Critical Realism and the Social Sciences : Heterodex Elaborations [1 ed.] 9781442684232, 9780802092151

Citation preview

CRITICAL REALISM AND THE SOCIAL SCIENCES: HETERODOX ELABORATIONS Edited by Jon Frauley and Frank Pearce

Critical realism is a distinct school of thought in philosophy and the social sciences that has been growing in significance over the past three decades. It offers important insights into aspects of our social and natural environments and the role of the social sciences by challenging conventional notions of the relationship between empirical experiences, actual events, and causal mechanisms. Critical Realism and the Social Sciences brings together contributors from both sides of the Atlantic, all of whom engage with tenets of critical realism, juxtaposing them with traditional representations of social scientific enquiry. United in the belief that the conceptual systems hitherto relied upon affect our styles of thought, ethical choices, political orientations, and so on, the contributors explore critical realism in relation to other currents of theoretical thought, thus suggesting a basis for evaluation and further elaboration. As a whole, the volume seeks to show how this particular approach provides a way to better understand many aspects of social existence, from human attributes to the interrelatedness of human activities and the natural world. In this much-needed study, critical realism is carefully examined, sympathetically assessed, and creatively developed by authors from diverse disciplinary backgrounds. is an assistant professor of Sociology in the School of Social Sciences at York University.

JON FRAULEY

FRANK PEARCE is

University.

a professor in the Department of Sociology at Queen’s

This page intentionally left blank

EDITED BY JON FRAULEY AND FRANK PEARCE

Critical Realism and the Social Sciences Heterodox Elaborations

U N I V E R S I T Y O F TO R O N TO P R E S S Toronto Buffalo London

www.utppublishing.com © University of Toronto Press Incorporated 2007 Toronto Buffalo London Printed in Canada ISBN 978-0-8020-9215-1

Printed on acid-free paper

Library and Archives Canada Cataloguing in Publication Critical realism and the social sciences: heterodox elaborations / edited by Jon Frauley and Frank Pearce. Includes bibliographical references and index. ISBN 978-0-8020-9215-1 (bound) 1. Social Sciences – Philosophy – Textbooks. 2. Critical realism – Textbooks. I. Pearce, Frank II. Frauley, Jon, 1972 – H61.15.C75 2007

300.1

C2007-903009-2

This book has been published with the help of a grant from the Canadian Federation for the Humanities and Social Sciences, through the Aid to Scholarly Publications Programme, using funds provided by the Social Sciences and Humanities Research Council of Canada. University of Toronto Press acknowledges the financial assistance to its publishing program of the Canada Council for the Arts and the Ontario Arts Council. University of Toronto Press acknowledges the financial support for its publishing activities of the Government of Canada through the Book Publishing Industry Development Program (BPIDP).

For Tara Milbrandt (Pearce) For Oliver, testament to the powers of non-human subjects (Frauley)

You propose then, Philo, said Cleanthes, to erect religious faith on philosophical scepticism; and you think, that if certainty or evidence be expelled from every other subject of enquiry, it will all retire to these theological doctrines, and there acquire a superior force and authority. Whether your scepticism be as absolute and sincere as you pretend, we shall learn by and by, when the company breaks up: we shall then see, whether you go out at the door or the window; and whether you really doubt if your body has gravity, or can be injured by its fall; according to popular opinion, derived from our fallacious senses, and more fallacious experience. And this consideration, Demea, may, I think, fairly serve to abate our ill-will to this humourous sect of the sceptics. If they be thoroughly in earnest, they will not long trouble the world with their doubts, cavils, and disputes: if they be only in jest, they are, perhaps, bad railers; but can never be very dangerous, either to the state, to philosophy, or to religion. – David Hume, Dialogues Concerning Natural Religion

This page intentionally left blank

Contents

Foreword by Ted Benton Acknowledgements Contributors

ix

xxi

xxiii

1 Critical Realism and the Social Sciences: Methodological and Epistemological Preliminaries 3 JON FRAULEY AND FRANK PEARCE

2 Bhaskar’s Critical Realism: An Appreciative Introduction and a Friendly Critique 30 FRANK PEARCE

3 For Realism and Anti-Realism

64

SERGIO SISMONDO

4 Critical Realism and God

74

GARRY POTTER

5 Rescuing Reflexivity: From Solipsism to Realism

97

ANTHONY WOODIWISS

6 More than Straw Figures in Straw Houses: Toward a Revaluation of Critical Realism’s Conception of Post-structuralist Theory 117 RICHARD DAY

viii Contents

7 Thinking across the Culture/Nature Divide: An Empirical Study of Issues for Critical Realism and Social Constructionism 142 RAYMOND MURPHY

8 Beyond Cognitive Critiques: Getting Real about Politics

162

JOSÉ JULIÁN LÓPEZ

9 Objectivity and Marxian Political Economy

179

ROBERT ALBRITTON

10 Why Is This Labour Value? Commodity-Producing Labour as a Social Kind 202 HOWARD ENGELSKIRCHEN

11 The Relation between Marxism and Critical Realism

224

HANS EHRBAR

12 Understanding Why Anything Matters: Needy Beings, Flourishing, and Suffering 240 ANDREW SAYER

13 The Expulsion of Foucault from Governmentality Studies: Toward an Archaeological-Realist Retrieval 258 JON FRAULEY

14 From Foucault’s Genealogy to Aleatory Materialism: Realism, Nominalism, and Politics 273 RONJON PAUL DATTA

15 Gadamer’s Minimal Realism 296 HOWIE CHODOS , BRUCE CURTIS , ALAN HUNT , AND JOHN MANWARING

Bibliography Index

347

317

Foreword

In their introductory chapter, the editors rightly note the great potential of critical realism (CR) as a stimulus to a renewal of social scientific enquiry. Given the evidence provided by the essays collected here of the immense diversity of current interpretations and uses of CR, I might be forgiven a few paragraphs of autobiography concerning its original context. Until the end of the 1960s polarized meta-theoretical commitments – positivism and hermeneutics – dominated philosophical thinking in and about the social scientific enterprise. This polarity could be represented by the revolt of ‘symbolic interactionism’ against Parsonian orthodoxy in the United States, or by Peter Winch’s devastating Wittgensteinian attack on the very idea of a social science. The revolt against ‘scientism’ in the social sciences gained further ammunition in the shape of philosophical histories of the natural sciences – the fashionable works of Kuhn, Feyerabend, and, for those of us interested in sources, the work of Koyré, Bachelard, and Canguilhem. Not only was empiricism (and therefore positivism) faulty as a philosophical theory of knowledge, but its precepts had no purchase on its paradigms of knowledge – the actual processes of scientific thought and progress. What were the implications of the critique of empiricism for our practice in the social sciences? For many it meant abandoning large-scale theorizing about ‘structures,’ and statistical analysis of large data-sets in favour of the more modest task of interpreting the small-scale interactions of social actors in specific action-situations. But the late 1960s and early 1970s were also a time of heightened political activism and of intellectual attention to its requirements. Politically radical thought relied on critical understanding of the sources

x Foreword

and nature of extant social inequalities, forms of social exclusion and stigmatization, feelings of frustration and lack of fulfilment, obstacles to desired social and economic change, and so on. If social research were to limit itself to a normatively and cognitively relativistic practice of interpretation of symbolic communication in small-scale actioncontexts, more or less all the critical resources of past radical thought would have to be abandoned. At the same time, the accumulated power of the critiques of empiricism/positivism was undeniable: indeed, it had its own radical implications, in rendering socially porous and sociologically analysable the very processes of scientific thought that had previously seemed beyond critical reach. In my own case, the prospect of a resolution to this dilemma emerged in the shape of the realist account of science then being developed by Rom Harré. Here was a non-empiricist view of science that both displayed the creative role of theory in the construction of scientific explanation and sustained a defence of its objectivity. What if the question of the possibility of objective knowledge in the social sciences were posed in relation to a non-positivist, realist account of the natural sciences? At least some of the arguments of the hermeneutic tradition would have to be either abandoned or re-thought. I worked on this project through the early years of the 1970s, and Russell Keat and John Urry, it turned out, were thinking on very similar lines. It was no coincidence, as we used to say, that Russell Keat and I were both graduate students at Oxford, and both strongly influenced by the work of Harré. Although, according to his own account, not influenced by Harré, Roy Bhaskar was also a graduate student at Oxford at around the same time. His Realist Theory of Science was published in the same year (1975) as Keat and Urry’s Social Theory as Science. My Philosophical Foundations appeared a couple of years later, and Roy’s application of his ideas to the social sciences (The Possibility of Naturalism) appeared the year after that. It was only at about this time that I became aware of Roy’s pathbreaking Realist Theory of Science – the book that I still see as his greatest intellectual achievement. Though they were developed quite independently of one another, and, indeed used rather different argumentative strategies, the three books converged in important respects: they drew on non-positivist and realist accounts of natural scientific knowledge. They posed the question of the possibility of a non-positivist science of society in the light of these accounts of natural science. They answered that question in the positive – though with varying degrees of qualification. Another common feature was the connection they made

Foreword xi

between a radical politics of the left and the defence of objectivity in the social sciences. This was expressed at the time in the repeated use of Marx’s Capital as an exemplar of a realist explanatory strategy, and an assumed affinity between the new realism and some form of Marxian or neo-Marxian critical theory. In my own work I took seriously the view (also shared by the three books) that the role of philosophy in the social sciences was that of ‘under-labourer’: in other words, that getting clear philosophically about the nature of scientific and social scientific theorizing and explanation was important in enabling new sorts of research practices, but could never substitute for that research. For me this meant ‘going native’ as a sociologist, abandoning the higher reaches of abstraction in favour of a more (eventually literally) ‘down-to-earth’ encounter both with the on-going research traditions in sociology and with substantive questions posed to inherited versions of the Marxian legacy. Several of these concerns intersect directly with the issues taken up by many of the essays in this volume. The first outcome was serious dissatisfaction with the qualified naturalism of Roy’s Possibility. To me this read as a qualified anti-naturalism, given the central role played in his argument by the dualism of ‘nature’ and ‘society.’ Frank Pearce gives an excellent account of my arguments against Bhaskar’s conception of social structure, since developed by Margaret Archer and himself, so I won’t repeat them here. My emphasis in that piece was on the methodological plurality of the natural sciences, suggesting that Bhaskar’s model of natural science not only fails to deal adequately with natural sciences in which experimental closure is not feasible but also ‘under-represents historicity and development as epistemologically significant characteristics of the objects of the objects of the natural sciences. Evolutionary biology, cosmology, geology, embryology are all natural sciences for which historicity and qualitative transformation pose epistemological and methodological problems which are in many respects directly comparable with those encountered in the sciences of human history, society and psychology’ (Benton 1981: 20). In other words, the methodological problems of the sciences do not stack up neatly on opposite sides of a categorical gulf between the natural and the social. The criticism went further than this. Bhaskar’s brilliant analysis of experimentation displayed the indispensability of human embodiment – of causally efficacious intentional action. But given that insight, why did the characterisation of social action and its relation to social structure

xii Foreword

empty the ontology of the social world of its material contents? Why were the facts of material embodiment and their psychic/social implications, the active role of material conditions of existence, of non-human living beings in the shaping of human social life and so on merely marginal presences in the ontology of The Possibility of Naturalism? At that stage, for me the issue was still one of finding an appropriate philosophical methodology to sustain a materialist understanding of social and historical processes. The aim was to develop a more adequate and sophisticated explanatory social theory that would, in turn, provide strategic orientation for a leftist politics. However, the residual Leninism implicit in this was soon to be eroded, partly by a succession of major political defeats, but more centrally by a growing awareness of very fundamental differences of value-standpoint within broadly conceived leftist/critical circles. How were issues of gender or racial oppression, stigmatized sexuality, cultural marginalization and ‘disablement’ to be normatively aligned with the emancipatory logic associated with the politics of class? Even more fundamentally (and, in retrospect, scandalously late on), I became aware of the salience of a radical green critique of capitalism and the deep challenge it posed to many of the unquestioned value orientations of the ‘traditional’ left. So, the hidden assumption that the purposes for which theory-based strategy could be taken as given was starkly exposed to questioning. If anything was to be rescued of an ‘emancipatory project,’ it would have to be one built from ‘below,’ recognizing value pluralism, respecting difference, but still seeking common ground through practice, selfcriticism, and dialogue. At this point we approach a key area of confusion surrounding the character of CR itself. Its significant achievement was critique and transcendence of the positivism/hermeneutics polarity in social science methodology. But this achievement was realized (perhaps I should say ‘actualized’) at a time of intense politicization of intellectual work, and the intellectual hegemony of various renditions of the Marxian heritage. This, and the epithet ‘critical,’ have led to the expectation that more can be asked of CR than mere underlabouring for the social sciences. Perhaps it can substitute itself for substantive scientific research? Perhaps it can draw from its own resources moral truths and a ready-made emancipatory project? Perhaps, it now seems, tell us about God and the spirit world? What can we make of all this? Was there some logical connection between realist philosophy of social science and the normative justification of an emancipatory political

Foreword xiii

project? What, exactly, was and is the ‘special relationship’ between Marxism and CR? How could an epistemically modest and fallibilist ‘underlabouring’ philosophy become transformed into an overwhelmingly ambitious and omnivorous religious/spiritual metaphysics? I’ll address the last of these questions first. The answer, I suspect, is only by virtue of some astonishingly weak arguments, hidden by a welter of impenetrable technical neologisms. Gary Potter has done us the great service of ploughing through this thicket to provide a clear and convincing demolition of the so-called spiritual turn. The importance of this critical work should not be underestimated. As Potter argues, the later metaphysical-religious writings of Bhaskar threaten to subvert the sources of intellectual liberation offered by CR in its original versions. This licensed and stimulated imaginative and creative theoretical work in the human sciences, welcomed plurality, and insisted on the provisional, fallible character of our understanding. This is sometimes a tough track to follow, and it is entirely different from the acknowledgement of higher sources of truth, obedience to doctrine and the struggle of faith against doubt that characterizes a religious orientation to the world. Of course, this is an orientation that many find attractive – even compelling – and they are entitled to its measures of comforts and pains. The only reservation I’ll choose to express here is that it should not claim, nor should its claims be accepted, to constitute a version, let alone development, of critical realism. To quote Potter (this volume): ‘Bhaskar (2000) identifies five moments in the history of the development of CR philosophy, each of these in turn identified with seminal text of his own authorship.’ Such flagrant self-promotion sits uncomfortably with Roy’s earlier view of CR as a collaborative research programme. But the problem is less with Bhaskar’s own view of his work than with the readiness of many followers and critics alike to see CR as somehow embodied in the person of Bhaskar (note how often this conflation appears even in this volume). Now, if CR is to retain its important intellectual role in sustaining independent and critical research, the earlier conception of CR as a diverse and collaborative research programme, not the intellectual trajectory of a single philosopher, no matter how brilliant, needs to be emphasized. The sheer diversity of the essays in this volume serves that purpose admirably. So, what of the other questions – the emancipatory project of CR, and the ‘special relation’ to Marxism? Let’s take ‘emancipation’ first. There are two main lines of argument that suggest an association

xiv Foreword

between CR and emancipatory politics. One works, but the other, in my opinion, doesn’t. The one that doesn’t is the idea advanced early on by Bhaskar of an ‘explanatory critique.’ At core this involves the claim that if a social structure can be shown to produce uncontroversially bad outcomes, it should be transformed. The standard example has been Marx’s critique of the wage form as generating false beliefs. Since it is clearly better to have true than false beliefs, then the wage form should be rejected. The particular power of this version of the argument derives form the evident performative contradiction involved in using language while denying the norm of truth-telling on which language depends. Of course, it can be conceded there may be situations in which the value of truth-telling is overridden by countervailing moral priorities, so the explanatory critique is hedged with the ceteris paribus clause. One trouble with this (there are quite a few) is that in situations where moral deliberation is required – for example, in trying to find an acceptable basis for collaborative action in politics – by definition, other things are not equal. We have not only to weigh conflicting descriptions of our situation and their rival moral claims against one another, but also to do so in dialogue with others who may make quite different choices. There are several other difficulties in this way of deriving an ‘ought’ from an ‘is,’ and some are well-explored in Andrew Sayer’s essay in this volume. Now, to an alternative, rather better, way of linking CR with emancipatory politics. Where CR differs from both empiricism and hermeneutics is in its licensing of explanatory theoretical work in the human sciences: there is no methodological barrier to attempts at explanation of patterns of events and experiences in terms of underlying structures, relations, and processes. If those patterns include ecological degradation, endemic poverty, and disease in conjunction with great affluence, gendered patterns of interpersonal violence, and so on, then they may be open to theoretical explanation in terms of deeper structural causes. This may be so quite independently of the moral or political status of the patterns to be explained. However, if there is seen to be something morally objectionable about economic inequality, domestic violence, or ecological degradation, the depth-realist mode of explanation will point to the likelihood that policies addressed at the level of the empirical patterns themselves, without changing their underlying structural causes will fail. Note that this is a very long way from supposing that emancipatory politics flows logically for an explanatory strategy. Nevertheless, it does suggest ways in which the outcome of substantive

Foreword xv

social scientific research might illuminate an open and democratic dialogue in politics. The case for deeper, structural change can be made, but, as Sayer has argued, such change needs to be open to scrutiny in terms of what might be lost along with the elimination of the objectionable state of affairs, and what other problems might be posed by new, alternative structures. A recent email debate in my own institution might illustrate the point here. A petition opposing U.K. government proposals on road pricing was circulated. My personal response was to ask, Why did the petition not call for alternative ways of addressing the international conflict, ecological destruction, and health damage caused by everescalating CO2 emissions? Some participants accepted the need to reduce emissions, while they and others objected to ways of doing it that increased surveillance and loss of privacy, or that placed the burden unfairly on those least able to pay, granting privileges to the affluent. So, here is an issue in which rival values and priorities are in competition, without any obvious or uncontroversial resolution. While a quick ‘explanatory critique’ would hardly help, a depth realist critical political economy can illuminate the question why environmental issues tend to be addressed in ways that enhance state power while also deploying economic incentives that privilege the affluent. Critical political economy can also be used to suggest possible structural changes that might go some way to combining social justice, civil rights, and ecological sustainability. But note that substantive, researchbased critical political economy might be able to do this, not the unmediated philosophical nostrums of CR. It is also worth mentioning that while the above sketch suggests that a tendency to advocate structural change is a likely consequence of adherence to a depth-realist explanatory strategy in social science, it still recognizes that normative orientations to such policies are not logically entailed by the analyses they are informed by. This opens up the gap between facts and values, ‘is’ and ‘ought,’ that Bhaskar, Collier, and some other critical realists have sought to foreclose. This is a central theme of the interesting essay in this volume by Andrew Sayer. If I understand his argument, it is that the human sciences generally fail to, but ought to, recognize human social actors as needy and desiring beings, capable of suffering or flourishing, and that normativity is already present in the ‘facts’ of social life that we social scientists study. This is common ground, and although he does not cite any of this literature, there is already a great deal of philosophical and sociological

xvi Foreword

work sustained by CR developing these ideas – from Kate Soper’s first book, On Human Needs (1981) onwards, including Maureen Ramsey’s Human Needs and the Market (1992), Soper’s Troubled Pleasures (1990), Ted Benton’s Natural Relations (1993), Carolyn New’s Agency, Health and Social Survival (1996), Ian Craib’s Importance of Disappointment (1994), Frank Pearce and Steve Tombs’s Toxic Capitalism (1998), Peter Dickens’s Society and Nature (2004), and many more. Oddly for a member of a sociology department, Andrew seems to derive his negative judgment of sociology’s contribution to this view of human being from its philosophical reflections, rather than through the substantive work of sociologists themselves (whether informed by CR or not). I would cite here the ‘classics’: Weber’s penetrating insights into the inner existential anxieties evoked by the Calvinist doctrine of predestination, or Durkheim’s powerful empirical demonstration of the individual’s need for normative integration and engagement in a wider sense of communal purpose in Suicide, or, subsequently, work by Sennett and Cobb on the Hidden Injuries of Class, or Arlie Hochschild’s investigation of the depth alienation involved in commercial ‘emotional labour’ in her Managed Heart. The rich content of explorations such as these far transcend the impoverished categories of either hermeneutic or positivist research methodology. Next, the question of CR’s special relation to Marxism. It seems clear that the coincidence in time of CR’s early development and the brief hegemony of critical theory and Marxism in the humanities and social sciences in some parts of Western Europe, the United States, and elsewhere has something to do with this. But is there more to it? If we take seriously the claim of CR to initiate a research programme in the human sciences, and define its role as that of ‘underlabourer’ rather than ‘masterbuilder,’ it follows that there can be no question of CR’s prescribing any specific theoretical orientation or research tradition – though it might provide resources for critical evaluation of some. The work of prominent critical realists such as Tony Woodiwiss and José López, including their essays in this volume, draws on Foucault rather than Marx, while other contributors make a case for realist readings of Foucault as well as of other post-structuralist writers, and Frank Pearce has made a convincing case for Durkheim’s status as a realist in his explanatory strategy. While conceding all this, I do think that there are good reasons for a continuing association of CR with Marxism of one sort or another. The significant contributions of Albritton, Engelskirchen and Ehrbar

Foreword xvii

to this volume exemplify the continuing fruitfulness of that connection. I will mention three broad considerations that tell in favour of the ‘special relationship’: 1 Marx’s own philosophical reflections on his approach to theoretical concept-construction and historical explanation (and Engels’s work on history and philosophy of science) were capable of sustaining depth-realist explanatory strategies, whereas other ‘classics’ were limited in their theorizing by the deforming effects of inappropriate philosophical representations. In my 1977 book I attempted to display elements of depth realist explanation in both Durkheim and Weber, notwithstanding its incompatibility with their professed philosophies of social science. However, Marx was alone among the sociological ‘founding figures’ for whom this was practiced consistently and in a philosophically self-aware way. 2 Albritton, in this volume, explains very clearly the distinction between a theoretically ‘purified’ account of capital and the provision of ‘middle-range theories’ or concrete historical analyses. The latter two levels of analysis are not mere deductions from the most abstract (and, for him, objective) level but require further analysis and theorization of the practices, institutional forms, and so on within which capital is instantiated and with which it interacts. The structural Marxist concepts of ‘relative autonomy’ and ‘overdetermination’ were developed to achieve a similar purpose, and there is a link here, too, to Bhaskar’s use of the notion of ‘open system.’ But, with these important barriers against economic reductionism in place, it remains the case that the structures and dynamic tendencies of capitalism as a mode of economic production have overwhelming causal importance in shaping modern historical processes, constraining, promoting, and ‘colouring’ all available patterns of social life (Woodiwiss, this volume). The critical account of the political economy of capitalism that Marx developed, and that has been reworked and developed since his death, remains by far the most systematic and powerful explanatory theory of capitalism available to us. For this reason, historical materialism (as a pluralistic, fallible, and open research programme developing and revising the Marxian heritage) is indispensable for any adequate understanding of our contemporary predicament and how we arrived at it. Several of the essays in this volume (Woodiwiss, López, Frauley, Datta) testify to the value of other research traditions – notably those

xviii Foreword

deriving from both the earlier ‘archaeology’ and later genealogical writings of Foucault. However, the work of Foucault, as well as other post-structuralist writers, lacks (and commonly refuses, on epistemological grounds) large-scale theoretical analysis of capitalism as an economic and social system, or associated research and theorizing on the dynamic relations between capital and its various state forms, cultural processes, social cleavages, and patterns of international ‘development’ and conflict. That this is a rather fundamental defect of these alternatives to historical materialist analysis is not a conclusion that could be derived a priori from critical realist philosophy. Rather, it is argued on the basis of substantive theoretical and empirical research sustained by CR philosophy. 3 The third reason for maintaining a close link between CR and Marxism is the urgent requirement for adequate ways of thinking about the relation between human social and economic forms and the rest of nature. There are two related aspects to this. One is the commitment of the various strands of CR to a ‘naturalistic’ methodology for the human and social sciences. As argued above, Roy Bhaskar’s early defence of this position made deep concessions to anti-naturalist hermeneutic philosophies of social science through his retention of a categorical opposition between human/social and natural beings. This opposition can be displaced by recognizing the great ontological diversity of natural beings, structures, and relationships and, as a consequence (for critical realism), the great methodological diversity of the sciences that take them as their various objects. Studying the social lives of other primates in the wild is a natural science, but many of its methodological problems are closely related to those of human social psychology; the methodological problems of meteorology have much in common with the other sciences – including sociology – which also deal with open systems. Arguing from the human side of the ontological divide, no serious sociological investigation of emotional work, of epidemiology, of the health implications of socio-economic status or unemployment, of gender divisions, or patterns of crime could get far without engaging seriously with work in the life sciences (for a fuller version of this argument, see Benton 1991, 1993). These arguments have been given added force by the rise to prominence of a political agenda that recognizes the deep and growing crisis in the relationship between globalizing capitalism and the rest of nature.

Foreword xix

Though climate change is the most widely recognised threat, the future of life on the planet is also jeopardized by other dynamic tendencies of the system. Some of these, like escalating international conflict and destabilization over access to fossil fuels, are directly causally connected to climate change. Others, such as shortages of fresh water and adequate sanitation, oceanic pollution, desertification, deforestation, and depletion of renewable resources have driving forces that are independent of climate change but still interact with it in potentially devastating ways. These changes have hitherto been seen as the domain of natural science, but there has been a growing awareness that they are the product of an inseparable intertwining of causal mechanisms, the understanding of which is fundamentally obstructed by the categorical division of the world into ‘nature’ and (human) ‘society.’ Raymond Murphy’s essay in this volume gives an admirably clear and persuasive account of the necessity of thinking across the culture/nature divide, and gives convincing illustrations in the shape of ‘natural’ disasters. Again, the tradition of historical materialism has distinctive strengths in providing conceptual resources to address these socio-ecological issues (though Murphy has elsewhere demonstrated the value of Weberian sociology in this respect – see, e.g., Murphy 2002a). The central concept of ‘mode of production’ conceptualizes social forms in terms of the structure of social relations through which human labour is deployed to work on non-human nature to produce means of consumption and the subsequent social division of the product. So, at the core of the theoretical approach is the socially organized ‘metabolism’ between society and its naturally given conditions, resources, and means. There is dispute about how far Marx himself got with exploring the implications of this for the ecological sustainability of capitalism, but the potential implicit in this and related ideas is being developed by numerous theorists and activists, some, such as Kate Soper, Peter Dickens, Frank Pearce, John O’Neill, Andrew Collier, and myself, directly influenced by CR. Ted Benton University of Essex

This page intentionally left blank

Acknowledgements

We would like to thank Sophia Papastavrou, Lynn O’Malley, and Andrew Stephens for their hard work in assisting the organizers of the Critical Realism and the Social Sciences International Symposium held at Queen’s University, September 2004. We are also grateful to the participants, particularly the contributors to this volume, as well as Linda Martín Alcoff, Ruth Groff, Nikolaos Liodakis, Eleanor MacDonald, George Rigakos, John Roberts, and members of the graduate Sociological Theory Class of 2004–5. Thanks also to Kristin Muszynski for her tireless copyediting efforts. A special thank-you to Virgil Duff at the University of Toronto Press for all his efforts, advice, and patience. It is truly appreciated. We would like to thank the following for financial support: the Canadian Federation for the Humanities and Social Sciences, Social Sciences and Humanities Research Council of Canada, Queen’s University Principal’s Visiting Scholar Fund, Blakeley Family Faculty Initiatives Fund, and the following departments, schools, and faculties of Queen’s University at Kingston: Sociology, Geography, Philosophy, Political Science, Urban and Regional Planning, and the Faculty of Law. Jon Frauley: I thank the Coffee Queen, Emily Flann, for being her precious self and for her unwavering and unconditional friendship and support; Christie Nazzer for, unknowingly, making a meaningful difference in my life, pushing me toward bigger and better things; Ronjon Paul Datta, Frank Pearce, Laureen Snider, April Gostlin, Alan Hunt, Steve Marmura, Kristin Muszynski, Lushious G (aka Georgie Blanchard), Ivan Stoiljkovic, and Noel Ward for their friendship and support over the years; and especially Leigh Currie for leaving me speechless, captivating me with her Shants Flip-Flop Shuffle, and Vulcan logic.

xxii Acknowledgements

Frank Pearce: I thank, for their friendship and for many dialogues about matters conceptual, ontological and epistemological, Anthony Woodiwiss, Jon Frauley, Ronjon Paul Datta, Danica Dupont, Roy Todd, Steve Tombs, Ivan Varga, Bill Stratton, Rob Albritton, Steve Gill, James Dickinson, Lorna Weir, Ted Benton, and Friedrich Sixel. Over many years I have received much support and access to a great deal of wisdom from Anita Harrington, Susan Egert, Ros Carne, Frankie Todd, Mike Woodiwiss, Laureen Snider, Ross Upshur, and Frances Newman. I am lucky to have a deep friendship with Tara Milbrandt, and I am particularly grateful to her for sharing with me the look of disbelief on her face after reading my arguments as to why the critical realist discussion of natural science was of key relevance to sociology. Finally, I again thank my sons, Rory and Blake, for being themselves.

Contributors

Robert Albritton, Professor Emeritus Political Science, York University Howie Chodos, Canadian Library of Parliament Bruce Curtis, Professor of Sociology, Carleton University Ronjon Paul Datta, Lecturer in Sociology, Mount Allison University Richard Day, Associate Professor of Sociology, Queen’s University Hans Ehrbar, Associate Professor of Economics, University of Utah Howard Engelskirchen, Lecturer in Philosophy, Iowa State University Jon Frauley, Assistant Professor of Sociology, York University Alan Hunt, Chancellor’s Professor of Law, Sociology and Anthropology, Carleton University José Julián López, Associate Professor of Sociology, University of Ottawa John Manwaring, Professor of Law, University of Ottawa Raymond Murphy, Professor Emeritus Sociology, University of Ottawa Frank Pearce, Professor of Sociology, Queen’s University Garry Potter, Associate Professor of Sociology, Wilfrid Laurier University Andrew Sayer, Professor of Social Theory and Political Economy, Lancaster University Sergio Sismondo, Associate Professor of Philosophy, Queen’s University Anthony Woodiwiss, Professor of Sociology and Dean of Social Sciences, City University of London

This page intentionally left blank

CRITICAL REALISM AND THE SOCIAL SCIENCES: HETERODOX ELABORATIONS

This page intentionally left blank

1 Critical Realism and the Social Sciences: Methodological and Epistemological Preliminaries JON FRAULEY AND FRANK PEARCE

Introduction Critical realism is a distinctive school of thought that has been growing in significance since 1975, when the philosopher of science Roy Bhaskar published his groundbreaking book, A Realist Theory of Science. Bhaskar’s next book, The Possibility of Naturalism (1978), focused more on the social sciences, and in 1989 he published Reclaiming Reality, a third crucial volume, where, in part through responding to critics of his work, he developed and consolidated his position. A Realist Theory of Science challenged the general representations of natural science while stressing that discipline’s capacity to produce valid knowledge. By reconceptualizing the nature of the relationship between empirical experiences, the wide range of actually occurring events, and the complex sets of generative mechanisms that produce experiences and events, Bhaskar developed what he termed ‘transcendental realism.’ In The Possibility of Naturalism he offered important insights into the social as well as the natural world while continuing his exploration of scientific enquiry, thereby developing a ‘critical naturalism.’ In Reclaiming Reality, he explicitly embraced what had become a general description of his position, namely ‘critical realism.’ Although there are many varieties of realism in the social sciences, we are concerned with this tradition that has emerged from the development of these early writings by Bhaskar, for, stemming from this philosophy of science and its subsequent incorporation into social scientific enquiry by British social scientists over the past quarter century, social scientific realism, especially this Bhaskarian variety, is now emerging as a formidable and major challenger to positivistic and phenomenological social science.

4 Jon Frauley and Frank Pearce

Critical realism has had relatively little impact on North American social theory or empirical research, and this is precisely why we think it important to promote discussion and debate about its fruitfulness as an alternative to dominant modes of social scientific reasoning and enquiry. We see critical realism as providing for a renewal of social scientific enquiry, and this anthology is intended as a contribution toward the renewal. Although there are subtle differences between some critical realists, and as José López and Gary Potter (2001a: 9) point out it is a ‘broad church,’ there are some central tenets. And though we make no pretension to exhaustiveness, this movement’s tenets are, broadly speaking, the following: 1 Reality exists independently of our knowledge of it, which is also to say that it exists independently of the mind of social actors. For example, the existence of systematic inequalities within the legalpolitical sphere does not depend on our knowing of their existence. Reality is always mediated through ‘perceptual filters,’ so it makes little sense to hold that our knowledge corresponds exactly to what exists. Rather, our categories and concepts help us to make reference to some aspect of a material referent. As our knowledge is fallible, our references will always undergo continual revision. The referent does not change, but rather our references to it and how we make those references does. 2 Objects are held to belong to a stratified reality independent of our perception and are the products of (at least partially) unobservable, constitutive processes and relations. Social objects, then, are held to not simply exist but to be emergent, arising from the intersection of a mass of tangled material and discursive relations. These relations require sorting out, especially in order to identify and separate the contingent relations from the necessary relations that are constitutive of the conditions under which objects emerge as social objects that can be known by researchers. This also means that inherent to critical realism is a concern with both social structure and social action and their articulation. As well, there is a concern to see empirically apprehendable phenomena as not exhaustive of that which actually exists, which means that partially obscured or unobservable entities may be inferred to exist from what can be seen or measured. 3 There are unobservable features of social life that can be known to some degree and must be revealed in order to plausibly explain the existence, reproduction, and transformation of empirically

Methodological and Epistemological Preliminaries 5

apprehendable social phenomena. For example, relations of power are not directly observable but can be inferred to exist from their effects in the social world. Such effects include class conflict, gender and racial inequalities, exploitation and domination. 4 Social structure pre-exists social action, as all human action is held to be situated activity. Social institutions such as the family, religion, education, work, and law pre-exist our birth and are relatively enduring or intransitive and constrain and enable social action. Our knowledge about these is transitive. Far from reifying social structure, that knowledge is held to be alterable and undergoing an incremental transformation. Social structure is not simply the sum of human interaction, although the former is reproduced and transformed through human interactions but also through non-human interactions as well as interactions between human and non-human entities. 5 Explanation is necessarily theoretical, and theoretical work is necessary for social scientific enquiry. It is active, conceptual work that is always tethered to an empirical referent. In other words, theoretical elaboration is tethered to an ongoing concern with the thing to be described and explained. 6 Critical realism is primarily concerned with ontology and so is ‘thing centred,’ meaning that it begins from questions about what exists (e.g., the conditions under which social objects such as ‘security’ emerge). It then moves to questions of epistemology, concerned with the production of knowledge about what exists (e.g., how can ‘security’ be investigated). This is to say that questions of epistemology are clearly distinguished from those of ontology. As critical realism is not a theory but a metatheory or a philosophy of (social) science it is compatible with a number of substantive theoretical positions (Layder 1990: 19; Sayer 1992: 4–5). For example, the work of Karl Marx (Brown, Fleetwork & Roberts 2002; Creaven 2000; Joseph 2002; Ehrbar, this volume; Engelskirchen, this volume; Albritton, this volume), Louis Althusser (Datta, this volume) Antonio Gramsci (Joseph 2002; Pearce & Tombs 1998), Émile Durkheim (Pearce 1989, 2001, and this volume), Jacques Derrida (Joseph & Roberts 2003; Norris 1987), Pierre Bourdieu (Potter 2000; Frauley 2006), Hans-Georg Gadamer (Chodos et al., this volume) and Michel Foucault (Pearce & Woodiwiss 2001; Woodiwiss; Frauley, this volume; Datta, this volume) have all been explored and situated in relation to critical realism. Likewise,

6 Jon Frauley and Frank Pearce

post-structuralism has also been found to be amenable (Stones 1996) as has critical race theory (Carter 2000; Frauley 2004) and feminist standpoint theory (Cain 1986, 1990; New 1998). Whether critical realists lean toward postmodernism (Stones 1996), hermeneutics and critical theory (Morrow 1994; Outhwaite 1983, 1987), are dealing with methodology (Carter & New 2004; Cruickshank 2002; Pawson 1989; Pawson & Tilley 1997; Frauley & Pearce, this volume; Sayer 1992), or are social psychologists (Manicas 1987; Layder 1990, 1993), philosophers (Norris 1987, 1997, 2000), legal theorists (Norrie 1993, 2005) geographers (Sayer 2000, this volume), or sociologists (Archer 1995; Benton 1977, Carter 2000; Carter & New 2004; Pearce & Woodiwiss, this volume), critical realism, as a metatheoretical position or set of methodological protocols, can and has been successfully and fruitfully employed for social enquiry. This collection brings together scholars from a range of disciplines from both sides of the Atlantic to systematically engage with and elaborate on critical realism, which offers an alternative to the dominant and long-standing reduction of social scientific enquiry to positivistic or interpretive-hermeneutic epistemologies. Contributors are united by a passionate belief that the conceptual systems that we deploy have serious consequences for our styles of thought, ethical choices, political orientations, forms of social analyses and modes of generating and evaluating empirical evidence. They explore these consequences in relation to realism and other currents of theoretical thought. Human attributes and capacities, societal relations and forces, and the interrelatedness of human activities and the natural world can all be investigated using the realist distinction between the empirical (the limited set of events and phenomena that human beings can experience), the actual (all the phenomena and events that have been produced), and the real (which includes not only the actual but also the stratified systems of generative structures and mechanisms that have the capacity, which may or may not be exercised, to produce these and possibly other events and phenomena). Realists recognize that ‘the objects investigated by science … exist and act independently of human activity [including that of scientists], and hence of both sense experience and thought’ (Bhaskar 1989: 12–13) and that these objects are an essential resource that needs to be shaped and modified in scientific and, above all, experimental activity. A distinction must be made between causal laws, which describe the tendencies of structures and mechanisms under specific conditions, and realized events. These latter are the outcomes of a confluence of disparate structures and mechanisms that happen to be parts of a spatially and

Methodological and Epistemological Preliminaries 7

temporally located complexus of relations operating under specific conditions that may lead to some causal powers not being exercised, some being exercised but their effects negated, and yet others active and effective but mutually determining outcomes. Critical Realism and Natural Science: A Brief Overview When Roy Bhaskar wrote A Realist Theory of Science he identified two dominant understandings of science: empiricism and idealism. The first can be traced back to Francis Bacon and then, more important, to David Hume. According to empiricism, the ultimate objects of knowledge are atomistic events, the constant conjunction of which constitutes the facts that provide the objective content of our ideas of natural necessity. Empiricism holds science to be conceived of as a kind of behavioural response to repeated factual stimuli, and the validity of scientific knowledge is assessed by scientists in relation to its contribution to the recognition and prediction of determined effects. The second position, which can be traced back to Immanuel Kant, is transcendental idealism, which suggests that the objects of scientific knowledge are artificial constructs – models, conceptions of natural order, etc. The experience of constant conjunctions is still necessary, but this position holds that such experience is always mediated through features characteristic of all individual minds and/or through concepts collectively produced by, and informing the perceptions and practices of, human groups. What we think of as the natural world becomes a construction of the human mind and/or of the scientific community (Bhaskar 1978: 24–25). Not surprisingly, this position is also often described as a form of ‘conventionalism’ (Keat & Urry 1975: ch. 3). In order to understand why Bhaskar focused on empiricism and idealism, it is useful to elaborate on the manifestations of these positions in the decades prior to his groundbreaking work. During this period and since, the major contemporary variant of empiricism was positivism, characterized by an ontology that holds that, belying the bewilderingly complex data of our experience, the universe comprises both a series of atomistic events and a determinate order; only things that can be observed can be known to be real, and for these things to be of scientific interest, claims about them must be true for all times and places. Its epistemology holds that knowledge is to be attained though the senses via experiments, that our sensory experience of a thing corresponds in some sense to what that thing is really like, and

8 Jon Frauley and Frank Pearce

knowledge is acquired in order to facilitate prediction and explanation of the world. This position is based on the idea that reality is simply a constant conjunction of events – that is, a series of related events that appear to us as regularities. Claims, therefore, are subject to verification or falsification to determine validity. Carl Hempel develops a sophisticated and influential variant of positivism that has been significant in discussions of both the natural and social sciences, and Bhaskar and other realists such Russell Keat and John Urry (1975) and Ted Benton (1977) pay close attention to his work and provide powerful realist critiques of it. In his covering law model of scientific explanation (Hempel 1965), Hempel’s goal is to provide the necessary and sufficient conditions for something to be properly regarded as a scientific explanation. That which has to be explained is the ‘explanandum statement,’ which has reference to some observable event and it is explained by the ‘explanans statement.’ The latter consists of antecedent observable conditions and the general laws that on this occasion come into play. If this is an adequate explanation, it will also allow us to predict an event that has not yet occurred, and as such it also allows us to explain the event after it has occurred. In such ’deductive-nomological’ laws there is therefore symmetry between prediction and explanation.1 Hempel suggests that in addition to deductive nomological laws there are also inductive statistical laws. In the case of these latter, some particular outcome is explained when it can be shown that included in a description of its conditions is a premise that under these conditions there is a high degree of inductive probability that this particular outcome is one of a statistically determinable and demarcated range of possible outcomes.2 Scientific theories consist of sets of highly general statements, the truth or falsity of which can be ultimately decided only by systematic observation and experiment that produce agreed observation-statements. Progress in science occurs when a theory emerges that can explain all a competing theory explained but can also explain more. Progress also occurs when the concepts, laws, and theories of one science are derived from another more basic science – for example, sociology from psychology, psychology from biology, biology from chemistry, and chemistry from physics (Hempel 1969). However, Hempel, like other positivists, agrees with David Hume’s scepticism about inductivism (1740/1962: 286–302) in that we cannot assume that just because some set of events have been followed by another set, however often that has been the case, there is any logical

Methodological and Epistemological Preliminaries 9

necessity that this will happen again. This leaves positivists in a difficult position, for there are then no grounds for asserting necessity in nature. In practice, a pragmatic resolution of this dilemma is to suggest that since there are always competing theories, and since, for a myriad of known and unknown reasons, the precise predictions of any scientific theory are only sometimes validated, a reasonable criterion for differentiating among a group of theories that try to explain some specific phenomenon is to give greater credence to the one most successful in its predictions. In this way progress in science is possible. A second more radical position is that proposed by Karl Popper. It accepts the Humean critique of inductivism (including his view that explanations are fictions, albeit useful ones) and suggests that while we can never know if a theory is correct – after all, some future prediction may be falsified – we can know that a theory is incorrect if one or more of its predictions has already been falsified. Falsificationists also typically argue that the scientific method does not first involve observations, then inductive generalizations from these to a theory, and then attempts to falsify it. Rather, it involves the hypothetico-deductive method whereby first a theory or hypothesis is formulated and then tested by making a series of potentially falsifying observations. Our conjectures tell us what to look for and what to test (Popper 1959, 1969). In this view, progress in science consists ultimately in the elimination of false theories and hence a narrowing of the field of competing potentially accurate accounts of the world. Those committed to both positions believe that science has progressed, albeit gradually, in that there has been a continuous increase in empirical knowledge, a related eradication of inaccurate representations, and a related rejection of false theorizing. And there has been a wider and more integrated plausible, but not definitive, theoretical understanding of the world. A problem for empiricists is that in order to explain phenomena they often invoke unobservable hypothetical or theoretical entities – for example, in the second book of Novum Organum (1620/1963) Francis Bacon suggests that heat is an effect of the usually unobservable expansive motion of microscopic parts of bodies, and Hempel endorses the use by science of concepts of unobservable electric, magnetic, and gravitational fields (Hempel 1965). But while positivists accept that they must on occasion make use of such entities in their explanations, they strongly differentiate these from observation terms and deny that any ontological commitment can be made to the real existence of such ‘theoretical entities.’ Instead, they assert that

10 Jon Frauley and Frank Pearce

observational language is ontologically and epistemologically privileged, although Hempel shows some disquiet over these distinctions (Hempel 1966: 81–82). Overall they retain the view that there is an absolute distinction between theories that offer explanations and that must be treated with scepticism, a scepticism magnified if they make use of theoretical entities, and observations that in principle can be established with certainty.3 The contrasting, conventionalist understanding of science, traced back to Kant’s transcendental idealism, yielded a powerful critique of empiricism in the 1960s. An influential indictment of many positivist arguments, The Structure of Scientific Revolutions, by Thomas Kuhn (1970), takes as its starting point the history of science. He makes a distinction between two moments in scientific endeavours, one he calls normal science and the other revolutionary science; during the ‘normal’ phase scientists work within a ‘paradigm.’ They assume that their theories provide an adequate general account of which entities, forces, and relations constitute the cosmos and that any difficulties they experience in applying and developing the established theories, including failures to produce predicted results, are to be explained as problems stemming from their personal inadequacies, deficiencies in their apparatus, or with how particular aspects of the theory have been elaborated. During normal science there is little concern with falsification but a great deal with puzzle solving. However, anomalous results may accumulate to the point that a general dissatisfaction with the established paradigm may develop. There may be a revolution in thinking and the establishment of a new paradigm. This change will involve the acceptance of a new general account of which entities, forces, and relations constitute the cosmos and will provide a new research agenda producing new puzzles that will now be assiduously solved. Kuhn makes a number of crucial claims about paradigm shifts like this. He believes that such paradigms are incommensurable; the scientists operating within different paradigms live in ‘different worlds’ (Kuhn 1970: 118). Further, the movement from one to the other does not necessarily mean that the new one will account for all the problems resolved by the previous paradigm (plus providing additional solutions) but, rather, it puts in place a whole different schema for evaluating what is important. The paradigms are not in accord about how they conceptualize and explain the physical world or about what constitutes an adequate explanation, or about what counts as a proof, etc. He also argues persuasively that observations are never theory neutral,

Methodological and Epistemological Preliminaries 11

that observers’ orientations, forms of instrumentation, and ways of interpreting results are always theory impregnated; hence, it is necessary to reject ‘the methodological stereotype of falsification by direct observation of nature’ (Kuhn 1970: 77). Predictions vindicated or falsified are no longer seen as determining the fate of theories, and the absolute distinction between observation statements and theoretical explanation is seen as no longer tenable. An even more radical version of scepticism or conventionalism is to be found in the work of Paul Feyerabend. In a series of articles and books he developed an ‘anarchist theory of knowledge’ challenging the notion that science was, could be, or ought to be a rational and objective enterprise. Feyerabend agreed with Kuhn that there are no theory-neutral observation languages or universal methodological rules; he agreed also that theories would not be rejected merely because of falsifications and that the history of science is characterized by change in meaning (Kuhn 1970; Feyerabend 1965: 168–172, 179–181). He concurred as well with revisionist Popperians such as Imre Lakatos (Lakatos and Musgrave 1970) that a crucial issue when evaluating rival sets of theories is to what extent they developed an empirically and theoretically fertile research programme. But he challenged the concept of ‘normal science’ – he was committed to a principle of proliferation that was opposed to any puzzle-solving normality. At the same time he was opposed to any model of scientific progress that assumed there should be a once-and-for-all replacement of one research programme paradigm by another newer one: ‘if it is unwise to reject theories the moment they are born because they might grow and improve, then it is also unwise to reject research programmes on a downward (degenerating) trend because they might recover and attain unforeseen splendour (the butterfly emerges when the caterpillar has reached its lowest stage of degeneration’ (Feyerabend 1975: 185). He believed that historically scientific progress had been possible because of ‘only one principle … anything goes’ (Feyerabend 1970: 26). For Feyerabend the implications of theses arguments were clear and positive. ‘Without universally enforced standards of truth and rationality we can no longer speak of universal error. We can only speak of what does, or does not, seem appropriate when viewed from a particular and restricted point of view, different views, temperaments, attitudes giving rise to different judgments and different methods of approach. Such an anarchistic epistemology … is not only a better means for improving knowledge, of understanding

12 Jon Frauley and Frank Pearce

history. It is also a more appropriate for a free man to use than are its rigorous and scientific alternatives’ (Feyerabend 1970: 21). Bhaskar’s response to these positions is to subtly change the terms of the debate. Early in The German Ideology, Marx and Engels show that even at the dawn of truly human history, human beings, whatever the mythological categories in which they cloaked their understanding, must have developed enough effective practical understandings to manipulate the world so effectively as to collectively reproduce themselves as material and cultural beings, day by day and generation by generation (Marx and Engels 1845/1976: 31). Somewhat similarly, Bhaskar starts with the fact that there has been successful scientific practice – think of the adequacy of the scientific knowledge that is a precondition for the aeroplanes that so many of us now routinely use to fly at thirty-five thousand feet and take off and land safely – and then reflects on the very possibility of science. In doing so he is engaging in the process of answering the critical metaquestion, ‘What are the conditions of plausibility of an account of science?’ (Bhaskar 1978: 8). He draws a distinction between two sides of scientific knowledge. First, science is a social product ‘much like any other, … which has its own craftsmen, technicians, publicists, standards and skills and which is no less subject to change than any other commodity.’ But second, scientific ‘knowledge is “of” things which are not produced by men at all: the specific gravity of mercury, the process of electrolysis, the mechanism of light propagation.’ ‘These ‘‘objects” of knowledge’ do not ‘depend upon human activity,’ for if human beings ‘ceased to exist sound would continue to travel, and heavy bodies fall to the earth in exactly the same way although ex hypothesi there would be no one to know it’ (Bhaskar 1978: 21). Scientific practice involves the modelling and representing of ‘intransitive objects of knowledge’ by producing ‘transitive objects of knowledge.’ But then, acknowledging but providing a somewhat Althusserian reframing of some of the issues raised by Kuhn and Feyerabend, Bhaskar points out that such practice always draws on ‘the antecedently established facts and theories, paradigms and models, methods and techniques of enquiry available to a specific scientific school or worker.’ The scientific theories thereby produced may work poorly or well and may be modified or discarded and replaced. Science as a practice always makes use of prior ‘transitive objects’ and is only conceivable if there are intransitive objects. To elucidate this latter point. it is useful to look at Bhaskar’s crucial insights about the intelligibility of experimental activity in the natural sciences.

Methodological and Epistemological Preliminaries 13 In an experiment the experimenter is a causal agent of a sequence of events but not of the causal law which the sequence of events enables him to identify. This suggests that there is an ontological distinction between scientific laws and patterns of events … To ascribe a law one needs a theory. For it is only if it is asked by a theory containing a model or conception of a putative causal or explanatory link that a law can be distinguished from a purely accidental concomitance … Now at the core of theory is a conception or picture of a natural mechanism or structure at work. Under certain conditions some postulated mechanisms can come to be established as real. And it is in the working of such mechanisms that the objective basis of our ascriptions of necessity lies … It is only if we make the assumption of the real independence of such mechanisms from the events they generate that we are justified in assuming that they endure and go on acting in their normal way outside the experimentally closed conditions that enables use to empirically identify them … Moreover it is only because it must be assumed, if experimental activity is to be rendered intelligible, that natural mechanisms endure and act outside the conditions that enable us to identify them that the application of known laws in open systems, i.e. in systems where no constant conjunction of events prevail, can be sustained. This has the corollary that a constant conjunction of events cannot be necessary for the assumption of the efficacy of a law. (Bhaskar 1978: 12–13)

There is, then, a need to make an ontological distinction between the intransitive causal powers of structures and mechanisms, which will only be activated and effective under specific conditions (and which, while still existing in potentia, may not be activated or effective under other conditions), the events of phenomena actually produced, and the small subset of these experienced by human beings. These objects can exist and act quite independently of human beings. At the same time, any attempts to produce knowledge of these intransitive causal powers and mechanisms necessarily depends on antecedent social products and involves human practices informed by socially situated forms of understanding and on the activation of the powers of at least some intransitive objects. These forms of knowledge may prove more or less adequate to their object and are always subject to revision. Here we have the essence of Bhaskar’s ‘transcendental realism.’ This introduction is no place to deal with Bhaskar’s treatment of these questions in detail; such is the task of the first substantive article in this volume and it is has been well explored by other realists such as Keat and Urry (1975), Benton (1977), and Sayer (1992; 2000). However, of immediate

14 Jon Frauley and Frank Pearce

interest here is that from these considerations Bhaskar argued for an ontological realism and an epistemological relativism. This latter statement might seem to give support to Kuhn and Feyerabend in their claim (cf. Kuhn 1970: 100–101; Feyerabend 1965: 168–171) that theories may be so radically different in meaning as to be literally ‘incommensurable.’ However, ‘if they were literally incommensurable, i.e. shared no element of meanings in common, it is difficult to see how scientists could have had grounds for preferring one to another. It is clear that at the moment of falsification, when one theory is replaced by another, some elements of meaning must be shared in common’ (Bhaskar 1978: 191). Bhaskar’s understanding of experimentation has important implications here. Any experiment depends on theorizing what mechanisms and structures need to be excluded, stabilized, and manipulated, and hence observations and experimental apparatuses are never theory neutral. However, as time passes, constant and elaborated testing of theories provides grounds for giving credence to particular practices and results. The sciences in general are continually consolidating and making relatively secure generalized claims, and scientific apparatuses are improved and better understood, an understanding that includes also specifying their limitations. Whatever their differences about particular issues at any time, scientists share a very large set of understandings. Moreover, these include not only scientific knowledge but also everyday ‘good sense,’ that which helps secure desired outcomes, and not mere common sense, which is often little more than prejudice. Kuhn’s claim that people in different paradigms literally ‘live in different worlds’ is an example of rhetorical excess (see also Keat & Urry 1975: 60–63). We have thus far discussed some important trends in the philosophy of science that have had a major impact on the social sciences.4 These trends have been important for the growth and development of social scientific methodologies, as well as the development of various techniques of data collection and interpretation. Humean empiricism and Kantian idealism have each found their way into the social sciences, manifesting as positivistic analyses on the one hand and conventionalist, phenomenological interventions on the other. The next section will systematically set critical realism apart from these methodologies in order to get at what is distinct and similar about a realist social science. This objective is met by means of attention to modes of reasoning.

Methodological and Epistemological Preliminaries 15

Modes of Reasoning, Methods, and Methodologies in Social Science Attention to methodology (modes of reasoning and explanation) as distinct from methods (techniques of data collection and interpretation) is important, but sadly this attention is more often than not considered ‘second’ order to the ‘real’ business of research (Benton 1977; Frauley 2004). However, as Sayer argues, the most crucial moment in any discussion of method is how we conceptualize. His view of method is one that ‘covers the clarification of mode of explanation and understanding, the nature of abstraction, as well as familiar subjects of research designs and methods of analysis’ (1992: 3). In other words, epistemology and ontology are key. A concern only with method or techniques of data collection can easily slide into scientism, which ‘uses an absurdly restrictive view of science, usually centring around the search for regularities and hypothesis testing, to derogate or disqualify practices such as ethnography, historical narrative or explorative research … ’ (Sayer 1992: 4). As was shown above, critical realism offers us an alternative strategy for investigating social phenomena, as it provides an alternative mode of reasoning and set of developed concepts through which to craft descriptions and explanations of social phenomena. As Norman Blaikie (1993) and Sayer (1992) have shown, research strategies (methodologies) are different from techniques of data collection (methods) in that they provide us with a mode of reasoning and sets of concepts that can be used to construct substantive theories and develop and evaluate methods. They offer us a way to understand the relationship between us and the things we seek to describe and explain. All research strategies (methodologies) offer a theory of what knowledge is and how it can be attained (epistemology) as well as offering a theory of the nature of social reality/objects of investigation (ontology). A crucial aspect of these strategies is that they are tools that allow us to conceptualize the production of knowledge and our objects of investigation, and they each offer us a different way of doing this, in that they use different analytical tools for the crafting of descriptions and explanations. In contemporary North American social science, the dominant research strategies stem from positivism (e.g., quantitative research) and phenomenology/conventionalism (e.g., hermeneutic, interpretive, and constructivist research) (see Benton & Craib 2001; Blaikie 1993; Delanty 1997; Keat & Urry 1975). It is pertinent, then, to briefly outline each in terms of its epistemological and ontological commitments.

16 Jon Frauley and Frank Pearce

As mentioned above, positivism is characterized by an ontology that holds that the universe is ordered by and comprises a series of atomistic and observable events that exist independently of our knowledge of them (but appear to us as regularities). It maintains that these events can be represented by universal truth claims, that only observable things can be considered real, that social reality is a set of complex causal relations between separate events, and that causes of human behaviour are external to individuals. The epistemology or theory of knowledge advanced by positivism holds that knowledge is to be attained though sensory perception via experiments, that our sensory perception of a thing corresponds to what that thing is really like, and that knowledge is gained toward prediction and explanation of the world. Prediction here is conflated with explanation. As there is a distinction made between theories that offer explanation and must be treated with scepticism, and observations that can be established with certainly, claims therefore are to be subject to verification or falsification. The ontology advanced by a conventionalist/phenomenological position holds that social reality consists of shared meanings and understandings. Here the process of interpreting our social reality rather than apprehending the physical world through sensory perception is central. Social reality is not held to be a thing interpreted in different ways but rather to be those different interpretations, and is therefore regarded as the product of our interpretations. The epistemology advanced holds that knowledge is derived from our everyday concepts and meanings, which are shared by members of our society. Researchers seek to understand the shared meanings about what is of interest and then attempt to ‘translate’ this into social scientific language using specialized categories (see Blaikie 1993). Falsification or verification is rejected as inadequate for determining validity. Realist ontology holds that objects exist independently of our knowledge of them, not unlike positivist ontology. Social reality is held to be stratified. That is, it is assumed to comprise three domains (outlined above): the real, the actual, and the empirical. For critical realists, causes of social phenomena are not located at the level of the empirical as they are for phenomenologists or positivists but in the ‘deeper’ level of the real (for Bhaskar 1975) and also the actual (for Sayer 1992). Social reality is held to be a constructed world – one comprising shared meanings – but underpinned by constitutive processes that require elucidation. The nature of the social world for realists is similar to that of conventionalists. However, unlike the latter, realism holds that there

Methodological and Epistemological Preliminaries 17

is a deeper reality underpinning what we can see or what we can know or interpret. What we see is an indicator of other things that exist that require explanation. A realist epistemology holds that social scientists need to build conceptual models in theorizing what underpins the social phenomena that are empirically apprehendable. These models are fallible. Whereas positivism attempts to find regularly occurring events or patterns in order to yield predictions, realism supposes that we can find tendencies or things that may or may not take place that may help us explain the event in question. Realism is often said to be ‘in between’ or a ‘third way’ between positivism and conventionalism. Although there has been some discussion of the compatibility of the research methods rooted in the traditions of positivism and phenomenology (conventionalism) (Bryman 1988; Layder 1993), because these are different philosophies there are irresolvable differences. Critical realism, as an alternative mode of reasoning and research strategy, shares some elements of both and has the capacity to yield knowledge that would not be arrived at if one was working from either a positivistic or phenomenological strategy of research, especially as it concerns the integration of analyses of social action and structure. Realism’s ontologically focused research questions have the potential to highlight new directions and forms of substantive research in the social sciences. It is highly unlikely that such questions would be generated or could be adequately dealt with by the methods informed only by positivism and phenomenology, as ontology is not explicitly attended to. One major feature of critical realism that differentiates it from positivism and phenomenology/conventionalism is its explicit concern with ontology, the ‘questions concerning the entities and structures that are constitutive of that region’ or field into which one is enquiring (Malpas 2001: 125). Although both positivism and phenomenology have provided the basis for North American social scientific methods for over a century, neither provides a basis for a rigorous theorization of the object of investigation. This distinction is important as our research strategies shape our styles of thought, forms of analyses, and our modes of generating and evaluating empirical evidence. Positivistic and phenomenological research designs emphasize how knowledge is to be generated without regard for what the object of investigation must be like in order for it to be known in the way proposed. As critical realism is primarily concerned with ontology. it is ‘thing centred’: it begins from questions about what exists (e.g., the conditions

18 Jon Frauley and Frank Pearce

under which social objects such as ‘security’ or ‘racism’ emerge). It then moves to questions of epistemology, concerned with the production of knowledge about what exists (e.g., how can ‘security’ or ‘racism’ be investigated). The demarcation of questions of ontology from epistemology in critical realism and the taking of the former as the starting point makes it unlike either of its competitors. This methodological difference reflects critical realism’s retroductive logic (Blaikie 1993). Positivistic research designs, such as those informed by middle range theory (MRT), operate via deductive reasoning while phenomenological designs, such as those informed by grounded theory (GT), operate via inductive reasoning. As the two currently dominant modes of reasoning are not retroductive in nature, critical realist research strategies will be fruitful for expanding the range of potential research questions, lines of enquiry, descriptions and explanations beyond what is available to researchers whose strategies of enquiry are rooted in either positivism or phenomenology. Given that a critical realist analytic can bear some similarities to more familiar social scientific research strategies, three caveats are worth entering. First, the suggestion that positivism and phenomenology have irresolvable epistemological differences that translate into an incompatibility of methods derived from these might be thought of as a claim that micro analysis is not compatible with macro analysis. Bryman (1988: 147) suggests that this understanding is common for the many researchers who link positivistic designs exclusively to macro analysis while linking phenomenological designs to micro-analysis. This understanding, however, is not a key issue and serves to divert out attention from more important matters. As Layder (1993: 110) has shown, and as realists in general assert, there is no necessary reason that investigation of social action and structure should be mutually exclusive, but this does not mean that the underlying modes of reasoning that animate positivistic and phenomenological modes of enquiry are compatible. Critical realism eschews both collectivism and individualism and emphasizes the articulation of social action and structure, holding that structure is antecedent and conditioning while action is reproductive and transformative. That is, there is an emphasis on the relation that obtains between situated and structured action (or human relationships) and a relatively enduring and conditioning set of social relations. The issue, then, is not whether macro-analysis is compatible with micro-analysis but with the compatibility of the different modes or logics of enquiry that inform the dominant research

Methodological and Epistemological Preliminaries 19

methods. As Carter (2000: 64) points out, critical realism ‘preserves the empiricist concern with a unity of method in the sciences, and its correlate emphasis on objectivity, whilst incorporating the interpretivist anxiety about the reification of structures by insisting on the crucial role of human agency in the maintenance and transformation of social structures.’ Second, it may seem from this that critical realism is simply a version of grounded theory or middle range theory (see Layder 1993). However, GT and MRT offer protocols for the collection and interpretation of data. MRT encourages research led by a clear theoretical idea formulated prior to research. With MRT, one collects data that fit with a preconceived hypothesis, as the aim is to verify or falsify that proposition. In this way it only generates new theory or concepts when propositions are falsified. Evaluating the adequacy of the proposition in relation to variables is primary, and the nature (qualities/characteristics) of these variables as social objects is not problematized. Also, MRT proceeds via deductive reasoning, unlike a realist retroductive mode of reasoning. Critical realism is not a form of grounded theory either. GT has a narrow, situated focus and aims to generate a specific substantive theory to explain the particular cases utilized as data. It proceeds via analytic induction (Layder 1993: 59–60) and is basically a ‘constant comparison’ that allows for the continual revision of a specific hypothesis. The generation of such a substantive theory remains constrained by the data collected, and so theory construction is limited to the specific data analysed. The critical realist question, ‘What must the object be like in order for it to be known in the way proposed?’ is animated by retroductive reasoning. Retroduction is concerned with discovery; inductive and deductive reasoning are concerned with confirmation and verification, or, more generally, justification. A retroductive analytic is used to create a theoretical model of what might exist and what, if true, could explain the existence of that which is observed. In other words, ‘This mode of inference in which events are explained by postulating (and identifying) mechanisms which are capable of producing them is called “retroduction”’ (Sayer 1992: 107). Knowledge of how something is produced, as opposed to correlations or predictions, is the concern of this mode of reasoning. It ‘proposes something that may not have been observed or could not be observed directly’ (Blaikie 1993: 165), and it is this ‘something’ (whether actual or potential) that, if it were true (as knowledge is always fallible), would explain the existence of the social objects that MRT and GT take for granted.

20 Jon Frauley and Frank Pearce

Retroductive reasoning is not unfamiliar to sociologists, and we can draw on classical sociological theory for illustrative purposes. Marx and Durkheim employed what could be characterized as retroductive reasoning. Marx inferred from empirically apprehendable class conflict the existence of an inherently contradictory process of capitalist production. That is, if Marx’s elaboration of a Capitalist Mode of Production were true, it could explain the emergence and reproduction of class conflict. Durkheim held that a social fact is the crystallization of constitutive and obfuscated social relations, and that these relations were what social scientists should be concerned with. Along these lines, the amount and kind of legal regulation in a society would refer to or indicate the presence of deeper constitutive relations; that is, for Durkheim, the level of social cohesion and the complexity of the division of labour can be inferred from the amount of legal regulation in a society. For Marx, from the level of exploitation and domination of one class by another it can be inferred that there is a degree of incongruence between the general lifeenhancing capacities of the means of production and their restrictive use determined by the extant social relations of production. The idea that empirically apprehendable phenomena are effects that are related in a complex way to underlying and sometimes obfuscated constitutive and sustaining processes, and that these effects can themselves generate further effects, is something that a critical-realist-informed research strategy would be explicitly concerned with, unlike positivistic and phenomenological research strategies. Third, it may be tempting to hold critical realism to simply be a version of an existing substantive realism such as criminological realism (left or right) (Lowman & MacLean 1992; Young & Matthews 1992a, b), legal realism (Hunt 1978; Leiter 1996), or feminist standpoint theory (e.g., Cain 1986, 1990; New 1998). These positions, however, offer substantive theories and attempt to be realistic in their descriptions of the empirical domains of concern. The realism that we are concerned with is very different from those that have been prominent in North American social science, especially sociology. Sayer (2001: 11, 70) captures this in his demarcation of ‘empirical realism’ from critical realism. Empirical realism ‘identifies the real with the empirical.’ That is, the domain of the real (tendencies and potentials) is held to be equivalent ‘with what we experience, as if the world just happened to correspond to the range of our senses and to be identical to what we experience’ (Sayer 1992). The ‘realistic’ theories do this and in doing so are empiricist. Critical realism, however, like positivism and phenomenology, is a

Methodological and Epistemological Preliminaries 21

metatheory or a philosophy of science, not a substantive theory. Metatheory is speculative and not determined by empirical data. As metatheory, critical realism offers transcendental arguments that must be ‘translated’ into workable social science frameworks with specific regard to the object of investigation. As critical realism is a metatheory, it is not the same or similar to the substantive realisms known to social scientists, nor is it the same as GT or MRT. Because it is a mode of reasoning that can be used to construct new substantive theories of particular objects and events, craft alternative research designs aimed at yielding alternative descriptions and explanations of social phenomena, and discover rather than justify propositions about the object of investigation, it offers a method different from the dominant research strategies currently employed in North American social science (Sayer 1992, 2000; Frauley 2004; Outhwaite 1983). A Brief Overview of the Chapters The first set of papers by Frank Pearce, Sergio Sismondo, and Garry Potter outline and assess crucial elements of the critical realist position. These papers show appreciation of its strengths, probe it to discover some of its weaknesses, and then offer some correctives. Pearce provides a detailed account of the early work of Bhaskar and particularly of the distinctions he makes, on the one hand, between the empirical, the actual, and the real and, on the other hand, between open and closed systems. These distinctions depend on the recognition that things are agents, with potential powers, the specific nature of which need to be specified, as do the conditions that help determine whether or not they will be exercised and with what consequences. Nature itself is stratified, in that there are different levels of sets of things, and the different levels at which the particular things interact are characterized by distinctive and emergent properties. Unconstrained nature is also generally unpredictable because the same ‘things’ are often operating under different and shifting conditions, hence producing a wide range of empirical outcomes. As Pearce shows, Bhaskar’s account provides another understanding: of the nature of science; of equivalences and divisions within the sciences; of overlaps between some social sciences and some natural sciences; and of features unique to specific social sciences. At the same time, Pearce identifies ambiguities and weaknesses in Bhaskar’s arguments and shows that they compromise the ability to break with humanist reductionism. In contrast, he suggests a more

22 Jon Frauley and Frank Pearce

strictly sociological line of development, which he illustrates by drawing on a nuanced reading of the Durkheimian tradition. Sergio Sismondo offers a succinct discussion of a deflationist conception of truth, drawing on the sociology of science, arguing, against the dominant thought, that this is assumed to be wholly compatible with realism. Bhaskar’s deflationist understanding of truth, asserts Sismondo, serves to confine realism to an ideological position on truth that negates the logical outcome of such a position, namely, pluralism. Sismondo makes his points by means of an investigation of three broad positions on truth, namely, realism, instrumentalism, and constructivism. Garry Potter stresses the strengths of critical realism but then offers both a philosophical and a socio-political critique of much of the later work of Bhaskar, arguing that it potentially compromises not only critical realism but the (social) sciences as well. Potter, commenting on five ‘critical moments’ in the development of Bhaskarian realism, highlights for us Bhaskar’s move away from realism and illustrates the ‘sociological fracture lines of thinking within Critical Realism.’ According to Potter, Bhaskar’s later work accomplishes not an extension of realist thought but ‘a wholesale re-definition of what realism is.’ Against this latter turn in Bhaskar’s thought, Potter argues that ‘critical realism’s initial role was an under-labourer for science and social science. This is, I still believe, its most important role. I also believe that not only critical realists but all intellectuals should engage politically with the world. However, propounding unconditional love, karma, God, universal self-realisation and reincarnation does not in fact facilitate such roles.’ Potter’s conclusion is that Bhaskar has become ‘an idealist propagating errors about the nature of realism.’ Having illustrated the socio-political consequences of the later Bhaskar, Potter turns to an examination of the ‘intellectual moment’ when Bhaskar moved away from realism. The second set of papers by Anthony Woodiwiss, Richard Day, Raymond Murphy, and José López strive to show the ways in which realism(s) can be deployed to strengthen or supplement sociological enquiry. Woodiwiss is concerned to rescue sociology from constructionist idealism, which downplays the concept of social structure, with its importance being ‘eclipsed by identity within social theory over the course of the last two decades of the twentieth century.’ As a way of moving away from this position, he argues for making use of a Bourdieu-inspired ‘realist reflexivity,’ thereby combatting the flourishing of the ‘humanistic idealisms of many weird and not so wonderful

Methodological and Epistemological Preliminaries 23

kinds.’ Social constructionism and postmodernism are included here. Advocating and using this realist reflexivity, drawing heavily on Michel Foucault’s Archaeology of Knowledge, Woodiwiss seeks to demonstrate that this shift away from social structure to ‘identity’ weakens sociology’s explanatory import because it premises the sociological enterprise on a humanist ontology, which negates the complexities and autonomous aspects of the social. It is this ‘unthought’ in much of the sociological enterprise that makes reflexivity necessary, and the reproduction of the weaknesses of a humanist epistemology that require this reflexivity to be of the realist variety. The supplanting of social structure with identity is argued to be a consequence of four major events: First, as itself an object, sociology became the product of an Atlanticist institutional setting that was continuously transforming itself as the United States changed from being one of two super powers to being the sole super power and also, in ideological terms, changed from being a modern to being a post-modern society. Second, as a style of talking and writing, sociology became ever more culturalist, psychologistic and reflexive. Third, the more specific concepts that both depended on and gave content to the social structure problematic were picked off one by one and replaced by those, such as culture and discourse, that created the possibility of, or gave content to, the identity problematic. And fourth, strategically, all these developments were encapsulated in an activist feminism and an academically militant postmodernism, both of which sought the total re-theorisation of social life in a way that completely marginalized the study of capitalism and its pathologies. (pp. 114–15)

Richard Day’s contribution explicitly takes up the position that poststructuralism has been wrongly and unfortunately been conflated with a poorly understood and stereotyped postmodernism, resulting in inappropriate dismissals. ‘Too many writers, from too many traditions, have for too long been dismissing so-called “postmodernism” without saying precisely what this term means to them, and without providing theoretical argumentation based on close readings of texts to support their claims. That is, “postmodernism” is all too often built up as a straw school before being casually burnt to the ground, with its occupants silently engulfed in the flames.’ Having said this, Day is clear that he is not defending eclectic and ad hoc invocations of Derrida or Jean-François Lyotard or others. He has in mind a defence of theorists

24 Jon Frauley and Frank Pearce

such as Foucault, Gilles Deleuze and Félix Guattari, and Jacques Lacan, achieved by demonstrating how their work cannot be adequately understood as examples of what he calls a ‘straw postmodernism.’ He thus hopes to make clear that some ‘so-called postmodernists are not discursive idealists at all, but share realism’s commitment to relative intransitivity and a depth ontology.’ He attempts ‘to show that postructuralist ontologies are like those of Bhaskarian-derived realisms, in that they are not classifiable as either positivistic or relativistic’ and that at the same time these are ‘unlike their realist counterparts.’ Drawing on Foucault, Deleuze and Guattari, and Lacan, Day advances an argument that ‘post-structuralism’ is dismissed too readily by its detractors but shows how this work can be used to rethink the way in which Bhaskarian realism has approached the issue of stratification as hierarchical, the agency-structure problematic, and the transitive and intransitive aspects of social and natural reality. Raymond Murphy engages with the constructivist position on truth in his paper, arguing that critical realism ‘offers a more complete analysis by situating local constructions in their context or nature’s constructions.’ This he demonstrates through a systematic critique of constructivism and through theorizing ‘across the culture/nature divide,’ illustrating the necessity of enquiring into the articulation of humans and non-human entities in the context of natural and technological disasters. These considerations also serve to strengthen his argument that in order to flourish, critical realism must ‘also develop a strong empirical dimension.’ By looking at the subject of disasters and the sociology of science, Murphy shows that a realist social science is better equipped to illustrate that nature is not ‘impotent’ but rather is dynamic. This is necessary in that social science must be able to effectively grapple with dynamism if it is to yield adequate explanations of what he calls ‘culture-nature hybrids’ and their impact on both the natural and social worlds. José López concentrates on what he finds to be a tension between the depth explanation preferred by critical realism and the emancipatory effects to follow from this explanation. Lopez encapsulates the issue thusly: ‘While I fully endorse the depth explanation model, I have a certain difficulty in accepting its alleged correlated emancipatory function. This is not because I believe that the knowledges produced by the social sciences cannot be deployed in attempts to produce social change, but because the emancipatory model, developed thus far by critical realism, fails to capture the processes and tensions involved in

Methodological and Epistemological Preliminaries 25

using social scientific knowledge to produce social change.’ To demonstrate his argument that ‘critical realism is, at the moment, poorly equipped’ to aid us in understanding the difficulties in deploying knowledge to foster social transformation, López engages with sociological and ethnographic analyses of American bioethics. These analyses illustrate what he views as the compelling depth-explanation aspect of critical realism as well as how the emancipatory aspect fails as it does not currently adequately address the pressures and constraints surrounding knowledge production and the deployment of knowledge toward social transformation. The fourth set of papers by Robert Albritton, Howard Engelskirchen, Hans Ehrbar, and Andrew Sayer each deal with political economy and its relation to realism. Albritton’s paper concerns the importance of elaborating ‘a more objective Marxian political economy.’ His elaborated conception of objectivity and ontology of capital is needed because, as Albritton stipulates, capital has played a central role in shaping modern history, and because of Marxian political economy’s role and aim in understanding this, the latter needs to be central to modern social science. As claims to objectivity have also been prominent within Marxist theory, some elaboration of an adequate conception of objectivity is necessary in order to provide an adequate ontology of capital, especially the latter’s ‘inner logic.’ The conception elaborated by Albritton ‘is sharply at odds with disembodied positivist conceptions, and while in some ways close to Bhaskar’s critical realism, it attempts to rethink ‘objectivity’ in relation to capital’s unique ontology.’ This uniqueness has to do with its ‘self-reifying properties.’ Thus, objectivity here has less to do with the epistemological notion of valueneutrality than with the nature and structure, or ontology, of capital as an object-like social entity and with the degree of the relatively enduring aspects of social structure. A theory of capital’s ‘inner logic’ is of concern to Albritton and it is the moving toward this logic that leads to his engagement and disagreement with Bhaskarian realism. Howard Engelskirchen’s contribution looks to social labour in a Marxian sense, taking note of how realist currents within Marx have been bequeathed by Aristotle. Discussing Marx’s critique of David Ricardo and his debt to Aristotle, Engelskirchen considers the commodity form of labour, getting at the ‘generative structure that organizes the development process of [the latter’s] actualization.’ Engelskirchen argues that Marx did break from classical political economy by going beyond Ricardo’s understanding of the value form that ‘constituted the product

26 Jon Frauley and Frank Pearce

of labour’ by posing the question of ‘why labour takes the form of value in the first place.’ It is this question, largely ignored by modern value form theory that is taken up by Engelskirchen, arguing that ‘Marx developed, deepened and clarified the theory of value he inherited from Ricardo, but he did not abandon the scientific object to which Ricardo referred.’ Thus there is only a ‘continuity of scientific reference’ left to tether Marx to Ricardo. The chapter uses Aristotle’s distinction between attributive and constitutive forms of entities – his concern was with ‘the potent form that ultimately organizes the power and processes of development of an entity’ (a formulation very compatible with Bhaskar’s understanding of a causal agent) – to illuminate Marx’s view that labour always exists as a particular socially and historically constituted form of labour when it transforms raw materials in any particular mode of production. Richard Boyd’s work is invoked to argue that while the world has an order, including causal powers irreducible to our projections on it, what is disclosed to us through our interaction with it is a mediated effect of both its structure and the way in which our causal powers are organized and deployed. As Engelskirchen shows, Boyd’s arguments are compatible with, clarify, and develop those in Bhaskar’s Realist Theory of Science and The Possibility of Naturalism. Thus the article builds on Bhaskar but innovatively uses Aristotle and Boyd to clarify some of Marx’s work and then shows its compatibility with an elaborated critical realism. Hans Ehrbar also addresses value in Marx, seeking to demonstrate through a close reading that Marx follows critical realist principles. At the same time, this reading illustrates how critical realist concepts can illuminate more systematically some of the obscurities present in Capital. Through his engagement with Marx and critical realism, he also is able to demonstrate one possible avenue for realist-informed social research. While some might object that Ehrbar is too ‘orthodox,’ in the sense that he accepts so much of both Marx and Bhaskar, in fact, Ehrbar provides an unusual in-depth and systematic consideration of the relation between Marx and Bhaskar. His reading of Capital, with a strong empirical orientation, shows how ‘critical realism can enrich Marxism and vice versa.’ Surprisingly, such close and careful readings of Marx are not so typical of critical realism, even though theorists like Andrew Collier or Sayer are appreciative readers of Marx. Given such attention to Marx and Capital (especially some of the more obscure arguments) is not typical of realists, in reality, Ehrbar’s approach is ‘heterodox’ rather than orthodox.

Methodological and Epistemological Preliminaries 27

Sayer attends to the implications of social scientific practice, arguing that the distancing of social scientists from the norms and practices of everyday life in the main leads to ‘producing alienated accounts of social life in which it is difficult to see why things … have significance for us and affect our well-being.’ That is, that some things matter more than others is something that has not been attended to. Although critical realism has been able to chart new inroads in the philosophy of social science, it has not been able to illuminate a distinction between ‘beings (characterized by needs, etc.) from objects that do not have needs and are capable of neither flourishing nor suffering.’ Thus, Sayer argues for a ‘needs-based conception of social being’ by advocating a ‘qualified form of naturalism’ and insisting that as social scientists we attempt to get at empirical questions about ‘what kinds of beings we are, and about how we mature and develop our capacities and susceptibilities.’ The final set of papers by Jon Frauley, Ronjon Paul Datta, and by Howie Chodos, with Bruce Curtis, Alan Hunt, and John Manwaring, all offer an engagement between realism and the work of other important social theorists such as Foucault and Gadamer. Frauley’s contribution attempts to illustrate Foucault’s realist and materialist leanings and situates this aspect of his thought within a prominent field of the social sciences, namely, governmentality studies. He takes issue with how Foucault has been taken up and applied in this field, arguing that prominent interpretations of Foucault’s work have affected his expulsion from a field in which his work is held to be foundational. By using realist metatheory and Foucault’s oft-ignored archaeological texts, Frauley seeks to demonstrate that the social analyses of governance offered by governmentalists would be strengthened and enriched if a more developed ontology, such as that provided for by realism, were to be utilized to bring out sympathetic elements in Foucault’s work. In turn, this would provide for a retrieval of Foucault and engagement with social structure. Ronjon Paul Datta, anchoring his work in Althusser’s aleatory materialism, provides a trenchant metacritique of Foucault’s Kantianism and its consequent nominalism, demonstrating how it leads to a displacement of politics by ethics. Foucault must be stood on his materialist feet since, as it is, he is standing on his nominalist head. At its core, Foucault’s position is that objects of knowledge do not exist independently of their constitution in a discursive formation: things do not have an independent knowable existence but are only objectifications that occur through discursive practices. Thus, objectifications of humans do

28 Jon Frauley and Frank Pearce

not refer to anything beyond their constitution in discursive formations; rather, they are effects of relations between knowers and known, the existence of which are determined by practices. Foucault ends up committing a modified form of the Kantian epistemic fallacy and hence cannot form a concept of politics because he is unable to coherently theorize the extra-discursive. Datta argues that the materialist-realist distinction between the real-concrete and the object of knowledge, present but attenuated in Foucault, provides the means for thinking about the relation between discursive and social formations in which the latter is shown to be the major determinant of the former. To think adequately about this relation requires theorizing political power beyond power, in terms of asymmetrical potentials and the material social conditions through which the capacities of subjects are constituted, creating a greater potential for particular actualizations of this potential relative to others. Datta provides solid theoretical grounds for displacing genealogy with aleatory materialism, thus reclaiming radical social science and politics against histories of the present. He also provides a helpful distillation of aleatory materialism. Chodos et al. cover broad terrain in their contribution, noting convergences and departures between Gadamer’s approach and that of critical realism, notably in relation to ontological realism, epistemological relativism, judgmental rationalism, and ontological truth. One particularly intriguing argument is that through emphasizing Gadamer’s ontology of the social, it is possible, contra the dominant understanding of Gadamer, to see that ‘he articulates explicitly a minimal realist stance.’ While Gadamer wishes to suggest, at times at least, that there is a very close relation between words and things, he also affirms that what exists in words does not, and cannot, exhaust what exists outside words. Gadamer’s account of the linguistic foundation of the hermeneutical phenomenon thus negotiates, in ways that are instructive, forms of nominalism in which only that which is named exists, and forms of realism in which an autonomous non-human world would cause humankind to know it in specific ways. However, even though ‘language is the ontological condition of human understanding’ for Gadamer, relativism is avoided and realism affirmed, in that ‘world views are inseparably views of the world in itself, and these views change without ever exhausting the world in itself.’ This is so, as Chodos et al. argue, because ‘Gadamer insists that language is not a barrier to our understanding of a world beyond our consciousness, but rather is the medium that embraces and makes

Methodological and Epistemological Preliminaries 29

possible all insights that we might have into such a world’ and because for Gadamer, there is an ‘ontological unity of subject and object through the medium of language,’ which means that the ways in which the world is present for us is as the world is in itself. Such a line of argument, as the authors point out, raises interesting questions about what is meant by ‘realism,’ which they explore through engaging with truth as theorized in Truth and Method and with what is taken to be Bhaskar’s ‘category mistake’ with his notion of alethic truth. Gadamer’s Minimal Realism’ provides an interesting dialogue between Gadamer and Bhaskar, and its basic argument, that there is a minimal realism in Gadamer, is sustained.

NOTES 1 For realists this formulation is underdeveloped in that it fails to differentiate between a mere correlation and an explanation where causal agents and processes are identified. 2 Realists point out that this does not explain a specific outcome and that in both cases, there is a failure to distinguish between providing grounds for expecting that an event will occur, and explaining why it occurs. 3 Realists would argue that this constitutes a metaphysics of experience since there is no logical reason to believe that what we can experience in any way exhausts the real. 4 There is no space here to list all of those philosophers who were a positive resource for Roy Bhaskar, but we would be remiss if we did not mention the works of Rom Harré, particularly (Harré 1970 and 1972), Harré and E.H. Madden (1973), and Harré and Secord (1972). In his later work Bhaskar continued to draw on Harré, but subsequently they have developed positions that in some ways differ significantly from each other’s (cf. Harré and Bhaskar 2001). 5 See Danermark et al. (1997) for an extended discussion of retroductive reasoning. For a general discussion of the philosophy of social science, see Benton and Craib (2001).

2 Bhaskar’s Critical Realism: An Appreciative Introduction and a Friendly Critique FRANK PEARCE

Introduction The purpose of this article is to provide a detailed introduction to the early writings of Roy Bhaskar, which were of tremendous significance in sociology, and subsequently in other fields of social science and in philosophy. They have been taken up by many scholars, including contributors to this volume, and they provide a backdrop to much of the development of ideas in their papers. Bhaskar’s early writings were and are of signal importance, because by providing a coherent alternative to both the positivistic and more hermeneutic traditions in social theory they show ways out of impasses still characteristic of social analysis. At the same time there are certain weaknesses in Bhaskar’s formulations that he never adequately addressed and that arguably opened up the possibility of the less fruitful directions of his thought in the last decade or so. Some of these weaknesses are identified in the latter part of the article, and suggestions are made to remedy them, and to sustain, and develop, the critical realist project. In A Realist Theory of Science and subsequently, Bhaskar changed the terrain of the debate about the nature of science and about how the social sciences can produce valid knowledge. In so doing, he provided a number of distinctions central to the social sciences; these include distinctions between the empirical, the actual, and the real and between open and closed systems. His arguments led to a unique understanding of what all sciences have in common and in what ways they differ from each other; of equivalences and divisions within the natural sciences; of certain overlaps between the social sciences and some of the natural sciences; of features specific to the social sciences.

Bhaskar’s Critical Realism 31

In these contexts he explores the stratification of nature and the existence of emergent properties. In particular, Bhaskar has proposed ways of conceiving of the social and ways of investigating it. It is worth noting that much of his work has a somewhat polemical edge, which means that he is often overly keen to differentiate himself from other philosophers of science and to stress their inadequacies rather than to show what is of value in their work and how his work builds on theirs. Here, my primary intention is to make clear the importance of the arguments in his first two books. There is coherence to this work, which gives it a certain unity – Andrew Collier’s book Critical Realism: An Introduction to Roy Bhaskar’s Philosophy (1994) provides a good, albeit insufficiently critical, guide to this work. In this article, somewhat more emphasis is placed on certain weaknesses and, as an alternative, a more Durkheimian sociology is developed. Bhaskar’s subsequent work shifts significantly and problematically. Most of those writing in this volume are in sympathy with this early work, but their relationship to the later work varies from endorsement to hostility. Bhaskar’s View of Science Bhaskar’s starting point, like that of other philosophers of science, is that it is human ingenuity that has created something definable as science, but he understands science in a specific way. He argues that while it is undoubtedly true that human beings can learn about features of the world through their personal experience of events – i.e., changes in things, and/or relations between things – these constitute only a necessarily limited selection from all similar events. This is so even when devices such as telescopes are used to extend human powers. While it may be true that a systematic empiricism can mount effective challenges to certain all-encompassing cosmologies and can reveal the spurious nature of some correlations, inevitably human beings will experience only a miniscule percentage of events. However, the problem with attempts to demonstrate any necessity to an observed constant conjunction of events is that, as David Hume showed so clearly (Hume 1777/1963: 25–39), the empiricist version of inductivism has no sound logical basis. Just as troubling, we might add, is the fact some events occur that have determinate effects on human life, though their existence can never be known directly but only inferred from such effects. There are yet others whose very existence and effects may be inaccessible to human experience. Experiences fail to exhaust the

32 Frank Pearce

domain of events, which in Bhaskar’s terms, are in the domain of the actual, the totality of events and phenomena that occur and exist in the universe. But, Bhaskar argues, what has been produced as the actual also fails to exhaust the totality of phenomena because it does not account for the structure of things, including their mechanisms, which have the power to generate events. Things here refer to a class wider than solid material objects, for it also ‘includes fluids, gases, electronic structures, fields of potential, genetic codes etc.’ (Bhaskar 1978: 98). These powers may be exercised, or they may not be exercised; and even if they are exercised, the role, if any at all, that they play in generating events will depend on the specific conditions under which they are operating. Countertendencies may block or negate their potential. Scientific explanation involves the hypothesizing and confirmation of intransitive and often unobservable things, of mechanisms and structures, that, interacting under theoretically specified conditions, can produce determinate effects – events and phenomena, with significant implications for understanding causality: The real basis of causal laws is provided by the generative mechanisms of nature. Such generative mechanisms are … nothing other than the ways of acting of things. And causal laws must be analyzed as their tendencies. Tendencies may be regarded as powers or liabilities of a thing which may be exercised without being manifest in any particular outcome. The kind of conditional we are concerned with here may be characterised as normic. They are not counterfactual but transfactual statements. Normic universals, properly understood, are transfactual or normic statements with factual instances in the laboratory (and perhaps as few other effectively closed contexts) that constitute their empirical grounds; they need not, and in general will not, be reflected in an invariant pattern or regularly occurring sequence of events. (Bhaskar 1978: 14)

The real here includes intransitive things and their potentiality, i.e., agents where an agent is ‘anything which is capable of bringing about a change in something (including itself)’ (ibid.: 109 and see 215) and the actual, events and phenomena that have been produced, and the empirical, the subset of the latter that have been experienced (ibid.: 56). It is important to note, however, that human beings may experience the world in ways that mean they only partially comprehend or misinterpret the nature of the actual. For example, common sense often treats the difference between an object being stationary or moving as

Bhaskar’s Critical Realism 33

an essential distinction, but for physics the essential distinction is based rather on ‘the principle of inertia, that only change in motion, and not motion in itself (i.e., change in position) requires explanation’ (ibid.: 81). This illustrates one break between commonsensical, experience-based reasoning and classical physics. The second distinction made by Bhaskar is between closed and open systems. In closed systems, many phenomena and causal processes are excluded. Others are controlled and stabilized, and some phenomena and interactions are manipulated within a measurable range. This produces the possibility of identifying the powers of particular structured things by stimulating them under at least partially known and controlled conditions. With the exception of certain astronomical phenomena, effective closure normally requires a humanly created experimental situation. An experiment involves human intervention, hence transitive practices, creating an ‘unnatural’ space and time and the stimulation of the ‘natural’ powers of intransitive ‘things’ that are thereby rendered observable and intelligible. These things and their powers are intransitive in that they persist and exist both within the closed experimental situation and also within open systems. In an open system, structured things may remain unstimulated and hence their powers unexercised. They may be stimulated and their powers exercised without producing any effects because such effects are nullified by the operation of other structures, or their powers may be stimulated and exercised with (significant) causal effect. ‘It is characteristic of open systems that two or more mechanisms, perhaps of radically different kinds, combine to produce effects; so that because we do not know ex ante which mechanisms will actually be at work (and perhaps have no knowledge of their mode of articulation) events are not deductively predictable. Most events in open systems must thus be regarded as “conjunctures” ’ (ibid.: 119). In open systems, then, effects are produced by the articulation of diverse mechanisms at particular junctures. Some sciences are able to make extensive use of closed systems – physics and chemistry, for example – and others can make some use of closed systems but also of open systems – biology, botany, astrophysics and cognitive psychology, for example. Others are essentially restricted to open systems – for instance, evolutionary biology, geology, and sociology. Such considerations make evident what Bhaskar describes as ‘the epistemic fallacy’: ‘the view that statements about being can be

34 Frank Pearce

reduced to or analyzed in terms of statements about knowledge; i.e. that ontological questions can always be transposed into epistemological terms’ (ibid.: 36). He is making reference here to how, in much thought, whether philosophical, scientific or commonsensical, there is an assumption that ‘whether something exists is reduced to the question whether we can know it exists … for example, A.J. Ayer’s claim that, since God is not constructible out of sense-data, the question is meaningless, and the question of his existence does not even arise’ (Collier 1994: 80). These are errors of actualism, of reducing reality to ‘the empirical world.’ The epistemic fallacy is also to be found in Kant’s view that we cannot know noumena, things in themselves, but only phenomena, appearances organized through universal categories of the human mind; hence, scientific knowledge creates artificial models, etc. What we take to be the natural world is a construction of the human mind; objects of scientific knowledge are artificial constructs. Thus the categories ‘allow only of empirical employment and have no meaning whatsoever when not applied to objects of possible experience’ (Critique of Pure Reason, B. 724), but as Bhaskar argues, ‘if the Kantian categories were adequate to the objects of scientific thought then they would continue to apply in a world without sense, and have a meaning in relation to that possibility’ (Bhaskar 1978: 37). Despite his critical words about this aspect of Kant’s thought, Bhaskar develops a transcendental argument modelled, in part, on the more persuasive aspects of the modes of reasoning found in Critique of Pure Reason. There Kant asks what must be the case for human subjects to have the kinds of experiences that we all have, and his answer is that this requires an enduring self, a synthetic unity of apperception. Thus, while agreeing with both positivism and transcendental idealism that the term ‘science’ is a distinctive knowledge-producing practice, Bhaskarian realism goes beyond both by examining science’s conditions of possibility. It is possible, Bhaskar argues, because nature contains intransitive elements – things with normic powers that are there whether or not human beings know about them – and transitive elements – the theoretical and material activities engaged in by conscious human beings that modify materiality and manipulate some intransitive things by activating their power to produce observable effects. This combination of the transitive and intransitive makes possible what can only ever provisionally be counted as knowledge. A full

Bhaskar’s Critical Realism 35

exploration of this transitive element – which involves social practices that order nature while using and transforming an ensemble of socially embedded theoretical and empirical ideas – requires an answer to the question, What must the world and society be like for science to be possible? (Bhaskar 1978: 57). Bhaskar’s answer is that ‘most things are complex objects … which … possess an ensemble of tendencies, liabilities, and powers’ by reference to which ‘the phenomena of the world are explained’ (ibid.: 51). Scientific Explanation Now, there are usually two somewhat distinctive forms of explanation in scientific work. First are what Andrew Collier describes as horizontal explanations, that is, attempts to describe what kind of effects are produced when specific kinds of agents – ‘things’ with particular kinds of causal powers – interact under specified conditions. Second, in the context of trying to explain the presence of these things with their causal powers, attempts are usually made to describe how they have been produced, and these, in Collier’s terms, are vertical explanations. Bhaskar provides an excellent discussion of the processes: The observable reactions of chemistry, which are represented in textbooks by formula such as 2Na + 2HCl = 2NaCl + H2, are explained by reference to the atomic hypothesis and the theory of valency and chemical bonding. The patterns which constitute the explananda of the theory of valency are needless to say by no means superficially obvious or readily available. Both the concepts and the substances and conditions had and have to be worked for, produced in the social activity of science. The theory itself sets out to describe the causal mechanisms responsible for the overt behaviour of the substances. Once its reality has been established (which justifies assuming that chemical boding occurs and the laws of chemistry hold outside the laboratory) and the consequences of the theory have been fully explored, the next task consists in the discovery of the mechanisms responsible for chemical bonding and valency. This has been explained in terms of the electronic theory of atomic structure. Once the reality of this explanation has been established, science moves on to the discovery of the mechanisms responsible for what happens in the sub-atomic microcosm of electrons,

36 Frank Pearce protons and neutrons; and we now have various theories of sub-atomic structure. The historical development of chemistry may thus be represented by the following schema. Stratum I 2Na + 2HCl = 2NaCl + H2 explained by, Stratum II theory of atomic number and valency Mechanism 1 explained by, Stratum III theory of electrons and atomic structure Mechanism 2 explained by, Stratum IV competing theories of sub-atomic structures Mechanism 3 (Bhaskar 1978: 169)

Reality, then, is vertically stratified, and ‘things’ and their powers at any particular level are determined by interactions between things at another more basic level. Thus ‘ontological depth appears as a condition of the development of the sciences, so that knowledge grows (as well as changes) as new and deeper layers of reality are progressively identified, described and explained’ (Bhaskar 1989: 40).1 But though things and their powers are so determined, their actual behaviour at a higher level is discovered through scientific work, and, indeed, any explanation of the effects of their interactions is likely to be post-factum. ‘Let us suppose that we could explain the emergence of organic life in terms of the physical and chemical elements out of which organic things were formed and perhaps even reproduce this process in the laboratory. Now would biologists lose their object of inquiry? Would living things cease to be real? Our apprehension of them unmasked as an illusion? No, for, in as much as living things were capable of acting back upon the materials out of which they were formed, biology would not be otiose. For knowledge of biological structures and principles would still be necessary to account for any determinate state of the physical world’ (Bhaskar 1978: 133, emphasis added). But the implications go far beyond this. The interactions of these ‘things’ at this higher level of organization occur according to principles that are not given in the lower level. ‘Whatever is capable of producing a physical effect are real and a proper object of scientific study … Living creatures qua causal agents determine the conditions under which physical laws apply; they cannot therefore already be manifest in the latter. Sentience determines the conditions of applicability of physical laws, but it is also subject to them … In short,

Bhaskar’s Critical Realism 37

emergence is an irreducible feature of our world, i.e. it has an irreducibly ontological character’ (Bhaskar 1978: 133). This point needs further elaboration. A glance at some aspects of the current ‘standard model’ of particle physics is illuminating here (cf. Treifl 1994; Weinberg 1994; Isaacs, Daintith & Martin 1999). According to this model, the building blocks of matter are a relatively small number of basic particles. First, there are the fermions: quarks and leptons (including electrons); each of these consists of three ‘generations,’ and each of these in turn contain two particles. Second, there are the bosons: the photon, the gluons, the weak W and Z bosons, and the gravitons; these mediate the forces – electromagnetism and gravity, the weak nuclear force, and the strong nuclear force – between the other particles and the various combinations of these. Some three of the quarks combine to form baryons: protons and neutrons are examples, and each of these is held together through the forces created by the exchange of gluons. Initially the universe was infinitesimally small and infinitesimally dense (a mathematical singularity). When the big bang occurred at a temperature above 1015 degrees Kelvin, the universe began to cool and expand homogeneously and isotropically. It is suggested, although by no means established, that a unified force existed that separated first into the gravitational force, then the strong nuclear force, followed by the weak nuclear force and the electromagnetic force. Quarks (and anti-quarks) and leptons and (anti-leptons) began to form, and then the quarks formed the hadrons (protons, neutrons, and pions). Sometime later helium nuclei were formed by fusion, and then atoms formed when protons combined with electrons. According to the electroweak theory: At high energies there is symmetry between the electromagnetic and weak interactions: electromagnetism and the weak force appear different to us at low energies because this symmetry is broken. Imagine a pencil standing on its end. When vertical it looks the same from all directions. A random air movement or passing lorry will cause it to topple: it could fall in any direction with equal probability. But when it falls, it falls some particular way picking out some specific direction. In the same way, the difference between electromagnetism and weak nuclear forces could be just happenstance, a chance consequence of how the high-energy symmetry was broken in our world … The laws of Physics also seem to fail to distinguish between matter and anti-matter. But we know that ordinary matter is much more common

38 Frank Pearce than anti-matter. In particular, we know that the number of baryons (protons and neutrons) exceeds the number of anti-baryons … The Russian physicist Andrei Sakharov … was able to suggest a mechanism which could produce a situation in which for every thousand million antibaryons in the early Universe, there were a thousand million and one baryons. When a baryon and an anti-baryon collide, they annihilate in a puff of electromagnetic radiation [i.e., photons]. In Sakharov’s model, most of the baryons would encounter anti-baryons, and be annihilated in this way. We would eventually be left with a universe containing thousands of millions of photons for every baryon that survives. This is actually the case in our Universe. The cosmic microwave background radiation contains billions of photons for every baryon. The explanation of this is a pleasing example of the interface between particle physics and cosmology. (Coles 2001: 71–73, emphasis added)

The extant universe, the actual universe that we inhabit, is only one possible outcome of a big bang. Furthermore, it is possible that this universe has properties that are present only within the system’s boundaries, for example, homogeneity and isotropy, and a constant speed for light. This raises the question, What, if anything, is beyond our universe, and what properties might be there? There are also unresolved problems linked to the Second Law of Thermodynamics, which states that entropy in a closed system never decreases and that the degree of disorder tends to increase. This ‘law is a macroscopic statement’ arising ‘from a microscopic description of atoms and energy states provided by detailed physical theories,’ but since the ‘laws governing these microstates are all entirely reversible with respect to time … how can an arrow of time emerge?’ (Coles 2001: 119). It is an emergent phenomenon. But this also brings up the crucial question of scale and totality. It is possible that micro ‘things’ and micro ‘powers’ depend for their existence on the continuing existence of the universe and its total powers, and it in turn might not be able to exist without the current structure of matter – from the smallest to the largest things and their powers. Any particular phenomenon must always be located in a wider set of relations and, implicitly at least, in the totality of known relations. To decide which things are more basic than others may be very difficult to achieve in practice.2 In discussing scientific activity, Bhaskar makes the point that ‘puzzles or problems are the concrete working data of the scientist’ (Bhaskar 1978: 193). Such scientific research takes place primarily in a

Bhaskar’s Critical Realism 39

closed system and is pursued according to what Bhaskar describes as the ‘DREI’ model: ‘Description of law-like behaviour; retroduction, exploiting analogies with already known phenomena; elaboration and elimination of alternative explanations; issuing (ideally) in the empirically-controlled identification of the causal mechanisms at work’ (Bhaskar 1986: 68). What this might mean can perhaps be made clear by consulting a textbook on chemistry written by Linus Pauling and Peter Pauling. It includes detailed discussions of the atomic and molecular structure of matter, and, indeed, of the chemistry of fundamental particles and nuclei. However, most of the text involves horizontal explanations, discussions of what is predicted and/or discovered about the outcomes of interactions between different elements and combinations of elements, under experimental conditions. It details the explanations offered for these and for other more unexpected outcomes – for example, that valence bonds seem to be ‘bent’ in such molecules as cyclopropane (Pauling & Pauling 1975: 157). The text also discusses the subfield of organic chemistry – the chemistry of the compounds of carbon – and the special field of biochemistry that deals with the chemical contribution to the particular properties of living organisms. These particular properties include responsiveness to the environment; self-reproduction through metabolism; sexual or asexual species reproduction (Pauling & Pauling 1975: 443). Clearly all forms of organic life are composed of organic and other chemicals. However, those that are to be found within living organisms will only be a small selection of such chemicals and will have combined to produce the building blocks of the functional components, the interrelationships of which make possible the properties of the organisms. Science and Open Systems In the work of Aristotle, William Harvey, Georges Cuvier, Claude Bernard, and beyond, the conceptualization of organic bodies as systems and structures has been a major focus of scientific theory and research, and although observation has remained of crucial significance, prediction has not been the major control of scientific adequacy. Further, the properties of organisms are not derivable from the juxtaposition of the properties of their isolated component elements. They are emergent and then only when the overall system is integrated and complete. Émile Durkheim, of course, creatively developed the organic

40 Frank Pearce

analogy for sociology – albeit sometimes falling into the error of confusing an analogy for a model. More recently, this focus on systems has reached new levels of sophistication with the contemporary autopoietic theory of Humberto Maturana and Francisco Varela (1980, 1987/ 1998) and, perhaps less successfully, by that of Niklas Luhmann (1990). Again, if we look at evolutionary theory, we find that early research did not conform to the DREI model, nor could it, because evolution has taken place within only partially closed systems, and it has taken place over staggering lengths of time. Charles Darwin’s fundamental argument is that over long periods some members of some species develop features that at specific times, in a context of scarce resources, make them more adaptive than rival members of their own species and than members of rival species. Thus, these members tend to win the struggle for survival. Hence, through their own disproportionate ability to survive and then reproduce themselves, their numbers have tended to increase relative to the others. Darwin provided no explanation why variations should be expected or how they were produced, although this was provided by Mendelian genetics leading eventually to the modern synthesis of evolutionary theory. An important implication of the successful development of evolutionary theory is that it is possible to generate valid knowledge of open systems. Bhaskar argues that in such systems explanation follows the RRRE model: ‘The four stages in the explanation of an open-systemic event may … be summarized as follows: (i) causal analysis (or resolution) of the event; (ii) theoretical redescription of the component causes; (iii) retrodiction via normic statements to possible causes of the components; (iv) elimination of alternative causes … In general the normic statements of distinct sciences, speaking perhaps of radically different kinds of generative mechanism, may be involved in the explanation of the event’ (Bhaskar 1978: 125). This argument also applies to the analysis of geological phenomena. The distinctiveness of DREI and RRRE, and the relational implications of this distinction, are well articulated by Bhaskar: ‘The importance of experimental activity in natural science conceived as a specific kind of conjunctural occurrence, allows us to stress that the predicates ‘natural’, ‘social’, ‘human’, physical’, ‘chemical’, ‘aerodynamical’, ‘biological’ ‘economic’ etc. ought not to be regarded as differentiating different kinds of events, but as differentiating different kinds of mechanisms. For in the generation of an open-systemic event several of these predicates may be simultaneously applicable’ (Bhaskar 1978: 119).

Bhaskar’s Critical Realism 41

Thus, the capacity to think and act is an integral aspect of the human mind which, in turn, depends on, but is not reducible to, the functioning of the brain. ‘An entity x may be said to possess a mind at time t if and only if it is the case that it has the capacity either to acquire or to exercise the acquired ability to creatively manipulable symbols … The capacities that constitute the mind, as so conceived, are properly regarded as causal, and … mind is a sui generis real emergent power of matter, whose autonomy, though real, is nevertheless circumscribed’ (Bhaskar 1979 & 1998a: 81). What this means is explicated by Bhaskar’s development of ‘Synchronic Emergent Powers Materialism.’ This is an argument against ‘Central State Materialism’ (cf. Armstrong 1968), that is, the view that neuropsychological states can explain psychological states and/or external behaviour. This argument necessarily presupposes a synchronic explanatory reduction, for it could only be plausible if (C1): … the individuals of the two kinds cannot be said to occupy the same place at the same time and one not be part of the other, and, (C2) … the terms of the two sciences either are at least partially intertranslatable or, if they possess well-developed independent taxonomies, overlap in at least some of their reference states. (Bhaskar 1979 & 1998a: 99)

The first criterion, (C1), highlights the problem of reducing mental or social phenomena to component parts; the second, (C2), highlights the problem of the loss of sense in moving from one system to another: ‘For it cannot be said that brain processes are about anything that they are meaningful, or that they are true or false, or that they are of or for something (as in the case of beliefs and desires respectively)’ (Bhaskar 1998a: 102). Further, reasons can be treated as possible causes of people’s actions, minimally in the sense that few bodily movements would occur the way that they do unless people have a reason to move their bodies in a particular way in a particular place at a particular time (Bhaskar 1998a: 92). In another discussion, Bhaskar (1991: 47–49) points out that Richard Rorty on occasion adopts the reductionist ‘physicalist’ position that ‘we shall someday be able, “in principle,” to predict every movement of a person’s body (including those of his larynx and his writing hand), by reference to microstructures within the body’ (Rorty 1980: 354). But, Bhaskar asserts, this presupposes that the human body is a closed

42 Frank Pearce

system, but it is surely not, because how it functions is dependent in part on its interactions with external stimuli. These include verbal statements by other human beings, and, further, it may respond differently to different verbal statements. However, as Collier notes, the difficulty is more profound than this, for an essential component of any observer’s explanation of a sequence of events that includes a verbal exchange between two people is that both they and the observer can in principle understand the meaning and significance of what is said; the different things being said may give us access to reasons capable of playing causal roles in the interaction (Collier 1994: 115). In another part of his critique of Rorty (Bhaskar 1991: 47), Bhaskar similarly asserts that in the same text Rorty claims that philosophy is a form of ‘conversational hermeneutics’ (Rorty 1980: 391) and cannot be thus reduced to mere physicality. Rorty’s philosophy is hence internally inconsistent and incoherent. Hermeneutics, after all, is concerned above all with meaning, significance, and understanding. Sociology and Society These considerations raise more general questions: What is society and what is sociology? Bhaskar’s response to the questions changes somewhat as his work develops. In A Realist Theory of Science, he begins his discussion by suggesting that for science as a particular kind of social activity to be possible, society ‘must satisfy the desiderata of being a structure irreducible but present only in its effects;’ that it ‘can only be known, not shown, to exist’; and, that ‘it exists only by virtue of the intentional activity of men but it is not the result (or the cause) of their intentional activity.’ He concludes that ‘sociology and psychology constitute distinct branches of science’; that the former ‘is not concerned with masses of individuals or mass behaviour; but with persistent relationships between individuals’; and that ‘such relationships would not exist without their relata but they do not depend for their efficacy upon any particular relata, any particular named individuals.’ This conception ‘preserves the status of human agency, but does away with the myth of creation’ and ‘of a methodologically individual reduction.’ It acknowledges that there is no necessity that a particular society continue but for it to do so people ‘must reproduce (or more or less transform) the structures (languages, forms of economic and political organization, systems of belief, cultural and ethical norms) that are given to them.’ How people ‘reproduce any particular society belongs

Bhaskar’s Critical Realism 43

to a linking science of social psychology.’ Over all, ‘society may be regarded as an ensemble of powers which exist, unlike other powers, only as long as they are exercised; and are continually exercised via (i.e. in the last instance through) the intentional activity’ of human being’ (Bhaskar 1978: 195–196). There is much insight but many ambiguities in this formulation; it seems to be an uneasy amalgam of Althusserian and humanist Marxism. Both of these positions within Marxism acknowledge the importance of actors’ interpretive orientation to the world but in quite different ways. For the former, these are created by interpellating ideological discourses, imbricated in apparatuses, and, for the latter, in forms of (mis-)understanding to which human beings have committed themselves. In practice, Bhaskar tends to rely on a model of intentional human subjectivity and, indeed, in the subsequent development of these arguments his positive resolution of some of these issues moved him further into the humanist camp. Thus, in The Possibility of Naturalism, first edition, 1979; third edition 1998) Bhaskar reiterates and develops some of the argument to be found in A Realist Theory of Science. He claims that ‘societies are irreducible to people’ and that ‘social forms are a necessary condition for any intentional act’; their ‘preexistence establishes their autonomy as possible objects of scientific investigation’; ‘their causal power establishes their reality’; ‘the preexistence of social forms … entail[s] a transformational model of social activity; ‘the causal power of social form is mediated through human agency’ (Bhaskar 1998a: 25–26). He develops these six or seven propositions through a particular representation and critical interrogation of different theorists: John W.N. Watkins as a ‘methodological individualist’; Max Weber as an ‘interpretative’ individualist; Émile Durkheim as a ‘collectivist’; Peter Berger and Thomas Luckman as ‘synthesizing’ ‘symmetrical dialectical’ theorists; and, finally, Marx as a ‘realist.’ He claims that Durkheim’s major contribution to sociological explanation is the idea that society constrains individuals but suggests that while he is wrong to not see society as ‘in principle enabling’ and in not recognizing that ‘people are the only moving forces in history’ (Bhaskar 1998a: 39). We will return to this issue below. Bhaskar cites Marx’s Grundrisse much more approvingly: ‘society does not consist of individuals [or, we might add, groups], but expresses the sum of the relations within which individuals [and groups] stand’ (Bhaskar 1998a: 26, Bhaskar’s interpolations). He emphasizes that ‘people do not create society … it always preexists

44 Frank Pearce

them and is necessary for their activity.’ He then conceptualizes the constitution of the social individual as being achieved through ‘socialization,’ by which people are equipped with the ‘skills, competences and habits’ to act in a way that both allows them to achieve some at least of their individual goals but to do so in a way that through its social organization achieves the ‘reproduction and/or transformation of society’ (Bhaskar 1998a: 25). This formulation treats conduct not so much as a response to the ways in which interactions have come to be organized in specific time-bound and place-bound contexts but rather as an expression of as a once-and-for-all-constituted subject. Here, he then again denies the relevance of collective phenomena for sociology: ‘Sociology is not concerned, as such, with the large-scale mass or group behaviour (conceived as the behaviour of large numbers, masses or groups of individuals) … Mass behaviour is an interesting social-psychological phenomenon, but it is not the subject matter of sociology.’ (Bhaskar 1998a: 28–29). He goes on to argue that ‘the essential movement in scientific theory’ consists ‘in the move from the manifest phenomena of social life, as conceptualised in the experience of the social agent concerned, to the essential relations that necessitate them,’ relations of which ‘the agents involved may or may not be aware’ (Bhaskar 1998a: 26). This assertion provides a useful segue into Bhaskar’s argument in The Possibility of Naturalism against absolute distinctions between facts and values. It is often claimed that social science is or should be neutral in that, ‘first, … its propositions are logically independent of, and cannot be derived from, any value position; second … value positions are logically independent of, and cannot be derived from, any social scientific proposition’ (Bhaskar 1998a: 54). Against this, he argues that rigorous social scientific analyses of some phenomenon – based on accurate epistemologies and ontologies, deploying elaborated, rigorous, and coherent theories and relying on valid and extensive empirical evidence – can demonstrate the error of certain other alternative forms of understanding, be they commonsensical or ‘social scientific.’ These rigorous analyses ought to displace the less valid theories. Moreover, the existence and persistence of invalid theories will also raise questions about the validity of any social order in so far as it can be demonstrated that such a social order depends for its very existence on some such inaccurate forms of understanding, for then these can be seen as not merely inaccurate but also ideological. Although, as Andrew Collier shows, Bhaskar’s argument is sometimes less than clear (Collier 1994: 176–179), the latter does provide some concise and accessible statements of his position.

Bhaskar’s Critical Realism 45 If, … one is in possession of a theory which explains why false consciousness is necessary, then one can pass immediately, without the addition of any extraneous value judgements, to a negative evaluation of the object (generative structure, system of social relations or whatever) that makes that consciousness necessary (and ceteribus paribus, to a positive evaluation of action rationally directed at the removal of the sources of false consciousness). Might it not be objected, however, that the fact/value distinction only beaks down in this way because one is committed to the prior valuation that truth is good, so that one is not deriving a value judgment from entirely factual (natural) premises? But that truth is a good (ceteribus paribus) is not only a condition of moral discourse. It is a condition of any discourse at all. Commitment to truth and consistency apply to factual as much to value discourse; and so cannot be seized upon to as a concealed (value premise) to rescue the autonomy of values from factual discourse, without destroying the distinction between the two, the distinction that is the point of the objection to uphold.’ (Bhaskar 1998a: 63)

Bhaskar summarizes these arguments by suggesting that a system of beliefs, I, may be characterized as ideological if, and only if, it is from the standpoint of a theory or consistent set of theories, T, that 1. Explain most or most significant phenomena, under its own descriptions, explained by I … 2. Explain in addition a significant set of phenomena not explained by I … 3. Explain the reproduction of I … In terms of a real stratification or connection … described in T but altogether absent from or obscured by I. 4. Explain or at least situate itself within itself … I must be unable to satisfy either of the following: 5. A criterion of scientificity, specifying the minimum necessary conditions for its characterization of a production as scientific; or 6. A criterion of domain-adequacy, specifying the minimum necessary conditions for a theory to sustain the historical of social (or whatever) nature of its subject matter.’ (Bhaskar 1998a: 67–68)

He makes these points powerfully by drawing on John Mepham’s (1972) discussion of Marx’s analysis of the fetishism of commodities. Marx’s labour theory of value, a critique of both vulgar and sophisticated political economy, reveals how capitalist exploitation occurs in

46 Frank Pearce

the production process – where workers add more value than they receive in wages which are ultimately determined by the value of labour power: ‘Real relations … located … in the sphere of production, generate phenomenal forms, characteristic of the spheres of circulation and exchange, which in turn are reflected in the categories of ideological discourse, … which sustain and underpin such ordinary commercial practices as buying and selling, wage negotiations etc. … These in turn are necessary for the reproduction of the real relations’ (Bhaskar 1998a: 69–70).3 Bhaskar’s argument certainly seems persuasive, particularly in his use of Marx’s analysis. It would also work well in the case of social structures that seem only to be able to survive through scapegoating mechanisms, such as claiming that disruptive problems like plagues are caused by witches maliciously using evil powers, for it is possible to demonstrate scientifically that such powers simply do not exist and that remediable unsanitary conditions generate contagious illnesses. Nevertheless, as we will see below, the argument is somewhat underdeveloped and, even in the case of Marx’s analysis, requires some elaboration and modification. But first, it is necessary to return to the main thrust of Bhaskar’s argument, for, from all of these considerations, he develops the Transformational Model of Social Activity or TMSA (see Fig. 2.1). ‘Society,’ he explains, ‘provides necessary conditions for intentional human action and intentional action are a necessary condition for it.’ Furthermore, ‘society is only present in human action, but human action always expresses and utilizes some or other social form … Neither can be identified with, reduced to, explained in terms of, or reconstructed from the other.’ Quite clearly, then, ‘there is an ontological hiatus between society and the people, as well as a mode of connection (viz. transformation)’ (ibid.: 37). This has significant implications: (i) Social structures, unlike natural mechanisms, do not exist independently of the activities they govern; (ii) Social structures, unlike natural structures, do not exist independently of the agent’s conception of what they are doing in their activity; (iii) Social structures, unlike natural structures, may be only relatively enduring (so that the tendencies they ground may not be universal in the sense of space-time invariant). (ibid.: 38)

Bhaskar’s Critical Realism 47 Fig. 2.1. Model IV: The Transformational Model of the Society/ Person Connection (Bhaskar 1998a: 36) Society Socialization

Reproduction/transformation Individuals

Bhaskar – Critiques and Reformulations Although Bhaskar’s formulation of the nature of societies, of social agents, and of their interrelations are useful, they have significant limitations. Ted Benton, for example, has raised issues about all three of Bhaskar’s statements about social structures. Thus, Bhaskar claims that ‘social structures … do not exist independently of the activities they govern.’ Benton points out that there can be power-relations where powers – for example, the full coercive powers of the state – can be available to certain agents and, even while not exercised on some set of human subjects or activities at some particular time, very much remain in reserve (Benton 1981/1998: 304). Bhaskar here (and elsewhere [Bhaskar: 1998a: 50]) also claims that social structures are present only in and though the activities of human agents. But, Benton argues, this would mean that ‘if any person “A” is the agent of any activity, “a,” then “A” must be the possessor of the power of which “a” is the exercise,’ and that, at best, ‘we can distinguish only between powers of agents possessed in virtue of their intrinsic natures, and powers possessed of agents possessed in virtue of their relational properties’ and that social structures are not ‘autonomous possessors of causal powers’ (Benton 1981/1998: 305). This entails a collapse into individualism. Bhaskar claims that ‘social structures do not exist independently of agent’s conceptions.’ Benton counters that while this might be true of some relationships, e.g., friendship, it is not for others; for example, ‘where society surrounds and sustains a relationship with sanctions, including coercive powers, social relationships can be, and are, sustained across great diversity and through immense changes in participating actors’ conceptions of what they are doing (employer/employee relationships, imperial domination, and marriage are three clear examples of such social structures).’ Of course, it is possible that changes in an actor’s

48 Frank Pearce

conceptions may have structural consequences, but again, they may not, and this has to be explored case by case and cannot be subsumed under a general rule (ibid.: 306). Bhaskar claims that unlike many natural structures, ‘social structures … may be only relatively enduring (so that the tendencies they ground may not be universal in the sense of space-time invariant’ but, as Benton shows, ‘natural mechanisms, like social ones are not eternal, but have definite conditions of existence which may or may not be present at any point in space or time’ (ibid.: 306). Indeed, many of the ‘laws of physics’ may have only existed since the big bang and may only function within this universe. In Reclaiming Reality, Bhaskar reiterates many of the arguments in the Possibility of Naturalism (Bhaskar 1989: 79), including the three statements criticized by Benton, and again uses his diagram TMSA. Arguably, however, they have some, but nevertheless insufficiently modified, significance, for, in a postscript to the second and third editions of The Possibility of Naturalism, Bhaskar conceded that ‘a structure of power may be produced without being exercised and exercised in the absence of any observable conflict’ but then added that this was true ‘only in so long as it is sustained by … the practices which reproduce or potentially transform it’ because ‘social structures exist materially and are carried or transported from one time-space location to another only by virtue of human practice’ (Bhaskar 1998a: 174). Margaret Archer (1995) suggests that this only partly deals with Benton’s objections because many aspects of social structures – demography, spatial patterns, distribution of resources – have a high degree of inertia and are the durable effects of the socially situated actions of long-dead social actors. She feels, nevertheless, that there is evidence here of further significant shifts in Bhaskar’s position, for in it he glosses the TMSA model somewhat differently. He now claims that this model ‘generates a clear criterion of historically significant events, namely those that initiate or constitute ruptures, mutations, or generally transformations in social forms’ (Bhaskar 1989: 77). He also develops an additional TMSA diagram, Structure and Praxis (see fig. 2.2). In this one, Archer suggests, he stresses ‘the prior emergence and current influence of structural properties … as the unintended consequences of past actions and unacknowledged conditions of contemporary activities,’ and he now recognizes that ‘their influence is to limit actors’ understanding of their social world which is compounded … by limitations of their self understanding, thus rendering the necessary production process (which is now introduced) the mediated product of

Bhaskar’s Critical Realism 49 Fig. 2.2. Structure and Praxis (Bhaskar 1989: 94) outcome reproduction

1

2 condition 3

production

4

1 reproduction/ transformation

1 = unintended consequences; 2 = unacknowledged conditions; 3 = unacknowledged motivation; 4 = tacit skills

agents who are far from highly knowledgeable about why they find themselves in the relations they do and why they do whatever they then do in these situations’ (Archer 1995: 155–156). Interestingly, in a 1983 article, Bhaskar had already begun to distance himself from any reductionist overemphasis on intersubjective interaction – he remarked that he had long been ‘inclined to give structures (conceived as transfactually efficacious)’ a strong ontological grounding and to place ‘emphasis on the pre-existence of social forms’ but now added that ‘theory need not be static, but can depict, in abstract fashion, flows, cycles and movements … tendentially applied to concrete historical situations.’ Thus, temporality is a necessity in sociological analysis, because ‘social structures are to be earthed in space and situated in time and space/time is to be seen as a flow’ (Bhaskar 1983: 85). Not surprisingly, then, ‘his refined diagram now contains mediating processes, that is, it deals with relations between positioned praxes which are not reducible to interpersonal interaction between occupants/incumbents’ (Archer 1995: 157). With some justification, Margaret Archer argues that this position has an essential compatibility with the morphogenetic approach, that she has developed. To simplify greatly, this suggests that contemporary human action always takes place under conditions shaped by previous actions, including those of previous generations and this is the foundation of the ontology of the social. It is worth noting that Andrew Sayer has further clarified some of the ways in which actions of human beings require reference to this ontologically distinctive realm. ‘Even though social structures exist only when people reproduce them, they have powers irreducible to those of individuals (you can’t pay rent to yourself). Explanation of the actions of individuals often therefore requires not a micro (reductionism) regress to their inner constitution

50 Frank Pearce

(though that may be relevant too) but a “macro regress” to the social structures in which they are located’ (Sayer 1992: 119). These reflections do clarify key aspects of the phenomenon of emergence in the social world. They are undoubtedly useful but still limited in that they fail to acknowledge the dynamism of both collectivities and non-human socio-structural agents (see below). Critical Discussion Here, it is important to raise some further questions about Bhaskar’s work in order to point to some ambiguities and problems with his arguments. The ones that will be highlighted are his formulations of: the distinction between the real, the actual and the empirical; the role of critique in critical realism; the epistemic fallacy; the negative implications of his somewhat narrow definition of sociology. Let us begin by revisiting the question of his treatment of the empirical, the actual and the real and their interrelations. In one formulation, Bhaskar suggests that the empirical is what we as human beings have experienced, the actual refers to all the phenomena and events that have actually occurred, including those we have experienced, and the real refers to the empirical and the actual, and the generative mechanisms and causal structures, which have contributed to their existence (Bhaskar 1978: 13; 56). It also refers to mechanisms and structures which have the unrealized potential of contributing to the existence of the empirical – some have been active but had their effects negated by other powers, and still other structures and mechanisms have remained dormant but nevertheless exist in potential. Thus, the real is the realm of possibilities or tendencies (the actual may or may not be produced). On other occasions, however, he sees things somewhat differently. For example, also in A Realist Theory of Science, Bhaskar writes, ‘The world consists of mechanisms and not events,’ but then in the next sentence he adds, ‘Such mechanisms combine to generate the flux of phenomena that constitute the actual states and happenings in the world’ (Bhaskar 1978: 46; see also Bhaskar 1998a: 9, 51). It is hard to see what events are if not ‘states and happenings.’ Then, in a discussion of features of ‘scientific laws’ he states that ‘for these features to be possible the world must be composed of agents’ (Bhaskar 1978: 109). Does he mean by this that agents alone are essential? The confusion continues in Reclaiming Reality. He writes: ‘Once we constitute a non-empiricist ontology that pitches invariance in nature

Bhaskar’s Critical Realism 51

at the level of structure, not event (and distinguish clearly between structure, event and experience), it becomes possible to discern in the historical development of the sciences a certain characteristic pattern or dialectic of discovery. In the schema that emerges, ontological depth appears as a condition of development of the sciences, so that knowledge grows (as well as changes) as new and deeper levels of reality progressively identified and explained’ (Bhaskar 1989: 40). There seems here an implicit belief that whatever produces the component parts of the things that together constitute a particular level of phenomena is more basic that the interactions between these things and their emergent properties, which seems to imply that deeper levels are more real than more accessible levels. However, while ‘things’ with powers exist at many different levels, and while they are sometimes inaccessible to observation, they may be accessible. The behaviour of many planetary bodies comes to mind, and many physical properties of bodies on the earth and many powers of the human body are equally open to informed inspection. In fact, the sets of things and their powers at a ‘higher’ level are inevitably a selection of the possibilities actualizable from a lower level and, indeed, they must constitute a particular level of the actual. Hence, the terms the real and the actual are relational between levels; they are not absolutes. Sayer succinctly makes a similar point about the relationship between the empirical, the actual and the real. ‘Some accounts of these concepts give the impression that the domain of the empirical can only refer to a subset of the actual and not the real, so that structures and powers are treated as unobservable. Unactivated powers or potentials are obviously not observable, but the structures in virtue of which they exist may be (the idle worker’s body, for example); observability is not restricted to what changes or moves, for at least some structures can be observed’ (Sayer 2000: 27). Furthermore, any view that the newly discovered ‘deeper’ structures and mechanisms are more fundamental or real than already known or more observable ones underestimates the extent to which in their experimental practice scientists are continually producing new and unexpected phenomena. True, these may require a reconsideration of another level of phenomena, but such reconsiderations are by no means essential; after all, horizontal explanations are continuously developed by elaborating and modifying existing theories. Now let us turn to Bhaskar’s concept of critique. He argues that if one has developed a theory that allows one to identify and explain a

52 Frank Pearce

false consciousness that one can pass immediately to ‘a negative evaluation of the object … that makes that consciousness necessary (and ceteribus paribus, to a positive evaluation of action rationally directed at the removal of the sources of false consciousness)’ (Bhaskar 1998a: 63). But critical realism recognizes the need for a certain epistemological caution, in that claims made for the truth of any theory must always be tentative, for a theory rarely has an exclusive claim to adequacy and may often be substantially modified, even rejected and superseded. Hence, we may know that there exists a mystification, and we may have an explanation of how it functions as an ideology, and we may argue for the possibility of alternative social arrangements, but we may not know what new problems might be generated by these alternative arrangements, and hence we cannot automatically claim that they are preferable. It may be argued that no society can function without imperfections, for example, illusions or ideology, and hence the most pertinent question is not the accuracy of this or that belief but the more utilitarian question of how well, on balance, do different societies actually provide for the needs of their members and constitute a democratic political sphere for disputes about what these might be. Such a sphere would require formal and substantive equality between members of the society, for these are the real conditions that make possible genuine democratic engagement and development. The chosen form of calculation of costs and benefits is, of course, premised on value commitments. What can be said confidently, however, is that if there has been made available a persuasive critical analysis of a society and how ideologies function within it, any implicit endorsement of the status quo is no longer an intellectually honest option because now an explicit moral choice is inescapable, and with that goes all the responsibilitues of working at creating the conditions for informed judgments. More adequate analysis widens the areas of moral responsibility. Thus, to support capitalism, after Marx’s critique, requires much more intellectual work than before, and it also makes one an even greater party to its disastrous consequences for humanity. These considerations lead back to the issue of the ‘epistemic fallacy,’ that is, ‘the view that statements about being can be reduced to or analyzed in terms of statements about knowledge’ (Bhaskar 1978: 36). We can demonstrate through reason that there exist phenomena that are intransitive in that they have an existence and an effectivity whether or not we are aware of them. We can also show that through transitive practices we can learn enough about these phenomena to develop theories about them and to even use some of their powers to avoid some of

Bhaskar’s Critical Realism 53

their effects. But such ontological reasoning is likely to be abstract and limited, and any theoretical knowledge is always provisional and revisable. In other words, it is important to follow more rigorously Marx’s distinction in the Grundrisse between real objects and thought objects (Marx 1973: 101; Athusser & Balibar 1970: 41; Bhaskar 1989: 127). But this needs to be extended further because our theoretical informed scientific research into the real is going to lead us to revise and render more sophisticated our understanding of ontology, for, as demonstrated by Benton (1981/1998) in his critique of Bhaskar’s The Possibility of Naturalism, our understanding as well as our particular substantive theories are also always susceptible to revision. It is crucial to distinguish, then, between the object of thought and the real object and also to recognize that the object of thought includes our epistemology and theories but also our ontology – as Alex Callinicos has argued in a debate with Bhaskar (Bhaskar & Callinicos 2003) and elsewhere (Callinicos n.d.). All of these remain open to question, qualification, and possible refutation both through philosophical debate – for example, about the status of Bhaskar’s transcendental argument – and also by new insights generated by reflection on scientific practice both old and new. The real object is intransitive in relation to all of these. A Durkheimian Realist Sociology Bhaskar has argued that much of Marx’s work is compatible with realism, and while certainly agreeing with this in principal, here I want to use an exploration of some of the sociological works of Durkheim to show why it is necessary to go beyond Bhaskar’s understanding of sociology in general and of Durkheim in particular. A good place to start is by attending to Bhaskar’s claim that an epistemological limit to naturalism in the social sciences is that its objects ‘only ever manifest themselves in open systems … where invariant empirical regularities do not occur’ and this means that ‘the social sciences are denied, in principle, decisive test situations for their theories’ and that ‘the criteria for the rational development and replacement of theories in social science must be explanatory and non-predictive’ (Bhaskar 1998a: 45–46). Through a succinct discussion of experimentation, Ted Benton provides a useful response to these arguments. It seems to be presupposed that … the constant conjunction of events associated with closure is necessary for prediction. But why should this be so? What is to rule out the calculation of the resultant effects of the joint

54 Frank Pearce operation of a plurality of mechanisms? Where very complex systems of interacting mechanisms, operating under conditions and initial states which may be known only approximately, outcomes may only be predictable as falling within a certain range of possibilities. It might be argued that … the outcomes of multi-mechanism (‘open’) systems are only predictable if it is possible to first isolate each constituent mechanism to examine its operation independently and in its relation with others … Durkheim, for example, in his classic work on suicide, uses elementary statistical comparisons in an effort to demonstrate that a definite coefficient of preservation or aggravation is associated with each of several different religious ways of life. The purpose of the statistical comparison in each case is to rule out the possibility that a given outcome (in this case, suicide rate), or given contribution to such an outcome, really is the result of religious confession, rather than the operation of some other mechanism (minority status in a society, persecution etc.). (Benton 1981/1998: 307–308)

It is useful to provide an additional rationale for Durkheim’s strategies in Suicide. He points out that each particular society tends to have relatively stable suicide rates that over time vary within a somewhat narrow range and that similar kinds of societies at a similar stage of development would be expected to have similar rates. Societies based on different forms of solidarity have distinctly different possibilities of development. Over the long run, these suicide rates may change a great deal, but such changes will likely be reflections of significant transformations in the social structure – institutional forms, interinstitutional relations, kinds of, and effectivities of, normative systems and belief systems, etc. But if the social structure stabilizes, so will the new suicide rates (Durkheim 1897/1951: 46–52). Hence, a stable social system in some ways approximates aspects of a closed system. Durkheim, then, has some good grounds for his naturalist claim that ‘considered in the light of the past’ human behaviour ‘is capable of being reduced to relationships of cause and effect’ and that these, in turn, ‘can then be transformed into rules of action for the future’ (Durkheim 1895/1982: 33).4 True, judgments as to what constitutes a specific kind of social structure and the relative significance of different elements of it can only be made on the basis of an elaborated theory of social types or species (ibid.: 97), but the discovery of correlations between variations in two different distinctive aspects of the society’s organization may suggest that one or other is a condition of existence

Bhaskar’s Critical Realism 55

of the other, or that each is a condition of existence of the other, and/or that both are dependent on some third social form. What is important here is that such system properties can sensitize researchers to where they might look for, and perhaps find, causal relations. Undoubtedly Durkheim’s categorizations of societies need developing and reworking (Pearce 2001), but in principle his sociological naturalism is more persuasive than Bhaskar’s. Bhaskar both underestimates the compatibility between Durkheim and key tenets of realism and also how much realism can learn from Durkheim. For example, he acknowledges that Durkheim successfully established ‘the autonomy of social facts using the criterion of externality’ but adds that ‘one must insist, against Durkheim, that the range of social facts depends upon (though it is irreducible to) the intentional activity of human beings’ for ‘everything that happens, happens in and through their actions’ (Bhaskar 1998a: 39–40). He agrees with Durkheim that society constrains individuals but suggests Durkheim is wrong in not seeing society as ‘in principle enabling’ and in not recognizing that ‘people are the only moving forces in history’ (ibid.: 39). The issue of Bhaskar’s poor conceptualization of the relationship between the individual actor and social structure has to some extent been addressed already; indeed, below it is argued that one can go one step further and suggest that social structures can also on occasion be agents. In fact, Bhaskar’s general understanding of Durkheim’s position is inadequate. He fails to recognize that Durkheim provides some strong realist formulations and provides a sociological account of phenomena that Bhaskar erroneously characterizes as psychological and not sociological when he asserts that sociology ‘is not concerned with masses of individuals or mass behaviour’ (Bhaskar 1978: 196). There is no intention here either to endorse the totality of Durkheim’s work or to claim that he is unequivocally a critical realist (for a detailed assessment see Pearce 2001), but his work provides many resources of developing a realist sociology. For that reason, it is useful to pay close by attention to some of Durkheim’s strong statements about the reality of the societal realm. He argued that while society cannot exist without conscious human individuals it is not merely their sum. In the same way that ‘in the living cells there are only molecules of crude matter’ and yet ‘they are in association’ and ‘this association’ causes ‘new phenomena which characterize life,’ the mode of association of human individuals also creates ‘a specific reality which has its own characteristics.’ Indeed, association can be constitutive of some of the

56 Frank Pearce

properties of its elements. Thus ‘the whole does not equal the sum of its parts; it is something different, whose properties differ from those displayed by the parts from which it is formed’ (Durkheim 1895/1982: 128). Each science has its own order of reality; the interconnectedness of the elements of this realm constitutes its determining principle and its province proper. Sociology has its own social facts, which are supraindividual although always existing in and through symbolic representations; they are only partially accessible to perception. Later he reiterated and developed this position further (in a realist direction) when he wrote that it was essential in scientific analysis ‘that the mind not be dominated by what appears to the senses, but instead to teach the mind to dominate it and to join together what the senses put asunder’ for once ‘man became aware of that internal connexions exist between things science and philosophy became possible’ (Durkheim 1912/1995: 239). Sociology’s ‘object’ is society, but a society’s social structure and associated social forces that determine its potential directions of development can vary greatly. Further, within a particular society there may also be present counterforces that inhibit such developments. The movement of societies characterized by organic solidarity – for example, in a meritocratic direction – is truncated if inherited wealth is not abolished (Durkheim 1883/1984: 310–323; 1950/1996: 213). In order to see the causal efficacy of forms of structural organization, it is worth briefly examining Guy Swanson’s The Birth of the Gods. In this work, Swanson examines data on fifty independent ‘simple’ societies (by this he means relatively structurally undifferentiated societies) to determine if there are correlations between aspects of their form of social organization and the specific content of their beliefs about supernatural forces. Following Durkheim, he suggests that particular social structures determine the contents of their religious beliefs, where religion is defined as ‘behaviors directed towards influencing the purposes of [personified and supernatural beings] – of spirits’ (Swanson 1964: 27). He suggests further that ‘people experience “supernatural properties” in social life not merely because men are unwittingly controlled by social norms … but because social relationships inherently possess the characteristics we define as supernatural’ (Swanson 1964: 22), and he is most concerned with the role played by group sovereignty. This he defines as an ‘original and independent jurisdiction over some sphere of life’ (ibid.: 20). Most societies have more than one level of sovereignty – for example, the nuclear family, however

Bhaskar’s Critical Realism 57

ordered, the unit of settlement (providing that participation within it entails obligations to the collectivity – and there may be a regional sovereign organization of the group of settlements, and so on. One of his findings is that ‘monotheism is positively related to the presence of a hierarchy of three or more sovereign groups’ and that it is not related to a ‘variety of other indices of social complexity’ (ibid.: 81). Social structures generate representations and the content of belief systems. There is no intention here to simply endorse Swanson’s work (for recent assessments, see Hammett 1984: 208–210; Peregrine 1995, 1996; Dow 2006). It needs reformulating somewhat – for example, rather than conceptualizing religions as ‘oriented behaviours’ it is better to understand them as discursive formations that help to order social practices and that interpellate subjects helping thereby to generate meaningful experiences (Foucault 1972b; Pearce 1998; Pearce & Woodiwiss 2001). It is also important to explore how some of these correlations can be explained through detailed causal chains rather than to rely on correlations alone. Nevertheless, what has been made available here is the discovery of a potential relationship between a number of different supra-human social forms, where some of these at least act as agents in that they are ‘capable of bringing about a change in something’ (Bhaskar 1978: 109 and see 215). What of constraint? Already, in the Rules of Sociological Method, Durkheim pointed to the enervating effects of society, since even while in his explicit theoretical formulation society is conceived of as coercive, in practice he also represents it as an enabling, creative, constitutive, collective stimulus. As Durkheim observes, ‘In a public gathering the great movements of enthusiasm, indignation and pity that are produced have their seat in no one individual consciousness … We have undergone the emotions much more than generated them’ (Durkheim 1895/1982: 4). In The Elementary Forms of Religious Life he developed these arguments in a discussion of the transformation of people in certain collective situations. He instances the night of 4 August 1789, when members of France’s new National Assembly abolished feudalism – ‘suddenly carried away in an act of sacrifice that each of its members had refused to make the night before’ (Durkheim 1912/1995: 212). The collective experience led them to do something that they would not have otherwise done. He also notes that a person’s language when addressing a crowd often becomes ‘high-flown in a way that would be ridiculous in ordinary circumstances; his gestures take on a an

58 Frank Pearce

overbearing quality; his very thought becomes impatient of limits and slips easily into every kind of extreme’ (ibid.) A transformation has been effected: ‘It is then no longer a mere individual who speaks but a group incarnated and personified’ (ibid.). Here, collective phenomena are clearly constitutive and energizing (see Pearce 2001: xiii-xix). In Suicide, through his use of the concept of social currents, and also through his understanding that collective representations exceeded what was in the consciousnesses of individuals, Durkheim made clear statements about the dynamic nature of social life. The social fact is sometimes so far materialized as to become an element of the external world. For instance, a definite type of architecture is a social phenomenon but it is partially embodied in houses and buildings of all sorts which, once constructed, become autonomous realities, independent of individuals. It is the same with the avenues of communication and transportation, with instruments and machines used in industry or private life which express the state of technology at any moment in history, of written languages, etc. … The material forms [legislation] assume[s] are … not merely ineffective verbal communications but active realities, since they produce effects which would not occur without their existence. They are not only external to individual consciousness, but this very externality establishes their specific qualities. Because these forms are less at the disposal of individuals, individuals cannot readily adjust them to circumstances, and this very situation makes them more resistant to change. (Durkheim 1897/1951: 313–314)

The theoretical implications of this set of formulations are rendered even clearer by explicitly considering the phenomenon of charisma. Weber tended to treat charisma as a real attribute of some individual, possibly ‘recognized’ by some other individual, or as a quality merely attributed to one individual by another (Weber 1978, 1: 215, 241) but in a Durkheimian view charisma is better understood as a social relation, a complex effect of social order. In opposition to individualistic explanations of social phenomena ‘that, taking the effect for the cause, … have … highlighted as causal conditions for social phenomena certain psychical states … which in reality are the consequence of the phenomenon’ (Durkheim 1982: 131). Thus, charismatic

Bhaskar’s Critical Realism 59

prophets, for example, must be seen to draw on and be constituted by collective representations, collectively specified criteria for recognition and socially prescribed sets of activities. More typically charisma is socially institutionalized. By means of hierarchical relationships, institutional rituals, and collective representations, an individual may experience himself or herself, and may be experienced by others, as possessing certain unique qualities. As a result, the individual is not only believed to have certain qualities but feels himself or herself to possess them and actually be energized by them. He or she has different powers because of intensified collective representations that institute him or her as a representative of the collective (cf. Lacroix 1979). Similarly, many social subordinates feel deferential and are subjectified by and subject to some person’s leadership (Pearce 2001: 31). These social relations might be filled by any number of social members, for people may become or cease to be suitable candidates as they are subject to socially ordered ebb and flow in these feelings and qualities, a societal phenomenon that Durkheim captured to some extent by the term ‘social currents’ and also by the use of the term mana (Durkheim 1912/1995: 215). What is also important here is that the social is acting on subjects even when they experience their individual selves and other individual selves as the motivating source of their responses. In such contexts Durkheim’s conceptualization of the role of individual human beings is somewhat similar to that of Louis Althusser when he conceives them as constituted as subjects by social relations that thereby render them into relatively autonomous agents (Althusser & Balibar 1970; Althusser 1971; Strawbridge 1982).5 But these Durkheimian reflections must also lead to a challenge to the overly instrumental rational actor model shared by both Weber and Bhaskar. In an explicit critique of Weber, Donald McIntosh (1983) has shown that by reconceptualizing affectual action as non-reflective but still social action, we can discriminate between expressive action that is understood to be culturally appropriate and adequate to the emotion it is expressing and that which is seen as merely bizarre and hence to be judged as irrational. Such social contextualization renders a far wider range of actions amenable to sociostructural analysis, including, incidentally, such ritual occasions as funerals. In line with a need to provide a richer ontology of the social then provided by Bhaskar, it is necessary to round out this description

60 Frank Pearce

somewhat. In many contexts individuals can only successfully inhabit the superordinate roles if they have acquired certain social orientations, attributes, and skills and if these are also routinely reinforced. The same will apply to those who are in subordinate roles. Either situation will require a subjection to socially organized disciplinary practices. If we look to the ‘schooling’ of monarchs and the rituals of the courts in Absolutist France we can see how members of the aristocratic caste were schooled for both subordinate positions (to God/the King) and superordinate positions (exercised over the rest of the population) (Elias 1983). But this does not exhaust the social: since formal official culture is only one source of the constitution of subjects, familial and other kinds of unofficial culture may produce conflicting subjectivities and offer social resources in cynicism, scepticism, and opposition.6 As well, a sociological socio-psychology would sensitize us to the fact that in ‘group’ situations, in addition to willing spontaneous subjection, subordinate individuals might feel collectively impelled to ‘fight or flight’ or ‘pairing’ (for ‘small’ groups, see Bion 1961 and for large groups see Kreeger 1975, and see the discussion in Pearce 2001: 151–152). Equally, some superordinate individuals might find it hard to feel the confidence to persuasively fulfil their ‘responsibilities’ (Louis XVI might be an example [Price 2002: 4]). But the development of different kind of personalities is inherently social – it depends, in part, on symbolic orders and socially constituted relationships, including prescriptions and prohibitions, involving children designated as male and female, mothers, fathers, siblings, uncles, aunts etc. (Althusser 1971; Hirst & Woolley 1982). It is worth noting, incidentally, that psychoanalysis has been explored from a realist perspective by a number of thinkers (Will 1986, Rustin 1993, Collier 1994). While a concrete event might benefit from different levels of sociological, socio-psychological, and psychoanalytic levels of analysis, the sociological levels are distinct and include causal levels that make no reference to ‘individual human agency.’ Conclusion This article focused on the arguments developed in the two foundational books for critical realism, Roy Bhaskar’s A Realist Theory of Science and The Possibility of Naturalism, and also made use of other early writings by Bhaskar. This corpus of work has been a crucial resource for other scholars also concerned to develop a non-positivistic

Bhaskar’s Critical Realism 61

and critical realism – Russell Keat and John Urry, Ted Benton, Andrew Sayer, Andrew Collier, Kate Soper, Anthony Woodiwiss, Margaret Archer, José López, and Gary Potter, to name but a few. At the same time, in developing their independent positions all these scholars have carefully delineated what they consider to be valuable and specified a number of problems and developed the arguments in various way while generally remaining grounded in this early work. A significant amount of space has been devoted to discussing Bhaskar’s view of science and its relevance to thinking about some aspects of contemporary physics, because his exploration of what makes the success of science is extremely original and avoids the pitfalls of positivist/empiricist and hermeneutic/conventionalist positions. Crucially, science is viewed not as a paradigmatic touchstone for all other forms of knowledge but rather as a source for developing an adequate and complex ontology and an understanding of the necessity of modesty in the case of epistemology. On the other hand, the consideration of Bhaskar’s arguments and formulations has also identified and problematized certain ambiguities within his formulations – the relationship between the empirical, the actual and the real and the question of the relationship between levels of causally effective systems, for example. The second part of the article focuses on Bhaskar’s approach to social science, and though it is also appreciative it is much more critical. The criticisms rely to some extent on the work of Benton and Archer but go beyond them, in part, by also drawing on aspects of the thought of Durkheim. Durkheim’s discussion of the ontology of the social and the causal efficacy he accords to it are major themes explored here. The reading of Durkheim’s work presented draws on, but is not reducible to, that developed in the interdiscursive text. The Radical Durkheim (Pearce 1989 and 2001) and this in turn draws upon insights consequent realist reflections on Foucault’s archeological and genealogical writings (Pearce 1988; Pearce & Woodiwiss 2001) and Althusser’s writings, including his aleatory materialist formulations (recently collected in Althusser 1993/2006). Crucially, it is argued, this tradition provides a better foundation for a critically realist sociology than that developed by Bhaskar himself. Overall, this article is intended to provide for readers of this book a reasonably clear account of those aspects of Bhaskar’s work that are a major reference point for all contributors to this volume. These latter, of course, make their own creative use of Bhaskar’s insightful writings.

62 Frank Pearce NOTES I wish to thank Tara Milbrandt, Steve Tombs, Roy Todd, Michal Bodemann, Jon Frauley, Paul Datta, and three anonymous reviewers for their comments on earlier versions of this article. I alone am responsible for any inadequacies in the text. 1 In a personal communication Roy Todd has suggested that one problem with Bhaskar’s theory of levels is that there is evidence that theories about the operation of weak radiation and gravity seems to imply operation at all four levels and beyond. Cf. Penrose (2004). There is a brief discussion in this article of the relation between components of the universe and it as a totality on pages 38, but I am making no claim to adequate scientific competence. 2 David Bohm (2002) has raised important questions about, among other ‘things,’ implicate and explicate orders, which are also questions about totalities. Again, I am not competent to judge their adequacy. However, some of his arguments about the significance of quantum theory have been treated very sympathetically by the realist philosopher Christopher Norris (2000). 3 A supplementary example of how an erroneous understanding can be seen to have ‘ideological effects’ is provided by the ways in which gender inequality is often explained and justified. It is seen as an effect of the different personality characteristics that normal men and women are assumed to exhibit ‘natural’ characteristics that equip men rather than women to take up most leadership roles in society and women to have primary responsibility for the domestic sphere. These different characteristics are understood as determined by people’s biological maleness or femaleness, which are assumed to be grounded in dichotomous physical characteristics such as the kinds of genitalia possessed. Such ideas have now been effectively discredited by decades of excellent feminist scholarship (cf. Henriques et al. 1983) and, indeed, more recently the very idea that human beings can be distributed according to some absolute physiological distinction between men and women has also been challenged (cf. Kessler & McKenna 1985). For a useful explicitly realist formulation of these issues see Dannemark et al. (2002: 62–64). 4 None of this to suggest that there are not significant problems with Durkheim’s analysis of suicide. For a balanced assessment of both the accomplishments of, and errors in, Durkheim’s strategies and techniques, see Selvin (1965) and for a critical realist reformulation see Taylor (1982) –

Bhaskar’s Critical Realism 63 in fact, arguably, Durkheim’s kinds of arguments would work better with attempts to understand such phenomena as birth rates and marriage rates. Nevertheless, it is important to stress that Suicide, particularly when put in play with other texts by Durkheim – those, for example, where he explores the negative structural consequences of inherited wealth in contemporary societies – is a rich resource for retheorizing social order (Pearce 2001). 5 There are a number of other examples of work inspired by Durkheim’s socio-relational understandings of ‘religious’ phenomena including Bataille’s theorization of how sacral ‘heterogeneous’ forces are, to different degrees, in constant tension with ‘homogenizing’ forces within societies (Bataille 1985; Pearce 2005). 6 The complexity of these relations in society as a whole has been explored by Caillois, Klossowski, and Foucault (cf. Pearce 2003) and social dynamism in particular by Caillois (Caillois 1939/1959).

3 For Realism, and Anti-Realism SERGIO SISMONDO

Why Realists Have Taken the Deflationist Road Theories about the nature of truth cluster into three broad regions in intellectual space that we can call realism, instrumentalism, and constructivism. Most important of these is realism, because it accords best with pre-theoretical intuitions about truths. In rough terms, realist theories see truths as fully dependent on the way the world is, as reflecting the facts. In the philosophy of science and the philosophy of language, realism has typically meant that true theories, sentences, or propositions correspond to the world. Thus, realism typically demands a relation of correspondence between true representation and that which it represents. Instrumentalist theories see truths as dependent on the way the world is but not as reflecting it. Rather, instrumentalism sees truths as drawn from our points of contact with the world but otherwise shaped by their relations with each other. Truths connect and package observations into consistent bundles, but those bundles shouldn’t be expected to mirror underlying structures. Constructivist theories see the world as at least partly dependent on truths. This is typically because constructivism questions whether it makes sense to talk of a ‘way the world is’ independently of our conceptions of it. Truth, then, is something that humans make – though not under conditions of their own choosing. World and truths correspond, but because the world mirrors truths, rather than the other way around. There is a fourth cluster, but it is quite different in character from the others. Deflationist accounts of truth deny that we can have a substantive theory of truth (e.g., Fine 1986; Horwich 1990). For the deflationist,

For Realism and Anti-Realism 65

truth is not to be analyzed in terms of relations to the world; to the extent that we can analyze truth at all, we analyze our use of the term or concept ‘truth,’ or we analyze how people decide what the truth is. Though it has a few able defenders (Sher 2004), correspondence realism has fallen into disrepute. The replacement is deflationism, which provides a good starting point here because it is the position Roy Bhaskar adopts at the end of A Realist Theory of Science: ‘[Epistemological relativism] entails the rejection of any correspondence theory of truth. A proposition is true if and only if the state of affairs that it expresses (describes) is real. But propositions cannot be compared with states of affairs; their relationship cannot be described as one of correspondence. Philosophers have wanted a theory of truth to provide a criterion or stamp of knowledge. But no such stamp is possible. For the judgement of the truth of a proposition is necessarily intrinsic to the science concerned. There is no way in which we can look at the world and then at a sentence and ask whether they fit’ (Bhaskar 1975: 249). For Bhaskar, there is no question of a relation between truths and the world, and therefore no question of our having a substantive theory of truth. But we can do perfectly well without either of those things and still have realism. Deflationism about truth draws its support from two sources. First, all efforts to arrive at substantive theories of truth have foundered. It appears impossible to define ‘correspondence’ in ways that are simultaneously non-vacuous and plausible. This is because language is nothing like the world it represents, and thus ‘correspondence to the facts’ looks like philosophical bureaucratise, like ‘operationally valid’ or ‘according with actuality.’ It appears equally impossible to elucidate constructivist or instrumentalist theories of truth that do not run afoul of basic norms of truth-talk: such theories have to deal with problems stemming from norms of fallibility and the stability of truths (e.g., Engel 2002). For example, it is a basic norm of truth-talk that truths about stable phenomena are stable, yet instrumentalism and constructivism would have it that changing modes of discourse change truths. Second, consider the simple equivalence ‘p’ is true iff p or, more concretely, ‘SSRIs increase suicidal tendencies’ is true if and only if SSRIs increase suicidal tendencies.

66 Sergio Sismondo

Such statements suggest that ‘is true’ does not add anything to a proposition or sentence, and therefore suggests that there may be nothing substantive to say about truth. The phrase ‘is true’ is nearly empty, at least when it is applied to simple putative truths. Thus, one should not seek a substantive theory of truth but rather recognize that statements are called ‘true’ when they are assertible – in conditions where the right amounts of integrity and evidence are thought to hold. Technically, deflationism is neutral with respect to realism and antirealism. Learning or deciding that there can be no substantive theory of truth is compatible with (a) the realist idea that truths are such because of the structure of the world, (b) the instrumentalist idea that truths are such because they organize our experience, or (c) the constructivist idea that truths create structure in the world. However, there is a sense in which deflationism looks most naturally like a realist position. If ‘p’ is true iff p, this is because ‘p’ is true iff it is a fact that p. Fact-talk, though, carries a presumption of realism. That is, realism is the standard pretheoretic assumption, continuous with common sense. While facts can be given anti-realist interpretations, those are clearly interpretations in a stronger sense than are realist ones. Interpretations of this stronger sort are clearly add-ons to deflationism, extras that go significantly beyond the position. A realist understanding of facts is in this sense the closest to the deflationist position by itself. This is, of course, good news for the realist. The failure to articulate a correspondence theory of truth tends to push us toward a different realism: it pushed us down to earth to the intuitively attractive position that truths represent things as they are, though without committing us to a theory of representation. While there is much right about this trajectory, the end position is unstable. By setting aside aetherial debates about the nature of correspondence, about the possibility of access to the way things are, and about the human contribution to any possible knowledge, we introduce more blunt questions about the relations between truths and their subject matters. And these questions draw attention to multiple ways in which truths relate to the facts. My goal here is to illustrate this claim with some mundane and relatively uncontroversial observations about the natural sciences. Deflationism suggests that we look to scientific practice to see how researchers arrive at truths. But if we look to practice, we find that truth in the natural sciences can be analyzed, though not in the terms of a single theory. Interestingly, the relations between truths and facts

For Realism and Anti-Realism 67

can be seen as sometimes more realistic, sometimes more instrumentalist, and sometimes more constructivist, though at a much more straightforward and grounded level than our original philosophical positions were. For Correspondence, Mechanized Let me first try to increase the good news for the realist. At least some of the time, not only is realism the obviously correct position, but something close to correspondence realism looks correct. I will make my case for this claim via an example. One of the key technical achievements in molecular biology in the past two decades has been the ability to sequence DNA, to read the nucleotides on pieces of material DNA. On one of the methods, a single strand of DNA is amplified, then placed in a mixture of tagged and untagged nucleotides. The nucleotides act as reagents and cut the DNA when they react. By comparing the lengths and types of tagged and untagged strands, one can build the DNA sequence. The fragment is then represented as a sequence of As, Cs, Gs, and Ts, for example by CCTAGTTACGCA … Obviously this representation does not correspond to the fragment in the sense of being a picture: the fragment is not made up of little As, Cs, Gs, and Ts. Nonetheless, to the extent that it is correct (error rates vary by procedure), each letter in the representation corresponds with a nucleotide in the fragment in a well-understood way. There is a quite precise mapping between the sequence and the nucleotides of the fragment. Moreover, this procedure can be fully mechanized! Sequencing machines simply spit out As, Cs, Gs, and Ts in response to an input of DNA. Interested laboratories can buy the machines off the shelf, or they can send DNA to specialized firms that charge by the nucleotide or by the sample. Human interpretation need play no role in the creation of sequences. We might have qualms about the exact shape of DNA, or about how it appears in bodies rather than how it appears under the crystallographer’s apparatus. But at this point it would be difficult to have qualms about the idea that there are four bases and that particular sequences of bases provide a kind of ‘instruction’ for, or more neutrally a step in, the production of particular amino acids. So when a machine translates a fragment of DNA into a sequence of letters, we can be confident that there is a fairly good correspondence between the letters and the

68 Sergio Sismondo

sequence of bases. The nature of that relation can be specified by our understandings of the protocols for turning DNA into letters, by our understandings of the relevant ‘inscription devices,’ to use Bruno Latour and Steve Woolgar’s (1986) term. There is slightly more. While sequencing procedures cannot literally be reversed, it is possible to turn sequences into DNA. And again, this can be mechanized. So if on the one hand we have inscription devices, and on the other hand we have materialization devices, we have a well-understood relation between these representations and the things they represent. I do not know what more we would want from correspondence. Such claims cannot be made about scientific knowledge in general. Mechanization (even involving humans) is the exception rather than the rule (e.g., Collins 1990). And even knowledge that depends heavily on mechanized data collection may not have a form that easily and obviously lends itself to being said to ‘correspond’ with something in nature. For Instrumentalism Instrumentalism finds truth not in correspondence to reality but in pragmatic virtues. Thus, on an instrumentalist account of science, a theory is valid because it can be used to make predictions, to unify a group of facts, as a base for further research, or as a platform for engineering. The case for a generalized instrumentalism relies on the gulf between subject and object, between human enquiry and the way the world really is. I am not convinced that this gulf is so wide – humans are not cut off from the material world in the way imagined by early modern philosophers. However, cases for more local instrumentalisms can be found in conflicts between realist and more pragmatic virtues, conflicts that are sometimes resolved in favour of the pragmatic ones. Following the work of Nancy Cartwright (e.g., 1983), I argue that for theories to be widely explanatory, applicable, and elucidating requires that they deviate from being realistic depictions. Good theories often require abstractions or other movements away from mere empirical reality. Again, I proceed by way of an example, R.H. MacArthur and E.O. Wilson’s (1963, 1967) equilibrium theory of island biogeography (IB), which can be treated as true but does not correspond to its objects. This counter-intuitive result is because IB’s truth stems from its abstraction and generality, whereas its falseness is falseness to concrete and particular facts.

For Realism and Anti-Realism 69

The central question MacArthur and Wilson addressed was, How many species of a given taxon (related grouping) should we expect to see on an island? Whatever island and taxa we are interested in, the factors affecting their numbers are immigrations, local extinctions, and speciations. Because speciation it is relatively uncommon, MacArthur and Wilson focused on understanding rates of immigration and extinction. What follows is a non-mathematical account of IB to show its intuitive appeal (see also Sismondo 2000). The important factor affecting immigration is degree of isolation. A measure of isolation, good for most purposes, is distance from sources of colonists. The closer an island is to a large shore, the more frequent new arrivals are. A second factor affecting immigration is the number of species already present on an island. The fewer species, the better the chance of a genuinely new immigration, because the odds are good that any new organism floating up to shore won’t already be represented. Chance of extinction is inversely related to population size. If an organism is rare, catastrophes, outbreaks of disease, predation, and random fluctuations in numbers have a much greater possibility of eliminating the breeding population altogether. On small islands, average population sizes tend to be small, so the smaller the island the greater the rate of extinction. Moreover, the more organisms of different species competing for the same resources, the smaller the number of organisms of each species there can be, so the greater the number of species, the greater the rate of extinction. Re-ordering these factors slightly, we have the following: 1 2 3 4

Immigration goes down with distance from sources of colonists. Extinction goes down with area of island. Immigration goes down with number of species. Extinction goes up with number of species.

Particular islands have specific sizes and distances from the mainland. Those fix parts 1 and 2 of our equation. And in the long run parts 3 and 4 balance each other off, establishing an equilibrium number of species. The exact point of their equilibrium is determined by 1 and 2. The equilibrium number of species is entirely determined by the size and distance from the mainland. According to IB, then, all else being equal it should be possible to specify the number of species of a taxon that any given island will

70 Sergio Sismondo

reach at equilibrium. Along the way the theory explains island extinctions, showing that small population sizes are a fact of life on islands. When approached from a conceptual direction, as above, IB is trivially true, and it has been treated as such by many ecologists. Moreover, IB has been a prominent tool for structuring experimental work and organizing data. It was a touchstone for hundreds of empirical articles especially in the 1970s and 1980s – IB could be applied to almost any ecological setting, even ones that are only metaphorical islands. Yet even fans of IB have recognized that it is not a realistic theory: ‘It has been tested repeatedly, often rejected, and not yet to my knowledge proven to be both necessary and sufficient to account for the diversity of a single insular biota … Nevertheless, the theory provides an exceptionally useful conceptual framework for investigating patterns of species diversity and the underlying mechanisms which produce these patterns’ (J.R. Brown 1981: 882). For its proponents, the pragmatic and conceptual virtues of the theory overwhelm its obvious lack of realism and its empirical inadequacy. For those who are not fans, the problems with IB are worse, because for any species distribution there is a particularist explanation that does not involve the variables of the theory, an explanation that focuses on real historical organisms and ecological qualities. IB tells us only about nameless, faceless organisms on featureless landscapes, so even if it is true, it is not true about any particular situation. Because of these failings, IB appears to be waning in popularity among biogeographers, conservation biologists, and field ecologists. This might be seen as an eventual victory for realism. However, theoretical ecologists are not abandoning IB in favour of more realistic theories but in favour of more multidimensional theories. The victory for realism is only partial at best. The lessons of the case apply widely: explanatory and conceptual success for a theory demands a certain amount of abstraction away from the particularities of the real world. Even for theories that have a good claim to truth, realism is only one virtue among many. For Constructivism Constructivism claims that truths do not stem from the world by itself but that they follow from human actions and interactions and in some sense help to constitute the world. As surprising as it may seem, there is a practice-based case for a constructivist interpretation of some scientific truths.

For Realism and Anti-Realism 71

Knowledge stemming from scientific experiments most directly describes situations that are distinctly non-natural, standing apart from nature in their purity and artificiality. Experimenters make sure that their inputs are refined enough to be revealing, and use those inputs to create relationships not found outside the laboratory. In the laboratory, scientists study interesting artefacts, things created or shaped by them. It is by studying these artefacts that scientists learn about a nature that they see as more basic or more fundamental than the immediately found world. Nature is messy. It can be unreliable and difficult to work with. Therefore, scientists construct systems to study controlled, cleaned, and purified phenomena, about which models can be more easily made. For example, for the past century, the laboratory mouse has been the organism of choice for cancer research, and has been a popular organism in many other areas, including experimental psychology, animal behaviour studies, and various medical fields. There are many different strains of lab mice, exhibiting different characteristics. A mouse strain is homozygous at virtually all loci, the results of multiple generations of sibling matings. Genetic impurities that could lead to idiosyncratic behaviour or properties are thus quite rare. This purification of the mice has useful features from the researcher’s point of view: most important, it means that the mice are standardized across experiments, allowing scientists to learn about environmental sources of variation. So here is the first place where nature takes a back seat to the interests of the experimenter. Natural mouse populations have some genetic diversity, but YBrDPA and other such mice are genetically identical one to the next, and thus make much better research tools than natural mice. Laboratory mice are also domesticated. Obviously the strains have not seen ‘the wild’ for at least the number of generations for which they have been systematically purified, but in addition, the individual mice are domesticated in the sense that their environment is a thoroughly artificial one. This would pose a problem if lab mice were representatives of wild mice, with all their variety and natural behaviours. However, few researchers are interested in mice per se, but rather are after much larger game: organisms or mammals or behaviour or immune responses or some other abstract and general entity. Not only are the mice purified and domesticated, they are often enhanced. DuPont’s famous OncoMouse is a line of mice genetically altered to reliably develop cancer. OncoMouse is a tremendous tool because it removes one source of uncertainty: for these unfortunate

72 Sergio Sismondo

mice, the cancerous state is the normal state, not an exceptional one. For other biomedical research, it is possible to buy mice that have been entirely raised in a sterile environment and are therefore free of particular germs. Such mice live their early lives behind airlocks, and when they are ready to be research subjects they are shipped in sterile containers to the labs needing them. There they can be exposed to certain diseases and treatments in the confidence that they have developed no immunities to those diseases, and that they aren’t infected with other diseases that might confuse the results. And experimental settings are invented, in the sense that objects are placed in situations in which they display hidden faces or properties. Lab mice encounter all manner of artificial diets, exposures to highly concentrated substances, strange stimuli, forced interactions, etc. Producing stable phenomena requires specific types of physical environments. Of course all this artificiality comes in the context of controls for artifactuality: experiments are supposed to display structures that are in some sense independent. In a strict sense, then, scientists rarely study nature. Or, as Karin Knorr Cetina (1981) says, nature is systematically excluded from the lab. Scientific knowledge that depends on experimentation is not true in a straightforward realist way, that it follows the contours of nature. Thus, in some contexts we should adopt a kind of constructivism, a view that emphasizes the contribution we make to knowledge. This is not to deny that experiment uncovers hidden structures. Experimentation, after criticism and refinement, can produce solid knowledge, knowledge that is difficult to improve on and is therefore stable, It would be entirely reasonable to say that experimentation can produce true knowledge. But there is always a sense in which experimentation produces the circumstances about which it is true, or the context in which its results can be said to be true. The staunch realist might argue that the non-naturalness of experimentation serves to reveal natural structures, and that this is seen in the ways that experimental results are discovered, and the fact that they subsist. To the extent that experimental knowledge is found and is stable, there is surely a sense in which this would be right – although I would insist that it cannot be the entire story. For example, there is a substantial price to be paid for accepting such an argument at face value, a price that stems from the fact that it is exactly analogous to the argument that sustains Platonism about mathematics. Unfortunately, Platonism is implausible if taken too strictly: a rich mathematical

For Realism and Anti-Realism 73

world does not simply exist independently of humans but needs to be constructed in order to so exist. Similarly, it seems from the above, the rich world revealed by experiment does not simply exist independently of humans but needs to be constructed in order to so exist. There is a further constructivist consequence of the experimental style of enquiry. Concrete effects flow from laboratory manipulations. For example, the discovery of ceramic high-temperature superconductors created more of those particular ceramics, and more variations on them; as a result, there are many more high-temperature superconductors in the universe today than there were in 1985. Or perhaps less trivially, the biomedical sciences shape and reshape the nature of health and illness to bring them in line with laboratory findings. On the basis of solid laboratory research, pharmaceutical companies define medical problems that their drugs can treat. They may medicalize disorders, or they may redefine them: David Healy (e.g., 2004) and other commentators on anti-depressants describe how patients having nervous problems talk about tension and stress before the mid-1980s, talk about panic attacks in the 1980s, and talk about moods in the 1990s. The transitions correspond with transitions in the scientific understandings of drugs and their subsequent marketing. The result is that people are ill in new ways, corresponding nicely with laboratory truths. Again we see that there is a practical sense in which the world follows truths, rather than the other way around. Conclusion If we really attend to scientific practice we learn that even when seen as true, scientific claims are amenable to philosophical interpretation, of a diverse, non-essentialist, and down-to-earth kind. There is nothing unusual about science in this respect: ordinary language and ordinary knowledge exhibits all the same features and more. Therefore, deflationist arguments do not license a straightforward realism, critical or otherwise – though the positions directly opposed to realism fare no better! When we study knowledge practices, realism, instrumentalism, and constructivism are interpretive stances that can provide metaphors and contexts for understanding knowledge and its uses. So to the extent that critical realism adopts a monolithic realism, it is adopting an ideological position that obscures other ways that truths operate. We should instead make room for both realism and anti-realism.

4 Critical Realism and God GARRY POTTER

Love of one is a piece of barbarism: for it is practiced at the expense of all others. Love of God likewise. It is not their love for men but the impotence of their love for men which hinders the Christians of today – from burning us. Objection, evasion, happy distrust, pleasure in mockery are signs of health: everything unconditional belongs in pathology. – Friedrich Nietzsche, Beyond Good and Evil If God existed it would be necessary to abolish him. – Mikhail Bakunin

Introduction In this chapter I wish to intellectually and historically situate what I believe is the single most significant issue of disagreement among critical realists: the existence or non-existence of a (loving) God (or Gods). Along the way of this outline I will also indicate some of the fault lines of other critical realist internal debate. Perhaps more important, I also intend to refute some of the arguments propounded by critical realist theists. Such refutation is by its nature mainly philosophical; however, I also intend to engage with some of the sociological and political aspects of the religious issues. One might well be inclined to pick other issues as likely candidates for the designation of the single most significant issue of disagreement among critical realists. Perhaps an epistemological issue, or something substantive concerning scientific practice? Yet there is not only considerable consensus concerning such issues but also no single area relative to them specifically demarcated as a matter of significant debate for

Critical Realism and God 75

critical realism as a whole. It is part of the argument of this chapter that there in fact ought to be – objection to the concept of ontological and/ or alethic truth – but this issue directly feeds into the wider issues of God and Love. It could be argued that among critical realists, theological (and/or anti-theological) issues are of secondary importance to their political concerns. And this is in some respects true. Critical realists, whether or not they believe in God, (and regardless of their different religious affiliations, or the lack thereof) share a common concern with human emancipation and the struggle against oppression. Certainly considerably more has been written by them concerning both political philosophy and practical politics than about God. Probably there has even been greater CR introspection concerning its relationship to Marxism than to religion. However, as shall be shown, such debates do not cut to the fundamentals of critical realist philosophy quite so significantly, nor do they have such profound socio-political consequences for the school of thought’s reception and intellectual credibility. There is an entirely different school of thought within theology that is also called critical realism. Well, I am calling it ‘entirely different’; it is a phrasing that indicates my bias on the issue. Margaret Archer emphasizes the considerable similarity between the independently reached conclusions in these two separately evolved schools of thought. It is perhaps time, she suggests, for them to become acquainted with each other (Archer et al. 2004: 22n1). There has been within CR in the past few years not only considerable discussion about the existence of God but also concerning the nature of God. There has been a lot of talk about ‘love’ and ‘unconditional love.’ One could even go so far as to say that there has been among many critical realists a celebration of religiosity; to use Archer’s phrase, theism has come out of the closet. The world, with the possible (and arguably not!) exception of Western Europe, is predominantly made up of believers. Yet atheism has been the default position of the academy. It is not exactly that atheists such as myself wish to push the believers back into the closet. It is that to assert atheism as the default position of the academy is mainly correct, at least for the social sciences, and that ‘God talk,’ ‘love talk,’ and religious sociology (as opposed to the sociology of religion) have a profoundly negative effect on the credibility of the entire critical realist (the philosophical as opposed to the theological) school of thought’s principal role as philosophical under-labourer for the sciences, social sciences and substantive humanities disciplines. A good many critical

76 Garry Potter

realists do not wish to get to know the other critical realism or engage in any theological discourse whatsoever. Many other critical realists feel even more strongly. They feel that if theology has become a central tenet of critical realism discourse, it is perhaps time to stop calling themselves critical realists. That atheism has become the default position of the social sciences Archer et al. (ibid.) argue is a direct inheritance of the Enlightenment. I agree. But she and her colleagues also argue that such a position results from that movement’s empiricist, and later positivist, ontologically impoverished epistemology. Here, I disagree. Atheism does not stand on positivism but rather on a realist (even, one could argue, critically realist) materialism. Theism (in most accounts) stands on idealism. The Enlightenment, as everyone today knows, had a certain dangerous arrogance and naivety, from which a good many unpleasant consequences followed. However, it was also an achievement, a collective human achievement. Perhaps the most significant of its many and varied accomplishments was with respect to knowledge and science. It dethroned God as the measure of all knowledge and placed the criteria of judgement within human capabilities, experiences, and reason. Anthropocentrism is considered a philosophical error from the point of view of theistic critical realists. But anthropocentrism is a historical achievement from the point of view of others. Yes, they would affirm with Nietzsche, God is dead! There is no philosophical need for Him in this world or in this school of thought; rather the need, with respect to religion, is to socio-politically understand it and to figure out how to deal with some of the unfortunate consequences of its practices. From the perspective of such critical realists, debates concerning the nature of God are as philosophically redundant as the scholastic’s debates about angels on the head of a pin. Archer (2004) and other critical realists1 argue that atheism is a belief in the sense that an absence of belief is logically symmetrical with its positive affirmation. Each needs to present arguments for its case. I would argue that this is no longer exactly so. I would argue instead that atheism had to ‘make its case’ against a human history where theism was so omnipresent that the very idea of there not being a God was virtually unthinkable. But it did make its case; and that is one of the reasons why the positions on today’s issue of God’s existence are not entirely symmetrical. The issue has already been decided. I do not mean by this that individuals have decided. Individuals each come to

Critical Realism and God 77

the issue (or not) in his or her own time and in his or her own way. Rather, I mean a community of thinkers has decided. Archer would actually agree with this; the problem is, she would argue that the decision is not rationally justifiable. The community I am referring to is not, of course, society as a whole. In society as a whole, the vast majority believe in some kind of God and a host of religious dogmas – even if they do extremely poorly at living according to the ethical guidelines of any religion’s teaching (religion of course has a religious explanation for this). No, the community in this case is far more of an elite grouping; it is the intellectual community that secularly evolved out of the Enlightenment. It is the community that accords a certain epistemological privilege to reasoned argument and empirical evidence. It accords a methodological privilege to respectful rational argument. Archer et al. are a part of this community but yet accuse it of injustice and hypocrisy vis-à-vis the religious believer. Atheism, they assert, is unfairly epistemologically privileged within this community. One may believe or disbelieve, but with respect to rational debate, the onus of justification is on the believer. This is part of the dark side of the Enlightenment and one they would reform. The other side of the debate would argue that the epistemological privileging of the atheist position as that of a rational default starting point is well deserved and that positivism does not really enter into it. The atheist (academic community?) certainly does believe that the burden of proof of God’s existence is on the theist. But that is not all. Today’s atheist is inclined to believe that the terms of the argument requiring proof have also shifted. This is because they have shifted their attention to what they consider more important issues. They would rather not debate it at all. It is not just that the burden of proof is on the theist but also a burden to provide new evidence and argument. Neither testimony of miracle, biblical scholarship, or new logical deductions hold much appeal. In fact, none of the arguments with which I shall engage in refutation later in this chapter hold much interest for me. I engage in them only because of my interest in the authors. This is what is at stake in this issue for critical realists. Do we still belong to the same community? Five Moments of Critical Realism Bhaskar (2000) identifies five moments in the history of the development of CR philosophy, each of these in turn identified with a seminal text of

78 Garry Potter

his own authorship. The first is ‘transcendental realism,’ identified with A Realist Theory of Science (1975/1997). The second is critical naturalism, identified with The Possibility of Naturalism (1978/1997). The two books truly are seminal texts among critical realists. They are for most, I suspect, the reason they came to CR in the first place. RTS provides a necessarily thorough critique of positivism and its Popperian variations. It properly circumscribes the constructivist aspects of scientific endeavour. It establishes a sound understanding of ontological realism and epistemological relativism. Above all, it clarifies the difference between epistemological relativism and judgemental relativism and establishes the manner in which a rational judgementalism might be exercised. The achievements of PON are no less impressive. PON achieves the promise of its title: the possibility of naturalism in social science is established. The conditions for the possibility of social science are to be found in the conditions of possibility of everyday life. PON offers a telling critique of the case – stemming chiefly from the hermeneutic tradition – that the uniqueness of our human relation to the subject matter of the social sciences makes the latter enterprise impossible. That case is turned on its head: it is just that uniqueness, that humanness to be found in social science and its subject matter, that makes its scientificity possible. The two books, I would argue – with some necessary supplementary work (fairly brilliant work!) done by scores of others – essentially complete the work of the philosophy of science and social science. They also form the basis of a near complete consensus to be found among CR thinkers concerning the fundamentals of these two areas of enquiry. Again, along with the supplementary work of some pretty impressive thinkers in their own right, the two books form, as Bhaskar argues, two of the five moments of CR’s history of development. I would assert that it is the potential of the two moments to benefit scientific practice, most particularly social scientific practice, wherein CR’s primary value lies. Notwithstanding the widespread agreement among critical realists concerning the central conclusions of these two moments, it is here that we must note the first, tiny hairline fracture of disagreement among them. I refer here to what I will call the ‘anti-humanist objection.’ This disagreement between variants of realism will perhaps highlight the different sorts of consequences of the serious disagreements we shall look at in a moment. That is, I would regard it as a socio-politically undamaging sort of debate to have within a school of thought – indeed, even one that could potentially be intellectually beneficial.

Critical Realism and God 79

It is not that the debate here is not important or in its own way fundamental. It cuts to the heart of our understanding of determination and agency. This anti-humanist line of thinking within CR is very much influenced by Michel Foucault’s (1972a) Archaeology of Knowledge and Emile Durkheim (1964). It emphasizes the intransitive nature of discursive structures.2 However, it is worth considering the extent to which Bhaskar would disagree with this line of thinking. I mean here the Bhaskar that conceived and authored the two previously referred to seminal texts of CR’s early moments (as the later Bhaskar, or so I would argue, is a different kettle of fish altogether). Though the early Bhaskar would disagree, this would reflect matters of degree and of emphasis concerning the possibilities of human agency. It would not be because he did not recognize an intransitive dimension to discourse as well. My own position on this issue reflects the kind of deliberate fence-sitting adopted by Pierre Bourdieu (1993: 25): ‘The degree to which the social world seems to us to be determined depends on the knowledge we have of it. On the other hand, the degree to which the world is really determined is not a question of opinion; as a sociologist it is not for me to be “for determinism” or “for freedom” but to discover necessity if it exists, in the places where it is’. No matter: the point I would make here is that this issue can be happily and productively debated within the CR school of thought without any socio-political negative consequences for that school of thought. My contention is that the later critical fault lines are of a different order. CR’s third moment is that of ‘explanatory critique’ and is marked by the book Scientific Realism and Human Emancipation. Or rather, it is for some. This leads us to the second sociological fracture line. That is, it brings us to the next significant issue on which critical realists are in dispute. It brings us to the question of Marxism’s relation to CR. This question has been debated at considerably greater length among critical realists3 than the first issue. Personally, I would mark the moment of explanatory critique as occurring long before CR. Explanatory critique moves us beyond merely logical or evidential grounds for evaluating theories or beliefs about the world – it directs us to the conditions that give rise to the beliefs. It is marked by that moment wherein Marx asserts that Ludwig Feuerbach has effectively completed the critique of religion, but then insists that we move beyond the critique of the religious notions of soul to the critique of a ‘soulless world’ – in other words, to the critique of political economy (Marx 1978b). Bhaskar reverses the direction of the critical trajectory of Feuerbach and religion to Marx and political economy, with a ‘return’ to religion.

80 Garry Potter

There is thus both agreement and disagreement here. Both the religious and irreligious membership of the CR school of thought are agreed on the necessity for political engagement. They all want a science and philosophy that is moral and political. They also – or they would not be critical realists – want their politics and morality to be philosophically grounded and justified. Bhaskar’s moral/political philosophy thus accomplishes (apparently) two crucially important things for critical realists. First, he directly overturns a long-standing, all-too-influential assumption in moral philosophy: that one cannot derive ought from is. Second, the politics that comes from his deriving ought from is seemingly links up well enough with the CR framing of Marx’s eleventh thesis – the point being to change the world. I am very much in sympathy with the intent of Bhaskar’s moral/ political philosophy. I too wish to derive ought from is, and elsewhere4 I provide a wholly atheistic manner of doing so. The problem is that his derivation of ought from is depends on a logical slippage. We will analyse it in some detail in the next section. But here is the essence of his argument, gleaned from the range of Bhaskar’s works since Plato Etc. First, we answer the question, What is the alethic (ontological) truth of the human condition? Next, we observe that the alethic truth of the human condition can be and usually is occluded by structural constraints to both knowledge and action. The philosopher’s task is to overcome such occlusion and constraint. Our knowledge of the alethic truth of the human condition tells us what is implicit in each and every action we undertake.5 The philosopher will make explicit what is only implicit in our actions; that is, the philosopher will articulate the truths within us, and this is what we should live by. And what is this explicitly, then? The alethic truth of each human being is God. God is unconditional love. Implicit in every action is the desire to absent constraints – a conatus to freedom and through our connections to others, to universal self-realization and the eudaimonistic society. Because we (on some level) know this already (except for the occlusions of structural constraints), we then have spontaneous ‘right’ action – carefree, thoughtful, and loving. Let us point out the flaws in this chain of reasoning. First, it begins with a notion of alethic truth. Beginning with alethic truth I will later argue is a category mistake. Indeed, I will argue that this mistake is the fundamental error on which all the later intellectual movements toward theism depend. Second, there is the assertion that structural

Critical Realism and God 81

constraints occlude our knowledge of the human condition. There is no problem with that, but next comes an essentially religious notion (supported only by faith) as to just what is being occluded by the structural constraints on knowledge. According to Bhaskar, what is occluded is the alethic truth of the human condition. The truth is that we are essentially God-like (or actually manifestations of God) and that God is unconditional love. It is our God-like unconditionally loving nature that provides us with a ‘conatus to freedom,’ a desire for the revolutionary transformation of society and individual attainment of nirvana simultaneously. But none of this is really necessary to feel bound to one’s fellow human beings, to have a desire to alleviate suffering and a determination to try to change the world. There are plenty of ‘this-worldly’ stimulations and secular moral justifications for such convictions and actions. However, my contention is not merely that the religious justifications are unnecessary but also that they are seriously logically flawed. Again, the principal problem is with the notion of ontological truth. The fourth moment of CR’s history has been labelled ‘dialectical critical realism’ and is marked by Dialectic: The Pulse of Freedom. We can also find much of the critical emphasis of this moment in another of Bhaskar’s works: Plato, Etc.: The Problems of Philosophy and Their Resolution (1994). We find here Bhaskar’s assertion of the importance of ontological monovalence as a, if not the, crucial error in the history of philosophy. We find here his continuing stress on the importance of absence. Note the connection here to Archer et. al.’s argument that was discussed in the introduction. It is at this moment that from the political sociology standpoint of this article, we find some considerable irony. Most Bhaskar critics of the ‘turn to the East’ deeply embraced this moment. Many who were politically and philosophically disturbed by From East to West saw Dialectic as a brilliantly insightful book, and there was little criticism of it when it first came out.6 CR’s fifth moment is ‘transcendental dialectical critical realism’ and is marked by From East to West: Odyssey of a Soul. Some of the collective unease felt by critical realists on publication of this work is indicated by the subtitle of a critical review that Gary MacLennan wrote: ‘a book too far?’7 However, I wish to make two strong contentions here: (1) that for many critical realists, Dialectic and Plato, Etc. were already ‘bridges too far’; and (2) that while most of the philosophical critique directed at From East to West has rested on the mistaken notion that the foundations

82 Garry Potter

provided by the ‘fourth moment’ were sound; actually, the truly serious philosophical errors (see below) occurred much earlier, and most of the ‘socio-politically objectionable contents’ of From East to West occur at the earlier moment as well.8 On one level, this is not too surprising a contention. Bhaskar did not suddenly spring his ‘Eastern religious ideas’ on us in a single work. The fifth moment of critical realism, in spite of the specifically Eastern turn of Bhaskar’s own work, is not necessarily Eastern at all. Rather, it is simply religious. I would argue that the work of others is also quite significant; most particularly I would reference the work of Margaret Archer (2003), Andrew Collier (2003b), and Doug Porpora (2002). The discussion is not only of the existence of God but also of the nature of God and most crucially His allegedly loving nature. This idea became respectable in the critical realist community at this moment; and while Bhaskar’s theology is both New Age and Eastern in its emphasis, Archer and Porpora are Catholic and Collier is Protestant. They make less specific theological claims than Bhaskar, or, perhaps more accurately, they make them more modestly. The principal arguments for their Christian God generally derive from all critical realism’s moments mentioned above; or rather, as I shall later argue, they derive from the same basic mistakes as made by Bhaskar. I shall present this critique later on. The principal argument of Archer et al.’s work Transcendence: Critical Realism and God is not, however, and most important, concerned with proving the existence of a Christian God (though some arguments with respect to that are advanced by them). Rather it is two-fold. On the one hand, they assert that the truth values of religious belief are potentially subject to reasoned debate and a rational judgement concerning their comparative validity. On the other hand, they argue that atheism is merely one position among many and should enjoy no epistemological privileging. It is rational, they argue, to base one’s belief in God (or not) on one’s experience (or not) of such. I will come back to this argument later and to the reasons why I reject it. But first I want to indicate some of the philosophical and sociological problems with respect to their former argument concerning the subjection of religious dogma to the criteria of rational judgementalism. Critical Realism and the Interfaith Dialogue – Rational Judgmentalism and Bad Argument Archer et al., and more specifically in this context Doug Porpora, as the author of their chapter on inter-faith dialogue, do admit of sociological

Critical Realism and God 83

and political considerations as possible constraints on debate about the truth claims of rival religious traditions. Porpora concludes that at this time the world simply is not ready for such a dialogue. Not being ready, he admits, creates a tension between the philosophic impulse of critical realism and the culture surrounding much of theology. I see the same tension in the internal debates of any theological discourse and the philosophical adjudication of truth as he does in interfaith debate. There is too much going of what Nietzsche would call the occasionally necessary ‘will to stupidity.’9 We have with Archer, Collier, and Porpora on the one hand the affirmation of rational judgementalism with respect to religious belief and yet on the other a subscription to the full Bhaskarian trajectory of argument from alethic truth to the Ultimata. But what would they say to my refutation of Bhaskar’s argument for reincarnation (below)? I would suspect they might think it too harsh a judgement. I suspect that critical realist Christianity’s rejection of reincarnation would be infinitely postponed on the philosophical field (we can leave that one to the atheists) because of a shared inclination to talk about unconditional love instead. Yet such arguments and their critiques are very simple and easily decidable. First the argument: Deduction of the necessity for reincarnation turns essentially on three features: first, that of universal causality; second, that of the emergence, i.e. causal and taxonomic irreducibility, of intentional states to the physical states through which they are manifest; and third, following on from the first and second, (a) the causal explicability of intentional phenomena, presupposing the pre-existence and (b) the causal efficacy of intentional states implying the post-existence of the being who is the subject of the intentional state. The continuant in question is commonly called the soul. (Bhaskar 2000: 60, my emphasis)

The error occurs at 3b. For an intention to be causally efficacious, it is indeed necessary for the subject of the intention to continue to exist after the initial formation of the intention, so as to be able actively to make the intended event (effect) come into being. It is true that intentions may be causally efficacious. But intentions are certainly not necessarily always efficacious. That intentionality may be causally efficacious is a real characteristic of it that may not be actualized in the form of an event in the future. There is thus no necessity for the subject of the intention to continue to exist. There need be no continuant. As the argument for

84 Garry Potter

reincarnation stands as a subset of this broader argument concerning intentionality and continuity, it too is refuted. There is thus no necessity for a ‘continuant’ (soul) to continue to exist after death, regardless of the fact that human beings are the subjects of intentionality. There may indeed be souls (thus this argument does not necessarily speak to Christianity’s insistence on such), but their existence (not to mention their continuity in bodily reincarnation) cannot be proved in this manner. Very well, he made an error in reasoning. But what would Archer or Porpora or Collier think about it? Notwithstanding Porpora’s socio-political reservations about the dangers and untimeliness for a true interfaith debate about the truth values of the different propositions and beliefs of different religions, surely it should be possible within critical realism? According to a shared critical realist understanding of rational judgementalism, such beliefs should be corrigible as a result of reasoned debate. Perhaps it is I rather than Bhaskar who has made the error in reasoning above and in the more fundamental philosophical arguments below. Perhaps. I do not doubt that were Archer et al. not to buy my argument, they would have no reservations concerning a public expression of the flaws in reasoning. But would the reverse be true? This may at first appear to be a very insulting suggestion. But it is insulting only insofar as there is a usually unstated mutual scepticism and feeling of epistemological superiority between atheists and theists. Outside of the ‘atheistic theists’ (whose beliefs in God are so abstractly transcendent, impersonal, and linked to the natural world that they are virtually unobjectionable to any atheist), Archer et al. are perhaps the most open-minded and reasonable Christians I have ever met. Rather, my suggestion is that there is an unacknowledged (even in their case of extremely rigorous self-reflexivity) element of irrational constraint to such debate. The religious brother- and sisterhood of shared concerns with respect to God and Love are such that they would be reluctant to pronounce so definitely on the contrary position. This, I am suggesting, would be true not only for the truth claims of a belief in reincarnation but for most all other religious dogma as well. I am suggesting an inbuilt socio-political tension to the entire project of attempting to philosophically ground theological dogma. If we were to compare science and religion in this respect, it is not that the passionate attachment to particular theories and paradigms is emotively weaker in science than that between different religious beliefs. Rather, I am suggesting the attachment is different in kind.

Critical Realism and God 85

And try as Archer et al. will, they simply cannot fully subject the canon (any canon!) of religious beliefs to the criteria of rational judgement. In making this assertion, I am also going against the entire project behind their book Transcendence: Realism and God. I am also going against what would appear to be some powerful contradictory evidence contained within that book. There are some very important Christian beliefs that they do submit to rational scrutiny and pronounce judgement on – two in particular: God’s omnipotence and an inerrant Bible. The former is considered from the standpoint of the evils of the world and justice. It is an old dilemma for believers. How, from the standpoint of personal suffering, either on an individual (for example, the death of one’s child, perhaps) or collective level (for example, humanity’s various genocides), can one reconcile the idea of there being a loving God or a just God that would allow such horror as is apparent from human history? The arguments on all sides of this issue are familiar to believers and non-believers alike. Logically the problem is not at all difficult. One cannot have both a just, loving God and an omnipotent one. Rejecting the logical exclusion of such a choice through various faith-based ‘getout clauses’ has been a frequent tactic of the believer. However, this would be to go against Archer et al.’s critical realist project. They accept the contradiction and the necessity for a choice. They choose the loving God. Similarly, they admit the conflicting accounts of the same stories frequently found in the Bible and thus are forced to conclude that it cannot all be true. So, I would appear to be mistaken. One after another the various conflicting claims of religious dogma could be (and from their critical realist standpoint should be) subjected to similar logical and evidential scrutiny and pronounced on. But, I am asking, could they really be? There are two problems with doing so; one is institutional and sociological; the other is philosophical. The very clarity of their position on the above issues (and others in the book) raises the issue of institutional boundaries to debate. Doug Porpora brought up quite sensible issues with respect to political and cultural constraints on such debate. But what about the constraints imposed by his own faith? Here Andrew Collier is better institutionally positioned than Archer and Porpora; faith is more individual and less institutionally policed for Protestants than Catholics. Collier can perhaps stretch his Christian dogma further than they, as he has no higher authority (other than God) to answer to. They, by contrast have an institutional hierarchy of spiritual authority to pronounce on the question of just how much

86 Garry Potter

deviance from accepted dogma is allowable for one still to be considered a Catholic. I would suggest that they are well over the line already. Will they be charged with heresy? Will they be ex-communicated? I highly doubt it. Many within the liberation theology movement, for example, have gone much further in their divergence from papal edict. However, it is also a part of their cultural heritage as Catholics that many have not gone nearly as far and received much harsher sanctions. My argument here is that somehow the very project of subjecting dogma to the criteria of rational judgement and evidence is counter to the very essence of the social institutional existence of their Church and a great many others. There is a philosophical problem as well. Their juxtaposition of rational judgementalism and epistemological relativism harmonizes them in such a fashion that one can, as it were, rationally critique the evidence and argument underpinning conflicting beliefs, yet in the end happily ‘agree to disagree.’ Ultimately, it is the different standpoint of personal experience that is the deciding factor. Thus, it may be rational for one person to believe one thing and another person to believe another because of their differing experience. Indeed, this is one of the principal arguments of their book. They assert a symmetry between the two positions, whereby it is rational for each to retain his or her beliefs because of the experience of God or the experience of absence with respect to Him. I will critique this idea of a symmetry of absence and presence when I return to Bhaskar, as it is one of his and a great many other critical realists’ crucial positions. But there is another problem with this position, and it manifests itself in the symmetry of evidential weight accorded to two ‘positive’ but different religious experiences. It undermines their previous solid clear judgements with respect to truth claims about the nature of God. One logically cannot have a loving God and an omnipotent one. They chose the former. But on what basis? Apparently on the spiritual experience of God’s love. But history and literature are full of people railing against an unjust God. Many, apparently, have experienced God’s malevolence. It would seem, then, that it would be equally rational for them to retain belief in a petty, spiteful, unjust God and by extension an omnipotent one as well, as the same problems (the problem of ‘Evil’ for example) do not manifest themselves for this position. It is not simply that beliefs are all fallible. Archer et al. correctly assert the importance of experience in determining belief. They are also correct insofar as they assert that experience alone cannot be the sole

Critical Realism and God 87

arbiter of rational judgement but must be placed in context with other evidence and criteria. They assert this correctly as a mainstay of critical realist difference from empiricism. My point here is that they cannot sustain it in practice. Rather, if we push any such debate far enough, it will reduce to the different religious experiences, because the experiences in the theological case are in fact always the ultimate trump card but not in the general condition of rational judgementalism. Why this is so will be explained in the next section. The Asymmetry of Presence and Absence with Respect to Evidence and Belief Roy Bhaskar attaches great importance to what he sees as the error of ontological mono-valence. His argument is that absence and presence are ontologically equivalent in value. They both can be objectively real; for example, the absence of food can be as real as the presence of food. The existence of that absence can be equally important to human beings as well. One can also speak of ‘absenting absences’; and again we could use food and hunger for an example. However, I believe that this is at root a semantic issue; and that though semantic issues can be of crucial importance with respect to many philosophical topics, in the case of theism it is actually somewhat misleading. Archer et al. argue for the possibility of ontological realism, epistemic relativism, fallibilism, and rational judgementalism appertaining to the question of the existence of God. Ontological realism would assert that either God does or does not objectively exist, regardless of our beliefs, knowledge, or experience of Her. Given the fallibility of human judgement generally, this too is a subject on which any of us could be mistaken. So far, so good. The absence of God’s existence, Its non-existence, as it were, could be an objective reality, no matter what the millions and millions of believers in the world believe to the contrary. The existence of God could very well be an objective condition, regardless of the hundreds of atheist academics’ lack of either an experience or evidence of such and a corresponding belief. Again, so far, so good. There is indeed an equivalence between the respective positions. Regardless of the fact that above I just questioned the practical possibilities of believers consistently practicing rational judgementalism, there is, at least in theory, no problem here for them. So that leaves epistemic relativism. Our beliefs (about anything) are all socio-economically, culturally, historically situated. We experience

88 Garry Potter

the world from different perspectives. How do the possibilities of rational judgementalism and epistemic relativism come together? While all beliefs can be subject to rational debate and critical evidential scrutiny, the particular case frequently cannot yet be publicly conclusive. The fact that we may come from very different standpoints or perspectives to begin with and that we may have different relevant personal experiences allows for the possibility that it may be rational for different individuals to hold different beliefs on such publicly unresolved matters simply because they have different experiences. Archer et al. hold, quite reasonably in my view, that in the absence of more compelling evidence and argument going against one’s own experience, it is quite rational to trust it. This is precisely where we disagree with respect to the issue of God’s existence, though. There actually is compelling evidence and argument not to trust at least the theistic interpretations of ‘religious experience.’ Archer et al. believe that the academy has unfairly privileged one side of a debate that is epistemologically symmetrical because of each side’s crucial dependence on personal experience as the deciding factor. But such is not so. They write that ‘atheism is the automatic default setting within the academic world.’ I accept that this is the case to some degree, though to a far lesser degree than they maintain – the Christians have not nearly so much been in the closet – but more important, I believe that to whatever degree atheism is the default position, it is in fact fully deserved. The positions are not symmetrical because it is differing interpretations of experience rather than simply experience where the difference lies. As critical realists, Archer et al. know that completely uninterpreted experience is not quite possible. Even perception itself is ‘theory laden’ to some degree. Intentionality projects, selects, adds, subtracts, and distorts our perception to varying degrees in particular cases; but, we can assert with confidence, it always does so, at least to some degree. The causal influences directing intentionality can be critically examined. We can understand not merely the existence of sensory perceptual distortion but also the direction and manner of such distortion in many particular cases. For example, the visual perception of a straight stick as bent when it is partially inserted into water is easily understood as being due to the behaviour of light passing through different media. The hallucinatory effects of LSD can cause a perception of there being letters of the alphabet coming in and out of focus and existence as we look at a shag

Critical Realism and God 89

carpet. This common experience among psychedelic drug users can be understood as deriving (at least in part) from a shared literate culture. Archer et al. would assert that the theist has the experience of God to base his belief in God’s existence, while the atheist has the experience of an absence. But this is not quite correct. It is not the theists’ experience that the atheist questions but the interpretation. Even if through intentionality, experience and interpretation are always intimately bound up, it is nonetheless relatively easy to analytically disentangle the epistemological issues concerning them. Let us use an old one of Sartre’s examples. We see a burning bush. A voice begins to talk to us, informing us that She is God. Do we accept this as the raw data of experience and a good rational basis for belief in God? Perhaps Moses was not really very self-analytical. Or perhaps it was simply that he was simply unaware of the range of possibilities of the effects of psychotropic drugs or of the delusional capacities of psychosis. However, a modern-day person experiencing the same should be aware of these and other possible explanations, rather than simply accepting that God had chosen to speak to him. Few Christians describe their feelings of transcendence in such direct, simple terms. It is interesting that even among Christians there is a strong tendency to preference a psychoanalytical explanation when they do. No, the much broader, vaguer experiences of ‘transcendence,’ or of an ‘oceanic feeling of love and belonging’ and such like, are more commonly used as the experiential base for theism. The atheist doubts the existence of such feelings usually not at all. I and many other atheists have experienced them ourselves, and though, yes, it is true, God and the burning bush are usually outside our experiences, even were we to experience that specific and intense Judeo-Christian religious moment, still we would question it. It is not the experience we would doubt but the interpretation. What is rejected is the equivalence of interpretations concerning the experience. It is here that we must consider the equivalence of presence and absence in terms of being epistemologically symmetrical with respect to the burden of proof. There are two lines of argument against the epistemological symmetry argued for by Archer et al. First, there is a kind of common sense reductio ad absurdum that can be applied. Archer et al. (2004: 4) write that ‘their absence of religious experience is itself a kind of experience.’ Let us rewrite their sentence: ‘their absence of alien-abduction experience is itself a kind of experience.’ And again: ‘their absence of dancingleprechaun experience is itself a kind of experience.’ Archer’s perspective

90 Garry Potter

is epistemologically very democratic. Why is it only the atheist who must assume an equality of burden of proof with the religious? Why is it not that everyone who does not believe everything that any crackpot on the street asserts as existing must assume an equal burden of proof to assert the loons as wrong? And the dismissal of such cannot be justified simply in terms of numbers of theistic believers versus the small number of loons. The numbers are not at all so small. Indeed, there are many more believers in ‘flying saucers’ in America than there are atheists. No, Archer et al. have to give other reasons for asserting a difference between the unicorn and God with respect to burden of proof. Archer et al. do not posit experience as the be all and end all of belief. They assert ‘that unless we have compelling grounds to discount a particular experience, our first impulse rationally is to trust it’ (2004: 4). Once again, though, we must stipulate contra them, that it is the interpretation that would be discounted rather than the wholly uninterpreted experience. We would agree that some evidence, some compelling grounds are required to discount experientially closely bound interpretations. But it is precisely here that absence and presence do not balance out epistemologically. If the experience can be explained alternatively without resort to the supernatural, then this itself is ‘compelling ground’ to doubt the validity of a supernatural explanation. Let me give an example to make clear just exactly what I am claiming here. Imagine witnessing a magic performance. The magician makes a supernatural or mystical claim concerning it. It is a ‘magical force’ that is the common causal mechanism of the events witnessed. This claim is not merely heard and registered by the witness, but one could say that in some sense she ‘experienced’ it (in my sense of closely bounded experiential interpretations). It ‘looked like’ ‘magic.’ She could not see any possible rational explanation without invoking some supernatural powers. From Archer et al.’s perspective we could just stop here – though I am sure they would feel a little uncomfortable with the substitution of magic for religion. But this is not all there is to the case. There are other compelling reasons for rejecting the magical explanation. We know of the existence of other magicians who can do all the same tricks and yet make no claim to ‘true magical powers.’ They admit there are ‘tricks’ whereby nature is all that is required for the causal mechanisms. Now, are we still justified in retaining our experientially based faith in magic? I would suggest not. Psychoanalysis, social science, Durkheim, Marx, Freud, and others have shown us the religious interpretive ‘tricks.’ They have causally

Critical Realism and God 91

explained the appeal of the religious explanation for the apparently religious (similar to apparently magical) experience’s interpretation. These explanations would provide for me the ‘compelling reasons’ for doubting even my own experience of being spoken to by God were I to experience such, but yet are seemingly insufficiently compelling for Archer et al. to doubt their religious interpretation. They are well aware of the above-mentioned theorists. They know Nietzsche well. But I wonder. They are impelled to question and doubt. Perhaps they feel the desire to set forth their reasons less to convince others and more to convince themselves. I would offer this amateur Freudian analysis: as critical realists they have a deeper faith in reason and evidence than their faith in God. The reason and evidence have been eating away at their Catholic and Protestant faith. But they can’t give it up either. As philosophers and sociologists they well understand the socialization of guilt. Part of that understanding is the understanding that reason is seldom enough to completely overcome what one has been thoroughly socialized into. This is where they make their mistake. It is not the different ‘experiences’ of the theist and atheist that grounds their different conclusions but the extent and intensity of their religious and theological socialization. When Archer and Porpora move from the vagueness of the experience of transcendence and the very broadest notions of God to the particulars of Catholic articles of faith, their project becomes increasing difficult to defend. The same is certainly true for Collier’s Protestantism or Bhaskar’s New Age, Buddhist, Hindu, all-religionembracing theism. They are much stronger on the ‘main line,’ so to speak, where their philosophical and theological commonalities are fast becoming the defining features of critical realism. As I am deeply unhappy with this spiritualization of the critical realist school of thought, I feel I must attempt to refute it at its source, so to speak. The source is not to be found so much in the work of Archer, Collier, and Porpora (either separately or together) or in that of the later Bhaskar. The significant error occurs much earlier: in the related notions of ontological and alethic truth. Ontological and Alethic Truth It is arguable that it is alethic truth that is the distinctive concept that sets CR apart from other variants of realism. What is now the Journal of Critical Realism was, after all, called Alethia for a couple of years prior

92 Garry Potter

to its current incarnation. In any case, to reiterate a basic point about the intellectual history of the CR school of thought: many critical realists never fully took on board the notion of alethic truth. Alethic truth is the cornerstone of Bhaskar’s ‘truth tetrapolity.’ The truth tetrapolity allegedly expresses the relationship between the different senses and categories of truth. First, there is the intentional aspect of truth: what Bhaskar calls the normative-fiduciary. Assertions are (or should be, to be truthful in this sense) made with a sincere belief in their truthfulness – a kind of individually subjective aspect of truth. Then there is the epistemological inter-subjective aspect. This is found in the various instances of consensus among the scientific community – significantly with respect to the production of scientific objects of knowledge and their agreed-on identifications with real objects (this can simply be some level of agreement about how the objects of research questions in relation to empirical indicators should be theoretically specified). Then there is objective truth, an ontological aspect wherein truth lies in the intransitive dimension of reality itself. Finally, there is alethic truth. It is a special kind of ontological truth – the most important such kind. It is in fact (allegedly) that which forms the dialectical ground of the truth tetrapolity – it holds it all together, so to speak. We shall examine this claim in a moment. First, let us look at Bhaskar’s definition of alethic truth. It is ‘a species of ontological truth constituting and following on the truth of, or real reason(s) for, or dialectical ground of, things, as distinct from propositions, possible in virtue of the ontological stratification of the world and attainable in virtue of the dynamic character of science’ (Bhaskar 1994: 251). Ruth Groff (2004: 73) expresses for us the implications of such a definition: ‘Alethic truths, then, are the underlying processes that both natural and social scientists seek to identify. To be the alethic truth (y) of x is to be the generative mechanism that gives rise to x. Moreover, just as, for any given alethic truth (y) of x, y is the underlying generative mechanism that causes x to be, alethic truth in general is the totality of real, causal powers that give rise to both actual and empirical events.’ Just as the truth tetrapolity has various levels, so too does our critique of Bhaskar’s understanding of truth. Alethic truth is ontological truth, and as such we can thus assert that this notion of truth is itself simply a category mistake. Truth is inescapably epistemological – by practical and common usage definition. ‘Things,’ ‘powers,’ ‘causal mechanisms,’ etc., are neither true nor false; they simply are. The manners in which we describe, explain, or identify them are what is true or false. To simplify

Critical Realism and God 93

both Bhaskar’s and Groff’s quotations above, we can say that alethic truths simply are generative mechanisms. When stated so baldly, the claim simply seems silly. We have a variety of possible truth-claims about generative mechanisms. It is the claims that are relatively true or false, not the mechanisms themselves. The mechanisms in question and their nature (provided that we have at least minimally established even their very existence) simply are the way they are. However, the above is (on one level) only a semantic technicality. One can, I suppose, define the word/concept ‘truth’ however one wishes. The real question is why we should wish to define it one way as opposed to another. What does it accomplish analytically to define it in a certain way? There are two rather different answers to this question in the case of alethic truth. First, it appears to solve a very old philosophical problem associated with other versions of realism. Most versions of realism are associated with a ‘correspondence theory of truth,’ a theory with long standing and seemingly intractable problems associated with it, no matter what its strengths. The differences between propositions (linguistic phenomena) and beliefs (mental phenomena) on the one hand, and reality (material and ideal phenomena) on the other, are rather fundamental. So what is meant by asserting that a belief or proposition ‘corresponds’ with reality? This problem has been raised in a number of ways, and in spite of a number of alleged resolutions of it (my own included),10 the issue remains controversial. The notion of ontological truth circumvents this problem. The correspondence theory of truth needs to clearly explain just what exactly is meant by perception x ‘corresponding’ in some fashion to some thing-in-itself that the perception is a perception of. The correspondence theory of truth needs to clearly explain what it means to say statement y or belief z ‘corresponds’ to the underlying reality of a phenomenon that we wish to explain. This takes us into the terrain of intentionality, the philosophy of language and much else that has been significantly problematic in the history of thought. This is the principal strength of, and the overt justification for, Bhaskar’s truth tetrapolity. Ontological truth seemingly avoids these problems. Aspects of the thing-in-itself, whether merely its surface or its deeper realities of generative mechanisms and fundamental characteristics, simply are the ontological truths (and in the latter case alethic truths) of things-inthemselves. There is no ‘correspondence’ problem because the ‘truths’ and the ‘things’ are of the same order of existence. However, we shall demonstrate that this does not actually solve the problem.

94 Garry Potter

In order for ontological truth to be able indeed to resolve the epistemological issues raised by realism generally, and rendered quite starkly by the correspondence notion of truth, alethic truth must really be capable of its alleged capacities in relation to the truth tetrapolity. It must bridge the gap between (on the one hand) objectivity and the intransitive domain and (on the other) the fallible propositions and beliefs found in the transitive domain. If, with some fairly extreme strain on our use of language,11 we re-define truth as falling within the former, we are still left with the problem of how one can move, problem-free, from one level of truth to another. That is, if the alethic truth is – to put it baldly – the real generative mechanism of a phenomenon, we still have to ask how our beliefs and/or propositions about that mechanism and phenomenon ‘correspond.’ The correspondence problem is not solved; it is merely shifted from the question of the correspondence between words and things to that between one set of words and another set of words, which conceals the fallibility of the alleged relationship to the things to which they refer. In essence, we have compounded, as well as mystified, the correspondence problem. Although the justification for alethic truth on the preceding grounds does not hold up, alethic truth does solve another problem. It is the first step in an argumentative chain that leads all the way to God. Logical Slippages in the Argumentative Chain Bhaskar offers us three definitions of God, each of which connects with the others in an interesting way. These definitions in turn lead us to certain other conclusions about the nature of human existence, as well as how we ought to live our lives. With respect to the latter, he presents us with an interesting linkage between individual enlightenment and what he calls ‘universal self-realization.’ Universal self-realization politically replaces the Marxist materialist notion of revolution. The first definition of God he offers is essentially irrefutable. It is, as it were, one of those atheistic (so would say any Christian or Islamic or Hindu fundamentalist anyway) scientific notions of God. God is the existentially constitutive ultimate categorial structure of the universe (Bhaskar 2000: 33). How could one possibly argue against the existence of this sort of God? Well, one could, of course, if one was not a certain kind of realist. That is, one could argue that ‘categorial structures’ do not exist in this sense. However, critical realists do understand that we tend to ‘cut reality at its joints,’ and that category in this sense is not

Critical Realism and God 95

simply epistemology but precisely where our knowledge of the structure of things coincides with the independent existence of that structure. The definition of God offered here is similar to such definitions as the ‘sum total of the natural laws of the universe’ – only it is better because it points out some real features of the universe. In short, it is a definition of God that I as an atheist have only one reservation concerning. Why do we want to call the most crucially important features of reality God? The answer lies in how this first definition connects with the second. The second is this: God is the ultimate ground or deepest categorial truth of all other things (Bhaskar 2000: 41). This is very similar to the previous definition – except that what was existence there is truth here. This is where ontological, and most particularly alethic, truth enter the picture. Let us repeat here Bhaskar’s definition of alethic truth; it is ‘a species of ontological truth constituting and following on the truth of, or real reason(s) for, or dialectical ground of, things, as distinct from propositions’ (Bhaskar 1994: 251, emphasis in original). We can clearly see how this notion of truth connects directly into our previously defined notion of God. God is the alethic truth of all things. As there are generative mechanisms and causes of things, there is a truth of things; and as there is, of course, a truth of things in this sense, then there is a God – with a very interesting sort of logical slippage taking Him, Her, or It from the realm of faith into the realm of philosophy and science. We have a third definition: God is unconditional love (Bhaskar 2000: 44).12 As such, God is the unifying, totalizing, liberating power of the universe: the fundamental ‘generative mechanism’ of our humanity. Unconditional love is our alethic truth. We can thus speak of realizing what we already are, of acting in accord with our most fundamental nature. This realization, which amounts to enlightenment, has as well a collective aspect, insofar as our unconditional love is the core of our social being (to give a Marxist political tinge to it), and thus we have a conatus not only to individual freedom but to universal self-realization and eudaimonia. It is (though rather abbreviated here) an interesting piece of reasoning. It seemingly retains for CR a morally grounded political project, while having shifted the terrain from the material to the ideal – and yet recognizes both science and materiality. The only problem with the argument is, of course, that it stands on a notion of truth that, as we have demonstrated, is unsound to begin with. It is perhaps comforting to believe that ‘unconditional love’ is the fundamental generative mechanism of our humanity and the deepest

96 Garry Potter

truth of the human condition. But that this is so about humanity rests on an assertion about God’s nature that in turn rests on a logical slippage underpinning the semantic identity of ultimate truth with fundamental generative mechanisms and categorial structures. We move from a slightly idiosyncratic, but otherwise unobjectionable, definition of God in terms no atheist could really quarrel with to a logical foundation for virtually all the world’s major religions. It is an interesting argument but one that is ultimately simply putting old wine in a new CR bottle. Theologians, have you heard the ‘good news’? But what is good news for theologians is not so for the rest of us.

NOTES 1 I am sure that in addition to her co-authors of Critical Realism and God, Andrew Collier and Doug Porpora, there are a great many other critical realists who would agree with her argument. 2 See, for example, López 2003b, Woodiwiss & Pierce 2001. 3 See, for example, Brown et al. 2002. 4 See Potter 2006. 5 There is a logical parallel here to Habermas’s ideal speech situation argument. See Hostettler & Norrie 2003. 6 Though there was, of course, significant criticism of it later on. See, for example, Hostettler & Norrie 2003. 7 Gary MacLennan, ‘From East to West: a book too far?’ http:// www.raggedclaws.com/criticalrealism/archive/gmac_ abtf.html 8 Much critique of From East to West tends to see it as a divergence from Dialectic. See, for example, Hostettler & Norrie 2003 or Hartwig 2001. 9 ‘To close your ears to even the best counter-argument once the decision has been taken: sign of strong character. Thus an occasional will to stupidity’ (Nietzsche 1990: 73). 10 See Potter 1999. 11 Groff (2004: 77) gives us the following example of such a strain. She correctly asserts that the propositions ‘Planetary orbit is the result of gravity’ and ‘The relationship between planetary orbit and gravity is a truthful one’ are not equivalent in any meaningful way. 12 This definition is invoked elsewhere, less directly.

5 Rescuing Reflexivity: From Solipsism to Realism ANTHONY WOODIWIS S

Introduction Reflexivity is a sine qua non of sophisticated social scientific work. Unfortunately, as currently practiced, it is also extremely hazardous for realists since it can all too easily lead to the subversion of realism’s foundational ontological postulate, namely that the social no more depends for its existence on individual human thought than the rest of nature. In what follows, an alternative to the current mode of reflexive practice is developed that is not subversive of this postulate. This alternative is then used to explain how the current mode of reflexivity achieved its dominance by understanding it as part of a wider discursive shift that saw the displacement of ‘social structure’ by ‘identity’ as the central problematic of sociology. In my view, Émile Durkheim was not only being a good realist but was also correct when he declared in his first public lecture that the foundational, substantive ontological assumption of sociology was that social life was a ‘sui generis’ object or ‘a natural product, like organisms or plants, which are born, which grow and develop through some internal necessity’ (Traugott 1978: 44–45). Indeed, it is hard to think of a stronger way of putting realism’s foundational ontological postulate. Durkheim was also being an acute defender of realism when, on the same occasion, and presumably referring to John Locke’s notion of the ‘social contract,’ he said that what had ‘radically prevented the establishment of sociology’ was the alternative assumption that ‘men began to live together because they found that it was useful and good: society is an invention which they thought up in order to better their condition a little.’ Again, it is hard

98 Anthony Woodiwiss

to think of a keener appreciation of the source of what I too regard as the principal threat to sociological realism, namely, a humanist ontology; that is, the belief basic to interpretivism/social constructivism, namely that individual human beings are the ultimate and only social stuff. What, of course, Durkheim could not have known was that such a humanist ontology would continue to undermine sociological progress into the twenty-first century. Like the other classical theorists, Durkheim was profoundly self-conscious or, as we would say today, ‘reflexive’ about his reasoning. He would, therefore, have been extremely surprised, not to say shocked, to discover that one of the most important conduits for the forces that have continued to subvert his project should be the recognition of the importance of reflexivity to effective theoretical practice. The core meaning of the term ‘reflexivity’ is the capacity to reflect on one’s thoughts or actions and alter what one thinks and/or does in the light of such reflection. In this broad sense reflexivity has always been part of the intellectual and social life of human beings, whether as individuals or groups. Thus, we have long been aware of our capacity to reflect on why we think or act in the way that we do, and we have valued this capacity as a means of improving or at least changing ourselves and our relationships to our surroundings. Within the realm of the social sciences, the most profound products of the exercise of this capacity are the various metatheoretical positions that guide our research strategies, namely, rationalism, empiricism, realism, interpretivism, and postmodernism. From my perspective as a realist, the problem is that over the past twenty or so years many sociologists, and especially those who regard themselves primarily as theorists, have been so impressed by the profundity of reflexivity’s achievements that an accidental and largely unconscious solipsism has spread among them. This spread has resulted in the widespread confusion of the ideas that are in theorists’ minds not only with the limits of knowledge but also with knowledge of the world, and – this is the issue that I will be particularly concerned with – with their own ideas. In other words, an intellectual climate has been created wherein talking and writing about theory has come to be regarded as an adequate substitute for engaging in research and making theory. So all-pervasive has this solipsism become that it has seduced even many of those who would regard themselves as realists, the most prominent victim being today’s justly pre-eminent realist social philosopher,

Rescuing Reflexivity 99

Roy Bhaskar. Bhaskar’s seduction was marked by a change in his selfdescribed position from ‘scientific realism’ to ‘critical realism.’ Inspired in part by Jürgen Habermas, Bhaskar apparently means by critical realism the following. Whereas the non-human world in no way depends on our thinking about it for its existence and is therefore to be understood as an ‘intransitive’ dimension of being, the human and especially the social world is so dependent, at least in part, and is therefore ‘transitive’ as well as ‘intransitive.’ For this reason, Bhaskar concludes that because the thoughts and actions of social scientists can affect the nature of the social world in a way that they cannot affect the nature of the nonhuman world, we ought to be committed to the removal of the sources of social injustice. I wholly agree with the goal of removing the sources of social injustice, and I even agree that this is a legitimate goal for a social scientist to subscribe to. However, what I cannot agree with is Bhaskar’s argument that what makes such commitment follow from one’s scientific work is the difference between the social world and the non-human world. This disagreement arises is for two reasons. First, I do not agree that social life should be ontologically distinguished from life in general: to believe that social structure as distinguished from life in general both depends on our thinking about it for its existence and does not so depend is either self-contradictory – now you see it, now you don’t – or no position at all, rather than the route to what some refer to as a ‘rich and complex ontology.’ At this point, Bhaskar adopts a neoKantian position and so contradicts his own avowed realism. Second, I do not think that scientific work provides an at all adequate basis for moral or political commitments, which on the contrary are more appropriately grounded in experience or, in its absence, a knowledge of ethical and political discourses, which of course may and probably should be articulated with one’s scientific work. On this issue, if on few others, I agree with Jean-François Lyotard (1979) when he argues that we should fear the moral arrogance of any science. However, to argue that social science is an insufficient basis for politics does not mean that social science in itself has no political significance. Of course it has, and I will endeavour to point out both why this should be the case and what this significance might be when I provide my account of the rise of sociological solipsism. All that said, despite my serious ontological differences with the later Bhaskar, I am in full agreement with him as regards epistemology. That is, I agree with and have always been guided in my practice by the alternative to empiricism that he outlines: namely observation

100 Anthony Woodiwiss

should be theory driven; causal-modelling and testing are better means of articulating theory and data than hypothesis testing for generalizations; and because of the irreducible difference between our minds and what they seek to comprehend, results are always ultimately fallible rather than certain. Taking reflexivity seriously need not, however, entail the subversion of realism. On the contrary, as Pierre Bourdieu (1992a) argued so eloquently, reflexivity is essential if one is to escape repeating commonsense ideas or what, following Michel Foucault, Bourdieu calls the ‘unthought,’ and so also avoid strengthening the biases, prejudices, and preconceptions that result from and reinforce inequalities in the disposition of power. In short, a reflexive practice is essential if we wish to become critically aware not simply of our own assumptions and preconceptions but also of the inter-relationships between knowledge and power that obtain in our fields of interest or expertise, let alone incorporate these insights and outsights into our practice – if, in sum, we wish to be able to avoid confusing the thoughts that are in our minds with our own thoughts. Reflexivity, then, should not be narrowly understood as simply a looking inward at one’s abstract assumptions, biases, and, à la Jacques Derrida (1977), one’s prior reading. Of course all these personal attributes and experiences are important and have powerful effects, but they do not mean that what we see and what we picture in our theories are simply artefacts of the interactive effects of these internal, mentalistic elements and influences alone. As realism insists, the external world must be of a certain character for us to be able to perceive it and interact with it as we do (experimentally, for example). And this is so no matter how various our perceptions may be, and regardless of the fact that we will never know what the external world is really like for itself, so to speak. The critically important aspect of what society’s character must be in the current context is that it must be not only external to us but alive and therefore in some sense ‘telling’ us what we are seeing and what we should think about it. The result is that the ‘sociologist is saddled with the task of knowing an object – the social world – of which he is the product, in a way such that the problems that he raises about it and the concepts he uses have every chance of being the product of the object itself’ (Bourdieu 1992a: 235). Thus, ‘for a sociologist more than any other thinker, to leave one’s thought in a state of unthought is to condemn oneself to be nothing more than the instrument of that which one claims to think’ (ibid.: 238).

Rescuing Reflexivity 101

What is needed, therefore, is some way of grasping the character of the unthought and bringing it to conciousness so that it too can be reflected on. In my view, this is what, building on a long line of sociological thought in the classical tradition concerning the relationship between power and knowledge, Foucault provided with what, somewhat controversially, I regard as his realism-compatible concept of a discursive formation (Woodiwiss 2001: ch. 4). As I understand it, a discursive formation is a specification of the social and intellectual conditions that create the possibility of, and indeed give content to, discourse; that is, by defining and naming an object of interest, and by providing us with ways of speaking about, conceptualizing and thematizing this object. In what follows I will illustrate what a realist reflexivity involves and the sort of results it produces by using my realist reading of Foucault’s concept of a discursive formation in order to explain the rise of the solipsistic variant of reflexivity and the associated eclipse of the concept of social structure by that of identity within contemporary sociological discourse. For me, four factors were critical in effecting this eclipse, and the discussion of them will take us deep into the structure of the discursive formation that governed the nature of the picturing machine that was social theory in the late twentieth century, albeit in a highly schematic way. First, within the context created by the signing of the North Atlantic Treaty in 1949 and the ensuing cold war, the institutional site wherein sociological thinking developed became Atlantic rather than European in the sense that the United States with its distinctively individualistic culture became the principal locus of sociology’s development (on the development of Atlanticism more generally, see Pijl 1997). Hence the narrowness of the nature of the ‘common heritage and civilisation’ specified in the preamble to the North Atlantic Treaty: The Parties to this Treaty reaffirm their faith in the purposes and principles of the Charter of the United Nations and their desire to live in peace with all peoples and all governments. They are determined to safeguard the freedom, common heritage and civilisation of their peoples, founded on the principles of democracy, individual liberty and the rule of law. They seek to promote stability and well-being in the North Atlantic area. They are resolved to unite their efforts for collective defence and for the preservation of peace and security. They therefore agree to this North Atlantic Treaty …

102 Anthony Woodiwiss

Note the absence of any mention of trade unions, social partnership, the welfare state, or of any other by then already distinctively Western European civilizational elements. Second, and relatedly, American sociology did not share the European interest in class as a major determinant and instantiation of social structure but instead was far more interested in the tensions between individuals and social order as demarcated by the terms ‘modernity, ’ ‘race, ’ ‘ethnicity, ’ and ‘gender.’ Third, following the ‘inward’ turn that was part and parcel of the United States’ entry into a condition of literal postmodernity in the late 1970s, the pre-eminence of such issues was reinforced by the rise of feminist and postmodernist critiques of what they termed ‘malestream’ or ‘foundationalist’ epistemologies respectively. The net result was that the concept of identity became central to the sociological picture of Europe too, and especially of Britain. Finally and ironically, if you accept my realist reading of Michel Foucault, one of the main instruments of this reinforcement was a misreading of Foucault’s theorization of discourse. What I intend to explain, then, is not why identity groups became so important in the late twentieth century global North, but why the concept of identity was available and why it was taken up so widely. From Capitalism to Modernity In order to appreciate the specificity of a discursive formation at a particular point in time, it is helpful to compare it with its state at another point in time. Consequently, I will try to bring out the distinctiveness of the state of the discursive formation pertaining to sociology as a whole in the 1990s by comparing it with its state in the 1930s. In the 1930s sociology was still primarily a European and even countryspecific area of intellectual activity within which efforts at synthesis and writing textbooks were few and far between. The two leading national traditions were those of France (Marxist or Durkheimian) and Germany (Marxist or Weberian). The lesser traditions such as those of Britain and the United States were for complex and different reasons (Abrams 1968; Anderson 1992; Bannister 1991; Turner 1994) less theory driven and much more empirically and policy oriented. Nonetheless, it is clear that even in Britain and the United States, the social-structural/ institutional complex within which sociology was emerging was seen as that which the classical theorists wrote about, namely, that formed by the nexus of the state, the capitalist economy and liberal-democratic

Rescuing Reflexivity 103

governmental discourses of rule. Thus, Karl Marx, Max Weber, and Durkheim became the principal authorities defining the field, with the result that the central object of interest to the discipline was demarcated and pictured as something called capitalism, which, in turn and to varying degrees, was understood to be animated by class tensions and conflicts. Accordingly, the vocabularies produced by these classical theorists defined much of the expert language that had to be used if someone wished to be taken seriously when speaking about social life. Similarly, these theorists also provided much of the discipline’s conceptual infrastructure, while the differences between them, especially at the metatheoretical level, stimulated the production of different strategic or overall theorizations. This, then, was the conjuncture with respect to sociology’s discursive formation that produced what we now refer to as classical sociology. However, as Ferdinand de Saussure intimated and in ways that Foucault made explicit, discursive formations and the ways of looking at the world they produce are always changing. Fittingly, what turned out to be the first significant post-classical school of non-Marxist social theory was initiated in the United States rather than Europe but by an American who had studied in Germany, Talcott Parsons. In his Structure of Social Action (1937), Parsons produced a distinctively American reading of what he was largely responsible for isolating as classical sociology. It was distinctively American in the sense that classical theory was read not as picturing the structure of capitalist society but, on the contrary, as contributing critical elements to the construction of exactly the kind of creative individualistic theory that according to Durkheim ‘radically prevented the establishment of sociology’(Traugott 1978), namely the ‘voluntaristic theory of action.’ Thereafter social theory developed in an intellectually as well as geopolitically Atlanticist institutional context within which the American side gradually became predominant. This said, there was a period – from the late 1940s to the mid-1960s – during which American social theory in general and Parsons’s theory more particularly morphed, in the sense of transformed itself, into an apparently European-style holistic theory. This theory, however, was only apparently sociologistic since, despite the predominance of the established empiricist tradition in the United States, it was produced not so much by empirically grounded theoretical reason as by Parsons, very likely stimulated by his wartime investigations of the social structures of Germany and Japan (Parsons 1937), turning his remarkable mind away from the

104 Anthony Woodiwiss

theory of creative individualism and toward the specifically American social conditions that he thought had facilitated the recognition of the theory and rationalistically developing an abstract language to describe these conditions. Soon afterwards others gave the label ‘modernity’ to the abstract representation that Parsons had projected onto the United States. Why ‘modernity’? Startlingly, and with consequences that I have discussed at length elsewhere (Woodiwiss 1990: ch. 7; 1997; 2001, ch. 3), few of those who, whether in the United States or Europe, took up the term so readily in the 1980s had enquired into the sociological genealogy of the term ‘modernity’ when it rather surprisingly resurfaced as an after-effect of the rise of postmodernism; and certainly no one had enquired into the developing discursive formation that accounted for its availability as a key element in contemporary theoretical discourse. Rather, those who took up the concept defined it for themselves, or so they thought. Thus, they defined it, depending on the context, by identifying it with either the emergence of Cartesian epistemology, the occurrence of the first and/or second industrial revolutions, or the appearance of a creative, self-reflexive individualism. In so doing they failed to realize that this selection or, as Foucault would put it, ‘dispersion,’ of factors precisely, if incompletely, replicated the unthought or discursive formation of Atlanticist social theory, as I will now show. It is commonly agreed that the discourse of modernism emerged as an aspect of a broad and therefore very loosely defined movement within the arts that was indeed informed and affected by all the developments just mentioned. In the course of its emergence, the initially negative connotations of the word ‘modern’ were gradually reversed as modernism slowly gestated during the nineteenth century, became widely visible around the turn of the century, and, in the Englishspeaking world, belatedly conscious of itself as ‘modernism’ during the 1930s (Bradbury & McFarlane 1991). What is far less commonly recognized is that the expansion of the meaning of ‘modernity’ to include a social rather than simply an aesthetic condition was primarily an American enterprise. In the course of the 1940s, and thanks in large measure to intellectual émigrés from fascist Europe, the United States became the centre of modernist experimentation in the arts (vide the displacement of Paris by New York as the artistic capital of the world with the rise of Abstract Expressionism – http://www. Artcyclopedia; search in ‘Movements’ – as the first major post-surrealist school of painting). Consequently, a stoical, world-weary reading of the otherwise

Rescuing Reflexivity 105

generally critical and progressive social diagnosis implicit in aesthetic modernism took hold of certain sections of the United States’ cultural intelligentsia. This was a reading to the effect that, qualified only by the existence of the New Deal or ‘social modernist’ (Woodiwiss 1993: ch. 1) package of reforms (a minimalist welfare state, the licensing of responsible trade unions, and a societal commitment to equality of opportunity), knowing self-reflexiveness, competitiveness, and changefulness were all that could or should be expected in or of the modern world – to want more risked the return of concentration camps. As the 1940s came to a close and the cold war commenced, this was a diagnosis that provided a new analytical vocabulary, and therefore a convenient resting place, for such soon to be highly influential sociologists as Daniel Bell and David Riesman on their journeys from ‘socialist estrangement’ to ‘reconciliation’ with the American status quo – journeys that they explicitly justified as having been made necessary by American socialism’s failure to measure up to the demands of what they termed ‘modernity’ (Bell 1967: 5; Brick 1986: passim; Gilman 2004; Wolff 1988: 66–68; Woodiwiss 1990: ch. 7 and 1993). The Decline of Modernity and the Rise of Identity When some thirty or so years later, and in the aftermath of postmodernism, some Western European social scientists enthusiastically imported the concept of modernity into their discourse, their picture of their own societies did not change because they accepted either the American empiricist project or Parsonianism, since they did not (see the discussion of Bauman, Habermas, Hall, Giddens, and Luhmann in Woodiwiss 2005: ch. 3). Rather, their picture of European societies changed because they belatedly, and I have to say carelessly, adopted the language of Parsonianism and Modernization Theory (that is, terms such as ‘tradition,’ ‘industrialization,’ ‘differentiation,’ ‘modernity,’ and ‘actor,’ for example) and thereby unknowingly bought into the Atlanticist unthought and accepted the humanistic ontology that underpinned these two bodies of theory (Woodiwiss 2001: ch. 2). In the United States this humanistic ontology and its symptomatic individualism had returned to the foreground when the appearance of social consensus fostered by the image of America as the modern society was shattered by the economic weakness and conflicts that marked the late 1960s and early 1970s and radically reduced the profitability of American corporations. For American sociologists like Bell (1976), the

106 Anthony Woodiwiss

resulting social turmoil simply confirmed the United States’ advanced modernity since it reflected the stresses and strains associated with the increasing size of the middle class (Bell 1973): ‘the classic trajectory of expectations … tells us that no society that promises justice and … slowly begins to open the way, can then expect to ride out the consequent whirlwind in a comfortable fashion’ (Bell 1976: 179). An alternative view, confirmed by subsequent developments (Sandage 2005), is that these were years during which the United States’ claim to have attained a state of advanced modernity began to look rather hollow, since this was a period of considerable class tension, as indicated by the high levels of official and unofficial employee actions in the industrial sector of the economy’ and increasing inequality. This tension arose because this period also saw the beginning of a successful and still ongoing effort to reduce the perquisites of the middle class, with the result that although the middle class as a whole grew larger, all but its upper echelon became relatively poorer (Woodiwiss 1993, pt. 2). In addition, unhappiness about the war in Vietnam interacted with the class tensions and a related set of inter-generational tensions to produce a whole raft of initially youth-oriented middle-class social movements pursuing such goals as women’s liberation, gay liberation, and ‘participatory democracy.’ However, as will become apparent below, the emergence of what was called at the time the ‘Movement’ did not mean that the influence of Parsons’s theory and of the intellectual and social forces that produced it came to an end – far from it. By the mid-1970s, the economy was on the upturn, and relatedly the labour force was quiescent, the Vietnam War had ended, and order had been more or less restored to the streets except for what proved to be an intensifying battle between ‘police and thieves’: By all accounts, students entering higher education in the mid-1970s became much more vocationally oriented … More noticeably, the already employed, the future ‘yuppies,’ began to seek meaning in the diffused and commercialized development of Movement themes: greater drug use, but for the sake of the ‘sharpness’ that comes from cocaine rather than the ‘mind expansion’ associated with marijuana and LSD; self-awareness, but through commodified and individualized therapies rather than ‘tribal’ rituals; feminism, but through the pursuit of a pragmatic Equal Rights Amendment rather than a romantic ‘liberation’; civic activism, but in the cause of ecology rather than civil rights; in sum, a ‘new age’ rather than a ‘revolution’ was sought. (Woodiwiss 1993: 97)

Rescuing Reflexivity 107

The return of relative tranquillity saw not only an inward turn on the part of many disappointed as well as reconciled social groups and individuals but also, as part of a strategy to restore corporate profitability, the beginning of the withdrawal of the state from its role as the ultimate guarantor of the values, policies, and programmes – principally those associated with the equalization of opportunities – that had hitherto confirmed the United States’ status as the modern society for many of its citizens. As a result, the United States became the first, and probably the only, literally postmodern society – no other society’s discourse of rule had at the time either used the term ‘modern’ to define itself or so clearly equated this claim to modernity with opportunity for all. However, even those self-described conservatives who had consciously rejected the social modernism of the post–New Deal period altogether nevertheless avoided acknowledging that anything had really changed: ‘Critics say that America is a lie because its reality falls far short of its ideals. They are wrong. America is not a lie; it is a disappointment. But it can be a disappointment only because it is also a hope’ (Huntington 1982: 262). One would have expected that the tortured logic of such statements would have been taken as clear evidence of an imminent ideological crisis. In any event, no such crisis occurred. Instead, what I will call ‘official America’ responded to the trauma of discovering that capitalism could not be made to deliver opportunity for all by experiencing a prolonged fit of forgetfulness or social amnesia. This fit continues today and takes the form of an involuntary repetition of the claim that the United States is the land of opportunity, combined with an equally involuntary forgetting of the promise that had been made by all administrations since the New Deal, namely that government would actively strive to make opportunity available to the whole population. In other words, after the mid-1970s official America gave up on even trying to be in fact, as well as in name, what it had itself defined as ‘modern.’ It also therefore gave up on trying to make what it continued to insist was one of its most ‘sacred’ words – opportunity – refer to anything that might presage a significant improvement in the lives of millions of poor Americans. However, this very sacredness has meant that no mainstream politician can call the perfection of American society into question and hope to be elected. As in the absolutist state during the baroque period, awe is the proper attitude, just as ‘shock and awe’ has become the preferred mode of exercising power. The result is that, politically at least, the United States has entered the condition that Jean Baudrillard (1995) has specified as

108 Anthony Woodiwiss

‘hyperreality.’ This is a condition in which the dominant understanding of how we come to see the world as we do is one in which images are understood to produce objects in the minds of people, with the result that, from breast implants to Disneyland, simulations are particularly valued as safer than, and even an improvement on, reality. Accordingly, ‘choice’ has today replaced ‘opportunity’ at the core of official discourse: ‘choice’ to, however, merely simulates ‘opportunity,’ since, although it is spoken of as if it means the same thing, it does not; ‘choice’ is socially safer than ‘opportunity’in that the former carries no implication that anything needs to be equalized and therefore nothing need change regarding the disposition of power or resources; and ‘choice’ is better than ‘opportunity’ in that it is impossible to improve on freedom. Moreover, once political discourse becomes a matter of producing and maintaining a simulation rather than promising and proving the attainment of measurables goals, it becomes part of the aesthetic realm, and therefore checking its veracity is both impossible and inappropriate. The result is that, again as in the baroque era, politics is personalized, and therefore only aesthetic or emotional responses are possible, in that citizens are only able to say whether or not they like, or what they feel about, a particular statement or, much more often, the person making it. Individual political and social withdrawal as well as governmental retrenchment occurred in Europe too but to a lesser degree and from much higher baselines of popular political involvement and social spending and therefore without the amnesiac after-effect; even the Blair/Giddens ‘third way’ was constrained to at least present itself as a variety of social democracy and as therefore still committed to the active pursuit of social justice. Accordingly, the exchange of ‘choice’ for ‘opportunity’ still seems to be a difficult idea to sell in Europe because choice remains strongly associated with ability to pay – who would freely choose an inferior house, school, or hospital? Finally, freedom is not the ethical absolute in Europe that it is in the United States, since both the Christian Democratic and Social Democratic discourses of rule allow constraints on individual freedom in the communal interest. Consequently, and despite a determined and largely successful effort on the part of European politicians of all stripes to turn a blind eye to the pathologies intrinsic to capitalism, political discourse has not escaped into the realm of the sacred and the aesthetic and therefore remains in the realm of the profane and the disputable. Consequently, charges of hypocrisy and lying related to a

Rescuing Reflexivity 109

politician’s governmental achievements remain far more potent political weapons than any allegations of personal sexual impropriety. Irrespective of the particular conditions they faced, it was nevertheless initially very difficult for sociologists on both sides of the Atlantic to know how to respond to these developments. Neither the classical paradigm nor its Parsonian or modernist offspring possessed, to use Foucault’s vocabulary, the enunciative, conceptual, or strategic resources to enable them to grasp what was going on. In such circumstances practitioners in any discipline commonly turn to neighbouring fields of enquiry for inspiration and even concepts. In this case, and especially in the American context, thanks to its understandable prominence in so individualistic a culture, the most obvious source of inspiration for puzzled sociologists as well as disappointed individuals and groups was psychology, and the concept that was borrowed was that of identity. Beginning in the 1930s, Freudian theory and its derivatives had turned individual identity, as a topic if not as a word, into a popular preoccupation, as instanced by the many films, novels, and self-help books concerned with this theme. By the late 1970s American sociologists, notably Richard Sennett (1977) and Christopher Lasch (1978), had produced historical and sociological explanations for the inward turn among their fellow citizens, which turn they both characterized in Freudian terms as ‘narcissistic.’ Deprived of their fathers by the demands of corporate America that had turned them into a ‘lonely crowd’ (Riesman 1950) of ‘white collar’ (Mills 1951) ‘organization men’ (Whyte 1963), and left alone with their housebound mothers (Friedan 1963), the post-war baby boomers grew up without having resolved their ‘Oedipus complexes.’ The result was that as adults the baby boomers remained uncomfortably dependent on others for their sense of self. Such dark thoughts did not, however, appeal to the majority of either the disappointed or those who sought to understand them. Significantly, the conception of the person in terms of which this majority had come to think about such issues had also changed during the postwar period. Specifically, this conception had become much less associated with the resistant, almost biological thing represented by the Freudian unconscious and had instead become associated with a set of responsive and flexible cognitive faculties that could be used actively and positively to shape an identity and so supposedly choose one’s self. Parsons’s theory played a direct and highly significant role in this ‘cognitive turn.’ One of Parsons’s psychologist colleagues at Harvard, Edward Tolman, was asked to contribute to the seminal book Toward a

110 Anthony Woodiwiss

General Theory of Action (Parsons & Shils 1951), and as a result he went on to become one of the founders of cognitive psychology. Reflecting on all the learning and adjustments individuals have to undertake in preparing themselves for, as well as performing, their roles, Tolman argued that one of the most important consequences of Parsonian theory for psychology was the new emphasis that it gave to ‘thinking,’ which, significantly, he understood as a form of reflexivity: ‘The process of thinking … seems to consist in some type of internal activity which enables an actor to bring into play the consequences of given potential types of behavior without, however, actually carrying out these behaviors. And as a result of these brought-into-play consequences he modifies or reformulates or expands his behavior space and his belief-value matrix … Satisfactory psychological studies of this process are as yet … to be written’ (Tolman 1951: 357, emphasis in original). For the disappointed, their sociological sympathizers, and those anyway always ready to ‘move on,’ the optimistic and modernistic cognitive psychologies that generalized Tolman’s point about the importance of thinking were much more appealing than the old behaviouralist and Freudian psychologies (for the effect of the rise of cognitive psychology on the psychoanalytic tradition in the United States, see Rustin 2004). In the case of the disappointed as well as the reconciled, this was because the new psychologies provided intellectual and even scientific backing for the personalizing of their aspirations. In the case of the sociologists, it was also because the ground had been prepared by a burgeoning of interest during the 1960s in a long-established, but hitherto marginal, cognate school of thought, symbolic interactionism, that was not only preoccupied with the production of the self but had already spawned much work on subcultural identities. The consequences of the changed nature of the sociological unthought that resulted from the cognitive nature of the inward turn were initially made visible in social theory by the renewed attention paid to the concepts of ‘action’ and ‘agency’ over those of ‘system’ and ‘structure.’ The net result of this shift in interest has been the move toward solipsism that I regretted in my introduction. This move is most clearly and pertinently exemplified by the transformation in the conception of social structure that occurred within two still very influential bodies of theory that regard themselves as the concept’s last best hope in what Charles Lemert (2004), bizarrely, has called the ‘time after the structures disappeared.’ The first of these is Anthony Giddens’s ‘structuration theory,’ wherein the dualism of subject and object or agent and

Rescuing Reflexivity 111

structure is replaced by what is claimed to be the ‘duality of structure.’ In a way that is both distinctly at odds with much of his earlier and indeed later work and largely unexplained, Giddens begins the book in which structuration theory is set out, The Constitution of Society (1984), by embracing the key propositions of what Lemert has called the ‘new sociologies’: 1 an emphasis on the ‘active, reflexive character of human conduct’ 2 the according of a ‘fundamental role to language and to cognitive faculties in the explanation of social life’ 3 ‘the declining importance of empiricist philosophies of natural science … [which means] that a philosophy of natural science must take account of just those social phenomena in which the new schools of social theory are interested – in particular language and the interpretation of meaning.’ (all passages are from Giddens 1984: xvi) On this basis he explains what he means by the ‘duality of structure’ in the following terms: human social activities, like some self-reproducing items in nature, are recursive. That is to say, they are not brought into being by social actors but continually recreated by them via the very means whereby they express themselves as actors. In and through their activities, agents reproduce the conditions that make these activities possible (ibid.: 2). Because of his uncompromising humanism (in this context the assumption that in the end all social stuff is reducible to individual human beings and their interactions), Giddens rules out in advance, because of their invisibility, the possibility that there may be forms of social life other than, or other to, human actors or entities reducible to them. Instead, all there is to social life is ‘agents … [moving] in physical contexts whose properties interact with their capabilities … at the same time as those agents interact with one another’ (ibid.: 112). This becomes the core of Giddens’s concept of structure except that, in a way reminiscent of Parsons’s inclusion of ‘means and conditions’ within the ‘unit act,’ he specifies ‘rules and resources’ as the ‘physical contexts’ of agents. But this is also the way in which social structure disappears from Giddens’s theory and solipsism takes over – that is, social structure is reduced to rules and resources made or assembled by other or preceding actors in time-space, and it is therefore ‘out of time and space, save in its instantiations and co-ordination as memory

112 Anthony Woodiwiss

traces’(ibid.: 25). If ‘memory traces’ in the minds of individuals are all that remain of social structure, there can be no other to individual human beings that would justify any idea of the ‘duality of structure’? Consequently, since all social life is in one’s mind, and as Giddens seems to have decided practically if not quite formally, there is no need to look beyond one’s own mind when trying to understand the social. The second body of theory of particular interest concerning the fate of ‘social structure’ is ‘neo-institutionalism,’ which continues to be one of the most influential ‘formal theories’ among American sociologists. The key thesis of the neo-institutionalists is what its adherents characterize as the mirror image of the ‘actor-centrism’ of the rational choice theory (for a parallel analysis to the present one focused on the rise of the concept of ‘rational choice,’ see Amadae 2003), which appears to have captivated large numbers of economic and political theorists. This is the idea that society should be understood by sociologists as consisting not of actors but of ‘organizations’ or institutionalized ‘knowledge and culture.’ Thus the causal connections neo-institutionalism is interested in operate at the ‘collective level and are cultural in nature – they feature processes occurring within and between institutions … the people implicated are … occupants of highly institutionally constructed roles, operating more in their cultural and professional capacities – that is, as agents of the cultural system – than as generic individual “actors”’ (Jepperson 2002: 233, emphasis added). Neo-institutionalism, however, is explicitly rooted in phenomenology, and, consequently, as it has developed, its conception of ‘organizations’ has tended to reduce to ‘knowledge and culture,’ and agents have tended to reduce to actors. This is hardly surprising since neo-institutionalism’s substantive research foci and understandings are every bit as Americentric as those of Parsonianism. More specifically, neo-institutionalism bases itself on the following sustantive axioms: first, the United States is the archetypal modern society; and second, Western history is to be understood as a process of modernization or convergence on the American archetype, thanks to the diffusion of American-style economic and political organisational forms – that is, ‘the secularization and elaboration of individualism … [is] … the main theme in Western history’ (Meyer 1986: 200). Thus, today’s leading neoinstitutionalists seem to have forgotten their opposition to ‘actor centrism,’ lost interest in institutions per se, and instead become preoccupied with the ‘contemporary identity explosion’ (Frank & Meyer 2000, quoted in Jepperson 2002: 250).

Rescuing Reflexivity 113

Despite the fact that the psychologically oriented resources represented by the identity theory that fertilized the ‘new sociologies’ and their distinctive stress on reflexivity were so close at hand, their take-up is not explainable by reference to broad social, structural, and purely intellectual developments alone, not least because many sociologists on both sides of the Atlantic, and not simply adherents to the classical tradition, strongly resisted any such take-up. Overcoming this resistance required both that existing concepts and strategies be discredited and that authorities prepared to carry out such a task and push the new conceptual agenda emerged. As commonly happens in such circumstances, new intellectual spaces were created that were relatively free from the constraints of the established discursive formation. In this case, the principal spaces were feminism and the overlapping, quasi-social movements known as postmodernism and cultural studies. Great interest and excitement were aroused by the appearance of these new spaces, not least on the part of a publishing industry ever hungry for new product. The largely unintended result, as was illustrated by what happened to Giddens and the neo-institutionalists, was that many established social theorists quite suddenly, and with little by way of explanation, took up the new issues and concepts, forgot about the old ones, and thereby joined the politicians in their state of amnesia. In this case, however, what was forgotten were not only the promises of politicians but also the problems that the politicians had promised to solve. Perhaps the single most effective and memorable slogan associated with the feminist movement of the 1970s was ‘the personal is political.’ Incomplete though the feminist project remains, its achievements are nevertheless very substantial, not least within the spheres of sociology and social theory more generally, wherein women and their concerns have long since ceased to be marginal. That is, the so-called private sphere of families and intimate relations more generally is now regarded as every bit as important a focus of sociological interest as the public sphere of, say, economic and political life. Likewise, the sources of gender inequality in the public sphere and the forces that maintain this inequality are now also a major focus of sociological interest. This said, it is also clear that, for a time at least, the rise of the ‘women question’ to social and sociological prominence contributed significantly to the subversion of the sociological project as Durkheim defined it. This was because, unlike the focus on racial and ethnic identities and for reasons I do not fully understand, the focus on women’s collective identity stimulated a radical challenge to the established metatheoretical

114 Anthony Woodiwiss

positions in the form of feminist standpoint epistemology (FSE) (Harding 1986). Unintentionally perhaps, FSE provided the conduit through which the personal did not become political so much as the political became personal. Moreover, insofar as FSE was associated with an often very effective social movement, the intellectual plausibility and potential political significance of the solipsistic variant of the reflexive turn was greatly strengthened and enhanced. Like feminism, postmodernism contributed to the displacement of social structure by identity through both its substantive and methodological interests. Again, I will restrict myself to few words on the methods side. These are that a funny thing happened to the concept of discourse on its way through postmodernism to the methodology textbooks; it became part of an anti-realism. This is how Alan Bryman, summarizing many others, characterizes discourse analysis in the latest edition of his very influential textbook, Social Research Methods: • It is anti-realist: in other words, it denies that there is an external

reality awaiting a definitive portrayal by the researcher and it therefore disavows the notion that any researcher can arrive at a privileged account of the aspect of the social world being investigated. Some discourse analysts, however, adopt a stance that is closer to a realist position, but most seem to be anti-realist in orientation. • It is constructionist: in other words, the emphasis is placed on the versions of reality propounded by members of the social setting being investigated and on the fashioning of that reality through their renditions of it. More specifically, the constructionist emphasis entails a recognition that discourse entails a selection from many viable renditions and that in the process a particular depiction of reality is built up. (Bryman 2004: 360) In other words, in the hands of the postmodernists, discourse became separated from its inherently social grounding in discursive formations and thereby lost the sui generis thingness it had had for Foucault and became instead the means through which social actors project their desires onto social life while thinking they are creating it. Social structure, then, was eclipsed by identity within social theory over the course of the last two decades of the twentieth century as a result of the largely unwilled effects of the interaction between the different elements of sociology’s unthought or discursive formation that I have outlined. First, as itself an object, sociology became the product of

Rescuing Reflexivity 115

an Atlanticist institutional setting that was continuously transforming itself as the United States changed from being one of two superpowers to being the sole superpower and also, in ideological terms, changed from being a modern to being a postmodern society. Second, as a style of talking and writing, sociology became ever more culturalist, psychologistic, and reflexive. Third, the more specific concepts that both depended on and gave content to the social structure problematic were picked off one by one and replaced by those, such as culture and discourse, that created the possibility of, or gave content to, the identity problematic. And fourth, strategically, all these developments were encapsulated in an activist feminism and an academically militant postmodernism, both of which sought the total re-theorization of social life in a way that completely marginalized the study of capitalism and its pathologies. Conclusion What, then, and apart from saving realism as a distinctive metatheoretical stance, is the significance of engaging in such an exercise in realist reflexivity? One reason for doing it is that by showing that the rise of solipsistic reflexivity may be explained socially rather than mentally, it provides an exemplification of how we might escape from solipsism, reconnect theory with empirical research, and avoid repeating what I regard as the disasterous errors of theoretical judgement made over the past twenty or so years. Another reason for doing it is that it enables us to have a clearer view of what we sociologists can and cannot do and therefore of what our role and responsibilities require of us. As sociologists we cannot affect the the nature of our object of interest – at least not unless we become what Foucault terms ‘authorities of delimitation’ and so capable of affecting the grid of specification that delimits our object of study and so defines its content. But even then the problematization or not of issues within the object field is, much more important, determined by precisely the ongoing social developments that sociology aspires to understand. Where of course we can have far greater direct effect on how society is understood is in the differences we can make to styles of talk, concepts, and strategies. This, then, demarcates not only our area of freedom but also our area of responsibility and our opportunity for virtue. That is, we are obligated to make our own judgements in these areas rather than allow them to be made for us.

116 Anthony Woodiwiss

This said, the effectiveness of the exercise of any such sociological virtue is likely to be limited because of, as Foucault argues, the varying functionalities of particular discourses in the fields of non-discursive practices and desire; unfortunately, a revived interest in the pathologies of capitalism would not be politically welcome, and, as we all know, desire tends to follow novelty. The result is that one has to learn to live with the fact that, like any other virtuous practice, sociological virtue does not guarantee that anyone will take notice. Nonetheless, hope springs eternal since the rules of formation of the discursive formation never cease changing.

NOTE 1 The present paper was derived from a longer text that was subsequently published as Scoping the Social (Woodiwiss 2005) and develops the argument in much more detail.

6 More Than Straw Figures in Straw Houses: Toward a Revaluation of Critical Realism’s Conception of Post-structuralist Theory RICHARD DAY

Introduction Proponents of Bhaskarian-derived realisms1 routinely claim that one of the most compelling aspects of their approach is that it provides an alternative to both (what I would see as) ‘modern’ positivism and (what they see as) ‘postmodern’ relativism. While I am sympathetic to this goal, the way in which ‘postmodernist’ theory is handled by most realists leaves much to be desired. Too many writers, from too many traditions, have for too long been dismissing so-called postmodernism without saying precisely what this term means to them, and without providing theoretical argumentation based on close readings of texts to support their claims. That is, postmodernism is all too often built up as a straw school before being casually burnt to the ground, with its occupants silently engulfed in the flames.2 l must admit that I have no interest in defending a postmodernism of this sort. However, I am interested in heading off the common extension of this pseudo-critique to post-structuralist theorists such as Michel Foucault, Gilles Deleuze, Felix Guattari, and Jacques Lacan. All these writers, as I will show, paid close attention to questions of epistemology and ontology and developed coherent answers to them. One goal of this chapter, then, is to show that some post-structuralist philosophies of science are like those of Bhaskarian-derived realisms, in that they are not classifiable as either positivistic or relativistic. Rather, they are relational in character, in that they acknowledge both the relative dependence and independence of human subjects with respect to other structures and systems, as well as the possibility of non-subjective structures and systems being tracked and affected by human subjects.

118 Richard Day

Sensing this commonality, some realists have attempted to integrate certain aspects of post-structuralist theory into their own paradigms. Carlos Prado (2006), Frank Pearce and Tony Woodiwiss (2001), and Norman Fairclough (1992), for example, have produced readings that acknowledge certain commonalities between their positions and those of Foucault. Their work is refreshing in its commitment to immanent critique and its avoidance of the stock tropes of dismissal discussed above. However, it seems to me that the Foucault who is created in many of these readings is a little too much of realist, that is, that certain crucial differences between the two paradigms are obscured. Hence the second goal of this chapter, which is to show precisely how such post-structuralist ontologies are unlike most of their realist counterparts, in that while they subscribe to a ‘depth ontology,’ they do not posit the existence of hierarchical strata of being, choosing instead to highlight the complex ways in which strata interlock and interact. The argument will begin by examining Foucault’s understanding of relations between discourses, practices, materialities, and subjectivities, with the aim of showing that he sees these structures as mutually conditioning and inseparable, that is, in a way that is different from relativism/idealism, and from those Bhaskarian-derived realisms that posit a hierarchically stratified world of things-in-themselves. Using Tony Wilden’s dependent-hierarchy notation, I will then show how Deleuze and Guattari’s theory/method of stratoanalysis differs from the approach taken by most Bhaskarian-derived realisms, in that it focuses on the analysis of stratified, but complex and non-hierarchical, systems of structures. Finally, I will use Lacan’s theory of the subject to indicate in a concise way precisely how it can be possible that subjects themselves are examples of just these kinds of structured systems, and how they are always already simultaneously in contact with, and alienated from, what both Lacan and the realists call ‘the real.’ The Elements of Straw Postmodernism Before commencing the main argument, I will first address some of those realist texts in which a straw postmodernism is created and destroyed. I want to treat these arguments carefully and seriously so as to avoid reproducing in my own work the problem of inadequate engagement that I have decried in the work of others. I will take a loosely Foucaultian approach, by observing and commenting on certain discursive regularities that are dispersed across these texts but not necessarily recognized within them. One of the most important of

Critical Realism’s Conception of Post-structuralist Theory 119

these is the perception of postmodernism as a vague, ethereal practice of some people called postmodernists, who are themselves equally wispy – they are rarely named, and their texts are apparently not only very difficult to read but in some cases impossible even to find. In a recent collection of realist essays, Ray Monk, for example, laments the fact that ‘postmodernist literary theorists, sociologists, historians, etc.’ very rarely state their ‘absurd doctrines’ in print, and ‘still more rarely do they argue for them. Rather, they are just assumed in the things they do write’ (Monk 2004: 34, emphasis in original), and espoused in ‘seminars, conferences, and pubs’ (ibid.: 35). Monk rightly points out that academic theoretical-philosophical practice must be based on argumentation, close readings of texts, and careful attribution of ideas and arguments. If it were the case that no self-identifying postmodernists have been willing to publish reasonably argued statements of their positions, one would indeed be forced into deficient practices such as basing most of one’s argument, as Monk does, on a ‘definition’ of postmodernism written for the Routledge Encyclopedia of Philosophy. However, if there are postmodernist texts with which one might engage productively, then it would be Monk and his colleagues who would be guilty of ‘just assuming’ what postmodernism is all about. To know whether there are any such texts, though, we need to know who the postmodernists in question might be; only then would it be possible to check if they have written anything. In Monk’s article, in addition to the dictionary-entry writer, the work of the sociologist Norman K. Denzin is introduced as an ‘example of the kind of thing with which one has to deal’ these days (Monk 2004: 37). Through Denzin, ‘recent structuralist and poststructuralist developments in critical theory’ (ibid.) are indirectly criticized, and some interpretations of some of the work of Jacques Derrida are addressed in a cursory fashion. After this, the article ends. The bibliography contains a reference to Derrida’s Positions, but no other works that could be seen as part of the postmodernist/post-structuralist canon. Indeed, only seven writers are mentioned in total. Does this mean that Derrida is the only person to contribute to ‘recent structuralist and poststructuralist developments in critical theory’? Surely he is not. Others who use the label ‘postmodernist’ as a mark of Cain usually include suspects like Foucault, Luce Irigaray, Judith Butler, Jean Baudrillard, Jean-François Lyotard, Deleuze and Guattari, Homi K. Bhabha … despite, in many cases, their consistent and angry protestations to the contrary. The list is long, covers pretty much every discipline, and amounts to rather a lot of books and articles, as anyone who has tried to treat these developments seriously can attest.

120 Richard Day

The same vague but potent method of indirect dismissal is deployed in another recent realist collection, edited by José López and Garry Potter (2001). In the introduction to the volume, the editors repeatedly dismiss postmodernism but do not mention a single author until page 8, where ‘Wittgenstein,’ ‘poststructuralism,’ and ‘the hermeneutic or interpretivist tradition’ are all loosely identified as exemplars. It is not only the most recent realist texts that deploy this trope. As Andrew Collier’s 1994 reference to ‘the sinister connotations [the slogan “knowledge is power”] has acquired in the covens of poststructuralism’ (1994: 12) shows, the trope is well established within the ranks of Bhaskarian-derived realism. Indeed, Roy Bhaskar himself, in his earliest work, was careful to distinguish his position from that of ‘the hermeneutical tradition,’ which he saw as too easily lapsing into ‘relativism’ and ‘idealism’ (1979: 202). At this point, Bhaskar did not make postmodernism a target of critique. In his later work, however, he did address ‘the poststructuralism of Derrida, Foucault, and Lacan’ (1994: 52), with what might be seen as characteristically idiosyncratic results. Derrida is repeatedly admonished for his aphoristic claim that there is nothing outside the text, which Bhaskar seems to take literally (1994: 15; 200). Foucault comes in for better – if not more thorough – treatment because of his theorization of ‘counter-conduct’ (Bhaskar 1994: 215). About Lacan we hear nothing more, so it is not possible to know what Bhaskar thinks about his brand of post-structuralism. I could go on, but I hope this discussion will be seen as adequate to my purpose, which is to show that many Bhaskarian-derived realists have had a tendency to reject what they see as postmodernism without any real engagement with representative writers or texts, and then to dismiss post-structuralism through guilt by association. Because of this lack of engagement, the results suffer from what Andrew Sayer calls a lack of ‘due care’ in making abstractions: ‘Abstractions should distinguish incidental from essential characteristics. They should neither divide the indivisible nor lump together the divisible and the heterogeneous’ (Sayer 1992: 88). Discourse, Practices, Subjectivity, and Materiality: Was Foucault a Latent Realist? Of course, some of those who work within the Bhaskarian-derived traditions are more forthcoming in naming their enemies and more diligent in backing up their claims. Andrew Sayer’s Realism and Social

Critical Realism’s Conception of Post-structuralist Theory 121

Science (2000) promises to be a text of this sort, in that an entire section of the book is devoted to ‘postmodern-realist encounters.’ Although the usual tropes of dismissal are deployed,3 Sayer also engages directly and in some depth with the work of some post-structuralist theorists, particularly that of Foucault. In a section on the linguistic turn, for example, Sayer claims that while this development has ‘alerted us to the constitutive role of language and textuality in understanding, it is strikingly ignorant of our ability to do things without language … Practice is conspicuously absent from postmodernist critiques of reference and representation’ (2000: 71). In a previous discussion of the notion of regimes of truth, Sayer acknowledges that ‘there is an important lesson in Foucault’s work regarding the way in which it is possible for particular discourses not merely to describe people and situations externally but to “construct” or at least mould them as particular kinds of subjects’ (2000: 44). This seems to me to be an acknowledgement that Foucault – whom Sayer associates with postmodernism and the linguistic turn – did in fact theorize practices, for it is only by way of practices that any construction or moulding of subjects could become visible and effective. To understand Sayer’s position, it would perhaps be helpful to deflate the language somewhat, to assume that what he probably means is not that postmodernists don’t theorize practice at all but that they do so in ways he finds inadequate. This interpretation is borne out by Sayer’s charge that Foucault conflates discourse and practices (2000: 45), and by his call for making a strong distinction ‘between discourses and their effects’ (ibid.). Carlos Prado has also read Foucault from a realist perspective, and also identifies him as ‘postmodernist’ and a ‘relativist’ (2006: 19). But his reading differs from Sayer’s in that he also sees in Foucault elements of what he calls a ‘tacit’ realism (2006: 96 ff.), and he acknowledges that, at least in some of his texts, Foucault does not in fact conflate discourse and practices. While Prado does see a ‘linguistic determinism’ in the early, archaeological Foucault, for whom he says ‘discourse is deemed to shape practice,’ in the later, genealogical Foucault, Prado argues, ‘discourse and practice are deemed to shape one another’ (ibid.: 91). This is definitely an advance over Sayer’s position, in that Prado here recognizes one of the key elements of Foucault’s ontology, i.e., his insistence on the mutual conditioning of discourses and practices. However, there are difficulties in Prado’s interpretation of Foucault’s epistemology. Prado is on solid ground when he notes that Foucault rejected correspondence theories of truth.

122 Richard Day

However, he takes a step in a dubious direction when he declares that ‘if truth is not mimetic with respect to objective or extralinguistic reality, then truth relates to nothing outside of discourse’ (ibid.: 81). This is a non sequitur, at least in the context of Foucault’s work, for truth can indeed ‘relate’ to something ‘outside’ of ‘discourse’ – that is, something not identical or reducible to the symbolic – if we break down the inside/ outside distinction by thinking in terms of interpenetrating rather than strictly dichotomous structures, and if we understand that for Foucault discourse is not reducible to language or the symbolic. Or, to use Prado’s terms, truth can be both ‘inside’ and ‘outside’ of discourse, if we understand the relation between truth and extralinguistic reality as one of mutual conditioning rather than of mimesis. In their article ‘Reading Foucault as a Realist,’ Frank Pearce and Tony Woodiwiss also discuss the relation between discourse and practices in various periods of Foucault’s work but produce a reading that is quite different from both Sayer’s and Prado’s. Where Sayer claims that Foucault conflates discourse and practice throughout his work, and Prado claims that the early Foucault suffers from linguistic idealism while the later Foucault does not, Pearce and Woodiwiss note the ‘satisfying’ way in which Foucault, in the Archaeology of Knowledge, refers to ‘the function that … discourse … must carry out in a field of non-discursive practices’ (Pearce & Woodiwiss 2001: 58–59). The reference to the ‘non-discursive’ is attractive to Pearce and Woodiwiss, it would seem, in that it appears to them to refer to a realm that is ‘external to our minds, subsisting independently of them in the wider realm of an equally external sociality’ (2001: 61).4 That is, it allows them to read the ‘early’ Foucault as more realist than the later. This is not to say, however, that Pearce and Woodiwiss see no elements of realism in the later Foucault. While they criticize some of the later work for its idealism, they also state that ‘the organization and content of the genealogical studies … are clearly structured in terms of the elements of the discursive formation and its conditions of existence’ (2001: 60). Here again it would seem that the reference to ‘conditions of existence’ represents for Pearce and Woodiwiss another indication that Foucault understood the extra- or non-discursive in a realist frame. What is the status of the ‘non-discursive’ in Foucault’s early work? This is a difficult question, since on close analysis Foucault deploys this term in apparently contradictory and confusing ways. In the Archaeology there are indeed passages where it is used in the way that Pearce and Woodiwiss contend it should be read, that is, as referring to

Critical Realism’s Conception of Post-structuralist Theory 123

a field of practices. In the passage they cite, Foucault provides several examples of how the discursive and the non-discursive interact. ‘Thus General Grammar played a role in pedagogic practice; … the Analysis of Wealth played a role not only in the political and economic decisions of governments, but in the … daily practice of emergent capitalism’ (Foucault 1972a: 68). It is helpful to read further, though, for it is not only practices to which Foucault refers when speaking of the nondiscursive, but ‘a whole non-discursive field of practices, appropriations, interests, and desires’ (ibid.: 69). This passage is important for two reasons. First, it specifies that the non-discursive is composed of something more than, something other than, practices; second, it is clear that all these non-discursive elements are heavily laden with the symbolic. That is perhaps why Foucault is careful to specify again and again that the ‘systems of formation’ of discursive regularities are ‘not static forms that are imposed on discourse from the outside’ (ibid.: 73). ‘These systems – I repeat – reside in discourse itself; … or rather, on its frontier’ (ibid.: 74). This is why Foucault says that what he calls the ‘preterminal regularities’ of a discourse ‘can certainly be qualified as ‘prediscursive,’ but only if one admits that this prediscursive is still discursive … one remains within the dimension of discourse’ (ibid.: 76). For Foucault, then, discourses are complex systems of interlocking and mutually conditioning practices, including practices of symbolization, of language. These practices are what ‘produce’ the conditions of the discourse itself, and of other, related, discourses. They also produce the subject positions that are made available within a discourse. ‘I wanted not to exclude the problem of the subject, but to define the positions and functions that the subject could occupy in the diversity of discourse’ (ibid.: 200). Clearly, for Foucault, it is not the case that ‘anything goes’ at the level of the subject, of what the subject might ‘know’ or ‘be’ – rather, discourses make available a diverse, but limited, set of subject positions that condition the possibilities of subjectivity. The same is true of objects: ‘We must no longer treat discourse as groups of signs [which represent an external reality] … but as practices that systematically form the objects of which they speak’ (ibid.: 49). Once again, one cannot posit the existence of just any object and achieve a truth effect. For Foucault, then, a discourse is a set of complexly linked and recursive practices that produce subjects, objects, regularities, desires, institutions … and of course practices themselves, hence discourses themselves. There is thus no point in seeking the ‘origin’ of a discourse, either in the sense of a set of logical conditions that

124 Richard Day

function a posteriori, or in the sense of a historical teleology. (Here we can see how the key insights of genealogy are in fact present in the archaeological Foucault.) But what of the question of hierarchy? How are these complex levels arranged with respect to one another? Foucault does refer, at times, to a ‘vertical system of dependencies’ that can structure a discourse, a ‘hierarchy of relations’ in which some levels are ‘authorized by anterior levels’ (ibid.: 73). He gives the example of natural history, in which ‘certain modalities of enunciation are excluded (for example the decipherment of signs), others are implied (for example description according to a particular code)’ (ibid.). However, Foucault does not posit a unidirectional hierarchy consisting of independent levels. He is careful to point out, on the contrary, that ‘relations are also established in a reverse direction. The lower levels are not independent of those above them’ (ibid.). This would seem to be a very explicit statement intended to ward off the interpretation that Foucault is positing the existence of a level that is external to our minds, subsisting independently of them. Indeed, he wrote, in the same text: ‘In the descriptions for which I have attempted to provide a theory, there can be no question of interpreting discourse with a view to writing a history of the referent … What we are concerned with here is not to neutralize discourse, to make it a sign of something else, and to pierce through its density in order to reach what remains silently anterior to it … What, in short, we wish to do is to dispense with “things”’ (1972a: 47, emphasis added). What then, are we to make of these various realist readings of Foucault? If we take them all seriously, we must conclude that both the ‘early’ and the ‘late’ Foucault contain elements of both realism and what realists see as idealism. This, of course, blurs the distinction between archaeology and genealogy on which the notion of an early versus a later Foucault depends. So one might say that he was both a realist and an idealist throughout his career. I would propose a different reading, however. Since Foucault neither separated the discursive from the material in the way that some Bhaskarian-derived realisms do, nor took up the straw postmodernist position of denying the existence of anything other than signification, it might be best to acknowledge that Foucault was not both a realist and an idealist, he was neither a realist nor an idealist. Rather, as I have noted, he was a discourse theorist who consistently saw discourses, practices, subjectivities, and materialities as mutually implicated and implicating – none of them ‘above’ or ‘below’ the others, none of them serving as a ‘ground’ or ‘referent’ for the others.

Critical Realism’s Conception of Post-structuralist Theory 125

There is, in fact, at least one realist who reads Foucault in a way that is quite close to this. Norman Fairclough notes that for Foucault, ‘discourse analysis is not to be equated with linguistic analysis, nor discourse with language’ (1992: 40). Fairclough’s ‘three dimensional concept of discourse’ allows him to analyse ‘any discursive event’ as ‘simultaneously a piece of text, an instance of discursive practice, and an instance of social practice’ (ibid.: 4). With regard to social ontology, Fairclough notes that ‘discourses do not just reflect or represent social entities and relations, they construct or ‘constitute’ them … in different ways, and position people in different ways as social subjects (ibid.: 3–4). Perhaps surprisingly for a realist, Fairclough also acknowledges that Foucault’s conception of the ‘objects’ that are constituted by discourse includes ‘the entities recognized in ordinary life’ (ibid.: 41). This, perhaps, opens the way to a realist understanding of a characteristic of a post-structuralist ontology that can also be found – with its own particularities – in the work of Lacan and Deleuze and Guattari, writers with whom, to my knowledge, Bhaskarian-derived realism has yet to engage. Before turning to them, however, it will be necessary to discuss a key realist concept, perhaps the key realist concept: I am referring here to stratification. Stratification in Bhaskarian Realism According to Roy Bhaskar, ‘the differentiation of the world implies its stratification, if it is to be a possible object of knowledge for us’ (1975: 19). Although there are several ways in which Bhaskar assumes ‘the world’ to be stratified, I will focus for the moment on its division into the domains of the real, actual, and empirical. According to Bhaskar’s original formulation, the domain of the real consists of ‘generative mechanisms’ that give rise to events in the domain of the actual, which may sometimes be experienced by human beings in the domain of the empirical (Bhaskar 1975: 13). Interestingly, Bhaskar says that while each of these domains is ‘distinct’ (ibid.), he also says that the domain of the real is ‘greater than,’ that is, in some way encompasses and exceeds, both the real and the actual; in a similar way, the actual encompasses and exceeds the empirical (ibid.: 228). This kind of apparently contradictory relationship can be rendered diagrammatically using what Anthony Wilden, following Gregory Bateson, calls a dependent hierarchy notation. Wilden uses the symbol to refer to a ‘level of complexity,’ and places these below one another to illustrate a relation of enabling, structuring

126 Richard Day

constraint across a permeable boundary. A concrete example will perhaps be helpful in understanding this notation: inorganic nature organic nature society

In this diagram, inorganic nature – quarks, atoms, molecules – is represented as providing a necessary basis for the emergence of organic nature: RNA, DNA, cells, organisms. Thus, it would be said that inorganic nature provides an environment for organic nature, just as we say that organic nature provides an environment for human society. Similarly, organic nature provides an environment for human social organization – if there were no human animals, there would be no human societies. It is important to note that the boundaries between levels are defined as permeable, that matter/energy and information flow across them in both directions – that is, the levels denote open systems.

It is also important to note that for Wilden, as we move from the top to the bottom of the diagram we are moving from orders of lesser to greater complexity. This hierarchical ordering has further implications that are summarized in what Wilden calls the Extinction Rule. ‘To test for the orientation of a dependent hierarchy,’ he writes, ‘mentally abolish each level in turn, and note which other levels will necessarily become extinct if it becomes extinct’ (1980: 4). Using this rule, one can tell if one has things the right way up, and it seems to apply nicely here: who would argue, given the evidence of the effects of global warming, for example, that human society can survive without nature? Or that organic life could do without proteins, nucleotides, and quarks? Wilden thus makes a strong realist claim: ‘Nature …

Critical Realism’s Conception of Post-structuralist Theory 127

belongs at the top of this dependent hierarchy, and its position there is the result of necessity, not of theory’ (1980: 4). Using Wilden’s method to represent Bhaskar’s ontological schema, we arrive at the following diagram:

the real (generative mechanisms) the actual (events) the empirical (experiences)

With this method of representation, we can see both how human experience of actual events depends on their production by generative mechanisms, and how the realm of the empirical is contained by, or better, is dependent on and constrained by, the actual and the real. Further, we have a nice illustration of the realist insistence that if human beings were to disappear – that is, if the realm of the empirical were to cease to exist – the generative mechanisms of the real would go on producing actual events. The commitment to a hierarchical ‘depth ontology’ is, as Andrew Sayer notes, one of the ‘characteristic claims of realism’ (1992: 5). Andrew Collier, for example, echoes both Bhaskar and Wilden in the following passage: ‘Mechanisms are, so to speak, layers of nature, and are ordered, not just jumbled up together … It appears that the material universe existed before there was organic life, and that living organisms can only exist as composed of and surrounded by matter. In this sense, matter may be said to be more ‘basic’ than life; life in turn may be said to be more basic than rationality (in the sense that we are rational animals), and hence than human society and its history’ (1994: 46, emphasis in original). There is, however, variation in realist accounts of stratification, and in some of these variants the lines between the real, actual, and empirical are blurred. In Sayer’s own work, for example, Bhaskar’s division of ‘the world’ into the levels of the real, actual, and empirical is partially reproduced through a vertical arrangement in which structures are said to give rise to mechanisms, which in turn

128 Richard Day

may generate events (1992: 117). Again using Wilden’s notation for clarity, Sayer’s model would be represented as follows:

structures mechanisms events

For both the early Bhaskar and Sayer, it would seem, events occur at a level that is discrete from the structured mechanisms that generate them. However, Sayer’s model puts structures and mechanisms at separate levels, whereas in Bhaskarian realism they coexist (and perhaps are the same sorts of ‘things’?). The most interesting difference between these two models, though, becomes evident in the following passage: ‘In the vertical dimension, some readers may want to add a fourth level above events [below, in the diagrams I am using] to cover meanings, experiences, beliefs, and so forth, but as these can form structures, function as causes, or be considered as events, I would suggest that they be taken as already included’ (1992: 116–117, emphasis added). The difficulty that arises as Sayer tries to fit what amounts to the discursive into his hierarchical model is telling. It would seem that the level of structures contains not only structures but meanings, experiences, and beliefs as well. What sense, then, is there in suggesting that it is a discrete level? The same goes for the level of mechanisms and, one might argue, for any level one might postulate. As soon as we acknowledge that there is no strict separation between structures on a single level, or between levels, that is, as soon as we take seriously the concept of open systems, the hierarchical nature of ‘the world’ starts to unravel, and it becomes apparent that these domains are, to some extent, integrated rather than independent. This kind of integration is in fact visible in Sayer’s text, through the theorization of what he calls a ‘horizontal dimension’ in which there are ‘a variety of structures, mechanisms, and events present in a complex system (1992: 116). In this model, there is no hierarchy: everything exists on a single plane. But still there are discrete entities, emergent properties, and so on. Indeed, as Sayer notes, emergence need not imply hierarchy. It can readily be explained ‘in terms of the distinction

Critical Realism’s Conception of Post-structuralist Theory 129

between internal and external relations’ (ibid.: 119) on one level or across non-hierarchically arranged levels. Indeed, a complex, nonstratified model appears to allow for everything realists need from an ontology, which implies that Bhaskar might not have been entirely correct: the differentiation of the world does not necessarily imply its stratification – it merely implies its differentiation. Sayer, however, is not so quick to let go of the stratified model. At one point in his text he acknowledges the existence of a ‘web ontology’ that informs the work of ‘some poststructuralists.’ This is an excellent insight, but Sayer misses the chance to take it further. He writes, ‘Interesting though the centreless web ontology is, it seems dogmatic and perverse to deny that there are not some structures that are extensive or widely replicated, such as those of patriarchy and capitalism’ (ibid.: 72). To deny the existence of ‘extensive or widely replicated’ social structures would indeed be dogmatic and perverse, and that is why we find in the work of Foucault concepts such as governmentality (1978/ 1991), biopower (1979/2003b), and racism (2003). It is why Deleuze and Guattari offered an in-depth analysis of heteronormativity in the Anti-Oedipus (1983), and of capitalism and the State form (and so much more) in A Thousand Plateaus (1987). It is why Derrida wrote Specters of Marx (1994) and The Other Heading (1992) – anyone who is familiar with the theoretical and activist work of the main figures of poststructuralist theory knows that they did not deny the existence of structured relations of power but, rather, focused almost all their attention on these relations. Why, then, the constant misunderstanding? Perhaps because post-structuralist thinkers believe it is possible for structures to be extensive and widely replicated, and to possess a degree of intransitivity, without their being independent and arranged into a strict hierarchy. To understand how this might be the case, I want to turn now to a discussion of Deleuze and Guattari’s method of stratoanalysis. Deleuze and Guattari on the Strata Feminists, radical-democratic political theorists, anarchists, and various practitioners of cultural studies have focused our attention on those aspects of Deleuze and Guattari’s work that involve ‘free intensities,’ ‘singularities,’ ‘flows,’ and ‘nomadic subjectivities.’ Rolando Perez, for example, in an early effort at working through links between schizoanalysis and what he calls an(archy), has argued that a poststructuralist anarchist politics must aim for ‘the replacement of poor

130 Richard Day

defenseless, guilt-ridden puppets in internal straight-jackets, with free, non-Oedipalized, uncoded individuals’ (1990: 28). This sounds very attractive, but given that ‘the individual’ is a discursive construct associated historically with Western bourgeois liberalism and its conception of the family, how could an ‘individual’ exist in a ‘non-Oedipalized, uncoded’ form? Indeed, how could any subject do as Perez suggests, and ‘destroy his or her own form of expression immediately, so as to make repetition and incorporation impossible?’ (1990: 57). Although Perez relies heavily on the work of Deleuze and Guattari, he seems to forget that repetition (in the Deleuzean sense) is unavoidable, and incorporation is always a danger to be warded off. When thinking about the possibilities of nomadic subjectivity, the final sentence of A Thousand Plateaus should always be kept in mind: ‘Never believe that a smooth space will suffice to save us’ (Deleuze & Guattari 1987: 500). We must be aware of the fact that flows are constantly being directed, singularities are being enveloped by molar structures, and nomads are adopting sedentary lifestyles. These tendencies are the result of specific processes of territorialization, capture, and, most important for the current discussion, stratification. In ‘10,000 B.C.: The Geology of Morals,’ Deleuze and Guattari provide a theory of stratification that bears a strong resemblance to realist accounts, in its attempt to develop an ‘axiomatics’ that deals with three ‘major types of strata’: the geological/chemical, the organic, and the linguistic (1987: 64).5 As in Wilden’s theorization of levels of complexity and Bhaskar’s conception of domains, Deleuze and Guattari understand strata as ‘layers’ or ‘belts’ that occur ‘at least in pairs, one serving as a substratum for the other’ (ibid.: 40). Particular strata are seen as displaying a certain sort of unity, but as always in their thought this unity is constantly in flux with multiplicity: ‘What accounts for the unity and diversity of a stratum? … There is no vital matter specific to the organic stratum [for example], matter is the same on all the strata. But the organic stratum does have a specific unity of composition, a single abstract Animal, a single machine embedded in the stratum, and presents everywhere the same molecular materials, the same elements or anatomical components of organs, the same formal connections’ (ibid.: 45, emphasis added). Unity of composition means that strata are capable of producing infinite variety within a context of limitation or constraint. It is in their theorization of this variety that Deleuze and Guattari make important advances over hierarchical ontologies.

Critical Realism’s Conception of Post-structuralist Theory 131

First, Deleuze and Guattari note that all strata are ‘doubly articulated’ in a number of different ways. For example, on the strata ‘interior and exterior exchange places … both are interior … the limit between them is the membrane that regulates the exchanges and transformations in organization’ (ibid.: 50). (Here the resonance with Foucault’s attention to the margins and limits of discourses is strong.) If we think only of ‘the interior’ for a moment, phenomena of double articulation can also be observed in the relations between substances and forms (ibid.: 40–1), and the ways in which time plays into, and in a sense becomes, space/structure: ‘In a geological stratum, for example, the first articulation is the process of “sedimentation,” which deposits units of cyclic sediment according to a statistical order … The second articulation is the “folding” that sets up a stable foundational structure and effects the passage from sediment to sedimentary rock’ (ibid.: 41). Internal complexity is also evident in the recursivity or self-similarity of strata, which can be seen as another kind of double articulation. ‘Each stratum is double (it itself has several layers)’ (ibid.: 40). Here the term ‘double’ must be taken as itself recursive, as implying doublings on doublings, to produce what Deleuze and Guattari call epi- and parastrata. These are mediating structures that occur in between other structures, outgrowths and offshoots that multiply and obscure sedimented relationships. ‘A stratum exists only in its epistrata and parastrata,’ Deleuze and Guattari declare, ‘so that in the final analysis these must be considered strata in their own right’ (ibid.: 52). In Deleuze and Guattari’s texts, then, terms like ‘level’ or ‘domain’ or ‘stratum’ have only a heuristic value and must be revoked as soon as they are invoked so that we do not obscure the immense complexity and internal differentiation that characterize any structure. Now that we have an understanding of the complexity of the strata as theorized by Deleuze and Guattari, it should be apparent that any charge of linguistic idealism that might be laid against them is misdirected, for linguistic idealism assumes that ‘everything’ can be comprehended as part of a single, undifferentiated stratum – that is, using just the sort of model that they so adamantly refuse. It is not necessary to rely only on this negative argument, however. In many passages Deleuze and Guattari directly address the question of whether signification occurs at all levels, and their answer is quite clear: it does not. Only on the linguistic level does what they call ‘translation’ occur, which is defined not only as the ability of one language to (more or less adequately) represent what is given in another but also as ‘the ability of

132 Richard Day

language, with its own givens on its own stratum, to represent all the other strata and thus achieve a scientific conception of the world’ (ibid.: 62). Here it must be noted that the representative function of language is explicitly acknowledged. At the same time, however, and in their usual way, Deleuze and Guattari are careful to avoid the assumption that because representation is possible, everything is representation. In a discussion of matter, content, expression, form, and substance, for example, they warn that ‘the net,’ or the complex of relations between these terms, ‘is not linguistic in scope or origin,’ since ‘language has a specificity of its own’ (ibid.: 43) that must always be respected. It is also important to note that unlike the straw postmodernists, they explicitly acknowledge the existence of what realists call the intransitive aspect of our interactions with the material world. This is visible, for example, in their discussion of the ‘nomadic’ sciences ‘that consist in following a flow in a vectorial field across which singularities are scattered like so many “accidents” (problems)’ (ibid.: 372). Picking up a rock that works as tool or discovering that food packed in snow does not go bad as quickly would be examples of nomad science in action.6 The more important point, though, is that the practice of tracking singularities clearly assumes the existence of ‘some thing(s) out there’ to be tracked, some thing(s) over which we don’t have total control, about which we don’t have total knowledge. Thus it can be seen that Deleuze and Guattari do not reject science as such. Rather, they carry out an analysis of the possible modes of science, with a focus on differentiating between nomadic science, as described above, and the royal sciences, which operate in the service of the states and corporations and have as their goal the domination of nature rather than the tracking of its particularities. As is so often the case, a careful reading of their texts reveals that both their postmodernist celebrators and their anti-postmodernist detractors are seeing only one side of a doubly articulated system: ‘What we have … are two formally different conceptions of science, and, ontologically, a single field of interaction in which royal science continually appropriates the contents of vague or nomad science while nomad science continually cuts the contents of royal science loose. At the limit, all that counts is the constantly shifting border line’ (ibid.: 367). Despite these similarities and points of compatibility, however, there are crucial differences between stratoanalysis and the hierarchical systems envisaged by realist ontologies. As I have shown, for Sayer and Bhaskar, levels or domains of complexity are always arranged in hierarchies, with one emerging out of the other, always in the same order, and with increasing complexity each time a boundary is crossed.

Critical Realism’s Conception of Post-structuralist Theory 133

For Deleuze and Guattari, this would be an expression of what they call ‘the illusion constitutive of man’ (ibid.: 63), which can be produced only at the level of language: ‘It is difficult to elucidate the system of the strata without seeming to introduce a kind of cosmic or even spiritual evolution from one to the other, as if they were arranged in stages and ascended degrees of perfection … If one begins by considering the strata in themselves, it cannot be said that one is less organized than another’ (ibid.: 69). Not only does the notion of increasing complexity go down, the Extinction Rule is rejected as well. ‘There is no fixed order,’ Deleuze and Guattari declare; ‘one stratum can serve directly as a substratum for another without the intermediaries one would expect there to be from the standpoint of stages and degrees … you can’t even tell in advance which stratum is going to communicate with which other, in what direction’ (ibid.). Thus, while Deleuze and Guattari agree that there is stratification in the real, they see any attempt to reduce its complexity to a hierarchical ordering as a human imposition. Deleuze and Guattari’s ontology can, then, be seen as compatible, in certain ways, with that of realism. They do work with a concept of ‘depth’ but reject any notion of a hierarchical stratification. One might say that what appears as ‘depth’ in realism appears as ‘complexity’ in post-structuralism. By this I mean the following: when observing that structure A has an effect on structure B, one can posit that A is ‘deeper’ in a hierarchy than B. Alternatively, one can posit that A and B are enmeshed in a complex set of relationships, one of which is that A has this effect on B. Thus, Deleuze and Guattari’s world, though similar to that of Wilden and Bhaskar in displaying relatively discrete levels with a certain unity of composition, is a much messier, livelier place. While the strata are reliant on processes of territorialization, or the formation of unities of composition, they are also constantly subject to deterritorialization, to intrusions that shake, fold, bend, and break the unities into relatively autonomous fragments. From where do these intrusions come? To answer this question, we need to pay greater attention, or perhaps a different kind of attention, to what Deleuze and Guattari call the ‘plane of consistency’ or ‘body without organs’ (BwO). Where Wilden’s levels of complexity rest on a stable, unquestioned foundation of ‘inorganic nature,’ and Bhaskarian-derived realisms stand firmly on the ‘generative mechanisms’ of the real, Deleuze and Guattari’s strata emerge out of, and subside into, a nameless void. They write, ‘The principal strata binding human beings are the organism, signifiance, and interpretation, and subjectification and subjection. These strata together are what separates us from the plane of

134 Richard Day

‘Stratum x’ ‘Stratum y’

PLANE OF CONSISTENCY

consistency and the abstract machine, where there is no longer any regime of signs, where the line of flight effectuates its own potential positivity and deterritorialization its absolute power’ (ibid.: 134, emphasis added). At the ‘centre’ of each stratum is a single abstract machine of capture/composition, yes, but Deleuze and Guattari insist that ‘you can’t reach the centre,’ that ‘it would be a mistake to believer that it is possible to isolate this unitary, central layer of the stratum, or to grasp it in itself, by regression’ (1987: 50). If you try, you will find nothing, or everything. That is, you will find yourself on the plane of consistency, where such distinctions are simply not possible. To render Deleuze and Guattari’s description diagramatically would require something like an n-dimensional network of what Pierre Bourdieu (1977a) called ‘structured structuring structures,’7 emerging out of a primordial, omnipresent, and eternal ‘soup’ of ‘unformed, unstable matters, flows in all directions, free intensities or nomadic singularities, mad or transitory particles’ (Deleuze & Guattari 1987: 40). However, it is possible to give some indication of how a ‘typical’ set of strata might look so as to highlight their ordered and disordered qualities, as well as their emergence out of, and subsidence into, the plane of consistency. It is, of course, impossible to represent the plane of consistency – here signified by dots – since it consists precisely of that which

Critical Realism’s Conception of Post-structuralist Theory 135

has not been captured. This plane, however, is not a ‘mere jumble.’ Rather, it contains an infinity of possibilities, a(n) (im)possible home for everything and everyone. A beckoning yet at the same time what Jacques Lacan would call a lack. Lack and the Lacanian Subject Although there are clearly incommensurabilities in their thought,8 Lacan’s ontology, like those of Deleuze and Guattari and Bhaskar, includes a realm that can never be known or controlled by human subjects. This realm he names – surely coincidentally – the Real. In it Lacan says we would discern, if we could discern anything at all, ‘things originally confused in the hic et nunc of the all-in-the-process-ofbecoming’ (Lacan 1968: 39). The Lacanian Real, then, is an unknowable stratum of non-being out of which all beings emerge. Let us once again use Wilden’s diagrams to represent it, thus:

Real (R) (hic et nunc)

Lacan posits two further levels of complexity. The first, developmentally speaking, is the Imaginary, which arises as the human subject (more precisely, the moi) passes through the mirror stage, thus gaining an awareness of itself as an object separate from other objects, a subject separate from other subjects. The Imaginary is a realm of images, the unconscious, binary oppositions. While it can be argued that the Imaginary partakes of the Symbolic, it is necessary to postulate it as a relatively discreet level in order to account for the possibility of a depth analysis of the psyche.

Real (R) (hic et nunc) Imaginary (I) moi, objet a

The next level is the Symbolic, out of which the speaking subject, or je, emerges, as a result of the passage through the Oedipal stage.

136 Richard Day

This is where signification occurs, keeping in mind, of course, that for Lacan the signified is itself composed of a network of signifiers and the category of the referent is approached only through the theory of fantasy.

Real (R) hic et nunc Imaginary (I) moi, objet a Symbolic (S) je, Autre

It is, of course, because of the denial not only of the referent but the signified as well that the Lacanian subject is generally thought to be a purely linguistic entity, a pure relativism, a ‘nothing.’ Lacan maintained, however, that the speaking subject retains access to the Real and the Imaginary. This is apparent, as I have already noted, in his assumption that something like psychoanalysis is possible – the whole point of his practice, he said, was ‘achieve the dealienation of the subject’ through understanding ‘the relationship of the subject’s moi to the je of his discourse’ (Lacan 1968: 68). And, as Slavoj Zizek has noted, in the Encore seminar Lacan works with a notion of the sign as not entirely contingent but as ‘preserving the continuity with the real’ (Zizek 1991: 39). Given the two-way communication across levels that is assumed in the dependent hierarchy diagram, the immediate rendering of the Real is explicable as the passing of information through a series of levels. But this is not how Lacan explained it. Like Deleuze and Guattari, he acknowledged the depth of the Real but did not assume its ordered stratification. That is, although the various levels of the subject might emerge in an order that is developmentally conditioned, they do not form strata of a strict hierarchical sort. We can see this in Lacan’s famous diagram of the emergence of the symptom, or subject, which is based on a Borromean knot. To arrive at this diagram from the dependent-hierarchy representation requires only a small amount of manipulation.

Critical Realism’s Conception of Post-structuralist Theory 137

First we close the brackets into circles … Body of the Real-Lamella (R) moi – objects (a) (I) je – representations – Big Other (S)

Then interlock them, and entwine them with the subject (I)

Σ

(R)

(S)

To arrive at Lacan’s diagram of the subject, or symptom: I Σ

R

S

138 Richard Day

During his trip to the United States in the mid-1970s, Lacan apparently spent hours before each lecture drawing this diagram on the blackboard in painstaking detail (Turkle 1978: 235). The key to understanding this figure is to note that, on one hand, none of the loops is ‘fundamental’ – each is entwined with the others. But at the same time, each is a necessary condition of the maintenance of the structure itself, in that undoing any of the knots undoes the whole. It seems clear that Lacan wanted to convey a sense of the mutual implication of the various levels he theorized, and the subject’s emergence out of, and potential access to, all of them. He wanted to make it clear that the subject as he conceived it was not a purely linguistic entity but a complex, multi-leveled emergent phenomenon, a structured structuring structure with dependencies but without stable lines of determination. A subject interacting not only with relatively discreet ‘internal’ structures (‘the psyche’), but with relatively discreet ‘external’ structures as well (‘the material world’). His rendering of the knotty subject thus brings together, in a compact and precise form, all the elements that are characteristic of a post-structuralist ontology: discursive materialism, complexity, and non-hierarchical stratification. It also signifies the key elements of a post-structuralist epistemology, in which subjects, objects, and relations between them (‘knowledge’) are seen as complex, mutually conditioning, and relatively intransitive. Conclusion I am aware that this analysis has moved far too quickly and has therefore skipped past many issues that would need to be addressed in a proper encounter between post-structuralism(s) and Bhaskarianderived realism(s). However, given the sorry state of relations between realism(s) and post-structuralism(s), I hope to have at least begun the work of dispelling some rather problematic assumptions while pointing to some interesting and important commonalities. Above all, I hope to have shown that some so-called postmodernists are not discursive idealists at all but share realism’s commitment to relative intransitivity and a depth ontology. This is to say that realism need not reject post-structuralist theory to the extent that some of its proponents do, via a non-immanent critique of so-called postmodernism, which is conflated with post-structuralism and then with certain theorists whose positions are in fact quite close to those of certain realists. Realism and post-structuralism, I have tried to show, overlap at several

Critical Realism’s Conception of Post-structuralist Theory 139

crucial and productive points. To be sure, the possibilities for more reflexive, politically engaged, and theoretically sophisticated analyses of natural and social structures and processes are often only alluded to in the work of theorists such as Deleuze and Guattari, who clearly preferred a philosophical, poetic approach and did not wish to generate anything like a ‘protocol’ for what they saw as Americanized approaches to social scientific ‘research.’ This does not mean, however, that the insights they generated cannot be of use to those whose interests incline more toward what are commonly known as the natural and social sciences.

NOTES 1 Since there are many divisions within the realist tradition inaugurated by Roy Bhaskar, e.g., the distinction between ‘critical’ and ‘ordinary’ realisms, I will define my object here as ‘Bhaskarian-derived’ realisms. My claim is that most of these share a tendency to set up and burn a straw ‘postmodernism.’ 2 To avoid an excess of scare quotes, I will no longer flag the term ‘postmodernist’ and its cognates. I will also avoid flagging the term ‘post-structuralist,’ although it too needs to be treated with caution. For example, I do not in any way want to suggest that the reading I produce here is the only possible reading of what a post-structuralist position with respect to realism might look like. Nor do I want to exclude the possibility that some realists might identify with some aspects of some texts commonly known as post-structuralist yet produce readings that are different from mine. It is my position that no synthetic label can substitute for close readings of particular texts, while at the same time, for purposes of intelligibility, one finds it necessary to use certain terms while working against them. This, I would suggest, can be seen as a fundamental insight/necessity of Derridean/ Heideggerian deconstruction: the necessity of working ‘under erasure,’ of ‘crossing through’ particular terms. 3 The term ‘postmodernists’ appears unqualified throughout all three chapters, and the discussion relies quite heavily on sources that, once again, offer relatively easy targets. The familiar trope of guilt by association is also used quite heavily, most obviously in the chapter on ‘postmodern flips.’ At the start of each of the first two sections, Sayer quickly mentions an idea commonly attributed to one of the major poststructuralist theorists, then goes on to discuss the work of others whom

140 Richard Day

4

5

6

7

he seems to think can stand in for the theorist he first mentions. In the section on the ‘pomo flip’ from foundationalism to idealism, he refers to ‘what Derrida terms “the metaphysics of presence”’ at the outset (2000: 68) but does not mention Derrida again in the ensuing discussion. In the section on grand narratives and local knowledges, Lyotard is handled in the same way. Because the text is appropriately laced with vague references to postmodernists and post-structuralists the casual reader can easily get the impression that Derrida and Lyotard – and all poststructuralists – are guilty of the charges Sayer levels against his stand-ins. Attempts to draw this kind of line are indeed characteristic of a realist position. Andrew Sayer, for example, says realism ‘insists on … the presence of a non-discursive, material dimension to social life’ (2000: 18). The problem with postmodernists is that they take ‘a turn to discourse and away from materialism’ (36). Here again we see the assumption that hard and fast lines can, and should, be drawn between the discursive and the material. Their means of defining the distinctions between these strata, though based on a more general theorization of form, content, matter, substance, and expression, are also similar to Wilden’s, in that a certain ‘unity of composition’ is sought. On the geological-chemical stratum, the basis for unity is found in the fact that molar, or human-scaled, forms can be said to express microscopic molecular interactions, as in the example of the crystal. The organic stratum is characterized by the linear, spatial form of coding found in the DNA-RNA system, which is exemplified in the form of the biological cell. Finally, the linguistic stratum displays temporal linearity and is the exclusive realm of signs and language. In Deleuze and Guattari’s discussion of the practice of the smith, one might see something similar to the realist notion of a realm of ‘tendencies and potentials,’ as distinct from ‘activated powers.’ What a smith does, perhaps, is activate powers that are, using Deleuze and Guattari’s terminology, ‘immanent’ to a particular ‘plane.’ Bourdieu’s work most definitely haunts this article but cannot be materialized within it, so to speak, without dealing with a large number of difficult questions that can (too easily) be summarized by the problem of whether or not he was a poststructuralist. The notion of structured-structuringstructure is too important to let go without at least a mention, however. It arises in Outline of a Theory of Practice (1977a), in the context of a discussion of habitus. Bourdieu writes, ‘The structures constitutive of a particular type of environment … produce habitus, … structured structures predisposed to function as structuring structures, that is, as principles of the

Critical Realism’s Conception of Post-structuralist Theory 141 generation and structuring of practices and representations’ (1977: 72). My intuition is that all structures are structured structuring structures, to some extent. 8 Although the Anti-Oedipus is sometimes seen as a total rejection of Lacanian psychoanalysis, Deleuze and Guattari actually say that Lacan has come up with an ‘admirable theory of desire’ (1983: 27, note). What they take isssue with is a ‘certain sort of Lacanism’ that they associate with some of ‘Lacan’s disciples’ (1983: 52). This to say only that one cannot simply dismiss any attempt to work across the texts of Deleuze and Guattari and Lacan on the basis of an assumed incommensurability of their approaches.

7 Thinking across the Culture/ Nature Divide: An Empirical Study of Issues for Critical Realism and Social Constructionism RAYMOND MURPHY

Introduction Social constructionism is impressive in its accomplishment of detailed empirical and ethnographic studies of a wide variety of phenomena. It is, however, less impressive in its tendency to bracket or suspend (ignore) nature’s dynamics from its analyses, leading its accounts to become decontextualized abstractions. Critical realism offers a more complete analysis by situating social constructions in their context of nature’s constructions. Critical realism must nevertheless also develop a strong empirical dimension if it is to flourish. The present paper seeks to document this weakness of social constructionism, to contribute to thinking across the culture/nature divide, drawing on the work of critical realists and some constructionists who have started to transcend that weakness, and to ground the critical realist approach by using it to conduct an empirical study analysing specific issues in relations between social constructions and nature’s constructions. Sociological Analysis That Takes into Account Nature’s Constructions: Simply Unthinkable Ted Benton (2001b: 137) argues that the presumption of categorical separateness between body, animals, biological and physical conditions (i.e., nature) on the one hand, and on the other, meaning, subjectivity, identity, and the human (i.e., culture) unfortunately ‘renders unthinkable’ the analysis of the interaction between them. Approaches attempting to restrict sociology to the culture side and sociologists condemning as unthinkable sociological work that endeavours to

Thinking across the Culture/Nature Divide 143

think across the culture/nature divide are found even in subareas directly bearing on culture/nature interaction. In the sociology of scientific knowledge (SSK), Harry Collins and Steve Yearley condemn as prosaic, conservative, conceited, and indeed epistemologically chicken any sociologist who dares to think across the culture/nature divide, such as Bruno Latour and Michel Callon. Collins and Yearley (1992a: 310) affirm that ‘the effect of SSK has been to show that the apparent independent power of the natural world is granted by human beings in social negotiation.’ How do they know that the independent power of the natural world is merely ‘apparent’ and that it is ‘granted’ by omnipotent human beings? Their proof is by epistemological assumption. ‘The methodological prescription that emerges from relativism is that explanations should be developed within the assumption that the real world does not affect what the scientist believes about it’ (Collins & Yearley 1992b: 372). When the assumption of impotent nature is fed into sociological accounts, it is not surprising that conclusions of feckless nature come out. Their assumption is based not on empirical findings but rather on their political agenda of keeping scientists at bay: ‘If nonhumans are actants, then we need a way of determining their power. This is the business of scientists and technologists’ (Collins & Yearley 1992a: 322). Their methodological prejudice of presupposing that nature’s dynamics do not affect the content of scientific theories renders them blind to the difference between the social construction of science and of parapsychology (Collins & Pinch 1979), and it undermines sociology’s ability to explain the power of modern science (Murdoch 2001; Schmidt 2001). In environmental sociology, constructionism brackets nature’s dynamics and only investigates how discourse and practices are socially constructed. Thus, books are written about ‘the social construction of nature’ and ‘Protestant nature’ (Eder 1996), and ‘contested natures’ (McNaghten & Urry, 1998). These writings claim to examine how sociocultural factors mediate between nature and accounts of nature. Since nature is suspended, however, analysis in this epistemic constructionism is done as if nature had no influence and only sociocultural factors were explanatory. The conception of the ‘social construction of nature’ reduces nature to discourse about it and obscures the effects of the autonomous dynamics of nature. The same weakness is found in the sociological analysis of risk, where social constructionists (Hannigan 1995: 38–39) limit their studies to threats that are talked about.

144 Raymond Murphy

The sociology of disaster has a paradigm where the physical origins of disaster are intentionally not studied – disaster is presumed social rather than physical (Quarantelli 1986): ‘there is no reference to disaster agents which implies that all disasters are socially caused’ (Dynes 1998: 113). Enrico Louis Quarantelli (1998: 259) contends that the ‘more we get away from our hang-up of including an agent or the physical environment as part of our conceptual view and focus on the social behaviour involved, the better off we will be,’ which results in the argument that ‘disaster is tightly linked to uncertainty that occurs when a danger, whether real or not, threatens a community, and this danger cannot be defined through causes or effects … It is the result of the upsetting in the system of meaning, and not the effect of the difficulty of solving problems of accidents or serious dysfunctions’ (Gilbert 1998: 17). The old premise ‘If men define a situation as a crisis, it will be a crisis in its consequences’ (Thomas & Thomas 1928: 572) is used to dismiss the importance of the reality of the danger. But false scares do not typically have the same gravity as real risk, whether measured in terms of fatalities, material damage, or duration of disorganization: compare the consequences of Orson Welles’s War of the Worlds with those of Hurricane Andrew. False alarms and pseudo risks are embarrassing but not catastrophic. The reality or falseness of claims of safety or danger is eventually revealed (by the dynamics of nature in cases examined here). Real safety and erroneous assurances of safety are different in kind. Understanding requires that this difference be captured in social theory. ‘We have no alternative but to attempt to assess the relative practical adequacy or objectivity of different social constructions’ (Sayer 1997a: 468). It is difficult to assess the reality of claims of safety or risk, but social theory is stronger if it deals with the difficult rather than sidestepping it. Wolf Dombrowsky (1998) argues that the exclusion of the physical environment results in ‘a misleading sociologism’ because disasters are generated not only by human interaction but also by our interaction with nature’s own autodynamic and self-organizing processes. Vulnerability analysis becomes weak if it neglects hazards analysis because, as Piers Blaikie et al. (1994) argue, it then lacks an explanation of how one gets from very widespread conditions such as ‘poverty’ to very particular vulnerabilities that link political economy to actual hazards people face. James Lewis (1999) concludes that by ignoring material hazards or assuming that development of any kind will make disasters go away, disasters could be made to increase (see also Murphy 2006b). .

Thinking across the Culture/Nature Divide 145

A Critical Realist Analysis of Interaction between Social Practices and Nature’s Dynamics Critical realism shares with social constructionism the study of how beliefs are socially constructed on the basis of power and sociocultural contingencies, but in addition CR includes two features that constructionists have intentionally ignored (euphemistically called bracketing, suspending, etc.). The first involves the unintended consequences of particular beliefs, for example, of assumptions of safety or risk. Claims were socially constructed on the basis of political, economic, and cultural contingencies that the 28 January 1986 Challenger Space Shuttle launch was safe, but there were serious ramifications of this assumption, including biophysical ones for the astronauts. Critical realism includes in its analysis the fact that competing claims have different material consequences, whereas relativist constructionism has tended to ignore and obscure that fact. Second, beliefs are influenced not only by sociocultural contingencies but also by biophysical ones. Interpretations are not constructed by pure spirits in a material vacuum. A constructionist sociology that abstracts embodiment and ecological embeddedness out of the analysis is all the weaker for it. Biophysical contingencies are included by realists to understand how conceptions are socially constructed in a material context. Realism examines ‘the way in which social order is embedded and conditioned by the natural order from which it is emergent and on which it in turn acts back’ (Bhaskar 1989: 173–140). It draws attention to the importance of investigating the complex processes of interaction and mutual constitution that tie together phenomena misleadingly dissociated from one another and restricted abstractly to one side or other of the nature/culture divide. ‘Only a theoretical breakthrough which precisely enables such thinking across the Nature/Culture divide, and in the process deconstructs it, could have any hope of grasping the underlying generative causes of our ecological predicament’ (Benton 2001b: 137). Critical realists envisage that a material problem can exist based on the autonomous dynamics of nature before risk is acknowledged and that this constitutes a latent social problem. This period of unperceived risk is particularly dangerous because no action is taken to transform social practices to correct the material problem. Current conceptions of nature only grasp some of its levels (Soper 1995: 158), yet

146 Raymond Murphy

proper concern about nature presupposes true beliefs about it (O’Neill 1994: 27). Nature is layered (Benton & Craib 2001: 125): causes often lie beneath surface appearances, and some knowledge claims are more valid than others (Rosa 1998). This layering of nature as well as the time lag (Adam 1998; Dickens 2001: 93–94) between the beginning of nature’s dynamics and effects visible (even with science) results in unperceived risk. Thus, it is necessary to avoid confusing ‘social constructs or interpretations with their material products or referents’ (Sayer 1997a: 482). The issue is not just the trivial terminological question of labelling perceptions ‘risks’ and material threats ‘dangers’ but rather the important issue of analysing the differing material consequences of divergent assumptions concerning nature and how biophysical consequences constitute contingencies that influence interpretations. The critical realist analysis of risk integrates biophysical risk, culture, and analytical tools (Rosa 1998: 28). Representations of safety or risk, including scientific ones, are social constructions that are fallible: they may or may not be accurate, with the environment being the testing grounds where expectations are publicly examined. This approach leaves conceptual space for unawareness: material hazards that are unperceived because of social, cultural, or political factors, or because the tools for identifying those hazards have not been developed. It does not reduce risk to its perception or assume that scientific hypotheses of risk are necessarily conterminous with biophysical risk. In the critical realist approach to environmental sociology, beliefs about nature – including scientific knowledge – are analysed as fallible social constructions that result in practices with different consequences, depending on their interaction with nature’s autonomous dynamics, and these consequences then affect discourse about nature. Peter Dickens (2003, 2004) theorizes that the social construction of efficient technology modifies nature physically, which turns back and transforms humans psychically, leading large groups to assume their invulnerability and omnipotence and to be obsessed with present consumption and oblivious to future needs (see also Murphy 2006a). Crashing into Nature’s Constructions: Fallible Expectations and Unacknowledged Realism Sayer asserts that the most compelling reason for accepting the basic realist premise of the independence or otherness of the world is the

Thinking across the Culture/Nature Divide 147

experience of making mistakes, of having one’s expectations confounded and of crashing into things unexpectedly – in other words, the experience of falsification (2001: 969). There are many more social scientists who carry out their research in terms of this basic realist premise than there are those who identify themselves as realists. For example, Diane Vaughan (1996) portrays her study of the Challenger Space Shuttle calamity as a constructionist analysis and never mentions critical realism; nevertheless, her sociological analysis of mistakes is based on the independence and otherness of the world, on the falsification of socially constructed expectations when the Challenger crashed into surprising dynamics of nature, and on the effect this extreme falsification had on subsequent expectations. When Charles Perrow (1984: 75) studied technologies that employed dangerous processes of nature or operated in locations where nature’s dynamics were unsafe, he showed the inadequacy of studying only how cultural expectations were constructed and contested, and he documented the necessity of examining material consequences of the interaction of practices based on such expectations with the dynamics of nature: ‘we [scientists at Three-Mile Island] acted in terms of our own designs of a world that we expected to exist, but the world was different.’ Barry Turner (1978) developed his analysis of ‘failures of foresight’ and the incubation of ‘man-made disasters’ of technology precisely on the basis of the discrepancy between socially constructed expectations about harnessed energy of nature and those real forces that slipped their leash. The anthropologist Anthony Oliver-Smith (1998) argues that nature’s constructions are exosemiotic, prediscursive processes that do not depend on human perceptions or interpretations yet affect both. He developed a political ecology of natural disasters that inhere in societal-environmental relations, arguing that ‘the question of how well a society is adapted to its environment must now be linked to the question of how well an environment fares around a society. The issue of mutuality is now at the forefront’ (1998: 193). Disasters increasingly indicate imbalances in such mutuality. Although there are social science currents that restrict themselves to the cultural side of the culture/nature divide, there are nevertheless important works such as the above that think across the divide. Far from being unthinkable, a rich body of realist sociological analysis already transcends the nature/culture divide.

148 Raymond Murphy

Moving Social Constructionism toward Realism by Incorporating Nature’s Dynamics Benton’s depiction of the analysis of the interaction between social constructions and those of nature as unthinkable is an apt description of most constructionism, but he (2001a) nevertheless demonstrates the realism that underlies some constructionist studies, notably those of Brian Wynne. Some other authors associated with constructionism are more explicit about their realism and have not ignored nonhuman actants. Prominent examples are Latour and Callon. Latour (2000: 117) states that ‘by insisting so much on hermeneutic loops, social scientists have got too easily out of the loop – leaving in the dark the myriad of non-human actants, so essential to the very definition of humanity.’ Latour (1991b: 110) seeks ‘to avoid the twin pitfalls of sociologism and technologism. We are never faced with objects or social relations, we are faced with chains which are associations of humans (H) and non-humans (NH).’ The methodology he uses is to study situations in laboratories where objects object to what is said about them. The present paper aims to demonstrate that another important set of such situations consists of disasters, both natural and technological. This extension is consistent with Latour’s (2000: 116) argument: ‘Natural objects are naturally recalcitrant; the last thing that a scientist will say about them is that they are fully masterable. On the contrary, they always resist and make a shambles of our pretensions to control.’ This is certainly true of natural objects involved in disasters. Latour (2000: 117) counsels social scientists ‘to discover those rare and sometimes dangerous situations where neither intentionality, nor consciousness, nor reflexivity defines humanity’ with a view to analyzing them. A disaster fits this description at least as well as laboratory experiments. It is an extreme event ‘when things strike back’ (Latour 2000). Suggestions for Analysing the Interaction and Mutual Constitution of Humans and Non-Humans Certain issues in the analysis of the interaction between humans and non-humans, culture and nature, require clarification. They will be examined here with reference to the study of a concrete event in order to ground the analysis empirically.

Thinking across the Culture/Nature Divide 149

What Are Non-Humans? In Latour’s illustrations, nonhumans are machines, speed bumps, European weighted hotel keys, etc., that control human activities according to their human designs or are translated to different objectives by other humans (see also Callon 1991: 136–137). Non-humans are incorporated into sociological analysis in a limited way: only nonhumans that are human creations and/or construct social order at the bequest of humans are considered. Non-humans are assumed to do what they are commanded to do by one network of humans or another, a very dubious assumption indeed. In his programmatic proposals, however, Latour suggests a broader and deeper conception of nonhumans and their actions. He (Latour 1991a: 17, 147) does not assume that the recomposition of society at an increased scale through quasiobjects like the electricity system occurs in terms only of active humans making social constructions using passive objects of nature. Rather, the social and the natural ‘mutually exchange and enhance their properties’ (Latour 1999: 125): they are both active in this process. Callon, Pierre Lascoumes, and Yannick Barthe (2001) argue that society is as uncertain as the non-human entities with which it shares its destiny. It is important to know what the ‘crowds of nonhumans’ (Latour 1996: viii) are and to have more precise conceptions than ‘non-humans’ and ‘things.’ In order to indicate that technological artefacts – from weighted keys to automobiles to nuclear reactors – consist of a rearrangement of nature’s dynamics and materials to accomplish particular goals and that nature’s dynamics embedded therein retain their potential to escape control, I have referred to them as forms of ‘recombinant nature’ (Murphy 2002b). Non-humans recombined in technological artefacts can produce consequences contrary to the objectives of the human recombiners. For example, Turner (1978) documented ‘man-made disasters’ resulting from the attempted harnessing of nature’s dynamics, which then slipped their lease. Perrow (1984) found ‘normal accidents’ in high-technology systems using nature’s dangerous materials. The recalcitrant character of recombinant nature and the tenuousness of controls were demonstrated in 2003 by the instantaneous cascading of power blackouts in northeastern North America and Italy. These are, however, just a fraction of non-humans. What I have called ‘primal nature’ (Murphy 2002b) constitutes a second type and

150 Raymond Murphy

refers to nature’s constructions that existed in pristine nature before humans evolved but that have become internalized into societies as the latter developed and population grew. Constructions of primal nature crash into human social constructions, often unexpectedly. Extreme weather events are examples. Nature retains its autonomy even as pristine nature is transformed into socially encompassed primal nature and even as technological constructions recombine some of its dynamics. I would suggest that two subtypes of primal nature can be distinguished: naturogenic and anthropogenic. The first refers to nature’s dynamics that have occurred and continue to occur without being significantly affected by human actions. For example, there have been earthquakes and volcanoes before and after humans emerged on the planet. They have significant effects on the social lives of humans. The second refers to the dynamics of primal nature as they have been modified or unleashed, usually inadvertently, by human activities. The depletion of the ozone layer by the use of CFCs thereby letting in more solar radiation constitutes an example. Another involves the increased frequency and intensity of extreme weather events predicted by meteorologists (Milton & Bourque 1999: 82; IPCC 2001; Webster et al. 2005) as human activities provoke global climate change. Constructions of primal nature can still be naturogenic, as they were before the evolution of humans, but now they are increasingly anthropogenic. Meteorological evidence in the case of global climate change is, however, only suggestive rather than definitive. The line of demarcation between naturogenic and anthropogenic primal nature is politically contested terrain. Non-human actants do not have intentionality and purposive motivation in the sense that human agents do. Attributing purposive behaviour to everything nonhuman would be misleading anthropomorphism. Nevertheless, these actants create, construct, and destroy in both forms of recombinant nature and primal nature. For example, a particular script to attain specific objectives was choreographed into the Challenger space shuttle by humans, but nature’s actants affected its launch in ways that violated the script, with fatal consequences (Vaughan 1996). The launch failed to attain its goals because nature had more autonomy than was assumed by the technical scribes who wrote the script. When non-humans are conceived of as materials and dynamics of primal and recombinant nature, Callon and Latour’s work about the co-construction by humans and nonhumans and Callon’s argument

Thinking across the Culture/Nature Divide 151

that society is as uncertain as its non-human associates complement anthropological work in the political ecology of disasters: ‘societalenvironmental relations are not relations between two separate entities, but between two mutually constitutive entities … people, economies, and nature are now in a process of co-evolution on a global scale, each influencing the others in unfamiliar ways and at scales that challenge our traditional understandings of structure and organization with serious implications for the adaptive capacities of people and societies’ (Oliver-Smith 1998: 193). The extreme weather event dealt with here occurred when unusually warm, moist air associated with El N~i no travelled across the North American continent in January 1998 and rained through the usual stagnant cold air mass in western Quebec, eastern Ontario, northern New York State and the New England states. The resulting five days of intense freezing rain created an ice loading on most surfaces that many structures could not support. Electrical lines and towers collapsed under the weight. Five million people were left without light and heat in the bitterly cold, dark winter, including 600,000 in Quebec for two weeks or more (Commission 1999: 34). Forty-seven people died as a direct result of the ice storm in Canada and the United States, and the death rate increased indirectly as well (Commission 1999: 46; Region I 1998: 8). Businesses and schools in Montreal had to be closed. Government offices in Canada’s capital, Ottawa, were shut down. Transportation was in chaos. Automatic teller machines and credit card machines did not function, so people were deprived of their own money. Televisions went black, some telephone lines went dead, and many cellular phones ceased operating when their electrically dependent towers no longer functioned. Modern technological infrastructures had proven vulnerable to intense, persistent freezing rain. The metropolis of Montreal temporarily ceased most of its activity, even if its residents survived. In Canada this ice storm, which paradoxically resulted not from unusual cooling but rather from abnormal warming, produced the most expensive disaster in the country’s history. Non-human actants of primal nature, such as extreme weather, are too socially important for social scientists to bracket. A Hybrid Disaster: Natural Yet Socially Constructed There are about twelve short freezing rain events a year in this part of the continent. Power companies learned from experience and

152 Raymond Murphy

constructed lines to withstand expected ice loadings. The January 1998 freezing rain went well beyond socially constructed expectations. People were not much affected by the freezing rain itself. They were seriously affected, however, when the electrical network that had been made essential was crushed by the resulting ice loading. In the past, decentralized wood and coal stoves provided heat no matter how intense or persistent the freezing rain, but they had been replaced by more comfortable and convenient means of heating dependent on electricity. This apparently natural disaster was unintentionally constructed by the recomposition of society on a wider scale using technoeconomic networks of transmission lines that bring power generated in a distant wilderness through the gravitational attraction on water to supply essential infrastructures of a modern metropolis (Murphy 2001). Dependence on this recombinant nature rendered modern society more fragile when it was faced with forces of primal nature in the form of persistent, intense freezing rain that exceeded expectations. Hence Latour’s (1991b) hypothesis that ‘technology is society made durable’ requires serious qualification when the full range of non-human actants are taken into account. That simple aphorism fails to capture the complexity of the situation. Not only technological artefacts but also this disaster was a hybrid, coconstructed by non-humans in the form of forces of primal nature and by modern practices of humans that exacerbated vulnerability. Nature’s Contingencies Inciting Patterns of Interpretative Similitude There is not a one-to-one relationship between nature’s dynamics and human conceptions and practices concerning them. Rather, there is interpretive flexibility (Pinch & Bijker: 1984): culture mediates relationships between nature’s processes and social practices. Nevertheless, there were many similarities in the disaster response of Quebec and the United States, such as federal compensation to spread out its cost so that communities hard hit could be resilient because of help from those not struck, politicians rushing in to show compassion for voters in order to receive their compassion in return, etc. Slight differences between societies should not blind us to the fact that nature’s disturbance activated a response in both, and the responses tended to be much alike because similar problems were being dealt with. On the federal levels the emergency legislations in the United States (the Robert T. Stafford Disaster Relief and Emergency Act) and Canada

Thinking across the Culture/Nature Divide 153

(the Emergencies Act) are remarkably similar, both calling for a bottomup response: municipal, and if insufficient, then provincial (state), and if that is inadequate, then federal assistance. The hypothetical possibility of interpretative flexibility did not suppress empirical patterns and structures of interpretative similitude. In both countries interpretations of the freezing rain shifted as its duration, intensity, and material consequences changed. It left a clear, thick coating of ice on all the branches of every tree and shrub. A northern New York State journalist recounted ‘waking up to a crystallized wonderland’ that turned out to be a ‘wintry nightmare’ (Ice Storm 1998: 11). A counsellor at the Brazilian embassy in Ottawa observed, ‘Another thing that struck me about the ice storm is the irony of the thing, because it’s so beautiful. In the first moments I even went out to take pictures because the trees looked as if they were made of glass. Normally catastrophes are so ugly’ (Ottawa Citizen 1998). The character of the extreme weather event itself led people from different cultures to the similar interpretation that this was a photogenic catastrophe. Newspapers produced not only stories but also photographs of the beauty as well as the destructiveness, emphasizing the contradiction between the two. A hardcover book of photographs (Abley 1998) became a best-seller in Canada. As the ice melted, broken trees, downed electrical lines, and fractured pylons replaced the crystallized wonderland. Expressions of the storm’s beauty were replaced by those of its destructiveness. The interpretation that was socially constructed by sensory humans was a product of their prior social and cultural contingencies and contingencies stemming from the physical disturbance of nature. It would be misleading to reduce it to only one or the other. Experience: Is It Always a Conscious Conception? Since humans are biophysical beings entrenched in nature’s constructions, we can experience phenomena without having a conscious conception of being affected and without interpreting that effect. For example, one of the major causes of fatalities and illnesses during the 1998 ice storm was carbon monoxide poisoning (Commission 1999). After people lost electrical power for their furnaces, they were desperate for heat in the frigid weather and some used gas generators or even gas barbecues in their houses. Invisible, odourless carbon monoxide poisoned them as they slept. They experienced death or illness without being aware of either. In this sense, culture did not mediate between

154 Raymond Murphy

carbon monoxide and death. Something can be experienced without being consciously perceived and without an interpretation by the people involved. This involves experience on the material level, but that is what so much of life is about for embodied humans in a biophysical world (the early stages of cancer, CJD, SARS, etc., are not consciously perceived, nor are sudden bomb fatalities). Of course, people who discovered the bodies then interpreted their findings. If mediation includes omission, however, then culture did mediate between carbon monoxide and its effects. A subculture that ignored the dangers led to this result. The fatal consequences were the unintended by-products of cultural beliefs inappropriate for the dynamics of nature. In this case, the risk-unaware subculture coexisted with a scientific culture that knew of the risk, but before these poisonings communication between the two was inadequate. No general inference of omniscient science is implied here. Science-based organizations can also be ignorant of nature’s emergent constructions. Vaughan (1996) showed that the Challenger space shuttle tragedy resulted not from reckless risk-taking but instead from a culture at NASA and at its subcontractors that failed to comprehend the risks of nature’s dynamics. Human cultural constructions, including scientific hypotheses, that attempt to depict nature’s autonomous constructions are fallible. This fallibility is an important characteristic of embodied humans living in a biophysical world. Cultural mediation and interpretation of biophysical experience is not a universal constant but rather a variable of different degrees and forms that requires investigation. Turner (1978) showed that when risk is accurately perceived, measures are taken that avoid calamities, whereas disasters involve failures of foresight. Culturally unmediated, little mediated, and mistakenly mediated biophysical experience should be theorized rather than ignored, especially at a time of global environmental change whose hidden effects may already be occurring without our knowing it. How Can There Be Unperceived Risk? ‘There is no way to separate the world from human interpretation of it’ is the refrain usually heard to deny the reality of a world independent of human interpretation or to be agnostic about it. However, because humans are embodied beings embedded in a biophysical world, the relationship between that world and its interpretation is a complex one

Thinking across the Culture/Nature Divide 155

not captured by that simplistic refrain that reduces the world to interpretations without referents. Humans interpret nonhumans, who as Latour (2000: 115) observed, can ‘object to what has been said about them,’ leading to a manifest disjuncture between the world and its human interpretation, hence often to revised interpretations. One sometimes hears the objection that if risk is unperceived, how would we ever know it? The question is perplexing only from an atemporal perspective. It can be answered by including time in the analysis. As Barbara Adam (1998) and Dickens (2001: 93–94) have shown with respect to mad cow disease (BSE-CJD), there is often a time lag between the beginning of nature’s dynamics producing material destructiveness and conceptions of risk. Visibility of physical effects is a variable that changes over time and influences assumptions of safety or risk, which are social constructions forged in a social and a biophysical context. Social and material contingencies affect what is seen as risky and the level of risk that is accepted. Risk is eventually known when material consequences become clearly visible: the worst way to learn about risk is by experiencing disaster. After a disastrous event, studies can determine whether the risk was perceived prior to it. Disaster studies (Turner 1978; Perrow 1984; Vaughan 1996) have documented that mistaken expectations, failures of foresight, and unperceived risk are often important sources of disasters. Taken-for-granted assumptions of safety and business as usual are difficult to contest until nature’s constructions produce visible effects. It is especially in the evidential grey zone where some scientific or experiential evidence suggests danger without being conclusive that there is political struggle over divergent risk perceptions. This is a major problem now, at a time of global climate change. A crucial issue when dealing with recombinant or primal nature remains nevertheless to perceive material risk accurately in time to mitigate it or at least adapt to it. With respect to the 1998 ice storm, meteorologists, power companies, emergency measures organizations, and laypeople did not perceive the risk of such intense, persistent freezing rain and were unprepared for it (Phillips 1998: 82; Region I 1998: 19; Commission 1999: 106). A predisposition to assume safety reigned, based not only on prior cultural assumptions but also on prior material experiences. Unawareness, not knowledge, was the significant phenomenon in these relations between social constructions and nature’s constructions, as Beck (1998: 90, 95) theorized. Adequate robustness was not built into the electrical network, and it collapsed. Unawareness was a source of disastrousness.

156 Raymond Murphy

The unperceived character of the risk was documented after the risk materialized into disaster (Commission 1999). Nature’s Constructions as Prompts to Social Constructions It is one thing to understand how nature’s constructions affect humans materially, but how do they influence cultural constructions like perceptions and beliefs? I would argue that non-humans as actants have material consequences for humans and that these are prompts for the social construction of interpretations, conceptions, and practices. Coconstructions occur when constructions by non-human actants prompt constructions by purposive, embodied human agents. ‘Prompt’ is a particularly useful concept for transcending the culture/nature divide because it captures the influence of nature’s constructions on human discourse and practices without implying that nature determines them, leaving a place for the mediation of culture and power. Prompts from nature’s dynamics can be classified according to three different types. Scientific prompts to discourse and practices result from ingenuous experiments and measurements that enable nature to object to what had been said about it, thereby leading to deeper scientific theories, new technologies, and the resulting recomposition of society at a different level. These prompts consist of findings that arouse an awareness of nature’s dynamics before they are otherwise visible, for example, meteorological forecasts. Imaginative scientific measurements made visible ozone-layer depletion, which incited the social construction of the Montreal Protocol. Reiner Grundmann (2001) emphasized clever measurements devised by ozone scientists and the cunning presentation of results that convinced politicians, negotiators, and the public there was a problem requiring social change to solve. However, Grundmann’s constructionist study does not relate this to whether there was a real material problem of ozone-layer depletion. Latourian realism would conclude that these clever measurements made visible hitherto invisible physical dynamics and thereby enabled the ozone layer to object to the assurances of safety by the CFC industry. Had negotiations depended only on rhetorical skill and social resources, the chemical industry – which certainly did not lack resources, intelligent scientists, and marketing specialists – could have countered with its own cleverly plausible discursive constructions. Thus, Grundmann’s study is convincing to the extent it is

Thinking across the Culture/Nature Divide 157

read as abandoning agnosticism concerning ozone-layer depletion. Translation of nature’s dynamics into discursive representations, not substitution of discourse unrelated to those dynamics, was the social process involved. Collins and Yearley (1992a; b) argue that nature’s dynamics must be excluded from explanations of discourse and social practices in order to deny natural scientists any role in those explanations. Scientific knowledge does provide a singular capacity to explain nature, thereby affecting discourse and practices, which SSK has not changed. But their worry ignores two other types of prompt coming from nature’s dynamics that influence everyone and can override scientific prompts. One involves extreme prompts, which occur through exceptional dynamics of either primal nature or recombinant nature (assumed harnessed but that slips its leash). These produce natural or technological disasters if societies are vulnerable, and their observation and/or experience can incite a modification of conceptions and practices: ‘At NASA, the crisis that precipitated a transformation of worldview and resulted in a paradigm shift was not the teleconference, but the Challenger Disaster’ (Vaughan 1996: 402). Such extreme prompts are so visible that they are a critical material contingency in the ‘closing of controversies, the non-negotiability of facts’ (Callon & Latour 1992: 355) and in arousing new controversies. They can instigate a cultural shift: ‘A situation which had been presumed to have one set of properties is now revealed as having different and additional properties which must be interpreted differently. The precipitating event draws attention to itself because of its immediate characteristics and consequences’ (Turner 1978: 75). In the case of the ice storm, scientific prompts had predicted little chance of severe, intense freezing rain. Scientific interpretation was, however, subverted by the extreme prompt of nature that incited new conceptions of risk and vulnerability among both scientists and laypeople (Denis 2002). The third type of prompt can be missed because it is so omnipresent for embodied humans embedded in nature’s processes, much as we fail to notice breathing. These everyday prompts consist of mundane observations and experiences of nature’s routine dynamics by humans. In an illustration often used to cling solely to the cultural side of the culture/nature divide, Keith Tester (1991) noted that it is possible to attribute to fish the property that they have wings and fly. Despite this hypothetical possibility, what is notable is the empirical pattern. The

158 Raymond Murphy

everyday material contingencies of observing fish and their actions have incited the overwhelming majority of sensory humans to construct conceptions of wingless, flightless fish. Exceptional possibilities should not be used to obfuscate the predominant pattern. Sensory humans attribute properties to nonhumans, and the attribution tends to be influenced by the biophysical properties observed and experienced when these are easily visible. Collins and Yearley’s (1992a; 1992b) pursuit of surprises and attempt to transcend common sense have led them to disregard the unsurprising and dismiss the common. They perceive only interpretations by scientists and ignore everyday and extreme prompts from nature that directly influence interpretations of everyone. In the time preceding the January 1998 freezing rain, everyday prompts of normal winter weather – including weak, short bouts of freezing rain – as well as scientific prompts of available meteorological data led both laypeople and meteorologists to perceive little risk of a massive ice storm. These socially constructed cultural expectations were built into the electrical transmission network in the form of inadequate robustness to withstand massive ice loading (Commission 1999). The three types of prompts can diverge. For example, the scientific prompt of warning by the IPCC (2001) about the dangers of global climate change resulting from human activities is at variance with the everyday prompt of well-being in wealthy societies, leading to contradictory discourse, a disjuncture between discourse and practices, and political conflict concerning environmental degradation if the Kyoto Protocol is not implemented versus economic degradation if it is. Dickens (2003: 104) argues that ‘it will take the breaking in of reality, in the form of, for example, a substantial transformation of weather systems, for the culture of narcissism to be transformed and for the reality principle to be restored.’ Thus, he ominously implies that it will take an extreme prompt of disaster to break the culture of consumerism based on the presumption of invulnerability. The Extreme Prompt of Disaster Expectations of well-being based on everyday prompting by typical weather were refuted very publicly by the extreme prompt of freezing rain from primal nature, which triggered a reassessment of cultural assumptions. Risk of such intense weather was only perceived as it

Thinking across the Culture/Nature Divide 159

occurred when nature’s disturbance crashed into society unexpectedly. The electrical failure resulting from the ice storm incited people to realize the extent to which their modern society depends on wires coming into homes and buildings. In Quebec, where hydroelectricity predominates as an energy source, the disaster led to the recognition that the wires ‘were surprisingly fragile, surprisingly mortal. In each strand of metal, perhaps, we should still be able to glimpse the northern rainfall and the swirling river, the turbines and transformers of our lives’ (Abley 1998: 128). It led to awareness that electrical lines are ‘icesensitive structures’ (Jones & Mulherin 1998: ii, 45) and more generally that technical and social constructions are ‘nature-sensitive structures’ that can be overwhelmed by the autonomous dynamics of primal nature. Being deprived of heat and light in a dark, frigid climate shocked sensory victims into temporarily abandoning their belief in modern technology’s omnipotence. Denis (2002: 29) documented that the ice storm disaster prompted a heightened sense of vulnerability. There is a long history of the Quebec power utility building greater robustness into its network only after being prompted by extreme disturbances of primal nature, such as lightning strikes, sunstorms, freezing rain, etc., that led to blackouts (Murphy 2002a). The natural world has a significant role in influencing beliefs about it that are socially constructed. As Lewis (1999: xiv) concluded, ‘The subject of disasters, “natural” or otherwise, is essentially event driven.’ Deaths and sicknesses due to carbon monoxide poisoning were additional extreme prompts that influenced the discourse and practices of the living. Fatalities, as well as illness of those fortunate enough to be rescued, motivated authorities to issue warnings against the use of risky sources of heating (Commission 1999). The relationship between nature’s prompts and socially constructed beliefs about nature is, however, not straightforward. Even after extreme weather, economic considerations can lead to decisions not to make the electrical network significantly more robust on the hope that such weather will not recur for centuries. The decision is a social construction based on the prompt of nature’s disturbance and on socioeconomic concerns. The problematic (rather than automatic) character of socially constructed decisions in a social context of culture and power must not be obscured, but neither should the influence of the prompts of nature’s dynamics in the material context of autonomous physical and biological forces.

160 Raymond Murphy

Conclusion This chapter has built on the work of social scientists who have gone beyond analysing exclusively culture and who have begun to theorize across the culture/nature divide. Those social scientists typically consist of critical realists but also include some social constructionists. They understand that nonhumans cannot be treated as passive resources and constraints. The active character of non-humans in constructing and destroying has to be comprehended. The chapter demonstrated that an account of interpretations solely in terms of sociocultural contingencies is too restrictive, incomplete, and misleading. A deeper analysis requires taking material contingencies into consideration. Visibility of physical effects varies and influences perceptions of risk. Thus, risk perceptions are social constructions assembled in and influenced by a context of sociocultural and biophysical contingencies. Nonhumans as actants have biophysical consequences for sensory, purposive human agents, and these consequences are prompts for their social construction of interpretations and practices. ‘Prompt’ is especially helpful for thinking across the culture/nature divide because it conceptualizes the influence of nature’s dynamics on discourse and practices without suggesting determination, and it recognizes the mediation of power and culture. Moreover, contradictory prompts from nature are part of the explanation of tension in society concerning environmental issues. Humans can socially construct whatever assumptions they want – those of safety or danger – but these lead to different practices with different consequences. As societies recombine nature’s dynamics in novel ways and inadvertently unleash new processes of primal nature, they face the quandary of confronting the unforeseeable. By unintentionally producing unfamiliar vulnerabilities to autonomous forces of nature, new technologies have forced perplexing decisions on society. Two sources of uncertainty abound: uncertainty about nature’s constructions; and uncertainty about which decisions will be socially constructed, in particular, to deal with nature’s constructions. Studies that transcend the culture/nature divide by investigating interaction between humans and non-humans and resulting culture/ nature hybrids are particularly important as we enter an age of global climate change.

Thinking across the Culture/Nature Divide 161 NOTE I would like to thank the Social Sciences and Humanities Research Council of Canada for a grant that supported this research.

8 Beyond Cognitive Critiques: Getting Real about Politics JOSÉ JULIÁN LÓPEZ

Introduction Roy Bhaskar has argued that critical realism represents an attempt to ‘re-orient the human sciences away from the positivist and instrumentalist goals of prediction and control to realist ones of depth explanation and human emancipation’ (Bhaskar 1989: 187). To retool the human sciences for depth explanation, Bhaskar and others have contributed to the development of an explanatory practice in which ‘observation should be theory-driven; causal modelling and testing are better means of articulating theory and data than hypothesis testing for generalizations; and because of the irreducible differences between our minds and what they seek to comprehend, results are always ultimately fallible’ (Woodiwiss 2001: 19). Casting the human sciences for a role in human emancipation has necessitated arguing for the possibility of extracting normative injunctions from factual propositions. Whereas the depth explanation model provides the basis for a powerful theoretical and explanatory ‘visuality’ (Woodiwiss 2001),1 its alleged correlated emancipatory function remains problematic. This is not because the knowledges produced by the social sciences cannot or should not be deployed in attempts to produce social change, but because the emancipatory model, developed thus far by critical realism, fails to address the complex social processes involved in mobilizing social scientific knowledge to produce social change. Very schematically, while Bhaskar’s earlier work developed an account of the relationship between scientific explanation and the conditions of possibility of emancipatory practices in general (Bhaskar 1986), his subsequent contributions have failed to distinguish between

Beyond Cognitive Critiques 163

abstract conditions of possibility and actual political practice. There can be little doubt that Bhaskar’s ‘lofty level of conceptual abstraction prevents him from exploring the nature of the structural constraints and enablements of emancipation in different systemic contexts’ (Creaven 2003: 74). However, in addition to the hazards of vertiginous abstraction, Bhaskar’s account of human emancipation also retreats from an engagement with the substantive exigencies of political practice via its overwhelmingly cognitive conceptualization of knowledge production and its social and political efficacy. Consequently, the social conditions that would mediate the movement from explanatory critique to political action remain undertheorized when not completely ignored. In this chapter, I begin by discussing the limitations of the critical realist conception of emancipation against the backdrop of this cognitive bias. I argue that this cognitive orientation is also found in critical realism’s account of the natural sciences; it represents a cognitive idealism that fails to attend to the materiality of knowledge, of the social relations, artefacts, and infrastructures that facilitate the movement and/or translation of knowledge from one social site to another. In order to illustrate the importance of considering this social knowledge infrastructure, I leave aside the grander objective of human emancipation and discuss a more modest affair: the attempt, by a group of critical social scientists, to produce a more critical and reflexive bioethics in the United States. Emancipation: Really? Bhaskar’s writing on emancipation has sought to dull the salience of key dichotomies in normative discourse – fact and value, theory and practice, explanation and emancipation, and science and critique – by forcefully claiming that the human sciences are ‘necessarily nonneutral’ and ‘inherently critical’ (Bhaskar 1998b: 409–10). This critical conatus exists in two modalities of explanatory critique. In the first, it is the identification of false beliefs and the elucidation of their generative mechanisms that introduce, under certain conditions, the normative injunction to engage in practices conducive to their elimination: ‘inasmuch as we can explain, that is show the (perhaps contingent) necessity for some determinate false consciousness, or perhaps just some determinate consciousness under the determinable ‘false,’ then the inferences to a negative evaluation of its source(s) and a positive

164 José Julián López

evaluation of action oriented towards their dissolution are ceteris paribus mandatory’ (Bhaskar 1989: 101). In the second, it is the identification of causal mechanisms responsible for generating suffering, oppression, or preventing the fulfilment of real needs that provides the normative impetus for social transformation (Bhaskar 1986: 191). Both modalities, however, are underpinned by the notion that ‘explanation enables us to decide how to judge the situation under investigation, hence indicating what critical standpoint should be taken’ (Sayer 1997b: 484). Andrew Sayer (1997b) has summarized with great clarity some of the fundamental shortcomings associated with this critical realist emancipatory imaginary. He concedes that although there is a certain appeal to the notion that ‘we should choose the values secreted by the superior explanation’ (ibid.: 475), this position glosses over the actual difficulties involved in choosing among competing explanations and the values they engender (ibid.: 475). For Sayer, these problems are correlated both with a lack of reflexivity with respect to the social location of explanatory practices and a modus operandi that is predominantly oriented toward arguing for the elimination of certain types of social arrangements without offering a robust account of how they will be replaced: ‘The claims of CSS [critical social science] regarding its emancipatory potential need to be moderated by recognizing the limits of its method. The main problems lie in justifying the standpoints of its critiques and of finding alternative social forms which generate fewer problems than those they replace, and hence lead to net improvement’ (ibid.: 484). Moreover, by privileging scientific explanation as the mechanism for deciding which critical standpoint should be adopted, ‘the difficulty and contestability of normative judgements are downplayed’ (ibid.: 484).2 Ultimately, for Sayer, the continued espousal of an emancipatory model that fails to attend to the type of objections he raises would seem to be explained by three broad cognitive errors. First, Sayer claims that critical realists do not take sufficient account of the complex interdependent nature of modern societies in which a normatively justified claim for change in one domain might generate normatively problematic claims in others. Second, the ability to achieve a consensus on values, presupposed by the critical realist explanatory critique, fails to register the obstacles generated by contemporary value pluralism. Third, he groups critical realism with the ‘older critical sciences’ (i.e., Marxism) that have failed, in the language of Zymunt Bauman, to

Beyond Cognitive Critiques 165

grasp the inherent limitations of the modernist project (ibid.: 485–486). Be this as it may, I would argue that a more fundamental shortcoming can be identified within the critical realist emancipatory problematic. Rather than a series of cognitive misapprehensions about the structure of contemporary society, the problem lies with critical realism’s cognitivist understanding of how knowledge is mobilized to produce social and political effects. To be clear, I am not claiming that critical realists understand emancipation as a process internal to thought. Notwithstanding the limitations of the rather abstract formulations found in Bhaskar’s (1988) TMSA (Transformational Mode of Social Activity) and subsequent models of social structure, critical realism’s general commitment to the reality of materially positioned practices cannot be put in doubt. Indeed, Andrew Collier insists that for critical realism ‘the main part of emancipation is not cognitive, but consists in toil and trouble, conflict, changes in power relations, the break up of some social structures and the building up of others’ (Collier 1998: 461). But here is the rub. While he draws our attention to the need to mobilize agents for action, he does not draw attention to how the knowledges produced through explanatory critiques move from the sites of their production to political arenas, or why agents would find the cognitive critiques acceptable or legitimate grounds for action in the first place. In the absence of a conceptualization of the social relations, practices, media, and institutional arrangements that would make possible the circulation of the knowledges produced by explanatory critiques, these knowledges appear to be able to propel themselves through social space in socially efficacious modalities fuelling social change. It is in this sense that the designation of cognitive idealism is appropriate. And it is for this reason that the legitimacy of the explanatory critiques and the values that they create can be taken for granted.3 It can be argued that this tendency toward cognitive idealism was already presaged in the critical realist perfunctory analysis of the social materiality of knowledge production in the natural sciences and was transferred to its understanding of the human sciences. Really Situating the Production of Knowledge In A Realist Theory of Science (RTS), Bhaskar argued that knowledge production was above all to be conceived as a social production: ‘science must be seen as a social process, irreducible to individual acquisition,

166 José Julián López

whose aim is the production of the knowledge of the mechanisms of the production of phenomena in nature’ (Bhaskar 1989: 18). Ted Benton (1981) and Alan Chalmers (1988) noted early on that although Bhaskar’s description of experimental practice and the ontological transcendental arguments that he drew were compelling,4 it assumed that all natural scientific knowledge production practices were based on experimentation and consequently failed to capture scientific knowledge production practices that were not thus based. Bhaskar replied that ‘the account developed in RTS [was] not intended to be an exhaustive account of physics and chemistry, let alone of all the sciences’ (Bhaskar 1989: 183). In this context, however, it is rather telling that a critical realist program of theoretical and empirical research, focusing on the social dynamics underpinning differing scientific knowledge production activities, has not emerged. This task has been taken up by ethnographers and scholars of the social studies of science, which have for the most part adopted a social constructionist perspective. The fact that critical realism seems to have been satisfied with a broad philosophical, and empirically thin, account of natural scientific practice reveals a lack of commitment to exploring the social relations in which knowledge production activities unfold.5 Knowledge for critical realists is to be grasped philosophically, not sociologically. Consequently, it is perhaps not surprising that a similar lack of interest should exist with respect to the social materiality of knowledge production in the social sciences. More important, the absence of an empirically and conceptually thick description of scientific practices within the critical realist tradition has barred it from fully participating and learning from current debates in the field of science studies. This is unfortunate. Science studies scholars have documented in great empirical richness and with considerable conceptual creativity the extent to which the flow of knowledge is enabled through complex networks where myriad political, social, economic, and cultural elements are mobilized. For science studies it is the elaboration of these networks as well as the development of artefacts and devices (e.g., academic articles, measurements, standards, Bruno Latour’s [1987] ‘immutable mobiles’) that make contextualized knowledge practices, located in laboratories and other knowledge production sites, mobile. It is this socio-technical infrastructure that allows scientific and technical knowledge to intervene in the natural and social world effectively, or not. One need not accept the radical social constructivist ontology held by some science studies scholars in order to appreciate to what extent the field has opened up

Beyond Cognitive Critiques 167

avenues for the exploration of the social materiality of the efficacy of knowledge, its production and dissemination, and thus undermined the implicit cognitive idealism that results from the failure to engage seriously with the sociality of knowledge not only in the natural but also the human sciences. For instance, in contemporary society, institutions of higher education are the nodal sites in which social science explanatory critiques are produced and through which they circulate to other sites. Underpinning the apparent frictionless circulation of knowledge is a whole range of artefacts, standards, rules, practices, and institutional arrangements. Knowledge is materially inscribed in syllabi, books, journal articles, or performed through the mediation of classes or conference papers. In the absence of these infrastructural arrangements (Bowker & Star 2000), knowledge could neither circulate nor produce desired social or political effects. Thus, to be defensible, any claim about the political efficacy or emancipatory potential of a particular knowledge needs to produce an account, or at the very least demonstrate an awareness, of the social modalities, practices, and circuits that enable it to exist as a socially efficacious form of communication. In the human sciences, it is Michel Foucault’s archaeological work, specifically his concept of the discursive formation, that has contributed decisively to making knowledge infrastructures theoretically and empirically visible.6 In order to illustrate its significance, I now turn to a discussion of how critical social scientists have attempted to modify the field of bioethics. I briefly summarize the nature and origins of bioethics and then report on the critiques developed by social scientists. I then draw on Foucault’s rules of discursive formation to provide an account of the social relations in which bioethics is embedded and through which it flows. This will allow me to draw attention to the importance of conceptualizing the social (human and nonhuman) ‘carriers’ of critical knowledges and the circuits through which they might move. Why Change Bioethics? Bioethics – as a discourse, a discipline, an institutionalized practice, and an aspiring profession – is just over three decades old. Few commentators could have foreseen in its embryonic stage that it would have the impact that it has had. At present, however, there can be little doubt that bioethics has been institutionalized as a mode of normative

168 José Julián López

analysis capable of legitimately identifying and resolving biomedical ‘ethical’ issues in the United States (López 2004; DeVries & Subedi 1998: xi). For Dan Callahan, a bioethics pioneer, the task of bioethics is the ‘determination, so far as that is possible of what is right and wrong, good and bad, about scientific developments and technological deployments of biomedicine. What are our duties and responsibilities in the face of those developments?’ (1999: 276). Bioethicists typically identify technological developments in biomedicine and allied areas as providing the impetus for the emergence of their field, i.e., the ICU (Intensive Care Units), NICU (Neonatal Intensive Care Units), organ transplantation, human genetics, etc. Consequently, it is claimed that ‘new intellectual resources were summoned to struggle with the new questions: What is benefit? What is harm? Who should live? Who should die? How should the expensive resources of health care be distributed? Who should decide?’ (Jonsen 1998: 1). The formative role given to technological determinism is well captured by the framing of the relationship between science and ethics in terms of an ‘ethics lag’: ‘All one need to do to see the depth of this belief is track any story about the ethics of any major new breakthrough in biology or medicine. It will not be many paragraphs before the writer notes either that ethics always seems to be lagging behind scientific advances or that biomedicine has outstripped the capacity of ethics and the law to keep pace. The “ethics lag” is a powerful presumption in American, European, and Japanese assessments of the future of biomedicine’ (Caplan 1999: 272). Naturally, bioethicists acknowledge that other elements have contributed to the teething of bioethics, such as the Nuremberg trials, the crisis of trust, and the questioning of authority and paternalism spawned by the social movements in the 1960s.7 Nonetheless, these elements have ultimately remained subordinated to the ethical challenge that technological change is believed to have produced. Unsurprisingly, this framing has been contested by social science scholars. The concern of sociologists and anthropologists to criticize the technological determinist account, however, does not arise merely from the will to correct a one-sided historical account. There is more at stake! Thus, it was with a sense of urgency that the medical sociologists Renée C. Fox and Judith P. Swazey famously claimed, ‘Bioethics is not just bioethics … Using biology and medicine as a metaphorical language and symbolic medium, bioethics deals … with nothing less than beliefs, values and norms that are basic to our society, its cultural tradition and its collective conscience’ (1984: 336). In other words, given the

Beyond Cognitive Critiques 169

central role of health and medical technology in the governance of contemporary societies and the soteriological function that medicine was increasingly embodying (Good 1994: 70), bioethics was contributing to a realignment of core societal values. Consequently, they were imploring critical scholars and bioethicists alike to delve into bioethics’ unthought’: e.g., to explore those normative questions that were being displaced and the societal implications of the displacements. The ‘Origins’ of American Bioethics Bioethics embodies a structural bias toward proceduralism, formalism, and a disembedded universalistic reasoning that can be traced back to the disciplinary matrices that created an epistemological space for its emergence: moral and ethical philosophy and law (Toulmin 1982; Caplan 1983; Wolpe 1998). The hegemonic paradigm principlism is true to this origin. For its advocates, principlism brings together a number of elements that are extremely useful for clarifying what is at stake in medico-moral dilemmas, a valuable ‘moral prize’ not to be foolishly squandered: ‘an analytical framework, elements of common moral language and a basic moral commitment to health care ethics that is neutral … and can be shared by all regardless of their background is surely too important a moral prize to be rejected carelessly, or for the fun of being a “Socratic gadfly”’ (Gillon 1994: 333). However, as I will show below, the existence of a common and neutral moral language, independent of background, capable of mediating ethical dilemmas in the diversity of social sites where biomedicine is practiced (e.g., the GP’s office, the ER, the NICU, the ICU, the laboratory, the hospice, screening programmes, the ethics committee, outreach programmes) is precisely what is put into question by critical social science scholars. Although principlism owes much to the epistemological space identified above, it has been equally overdetermined by the social and political space in which it erupted, viz, the National Commission for the Protection of Human Subjects of Biomedical and Behavioural Research.8 The commission, which was set up by the United States Congress in 1974, published an influential report in 1979: The Belmont Report. The remit of the commission was to identify ‘the principles that should underlie the conduct of biomedical and behavioral research involving human subjects, and to develop guidelines which should be followed to ensure that such research is conducted in accordance with those principles’ (Gert et al. 1997: 73). At the time, there were a variety

170 José Julián López

of rules dealing with human research (e.g., the Nuremberg Code of 1947, the Helsinki Declaration of 1964, and the 1971 United States Department of Health, Education and Welfare rules). However, there was no clear understanding of how these rules were to be applied, the extent of their consistency, or the broader ethical principles in which they were anchored. Consequently, the commission was given the task of clarifying the situation (Gert et al. 1997: 73–74). In the words of the bioethicist Albert Jonsen (a member of the commission), the success of the commission rested on a new way of ‘doing ethics’ that involved ‘going beyond the often chaotic debate that rages around moral issues and beyond the private ruminations of scholars. In this new way, a public moral discourse began to evolve, in which a group of citizens seeks the facts of the case, asks for scholarly advice, and enters a debate with a view to resolution’ (1998: 101.) The commission identified a series of principles that would serve ‘as rational justification for decisions and policies’ relating to research on human subjects (ibid.: 103). The three principles considered germane to research on human subjects and embedded in the American ‘cultural tradition’ were ‘the principles of respect for persons, beneficence and justice’; these were to be translated into actual practice as ‘informed consent, risk/benefit assessment and the just selection of the subjects for research’ (ibid.: 103–104). Tom Beauchamp and James Childress, the intellectual entrepreneurs of principlism, would add a fourth principle, nonmaleficence, in their immensely influential text Principles of Biomedical Ethics. In his discussion of the 1974 Commission, Jonsen stresses that the commission demonstrated the feasibility of identifying principles capable of informing public decision making independently of disagreements at the level of ideology and abstract moral theory (1998: 99–106). In other words, he points to the emergence of a new modality of social communication capable of working across a variety of social spaces. Similarly, the classic elaboration of principlism by Beauchamp and Childress, in their four-times-reissued and widely read seminal text, is also premised on the notion that it is not necessary to arrive at agreement at the level of overarching ethical theories (Beauchamp is a utilitarian and Childress a deontologist!) or to build an understanding of ethics on a case-by-case basis to arrive at decisions of the moral acceptability of actions. Instead, they claim that what characterizes reflective moral behaviour is its appeal to ‘universal’ principles that any rational person would accept. Thus, they argue, ‘we conclude that

Beyond Cognitive Critiques 171

although flexibility and diversity in judgement are ineliminable, judgement generally should be constrained by the demands of moral justification, which typically involves appeal to principles’ (Beauchamp & Childress 1994: 105). The proposed principles, sometimes referred to as the Georgetown Mantra, are autonomy, beneficence, nonmaleficence, and justice. They, it is argued, provide a rational and universal ethical grid that can be deployed to inform moral decision making. What Is Wrong with Bioethics? For Beauchamp and Childress, the relationship between the four principles is conceptualized non-hierarchically: ‘in stubborn cases of conflict there may be no single right action, because two or more morally acceptable actions are unavoidably in conflict and yet have equal weight in the circumstances. Here we can give good but not decisive reasons for more than one action’ (1994: 105). However, as Paul Wolpe has demonstrated, in practice ‘autonomy has emerged as the most powerful principle in American bioethics, the basis of much theory and most regulation, and has become the “default” principle of applied principlism, the principle to be appealed to when principles conflict’ (1998: 43). It is this excessive modulation of the field by the principle of autonomy that many critics have found profoundly worrisome. For instance, Fox maintained that ‘from the outset, the conceptual framework of bioethics has accorded paramount status to the value complex of individualism, underscoring the principles of individual rights, autonomy, self-determination, and their legal expression in the jurisprudential notion of privacy’ (1989: 229). Unsurprisingly, bioethics has had little to say about the social and cultural conditions that ground ‘autonomous’ action. As most social scientists would agree, autonomous action is not the property of individuals but of the complexly ordered and multi-textured social relations in which they are positioned. Thus, it is not the case that physicians or other health professionals will be in a position to secure the conditions for autonomous actions for their patients or research subjects in the absence of wider structural figurations.9 In this context, a number of ethnographic studies have sought to document the types of structural conditions that undermine the viability of autonomy, precisely when autonomy (in the form of informed consent) is being pursued. For instance, Robert Zussman, in his conclusion of a study on Intensive Care Units, writes, ‘Even more, we might argue that physicians have turned the

172 José Julián López

imperatives of formal consent, whether to the initiation of new procedures or to the discontinuation of old ones, to their own purposes. Borne of and sustained by a doctor-patient relationship with neither intimacy nor a great deal of trust, formal procedures may become a means by which physicians withdraw from intense involvement with their patients. Protestations of patient autonomy in this context become little more than excuse for disengagement’ (1992: 221). The problem with the centrality of autonomy in American bioethics is not only related to the naivety of the notion that ‘ethical analysis leads effectively to ethical practice’ (Bosk 1999: 55), but also the assumption that the autonomous subject position is to be found in all social contexts and cross-culturally (see, for instance, Beeson & Doksum 2001; Kleinman 1997; Fox & Swazey 1984; Gervais 1998; Good 1994; Light & McGee 1998; Lock 2001). Moreover, sociologists and anthropologists have also sought to highlight the extent to which the focus on autonomy has marginalized other issues that might have rightfully fallen under its remit. To take an example, although bioethics has explored in great detail the ethical dilemmas that emerge in the context of treating extremely premature babies (Rothman 1991: 190–221), ‘relatively little attention has been paid to the fact that a disproportionately high number of the extremely premature, with very low birth weight infants, many with severe congenital abnormalities, cared for in the NICUs are babies born to poor, disadvantaged mothers, many of whom are single nonwhite teenagers’ (Fox 1989: 231). A related and extremely glaring omission is the way in which issues relating to gender and ethnicity have been entirely absent from bioethics’ field of ethical vision (Wolf 1996). Many scholars who have been critical of the single-mindedness with which bioethics has pursued the question of autonomy have linked this quest with the conceptual and discursive carrier of bioethical discourse, namely analytical philosophy (Anspach & Beeson 2001; Bosk 1999; Chambliss 1993; Fox & Swazey 1984; Fox 1989; Hoffmaster 1992; Hoffmaster 2001: 1–11; Light & McGee 1998; Kleinman 1997: 41–67; Wolf 1996). To the extent that the ideal of analytical philosophy is premised on the alleged ‘rational,’ ‘disinterested,’ and disembodied activity of logical argumentation,10 it has projected this vision on the social world itself. Thus, ‘following the lead of philosophical ethics, justification is regarded as a matter of providing “good reasons” for judgements, and that, in turn, is taken to require an appeal to moral rules or principles of moral theories. Bioethics, in this view, is situated in

Beyond Cognitive Critiques 173

rationality and generality. It prescinds the messy details and attachments that give our lives meaning and vigor, the nagging contradictions that make us squirm and struggle, and the social, political and economic arrangements that simultaneously create and constrain us’ (Hoffmaster 2001:1). Finally, equally problematic has been bioethics’ persistent inability to reflect on its own social, political, and economic role. Although the impulse behind bioethics was to question medical authority and technological development, in the end it is maintained that bioethics has reinforced and granted legitimacy to that which it purportedly questioned (Bosk 1999; Elliot 1999; Evans 2002; Fox 1989; Imber 1998; Rosenberg 1999; Kleinman 1997:41–67; Stevens 2000). The Limits of Cognitive Critique One of the elements that reproduces itself, almost with a viral logic, within the social scientific critique of bioethics is the latter’s inability to produce an empirically informed contextual analysis of the social sites and structural contexts in which ethical dilemmas emerge and the processes through which they become settled. This critical strategy creates a place of methodological and epistemological privilege for the practice of ethnography. As one of ethnography’s most frequently cited advocates argues, ‘the results of ethnographic investigations challenge both the dogmas that pervade the received view of medical ethics and the underlying philosophical model upon which “applied ethics” is predicated’ (Hoffmaster 1992: 1425). In other words, it is the ability of the social sciences to grasp and explain the dynamics of ethical and moral negotiation that should guarantee them a legitimate stake in the domain of bioethics. However, the attempt to open up a space for the social sciences in the bioethics discursive formation by arguing that they provide a cognitive mode of analysis that is more adequate to the task has been spectacularly unsuccessful. This absence of success is no doubt related to the fact that the status of bioethics as a legitimate tool for extracting ethical knowledge does not rest with the validity of its cognitive claims alone. It would be patently unfair to say that the sociologists and anthropologists working in the field are unaware of the wider structural context from which the claims of bioethics derive their legitimacy. However, if they recognize that it is neither method nor cognitive acumen alone that sustains bioethics, it is improbable that the

174 José Julián López

more adequate method of the social sciences will be sufficient to provide them with legitimacy or social traction. Elsewhere (López 2004) I have argued that the framework developed by Foucault in The Archaeology of Knowledge not only provides a conceptual schema for investigating the rules or relations that make knowledge claims legitimate through boundary work (Gieryn 1999); it also provides a way of locating these rules in a wider social space and political context. Thus, it facilitates the exploration of the social carriers, infrastructures, and channels that enable the circulation of knowledge claims and provide them with efficacy in different social terrains. It provides tools for investigating the materiality of the movement of knowledge practices. In The Archaeology of Knowledge, as is well known, Foucault describes the discursive formation in terms of four types of rules or relations: the formation of objects (What is the object of knowledge? Which are the social sites in which it emerges? Who is authorized to define it?), enunciative modalities (Who is allowed to produce knowledge of this object? From which institutional sites? What is the relationship between the producer of knowledge and the object?), concepts (What are the logical and methodological rules used for the production of knowledge? On what criteria are statements accepted or discarded? Which concepts are mobilized from other discursive formations?), and finally strategies (Which overarching theories or themes are deployed? Which theories from other discursive formations do they articulate with? What functions does the discourse play in related fields of non-discursive practices?). In my analysis of the emergence of bioethics in the United States (López 2004), I use the rules of discursive formation to explore the social processes and arrangements that enable the apparent frictionless movement of bioethics as a knowledge practice. It is beyond the scope of this paper to repeat the argument developed there, but a brief discussion of some of the rules will prove useful. For instance, if we take the rules for the formation of objects, it is clear that the grid of specification used to define the object of bioethical knowledge is constituted through the application of the normative templates of autonomy, justice, beneficence, and malificence to individuals in medical and biomedical encounters. However, as is evident from the brief summary on the ‘origins’ of bioethics above, the development of this grid is inseparable from the social surfaces in which bioethical dilemmas emerged. These would include, for instance, the state regulation of biomedical

Beyond Cognitive Critiques 175

research, the regulation of the medical and health care professions, and the individual as the locus of liberal autonomy. In turn, these are not independent of those positions from which individuals are socially sanctioned to describe particular instances as bioethical events, e.g., the bioethicist, the policy maker, the researcher, the lawyer, the nurse. Similarly, the rules for the formation of objects are intricately related to those concerned with enunciative modalities. Thus, the institutional sites in which bioethical knowledge is created – e.g., presidents’ commissions, expert panels, regulatory bodies, bioethics centres, IRBs – contribute to the moulding of bioethical objects as useful instruments of bureaucratic oversight. Moreover, if we examine the rules for the formation of concepts, it becomes apparent that bioethics’ shared methodological protocols with the discourse of law and medicine further strengthen its bureaucratic function while providing it with some of the legitimacy associated with these discourses. More broadly, if we look at the rules of formation of theories (or strategies), bioethics has no doubt participated in the institutionalization of the close, if contingent, relationship identified by Jürgen Habermas, among others, that links contemporary notions of freedom with the idea that individuals exercise this freedom through access to scientific and technological facilities (Habermas 2003: 24). Although highly schematic and condensed, this description of bioethics as a discursive formation is sufficient to illustrate both the limitations of the cognitive critique developed by sociologists and anthropologists and the challenges of using social scientific knowledge to pursue social change. To the extent that the cognitive critique focuses on transforming the rules of formation of concepts (i.e., methodology) and those concerning the formation of objects (i.e., What is an ethical dilemma? Where does it emerge?), it fails to locate itself within the wider social conditions on which bioethics’ legitimacy rest. Bioethics puts in circulation a highly mobile and flexible knowledge that moves through a number of different social domains (e.g., the laboratory, the hospital, bureaucratic apparatuses, the media, and the research centre). It provides a type of ethical capital that can be traded for social and democratic legitimacy, scientific morality, and bureaucratic restraint and can operate within a variety of discursive formations (e.g., medicine, law, the governance of science). A number of sociologists have claimed that ethics was a concern around which the emergence of classical sociology coalesced (Levine 1995; Shilling & Mellor 2001), and have used this to claim a stake in the

176 José Julián López

field of ethics. While there is some validity in this position, it does not follow that sociology is currently positioned as a socially sanctioned producer of ethical knowledge. The question, then, is not only whether the critical social sciences can produce better explanations of ethical decision making, but whether they can produce knowledges or contribute to the production of knowledges that will be able to circulate as widely, as socially and politically effectively, as bioethical knowledge does. For instance, is it possible to think about how the knowledge produced by the social sciences can link to new modalities of political engagement and discourses (e.g., human rights, cosmopolitanism, deliberative practices) as a way of democratically securing legitimacy for themselves? Cognitive critique is a legitimate and effective practice within the academy, but what are the modalities of practice that can be pursued to widen its scope of effectivity beyond the academy? Conclusion In this chapter I have argued that while the depth explanation component of the critical realist project is compelling, its emancipatory component is less satisfactory. As a number of critics have noted, it remains too broad and disconnected from the complex dynamics of social life to be useful or relevant. In addition, I have drawn attention to its cognitive idealism. That is to say, its failure to engage with the social infrastructure that enables communication, hence myriad forms of social and political effectivity. As a result, the critical realist conception of emancipation never steps beyond academic practice (i.e., cognitive critique) into political practice. That human emancipation is conceivable in the absence of politics is, to say the least, highly dubitable. I have drawn on the critiques of bioethics developed by critical social science scholars in order to illustrate how the social embeddedness of knowledges in discursive formations frequently make cognitive critiques relatively inefficient vehicles for the pursuit of social change. The point, however, is not to show that cognitive critiques will not suffice because false beliefs, oppressive practices, or ideologies are sedimented in social structures or institutions, which are difficult to shift, as is clearly the case with bioethics. Rather, it is to make clear that social life is not constituted by a special medium that privileges the circulation or effectivity of truthful knowledge, nor will this be achieved at some point in the future when all ideological distortions and social constraints have been removed, the Habermasian dream.

Beyond Cognitive Critiques 177

Social infrastructures, relations, and practice are necessary for the flow, communication, and effectivity of knowledge; truth, just like ideology, has broader conditions of existence and effectivity. If this is not taken seriously, it is unlikely that critical social science will provide the starting point for the politics of social change, let alone emancipation.

NOTES 1 See Sayer (1992). 2 Similar arguments have recently been developed in response to Tony Lawson’s (1999, 2003a, 2003b) attempt to incorporate critical feminist economics within the critical realist emancipatory sphere. See Barker (2003), Harding (1999, 2003); Nelson (2003), and Peter (2003). 3 It might be suggested that Bhaskar’s subsequent development of critical realism in a dialectical direction (Bhaskar 1993) may represent an opportunity to overcome this situation. This is, however, unlikely. Sayer, for instance, argues that this work represents a further retreat from substantive analysis, thus ‘the resulting impression is one of pulling global salvation out of the critical realist hat’ (1997b: 487n10). Similarly, John Roberts claims that ‘the “underlabourer” status of critical realism – the idea that critical realism should be viewed primarily as a method seeking to remove distortions about our knowledge of the world – severs methodological and theoretical endeavour from the system within which knowledge is sought’ (2001: 670). Moreover, I take Ruth Groff’s (2000) critique of Bhaskar’s concept of alethic truth as demonstrating how far Bhaskar is willing to go in order to empty epistemology of a social context by grounding truth in the very nature of being. 4 Bhaskar claimed that to the extent that experimental practice was concerned with isolating the operation of specific causal mechanisms, it was possible to argue that the world had to exist independently of our (transitive) conceptions of it as a complexly structured (intransitive) open system. 5 It is worth noting that those realists who have explored the complex mediations between social and natural structures (e.g., Benton 1993, Murphy 2002c, and Dickens 1992) are also those who are not affiliated with the critical realist emancipatory program. 6 For science studies’ turn to Foucault, see Pels (1997). 7 Tina Stevens in her critical history of the emergence of bioethics, however, claims that ‘considered within the larger narrative of the 1960s protest culture, the role of bioethics from the 1960s to the 1970s may be seen as one in

178 José Julián López the transition from critique to management. The cultural politics of bioethics is a history of the waning of the sixties milieu of challenges’ (2000: 151). 8 The picture is, of course, much more complex than can be adequately conveyed here. In addition, there were other key events such as exposés of highly questionable research practices as well as important structural trends, as in the case of the hierarchization of medicine around scientific and hospital medicine, the changing foundation for doctor-patient trust relationships, and the increase of state support for biomedical research. See Rothman (1991), Evans (2002) and Stevens (2000). 9 Informed consent is seen as one of bioethics’ most important achievements with respect to autonomy; however, it is not clear that informed consent promotes autonomy in the way envisioned by bioethics. See, for instance, Corrigan (2003), O’Neil (2002: ch. 2), and Pilnick (2002). 10 See Chambers (1999) for an analysis of the discursive and rhetorical devices through which the effects of rationality, disinterest, and logical argumentation are produced in writing.

9 Objectivity and Marxian Political Economy ROBERT ALBRITTON

Introduction ‘Objectivity,’ as in ‘objective knowledge,’ has acquired a bad name in recent years because it has sometimes been used to make rather authoritarian knowledge claims that deny the role of subjectivity in knowing or that overstate the knowledge claims appropriate to various types of object. All too often claims to objective knowledge do not stand up to close scrutiny or seem to imply some sort of out-of-body knowing. Yet, in the social sciences, theorists do continually try to replace less objective knowledge with more objective knowledge even if they do not explicitly state this particular dimension of their intellectual efforts. Indeed, the more objective a theory, the more adequate it is to understanding its particular object of knowledge, or, in other words, the more accurate or truthful it is. It is perhaps healthy for social scientists to be cautious in making claims about being objective, but, at the same time, debate could be watered down if intellectual cowardice led to running away from claims to increase objectivity. As Andrew Collier (1994: 13) puts it, ‘To claim objective truth is to lay one’s cards on the table, to expose oneself to the possibility of refutation.’ Claims to objectivity have certainly been abused in the history of Marxist theory, but that is all the more reason to openly confront the issue. Because of capital’s central role in shaping modern history, Marxian political economy, which aims at understanding the nature of capital and capitalism, must, to the extent that it is successful, be central to modern social science. In this paper I want to argue for a conception of objectivity that is sharply at odds with disembodied positivist conceptions and, while in some ways close to Roy Bhaskar’s critical

180 Robert Albritton

realism, attempts to rethink ‘objectivity’ in relation to capital’s unique ontology. This ontology, it will be argued, enables us to achieve ‘complete’ objectivity at the level of the theory of capital’s ‘generative mechanisms,’ while at more concrete levels of analysis only a degree of objectivity is possible, because the theory of capital’s ‘inner logic’ must be translated into the more concrete language of institutional practices where economic practices are mixed with other relatively autonomous practices, and where a less coherent and less necessary ‘externalized’ logic opens the possibilities for history altering collective agency. Bhaskar’s critical realism is my starting point, because in contrast to postmodern relativism, he argues that all science depends on the existence of objective structures and their accompanying causal powers and that the aim of science is to gain as much reliable knowledge of these structures as possible.1 Next, I agree with his claim that it is the ultimate nature of the object that determines its cognitive possibilities.2 And thus it follows that ontology must be the basis for epistemology, and that different types of objects may have different ontologies. Further, I agree with his claim that objects may be ontologically stratified in relation to each other or internally. And finally, I agree with his claim that social science should be critical in the sense that integrates both normative and empirical considerations.3 Although the three dimensions of science according to Bhaskar – real objects of which we attempt to improve our knowledge (the real), events caused by the powers of real objects (the actual), and our experience of events (the empirical) – is perhaps a good first approximation, I shall argue in this paper that the unique ontology of capital as an object of knowledge facilitates constructing a completely objective theory of its inner mechanisms along the lines achieved by Marx in Capital, and this is because social relations along with the concepts through which we think and reproduce them tend to become fully objectified when we assume for the sake of theoretical clarity and precision, as Marx did, that the commodity form is completely in charge of economic life. In other words, complete objectivity is achieved when we allow the historical tendency for the commodity form to dominate economic life to complete itself in theory. Reification Reification, or the becoming thing-like of social relations, occurs throughout society and history. Thus, for example, economies, states,

Objectivity and Marxian Political Economy 181

families, laws, and religions all have their own ways of becoming to some extent thing-like in different times and places. Generally speaking, a social structure or social relation becomes thing-like to the extent that it becomes like a natural fact: something that we think we have to accept or learn to live with as relatively unalterable. An example of the extreme case of an unalterable natural fact that we must accept is our inability to fly unaided. Reification is generally used to refer to social ways of doing things that are in principle alterable, but in the short term are treated as relatively unalterable either because, as they are deep structures, changing them would be difficult, or because they have become so second nature to us, it does not occur to us to alter them. The term ‘reification’ as used by social scientists generally has negative connotations since it implies treating socially constructed institutions uncritically as unalterable natural facts. Such reification makes it possible to accept even extreme injustices with statements accompanied with the appropriate sigh, like ‘alas, that is just the way things are.’ But not all reification feeds such uncritical surrender to what is. Some types may even be empowering. The core rules of grammar, for example, are reified in the sense that it would be difficult to alter them by conscious action, and yet they make communication possible. It follows that reification may not always be a bad thing. Indeed, exploring different types and degrees of reification, their interrelations, and their progressiveness in advancing human flourishing could be an extremely fruitful research programme for the social sciences. In fact, this research programme would be a kind of structuralism if we consider a structure to be a way of doing something that has become reified to some extent. Many kinds of structuralism, however, consider patterns of social relations to be either structures or not, but in contrast what I am suggesting is that social structures exist as a matter of degree having to do with degree of reification. In this conceptualization, social relations are more structural or less, depending on how difficult they are to alter by conscious human agency. In other words, the more structural a social relation is, the more enduring. In this paper, my aim is only to begin to explore this potential research programme with regard to theorizing capital and capitalism. In order to forestall unnecessary misinterpretation, I want to make it clear that I am not using ‘reification’ as Georg Lukacs does. I would not bother making this point but for the great influence of Lukacs’s usage. For Lukacs, ‘reification’ is the centre of a capitalist totality (‘expressive’

182 Robert Albritton

in Louis Althusser’s sense) in which arguably all social forms are simply different forms of reification. In contrast, I would argue that the high degree of reification that he imagines never comes close to existing in any historical capitalism but instead exists only as the thought experiment that is Marx’s theory of capital’s inner logic. One of the most remarkable achievements of Marx’s Capital is the clarity that it sheds on the unique reifying properties of capital, or, what I (1999) have elsewhere referred to as the ‘unique ontology of capital.’ Many critical realists have noted that generative mechanisms are of different types relative to different sciences. What I have argued at length elsewhere (1999), and want to claim here, is that the selfreifying properties of capital give it a unique ontology. To the extent that this is true, it would follow that it would be a mistake to think that there is only one ontology common to all social sciences. Exploring this uniqueness in depth is beyond the scope of this paper, but it is worth indicating, for example, that from the point of view of capital, individuals need only be recognized as legal persons capable of property transactions, and that in a purely capitalist society all individuals are swept up in the incredibly powerful and relentless dynamism of a commodity-economic logic that takes the form of ‘self-expanding value.’4 And this commodity-economic logic seizes hold of our most basic practices of material self-preservation and subsumes them to profit maximization even if this subsumption makes material self-preservation itself highly problematic for the majority. To give just one example of Marx’s thinking on reification from Capital: ‘The more acute and frequent these revolutions in value [economic crises] become, the more the movement of the independent value, acting with the force of an elemental natural process, prevails over the foresight and calculation of the individual capitalist … These periodic revolutions in value thus confirm what they ostensibly refute: the independence which value acquires as capital, and which is maintained and intensified through its movement’ (1978a: 185). And of course each crisis, by temporarily expanding unemployment and lowering wages, contributes to exacerbating the highly insecure life situation for workers. In a purely capitalist society, where it is assumed that all production is governed by a commodity-economic logic and that capital is therefore self-expanding value, all capitalists compete in trying to maximize their profits, and capital is totally indifferent to the human qualitative differences between capitalists, who are simply place holders in the society-wide meshing of capitalist markets and production processes.

Objectivity and Marxian Political Economy 183

As property owners, capitalists are legal persons by virtue of which they enter exchanges to maximize their profits. As legal persons, they can propose prices, but if their proposed price is too out of line (i.e., too high) with other prices for the same commodity, it will not sell. Thus, they are capable of agency as in the offer to sell a coat for $100, but if society’s markets deem the coat to be worth only $50, the sale will not take place. As the saying goes, Capitalists may propose but the market disposes. In a purely capitalist society without state intervention or other forms of human intervention, the only economic function of humans is to provide the energy for commodities to change hands and to move in a capitalistic rational way according to prices determined by markets. It is as if the structures both draw their power from human agency and at the same time use this power to dictate to agents what paths they must take in order to profit (for capitalists) or to maintain subsistence (for workers). Since in a purely capitalist society, economic agency is always trumped by economic structures that the agency makes dynamic, it follows that economic categories are reified social relations. Or in the clarity of Marx’s (1971: 483) writing: ‘In this quite alienated form of profit, and in the same measure as the form of profit hides its inner core, capital more and more acquires a material form, is transformed more and more from a relationship into a thing, but a thing which embodies, which has absorbed, the social relationship, a thing which has acquired a fictitious life and independent existence in relation to itself, a natural-supernatural entity … ’ If we imagine, as Marx did, a society in which economic life is totally governed by a commodity-economic logic, then all socio-economic relations will be mediated by the commodity-form in its various incarnations, and this mediation will displace direct person-to-person relations.5 The primary language becomes the purely quantitative language of price, as social relations are not only mediated by the commodity but are determined by it. The language of qualitative difference is only significant insofar as it matters quantitatively (i.e., makes a difference to profits), so that qualitative differences are never of interest in themselves to capitalists. The general formula for capital (M-C-M′: money-commodity-more money) implies that capitalists are simply place holders in a system of profit-oriented exchanges, that the commodity mediates all socioeconomic relations, that the agency associated with capitalistic legal personhood is strictly limited, that such agency is dictated by market prices and the quantitative exigencies of profit maximization, and that

184 Robert Albritton

the language of quantity subsumes the language of quality. The problem with the general formula M-C-M′ is that it does not set forth the basic source for the profit. In a closed system of equal exchanges, it is not possible for everyone to buy cheap and sell dear. The formula therefore requires a C that the capitalist can make use of to create more value than the C costs. If other means of production can only pass on the value they already have, it follows that only labour-power can create more value than it costs. Of course, the commodification of labourpower produces a sharp class opposition between capital and labour. Thus the expanded formula M-C … P … C′-M′ (where C = means of production + labour-power) makes it clear that M-C-M′ must subsume the labour-and-production process in order to locate the source of expansion within itself. It is only with this formula that capital becomes self-expanding value, ultimately able to subsume the major economic variables: land, labour, money, capital, technological change, etc., strictly to a commodity-economic logic. This enables capital in its purest forms to subsume the material reproduction of an entire society and in its reifying force to both expand and deepen its hold on the world. The basic social forms of capital can be most fully theorized if we assume that M-C … P … C′-M′ (and its related forms of the circuit) totally governs the fundamental material reproduction of a society. Furthermore, since the source of profit clearly lies within the formula, there is every possibility that the commodity form as a powerful reifying force will both be expansive and will penetrate more and more deeply into other social forms. Finally, to the extent that this formula can take on a life of its own (as in ‘self-expanding value’), it is like a subject, but an extremely one-sided and limited subject. It directs society but can think only quantitatively and can aim only to maximize profits. The tragedy of capitalist history is that this extremely limited subjectivity has been allowed to lord it over human beings who potentially may value many things and may revel in the qualitative differences of the world but can do so only by resisting or ignoring the pressures of capital. While most social structures have their peculiar reifications, the reifications associated with capital are perhaps unique in their dynamism and their forcefulness in shaping other reifications in the long run. This is not to say that other structures may not be relatively autonomous and relatively resistant to capital’s reifying force, but even an institution so deeply imbued with direct person-to-person relations as the

Objectivity and Marxian Political Economy 185

family has in the course of capitalist history become more and more shaped by the commodity form and its associated cash nexus. Using the language of critical realism, we might say that the theory of a purely capitalist society is the theory of the ‘generative mechanisms’ of capitalism, but in this case the generative mechanisms are encompassed by a commodity-economic logic that subsumes the fundamental use-values (land, labour, capital, etc.) that must be subsumed for the logic to operate the economy unaided. Collier (1994: 37) defines an experiment as an ‘artificial closure,’ but in this case the closure is not artificial since we allow the self-reifying and self-abstracting tendencies of capital in history (e.g., increasing commodification of labour-power up to a point) to complete themselves in theory. In contrast to the artificial closure typical of experimentation in the natural sciences, I will call this a ‘dialectical closure.’6 And the generative mechanism or deep structure that we isolate is the entire inner logic of capital as a set of necessary inner connections between economic categories. Because dialectics is important to the claim of objectivity in this case, I need to say at least a few things about it. Dialectical Knowledge Marx’s Capital is influenced by his knowledge of dialectical reason, even if, as formulated by Marx, it is not rigorously dialectical. It is the recent work of Tom Sekine (1997) that has drawn out and sharpened capital’s dialectical logic. A rigorous dialectical logic can only be formulated of a set of social relations that by its own structural dynamic can form a self-objectifying totality. Further, I believe that capital is the only such set of social relations because the perfected commodity form has the peculiar properties summarized in Marx’s formulations: selfexpanding or self-valorizing value. Since up to a point the commodity form tends more and more to subsume capitalist economic life to itself in history, it is possible in theory to extend the more capitalist to the most capitalist, where the commodity form is completely in control. Such a theory can, in principle, present a fully objective account of capital’s deep structure. The rigorous kind of dialectic that I have in mind requires an object that can be theorized as a set of necessary inner connections between categories that are simply different forms of the same substance (abstract labour). Further, there must be a primary category containing a fundamental categorical contradiction (mutually exclusive and mutually

186 Robert Albritton

dependent) that can propel reasoning forward until all basic categories required for the object to have a complete logic can be internalized into the primary category as simply different forms of it. The primary category, which Marx refers to as the ‘cell form,’ as has been argued, is the commodity form with its inner contradiction between value and use value. For Marx dialectical reasoning utilizes contradiction to drive forward a logic that moves from less specified economic categories to more specified until the interconnections between all the basic economic categories of capital are clarified. The least specified, most general economic category, and therefore the starting point that can encompass the whole that is to be known while requiring the fewest presuppositions is the commodity form with its internal contradiction between value and usevalue.7 It is this contradiction that drives the dialectic forward from the most abstract and indeterminate to ever more concretely specified economic categories. Thus, the dialectic can be conceived as value overcoming and subsuming successive use-value obstacles until value can, as self-valorizing value, achieve indifference to use-value precisely in the sense that capital can shift from producing one use-value to another with only short-term-profit criteria in mind. And this indifference is maximally expressed in M-M’, the form of interest-bearing capital, which represents the commodification of capital itself, thus closing the circle that started with the commodity form and making it appear that value in its externalized form of money can really expand itself indefinitely and think of itself only in mathematical terms without any reference to social relations.8 In a dialectic we know in a general way where we want to go conceptually, but instead of simply positing concepts, we try to let their necessary inner connections emerge as a logical unfolding. For example, the class relation emerges necessarily from the opposition between value and use-value when it becomes clear that self-valorizing value requires the exploitation of labour.9 By letting categories emerge in their necessary sequence (e.g., ‘money’ cannot come before ‘commodity’), dialectical reasoning at the same time fills in the meaning of previous categories. We start with the commodity form, but we only subsequently learn that for it to be a capitalistic commodity form, both the money form and capital form are required. Indeed, at the end of the theory we learn that not only are all the basic economic categories necessary in order to completely fill in the commodity form if it is to be fully capitalist, but also that all these categories have necessary ways of interrelating. And the necessity may be quite strict, as in the commodity

Objectivity and Marxian Political Economy 187

form strictly requiring the money form, or it may be looser, as in the rate of profit necessarily constraining movements of the interest rate, or the need to reproduce a productive working class necessarily constraining the movement of the wage-rate.10 Marx’s reference to the commodity form as the economic cell-form suggests an analogy that the cell is to biology as the commodity is to economics. And while as basic units of analysis of their respective sciences commodity and cell have certain similarities, here I want to emphasize their differences in order to clarify the unique ontology of capital. In order for a dialectical inner logic of biology to be possible, the following conditions would need to hold. 1 All the basic categories of biology could be encompassed within a theory that could understand the deep structure of biology as a necessary unfolding of the cell form such that each category would be a differentiation within the cell form proceeding in a necessary sequence from the simplest to the most complex. 2 There would be some tendency in history for all life forms to be increasingly subsumed to the motion of this cell logic, enabling us to theorize living organisms as in principle subsumed completely to this cell logic. 3 The cell form would contain within itself a basic contradiction of mutually exclusive and mutually dependent aspects such that the entire theory could be understood as a necessary unfolding of the cell form as resolutions of contradictions drive the logic forward through all the basic categories of biology. As a result, the theory of biology’s inner logic would have a necessary beginning, unfolding, and closure. 4 Since the basic categories of biology would primarily differ quantitatively and not qualitatively, their necessary inner connections could be represented numerically. The result would be a numerical theory of the necessary inner connections among all basic biological (i.e., cell-formed) categories. Though not a biologist, I doubt very much that a dialectical logic of this sort would make any sense in biology or any other natural science, and probably not in any social science other than the theory of capital’s deep structure. The fact that the unique ontology of capital lends itself to dialectical reasoning means that in principle it is the most objective theory possible.

188 Robert Albritton

‘Self’-Objectification of Social Relations We can never be certain that our knowledge of an other is objective, but objective knowledge is at least a possibility where we are dealing with a society-wide self-objectification that has an inner logic.11 In history it is never the case the all the inputs and outputs of capitalism are totally and securely commodified, as is assumed in pure capitalism, and yet the commodity form permeates social life to varying degrees, and as a result social relations may become increasingly woven into a cash nexus. But this means that social relations themselves become quantitative in the sense that they are governed by commodities that capital relates to purely quantitatively. Social positionality, then, becomes increasingly determined by its relation to commodities, and from the point of view of capital only those positionalities determined by self-expanding value are recognized. In this sense, the motion of objects constitutes a system, the result of which is continually moving prices, which come to govern socio-economic life, and to this extent, social life can be said to be objectified. In other words, the order of the system of objects (price signals) governs the order of social relations. Similarly, since the objects are related to purely quantitatively, social relations become abstract in the sense that they are abstracted from all qualitative differences. Now if we accept that capital as self-expanding value has a dynamic that forcefully expands and deepens the commodity form, it follows that a capitalist society is at least to some extent self-objectifying and self-abstracting. And it is perhaps for this reason that modern social science originates in political economy as a science based on the objectification of social relations through their quantification and on the abstraction of social relations from their qualitative differences. To some extent, Jacques Derrida’s concept différance can be viewed as a reaction to the injustices that arise from capital’s total indifference to qualitative difference except as it figures in profits. With concepts like commodity, money, capital, wage, profit, rent, and interest arising out of the practices of economic life, it becomes possible to begin thinking about their interrelations, because as these quantitative categories gain more purchase on economic practices, those practices become increasingly objectified and abstract, to the extent that one might say they are asking to be theorized, or perhaps that they have already gone partway to theorizing themselves. I know that this way of speaking seems to give objects the properties of subjects, but perhaps it is not so strange if we consider that the reifying

Objectivity and Marxian Political Economy 189

force of capital tends to objectify subjects and subjectify objects. Of course, capital cannot literally theorize, but it can prepare the paths along which human thought can be invited to flow. If capital is self-abstracting and self-objectifying, it might seem relatively easy to theorize its inner logic by simply completing these tendencies in theory, even though they are never complete in empirical history. We would then have a theory that is objective in the sense that the necessary inner connections among all the basic economic categories required for capital in the abstract and in general (i.e., without reference to the specificities of particular capitalisms in particular contexts) to reproduce and expand would be fully exposed. And while it is true that capital by its own logic helps us along in theorizing, Marx also recognized that this same logic is far from being easily grasped, as his theory of fetishism demonstrates. In my reading, Marx uses the concept ‘fetishism’ to explore all the ways in which political economy has been led astray by circulation forms that lack the agoraphobia of production relations. It seems to me that the principal reasons why the theory of pure capitalism is far from transparent are the following. First, we do not like to think that we have allowed ourselves to be objectified. It is much more congenial to our self-esteem to see commodities as objects fully subordinate to sovereign individuals and not as historically specific objectifications of us. Second, in actual history, objectification is never complete, allowing us enough agency to give us the illusion of being in control. Third, buying and selling seems to be both free and fun, and hence often ends up being at the centre of economic theory. Production as a dirty, noisy, grinding, oppressive place is best ignored. Fourth, by focusing on value, the capitalist economy can be pictured as harmonious (of course Marx does not do this), such that the qualitative use value resistances that value must subdue in order to achieve indifference to use value can in principle be ignored. Fifth, the way in which capitalism presents itself in daily life produces a series of misrecognitions that Marx referred to as ‘fetishism of commodities.’ Sixth, even if the above problems are avoided, there is still room for debate on how best to theorize the necessary inner connections among the core categories of capital, even though Marx offers us a good start in sorting this out. Thus the self-abstraction that follows on self-objectification is far from being a transparent reflection of empirical reality, nor is it simply a theoretical production that is not helped along at all by empirical

190 Robert Albritton

reality (it follows that both reflection and production theories of knowledge are incorrect). It takes ongoing and collective theoretical work over centuries for capital’s inner logic to be presented in any kind of adequate theory. Marx made great strides in this respect but could not have without the work of previous political economists. For example, he often reflects both on how Adam Smith and David Ricardo advanced political economy, while falling short of adequately understanding capital’s inner logic. In volume 2 of Theories of Surplus Value, Marx (1968: 106) writes: Adam Smith, as we saw above, first correctly interprets value and the relation existing between profit, wages, etc. as component parts of this value, and then he proceeds the other way round, regards the prices of wages, profit and rent as antecedent factors and seeks to determine them independently, in order to compose the price of the commodity out of them. The meaning of this change of approach is that first he grasps the problem in its inner relationships, and then in the reverse form, as it appears in competition. Ricardo, on the other hand, consciously abstracts from the form of competition, from the appearance of competition, in order to comprehend the laws as such. On the one hand he must be reproached for not going far enough, for not carrying his abstraction to completion, for instance, when he analyses the value of the commodity, he at once allows himself to be influenced by consideration of all kinds of concrete conditions. On the other hand one must reproach him for regarding the phenomenal form as immediate and direct proof or exposition of the general laws, and for failing to interpret it. In regard to the first, his abstraction is too incomplete; in regard to the second, it is formal abstraction which in itself is wrong. (emphasis in original)

As I understand this quotation, the problems of abstraction have to do with value’s seeming autonomy from use value or with the market’s seeming separation from production on the one hand, and lack of attention to the internal integrity of a particular level of abstraction on the other. The first leads the theorist to treat economic categories formally and mathematically either cut free from their underlying social relations or connected with them haphazardly. And the second confuses the theory of the ‘laws as such’ by directly mixing with it empirical material. Smith falls into the first error by taking the economic appearances of everyday life and raising them to the level of abstract theory without thinking through their necessary inner connections so

Objectivity and Marxian Political Economy 191

that quite conveniently land, labour, and capital get their rent, wage, and profit in proportion to what they contribute to the economic pie. In this Smith completely fails to think through the necessary socioeconomic connections between the use-values and their quantitative returns. In contrast, Ricardo confuses levels of abstraction by injecting concrete material at an abstract level where it does not belong; and by abstracting directly from empirical material, he produces formal abstractions that fail to adequately develop the mediations required for their accuracy, thus manifesting an extreme deductivism between theory and history coupled with a lack of theoretical mediations connecting the basic categories of economics as a necessary movement from the simplest to most complex categories. Here Marx no doubt has in mind Ricardo’s failure to problematize the transformation of values into prices by distinguishing surplus value from the various forms of profit, interest, and rent. To many the term ‘self-objectification’ may immediately trigger alarm bells of humanist essentialism or subject-centred theory. But here I am using the term in a structuralist way as resulting from a specifically capitalist form of reification. I mean by the concept a collective objectification of social relations as mediated by the commodity form. While it is capital that does a job on us, we can do a cognitive job on capital precisely by letting it have its way with us in theory. In having its way, capital reveals its deep character structure, and it is by understanding this deep structure and its impact on us that we can develop the most clearly informed and effective strategies for separating ourselves from it. Because we become partially objectified subjects, we can understand capital as a partially subjectified object. What makes this particular object of knowledge capable of being known objectively is not only that it is a reification of our own socioeconomic relations but also that it is a reification that results in a complete inner logic (or what Marx calls ‘the laws as such’). We become objectified because we become the energizers of a force field that dictates the main outcomes of our economic activities. And by letting the force field take over completely in theory, we can arrive at an objective account of capital’s logic. The logic starts with the commodity form and ends when that form subsumes capital itself in the form of interest-bearing capital. The fundamental commodity logic that interconnects labour, land, and capital is developed to the point where we can clearly grasp how, by its own dynamic, it can reproduce itself and expand.

192 Robert Albritton

The theory of capital’s inner logic (or deep structure, or laws as such, or set of necessary inner connections) is essentially the logic of a purely capitalist society where the rule of the commodity form through competition is allowed to display itself without interference. At the same time as he constructed his theory of capital, Marx was fully aware that pure capitalism had never existed in history, and would never exist in history, even though historical capitalism displayed strong self-purifying tendencies. It might seem, then, that the high degree of objectivity achieved in this case is achieved at the cost of being divorced from history. But this is not the case, since the inner logic consists simply in extending to completion in theory self-purifying tendencies that are readily apparent in history (increasing commodification of economic life). In what sense, then, might the theory of capital’s inner logic correspond to Bhaskar’s conception of the role of the ‘real’ in scientific theory? Or how does Bhaskar, given his conception of science, interpret Marx’s Capital? In The Possibility of Naturalism, Bhaskar (1979: 65) argues that Marx’s Capital is ‘an attempt to establish what must be the case for the experiences grasped by the phenomenal forms of capitalist life to be possible; setting out, as it were, a pure schema for the understanding of economic phenomena under capitalism, specifying the categories that must be employed in any concrete investigation.’ A critical analysis of this quotation will help to further clarify the differences between Bhaskar’s position and mine. If by ‘the phenomenal forms of capitalist life’ Bhaskar means the phenomenal forms taken by the law of value in a purely capitalist society, I would agree with this clause, but since ‘capitalist life’ implies to me concrete historical life within capitalist societies, I am most likely in strong disagreement with this quotation. Understood in this latter concrete sense, ‘phenomenal forms’ might include a law against embezzlement aimed at cottage weavers (in England in the eighteenth century), the non-recognition of married women as legal subjects prior to the late nineteenth century, hyperinflation in post–First World War Germany, the cold war, the stagflation of the early 1970s, etc. One question that emerges right away with these examples is, Which ones are to be considered capitalist economic phenomena? Outside of the theory of a purely capitalist society, and at more concrete levels of analysis, the categorization of phenomena as ‘economic’ or as ‘capitalist’ is often unclear, as many phenomena may be partially capitalist and partially economic. Indeed, arguably the best way to sort out these issues is to

Objectivity and Marxian Political Economy 193

refer to the more abstract level as a benchmark, which can serve as a basis for determining the extent to which various economic phenomena are influenced by capital’s deep structure. Furthermore, Bhaskar’s use of the concept ‘setting out’ makes it seem as if Marx is simply constructing a Weberian pure type. As I have argued above, it is not simply a question of the theorist ‘setting out’; rather, Marx explores and extends in theory a process of objectification that he is caught up in. To use a distinction developed by Alfred SohnRethel (1978), the pure schema is not simply a mental abstraction but is a mental abstraction that extends a real abstraction (exchange really abstracts from qualitatively distinct commodities). Or, in other words, the self-abstracting tendencies of capital that are only partially realized in history are helped along by the theorist. And this helping along may require conceptual innovation beyond the concepts produced by capital’s discursive practices, as in Marx’s use of the concept ‘surplus value.’ In contrast to Weber’s ‘artificial’ ideal-types, I would argue that the concepts of capital’s inner logic are dialectically necessary. Finally, I would argue that the aim of the theory of pure capitalism, is not to directly understand the economic phenomena under capitalism, many of which may not be fully capitalist or fully economic, but rather it is to understand which economic phenomena are most fundamentally capitalist and how their most fundamental capitalist forms are to be theorized. While the categories of capital’s inner logic may inform any concrete study of capitalism, categories that arise in a context of total reification should not be employed without modification in a ‘concrete investigation.’ I would say that the analysis of any concrete capitalist context would most often require numerous categories not to be found in the theory of capital’s inner logic, and that the categories of the inner logic should never to applied without modification directly to concrete contexts. For example, in order for money to be a commodity economically regulated, it must be gold and convertible paper, but the modern state has for various reasons come to maintain the commodification of money and to substitute fiat money for gold and convertible paper. Thus, being clear about the basic nature of capitalist money can inform our thinking about the diverse types of monetary system in history where often the state steps in to varying extents and in varying ways to take over the regulation of money from capital’s commodity-economic logic. In order to avoid the reductionist application of inner-logic concepts to concrete history, it is important to treat historical analysis as a distinct level of theory requiring concepts appropriate to this level.

194 Robert Albritton

Objectivity and History Objectivity in the case of the theory of pure capitalism does not refer to anything like an objective Archimedean fact but to an objective inner logic quantitatively connecting the basic capitalist economic categories as reified social relations. The theory of capital’s inner logic can clarify for us exactly what a capitalist commodity is and, by virtue of its basic properties, how it behaves in relation to other economic categories that assume a commodity form. On the one hand, this clarity can be an enormous help in theorizing concrete situations where the capitalist, quasi-capitalist, pre-capitalist, post-capitalist, non-capitalist, and prefigurative socialist are jumbled together (e.g., domestic labour, petit bourgeois labour, voluntary labour, forced labour, cooperative labour). On the other hand, the categories of this theory cannot be directly applied to concrete situations where reification and commodification are partial and are continually supported, resisted, or transformed by social agents and social forces. For lack of a better concept, we might say that the theory of capital’s logic can ‘inform’ our analysis of concrete situations as long as due attention is given to the sorts of refractions and translations required between distinct levels of analysis (see below). In this regard, I would modify Bhaskar’s distinction between real, actual, and empirical. It is not simply that the causal powers of the real in this case may be partially or totally blocked by other causal powers, but rather at more concrete levels of analysis the logic of the real itself must be translated into institutional practices that have less of a logic about them, since at the level of the actual we are studying a less, reified environment. This also means that objectification is less, and our claims to objectivity must therefore become limited by and relative to our particular subject matter, which may include an articulation of a variety of types and degrees of reification and resistances to reification. It follows that the theory of the real and the theory of the actual need to be distinct levels of analysis, since the causal powers of the real must be rethought, retheorized, and partially transformed at the level of the actual in connection with an array of relatively autonomous practices. The theory of pure capitalism presents the deep structure of capital as having a certain directionality, but this directionality is too abstract to be directly translated into a theory of concrete historical development. For example, the abstract theory tells us that capital will tend to concentrate and centralize, but we can have no idea what forms this

Objectivity and Marxian Political Economy 195

may take or how rapidly it may occur. We need a more concrete midrange theory to explore the types of concentration and centralization that may occur at different phases of capitalist development. For example, a whole variety of institutional arrangements must be in place before the sort of merger movements that occurred in the United States and Germany between 1870 and 1914 can be explained. Since arguably the most successful and typical modes of capital accumulation characteristic of this phase occurred in these two countries, they can serve as reference points for developing the material-type (not ideal-type) concepts needed to outline the patterns of capital accumulation typical of this phase.12 Like the theory of pure capitalism, this mid-range theory is primarily synchronic. But because historically specific economic, ideological, and political practices may have relatively autonomous temporalities (i.e., they do not change in lockstep with each other), the more autonomous the practice from capital’s structural dynamic, the more the synchrony at this level takes on a certain artificiality and refers not to a time slice at a moment of time but to a time chunk, a sort of compressed diachrony, or an extended period of time thought structurally. For example, both Germany and the United States develop oligopoly and forms of finance capital between 1870 and 1914, but state welfare policies have very different temporalities in the two countries. It follows that it is not easy to decide whether or not to include a degree of state welfare as being typical of this phase of capitalist development. This problem points to the need for historical analysis that enables us to study diverse temporalities as they interact in processes of change. Here structured agencies and agencied structures interact in continual processes of change and resistance to change. At the level of historical analysis, we can try to sort out the relative weight of various causal factors such as the specific weight of the SPD (Social Democratic Party) in forcing Bismarck to adopt a program of sozialpolitik. We study dominant types of agency in connection with dominant types of structural dynamic at the level of mid-range theory. These types in turn help us to sort out historical detail in ways that lead to the most effective causal analysis. In a sense capital’s logic is present at all three levels. At the most abstract level, capital’s inner logic is presented in its totality. At the level of mid-range theory, this logic is translated into dominant types of institutional practices characteristic of different phases of capitalist development with their phase-specific types of use-value obstacle. For

196 Robert Albritton

example, at this level crises’ tendencies may manifest typical forms that in turn are countered by particular types of policy options available to the state. At the level of historical analysis, the particular causes of particular crises can be studied. Levels of analysis enable us to think diachrony (i.e., interconnections across time) without destroying the synchronic analysis (i.e., interconnections across space) that are so important to the strength of structural analysis. Similarly, instead of limiting ourselves to general formulations about the relation between structure and agency, we can think these relations at different levels of abstraction where different degrees and types of reification and objectification are present. For example, we move from a level of analysis that converts individuals into objects that can speak and act and imagine themselves to be sovereign subjects, to a level where we consider types of subjectification and objectification in relation to typical forms of phase-specific capital accumulation, and finally to a level where structurally conditioned and partially objectified individuals and collectivities can assert their subjectivity by transforming structures. The concept of levels of analysis not only enables us to think types and degrees of objectivity relative to various theoretical objects but also enables us to make significant headway in sorting out seemingly intractable dualisms between structure and agency or between synchrony and diachrony. If I am right about this, then the notion of levels of analysis has not been given anywhere enough attention in Marxian political economy or in social science more generally. Objectivity and Morality The kind of objectivity sought by positivist social science requires the removal of all normative considerations and the ideal of an Archimedean standpoint where all the possible biases associated with any concrete social formation are more or less neutralized. It follows that for positivists, all forms of normative theory should be maximally separated from social science. But where the kind of objectivity is an objectification of social relations, it is quite easy and almost natural to integrate cognitive with moral considerations. After all, many forms of social self-objectification can in principle be altered, and we cannot help being intensely interested in forms of objectification that can both be altered and that constrain or block human flourishing.

Objectivity and Marxian Political Economy 197

Given that the inner logic of capital would subordinate all human values to short-term-profit considerations, the moral deficiencies of capitalism are legion. Indeed, a great deal of what is referred to as ‘civil society’ in capitalism is a desperate scrambling to deal with the destructiveness continually generated by capital’s logic. A brief list of some of the more morally problematic consequences of capitalism will illustrate this point. First, the competitive nature of capitalism tends to overdevelop the egotistical side of humans while underdeveloping the human capacity for cooperation, generosity, and care. The generosity that still exists in capitalist society is not because of capitalism but despite it, and it is often motivated by the guilt associated with exaggerated egotism or with the image problems of large corporations who must maximize profits even when the long-term social costs associated with such profit-making may be immense. Arguably, humans have an immense need for generosity, which capitalism continually thwarts. Second, the highly unequal distribution of wealth under capitalism creates enormous moral problems. Capitalism has worsened the qualityof-life for the majority of people in the world who daily face every kind of quality of life shortage from adequate food to clean air, from quality education to clean water, from quality health care to a secure source of adequate income. Capitalism makes a total mockery of any theory of distributive justice. Third, the efforts by dominant capitalist powers to avoid the pressures of periodic crises lead to intense inter-imperialist rivalry resulting in almost continual warfare of one sort or another. The need to maintain profit rates results in transfers of wealth from the public sector to the private sector, from the developing to developed countries, from competitive to monopoly sectors, from the poor to the rich, and from butter to guns. It is the same pressure that blocks environmental policies that might inhibit profit making no matter what the long-term environmental costs. Finally, business cycles breed continual insecurity and anxiety of the sort that undermines ethical behaviour. Fourth, the subordination of quality (use-value materiality) to quantity (profit) undermines human concerns for the environment, for beauty, and for truth, while making otherness as qualitative difference threatening. In short, instead of integrating quantity and quality in ways that maximize the quality of life for the majority of humans, capitalism subsumes quality to quantity in order to maximize the

198 Robert Albritton

profits of the few. As a result, when profits are threatened, any other human value may be sacrified if by doing so profits might recover. Fifth, reification, by handing control of our economic life to markets, promotes a formalist type of democracy, giving little scope and encouragement to real citizenship.13 Indeed, the limited democracy that has been consistent with capitalism in the past is daily more and more undermined by the power of big capital. While each of these points could be expanded on, for now I leave this issue for future writings. A social science that focused primarily on types and degrees of reification and their interactions could potentially be a social science of responsibility, in the sense that it would help us to consciously adopt the degrees and types of reification most conducive to human flourishing, or, in other words, take responsibility for our social relations. Given the current resources put into education globally, such a reorientation of social science could bring enormous benefits. And yet, while we live in the type of capitalist world that we do, such a reorientation is highly unlikely. Capitalism needs a social science of control and manipulation and not of human flourishing and moral responsibility. Conclusion What conclusions about objectivity in Marxian political economy can be drawn from what I have been arguing? First, I have argued that it is possible to have an objective theory of capital’s inner logic by letting that logic in theory objectify us. In a purely capitalist society, we provide the energy that moves commodities, and commodities provide the signals that direct this energy. Second, this theory is objective in the sense that no other theory of capital’s inner logic is possible. This follows from the fact that the inner connections among the core economic categories are necessary connections. Of course, details on how best to formulate these necessary inner connections may differ from theory to theory, but the basic necessity of the connection between categories constrains theoretical divergences. Third, it is objective in the sense of being a complete theory of capital’s deep structures. If a commodity-economic logic is to materially reproduce a society, then nothing must be left out that is in general required for accomplishing this task. The theory must show that all the basic elements of economic life such as land, labour, tools, money, funds, etc. can in principle be managed by a consistent and workable commodity economic logic. The theory in this case also demonstrates

Objectivity and Marxian Political Economy 199

that land, labour, technology, and money raise special problems for a commodity-economic logic. And that is why in actual capitalist history their commodification is seldom total, and the degree that does exist nearly always requires extensive political and ideological supports. Fourth, it is objective in the sense of being real. And how do we know this? Because it emulates capital’s self-reifying force to complete in theory what capital can never complete in history. But the degree of objectivity achieved at the level of the theory of pure capitalism cannot be achieved at more concrete levels of analysis. The recalcitrance of use-values at the level of history assures that the real in Bhaskar’s sense can never be fully realized in the actual or empirical. Furthermore, though capital’s inner logic may be said to exist as the real of capitalism, at the level of history it becomes a particularly strong causal force articulated with other relatively autonomous forces. It follows that we cannot make judgements about the causal efficacy of capital’s inner logic at the level of history without studying it in connection with other types of reifying and objectifying forces. Fifth, while the theory of pure capitalism can be stated without moral judgement, it urges upon us a consideration of morality. It does so because the objectification entailed in capital’s logic is so obviously impoverishing relative to the rich possibilities of human flourishing. In the case of Marxian political economy, we have a form of layered objectivity across distinct levels of analysis. The most abstract level of analysis is the most objective, and by informing mid-range theory this less objective level of analysis can improve its degree of objectivity, as can historical analysis in so far as it can be informed by the two more abstract levels. The theory of pure capitalism, by presenting a complete inner logic, is more powerful than Bhaskar’s real that can never be completely known. Both the actual and the empirical would only exist at my level of historical analysis, where our knowledge of the actual would be more adequate or less adequate depending on the strength of our theories. In so far as the actual refers to complex historical events, we would not expect our theories to bring us close to anything like the degree of objectivity possible in the theory of capital’s logic. At this level questions of causality are always open to further elaboration and refinement as we learn to better weigh and trace the resonances between an array of contributing causes. Sixth, the conception of levels of analysis that I am advocating constructs effective mediations between abstract levels of theory and practice. It avoids economistic strategies that derive from a too deductive

200 Robert Albritton

approach to abstract theory and voluntaristic strategies that in reaction to the politics of the theoretically deduced ‘correct line’ tend to downgrade the potential informative powers of theory in favour of the spontaneity of the masses. Good strategic thinking is essentially a kind of future-oriented historical analysis that attempts to find a balance between more short-term, reforming, and unifying demands and more long-term, revolutionary demands that can grow out of previous radicalized mass mobilizations. Above all, I want to assert the importance of a more objective Marxian political economy against the disarming of the left that flows from the anti-theoretical and ultimately anti-political spontaneism that has become such a trendy consequence of post-structuralism.

NOTES 1 ‘… there are enduring structures and generative mechanisms underlying and producing observable phenomena and events … These structures are not spontaneously apparent in the observable pattern of events; they can only be identified through the practical and theoretical work of the social sciences’ (Bhaskar 1989: 2). 2 ‘For realism, it is the nature of objects that determines their cognitive possibilities for us …’ (Bhaskar 1989: 25). 3 See Bhaskar, 1986. 4 Marx frequently refers to capital as ‘self-expanding value.’ 5 See Albritton (2003) for a brief defence of this interpretation of Marx’s Capital backed by numerous citations. 6 I believe that the inner logic of capital is a dialectical logic and that this adds a great deal to the claim that this logic is ‘real’ and ‘objective.’ Given the complexity of arguing for this, I only hint at it in this paper. See Sekine (1997), Kourkoulakos (2003), and Albritton (1999) for extended argumentation on this issue. 7 ‘As the commodity-form is the most general and the most undeveloped form of bourgeois production …’ (Marx 1976: 176) ‘… the exchange of commodities implies contradictory and mutually exclusive conditions. The further development of the commodity does not abolish these contradictions, but rather provides the form within which they have room to move’ (ibid.: 198). 8 ‘We started with money as the converted form of the commodity. What we arrive at is money as the converted form of capital … (Marx 1971: 467). ‘The

Objectivity and Marxian Political Economy 201

9

10

11

12 13

formation of interest-bearing capital, its separation from industrial capital, is a necessary product of the development of industrial capital … (ibid.: 471, emphasis in original). ‘As soon as production by way of wage-labour becomes general, commodity production must be the general form of production.’ (Marx 1978a: 119) ‘… capitalist production is commodity production as the general form of production, but it is only so, and becomes ever more so in its development, because labour itself here appears as a commodity’ (ibid.: 196). Many reject Marx’s theory for the supposed incoherence of the labour theory of value. See Sekine (1997) for an explanation of the necessity of the labour theory of value. ‘Self-objectification’ is a little misleading, but I do not know of an expression that means social-relations-objectifying-themselves. ‘Self-objectification’ seems to put objectification under the category ‘subject,’ but clearly I do not mean it is this sense. For an extended discussion of the difference between an ideal and material type, see Albritton 1999: ch. 4. See Kathryn Dean’s (2003) important book on capitalism and citizenship.

10 Why Is This Labour Value? Commodity-Producing Labour as a Social Kind HOWARD ENGELSKIRCHEN

Introduction First, I don’t ask, Why is this ‘labour value’? but instead Why is this labour value? The latter question, I hope to show, is decisive for social enquiry. It opens on an understanding of the social composite of labour and form, of enformed social labour, that is the basis for commodity production. As a result, an answer to it can ground our understanding of those societies that depend on the production and exchange of commodities. For Marx, the question’s significance seems to me to reflect the influence Aristotle had on his thought. Moreover, what he did with it prefigures contemporary methodologies of critical scientific realism: we may characterize the form of social labour that produces commodities as a natural kind and offer a real definition of it. As a consequence, asking, Why does labour take the form of value? shows not only the continuing relevance of Marx’s work as a social theorist but also how it continues to anticipate what we may ask of social science today. The Critique of Ricardo We can begin with Marx’s relation to David Ricardo, an issue that has become something of a hot topic in the study of Marx and one that goes to the nub of how we read him today (Milios et al. 2002). Marx judged that Ricardo had made an enormous scientific advance by showing clearly that labour was the substance of value. But he thought also that Ricardo had ignored questions of social form (1971: 137; 1967a: 80–81). From this, over the past quarter century a powerful

Why Is This Labour Value? 203

tradition has developed emphasizing that the understanding of capitalism must begin with the analysis of the social form of value. According to this tradition – called Value Form Theory or ‘new’ and also ‘systematic’ dialectics – value is typically understood as a category of exchange reflected in the forms of valorization, exchange value, and money, and the idea that value may be characterized as the labour materially embodied in commodities is rejected. ‘Substance,’ conceived of as labour, and labour conceived of as a physiological expenditure of effort, a natural phenomenon, are counterposed to social form and rejected as the basis of value analysis; instead, value, often understood as empty of content except as constituted by money in exchange, is taken as primary. By this measure, Marx is considered to have broken incompletely with classical political economy, and his analysis in Capital is thought to be still clumsily burdened with Ricardian notions of embodied labour; as a result, Marx fails to present systematically or coherently a thoroughgoing form analysis of the capitalist mode of production. Much reconstruction is required, and we get new theories of labour, value, exploitation, and so forth. (Arthur 2002, 1993; Taylor 2000; Reuten 1993; Reuten & Williams 1989; Eldred & Hanlon 1981). Once your attention is called to it, no one can read the first chapters of Capital without recognizing the significance of ‘form’ to Marx’s analysis: there’s the commodity form, the value form, the natural form, the elementary form, the accidental form, the equivalent form, the general form, the money form, and so on. ‘Form determination’ is essential to his investigation. But, in fact, while calling attention to the question of form, Value Form Theory has pretty systematically noticed only one-half of Marx’s critique of Ricardo, and this has muddied everything. Christopher Arthur (1979: 68), for example, wrote, ‘By the value form Marx means the form of appearance of value.’ But Marx’s critique of Ricardo was double: the failure to develop value’s forms of appearance was part of it, but, even more fundamentally, Marx also noticed that Ricardo had not asked why labour takes the form of value in the first place. Here is Capital’s presentation of Marx’s fundamental point: ‘Political Economy has indeed analyzed, however incompletely, value and its magnitude, and has discovered what lies beneath these forms. But it has never once asked the question why labour is represented by the value of its product and labour time by the magnitude of that value’ (1967a: 80). This is an interesting question. It does not ask how it is that value appears in exchange in the form of money.

204 Howard Engelskirchen

It is connected to that question, but it asks, more fundamentally, how it is that value comes to be what it is. Marx and Aristotle At some point, reflecting on Marx’s challenge to Ricardo, I realized I’d heard the question Marx posed before, or at least something very close to it. In German the question is specifically, Why does this content assume that form? (‘Sie hat niemals auch nur die Frage gestellt, warum, dieser Inhalt jene Form annimmt’ [1970: 95]). I realized the place I’d heard it was in the decisive chapter of the pivotal book of Aristotle’s Metaphysics, chapter 17 of Book Zeta. It was the fundamental question posed, in a way the fulcrum on which the whole structure of the Metaphysics turned. Scott Meikle calls Marx an ‘Aristotelian in metaphysics’ (Meikle, 1991: 296); this echo, therefore, seemed an important link to follow. Aristotle argues that ordinarily when we explain, we try to show how we predicate one thing on something else. Why do animals have joints? Why is there noise in the clouds? But we often mislead ourselves when the predication is implicit – constitutive rather than attributive – and we then ask something Aristotle thought unhelpful. What is a man? he thought such a question. Here we are not asking about some attribute or property that a person has, like smooth skin or dark hair. Instead, we are asking what it is to be human. We are asking what it is that constitutes a human person as human.1 For this, Aristotle thought, we needed a question that located the generative basis of fundamental change. We want to know how this thing that is human comes to be so. We are not investigating changes that we attribute to a subject the way dark hair turns grey or skin wrinkles. We’re interrogating instead the kind of change that causes the subject to be what it is in the first place. For example, we shingle a roof and notice the house has changed, but what has changed when fire burns it to the ground? If we say the timber has changed to ash, this is true, but the timbers are not the house. They are only the material for the house. The question we need to ask is why these materials constitute a house. If we ask, What is man? we want to know not that man is a thing of flesh and bones, but instead why these flesh and bones constitute a man. Why, in other words, does this content assume that form? And Aristotle’s answer to this question provided for him the key to the investigation of the Metaphysics: the basic reality of things, primary substance, primary being in

Why Is This Labour Value? 205

the most fundamental sense, is the constitutive form of things. A thing’s causally potent constitutive form is the most fundamental content it can have. In other words, and to abbreviate necessarily, Aristotle was after the underlying structure or principle of an entity that determined and conditioned or organized its development (Lear 1988: 19). But to make this clear, he found he had to differentiate two ways in which the substratum could be understood to be underlying: ‘for example, an animal may be the subject to which attributes are referred, or [the substrata] are materials to be given actualization’ (Aristotle 1960: 158–159 [1038b]).2 If we study forms of appearance we study the first of these – how a subject as substratum manifests itself in its attributes. If we study the second, we study the way the materials which make up an entity are formed so as to constitute it; in a bit of Aristotle congenial to the street, we study the ‘what-it-is’ of a thing (1960: 134 [1029b]). Now in the second case, for the things of nature, matter is substratum, but as such it is not the primary reality – for itself, formless, matter can’t be considered more than ‘undetermined possibility’ (Reale 1980: 212) and is impossible really to think of except hypothetically. As it actually appears, it always exists as constituted by form, enformed, and as enformed, it is causally potent, a power in the course of actualization. Thus, the significance of the question at the end of Book VII of the Metaphysics lies in this – it gets at the potent form that ultimately organizes the power and processes of development of an entity. It does this by locating that power in its fundamental causal structure or form. In the Metaphysics, Reale argues, ontology, the study of being, just reduces to ousiology, the study of substance. And primary substance is, in the most basic sense, an aetiology, the study of a thing’s generative essence or constitutive cause (1980: 358). As a consequence, it is important never to reduce our understanding of the Aristotelian concept of substance, ousia, to an ontology of things (Kosman 1994: 212). The most basic reality of the world for Aristotle must be understood as structure or power or activity (Moravcsvk 1991; Kosman 1994). It is always an underlying nature in the process of actualization, an organizing principle that can account for how an entity might develop or persist as the kind of thing it is. This distinction between the constitutive and attributive form of an entity is exactly the sort of distinction Marx thought political economy had to make. Ricardo had discovered labour as the substance of value, but he did not know how it was enformed. He argued that the value of

206 Howard Engelskirchen

a thing could be attributed to the amount of labour expended on it, but he didn’t show how the social form of the activity of labour was constituted. By contrast, for Marx the concept of a bare, physiological expenditure of labour is best understood the way Aristotle understood matter. As the expenditure of labour in general, it can only be conceived of, like matter, as formless, no more than undetermined possibility. That is, labour only ever occurs historically within specific social forms, as enformed. And just as to understand a thing we must explain how it is constituted by matter and form, how it is enformed, so too if we wish to understand a mode of social labour we must locate the generative structure that organizes the developmental process of its actualization. This is its constitutive form, the historically specific underlying structure that makes it what it is. Labour’s constitutive form is not a property of it the way sweat is a property of a person expending physical energy. Labour’s constitutive form is not something it has, the way a physical product has density or texture, but the form it is, the structure of the actual process of production that makes the product a commodity in the first place. The point can be underscored by recalling Marx’s claim at the beginning of section 2 of chapter 1 of Capital that he ‘was the first to point out and to examine critically this two-fold nature of the labour contained in commodities.’ Yet Ricardo had distinguished use value from exchange value. But for Ricardo exchange value was an attribute of the product measured by the quantity of labour expended on it, and he did not investigate social labour’s constitutive form. For Marx on the other hand, the products of labour, insofar as they were values, expressed a relation – a relation of persons to their labour expenditure3 – and he then went on to make precise exactly what the relation was that could account for the necessity of comparing the labour of individuals as values. We turn now to the characterization of this social relation. The Social Form of Commodity-Producing Labour How shall we characterize the social composite of activity and form that constitutes commodity-producing labour? Elsewhere I have shown how Marx’s analysis in section 2 of the first section of Capital I’s first chapter presents the social form of commodity labour (2006).4 Here I want to show how this is worked out in a section of the Grundrisse, a draft for Capital. In discussing the form determinations of capital, Marx looks back to what he calls the ‘simple determination’ that generates

Why Is This Labour Value? 207

exchange value. He writes, ‘In the first positing of simple exchange value, labour was structured in such a way that the product was not a direct use value for the labourer, not a direct means of subsistence. This was the general condition for the creation of an exchange value and of exchange in general. Otherwise the worker would have produced only a product – a direct use value for himself – but not an exchange value’ (1973: 266–267; 1986: 197). The point is decisive: this is a structuring of labour itself. Moreover, it is a causally potent structuring. The labourer produces use values that are of no direct use to herself. Because the labourer here is also assumed to produce separately from others, fully responsible for her own production, in order to obtain the means to sustain her own existence and to produce again, she must resort to exchange.5 The structure therefore is one that relates the labouring subject necessarily (1) to nature – she produces products not useful to her – and (2) to others – she produces independently. I’ve characterized this structure as a form of ‘interdependent autonomy.’6 It is a relation of production, the form determination we’re looking for. It is the generative causal structure that forms the basis of value and the commodity form. Notice that by abstracting in this way to the simplest determination of the commodity form itself, we lose the level of precision required to locate the actual relation of direct possession connecting the labouring producer to the means of production other than in the two respects just specified. Here’s what I mean: Marx’s reference to ‘the worker’ and my use of the gendered personal pronoun can both mislead insofar as they tend to evoke the image of an isolated individual labourer who is part of a mode of petty commodity production. But to suppose that the simplest determination of the commodity implies necessarily a ‘simple commodity mode of production’ would be to confuse levels of abstraction. We here characterize the underlying structural relation that characterizes the commodity form, not the social totality referred to by the concept of a mode of production. Thus, the commodity form may have its source in units of production as diverse as the farm of an isolated peasant, an artisan’s small business, a patriarchal estate, a slave plantation, a capitalist enterprise, or a worker collective during the transition to socialism. That is, in his analysis of the commodity Marx abstracts to two bare facts only: whatever the relation of direct possession that connects the working individual to the means of production, we know actual living labour is executed under conditions of separation from other units of production and occurs in conditions where the product is not useful except to others (Bettelheim 1975). From the existence of a

208 Howard Engelskirchen

commodity we know living labour has been carried on separately and for others. To go further, to know, for example, how the commodity becomes the universal form of the product of labour requires an additional, different, and more concrete analysis. Nonetheless, we do have the composite of social labour that constitutes the commodity form of the product of labour. Recall Marx’s observation in the preface to the first German edition of Capital: ‘But in bourgeois society the commodity form of the product of labor – or the value form of the commodity – is the economic cell form.’ (Marx 1967a: 19). The structure of living labour identified in the passage from the Grundrisse above is the underlying causal structure that accounts for this. From Aristotle to Contemporary Scientific Realism For Aristotle, the things of the world were composites of matter and form, and Marx applied this to the study of the historical forms of labour: he found the explanation for value and its expressions in exchange in a particular configuration of living labour. I turn now to show that this structure of labour may be considered a social kind. Preliminarily, a word on terminology: I take it that if natural kind methodology can be extended to the study of social life, then we can refer to the causal mechanisms of society as falling under natural kinds. Typically, it is convenient to speak of social kinds, and I will do so here, but this is not meant to mark off a difference other than of subject matter. The naturalism I appeal to is unqualified. That is, we can speak indifferently of natural kinds in the social or natural world. As an additional point, if we do succeed in identifying and describing the causal structure of a natural kind, we say we have offered a real definition of it. Perhaps two emphases characterize most fundamentally work on kinds in the contemporary philosophy of science. First, there is a recognition that natural-kind terms can serve the objectives of science by enabling us to pick out the fundamental causal structures studied by different sciences. In this respect, natural-kind theory reflects a blend of both the attention of working scientists to the dispositional properties of things (Wagner 2001) and also the attention of philosophers of language to the causal theory of reference (Kornblith 1993). Second, there is a recognition that natural-kind terms function to accommodate the demands of our practice – scientific, productive, or political – to the

Why Is This Labour Value? 209

causal structures of the world (Boyd 1999a, b). That is, our practice, natural or social, requires that we get it right about the way the world is and that we be able to communicate accurately about this. To put it in terms familiar to progressive social theory, we interpret the world to change it. When we get it right, natural-kind terms facilitate our ability pick out the causal structures of the world and make our use or transformation of them possible. According to Stathis Psillos (1999), the concept of ‘kind-constitutive properties’ best characterizes the features of an entity to which we refer in our use of natural-kind terms. This concept identifies an entity’s essential causal description. Thus kind constitutive properties like the combination of hydrogen and oxygen in a water molecule allow us to explain how an entity behaves and how it persists as the kind of thing it is. Moreover, they also account for, and causally produce, the body of information we have about the thing (1999: 287– 288). As Psillos explains, the causal origin of the information that water is a colourless, odourless liquid that becomes a solid at zero degrees celsius is found in the chemical bonds joining atoms of hydrogen with those of oxygen. Commonly we expect the sensible properties of a thing and the generative mechanisms that account for them to differ fundamentally – the atoms of hydrogen and oxygen that make up a water molecule aren’t wet.7 That is, while the manifest properties of a thing may trigger our initial identification of it – we notice tigers have stripes – generally such properties are inadequate to secure the reliable extension of the term to all members of a kind. A liquid, Psillos reminds, can be colourless and odourless but kill you if it is not H2O (1999: 285). In other words, it is the fundamental causal properties of a thing that determine how it will behave. As Günter Wagner has recently suggested, the natural-kind problem is essentially a search for the ‘fundamental units [that] play a causal role in a process’ (2001: 6). Richard Boyd refers to the homeostatic clustering of causally important properties or mechanisms – those responsible for a thing’s ability to maintain and reproduce itself as the kind of thing it is. A comparable distinction between the underlying causal structures of social life and the manifest appearances they generate was of course an abiding theme of Marx’s work. In ‘Notes on Wagner,’ his last writing on political economy, he made fun of the abstract method of German academicians such as the economist Adolph Wagner precisely because they thought they understood a thing when they had

210 Howard Engelskirchen

observed its manifest properties (Marx 1975: 201). In this they acted like ‘the old chemists before the science of chemistry: because cooking butter, which in ordinary life means butter pure and simple … may have a soft state they called chlorides zinc-butter, antimony-butter etc … [emphasis in original].’ The common word ‘butter’ comes to be applied to all manner of substances with a soft malleable consistency. But to refer to the sensible features that compounds such as zinc chloride and antimony trichloride share with edible butter tells nothing about their constitutive structure, nor can we use them to give a reliable account of what such compounds are or how they behave. For that we need an understanding of their molecular structure, knowledge ultimately embedded in background chemical theories. Marx drew this lesson from his study of science and found it applicable to the study of political economy. Recall his reference in the ‘Afterword to the Second German Edition’ of Capital to tracing out the ‘inner connexion’ of the forms of development of the material studied (1967a: 19). Making a distinction between the manifest properties of a thing and its fundamental causal structure invited a consequence that Marx was fully aware of – there is no way to identify structures that are essential except through and by means of background theories and hypotheses. Psillos writes, ‘Only theories can tell us in virtue of what internal properties or mechanisms, as well as in virtue of what nomological connections, a certain substance possesses the properties and displays the behavior it does. Similarly, only theories can tell us in virtue of what internal properties an item belongs to this rather than that kind. And only theories can tell us whether a certain collection of entities, samples or items is a candidate for a natural kind’ (1999: 288). A qualification on the use of ‘internal’ here: the basic molecular structures of kinds in, for example, chemistry may be thought of as internal, but this does not generalize to kinds, natural or social, that can only be characterized by means of relations that are external, nor is this a mistake Marx made.8 But the main point Psillos here makes definitely does generalize: always our grasp of something, whether of water or value, depends on a background of theoretical knowledge. Enthusiasm for Social Kinds Despite the appeal of natural kind theory to many working scientists (Rieppel 2004; Wagner 2001; Keller et al. 2003), the extension of kind

Why Is This Labour Value? 211

methodology to the study of social life has been halting. For natural scientists, the appeal no doubt rests at least in part on their demonstrable success in identifying the physical and chemical make-up of things. In fact, for the natural sciences, a sense that the constitutive causal structures they study make it possible for us to characterize what they are and how they behave has entered popular consciousness – everyone knows water is H2O, and if you explain to someone that the real definition of water is H2O you will have no difficulty explaining that this means that the chemical formula captures the nature of water such that anything that is H2O is water and anything that isn’t, isn’t. But there is no comparable recognition for the achievements of social science. No doubt there are a variety of reasons for this, not the least of which must include the ideological and even apologetic character of much social theory. In the space available, I can only address briefly two important tendencies often thought to compromise the extension of kind methodology to the study of social life. First, traditional thinking about natural kinds tends to assume models from physics and chemistry, and such models are often thought to reflect necessary and sufficient membership conditions that are exceptionless, ahistorical, and unchanging, Thus, working at least in part on the basis of such assumptions, philosophers of social science have often concluded that the categories of human action do not lend themselves to treatment as kinds.9 A most uncompromising expression of this traditional view is found in the recent work of Brian Ellis (2001, 2002). The ‘new essentialism,’ he explains, holds that kinds must be unambiguously distinct from one another and characterized by properties independent of history, location, or circumstance the way spin, charge, and mass characterize an electron regardless of when or where it is. Further, the way we characterize a natural kind, according to Ellis, is dictated exclusively by nature; for his essentialism, kind definitions must be independent of the investigator’s interest or practice. Plainly, if these views capture what we mean by natural kinds, then necessarily there are no such things in the human sciences. Ellis (2001: 187) writes that ‘the economy of a country is not an object that belongs to a natural kind, and the processes of the market are not instances of natural kinds of processes.’ As he (2001: 178) explained a few pages earlier, ‘Human laws, institutions, social structures, cultures, and political organizations, and so on are not members of natural kinds. But, according to scientific essentialism, all causal laws are grounded in the

212 Howard Engelskirchen

essential dispositional properties of natural kinds. Consequently there can be no causal laws of social behavior or development that are specific to the kinds of things that are the subject of social theory.’ The argument, however, needs to proceed the other way around. Our first objective must be to identify what potent structures exist in the world, including the social world, and then establish how they are to be best characterized.10 Richard Boyd’s ‘enthusiasm for natural kinds’ (1991) rejects this traditional baggage for the study of both nature and social life. He argues instead that the insistence that kinds be unchanging, ahistorical, intrinsic, foundational, and exceptionless with necessary and sufficient membership conditions is itself neither necessary nor sufficient to explain the way we use natural kind terminology across the diverse domains of science: ‘The natural kinds that have unchanging definitions in terms of intrinsic necessary and sufficient conditions that are the subjects of eternal, ahistorical, and exceptionless laws are an unrepresentative minority of natural kinds (perhaps even a minority of zero)’ (1999a: 169). Like social kinds, biological kinds, for example, often reflect fuzzy boundaries and must be understood historically and in relation to locally specific environments. Moreover, Boyd argues, there is no one true way to understand living organisms or many other natural kinds, including social kinds. But none of these features gets in the way. It is a matter of being clear on what it is our reference to kinds is meant to do. For example, in explaining that there is a ‘necessary indeterminacy’ in the extension of species terms, Boyd recalls that for Darwin, ‘speciation depends on the existence of populations which are intermediate between the parent species and the emerging one.’ As an important consequence, ‘Any “refinement” of classification which artificially eliminated the resulting indeterminacy in classification would obscure the central fact about heritable variations in phenotype upon which biological evolution depends, and would thus undermine the accommodation of the classificatory resources of biology to relevant causal structures’ (1999b: 68). On the one hand, Boyd insists, causation is not a social construction: ‘we do not make causal relations, except in so far as we ourselves function as ordinary causal phenomena’ (ibid.). On the other hand, the way we appropriate the world is constructed – it is always relative to our practice and how we interpret it. ‘Natural kinds,’ he writes, ‘are features, not of the world outside our practice, but of the ways in which that practice engages with the rest of the world’ (ibid.: 66). Thus,

Why Is This Labour Value? 213

they explain a kind of achievement. By contributing to our success in picking out causal structures essential to the demands of practice, kind terms help to accommodate our practice to them. In sum, ‘Kinds are practice dependent but … the world is not’ (ibid.: 55). We carve the world at its joints, surely – it offers us no alternative – but we do not always carve at the same joints or in the same way or for the same purposes. We learn to understand its causal structures, including our own, in order to align our practice with them – again, we interpret the world in order to change it. A second worry compromising the extension of natural-kind theory to social life follows from recognition of the mind dependence of social phenomena. This leads to conclusions that seem distinctly un- or antinaturalist of the sort that suggest social kinds are constituted by ideas (Wendt 1999) or that they are compromised by the looping effects described by Ian Hacking (1995). A news item from the late spring of 2006, for example, reports that 17 per cent of students at Cornell and Princeton practice ‘self-injury’ in the sense of self-mutilation, cutting, burning, and so forth: the person intends to hurt him- or herself but does not intend suicide.11 The label seems new. But once classified in that way, people may relate differently to a person so labelled, see him or her differently, act differently toward him or her, and the person labelled may see him- or herself differently also, reinterpreting the past, looking differently on the future, and so forth. Some worry that media coverage may contribute to instances of such injury. Alexander Wendt (1999: 71) suggests that even the causal theory of reference seems to break down in such instances, insofar as it can be the very act of naming that may create the entity studied rather than a structure independently regulating one’s reference to it. As the critical realist Roy Bhaskar (1989: 84) argues, ‘the social sciences are part of their own field of inquiry, in principle susceptible to explanation in terms of the concepts and laws of the explanatory theories they employ; so that they are internal with respect to the subject matter in a way in which natural science is not. This qualifies the sense in which the objects of social scientific investigation can be said to be intransitive, or exist and act independently of it.’ Naturalism is thus qualified by an ontological limitation: ‘social structures, unlike natural structures do not exist independently of the agents’ conceptions of what they are doing in their activity’ (1989: 85). Bhaskar’s contribution to underlabouring for social theory has, it seems to me, been enormous, and it is in virtue of that contribution that

214 Howard Engelskirchen

I want to question the suggestion that mind dependence constitutes in any way a limit on naturalism as applied to the study of social life. For Bhaskar, naturalism is qualified because social structures are not ontologically intransitive – a term he uses to refer to the way the things of the world act independent of our knowledge, thought, or description of them. By arguing instead for an unqualified naturalism, I mean that the objects of social life can be known without recourse to ghostly or supernatural considerations, of course, and also that they can be known without recourse to a priori beliefs or foundational principles. Knowledge of social kinds, like that of science generally, is of causal structures discovered a posteriori that exist and act independent of our theorizing. It is important to see why this position is not paradoxical with respect to the mind dependence of social kinds. One way to do so is to step back a minute from Bhaskar’s assertions. As creatures, as things of nature, as natural causal structures, we are entities with a mind. Intentionality, as Bhaskar has shown, is integral to what it means for us to act causally. It is part of the causal structure we are. So of course insofar as social structures are constituted by the causal activity of human individuals, they are not independent of mind. But does this render social structures unlike natural structures? If we were to make an appropriate comparison, we would have to ask whether natural structures are independent of the causally potent components that constitute them. H2O is not a causal structure that exists independent of the molecular and intermolecular bonds manifested in the action of hydrogen and oxygen on one another reciprocally. That is, given that mind is part of the causal structure we are, we seem forced to read Bhaskar’s proposition to make the absurd assertion that natural structures, unlike social structures, do exist independent of the causally potent components that make them up. Without doubt, the human sciences are distinctive in that they include the study of mental phenomena as features of their domain. But these are not qualifications or limits on the possibility of naturalism – the constitutive causal structures in play are the object of study in all sciences. Whether we distinguish psychology from sociology or chemistry from physics, the basis for differentiation lies in how we classify just such structures. The real point with respect to the question of mind dependence is instead addressed by Boyd’s ‘metaphysical innocence thesis.’ Realists agree, Boyd argues, that mental conceptions like the adoption of

Why Is This Labour Value? 215

particular theories may have tremendous causal effects on their target social objects. But when constructivists and others suggest that this limits the extension of naturalism to social life, realists disagree: particular theories may have a causal effect, but the theories and beliefs we adopt make no non-causal, logical, or other conceptual contribution to the thing investigated. They will be causally efficacious just to the extent that, and only to the extent that, they engage ordinary causal mechanisms: ‘for realists, human social practices, like the adoption of theories and classificatory schemes, are metaphysically innocent: they affect the causal structure of the world only via the operation of the intermediary causal mechanism which supervenes on the causal structures studied by the special sciences, and not also in some additional way studied only by philosophers practicing conceptual analysis’ (1991: 144). Reasons cause action, but to do so they must supervene on ordinary causal mechanisms. The Kind-Constitutive Properties of the Commodity Form Undoubtedly the extension of natural-kind theory to the study of social life has also been hobbled by the abstract character taken by many theoretical discussions of social kinds. Here especially attention to Marx’s work, so rich in relevant examples, can make a helpful contribution. Let me summarize the example I presented above. To begin, it is important to notice that it is not possible to start by giving a real definition of value; in fact, the nature of value cannot be understood except as a consequence of the social form of commodityproducing labour. Value as such must be explained as a social relation of persons with respect to the expenditure of their labour. But this cannot be understood without spelling out that the social relation instantiated in this quantitative relation is a specific form of the relation of labouring persons to nature and to each other. The kind-constitutive properties of that specific form of social labour are two: persons labour independently, and they do so as part of the social division of labour – they produce goods useless to them for private exchange. This is a real definition of the enformed social labour that produces commodities. When labour takes that form, reciprocally related quantities of labour expenditure in society will tend to take the value form, and value will tend to be expressed by means of commodities and money in the quantitative ratios in which goods exchange. The social form of commodityproducing labour is a social entity, the real definition of which identifies

216 Howard Engelskirchen

for us the causal structure that accounts for the commodity form of the product of labour and the value form of the commodity. Notice that we discover these essential characteristics of the commodity form through investigation a posteriori, and we can test them empirically: if labour is characterized by the separate production of products not useful to the producer, there will be a tendency for the product to take the form of a commodity, and it will tend to be exchanged for value in the market. Moreover, this will be true under a wide variety of social circumstances.12 Under capitalism the commodity is the general form of the product of labour, but of course historically this has not always been true. Nonetheless, even where it is not true, the causal tendencies generated by enformed commodity labour may be incompletely realized or even deformed as they are overridden by the effects of other and conflicting social structures. Still, if the constitutive properties of this living structure exist, it may be possible to mark the traces left by their causal force. Ancient Rome’s forms of law, so readily adaptable to the exigencies of bourgeois life, may offer one dramatic example of such residue. It is important to note that, the kind constitutive properties of commodity-producing labour identify a causal structure that exists independent of our theorizing. Thus, where production is exclusively of use values useless to the producers of them – an activity that certainly depends on the conception producers have of what they are doing – then resort to exchange is compelled. Exchange in turn functions homeostatically to provide each producer with the goods and services necessary for subsistence and new production. In so doing, the original causal structure of social labour is reproduced. It should be added here, in part in response to Brian Ellis (2001), who argues that metaphysical necessity has no meaning in reference to social laws, that the particular form of natural necessity appropriate to enformed commodity labour may be styled ‘reproductive necessity.’ That is, it will be contingent whether a particular social form is to be reproduced, but if it is, this cannot be done unless critical features of the processes essential to it are present. For example, the independence of productive entities, which is the material basis not only for value but also for the legal relation of private property, requires the prohibition of theft. This requirement doesn’t mean that the social form of commodity-producing labour cannot be reproduced if any theft occurs. Neither does it mean that all thieves will be caught or every theft punished – value does not generate exceptionless laws. It does mean that the commodity form is

Why Is This Labour Value? 217

inconsistent with theft and that some qualitatively significant level of suppression must obtain if this form is to be reproduced. It is also the case that no matter what we think or theorize, the kindconstitutive properties of commodity-producing labour will tend to persist and reproduce this social form until they are themselves transformed. This task – as the persistence of commodities and monetary calculations in the former Soviet Union illustrates (Bettelheim 1975) – depends on more than wishing away the manifestations of an underlying social structure. Boyd’s metaphysical innocence thesis applies. Regardless of our theories or ambitions, transforming the causal structure that accounts for the commodity form requires engaging intermediate causal mechanisms that can make theory a material force. For one, large-scale social organization capable of overcoming the autonomous separation of productive units is required. And that in turn requires understanding what such overcoming could mean. On both these points, see Charles Bettelheim’s seminal work, Economic Calculation and the Forms of Property (1975). In an essay responding to Ian Hacking’s discussion of natural kinds, Boyd (1991) identified three key features of ‘the central core conception of natural kinds’ and argued that, properly understood, each applied fully to social kinds. The foregoing analysis illustrates his point. First, he argues that the definition of a kind must be arrived at a posteriori rather than by social convention. I noted above that we discover the social form of commodity-producing labour by investigation a posteriori. Second, the kind must be defined by properties that characterize a causal structure that exists independent of our theorizing. This, as I explained immediately above, does characterize the composite of activity and form that constitutes commodity-producing labour. Third, there must be a causal relation between the instantiation of the kind and the use of the kind term in reference. That is, as Stathis Psillos remarked, the origin of the information we have about a kind must be located in the constitutive properties of it, and thus it is the kind’s causally efficacious structure that in the end regulates how we deploy the term (Psillos 1999). This last point I think is demonstrably true of Marx’s use of terms associated with commodity-producing labour, including, above all, the term ‘value.’ Opposed to empty and vague metaphysical appeals to ‘value,’ Joan Robinson claimed some years ago that value ‘is just a word’ (1962: 47). By contrast, Marx has given an account of the term that roots our use of it in our understanding of an underlying causal

218 Howard Engelskirchen

structure capable of accounting both for the social relation of persons that he characterizes as the value relation and also for the fetishistic expression of that social relation by means of products in exchange. Marx and the Continuity of Scientific Reference If enformed commodity labour provides a real definition of the social form of commodity-producing labour, this offers a context for evaluating Value Form Theory’s charge that Marx’s break with Ricardo was wavering and incomplete. The question here is one of the continuity of scientific reference. Consider two different examples from the history of science. (1) Antoine Laurent Lavoisier’s contribution to the science of chemistry constituted a decisive rupture with that discipline’s earlier understanding of combustion insofar as he rejected the category phlogiston and proposed instead the existence of a new chemical element, oxygen. The category to which Joseph Priestly clung, phlogiston, in fact did not refer to anything. According to older chemical theory, phlogiston was a substance given off during the course of combustion. But there is nothing in nature that corresponds to this account – oxygen combines with other elements in combustion to form oxides. (2) By contrast, Psillos explains that nineteenth-century references to the luminiferous ether did refer to the same entity that later science came to call the electromagnetic field. In that case, although the conceptions that earlier physicists held were inaccurate, still the core causal description they gave of the ether did refer to an energy-filled medium of the kind required for the propagation of light waves, and these features, more fully understood, characterized the core causal properties attributed to the electromagnetic field (1999: 290–298). A distinction Psillos draws from Larry Laudan clarifies the difference. That two scientific theories possess a ‘shared explanatory agenda’ is not enough for the continuity of scientific reference. There must be also a ‘shared explanatory ontology.’ For example, Aristotle’s theory of natural place, Isaac Newton’s theory of gravitational action at a distance, and Albert Einstein’s theory of the curvature of space time refer to a shared explanatory agenda – they want to solve the same problem – but they do not refer to some one causal structure. In other words, we cannot reduce the continuity of scientific reference to the problem being explained, but rather must insist on continuity in the causal agent accounting for the explanation. Suppose a discarded theory is incomplete and also wrong in significant particulars.

Why Is This Labour Value? 219

Still, there may be referential continuity if the fundamental causal properties to which scientists now refer show some important overlap with the causal description that earlier scientists used. This did not occur in the case of phlogiston because there is no substance in nature given off in combustion, and thus no causal properties that function in the way phlogiston theory required. In the preface to the second volume of Capital, Engels raised such questions about Marx’s place in the history of political economy. ‘What is new in Marx’s utterances on surplus-value?’ he asked, and offered an explanation that compared Marx’s achievement with that of Lavoisier in the science of chemistry (Marx 1967b: 14–16). Priestly had actually produced oxygen in the laboratory, but because he could not escape the old theoretical categories, he did not know what he had found. Lavoisier, by contrast, subjected the categories of chemistry to reevaluation and concluded that Priestly had in fact stumbled on a new element. Engels continues: Marx stands in the same relation to his predecessors in the theory of surplus value as Lavoisier stood to Priestley and Scheele. The existence of that part of the value of products which we now call surplus-value had been ascertained long before Marx … With this fact as his starting point he examined all the economic categories which he found at hand, just as Lavoisier proceeding from oxygen had examined the categories of phlogistic chemistry which he found at hand. In order to understand what surplus-value was, Marx had to find out what value was. He had to criticize above all the Ricardian theory of value. Hence he analysed labour’s value producing property and was the first to ascertain what labour it was that produced value, and why and how it did so. He found that value was nothing but congealed labour of this kind … (1967b: 15) (emphasis in original)

Lavoisier’s revolution in chemistry required that he abandon the concept of phlogiston, which did not refer, in order to fashion a category that did. Is Marx’s relation to Ricardo the same? Not in this respect. There is no doubt that Marx revolutionized the categories of economics, but he did not abandon a concept that failed to refer for one that did. Engels pointed out earlier in the preface that Ricardo had developed a theory of value that ‘became the starting point of all subsequent economic science’ (1967b: 10). Plainly, Ricardo’s theory of value could not have become the starting point for subsequent economic science

220 Howard Engelskirchen

if it did not refer. By contrast, Priestly had physically produced oxygen but did not produce a theoretically coherent theory of it. Theoretically he offered nothing that could serve as a starting point for any form of chemistry. Marx’s relation to Ricardo was therefore not really like that of Lavoisier to Priestly. Marx developed, deepened, and clarified the theory of value he inherited from Ricardo, but he did not abandon the scientific object to which Ricardo referred. Marx was not the first to use the term ‘labour’ to refer to the causal agent that produced value. Instead, Engels argues, he ‘was the first to understand what labour it was that produced value, and why and how it did so. He found that value was nothing but congealed labour of this kind.’ Notice the distinction implicit here. Value is congealed labour, but the underlying nature of this, the causal structure capable of accounting for value as a social kind, must be found in the historically and socially specific constitutive properties of commodity-producing labour. That Ricardo had no grasp of the form of labour that constituted the product of labour as value does not compromise either the continuity of scientific reference, nor does acknowledging such continuity compromise the significance of Marx’s theoretical revolution. On the contrary, Psillos argues, such progress is normal in the course of science: ‘This is as it should be. It signifies the open-ended character of scientific concepts and of scientific inquiry itself. A fuller characterization of the putative entity can be discovered only by further scientific investigation. Nor is it reasonable to argue that before a full characterization becomes available the scientists who employ the relevant term do not refer to anything at all’ (1999: 295). Conclusion Recall once more the critique of Ricardo with which I began: if between Ricardo and Marx there is referential continuity – if both Marx and Ricardo refer to the same causal agent – then the advance represented by Marx’s work is not well characterized as a rupture. As profound as his analysis of the social form of labour was, both he and Ricardo referred to the same causal agent. The distinction between a shared explanatory agenda and a shared explanatory ontology must be underscored. Smith, Ricardo, Marx, neo-classical theory, Piero Sraffa, Value Form and others do share a common explanatory agenda. All offer explanations for the ratio in which goods exchange.

Why Is This Labour Value? 221

The question is whether they present a shared explanatory ontology – do they refer to the same causal forces of social life? If, as I believe, Marx and Ricardo do, the rupture on offer will likely break not just with classical political economy but with Marx as well. In the end, Value Form Theory proposes a different scientific object. I do not have space to pursue the critique of Value Form Theory implied by the foregoing analysis and can only, by way of conclusion, offer a promissory note sketching a few essential features of the task. Beginning with the seminal 1969 article of Hans-Georg Backhaus (1980), Value Form Theorists have not understood the distinction presented at the beginning of Capital between value and its nature (Engelskirchen 2006). In effect, they missed the distinction Marx took over from Aristotle between substance, in the sense in which a substantial thing may be the subject of predication (this dog, Fido; this man, Socrates), and primary substance, ousia, the constituting or substantial form of that subject as a kind.13 In the first case, we refer to the commodity as a product of private labour produced for exchange and that exchanges in a ratio regulated by the quantity of value-producing labour expended on it; in the latter case, we refer to the kind-constitutive properties that characterize the causal structure of social labour giving rise to this process. In consequence, the value-form tradition tends to substitute theories of value’s forms of appearance for value’s constitutive form. In the event, an inversion occurs: value’s forms of appearance –, exchange value and money –, are called upon to play the role actually accomplished by value’s kind-constitutive properties. At the limit because no sense can be made of the substantial content of value, value comes to be considered ‘pure form’ (Arthur 2002: 155; Reuten & Williams 1989: 65) and as such ‘without content’ (Arthur 2002: 155). There no longer appears to be an underlying causal structure for scientific reference to pick out (Murray 1993: 51, but see Murray 1997). Alternatively, value is understood as constituted by exchange (Arthur 2002: 96), and in contradistinction to Marx (‘money does not arise by convention’ [1973: 165; 1986: 102]), we are offered explanations of money as conventional, a fiduciary store of value, and as itself without value (Reuten & Williams 1989). Marx (1967a: 60) anticipated the underlying error: ‘Our analysis has shown, that the form or expression of the value of a commodity originates in the nature of value, and not that the value and its magnitude originate in the mode of their expression as exchange value.’ ‘Nature’ here can be understood as value’s essence, the causal structure of

222 Howard Engelskirchen

labour that, as the commodity’s simplest determination, forms its organizing principle. Value Form Theory shifts the object of social investigation instead: emphasizing the constitutive role of exchange, the theory sets labour’s potent material structure aside. As a result, the touted reconstructions of Marx get situated on terrain he left behind. What we have made we can often unmake and refashion, but to do so we need references to the world that accurately present the task confronting us. Bettelheim (1975) showed many years ago that overcoming the persistence of commodity categories during the transition to socialism requires, among other things, overcoming the separation of productive units from one another. In both ordinary language and science, we interpret the world using the terminology of natural kinds in order to get right our efforts to change it.

NOTES Richard Boyd and Anthony Preus have made invaluable comments on drafts of this paper; I am deeply grateful to them. Thanks also to Andrew Brown, Paul Bullock, and Ruth Groff, for their help. 1 Wendt (1999) discusses the distinctive character of asking constitutive questions and offers an example from the field of international relations: ‘what is sovereignty?’ 2 I draw here on suggestions from D.K. Modrak (1985). 3 ‘When, therefore, Galiani says: Value is a relation between persons – [Italian omitted] – he ought to have added: a relation between persons expressed as a relation between things’ (Marx 1967a: 74). 4 Marx gives a complete but succinct definition of the social form of commodity-producing labour in §2: ‘Only such products can become commodities with regard to each other, as result from different kinds of labour, each kind being carried on independently and for the account of private individuals’ (Engelskirchen 2006: 42). 5 ‘All commodities are non-use-values for their owners, and use-values for their non-owners. Consequently, they must all change hands’ (Marx 1967a: 85). 6 In Engelskirchen (1997, 2003); the characterization seems paradoxical because the social form to which it refers is a contradictory one – a producer produces independently but as part of a social division of labour on which she depends.

Why Is This Labour Value? 223 7 Searles makes this observation in his lectures on the Philosophy of Mind (Searles 1996). 8 Marx was careful with the distinction. For one thing, his analysis of commodity fetishism consists precisely in a critique of the illicit tendency to think that properties that depend on a thing’s relations to its environment belong to it intrinsically. Additionally, there is the Sixth Thesis on Feuerbach: ‘the essence of man is no abstraction inherent in each single individual. In its reality it is the ensemble of social relations’ (Marx 1998: 570). 9 Alan Nelson (1990), for example, takes commodities to be fundamental to the description of economic phenomena but concludes that they cannot be thought of as natural kinds in any society; see also Alexander Rosenberg (1983), who finds kinds lacking among the phenomena of the market. 10 See Groff (2004) for a critical realist discussion of Ellis’s rejection of social kinds. 11 Lockard (2006). 12 ‘The mode of production in which the product takes the form of a commodity, or is produced directly for exchange, is the most general and most embryonic form of bourgeois production. It therefore makes its appearance at an early date in history, though not in the same predominating and characteristic manner as now-a-days’ (Marx 1967a: 82). See also ‘The Method of Political Economy,’ where Marx observes that ‘as a category exchange value leads an antediluvian existence’ (1986: 38). 13 ‘A material composite is not the same as its essence,’ (Code 1984a: 117), and thus we distinguish in Aristotle between substance as the subject of attributive predication and primary substance as a thing’s constitutive or substantial form (1984b). Socrates is a man and a member of the human species because the matter of which he is compounded is constituted into a particular individual by that which organizes what it is to be human (1986).

11 The Relation between Marxism and Critical Realism HANS EHRBAR

Realism and the Materiality of Value It has often been claimed that Marx, when writing Capital, followed critical realist principles before critical realism even existed. The present article looks at the evidence to indicate whether this claim is justified. It examines Marx’s famous derivation leading from the commodity to congealed abstract labour, as one can find it, e.g., at the very beginning of Capital, but that exists in somewhat different versions also in the Contribution to the Critique of Political Economy and Value, Price, and Profit. Marx’s derivation has always been controversial. But if this derivation is viewed in critical realist terms, many of the criticisms and doubts raised about it can be answered. Some of Marx’s steps can be considered almost a schoolbook-type application of things written in the Realist Theory of Science (Bhaskar 1975/1997), Possibility of Naturalism (Bhaskar 1999), and Dialectic (Bhaskar 1993). A look at Marx through the spectacles of critical realism has three main implications: • Marx’s Capital is an example how to do critical realist social research. • The systematic apparatus provided by critical realism makes Marx’s

concepts and transitions more precise and accessible. • The comparison with Marx’s method points to areas where critical

realism needs development. Four topics will be investigated in detail: Marx’s starting point, the character of the relationship between use value and exchange value, Marx’s dialectical method, and his emphasis on the quality of valuecreating labour.

Marxism and Critical Realism 225

Is There a ‘Right’ Starting Point for Political Economy? Marx Does Not Start with the Individual Most modern economics textbooks begin with individuals and their utility functions, in keeping with the principles of ‘methodological individualism’ or ‘microfoundations.’ Marx rejects this starting point. One of his contemporaries, the economist Adolph Wagner, wrote a textbook of economics (1879) in which he tried to give a foundation of value based on individuals assigning value to things based on their needs, similarly to the existence of a modern utility function. Marx argues in his Notes on Wagner (Marx & Engels 1975: 538) that these individuals of course live in a society, therefore ‘as a starting-point the specific character of this social man must be presented, i.e. the specific character of the community in which he lives, since in that case production, i.e. the process by which he makes his living, already has some kind of social character.’ This is very similar to the critical realist critique of methodological individualism. Bhaskar writes in the Possibility of Naturalism (1999: 28) that ‘the predicates designating properties special to persons all presuppose a social context for their employment.’ The rejection of methodological individualism is one of the main planks of critical realism. Also the proposed alternatives to methodological individualism seem similar between Marxism and Critical Realism. In Grundrisse (Marx 1986: 195), Marx writes: ‘Society does not consist of individuals, but expresses the sum of connections, relations, in which these individuals stand with respect to each other.’ The Possibility of Naturalism (Bhaskar 1999: 26) uses this exact Marx quotation to explain the relational concept of society in the Transformational Model of Social Activity (TMSA). According to the TMSA, the social structure pre-exists any living individual, and although individual actions reproduce and modify this structure, it cannot be considered the product of individual activity. Social relations must therefore be studied in their own right before the actions of the individuals filling the relations can be understood. Starting Point Important for Marx Marx considered the choice of a starting point to be an important scientific issue. His two methodological manuscripts, the Introduction to Gründrisse and the Notes on Wagner, pay close attention to the

226 Hans Ehrbar

question of where an investigation of capitalist society should begin. (See Carver 1975: 89). Does critical realism share this concern about the starting point? As far as the broad lines of the argument are concerned, the starting point is extremely important for critical realism as well. Bhaskar revolutionized the philosophy of science with his unique starting point. Departing from the judgement that ‘science is possible,’ i.e., that the social activities that usually pass for science are successful in uncovering information about the world, Bhaskar addresses the ontological question, What must the world be like for science to be possible, before answering any epistemological question such as, How should science be done in order to give us information about the world? Marx’s broad approach in Capital is similar. Departing from the unspoken presupposition that capitalism is possible, i.e., that an economy mediated by market relations can lead to the concentration of wealth, Marx investigates the character of the underlying social relations that must be in place for this outcome to occur. Only after these underlying relations are known does it make sense, according to Marx, to look in greater detail at the market transactions themselves, i.e., at the ‘sphere of competition.’ For instance, in chapters 4–6, he investigates systematically how the buying and selling of commodities at their values can be used to turn money into more money; in chapter 12, he asks how capital can continue to increase profits even if the length of the working day is fixed; and in chapter 25, he looks for economic mechanisms enabling capital to grow at a greater rate than the population without being stifled by labour shortages. But Marx is also very picky regarding his specific starting point. He is adamant that one has to begin with the commodity in order to properly understand capitalism. Such a narrowing-down to a specific point of departure can also be found in Hegel, but not in critical realism. Let us therefore see what can be said about this particular starting point from a critical realist perspective. Why Does Marx Begin with the Commodity? The reason why Marx begins with the commodity is often misunderstood. To someone steeped in methodological individualism, the famous remark in Capital (1996: 7) about the commodity being the cell form, and Marx’s emphasis on the prevalence of the commodity in capitalism (1996: 45, 179), suggest that the commodity is studied first

Marxism and Critical Realism 227

because it is the atom that everything else is composed of. This would be methodological individualism starting with the individual commodity instead of the human individual, and it is not Marx’s reason for studying the commodity. Marx’s own reasons for starting with the commodity (as I understand them) are, by contrast, very compatible with critical realism. He looks at the everyday practical activity of the economic agents on the surface in order to make inferences, through second-order arguments, about the invisible social relations enveloping these agents that both enable and necessitate the observed surface activity. This is exactly the method for social sciences recommended in the Possibility of Naturalism (Bhaskar 1999: 26). Material commodities are Marx’s specific point of departure because they play a prominent role in this practical activity. Commodities are ubiquitous in capitalism. People living in capitalism handle them every day. Marx does not explicitly declare that this is his procedure, but a critical realist reading Marx cannot fail to notice that many of his conclusions are second-order arguments of the type just described. The clearest explanation of the character of this starting point can perhaps be found in Marx’s Notes on Wagner, (Marx & Engels 1975: 544): ‘What I proceed from is the simplest social form in which the product of labor presents itself in contemporary capitalist society, and this is the “commodity.” This I analyze, initially in the form in which it appears’ (emphasis in original). The original meaning of the word ‘analyse’ is ‘decompose into its parts.’ But since the commodity is already simple, there is nothing to decompose. What Marx really does is to use a series of second-order arguments to draw conclusions from the practical surface activity with commodities. But since Marx does not have the concept of second-order argument, he mislabels this procedure as an analysis of the commodity. Of course, if prevalence in practical activity were the only criterion, Marx would have to begin with money, like the post-Keynesians. Marx chose the material commodity instead of money because of its simplicity. A material commodity is easy to define: it is something produced for exchange. To the readers of Capital, who are as immersed in capitalist society as the surface agents themselves, it is immediately obvious what the individuals have to do with their commodities. Money cannot be a starting point because it is not immediately clear why it has the properties that make it so indispensable for practical activity.

228 Hans Ehrbar

Profits or capital cannot be starting points because they cannot even be defined if one does not know what money is. A third point that can be adduced in favour of starting with the commodity is the centrality of the commodity relation: money under the gold standard, wage labour, and capital are commodities. Marx argues that some properties of the latter can already be found, in undeveloped form, in the former; he says, for instance, in Capital (1996: 102), ‘The difficulty lies not in comprehending that money is a commodity, but in discovering how, why and through what a commodity is money.’ However, this centrality cannot be obvious at the beginning; it is an after-the-fact confirmation that the commodity was the right starting point. What Can Critical Realism Learn from Marx’s Starting Point? If one looks at this specific starting point through the lenses of critical realism, certain omissions in critical realism become apparent: 1 With its stratification of reality, critical realism knows that certain things are more basic than others. It has the concept of ‘vertical’ causality or emergence, and it also knows that reductionism is a fallacy. Every critical realist would nod and smile knowingly when reading the following passage from Grundrisse (Marx 1986: 190): ‘In order to develop the concept of capital, it is necessary to begin not with labor but with value or, more precisely, with the exchange-value already developed in the movement of circulation. It is just as impossible to pass directly from labor to capital as from the different human races directly to the banker, or from nature to the steam engine.’ On the other hand, there is a difference between ‘basic’ categories and ‘simple’ categories, and critical realism does not have the concept of ‘simple’ categories. The tension between those social relations that are simple, and can therefore serve as starting point for the theoretical appropriation of the subject, and those social relations that are basic in reality is discussed at length in Marx’s methodological Introduction to Gründrisse, (1986: 39). Marx observes that historically, often the simpler concepts come first – for instance, money and commodity production preceded capital for a long time. But it can also happen that the simple concepts come last, that a long historical evolution is necessary to distil a concept down to its essentials. For example, labour has always been the expenditure of human brain,

Marxism and Critical Realism 229

muscle, nerves, etc. (Marx 1996: 82; 68), but the mode of production that turned almost every product into a commodity, and therefore bases its relations of production on labour in the abstract instead of specific kinds of labour, comes late. In the introduction to Grundrisse, Marx summarizes that the simpler category may express predominant relations in an immature entity, or subordinate relations in a more advanced entity. This necessary lack of isomorphism between the real world and its appropriation by the human mind, which leads in Marxism to the tri-partition between the so-called logical development, the historical development, and the structure of the capitalist system at any given point in time, constitutes a circle of problems that critical realism, as far as I can tell, has not addressed specifically. 2 Critical realism lacks the concept of abstraction, which is often considered the main ingredient of Marx’s method. The starting point for Marx’s critical representation of capitalism is not any kind of practical activity of the individual agents in capitalism. Instead, he uses his powers of abstraction to cut away all those behaviours that cannot yet be explained at the beginning and only asks what practical agents do with commodities. For instance, although he knows that even a weakly developed commodity production immediately leads to the development of money, he disregards money altogether in his beginning pages of Capital and acts as if commodities were directly bartered against each other. His starting point is therefore not the raw empirical experience with capitalism, but this empirical experience is first boiled down to its simplest elements. Since critical realism does not have the concept of ‘simplicity,’ it also has not thematized the concept of abstraction. 3 Marx uses dialectical and developing categories right from the beginning. Critical realism, by contrast, started with the Realist Theory of Science (Bhaskar 1975/1997) and Possibility of Naturalism (Bhaskar 1999), which were implicitly but not explicitly dialectical, and were dialecticized only later. The late arrival of dialectic shifts the internal development of categories into what Bhaskar calls the ‘second edge’ (2E), although the non-static nature of categories should have been part of realist philosophy from the very beginning. In critical realism, therefore, dialectics and development can take off only from an already complex system, while Marx gives the example of a simple starting point evolving dialectically into a rich totality.

230 Hans Ehrbar

The Relation between Use Value and Exchange Value The first thing the practical agents need to know about commodities is that commodities have a double character: use value and exchange value. Marx begins with this double character. But this double character is more than only a starting point. It is apparent from Marx’s further development that the separation between use value and exchange value goes very deep. It is like a crevasse immediately visible from the surface that goes all the way down to the bottom of the glacier. One of Them Cannot Be Reduced to the Other This separation between use value and exchange value implies that one cannot be reduced to the other, i.e., exchange value does not depend on use value. Marx does not always present this as an explicit step, although a very early version of his beginning arguments in Grundrisse, (1987b: 252), does. In Capital, he makes occasional remarks to this effect, for instance on page 48. Whether or not spelled out explicitly, this independence is built into the very structure of Marx’s derivation. Use value as such is dismissed as not relevant at the beginning, and Marx proceeds to analyse exchange value independently of use value. However, specific use values that have an impact on economic relations, notably the use value of the commodity labour power, will be discussed later. The evidence for the independence of exchange value from use value is mixed. Some evidence seems to support independence: water is cheaper than diamonds, although it is more useful than diamonds. Other evidence seems to deny independence: in practical life, the more desirable things often have higher prices, and prices rise if demand exceeds supply. With his assertion that exchange value is independent of use value, Marx therefore disregards an important part of empirical evidence. But he comes back to the conflicting evidence later. On pages 111–112 he explains why it is a necessity in capitalism that prices deviate from values if the prices determined by labour values do not clear the market. As long as Marx is able to eventually explain the conflicting evidence, his disregard of this evidence at the beginning should not bother us. The subject of his investigation is a totality complex enough that it can easily generate contradictory evidence. Only one side of the contradictory evidence allows the researcher to understand the underlying

Marxism and Critical Realism 231

mechanisms; the other side is ‘out of phase’ with them, as Bhaskar would say, and therefore misleading. It is part of scientific education to learn which evidence is relevant and which evidence leads to dead ends. In the first footnote of chapter 11 of Capital (1996: 309), for instance, Marx denounces the law of demand and supply as a misleading entry point into political economy. They Are Not Two Sides of the Same Thing The separation of use value and exchange value also has a second implication. Despite a terminology that might suggest otherwise, use value and exchange value are not two sides of the same thing. There is no category ‘value’ of which they are special cases. In Grundrisse, (1986a: 197), Marx devotes a long footnote to the question, ‘Is not value to be conceived as the unity of use value and exchange value?’ He argues that in simple exchange, use value, however important it may be to the individual making the exchange, is not thematized in the social exchange relations: ‘Use value presupposed even in simple exchange or barter. But here, where exchange takes place only for the reciprocal use of the commodity, the use value, i.e. the content, the natural particularity of the commodity has as such no standing as an economic form. Its form, rather, is exchange value. The content apart from this form is irrelevant; is not a content of the relation as a social relation.’ Similarly, he says in Capital that ‘it is exactly the abstraction from the use-values of the commodities which evidently characterizes their exchange relation’ (1996: 47). Something similar can also be found in Contribution, (Marx 1987b: 283). Not only does use value not enter the exchange relation, but commodities must be non–use values for the traders in order to be exchangeable. Whenever Marx uses the word ‘value,’ he understands it therefore to be the underlying relation of which exchange value (but not use value) is the form, and does not consider it as the general concept of which both use value and exchange value are particulars. Development of Their Relation in the Commodity Despite the independence of use value and exchange value, and the dismissal of use value at the beginning, the footnote in Grundrisse as well as the Notes on Wagner emphasize that use value can have an important economic role: the use value of gold mirrors the qualities of

232 Hans Ehrbar

value, and labour power has the use value to produce more value than its own, etc. That is, there is a relationship between use value and exchange value. This relationship does not come from them both being two aspects of the same thing, but it develops only after both are chained together in the commodity. In Capital, chapter 1, section 3, Marx says that the forms of value are the externalization of the immanent counterposition (Gegensatz) between use value and value in the commodity (1996: 71), and the development of the forms of value is also a development of this counterposition (1996: 78). In the first edition of Capital, (1983: 51), Marx says, when he gets ready to discuss the exchange process, ‘The commodity is immediate unity of use-value and exchange-value, i.e., of two opposite moments. It is, therefore, an immediate contradiction. This contradiction must develop as soon as the commodity is not, as it has been so far, analytically considered once under the angle of use-value, once under the angle of exchange-value, but as soon as it is placed as a whole into an actual relation with other commodities. The actual relation of commodities with each other, however, is their exchange process’ (emphasis in original). According to Marx, therefore, the relationship between use value and exchange value does not come from an original unity between them, but it only develops after the two originally independent relations are combined in the commodity. The Error of Central Conflation Adolph Wagner (1879) tries to derive use value and exchange value from a general overarching concept of value. In Wagner’s theory, products have values because it is a natural striving of humans to make themselves aware of and to measure the relations in which the means to satisfy their needs stand to their needs. This is very close to saying that products have values because humans have utility functions (or naturally strive to build utility functions for themselves). Marx brings three arguments against this approach: 1 As already discussed in connection with methodological individualism, human needs, without reference to the social context in which these humans stand, cannot be a starting point. 2 Humans do not first make themselves aware of the relations between their needs and the outside world and then use this awareness to guide their actions, but humans first act (eat, drink, produce)

Marxism and Critical Realism 233

and through this action establish a relationship to the outside things, then make themselves aware of the relationship because they must communicate with other humans about these things. 3 This awareness leads to the distinction between use values or ‘goods’ and other outside things, but it is no more the basis for the exchange values of things as it is the basis for the chemical valence of things. If one tries to deduce use value and exchange value from one original concept of ‘value,’ i.e., if one commits the error for which critical realism has the coined the term ‘central conflation,’ (Bhaskar 1999: 32; Archer 1995), one cannot see the specific role of the commodity but comes to the conclusion that commodity-like relations are the natural state of the economy. Seeking Out Contradictions Marx’s Capital is a thoroughly dialectical work, although the dialectic is often woven into the substantive development of the subject in such a way that an untrained reader may not even notice it. The Contradiction Implied in the Exchange Relations At the beginning, use value gets only fleeting mention, and Marx quickly concentrates on exchange value. It is often overlooked that this discussion of exchange value begins with a contradiction. First, on page 46, exchange value is introduced as something attached to the commodities; it is a second property that commodities have in addition to use value. Marx calls the use value the ‘carrier’ of exchange value, because it is a necessary condition for it – if the commodity falls to the floor and breaks, not only its use value but also its exchange value disappears. On the other hand, as was just discussed, exchange value cannot be derived from use value. But in the next paragraph on page 46, in the very next sentence after the his first use of the term ‘exchange value,’ Marx begins his discussion of exchange value with the observation that exchange value manifests itself as the proportions in which commodities are exchanged, i.e., it is not attached to one commodity but it is a relation between commodities. In addition, this exchange proportion varies with time and place, a fact that would suggest that exchange proportions depend on the circumstances of the exchange and are not inherent in the commodities.

234 Hans Ehrbar

Sigma-Transforms Although this is a contradiction, it is not a blatant contradiction that would jump out at the reader or the agent in capitalism. Most people spend their whole lives in a commodity society without ever considering the exchange relationship to be contradictory. The contradiction is something that has to be discovered. Bhaskar in Dialectic (1993) calls the discovery of such contradictions ‘sigma-transforms,’ as opposed to the dialectical resolution of these contradictions which are ‘tautransforms.’ He writes (1993: 26) that the real work of the dialectic is done by these two transforms. Such sigma-transforms, i.e., the uncovering of non-obvious contradictions, can be found in Capital several times. For instance, Marx points out contradictions on pages 49 and 704, among others. They are exactly the places where modern readers, who are trying to follow Marx’s arguments closely but who are typically not schooled in dialectical thinking, scratch their heads and wonder, What in the world is Marx doing now? After pointing out that he has run into a contradiction, Marx makes a new start on p. 46 with the words ‘let us consider the matter more closely.’ This is a recurrent phrase in Capital. On pages 96 and 203, he uses almost exactly the same phrase ‘let us take a closer look.’ Marx acts here as if he was following the advice given by Bhaskar over a hundred years later in (1993: 378–379): ‘A logical (or other) contradiction is not something to fear and/or to seek to disguise, cover up or isolate. Rather it should be taken as a sign that the existing conceptual field is incomplete in some relevant respect.’ Here is another quotation from Bhaskar (1993: 20): ‘For it is the experience of what in non-dialectical terms would be a logical contradiction which at once indicates the need for an expansion of the universe of discourse or thought and at the same time yields a more comprehensive, richly differentiated or highly mediated conceptual form.’ Or ‘the contradiction becomes the signalling device for the expansion of the conceptual field or the universe of discourse’ (1993: 31, emphasis in original). Finally, he remarks that such contradictions are often not obvious: ‘… is a great advance on the pre-reflective reasonableness of ordinary life, which readily tolerates contradictions without finding anything problematic in them’ (1993: 21). Exchange Value Is the Expression of a Relation of Production Reality is stratified, and Marx uses contradictions as indicators that the linear development, which pursues the relations in one given stratum,

Marxism and Critical Realism 235

has reached its limits and that it is necessary to look at the influence coming from other strata. The surface contradiction involving the exchange value prompts Marx to dig below the surface. The details of this derivation will not be discussed here; they can be found in Ehrbar (2005). Marx comes up with the following resolution: 1 Exchange-value seems to be associated with a commodity because it is the surface expression of some substance, called ‘value,’ which is inside the commodities, but which is not generated in the sphere of exchange, but in the underlying sphere of production. 2 Exchange-value seems to be relative because this surface expression of value takes the form of a relation between different commodities. In this way, both sides of the contradiction can be satisfactorily explained. In order to get this explanation, Marx had to expand his field of vision; he could no longer limit himself to the sphere of exchange but had to dive into the sphere of production. The Material Basis of Value The double character of labour is, according to Capital (Marx 1996: 71), the pivot around which the understanding of the political economy of capitalism revolves. Marxists usually don’t question this claim; they don’t want to admit that they haven’t understood something that Marx considers so basic and important. Critical realism can shed light on the reasons why Marx put so much emphasis on the double character of labour. Since value is real, i.e., value is a causal agent with its own dynamic, Marx was looking for some real object from which value draws its energy. Ghosts as Metaphors Since in principle every use value can be exchanged against every other (as long as the exchange proportions are right), Marx concludes that for the purpose of the exchange relations, each use value is as good as any other, the only difference being quantitative. In a draft version, published in Marx, 1987a: 4, for a paragraph on page 48 of Capital, Marx writes (my translation): ‘One commodity looks now like any other. All that remains is the same ghostlike materiality of what? Of undifferentiated human labour, i.e., of expenditure of human labour power, without regard to the particular useful determinate form of its expenditure. These things

236 Hans Ehrbar

no longer represent anything at all except that in their production human labour power has been expended, human labour has been accumulated. As crystals of this social substance that they have all in common they are – values’ (emphasis in original). This value materiality is rarely mentioned by modern commentators of Marx. They are too embarrassed. Even Marx himself got in trouble for it. The first edition of Capital (1983: 30), described the quality of this materiality with the following words: ‘In order to fix linen as material expression of mere human labour, one must disregard everything that actually makes it an object. The materiality of human labour that is itself abstract, lacking further quality and content, is, of necessity, an abstract materiality, a thing made of thought. Thus, cloth woven from flax becomes a phantom spun by the brain’ (emphasis in original). This vivid and memorable passage did not make it into the second edition, probably because, at the GDR-editors of MEGA2 surmised, it might have ‘raised doubts about the materialist character of value theory’ (Marx 1987: 23*). Also Alain Lipietz dismisses Marx’s value materiality as ‘the major, “substantialist” weakness of vulgar Marxism – which reduced value to a sort of immaterial yet quantifiable product of human labour, incorporated in commodities’ (1983: 4) or in Lipietz 1983: 21, he speaks about the interpretation of value as ‘a mysterious quantity assigned to the product, which enables it to be exchanged with other products.’ Critical Realism to the Rescue Ironically, Marx is rejected where he is most realist. The apparatus of critical realism can clarify things, since it allows us to frame Marx’s ideas in a more systematic and less metaphorical way than his own Hegelian formulations. This requires the following steps: 1 If people exchange their commodities following a consistent and predictable pattern of exchange proportions, then they respond to, and also reproduce or transform, an invisible network of social relations involving these commodities. Marx calls this network the ‘exchange relations’ of the commodities. Of course, the decisions what to exchange for what are individual decisions, but the proportions in which these things can be exchanged are determined by the social exchange relations. The idea that these relations are real and distinct from the individual actions in which they manifest themselves is one of the basic staples of the social ontology of critical realism.

Marxism and Critical Realism 237

2 But Marx’s social ontology has an additional twist. Not only are these relations real, but they furthermore have the character of an immaterial substance inside the commodities. The following subpoints (a) and (b) develop the argument for this: (a) In a conclusion familiar to neoclassical economists it can be shown that these exchange relations, which prescribe the proportions in which the individual agents can exchange their wares, must be ‘transitive’ in order to withstand arbitrage attacks, and therefore can be described by a metric or a numeraire. One knows all there is to know about the status of these relations if one knows how many units of a certain fixed numeraire commodity can be exchanged for each given commodity. (b) So far there is no disagreement between Marx and neoclassical economics. But then Marx takes an additional step that neoclassical economics refuses to take. Marx considers this numeraire not merely as a way to describe the many motley pairwise relationships that together form the network of exchange relations, but in Marx’s theory the exchange relations are generated by a numeraire-like substance, which Marx calls Wertgegenständlichkeit (value materiality). Marx uses his example of the polygons to argue this step. 3 The next step is in tune with one central aspect of critical realism that is often not taken seriously enough. Bhaskar (1975; 1997: 14) says that generative mechanisms are the ways of acting of things. Let us apply this to the present situation. We have found an obviously active generative mechanism – it is the value residing in the commodities, which generates the exchange relations between commodities. But we still have to find the thing whose activity drives this generative mechanism. Marx uses the word ‘value materiality’ (Wertgegenständlichkeit) for this thing. The expectation that such a thing exists is expressed in Marx’s seemingly simple-minded utterances such as ‘So far no chemist has ever discovered exchange-value in pearl or diamond’ on p. 94. 4 The search for such a value materiality has mixed success: • No common substance can be found in the physical bodies of the

commodities themselves. • On the other hand, the production processes from which these

commodities spring have a physical, tangible commonality: all such production processes are the expenditures of human labour power.

238 Hans Ehrbar • But unlike the concrete labour, which is materialized in the use

value of the product, the abstract labour, i.e., the fact that labour is the expenditure of human labour power, is not reflected in the physical make-up of the commodity itself. This is why Marx concludes that this value materiality is purely social. Although one might think that we did not make any progress, since we did not find a material basis, social relations are indeed real enough to do the job. Marx says, for instance, that as value, the commodity represents nothing except that labour is materialized in it. Although this is a social relation rather than a material physically incorporated in the body of the commodity, it is indeed sufficient to explain the causal powers of value. Somebody has produced this commodity, and that person must watch over it that he or she receives reward for the labour placed in that commodity. That is, society remembers how much abstract labour was placed in that commodity, even if this fact is not inscribed in the physical body of the commodity itself. 5 This is not yet the end of the story. Although the purely social value materiality is sufficient as the causal force that anchors the values of the commodities and therefore keeps their exchange relations in place, it is insufficient for the practical activity of the commodity producers. These commodity producers are in the following dilemma: they put their labour into a product that they cannot use and go to the market in order to exchange their product for something they can use. One might say that they try to pull the value materiality out of their product in order to make it useful for them. Since this value materiality is purely social, they must hunt after it in the social relations of commodity to commodity (1996: 57). In section 3 of the first chapter of Capital, Marx shows that the inner dialectic of the value relations will not rest until an independent material form of existence has been developed for this social value materiality – in money. In this way, the search for a tangible value materiality, which is separate from the use value of the commodities, comes to fruition: this tangible value materiality is money. 6 With this independent body, namely money, serving as centre of attraction and reference point, the causal powers of value evolve into the overwhelming vampire-like self-activity of capital. Marx describes here a process of emergence, in which the needs of circulation unwittingly activate a powerful generative mechanism, which previously lay disarmed for lack of a tangible value materiality.

Marxism and Critical Realism 239

Conclusion A much more detailed interpretation of Marx can be found in Ehrbar 2005. The highlights given here were chosen to show how critical realism can throw light on some of Marx’s more obscure arguments in Capital, while at the same time rescuing Marx from the Hegelian embrace. It is my hope that this will make Marx accessible to a broader audience than the devotees who have to treat Capital like the Bible because they never fully understand it. On the other hand, it seems that Marxism can also give valuable input to critical realism.

12 Understanding Why Anything Matters: Needy Beings, Flourishing, and Suffering ANDREW SAYER

Introduction In everyday life, the most important questions we face tend to be normative ones. To translate them into generalities, they are about how to evaluate what is going on, how our lives and those of others are faring, and what to do for the best. They can be about the minutiae of social life, like what to say next in a conversation, or about individually significant events like divorce or changing our job, or about more general matters of politics or lifestyle. As social scientists, however, we are trained to block out normative matters and focus on positive questions of description and explanation. This estrangement of the positive and the normative in social science has come about gradually and unevenly over the past two centuries, and is evident in the widespread acceptance of dualisms of fact and value, reason and emotion, positive and normative. It has had two unfortunate effects. First, it has meant that unless they happen to have studied some political or moral philosophy, social scientists have no particular expertise in normative matters, so they are not particularly sensitive to normative distinctions. Second, they tend unknowingly to project their special intellectual relation to the world onto those they study (the scholastic fallacy), ignoring not only the predominantly practical character of experience (as Pierre Bourdieu noted), but its normative character, and hence fail to do justice to how and why anything matters to people (Bourdieu 2000). Much of recent social theory assumes action to be either merely interest-driven – which implies a very narrow form of normativity – or habitual, or a product of wider discourses and institutions.

Understanding Why Anything Matters 241

These assumptions give much of social science a curiously alienated and sterile character, which can be a shock to beginning students who might have hoped that it would illuminate their normative concerns. Sometimes positive science does illuminate such concerns indirectly, for example by showing that things which we already believe to be good or bad are differently distributed from what we had assumed – for example, that poverty and oppression are more extensive. But it is weak at identifying why, for instance, relations that are called ‘oppressive’ should be a problem, why they should matter. Rather, to the extent that it uses such terms it takes their justification for granted. To be sure social scientists need to learn how to distance themselves from familiar social norms and practices in order to recognize their constructed character, but a crucial aspect of life is that it has import for actors and is not merely a contingent state of affairs of which they happen to be a part. The distancing and de-naturalization are useful but appear to be bought at the cost of producing alienated accounts of social life in which it is difficult to see why things – including contingent social forms – have significance for us and affect our well-being. Of course science would be redundant if it merely reproduced everyday knowledge, and it strives to find deeper explanations, identifying processes and phenomena not noticed in everyday life, but this should not result in the neglect of import. Critical realism has not done much better than other philosophies of social science in illuminating why anything matters. Its way of combining causal and interpretive elements has enabled important progress in the philosophy of social science, but while these are necessary conditions for understanding import, they are not sufficient. It has also challenged the estrangement of fact and value and defended a critical approach to social science, arguing that social science can and should be critical of the phenomena it studies, and have emancipatory goals (Bhaskar 1979; Collier 1994). However, the accounts of this critical approach to social research assume rather than explain import, and do not justify their own normative standpoints (Sayer 2000). Indeed, in part it is thanks to the realization of the need to provide such justifications that critical realists have begun to take an interest in ethics (e.g., Collier 1999, 2003b; K. Dean 2003). I shall begin by defending these claims in relation to critical realism by showing how its twin emphases on causality and meaning do not adequately address need or import. This points to the necessity of adopting a needs-based conception of social being. I shall then outline

242 Andrew Sayer

a possible interpretation of these matters by sketching a qualified form of naturalism, which takes seriously the social and cultural character of human being. This involves a critique of dualisms of fact and value, reason and emotion, normative and positive thought, and in particular the associated de-rationalization of values. I shall then summarize the main possible reasons for the failure to address need and import, and conclude by arguing that critical realism needs to look beyond the model of emancipatory critique proposed in Bhaskar’s early work. Critical Realism, Causality, and Meaning: What’s Missing? Critical realism has a distinctive view of causation. Instead of theorizing it in terms of empirical regularities in events, it centres on the causal powers or susceptibilities1 of objects. According to circumstances, these may or may not be activated; when they are, the effects they produce also depend on contingently related conditions. The emphasis on the contingency of the activation and effects of causal mechanisms combines with a view of most systems, indeed all social systems, as open rather than closed, and hence lacking enduring, precise regularities. This account is one of the most distinctive and important contributions of critical realism, and one of its main challenges to positivism. However, as an account of causality in the biological and social worlds, it can lead us to overlook distinctive features of living beings. Some of their constitutive powers must act continually as a condition of their existence; some causal processes are continuously active and resist suppression (as we find when we try to hold our breath), and some appear to be teleological. We are needy beings, characterized by lack and desire, and beings capable of flourishing and suffering: we cannot live without attending to needs, lacks, desires. Causal powers and liabilities are presupposed by needs, lack, suffering, and flourishing, etc., but the former concepts are not sufficient to identify what distinguishes beings (characterized by needs, etc.) from objects that do not have needs and are capable of neither flourishing nor suffering. Similarly, needs or lack, which by definition are only contingently met, presuppose the openness of systems. Individual autonomy is also predicated on lack and open systems, as is normativity, for if the future is pre-determined, there is no scope for autonomy and choice or normative deliberation. Yet open systems are not a sufficient condition for these – for example, geological systems can exist without them.2 While critical realism’s account of emergence and stratification helps us to

Understanding Why Anything Matters 243

understand such matters, it has paid little attention to the fact that the resulting emergent behaviours involve causal powers and liabilities that differ from those of lower physical strata, such as those of water or wood, in that they are related to needs, lack, and desire, and to flourishing and suffering. Whatever their specific content, needs, desires and drives have a dual character in being related both to existing circumstances and a different state that is sought, whether consciously or unconsciously.3 The other important feature of critical realist views of social science concerns meaning and interpretation. Social life is not merely externally but intrinsically meaningful; it is concept dependent and in part conceptual, and the meanings thereof have to be understood. Critical realism provides qualified support for the interpretivist critique of positivism’s inability to see this. It acknowledges that without an understanding of what the social world means to actors, we simply cannot understand what is going on. As part of its social ontology, critical realism therefore accepts that we are not only causal agents but ‘meaning-makers’ – self-interpreting beings capable of meaningful action. Bhaskar’s seminal work argued that reasons can be causes, that is the things which bring about change, though, like physical causes, the effects depend on context (Bhaskar 1979). We could add that not just reasons but any discursive phenomena can be causes; for example, the words and tone that a speaker uses can produce effects through signalling friendliness or hostility (Fairclough, Jessop & Sayer 2002). The point of much communication is to produce or prevent change. In approaching meaning in social life, we should therefore seek both to understand it and assess what changes it brings about, thereby combining causal explanation and interpretive understanding. One of the useful qualifications Bhaskar made in The Possibility of Naturalism to interpretivist arguments was to note that, in practice, communication presupposes incomplete understanding of the other (otherwise there would be no need for it), but like many philosophers of social science, he didn’t (at that time) say why incomplete understanding should matter, that is, why we should need to gain a more complete understanding or need to reason about anything, or indeed what is about us that makes us need anything (Bhaskar 1979).4 Some accounts of interpretivism emphasize the intentional character of action, but even here the fact that we are beings capable of having intentions often tends to be taken as given rather than investigated (perhaps because to do so would imply a naturalism alien to analytical

244 Andrew Sayer

approaches to meaning), emphasis being given instead to their relation to beliefs, rules, and actions without reference to lack, needs, and desire (e.g., Winch 1958). (It is also possible to develop accounts of interpretive understanding that are silent on intentionality, particularly where its discursive or textual character is emphasized, though in practice, to understand others we have to appreciate that they have desires and intentions.) Intentionality presupposes will, which is more than mere choice, because we can continue to will that something will happen after we have made our choice. ‘I must find a job’ is not merely a statement of intentions, but a statement of what I want to happen, in relation to what I believe myself to lack and need. Within the interpretive dimension of critical realism and other postpositivist social theory, the element of empathy in understanding was also sometimes dropped, the argument being that rendering verstehen as empathy missed the linguistic character of understanding and the fact that some kinds of understanding do not require empathy (e.g., Giddens 1976). This is an important point, but it does not warrant ignoring empathetic understanding or ‘fellow feeling’ as preconditions of social life. Some empathy appears to be pre-linguistic: babies can be affected by others’ moods even before they learn any language. It is hard to see how social life in general would be possible without it, and indeed we regard those who are deficient in it (i.e., autistic) as impaired. Such non-linguistic communication or interaction tends to be ignored in philosophy of social science and much social theory, again perhaps because of the latter’s scholastic privileging of reason and discourse, though there are important exceptions. To empathize with another is not merely to understand the meaning of their words and gestures but to understand their concerns, what is of import to them, including what they believe they need and want, what they fear or welcome, and how they value things generally.5 In communicating we do not merely describe, or indeed perform, we also infer and (intentionally or unintentionally) signal judgements of import, that is, of how things impinge on what we care about. What matters about communication is not merely the signifier and the signified, or indeed the referent, but the significant. Sometimes when we talk about the meaning of something we are drawing attention to its import for us (e.g., ‘what global warming means for us’) (Sayer 2006). As is particularly clear in anthropology, if we do not understand why certain things are significant to others, we cannot understand what they say and do, even though we may be familiar with the meanings of their words.

Understanding Why Anything Matters 245

Empathetic understanding and recognition of import are key features of emotional responses, though not unique to them. To appreciate this we need to recognize the cognitive aspect of emotions and to avoid counterposing emotion to reason. As Martha Nussbaum argues, emotions are evaluative judgements of circumstances believed to affect or to be likely to affect our well-being or that of others or things we care about (Nussbaum 2001; see also Archer 2000; Helm 2001; Oakley 1992). Sentiments such as pride, shame, envy, resentment, compassion, and contempt are evaluative judgements of how people are faring or being treated as regards what they value, that is, things they consider to affect their well-being. They are forms of emotional reason. They are fallible judgements, but then so too are those of unemotional reason. Although it is common to refer to emotion as ‘affect,’ there is a danger that this may affirm an opposition of reason and emotion, and a detachment of emotion from what happens to people, so that it becomes ‘merely subjective’ and lacking any rational content, and so that it becomes unclear how concerns can be about anything.6 At the same time as contemporary social science has de-rationalized emotions and values, the role of reason in everyday action has itself often become attenuated and abstracted from practical, normative experience, and at worst reduced to the application of deductive logic or instrumental rationality (Archer 2000). This is most evident in contemporary economics but sociologists such as Weber have reinforced it too (MacIntyre 1981). When we say things like ‘we have reason to be angry’ (because someone has deceived us, slandered us, or whatever), we don’t mean merely that we have a logical argument for being angry,7 or that being angry is a means to an end. Rather, citing such reasons also indicates that reason can be related to needs, desire, commitments, and practical matters of welfare or flourishing and suffering.8 As Margaret Archer’s recent research shows, people’s internal conversations include thinking and worrying about their commitments, weighing them up in a way that involves a kind of practical reason or substantive rationality, dealing with the valuation of ends and concerns themselves, not merely means toward them (Archer 2003). Accordingly, neither the causal nor the interpretive element of critical realist philosophy of social science, nor indeed the two in combination, are sufficient to do justice to need and import. Critical realists are used to asking questions of the form: what must X be like for Y to be possible? (e.g., what must the world be like for science to be possible?), but to my knowledge, with the exception of Archer and Kathryn Dean,

246 Andrew Sayer

critical realists haven’t got far in asking what it is about humans that makes them needy and capable of flourishing and suffering (Archer 2000; K. Dean 2003). We can summarize the argument so far by reference to four models of social being and action. The first model, exemplified by behaviourism and positivism, has a material causal conception of society, focusing only on material states and processes, ignoring the constitutive character of meaning in social action, hence reducing action to behaviour. The second has a hermeneutic conception of society and focuses purely on understanding actors’ understandings, hence reducing action to meaning and ignoring the materiality of social processes; some variants of interpretive and discourse analytic approaches do this. The third model combines causal and hermeneutic approaches by regarding reasons and discourses as possible causes of change. While Weberian and critical realist approaches do this consciously, in practice all social scientist have to do this, even if unknowingly. None of these three approaches enables us to understand why anything matters to people. To do so, we have to turn to a fourth, neglected, model, which might be described as having a ‘needs-based’ conception of social being and action, viewing actors as needy, desiring beings (characterized by lack), capable of flourishing or suffering. ‘Needs’ here is a shorthand that also covers lack, wants, and desire. Needs in this broad sense include what might be termed ‘culturally emergent needs’ such as that of the religious to worship. This fourth model subsumes the third model but goes beyond it in a way that enables us to understand import. Failure to acknowledge human neediness and vulnerability invites misattributions of causality or responsibility, so that, for example, discourses are treated as capable on their own of motivating people, which of course fails to explain why things like rocks or plants cannot be motivated thus. Is it not remarkable that the literature on social ontology rarely gets beyond discussions of structure and agency and fails to recognize humans as needy and vulnerable social beings, capable of flourishing or suffering, and hence as evaluative beings?9 Needs, Lack, and Desire: Toward a Qualified Naturalist Interpretation This fourth conception of being implies a kind of ‘ethical naturalism,’ both for lay or folk ethics and for philosophical normative ethics,

Understanding Why Anything Matters 247

though in some respects the term is a misnomer because the criterion for evaluation is not nature but flourishing. Our natural (in the sense of physical) powers are a precondition of our social powers, though they can be modified, developed, or sapped by the latter, but many natural states do not amount to flourishing. Needs, lack, desire, and flourishing and suffering can take many forms. They may be primarily physical, such as hunger, but even then, how they are experienced and evaluated is culturally mediated, according to cultural valuations. The body itself is conditioned by contingent forms of material experience so that it may become tough or soft, fit or unfit, and so on. But the more interesting cases are those that are oriented to valuations of cultural goods themselves,10 particularly where they involve commitments, such as that of the religious person to worship or the political activist to a cause.11 There appear to be crosscultural needs such as for sociability, recognition, and self-respect, and for forming commitments, but cultures always provide definitions of how these are to be achieved, that is, what counts as worthy of recognition, respect, and commitment, though cultures rarely do so in unitary, internally consistent ways, and hence contestation is common. Flourishing and suffering are therefore not merely physical states but depend on ways of relating to others in terms of how practices and behaviours are culturally valued and on the formation and pursuit of commitments. In this regard, sociology has contributed to the explanation of the social production of human subjects with a wide range of capacities and sensibilities.12 Yet actors’ behaviour is not wholly determined by cultural scripts and prevailing norms, for they are reflexive beings, capable of ‘strong evaluation,’ that is, evaluating not only how to achieve goals and respond to norms but the goals and norms themselves (Archer 2003). Self-respect and shame are forms of flourishing and suffering that are influenced by how people feel they have acted and are faring with respect to the particular cultural norms that they have come to accept. A musician may be mortified to learn that a peer thinks she has no talent, or may challenge that judgement. An accountant may feel pride in having got top marks in her accountancy exams, or feel it wasn’t worth it. Men of an older generation may have felt ashamed if their wives went out to work, while their sons may feel embarrassed if their female partners don’t. Similarly, people can form a remarkable diversity of commitments to others and causes, sometimes valuing them more than their own lives. While the capacities for emotional responses and for forming commitments appear to be universal

248 Andrew Sayer

and have certain general structural properties, theorized by moral philosophers (e.g., Nussbaum 2001; Calhoun 2003), their particular contents vary across a wide range. However, despite the remarkable diversity of cultural goods and norms, it does not follow that no element of naturalist explanation is required. One reason for this is that only particular kinds of beings can be enculturated in these ways. Strong versions of social constructionism overlook this. There must be something about humans (and those other species that appear to be capable of cultural variation) that makes this variety possible. In this way, as Andrew Collier points out, far from cultural variety refuting naturalism, it presupposes it (Collier 2003a). To be sure, our mental development and behaviour depend on constitutive processes of socialization and language acquisition, and indeed what we take to be characteristically ‘human’ depends on them. Here we have to steer a course between the ‘Standard Model of Social Science,’ or ‘blank-slateism,’ and sociobiological determinism. Both are products of disciplinary imperialism – the stretching of forms of explanation beyond their appropriate domains to others where they can only mislead. The former treats the mind as a blank slate on which anything cultural can be written, and renders inexplicable how it is that not just any object can be enculturated. In the social constructionist version, it also renders phenomena like abuse and injury incomprehensible, for there is nothing extra-discursive or extra-cultural to violate (Soper 1995). On the other side, the reductionist projects of sociobiology and evolutionary psychology treat contingent cultural forms as genetically based (Dupré 2001). Social development and enculturation are emergent properties dependent on but irreducible to biological and social influences. What we call the mind is a co-product of physiology, culture, and language. Thus, to talk of human nature is not usually to refer just to the newborn’s powers and liabilities, or indeed his or her needs and drives, but to the dependence of those needs and drives on society for their development and supplementation. The role of the social environment is not merely as a source of ‘triggers’ that set in motion specific already-constituted powers but a co-constituent of a vast range of powers. Like John Dupré, I do not want to privilege internal over external causes a priori here, or indeed, vice versa. Second, naturalism is also needed to acknowledge the fact that, although cultures provide conceptions of the good, they are fallible and often blatantly ideological, typically legitimizing forms of domination.13

Understanding Why Anything Matters 249

While there are indeed many forms of flourishing, not just anything can be successfully passed off as flourishing. Despite the remarkable variety of cultures, we are not infinitely malleable (for example, genital mutilation and solitary confinement cause objective harm, whatever particular cultures may say to the contrary) (Midgley 1972). Conceptions of needs, wants, and desires are fallible; indeed, cultures are fallible. (For example, as a Westerner I would argue that Western culture has promoted mistaken conceptions of the good, such as the pursuit of money as an end in itself.) As Amartya Sen (1999; and Nussbaum 2000, 2001) argue, there are certain forms of being and doing that appear to be important for human flourishing cross-culturally, many of which are not currently available to large numbers of people. The concept of ‘drives’ quite rightly tends to set alarm bells ringing in sociologists’ minds, because of its association with sociobiology and psychologism, that is, with reductionist accounts that attribute what is due to culture or discourse to biology and psychology, often thereby legitimating behaviours that are in fact contingent and historically specific as universal and necessary. But the relationship between drives and the particular practices they are involved in and their consequences is contingent. Drives are partly developed (or suppressed) and given content and direction by culture, and hence their possible effects are open and multiple rather than singular. As Dean puts it: ‘The drive is the “interface” between the potentially human, which is the peculiarly incomplete biology of the unformed human organism and the realized humanity of the organism once it has been culturally formed.’ (Dean 2003: 19). It is not only that, as critical realism rightly claims, social systems are open, but that actions and practices are shaped by human incompleteness, lack, and pre-maturity, which drive actions toward completion and maturity. Ironically for a discipline whose most distinctive object of study is social relations, sociology can only give a partial and anemic understanding of them because of its avoidance of the psychological and developmental character of sociality. As Adam Smith recognized, we are not merely contingently socialized and economically dependent on others but psychologically and physically dependent on them, and liable to suffer if denied company or recognition (Smith 1759). Understanding needs, drives, lack, desire, flourishing and suffering, including their specific forms such as fulfilment, health, illness, loneliness, oppression, and disrespect, requires us to realize that these are matters that are neither purely positive or normative but both. It is

250 Andrew Sayer

striking that some of the most important aspects of life fail to fit into the categories of fact or value. Just what ‘suffering’ and ‘flourishing’ are, is both an empirical and a normative matter, encompassing many varieties, including some we have not yet discovered or devised.14 In positive thought, we assume that if our ideas fail to fit the world, we should change them to fit the world. By contrast, in normative thought, when we perceive a mismatch between our thinking and how the world seems to be, we assume that the world needs to be changed to fit our ideas (Helm 2001). So positive thought is ‘world guided,’ and normative thought ‘world guiding.’ But concepts such as ‘needs’, ‘desire’, ‘lack’, ‘flourishing,’ and ‘suffering’ have both descriptive and evaluative content; indeed, the two cannot be separated. Lack, needs, and desires are not merely markers of the difference between the ideas of an outside observer and the world but part of the world that strives to go beyond its existing state at any given time. In recognizing needs, I simultaneously adjust my thinking to the world and (ceteris paribus) think that the world ought to be changed to meet them. This is not to say that such recognition is infallible. The relationship between what we feel we lack and desire and what will actually satisfy us and enable us to flourish is something we may struggle to discover. It is not an apodictic relationship, and it would be absurd to expect it to be. Some readers may smell a rat here regarding the relationship between positive and normative discourse. Many philosophers from Hume onward have argued that we cannot logically deduce an ‘ought’ from an ‘is,’ and that to imagine that we can is to fall foul of the ‘naturalistic fallacy.’15 Even the statement ‘X is starving’ supposedly does not logically entail the statement ‘X should be fed.’ At one level it is indeed fallacious to imagine that an is statement logically entails an ought statement. But in the realm of practice, in terms of the relation between circumstances and practices, other things being equal, it would be bizarre to deny X food when we could easily give them it. But then logic is just about the relations between statements, not circumstances and practices. When I feel hungry my desire to eat does not require a logical warrant. If we recognize something as having major implications for our well-being, then whatever words we choose to describe this, the description is already ‘valuey,’ as Collier puts it (1994). Needs, etc., are certainly already valuey, but values are not merely subjective states; they are fallible but possibly rational valuations of such things. Their fallibility derives from the fact that they are about states that can exist independently of their description, i.e., objectively, and hence

Understanding Why Anything Matters 251

about which they may be mistaken. Valuations of matters of import are therefore both objective and subjective: they are objective in that they are about things that exist or can exist independently of the valuer and the valuation, such as the harm or benefit that may arise from some action or institution; and they are subjective in that they are made by subjects, albeit subjects who are objectively capable of flourishing and suffering according to their (partly acquired) powers and susceptibilities. Factual accounts of life are valuey because we are vulnerable beings who need to evaluate in order to flourish; in life, not valuing the world is no more possible than choosing not to breathe. And valuing the world is a form of reasoning about it, not something quite unlike reasoning. In effect, therefore, the fact-value, positive-normative distinction splits us in two. It might be noted at this point that the stance I am advocating opposes the general thrust of post-structuralism, though I acknowledge that the latter has brought some advances in understanding society and can in parts be reinterpreted in a way compatible with critical realism (e.g., Joseph 2004; Marsden 1999). In particular, critical realism opposes post-structuralism’s overreaction against essentialism, and its ‘cryptonormativity.’ Regarding the first of these, we can acknowledge that humans are not uniform, though without denying that they have much in common. As we have just argued, this similarity is a condition of their being similarly susceptible to specific forms of enculturation.16 We can also accept that what humans (can) do is not predetermined by their nature, without abandoning a conception of human causal powers and susceptibilities, or affordances and constraints, so that they become merely a product of external construction. Even the many powers and susceptibilities that we acquire through socialization or ‘subjectivation’ are themselves constrained and enabled by our nature, and what we can acquire at any given time is constrained and enabled by previous rounds of subjectivation. Again, if we deny that people have any particular natural properties as human beings, as organic bodies, and assume instead that these properties are no more than cultural or discursive constructions, we lose all critical purchase on any oppressive exercise of power, particularly through torture, mutilation or abuse (Soper 1995: 138). This excessive anti-essentialism is generally coupled with what Jürgen Habermas, in his critique of Michel Foucault, termed ‘cryptonormativity,’ that is, the tendency merely to insinuate that some object – particularly power – is problematic without giving any indication why, indeed while refusing judgement, so that the reader is left unsure of

252 Andrew Sayer

what is the problem.17 This has become common in post-structuralism. Sometimes the associated normative disorientation it produces is even valued, as if there were some virtue in this.18 But this disorientation is not the only problem, for given the fusing of positive and normative in matters of needs and well-being, crypto-normative approaches are also crypto-positive, since they sacrifice descriptive richness through their refusal of valuation; in avoiding comment on whether something causes suffering or flourishing we miss important positive information. Thus, in response to vaguely dystopic accounts of ubiquitous power and subjectivation we may be equally tempted to ask ‘is everything bad?’ or ‘so what – why do they matter?’ It may help at this point to summarize the possible reasons for the extraordinary neglect of needs and the like by critical realism and other philosophies of social science. First, this deficiency in critical realism bears the marks of its response to analytical approaches to philosophy, which are arguably themselves the product of the scholastic fallacy and the estrangement of values and reason, value and fact.19 Second, in sociology, the deficiency of realist – and many other – social ontologies in ignoring lack, neediness, desire – could be a product of sociologists’ aversion to psychology and the sociobiological associations of ‘drives,’20 and to what they see as ‘essentialism’ or ‘universalism,’ which allegedly always naturalize historically and culturally specific phenomena and hence marginalize or pathologize those who do not fall under their description. This last point is related to a third: the retreat from naturalism, along with humanism and philosophical anthropology. Naturalism is widely rejected, particularly in post-structuralism, as allegedly inevitably producing a naturalization of contingent forms of social organization that serve to legitimize oppression. I would argue that the problem here is not naturalism in general but inappropriate or mistaken forms of naturalism. An example is racism. It is not that there are absolutely no physical differences between ‘races,’ but that those differences are trivial and insignificant and have no bearing on the objective properties and capabilities of people or their moral worth.21 Racism is typically based on a misused or misplaced naturalism.22 If instead of correcting this misused naturalism we refuse any kind of naturalism, as is common in cultural studies, and ignore the objective nature of flourishing and suffering, then it is open house for oppression, for it can be denied that there are any natural vulnerabilities to suffering. (One wonders what an anti-naturalist medical ethics would look like!)

Understanding Why Anything Matters 253

Fourth, as Pierre Manent argues, whereas philosophy assumes the viewpoint of the actor, sociology assumes the viewpoint of the spectator and often accords no real initiative to actors, which creates a gulf between how sociologists describe the world and how we (including sociologists when off duty) experience it (Manent 1998).23 Iain Craib argued that sociologists often exhibit what psychologists call a ‘normotic personality’ (Craib 1998), that is, a refusal to entertain the subjective element of life, so that, for example in post-structuralism, the inner life is just the product of the outer life and does not generate anything new (see also Archer 2000, 2003; Craib 1997).24 When sociologists do grasp the motive of a behaviour they tend not to see it as a product of judgement but as either a product of an internalized cultural script or of a ‘value,’ which in turn has no rational relation to the world. ‘Thus the sociology of values, which claims to give the actor’s viewpoint its due, leaves intact the sovereignty of the spectator’s viewpoint’ (Manent 1998: 75). Thus, ‘the arrogant philosopher is superior to other men, but the modest sociologist is superior to the humanity of man’ (77). However, as we noted earlier, although philosophy assumes the viewpoint of the actor, it has often done so in a way that privileges discourse and reason. It also tends to overlook the institutional context of actions, hence attributing more responsibility to individuals than is warranted. Conclusion: Implications for Critical Realism Critical realism is distinctive not only in adopting our third conception of action as requiring both interpretive understanding and causal explanation but in being critical in the strong sense of seeking to critically evaluate social life itself with a view to facilitating emancipation. By way of conclusion, I shall argue now that it will be inhibited in this respect as long as it fails to embrace the fourth, needs-based, conception of action. First, to recap: little of social science tells about why we care about anything. One reason for this is its de-rationalized view of normativity and values and its abstraction of reason from practice and from our condition as needy, vulnerable, social beings. We thus have on the one side reason, which does not seem to matter or motivate, and on the other values and emotion, which matter and motivate but from no apparent reason. This de-rationalization of values is at odds with the fact that when necessary, as in the case of perceived injustices, we do reason about

254 Andrew Sayer

values, and not merely by appeal to personal preferences or mere convention. Rationality is not wholly separate from emotion and social, practical life, and it often concerns what we care about, what matters for well-being, hence needs, wants, lack, desire, flourishing, and suffering. The divorce of normative and positive thought in social science has rendered much of so-called critical social science unable to identify not only its own normative standpoints but the normative concerns, distinctions, and valuations that figure so prominently in the lives of the people it studies. As with ethics, in developing a philosophy of social science, we have to go beyond philosophy to empirical questions of psychology (some of them nowhere near answered yet), about what kind of being we are, and about how we mature and develop our capacities and susceptibilities.25 Without a consideration of import and flourishing, critical social science tends to resort to a somewhat empty, libertarian, and masculinist concept of emancipation as freedom from (unwanted) determinations, in which the primary reference point is an ideal society in which the freedom of each individual is the condition of the freedom of all. This is an appealing slogan but has little use, for like liberalism it evades issues of the good or flourishing and our capacities and neediness as social beings. The implication of Bhaskar’s early models of emancipatory explanation is that by identifying the causes of suffering and by removing them, emancipation can be achieved (Bhaskar 1979). Just what is suffering or otherwise problematic is reduced to the absence of freedom. All problems apparently derive from domination implicitly conceived as a form of restriction of liberty, or from ideological repression of the truth. But social problems also derive from other sources, for example, from the failure of certain individuals or groups to acknowledge their responsibilities for others, leading to deficiencies in care, so that some flourish at the expense of others. It would be better to try to devise forms of social organization in which the flourishing of each can be a condition of the flourishing of all and to focus on just what does constitute flourishing.26 Although Bhaskar’s later work mentions flourishing and occasionally some of its preconditions, such as trustworthiness, sympathy, and care, this is not a substitute for a naturalistic ethics with substantive content, such as that developed by Sen and Nussbaum (Bhaskar 1993; Sen 1999; Nussbaum 2000).27 We cannot realize the emancipatory project of social science purely by attempting to drive forward by looking only in the rear-view mirror, moving away from problems, with little argument about what

Understanding Why Anything Matters 255

exactly is problematic or what constitutes failure to flourish and without any conception of where would be better to go. This is not only because moving away from known problems may lead us to bigger ones (Sayer 2000), but because emancipation depends on achieving deeper and more inclusive forms of flourishing, and contemporary social scientists are generally unaccustomed to thinking about such matters. These are undoubtedly difficult issues, but that makes the revival of normative thinking all the more important. Like Sen and Nussbaum, critical realists need to proceed in a neo-Aristotelian direction to consider what the elements or forms of flourishing are, and this will require a much fuller analysis of psychological and social being than has been provided thus far. It will also require closing the gap between positive social science and normative moral and political philosophy, whose tendency to abstract from social structures, subjectivation and the concrete biographies of individuals is itself a product of the estrangement of positive and normative thought.

NOTES 1 I use this term in preference to Bhaskar’s causal ‘liabilities’ (Bhaskar 1975). 2 This implies the critical realist stratified ontology needs more specification with regard to the relationships between biological and psychological/ social strata, indeed regarding whether the latter are distinct. 3 I have not followed Bhaskar (1993) in theorizing lack as ‘absence,’ because in my view the latter is anthropocentric and implies an epistemic fallacy; i.e., it confuses an epistemic matter (what we expect) with an ontological matter (what exists in terms of biological and psychological drives). Lack implies not absence but the presence of a pressure for a different state than that which holds at present. Absence also implies the possibility of nothingness. When something dies or ceases to exist, all that happens is that it is transformed into something else, a different kind of presence. The law of the conservation of matter holds: the ozone hole is not an absence, just the presence of something other than ozone. ‘Absence’ invites us to imagine there is nothing, emptiness, when there is actually just something else there from what we expected. Absence is thus either just different presence or presence in a different place from the one expected. Nor is absence necessarily associated with lack, desire, or need. The demise of dinosaurs has not created a situation that needs to be remedied.

256 Andrew Sayer 4 In his often impenetrable Dialectic: The Pulse of Freedom (1993), Bhaskar introduces lack and desire and the need for a naturalist ethics, but the content of such an ethics is not fleshed out, and as the subtitle indicates, while flourishing is mentioned at various points, the dominant motif is freedom. 5 See especially Adam Smith’s analysis of fellow feeling or ‘sympathy’ (Smith 1759). 6 It is perhaps significant that when we speak of ‘affectations’ and ‘affecting’ a certain manner, we mean precisely to draw attention to their simulated – or rather dissimulated and false – character. 7 Anti-naturalists would refuse such a possibility by denying that is can ever entail ought. This is not only flawed in its own terms, as Bhaskar has demonstrated (Bhaskar, 1979), but it argues on the wrong terrain, for it misidentifies relations between substantial processes, such as being assaulted, insulted, or deceived and the consequent effects on one’s well-being and state of mind, as logical relations between statements. 8 This also implies that reason itself has ‘shoving power’ (Archer 2003; Collier 2003a). 9 The treatment in actor-network theory of people as ‘actants,’ no different from other objects with causal powers, disastrously tries to make a virtue of this oversight. 10 Culture ‘occupies’ ‘the biological by way of the emergent subject’s more or less active response to an ‘invitation’ to attend to a part of the world in a particular manner’ (Dean 2003: 19). 11 I do not need to distinguish between needs and wants for the purposes of the current argument. 12 For example, Pearce demonstrates this in Durkheim’s work (Pearce 2001). 13 Often they use a false kind of naturalism, by naturalising contingent social forms. 14 As I shall argue shortly, it is probable that the estrangement of fact and value and the belief that relating them involves a fallacy (the ‘naturalistic fallacy’) have contributed to social science’s indifference toward these (literally) vital phenomena. 15 My thanks to John O’Neill for discussion of these matters. 16 Thus we would expect a baby born in the West who was immediately adopted into, say, Japanese society, to grow up to become very similar in beliefs and behaviour to those born into his/her adoptive society, and vice versa for a Japanese newborn immediately transported to the West and brought up there. To deny this similarity of human nature would invite a kind of racism.

Understanding Why Anything Matters 257 17 Habermas (1987, chapter 10), reviewing a wide range of Foucault’s work. On the debate between Habermas and Foucault, see Kelly (1998). See also Fraser (1981). 18 This complements a rather adolescent tendency to value transgression or ‘unsettling’ in itself, regardless of what is being transgressed or unsettled. 19 For Dean, there’s a double estrangement – of rationality from affect, and rationality from sociability, the latter being a product of deepening of divisions of labour. They also involve, I would argue, a refusal of practical, naturalistic bases for logic and reason. On this see Archer, 2000: 145–152, and Harré & Madden, 1975). 20 These aversions are the flip side of sociological imperialism (see Archer 2000). 21 Similarly with sexism, it is not that there are no physical differences between men or women, so that they only exist in our imaginations, but that they do not justify the domination of women by men. Unlike the case of ‘race’, some of the differences between men and women, like the distribution of the capacity for childbearing, are real, but they do not warrant patriarchy (New 2004). 22 There can also be cultural racism, responding to ethnicity. 23 I am indebted to Paul Fletcher for bringing Manent’s book to my attention. See also Archer (2000; 2003). 24 Craib, who practised psychotherapy as well as sociology, described strong versions of social constructionism as a form of social psychosis (Craib 1997). 25 One of the strengths of Adam Smith’s empirical study of moral sentiments is that it incorporates a rudimentary developmental social psychology and gives a strong sense of the social character of the human being, which enables him to avoid the alienated treatment of morality that now dominates sociology (Smith 1759). 26 See note 3. 27 In Dialectic, having mentioned such preconditions of flourishing, Bhaskar reverts to a model of emancipation as liberation from ‘all ills and constraints’ [as if they were always independent of goods and opportunities and enablements] and hence their explanatory critically identified [sic] causes … And from here it is, in theory, but a short hop to the free society, which satisfies or approximates or approaches the formal criterion of the free flourishing of each and all … ’ (1993: 287). The dominant theme is ‘absenting’ constraints, rather than constructing ways of meeting needs and enabling flourishing.

13 The Expulsion of Foucault from Governmentality Studies: Toward an Archaeological-Realist Retrieval JON FRAULEY

In effect, what defines a relationship of power is that it is a mode of action which does not act directly and immediately on others. Instead it acts upon their actions: an action upon an action, on existing actions or on those which may arise in the present or the future (emphasis added). – Michel Foucault, ‘The Subject and Power,’ 1983 We argue that an analysis of modern ‘government’ needs to pay particular attention to the role accorded to ‘indirect’ mechanisms for aligning economic, social and personal conduct with socio-political objectives (emphasis added). – Peter Miller and Nicolas Rose, ‘Governing Economic Life,’ 1990 It is for these reasons that we have suggested the need for the analysis of the ‘indirect’ mechanisms of rule that are of such importance in liberal democratic societies: those that have enabled, or have sought to enable government at a distance (second emphasis in original). – Miller and Rose, ibid. Liberal doctrines on the limits of power and the freedom of subjects under the law were thus accompanied by the working out of a range of new technologies of government, not having the form of direct control by authorities … a key characteristic of modern government: action at a distance (emphasis in original). – Miller and Rose, ‘Political Power Beyond the State,’ 1992

Introduction Inspired by Foucault’s work on indirect rule and the decentralized nature of modern power and regulation, studies of ‘governmentality’ have become prolific, appearing in sociology, criminology, legal

The Expulsion of Foucault from Governmentality Studies 259

studies, cultural studies, and many other disciplines.1 In an ironic twist, however, although it is Foucault’s (1978/1991) governmentality paper that is the ‘source’ of this prolific scholarly activity, it is the influential efforts of Rose and Miller (Miller & Rose 1990; Rose & Miller 1992)2 to bring Foucault’s work to bear on the study of social action as it relates to the intersection of expertise and the regulation of conduct that is foundational to the field. It is Rose and Miller’s interpretation of Foucault’s understanding of indirect rule that is widely employed by governmentalists, for instance, when they use the field’s catchphrase, ‘government at a distance’ (see Frauley 2005b). The ‘Rose-Miller formulation’ of indirect rule, however, offers an empiricist ontology and a humanist epistemology, against that of Foucault, whose work is arguably materialist and sympathetic to a realist ontology (Arac 1980; Cain 1993; Cavallari 1980; Fairclough 1992; Frauley 2005a; Hook 2001; Lecourt 1970/1975; Olssen 2003; Pearce & Tombs 1998: 142–144; Pearce & Woodiwiss 2001; Purvis & Hunt 1993; Sprinker 1980; Veyne 1971/1997; Woodiwiss 2001, this volume). A major weakness of the dominant current within governmentality studies is the lack of a ‘depth ontology.’ Critical realism offers a model for theorizing a depth ontology, which is needed in order to go beyond the description of what are the surface effects of processes of ordering and not simply the actions of human agents, as the dominant position holds. Assuming governance to be centrally concerned with the actions of experts and authorities leads to a failure to account for social action as ‘situated activity’ and with how this is structurally enabled and constrained, and, in turn, with how ‘situated activities’ reproduce or transform the structure and organization of social fields. Governance and social ordering are reduced to the actions of human beings, actions that are held to be constitutionally distant (Rose 1994: 364–365), in that they are indirect interventions by experts and authorities into some sphere of activity, and spatially distant (Rose 1994: 364–365), referring to the ‘reach’ of these indirect interventions (see also Rose 1999: 50–51). Rose and Miller’s understanding of social fields and objects is not congruent with that of Foucault, which I will demonstrate below. Moreover, I argue that their formulation of indirect rule offers an ‘analytics of government’ from which Foucault has been expelled. As much of the governmentality literature follows the Rose-Miller formulation of the nature of distance and governance, Foucault has been expelled from much of governmentality studies, despite the many references to his texts. As Pat O’Malley, Lorna Weir, and Clifford Shearing have duly noted, ‘there has been a lack of (published) debate within the literature about major

260 Jon Frauley

issues central to governmentality work’ (1997: 505) – namely, on the lack of a critical politics, disregard of social relations, and the field’s idealist leanings. We might also add to this, a disregard for the place and role of crafting and/or refining analytic concepts. Frank Pearce and Steve Tombs (1998: 568) echo this concern over a disregard for social relations, and in relation to Rose and Mariana Valverde, and I believe of governmentality studies more generally, they argue that the appropriations of Foucault are ‘in danger of assuming the mantle of an orthodox interpretation of his work, … which in our view entails an unhelpful closure and so denial of much that his work offers to us as social scientists.’ In this chapter I argue that Foucault has been expelled from governmentality studies and that in order to provide for a richer and more nuanced concept of ‘governance’ it is necessary to retrieve both Foucault’s emphasis on ontological depth and the structured and situated nature of social action. In doing so I enlist some of Foucault’s archaeological works (1963/1973, 1968/1991, 1972b, 1970/1972) and draw on realist metatheory. Much contemporary critical realist work grows out of contributions made in the philosophy of science by Roy Bhaskar (1975, 1979; see also Collier 1994). Although Bhaskar’s work is foundational, social scientists have done much to expand on and, in some cases, reformulate his work with specific regard to sociology and sociological research. Therefore, my usage of critical realism in this chapter owes more to those sociologists who have worked over the issues and debates raised by Bhaskar than Bhaskar’s own work (e.g., Layder 1993; López & Potter 2001; Pearce & Woodiwiss 2001; Sayer 1992, 2000; Woodiwiss 1990, 2001). Drawing on a sociologically informed critical realism will bring alternative theoretical resources to bear on governmentality studies, which is a relatively insular field of study. These resources will aid in advancing the argument that Foucault can be read as a materialist and realist, and in turn will serve to open up the field to wider debate about the nature of governance as both a social scientific concept and as a complex social phenomenon. Realism and the Empiricism of the Rose-Miller Formulation Studies of governmentality practise a certain kind of empiricism … The attention to the empirical undertaken here has nothing at all to do with a valorization of experience, with a denigration of theory and the like …

The Expulsion of Foucault from Governmentality Studies 261 The kind of empirical analysis that is involved here is not hermeneutic. It is not a question of decoding or interpreting a particular strategy to discover hidden motives, of critiquing a particular alignment of forces to identify class interests or of interpreting a particular ideology to discover the real objectives that lie behind it … Against interpretation, then, I advocate superficiality, an empiricism of the surface, of identifying the differences in what is said, how it is said, and what allows it to be said and to have an effectivity. (Rose 1999: 56–57)

This concern with surfaces is reflected in the widespread belief among governmentalists that objects and governors are empirically apprehendable and that the latter are human beings who affect control of the former indirectly through the use of calculated tactics, techniques, and strategies. This, I argue, stems from the untheorized use of the term ‘distance’ within this field, which aids in reproducing the shallow ontology that underpins governmentality studies. Despite the growing number of critical appraisals of governmentality studies (Brown 2001; Curtis 1995; Dupont & Pearce 2001; Frankel 1997; Garland 1997; Hinkson 1998; Kerr 1999; Malpas 2001; O’Malley, Weir & Shearing 1997; Pearce & Tombs 1998: 39–40; Stenson 1998; Stenson 2001; Woodiwiss 1998), the role and operation of the category, distance, in governmentality theory has remained unexamined, despite the prolific use of the phrase, ‘government/action at a distance.’ Examples of the proliferation of the ‘at a distance’ maxim are plentiful (e.g., Dean 1994b: 151; Dillon 1995: 325; Hermer & Hunt 1996: 455; O’Malley, Weir & Shearing 1997: 503; Cooper 1998: 12; Hogg & Carrington, 2001: 50; Purvis, 2002: 31). Each, either overtly or implicitly, follow the Rose-Miller formulation on the nature of distance and do not engage with the above remarks of Foucault on the governing of actions ‘which may arise in the present or the future,’ nor do they attend to just how this might be accomplished. Moreover, governors are held to be only human beings despite the situated and structured nature of human action; that is, despite the governing capacity of social structure. What Foucault hints at is a process that realists term ‘emergence.’ This refers to the presence of potentialities, made possible by the structure and organization of a thing, which may be activated or impugned depending on how the causal powers of this structure are related to its external conditions. The following remarks by Andrew Sayer provide an explanation.

262 Jon Frauley The particular ways-of-acting or mechanisms exist necessarily in virtue of their object’s nature. The nature or constitution of an object and its causal powers are internally or necessarily related: a plane can fly by virtue of its aerodynamic form, engines, etc.; gunpowder can explode by virtue of its unstable chemical structure … (Sayer, 1992: 105) Whether a causal power or liability is actually activated or suffered on any occasion depends on conditions whose presence and configuration are contingent … Whether gunpowder ever does explode depends on it being in the right conditions – in the presence of a spark, etc. So although causal powers exist necessarily by virtue of the nature of the objects which possess them, it is contingent whether they are ever activated or exercised. (Sayer 1992: 107) Gunpowder has the ‘causal powers’ to explode in virtue of its unstable chemical structure. Copper can conduct electricity because of the presence of free ions in its chemical structure. Whether either of these causal powers are ever ‘realised’ or ‘activated’ depends upon contingently related conditions, such as the presence of oxygen, low humidity and a spark in the first case, and an electric current in the second. Because the conditions are independent of the causal powers … it is contingent that gunpowder ever explodes, but in certain conditions it will do so necessarily.’ (Sayer 1981: 8)

The conception of governance worked with by the vast majority of governmentalists does not and cannot capture this process of emergence or of the governing capacity of non-human entities. One implication is that the relations at work that govern the situated and structured actions of experts and authorities and that combine to produce objects that can be subjected to social and/or legal regulation cannot be identified or explained. As Russell Hogg and Kerry Carrington (2001: 46) note in passing, many categories in the governmentality literature that are used in an explanatory capacity are left undeveloped and are little more than metaphorical. According to Sayer (2000: 13), ‘Both lay and scientific thought requires not only logic but metaphors and associational thinking.’ However, as Louis Althusser (1971) and Anthony Woodiwiss (1990) argue, metaphorical descriptions are not explanations, so they must be used to further the development of analytic concepts and models that will then allow for explanation and the production of new,

The Expulsion of Foucault from Governmentality Studies 263

more penetrating, descriptions. José López (2003a: 80) puts it this way: because tendencies and constitutive processes are not necessarily empirically manifest, ‘realists have emphasised the need to explore how mechanisms, which are not readily empirically apprehended in constant conjunctions, can be modelled or “visualised.”’ As a metaphor must be accepted or rejected, it does not afford the (necessary) corrigibility of an analytic concept (Danermark, Ekström, Jakobsen, & Karlsson, 2002: 123). As we require language for this modelling, and as language itself is metaphorical, we cannot eliminate the use of metaphor in social science. We can, however, attend to the above weaknesses and distinguish between strong and weak metaphors for social science. As López (2003a: 80, see also 2003b) has argued, drawing on Sabine Maasen, ‘the metaphor should not be understood merely in a descriptive sense.’ López discusses two operations for metaphor that will help us distinguish between strong and weak: transfer and transformation. Only the latter is adequate to social science because with the former, ‘The epistemic content of the new concepts and theoretical strategies remain dependent on the semantic ties established between both domains [the old domain from which the term has emerged and the new domain to which it is being imported]. Consequently, networks of concepts will not be adequately integrated, leading to both conceptual and theoretical instability’ (López 2003a: 81, drawing on Maadsen). With transformation, new concepts and theoretical strategies are produced in the domain that hosts the metaphor (López 2003a: 81). This transformation or ‘reformulation,’ Pearce (1989: 8) argues, ‘will not represent a mere tidying up or correction of the old texts. The original texts will be displaced by the theoretical work expended upon them producing conceptual systems with significantly different substantive implications.’ When transformation or reformulation do not occur, what we are left with is a category that ‘cannot bear the explanatory weight many researchers have been tempted to put upon it’ (Sayer 2000: 19). It should be noted that the metaphor itself is not transformed but rather has transformative effects. Woodiwiss echoes these concerns, stipulating that ‘there is a danger that once one has provided a reasonably detailed account of the pertinent ‘deliberations, strategies, tactics, and devices’ that comprise governmentality … one might think that one has fully explained the phenomena involved’ (Woodiwiss 1998: 26, reference omitted). This suggests to the need to get beyond transference in order to generate

264 Jon Frauley

new descriptions of social objects and fields and explanations of the constitutive processes immanent to empirically apprehendable social phenomena. Conflating transfer and transformation has led many governmentalists to ignore the vertical order of relations and the need to develop a ‘visuality’ that can aid us in deepening our appreciation of this order of constitutive relations: ‘By and large, the partisans of governmentality … share a privileging of vision and the visible … The result is that they eschew any attempt to create a new visuality, in the sense of a theoretical structure which could make a new way of seeing possible, and which therefore could throw light on the hitherto invisible and thus deepen our understanding of the pertinent causal mechanisms’ (Woodiwiss 1998: 26). In this way a great many governmentalists succumb to what Althusser (1971) has termed ‘descriptive theory’; no attempt is made to clarify what is, to refer to Marx, an ‘abstract conception’ of distance, and this operates as an ‘epistemological obstacle’ (Althusser 1971: 134) to the development of a fuller and richer Foucauldian informed theory of governance and social ordering. Althusser elaborates through his problematizing of the determinism represented by the Marxian base-superstructure metaphor as an epistemological obstacle to be overcome or surpassed. The overcoming or surpassing of such on epistemological obstacle requires that transformative effects be furnished by this metaphor within the specific domains to which it is being applied. For this reason, Althusser introduces the analytic concepts of relative autonomy (of the political-legal and ideological realms) and reciprocal action (the conditioning aspect of these realms on the economic base or conditions of production). Althusser, takes base-superstructure as a starting point, the first step toward explanation, considering the legal-political and ideological realms that make up the superstructure as essential for the reproduction of the base rather than simply as epiphenomenal. ‘It now seems to me that it is possible and desirable to represent things differently. NB, I do not mean by this that I want to reject the classical metaphor, for that metaphor itself requires that we go beyond it [because this is the “first phase of every theory,” p. 132]. And I am not going beyond it in order to reject it as outworn. I simply want to attempt to think what it gives us in the form of a description. I believe that it is possible and necessary to think what characterizes the essential of the existence and nature of the superstructure on the basis of reproduction’ (Althusser 1971: 130–31, emphasis in original). The metaphor is taken up for the innovations it can furnish, which in turn requires creative theorizing and

The Expulsion of Foucault from Governmentality Studies 265

theoretical resources that will then afford reformulation rather than simply a straight transference. As Pearce (1989: xiii, 2, 4–8) has argued, putting theory or sets of concepts to good use requires a creative reading and reformulation as opposed to only acceptance or rejection. The problem with governmentality studies is the dearth of creative theorizing due to the field’s insular nature; alternative theoretical resources are not sought out, and metatheoretical coherence is lacking. Just as descriptive theory is necessary for a beginning but often ‘runs the risk of ‘blocking’ the development of the theory … ’ (Althusser 1971: 134), the conception of distance as a description of the reach of experts and other human governors has blocked the theorizing necessary to elevate distance to that of analytic concept. Once it is recognized that descriptive theory and transference make for a weak explanatory instrument, it becomes possible to make the necessary move toward a theoretical practice that can operate to reveal the vertical order of relations immanent to governance and thus provide a more developed understanding of distance as involving depth as well as horizontal reach. As Rose makes clear, the preoccupation of governmentalists is with surfaces and thus with the products of constitutive process, not with those processes themselves. This understanding of the objects of governance as immediately apprehendable discards the constitutive relations and orders of objects that are not directly observable but that are themselves objects of social ordering. What is actually delimited is not actions per se – we are not speaking of prohibitions – but the emergence of unruly conduct; hence, the process of emergence is central to a consideration of both the concept governance as well as governance as a complex and multidimensional social phenomenon. The dominant conception of distance at work within this field is not only empiricist but it places emphasis on the wrong ‘things’ in its attention to the empirically apprehendable and negation of the structure of constitutive relations. This provides for a weak account of both the conditions under which objects of governance become possible and the ontological nature of those objects. An analysis of what is governed, including the relations that comprise the conditions of governance, cannot be approached with much success by focusing only on the horizontal order of relations. To this end, they fail to employ Foucault’s method of analysis, which would involve not only analysis and critique of hegemonic forms of power-knowledge (i.e., discourses) but also an archaeological investigation. This is to say, as Foucault does, that a

266 Jon Frauley

genealogical approach always involves archaeology (Foucault 1976/ 1980: 85; 1997/2003b: 58–59; 1984/1985: 11–12; 1984/2003c: 138). Governmentalists, in discarding Foucault’s methodology, adopt a form of empiricism that leads them to eschew the layered and relational nature of that which is constitutive of the empirically apprehendable. The empirical or experiential realm of social life is conflated with that which actually exists and is held to be exhaustive of the latter. We can understand the empiricism of governmentality studies to be one that holds social reality to be without depth and governance to concern only the actions of human governors. To borrow from Marx again, a ‘concrete concept’ rather than a ‘chaotic conception’ of distance would include reference to a vertical order of relations. This would allow for a retrieval of Foucault, and this is where the theoretical resources of critical realism and Foucault’s early texts come in. For realists, the world comprises overlapping and interacting layers or what I will term orders: the real (the realm of tendencies or potentialities), the actual (the realm of actualized or realized potential, which in turn can generate further effects), and the empirical (the realm of sensory perception or experience). ‘In distinguishing the real, the actual and the empirical, critical realism proposes a “stratified ontology” in contrast to other ontologies which have “flat” ontologies populated by either the actual or the empirical, or a conflation of the two’ (Sayer 2000: 12).3 Foucault’s mode of analysis, with its attention to differing orders, can be seen as a type of realist ontology and materialist treatment of discourses (and the power and knowledge that these carry). Foucault and the Vertical Order of Relations In The Birth of the Clinic, Foucault gets beyond the surface of spatial metaphors to a ‘deeper level’ (1963/1973: 4) to interrogate the eighteenth-century ‘configuration of disease.’ ‘This organization treats localization in the organism as a subsidiary problem, but defines a fundamental system of relations involving envelopments, subordinations, divisions, resemblances. This space,’ Foucault (1963/1973: 5; emphasis added) continues, involves a vertical and a horizontal dimension; it involves ‘a deep space, anterior to all perceptions, and governing them from afar.’ Implying a need to consider the ontology of disease, Foucault (1963/1973: 199; emphasis added) concludes this text with the following: ‘The formation of clinical medicine is merely one of the

The Expulsion of Foucault from Governmentality Studies 267

more visible witnesses to these changes in the fundamental structures of experience; it is obvious that these changes go well beyond what might be made out from a cursory reading of positivism. But when one carries out a vertical investigation of this positivism, one sees the emergence of a whole series of figures – hidden by it, but also indispensable to its birth – that will be released later, and, paradoxically, used against it.’ In his next book, The Archaeology of Knowledge, he also speaks of formations and vertical relations. In referring to a ‘unity of discourse,’ that is, a ‘discursive formation’ founded in a ‘group of rules’, he stipulates, ‘… if there really is a unity, it does not lie in the visible, horizontal coherence of the elements formed; it resides, well anterior to their formation, in the system that makes possible and governs that formation’ (Foucault 1972b: 72). This ‘system of formation’ of the orders that make up a discursive formation – objects, enunciative modalities or positions, concepts, and strategies – concerns the vertical dimension of discursive formations: ‘In fact, the different levels thus defined are not independent of one another … In this way, there exists a vertical system of dependencies …,’ which Foucault speaks of as a ‘whole hierarchy of relations’ (1972b: 72–73) that make up the orders of a discursive formation. ‘By system of formation, then, I mean a complex group of relations that function as a rule: it lays down what must be related, in a particular discursive practice, for such and such an enunciation to be made, for such and such a concept to be used, for such and such a strategy to be organized’ (1972b: 74). Foucault is discussing the emergence of these ensembles or apparatuses; or, if one likes, the formation of formations. A discursive formation necessarily concerns a vertical order of relations that intersect to produce that which emerges to be taken as a unity that can be subject to regulation or one sort or another. Importantly, it is the ‘system of formation’ that we can gain access to through studying emergent objects and their domains (Foucault 1972b: 47), but only if we are working with a depth ontology. Identifying discursive formations and describing the formation of discourses requires an analytical model not well developed in governmentality studies. As we can see, and as Woodiwiss (2001: 152–153) has argued, Foucault’s work ‘was premised upon a gradually emerging but in the end very sophisticated, non-humanist metatheory that has all the hallmarks of realism: an ontological insistence on the nonminded and material character of social reality that because of its ontological depth subsists as structures that, in terms of epistemology, cannot be observed spontaneously or directly but instead have to be

268 Jon Frauley

modelled and tested before they can be spoken of with any and only ever a provisional confidence.’ From the archaeological texts, we see that Foucault does have a conception of the vertical and the oftenunobservable constitutive relations that produce and reproduce that which is empirically apprehendable. We can find in his texts a conception of a depth ontology that can be employed to deepen our understanding of both the concept and material forms of social ordering and for understanding governance as a process that also always concerns social relations. Moreover, he made a distinction between that which actually existed and that which was immediately observable. Additionally, he seems to be suggesting that that which is ‘indispensable’ to the birth of clinical medicine consists of the tendencies of a ‘series of figures’ that may actualize or impugn the exclusion, limitation, and social appropriation of knowledge and social power. In the main, governmentalists have yet to identify or take up these issues, instead employing Foucault’s genealogical works in an instrumental and unreflective manner. Realism, Structural Orders, and Situated Action The shadow of empiricism that is cast over much of this literature impresses on researchers a misguided need to enquire into governance after it has emerged as such – enquiries that assume and treat the objects of rule as overt to the neglect of constitutive and unobservable social processes and relations (cf. O’Malley et al. 1997; Woodiwiss 1998: 25–28; Hinkson 1998: 135). This is not to say that a consideration of empirically apprehendable objects subjected to regulatory programmes is not important or useful. Rather, it is to say, as I have shown, to solely focus on these is to seriously delimit the purview or scope of enquiry and to foreclose on the developing of an ontology of governance. It is also to break with Foucault’s methodological position, which constitutes a tacit rejection and expulsion of Foucault rather than transformation or reformulation and development of his concepts. I am arguing, therefore, for a shift in focus toward ontology in order to include the vertical order immanent to social ordering. If governance is indeed ‘an action upon an action, on existing actions or on those which may arise in the present or the future,’ as Foucault stipulates at the opening of this chapter, it must be much more than the regulation of observable, horizontally distributed social objects; it must also be to act on existing or potential relations, on vertical orders of

The Expulsion of Foucault from Governmentality Studies 269

relations and domains and processes of emergence; this would have a direct bearing on what could be actualized or realized as an object of governance (or on what would emerge as something that could be observed, objectified, and subjected to social regulation of some kind). Foucault further stipulates, ‘The exercise of power consists in guiding the possibility of conduct and putting in order the possible outcome. Basically power is less a confrontation between two adversaries or the linking of one to the other than a question of government … modes of action, more or less considered and calculated, which were destined to act upon the possibilities of action of other people. To govern, in this sense, is to structure the possible field of action of others’ (1983: 221, emphasis added). Returning to The Birth of the Clinic, we might conceptualize ‘changes in the fundamental structures of experience’ as a reference to the structure of a discursive formation. Pierre Bourdieu (1977b, 1980/1990) provides us with the concepts habitus and field, and his illustration of their articulation is useful here (see also Frauley, 2005b, chapter 3). Habitus, as the cognitive structure of a subject, is the product of alignment to and interiorization of the structural arrangements of social fields. Habitus is an object-like aspect of social subjects that allows for their reproduction and transformation of their structured settings. Following this, indirect rule as the ‘conduct of conduct’ (as it is often referred to by governmentalists) concerns the transformation or governance of the habitus, including its alignment with the larger social field. This process therefore entails the ordering of social fields, institutional arrangements, and subject positions. Given this, it would be incorrect to argue, as many governmenatalists do, that governance concerns only indirect interventions of experts and authorities. Foucault is clear that action is important to a consideration of governance but does so in a nonhumanist way. Posibilities or potentialities are also a crucial consideration for the study of government, making it necessary to focus on the real and actual conditions from which objects arise and not simply on that which is empirically apprehendable. To do this requires a conception of depth, and as I have shown, Foucault’s genealogy (concerned with the formation of fields of social objects, including discourses) and archaeology (concerned with the depth of relations of the field of social objects) indicates an awareness of the importance of a ‘depth ontology.’ Exercises of power and resistance, then, must be understood as the actualizing of what is already possible given the field of relations. That is, to not exercise power does not mean that power does not exist, it simply means it is unexercised or that the potential is unrealized/

270 Jon Frauley

unactualized. Foucault may sometimes leave us with the impression that social action is primary where power is concerned: ‘… something called Power, with or without a capital letter, which is assumed to exist universally in a concentrated or diffused form, does not exist. Power exists only when it is put into action, even if, of course, it is integrated into a disparate field of possibilities brought to bear upon permanent structures’ (Foucault 1983: 219). However, considering that his ‘fundamental structures of experience’ resembles Bourdieu’s habitus, we must see this statement as one that concerns the activating or actualizing of potential, given the field of relations in which action takes place: ‘Power is relations; power is not a thing, it is a relationship between two individuals … ’ (Foucault 1981/1997: 156–57). As ‘individuals’ refers to social subjects in field of structured positions (Foucault 1976/1980: 98), power relations obtain between two social subjects – subject positions – within a structured context. Because of this, power is part of a system, a ‘net-like organisation,’ a chain or matrix of relations (Foucault 1976/1980: 98) and is not exhausted by action but rather is potential that may be actualized, a passive or latent power that might be activated. This formulation has to do with the process of emergence and concerns both interpersonal relations as well as structural conditions in how, where, and what form power becomes manifest or actualized in programmes of governing: ‘In other words, when one sees what power is, it is the exercise of something that one could call government in a very wide sense of the term’ (Foucault 1981/1997: 156–57; emphasis added). He (1981/1997: 156–157) goes on to stipulate that governmentality is ‘the group of relations of power … and techniques which allow these relations to be exercised’ – relations as it is used here can be read to refer to potential while techniques can be read as referring to mechanisms, ‘instruments,’ or structured positions that facilitate the activating or actualizing of potential. To ‘structure the possible field of action of others,’ then, would not be so much a concern with social action as in the ‘interaction of human beings’ as it would be with the intersection of techniques, subjectpositions, and a concern with delimiting the activating or actualizing of potential through structural arrangements and the process of aligning habitus to fields. This latter point may be one reason why Foucault refuses a juridical notion of power, as it would hold that power is repressive or prohibitive of what has already been actualized. Fundamental here is the process of emergence – the actualizing of potential – and the indication of a vertical order of relations. It is important to

The Expulsion of Foucault from Governmentality Studies 271

realize that in stipulating that power may be exercised through certain techniques (mechanisms), at which time power becomes observable as ‘government,’ Foucault is delineating differing orders, distinguishing between power as potential, power as actualized (exercised), and power as something that is empirically apprehendable. This view of power and its operation is compatible with a realist ontology: ‘On the realist view, causality concerns not a relationship between discrete events (“Cause and Effect”), but the “causal powers” or “liabilities” of objects or relations, or more generally their ways-ofacting or “mechanisms” … Often the causal powers inhere not simply in single objects or individuals but in the social relations and structures which they form … Powers and liabilities can exist whether or not they are being exercised or suffered’ (Sayer 1992: 105, emphasis added). Distinguishing between these orders of power, attending to the relational nature of action and structure, and power’s operation via concrete techniques and mechanisms allow us to more clearly specify the components of the concept governance, as well as to indicate the importance of a depth ontology for understanding governance as a process that works ‘to structure the possible field of action of others.’ To govern concerns not only what some thing does, but also what it may do, and realists offer us a way to theorize what things are capable of despite how they are observed to operate. A realist ontology can be combined with Foucault’s archaeological approach and can be used to supplement underdeveloped conceptions in governmentality theory. ‘Causal powers’ can be conceptualized as capacity to do something, to be productive, while liabilities can be thought of as unactualized or perhaps passive powers, as ‘specific susceptibilities to certain kinds of change’ (Sayer 2000: 11). As governmentalists do not work with an ontology that allows for the investigation of these potentials or the vertical orders of relations, much of governmentality studies effects an expulsion of Foucault. However, as I have argued, Foucault can be retrieved by attending to his emphasis on emergence and structure, most clear in his Archaeology of Knowledge. These implications of the empiricist ontology and humanist epistemology of the dominant interpretation of Foucault in governmentality studies explains, at least in part, not only the neglect of vertical relations of distance but also the neglect of social structure. We can see, then, that the dominant humanist epistemology and empiricist ontology of governmentality studies is not adequate to the complexities of the way that social orders are governed.

272 Jon Frauley

Conclusion From the foregoing it is clear that Foucault conceived of social orders as complex and as having depth. This ontology is not represented in the dominant interpretation of his work in governmentality theory. Moreover, Foucault emphasized social structure and the structuring and conditioning aspect of social processes and relations that, again, are not represented within the dominant current within governmentality studies. This absence, I have argued, is because much of the extant literature approaches Foucault through the work of Rose and Miller, which, ironically, has effected an expulsion of Foucault and constitutes an epistemological obstacle to be reformulated and surpassed. The analytic concepts developed by realists, and those furnished by Foucault, are more adequate to an analysis of governance than those currently employed and can be usefully combined to advance a more nuanced visuality for understanding, describing, and explaining the operation of governance and social ordering. Critical realism and the archaeological Foucault can be drawn on to theorize that ‘[t]here is more to the world, then, than patterns of events. It has ontological depth: events arise from the workings of mechanisms which derive from the structures of objects, and they take place within geo-historical contexts’ (Sayer, 2000: 15). To stop at the level of tactics or techniques of government, which most governmentalists do (Woodiwiss 1998: 26), is to be unable to explain governance in that one would never get to orders of power other than that which is expressed at the empirical level as social action.

NOTES 1 Some recent books include, in legal studies: Hunt and Wickham 1994, Cooper 1998, Wickham and Pavlich 2001; in criminology: Smandych 1999; in political sociology: Barry et al. 1996, Rose 1999, Dean 1999; in cultural studies: Bratich et al. 2003. 2 For a rejoinder to Rose and Miller 1992 (above: 180), see Curtis 1995. 3 Following Sayer 1992, these categories do not refer to a hierarchy or to distinct levels so much as domains or overlapping layers (see also Layder 1993).

14 From Foucault’s Genealogy to Aleatory Materialism: Realism, Nominalism, and Politics RONJON PAUL DATTA

Introduction Recent debates about Michel Foucault’s epistemology and ontology have highlighted a tension in his work between his explicit nominalism/ antirealism (e.g., Rose & Miller 1992; cf. Veyne 1978/1997) and his implicit realism (Pearce & Woodiwiss 2001). Critical realists have yet to engage with the question of Foucault’s nominalism, dismissing poststructuralism as ‘idealism’ (Bhaskar 1989: 2, 188–191; Sayer 2001: 32). Also, contemporary Foucauldians have neglected the explication of their metatheoretical commitments. I aim to clarify and intervene in this debate by drawing on Roy Bhaskar’s identification of the epistemic fallacy in Immanuel Kant, a fallacy that leaves traces in Foucault’s work, producing an unhelpful and unecessary limitation on it for radical sociology and politics. I argue that Foucault’s Kantianism and consequent nominalism, which I call critical nominalism (hereafter CN), constrains his analysis of the political because it is unable to coherently conceptualize ‘societies.’ This inability leads to his recommended displacement of politics by ethics. ‘Ethics’ refers to the process by which a subject takes itself as an object to be transformed by discursive practices in order to achieve a goal chosen by the subject, the best goal being perpetual creativity and experimentation with living both for the self and for others (Foucault 1984/2003b; 1984/1996). Thus, the question I pose to Foucault’s work is, Why displace politics with ethics? This question is important because it forces us to assess Foucault’s reasoning about what the target and mode of transformation should be. If his reasons are implausible, and one can show how this implausibility is produced, then there are grounds for reformulating the terrain, a task I take up here.

274 Ronjon Paul Datta

In reading Foucault symptomatically, I explicate and assess his CN problematic, his dominant means of posing questions and producing accounts (Althusser 1965/1970a; Pearce 2001; Negri 1995). Contra Paul Rabinow and Nikolas Rose (2003: xv), this process does not imply the search for ‘hidden interests’ but means reading Foucault exactly to the letter. Now, while I accept the argument that Foucault is a realist about discursive formations (Pearce & Woodiwiss 2001), I wish to account for why Foucault is not an epistemological realist about anything else without contradicting CN. The result is the persistent problem of the extradiscursive dimension. Reading Foucault symptomatically allows one to assess why he argues that the human sciences, like Althusserian Marxism, are part of the problem and not a part of a viable solution (Foucault 1997/2003a: 11; Foucault 1984/2003b: 104; Rabinow & Rose 2003: xv; Rose & Miller 1992: 177). At the same time, Foucault’s genealogical writings are traversed by a materalist-realism/‘ordinary realism’ (Pearce & Woodiwiss 2001) about societies much indebted to Louis Althusser, one compatible with materialist-realism but incompatible with CN. The ‘discovery’ of aleatory materialism (hereafter AM), the major statements of which have recently been translated into English as Philosophy of the Encounter (Althusser 1993/2006) and fruitful re-engagement with Althusser in Judith Butler (1997) and Alain Badiou (1998/2005) indicates that the common dogmatic rejection of Althusser is specious (McInerney 2005; Terray 1995; Negri 1995; Matheron 1998: 27; Elliot 1998; Navarro 1998: 94; cf. Pearce & Dupont 2001: 151).1 Furthermore, I find Roy Bhaskar’s critical realism (1989), with its humanism, dialectics, and metaphysics of ‘deep structures,’ theoretically and practically retrogressive. So while I agree with linking realism and emancipatory politics, the AM emphasis on forming a theory of politics and social formations, as structured surface effects (cf. Deleuze 1972/ 2004; Patton 2000: 45), provides an alternative normative justification for realist-materialist social science, distinct from critical realism. Working on Foucault and Althusser, while deploying Bhaskar’s theoretical critiques, is, I think, a fruitful path for opening a viable alternative to Foucault and Bhaskar in AM – a ‘heterodox realism.’ Initial Definitions CN produces a consistency in Foucault’s oeuvre, from his early transformation of Kant to his later work on ethics. The term ‘critical’ is indicative of Foucault’s debt to Kant, since for both, the task of philosophy

From Foucault’s Genealogy to Aleatory Materialism 275

requires that we ‘know knowledge’ (Foucault 1997/2003b; cf. 1982/ 2003: 128). Foucault, however, also transformed Kant, and in the process the meaning of ‘critical.’ The result is Foucault’s nominalism, in which human existence ‘in-itself’ is not taken as an independently existing ‘thing.’ Rather, he argues that there are only objectifications of unformed bodily capacities produced by discursive practices (cf. Rajchman 1985: 74). AM is a theory of politics, society, and history, and it offers a distinctive justification for theoretical practice. Its chief concern is with the real-concrete objects of social formations and real mechanisms of political transformation. Its central theorem is that politics happens because of a chance (‘aleatory’; ‘contingent’) encounter between virtue and fortune (Althusser 1999: 74). ‘Virtue’ refers to (i) the theoretical task of explaining ‘things as they are in actual truth’ (7); (ii) offering an account of how things might be otherwise (i.e., theory as political imagination), a radical intervention involving thinking what is possible yet deemed impossible according to dominant conceptions (Negri 1995: 54); (iii) the courage to act in order to actualize the possible (Althusser 1999: 84); and (iv) the massive organization of social forces that enable action, potentially shifting the balance of social forces (Althusser 1999: 11, 25, 50). ‘Fortune’ refers to the singular structure of a social formation under its conjuncture that is both overdetermined by some dominant instance that gives social formations their structure, and underdetermined by the exceptional quality of politics in relation to a state of rule (Althusser 1999: 17–18; Montag 1998; Matheron 1998: 31). In realist terms, I take underdetermination to mean that the potentials within a social formation are not exhausted by the actually existing arrangement of social forces (Sayer 2001: 11–12). The difference between the actual and potential is the space/void in which the battle between politics and hegemony ‘takes place’ (Althusser 1982/2006: 172ff; 1999: 68). The task of theory is to name this void (cf. Matheron 1998). Thus, politics refers to the exceptional event that ‘proves’ that there is ‘rule.’ Politics irrupts in history when there is a chance encounter between virtue and fortune that allows a new structuring of a social formation, a new chance encounter that lasts (Althusser 1999: 20–21; Matheron 1998: 33). So, while CN engages in writing ‘a history of the present’ (Foucault 1975/1979: 31), AM is concerned with the present and the future; a future that lasts is at stake in present politics. The key difference between CN and AM concerns the materialist position on the knowledge-object and the materiality of the real, social

276 Ronjon Paul Datta

object. Bhaskar’s realism is useful for clarifying this difference. In CN, the knowledge-object and the real, social object are the same. For both Bhaskar’s realism and AM on the other hand, the object (the realconcrete) is external to the ‘object of knowledge.’2 The difference between the real-object/concrete-object that exists independently of the production of knowledge about it (thought-concrete), an object that exists before as after the process of knowledge production is also the point of contact between Althusser’s materialism and Bhaskar’s early realism (Althusser 1988/2006: 274; 1965/1970a: 41–42, 87; 1972/1976: 54). For the early Bhaskar, knowledge of the social world is transitive, it changes and develops, whereas the real object of the social world is (relatively) intransitive (López & Potter 2001: 12–14): the production of knowledge of the world does not change the nature of the object. Thus, AM, like Bhaskar’s early realism, requires that we put knowledge ‘in its place’ within a social formation. For CN, on the contrary, there is no difference between the transitive and intransitive dimensions of knowledge: the process of producing knowledge about ‘humans’ itself produces the reality of this object. A second difference lies in their respective orientation toward the present. CN writes histories of the present to show how what we take as self-evident and natural is an effect of contingently assembled diverse elements that have stablized but remain ‘open’ (Foucault 1975/1979: 31; 1980/2003). AM, on the other hand, endeavours ‘to think theoretically from the point of view of a task to be accomplished and not from the angle of the necessity of an accomplished fact’ (Matheron 1998: 32). Nevertheless, both CN and AM emphasize contingency, although in AM, in a manner similar to realist social science, contingency is always conceptualized alongside necessity (Elliot 1998; Sayer 2001: 10). Foucault’s Concerns Foucault’s main object of investigation and concern is with ‘the history of the different modes by which, in our culture, human beings are made subjects … modes of objectification that transform human beings into subjects’ (1982/2003: 126). His work elucidates the contingency and arbitrariness of subjectification (the practices and their conditions of existence constitutive of forms of subjectivity) (1984/2003b: 33) through which subjects render existence intelligible and that also enable and constrain their capacities to act (1983/ 1996a: 359).

From Foucault’s Genealogy to Aleatory Materialism 277

Contingency supplies Foucault’s optimism about ethics: if something contingent and small managed to constitute our present, it is likely to happen again. Nevertheless, Foucault stresses the extent to which ‘modes of being’ do not change since ‘the present’ has lasted a long time, from roughly the mid-nineteenth century until now (1984/2003b: 26). Thus, Foucault’s histories describe how madness, illness, crime, etc., have, mutatis mutandis, remained fundamental experiences in modern Western civilization. Much of what we are today is still being made through the same practices (and their constitutive conditions of existence) that congealed in the mid-nineteenth-century West.3 While Foucault is optimistic about ethics, he viewed the linkage of the human sciences and politics as central to disasters like Nazism and Stalinism (Foucault 1982/2003: 127). The dominance of the will to truth about humans, constituted through the production of veridical discourse about humans (judgements about what is true and what is false), which in turn produces effects of jurisdiction (judgements about what is to be done), is held to be the defining feature of the present (Foucault 1980/2003: 248). As individuals and collectives, we have come to think of and treat ourselves (i.e., ‘problematize’ existence) in and through the terms of the human sciences, e.g., as healthy, normal, reasonable beings. This conception exists regardless of capitalism, state-form or political ideology: American liberalism, Stalinism, or Nazism are all reliant on the techniques found in the human sciences for their programs (Foucault 1977/1990: 180, 182; 1974/1996; 1972/1996). Foucault’s Transformation of Kant: The A Priori and Experience Foucault’s position is determined by his adoption of Kantian concerns, and he poses the same questions as Kant concerning ‘Man,’ knowledge, limits, experience, and ‘the present’ (Foucault 2003: 1; 1982/2003: 128; cf. 1984/2003c: 56; 1966/1971: 340–1; Han 2002: 3). Foucault’s debt to Kant requires that we attend to the consequences of the Kantian Kampfplatz (‘battleground’) that Foucault’s work occupies, since it structures his problematic. Kant sought to account for the conditions of possibility of true/legitimate knowledge and sought a concept of the mechanism through which the welter of sensations is rendered intelligible. Since intelligibility is itself not given by the sources of stimuli beyond the subject (the world as it is in-itself), the source of the intelligibility of sensations that are the ‘stuff’ from which experience is made

278 Ronjon Paul Datta

was argued to exist a priori, existing independently of all experience. Sense suggests that there is something active in, and transcendent of experience, namely, the structure of the human mind. Crucially, the object of knowledge is phenomenal (its form is determined by how it is perceived), since it is impossible to have direct cognitive access to things in-themselves. Rather, knowledge of the world is the synthetic product of the a priori and the external world, irreducible to either mind or world. A priorism has important implications for understanding CN, since for Kant and Foucault the object and the reality to which it belongs exist in the field of knowledge, not external to it. Appreciating Kant’s account of experience and the status of the object of knowledge also allows us to understand his and Foucault’s attention to discursive and subjective formations: for both, an a priori forms and orders sensations into perceptions (experience) and allows us to make judgements about perceptions, thus creating conceptions (thoughts and judgements about perceptions). So form shapes content (the world as it is in-itself) to render our ‘reality.’ Kant’s postulation of the a priori allowed him to ‘know knowledge,’ allowing reason to legislate the appropriate limits of knowledge. Foucault’s transformation of Kant occurs through two main moves. First, the human becomes a historical explanadum and not an explanans of knowledge. Second, the a priori is not taken as a universal but as something historically specific (Foucault 1977/2003a: 316; 1984/2003c: 53–54). Nevertheless, a Kantian sensibility remains because Foucault argues that experience is produced by an a priori, albeit a historical one called ‘the archive’ (Foucault 1972a: 27). So Foucault, like Kant, theorizes how an a priori structures experience or perception. The archive comprises the set of statements actually pronounced deemed ‘true,’ and these statements are material in that they ‘have a substance, a support, a site and a date’ (Foucault 1972a: 26–7, 131, 101; cf. Lecourt 1970/1975: 195). Archaeology aims to explain the regularity of these archival statements. Discursive formations are ‘structures’ productive of the archive (historical a priori) because they are the production condition of intelligibility (Foucault 1984/2003a: 61; 1980/ 2003: 254). As such, the archive is the primary but insufficient condition for the existence of a discursive formation because the activity of knowing requires institutional conditions and the constitution of knowers and known. Foucault’s archaeologies of knowledge about humans start with certain forms of experience (1984/2003a: 59–60; Rabinow & Rose 2003: viii). These experiences include ‘madness, illness, transgression of laws,

From Foucault’s Genealogy to Aleatory Materialism 279

sexuality, self-identity … suffering, death, crime, desire, individuality’ (Foucault 1979/2003a: 190–191), and they can be individual or collective (Foucault 1984/2003a: 60; 1983/1996b: 408). Experiences orient and govern individual and collective life in the present, since they are the basic material for making judgements of veridiction and jurisdiction. A Foucauldian concept of experience can thus be defined as the form through which both subject and object come to be related within a knowledge terrain through practices of seeing and saying combined in systems of thought. This allows Foucault to delineate how subjects apprehend their present, orient themselves to it, and analyse how this apprehension also partially constitutes the subjective capacity to render experiences of being human intelligible and hence be ‘subjects’ of their historical world. What Foucault rejected was Kant’s concept of the a priori, which located it within the universal of human subjectivity (Foucault 1966/1971: 344; 1972b: 12). The central objects of the archaeology of the human sciences are savoirs, the condition of possibility for the practical know-how of subjects engaged in knowing humans and making statements about humans as objects, governed by the immanent, varied, and uneven criteria for making true statements about a discourse object found in discursive formations. Savoirs include the activities of doctors, police, teachers, etc., their record keeping, training manuals, and interventions done in the name of science and humanity. Thus savoirs are what leave the sediments of statements behind (Foucault 1972b: 15). So the concept of discursive formation answers the Kantian/critical injunction to ‘know knowledge.’ Central to these practices of objectification of humans are the mechanisms productive of illumination that render possible objects of knowledge visible/perceptible. Seeing the object is made possible by the conditions that illuminate the object and allow it to be spoken about (e.g., the teaching hospital or the panoptic prison). The statement is the active, actual ‘expression’ of the power of a discursive formation giving objects their determinate form, actually making them ‘objects’ (Deleuze 1986/1988: 50ff.). Crucially, saying refers to the object that is seen and not to a world that exists independently of the discursive formation; knowing happens within the terrain of knowledge. So, as in René Magritte’s painting Ceci n’est pas une pipe, the script (saying) refers to the drawing of the tobacco pipe (seeing) and vice versa, not to a real-concrete pipe ‘beyond’ the canvas: the whole apparatus of modern sense exists on the plane of the canvas; perception and judgements about it occur in ‘systems of thought.’

280 Ronjon Paul Datta

Thus, discursive formations are mechanisms of objectification, and since Foucault is concerned with the human sciences, discursive formations are mechanisms productive of modes of objectification of subjects, giving them a determinate mode of existence. I define ‘mode’ as referring to the synthesis of ‘form’(i.e., discursive practices) with ‘content’ (‘pre-discursive referents’ or human life) (Foucault 1972b: 7). Discursive formations also constitute subjects whose statements can be recognized as either true or false, according to the constituent ‘rules of formation’ (Foucault 1972b: 40). In this sense, discursive formations can usefully be thought of as material modes of production of the transitive dimension of knowledge (cf. Althusser 1965/1970a: 42). Discursive formations, like social formations, have a specific kind of structural causality since their causes are not ‘external to phenomenal effects’ (Althusser 1965/1970b: 188); rather, ‘the structure is immanent in its effects in the Spinozist sense of the term, that is, the whole existence of the structure consists in its effects, in short is merely a specific combination of its peculiar elements’ (189; Foucault 1972b: 107; Lecourt 1970/1975: 195). Discursive formations as causal mechanisms are present in their effects as discursive practices and are constitutive of modes of objectification. Foucault’s account of objectification allows him to displace the ‘science versus ideology’ problem in the terrain of the human sciences. Since ‘man’ is an invention of the modern epoch (Foucault 1966/1971), ‘man’ is not a thing external to knowledge and hence cannot function as an external referent to be wrong about – i.e., realist fallibalism is untenable (cf. López & Potter 2001: 14). Since the object is always already produced within a terrain governed by the true, there is no question of ‘ideology’: the referent and the statement about the object are produced in one and the same process. To dissolve this difference is to dispense with the ground on which the problem of ‘science versus ideology’ is predicated. However, Foucault’s answers to his Kantian questions lead logically to his nominalism and then to his displacement of politics by ethics, the moves to which I now attend. Foucault’s Nominalism: A Critique Foucault’s attention to the contingency and historical variability of experiences led him to his nominalism, a label he adopted: there is, he claims, no real, independent existence of subjectivity, self, society, state, etc. (Foucault 1980/2003: 349; Rajchman 1985: 51; Veyne 1971/1997: 155, 157, 167). He states that ‘we wish to dispense with things’ and ‘to

From Foucault’s Genealogy to Aleatory Materialism 281

define these objects without reference to the ground of things, but by relating them to the set of rules that enable them to be formed as objects of a discourse and thus constitute the conditions of their historical appearance’ (1972b: 47–48). CN protocols hold that the object of knowledge does not have an existence independent of its constitution in a discursive formation: ‘things’ do not exist but only objectifications do, ones that occur through discursive practices, i.e., the material relation obtaining between the knower and the known (who happens to be a human subject). These objectifications do not refer to ‘anything’ beyond their constitution in discursive formations; rather, they are effects of relations, the existence of which are determined by practices. Foucault’s nominalism about human existence is thus a result of his critical histories, and at the same time it consolidates his position (cf. Foucault 1980/2003: 258). In short, CN is an anti-realism when it comes to human existence. Foucault explicitly identified with nominalism from the writing of The History of Sexuality onward. There he writes, ‘One needs to be nominalistic, no doubt: power is not an institution, and not a structure, neither is it a certain strength we are endowed with; it is the name that one attributes to a complex strategical situation in a particular society’ (1976/1994: 93). This nominalism also subtends his work on liberalism (1979/2003a,b: 202) and is linked with his Kantianism (1984/2003c: 200). However, are human modes of being and knowing one and the same? The issue hinges on materialism and the limits that Foucault places on it as a consequence of CN. The realism that Althusser and Bhaskar take from Marx is based on the distinction between the real-concrete and the concrete-in-thought. As Althusser explicates, ‘This distinction is contained in the very phrase in which Marx deals with the process of knowledge. As a materialist, he argues that knowledge is knowledge of a real object (Marx says a real subject) which (I quote) “remains, after as before, outside the intellect and independent of it.” And a little later, in reference to the subject of investigation, society, he writes (I quote) that it “must always be envisaged therefore as the precondition of comprehension” … if this necessary distinction is not solidly supported it may lead to nominalism’ (Althusser 1975/1990: 227–8; cf. 1988/2006: 274; 1965/ 1970a: 40–43). But CN exactly collapses the two: the knowledge production process of the human sciences are argued to be what is constitutive of the real object, i.e., modes of human existence. The problem is not ‘idealism’, since Foucault’s account of the knowledge production process is properly materialist and structural, but is the

282 Ronjon Paul Datta

limited scope placed on this materialism of the human sciences. The problem concerns the ontological status of the object to be transformed (i.e., forms of subjectivity): ‘naming and knowing what is called human’ is taken to constitute human reality. Thus, the knowledge object and the real object are collapsed together in the same process of objectification that takes place in discursive formations. For Foucault, then, the question of transforming the present requires that we concern ourselves with the ‘naming’ of human existence that objectifies the subject, thus subjugating subjects to the dominance of the human sciences (and their conditions of existence) and uses. To change the subjugations of the present, i.e., to transform modes of being, we must change the discursive practices so that we are not subjected to the will to truth that constrains the present. Foucault’s normative recommendations follow logically from CN, hence his endorsement of ‘subjugated knowledges,’ ethics, arts of existence, and ‘experimentation,’ none of which are governed by the will to truth (1997/2003a: 7; 1984/2003c: 56). Significant limitations emerge from CN, and they follow from Foucault’s Kantian Kampfplatz. Bhaskar’s critique of the Kantian position is helpful here, since Foucault commits his own version of the Kantian epistemic fallacy. Kant committed the epistemic fallacy through his critical approach to the question of knowledge and its limits. The epistemic fallacy ‘consists in the view that statements about being can be reduced to or analysed in terms of statements about knowledge; i.e., that ontological questions can always be transposed into epistemological terms,’ and ‘Kant committed it in arguing that the categories ‘allow only of empirical employment and have no meaning whatsoever when not applied to objects of possible experience; that is to the world of sense’ [Kant, Critique of Pure Reason B 274] … More generally, the epistemic fallacy is manifest in a persistent tendency to read the conditions of a particular concept of knowledge into an implicit concept of the world’ (Bhaskar 1998a: 27, 28–29). As Bhaskar has argued, all knowledge claims presuppose some conception of what the world must be like such that knowledge is possible and relies on an ontology, some conception of a world external to knowledge over which knowledge claims compete (1989: 13). The correlate is that Kant is an agnostic realist: we must postulate the existence of the world as it is in-itself although we cannot know it; we can only know knowledge (i.e., phenomena). However, Foucault’s acceptance of the Kantian Kampfplatz leads him to commit a fallacy similar to Kant’s that can be called the onticodiscursivist fallacy. It goes something like this: since knowledge of

From Foucault’s Genealogy to Aleatory Materialism 283

humanity is produced through discourse about humanity (discursive formations that objectify human existence, giving it its form), all we can know and need know about humanity (know about modes of being) are discourses, and since discursive practices produce the object that is ‘human’, we know the material object ‘human,’ by knowing the discursive practice. This process also leads to the solution that the transformation of the object that is ‘human’ is possible if we take up discursive practices that are not governed by the will to truth (e.g., ethics as the ‘thoughtful’ arts of existence). Thus, according to Foucault’s reasoning, all that we need know is given in the archive since the archive is the a priori of discursive practices that give existence determinate form (i.e., objectifies human life), constituting the present. Hence the justification for analysing systems of thought as the form that being critical must take. Foucault’s position is reductive in two senses: ontologically, it reduces modes of human existence to its formal determination qua objectification in a discursive practice; and normatively, it reduces the transformation of human existence to ethics. There is also a resulting circularity stemming from Foucault’s Kantian Kampfplatz. Foucault takes the point of view of the problem of thought, knowledge, and experience (by historicising the a priori), and these then come to be seen as the problem, i.e., the techniques and conditions of the human sciences are what has subjugated us. The terrain of his problematic is self-referential: the epistemological point of view on possible knowledge of human existence produces his ontology, and his nominalist ontology is concerned with the conditions that limit the capacities of subjects. The nominalist problematic leads to the normative position of targeting the transformation of the human sciences. Thus, ‘ethics’ is the logical normative conclusion to be drawn from CN since it aims at a new kind of experimental thought of existence. But this conclusion holds only if one believes that we can only ‘know knowledge’ and that we cannot know anything about what is beyond knowledge, i.e., what is beyond the field of discursive practices. Were we to do so, we would need to put knowledge in its real place. CN leads to another problem that plagues Foucault’s Kantianism, that of the extra-discursive that continually emerges and is never resolved. Foucault’s explicit remarks about the category of the extra-discursive are rather thin, even if they populate the margins of his works. They amount to descriptions of the relations between the order of discourse and the rest of human existence. But he fails to specify this relationship

284 Ronjon Paul Datta

and ‘the mechanism governing the relations of those two systems’ (Lecourt 1970/1975: 209). In one very explicit statement on the extradiscursive, Foucault notes that historical analyses of discourse must involve an analysis of the interrelations between discourse and a broader range of transformations. These include ‘Extra-discursive dependencies (between discursive transformations and transformations outside of discourse: for example, correlations studied in Histoire de la Folie and Birth of the Clinic between medical discourse and a whole play of economic, political and social changes’ (1968/1991: 58; cf. 1972b: 40; 1973/1996). For example, in ‘The Birth of Social Medicine’ (1974/2003) he describes how medicalization was used by the bourgeoisie to target the urban poor. He links the biopolitics of social medicine to economic changes, property regimes, economic production, and demographic changes in the population. Furthermore, doing archaeology involves consideration of correlations between discourses like medicine and psychiatry and ‘an institutional field, a set of events, practices, political decisions, a sequence of economic processes that also involve demographic fluctuations, techniques of public assistance, manpower needs, different levels of unemployment, etc.’ (Foucault 1972b: 157). Moreover, as Lecourt argues, ‘“Discursive” connexions are said to be secondary with respect to certain connexions said to be “primary” which “independently of all discourse or all objects of discourse, may be described between institutions, techniques, social forms, etc.” [Foucault 1972b: 45]’ (Lecourt 1970/1975: 199), and the connections between the social/extra-discursive domain ‘are not present in the object’ (Foucault 1972b: 45). Thus, Foucault only juxtaposes the discursive and extra-discursive, indicating that there is an ‘interlocking’ between these two kinds of practices and does not specify exactly what kind of relationship obtains between them (Lecourt 1970/1975: 204). So just as Kant rendered the world initself unknowable, so also is the extra-discursive: Foucault asserts the existence of the socio-historical world beyond the order of discourse but has no protocols for justifying reference to it or means of specifying how they affect each other. Just as Kant was an agnostic realist, Foucault is an agnostic sociologist. Indeed, Foucault actively sought to distance his work from the social sciences especially as concerns the object ‘society’ (e.g., 1980/2003: 257; 1984/2003a: 58). But there are no sound arguments for why Foucault could not offer realist accounts of extra-discursive social determinants. This is for two reasons. First, his account of knowledge was materialist and implicitly

From Foucault’s Genealogy to Aleatory Materialism 285

realist (Pearce & Woodiwiss 2001). Second, the ‘stuff’ that makes up the extra-discursive social world is exactly the same kind of stuff (practices, institutions, etc.) employed in his materialist and realist account of discursive formations allowing him to explain the production of the archive, savoirs, objectifications of humans, etc. (cf. Lecourt 1970/1975: 207–208). Both permit of knowledge of the extra-discursive. The only ‘reason’ Foucault has for limiting his materialist methodological protocols to the domain of knowledge is his Kantian constraint. Genealogy attempts a solution. Beyond Knowledge: Power, Politics, and Social Formations Many believe that genealogy, by introducing the concept of power, allowed Foucault to incorporate the extra-discursive into his work and so integrate a concern with politics. Power, defined as ‘relations of force’ constituting the capacities of bodies imbricated in a singular position, is the ‘content’ that is formed, shaped, directed, and organized by discursive practices: power, the ‘unformed,’ is outside of knowledge (Deleuze 1986/1988).4 This definition of power is why Foucault examines the daily, small, mundane exercises of domination that abound in schools, prisons, factories, hospitals, barracks, etc. instead of focusing on direct violence or coercion: these exercises of power are what are responsible for constituting actual happenings. On this basis, Foucault defines politics as ‘a more-or-less global strategy for co-ordinating and directing those relations’ (1977/1980c: 189). However, the concept of power and its corresponding analytic brings to light the limitations of CN. A discrepancy emerges when Foucault poses the question of politics, because it requires that one account for how power is organized, coordinated, and directed. Thus, for Foucault, ‘what makes the domination of a group, a caste, or a class, together with the resistance and revolts that domination comes up against, a central phenomenon in the history of societies, is that they manifest in a massive and global form, at the level of the whole social body’ (1982/2003: 143–4; cf. Hunt 1993: 276–279). The fault line that appears concerns the concept of ‘the whole social body’ producing the place of dominance through which such an organization can occur. This discrepancy pushes Foucault beyond his nominalist conception of the object of human existence toward a materialist sociology because it is no longer simply a question of objectifications of subjects but of the constitution of positions on which such practices rest in a broader

286 Ronjon Paul Datta

social formation. Moreover, Foucault ‘uses’ concepts from Marxism like ‘overdetermination’ (1977/1980b: 195), ‘the class that finds itself in a ruling position’ (203), ‘unequal and relatively stable relations of forces’ (201), ‘inequality’ (1982/2003: 135) and ‘social hegemonies’ (1976/1994: 93), introducing thought-objects incompatible with his nominalist rejection of ‘society’ as an independently existing, objective ‘thing’ (cf. Balibar 1989/1992: 51). What is remarkable here is that all of a sudden, discursive practices are no longer the sole constituent of modes of existence: the more determinant element concerns the constitution of inequalities in a society. Thus, I suggest that what was previously held to form human existence (i.e., discursive practices) becomes the content in a place in a structure of places forming this power-knowledge content in a determinate way. In short, the materialist (and structuralist) position of the difference between the position/ place and what holds/actualizes the place now appears: power and knowledge become decentred in a social formation. Foucault has unwittingly changed terrain. But he cannot think this ‘new object’ without contradicting CN. Next, I draw out the consequences of this tension between nominalism and realism in genealogy. The genealogical concepts of dispositifs (‘social apparatus’) and strategy attempt to grapple with the relation between thought and life constitutive of the historically contingent assembling of existence. So while archaeology concerns the a priori of experience, genealogy concerns the a priori of existence, the structuring of thought and life and the way they come to be attached. The question I pose to Foucault’s work here is, What must social existence be like such that strategies come to be massively effective? Hegemonic strategizing on Foucault’s own terms thus needs to be accounted for, since it is apparently the condition constitutive of modes of existence. If we wish to transform modes of existence, then a new target of transformation has emerged. The decisive political factor now appears as the structure of a society such that positions of asymmetrical potentials are produced, potentials that constitute subjects with the capacity to attempt the implimentation of their discursively formed dreams and schemes with greater ‘effectiveness’ than others in intransitive conditions. By ‘asymmetrical potentials,’ I mean the material social conditions through which the capacities of subjects are constituted, producing a greater potential for actual exercises of this potential relative to others. As such, asymmetrical potentials refer to the condition of politics as the capacity to ‘structure the possible field of action of other people,’ to borrow Foucault’s own theorem (1982/2003: 138).

From Foucault’s Genealogy to Aleatory Materialism 287

Foucault defines power as ‘the multiplicity of force relations’ referring to the actualized capacities of living human bodies imbricated in connections with other bodies. Power is about what bodies do, what bodies can do, and how actions are constituted. Neither power nor the body are ‘things’ and hence cannot be possessed (1976/1994: 92; cf. 1977/1980c: 189). For Foucault, this means that we must see power as productive, as exercised, and as something that passes through bodies (Deleuze 1986/1988: 89). Power as relations of force, or ‘action upon an action,’ consists of relations of dominance and subordinance: some relation comes to dominate another; it does not destroy it or capture other forces but affects them (Foucault 1982/2003: 137–138). The emphasis on relations shows why Foucault is a nominalist about power and bodies: there are no ‘things’ involved, only an extant, contingent, singular encounter of force relations (1976/1994: 93). ‘Force’ however, is nothing but the dynamism of life itself, as indicated by his remark that the ultimate form of resistance to being ‘dominated’ or ‘governed’ by another is to kill oneself or kill the other (1984/2003b: 34). Foucault thus transposes Kant’s dualism of will and idea into one of materialism (of thought) and vitalism (of life) (Deleuze 1990/1995: 91; Deleuze 1986/1988: 91–93; Foucault 1976/1994: 144; Balibar 1989/ 1992: 55). Consequently, power relations are not ‘knowns’ but are the ‘other thing,’ the ‘outside’ of knowledge, and cannot be known ‘for the power relation has no form itself’ (Foucault 1972b: 86; Deleuze 1986/ 1988: 84, 86, 74). As such, power and bodies are ‘infranoumenal’ (Wahl in Balibar 1989/1992: 56; Deleuze 1986/1988: 83), and genealogy is based on a vitalist metaphysics. The concept of ‘power-knowledge’ attempts to show how they are mutually implicated (Foucault 1975/1979: 27). Power is what makes ‘things/objectifications’ happen, compelling subjects ‘to see and to speak’ (Deleuze 1986/1988: 82). Two points need to be noted. First, knowledge (or thought) is what organizes and directs power relations. Second, the unknowability of power relative to knowledge means that we ‘know’ power relations by the way they are registered in knowledge and thought as ‘problems.’ On the first point, thought is like ‘software,’ an ‘abstract machine’ for organizing power relations that give actual form to power (Deleuze 1986/1988: 73, 77). For example, the panoptic diagram is ‘the fixed form of a set of relations between forces’ (89). Panopticism is a strategic thought of how to use and deploy the effects of discipline and normalisation on ‘inmates,’ be they schoolgirls or criminals. Thus, there is no organization of power without knowledge and

288 Ronjon Paul Datta

the thought of how to use the effects of power (Foucault 1976/1994: 101). Crucially, power-knowledge is not a parallelism. Rather, differential/unequal force relations are expressed in the rarity of statements, and when these statements (and their conditions qua discursive formation) concern human life, they directly form and shape force relations, i.e., objectify them. On the second point, knowledge of power is not dependent on ‘epistemology.’ Rather, the realm of power as outside and in excess of knowledge is directly recorded in knowledge as a ‘problem’ requiring a solution. Excellent examples of this process are the records of lettres de cachet analysed in Foucault’s ‘Lives of Infamous Men’ (1977/2003b). He shows how we have come to have knowledge of mundane, daily social life in the Classical Age and all the small mechanisms of power within it (e.g., between family member, neighbours, etc.). Our knowledge comes through their recording in lettres de cachet, a legal instrument through which almost anyone could call on sovereign power to seek the incarceration of someone deemed an annoyance, a ‘problem.’ Knowledge, then, is like the Nietzschean ‘interpretation’: it is a sign of a will to power, the ‘dust kicked up by such a contest’ (Deleuze 1986/ 1988: 29). However, power relations are not social relations but are rather their substratum (Foucault 1976/1994: 94; 1984/2003b: 34). Thus power relations are not about ‘the social’ but concern the decentred multiplicities of life. But this has the effect of redoubling the problem of the ‘extradiscursive’ elements that Foucault discusses: the discursive-extradiscursive juxtaposition of his archaeological works is transposed into the genealogical power-knowledge-‘extra-discursive’ juxtaposition. Thus, the question about what relationship exists between a hegemonized ‘society’ and power-knowledge is not yet resolved. The concept of dispositif attempts a resolution (Foucault 1976/1994: 93, 94, 127; Balibar 1989/1992: 51). A dispositif is the immanent, contingent structure combining the multiplicities of thought and life in a stabilized way, i.e., with how thought targets bodily capacities in their actuality and thinks about how they can be used, shaped, and organized (Foucault 1975/1977: 218; cf. Brenner 1994). A dispositif constitutes ‘a heterogeneous ensemble consisting of discourses, institutions, architectural forms, regulatory decisions, laws, administrative measures, scientific statements, philosophical, moral and philanthropic propositions – in short the said as much as the unsaid. Such are the elements of the apparatus. The

From Foucault’s Genealogy to Aleatory Materialism 289

apparatus itself is the system of relations that can be established between these elements’ (Foucault 1977/1980b: 194, my emphasis; cf. Deleuze 1989/ 1992). Thus, a dispositif is a contingent assemblage of various means of production of existence and experience: they make a ‘present.’ That is to say that a dispositif is a new figure of the a priori present in and as its effects, structuring modes of being. Foucault also argues that dispositifs are stabilizations of strategies. Strategies consist of the thought about the way in which power relations, discourse, their concrete conditions of existence, their uses and effects can be taken up to achieve a goal or objective retroactively created or used by a dominant class (1976/1994: 98–102; 1977/1980b: 195; 1977/1980a; cf. Jessop 1990: 220–247; Hunt 1993). Strategies involve attempts to align and coordinate the effects of power relations, connecting them together, borrowing diagrams of power, knowledge, institutional models, and the like in order to achieve a goal. Thus, the effects of micro-dominations of any particular power relation become used as tactical supports for an overall strategy (1977/1980b: 199; 1977/1980c: 142). Discourses are the decisive, constitutive element of a strategy (1977/1980b: 196; 1976/1994: 97–101). For example, the bourgeoisie uses the models and savoirs of discipline (borrowed from the military and the monastery), medical discursive practices, normalization, policing, and surveillance as tactical elements in a dominant strategy targeting the working class by actually constituting the productive capacities of individual and collective bodies with the aim of creating a productive, employed, domesticated, regularized, healthy body. As he states, ‘A dominant class isn’t a mere abstraction, but neither is it a pregiven entity. For a class to become a dominant class, for it to ensure its domination and for that domination to reproduce itself is certainly the effect of a number of actual pre-meditated tactics operating within the grand strategies that ensure this domination’ (1977/1980b: 203). It is the combining of numerous instances of power-knowledge that gives strategy its political character. As such, a strategy requires the capacity to think ‘a difference of potentials’ (ibid.: 200–201). Foucault, however, must now account for the way a collection of dominant subjects comes to be in a place of strategic dominance. The introduction of words like ‘potential’ and ‘possible’ are symptomatic of a slippage in the CN terrain: we are no longer in the terrain of the actuality of the present or actual exercises of power – there is something else. The slippage appears when he introduces the ‘minimal difference’ (cf. Zizek 2004: 92) between what holds the place (power relations) and

290 Ronjon Paul Datta

condition constitutive of the place/position. So something else is constitutive of the ‘asymmetrical’ potentials of positions within a ‘society’ or ‘objective capacities’ (Foucault 1984/2003b: 35; 1982/2003: 135). Foucault precisely at this point leaves CN. This ‘something else’ is that which constitutes the asymmetrical potentials of places/positions. Here it is useful to think about these places in the nonessentialist/ realist AM terms of a social formation as the contingent immanent structuring of asymmetrical, articulated positions. Thinking about ‘society’ in Foucault’s implicit AM terms also allows us to highlight a crucial aspect of ‘politics’ (in Foucault’s terms). Thus, the political point to stress about a dispositif is not that it is a heterogeneous, contingent ensemble that can be analytically broken down by CN histories into its various elements, each with its own specific history, hence denaturalizing them. Neither is it a question of how a heterogeneous ensemble produces regularized effects on modes of existence. Rather, the political point is that something (something dominant) was able to combine heterogeneous elements in such a way that the combination becomes stabilized and intransitive. The fundamental political issue is not that power-knowledge networks exist: a necessary condition of politics is the asymmetrical condition of existence necessary to making or establishing a network. Thus, the ‘politics’ happens in the networking or the combining of elements, not in the facticity of the actual network or the dispositif. This means going beyond critical analysis to accounting for how such disparate elements could be and were put together. The pertinent question now is, What must a society be like such that dreams and schemes of rule come to be massively effective? In other words, there is no dispositif without some dominant position (or bloc of positions) providing the potential for the activity of assembling. This question has consequences for Foucault’s displacement of politics by ethics. His first recommendation for changing a dispositif is strategic resistance (1976/1994: 96). This recommendation follows from his notion of the “political field” as composed of micro-powers taken up as tactical supports in an overall strategy (Foucault 1977/1980c: 189). So shattering supports undermines the conditions that strategies and dispositifs are reliant on. But this is not the whole story, since we are dealing with asymmetrical positions of potentials. Moreover, the coordination of resistance does not have the same characteristics of political thought, since Foucault does not say that resistance requires the thought of a possible future, or goals, in other words, a political ‘programme’ –

From Foucault’s Genealogy to Aleatory Materialism 291

indeed, he was explicitly averse to such (cf. Foucault 1983/2000: 386). Thus, resistance is limited because of only consisting of attempts to destablize and disrupt the hegemonic supports. Foucault recognized this and hence sought an alternative in ‘ethics’ (Deleuze 1986/1988: 94). However, Foucault sees ethics as a means of escape only because the subject chooses from available means of thought and techniques of working on the self and caring for others and chooses its own goal instead of being subjugated to the veridical-juridical linkage of the human sciences on which contemporary hegemonies depend. It is my view that Foucault’s failure to theorize counter-hegemonic politics according to his own definition of politics as ‘a more or less global strategy for co-ordinating and directing [power] relations,’ (1977/1980c: 189) means that he abandoned the activity of politics to the ruling classes. This failure leads to a weak conception of the transformation of existence whereby the multiplication of ethical subjects leads to the downfall of the dispositifs that hegemonize societies (Foucault 1984/2003b: 50). Here we can use Foucault’s microphysics against him, since the question is one of the balance of forces: on the one hand, the massive force relations that are the conditions of existence of dispositifs versus a collection of individuals engaged in making themselves and caring for others. If the latter is enough to surpass the quantities of force in a dispositif, Foucault’s conception of transformation is reduced to individual practices and is inconsistent with his conception of the structural potentialization of constituted subjects. We now come to the practical limitations of CN. First, ethics does nothing to change the place of the practising subject: it changes only what happens ‘in a place.’ Ethics is concerned with what is possible from the position occupied by an individual, and thus neither with what other position one might occupy nor with what it is that determines the structuring of positions that subjects can occupy. As concerns ethical subjects, there are no good reasons to assume that their effects would not become supports of a dominant governmentality or strategy, since governing is exactly concerned with using, coordinating, and organizing the actually existing activities of subjects. Indeed, contemporary capitalism thrives on this circumstance: self-improvement and home renovations do nothing to challenge the social place in which subjects do what they do in such projects of transformation. Ethics can only be the solution if one forgets the fundamentally political matter, namely, that it is the potential to network and engagement in doing so that is the social condition of hegemony (cf. Foucault 1977/1980b: 202–203).

292 Ronjon Paul Datta

Transforming existence and escaping the limitations of the present, in Foucault’s own (contradictory) terms, means that the will to truth of ‘Man’ can no longer be the sole target. Rather, the targets are the dominant elements that structure the asymmetrical arrangement of potentials (i.e., the social formation). To recognize that the placement of subjects determines their field of possible actions is to implicitly recognize that societies are real and material, constituting the places. What must be done, then, is to displace the dominant place and what makes that place, rather than to resist. Toward AM: Three Theorems The move from genealogy/CN to AM, reclaiming politics, society, and social science can now be distilled. First, put knowledge in its place. Discursive formations are not the sole constitutive element of modes of existence (subjectifications and objectifications of life). Placing knowledge means affirming the materialist argument that the ‘real-concrete object’ of a social formation exists independently of its constitution as an object of knowledge and is the target of transformation: changing discursive practices does not change the real-concrete object of the social formation actually constitutive of the place of power-knowledge. By introducing the terms ‘asymmetrical,’ ‘potentials,’ and ‘structuring the field of possible actions of others,’ Foucault refers to the reality of social formations beyond the discursive objectifications of human existence: he has unwittingly decentred what he had taken to be constitutive. Decentring knowledge and dispositifs in a social formation puts one back in the realist-materialist camp. Putting knowledge in its place justifies engaging in the science of social formations and hence the need for theoretical work that refers to the independently existing reality of the intransitive socio-historical world (i.e., the knowledge production process itself does not have the power to change the world). Thus, there is a social reality to be wrong about and hence fallibalist criteria are back in, correcting the ontico-discursivist fallacy. Second, we must think about ‘political power beyond power’ in contrast to the Foucauldian formula of “political power beyond the state” (Rose & Miller 1992). Here, AM helps us think about the social bases of power instead of focusing on what is ‘taking place’ between subjects (Foucault 1982/2003). The power-knowledge nexus is not solely constitutive of modes of being: power-knowledge is itself placed in a social formation under its conjuncture (cf. Montag 1998: 72). Thus,

From Foucault’s Genealogy to Aleatory Materialism 293

power-knowledges are the contents positioned by social formations in a structure of unequal potentials and their combination as a singular social formation. AM (per contra CN) refers to social relations, in particular, the specific modalities by which modes of social organization constitute actual existence. Social relations have primacy over power relations since relations determine the place of the exercising of power actualizing the asymmetrical potentialization present in-and-as-itseffects (cf. Balibar 1989/1992: 52). It is the structure of the social formation that allows us to account for the place of dominance, allowing us, then, to account for how strategizing is possible and effective and hence constitutive of modes of existence. Thus, actual modes of existence are not reducible to the practice of subjects. Concerning politics, what is decisive is the organization of the ‘masses’ (those not in their place), since this is the basis for a potential shift in the balance of social forces. For AM, this is the anti-place of politics, that which cannot be placed/hegemonized. Third, politics, in AM terms (i.e., the chance encounter between the two multiciplicities of virtue and fortune) is the mode of transformation (it is not enough to have the connecting together of resistances or to displace politics by ethics). Here the concept of dispositif can be rearticulated with AM: the virtuous collective political subject exists because constituted by a dispositif of politics (i.e., the aleatory encounter of the independent series of virtue, noted above). Perhaps Foucault should be read as the theorist of bourgeois virtue who has forgotten its fortune, and is not governmentality (the thought of how to use and shape what others are doing for some other ends) the name for bourgeois virtue? Conclusion To return to the original question, Foucault’s problematic is structured by his critical nominalism, and it leads him to displace politics by ethics as the answer to how to transform existence. Ethics is the normative conclusion logically drawn from his nominalism that perpetually only juxtaposes the relation between knowledge, power, and social formations. Foucault’s own concept of politics requires him to think in terms of real social formations, but his nominalism cannot account for them. His terrain can be changed by (i) putting knowledge in its place; (ii) thinking political power beyond power; and (iii) defining politics as the organizing of social relations of force necessary to displace the

294 Ronjon Paul Datta

dominance that structures a social formation. Arguably, the problem with Foucault’s nominalism is that it stops short, and it is rather Althusser’s AM that is the truly radical, materialist, nominalist position, the position that affirms the contingency of existence. Althusser argued that nominalism is the only materialism (1988/ 2006: 265), but this means being nominalist about the singularity of a social formation under its conjuncture (Montag 1998: 67; Althusser 1999: 16; Matheron 1998). Thus, social formations do not have ‘essences,’ and neither can there be one ‘true’ politics: politics and thinking about what can be done must continually be produced. This circumstance demands that materialists must always think the actual state of things that exist independently of the awareness or knowledge of them. The AM justification for theoretical work (i.e., thinking things as they actually are) lies in imagining what is possible through a rearrangement of social relations of force since actual force relations do not exhaust their potential. Theory is a political intervention because it thinks the difference between the actual and potential, allowing for a thought of another possible arrangement of potentials. Theory thus is an intervention for the possible, attempting to interpellate a political subject (cf. Althusser 1999: 21–25). Producing a thought of the (im-) possible creates a new ‘experience’ of the present, produced through the thought of a possible future (though without any guarantee of the destination). This thought about and for the future is in contrast to the CN justification of writing histories of the present, disturbing the experiences we take as self-evident. In the move from genealogy to aleatory materialism, much depends on how one ‘experiences’ and thinks (through some discourse or other) the present and the making of a future. It hinges on the difference between the ‘is’ and ‘the possible’ – the ‘is’ of existence and the ‘possible’ of politics: the thought of a future.

NOTES Research funding for this project was provided by the Social Sciences and Humanities Research Council of Canada Doctoral Fellowship. My thanks to Frank Pearce, Alan Hunt, Paul Giovanetti, Bela Egyed, Lorna Weir, Kelly Gorkoff, and Bruce Curtis for being generous interlocutors on these matters. 1 For discussion of critiques of Althusser, see Elliot (1987), and for a productive response to them, see Resch (1992).

From Foucault’s Genealogy to Aleatory Materialism 295 2 Bhaskar claims that Althusser rejected ontology, trapping him in the epistemic fallacy (1989: 208), but Bhaskar fails to read him from his problematic, i.e., Althusser’s concern with the structuring of concrete social formations. For a discussion of the relationship between Althusser and Bhaskar, see Resch (1992: 198, 200, 377). I do recognize that AM remains theoretically underdeveloped and that a more rigorous inspection of it, from a realist position, is likely a necessary project. 3 One could say then that Foucault’s work on discursive formations and the production of experience highlights the intransitive structure of the ‘transitive’ dimension of knowledge, since the conditions of modern knowledge of ‘humans’ have not changed since the mid-nineteenth century. This permits Pearce and Woodiwiss’s (2001) rendering a legitimate argument that The Archaeology of Knowledge is an implicitly realist text. 4 Foucault’s concept of power is deeply indebted to Deleuze’s works on Nietzsche and Spinoza (1962/1983, 1970/1980). The virtue of Deleuze’s reading of Foucault is that he understands and explicates the Kantianism subtending Foucault’s work. But there is a fundamental metatheoretical difference between them because Deleuze starts with ‘being’ instead of ‘knowing.’ Moreover, Deleuze’s transcendental empiricism is incompatible with Foucault’s nominalism. This distinction renders specious the eclectic combination of Foucault and Deleuze, as found in Rose (1999), for instance.

15 Gadamer’s Minimal Realism HOWIE CHODOS, BRUCE CURTIS, ALAN HUNT, AND JOHN MANWARING

Introduction A few words about the genesis of this chapter are in order. Unusually for theoretical papers of this sort, it is a collective effort. In fact, it is the product of a discussion within a social theory reading group that has been meeting regularly for eight years. The chapter represents an attempt at producing a work in common; and as is to be expected, we each probably stress some aspects of the argument over others. However, there are some core propositions that we do share. 1 Hans-Georg Gadamer is a realist, although in a minimal sense. 2 There is much to be said for Gadamer’s understanding of the centrality of language to humankind. 3 We stress his insistence that the ‘fundamental prejudice of the Enlightenment is the prejudice against prejudice itself.’ 4 His approach to truth provides a strong basis from which to negotiate a path between the Scylla of scientism and the Charybdis of relativism. 5 The lack of sociological depth of some of Gadamer’s key concepts, notably that of ‘tradition,’ needs to be corrected. 6 An acceptance of Gadamer’s basic approach to truth leads to the conclusion that the (dialectical) critical realist notion of a ‘truth of things, not propositions’ embodies a category mistake. Gadamer’s Realism In one of the Supplements to his Truth and Method (2003: 528), Gadamer observed that ‘one thing common to all contemporary criticism of

Gadamer’s Minimal Realism 297

historical objectivism or positivism’ is ‘the insight that the so-called subject of knowledge has the same mode of being as the object, so that object and subject belong to the same historical movement.’ We focus on the implications of this rejection of the subject-object divide for Gadamer’s realism. By realism, we refer simply to the proposition that there exists an order of entities and relations that is external to human will, knowledge, and intentions, and that these both constrain and enable all aspects of human life. Our point of departure is Gadamer’s insistence on the foundationally linguistic character of the being of humankind. While he wishes to suggest that there is a close relation between words and things, he also affirms that what exists in words does not, and cannot, exhaust what exists outside words. Gadamer’s account of the linguistic foundation of the hermeneutical phenomenon thus negotiates, in instructive ways, forms of nominalism in which only that which is named exists, and forms of realism in which an autonomous non-human world would cause humankind to know it in specific ways. Gadamer’s position is that language is a constitutive medium of human experience; the world is linguistic in nature, and language expresses the world. Language is not an instrument that humankind chooses to employ in order to master a non-human world, for there is no pre-conceptual universe. For Gadamer, there cannot be a world without language, and our engagement with the world is always already conceptual and linguistic. Nor is there a language without a world: ‘this world is verbal in nature,’ and ‘language has no independent life apart from the world that comes to language within it’ (2003: 443). ‘Language, ’ he writes, ‘is not just one of man’s possessions in the world; rather, on it depends the fact that man has a world at all. ’ (2003: 443). Moreover, ‘to have a world means to have an orientation (Verhalten) toward it.’ (2003: 443) Our being in the world is thus ‘primordially verbal,’ and it is for this reason that we have ‘a world’; all other creatures live in an environment. Language makes it possible to objectify our environment as an object of knowledge and understanding. As Gadamer puts it, ‘[T]hat a thing behaves in various ways permits one to recognize its independent otherness, which presupposes a real distance between the speaker and the thing’ (2003: 445). Parenthetically, we can note here a first point of intersection with critical realism and the way in which Roy Bhaskar (1994:16) uses the concept ‘referential detachment.’ He writes, ‘… it is easy to construct a prima facie case for ontology – through the notion of referential detachment. This

298 Chodos, Curtus, Hunt, and Manwaring

is the detachment of the act by which we refer to something from that to which it refers, or of reference from referent. Discourse must be about something other than itself, at the very least potentially, for us to be able to refer at all’ (emphasis in original). At the same time, for Gadamer, the variability of linguistic expression means that different linguistic communities will conceptualize things differently. Language itself is inescapably historical and necessarily variable. Yet Gadamer insists that language is not a barrier to our understanding of a world beyond our consciousness, but rather is the medium that makes possible all insights we have into the world. Languages, despite their variability, are all bound together by their verbal character, and Gadamer argues that the insights that each contain are potentially accessible to anyone. However, such a position makes the concept of ‘a world in itself’ problematic: variability implies that the view we have of the world is not determined by the world itself but exists in language and tradition (2003: 447). However, Gadamer is clearly not arguing that language determines the world outside language; he articulates an explicitly minimal realist stance (2003: 447): ‘No one doubts that the world can exist without man and perhaps will do so. This is part of the meaning in which every human, linguistically constituted view of the world lives. In every worldview the existence of the world-in-itself is intended. It is the whole to which linguistically schematized experience refers. The multiplicity of these worldviews does not involve any relativization of the “world.” Rather, what the world is is not different from the views in which it presents itself. The relationship is the same in the perception of things.’ Language is the ontological condition of human understanding, and only a god or a devil could move beyond the conceptual medium of language and grasp the ‘world-in-itself’ or ‘being-in-itself’ in a way not mediated by a horizon of language and tradition. At the same time, Gadamer insists, the variety of world views is not equivalent to the relativization of the world. World views are inseparably views of the world-in-itself, and these views change without ever exhausting the world-in-itself. Gadamer does speak of the ongoing interplay of horizons of understanding as a process of enlarging or enriching our understanding, of bringing more of the world into language, but, because of the finiteness of human understanding, this process can never end. However, he argues that the ontological unity of subject and object through language means that the world in itself is as it is in the

Gadamer’s Minimal Realism 299

diversity of ways that it are present for us. Gadamer insists that the world does not become the object of language (2003: 450). This position would be akin to narrow linguistic determination; humankind’s existence in the world is primordially linguistic, but the world is not language. ‘Rather, the object of knowledge and statements is always already enclosed within the world horizon of language’ (2003: 450). The unity of subject and object means that both are in language: the subject does not stand outside language, using it as a tool to make the world an object. On the whole, we are convinced by this stance, but it does contain one troubling element, namely, the seemingly incompatible ways in which Gadamer uses the concept ‘world.’ He introduces the fundamentally linguistic character of human beings by introducing a distinction between ‘world’ and ‘environment.’ ‘Environment’ refers to the condition in which entities not possessed of language find themselves; while they may be able to communicate to one another, they are unable to treat matters around them as ‘matters of fact.’ We imagine the capacity of one wolf to communicate to others in its pack the presence of prey at some location. Presumably such forms of communication also imply elements of memory. Human beings, by contrast, through language can make conditions in the world into objects of reflection, contemplation, and analysis. We detach experience from its immediacy and invest it in forms that give it new kinds of objectivity. It is this capacity that separates us from environment and makes environment for us a specifically human world. And yet, Gadamer repeatedly refers to world-in-itself, and in English translation, at least, the concept is used interchangeably with being-initself. As cited above, he takes it as a given that the world might exist without humankind. The claim seems to us to render somewhat ambiguous the distinction between environment and world that grounds the linguistic substance of a human world and to be inconsistent. Gadamer takes for granted that modern science can indeed dominate being-in-itself, that it can enlist ‘nature’ successfully. He wishes to argue that modern physical science is able to dominate being-in-itself because the particular world view it embodies, paradoxically, makes that being accessible to the domination of will by treating it as something that exists beyond will. The factual character of being-in-itself and its resistance to will are certified in the scientific world view. Gadamer does not argue that this world view is wrong; rather, he tells us that its conditions of possibility are conceptual and linguistic.

300 Chodos, Curtus, Hunt, and Manwaring

Knowledge for domination is one possible horizon of interpretation. It would, however, be mistaken to conclude that the capacity for domination is equivalent to ‘determining the real meaning of being-in-itself.’ This is the case because the physical sciences are not themselves what they study or what they perceive themselves to be. The particular forms of being-in-itself of any science are relative to its research project; ‘each science, as a science, has in advance projected a field of objects such that to know them is to govern them’ (2003: 451, 452). In this part of his argument, then, Gadamer holds consistently to his refusal of any separation of the subject and object of knowledge and to the inescapably linguistic character of the world. Language does not create a new world; rather, for humankind, language is the primordial ontological condition. Thus, he concludes that ‘language is a medium where I and world meet or, rather, manifest their original belonging together’ (2003: 474). Moreover, ‘the coming into language of meaning, points to a universal ontological structure, namely to the basic nature of everything toward which understanding can be directed. Being that can be understood is language’ (2003: 474, emphasis in original). One implication is that the question, Is Gadamer a realist? ceases to be a real question. Gadamer could only be a realist if we accept the separation of subject and object, such that the world outside humankind could determine humankind. From Gadamer’s position that humankind alone has a world, determination could only be one of the modes of being human. So while we continue to hold that Gadamer should retain his place in the realist camp, his inclusion raises important questions about what we mean by ‘realism.’ Where Is the Truth in Truth and Method? Parallel to the contestation of the subject/object divide, Truth and Method challenges the traditional dichotomy of truth and realism on one side and relativism on the other. Gadamer’s analysis suggests that neither side can ‘win’ this debate because it requires that we make impossible choices. We therefore need a better way of thinking about these problems, which is what Gadamer strives to give us. In Truth and Method, talk about truth and that which is true pervades the text, but Gadamer never really defines what he means by the concept of ‘truth.’ As Richard Bernstein notes: ‘It might seem curious (although I do not think that it is accidental) that in a work entitled Truth and Method, the

Gadamer’s Minimal Realism 301

topic of truth never becomes fully thematic and is discussed only briefly toward the very end of the book … Even more problematic and revealing, if we closely examine the way in which Gadamer appeals to “truth,” we see that he is employing a concept of truth that he never makes fully explicit’ (1983: 151–152). In the essay entitled ‘What Is Truth?’ which preceded Truth and Method, Gadamer discusses the concept of truth, but this earlier discussion is no more conclusive. He says there that ‘truth must be won as if it were a robbery [Raub] from the concealment and hiddenness of things.’ (1994: 36) and that ‘truth is unconcealedness’” (1994: 36). He concludes by stating that ‘what we mean by truth, the revealedness and unconcealedness of things, thus has its own temporality and historicity’ (1994: 46). The argument that truth has ‘temporality and historicity’ fits with Gadamer’s analysis of understanding as an event in which both the knower and the object of knowledge participate. Truth is a relationship. It does not reside in the thing, but rather in the language that the knower uses to express her knowledge about the thing. Truth outside language is no more conceivable than is ‘world’ outside language. The thing does not interrogate its own truth. Rocks do not worry about their history and composition. Texts do not question their own authenticity. It is here that we can locate a basic difference between Gadamer’s approach and that of dialectical critical realism, which supports a notion of alethic truth, defined as ‘the truth of things, not propositions.’ For the moment we will just flag this issue, postponing a fuller consideration until we examine Bhaskar’s approach in more detail. We hope to show that Gadamer provides us with grounds for thinking that Bhaskar’s capsule definition of alethic truth constitutes a category mistake. We suggest that Bhaskar’s notion of alethic truth collapses truth back into ‘things’ and thereby opens the door to a correspondence theory of truth. For Gadamer, however, the truths produced through the event of understanding change across time as the questions asked change. Truth is dependent on perspective; the location of the knower determines what aspect of the object she will see. The experience of truth is akin to getting lost in conversation (Dostal 1994: 63). The conversation may be with another human or with the tradition within which one is located. The concepts of ‘tradition’ and ‘prejudice’ are fundamental to Gadamer’s perspectivist definition of truth; no one can seek understanding without the foreknowledge or prejudgements that make us human.

302 Chodos, Curtus, Hunt, and Manwaring

Reinstating the concepts of ‘tradition’ and ‘prejudice,’ which have been marginalized, is central to Gadamer’s project. He puts this point both forcefully and with great beauty: ‘… there is one prejudice of the Enlightenment that defines its essence: the fundamental prejudice of the Enlightenment is the prejudice against prejudice itself, which denies tradition its power’ (2003: 270). Gadamer is intent on refusing the negative connotations that the Enlightenment has attached to ‘prejudice.’1 Rather, for him, prejudice helps us to organize our experience of the world and is therefore one of the conditions for our having knowledge of it. The fact that we cannot know without prejudice means that prejudice must be seen not only as a potential constraint on knowledge but also as one of its indispensable enabling conditions. We suggest that Gadamer provides us with a way to reclaim ‘prejudice’ as a component of all understanding, without simultaneously undermining the possibility of judgement that would follow from a wholesale descent into relativism. An ontology of human understanding must incorporate prejudice. Gadamer insists that the interpreter must make a conscious effort to be aware of the assumptions and prejudices brought to the interpretive task. The more fully we acknowledge our prejudices, the better we grasp the limits of our historically grounded understanding. Nor does he seek to invest tradition with a self-evident authority merely because it is old or familiar, but rather invokes its assistance in the quest for ‘objective knowledge of the historical world’ (2003: 275). So the question now becomes, ‘Does being situated within tradition really mean being subject to prejudices?’ (2003: 276). Gadamer’s response is that ‘the prejudices of the individual, far more than his judgements, constitute the historical reality of his being’ (2003: 276– 277). The way to secure an appreciation of tradition without being its slave is to regard tradition as ‘a genuine partner in dialogue’ (2003: 358). The positive incorporation of tradition is essential because we are ‘always situated within traditions’ (2003: 282), and there is no position outside tradition, any more than there can be a position outside language. Crucially, he contends that ‘to be situated within a tradition does not limit the freedom of knowledge but makes it possible’ (2003: 361). Moreover, ‘linguistic tradition is tradition in the proper sense of the word – i.e. something handed down’ (2003: 389). The positive sense of tradition he styles as ‘living tradition,’ a formulation that resonates with Michel Foucault’s concern with ‘the history of the present’ (1977: 31) in that the questions we pursue are motivated by

Gadamer’s Minimal Realism 303

the present (2003: 284). Reason and tradition are not antithetical, but one must never succumb to accepting tradition’s authority simply because it claims it. Gadamer is fully aware of the conservative dangers of such a stance. Nonetheless, knowledge for him is always partial and incomplete. Humans cannot achieve a god’s-eye point of view and cannot see the truth with such certainty that the conversation is brought to a conclusion that is no longer open to historical revision. We can only ever achieve the truth for us, here and now. Truth reveals and, in revealing, must conceal some aspect of that which is revealed. It is worth noting that this also applies to natural scientific truths as to any other. Prigogine and Stengers (1985: 225) capture this with reference to the basic atomic structure studied by quantum mechanics. No single theoretical language articulating the variables to which a well-defined value can be attributed can exhaust the physical content of a system. Various possible languages and points of view about the system may be complementary. They all deal with the same reality, but it is impossible to reduce them to one single description. The irreducible plurality of perspectives on the same reality expresses the impossibility of a divine point of view from which the whole of reality is visible. However, the lesson of the principle of complementarity is not a lesson in resignation; it follows from both the finite and imperfect language that we use to reveal truths and from the fact that humans cannot see all sides of the object simultaneously. Truth is also multiple because the truth that is revealed in the conversation depends on the questions asked. Questioning is fundamental to truth because it is through the interrogation of other human beings, traditions, texts, etc., that the knower is able to experience the enlightenment of truth. But the knower’s experience of truth is always in relation to the question asked, which means that different knowers will experience different truths by asking different questions. Thus, Gadamer uses a perspectival, historical, and pluralist definition of truth as that which is revealed in the event of understanding that occurs in the fusion of horizons arising in the dialogue of questioning. His theory is also fallibilist in that he recognizes the finitude of human understanding, since all knowledge is subject to revision as the questions asked and the understandings developed change over time. While Gadamer retains a minimal realism in Truth and Method, he clearly rejects the theory of truth often associated with strong realism: the correspondence theory of truth. He states this unequivocally in

304 Chodos, Curtus, Hunt, and Manwaring

discussing art: ‘The world that appears in the play of presentation does not stand like a copy next to the real world, but is that world in the heightened truth of its being’ (2003: 137). Truth is not a quality of the object that pre-exists the dialogue with the knower, but rather is an experience that occurs in the course of the interaction. Truth is not an object that a knower possesses but is a dynamic relationship, an event. Gadamer’s historical understanding of truth also undermines a strong coherence theory of truth, since coherence can only ever be temporary. What is coherent at one point in time will not be coherent at another, given that the texts being interrogated and the questions being asked will be different. Both the self and the tradition may not constitute a coherent whole. Coherence may be a precondition of rationality, but it cannot guarantee the truth of any understanding. Gadamer also rejects a pragmatist theory of truth as what works; he associates this theory of truth with the natural sciences and the desire of the scientist to dominate the world. He also rejects voluntarist or will theories of truth. We are all located within traditions, and it is only because we come into being in language that we can have a self in any important sense. Our ability to have a self and to understand this self requires experience and foreknowledge. One can make tradition and prejudice objects of knowledge, but one cannot stand outside tradition and prejudice, or outside history, and simply will truth into being. We can have a world only in language, and because we come into being in language, our self happens to us before we can engage in self-description. The self is a ‘linguistically mediated reality … both determined and awaiting further determination and definition … The self is neither an indeterminate “nothing” nor a fully determinate “essence”’ (Wachterhauser 1986: 225). The knower cannot completely transcend the linguistic world into which she is born, however much she modifies it through selfknowledge and understanding. It is precisely the fact that knowers are located historically and linguistically within a tradition that they have the freedom to reflect critically on who they are and who they want to become. Truth is historically effected in two ways. First, all truth is a product of history and is given to us by history. Second, we work out our truths in time and over time. ‘Experience teaches us to acknowledge the real. The genuine result of experience, then – as of all desire to know – is to know what is. But “what is,” here, is not this or that thing, but “what cannot be destroyed”’ (Ranke quoted in Gadamer 2003: 357).

Gadamer’s Minimal Realism 305

The Question of Relativism From these considerations it becomes clear that Gadamer’s understanding of truth has little in common with either a theory of absolute truth or a purely relativist theory of truth. He sketches his alternative by drawing on the experience of the truth of art. While art can be a field of enquiry, the experience of art – a painting or a play – shares something of the quality of religious revelation. To what extent is Gadamerian truth experienced as revelation? His definition of truth as ‘revealedness and unconcealedness’ suggests the link (2003: 150–151). His argument that understanding is an event (2003: 485) suggests that truth has a quasi-mystical character. If this is a valid interpretation of Gadamer’s theory of truth, he appears to be teetering on the brink of a profoundly irrationalist and relativist theory of truth. For this reason his work has lent itself to anti-foundationalist interpretations. Richard Rorty, who is often cited as the paradigmatic relativist, offers one such interpretation.2 As Georgia Warnke notes, Rorty understands hermeneutics as conceiving of the hope of overcoming disagreements through conversation ‘as functioning without preexisting grounds … In Rorty’s analysis, hermeneutics replaces the goal of grounding cognition with that of Bildung, or edification’ (Warnke 2003: 105). Hermeneutics, on this reading, frees us from the sway of epistemology and allows us to get on with the task of developing our understanding without worrying about whether the understandings are objectively true. While this is a possible understanding of the lessons we can learn from Truth and Method, Gadamer’s own language throughout suggests that he is not willing to jettison the concepts of objectivity and universality. The question we need to ask is therefore whether Truth and Method allows for a theory of truth that avoids relativism and subjectivism while still using a perspectivist and pluralist theory of truth. Gadamer expresses impatience with the charge of relativism that was levelled against Heidegger: ‘However clearly one demonstrates the inner contradictions of all relativist views, it is as Heidegger has said: all these victorious arguments have something of the attempt to bowl one over. However cogent they may seem, they still miss the main point. In making use of them one is proved right, and yet they do not express any superior insight of value. That the thesis of skepticism or relativism refutes itself to the extent that it claims to be true is an irrefutable argument’ (2003: 344). However, Gadamer is not arguing that we should embrace

306 Chodos, Curtus, Hunt, and Manwaring

relativism, but instead that the refutation of relativism does not prove that there is a single objective, universal ‘Truth.’ Gadamer’s project is to free the human sciences from the tyranny of the scientific method and its search for ‘universal truth.’ He wants to show that in spite of the fact that the human sciences, and in particular history and philosophy, do not use the same method as the natural sciences, they still produce truth. He states in the introduction to Truth and Method that his investigations ‘are concerned to seek the experience of truth that transcends the domain of scientific method wherever that experience is to be found, and to inquire into its legitimacy. Hence the human sciences are connected to modes of experience that lie outside science: with the experiences of philosophy, of art, and of history itself. These are modes of experience in which a truth is communicated that cannot be verified by the methodological means proper to science.’ (2003: xxii). His impatience with the charge of relativism reflects his desire to transcend the ‘ … restricted horizon of scientific theory and its methodology’ (2003: 552). However, his rejection of the natural science model of knowledge does not entail the rejection of the possibility of knowledge. As we have insisted from the outset, Gadamer remains a realist (although a minimal one) and shares the view that the world exists independently of language and thought. As noted above, ‘[n]o one doubts that the world can exist without man and perhaps will do so’ (2003: 447). Moreover, he does not push his argument to the point of making a ‘social construction of everything’ argument (Hacking 1999; Longino 2002). It should be noted that in the afterword to Truth and Method he does suggest that the natural sciences cannot escape hermeneutics because scientific methods cannot determine the questions that scientists will ask. He argues there that ‘the viewpoints that select the relevant topics of inquiry and foreground them as subjects of research cannot themselves be derived from the logic of investigation’ (2003: 555). Scientific knowledge and the truths it produces are not the only possible forms of knowledge and truth; it also requires that the validity of other forms of truth and knowledge be recognized. A fusion of horizons that occurs as one engages with a tradition does not produce an absolute truth. Rather, it produces a truth for that experience or event of understanding – a truth like that produced by great works of art that tell us something profound about the world in the moment we experience them, even though we cannot always say what that truth is.

Gadamer’s Minimal Realism 307

However, if tradition lends itself to many understandings, all of which contain truth, is not truth relative to the question asked, the language used, or the world view brought to bear in formulating a question and bringing horizons into contact? Are we not back at the fork in the road where we are forced to choose between relativism and correspondence theory? The point to be emphasized is that tradition and prejudice which form our world, and the language through which we know the world, cannot prevent disagreement about the meaning and significance of that tradition. For Gadamer, ‘the fact that our experience of the world is bound to language does not imply an exclusiveness of perspectives. If, by entering foreign language-worlds, we overcome the prejudices and limitations of our previous experience of the world, this does not mean that we leave and negate our own world. Like travellers we return home with new experiences’ (2003: 448). Perspectivism does not mean that we retreat into a world of multiple and incompatible truths; rather, the engagement in conversation allows a fusion of horizons that yields a net increase of being. Gadamer thus avoids relativism by arguing that understanding involves growth and progress such that knowledge is developed over time. The reason that this idea eludes the relativist trap is that it allows for the possibility that we can identify criteria that will allow us to distinguish among competing truth claims. Since all knowledge evolves across time and space, any such criteria must themselves be understood to be subject to historical, cultural, and temporal evolution. In this sense, the simple acknowledgement of the possibility that some accounts may be better, even truer, than others marks a dividing line from a fully relativist stance. Still, while philosophy shows us that it may be possible to make judgements with regard to competing truth claims, it cannot define a priori eternal, unchanging criteria for making those judgements. It is the activity of truth-seeking that makes truth possible. It is thus the commitment to the idea that some answers, some projects, will be better than others (according to criteria themselves subject to refinement) that prevents a descent into full-blown relativism. The best versions of truth will emerge in an unending contest to discover them, a contest that involves both confronting different theories with one another and testing them ‘against the world.’ Of course, success never guaranteed, we can never expect ‘the world’ to give us unambiguous responses. All our knowledge is linguistically based and marked by ‘prejudice,’ but this does not mean that it is only language, that it is not in some sense also knowledge of the world. Under

308 Chodos, Curtus, Hunt, and Manwaring

some circumstances prejudice can mislead us, but as we have seen, prejudice is also necessary to our capacity to gain superior insights about the world around us. Language and prejudice not only constrain, they also enable. At the same time, however, the idea that one account can be better than another does not return us to a correspondence theory of truth as long as there is no pre-given definition of what constitutes a ‘better’ explanation. We can never know in advance everything that could constitute truth. But the insistence that there are reasonable criteria (albeit historically, culturally, and socially relative ones) that allow a rational choice to be made from among competing accounts, it is nonetheless sufficient to prevent a slide into a fully fledged relativism. Questions, Not Methods A particularly attractive feature of Gadamer’s hermeneutics is the priority he accords to questions. He starts from the contention that ‘we cannot have experiences without asking questions,’ such that ‘the path of all knowledge leads through the question’ (2003: 362, 363). It is the posing of a question that opens up enquiry. At the same time, according priority to questions reveals the limited role of method in the pursuit of knowledge. ‘There is no such thing as a method of learning to ask questions’ (2003: 365). The inherent weakness of the preoccupation with method in the social sciences is that it focuses attention on the procedures of enquiry rather than on the direction of enquiry. Such an approach can easily tempt one to fix a given ‘method’ as the only way to guarantee access to ‘truth.’ Gadamer draws heavily on R.G. Collingwood for the development of the logic of question and answer (Collingwood 1939: 29–43). A text can be understood only when our enquiry starts out to discover the question that it answers, or, perhaps better, the question that it sought to answer. Since Gadamer’s hermeneutics is not restricted to textual interpretation, it is only possible to understand historical events if we reconstruct the questions to which the actions of participants were the answer. Of course, events in history do not manifest any necessary agreement with the subjective ideas of actors; similarly, the text may reach beyond what its author originally intended. Gadamer does not doubt the accessibility of the intentions of actors but chooses to address the issue of ‘unintended consequences’ in that he insists that we must not assume

Gadamer’s Minimal Realism 309

that outcomes are those intended by actors. Gadamer seems to approach something close to a critical scepticism regarding intentions by refusing any hermeneutical reduction to the author’s own meaning. ‘The most important thing is the question that the text puts to us’ (2003: 373). For Gadamer, ‘understanding is always more than merely re-creating someone else’s meaning. Questioning opens up possibilities of meaning’ (2003: 375). The aim should be to understand how particular questions come to be asked in specific historical circumstances. It is worth noting that this line of argument suggests a certain similarity to Foucault’s concept of problematization, which proposes a research strategy that asks how social phenomena come to be viewed as problems, and why it is that the specific phenomenon, and not some other, comes to be constituted as a problem. Foucault described his approach to history: ‘I will say that it’s the history of problematizations, that is, the history of the way in which things become a problem. How, why and in what exact way, does madness become a problem in the modern world, and why has it become an important one’ (1996: 414, emphasis in original). Tradition Revisited Our primary criticism of Gadamer’s conception of tradition – which is at the centre of his project – is that it is sociologically weak. He tends to speak as if we are born into a single tradition, world view, and language, and that each language can express only a single world view. We hasten to add that he nowhere explicitly states this, and there is nothing in Truth and Method that indicates that Gadamer would necessarily be opposed to recognizing that traditions and languages are more complex, contested, and conflicted. Unfortunately, though, by not considering the possibility that traditions are anything other than homogeneous national entities, Gadamer ignores many important questions about internal variation within linguistic traditions – a phenomenon well established with respect to class variation (Bernstein 1971; Willis 1977), gender (Gilligan 1983) and other intersecting social differentiations. Moreover, these varied forms of lived worlds do not stand apart in isolation from one another and can, in varying degrees, understand and communicate with other subject positions. Coming to grips with the plurality of situated understandings is a precondition of an adequate approach to the hermeneutic problem that Gadamer sets out to address.

310 Chodos, Curtus, Hunt, and Manwaring

In important ways, then, because it remains insufficiently developed and lacking in sociological depth, Gadamer’s account of tradition is not able to sustain the burden that he imposes on it. He is not without sociological imagination, as is attested by his acute observation that the speed-up of postal delivery did nothing to improve the quality of letter writing but had the reverse effect (2003: 369). However, he has nothing to say about the conditions that might dictate whether a tradition becomes frozen or, alternatively, manifests a capacity for critical selftransformation. He is blind to processes such as the invention of tradition that make tradition both a construct and a legacy (Hobsbawm & Ranger 1983). There is no need to follow the logic of identity politics by positing an ever-expanding plurality of positions. What is needed is something closer to Pierre Bourdieu’s treatment of the interconnection of doxa and habitus that is able to identify both elements of commonality and features of differentiation and cleavage within communities (Bourdieu 1992). Within a single nation that speaks one language and shares a relatively homogeneous culture, there can be multiple traditions – traditions of union militancy, political activism, feminist struggles, and corresponding intellectual traditions such that within any single one there are conflicting understandings that compete for hegemony. But Gadamer never deals with complexity and conflict internal to a tradition. Nor does he deal with complexity and conflict among traditions. When we are learning another language or studying another tradition, we may come into conflict with world views that are mutually incompatible. Militant fundamentalisms, whether political or religious, often view the Other as something to be either converted or destroyed. German fascism viewed the Jews as a ‘disease’ that had to be eradicated. Fundamentalist Islam and fundamentalist Christianity confront each other in a ‘clash of civilizations.’ It seems obvious that such fundamentalisms do not seek a fusion of horizons. Gadamer thus avoids some of the relativist implications of his hermeneutics by repressing the potential for conflict and for multiple understandings of the world. At the forefront of his mind would seem to be the self-evident sense in which we ‘share language,’ so that the idea of a uniform language tradition does not make him uneasy. On this basis he proceeds to the further assumption that there is some direct connection among language, ethnicity, and nation. It is evident that the case of Germany informs much of his argumentation.

Gadamer’s Minimal Realism 311

Thus, Germans speak German, they share German ethnicity, and they communicate within a German tradition. Being an intellectual who lived through the Second World War, Gadamer more than others might have been expected to have attended to the vagaries inherent in the concept of linguistic tradition. As well, it is perhaps a little surprising that he does not exhibit some hesitation, as Germany was one of the later European nations to bring together nation with language and culture, and, in addition, many other ethnicities shared the German language. This lateness is not unconnected with the fact that unity was subsequently locked into a particularly severe legal form that continued, down to the very recent past, to deny citizenship to German-speaking residents not of German parents (blood), and paradoxically conferred the right of return (mainly from Eastern Europe) on non-German speakers of ‘German parents.’ Gadamer does offer dynamic potential with his concept of the ‘fusion of horizons,’ which recognizes that out of the interaction of different perspectives or horizons, understood as the situated range of vision or vantage point (2003: 302), comes the possibility of a new perspective and thus new understanding that can be generated when one person is striving to understand another. These elements that provide for the possibility of change are valuable. But they require a more robust concept of tradition. Gadamer shares with Jürgen Habermas a dependence on a presumed uniformity in the conditions of communication and understanding; in Habermas’s case, an ‘ideal speech situation’ in which the subject emerges as a rational, autonomous agent, but one who, like Gadamer, is historically located within social relations. Their difference is that Habermas insists on the need to take account of relations of domination (1977). Gadamer, Critical Realism, and Social Change As we have noted at several junctures, there are points of both overlap and divergence between Gadamer’s approach and critical realism, in its various incarnations. There are, in fact, a greater number of points of convergence, but the issue on which the two approaches part company is a significant one. Gadamer’s reasoning parallels three key elements in the philosophical armoury of critical realism – ontological realism, epistemological relativism, and judgemental rationalism. First, as we show, Gadamer deploys a minimal realism that relies on at least a significant element of the notion of ontological realism that

312 Chodos, Curtus, Hunt, and Manwaring

Bhaskar defends. As well, there can be no doubt that Gadamer argues for something akin to epistemological relativism by making the historical and socio-cultural variability of what counts as knowledge a centrepiece of his approach. Finally, the injunction that emanates from Truth and Method to compare and evaluate competing explanations can be seen as an argument in favour of what Bhaskar calls judgemental rationalism, or the ability to choose rationally between options. However, as we noted earlier, agreement ceases when it comes to the notion of ontological truth. We argued that Gadamer would have opposed such a concept on the grounds that it would not make sense to him to speak about the ‘truth of things, not propositions.’ And we hinted that we sided with Gadamer on this question. The idea of a ‘truth of things’ embodies, as we argued earlier, a category mistake. There may be truths about ontologies, but these are always truths in a human sense, that is, truth constrained (and enabled) by prejudice. Things may exist outside human awareness, but truths about these things do not. Truth as a category belongs to us as human beings, even when we describe truths about phenomena that do not depend on us for their existence. There is no such thing as ontological truth, although this does not invalidate ontology as the category of that which exists independently of our thinking human thoughts about it. To conflate truth and ontology must ultimately lead back toward correspondence theories of truth. The only possible requirement for a concept of the ‘truth of things’ would be to avoid the spectre of relativism. But, as we argued earlier, it is not necessary to refer to a reductionist ‘thing-in-itself’ in order to avoid sliding into the relativist trap. All that is needed is the recognition that over time, the criteria we use to distinguish truths are themselves subject to refinement. We can then argue for historically, and socioculturally, grounded reasons for preferring one option to another. We do not need to reinstate a notion of trans-historical truth, or of ‘objective’ certainty, but we must recognize that the criteria for ranking knowledge cannot be entirely arbitrary. Rather, they help identify that which constitutes a ‘truer’ understanding of the world around us (both social and natural), while at the same time embracing the ‘truth’ that all knowledge is nonetheless thoroughly infused with ‘prejudice.’ Nor do we need to seek a single way in which truths about social life can be argued for, or a single ‘scientific’ method by which we conduct our investigations. It is no more possible to establish a definitive scientific method that applies outside time and history than it is to establish a definitive truth about things.

Gadamer’s Minimal Realism 313

In many ways, then, contrary to Bhaskar’s claim that alethic truth helps solve the problems of philosophical aporiae, it merely leads us straight back into them. If alethic truth, that is, truth ‘true to, for and of itself’ (1998a: 568), is to be anything more than a repository for quasiPlatonic forms, it must be possible for it to have a direct causal influence on human action. Bhaskar recognizes the importance of this issue for his philosophical enterprise, asserting that it is the ‘grounding of the ethical in the ontological that distinguishes Dialectical Critical Realism from other contemporary philosophies’ (1998a: 568). But what could it mean to ‘ground the ethical in the ontological’? Does it mean that alethic truth must in some sense cause us to do the things that we do, cause us to have the values that we have, independently of how we think about that truth or how others interpret it for us? It is hard to avoid this interpretation. Otherwise, truth ceases to be ontological and is either mediated by (historically, culturally, and linguistically specific) human intervention, or it exists in a world unto itself that cannot intersect with actual human practices. But what could causality in human affairs dissociated from human intervention look like? Are we headed back to some form of determinism? In our view, Bhaskar never successfully builds a bridge between alethic truth and human action, because such a bridge cannot be built. In fact, we would argue that Bhaskar is asking the wrong question. The very project of seeking a ‘ground’ for ethics in ontology either places at least some determinants of ethical values outside history, language and culture, or it is simply another way of stating that ethics is rooted in temporally and spatially defined communities? The first possibility is one that, following Gadamer, we would reject. The second cannot be a secure ‘ontological ground’ since that ground shifts with the nature of the communities in question. So while Bhaskar is explicit in his desire to provide a philosophical foundation for critical and left-wing politics, it is somewhat ironically a Gadamerian stance that, in our view, provides a better basis for such a project. We believe that our reading supports what Linda Alcoff has called a left-wing Gadamerian approach. She tells us that she has embraced ‘an avowedly “left-wing Gadamerianism.” No less than Hegel, Gadamer’s work spawns multiple political interpretations. There is a tension in Gadamer’s work between his invocation of the permanent truth of tradition, on the one hand and his insistence on the impermanence, change, context-dependence, and historical mobility of meaning on the other’ (2004: 234). For us, Gadamer’s

314 Chodos, Curtus, Hunt, and Manwaring

reaffirmation of prejudice as a condition for acquiring true knowledge and not an impediment to doing so points to the possibility of sustaining the feasibility of significant social change without succumbing to reintroducing a false scientistic certainty. In other words, Gadamer helps us to travel a path that eschews the Scylla of scientism and the Charybdis of relativism. He does not take us all the way there, however, because he does not explore key sociological and political implications of his argument. One could argue, in fact, that the political terrain is necessarily productive of ‘fusions of horizons.’ The political process is a complex one, and just as science itself is fallible and prone to being influenced by the biases and interests of scientists themselves, so too politics is a highstakes game of power where truth all too often takes a seat at the very back of the bus. But despite this, it is the only terrain on which it is conceivable to blend theory and practice, that is, to hear the testimony of those who have a direct experience of the matters being adjudicated, and to be able to pitch our theories and hypotheses about the significance of that testimony against the social world and subject them to the scrutiny and criticism of others. We do not need to know everything about everything, since such a goal is not realizable. Because there can be no universal theory of truth, we must engage in an unending contest over both what is to count as truth and how we should go about seeking it. What needs to be reconciled is therefore the impossibility of certainty and the necessity of action. In terms of social action, the resolution of this tension lies in fostering the knowing acceptance of the risks involved in a project of social change by those who undertake it. People must be convinced that they are in possession of sufficient truth in order to be able to act.

NOTES 1 Of course, Gadamer is not using prejudice in the way it is often used to designate forms of hate and bias but is referring to the forms of explicit and implicit knowledge that make understanding of the world possible. These forms of knowledge do not necessarily involve hatred and bias in the human rights sense (although they can). His argument is not that hatred, xenophobia and other forms of prejudice are inevitable, but rather that we

Gadamer’s Minimal Realism 315 always already have a framework or world view expressed through tradition that enables us to understand the world around us. 2 See Norris, who targets Rorty in his attack on relativism and includes hermeneutics, the hermeneutic circle, and Gadamer in the category of strong relativists (1997: 81, 147).

This page intentionally left blank

Bibliography

Abley, M. (1998). The ice storm. Toronto: McClelland & Stewart. Abrams, P. (1968). The origins of British sociology, 1834–1914. Chicago: University of Chicago Press. Adam, B. (1998). Timescapes of modernity. London: Routledge. Albritton, R. (1991). A Japanese approach to stages of capitalist development. London: Macmillan. Albritton, R. (1999). Dialectics and deconstruction in political economy. London: Palgrave. Albritton, R. (2003). Returning to Marx’s Capital: A critique of Lebowitz’s Beyond Capital. History of Economic Ideas, 11(3). Alcoff, L.M. (2004). Gadamer’s feminist epistemology. In L. Code (Ed.), Feminist interpretations of Gadamer. University Park: Pennsylvania State University Press. Althusser, L. (1971). Ideology and ideological state apparatuses. In Lenin and philosophy and other essays. London: New Left Books. Althusser, L. (1965/1970a). From Capital to Marx’s philosophy. In Reading Capital (pp. 11– 6). London: New Left Books. Althusser, L. (1965/1970b). Marx’s immense theoretical revolution. In Reading Capital (pp. 182–193). London: New Left Books. Althusser, L. (1972/1976). Reply to John Lewis. In Essays in self-criticism (pp. 35–77). London: New Left Books. Althusser, L. (1975/1990). Is it difficult to be a Marxist in philosophy? In Philosophy and the spontaneous philosophy of the scientists (pp. 203–240). London: Verso. Althusser, L. (1982/2006). The underground current of the materialism of the encounter. In F. Materon & O. Corpt (Eds.), Philosophy of the encounter (pp. 163–207). London: Verso. Althusser, L. (1988/2006). Philosophy and Marxism. In F. Matheron & O. Corpet (Eds.), Philosophy of the encounter (pp. 251–289). London: Verso.

318 Bibliography Althusser, L. (1993/2006). Philosophy of the encounter: Later writings, 1978–1987 (F. Matheron & O. Corpet, Eds.). London: Verso. Althusser, L. (1999). Machiavelli and us. New York: Routledge. Althusser, L. & Balibar, E. (1970). Reading Capital. London: New Left Books. Amadae, S. (2003). Rationalizing capitalist democracy: The cold war origins of rational choice liberalism. Chicago: University of Chicago Press. Anderson, P. (1992). The Peculiarities of the English. London, Verso. Anspach, R.R., and Beeson, D. (2001). Emotions in medical and moral Life. In B. Hoffmaster (Ed.), Bioethics in social context. Philadelphia: Temple University Press. Arac, J. (1980). The function of Foucault at the present time. Humanities in society, 3(1), 73–86. Archer, M.S. (1995). Realist social theory: The morphogenetic approach. Cambridge: Cambridge University Press. Archer, M.S. (2000). Being human: The problem of agency. Cambridge: Cambridge University Press. Archer, M.S. (2003). Structure, agency and the internal conversation. Cambridge: Cambridge University Press. Archer, M.S., Bhaskar, R., Collier, A., Lawson, T. & Norrie, A. (Eds.). (1998). Critical realism: Essential readings. London: Routledge. Archer, M.S., Collier, A., & Porpora, D. (2004). Transcendence: Critical realism and God. London: Routledge. Aristotle. (1960). The Metaphysics (R. Hope, Trans.). Ann Arbor: University of Michigan Press. Armstrong, D.M. (1968). A materialist theory of the mind. London: Routledge and Kegan Paul. Arthur, C.J. (1979). Dialectic of the value form. In D. Elson (Ed.), Value: The representation of labour in capitalism (pp. 67–81). London: CSE Books. Arthur, C. (1993) Hegel’s Logic and Marx’s Capital. In F. Moseley (Ed.), Marx’s method in Capital: A re-examination (pp. 63–97). Atlantic Heights, NJ: Humanities Press. Arthur, C.J. (2002). The new dialectic and Marx’s Capital. Boston: Brill. Backhaus, H.G. (1980). On the Dialectics of the value-form. Thesis Eleven, 1, 94–120. Originally published as ‘Zur Dialektik der Wertform’ in A. Schmidt (Ed.), Beitraege zur marxistischen Erkenntnistheorie. Frankfurt a.m. (1980). Bacon, F. (1620/1963). Novum Organum. In The complete essays of Francis Bacon including the New Atlantis and Novum Organum. New York: Washington Square Press. Badiou, A. (1998/2005). Metapolitics. New York: Verso. Balibar, E. (1989/1992). Foucault and Marx: The question of nominalism. In Michel Foucault: Philosopher (pp. 38–57). New York: Routledge.

Bibliography 319 Bannister, R.C. (1991). Sociology and scientism: The American quest for objectivity. 1880–1940. Chapel Hill: University of North Carolina Press. Barker, D.K. (2003). Emancipatory for whom? A comment on critical realism. Feminist Economics, 9(1), 103–108. Barry, A., Osborne, T., & Rose, N. (Eds.). (1993). Liberalism, neo-liberalism and governmentality. Economy & society 22(3) (Special Issue). Barry, A., Osborne, T., & Rose, N. (Eds.). (1996). Foucault and political reason: Liberalism, neo-liberalism and rationalities of government. Chicago: University of Chicago Press. Bataille, G. (1947/1988). The accursed share. New York: Zone Books. Bataille, G. (1985). Visions of excess: Selected writings 1937–1939. (Alan Stoekl, Ed. Trans.). Minneapolis: University of Minnesota Press. Baudrillard, J. (1995). Simulacra and simulation. Ann Arbor: University of Michigan Press. Beauchamp, T.L., & Childress, J.F. (1994). Principles of biomedical ethics (4th ed.). London: Oxford University Press. Beck, U. (1998). Democracy without enemies. Cambridge, MA: Polity Press. Beeson, D., & Doksum, T. (2001). Family values and resistance to genetic testing in Japan. In B. Hoffmaster (Ed.), Bioethics in social context. Philadelphia: Temple University Press. Bell, D. (1960). The end of ideology. Cambridge, MA: Harvard University Press. Bell, D. (1967). Marxian socialism in the United States. Princeton: Princeton University Press. Bell, D. (1973). The coming of postindustrial society. New York: Basic Books. Bell, D. (1976). The cultural contradictions of capitalism. New York: Basic Books. Benton, T. (1977). Philosophical foundations of the three sociologies. London: Routledge and Kegan Paul. Benton, T. (1981/1998). Realism and social science: Some comments on Roy Bhaskar’s The Possibility of Naturalism. In M.S. Archer et al. (Eds.), Critical realism: Essential readings. London: Routledge. Benton, T. (1981). Realism and social science: Some comments on Roy Bhaskar’s The Possibility of Naturalism. Radical Philosophy, 21, 13–21. Benton, T. (1991). Biology and social science. Sociology, 25, 1–25. Benton, T. (1993). Natural relations. London: Verso. Benton, T. (2001a). Environmental sociology. Sosiologisk tidsskrift, (9), 5–48. Benton, T. (2001b). Why are sociologists naturephobes? In J. Lopez & G. Potter (Eds.), After postmodernism. London: Athlone. Benton, T., & Craib, I. (2001). Philosophy of social science: The philosophical foundations of social thought. Hampshire: Palgrave.

320 Bibliography Berger, J., & Zelditch, M. (Eds.). (2002). New directions in contemporary sociological theory. Lanham, MD: Rowman & Littlefield. Bernstein, B. (1971). Class, codes and control. London: Routledge. Bernstein, R.J. (1983). Beyond objectivism and relativism: Science, hermeneutics and praxis. Philadelphia: University of Pennsylvania Press. Bettelheim, C. (1975). Economic calculation and forms of property. New York: Monthly Review Press. Bhaskar, R. (1975). A realist theory of science. Leeds: Leeds Books. Bhaskar, R. (1975/1997). A realist theory of science. London: Verso. Bhaskar, R. (1978). A realist theory of science. Brighton: Harvester Press. Bhaskar, R. (1978/1997). The possibility of naturalism: A philosophical critique of the contemporary human sciences. London: Routledge. Bhaskar, R. (1979). The possibility of naturalism: A philosophical critique of the contemporary human sciences. NJ: Humanities Press. Bhaskar, R. (1979/1988). The possibility of naturalism. Hemel Hempstead: Harvester Wheatsheaf. Bhaskar, R. (1983). Beef, structure and place: Notes from a critical naturalist perspective. Journal for the Theory of Social Behaviour, 13. Bhaskar, R. (1986). Scientific realism and human emancipation. London: Verso Books. Bhaskar, R. (1989). Reclaiming reality: A critical introduction to contemporary philosophy. London: Verso Books. Bhaskar, R. (1991). Philosophy and the idea of freedom. Oxford: Blackwell. Bhaskar, R. (1993). Dialectic: The pulse of freedom. London: Verso. Bhaskar, R. (1994). Plato etc.: The problems of philosophy and their resolution. London: Verso. Bhaskar, R. (1998a). The possibility of naturalism: A philosophical critique of the contemporary human sciences (3rd Ed.). London: Routledge. Bhaskar, R. (1998b). Facts and values: Theory and practice. In M.S. Archer et al. (Eds.), Critical realism: Essential readings (pp. 409–443). London: Routledge. Bhaskar, R. (1998c). General introduction. In M.S. Archer et al. (Eds.), Critical realism: Essential readings (pp. ix–xxiv). London: Routledge. Bhaskar, R. (1999). The possibility of naturalism: A philosophical critique of the contemporary human sciences. London: Routledge. Bhaskar, R. (2000). From east to west: Odyssey of a soul. London: Routledge. Bhaskar, R., & Callinicos, A. (2003). Marxism and critical realism: A debate. Journal of Critical Realism, 1(2). Bion, W.R. (1961). Experiences in groups and other papers. London: Tavistock. Blaikie, N. (1993). Approaches to social enquiry. Cambridge, MA: Polity Press.

Bibliography 321 Blaikie, P., Cannon, T., Davis, I. & Wisner, B. (1994). At risk. London: Routledge. Bohm, D. (2002). Wholeness and implicate order. London: Routledge. Bosk, C.L. (1999). Professional ethicist available: Logical, secular and friendly. Daedalus, 128(4), 47–68. Bourdieu, P. (1977a). Outline of a theory of practice. Cambridge, MA: Harvard University Press. Bourdieu, P. (1977b). Structures and habitus (R. Nice, Trans.). In Outline of a theory of practice (pp. 72–95). New York: Cambridge University Press. Bourdieu, P. (1980/1990). The logic of practice (R. Nice, Trans.). Stanford: Stanford University Press. Bourdieu, P. (1992a). The practice of reflexive sociology. In P. Bourdieu and L. Wacquant, An invitation to reflexive sociology. Chicago: University of Chicago Press. Bourdieu, P. (1992). Doxa and common life: Interview with Terry Eagleton. New Left Review, 191, 111–121. Bourdieu, P. (1993). Sociology in question. London: Sage. Bourdieu, P. (2000). Pascalian meditations. Cambridge: Polity. Bourdieu, P., & Wacquant, L. (1999). On the cunning of imperialist reason. Theory, culture and society, 16(1), 41–58. Bowker, G.C., & Star, S.S. (2000). Sorting things out: Classification and its consequences. Cambridge, MA: MIT Press. Boyd, R. (1991). Realism, anti-foundationalism, and the enthusiasm for natural kinds. Philosophical Studies, 61, 127–148. Boyd, R. (1999a). Homeostasis, species, and higher taxa. In R.A. Wilson (Ed.), Species: New interdisciplinary (pp. 141–186). Cambridge, MA: MIT Press. Boyd, R. (1999b). Kinds as the ‘Workmanship of men’: Realism, constructivism, and natural kinds. In J. Nida-Rumelin (Ed.), Rationality, realism, revision (Perspectives in analytical philosophy v. 23) (pp. 52–89). Berlin: Walter de Gruyter. Bradbury, M., & McFarlane, J. (1991). Modernism: 1890–1930. Harmondsworth: Penguin. Bratich, J., Packer, P., & McCarthy, C. (Eds.). (2003). Foucault, cultural studies, and governmentality. Albany: SUNY Press. Brenner, Neil. (1994). Foucault’s new functionalism. Theory and Society, 23, 679–709. Brick, H. (1986). Daniel Bell and the decline of intellectual radicalism: Social theory and political reconciliation in the 1940s. Madison: University of Wisconsin Press. Brown, A., Fleetwood, S., & Roberts, M. (2002). Critical realism and Marxism. London: Routledge.

322 Bibliography Brown, D. (2001). Governmentality and law and order. In G. Wickham & G. Pavlich (Eds.), Rethinking law, society and governance: Foucault’s bequest (pp. 109–121). Portland: Hart Publishing. Brown, J.H. (1981). Two decades of homage to Santa Rosalia: Toward a general theory diversity. American Zoologist, 21, 877–88. Bryman, A. (1988). Quantity and quality in social research. New York: Routledge. Bryman, A. (2004). Social research methods. Oxford: Oxford University Press. Butler, J. (1990). Gender trouble: Feminism and the subversion of identity. London: Routledge. Butler, J. (1993). Bodies that matter: On the discursive limits of ‘sex.’ London: Routledge. Butler, J. (1997). The psychic life of power. Stanford: Stanford University Press. Butler, J. (1998). Merely cultural. New Left Review, 227, 33–44. Caillois, R. (1939/1959). Man and the sacred. New York: Free Press. Cain, M. (1986). Realism, feminism, methodology, and law. International Journal of the Sociology of Law, 14(3–4), 255–267. Cain, M. (1990). Realist philosophy and standpoint epistemologies or feminist criminology as a successor science. In L. Gelsthorpe & A. Morris (Eds.), Feminist perspectives in criminology (pp. 124–140). Philadelphia: Open University Press. Cain, M. (1993). Foucault, feminism and feeling: What Foucault can and cannot contribute to feminist epistemology. In C. Ramazanoglu (Ed.), Up against Foucault: Explorations of some tensions between Foucault and feminism (pp. 73–98). New York: Routledge. Calhoun, C. (2003). An apology for moral shame. Journal of Political Philosophy, 11(2), pp. 1–20. Callahan, D. (1999) The social sciences and the task of bioethics. Daedalus (128)4: 275–94. Callinicos, A. (n.d.). Critical realism and beyond: Roy Bhaskar’s dialectic. n.p. Callon, M., (1991). Techno-economic networks and irreversibility. In J. Law (Ed.), A sociology of monsters (pp. 132–161). London: Routledge. Callon, M., & Latour, B. (1992). Don’t throw the baby out with the Bath school! In A. Pickering (Ed.), Science as practice and culture (pp. 343–368). Chicago: University of Chicago Press. Callon, M., Lascoumes, P., & Barthe, Y. (2001). Agir dans un monde incertain. Paris: Seuil. Caplan, A. (1983). Can applied ethics be effective in health care and should it strive to be? Ethics, 93, 311–19. Caplan, A. (1999). Can ethics help guide the future of biomedicine? In The American medical revolution in ethics. R.B. Baker, A.L. Caplan, L.L. Emanuel, & S.R. Latham (Eds.). Baltimore: Johns Hopkins University Press.

Bibliography 323 Carter, B. (2000). Realism and racism: Concepts of race in sociological research. London: Routledge. Carter, B., & New, C. (2004). Making realism work: Realist social theory and empirical research. London: Routledge. Cartwright, N. (1983). How the laws of physics lie. Oxford: Clarendon. Carver, T. (1975). Karl Marx: Texts on method. Oxford: Basil Blackwell. Cavallari, H.M. (1980). Savoir and pouvoir: Michel Foucault’s theory of discursive practice. Humanities in Society, 3(1), 55–72. Chalmers, A. (1988). Is Bhaskar’s realism realistic? Radical Philosophy, 49, 18–23. Chambers, T. (1999). The fiction of bioethics. London: Routledge. Chambliss, D.F. (1993). Is bioethics irrelevant? Contemporary Sociology, 22, 649–652. Chodos, H., & Curtis, B. (2002). Pierre Bourdieu’s masculine domination: A critique. Canadian Review of Sociology and Anthropology, 39(4), 397–412. Code, A. (1984a). On the origins of some Aristotelian theses about predication. In J. Bogen and J.E. McGuire (Eds.), How things are (pp. 101–131). Boston: D. Reidel. Code, A. (1984b). The aporematic approach to primary being in metaphysics Z. Canadian Journal of Philosophy, 10, 1–20. Code, A. (1986). Aristotle: Essence and accident. In R.E. Grandy and R. Warner (Eds.), Philosophical grounds of rationality: Intentions, categories, ends (pp. 411–439). New York: Oxford University Press. Coles, P. (2001). Cosmology: A very short introduction. Oxford: Oxford Paperbacks. Collier, A. (1994). Critical realism: An introduction to Roy Bhaskar’s philosophy. London: Verso. Collier, A. (1998). Explanation and emancipation. In M.S. Archer et al. (Eds.), Critical realism: Essential readings (pp. 444–472). London: Routledge. Collier, A. (1999). Being and worth. London: Routledge. Collier, A. (2003a). In defence of objectivity. London: Routledge Collier, A. (2003b). On Christian belief: A defense of a cognitive conception of religious belief in a Christian context. London: Routledge. Collingwood, R.G. (1939). Question and answer. In An autobiography (pp. 29–43). Oxford: Oxford University Press. Collins, H., & Pinch, T. (1979). The construction of the paranormal: Nothing unscientific is happening. In R. Wallis (Ed.), On the margins of science (pp. 237–70). Keele: University of Keele. Collins, H.M. (1990). Artificial experts: Social knowledge and intelligent machines. Cambridge, MA: MIT Press. Collins, H.M., & Yearley, S. (1992a). Epistemological chicken. In A. Pickering (Ed.), Science as practice and culture (pp. 301–26). Chicago: University of Chicago Press.

324 Bibliography Collins, H.M., & Yearley, S. (1992b). Journey into space. In A. Pickering (Ed.), Science as practice and culture (pp. 369–89). Chicago: University of Chicago Press. Commission scientifique. (1999). Pour affronter l’imprévisible. Québec: Publications du Québec. Cooper, D. (1998). Governing out of order: Space, law and the politics of belonging. London: Rivers Oram Press. Corrigan, O. (2003). Empty ethics: The problem with informed consent. Sociology of Health and Illness, 25(7), 768–792. Craib, I. (1994). The importance of disappointment. London, etc.: Routledge. Craib, I. (1997). Social constructionism as a social psychosis. Sociology, 31(1), 1–18. Craib, I. (1998). Experiencing identity. London: Sage. Creaven, S. (2000). Marxism and realism: A materialistic application of realism in the social sciences. New York: Routledge. Creaven, S. (2003). Marx and Bhaskar on the dialectics of freedom. Journal of Critical Realism, 2(1), 64–93. Cruickshank, J. (2002). Realism and sociology: Anti-foundationalism, ontology and social research. New York: Routledge. Curtis, B. (1995). Taking the state back out: Rose and Miller on political power. British Journal of Sociology, 46(4), 575–589. Dannemark, B., et al. (2002). Explaining society: Critical realism and the social sciences. London: Routledge. Dean, K. (2003). Capitalism and citizenship: The impossible partnership. London: Routledge. Dean, M. (1994a). Critical and effective histories: Foucault’s method and historical sociology. New York: Routledge. Dean, M. (1994b) A social structure of many souls: Moral regulation, government, and self-formation. Canadian Journal of Sociology, 19(2), 145–168. Dean, M. (1999). Governmentality: Power and rule in modern society. Thousand Oaks, CA: Sage. Delanty, G. (1997). Social science: Beyond constructivism and realism. Buckingham: Open University Press. Deleuze, G. (1962/1983). Nietzsche and philosophy. New York: Columbia University Press. Deleuze, G. (1970/1980). Spinoza: Practical philosophy. San Francisco: City Lights Books. Deleuze, G. (1972/2004). How do we recognize structuralism? Desert Islands, NY: Semiotext(e). Deleuze, G. (1986/1988). Foucault. Minneapolis: University of Minnesota Press.

Bibliography 325 Deleuze, G. (1989/1992). What is a dispositif? In Michel Foucault: Philosopher (pp. 159–168). New York: Routledge. Deleuze, G. (1990/1995). Negotiations: 1972–1990. New York: Columbia University Press. Deluze, G., & Guattari, F. (1983). Anti-Oedipus: Capitalism and schizophrenia. Minneapolis: University of Minnesota Press. Deleuze, G., & Guattari, F. (1987). A thousand plateaus: Capitalism and schizophrenia. Minneapolis: University of Minnesota Press. Denis, H. (2002). Stress et catastrophes. In D. Maltais (Ed.), Catastrophes et état de santé des individus, des intervenants et des communautés (pp. 23–33). Chicoutimi, QC: UQAC. Derrida, J. (1977). Of grammatology. Baltimore: Johns Hopkins University Press. Derrida, J. (1992). The other heading: Reflections on today’s Europe. Bloomington: University of Indiana Press. Derrida, J. (1994). Specters of Marx. London: Routledge. DeVries, R., & Conrad, P. (1998). Why bioethics needs sociology. In R. DeVries & J. Subedi (Eds.), Bioethics and society: Constructing the ethical enterprise. Upper Saddle River, NJ: Prentice Hall. DeVries, R., & Subedi, J. (Eds.). (1998). Bioethics and society: Constructing the ethical enterprise. Upper Saddle River, NJ: Prentice Hall. Dickens, P. (1992). Society and nature. Philadelphia: Temple University Press. Dickens, P. (2001). Linking the social and natural sciences. Sociology 35, 93–110. Dickens, P. (2003). Changing our environment, changing ourselves. Interdisciplinary Science Review 28, 95–105. Dickens, P. (2004). Society and nature. Cambridge, MA: Polity Press. Dillon, M. (1995). Sovereignty and governmentality: From the problematics of the ’New World Order’ to the ethical problematic of the world order. Alternatives, 20, 323–368. Dombrowsky, W. (1998). Again and again: Is a disaster what we call a disaster? In E.L. Quarantelli (Ed.). What is a disaster? (pp. 19–30). London: Routledge. Dostal, R.J. (1994). The experience of truth for Gadamer and Heidegger: Taking time and sudden lightening. In Brice Wachterhauser (Ed.), Hermeneutics and truth. Evanston, IL: Northwestern University Press. Dow, J. (2006). The evolution of religion: Three anthropological approaches. Method and theory in the study of religion (forthcoming). Dupont, D., & Pearce, F. (2001). Foucault contra Foucault: Rereading the ’Governmentality’ papers. Theoretical Criminology, 5(2), 123–158. Dupre, John. (2001). Human nature and the limits of science. Oxford: Clarendon Press. Durkheim, E. (1883/1984). The division of labour in society. London: Macmillan.

326 Bibliography Durkheim, E. (1884/1984). The rules of sociological method and selected texts on sociology and method. London: Macmillan. Durkheim, E. (1893/1997). The division of labour. New York: Free Press. Durkheim, E., (1895/1982). The rules of sociological method and selected texts on sociology and its method. S. Lukes (Ed.). London: Macmillan. Durkheim, E. (1986/1952). Suicide. London: Routledge & Kegan Paul. Durkheim, E. (1897/1951). Suicide. New York: Free Press of Glencoe. Durkheim, E. (1950/1996). Professional ethics and civic morals. London: Routledge. Durkheim, E. (1964). The rules of sociological method. New York: Free Press. Durkheim, E. (1912/1995). The elementary forms of religious life. New York: Free Press. Dyer-Witheford, N. (1999). Cyber-Marx: Cycles and circuits of struggle in hightechnology capitalism. Urbana: University of Illinois Press. Dynes, R. (1998). Coming to terms with community disaster. In E.L. Quarantelli (Ed.), What is a disaster? (pp. 109–126). London: Routledge. Eder, K. (1996). The social construction of nature. London: Sage. Ehrbar, H.G. (2005). Annotations to Karl Marx’s Capital. Draft version under development. Available at http://www.econ.utah.edu/ehrbar/akmc.htm. Eldred, M., & Hanlon, M. (1981). Reconstructing Value-Form Analysis. Capital and Class, 13, 24–60. Elias, N. (1983). The court society. Oxford: Blackwell. Elliot, C. (1999). Bioethics, culture and identity: A philosophical disease. London: Routledge. Ellis, B. (2001). Scientific essentialism. Cambridge: Cambridge University Press. Ellis, B. (2002). The philosophy of nature: A guide to the new essentialism. Montreal: McGill-Queen’s University Press. Elliot, G. (1987). The detour of theory. New York: Verso. Elliot, G. (1998). The necessity of contingency: Some notes. Rethinking Marxism, 10(3), 74–79. Engel, P. (2002). Truth. Montreal: McGill-Queen’s University Press. Engelskirchen, H. (1997). Consideration as the commitment to relinquish autonomy. Seton Hall L. Rev, 27, 490–573. Engelskirchen, H. (2003). Value and contract formation. In J. Joseph & J.M. Roberts (Eds.), Realism, discourse and deconstruction (pp. 111–134). New York: Routledge. Engelskirchen, H. (2006). On the clear comprehension of political economy: Social kinds and the significance of section 2 of Marx’s Capital. Unpublished manuscript. Evans, J.H. (2002). Playing God? Human genetic engineering and the rationalization of public bioethical debate. Chicago: University of Chicago Press.

Bibliography 327 Fairclough, N. (1992). Michel Foucault and the analysis of discourse. In Discourse and Social Change (pp. 37–55). Cambridge, MA: Polity Press. Fairclough, N., Jessop, B., Sayer, A. (2002). Critical realism and semiosis, Journal of critical realism, 5(1), pp. 2–10. Feyerabend, P.K. (1965). Problems of empiricism. In R.G. Colodny (Ed.), Beyond the edge of certainty: Essays in contemporary science and philosophy. Engelwood Cliffs, NJ: Prentice Hall. Feyerabend, P.K. (1970). Consolations for the specialist. In I. Lakatos & A. Musgrave (Eds.), Criticism and the growth of knowledge. Cambridge: Cambridge University Press. Feyerabend, P.K. (1975). Against method. London: Verso. Fine, A. (1986). The shaky game: Einstein, realism and the quantum theory. Chicago: University of Chicago Press. Foucault, M. (1963/1973). The birth of the clinic: An archaeology of medical perception. New York: Vintage. Foucault, M. (1966/1971). The order of things: An archaeology of the human sciences (Unknown, Trans.). New York: Vintage. Foucault, M. (1968/1991). Politics and the study of discourse. In G. Burchell, C. Gordon, & P. Miller (Eds.), The Foucault effect: Studies in governmentality (pp. 53–72). Chicago: University of Chicago Press. Foucault, M. (1970/1972). The discourse on language (A. Sheridan, Trans.). In The archaeology of knowledge (pp. 215–237). New York: Pantheon. Foucault, M. (1972a). The archaeology of knowledge. London: Routledge. Foucault, Michel. (1972b). The archaeology of knowledge. New York: Pantheon. Foucault, M. (1972/1996). Confining Societies. In S. Lotringer (Ed.), Foucault live (pp. 83–94). New York: Semiotext(e). Foucault, M. (1973/1996). Equipments of power. In S. Lotringer (Ed.), Foucault live (pp. 105–112). New York: Semiotext(e). Foucault, M. (1974/2003). The birth of social medicine. In Paul Rabinow & Nikolas Rose (Eds.), The Essential Foucault (pp. 300–318). New York: New Press. Foucault, M. (1974/1996). On Attica. In S. Lotringer (Ed.), Foucault live (pp. 113–121). New York: Semiotext(e). Foucault, M. (1975/1979). Discipline and punish: The birth of the prison (A. Sheridan, Trans.). New York: Pantheon. Foucault, M. (1975/1980). Body/power. In C. Gordon (Ed.), Power/knowledge (pp. 55–62). Toronto: Random House. Foucault, M. (1976/1980). Two lectures. In C. Gordon (Ed.), Power/knowledge (pp. 78–108). New York: Pantheon. Foucault, M. (1976/1994). The history of sexuality. New York: Vintage. Foucault, M. (1976/2003). 17 March 1976. In Society must be defended: Lectures at the Collège de France 1975–1976 (pp. 239–264). New York: Picador.

328 Bibliography Foucault, M. (1977). Nietzsche, genealogy, history. In D. Bouchard (Ed.), Language, counter-memory, practice (pp. 139–164). New York: Cornell University Press. Foucault, M. (1977/1980a). Power and Strategies. In Power/knowledge (pp. 134–145). New York: Vintage Books. Foucault, M. (1977/1980b). The confession of the flesh. In Power/knowledge (pp. 194–228). New York: Vintage Books. Foucault, M. (1977/1980c). The history of sexuality. In Power/knowledge (pp. 183–193). New York: Vintage Books. Foucault, M. (1977/1990). Confinement, psychiatry, prison. In L.D. Kritzman (Ed.), Michel Foucault: Politics, philosophy, culture (pp. 178–210). New York: Routledge. Foucault, M. (1977/2003a). Truth and power. In P. Rabinow & N. Rose (Eds.), The essential Foucault (pp. 300–318). New York: New Press. Foucault, M. (1977/2003b). Lives of infamous men. In P. Rabinow & N. Rose (Eds.), The essential Foucault (pp. 279–293). New York: New Press. Foucault, M. (1978/1991). Governmentality. In G. Burchell, C. Gordon & P. Miller (Eds.), The Foucault effect: Studies in governmentality, with two lectures by and an interview with Michel Foucault (pp. 87–104). London: Harvester Wheatsheaf Press. Foucault, M. (1979/2003a). ‘Omnes et singulatim’: Toward a critique of political reason. In P. Rabinow & N. Rose (Eds.), The essential Foucault (pp. 202–207). New York: New Press. Foucault, M. (1979/2003b). The birth of biopolitics. In P. Rabinow & N. Rose (Eds.), The essential Foucault (pp. 202–207). New York: New Press. Foucault, M. (1980/2003). Questions of method. In P. Rabinow & N. Rose (Eds.), The essential Foucault (pp. 246–258). New York: New Press. Foucault, M. (1982/2003). The subject and power. In P. Rabinow & N. Rose (Eds.), The essential Foucault (pp. 126–144). New York: New Press. Foucault, M. (1983/1996a). How much does it cost for reason to tell the truth? In S. Lotringer (Ed.). Foucault live (pp. 348–362). New York: Semiotext(e). Foucault, M. (1983/1996b). What our present is. In S. Lotringer (Ed.), Foucault live (pp. 407–415). New York: Semiotext(e). Foucault, M. (1981/1997). What our present is. In J. Fleming & S. Lotringer (Eds.), The politics of truth (pp. 147–168). New York: Semiotext(e). Foucault, M. (1983/2000). What is called ‘punishing’? In J.D. Faubion (Ed.), Power (pp. 382–393). New York: New Press. Foucault, M. (1983). The subject and power. In H. Dreyfus & P. Robinson, Michel Foucault: Beyond structuralism and hermeneutics (pp. 208–226).

Bibliography 329 Foucault, M. (1983). On the genealogy of ethics: An overview of a work in progress. In H. Dreyfus & P. Rabinow (Eds.), Michel Foucault: Beyond structuralism and hermeneutics (pp. 229–252). Chicago: University of Chicago Press. Foucault, M. (1984/1985). The use of pleasure: Volume two of the history of sexuality (R. Hurley, Trans.). New York: Vintage. Foucault, M. (1984/1996). An aesthetics of existence. In S. Lotringer (Ed.), Foucault live (pp. 450–454). New York: Semiotext(e). Foucault, M. (1984/2003a). Preface to the history of sexuality, volume two. In P. Rabinow & N. Rose (Eds.), The essential Foucault (pp. 58–63). New York: New Press. Foucault, M. (1984/2003b). The ethics of the concern for the self as a practice of freedom. In P. Rabinow & N. Rose (Eds.), The essential Foucault (pp. 25–42). New York: New Press. Foucault, M. (1984/2003c). What is enlightenment? In P. Rabinow & N. Rose (Eds.), The essential Foucault (pp. 43–57). New York: New Press. Foucault, M. (1997/2003a). Society must be defended. New York: Picador. Foucault, M. (1997/2003b). What is critique? In P. Rabinow & N. Rose (Eds.), The essential Foucault (pp. 263–278). New York: New Press. Foucault, M. (2003). Foucault. In P. Rabinow & N. Rose (Eds.), The essential Foucault (pp. 1–5). New York: New Press. Foucault, M., & Deleuze, G. (1972/1977). Intellectuals and power. In D. Bouchard (Ed.), Language, counter-memory, practice. New York: Cornell University Press. Fox, R.C. (1989). The sociology of medicine. Englewood Cliffs, NJ: Prentice Hall. Fox, R.C., & Swazey, J.P. (1984). Medical morality is not bioethics: Medical ethics in China and the United States. Perspectives in Biology and Medicine, 35, 336–360. Frank, D., & Meyer, D. (2000). The Contemporary Identity Explosion: Individualizing Society in the Postwar Period. Mimeo. Frankel, B. (1997). Confronting neoliberal regimes: The post-Marxist embrace of populism and realpolitik. New Left Review, 226, 57–92. Fraser, N. (1981). Foucault on modern power: empirical insights and normative confusions. Praxis International. 1(3), 272–87. Reprinted in (1989) Fraser’s unruly practices: Power, discourse, and gender in contemporary social theory. Minneapolis: University of Minnesota Press. Frauley, J. (2004). Race, justice, and the production of knowledge: A critical realist consideration. Canadian Journal of Law and Society, 19(2), 177–197. Frauley, J. (2005a). Architextures of governance: Toward an archaeologicalrealist ontology of social ordering. Unpublished PhD dissertation, Queen’s University, Kingston, Ontario.

330 Bibliography Frauley, J. (2005b). Representing theory and theorising in criminal justice studies: practising theory considered. Critical Criminology: An International Journal, 13(3), 245–265. Frauley, J. (2006). ‘Temporality and non-humanist agential power: A Bourdieuan contribution to Foucauldian governance.’ Unpublished manuscript. Frauley, J. (2007). Towards an archaeological-realist Foucauldian analytics of government. British Journal of Criminology. Friedan, B. (1963). The feminine mystique. New York: Dell. Friedland, R., & Mohr, J. (2004). Matters of culture: Cultural sociology in practice. Cambridge, MA: Harvard University Press. Gadamer, H.-G. (1994). What is truth? In B.R. Wachterhauser (Ed.), Hermeneutics and truth. Evanston, IL: Northwestern University Press. Gadamer, H.-G. (2003). Truth and method (2nd ed.). New York: Continuum. Garland, D. (1997). Governmentality and the problem of crime. Theoretical Criminology, 1(2), 173–214. Gert, B., Culver, C.M., & Clouser, K.D. (1997). Bioethics: A return to fundamentals. New York: Oxford University Press. Gervais, K.G. (1998). Changing society, changing medicine, changing bioethics. In R. DeVries & J. Subedi (Eds.), Bioethics and society: Constructing the ethical enterprise. Upper Saddle River, NJ: Prentice Hall. Giddens, A. (1976). New rules of sociological method. London: Hutchinson. Giddens, A. (1984). The constitution of society. Cambridge: Polity. Huntington, S. (1982). American politics: The promise of disharmony. Cambridge, MA: Harvard University Press. Gieryn, T.F. (1999). Cultural boundaries of science. Chicago: University Chicago Press. Gilbert, C. (1998). Studying disaster. In E.L. Quarantelli (Ed.), What is a disaster? (pp. 11–18). London: Routledge. Gilligan, C. (1983). In a different voice: Psychological theory and women’s development. Cambridge, MA: Harvard University Press. Gillon, R. (1994). The four principles revisited: A reappraisal. In R. Gillon (Ed.), Principles of health care ethics. Chichester: John Wiley and Sons. Gilman, N. (2004). Mandarins of the future: Modernization theory in cold war America. Baltimore: Johns Hopkins University Press. Good, B.J. (1994). Medicine, rationality and experience. Cambridge: Cambridge University Press. Groff, R. (2000). The truth of the matter: Roy Bhaskar’s critical realism and the concept of alethic truth. Philosophy of the Social Sciences, 30(3), 407–435. Groff, R. (2004). Critical realism, post-positivism, and the possibility of knowledge. London: Routledge.

Bibliography 331 Grundmann, R. (2001). Transnational environmental policy. London: Routledge. Habermas, J. (1977). A review of Gadamer’s Truth and Method. In Fred R. Dallmayr & Thomas A. McCarthy (Eds.), Understanding and social inquiry. Notre Dame: University of Notre Dame Press. Habermas, J. (1987). The philosophical discourse of modernity: Twelve lectures. Cambridge, MA: MIT Press. Habermas, J. (2003). The future of human nature. Cambridge: Polity Press. Hacking, I. (1995). The looping effects of human kinds. In D. Sperber, D. Premack, & A. Premack (Eds.), Causal cognition: A multidisciplinary approach. Oxford: Clarendon Press. Hacking, I. (1999). The social construction of what? Cambridge, MA: Harvard University Press. Hammett, I. (1984). Durkheim and the study of religion. In Steve Fenton (Ed.), Durkheim and modern sociology. Cambridge: Cambridge University Press. Han, B. (2002). Foucault’s critical project. Stanford: Stanford University Press. Hannigan, J. (1995). Environmental sociology. London: Routledge. Harding, S. (1986). The science question in feminism. Ithaca, NY: Cornell University Press. Harding, S. (1999). The case for strategic realism: A response to Lawson. Feminist Economics, 5(3), 127–133. Harding, S. (2003). Representing reality: The critical realism project. Feminist Economics, 9(1), 151–159. Hardt, M., & Negri, A. (2000). Empire. Cambridge, MA: Harvard University Press. Harré, R. (1970). Principles of scientific thinking. London: Macmillan Harré, R. (1972). Philosophies of science. Oxford: Oxford University Press. Harré, R., and Bhaskar, R. (2001). How to change reality: Story vs. structure – a debate between Rom Harré and Roy Bhaskar. In J. Lopez and G. Potter (Eds.), After postmodernism: An introduction to critical realism. London: Athlone. Harré, R. and Madden, E.H. (1973). Natural powers and natural natures. Philosophy. 45 Harré, R., and Madden, E.H. (1975). Causal power. Oxford: Blackwell. Harré, R., and Secord, F. (1972). The explanation of social behaviour. Oxford: Blackwell. Hartwig, M. (2001). New left, new age, new paradigm? Roy Bhaskar’s from East to West. Journal for the Theory of Social Behaviour, 31(2), 139–66. Healy, D. (2004). Shaping the intimate: Influences on the experience of everyday nerves. Social Studies of Science, 34, 219–245. Heller, T., Sosna, M., & Wellbery, T. (Eds.). (1986). Reconstructing individualism. Stanford: Stanford University Press.

332 Bibliography Helm, B.W. (2001). Emotional reason: Deliberation, motivation and the nature of value, Cambridge: Cambridge University Press. Hempel, C. (1965). Aspects of scientific explanation. New York: Free Press. Hempel, C. (1966). Philosophy of natural science. Engelwood Cliffs, NJ: Prentice Hall. Hempel, C. (1969). Logical positivism and the social sciences. In P. Achinstein & S.F. Barker (Eds.), The legacy of logical positivism. Baltimore: Johns Hopkins University Press. Henriques, –, et al. (Eds.). (1983). Changing the subject: Psychology, social regulation and subjectivity. London: Routledge. Hermer, J., & Hunt, A. (1996). Official graffiti of the everyday. Law & Society Review, 30(3), 455–480. Hinkson, J. (1998). Subjectivity and neo-liberal economy: The poverty of governmentalist approaches. Arena Journal, 11, 119–144. Hirst, P.Q., & Woolley, P. (1982). Social relations and human attributes. London: Tavistock. Hobsbawm, E.J., & Ranger, T. (Eds.). (1983). The invention of tradition. Cambridge: Cambridge University Press. Hochschild, A.R. (1983). The managed heart. Berkeley: University of California Press. Hoffmaster, B. (1992). Can ethnography save the life of medical ethics. Social Science and Medicine, 35(12), 1421–1431. Hoffmaster, B. (Ed.). (2001). Bioethics in social context. Philadelphia: Temple University Press. Hogg, R., & Carrington, K. (2001). Governing rural Australia: Land, space and race. In G. Wickham & G. Pavlich (Eds.), Rethinking law, society and governance: Foucault’s bequest (pp. 43–60). Portland: Hart Publishing. Hook, D. (2001). Discourse, knowledge, materiality, history: Foucault and discourse analysis. Theory and Psychology, 11(4), 521–547. Horwich, Paul. (1990). Truth. Oxford: Blackwell. Hostettler, N., and Norrie, A. (2003). Are critical realist ethics foundationalist? In J. Cruickshank (Ed.), Critical realism: The difference it makes. London: Routledge. 2003. Hume, D. (1740/1962). On human nature and understanding. In An enquiry concerning human understanding. London: Crowell-Collier Publishing. Hume, D. (1777/1962). Enquiries concerning the human understanding and concerning the principle of morals. Oxford: Oxford University Press. Hunt, A. (1978). The sociological movement in law. London: Macmillan. Hunt, A. (1993). Foucault’s expulsion of law. In Explorations in law and society (pp. 267–300). New York: Routledge.

Bibliography 333 Hunt, A., & Wickham, G. (1994). Foucault and law: Towards a sociology of law as governance. Boulder: Pluto Press. Imber, J.B. (1998). Medical publicity before bioethics: Nineteenth century illustrations of twentieth century dilemmas. In R. DeVries & J. Subedi (Eds.), Bioethics and society: Constructing the ethical enterprise. Upper Saddle River, NJ: Prentice Hall. IPCC (Intergovernmental Panel on Climate Change). (2001). Climate change 2001. Cambridge: Cambridge University Press. Isaacs, A., Daintith, J. & Martin, E. (1999). A dictionary of science. Oxford: Oxford University Press. Jepperson, R. (2002). The development and application of neo-institutionalism. In J. Berger & M. Zelditch (Eds.), New directions in contemporary sociological theory. Lanham, MD: Rowman & Littlefield. Jessop, B. (1990). State theory. University Park: Pennsylvania University Press. Jones, K., & Mulherin, N. (1998). An evaluation of the severity of the January 1998 ice storm in northern New England. Hanover, NH: U.S. Army Laboratory. Jonsen, A.R. (1998). The birth of bioethics. New York: Oxford University Press. Joseph, J. (2002). Hegemony: A realist analysis. London: Routledge. Joseph, J. (2004). Hegemony in the fourth dimension. Journal for the Theory of Social Behaviour, 31(3), 261–277. Joseph, J., and Roberts, J. (2003). Realism, discourse and deconstruction. London: Routledge. Keat, R., & Urry, J. (1975). Social theory as science. London: Routledge and Kegan Paul. Keller, R., Boyd, R., & Wheeler, Q. (2003). The illogical basis of phylogenetic nomenclature. The Botanical Review 69(1), 93–110. Kelly, M. (1998). Critique and power: Recasting the Foucault/Habermas debate. Cambridge, MA: MIT Press. Kendall, G. (1997). Governing at a distance: The colonisation of Australia. In C. O’Farrell (Ed.), Foucault: The legacy (pp. 90–103). Kelvin Grove, Australia: Queensland Institute of Technology. Kerr, D. (1999). Beheading the king and enthroning the market: A critique of Foucauldian governmentality. Science & Society, 63(2), 173–202. Kessler, S.J. & McKenna, W. (1985). Gender: An ethnomethodological approach. Chicago: University of Chicago Press. Kleinman, A. (1997). Writing at the margin: Discourse between anthropology and medicine. Berkeley: University of California Press. Knorr Cetina, K.D. (1981). The manufacture of knowledge: An essay on the constructivist and contextual nature of science. Oxford: Pergamon Press.

334 Bibliography Kornblith, H. (1993). Inductive inference and its natural ground: An essay in naturalistic epistemology. Cambridge, MA: MIT Press. Kosman, A. (1994). The activity of being in Aristotle’s metaphysics. In T. Scaltas, D. Charles, & M.L. Gill (Eds.), Unity, identity, and explanation in Aristotle’s metaphysics (pp. 195–213). Oxford: Clarendon Press. Kourkoulakos, S. (2003). The specificity of dialectical reason. In R. Albritton & J. Simoulidis (Eds.), New dialectics and political economy. London: Palgrave. Kreeger, L. (Ed.). (1975). The large group: Dynamics and therapy. London: Constable. Kuhn, T. (1970). The structure of scientific revolutions (2nd ed.). Chicago: University of Chicago Press. Kymlicka, W., and Matthews, J. (Eds). (1998). The Reagan revolution. Chicago: Dorsey Press. Lacan, J. (1968). Speech and language in psychoanalysis. Baltimore: Johns Hopkins University Press. Lacroix, B. (1979). The elementary forms of religious life as a reflection on power (objet pouvoir). Critique of Anthropology, 5. Lakatos, I., and Musgrave, A. (Eds.) (1970). Criticism and the growth of knowledge. Cambridge: Cambridge University Press. Lasch, C. (1978). The culture of narcissism. New York: Norton. Latour, B. (1987). Science in action. Cambridge, MA: Harvard University Press. Latour, B. (1991a). Nous n’avons jamais été modernes. Paris: La découverte. Latour, B. (1991b). Technology is society made durable. In J. Law (Ed.), A sociology of monsters (pp. 103–131). London: Routledge. Latour, B. (1996). Aramis or the love of technology. Cambridge, MA: Harvard University Press. Latour, B. (1999). Pandora’s hope. Cambridge, MA: Harvard University Press. Latour, B. (2000). When things strike back. British Journal of Sociology, 51, 107–123. Latour, B., & Woolgar, S. (1986). Laboratory life: The construction of scientific facts (2nd ed.). Princeton, NJ: Princeton University Press. Lawson, T. (1999). Feminism, realism and universalism. Feminist Economics, 5(2), 25–59. Lawson, T. (2003a). Ontology and feminist theory. Feminist Economics, 9(1), 119–150. Lawson, T. (2003b). Theorizing Ontology. Feminist Economics, 9(1), 161–169. Lawson, T., & Norrie, A. (Eds.). Critical realism: Essential readings. New York: Routledge. Layder, D. (1990). The realist image in social science. London: Macmillan. Layder, D. (1993). New strategies in social research: An introduction and guide. Cambridge, MA: Polity Press.

Bibliography 335 Lear, J. (1988). Aristotle: The desire to understand. Cambridge: Cambridge University Press. Lecourt, D. (1970/1975). On archaeology and knowledge. In Marxism and epistemology (pp. 187–213). London: New Left Books. Leiter, B. (1996). Legal realism. In D. Patterson (Ed.), A companion to philosophy of law and legal theory (pp. 261–280). Malden, MA: Blackwell. Lemert, C. (2004). Sociology: After the crisis (2nd ed.). Boulder: Paradigm Publishers. Levine, D.N. (1995). Visions of the sociological tradition. Chicago: University of Chicago Press. Lewis, J. (1999). Development in disaster-prone places. London: Intermediate Technology. Light, D.W., & McGee, G. (1998). On the social embeddedness of bioethics. In R. Devries & J. Subedi (Eds.), Biothics and society: Constructing the ethical enterprise. Upper Saddle River, NJ: Prentice Hall. Lipietz, A. (1983). The enchanted world: Inflation, credit and the world crisis. London: Verso. Lock, M. (2001). Situated ethics, culture, and the brain death ‘problem’ in Japan. In B. Hoffmaster (Ed.), Bioethics in social context. Philadelphia: Temple University Press. Lockard, B. (2006, June 10). Cornell study examines surge in self-injury. Ithaca Journal, pp. 1B-2B. Longino, H.E. (2002). The fate of knowledge. Princeton, NJ: Princeton University Press. López, J. (2003a). Critical realism: The difference it makes, in theory. In J. Cruickshank (Ed.), Critical Realism: The difference it makes (pp. 75–89). New York: Routledge. López, J. (2003b). Society and its metaphors: Language, social theory and social structure. London: Continuum. López, J. (2004). How sociology can save bioethics … maybe. Sociology of Health and Illness, 26(7), 875–896. López, J., & Potter, G. (2001a). After postmodernism: The new millennium. In J. López & G. Potter (Eds.), After postmodernism: An introduction to critical realism (pp. 3–16). London: Athlone Press. López, J., & Potter, G. (Eds.). (2001b). After postmodernism: An introduction to critical realism. London: Athlone Press. Lowman, J., & MacLean, B. (1992). Realist criminology: Crime control and policing in the 1990s. Toronto: University of Toronto Press. Luhmann, N. (1990). Essays on self reference. Chicago: University of Chicago Press.

336 Bibliography Lyotard, J.F. (1979). The postmodern condition. Manchester: Manchester University Press. MacArthur, R.H. & Wilson, E.O. (1963). An equilibrium theory of insular zoogeography. Evolution, 17, 373–87. MacArthur, R.H., & Wilson, E.O. (1967). The theory of island biogeography. Princeton, NJ: Princeton University Press. McIntyre, A. (1981). After virtue. London: Duckworth. MacNaghten, P., & Urry, J. (1998). Contested natures. London: Sage. Malpas, J. (2001). Governing theory: Ontology, methodology and the critique of metaphysics. In G. Wickham & G. Pavlich (Eds.), Rethinking law, society and governance: Foucault’s bequest (pp. 125–140). Portland: Hart Publishing. Manent, P. (1998). The city of man. Princeton, NJ: Princeton University Press Manicas, P.T. (1987). A history and philosophy of the social sciences. New York: Blackwell. Marsden, R. (1999). The nature of capital: Marx after Foucault. London: Routledge. Marx, K. (1967a). Capital, v. 1. New York: International Publishers. Marx, K. (1967b). Capital, v. 2. New York: International Publishers. Marx, K. (1968). Theories of surplus value, v. 2. Moscow: Progress. Marx, K. (1970). Das Kapital, Buch 1. In Karl Marx & Friedrich Engels. Marx Engels Werke. Band 23. Berlin: Dietz Verlag. Marx, K. (1971). Theories of surplus value, v. 3. Moscow: Progress. Marx, K. (1973). The Gründrisse (M. Nicolaus, Trans.). Harmondsworth: Penguin. Marx, K. (1975). Notes on Wagner. In T. Carver (Ed.), Texts on method. New York: Harper & Row. Marx, K. (1976). Capital, v. 1. New York: Penguin. Marx, K. (1978a). Capital, v. 2. New York: Penguin. Marx, K. (1978b). Contribution to the critique of Hegel’s Philosophy of right: An introduction. In R.C. Tucker (Ed.), The Marx–Engels reader. London: Norton. Marx, K. (1981). Capital, v. 3. New York: Penguin. Marx, K. (1983). Das Kapital, Kritik der politischen ¨Okonomie, Erster Band, Hamburg 1867. Text/Apparat, volume 5 of MEGA Abt. 2: Das Kapital und Vorarbeiten. Dietz Verlag, Berlin. First edition of ‘Das Kapital.’ Marx, K. (1986). In Marx Engels Collected Works, v. 28. New York: International Publishers. Marx, K. (1987a). Das Kapital, Kritik der politischen ¨Okonomie, Erster Band, Hamburg 1872. Text/Apparat, volume 6 of MEGA Abt. 2: Das Kapital und Vorarbeiten. Dietz Verlag, Berlin. Second edition of “Das Kapital” together with preparatory materials. Marx, K. (1987b). Economic Works 1857–61. In Marx Engels collected works, v. 29. New York: International Publishers.

Bibliography 337 Marx, K. (1996). Capital, v. 1. In Marx Engels collected works, v. 35. New York: International Publishers. Marx, K. (1998). The German ideology. Amherst, NY: Prometheus Books. Marx, K., & Engels, F. (1845/1976). The German ideology. London: Lawrence and Wishart. Marx, K., & Engels, F. (1975). Works 1874–83. In Marx Engels collected works, v. 24. New York: International Publishers. Matheron, F. (1998). The recurrence of the void in Louis Althusser. Rethinking Marxism, 10(3) 22–37. Maturana, H. & Varela, F. (1980). Autopoiesis and cognition: The realization of the living. Boston: D. Reidel. Maturana, H. & Varela, F. (1987/1998). The tree of knowledge: The biological roots of human understanding. Boston: Shambhala. McInerney, D. (Ed.). (2005). Special Issue on Althusser. Borderlands e-journal, 4(2). McIntosh, D. (1983). Max Weber as a critical theorist. Theory and Society, 12. Meikle, S. (1991). History of philosophy: The metaphysics of substance in Marx. In T. Carver (Ed.), Cambridge Companion to Marx (pp. 296–313). Cambridge: Cambridge University Press. Mepham, J. (1972). The theory of ideology. Capital Radical Philosophy, 2. Meyer, J. (1986). Myths of socialization and personality. In Heller et al. (Eds.) Reconstructing individualism. Stanford: Stanford University Press. Midgley, M. (1972). Is ’moral’ a dirty word? Philosophy, 47(181), 206–228. Miller, P., and Rose, N. (1990). Governing economic life. Economy and Society, 19(1), 1–31. Mills, C.W. (1951). White collar: The American middle class. New York: Oxford. Milton, J., & Bourque, A. (1999). A climatological account of the January 1998 ice storm in Quebec. Ottawa: Environment Canada. Milios, J., Dimoulis, D., & Economakis, G. (2002). Karl Marx and the classics. Burlington, VT: Ashgate Publishing. Modrak, D.K. (1985). Forms and compounds. In J. Bogen & J.E. McGuire (Eds.), How things are (pp. 85–90). Dordrecht: D. Reidel Publishing. Mohanty, C. (2003). Feminism without borders: Decolonizing theory, practicing solidarity. Durham: Duke University Press. Monk, R. (2004). Objectivity, postmodernism and biographical understanding: Andrew Collier on R.D. Laing. In M. Archer & W. Outhwaite (Eds.), Defending objectivity: Essays in honour of Andrew Collier. London: Routledge. Montag, W. (1998). Althusser’s nominalism: Structure and singularity. Rethinking Marxism 10(3), 64–73.

338 Bibliography Moravcsik, J.M. (1991). What makes reality intelligible? Reflections on Aristotle’s theory of Aitia. In L. Judson (Ed.), Aristotle’s physics: A collection of essays (pp. 31–470). Oxford: Clarendon Press. Morrow, R. (1994). Critical theory, poststructuralism, and critical realism: Reassessing the critique(s) of positivism. Current Perspectives in Social Theory, 14, 27–51. Murdoch, J. (2001). Ecologising sociology. Sociology, 35, 111–133. Murray, P. (1993). The necessity of money: How Hegel helped Marx surpass Ricardo’s theory of value. In F. Moseley (Ed.), Marx’s method in Capital: A re-examination (pp. 37–61). Atlantic Highlands, NJ: Humanities Press. Murray, P. (1997). Redoubled empiricism: The place of social form and formal causality in Marxian theory. In F. Moseley, and M. Campbell (Eds.), New investigations of Marx’s method (pp. 38–65). Atlantic Highlands, NJ: Humanities Press. Murphy, R. (1994). Rationality and nature. Boulder: Westview. Murphy, R. (1997). Sociology and nature. Boulder: Westview. Murphy, R. (1999). Unperceived risk. Advances in Human Ecology, 8, 99–123. Murphy, R. (2001). Nature’s temporalities and the manufacture of vulnerability. Time and Society, 10, 329–348. Murphy, R. (2002a). Ecological materialism and the sociology of Max Weber. In R. Dunlap, F. Buttel, P. Dickens & A. Gijswijt (Eds.), Sociological theory and the environment (pp. 73–89). New York: Rowman and Littlefield. Murphy, R. (2002b). Extreme weather and the energy metabolism of the city. Environment and History, 8, 43–64. Murphy, R. (2002c). The internalization of autonomous nature into society. Sociological Review, 50, 313–333. Murphy, R. (2004). Disaster or sustainability. Canadian Review of Sociology and Anthropology, 41, 1–18. Murphy, R. (2006a). Environmental realism: From apologetics to substance. In Nature and Culture. 1(2), 49–72. Murphy, R. (2006b). The challenge of disaster reduction. In Aaron M. McCright & Terry N. Clark (Eds.), Community and Ecology: Dynamics of Place, Sustainability, and Politics. Boston: JAI/Elsevier, Chapter 7. Murphy, R. (forthcoming). Leadership in disaster: Learning for a future under global climate change. Montreal and Kingston: McGill-Queen’s University Press. Navarro, F. (1998). An encounter with Althusser. Rethinking Marxism, 10(3), 93–98. Negri, A. (1995). Notes on the evolution of the thought of the later Althusser. In A. Callari & D.F. Ruccio (Eds.), Postmodernism and the Future of Marxist Theory (pp. 51–68). Hanover: Wesleyan University Press.

Bibliography 339 Nelson, A. (1990). Are economic kinds natural? In W. Savage (Ed.), Scientific theories: Minnesota studies in the philosophy of science, 14 (pp. 101–135). Minneapolis: University of Minnesota Press. Nelson, J.A. (2003). Once more, with feeling: Feminist economics and the ontological question. Feminist Economics, 9(1), 109–118. New, C. (1992). Agency, health and social survival. London: Taylor & Francis. New, C. (1998). Realism, deconstruction and the feminist standpoint. Journal for the Theory of Social Behaviour, 28(4), 349–372. New, C. (2004). Proceedings from IACR Conference ’04: Sex and gender: A critical realist approach. Cambridge: Cambridge University Press. Nietzsche, F. (1990). Beyond good and evil. London: Penguin. Norrie, A. (1993). Crime, reason and history: A critical introduction to criminal law. London: Weidenfeld and Nicolson. Norrie, A.W. (2005). Law and the beautiful soul. London: GlassHouse. Norris, C. (1987). Derrida. London: Fontana. Norris, C. (1997). Against relativism: Philosophy of science, deconstruction and critical theory. Oxford: Blackwell. Norris, C. (2000). Quantum theory and the flight from realism: Philosophical responses to quantum mechanics. London: Routledge. Nussbaum, M.C. (2000). Women and human development: The capabilities approach. Cambridge: Cambridge University Press. Nussbaum, M.C. (2001). Upheavals of thought: The intelligence of emotions. Cambridge: Cambridge University Press. Oakley, J. (1992). Morality and emotions. London: Routledge. Olssen, M. (2003). Structuralism, post-structuralism, neo-liberalism: Assessing Foucault’s legacy. Journal of Education Policy, 18(2), 189–202. O’Malley, P., Wier, L., & Shearing, C. (1997). Governmentality, criticism, politics. Economy and Society, 26(4), 501–517. O’Neil, O. (2002). Autonomy and trust in bioethics. Cambridge: Cambridge University Press. O’Neill, J. (1994). Humanism and nature. Radical Philosophy, 66, 27. Oliver-Smith, A. (1998). Global changes and the definition of disaster. In E.L. Quarantelli (Ed.), What is a disaster? (pp. 177–194). London: Routledge. Outhwaite, W. (1983). Toward a realist perspective. In G. Morgan (Ed.), Beyond method: Strategies of social research (pp. 321–330). Beverly Hills, CA: Sage. Outhwaite, W. (1987). New philosophies of social science: Realism, hermeneutics and critical theory. London: Macmillan. Outhwaite, W. (1987/1998). Realism and social science. In M.S. Archer et al. (Eds.), Critical realism: Essential readings, (pp. 282–296). London: Routledge. Ottawa Citizen. (1998, January 19). The great ice storm of January 1998, G5.

340 Bibliography Parsons, T. (1937). The structure of social action. New York: McGraw Hill. Parsons, T. (1951). The social system. London: Routledge. Parsons, T., & Shils, E. (Eds.) (1951). Toward a general theory of action. New York: Harper & Row. Patton, P. (2000). Deleuze and the political. New York: Routledge. Pauling, L., & Pauling, P. (1975). Chemistry. San Francisco: W.H. Freeman and Co. Pawson, R. (1989). A measure for measure: A manifesto for empirical sociology. London: Routledge. Pawson, R., & Tilley, N. (1997). Realistic evaluation. London: Sage. Pearce, F. (1988). The struggle for Foucault: Review essay on Michel Foucault. International Journal of the Sociology of Law, 16(2), 256–269. Pearce, F. (1989). The radical Durkheim. London: Unwin Hyman. Pearce, F. (1998). Gifts of blood: An aleatory materialist exploration of the confrontation between the Aztec/Mexica and the Spanish Conquistadores. International Sociological Association Conference, Montreal. Pearce, F. (2001). The radical Durkheim (2nd ed). Toronto: Canadian Scholars Press. Pearce, F. (2003). ‘Off with their heads’: Public executions with Klossowsi, Caillois, Foucault. Economy and Society, (32)1. Pearce, F. (2005). The Collège de Sociology. In George Ritzer (Ed.), Encyclopedia of Social Theory, v. 1. Thousand Oaks, CA: Sage. Pearce, F., & Dupont, D. (2001). Foucault contra Foucault: Re-reading the ‘Governmentality’ papers. Theoretical Criminology, 5(2), 123–158. Pearce, F., & Tombs, S. (1998). Toxic capitalism: Corporate crime and the chemical industry. Brookfield, VT: Ashgate. Pearce, F., & Woodiwiss, A. (2001). Reading Foucault as a realist. In J. López & G. Potter (Eds.), After postmodernism: An introduction to critical realism (pp. 51–62). New York: Athlone Press. Pels, D. (1997). Mixing metaphors: Politics or economics of knowledge? Theory and Society, 26(5), 685–717. Penrose, R. (2004). The road to reality. London: Jonathan Cape. Peregrine, P. (1995). The birth of the Gods and replications: Background to the data and codes. World Cultures, 9(2), 56–61. Peregrine, P. (1996). The birth of the Gods revisited: A partial replication of Guy Swanson’s (1960) cross-cultural study of religion. Cross Cultural Research, 30(2), 84–112. Perez, R. (1990). On an(archy) and schizoanalysis. New York: Semiotext(e). Perrow, C. (1984). Normal accidents. New York: Basic. Peter, F. (2003). Is critical realism as useful ontology for feminist economics? Feminist Economics, 9(1), 93–101.

Bibliography 341 Phillips, D. (1998). Blame it on the weather. Toronto: Key Porter. Pijl, K. van der. (1997). The making of an Atlantic ruling class. New York: Norton. Pilnick, A. (2002). What ‘most people’ do: Exploring the ethical implications of genetic counselling. New Genetics and Society, 21(3), 339–50. Pinch, T.J. & Bijker, W.E. (1984). The social construction of facts and artefacts. Social Studies in Science, 14, 399–441. Popper, K. (1959). The logic of scientific discovery. London: Hutchinson. Popper, K. (1969). Conjectures and refutations. London: Routledge and Kegan Paul. Porpora, D.V. (2002). Landscapes of the soul. Oxford: Oxford University Press. Potter, G. (1999). The bet: Truth in science, literature and everyday knowledges. Aldershot: Ashgate. Potter, G. (2000). For Bourdieu, against Alexander: Reality and reduction. Journal for the Theory of Social Behaviour, 30(2), 229–246. Potter, G. (2003). Critical realist strengths and weaknesses. Journal of Critical Realism (2)1. Potter, G. (2006). Reopening the wound: Against God and Bhaskar. Journal of Critical Realism, (5)1. Prado, C. (2006). Searle and Foucault on truth. Cambridge: Cambridge University Press. Price, M. (2002). The fall of the French monarchy: Louis XVI, Marie Antoinette and the baron de Breteuil. London: Macmillan. Prigogine, I., & Stengers, I. (1985). Order out of chaos. London: Fontana. Psillos, S. (1999) Scientific realism: How science tracks truth. London: Routledge. Purvis, T. (2002). Regulation, governance and the state. In M. MacNeil, N. Sargent and P. Swan (Eds.), Law, Regulation and Governance (pp. 28–53). New York: Oxford University Press. Purvis, T., and Hunt, A. (1993) Discourse, ideology, discourse, ideology. British Journal of Sociology, 44(3), 473–499. Quarantelli, E.L. (1986). Planning and management for the prevention and mitigation of natural disasters. Planning for crisis relief. Nagoya: UN. Quarantelli, E.L. (1998). Epilogue. In E.L. Quarantelli (Ed.), What is a disaster? (pp. 234–273). London: Routledge. Rabinow, P. (1988). Beyond ethnography: Anthropology as nominalism. Cultural Anthropology, 3(4), 355–364. Rabinow, P. & Rose, N. (2003). Introduction: Foucault today. In P. Rabinow & N. Rose (Eds.), The essential Foucault (pp. vii–xxxv). New York: New Press. Rajchman, J. (1985). Michel Foucault: The freedom of philosophy. New York: Columbia University Press.

342 Bibliography Ramsey, M. (1996). Human needs and the market. Aldershot: Avebury. Reale, G. (1980). The concept of first philosophy and the unity of the metaphysics of Aristotle. Albany: State University of New York Press. Region I. (1998). Interagency hazard mitigation team report, January 1998 ice storm, Maine, FEMA DR-1198–ME. Boston: FEMA. Resch, P. (1992). Althusser and the renewal of Marxist social theory. Berkeley: University of California Press. Reuten, G. (1993). The difficult labor of a theory of social value: Metaphors and systematic dialectics at the beginning of Marx’s Capital. In F. Moseley (Ed.), Marx’s method in Capital: A reexamination (pp. 89–113). Atlantic Highlands, NJ: Humanities Press. Reuten, G., & Williams, M. (1989). Value-form and the state. London: Routledge. Rieppel, O. (2004). Monophyly, paraphyly, and natural kinds. Bio. and Phil., 19, 1–23. Riesman, D. (1950). The lonely crowd. New Haven: Yale University Press. Roberts, J.M. (2001). Critical realism and the dialectic. British Journal of Sociology, 52(4), 667–685. Robinson, J. (1962). Economic philosophy. Harmondsworth: Penguin. Rorty, R. (1980). Philosophy and the mirror of nature. Princeton, NJ: Princeton University Press. Rosa, E. (1998). Metatheoretical foundations for post-normal risk. Journal of Risk Research, 1, 15–44. Rose, N. (1994). Expertise and the government of Conduct. Studies in Law, Politics and Society, 14, 359–397. Rose, N. (1999). Powers of freedom: Reframing political throught. Cambridge: Cambridge University Press. Rose, N., & Miller, P. (1992). Political power beyond the state: Problematics of government. British Journal of Sociology, 43(2), 173–205. Rosenberg, A. (1983). If economics isn’t science, what is it? Philosophical Forum, 14(3–4), 296–314. Rosenberg, C.E. (1999). Meanings policies and medicine: On the bioethical enterprise and history. Daedalus, 128(4), 27–46. Rothman, D.J. (1991). Strangers at the bedside. New York: Basic. Rustin, M. (1993). The good society and the inner world: Psychoanalysis, politics and culture. London: Verso. Rustin, M. (2004). Private solutions to public problems. Mimeo, Department of Cultural Studies, University of East London. Sandage, S. (2005). Born losers: A history of failure in America. Cambridge, MA: Harvard University Press.

Bibliography 343 Saussure, F. de. (1974). Course in general linguistics. London: Fontana. Sayer, A. (1981). Abstraction: A realist interpretation. Radical Philosophy, Summer, 6–15. Sayer, A. (1992). Method in social science: A realist approach (2nd ed.). London: Routledge. Sayer, A. (1997a). Essentialism, social constructionism, and beyond. Sociological Review, 45(3), 453–87. Sayer, A. (1997b). Critical realism and the limits to critical social science. Journal for the Theory of Social Behaviour, 27(49), 473–488. Sayer, A. (2000). Realism and social science. London: Sage. Sayer, A. (2001). Reply to Holmwood. Sociology, 35, 967–84. Sayer, A. (2006, forthcoming). Language and significance – or the importance of import: Implications for critical discourse analysis. Journal of Language and Politics. Schmidt, V. (2001). Oversocialised epistemology. Sociology, 35, 135–57. Searle, J. (1996). The philosophy of mind. Springfield, VA: Teaching Company. Sekine, T. (1997). An outline of the dialectic of capital (2 vols). London: Macmillan. Selvin, H.C. (1965). Durkheim’s Suicide: Further thoughts on a methodological classic. In Robert Nisbet (Ed.), Emile Durkheim. Engelwood Cliffs, NJ: Prentice Hall. Sennett, R. (1977). The fall of public man. New York: Norton. Sen, A. (1999). Development as freedom. Oxford: Oxford University Press. Sennett, R., & Cobb, J. (1977). The hidden injuries of class. Cambridge: Cambridge University Press. Sher, G. (2004). In search of a substantive theory of truth. Journal of Philosophy, 101, 5–36. Shilling, C., & Mellor, P.A. (2001). The sociological ambition. London: Sage. Sismondo, S. (2000). Island biogeography and the multiple domains of models. Biology and Philosophy, 15, 239–258. Smandych, R. (Ed.). (1999). Governable places: Readings on governmentality and crime control. Aldershot: Ashgate/Dartmouth. Smith, A. (1759 /1984). The theory of moral sentiments. Indianapolis: Liberty Fund. Sohn-Rethel, A. (1978). Intellectual and manual labour. London: Macmillan. Soper, K. (1981). On human needs. Brighton: Harvester. Soper, K. (1990). Troubled pleasures. London: Verso. Soper, K. (1995). The concept of nature. Oxford: Blackwell. Spivak, G. (1988). Can the subaltern speak? In C. Nelson & L. Grossberg (Eds.), Marxism and the interpretation of culture (pp. 271–313). Urbana: University of Illinois Press.

344 Bibliography Sprinker, M. (1980). The use and abuse of Foucault. Humanities in Society, 3(1), 1–22. Stenson, K. (1998). Beyond histories of the present. Economy and Society, 27(4), 333–352. Stenson, K. (2001). Reconstructing the government of crime. In G. Wickham and G. Pavlich (Eds.), Rethinking law, society and governance: Foucault’s bequest (pp. 93–108). Portland: Hart Publishing. Stevens, T. (2000). Bioethics in America: Origins and cultural politics. Baltimore: Johns Hopkins University Press. Stones, R. (1996). Sociological reasoning: Towards post-modern sociology. London: Macmillan. Strawbridge, D. (1982). Althusser’s theory of ideology and Durkheim’s theory of religion: An examination of some striking parallels. Sociological Review, 30, 125–40. Swanson, G. (1964). The birth of the Gods: The origin of primitive beliefs. Ann Arbor: University of Michigan Press. Taylor, N. (2000). Abstract labour and social mediation in Marxian value theory. Unpublished bachelor’s thesis, Murdoch University School of Economics, Western Australia. Taylor, S. (1982). Durkheim and the Study of Suicide. London: Macmillan. Terray, E. (1995). An encounter: Althusser and Machiavelli. In A. Callari & D.F. Ruccio (Eds.). Postmodernism and the future of Marxist theory (pp. 257–277). Hanover: Wesleyan University Press. Tester, K. (1991). Animals and society. London: Routledge. Thomas, W., & Thomas, D. (1928). The child in America. New York: Knopf. Tolman, E. (1951). A psychological model. In T. Parson & E. Shils (Eds.), Toward a general theory of action. New York: Harper & Row. Toulmin, S. (1982). How medicine saved the life of ethics. Perspectives in Biology and Medicine, 25(4), 736–750. Toulmin, S. (1988). Medical ethic in its American context: A historical survey. In D. Callahan & G.R. Dustan (Eds.), Biomedical ethics: An anglo-American dialogue. New York: New York Academy of Sciences. Traugott, M. (1978). Émile Durkheim on institutional analysis. Chicago: University of Chicago Press. Treifl, J. (1994). From atoms to quarks: An introduction to the strange world of particle physics. New York: Anchor Books. Turkle, S. (1978). Psychoanalytic politics: Freud’s French revolution. New York: Basic. Turner, B. (1978). Man-made disasters. London: Wykeham. Turner, J. (1994). The failure of sociology to institutionalize cumulative theorizing. J. Hage (Ed.), Formal theory in sociology: Opportunity or pitfall. Albany, NY: SUNY Press.

Bibliography 345 Vaughan, D. (1996). The challenger launch decision. Chicago: University of Chicago Press. Veyne, P. (1971/1997). Foucault revolutionizes history. In Foucault and his interlocutors (pp. 146–182). Chicago: University of Chicago Press. Wachterhauser, B. (1986). Must we be what we say? Gadamer on truth in the human sciences. In B. Wachterhauser (Ed.). Hermeneutics and modern philosophy. Albany: State University of New York Press. Wachterhauser, B. (2002). Getting it right: Relativism, realism and truth. In R. Dostal (Ed.), The Cambridge companion to Gadamer. Cambridge: Cambridge University Press. Wagner, A. (1879). Allgemeine oder Theoretische Volkswirthschaftslehre. Leipzig: Erster Theil. Wagner, G. (2001). The character concept in evolutionary biology. London: Academic Press. Warnke, G. (2003). Rorty’s democratic hermeneutics. In C. Guignon, & D.R. Hiley (Eds.), Richard Rorty. Cambridge: Cambridge University Press. Weber, M. (1978). Economy and society, v. 1. Berkeley: University of California Press. Webster, P.J., et al. (2005). Changes in tropical cyclone number, duration, and intensity in a warming environment. Science, (309)5742, 1844–1846. Weinberg, S. (1994). Dreams of a final theory. New York: Vintage. Wendt, A. (1999). Social theory of international politics. Cambridge: Cambridge University Press. Whyte, W.F. (1963). The organization man. Harmondsworth: Penguin. Wickham, G., & Paulich, G. (Eds.). (2001) Rethinking law, society and governance: Foucault’s bequest. Portland: Hart Publishing. Wilden, A. (1980). System and structure: Essays in communication and exchange. London: Tavistock. Will, D. (1986). Psychoanalysis and the new philosophy of science. International Review of Psychoanalysis, 13. Willis, P.E. (1977). Learning to labour. London: Saxon House. Winch, P. (1958). The idea of social science. New York: Routledge and Kegan Paul. Wolf, S.M. (Ed.). (1996). Feminism and bioethics: Beyond reproduction. Oxford: Oxford University Press. Wolff, A. (1988). ‘The cultural sources of the Reagan revolution.’ In W. Kymlicka & J. Mathews (Eds.), The Reagan Revolution. Chicago: Dorsey Press. Wolpe, P.R. (1998). The triumph of autonomy in American bioethics: A sociological view. In R. DeVries & J. Subedi (Eds.), Bioethics and society: Constructing the ethical enterprise. Upper Saddle River, NJ: Prentice Hall.

346 Bibliography Woodiwiss, A. (1990). Social theory after post-modernism: Rethinking production, law, and class. London: Pluto Press. Woodiwiss, A. (1993). Postmodernity USA: The crisis of social modernism in the postwar United States. London: Sage. Woodiwiss, A. (1998). Globalisation, human rights and labour law in Pacific Asia. Cambridge: Cambridge University Press. Woodiwiss, A. (2001). The visual in social theory. London: Athlone Press/ Continuum. Woodiwiss, A. (2005). Scoping the social: An introduction to the practice of social theory. Maidenhead: Open University Press. Young, J., & Matthews, R. (1992a). Issues in realist criminology. London: Sage. Young, J., & Matthews, R. (1992b). Rethinking criminology: the realist debate. London: Sage. Zizek, S. (1991). Looking awry: An introduction to Jacques Lacan through popular culture. Cambridge, MA: MIT Press. Zizek, S. (2004). Organs without bodies. New York: Routledge Zussman, R. (1992). Intensive care: Medical ethics and the medical profession. Chicago: University of Chicago Press.

Name Index

Archer, M. xi, 48–9, 61, 75–7, 81–91, 245 Aristotle 25, 26, 39, 202, 204, 205–6, 208, 221, 223 Badiou, A. 274 Bataille, G. 63 Bauchamp, T.L. 170–1 Baudrillard, J. 107–119 Beck, U. 155 Bell, D. 105 Benton, T. xvi, 8, 13, 47–8, 53, 61, 142, 148, 166 Bhaskar, R. x–xi, xiii–xv, xvii–xviii, 3, 7–8, 12–14, 21–2, 24–6, 29–36, 38–51, 53, 55, 59–62, 65, 77–84, 86–7, 91–5, 99, 117–18, 120, 124–5, 127–30, 132– 3, 135, 138–9, 162– 3, 165–6, 177, 179–80, 192–4, 199, 213–14, 225–6, 229, 231, 234, 237, 242–3, 254–7, 260, 273– 4, 276, 281–2, 295, 297, 301, 312– 13 Blaikie, N. 15 Bohm, D. 62 Bourdieu, P. 5, 22, 79, 100, 134, 140, 269–70, 310

Boyd, R. 26, 209, 212, 214, 217 Butler, J. 119, 274 Cain, M. 119 Callahan, D. 168 Callon, N. 143, 148–50 Carter, B. 19 Chalmers, A. 166 Childress, J. 170–1 Chowdos, H. 27–8 Collier, A. xv, xix, 26, 31, 35, 42, 44, 61, 82–5, 91, 120, 127, 165, 179, 185, 248, 250 Collins, H. 143, 157–8, Craib, I. xvi, 153, 257 Curtis, B. 27 Deleuze, G. 24, 117–19, 125, 129–36, 139–41, 295 Denis, H. 139 Derrida, J. 23, 100, 119–20, 129, 140, 188 Dickens, P. xvi, xix, 146, 155, 158 Dombrowsky, W. 144 Durkheim, E. xvi–xvii, 20, 22, 31, 39, 43, 53–9, 61–3, 79, 90, 97–8, 102– 103

348 Index Ehrbar, H.G. xvi, 25–6, 235, 239 Engelskirchen, H. xvi, 25–6, 222

Luhmann, N. 40, 105 Lyotard, J.F. 23, 99, 119, 140

Fairclough, N. 118, 125 Fayerabend, P.K. ix, 11–12, 14 Foucault, M. xvi, xviii, 5, 23–4, 27–8, 61, 79, 100–4, 109, 114–25, 129, 131, 167, 251, 258–61, 265–95, 302, 309 Fox, R.C. 168, 171 Frauley J. xvii, 27, 258

Marx, K. xi–xiv, xvi–xix, 5, 12, 20, 25– 6, 43, 45–6, 52–3, 75, 79–80, 90, 94– 5, 102–3, 129, 164, 179–80, 182–3, 185–7, 189–93, 196, 198–200, 202, 210, 215, 217–31, 264, 266, 274, 281, 286 Maturana, H. 40 Murphy, R. xix, 22, 24, 142

Gadamer, H.G. 5, 27–9, 296–314 Giddens, A. 105, 108, 110–13, 244 Grundmann, R. 156

New, C. xvi Nietzsche, F. 76, 83, 91, 288, 295 Norris, C. 62

Habermas, J. 99, 105, 175–6, 251, 257, 311 Hacking, I. 213, 217 Harré, R. x, 29 Hempel, C. 8–9 Hume, D. v, 7–9, 14, 31, 250 Hunt, A. 27, 296

O’Malley, P. 259 Oliver–Smith, A. 147

Jonsen, A.R. 170 Keat, R. x, 8, 13, 61 Knor-Cetina, K.D. 72 Kuhn, T. ix, 10–12, 14

Parsons, T. 103–104, 106, 109, 111 Pauling, L., and Pauling, P. 39 Pearce, F. xi, xvi, xix, 21, 30, 55, 61, 118, 122, 260, 263, 265, 295 Perrow, C. 147, 149 Popper, K. 9, 11, 78 Porpora, D.V. 82–5, 91, 96 Potter, G. xiii, 4, 21–2, 61, 74, 120 Prado, C. 118, 121–2 Quarantelli, E.L. 144

Lacan, J. 24, 117–18, 120, 125, 135–8, 141 Latour, B. 68, 145, 148–9, 150, 152, 155–6, 166 Lawson, T. 177 Layder, D. 18 Lecourt, D. 284 Lewis, J. 144, 159 López, J.J. xvi, 4, 22, 24, 25, 61, 120, 162, 179, 268

Rabinow, P. 274 Riesman, D. 105 Rorty, R. 41–2, 305, 315 Rose, N. 259–61, 265, 272, 274 Sayer, A. xiv–xv, 13, 15, 20, 25–7, 49, 51, 61, 120–2, 127–9, 132, 139–40, 146, 164, 177, 240, 261–2, 272 Searle, J. 223

Index 349 Sekine, T. 185 Sismondo, S. 21, 22, 64 Smith, A. 190–1, 220, 249, 257 Soper, K. xvi, xix, 61 Swanson, G. 56–7 Swazey, J.R. 168

Varela, F. 40

Turner, B. 147, 149, 154

Weber, M. xvi–xvii, xix, 43, 58–59, 102–3, 193, 245–6 Wilden, A. 118, 125–28, 130, 133, 135, 140 Woodiwiss, A. xvi, 22–3, 61, 97, 118, 122, 262–3, 267, 295

Urry, J. x, 8, 13, 61

Zizek, S. 136, 289