, 258 Culture 34, 42, 48–53, 70, 82, 108, 113, 175, 239, 251, 272, 311, 530, 539, 554, 559, 603, 612, 632, 640, 669

D

Decision-making 107, 174, 175, 277, 355, 358, 440, 441, 576, 589, 659, 666 Deliberation 165, 614, 619, 621–623, 626–628 Democracy 58, 131–133, 187, 251, 255, 259, 362, 364–367, 388, 502–504, 548 Democratic 123, 133, 219, 252, 255, 261, 349, 364–366, 388, 419, 489, 502–504 Democratization 111, 254 Determinism 44, 46, 186, 187, 674

Determinist 96 Dignity 37, 40, 62, 251–255, 257–265, 338, 516, 593 Dualism 80, 81, 307, 325, 476, 493–497, 501

E

Economic behavior 38, 112, 393 Economic calculation 162, 172–174, 649 Economic order 24, 25, 327, 425, 673 Economics 3–9, 12, 13, 17–19, 35, 37–39, 42, 61, 71, 72, 89, 91–93, 98, 105, 113, 143, 162, 166, 169, 170, 177, 178, 280, 298, 299, 311, 329, 348, 351, 352, 393, 398, 413, 437–439, 444, 445, 447–449, 469, 475, 480, 535, 568, 570, 572, 574, 575, 589, 594, 610, 637, 638, 641–643, 645, 655, 656, 658–661, 665, 667, 668, 671–674, 685, 687, 697 Economist 3, 4, 7, 8, 13, 35, 36, 38, 80, 92, 97, 98, 162, 165, 185, 194, 352, 434, 437, 444, 445, 447, 458, 571, 572, 574, 575, 642, 655, 659–667, 670–674, 686, 702 Egoism 58, 241, 516, 520, 524, 525 Emergence 3, 9–12, 62, 72, 82, 83, 89, 167, 236, 259–261, 287, 296, 307, 311, 324, 350, 356, 359, 362, 363, 371, 396, 398, 407, 409–412, 415, 417, 422, 486, 497, 513, 517, 520, 532, 545, 600, 601, 672, 674, 684, 687, 695 Emergentism 296, 312, 722 Emergentist 296, 312–315, 690, 718, 720

Index

Emergent properties 313, 396, 545, 687, 719 Emotion 21, 133, 141–157, 170, 233, 234, 246, 355, 556, 651 Emotional 142, 147, 152, 232, 340, 355, 611 Empirical 8, 11, 33, 34, 39, 42, 46, 47, 49, 64, 68, 69, 91, 96, 97, 119, 132, 133, 164, 167, 168, 170, 181–184, 188, 192, 193, 197, 198, 210, 279, 299, 301, 302, 304, 308, 310, 325, 326, 328, 332–337, 340, 342, 350, 353, 354, 357, 364, 382–385, 387, 388, 390, 391, 395, 398, 399, 414, 422, 465, 486, 487, 489, 491, 493, 494, 498–508, 514, 517, 518, 528, 535, 547, 551–553, 557, 569–574, 578, 580–582, 592–594, 603, 646, 647, 660, 663, 682, 700, 703, 704, 719, 721, 722 Empiricism 298, 307–310 Empiricist 50, 307, 309, 310, 557 Epistemological 79, 86, 161, 163, 170, 182, 184, 185, 188, 191, 193, 272–274, 285, 297, 308, 322, 407, 409, 494, 557, 580, 581, 628, 643, 682, 695, 703, 714, 721 Epistemology 39, 80, 163, 297, 301, 304, 314, 322 Equality 106, 251, 252, 255, 258–260, 263, 265, 365, 517, 529, 674 Equilibrium 9, 13, 92, 128, 410, 413, 423, 437, 440, 441, 443, 445, 447–449, 461, 470, 471, 473–476, 479, 480, 538, 719 Ethical individualism 119, 120 Ethnic 177, 351, 353, 599, 638 Ethnicity 240, 504 Eusociality 415, 416, 419–421, 426

745

Evolutionary 79, 80, 82, 83, 98, 203, 273, 280, 293, 294, 296, 415, 416, 420, 425–427, 479, 667 Evolutionary game theory 422, 424 Evolutionism 37, 287, 296 Explanation 7, 10, 11, 19, 24, 26, 31, 32, 40, 44–47, 51–54, 60, 64, 69, 71–73, 77–79, 83, 85–87, 89, 90, 93–96, 105, 106, 112, 113, 123, 128, 129, 131, 134, 141–143, 146, 150, 154, 166, 182, 183, 185–195, 199, 204–211, 217, 222, 223, 234, 235, 239, 242, 246, 252, 253, 280, 285, 297, 302, 304, 310, 312, 314, 333, 347, 349–351, 356, 358, 363, 364, 369, 371, 378, 379, 384, 389–393, 395, 397, 398, 410, 412, 434, 457, 464, 492, 496, 556, 568, 610, 611, 615, 617–620, 623–627, 632–637, 640–645, 648–651, 655, 659, 672, 683, 685, 688, 689, 691, 694, 700, 701, 711, 713, 714, 722 Explanatory 66, 80, 93, 95, 96, 111, 118, 123, 129, 131, 143, 182, 188, 189, 193, 194, 204, 205, 208, 210, 211, 247, 253, 303, 304, 312, 313, 349, 385, 394, 398, 496, 497, 499, 500, 507, 535, 536, 541, 546, 568, 618, 636, 645, 650, 672, 682, 683, 688, 689, 691, 694, 695, 698, 700, 702, 705, 711, 713, 723 Explanatory model 22, 154, 244, 281, 526, 574, 682–684, 703, 712, 714, 719, 722

F

Fable of the bees 359, 416–418

746

Index

Fallibilism 721 Freedom 31, 38, 40, 43, 46, 53, 58, 110, 123, 130, 219, 262, 295, 370, 421, 444, 517, 524, 544, 559, 595, 596, 603, 690 Free will 40, 507 Functional 60, 95, 120, 123, 131, 205–208, 237, 295, 300, 304, 306, 311, 313, 410–412, 505, 506, 515, 538–543, 545, 552, 554, 556 Functionalism 203, 204, 207, 208, 210, 243, 410, 473, 539–541, 551–553, 555–557, 613, 627 Functionalist 203–207, 256, 473, 475, 552, 555, 595

G

Game theory 92, 424, 475, 478, 479, 686 General equilibrium 119, 448, 449 Good reason 21, 171, 182, 184, 191, 193, 264, 303, 603, 645, 713 Group agency 129, 133

H

Habit 33, 233, 360, 523 Habitus 111, 230, 234, 293, 304, 356, 714 Hermeneutic 51, 70, 181, 191–199, 246, 465, 577, 580, 581, 721 Heuristic 97, 181, 283, 300, 334, 541, 604, 606, 664, 704 Historical 12, 20, 32, 34, 37–42, 45–52, 54, 65, 68, 69, 71, 78, 97, 98, 119, 120, 147, 150, 161, 164, 167–173, 178, 179, 182, 187, 188, 192, 193, 198, 203, 254, 259, 300, 305, 333, 334, 336, 338, 341, 366, 393, 427, 463, 475, 519, 527, 530,

537, 538, 542, 543, 545, 555, 558, 579, 610, 611, 633, 643, 671, 690, 693, 719–722, 724 Historicism 68, 70, 296, 298, 308, 462 Historicist 65, 68, 170, 187, 293 History 3, 12, 32, 34–37, 45, 46, 48, 49, 51, 52, 62, 67, 68, 70, 73, 78, 82, 86, 106, 110, 117, 122, 123, 136, 156, 167–169, 171, 172, 177, 205, 206, 215, 216, 228, 259, 260, 280, 285–287, 300, 323, 336, 364, 392, 409, 416, 418, 434, 445, 457, 460, 465, 475, 547, 557, 569, 571, 591, 614, 631, 632, 660, 683–685, 688, 719, 722 Holism 69, 78, 84, 85, 104, 129, 293–298, 301, 303, 306, 309, 311, 313–315, 322, 419, 466, 492, 495–497, 526, 535, 649 Holistic 7, 19, 78, 105–108, 110–114, 203, 263, 294–297, 322, 527, 610, 614, 615, 645, 724 Homo economicus 4, 19, 25, 245, 246, 573, 592, 656, 657 Human action 3, 6, 9, 10, 22, 32, 38–40, 43, 44, 46, 47, 53, 60, 63, 65–67, 69, 70, 80, 85, 163, 164, 166, 167, 169, 178, 182, 191–193, 195, 198, 204, 207, 208, 233, 305, 323, 325, 336, 421, 422, 437, 451, 460, 462, 464, 469, 471, 477, 479, 480, 523, 528, 558, 569–572, 575, 590, 596, 611–614, 616–619, 621, 622, 624–626 Humean 262, 263, 297, 302, 621 Humeanism 308 Hypostasized 62, 64, 327

Index I

Ideal 44, 49, 54, 89, 112, 168–172, 300, 325, 328–331, 333–336, 342, 364, 365, 394, 437, 440, 531, 559, 568, 572, 576, 581, 582, 587, 588, 658, 682 Ideal types 576 Ideal typical 112, 170, 246, 271, 328, 330, 334, 336, 631, 686, 722 Identity/identities 81, 95, 96, 107, 251, 254, 335, 348, 447, 497, 559, 592, 594, 599, 632, 633, 668, 669 Ideology 19, 59, 108, 114, 175, 176, 181, 324, 560, 632–634, 637, 640–642, 644, 645, 650 Income inequality 378, 385, 390, 396 Inequality 337, 378–381, 384–390, 554 Institution 4, 6–8, 10, 12, 13, 20, 22, 51, 58–62, 65–67, 71–73, 78, 80, 85, 90, 108, 117, 125, 127, 134, 203, 205, 207, 219, 233, 254, 258–261, 295, 350, 352–354, 358, 362–364, 367, 368, 377, 390, 393, 419, 422, 425, 426, 437, 441, 442, 447, 457, 460–481, 505, 538, 539, 541, 542, 554, 596, 612, 613, 616, 626, 627, 658, 661, 666, 667, 669–674, 691 Institutional 8, 9, 11, 60, 127, 301, 305, 347, 367, 388, 393, 407, 444, 461, 465, 466, 468–470, 472, 475–480, 542, 543, 545, 554, 558, 612, 614, 624, 626, 672–674, 683, 691 Institutional analysis 13, 393, 475, 671, 672 Institutional individualism 204, 457, 460, 461, 465–467, 472–475, 480

747

Institutionalized 339, 359, 479 Instrumental 162, 163, 166, 178, 185, 228–233, 241–243, 247, 252, 263, 264, 272, 280, 303, 355, 356, 358, 362, 539, 588, 603, 624, 650, 672, 712 Instrumentalism 162 Instrumentalist 228 Interaction 3, 4, 8, 9, 20, 22, 72, 80–82, 84, 86, 117, 126, 127, 129, 132, 134, 135, 142, 152, 155, 156, 203, 237, 271, 295, 300, 312, 328, 331–333, 359, 385, 391, 393, 396, 408, 409, 417, 418, 423, 425, 437, 441, 457, 464, 477, 480, 488, 490–492, 494, 496, 500, 501, 544–547, 554–556, 559, 573, 574, 580, 581, 590, 591, 595–597, 599, 601, 602, 605, 606, 617, 646, 647, 655, 660, 663, 671, 682, 684, 687–689, 691, 694, 696, 702, 722 Interactionism 79–82, 85, 98, 352, 540 Interactionist 168, 558, 589 Interdependence 108, 237, 238, 296, 353, 357, 359–361, 515, 589, 591, 595, 597, 602, 604, 605, 637, 685, 686, 697, 699, 700, 702, 703 Interpretative 191, 192, 194, 198, 262, 263, 265, 278, 309, 312, 339, 581, 628, 689, 719 Interpretive 31, 135, 191, 193–195, 199, 304, 308, 309, 323, 324, 337, 459, 569, 577, 581, 674, 711, 714, 723 Invisible hand 3, 7–9, 12, 13, 62, 359, 417, 421, 424, 440, 444 Irrational 38, 43, 97, 111, 113, 114, 147, 148, 156, 163–167, 212, 214, 217, 222, 277, 305, 438, 472, 568, 594, 642, 643, 645

748

Index

Irrationality 37, 38, 43, 165, 193, 220, 221, 275, 280, 361, 438, 529, 594, 644

307–310, 334, 353, 355, 357, 368, 389, 468, 469, 491, 501, 503, 567, 572, 604, 606, 638–640, 646, 700

K

Knowledge 6, 11, 33, 35–37, 41, 42, 45–48, 51, 82, 98, 104, 111–113, 163–165, 168, 176, 181, 183, 185, 187, 188, 191–195, 198, 199, 212, 215–217, 228, 230, 232, 240, 243, 252, 260–262, 273, 274, 276, 283, 300, 304, 305, 312, 322, 326, 334, 340, 365, 382, 389, 398, 411, 424–426, 442, 443, 459, 464, 465, 472, 477, 488, 506, 540, 544, 556, 560, 567, 569, 572, 573, 603, 617, 618, 631–634, 636–640, 644–651, 682, 712, 720

L

Labor theory of value 121, 642 Law 13, 20, 34, 35, 38, 39, 43, 44, 46, 61, 71, 72, 79, 87, 93, 94, 96, 121, 123, 186, 187, 189, 230, 239, 255, 257, 259–261, 335, 340, 359, 362, 367, 368, 385, 387, 388, 413, 418, 421, 422, 427, 434, 435, 440, 441, 444, 450, 451, 460, 500, 502, 521, 522, 536, 545, 574, 610, 613, 638, 639, 642, 669 Liberal 254, 261, 417, 435, 514, 520 Liberalism 20, 119, 254, 259, 351, 417, 641 Logical 32, 34–37, 42, 44, 45, 47–50, 70, 120, 123, 132, 161, 164, 165, 168, 177, 178, 181–184, 189, 192, 193, 197, 198, 204, 219, 275, 303,

M

Macroeconomic 351, 433–444, 447–452, 686 Macrological 349 Macrosociological 349–355 Macro-to-micro 379, 380, 384–387, 395, 397, 701, 702, 723 Madness 209, 214, 222, 275, 305 Market 3–7, 9, 11, 13, 25, 66, 71, 105, 111, 124, 126, 128, 132, 136, 169, 172–174, 284–287, 323, 327, 329, 332, 342, 350, 359, 361, 363, 377, 393, 422, 425, 434, 437–442, 445–451, 461, 554, 589, 637, 655, 657–659, 662, 665–667, 672, 674 Marxian 175–177, 537, 540, 558, 640 Marxism 46, 68, 69, 176, 558, 627 Marxist 69, 176, 239, 293, 537, 641 Materialism 46, 81, 82, 494, 497, 498, 531, 640 Materialist 34, 81, 216, 530 Meaning 6, 31, 33, 42, 49–53, 62, 103–106, 111, 114, 131, 135, 144, 148, 155, 164, 170, 185, 191, 193, 195, 196, 234–237, 252, 260, 262, 263, 271–273, 275, 276, 281, 298, 300, 303–305, 308, 314, 323–325, 328, 332, 334, 335, 340, 347, 358, 361, 368, 369, 383, 390, 392, 394, 437, 440, 442, 444–447, 527, 529, 538, 554, 567–571, 574, 576, 577, 580, 581, 591, 595, 612, 614, 615,

Index

617, 620, 623, 648, 689, 692, 699, 703, 711–714 Meaningful 237, 305, 307, 309, 323, 410, 593, 621, 638 Means-ends 167, 183 Mechanism 8, 20, 62, 88, 108, 125, 126, 143, 149, 150, 152, 154–157, 186, 188, 283, 295, 331–333, 355, 359, 361, 417, 418, 435, 439, 442, 445–447, 491, 538, 545, 655–657, 665, 670, 686, 689, 702, 711, 713 Mental 13, 62, 80–82, 90, 124, 142, 152, 156, 157, 169, 176, 178, 196, 271–273, 275, 276, 279, 282, 284, 310, 312, 369, 449, 468, 476, 480, 481, 488–490, 493–500, 506, 507, 528, 572, 573, 591, 613, 615, 616, 620–622, 632, 633, 636, 637, 665, 670–674 Metaphysical 33, 40, 47, 53, 79, 80, 92, 93, 199, 216, 246, 307, 308, 417, 494, 610, 614, 615, 628, 647, 649 Metaphysics 39–41, 79, 246, 293, 302, 308, 313 Methodology 20, 31, 35, 36, 42, 78, 79, 92, 95, 97–99, 105, 107, 108, 110, 113, 127, 204, 246, 299, 322–324, 326, 328, 341, 434, 458–460, 469, 514, 528, 536, 537, 553, 555, 565–567, 576, 577, 580, 613, 614, 631, 634, 645, 660, 661, 665, 666, 691 Microeconomic theory 92, 435, 436, 668 Microfoundation(s) 118, 123, 132, 434, 435, 438, 439, 445, 447, 449–451 Micrological 350, 351, 353, 356

749

Micro-macro 298, 322, 323, 326, 327, 341, 342, 355, 378–381, 384, 386–396, 398, 416, 437 Microsociological 353 Mind 4, 5, 9, 10, 22, 23, 32, 38, 40, 41, 43, 51, 63, 64, 78–82, 84, 98, 104, 124, 129, 133, 147, 157, 163, 167–169, 177–179, 191, 193, 197, 198, 215, 220, 229, 252, 260, 273, 276, 279, 283–285, 300, 304, 305, 312, 336, 418, 420, 427, 467, 468, 470, 473, 476, 477, 479, 486, 494, 497, 500, 520, 521, 609, 617, 620, 621, 638–640, 684 Model 17, 18, 23, 59, 63, 79, 86–98, 112, 119, 181, 182, 186–188, 191, 193, 199, 204, 210, 211, 214, 216, 222, 223, 227, 228, 230, 232, 234, 239, 241, 242, 245–247, 256, 257, 272, 274, 279, 280, 285, 302–304, 307, 311, 312, 322, 324, 329, 330, 335, 341, 342, 359, 362, 363, 366–368, 378–382, 384, 387–390, 392–394, 398, 411, 413–415, 422, 424, 438–441, 448, 449, 465, 474, 524, 536, 537, 540, 541, 544, 545, 547, 549, 557, 567, 568, 571–576, 578, 582, 588, 591, 593, 594, 598–600, 602–604, 610, 611, 613, 614, 617, 621, 624, 625, 627, 628, 634, 643, 645, 655–658, 660–663, 665, 666, 670–674, 681–684, 686–689, 691, 693, 694, 696–700, 702–704, 711–714, 717–720, 722–724 Modernization 108, 255 Money 8, 9, 27, 73, 171, 173, 207, 211, 229, 234, 351, 360, 363, 366, 422, 437, 440, 442, 443,

750

Index

445–448, 452, 469, 471, 614, 635, 659, 667 Monism 46, 476, 480, 481, 493–495, 507, 508 Moral 5, 19, 20, 32, 103, 118, 123, 147, 149, 151, 162, 205, 222, 228, 230, 232, 251–265, 295, 367, 417, 418, 421, 427, 462, 467, 477, 515–527, 529–531, 594, 605, 623, 632 Moral individualism 514, 516–520, 524 Morality 171, 240, 253, 254, 257, 258, 260–263, 513, 516, 520, 522 Moral norm 126, 517, 527, 530 Moral sentiments 147, 149, 151, 162 Motivation 22, 51–54, 77, 104, 105, 110, 144–146, 153, 161, 167, 168, 170, 171, 177, 184, 204, 213, 275, 277, 303, 304, 314, 382, 391, 477, 478, 523, 526, 543, 569, 570, 592, 593, 595, 596, 612, 617, 698, 712 Motive 8, 12, 52, 54, 65, 78, 177, 185, 281, 323, 328, 330, 331, 334, 336, 338, 340, 341, 359, 360, 565, 567, 570, 574, 576, 612, 616, 619, 620, 641, 673 Multi-agent 245, 312, 414, 425, 713, 714 Multiple-realizability 311, 312 Multiple-realization 306, 499–501, 722

N

Naturalism 238, 241, 307 Neo-Kantian 33, 34, 302, 569, 576 Neo-Kantianism 329, 576 Network 109, 127, 132, 216, 263, 331, 347, 356, 368, 393, 409,

410, 412, 414, 536, 547, 548, 691, 699 Nominalism 131, 723, 724 Nomological 40, 41, 44–46, 48, 51, 96, 183, 186–188, 190, 191, 199, 305, 569, 713 Nomologico-deductive 181 Norm 4, 8, 33, 37, 85, 124–127, 135, 205, 233, 234, 256, 258, 278, 287, 329, 332, 333, 337, 338, 340, 356, 358, 359, 361, 362, 367, 368, 370, 382, 384, 385, 391, 395, 425–427, 475–479, 524, 528, 542, 556, 613, 617–619, 621, 624, 666, 667, 669–671, 673, 674 Normative 120, 131, 148, 164, 228–230, 251–253, 255–260, 262–265, 272, 287, 303–305, 331, 332, 338, 341, 367, 368, 427, 478, 519, 522, 529, 543, 557, 589, 591, 594, 621, 669, 712, 714

O

Objectivity 37, 49, 195–197, 252, 255–258, 262–265, 577, 582 Ontological 38, 40–43, 47, 49, 52, 57, 78, 80, 84, 86, 246, 294, 301, 302, 306–308, 313, 315, 322–324, 458, 466, 476, 481, 487, 492–497, 500, 501, 508, 581, 615, 616, 683, 692–696, 723, 724 Ontological individualism 458, 466, 481, 492–494, 496, 501, 508, 694, 724 Ontology 78, 79, 246, 301, 307, 308, 313, 314, 399, 458–460, 467, 487, 497, 499, 501, 507, 557, 574, 694, 695, 724 Organicism 298, 419

Index

Organization 63, 211, 296, 298, 307, 313, 347, 349, 353, 363, 366, 411, 412, 415, 416, 419–422, 425, 426, 478, 491, 503, 538, 539, 551, 554, 574, 575, 588–592, 595–597, 601–606, 614, 615, 633, 640, 663, 701 Organizational 293, 296, 307, 312, 587–596, 600–606

P

Paradox 125, 278, 279, 283, 355, 356, 358, 364, 418, 515, 573, 580, 642 Pareto-optimality 119 Part-whole 486 Passions 21, 143–148, 150, 151, 153, 154, 162, 165, 171, 255, 417 Personality 39, 82, 295, 297, 443, 516, 519, 525–527 Phenomenology 540, 558, 565–567, 570, 571, 574, 576–579, 581 Philosophy 32–35, 37, 50, 51, 58, 78, 79, 91, 120, 146, 147, 151, 245, 260, 274, 296, 302, 303, 308, 310, 311, 313, 314, 399, 418, 419, 435, 457, 460–462, 465, 467, 469, 472, 485–488, 490, 492, 497, 499, 500, 507, 514, 518, 520, 535, 537, 577, 614, 615, 622, 647, 719, 722, 723 Philosophy of mind 84, 311, 467, 480, 486, 487, 489, 490, 493–500, 506, 507 Physicalism 306, 307, 311, 313, 315, 480, 481, 493, 494, 497–499 Physicalist 84, 297, 306, 307, 310–314, 499

751

Political economy 32, 37, 38, 84, 165, 176, 444 Political individualism 119, 120 Political science 104, 697 Polylogism 162, 172, 175–178, 637–640 Positivism 197, 238, 243, 256, 310, 536, 557, 645 Positivist 197, 228, 239, 240, 297, 301, 303, 306, 314, 556, 557, 645 Praxeology 163, 166–169, 181, 182, 184, 191–193, 572, 575 Prediction 68, 69, 87, 89, 239, 279, 351, 357, 656, 659, 662, 663, 668 Pre-social 520, 521, 524–528 Price-theoretic approach 437, 440, 442, 443, 446, 450–452 Problem-solving 182, 639 Psychic 41, 43, 50, 286, 309 Psychical 295, 538 Psychological 43, 47, 78, 90, 119, 123, 135, 204, 223, 239, 245, 298, 309, 310, 323, 338, 351, 352, 381, 391, 411, 412, 438, 459, 462–464, 466, 467, 485, 488, 489, 493, 498, 504, 525, 543, 545, 546, 589, 593–596, 612, 614, 623, 625, 644–646, 659, 666 Psychological individualism 78, 85, 90 Psychologism 65–67, 78, 204, 323, 324, 460–467, 474, 480

R

Radical 31, 38, 81, 107, 240, 307, 494, 495, 554, 566, 576, 582, 595, 596, 604, 619, 627, 646 Radicalization 389 Rational 24, 25, 42, 43, 52, 68, 107, 111, 113, 119, 125, 128, 129,

752

Index

143–148, 150, 151, 154–157, 162–165, 167, 169–175, 177, 182, 183, 185, 190–193, 204, 207, 211, 212, 214, 216, 217, 220–223, 233, 238, 240, 241, 252, 256, 262, 265, 272, 282, 287, 302, 304–306, 312–315, 325, 326, 329–332, 334–342, 354, 355, 357, 358, 361, 362, 382, 414, 418, 419, 422–424, 426, 427, 438, 472, 474, 543, 545, 568, 573, 590, 594, 611, 614, 619, 624, 626–628, 634, 638, 641, 642, 645–650, 656–660, 663, 666, 711–714 Rational choice 13, 19, 23, 184, 279, 326, 329, 359, 362, 368, 381, 383, 418, 550, 670, 693 Rational choice theoreticians (RCT) 227, 243, 245, 304, 327, 362, 381–383, 422, 537, 570 Rational emotion 146 Rationalism 161, 162, 399, 462, 472 Rationalist 217, 352, 354, 424, 457, 466, 641 Rationality 43, 52, 60, 70, 79, 88, 91, 93, 94, 97, 114, 145, 146, 148–152, 161–172, 174–176, 178, 179, 181, 182, 184, 185, 190, 198, 204–206, 208, 209, 211, 212, 214–217, 221–223, 227, 228, 230–235, 238, 239, 241, 243–246, 251–253, 255–257, 259–265, 271, 272, 274, 276–281, 283, 286, 287, 303–306, 314, 321, 323, 329, 330, 354–356, 358, 362, 382, 417, 422–424, 427, 438, 461, 468, 472, 474, 475, 568, 570, 573, 580, 588, 590, 594, 596, 600, 603, 605, 611, 619, 621, 624, 637, 644–647, 649, 651,

656, 658, 660–666, 670, 689, 711–714 Rationalization 234, 259, 261, 339, 590, 647 Realism 61, 63, 64, 104, 299, 301–303, 312, 410, 497, 647, 659, 712 Realist 125 Reason 7, 12, 20, 21, 27, 32, 38, 43, 47, 57, 60, 62, 68, 69, 78, 79, 82, 88, 91, 93, 97, 98, 104–107, 111–114, 124, 130, 142–153, 156, 162–165, 175, 177, 179, 182–186, 193, 194, 198, 227–244, 246, 247, 253, 254, 256, 257, 262–265, 272–276, 278, 282, 284, 286, 294, 303, 305, 310, 314, 332, 336, 356, 369, 383, 390, 417, 418, 425, 427, 434, 435, 438, 444, 445, 463, 465, 468, 477, 489, 492, 501, 502, 504, 513, 529, 531, 532, 567, 569, 576, 577, 588, 593, 594, 599, 602, 611, 612, 616, 617, 619, 622–625, 631, 635, 637–641, 644–646, 648–651, 658, 659, 669, 673, 686, 689, 691, 693, 695, 696, 700, 704, 711–714, 719 Reasoning 9, 12, 147, 174–176, 178, 182, 194, 222, 223, 255, 257, 258, 260, 262–265, 279–281, 284, 303, 356, 370, 398, 433, 446, 486, 592–595, 609, 619, 622, 624, 634, 635, 637, 642, 643, 645, 647, 684, 688–690, 692, 712, 717, 718, 720, 721, 723 Reduction 53, 77, 105, 229, 264, 271, 296, 299, 307–311, 327, 329, 463, 468, 497, 499, 501, 580, 626

Index

Reductionism 82, 83, 86, 118, 245, 296, 297, 300, 307–311, 315, 325, 378, 397, 398, 408–410, 690, 700, 711, 718–720 Reductionist 83, 297, 298, 301, 303–315, 397, 437, 462, 494, 495, 545, 625, 626, 685, 690, 694, 718–721, 723, 724 Relativism 37, 223, 240, 259, 264 Relativist 252 Religion 71, 123, 124, 210, 240, 251, 253, 323, 351, 357, 368, 418, 422, 516, 538, 613, 641 Religious 67, 106, 108, 123, 124, 171, 186, 209, 227, 228, 255, 275, 349, 351, 353, 362, 364–367, 516, 525, 532, 554, 611–613, 632, 648, 650 Rules 22, 36, 46, 60, 66, 92, 121, 130, 149, 150, 188, 234, 239, 278, 305, 334, 335, 339, 361, 367–369, 421, 441, 444, 449, 461, 464, 468, 470–475, 477–480, 492, 522, 528, 538, 603, 614, 616, 618–620, 724

S

Say’s Law 434–436, 438, 440, 444–449, 451, 452 Science 4–6, 8, 10, 31–34, 37–54, 61–72, 77–80, 86, 88–90, 92–98, 103, 105, 111–113, 131, 136, 141, 143, 161–163, 167, 169, 181, 184, 186–188, 190, 194, 197–199, 204, 208, 241, 257, 272, 274, 280, 293, 298, 299, 303–306, 308, 310–312, 314, 327, 347, 349, 352, 363, 378, 393, 396, 397, 407–413, 457, 459–462, 465, 467, 471, 472, 474, 486, 487, 489–491, 552, 557, 565–567, 570, 574, 580, 589, 610, 613,

753

614, 626, 627, 631, 633, 640, 641, 644, 646, 686, 688, 689, 699, 705, 712, 714, 718, 719, 721–724 Scientific 7, 9, 32–36, 38, 40–44, 46, 48, 51–53, 68, 70, 74, 79, 80, 87, 89, 92–94, 96, 120, 142, 147, 152, 163, 164, 166, 175–177, 181, 182, 189–192, 195, 197–199, 204–206, 213, 214, 217–221, 223, 232, 240, 247, 256, 257, 260–262, 264, 265, 274, 280, 298, 300, 301, 307, 308, 310, 314, 322, 326, 327, 331, 336, 348, 398, 407, 410, 411, 436, 440, 444, 457, 459, 461, 462, 469, 471, 476, 480, 486–488, 490, 491, 493, 495, 498–502, 507, 541, 567, 568, 570–578, 580–582, 621, 632, 633, 637, 639, 640, 643–650, 655, 672, 722, 723 Scottish Enlightenment 6, 421, 719 Self-interest 5, 13, 19, 25, 144, 166, 167, 184, 325, 417, 418, 436, 516, 524, 656, 657, 661, 662, 670 Sensualism 238, 240 Situational analysis 60, 78–80, 86, 88, 90–92, 94, 97, 98, 188–190, 459, 464, 465 Social change 206, 217, 236, 360, 529, 531, 613, 661, 666, 667, 671 Social control 519, 539 Social form 527 Social influence 177 Socialization 278, 554, 589, 596 Socialized 113, 114, 130, 596 Social movement 105, 106, 336, 383 Social need 124, 593 Social norm 23, 124, 126, 127, 191, 246, 359, 361, 364, 367, 368,

754

Index

426, 427, 617–619, 621–623, 627, 632, 691 Social structure 9, 52, 108, 110, 111, 114, 240, 295, 299, 313, 324, 341, 418, 544, 552, 554, 556, 603, 627, 687, 711 Social value 207, 417 Solidarity 206, 254, 369, 515–517, 520, 527, 531, 554 Spontaneous order 6, 8, 409, 412, 414, 417–419, 422, 426 Structural functionalism 554, 555 Structural reasons 134, 135 Structuration 321, 537 Structure 8, 12, 32, 52, 64, 65, 94, 96, 105, 107–111, 114, 118, 122, 123, 153, 167, 168, 170, 176–179, 184, 192, 198, 205, 239, 258, 293–297, 300, 304, 306, 307, 310, 312, 313, 325, 328, 335, 337, 347, 349, 353, 354, 358, 361, 366, 378, 379, 389, 393, 396, 407, 408, 410, 412, 413, 415, 416, 424, 425, 438, 443, 469, 475, 478, 487, 488, 490, 491, 502, 503, 527, 537, 539–542, 544, 545, 554, 556, 566, 577–581, 588–590, 596, 599–605, 619, 632, 639, 640, 656, 671, 685, 689, 698, 700–702, 711–714, 723 Subjective theory of value 5, 164 Subjectivism 4, 64 Subjectivity 311, 557, 605, 714 Suicide 112, 213, 222, 350, 356, 368–370, 518, 520, 525, 526, 528 Supraindividual 120, 121, 127, 701 Sympathetic individualism 19, 24 Sympathy 5, 10, 20, 21, 25, 26, 106, 119, 147, 418, 516 System 6, 8, 9, 11, 12, 23, 24, 44, 64, 68, 69, 73, 83, 88, 89, 97, 105, 107, 109, 111, 121, 124,

129, 162, 172, 173, 186, 205, 206, 208, 216, 228–230, 233, 235, 238, 256, 257, 263, 273, 278, 295, 310, 312, 321, 349–351, 357, 364, 389, 395, 396, 409–411, 424, 425, 443, 447, 460, 462, 468–470, 476, 477, 487, 488, 518, 538, 541, 543, 544, 549, 552, 554, 567, 598, 614, 624, 636, 638, 640, 642, 648, 650, 659, 666, 673, 686, 689, 691, 700, 701, 712, 713, 722 Systemic 186, 349, 350, 425, 485, 538, 543, 597, 686, 690, 694, 718, 720, 722

T

Three-worlds 79–82, 90, 98 Thymology 167, 168, 171, 179 Totalitarian 104, 638 Totalitarianism 462 Tradition 13, 113, 147–149, 170, 179, 181, 191, 193, 203, 234, 254, 271, 272, 281, 306, 329, 338, 339, 354, 357, 361, 369, 386, 421, 458, 497, 520, 526, 535, 553, 555, 569, 592, 617, 620, 640, 641, 645–648, 655, 661, 671, 698, 719, 721

U

Unconscious 124, 130, 170, 171, 196, 204, 305, 314, 515, 627, 670 Understanding 6–8, 12, 13, 22, 31, 32, 36, 40, 44, 45, 48, 50–54, 62, 69, 71, 72, 79, 80, 92, 94, 98, 124, 126, 154, 157, 161, 163, 166–172, 174, 188, 189, 191, 192, 196, 197, 199, 211, 228, 236, 237, 240, 242, 253,

Index

255, 257, 271–273, 276, 278, 299–301, 303–305, 309, 310, 314, 323, 325, 342, 433–435, 437, 440, 442, 445, 471, 486, 502, 513, 523, 536, 556, 567, 568, 576, 580, 589, 591, 592, 601, 605, 617, 620–622, 638, 648, 660, 663, 668, 671, 674, 683, 687, 690–694, 697, 700, 703, 704, 712–714, 720, 723, 724 Understanding explanation 31, 48, 51–53 Understanding interpretation 310 Understanding sociology 299, 304, 305 Unexpected 66, 89, 142, 154, 208, 237, 348, 349, 356–360, 365, 503, 506, 558 Unintended 13, 19, 20, 59, 60, 65–67, 69–73, 154, 181, 204, 207, 208, 247, 358, 360, 417, 418, 422, 460, 469, 471, 473, 479, 480, 603, 605, 612, 705 Universalism 640 Universalization 254, 257 Utilitarian 163, 166, 170, 178, 232, 243, 246, 418, 427, 514, 518,

755

520, 521, 524, 526–529, 576, 594, 712, 719 Utilitarian individualism 514, 524, 526, 527 Utilitarianism 162, 171, 245, 348, 427, 514, 518–520, 526, 527

V

Value 4, 5, 19, 31, 33, 34, 37, 40, 45, 49–51, 61, 117, 121, 122, 163–165, 167, 170–172, 178, 204, 214, 215, 251, 253–265, 298, 305, 325, 329–332, 335, 336, 338–341, 354, 356, 358, 364, 381, 408, 409, 417, 438, 446, 524, 540, 542, 581, 589, 611, 612, 623–625, 641–643, 659, 664, 665, 673, 688, 704 Value-rational 325, 330, 331, 335, 336, 338–341 Value rationality 170, 611 Verstehende 168

W

World 3 70, 80–82, 84, 85, 90, 304, 305, 461, 465, 476, 477, 479–481