Handbook of Potentiality 9402412859, 9789402412857

Citation preview

Kristina Engelhard Michael Quante Editors

Handbook of Potentiality

Handbook of Potentiality

Kristina Engelhard Michael Quante •

Editors

Handbook of Potentiality

123

Editors Kristina Engelhard Technische Universität Dortmund Dortmund Germany

Michael Quante Westfälische Wilhelms-Universität Münster Germany

ISBN 978-94-024-1285-7 ISBN 978-94-024-1287-1 https://doi.org/10.1007/978-94-024-1287-1

(eBook)

Library of Congress Control Number: 2017963989 Chapter “Potentiality in Biology” has a permission form for re-use of figures. © Springer Science+Business Media B.V., part of Springer Nature 2018 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Printed on acid-free paper This Springer imprint is published by the registered company Springer Science+Business Media B.V. part of Springer Nature The registered company address is: Van Godewijckstraat 30, 3311 GX Dordrecht, The Netherlands

Acknowledgements

We would like to thank Anna May Blundell for her profound proofreading and translation work. Without her help, we would not have been able to complete this project.

v

Contents

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Kristina Engelhard Part I

1

The Concept of Potentiality in the History of Philosophy—Aristotle

Potentiality in Aristotle’s Metaphysics . . . . . . . . . . . . . . . . . . . . . . . . . . Anna Marmodoro

15

Potentiality in Aristotle’s Physics and Biology . . . . . . . . . . . . . . . . . . . . Stephen Makin

45

Potentiality in Aristotle’s Psychology and Ethics . . . . . . . . . . . . . . . . . . Frans A. J. de Haas

71

Part II

The Concept of Potentiality in the History of Philosophy—Medieval Philosophy

Potentiality in Classical Arabic Thought . . . . . . . . . . . . . . . . . . . . . . . . Taneli Kukkonen

95

Potentialities in the Late Middle Ages—The Latin Tradition . . . . . . . . . 123 Stephan Schmid Part III

The Concept of Potentiality in the History of Philosophy—Early Modern Philosophy

Potentiality in Rationalism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157 Michael-Thomas Liske Potentiality in British Empiricism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199 Katia Saporiti

vii

viii

Part IV

Contents

The Metaphysics of Potentials

Real Potential . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229 Jennifer McKitrick Powers and Potentiality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261 Stephen Mumford and Rani Lill Anjum From Potentiality to Possibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279 Barbara Vetter Part V

Potentiality in Specific Fields of Philosophy

Potentialities in the Philosophy of Mind . . . . . . . . . . . . . . . . . . . . . . . . . 305 Frank Hofmann Potentiality in Bioethics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327 Marco Stier Part VI

Potentiality in the Sciences

Potentiality in Physics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 353 Max Kistler Aspects of the Concept of Potentiality in Chemistry . . . . . . . . . . . . . . . . 375 Paul Needham and Robin Findlay Hendry Potentiality in Biology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 401 Andreas Hüttemann and Marie I. Kaiser Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 429

Editors and Contributors

About the Editors Kristina Engelhard is a postdoc Research Fellow at the DFG-funded research group “Inductive Metaphysics”, located at the Technical University of Dortmund, Germany. Her current research is on dispositions, laws of nature, methods in metaphysics, metametaphysics, classical German Philosophy and early modern philosophy. Her publications include „Das Einfache und die Materie. Untersuchungen zu Kants Antinomie der Teilung“ (De Gruyter 2005), “Categories and the ontology of powers. A semi-dualist account of pandispositionalism” (in: A. Marmodoro (ed.), “The Metaphysics of Powers. Their Grounding and their Manifestations” (Routledge 2010), „Können Dispositionen das Realismus-Problem des transzendentalen Idealismus lösen?“ (in: M. Egger (ed.), „Philosophie nach Kant. Neue Wege zum Verständnis von Kants Transzendental- und Moralphilosophie“ (De Gruyter 2014). Michael Quante is Professor for Practical Philosophy at the Department of Philosophy at Westfälische Wilhelms-University of Münster, Germany. His research areas are action theory, philosophy of person, ethics and applied ethics. His most important publications are “Hegel’s Concept of Action” (Cambridge University Press 2004), “Person” (de Gruyter 2012), “Personal Identity as a Principle of Biomedical Ethics” (Springer 2016).

Contributors Rani Lill Anjum School of Economics and Business, Norwegian University of Life Sciences, Ås, Norway Kristina Engelhard Technische Universität Dortmund, Dortmund, Germany Frans A. J. de Haas Ancient and Medieval Philosophy, Leiden University, Leiden, The Netherlands Robin Findlay Hendry Department of Philosophy, Durham University, Durham, UK Frank Hofmann University of Luxembourg, Esch-Belval, Luxembourg Andreas Hüttemann University of Cologne, Cologne, Germany Marie I. Kaiser University of Bielefeld, Bielefeld, Germany

ix

x

Editors and Contributors

Max Kistler IHPST, Université Paris 1 Panthéon Sorbonne & CNRS, Paris, France Taneli Kukkonen New York University Abu Dhabi, Abu Dhabi, United Arab Emirates Michael-Thomas Liske University of Passau, Passau, Germany Stephen Makin University of Sheffield, Sheffield, UK Anna Marmodoro Department of Philosophy, Durham University, Durham, UK; Faculty of Philosophy, University of Oxford, Oxford, UK Jennifer McKitrick Department of Philosophy, University of Nebraska, Lincoln, USA Stephen Mumford Department of Philosophy, Durham University, Durham, UK Paul Needham Department of Philosophy, University of Stockholm, Stockholm, Sweden Katia Saporiti University of Zurich, Zürich, Switzerland Stephan Schmid University of Hamburg, Hamburg, Germany Marco Stier Institute for Medical Ethics, History and Philosophy of Medicine, University of Münster, Münster, Germany Barbara Vetter Freie Universität Berlin, Berlin, Germany

Introduction Kristina Engelhard

“The world is full of threats and promises” (Goodman 1983, 40); not only do we want to know which threats and promises there presently are, but also which ones there will be—maybe to a certain degree of certainty—in the future. If something is dangerous, it is disposed to cause harm under certain conditions at the present moment. However, if something has the potential to be or to become dangerous, we often want to say that it is not dangerous at present, but will become dangerous in the future under certain conditions. Physicians say that a certain virus has a high human-pathogenic potential if it is a virus that normally affects animals only but a small mutation could turn it into a variant of the virus that may very easily affect humans.1 Potentials may also be promising: a violin teacher might be convinced that her six year old pupil Helen has the potential to play the violin on a professional level and teaches her accordingly. Investors might be convinced that Commerce Inc. has the potential to be or become a global player on the market and invest accordingly. Potentials may also be ethically relevant. According to the so

1

E.g. some new types of avian influenza viruses (HN-viruses) are suspected of maybe turning into very harmful variants of HN-viruses if they undergo mutations that allow them to propagate under lower body temperatures than they do at present. Birds have a body temperature of about 38–42 °C, while humans have around 37 °C, e.g. H10N8. If H10N8-viruses were to undergo a mutation that would allow them to propagate at 37 °C an infection from human to human would be possible and hence it would be possible that avian flue would cause a pandemic (Nunes-Alves 2014). However, sometimes physicians speak of a human pathogenic potential also, if certain bacteria or viruses are presently disposed to cause severe illness if they infect humans. K. Engelhard (&) Technische Universität Dortmund, Dortmund, Germany e-mail: [email protected] © Springer Science+Business Media B.V., part of Springer Nature 2018 K. Engelhard and M. Quante (eds.), Handbook of Potentiality, https://doi.org/10.1007/978-94-024-1287-1_1

1

2

K. Engelhard

called ‘argument from potential’ human embryos and fetuses have the potential for being a person and hence are at present bearers of full or some lower moral status.2 Potentials are at least practically relevant: we discriminate entities by their alleged potentials and invest all kinds of resources for dealing with potentials: we either promote the manifestation of a potential if we think of its manifestation as something positive, or we suppress it, if we evaluate the manifestation of a potential as something negative. The term ‘potential’ and all its forms differ in meaning. Some uses are predicative and hence at least seem to ascribe a kind of property to an object. A potential then is a property that an object may instantiate that involves either the mere possibility of it acquiring some other property or the disposition to acquire other properties. Or it may mean some property conferring a potential to its bearer due to its identity. Some uses, most of all as adverbs or adjectives, as in ‘Hepatitis is a potentially fatal disease’, mean some form of modality: either an epistemic or a metaphysical possibility.3 In this example we normally want to say that if somebody is infected with hepatitis it is possible that he will die of hepatitis, if certain conditions obtain. We can understand this case as referring to the disposition of the hepatitis virus to kill human organisms under certain conditions. There also seem to be slight differences in use between modern languages even of common origin like English and German. Whereas in English it seems more common to take the adverbial e.g. ‘x has the potential to be dangerous’ in the modal sense, normally meaning that x is possibly dangerous at present; whereas in German there is the liability to use the word ‘Potenzial’ only in cases that indicate the ascription of a property that is responsible for acquiring another causally relevant property in the future. By ‘x hat das Potenzial, gefährlich zu sein’, a speaker of German usually wants to say that x has some property that is responsible for it to turn into something dangerous in the future or if it is placed in different conditions x will be dangerous. So, the use of ‘Potenzial’ seems to be more restrictive in German than ‘potential’ in English; although there might be examples of the contrary also. This might also be a source of different intuitions with respect to the concept of potentiality in philosophical debates. However, it is not at all clear what potentials are metaphysically. What do we refer to when we attribute a potential to something? What is it that a bearer of a potential has in contrast to something that lacks the potential, e.g. the potential for being a professional violinist? What are potentials, what is potentiality? And what are the truth-makers of propositions containing some form of potentiality vocabulary? And which role do potentials play in the different fields of philosophy and in 2

According to some philosophers the potential for personhood is relevant for ethical problems of stem-cell research, stem-cell therapy, abortion, prenatal diagnostics and euthanasia. However, this is controversial (cf. Stier, this volume). 3 Questions of reduction, e.g. whether the phrase also refers to a property, e.g. whether the Hepatitis pathogen has the potential to cause the death of the infected organism, will not be discussed. This is relevant for accounts that try to reduce possibility to dispositions resp. potentials (Borghini and Williams 2008; Vetter 2015).

Introduction

3

the sciences. These metaphysical questions are the main issues of this book. It is not concerned with issues of language and its use. The basic idea of this volume is to give an overview of the metaphysics of potentiality on the one hand in the history of philosophy and on the other hand in several fields of contemporary philosophy and in the sciences. The terms ‘potential’ and ‘potentiality’ reach back to ancient philosophy; the most influential framing of these terms to the present has been given by Aristotle. In Aristotle the concept of ‘dynamis’, as a complementary term to ‘energeia’, plays a central role in his metaphysics. It was primarily introduced to explain phenomena of identity and change. The ancient Greek term ‘dynamis’ allows at least for two different translations into modern languages like English and German—mirroring the options for understanding the truth-maker of propositions containing potentiality vocabulary today: there is a merely modal sense by which ‘dynamis’ can be translated by possibility or probability. And there is a predicative sense by which ‘dynamis’ means ‘capacity’, ‘faculty’, ‘disposition’ or ‘power’. These two senses of ‘dynamis’, the merely modal sense of possibility and the metaphysical sense of some property with modal force have already been attributed to the ancient term dynamis by Aristotle himself in book 5 of the Metaphysics (1019a f.). Aristotle also forms the concept through examples: the acorn that has a potential to become an oak tree, wood which has the potential to become a casket, the potential of a human sperm to become an adult human being, or the potentiality of children to develop properties inherited by their parents. However, Aristotle does not distinguish in principle between potentials and other kinds of dispositional properties, common natural dispositions—as fragility, malleability, heatability, freezability—or cognitive capacities of animals and humans—perception or epistemic capacities—or ethical capacities—like the virtues as capacities to act in a certain way. He does not seem to draw a principled distinction between examples of potentials in the narrow sense and common dispositions as the articles in this volume exhibit. Hence, a further problem that can be traced back to Aristotle concerns the connection between dispositions and potentials. While it is a common opinion that potentials are at least a subclass of dispositions, it remains an open issue whether potentials just are dispositions, i.e. they are simply identical with dispositions, or whether there are features of potentials that make them a decisive subclass of dispositions. The acorn that has a potential to become an oak tree, wood which has the potential to become a casket, the potential of a human sperm to become an adult human being can be taken to be potentials in the narrow sense, while potentials in the wide sense include common examples of dispositions, like elasticity, flammability etc. The articles in this volume take different stances toward this question. Some authors argue that potentials are synonymous with dispositions, others argue for the claim that potentials are grounded in powers, i.e. causally efficacious dispositional properties (Mumford/Anjum). Hüttemann/Kaiser think that there are features of potentials, in this case in biology, that reveal that there are characteristics of dispositions in general that are easily overlooked when only taking into account the common examples of dispositions like fragility, flammability, elasticity, mass or

4

K. Engelhard

charge. These features are connected with the complexity of the manifestation process of potentials. Biological dispositions, according to Hüttemann/Kaiser, manifest by a process that has several more or less clearly defined stages, every step of which involves various dispositions in turn and which has certain starting conditions and termination conditions. Each step is also dependent on very specific environmental conditions. The manifestation of a biological disposition or potential involves changes of the entity of the potential. Makin finds in Aristotle the idea that potentials are not reducible to the dispositions the complex manifestation process breaks up into; a potential is not only a term for the collection of dispositions that participate in the complex manifestation process, but that there has to be some power on top directing the manifestation process to its specific goal because each disposition is multi-track. Other authors also deal with one or several of these features as candidates for a discriminating criterion for potentials—McKitrick, Mumford/Anjum, Hofmann, Kistler, e.a. McKitrick takes potentials to be dispositions to acquire a new property or to become a member of a certain kind. Another idea is to take them as second-order dispositions, i.e. dispositions to acquire yet further dispositions or as iterated dispositions, e.g. Vetter. Kukkonen shows that Avicenna held a similar idea in differentiating between proximate and remote potentials, with remote potentials being potentials the manifestation of which affords several steps that have to be taken until the final goal can be reached. However, all authors conform in the diagnosis that it is not possible to make the distinction precise. Today, the terms ‘potential’ and ‘potentiality’ are used in a wide variety of senses in different areas of discourse, in philosophy, the empirical sciences and in language of everyday life. These different senses of the term ‘potential’ in philosophy are reflected by the contributions of this book. One aim of this book is on the one hand to highlight the historical sources of this term and its interpretation within those strands in the history of philosophy that are relevant for the evolution of the concept of potentiality and potential. And on the other hand to show how the concept is framed in the different fields of contemporary metaphysics including the metaphysics of the sciences. Since Aristotle has most influentially coined the term in ancient philosophy the first chapter of this volume comprises three articles on Aristotle’s thought on the concept of dynamis in different fields of his philosophy, first in metaphysics, second in his philosophy of physics and biology and third his theory of the soul and his ethics. In the second chapter the volume is concerned with the tradition of the concept in the Latin and Arabic tradition of Aristotelian thought. The third chapter of the book is dedicated to the views on the concept of potential and potentiality in early modern philosophy. Here the concept is interpreted and evaluated along contrary lines in the empiricist and rationalist movements of that period. The following parts concern the use of the term in contemporary metaphysics, in specific fields of philosophy as the philosophy of mind, and bioethics. The last chapter is about the concept of potentiality in the sciences, in physics, chemistry, and in biology.

Introduction

5

In her article “Potentiality in Aristotle’s metaphysics” Anna Marmodoro investigates Aristotle’s use of the concept of ‘dynamis’ and shows, that his views on dynamis and its complement energeia should be understood as a metaphysics of powers. More specifically she points out that it is similar to contemporary pure powers theories, i.e. powers need not have a causal or categorical base and she thinks that Aristotle held a dispositional monism, i.e. the view that at least all fundamental properties are dispositional or they are powers. She delineates Aristotle’s theory according to which the manifestation of powers needs complementary powers as manifestation conditions and that his model of manifestation is similar to a mutual manifestation model according to which powers are not triggered to become manifest, such that the trigger is just an activator of the power, but does not itself participate in the manifestation itself; on the contrary the manifestation of powers comes about by contact such that the manifestation is an ensemble performance of all participating powers. She reconstructs Aristotle’s views on powers in detail and shows that Aristotle’s theory even makes a unique contribution to the contemporary landscape of theories of powers: it establishes a model of powers according to which a power is in a sense identical with its manifestation. Powers are monadic non-relational properties. The power is wholly present when it is manifesting. The manifestation of a power is not a transition from one fact containing property F say to some numerically different fact containing property G say. But it is the transition of the very same property from one status to another, from a non-actualised state to its actualised state. In his physics and biology Aristotle makes use of his metaphysical framework of potentials for explaining natural phenomena. However, Aristotle’s model of potentials has been criticised in early modern philosophy. In his article on “Potentiality in Aristotle’s physics and biology” Stephen Makin argues that this critique is not eligible and analyses the structure of physical and biological potentials in detail. He differentiates between two models of the transition from an inactive, dormant potential to its exercise. One model is binary, it takes the exercise of a potential to be an all or nothing affair. This model seems appropriate for those potentials the manifestation of which needs ideal manifestation conditions to obtain. Once the ideal conditions obtain, the potential exercises. The second is scalar, it takes the transition to be gradable. Now, Makin argues that in most cases of biological or physical potentials exercising in Aristotle the second model seems more adequate. Though, particularly the model of scalar potentials is open for the objection that potentials are not explicative because making reference to them to explain some fact that x F’s is nothing else than saying that x F’s because x can F, because the identity of the potential is fixed with reference to its manifestation only. This objection is not pressing for binary potentials, as Makin argues, because their identity is also fixed by their ideal manifestation conditions the identity of which can be fixed independently of the potential. It is however particularly pressing for scalar potentials. In Aristotle this objection can be rebutted because the manifestation of a potential breaks down into a series of underlying processes that are adequately taken to fulfil the binary model of potentiality manifestation that is not subject to this objection. Nevertheless the manifestation of the process has to be

6

K. Engelhard

governed by the scalar potential, because it is necessary to direct the underlying partial processes to a specific outcome. Frans de Haas shows in his article “Potentiality in Aristotle’s theory of the soul and his ethics” that Aristotle’s views on potentiality stem from distinctions that Plato made in his ethics. The potentials of the soul are bound up with the capacity to acquire knowledge that is in turn a potential. As de Haas shows, according to Aristotle’s theory two series are bound up with potentials: a series from not yet having, but capable of acquiring to having acquired and a series from possessing but not using to using. He then explores the various potentials of the soul, how they are interrelated and how this relates to Aristotle’s virtue ethics. Aristotle’s theory of a hierarchy of living beings from primitive animals up to the human being is structured by the potentials that these beings have. The human being has cognitive potentials the activation of which enables it to know by learning. In Aristotle the process of learning is a natural process, because it is a final goal of the human being to become a being capable of knowing. Furthermore, the human being has potentials of character and intellect, the potential to acquire virtues, a disposition to act in a way that enables it to achieve a happy life. The acquisition and exercise of the virtues is the basis of Aristotle’s ethics. Taneli Kukkonen starts his investigation of the interpretation of the terms ‘potential’ and ‘potentiality’ in the Arabic Aristotelian tradition with the early Arabic thinkers of this movement who mainly took up the Aristotelian conception without reframing it in any way. But then Avicenna and Averroes made innovative contributions to the Aristotelian framework. Their views were concerned with reconciling the Aristotelian metaphysics of dynamis and energeia with Islamic theological doctrines, first of all the omnipotence of God; hence it is God who bestows his creatures with powers and capacities. Avicenna understands the potential to have a lesser degree of being than the exercising of it. Nevertheless, he emphasises that the potential is the condition on which processes of change take place. This condition is prime matter. He distinguishes between proximate and remote potentialities; the sperm has the remote potentiality to turn into a man because it has to pass the state of boyhood. A proximate potential can immediately manifest. Averroes on the other side was more concerned to defend orthodox Aristotelianism against those thinkers who wanted to develop it further. The concept of potential is most central in his struggle: Averroes understands Aristotle to think that the identity of substances is fixed by their potentials. Since the substances are dependent on their potentials and since the potentials are dependent on their actuality and since the full actuality is only with God—as Averroes takes it—all substances depend on God. Stephan Schmidt shows on the basis of Aquinas, Scotus and Ockham that during the medieval period in western philosophy the Aristotelian concept of potentiality was transferred from natural substances to the rational potentialities of God. The concept of natural potentials lost explanatory power while the concept of the rational potentials of God where introduced to overtake its explicative function. As a result in the early modern period the concept of potentiality was dismissed by Descartes and others because potentials were supposed to be epistemically

Introduction

7

inaccessible and explanatorily idle. The concept of potentiality was introduced to western philosophy through the Arabic thinkers who made Aristotelian thought initially accessible. In Aquinas potentials of natural substances explain natural possibility, contingency and teleology in nature. Following Aristotle Aquinas differentiates between natural and logical possibility. Potentials are only apt to explain the first while logical possibility concerns the compatibility of concepts. According to Schmidt Aquinas has a normative concept of contingency grounded in fallible powers, i.e. powers the manifestation of which can be interfered with even if the manifestations conditions obtain. Scotus and Ockham however explain natural phenomena with reference to the rational capacities of God alone. Scotus’s theory of modality takes possibility to follow from logical consistency only. In his article on potentiality in rationalism Michael-Thomas Liske shows which theorems led most rationalists, esp. Descartes and Spinoza, to dismiss the concept of potentiality. The main reason for these two thinkers is their view of the relation between time and substance. Descartes dismisses potentiality due to his atomism with respect to temporal existence. This atomism leads him to reject a temporally persisting substance and hence the view of a substance realising its dispositions or potentials while persisting in time. Spinoza’s monism of substance leaves no room for potentials of individual substances besides the one omniferous substance. And since this one substance is pure actuality, potentials are merely a descriptive affair of phenomena. Liske then points out why and how Leibniz, contrary to Descartes and Spinoza, retained the concept of potentiality. However the specific determinism Leibniz in effect is committed to, according to Liske, is contrary to at least traditional features of potentiality, including the metaphysical possibility of non-realised potentials. On the contrary, a Leibnizian monad necessarily realises all its potentials, due to its identity and also all a substance’s future states are enclosed within its present state; there is no substantial change within a monad, which is contrary to common concepts of potentials, as Liske points out. In her inquiry of potentiality in the British empiricists Katia Saporiti differentiates the views of the central thinkers Locke, Berkley and Hume. Although the common view is that unmanifested potentials are not perceivable and though this excludes them from being possible objects of knowledge in an empiricist framework, Saporiti shows that Locke accepts dispositional properties. Due to the close relation between potentials and dispositions this is relevant for potentials in the same way. Because Locke thinks that although we have no direct access to substances we can be justified to infer the existence of substances and think of their perceivable qualities as powers. Locke even assumes that we have indirect perceptual access to the powers of substances by their manifestations. Berkley on the contrary dismisses the concept of power in nature as vacuous, but accepts it in the mental realm. Only spirits can be causally efficacious and hence can have powers. Going even further Hume does not accept dispositions or potentials in his ontology at all. Basically it is his views on causation that goes against the acceptance of potentials: Hume’s analysis of the content of our perceptions reveals—according to him—that there is no objective trace of the causal production of anything. Since sensation and reflection are the only possible sources of knowledge we cannot

8

K. Engelhard

know anything about necessary connections; we have no knowledge about the causes of change in nature. The idea of necessity originates from reflection. Hence Hume thinks that the ascription of powers to objects in nature is not based on powers in objects but on the constant conjunction of similar objects in the past. Hume’s epistemic worries and probably also Kant’s critique of traditional metaphysics pushed the concept of potentiality into the background of philosophical thinking during the late 18th and 19th century. It, however, reappeared in contemporary philosophy a few decades ago in the context of metaphysical debates about the nature of properties, about what the laws of nature are. The second part of the book is concerned with the role the concept of potentiality plays with respect to these debates. In her contribution “Real Potential” Jennifer McKitrick takes a realist position of potentials: she defends the thesis that the basis of something’s having a potential is that this thing has a dispositional property. Since McKitrick is also realist with respect to dispositional properties this means that if something has a potential it has a certain kind of property, it can instantiate its dispositional property without manifesting it, because it takes certain conditions for its dispositional properties to become manifest. McKitrick argues for the causal efficacy of potentials qua dispositional properties. The contrary position to realism of potentials is anti-realism of potentials according to which the attribution of a potential can be true only in virtue of some non-dispositional property or some other entity. If potentials are dispositions then reductive analyses of potentials fail as well as analyses of dispositions. Potentials are, according to McKitrick, dispositions to acquire a property or become a member of a kind. She also argues for the thesis that there are extrinsic potentials, i.e. there are cases of potentials with respect to which it depends on circumstances whether something has the potential or not; two intrinsically exactly alike duplicates can either have or lack a potential just because they are in different environments. In their paper Steven Mumford and Rani Lill Anjum argue for the claim that powers explain or metaphysically ground potentiality. This account includes the thesis that some potentials can be scientifically investigated because they are epistemically accessible. It also shows that and how the ascription of potentials is constrained, such that not everything has a potential for everything; however, something has a potential only if there is a power for it. Mumford’s and Anjum’s emphasis is on how powers jointly bring about their manifestations. They argue for a mutual manifestation model of powers, contrary to a stimulus response model that includes the powers causing their manifestation. By way of contrast to C.B. Martin’s mutual manifestation model, which is a mereological model of manifestation, i.e. the manifestation of a power comes about merely by the coming together of the required powers to some specific manifestation. This underscores the temporal extension of the manifestation process that is specific to potentials, the manifestation of a potential often takes some time, from the moment of beginning to its completion. They furthermore argue that Martin’s mutual manifestation model has to be amended with respect to the way the coming together of the powers is modelled. Contrary to Martin they think that the interaction of powers is a non-linear composition.

Introduction

9

Barbara Vetter explores how potentials are explicative of natural modality. She shows how a theory of modality can be grounded on potentials. The common opinion sees the direction of explanation reverse; potentials are to be explained by possibility. Her concept of potentiality is very broad such that it includes common dispositions and potentials in the more narrow sense. This makes it possible to erect a theory of possibility on the concept of potential. On this account p is possible, if and only if something has the potential for p to be the case. The rationale behind this is the view that we are more acquainted with the concrete potentials of things in nature than with abstract possibilities. Nevertheless the account is not supposed to be reductive, because potentials contain modal features. Vetter shows that her analysis covers all relevant cases of possibility, if one accepts iterated potentials, i.e. potentials to acquire yet further potentials; one of her examples is my potential for my great-granddaughter be a painter. She then argues that it is plausible to assume that there are potentials such as these. Part fife of the book deals with the different functions the concept of potentiality can play in different fields of philosophy. While it has an explicative function in the philosophy of mind it is used as basis of arguments in bioethics. Frank Hofmann shows, that it is a common thesis in the philosophy of mind that the mind is nothing but potential, the so-called potentiality thesis. This is understood as the claim that the mind is entirely constituted by its faculties that bestow dispositions. Hofmann shows that although there might be features that preclude potentials from being dispositions it turns out that there is no principled difference. However, the potentiality thesis is afflicted with difficulties as Hofmann shows, although potentials play a very important role for understanding the mental. A central feature of the mental is its normativity. According to Hofmann this cannot be accounted for by dispositions. Also, context-externalism, the claim that in order to have certain kinds of intentional mental states it is necessary to be related to the external world in the right way, is incompatible with the potentiality thesis, because according to Hofmann the relation is an actual relation, not a potential. Finally, phenomenal consciousness cannot be accounted for with reference to potentials because it is manifest in a way that involves features that are not dispositional but actual. Marco Stier discusses arguments in bioethics for which potentials play a central role. One argument takes potentials of human embryos to account for their special moral status. This argumentative use of the concept of potentials is made in arguments against abortion, embryo or stem cell research and artificial reproduction. The core premise of one version of this argument from potential runs thus: Because a human embryo has the potential to become a person in the future—or is a person with potentials—it now has the moral status of a person. Stier points out that there are two different issues that have to be taken apart. One issue is with the metaphysics of the relevant potentials and the other is with their normative relevance. He discusses the differing meanings of the concept of potentiality in bioethics and which consequences follow from them for the argument from potential. Stier discusses the numerous objections against the argument. His thesis is that since the concept of potentiality in the special case of potentials that are related with personhood has normative consequences, the concepts of ‘potential to become a person’ or ‘person

10

K. Engelhard

with a potential’ are themselves normative concepts in bioethics. Potentials also play a role in debates about human persons in special situations, like brain-dead patients and about special beings like human-animal chimeras. The articles in the sixth part of the book are devoted to different sciences for which the concept of potentiality is relevant. These articles reveal that the concept of potential or potentiality has different meanings and plays very different roles in physics, chemistry and biology. Max Kistler shows that ‘potential’ is a technical term in physics whereas scientists do not use ‘potentiality’. However the term ‘potentiality’ is also helpful in the metaphysics of physics. Classical physics includes the term ‘gravitational potential’ and the capacity of heat, which can be categorised as a potential. In quantum physics Heisenberg used the term in his interpretation of quantum mechanical systems in superposed states. Potentials in physics are states of systems; a system has a potential if all its forces are conservative; potentials in physics are scalar quantities. Some potentials in physics are convenient ways to represent fields and forces like the gravitational potential. There is however good reason to think that there are cases of potentials in which the potential is a real entity because it is causally efficacious. As Kistler points out potentials are dispositions of a certain kind—according to Kistler the demarcation between the terms ‘disposition’ and ‘potential’ is pragmatic; they are dispositions the manifestation of which takes more or less time than with common dispositions and they make their manifestation more or less probable but not necessary. Kistler however also registers differences between the use of disposition terms and potential terms: if a potential is manifested it is less appropriate to still attribute the potential. In his discussion of Heisenberg’s interpretation of quantum mechanical systems in superposed states Kistler argues that quantum mechanical phenomena underscore a certain theory of the difference between categorical and dispositional properties, namely that they are two ways to refer to properties rather than two numerically distinct kinds of properties. In chemistry potentiality is at stake in the very central problem of the identity of chemical elements over time, as Robin Hendry and Paul Needham show: how are we to account for the elements within a compound substance: are they actually or potentially present? While atomism affirms the first disjunct Aristotelians affirm the second. However, what does ‘being potentially present’ mean? Hendry and Needham examine whether and to which extent modern chemical theories are able to underscore the atomist model. Their thesis is that they only partially confute the Aristotelian model. The Aristotelian model, according to which the elements are potentially present within a compound substance, assumes that the elements are potentially present in the sense that they can be recovered by separation of the elements. Atomism on the other hand has to maintain that the separation of the elements is a rearrangement of parts of matter such that they persist throughout the changes that they undergo. However, modern chemistry shows that the transformation of substances do not underscore this view, the portions of matter before being transformed into the compound substance cannot be identified with those portions that result from the separation process. This casts interesting light on processes of change and the role of potentials within these processes.

Introduction

11

In biology potentials form an important group of properties of biological systems that are a subtype of dispositions as Andreas Hüttemann and Marie Kaiser argue; there are biological potentials in molecular biology, cell biology, developmental biology, evolutionary biology, and ecology. According to Hüttemann and Kaiser biological dispositions share two features: first, their manifestation process is a complex process of consecutive stages and second, that the manifestation conditions are highly specific collections or series of several elements. Furthermore, Hüttemann and Kaiser show that the concept of potentiality is a useful tool to describe the explicatory practices of biologists. They also argue for the claim that a dispositional account of natural properties provides a conceptual framework that makes biology and physics comparable sciences. Explanations by cp-laws, as in physics, and mechanisms, as in biology, can both be regarded as referring to dispositions. They analyse examples of biological potentials in the different fields of biology and show which influence complexity has on the manifestation conditions and the manifestation process of potentials. It turns out that biological potentials involve not only trigger conditions and background conditions for their manifestation like common dispositions but also sustaining conditions, conditions that have to obtain at very specific stages of the manifestation process and that are causally relevant for the temporally extended process to continue. Hüttemann and Kaiser also underscore the thesis that there are extrinsic potentials in biology.

References Borghini, A., & Williams, N. E. (2008). A dispositional theory of possibility. Dialectica, 62, 21–41. Goodman, N. (1983). Fact, fiction and forecast. Cambridge, MA: Harvard University Press. Nunes-Alves, C. (2014). The pandemic potential of H10N8. Nature Reviews Microbiology, 12, 461. (http://www.nature.com/nrmicro/journal/v12/n7/full/nrmicro3301.html). Vetter, B. (2015). Potentiality: From dispositions to modality. Oxford: Oxford University Press.

Author Biography Kristina Engelhard is a post-doc Research Fellow at the DFG-funded research group “Inductive Metaphysics“, located at the Technical University of Dortmund, Germany. Her current research is on dispositions, laws of nature, methods in metaphysics, metametaphysics, classical German Philosophy and early modern philosophy. Her publications include „Das Einfache und die Materie. Untersuchungen zu Kants Antinomie der Teilung“ (De Gruyter 2005), “Categories and the ontology of powers. A semi-dualist account of pandispositionalism” (in: A. Marmodoro (ed.), The Metaphsics of Powers. Their Grounding and their Manifestations, Routledge 2010), “Können Dispositionen das Realismus-Problem des transzendentalen Idealismus lösen?” (in: M. Egger (ed.), Philosophie nach Kant. Neue Wege zum Verständnis von Kants Transzendental- und Moralphilosophie, De Gruyter 2014).

Part I

The Concept of Potentiality in the History of Philosophy—Aristotle

Potentiality in Aristotle’s Metaphysics Anna Marmodoro

In this paper I will argue that Aristotle built his ontology solely on powers.1 On my reading, Aristotelian powers are pure powers. That is, all there is to a power is its powerfulness; nothing inert, or impotent is needed in the power’s nature to anchor the power to reality.2 But from this—namely that all there is to a pure power is its powerfulness—it does not follow that all there is to a pure power is potentiality. This latter position, which one encounters in contemporary accounts of powers, has the unwelcome consequence that a world of pure powers only is a world of potentialities only. I will argue that in Aristotle’s system pure powers are actualised, not by a transition to different potentialities, but by a transition to a different status of the powers themselves. Ontologies of pure powers are invariably construed as relational ontologies— powers in potentiality are taken to be essentially related to further powers in potentiality, namely, to their manifestations (see e.g. Bird 2007). Aristotle’s power ontology, by contrast, is not relational. On the one hand, a power in potentiality is the same power as its manifestation—so being manifested does not relate a power The European Research Council and the British Academy have supported two different stages of the research leading to the preparation of this article. 1

This interpretation makes a radical departure from the traditional ways Aristotle has been understood. I articulate it and defend it more fully than the limits of the present paper allows in Chap. 1 of in my monograph Aristotle on Perceiving Objects (2014). 2 Contrast with views on which a power has a categorical basis (e.g. Ellis 2010), or is qualitative as well as powerful (e.g. Heil 2003). Present Address: A. Marmodoro (&) Department of Philosophy, Durham University, Durham, UK e-mail: [email protected] A. Marmodoro Faculty of Philosophy, University of Oxford, Oxford, UK © Springer Science+Business Media B.V., part of Springer Nature 2018 K. Engelhard and M. Quante (eds.), Handbook of Potentiality, https://doi.org/10.1007/978-94-024-1287-1_2

15

16

A. Marmodoro

to a further power in potentiality. On the other hand, although Aristotle’s powers are dependent on other powers in order to be activated, they are not related to these other powers through polyadic relations, such as ‘x being the father of y’. For Aristotle ontological dependence is grounded on monadic properties, such as ‘y being a father’ and ‘x being an offspring’, that belong to interdependent entities. Thus, neither the manifestation of powers nor the interdependence among powers require introducing polyadic relations in Aristotle’s ontology. The ontological interdependence between powers structures them into a (non-relational) nexus, which is the bedrock of reality for Aristotle. Some of these powers are in potentiality and some in actuality. Powers in actuality are activated powers, exercising their powerfulness; they do not cease to be powerful while activated, nor is their powerfulness reducible to mere potentiality. Thus, powerfulness is either the potentiality to bring about change, or the actuality of bringing about change. Thus Aristotle’s ontology is a structure of interdependent powers at different stages of activation. Aristotle allows for three different ontological states of a subject in reference to a power: a subject s can possibly acquire a power (s can acquire the power of playing chess, or of heating other objects); s can have a power in first actuality (s has acquired the power to play chess and is not playing chess, or the power to heat other objects but is not heating any objects); and s can have a power in second actuality (s is playing chess, or is heating other objects). On my reading, for Aristotle both the first and the second actuality statuses are compatible with a power retaining its powerfulness, even if not in all cases its potentiality. This is why powers remain powerful while in actuality. This reading of Aristotle shows his account of powers to be importantly different from the contemporary accounts of powers, in that Aristotle does not identify the powerfulness of a power with its potentiality (or, in other words, to its dispositionality) [e.g. Ellis (2001: 127); also, Bird (2007) and Mumford (2010)]. Rather, for Aristotle, only the first actuality of a power possessed by a subject is potential, while powerfulness extends also to the second actuality, the activated power; in both states, the power is powerful with reference to change—it is capable of bringing about change, or is bringing about change. The appropriate conditions for the activation of a power by another power are generically described by Aristotle as ‘contact’ between the two powers. Contact could be thought of as a relation between powers, but Aristotle does not think of it in such terms. Rather, he analyses contact in terms of place and limits, without there being a connection between a limit and what converges on it. Aristotle thinks of powers as differentiated into active and passive powers, where an active power ‘moves’ or somehow ‘operates on’ the passive power. In most cases he thinks in fact that each power is at the same time both active and passive, since powers operate on each other. It is helpful to think, with Aristotle, of the operation of a mover on a movable as the transmission of the form of a power, e.g. the hot, onto a passive power. This should not be thought to be a literal description, but a figurative account that explains what results from the mutual activation of an active and a passive power. Thus, it should not be thought that when an active power activates a passive power there is an underlying physical mechanism for the

Potentiality in Aristotle’s Metaphysics

17

transmission of any item, such as the form of the agent to the patient power.3 The transmission of form is a way of describing what is brought about by causation, as though the patient power received the form of the agent power (although nothing is actually transferred from the agent to the patient). But as we shall see, what happens is not a transmission, but an activation of the patient power by the active power.4 Causal change, in Aristotle, results from the mere proximity of interdependent powers. In the above I have been discussing fundamental powers rather than complex macro-powers, such as the power to build a house or to carve a statue. In the case of macro-powers it might appear that matter is transferred from the agent to the patient, e.g. the builder lays bricks to build the house. But in fact the builder is building by simply passing on the form of the house to the bricks through her movements. (Again this is a figurative description, since the builder is not actually transferring ‘a form’ to the bricks.) The only thing that happens in the exercise of the building power is the mutual activation of powers, whether this mutual activation is between the powers of the builder’s arms and of the bricks, or between the powers of the bricks themselves, etc. For Aristotle there is nothing exchanged between the powers. Talking of the transmission of the form is only a way of describing the type of causal action of a power upon another: as if the form of the one was transmitted onto the other power. Before coming to a more detailed discussion of Aristotle’s position, it might be helpful to put my interpretation in a nutshell, to contrast Aristotle’s views, as I understand them, to his predecessors’, and to bring out the novelty of his account, not only in relation to ancient metaphysics, but for contemporary philosophy too. The Pre-Socratics described the world by introducing principles that were aimed at explaining change in nature. Movement or force gives rise to generation and change in some type of stuff or other; e.g. the hot and the cold, or condensation and rarefaction change water, or air, or amorphous stuff, or atomic formations, etc. Plato, on the other hand, focused on a problem that did not arise in the Pre-Socratics, namely the instantiation of universals in his ontology; but he

3

Aristotle understands perception too, which is a case of causation, as being the reception of form without matter by the sense organ. For a fuller discussion of the case of perception, see Marmodoro (2014). 4 In contemporary physics, to explain how elementary particles act on one another virtual particles are posited, as force-carriers (in effect, force-instances). Thus, elementary particles exert forces on each other by exchanging such virtual particles—e.g. the gauge bosons. One might think that, by introducing virtual particles to carry forces from particle to particle, e.g. the electromagnetic force or the weak force, contemporary physics has solved the problem of causal efficacy by replacing causal efficacy with addition to, or subtraction from the constitution of the particles (e.g. more, or less, weak force) rather than with interaction between particles. But there are reasons to think that this is not the case. Virtual particles of different types interact with one another, too. For instance, in the Standard Model, vector bosons couple with fermions, and W bosons couple with a photon or a Z0 boson (Couchman 2000). Such couplings between virtual particles happen due to the effect of gauge bosons on gauge bosons of a different type. Such primitive effect between forces is what Aristotle, too, assumed in his theory of causation.

18

A. Marmodoro

neglected to provide a metaphysical account of generation and change. He reified instantiation as the relation of participation in the Forms, which take on the role of universals among other roles. But change receives brief explanation in Plato, metaphysically as the occurrence of participation in Forms (Phaedo 100c–d; Aristotle, Metaphysics 991b3–9; Generation and Corruption 335b7–24), and physically as the rearrangement of elements (Timaeus 56c–e) (which is akin to Pre-Socratic types of explanation of change). Aristotle, by contrast, does not reify the instantiation of universals into a relation, nor does he reify universal forms as such, nor matter as such—and this has important implications, to be drawn in what follows, for understanding his hylomorphism. The ultimate constituents of hylomorphic compounds are not matter and form held together by relations between them. The ultimate constituents are powers, which become activated. More graphically, the bedrock of Aristotle’s reality is not a two-tier hylomorphic “clasp” between matter and form; but a single tier of powers that are either in potentiality or are activated. If we are to think of a ‘cosmic generator’ of creation and change for Aristotle’s world, it is not the coming together of matter and form into hylomorphic compounds; but rather, it is the activation of powers from potentiality to actuality [see Marmodoro (2013)]. For Aristotle, change is the activation of a passive power, whose nature is to suffer the agency of an active power. A passive power’s activation (actualisation, realisation)5 may be a process, such as becoming hotter; or an activity, such as seeing. For Aristotle, there is change in the case of the process only, since the resulting state from the process is qualitatively different from the initial state—as for instance in the case of heating (process), but not of seeing (activity).6 An important difference between Aristotle’s account of powers and the contemporary ones is the following. In contemporary theories, the manifestation of a power is a new power [e.g. for Mumford and Anjum (2011)] or a new property (e.g. for Bird (2007) that comes about, e.g. an ice cube’s power to cool the lemonade in the glass is manifested in the new—lower—temperature of the lemonade.7 But for Aristotle, this is not the case. The actualisation of a power is not a new property that comes about. Rather, it is the activation of the power, either as it is exercising its influence on the passive power or as the passive power is suffering that influence. For example, if a mango has the power to ripen in the heat, the ripening is the actualisation of active and passive powers at play. The ripe state of the mango that comes about is the ‘aftermath’ of the activation of the powers, not their manifestation. The powers are manifested in their activity with each other, not in the state that results from their activation. I will use the terms ‘activation’ and ‘realisation’ of powers interchangeably in what follows, to describe a power’s reaching the end that defines its nature. 6 In our common sense conception of change, both process and activity count as changes. What Aristotle wishes to capture by treating only process as change is that activity does not alter the constitutional make up of the active agent, but only puts the existing constitution to work. 7 For instance, see Bird (2007: 7): ‘Potencies are characterized in terms of other properties (their stimulus and manifestation properties)’. 5

Potentiality in Aristotle’s Metaphysics

19

All that happens in Aristotle’s world is that powers in potentiality come to be actualised, either as agents of change or as patients of change. Change involves the mutual activation of agent and patient powers, brought about by the contact between ontologically interdependent pairs of powers, such as e.g. what can heat and what can be heated. Powers are monadic properties, since Aristotle does not reify polyadic relations in his ontology. But powers can be ontologically dependent on each other, where ontological dependence is not a ‘connecting bridge’, i.e. a relation or an extra entity connecting an agent power and a patient power. In sum, nature is a cluster of interdependent powers that, when in contact, activate one another; this may result either in activity, e.g. seeing, or in a process of change, e.g. being heated.

1 Potentiality for Change For Aristotle, there is a primary sense of potentiality from which the other senses are derived. This is the capacity to bring about change: All potentialities (dynameis) that conform to the same type are starting points of some kind, and are called potentialities in reference to one primary kind, which is a starting point of change in another thing or in the thing itself qua other. (Metaphysics 1046a9–10, my emphasis)8

This primary sense of potentiality is that of the capacity to bring about change in another thing or in the same thing as if it were another. The former case is the standard case of causing change, such as fire heating an object or an object in motion setting something else in motion, etc. Aristotle’s qualification of causing change in the thing itself as if it were another is aimed at including complex entities which have the capacity of bringing about a change in a part or the whole of themselves, e.g. an athlete training herself. Of course the criterion of being a cause that is the originative source of change would need to be relativised to a context, for otherwise one could endlessly trace back origins of change. Aristotle does not mention this issue in the passage quoted, but it is a consideration he is sensitive to, as we see from his subsequent discussion of matter and the introduction of the concept of proximate matter: It seems that when we call a thing not something else but ‘of’ that something else (e.g. a casket is not wood but of wood, and wood is not earth but made of earth …), that something is always potentially (in the full sense of that word) the thing which comes after it in this series. E.g. a casket is not earthen nor earth, but wooden; for wood is potentially a casket and is the matter of a casket. (Metaphysics 1049a18–23; my emphasis)

Aristotle’s point here is two faceted. On the one hand he is introducing a strong sense of potentiality; on the other, he is associating it with a semantic/grammatical

8

All translations of Aristotle’s text are from Barnes (1995).

20

A. Marmodoro

phenomenon. The strong sense of the term ‘potentially’ is to be found in the case of adjacent items in the series of changes—e.g. when the wood becomes a casket. The earth is not potentially a casket in this sense, because there is an intermediate step between earth and casket in the series of changes from earth to wood to the casket. So it is the wood that is strictly speaking potentially the casket in the present context, despite the fact that the wood comes from earth, and hence the casket comes from earth. This is the reason, Aristotle explains, for the corresponding semantic/grammatical point, that is, why we call the casket ‘wooden’ but not earthen. For corresponding reasons, in any given change, the origin of the change, i.e. its cause, will be taken to be the immediate cause of this particular change in question, rather than an antecedent one in the causal history of this change. A second type of potentiality Aristotle includes in his ontology is the capacity to suffer change: For one kind is a potentiality for being acted on, i.e. the principle in the very thing acted on, which makes it capable of being changed and acted on by another thing or by itself regarded as other. (Metaphysics 1046a11–13)

We might not be immediately ready to acknowledge the capacity to suffer change as a power; but it only takes some reflection to see that we do have some terms in everyday language that pick out just such capacities or powers, e.g. ‘fragility’, or ‘malleability’, or ‘flexibility’, etc. For Aristotle being able to change is as much a capacity or power as being able to effect change: In a sense the potentiality of acting and of being acted on is one (for a thing may be capable either because it can be acted on or because something else can be acted on by it), but in a sense the potentialities are different. For the one is in the thing acted on; it is because it contains a certain motive principle, and because even the matter is a motive principle, that the thing acted on is acted on … for that which is oily is inflammable, and that which yields in a particular way can be crushed; and similarly in all other cases. But the other potency is in the agent, e.g. heat and the art of building are present, one in that which can produce heat and the other in the man who can build. (Metaphysics 1046a19–28)

A notion that is peculiar to Aristotle’s account is conceiving of passive powers as originative sources of change (1046a11–13; a23). It is natural for us to think that an originative source of change is a power to bring about change; but it is not as natural to think that an originative source of change is a capacity to suffer change. Yet Aristotle sees both active and passive powers as originative sources of change, the one as a source that changes something, and the other as a source of suffering change. In fact, Aristotle gives several examples of originative sources of suffering change to make his point clear, such as for example, oil or brittle matter. The distinction between the active and the passive capacities or powers also serves to set up the conditions under which change takes place. This is determined in the definition of a capacity. What is specified in the definition of a capacity is: the type of capacity the capacity in question is, namely what it is that it can do, i.e. bring about or suffer; the appropriate occasion in which the capacity can do this; the way in which it can do it; and any other conditions that need to obtain for it to do what it does. When all the conditions set out in the definition are met, including the

Potentiality in Aristotle’s Metaphysics

21

active and passive powers coming in contact, in the relevant sense of contact for the type of power they are, then necessarily the agent power acts on the passive power and brings about its effect: Since that which is capable is capable of something and at some time in some way – with all the other qualifications which must be present in the definition–, … as regards potentialities of … [those things that are non-rational; e.g. the fire] … when the agent and the patient meet in the way appropriate to the potentiality in question, the one must act and the other be acted on … For the non-rational potentialities are all productive of one effect each. (Metaphysics 1047b35–1048a8; my emphasis)

The necessity is natural necessity, stemming from the nature of the capacities/ powers themselves, on satisfaction of the conditions in the definition of their natures. Aristotle did not talk of laws of nature. Yet, it is clear from the normativity expressed in this passage, that the definitions of powers determine the conditions whose satisfaction is the instantiation of laws of nature.

2 Causal Agency We must now come to examine and try to comprehend the nature of the agency of the capacity/power of a mover on the capacity/power of the movable. There are two aspects of a mover’s causal agency that reveal its nature. The one is what it brings about, and the other is how it achieves it. Aristotle describes what the mover does to the movable in terms of the transmission of the form of the moving power: The mover will always transmit a form, either a ‘this’ or such or so much, which, when it moves, will be the principle and cause of the motion, e.g. the actual man begets man from what is potentially man. (Physics 202a9–12)

So the form could be a substantial form, as in the case of the transmission of the form of a human being to the menstrual fluids in the generation of an embryo; or it could be a quality, a suchness, as for instance heat or weight, etc. But what does it mean to say that the form of the moving power is transmitted? Aristotle wants to find a way to explain the change that is brought about by the moving power. In his example above, the generation of a new human being is accounted for by the transmission of the form of a human being, which is the principle and the cause of the motion. The form transferred is the form that determines the end/goal (telos) of the potentiality in the moving power’s definition. Thus a parent has the potentiality to generate a human being, and a painter the potentiality to generate a painting on canvas. These are the ends that the movers’ powers are directed towards, in their potential state, e.g. the ends that the parent and the painter have respectively. They express what the powers can bring about when actualised. Aristotle’s explanation of causation in terms of the transference of a form from the moving power to the passive one should not be taken as a literal but a figurative description. Aristotle is not reifying the form of the power into an active agent of its own, over and above the power. There is no homunculus-form that is transmitted from the parent to the

22

A. Marmodoro

offspring. There are only motions transmitted from the parent to the menstrual fluids by the sperm that is implanted in them; the heat in the parent’s sperm generates the motions in the fluids, which gradually shape the embryo, as Aristotle tells us explicitly (See Aristotle, Generation of Animals II.2–3). Similarly, there is no form of a statue that is literally transferred from the sculptor to the marble; a sculptor transfers the form of a figure in her mind to the marble through the movements of her hands and chisel. Nevertheless, talk of transmitted forms is the best way to describe collectively the type of effect that the respective moving powers have on the passive ones. The movements generated from the heat of the sperm in the first case, and from the hands of the sculptor in the second bring about changes of particular types, which are determined by the kind of moving power that is acting on the passive one. The resulting change is as if the sperm transferred a form onto the menstrual fluids, which en-formed them and shaped them into an embryo; and as if the sculptor transferred a form onto the marble which en-formed it into a statue. There is no such magic; Aristotle’s account is cogent and intuitive. Macro-changes emerge from micro-changes brought about by the fundamental powers (i.e. as we will see, the hot, the cold, the wet and the dry), which affect their passive correlates. Even if one took Aristotle to be saying that, literally, there is a (reified, matter-less) form that is transmitted to the passive power, this would still not explain how causation takes place. We would want to know how that form does it; what causal efficacy a form can have on a passive power. Assuming Aristotle is looking for an answer as to how a power affects another, adding a further item to the causal series would not offer an explanation. It would only continue the regress generated in the search for the mechanism of causal efficacy. Then how does causal efficacy operate? Even if macro-level powers depend on micro-level powers to bring about their effects, how do micro-level powers exert their causal efficacy on other micro-level powers? As we shall see, Aristotle avoids the regressive series of introducing further intermediaries by assuming the efficacy of a moving power on a passive one; all that happens is that ‘when the agent and the patient meet in the way appropriate to the potentiality in question, the one must act and the other be acted on’ (Metaphysics 1048a6–7). The transference of the form of the moving power to the passive one is not a description of the mechanism of causal efficacy, but only a collective description of the type of qualitative change that takes place in the passive power. Examining closely the behaviour of active and passive powers will help us understand how this change is effected.

3 Stimulus and Appropriate Conditions Aristotle acknowledges that there is a variety of what we would call enabling conditions for the activation of a power, pertaining to the right time, the right situation, the right external conditions; he summarises them in saying that the mover is capable of something ‘at some time in some way (with all the other qualifications which must be present in the definition)’ (Metaphysics 1048a1–2).

Potentiality in Aristotle’s Metaphysics

23

On the other hand, he collectively describes the stimulus that triggers powers that are in the right circumstances into causal activity by the general condition of ‘contact’ between them: To act on the movable as such is just to move it. But this it does by contact, so that at the same time it [the mover] is also acted on. Hence motion is the fulfilment of the movable as movable, the cause being contact with what can move, so that the mover is also acted on. (Physics 202a5–9)

So contact is the triggering condition, along with all the other conditions mentioned in the definition determining the enabling conditions for causal efficacy to take place. It is therefore important to come to understand what is involved in the contact between the active power and the passive power it operates on. Aristotle tells us that: Things are said to be in contact when their extremities are together. (Physics 226b23)

He further explains that Things are said to be together in place when they are in one primary place and to be apart when they are in different places.9 (Physics 226a21–3)

So things that are in contact have their extremities in the same place. For the purposes of causation, having the extremities in the same place will have to be understood as either touching or being in proximity, since there appear to be cases where there is causal impact even when things are merely proximate, namely, in the same place in the sense of same spatial region. For example, proximity to a fire is sufficient for heating, and even for catching fire. We saw above that for Aristotle to act on the movable is to move it, and that this is achieved by contact. In examining his concept of contact as defined, we see that the contact between mover and movable involves the coincidence of place of the extremities of the mover and the movable, e.g. the chisel of the sculptor and the wood, or the flames of the fire and the pot resting on it. But once contact is achieved, what does the operation of the mover on the movable involve? Is it something over and above the contact between them, and if so, what? Is there a relation established from the mover to the movable that is responsible for the change the movable suffers, or is there a different mechanism of causal interaction? The answer to the question of how the mover operates on the movable and brings about change will not for Aristotle involve positing any ‘bridge’, relation or connection from the mover to the movable. He grounds his theory of causation in his account of relations in terms of monadic properties and counterfactual dependence. To this I now turn.

Aristotle defines place as ‘the innermost motionless boundary of what contains’ (Physics 212a20– 21).

9

24

A. Marmodoro

4 Non-relational Powers We know from Aristotle’s Categories (Chap. 1) and from the Metaphysics (book VII Chap. 4) that even accidental properties (e.g. being pale, or being hot) have essences and definitions. Furthermore, properties cannot exist unattached, on their own, but they have to belong to a subject (see Categories Chap. 2). If we then consider a relation between two things, for example, Marco being the father of Pietro, and we try to think of this relation as a single polyadic property that conjoins the two, Marco and Pietro, decisive difficulties follow. On the one hand, this polyadic property would belong to both subjects, since it can only exist by belonging to something(s) as subject, and both subjects have a claim to it by being conjoined by it. On the other hand, although Marco is related to Pietro as a father, Pietro is not related to Marco as a father, but as a son; hence, either the polyadic property would belong to Pietro without being true of him; or the polyadic property would have two different natures, endowing each of the two conjoined entities with different qualifications, which is incompatible with the property being one and the same. The asymmetry of the relation introduces plurality of natures in it, and this plurality undermines its oneness. Conceiving of relations as polyadic properties was not even entertained by Aristotle. For him relational predicates are accounted for in terms of monadic properties—monadic properties of a special kind, which he called the pros ti (the ‘toward something’) type of property: such properties are in themselves pointing toward something other than themselves. Thus, Aristotle says: We call relatives all such things as are said to be just what they are, of or than other things, or in some other way in relation to something else. For example, what is larger is called what it is than something else (it is called larger than something); and what is double is called what it is of something else (it is called double of something); similarly with all other such cases. (Categories 6a36–b3)

(Aristotle does not distinguish between relatives and relations. I take it this is for the reason given above: that neither relatives nor (asymmetric) relations can be single polyadic properties with a single nature each that belongs to (is true of) the two relata.) What does Aristotle mean by taking relatives to be pros ti—toward something? He explains it as follows: All relatives are spoken of in relation to correlatives that reciprocate. For example the slave is called slave of a master and the master is called master of a slave (Categories 6b28–30)

Pros ti properties are monadic properties such that their manifestation or activation depends counterfactually on the activation of their correlatives. Someone is actually a master only if there is a slave of whom he is master; and vice versa for the

Potentiality in Aristotle’s Metaphysics

25

slave.10 Reciprocating correlatives are not related to one another but are ontologically interdependent, as Aristotle states clearly: If there is no master, there is no slave either … When there is a slave there is a master; and similarly with the others [sc. other relatives] … Also, each carries the other to destruction; for if there is not a double there is not a half, and if there is not a half if there is not a double. So too with other such cases. (Categories 7b6–22)

So the ‘pointing’ nature of relatives is Aristotle’s way of depicting ontological dependence. This is what binds monadic properties into reciprocal pairs for their activation, e.g. being a master and being a slave. But ontological dependence is not a polyadic relation between relata. Just as there is no polyadic connection binding a species to its genus, in spite of their ontological interdependence, similarly, for Aristotle, there is no polyadic connection binding one activated monadic property to its correlative.11 A species is ontologically dependent on its genus, and yet there is no entity stretching over to bind them; the same holds for relata even if their ontological interdependence is different in kind than the genus-species one. And the same holds for matter and form, and subject and property, where Aristotle is explicit in saying that there is no (polyadic) entity unifying them into one (See Metaphysics 1045b8–16; b21–22).

5 The Dependence of the Mover on the Movable The reason why we examined Aristotle’s account of relatives in the foregoing section is that Aristotle considers causes as relatives; namely, the agent and patient in a causal pair are causal relatives. In Metaphysics book V Aristotle explains the term ‘relative’ or ‘relation’ as follows: Things are relative [pros ti] (1) as double to half, and treble to a third, … and that which exceeds to that which is exceeded; (2) as that which can heat to that which can be heated, and that which can cut to that which can be cut, and in general the active to the passive; (3) as the measurable to the measure, and the knowable to knowledge, and the perceptible to perception. (1020b26–31)

Causal examples such as heating and being heated are included in the above among relatives such as double and half, and they are collectively grouped under the description of being active and being passive. It follows that generally, for Aristotle, the mover and the movable, namely the active and the passive powers in causal interactions, are causally engaged through two monadic properties rather 10 Aristotle’s account of relatives is different from Bennett’s (2011a) account of superinternal relations. For Bennet superinternal relations are such that the intrinsic nature of one relatum grounds the obtaining of the relation, as well as the existence and nature of the second relatum. For Aristotle it is the nature of both correlative monadic properties that grounds the nature of their interdependence. 11 See also pages 17 and ff. below.

26

A. Marmodoro

than a polyadic one. In the causal cases, the two monadic properties are the powers themselves, namely the moving power and the passive power. Using the example just given, it would be the heating power and the power to be heated, the cutting power and the power to be cut, the power of perception and the power of being perceived, and generally the active and the passive powers. According to Aristotle’s account, then, the active and passive powers are monadic properties that are ontologically interdependent. We saw that Aristotle explained the ontological dependence between relata as an existential dependence; e.g. nobody is a slave if there is no master. Similarly here, the ontological dependence between the causal relata entails that there is no mover if there is nothing movable. But here the question arises whether the ontological dependence determines the potential or the activated state of the powers in question. Clearly, if there is nothing that can be affected, then the mover will not bring about any effect; there is no actual moving—changing—if there is nothing that can be moved—changed. For instance, if nothing can be heated, no heating will take place either. But can there be a mover in potentiality, even if there is nothing that could be moved? Could there be a knife in a world where nothing could be cut? We already saw that the definition of a power mentions the end toward which the power is directed: what the power is capable of bringing about (Metaphysics 1047b35– 1048a8). But if there is nothing that can be so affected, how can there be a power whose nature is to bring about that effect? Aristotle believes in some form of the Principle of Plenitude, namely, that what is possible will happen [see Makin (2006: 84)]. If so, then it follows that he believes that the end of each power in potentiality must be realisable. So, there could be no mover, even in potentiality, in a world where it was not possible for it to move anything. Hence, we must assume that the ontological dependence between active and passive powers applies to their potential state, as much as to their activated state.

6 The Effect of the Mover on the Movable How does the active power bring about an effect on the passive power? What does Aristotle mean when he says that the mover acts or operates on the movable? How are we to understand the activity that takes place when causal powers are being efficacious? He explains that To act on the movable as such is just to move it. But this it does by contact, so that at the same time it is also acted on. Hence motion is the fulfilment of the movable as movable, the cause being contact with what can move, so that the mover is also acted on. (Physics 202a5–9)

We saw above that the definition of a power mentions the various conditions that must obtain for the power to be activated. If these conditions obtain, when there is contact, the power is activated, bringing about a change in the passive power. (In most cases, a moving agent is also moved itself in return by the contact with the

Potentiality in Aristotle’s Metaphysics

27

agent that is moved, e.g. as in pressing against a hard surface.) We further saw that Aristotle also says that The mover will always transmit a form, either a ‘this’ or ‘such’ or ‘so much’, which, when it moves, will be the principle and cause of the motion, e.g. the actual man begets man from what is potentially man. (Physics 202a9–12)

But we found good reason not to take this as a description of some item actually transmitted from the mover to the movable, as though a form that was in the mover is transferred to the movable in the way that a book can be transferred from one shelf to another. We saw that the transference of the form is an explanatory principle of the type of change that takes place in the movable. Hence, the contact does not engender the transference of something, but rather the occurring of a change, whose type is explained by the form of the active power, namely, the type of agent the active power is. My understanding of Aristotle’s conception of the operation of an active power on a passive power is that if the conditions specified in the definition of the active power are met, contact between them triggers the change in the passive power. Aristotle says: All things that are capable of affecting and being affected, or of causing motion, and being moved, are capable of it not under all conditions, but only when they are in a particular condition and approach one another: so it is on the approach of one thing to another that the one causes motion and the other is moved. (Physics 251b1–4)

When the agent and patient fulfil the respective conditions and are appropriately near one another then the passive power suffers the effect of the active power: That which can be hot, must be made hot, provided the heating agent is there, i.e. comes near. (On Generation and Corruption 324b7–9)

Nothing else happens other than the proximity of the agent and patient under the right conditions, as specified in the definitions of their powers. It is in the nature of the powers to interact thus, when the enabling conditions are satisfied and their contact triggers the change. Aristotle describes the nature of an agent as having a principle of movement: That which contains the origin of the motion is thought to impart motion (for the origin is first amongst the causes). (On Generation and Corruption 324a26–28)

What we are to understand by this is that in the right conditions, the principle of the origin of motion applies, and change occurs. Things in the world are made in such a way that they change when in proximity to some further things of a special type. The change is qualitative, not a transference of matter. Transferring matter or a quantum of energy or any other item does not explain causal efficacy, since we still need a further account of how the transferred item interacts with the environment it enters. For Aristotle there is transference of form from agent to patient, but this is only a qualitative description of the type of change that occurs. So, ‘contact’ for Aristotle is a key factor for causal efficacy. It does entail a type of proximity or sameness of place, but more importantly, in a causal context, it has

28

A. Marmodoro

come to mean, for him, triggering change, allowing ‘touching’ even in situations where the touching is not physical and not even reciprocal: If anything imparts motion without itself being moved, it may touch the moved and yet itself be touched by nothing-for we say sometimes that the man who grieves us touches us, but not that we touch him. (On Generation and Corruption 323a31–33)

7 The Primary Powers Aristotle aims at a rational explanation of the world all the way down to the bedrock of reality. He says: ‘the differences are reasonably distributed among the primary bodies, and the number of the latter is consonant with theory’ (On Generation and Corruption 330b6–7). Accordingly, he offers his famous analysis of the simple elements, namely fire, air, water, and earth. In allotting the fundamental properties that characterise these elements, Aristotle narrows down the candidates to the tangible contrarieties (329b6–9), and of those, the ones that are powers: ‘capable of acting and being affected … said of things in virtue of their acting upon something else of being acted upon by something else’ (329b20–21). He goes through an analysis of the list of contrary powers of what is tangible, and concludes that they are all reducible to four primary, fundamental powers: It is clear … that all the other differences reduce to the first four, but that these admit of no further reduction. … Hence these must be four. (330a24–29)

These primary powers are heat, cold, wetness and dryness. The four simple elements are constituted of the four primary powers: Fire is hot and dry, whereas Air is hot and moist … and Water is cold and wet, while Earth is cold and dry. (330b3–5)

There are no further fundamental properties that any of the simple elements possess other than two contrary powers each. The simple elements can reciprocally transform into one another by gaining or losing their powers—for instance: There will be Air, when the cold of the Water and the dry of the Fire have passed-away (since the hot of the latter and the moist of the former are left); whereas, when the hot of the Fire and the moist of the Water have passed-away, there will be Earth, owing to the survival of the dry of the Fire and the cold of the Water. So, too, in the same way, Fire and Water will result from Air and Earth. For there will be Water, when the hot of the Air and the dry of the Earth have passed-away (since the moist of the former and the cold of the latter are left); whereas, when the moist of the Air and the cold of the Earth have passed-away, there will be Fire, owing to the survival of the hot of the Air and the dry of the Earth-qualities constitutive of Fire. (331b14–24)

What Aristotle is describing is the constitution of the most fundamental level of reality. There are four types of primary power, which do not exist separately each on its own, but pair up and constitute the simple elements. The simple elements,

Potentiality in Aristotle’s Metaphysics

29

earth, water, air, and fire, are the most fundamental separable things in nature.12 Each element has two contrarieties, and when they come in contact the interaction between them results in the heat of fire overpowering the coldness of the water while the wetness of water overpowers the dryness of fire, giving rise to what is hot and wet, namely air. Aristotle distinguishes the underlying matter, the contrary properties, and the composite of the two, namely the simple elements: Our own doctrine is that although there is a matter of the perceptible bodies (a matter out of which the so-called elements come-to-be), it has no separate existence, but is always bound up with a contrariety. A more precise account of this has been given in another work; we must, however, give a detailed explanation of the primary bodies as well, since they too are similarly derived from the matter. We must reckon as a principle and as primary the matter which underlies, though it is inseparable from, the contrary qualities: for the hot is not matter for the cold nor the cold for the hot, but the substratum is matter for them both. Thus as principles we have firstly that which is potentially perceptible body, secondly the contrarieties (I mean, e.g. heat and cold), and thirdly Fire, Water, and the like. For these bodies change into one another (they are not immutable as Empedocles and other thinkers assert, since alteration would then have been impossible), whereas the contrarieties do not change. (Generation and Corruption 329a28–b2)

Properties are not subject to change, any more than e.g. numbers are—the reason being that qualitative alteration of a property or a number (if it were possible) would give them a different identity; properties and numbers are what they are, and all they are, essentially.13 But the composites that result from the matter and the properties, namely earth or air etc., are subject to change.

12

They are separable at least in principle, since they are not found in pure form in nature, but mixed between them. See Generation and Corruption 330b21–23: Fire and air, and each of the bodies we have mentioned, are not simple but combined. 13 A qualification of this assertion is in place. Aristotle explains mixing by saying that when two items, e.g. wine and water, mix, Then each of them changes out of its own nature towards the dominant one … Thus it is clear that only those agents are combinable which involve a contrariety (for these are such as to suffer action reciprocally). (Generation and Corruption 328a28–33) But mixing requires that: Each of them may still be potentially what it was before they were combined, and both of them may survive undestroyed. (Generation and Corruption 327b25–26) In giving this account of mixing, Aristotle does not assume that the contrarieties themselves change, without being destroyed. It is the things mixed which change without being destroyed, and therefore their complex natures that somehow remain and yet are compromised—in a way that we might compare to the way the nature of a substance remains, increasingly compromised, while the substance is deteriorating towards the end of its life span. The identity criteria of complex natures need not be the same as those of simple properties.

30

A. Marmodoro

Since properties themselves do not change, when change occurs it is the entity qualified by a property that is changing, by acquiring a new property in place of the former one. Thus when the hot is replaced by the cold, it is what is hot that changes, not the property of being hot, which is in fact lost. So, when air transforms into water it loses its primary power of heat, which is replaced by the power of cold. Thinking of properties as enmattered powers explains the particle-language that Aristotle uses in the passage quoted above, which is suggestive of constituents which do not change but transform (Generation and Corruption 331b14–24). He is considering the coming together of the simple elements, in this case water and fire, about which he says that the cold of the water and the dry of fire are destroyed (phtharēnai), while the heat of the fire and the wetness of water remain (leipetai). The deictically-referred-to items that remain, the enmattered powers, are the constituents that now come to constitute the new simple element, in this case air; and so forth. I examined in some detail the immutability of primary powers, the contrarieties, because it is a feature they share with (Democritean) atoms. In addition to immutability there is another important similarity between primary powers and atoms: primary powers are fundamental; they are not constituted of any further elements, as their building blocks, any more than the atoms are constituted by any elements as building blocks. Rather, primary powers are grounded in matter that is pure potentiality, with no specific features of its own. Therefore, a primary power (cold, hot, wet, dry) is the form of the power of cold, or hot, wet, or dry, enmattered in pure potentiality. There are no constituting elements of primary powers, and therefore there are no further elements constituting the simple elements—air, water, earth, and fire—apart from their primary powers. In view of the fact that for Aristotle everything in physical nature is built out of the four simple elements and their mixtures, it follows that all there is in nature is built out of primary powers grounded on pure potentiality. In light of the above considerations, it is helpful to think of Aristotle’s account of the most fundamental building blocks of reality, the four primary powers (hot, cold, wet, dry) as power-atoms. They are not atoms of the standard kind, in so far as primary powers can be destroyed, while atoms cannot; also, a primary power cannot exist separately in nature, but must always be paired up with another primary power. But the primary powers are like atoms in that they are not the subjects of change (from what they are), and they are the fundamental elements from which everything else is built. Furthermore, all physical changes in nature derive from changes in the combinations of the primary powers. As we saw, Aristotle reduces the principles of perceptible bodies to tangible contrarieties only, and reduces all tangible contrarieties to the four primary powers (Generation and Corruption Book II, Chap. 2). Thus, any change in the perceptible bodies derives from a change in the primary powers. There is a difference between atoms and Aristotle’s power-atoms to be considered in further depth. Atoms are independent of one another; but Aristotle’s

Potentiality in Aristotle’s Metaphysics

31

power-atoms are not. Powers are dependent on one another—active and passive— for their fulfilment/activation/realisation.14 Aristotle analyses relational predicates in terms of monadic properties that are ontologically dependent on one another (see above). Primary powers are power-atoms whose ontological interdependence can be explained by Aristotle as monadic interdependent properties. Thus, the picture that emerges is that for Aristotle the most basic stratum of reality consists in monadic properties, the primary powers, composing the four elements (water, air, fire, earth), and arrayed in a structure of ontological interdependence on one another for their actualisation. This structured layer of interdependence is the ground of all there is in nature.15

8 Actualised Powers We need now to examine whether Aristotle’s world of powers is a world of mere potentiality. The problem I am addressing here is whether Aristotle’s power-ontology is committed to the position that all there is or can be is potential, and that change is simply a transition from one potential state of the world to another such state. This is a problem faced by many contemporary power-ontologies, sometimes referred to as the “Always packing, never travelling” problem.16 David Armstrong (1997: 80), following C. B. Martin (1993: 68), formulates the problem thus: Given purely dispositionalist accounts of properties, particulars would seem to be always re-packing their bags as they change properties, yet never taking a journey from potency to act.

The problem stems from the position held by contemporary power ontologists that the manifestation of a power is another power. This position commits them to a network of powers in potentiality, as the actualization of each power is a transition to another power in potentiality. Aristotle distinguishes the activation of a power from the realisation of the power’s end. The end of a power is given in the power’s definition: That which is capable is capable of something and at some time in some way – with all the other qualifications which must be present in the definition. (Metaphysics 1047b35– 1048a2)

I will use the terms ‘fulfilment’, ‘activation’ and ‘realisation’ of a power interchangeably in what follows, to describe that the end that defines the nature of the power comes to be. 15 I discuss further the ground level of powers in elemental transformations in my paper ‘Reciprocity without symmetry in causation’, Marmodoro 2017. 16 It is Molnar (2003: 173) who called it the “Always packing, never travelling” argument. 14

32

A. Marmodoro

For Aristotle, some powers have an end which is an activity, and some have an end which is a process. The former are powers whose end is realised instantaneously, such as the power to see; at any one moment you see and you have seen; Aristotle calls these “activities” (energeia, praxis). The latter are powers whose end is realised in stages in a process, such as the power to build a house; when you are building a house you have not built a house; Aristotle calls these “changes” (or motions, kinēsis). Changes have a natural completion point, when the end of the process is reached, such as the completion of the house; activities do not have a natural completion point, as e.g. in the case of seeing. Thus a power can realise its end either through a change or through an activity; changes and activities are the two ways in which powers are actualised: Since of the actions which have a limit none is an end but all are relative to the end, e.g. the process of making thin is of this sort, and the things themselves when one is making them thin are in movement in this way (i.e. without being already that at which the movement aims), this is not an action or at least not a complete one (for it is not an end); but that in which the end is present is an action. E.g. at the same time we are seeing and have seen, are understanding and have understood, are thinking and have thought: but it is not true that at the same time we are learning and have learnt, or are being cured and have been cured. At the same time we are living well and have lived well, and are happy and have been happy. If not, the process would have had sometime to cease, as the process of making thin ceases: but, as it is, it does not cease; we are living and have lived. Of these processes, then, we must call the one set movements (kinēseis), and the other actualities (energeias). For every movement is incomplete – making thin, learning, walking, building; these are movements, and incomplete movements. For it is not true that at the same time we are walking [to a destination] and have walked [to the destination], or are building and have built, or are coming to be and have come to be – it is a different thing that is being moved and that has been moved, and that is moving [to a location] and that has moved; but it is the same thing that at the same time has seen and is seeing, or is thinking and has thought. The latter sort of process, then, I call an actuality (energeia), and the former a movement (kinēsis). What, and what kind of thing, the actual is, may be taken as explained by these and similar considerations. (Metaphysics 1048b18–36)

Powers are actualised, according to Aristotle, into either activities or changes. The difference between them is that change has a beginning and an end which are different from each other, so completing the realisation of the end requires different stages in a process; while in activity the beginning and the end are the same, in a continuous realisation of the end. Since while change is taking place it has not reached its end point yet, it can be thought of as a potentiality in the process of being actualised, which is how Aristotle thinks about it. Namely, a change is an actuality, because the unfolding realisation of its different stages is happening; but at the same time it is not fully realised, in so far as it has not reached its end yet. In that sense a change is an actual process in progress, realising its remaining potential stages: The actuality of the potential, qua potential, is change – e.g. the actuality of what is alterable as alterable, is alteration; of what is increasable and its opposite, decreasable (there is no common name for both), increase and decrease; of what can come to be and can pass away, coming to be and passing away; of what can be carried along, locomotion. That this

Potentiality in Aristotle’s Metaphysics

33

is what change is, is clear from what follows: when what is buildable, in so far as we call it such, is in fulfilment, it is being built, and that is building. (Physics 201a9–18)

Some confusion might arise in reading the above passage: it might appear that a power is potential before it is actualised, and again potential after it is actualised, as if there is unactualised and actualised potential. To avoid confusion it is important to distinguish between the activation of a power and the completion of the process of its realisation. To use Armstrong’s terminology, a power makes the “journey from potency to act” when it is activated, and remains in actuality during its second “journey” from the beginning of the process of realisation of its end to its completion. Thus, the power of house-building becomes actual when activated at the beginning of the house-building process, and continues to be in actuality until all the stages of house-building are completed. Thus, although in activities the end is reached as soon as the activity is actualised, in the case of changes there are two ends: the actuality of the process itself while it is taking place, and the actuality of the end achieved on completion of that process: While in some cases the exercise is the ultimate thing (e.g. in sight the ultimate thing is seeing, and no other product besides this results from sight), but from some things a product follows (e.g. from the art of building there results a house as well as the act of building), yet none the less the act [of seeing] is in the former case the end and in the latter [house-building is] more of an end than the mere potentiality [to build] is [even if it is less of an end than the completion of the house]. (Metaphysics 1050a24–27)

The contrast is between the potentiality for building a house (when nothing is being built) and the potentiality for a complete house (while a house is being built). In the latter case the potentiality is more of an end than the mere potentiality to build, because the power to build is actualised in the process of building the house, even if the house is not complete yet. This is what the actuality of the potential qua potential is—the actual process of building the house. During the building process, the power to build is as activated (and as actual) as is the power to see when one is seeing. Thus, the actuality of a power is best understood as the activation of the power. When the power is actively doing what it is in its own nature capable of doing, then the power is actualised. Prior to this it exists but in a potential state. Thus the actuality of a power, whether for an activity or a process, is the activation of the power. That which is in the primary sense potential is potential because it is possible for it to become actual; e.g. I mean by ‘capable of building’ that which can build, and by ‘capable of seeing’ that which can see (Metaphysics 1049b12–15)

(By contrast, the actuality of the outcome resulting from a change, e.g. of a house, is something other than—beyond—the change itself.) Thus powers for activities and powers for changes can be either in potentiality or in actuality, when activated. A power in potentiality is the same power as that power in actuality, i.e. when it is activated. The difference between potential and actual power is not a numerical

34

A. Marmodoro

difference. As a consequence, Aristotle’s powers are not relational by each being directed toward another power as its manifestation, which is the usual position in contemporary ontologies of powers.17 In Aristotle’s system, we should understand manifesting as internal to a power, just as a girl’s maturing into womanhood is internal to her. Apart from manifesting, a power is also interdependent with other powers, i.e. active with passive powers. This too is intrinsic to powers, being specified in the definition of a power, which gives the enabling conditions under which the power can be activated. These conditions involve pairing up each power with its correlative power for the possibility of their mutual realisation. Thus, the house-building power depends for its activation on there being buildable material, which will suffer the changes brought about by house-building. The actuality of the house-building power is the activity of building; but this activity essentially requires something external to the power itself in order to obtain, i.e. the availability of materials that have the power to be built. It is this essential dependence of each power on external conditions (such as its correlative power) for its activation that makes Aristotelian powers part of a structured network of powers.18 But being thus structured does not turn Aristotelian powers into polyadic relations, since, as we have seen, Aristotle’s properties are monadic properties. The numerical oneness of a potential power and its activated state—its manifestation—shows that Aristotelian powers in actuality are still powers. It is not the case that a power in actuality becomes inert, losing its powerfulness and turning into a categorical property. E.g. the actualised power to see is exercising seeing, and the actualised power to build is exercising building. In fact, far from being inert, activated powers are at their most powerful while they are exercising their powerfulness. The powerfulness of a power before and after it is triggered into activity is a different matter from the issue of whether a potentiality is preserved when it is actualised (as e.g. in the case of the potentiality to solve mathematical exercises), which we are not examining here. The former issue is whether the powerfulness of a power is essentially tied to the power’s capacity to be triggered into activity—a capacity which is lost when the triggering occurs and the power is actualised (at least the potentiality for this triggering is lost)19; or alternatively, whether powerfulness is not tied to being triggered but is exhibited all along in the exercise of the power after triggering occurs, for as long as the power is activated. My claim is that there is no indication that Aristotle thought that powerfulness is only the 17

See e.g. Bird (2007) Chaps. 5 and 6; Psillos (2006), and my discussion of his paper in Marmodoro (2009). 18 Thus, reality is a structured network of powers; as opposed to an aggregate of powers with no internal connectivity. 19 Even if the potentiality to be actualised is preserved when the power is actualised, as e.g. in the case of mathematical abilities, the claim here is that an actualised mathematical ability is powerful because problems are being solved, not because it can be put to further use to solve problems when another occasion arises.

Potentiality in Aristotle’s Metaphysics

35

potentiality for being triggered. Contemporary power ontologies seem to consider triggering as the only mark of powerfulness—or at least they treat the exercise of a power, while it is bringing about a change, as instantaneous, without duration.20 On the contrary, there is reason to believe that Aristotle thought that the powerfulness of a power is manifested both in its potentiality for activation, and in its exercise while bringing about a change: A thing is capable of causing motion because it can do this, it is a mover because it actually does it. But it is on the movable that it is capable of acting. (Physics 202a16–18, my emphasis)

This is the solution that Aristotle’s power ontology can give to the “Always packing, never travelling” argument. Aristotle’s ontology is not an ontology of potentiality only, because powers continue to be powerful while in actuality. Powers in actuality are the backbone of nature. Powers in potentiality are grounded on activated powers. There is no need, therefore, to introduce categorical properties into the ontology for the sake of grounding powers in potentiality; powers in actuality provide this support. Consequently, there is no need for Aristotle to introduce anything other than pure powers to ground his ontology. His is a world of powerful potentialities and actualities. The actuality of a power is its activation, bringing about activity (e.g. seeing), or change (e.g. house-building). The activation of a power requires its correlative (active or passive) power, where both powers are monadic and yet interdependent properties. We saw above that what triggers the activation of correlative powers is contact between them, when other conditions relevant to those specific kinds of powers obtain. Thus, the mutual actualisation of two powers is the activation of two monadic properties that are dependent on one another; each of them undergoes a transition from potentiality to actuality, while remaining the very same monadic property. Hence, the metaphysics of Aristotle’s nature is at the bedrock of reality a metaphysics of monadic primary powers, which together comprise a structure of interdependent actualities and potentialities.

9 The Potentiality of the Four Causes Our examination so far has focused on powers that are the origins of change— movers that act on movables. We have found that all such powers are monadic interdependent properties, in potentiality or in actuality. The primary powers, to which the other powers are reducible,21 interact with each other in the cyclical

20

See for example Martin (2008: 51), discussed by Mumford in his contribution to the present volume ‘Mutual manifestation and Martin’s two Triangles’. 21 See discussion of Generation and Corruption 330a24–29, p. 13 above.

36

A. Marmodoro

transformations of the primary elements they constitute,22 thereby making up a structure of interdependent powers that is the foundation of all there is in nature. In his Physics and the Metaphysics, Aristotle developed his theory of the Four Causes, which account for the constitution of each thing in nature. The four causes are the principles that explain what substances are, and give a scientific understanding of the nature and makeup of each substance, namely what the physicist or the biologist will be looking for in studying the substances in their respective domains. The four causes are: the material cause, the formal cause, the final cause and the efficient cause of a substance. So far in our discussion we have focused on the efficient causes, namely physical powers, which Aristotle explained in terms of potentiality and actuality. We shall now follow Aristotle in examining the other three types of cause, to see what it is that they have in common, as causes, with the efficient causes, and whether this extends the notion of potentiality to apply to these further causes, too. In Metaphysics Book IX Aristotle applies the account he gave of efficient causes in terms of potentiality and actuality to explain the other two principles of being: matter and form (material and formal causes). With this achieved, potentiality and actuality become the most fundamental principles of being in terms of which the metaphysics of nature, according to Aristotle, is to be understood. Aristotle introduces the discussion in Metaphysics Book IX by giving in outline the research questions he will pursue: Since ‘being’ is in one way divided into ‘what’, quality, and quantity, and is in another way distinguished in respect of potentiality and fulfilment, and of function, let us now add a discussion of potentiality and fulfilment. First let us explain potentiality in the strictest sense, which is, however, not the most useful for our present purpose. For potentiality and actuality extend further than the mere sphere of motion. But when we have spoken of this first kind, we shall in our discussions of actuality explain the other kinds of potentiality. (1045b32–1046a4)

Aristotle extends his discussion of potentiality and actuality from the causes of change to matter and form, and through the issue of form, to all the different kinds of being, in the second half of Book IX, Chaps. 6–10. He does not offer conceptual analyses of potentiality and actuality. Rather he utilises many different kinds of example through which he builds an analogy between the primary sense of potentiality and actuality and their extended senses. The primary one is the sense in which potentiality and actuality apply to efficient causation, relating to the domain of change, which Aristotle explains in Book IX, Chaps. 1–5 and his other works. His analogies are the following:

22 ‘All the other things—the things, I mean, which are reciprocally transformed in virtue of their qualities and their powers, e.g. the simple bodies—imitate circular motion. For when Water is transformed into Air, Air into Fire, and the Fire back into Water, we say the coming-to-be has completed the circle, because it reverts again to the beginning’. Generation and Corruption, 337a2–6.

Potentiality in Aristotle’s Metaphysics

37

Since we have treated of the kind of potentiality which is related to movement, let us discuss actuality, what and what sort of thing it is. In the course of our analysis it will also become clear, with regard to the potential, that we not only ascribe potentiality to that whose nature it is to move something else, or to be moved by something else, either without qualification or in some particular way, but also use the word in another sense, in the pursuit of which we have discussed these previous senses. … Our meaning can be seen in the particular cases by induction, and we must not seek a definition of everything but be content to grasp the analogy,– that as that which is building is to that which is capable of building, so is the waking to the sleeping, and that which is seeing to that which has its eyes shut but has sight, and that which has been shaped out of the matter to the matter, and that which has been wrought to the unwrought. (1048a25–b4)

The matter Aristotle is talking about here can be understood either as the original matter from which a substance is made, such as the logs of wood from which a ship is made, or the underlying matter of the ship which is the entity that remains if we abstract away the form of the ship. Further on, Aristotle explains that: Matter exists in a potential state, just because it may attain to its form; and when it exists actually, then it is in its form. (1050a15–16)

The form is any of the types of being that would fall under the categories of being, in terms of which Aristotle explains the changes that take place in generation and destruction or in alteration. Clearly the sense in which matter is in a specific form, whether Aristotle is talking of original matter or underlying matter, is by being en-formed by this form; namely, the form is instantiated in that matter. One important feature of the potential is that it can be triggered into actuality. Aristotle does not mention that this feature of the potential does not apply to matter as potential. Neither the logs of wood from which the ship is made, nor the wood constituting the ship are triggered into becoming en-formed, in the way that powers which are efficient causes are. Aristotle does indeed tell us that the extended sense of potentiality and actuality is different from the primary one, but he does not specify what the differentiating feature is. Yet, this feature differentiates in an important way the potentiality of, say, the wood in the ship from the potentiality of fire. If then it is not through the role that contact plays, or more generally through the role of the stimulus conditions in efficient causation, that the two kinds of cause, matter and form, can be shown to be cases of potentiality, what is it that they have in common with efficient causes? Aristotle does not specify what the grounds of the analogy between the two types of potentiality and actuality are. But I will here propose what I take them to have in common, which allows Aristotle to talk of potentiality and actuality in relation to the Four Causes generally. The feature I want to focus on is that of incompleteness, which Aristotle introduces in Book IX Chap. 6. He uses it to distinguish change from activity: Of these processes, then, we must call the one set movements, and the other actualities. For every movement is incomplete – making thin, learning, walking, building; these are movements, and incomplete movements. For it is not true that at the same time we are walking and have walked, or are building and have built, or are coming to be and have come to be – it is a different thing that is being moved and that has been moved, and that is

38

A. Marmodoro moving and that has moved; but it is the same thing that at the same time has seen and is seeing, or is thinking and has thought. (1048b28–34)

But Aristotle explains the incompleteness of change as an instance of the incompleteness of potentiality: Motion is thought to be a sort of actuality, but incomplete, the reason for this view being that the potential whose actuality it is is incomplete. (Physics 201b31–33, my emphasis)23

This is Aristotle’s key conception of potentiality: incompleteness. It is hard to overemphasise the importance of this tenet within Aristotle’s metaphysics. It is a conception that will allow us to understand fundamental aspects of his system, and in particular aspects that we find counterintuitive from our contemporary perspective. The most fundamental role of incompleteness in Aristotle’s system is that it helps us understand the ‘dynamism’ associated with the notion of potentiality. Aristotle explains the notion of the incompleteness of a process as incompleteness in achieving a task that is achievable in time: Since of the actions which have a limit none is an end but all are relative to the end, e.g. the process of making thin is of this sort, … (i.e. without being already that at which the movement aims), this is not an action or at least not a complete one (for it is not an end); but that movement in which the end is present is an action. E.g. at the same time we are seeing and have seen, … while it is not true that at the same time we … are being cured and have been cured. (1048b18–25)

It is easy enough to understand the incompleteness of a process that has different stages of development, while it has not reached the end. What makes it easy is that the process that has been achieved thus far is self-standing. Even if the process stops midway, it can stand incomplete, as the completeness is one that is achieved over time. But it is more difficult to understand cases of incompleteness in which what is incomplete cannot be observed but is accessible only through abstraction. Such is the case of constitutional incompleteness, e.g. of matter and form in a substance, which do not occur in nature in the way that a half-built house does, but always as composites of matter and form.24 Nevertheless, Aristotle’s account of potentiality requires that we comprehend the incompleteness of substances when conceived of without one of their constitutive principles.

23

Aristotle makes the same point verbatim in the Metaphysics: Movement is thought to be an actuality, but incomplete; the reason is that the potential, whose actuality it is, is incomplete. (1066a22–23)

Cf. also Physics 257b9: ‘Motion is an incomplete fulfilment of the movable’. 24 The matter, such as the marble for a statue or the pieces of wood for a house, are not the matter that remains when we abstract the form of a statue or a house, because lumps of marble and planks of wood do have forms of their own, even if they are thought of as privations (because non-substantial). The matter that remains is an abstract entity.

Potentiality in Aristotle’s Metaphysics

39

Globally in his system, we find Aristotle using the notion of potentiality in relation to two cases: for the process of change, whether generation or alteration; and for the constitution of an entity. In cases of change, the process can be observed unfolding in time as it is completed, by achieving its stages of development. But in the cases of constitution, incomplete entities such as constituent matter and form have to be conceptualised through abstraction of the primary principles making up the entity. In the cases of generation, incompleteness can be understood in time rather than through abstraction, as the incompleteness of the form of a child. Further cases—and possibly the paradigmatic cases—of constitutionally incomplete entities are the efficient causes we have been examining above. They exist incompletely as potentialities in nature; they are completed constitutionally by contact with their correlate powers in the right circumstances, at which stage they are actualised by being activated into exercising their powerfulness. Thus the power of heat is potentially a heater, being incomplete by lacking the form of heating, but it comes to be actually en-formed by the form of heating when a heatable object comes in contact with it. Efficient causal powers are not observable in their incomplete, potential, state; for, observation is causal interaction, which activates the powers. In the case of the constitutional incompleteness of the matter and form of a substance (the topic to which the second half of book IX of the Metaphysics is devoted), matter and form can be thought of as being potential—not because they are dynamic in the sense that they can bring about change; but because their constitutional state when considered by themselves is a state of incompleteness (matter lacking form, or form lacking matter), and as such it is a state that is in need of completion. We can comprehend the incompleteness of the matter or form of a substance by abstracting the corresponding constituents of the substance; the corresponding abstract entities of matter and form need completion to occur in nature.25 The significance of incompleteness in Aristotle’s metaphysics is, I submit, far greater than has so far been appreciated. On account of the notion of constitutional incompleteness, the focus of the inquiry into the power to change shifts: from change being a primitive capacity in things (e.g. between opposites), to an explanation of change being fuelled by the mutual dependence between the building blocks of reality on each other. The constitutional incompleteness of potentiality, I argue, is the key to Aristotle’s conception of ontological dependence. It is why a power depends on a correlative power to receive its form and change26; why matter depends on form to be qualified; why form depends on matter to be instantiated; why a master depends on a slave to rule, etc. In all cases, a single item (concrete or

25

There are secondary senses of lacking e.g. form, as in the case of an object whose form is a privation, e.g. a lump of bronze, but I will not expand the discussion in this area. 26 Although there is nothing physical that is transmitted between the powers, the reception of the form of the active power by the passive power indicates the respective metaphysical incompleteness of the powers, which is exhibited as the efficacy of the active power on the passive power.

40

A. Marmodoro

abstract) is tied to another single item through the need to constitutionally complete one another. Incompleteness is the key to understanding Aristotle’s final causation in a way which undermines the criticisms his account has received in modern philosophy, on account of allowing for backward causal efficacy. On Aristotle’s account of final causation,27 as I propose to understand it, the final end is not reified so as to be efficaciously operative, backwards, on an entity or process which itself is directed towards its final end. Rather, the final end represents the type of incompleteness that the entity or the process has. The incompleteness is relative to an occurrent capacity; the capacity is present in the entity or process themselves, or in their environment. For instance, the final end of a human embryo is the state of a mature human being that the embryo is aiming toward; the embryo is incompletely developed by comparison to that state; yet, there is a capacity in the movements that the sperm is transferring to the embryo (deriving from the parent’s movements) to change and develop the embryo into a mature human being. In Metaphysics IX Aristotle says: Nature also is in the same genus as potentiality; for it is a principle of movement – not, however, in something else, but in a thing itself qua itself. (1049b8–10)

So the final end represents the respect in which the entity or process that has this end is incomplete; their incompleteness is relative to an occurrent capacity for the final end, whether this capacity is their nature, or in their environment.28 In his discussion of potentiality and actuality, Aristotle considers different types of entity that are potential and can be actualised. The primary one is a power in potentiality and its activated state. We saw that a power in potentiality is incomplete with respect to the form of the activity it is directed towards, e.g. heating, which the power comes to actually possess when activated. The activation of such a power is an instance of efficient causation. Further types of entity Aristotle examines in his discussion of potentiality and actuality are the constituent matter, and the form that en-forms it in a substance. These are potential in a different sense than the sense in which potentiality applies to change: All things are not said in the same sense to exist actually, but only by analogy… for some are as movement to potentiality, and the others as substance to some sort of matter. (1048b4–6)

This different sense of actuality and potentiality applies to the non-efficient causes. We saw above that on this extended sense, the potential is not triggered into actuality. But what is common between this analogous sense of potentiality and the

27

E.g. that every coming to be is for the sake of an end (See Metaphysics 1050a7–9). Generally speaking, there may be many—even arbitrary—ways in which something incomplete can be completed. Aristotle’s teleology selects one of those ways as describing the type of incompleteness an entity or a process has on the basis of the systematically occurring capacities present in the entity or process to complete entities or processes of that type, if nothing prevents.

28

Potentiality in Aristotle’s Metaphysics

41

sense in which efficient causes are potential is that they all have some kind of incompleteness. Thus, matter can be thought of as being potential: Matter exists in a potential state, just because it may attain to its form; and when it exists actually, then it is in its form. (1050a15–16)

Read constitutively, this claim tells us that the matter of a substance, when considered in abstraction without its form, is potential because it is incomplete, lacking form. It is actualised when it has its form, thus constituting the substance. The original matter of a substance (rather than the abstracted matter which constitutes the substance), when considered as such, is also potential; thus for example, an embryo which develops into a human being. The incompleteness of the embryo is its lack of the form of a human being. Since the embryo is a concrete particular, as opposed to the abstracted matter of a substance, the sense in which it lacks the form of a human being is different from the sense in which the abstracted flesh and bones considered in abstraction do. Nevertheless, the embryo is the original matter from which the human being is generated, and as such it is potentially the human being into which it will be transformed. Matter as potential then, whether original matter or abstracted matter, that comes to be in its form when in actuality, is an instance of material causation. Analogously, the form, considered in abstraction as non-enmattered, is incomplete too, and thus in a potential state, in that it can come to be enmattered. It should not be thought that the fact that Aristotle speaks of the form as an actuality conflicts with the potentiality status of the form; not any more than the status of a change being in actuality conflicts with its potentiality status (Physics 201a10–11). The sense in which a form is an actuality is compatible with its being in potentiality, when considered in abstraction as non-enmattered; the form is an actuality because it stands for what a substance is, as stated in the substance’s definition, whether the form is enmattered or not; while the form is in actuality because it is enmattered. Form as potential then, which comes to be in its matter when in actuality, is an instance of formal causation or final causation: Why are these things, i.e. bricks and stones, a house? Plainly we are seeking the cause. And this is the essence (to speak abstractly), which in some cases is the end, e.g. perhaps in the case of a house or a bed, and in some cases is the first mover; for this also is a cause. But while the efficient cause is sought in the case of genesis and destruction, the final cause is sought in the case of being also. (Metaphysics, 1041a26–32)

We see therefore that the four causes all involve some form of metaphysical incompleteness, which accounts for their potential status, and explains what it is for each one of them to attain actuality. There are significant differences between the four causes, and in particular, between those that involve change, and those that involve abstraction from the constitution of substances. There are corresponding differences in the types of incompleteness in each case; in the dependence on internal or on external factors for their completion; in the completeness being achieved by change or by being constituted, involving development or abstraction, and so on. But what is significant for understanding their role as causes is that in

42

A. Marmodoro

each case, there is a sense in which there is a metaphysical incompleteness, which grounds potency, which explains their causal role. Thus in Aristotle’s system, all the fundamental principles of being are to be thought of as instances of some type of incompleteness, and therefore of potency, with the corresponding ontological interdependencies between them.

Bibliography

Aristotle’s texts are all quoted from Barnes, J. (Ed.). (1995). The complete works of Aristotle. Princeton: Princeton University Press.

Other works cited Armstrong, D. M. (1997). A world of states of affairs. Cambridge: Cambridge University Press. Bennett, K. (2011). Construction area (no hard hat required). Philosophical Studies, 154, 79–104. Bird, A. (2006). Potency and modality. Synthese, 149, 491–508. Bird, A. (2007). Nature’s metaphysics. Oxford: Oxford University Press. Couchman J. P. (2000). Trilinear gauge boson couplings and W-Pair production. In: A measurement of the triple gauge boson couplings and W boson polarisation in W-pair production at LEP2, Ph.D. thesis, University College London. Available at http://www.hep. ucl.ac.uk/*jpc/all/ulthesis/node40.html. Ellis, B. (2001). Scientific realism. Cambridge: Cambridge University Press. Ellis, B. (2010). Causal powers and categorical properties. In A. Marmodoro (Ed.), The metaphysics of powers: Their grounding and their manifestations. New York: Routledge. Heil, J. (2003). From an ontological point of view. Oxford: Oxford University Press. Jacobs J. D. (ed.). (2017). Causal powers. Oxford: Oxford University Press. Makin, S. (2006). Aristotle: Metaphysics book H. Oxford: Clarendon Press. Marmodoro, A. (2009). Do powers need powers to make them powerful? From pandispositionalism to Aristotle. History of Philosophy Quarterly, 26(4), 337–352. Marmodoro, A. (Ed.). (2010). The metaphysics of powers: Their grounding and their manifestations. New York: Routledge. Marmodoro, A. (2013). Aristotle’s hylomorphism without reconditioning. Philosophical Inquiry, 37(1/2), 5–22. Marmodoro, A. (2014). Aristotle on perceiving objects. Oxford: Oxford University Press. Marmodoro, A. (2017). Aristotelian reciprocity without symmetry in causation. In J. D. Jacobs (Ed.), Causal powers. Oxford: Oxford University Press, 57–76. Martin, C. B. (1993). The need for ontology: Some choices. Philosophy, 68(266), 505–522. Martin, C. B. (2008). The mind in nature. Oxford: Oxford University Press. Molnar, G. (2003). Powers: A study in metaphysics. Oxford: Oxford University Press. Mumford, S. (1998). Dispositions. Oxford: Oxford University Press.

Potentiality in Aristotle’s Metaphysics

43

Mumford, S. (2010). Ascribing dispositions. In G. Damschen, R. Schnepf & K. Stüber (Eds.), Debating dispositions: Issues in metaphysics, epistemology, and philosophy of mind (pp. 168–185). Berlin: DeGruyter. Mumford, S., & Anjum, R. L. (2011). Getting causes from powers. Oxford: Oxford University Press. Psillos, S. (2006). What do powers do when they are not manifested? Philosophy and Phenomenological Research, 72(1), 137–156.

Author Biography Anna Marmodoro holds the Chair of Metaphysics in the Department of Philosophy at the University of Durham, United Kingdom, and is concomitantly a Research Fellow and an Associate Faculty Member in Philosophy at the University of Oxford. She specializes in two research areas: metaphysics on the one hand, and ancient, late antiquity and medieval philosophy on the other. She has also strong research interest in the philosophy of mind and the philosophy of religion. Anna has created and directed for the past ten years in Oxford a large-scale multidisciplinary research group funded by the European Research Council, the Templeton World Charity Foundation and other funders. Her recent publications include “Aristotle on Perceiving Objects” (Oxford University Press 2014), “Everything in Everything. Anaxagoras’s Metaphysics” (Oxford University Press 2017), “The Metaphysics of Relations” (Oxford University Press 2016, coedited with David Yates).

Potentiality in Aristotle’s Physics and Biology Stephen Makin

The most essential differences between bodies are differences in properties and functions and powers. ἐpeὶ dὲ jtqiώsasai diaφoqaὶ sῶm rxlάsxm aἵ se jasὰ sὰ pάhη jaὶ sὰ ἔqca jaὶ sὰ1 dtmάlei1. — Aristotle, de Caelo 3.8, 307b19–20

1 Physics, Biology and Psychology This is a paper about the use Aristotle makes of the notion of potentiality.1 But it is limited to certain areas of Aristotle’s thought—physics and biology—while Aristotle himself also uses the notion in other contexts. The preceding chapter of this volume was on Aristotle’s metaphysics, and the chapter following is on Aristotle’s theory of the soul (his psychology) and his ethics. However the contrast between Aristotle’s appeal to potentialities in physics and biology and his use of the notion in psychology in particular calls for some comment. The notion of a potentiality is certainly integral to Aristotle’s treatment of the central areas of psychology. The account of the soul in de Anima 2.1 is cast in terms

1

For readers who are non-specialists in Aristotle, there are a range of introductions and surveys to consult. For two excellent volumes see Lear (1988) and Shields (2007). Highly recommended online resources are the Routledge Encyclopedia of Philosophy article on Aristotle by Terry Irwin, and the Stanford Encyclopedia of Philosophy entry on Aristotle by Christopher Shields. For a volume of non-introductory reflections on different aspects of Aristotle’s philosophy see Shields (2013). S. Makin (&) University of Sheffield, Sheffield, UK e-mail: s.makin@sheffield.ac.uk © Springer Science+Business Media B.V., part of Springer Nature 2018 K. Engelhard and M. Quante (eds.), Handbook of Potentiality, https://doi.org/10.1007/978-94-024-1287-1_3

45

46

S. Makin

of the distinction between actuality and potentiality.2 Indeed casting an account of the soul in that way is claimed to render unnecessary the question of whether soul and body are or are not one and the same.3 Much of the detailed and nuanced work in de Anima proceeds with reference to the notions of what is potential and what is actual. The discussion in de Anima 2.5, which seeks to adapt the general account of action developed in Generation and Corruption 1.7 to the case where observable entities in the world act on appropriately placed and structured subjects to give rise to perceptions, turns on a careful contrast between different levels of potentiality and actuality.4 The account of the different senses at de Anima 2.7–11, with a general summary at de Anima 2.12, relies heavily on the notions of what is potential and what is actual.5 The same notions are pressed into service in Aristotle’s account of thought and knowledge.6

2 See for example de An 2.2, 412a27–28 ‘That is why the soul is an actuality of the first kind of a natural body having life potentially in it. The body so described is a body which is organised’. All translations of Aristotle’s texts in this chapter are taken from the Revised Oxford Translation of Aristotle’s works (see Aristotle 1984). For the convenience of those who are not specialists in Aristotelian studies I have generally quoted (often in footnotes) translations of Aristotelian passages to which I have referred. These translations invariably reproduce the Revised Oxford Translation. In addition to the Revised Oxford Translation there are also alternative individual translations of a number of the texts I will refer to, along with extended philosophical commentaries, in the Clarendon Aristotle series. For details see references in the bibliography to Aristotle Physics I and II translated with commentary by Charlton (1992); Aristotle Physics III and IV translated with commentary by Hussey (1983); Aristotle Physics VIII translated with commentary by Graham (1999); Aristotle De Generatione et Corruptione translated with commentary by C. J. F. Williams; Aristotle On The Parts of Animals I–IV translated with commentary by Lennox (2001); Aristotle Metaphysics Book H translated with commentary by Stephen Makin. 3 de An 2.2, 412b3–9 ‘If then we have to give a general formula applicable to all kinds of soul, we must describe it as an actuality of the first kind of a natural organised body. That is why we can dismiss as unnecessary the question whether the soul and the body are one; it is as though we were to ask whether the wax and its shape are one, or generally the matter of a thing and that of which it is the matter. Unity has many senses (as many as ‘is’ has), but the proper one is that of actuality.’ 4 This is often presented as a contrast between first level potentiality (the sense in which a human baby is a potential knower of grammar, in virtue of being human, and by contrast with e.g. a kitten), second level potentiality (the sense in which a educated yet silent adult human is a potential knower of grammar, in that they could speak and put their knowledge into action if they so chose; some commentators have attached the name ‘first actuality’ to this level, although it’s not clear that label has any Aristotelian credentials), and second level actuality (the educated human actually speaking and using/manifesting their grammatical knowledge). There is a recent literature on this chapter of de Anima following on from Burnyeat (2002), see also Heinaman (2007) and Bowin (2011). 5 See for example de An 2.11, 423b30–424a3 ‘The organ for the perception of these [hot, cold, dry, moist] is that of touch—that part of the body in which primarily the sense of touch resides. This is that part which is potentially such as its object is actually; for all sense-perception is a process of being so affected; so that that which makes something such as it itself actually is makes the other such because the other is already potentially such.’ 6 See de An 3.8, 431b24–28 ‘Knowledge and sensation are divided to correspond with the realities, potential knowledge and sensation answering to potentialities, actual knowledge and sensation to

Potentiality in Aristotle’s Physics and Biology

47

Further, there are various places in the Aristotelian corpus where Aristotle offers a schematic outline of what he intends his discussion of the physical sciences to cover. One significant passage is Meteorology 1.1. That opening chapter of Meteorology looks back over the Physics, de Caelo and Generation and Corruption.7 After outlining the topics to be covered in the study of meteorology,8 Aristotle points forward to the investigation of plants and animals—including, presumably, human animals—and the study of the latter will involve both biology and psychology. Once all that is done the investigation—which is presumably the investigation started at the beginning of the Physics—will have been completed.9 So we see that natural science for Aristotle incorporates questions about the nature of time or the existence of void, accounts of the four sub-lunar material elements, the structure and behaviour of the heavens, and the full range of psychological capacities possessed by animals from the simplest up to humankind.10 The first book of Parts of Animals is also clear that psychology, the study of the soul, falls within natural science, so that physics incorporates both biology and psychology.11

actualities. Within the soul the faculties of knowledge and sensation are potentially these objects, the one what is knowable, the other what is sensible. They must be either the things themselves or their forms. The former alternative is of course impossible: it is not the stone which is present in the soul but its form.’ 7 Meteorology 1.1, 338a20–23 ‘We have already discussed the first causes of nature, and all natural motion, also the stars ordered in the motion of the heavens, and the corporeal elements—enumerating and specifying them and showing how they change into one another—and becoming and perishing in general.’ 8 Meteorology 1.1, 338a23–338b21 ‘There remains for consideration a part of this inquiry which all our predecessors call meteorology. It is concerned with events that are natural, though their order is less perfect than that of the first of the elements of bodies.’ 9 Meteorology 1.1, 339a6–9 ‘When the inquiry into these matters [thunderbolts, whirlwinds etc.] is concluded let us consider what account we can give, in accordance with the method we have followed, of animals and plants, both generally and in detail. When that has been done we may say that the whole of our original undertaking will have been carried out’. 10 At the opening of de Anima Aristotle is explicit that the study of nature incorporates (as a valuable part) the study of the soul; see de An 1.1, 402a1–7 ‘Holding as we do that, while knowledge of any kind is a thing to be honoured and prized, one kind of it may, either by reason of its greater exactness or of a higher dignity and greater wonderfulness in its objects, be more honourable and precious than another, on both accounts we should naturally be led to place in the front rank the study of the soul. The knowledge of the soul admittedly contributes greatly to the advance of truth in general, and, above all, to our understanding of nature.’ 11 Part An 1.1, 641a17–26 ‘If now the form of the living being is the soul, or part of the soul, or something that without the soul cannot exist; as would seem to be the case, seeing at any rate that when the soul departs, what is left is no longer an animal, and that none of the parts remain what they were before, excepting in mere configuration, like the animals that in the fable are turned into stone; if, I say, this is so, then it will come within the province of the natural scientist to inform himself concerning the soul, and to treat of it, either in its entirety, or, at any rate, of that part of it which constitutes the essential character of an animal; and it will be his duty to say what a soul or this part of a soul is; and to discuss the attributes that attach to this essential character, especially as nature is spoken of—and is—twofold, as matter and as substance; nature as substance including both the motor cause and the final cause.’

48

S. Makin

However I am going to concentrate in this chapter on Aristotle’s use of the notion of potentiality in physics and biology narrowly conceived, namely in such a way as to exclude the discipline which we would now distinguish as psychology. This narrowing should not matter, though, since the aim in what follows isn’t to offer a comprehensive or summary report of Aristotelian explanations of specific physical or biological—or indeed psychological—phenomena. But given that my main interest isn’t in the way in which Aristotle appeals to the notion of a potentiality in explaining specific phenomena, then what is there to say about Aristotle’s use of the notion of potentiality in physics and biology?

2 Potentiality: Three Types of Appeal Aristotle makes a lot of references to potentiality and related notions in the course of his physics and biology, and it would perhaps be helpful to bring some order to the various types of appeal that he makes. A rough distinction can be drawn between three ways in which the notion of a potentiality enters into Aristotelian natural science. First there are ground level explanations of specific physical phenomena. For example, oily materials are flammable because they possess a particular passive potentiality (Metaphysics H 1, 1046a24–25). Children inherit characteristics from both their paternal and maternal ancestral lines because there are various potentialities which are either carried by a father’s seed or which fail to be mastered by a father’s seed (GA 4.3). The sea is warm because everything which is exposed to fire contains heat potentially, and the sea is exposed to the sun (Meteorology 2.3, 358b8–9). Different colours come about because there is a mixing of white and black, which mixing is conceived on the model of different constituents being mixed together and present potentially in a complex material stuff.12 Ground level explanations such as these invite two questions from the contemporary reader. The first is whether each is a good explanation of the phenomenon it is intended to explain. The second is whether there is some general barrier to this type of ground level explanation being satisfying. I’m not going to say much about the first question, mainly because that sort of question is as close as you might like to being a straightforward physical enquiry Our interest is not in Aristotelian physics and biology as live contributions to physics and biology. I doubt anyone is going to deny that we typically have better explanations of a wide range of physical and biological phenomena than were available to Aristotle. But the second question is a very different matter. Some may suspect that, whatever the details, no ground level explanation of a physical or biological phenomenon which is cast in terms of potentialities could possibly be a satisfying explanation. Indeed it is perhaps this

12

de Sensu 3, 440a31–440b4. For Aristotle’s account of the mixing of simple physical ingredients into complex material stuffs see GC 1.10 (with further comment on this issue to follow below).

Potentiality in Aristotle’s Physics and Biology

49

suspicion which leads one to think that not much is lost by failing to look at the details of any specific Aristotelian explanation cast in these terms. This sort of general concern about ground level explanations in terms of potentiality is well known and important, and I will say more in detail about it presently.13 Second, there are the uses Aristotle makes of the notion of a potentiality in providing analyses of concepts which are crucially important to the framework of physics and biology. There may be some artificiality in separating this level of appeal to potentialities from the appeal to potentialities in providing ground level explanations of specific phenomena. For explanations of particular phenomena tend to generalise. Suppose I ask why Deborah has black hair, when neither her father nor mother do. Aristotle’s answer will rely on a general explanation of how it is that children can inherit characteristics through their maternal ancestral line. Now at that point we are still within the realm of explaining phenomena, whether specific or general. The second type of use Aristotle makes of the notion of a potentiality appears not here, when moving from specific to general explanations, but rather at the point at which we move from using explanatory notions in accounting for phenomena to analysing concepts which are fundamental to physics and biology. For example, what is characteristic of the realm of physics for Aristotle is that physical things move and undergo change. So something is going to remain unclear about physics if we can’t provide an account of what change and motion are. Aristotle does indeed provide such an account—albeit one which is notoriously opaque—in Physics 3.1, and it is in terms of the notion of a potentiality.14 Or again, it is fundamental to Aristotle’s biology that living things are nourished and grow. Aristotle offers an account of nourishment and growth in GC 1.5, and that account involves the contrast between, for example, what is potentially flesh and what is actually flesh. If there was no coherent account to be provided of the phenomenon of growth then biology would be in serious trouble. And Aristotle’s account of growth is built on his account of another basic physical phenomenon: the combination or mixture (what Aristotle calls mixis) of simple ingredient materials into new more complex stuffs.15 If there were no such combination or mixture then there

13

See Sect. 5. See Physics 3.1, 201a10–12 ‘We have distinguished in respect of each class between what is in fulfilment and what is potentially; thus the fulfilment of what is potentially, as such, is motion’; also 201a28–31 ‘It is the fulfilment of what is potential when it is already fulfilled and operates not as itself but as movable, that is motion. What I mean by ‘as’ is this: bronze is potentially a statue. But it is not the fulfilment of bronze as bronze which is motion. For to be bronze and to be a certain potentiality are not the same.’ 15 See the following two passages from GC 1.5: 321a33–321b2 ‘Perhaps the explanation is that the substance of the one remains unchanged, but the substance of the other (viz. of the food) does not. For indeed, even in the mixture of wine and water, it is the prevailing ingredient which is said to have increased in volume. We say e.g. that the wine has increased, because the whole mixture acts as wine but not as water. A similar principle applies to alteration.’ GC 1.5: 322a4–13 ‘One might discuss what must be the character of that whereby a thing grows. Clearly it must be potentially that which is growing—potentially flesh e.g. if it is flesh that is growing. Actually, therefore, it must be other than the growing thing. This then has passed-away and come-to-be flesh. But it has 14

50

S. Makin

would be a real sense in which there were only four basic types of stuff in the physical universe (the four Aristotelian elements: earth, air, fire and water). Any other apparent types of stuff would be mere agglomerations of discrete portions of those basic elements—as when dry sugar and salt are shaken together: they do not mix and produce a new stuff, but we have just an agglomeration of sugar and salt particles, adjacent to, but not mixing and combining with, one another. Absent genuine mixture there wouldn’t really be complex organic materials such as wood, flesh, blood or bone. Aristotle offers an account of combination mixture in GC 1.10, and that account also relies centrally on the contrast between what is potential and what is actual.16 Third and finally there is the explicit and carefully developed account of a potentiality connected with change that Aristotle offers in Metaphysics H 1–5—his canonical account of potentialities for change. For the present there are two points to note about Aristotle’s analysis in Metaphysics H of the notion of a potentiality. The first is that the notion of a potentiality is analysed in a way which twins it with the notion of a nature (a phusis), the potentiality-nature pairing being definitive of the very discipline of physical science. And the second is that it is the explicit analysis of the notion of a potentiality in Metaphysics H 1–5 which generates the important constraints on that notion which allow Aristotle to avoid the charge that there is something intrinsically vacuous about explanations framed in terms of potentialities.17 I will say something about the first of these points in the following section, and return to the second in Sect. 5.

not been transformed into flesh alone by itself (for that would have been a coming-to-be, not a growth); rather, the growing thing has done so by the food. In what way, then, has the food been modified by the growing thing? Perhaps we should say that it has been mixed with it, as if one were to pour water into wine and the wine were able to convert the new ingredient into wine. And as fire lays hold of the inflammable, so the active principle of growth, dwelling in the growing thing (i.e. in that which is actually flesh), lays hold of an acceding food which is potentially flesh and converts it into actual flesh.’ 16 See GC 1.10, 327b23–27 ‘Since however some things are potentially while others are actually, the constituents can be in a sense and yet not-be. The compound may be actually other than the constituents from which it has resulted; nevertheless each of them may still be potentially what it was before they were combined, and both of them may survive undestroyed’. 17 Aristotle’s main purpose in Metaphysics H is not, however, to analyse the notion of a potentiality for change for its own sake, or for the sake of the light it casts on the physical sciences. He has a much broader concern which takes him beyond the realm of physical science and on to more abstract metaphysical issues. I am not for the most part going to say anything about those issues here.

Potentiality in Aristotle’s Physics and Biology

51

3 Potentiality: The Terminology and the Canonical Account So much for marking out the concerns of this chapter. Now to more detail. The first issue to address is that of terminology. Where does the word ‘potentiality’ come from? Anyone familiar with the notions clustered round this area of thought is likely to be aware of a range of terms in English. There is a noun potentiality (‘fire has the potentiality to heat’; ‘this book is called Handbook of Potentiality’). There is another noun potential (‘she doesn’t have any potential for the job’; ‘the situation in Iran has the potential to trigger a serious international conflict’). There is the same term potential used as an adjective (‘she’s a potential head of department’; ‘that’s a potential venue for my bridal party’). And there is an adverb potentially (‘the situation is potentially dangerous’). It may also be well known that these terms (at least in their developed use) have their origin in Aristotelian philosophy. So a natural question for the non-ancient-specialist to ask is: how do those (English) terms relate to the (Greek) terminology from which they originate? ‘Potentiality’ and its cognates are derived from Latin, and the Latin in turn translates Aristotle’s Greek. Aristotle has three related terms. First there is a noun dunamis (‘might, strength; a power, an ability, a capacity, a faculty; a potentiality’). Second the dative case of that noun (dunamei) serves as an adverb (‘potentially). Finally, there is a related adjective dunaton (‘capable, able; potential’). In addition the notion of a potentiality comes as one side of a dichotomy: there is the contrast between what is potential and what is actual, what has the capacity to do something and what is actively doing that thing, what obtains in potentiality and what obtains in actuality. These contrasts cover a great deal for Aristotle, and the notions of what is actual or active are the subject of much scholarly debate, although for the most part I will bypass that debate here.18 As noted above, Aristotle’s richest theoretical explication of the notion of a potentiality for change is to be found in Metaphysics H 1–2 and 5. He makes the typically Aristotelian observation that the notion exhibits considerable complexity,19 and then focuses on the concept of potentiality which is correlated with change. That concept itself is then analysed in terms of a core case, that of an active potentiality (or, in more natural English, an active capacity).20 An active potentiality is what it is about one thing which explains how it is that that type of thing

18

Aristotle uses two neologisms energeia and entelecheia for the contrast with dunamis/potentiality. For some summary remarks on these neologisms see Makin (2006 pp. xxvii–xxx). For considerably more detail see Menn (1994). For a recent extended discussion see Beere (2009), especially Chapters 8 and 9. 19 Metaphysics H 1, 1046a5–6 ‘We have pointed out elsewhere that ‘potentiality’ and the word ‘can’ have several senses.’ The reference is to Metaphysics D 12. 20 Metaphysics H 1, 1046a9–11 ‘But all potentialities that conform to the same type are starting points (archai) and are called potentialities in reference to one primary kind, which is a starting point of change in another thing or in the thing itself qua other’.

52

S. Makin

can produce a certain sort of change in another type of thing. The heat of a fire, the hardness of a bullet, the sharpness of a knife are all active potentialities possessed by inanimate things; the skill of a doctor, the ability to play the flute and the art of strategy are active potentialities possessed by animate and rational agents. It is the sharpness of the knife, rather than say its colour or weight, which explains why a knife can, while a twig cannot, cut through materials and sunder them apart. It is the art of strategy possessed by the general which explains why she can, while the military novice cannot, devise and execute a plan sufficient to defeat the enemy. Once the notion of an active potentiality is on the table, other correlative notions follow. The fact that a knife is sharp doesn’t fix that a knife can cut steel, and a skilled general cannot organise and deploy a flock of sheep. Active potentialities are paired with correlative passive potentialities. A sharp knife can, while a blunt knife cannot, slice though cuttable materials, a skilled strategist can, while a military novice cannot, manoeuvre those who are able to understand and follow orders.21 As noted earlier in this chapter, and as is clear from something Aristotle says in Metaphysics H, this notion of an active potentiality—an active capacity—is twinned with the notion of a nature (a phusis).22 I appeal to an active capacity in explaining why one thing produces a change in another: the knife cuts the paper, the general manoeuvres his army. But many things also produce changes in themselves: a fish feeds and nourishes itself, a cat regulates its own temperature. Those changes are due to something’s own nature. Aristotle’s canonical account of nature in Physics 2.1 is parallel to the account of an active potentiality in Metaphysics H 1. According to Metaphysics H 1, an active potentiality is ‘a starting point of change in another thing or in the things itself qua other’(Metaphysics H 1, 1046a10–11). And according to Physics 2.1, a nature is ‘a principle or cause of being moved and being at rest in that to which it belongs primarily, in virtue of itself and not accidentally.’23 Each of these accounts includes a qualification intended to exclude cases in which it is merely accidental that the item from which the change arises and the item to which the change happens are one and the same: there is reference in the Metaphysics H 1 account of an active potentiality to ‘a thing itself qua other’, and reference in the account of a nature in Physics 2.1 and Metaphysics H 8 to ‘a thing itself qua itself.’ We should not be thrown by this terminology. Aristotle himself gives an example which illustrates the point splendidly. A doctor might heal herself in the sense that she (as doctor) prescribes

Metaphysics H 1, 1046a11–15: ‘For one kind is a potentiality for being acted on, i.e. the principle in the very thing acted on, which makes it capable of being changed and acted on by another thing or by itself regarded as other; and another kind is a state of insusceptibility to change for the worse and to destruction by another thing or by the thing itself qua other.’ 22 Metaphysics H 8, 1049b5–10: ‘And I mean by potentiality (dunamis) not only that definite kind which is said to be a principle of change in another thing or in the thing itself regarded as other, but in general every principle of movement or rest. For nature (phusis) is also in the same genus as potentiality (dunamis); for it is a principle of movement (archê kinêtikê)—not, however, in something else but in the thing itself qua itself’. 23 Physics 2.1, 192b21–23. 21

Potentiality in Aristotle’s Physics and Biology

53

medicine to herself (as patient). Nevertheless it seems intuitively plain that this would not be a natural return to health, as would be the case if the doctor’s own immune system had dealt with the situation unaided, but a recovery due to medical skill (Physics 2.1, 192b24–25). Aristotle has to work hard to delineate the contrast between what can and what cannot produce natural changes in itself. It would be tempting to suppose that the contrast maps neatly onto that between things which are themselves natural and things which are the products of artifice and skill. But as Physics 8.4 makes plain the situation is more complex, in particular due to the existence of inanimate natural stuffs—for example, the material elements. Those elements are not alive, and so it is difficult to see how they can have sufficient complexity to move themselves. On the other hand, while it seems clear that a heavy stone moving upwards has been thrown by something else, the same stone moving naturally downwards does not seem to Aristotle to have any obvious external source of its motion.24 Again, I will bypass the details of that debate here. Nevertheless there are two reasons why it is important to flag the distinction, and the relations, between a capacity in virtue of which one thing changes another and a nature in virtue of which one thing in some way gives rise to changes in itself. One is that there are potentialities in biology that look like paradigm cases of natural changes (for example, reproduction and nutrition). The second is that I will mention, in what follows, Aristotle’s reference to potentialities in explaining the mutual transformations between the material elements (GC 2.1–4) and the combination of simpler into more complex material stuffs (GC 1.10); and it is precisely the case of the material elements which puts most strain on Aristotle’s discussion of self-motion in Physics 8.4.

4 Two Ways of Thinking About Potentialities and Their Exercise At this point I want to introduce a contrast between two ways in which we could think of the relation between a potentiality and its exercise. The contrast concerns different ways in which we might think of the transition of a potentiality away from its inactive state.

24 Physics 8.4, 255a3–7: ‘When these things [light and heavy things] are in motion to positions the reverse of those they would properly occupy, their motion is violent; when they are in motion to their proper positions—the light thing up and the heavy thing down—their motion is natural (phusei—i.e. by phusis]; but in this case it is no longer evident, as it is when the motion is unnatural [para phusin—i.e. against phusis] whence their motion is derived. It is impossible to say that their motion is derived from themselves: this is a characteristic of life and peculiar to living things.’ The idea of a thing moving itself marks a significant contrast between Aristotelian and Newtonian science. For a collection of papers on this issue see Gill and Lennox (1994).

54

S. Makin

[A] The transition of a potentiality away from its inactive state is thought of as a binary transition. The potentiality is either inactive or it is active. If it is active then some correlated change occurs, and there exists some change to which that potentiality has given rise. [B] The transition of a potentiality away from its inactive state is thought of as gradual or scalar. The potentiality can be more or less active—further from or nearer to its inactive state, as it were. A good way to appreciate what the contrast comes to is by considering one type of example for which [A] is very natural, and then another for which [B] is more appealing. First [A]. Take a doctor’s (learned) potentiality to heal or an oak tree’s (natural) potentiality to shed its leaves in winter, and consider how the state of the world could impinge on the exercise of those potentialities.25 First there seem to be optimal conditions for their exercise. These would be conditions under which the potentiality is successfully exercised: the doctor does heal (the patient recovers), and the oak tree does shed its leaves (the leaves drop to the ground). Second there are also non-optimal conditions. In those conditions the potentiality is indeed exercised but not successfully. For example, the doctor administers a treatment, but the patient doesn’t respond as expected and dies; or the tree is infected and doesn’t shed its leaves, but rather produces an unusual secretion of sap around the leaf bases. Now these two types of case taken together are to be contrasted with those in which the relevant potentiality is simply inactive; for example, cases in which the doctor does nothing, or the time is wrong for an oak to shed its leaves (as in summer). So there are in effect two options: either the potentiality is inactive or it gives rise to some change (be it in optimal or non-optimal conditions). That is the view of potentiality exercise expressed by [A]. Now given that a potentiality is exercised rather than inactive, then we can as noted differentiate the types of change which could result according as the potentiality is exercised in optimal, privileged conditions or in non-optimal, interfering conditions. We identify the potentiality in question—we fix which potentiality it is—by reference to the type of change it produces in optimal conditions. Medical skill is a potentiality to heal, a tree which is deciduous has the potentiality to shed As it happens the first of these is what Aristotle calls a rational potentiality. It is a potentiality which is constituted by a body of knowledge (i.e. of what health is and the ways to achieve it; see Metaphysics Z7, 1032b4–22 for a schematic account of how the doctor’s knowledge of what health is generates a particular course of healing). Rational potentialities can be possessed only by rational agents, and an agent possessing a rational potentiality possesses it in virtue of being rational—were a doctor to lose her rational abilities, for example as a result of dementia, then she would also lose her medical skill. In Metaphysics H 2 and 5 Aristotle emphasises that rational capacities can be exercised in either of two opposed ways, depending on the desires of the agent: a trained doctor can, if she wishes, deliberately harm her patients. But the fact that medical skill is a rational capacity is irrelevant to the example here. The example of a trained sniffer dog’s potentiality to detect contraband would have served equally well. But that’s clearly not a rational potentiality, and nor could a trained sniffer dog deliberately steer its handler away from a stash of illegal imports.

25

Potentiality in Aristotle’s Physics and Biology

55

its leaves. Nevertheless it is the very same capacity which is exercised non-standardly in non-optimal conditions. The patient died because while the doctor administered treatment appropriately and expertly (i.e. exercised their medical skill), the effect of the treatment on the patient’s immune system lead to secondary infections which the patient was no longer able to survive. The tree produced excessive levels of sap at the leaf bases because the presence of disease in the oak during the winter, when the capacities characteristic of deciduous trees were exercised, knocked things off course and produced sap secretion instead of leaf loss.26 Now to the contrast with [B]. Think in this case about a hot block of iron heating some cold water, or a spoonful of milk binding some dry flour. These potentialities, to heat and to bind, could also simply be inactive. The hot iron inside a vacuum chamber will not heat anything at all, the moist milk in a jug won’t bind anything dry. The reason why the potentialities aren’t exercised in those cases is that there isn’t a suitable patient present in the appropriate relation. If a hot block of iron is to do any heating, then there needs to be contact with something heatable, and milk can only bind dry powders which are there to be bound. But it would be peculiar to suppose that the presence of something heatable or something dry are optimal conditions for the exercise of the potentialities to heat and to bind, or to suppose, correlatively, that we could differentiate between the successful and non-successful exercises of these potentialities. The reason that would be peculiar is that in this sort of case the exercise of a potentiality necessarily involves a mutual interaction between agent and patient: as the iron heats the water, the water cools the iron. The notion of mutual interaction is what carries the weight of motivating [B] as a way of thinking about the exercise of a potentiality, and I will say more about that notion shortly. But first it is important to spell out more carefully why it is that if the exercise of a potentiality involves the sort of mutual interaction manifest in cases such as heating and cooling then it is more appealing to think about the exercise of that potentiality along the lines of [B] rather than along the lines of [A]. The reason is that if the exercise of a potentiality involves such mutual interaction, then there will not be optimal conditions for the exercise of the capacity. For this sort of mutual interaction between agent and patient requires that agent and patient each have causally significant properties which lie along a single continuum—a range of properties standing between a pair of opposed termini (as different shades of gray stand between black and white, or different degrees of viscosity between solid and liquid).27 Given that fact, could an agent (e.g. a hot block of iron) and patient

26

There is more on the significance of this point in Sect. 5. Aristotle makes this point in GC 1.7, in the course of adjudicating on the question of whether an agent and patient should be like one another or unlike one another. See GC 1.7, 324a4–8: ‘Hence agent and patient must be in one sense identical, but in another sense other than (i.e. unlike) one another. And since agent and patient are generically identical (i.e. like) but specifically unlike, while it is contraries which exhibit this character: it is clear that contraries and their intermediates are such as to suffer action and to act reciprocally.’ See later in this section for some further remarks on this idea. 27

56

S. Makin

(e.g. some water) be located at points on that continuum which were optimal for each exercising its appropriate potentiality? For example, could there be two temperatures such that optimal conditions for the exercise of a potentiality to heat would be a hot thing at one of those temperatures and a cold thing at the other? Here is an argument that there couldn’t be. If agent and patient were minimally distinct then there would be no exercise of the potentialities in question. The upshot of putting a pan of water on a block of iron of exactly the same temperature is inactivity, no heating and no cooling. Equally though it would be strange to think that optimal conditions for the exercise of a potentiality to heat would involve an agent and patient which are maximally distinct: for example, a pan of water that is as cold as possible standing on something that is as hot as possible. If we think of a cold pan of water standing on a hot block of iron, it would be odd to think that the colder the water placed on the iron, the more successfully the iron can exercise its potentiality to heat the water. In fact it seems quite the contrary. The colder the water placed on the hot iron, the more the water will cool the iron, the less hot the iron will be, and therefore the less effective an actual heater the iron can be. And since the potentiality to heat isn’t optimally exercised when agent and patient are either minimally distinct or maximally distinct, then it would seem arbitrary to identify some intermediate degree of distinction as optimal for the exercise of the capacity to heat. Now it is the fact that there are no optimal conditions for an exercise of the potentiality to heat which makes it very appealing to think of that sort of case along the lines of [B] rather than [A]. Suppose we approach this sort of case with [A] in mind, and seek to accommodate it in the same sort of way as we did the earlier examples of a doctor’s potentiality to heal or an oak tree’s potentiality to drop its leaves in winter. Just as in those cases there are certainly differences between different exercises of the potentiality to heat which we want to say something about—for example, the difference between a very hot thing heating some water and a less hot thing heating that same water.28 But, as has just been argued, it is not appropriate to think of those differences as being between a successful and an

28 Aristotle discusses the different ways in which one thing may be hotter than another at PA 2.2, 648a19–649b9 (see also PA 2.3, 649b10–21 for a shorter parallel discussion of the different ways in which something may be called wet or dry). The point Aristotle is keen to make there is that there is no genuine contradiction involved in supposing that X is hotter than Y and that Y is hotter than X, so long as the way in which X is hotter than Y is different from the way in which Y is hotter than X. To take an example from PA 2.2, 648b33, there need be no single determinate answer to the question of whether water is hotter than iron. On the one hand water heats up more quickly than iron, so that it is easier to heat water than to heat iron. On the other hand, hot iron burns more intensely (and cools more slowly) than hot water. However the complexity noted there doesn’t rule out, as overly naive, the example envisaged here of a more hot thing and a less hot thing heating some water, since here we are comparing two lumps of iron and contrasting them precisely as more and less hot in the same way (i.e. relative to how much hotter each can make a pan of cold water).

Potentiality in Aristotle’s Physics and Biology

57

unsuccessful exercise of one and the same capacity. How should we express the differences then? What it is appealing to say is that the very hot iron heats more than the less hot iron. And it is precisely that notion, of two things both exercising their potentiality to φ, but with one φ-ing more than the other, which [B] is intended to capture. Indeed, the contrast between [A] and [B] could be explicated in just the following way. [A] is appropriate for potentialities to φ such that there is a distinction between one thing exercising the potentiality successfully and another thing exercising the potentiality unsuccessfully. [B] is appropriate for potentialities to φ such that there is a distinction between one thing φ-ing more and another thing φ-ing less.29 Before proceeding any further I should return to the notion of mutual interaction, which was crucial to the examples used to illustrate [B]. Heating and cooling, binding and damping were offered as a paradigm cases of mutual interaction. But it would be helpful to say something more general about the notion of mutual interaction, beyond pointing to those paradigms. Intuitively mutual interaction is intended to pick out those cases in which there are a pair of items X and Y such that the interaction between them is just as much an instance of X producing a change in Y as it is an instance of Y producing a change in X. Quite what that involves can perhaps be clarified through some contrasts with a range of different cases which fail, more or less obviously, to be genuine instances of mutual interaction. In one type of case X gives rise to a change in Y without there being any change in X. Aristotle’s discussion at GC 1.7, 324a29–324b12 suggests an example. Suppose Dr. Caroline cures Deborah of her cold. Caroline produces this change in

29

It is important to keep the notion of a potentiality being more or less active—something’s possessing the potentiality to φ exercising that potentiality in φ-ing more or less—distinct from the notion of a change being more or less complete. Many (but not all) of the changes for which [A] is appropriate are progressive, i.e. directed towards a terminus (housebuilding towards houses, healing towards health); and once that terminus is achieved then the change is completed or finished. So a change like that could be closer to or further from that terminus. That might suggest an intuitive sense in which a builder, for example, could be more or less active. A more active builder will build more continuously, spending more time building, less time gazing off into space. And it is unsurprising that that intuitive sense of greater activity is connected with the degree to which any specific change is complete. If the builder spends more time building, less time gazing into space, then ceteris paribus the house she is building will be completed sooner rather than later. But that’s not the notion of something φ-ing more or less which [B] is intended to express. Indeed it is easy to see how the intuitive contrast between (for example) builders who are more or less active is accommodated by [A]. The builder who spends more time gazing into space is precisely the builder who spends less time exercising her building skill. The intuitive notion of the more active builder is the notion of someone whose exercise of their potentiality to build is more continuous, less interrupted. And that notion aligns nicely with [A], since thinking of the exercise of a potentiality along the lines of [A] lends itself very naturally to the idea of episodic φ-ing. A lazy builder exercises her capacity, then lapses back into inactivity, then exercises her capacity again, then lapses back into inactivity etc. But each episode of activating the capacity is metaphysically on a par with every other. What we have is temporally interrupted instances of φ-ing, as opposed to the genuine cases of φ-ing more or less which [B] is intended to express.

58

S. Makin

virtue of her medical skill, and it is her medical skill which is most properly the source of the change.30 But Caroline’s medical skill is utterly unaffected by its giving rise to healing in Deborah. Caroline knows just as much about medicine after she heals Deborah as she did before she healed Deborah. In a second type of case X’s doing something to Y does require that Y does something to X, but there is some sort of priority of what X does over what Y does. For example, if my dog Fido digests the meat then it must also be the case that the meat nourishes Fido—that’s why eating something that could not possibly nourish me, such as plastic, is eating something indigestible, and why one would say that I ingested rather than digested the poison. Here we have a pair of items (dog, meat) such that it is non-accidentally the case that the first digests the second if and only if the second nourishes the first. But this isn’t the notion of mutual interaction that we are after.31 For it seems plausible to think as regards this case that it is facts about one side of the interaction (the dog’s digesting) which fix that we have a case of digesting-nourishing: to think, that is, that the meat is nourishing because or in virtue of the fact that the dog digests it, rather than vice versa. In cases of genuine mutual interaction by contrast there is no such priority. If I put a pan of cold water on a hot block of iron then it is no more the case that the iron heats the water than it is the case that the water cools the iron, and it would sound odd to insist, for example, that the water cools the iron because the iron heats the water rather than vice versa—or that the heating occurred in virtue of the cooling rather than vice versa. A third type of case highlights another feature of mutual interaction. In cases of mutual interaction it is not merely that each item does something to the other. It should also be the case that what X does to Y is in some sense the opposite of what Y does to X. Not every description of a purely symmetrical interaction between X and Y reveals this feature. Two people marry one another, whereas the mutual interaction of cold water and hot iron does not consist in any single thing that they do to one another. As Aristotle points out, the cold water and hot iron must indeed touch one another32—not much is going to happen unless the pan of cold water is in contact with the hot iron. But that mutual contact is not what their mutual interaction consists in. There would be the very same touching and contact between cold water and cold iron, but no mutual interaction in that latter case. How should we get at the idea that what X does to Y is the opposite of what Y does to X? Recall the Aristotelian context in which we are thinking of mutual interaction. X’s ‘doing

Compare Physics 2.3, 195b22–25: ‘In investigating the cause of each thing it is always necessary to seek what is most precise (as also in other things): thus a man builds because he is a builder, and a builder builds in virtue of his art of building. This last cause then is prior; and so generally.’ 31 This is as it should be, given the argument in this section that there are no optimal conditions for the exercise of potentialities which involve mutual interaction, along with the fact mentioned in the following section that there are optimal and non-optimal conditions for the exercise of the human digestive capacity (optimal conditions would be good health; non-optimal conditions would include diseases of the intestine). 32 Aristotle therefore provides a treatment of the type of contact involved in interaction at GC 1.6, 322b25–323a34. 30

Potentiality in Aristotle’s Physics and Biology

59

something’ to Y is the exercise of a potentiality, as is Y’s ‘doing something’ to X. And those potentialities have a content, something which identifies them as the potentialities they are—in our running example, as the potentiality to heat and the potentiality to cool. So the basic thought is that X’s φ-ing Y and Y’s w-ing X is an instance of mutual interaction when the potentialities for φ and w are such that [i] it is not possible that X φ Y and X w Y at the same time (e.g. nothing can both heat and cool something else at the same time) [ii] it is not possible that X φ X (e.g. nothing can heat itself) [iii] necessarily (X φs Y iff Y ws X) (e.g. whenever one thing heats another, the second thing cools the first) And this way of putting the point gets at the idea that mutual interaction involves pairs of changes (and therefore pairs of potentialities for those changes) which are both intertwined and opposed.33 A final point before moving to the next section, where the significance of the contrast between [A] and [B] as ways of thinking about the exercise of a potentiality will become clearer. The contrast between [A] and [B] is not straightforwardly a difference between potentialities in themselves. Consider the potentiality to heal. As Aristotle notes, we attribute this potentiality both to the medically skilled agent who initiates courses of treatment and to the medicines, wine or food which are administered to a patient in order to bring about healing: A similar distinction holds also of the agent; for we speak of both the doctor and the wine as healing… For if things have not the same matter, the agent acts without being affected; thus the art of healing produces health without itself being acted upon in any way by that which is being healed. But the food, in acting, is itself in some way acted upon: for, in acting, it is simultaneously heated or cooled or otherwise affected. Now the art of healing corresponds to an origin, while the food corresponds to the last (i.e. contiguous) mover. GC 1.7, 324a29–30, 324a33–324b3

A doctor’s exercise of her potentiality to heal is appropriately thought of in terms of [A], precisely because there is a difference between her successful and her non-successful healing. And in that case we might, at a stretch, think of paracetamol’s potentiality to heal in terms of [A] also. Nothing would prevent us doing so, since it is certainly not insofar as the paracetamol is healing that the exercise of its potentiality involves mutual interaction. The thought is rather that an exercise of the paracetamol’s potentiality to heal is at the most basic level its exercising other potentialities—heating and cooling—which do involve mutual interaction, for which there is no sensible distinction between conditions which are optimal or 33

The intertwining is secured by condition [iii]. It corresponds to the idea noted earlier in this section, that when two items mutually interact they must have causally significant properties which lie along a single continuum (compare Aristotle’s reference to ‘one and the same matter’ underlying mutually interacting items at GC 1.7, 324b5–9). If we think of pairs of items which intuitively do not mutually interact—for example something with the potentiality to darken and something with the potentiality to moisten—it is plain that they fail condition [iii]: it is not the case that one thing darkens another iff the second moistens the first.

60

S. Makin

non-optimal for their exercise, and which are therefore appropriately thought of in terms of [B].34 And we might also attribute the potentiality to heat to higher level agent-like items, whose exercise of the potentiality doesn’t involve them in mutual interaction. Imagine that there is an electrically powered device in the rocket engine which serves to heat the fuel before it is injected into the combustion chamber. That device has the potentiality to heat the fuel, and it might exercise that potentiality by providing electrical current to warming filaments over which the fuel passes. And in that case the fuel will cool the filaments as the filaments warm the fuel, but the fuel-warmer considered as a single device doesn’t mutually interact with—isn’t cooled by—the fuel it warms at all.

5 A Familiar Objection and a Response What is the significance of the contrast between [A] and [B]? One way of answering that question is to think further about a familiar objection levelled against the Aristotelian appeal to potentialities in science. A basic fact about Aristotle’s appeal to potentialities in physics and biology is that he takes them to be explanatory. A potentiality is an archê: a source, origin or principle of the changes with which it is correlated. But the idea that something’s changing in a certain way can be explained by reference to a potentiality has been the subject of much criticism. One claim that has often been made is that any explanatory appeal to a potentiality will be empty and uninformative. This criticism comes in a number of forms.35 One version can be stated abstractly as follows. Suppose I explain the fact that X φ’s by saying that X has some particular potentiality. If this is to be explanatory then at the very least I have to succeed in referring to some specific potentiality. A potentiality is identified in part by its content, by what it is a potentiality to do.36 So which potentiality is it that explains X’s φ-ing? If the answer to that question is that it is a potentiality to φ, and if the ground for giving that answer is that the potentiality must be a potentiality to φ since the change it gives rise to is an instance of φ-ing,

Compare Aristotle’s remark in passing at Metaphysics H 2, 1046b17–19: ‘the healthy produces only health, and what can heat only heat, and what can cool only cold, but the scientific man, on the other hand, produces both the contrary effects.’ The healthy (e.g. curative medicine) may well produce health in the patients to whom it is administered, but it will do so by means of heating and cooling, moistening and drying, as suggested by the passage from GC 1.7 cited above. 35 See, for example, Henry (2008) (esp. pp. 51–54), who distinguishes five possible readings of the objection. For another recent rehearsal see Beere (2009, 63–64). 36 This is the point behind Aristotle’s claim in Metaphysics H 8 1049b12–17 that actuality is prior to potentiality in account (logôi). It is also common sense. Suppose all and only items which are tangible are visible. Still, there’s a difference between being visible and being tangible because there’s a difference between seeing and touching. 34

Potentiality in Aristotle’s Physics and Biology

61

then reference to the potentiality to φ is explanatorily idle. It does indeed come to no more than saying that X φs because X can φ. Aristotle’s treatment of potentialities in Metaphysics H, outlined in Sect. 3, gives him the resources to resist this objection. One of the main conclusions Aristotle wants to establish in Metaphysics H is that actuality is prior in various ways to potentiality.37 Precisely what this conclusion means in its broadest sense is opaque, and the subject of much work.38 In part this opacity is due to the fact that Metaphysics H 8 concerns not merely the notion of a potentiality as it turns up in the physical sciences (a potentiality for change), but in addition an extended and novel metaphysical concept. But if we limit our attention to potentiality just in so far as it is connected with change, then the issues are slightly more tractable. In particular two claims emerge from Aristotle’s treatment of potentialities for change in Metaphysics H which, in some way or another, contribute to his general conclusion about the priority of actuality over potentiality. These are [O] that there are Optimal conditions for the exercise of a potentiality and [I] that the necessary exercise of a potentiality in those optimal conditions as an instance of φ-ing is what fixes the Identity of the potentiality in question as a potentiality to φ. The familiar objection mentioned earlier threatens so long as a potentiality claimed to give rise to φ-ing is to be identified as a potentiality to φ in virtue of giving rise to an instance of φ-ing. The upshot of [O] and [I] is that it would be a serious error to think of the identification of potentialities in that way. If that is indeed an error then it has to be possible for a potentiality to φ to be exercised in a change which is not an instance of φ-ing. If the only way in which a poison’s potentiality to paralyse could be exercised was in paralysing, then it would be explanatorily redundant to appeal to that potentiality in explaining why someone who ingested the poison was paralysed. But in fact that’s not how it is. Perhaps an animal has developed a sensitivity to paralysing poisons such that they trigger reflex vomiting. In that case the poison’s potentiality which, absent the vomiting reflex, would have resulted in paralysis, in fact results in stomach contractions and expulsion of the poison from the body. The point underlying [O] and [I] is that the world could be such as either to get in the way of a potentiality’s being exercised or to conduce to its being exercised. As a human being I have the potentiality to digest food. Various illnesses interfere with the

The longest chapter in Metaphysics H is Chap. 8, which starts (1049b4–12) ‘We have distinguished the various senses of ‘prior’ and it is clear that actuality is prior to potentiality … To all such potentiality, the, actuality is prior both in formula and in substance; and in time it is prior in one sense, and in another not’. The chapter closes with (1051a2–3) ‘Obviously, then, actuality is prior both to potentiality and to every principle of change.’ 38 For a recent book length treatment see Peramatzis (2011). 37

62

S. Makin

exercise of that potentiality, while good health is conducive to its exercise. [O] relies on its being the case that some of the conditions in which a potentiality is exercised are optimal for the exercise of that potentiality, while other conditions either interfere with or prevent the exercise of a potentiality. Good lighting is among the conditions which are optimal for me exercising my visual powers, dim lighting interferes and darkness prevents me seeing at all. So now we have the following line of thought. A potentiality can persist through time and be exercised in a wide variety of conditions. That’s to say that depending on conditions there can be an open ended variety of changes F, G, H… deriving from a given potentiality. Nevertheless in each case it is the same potentiality giving rise to different changes in different conditions. Since the identity of a potentiality depends in part on what it is a potentiality for, there has to be some privileged specification of its content, some way of specifying that it is the potentiality to φ which is variously exercised in F-ing, G-ing, H-ing… How is that privileged specification to be fixed? Aristotle thinks of the relation between a potentiality and the change of which it is the source in teleological terms.39 So, given his commitment to the idea that there is some change to which the potentiality is directed, and which fixes its identity, any account of the identity of potentialities will in effect be required to place some constraints on the variety of conditions in which the potentiality may be exercised. And that is to say that there will have to be some conditions which are particularly revealing of the potentiality. These are precisely those conditions which are optimal for the potentiality in question. And this is just what [I] says: what fixes something as a potentiality to φ is that any exercise in optimal conditions is (inter alia) an instance of φ-ing.40 So [O] and [I] give Aristotle the resources to respond to the familiar objection; and [O] and [I] are prominent in, and can be extracted from, Aristotle’s canonical account of potentialities for change in Metaphysics H. But now the contrast introduced in the preceding section between potentiality exercises for which [A] is appealing and those for which [B] is appealing, assumes considerable significance. For [O] and [I] make essential reference to optimal conditions for the exercise of a potentiality. But, as argued in the preceding section, examples of potentiality exercise for which [B] is appealing are precisely those for which it is implausible to differentiate among optimal and non-optimal conditions for exercise. So it might

See for example Metaphysics H 8, 1050b10–12: ‘For it is not that animals see in order that they may have sight but they have sight so that they may see, and likewise too they possess the building craft in order that they may build.’ 40 There are two points to note here. (i) The qualification inter alia is included in order to accommodate the obvious point that any action will fall under a variety of descriptions; (ii) What I have said about [I] isn’t strictly correct, since Aristotle thinks it fails of rational potentialities such as skill in military strategy; but those can be put aside for my purposes here, since they won’t be relevant to thinking about Aristotle’s physics and biology. For more detail on these issues see Makin (2006), Chap. 2 commentary. 39

Potentiality in Aristotle’s Physics and Biology

63

now seem that we cannot rescue those potentialities for which [B] is plausible from the familiar objection by referring to the conceptual resources developed in the canonical treatment of potentiality for change in Metaphysics H.41

6 Heating and Cooling Is this a pessimistic conclusion? Is it a conclusion which should give an adherent of Aristotelian science cause for concern? These are the questions I will finish on—but about which I will be able to say very little in detail here. Potentialities such as those to heat, to cool, to moisten and to dry play a crucial role in Aristotle’s account of what goes on at the material level when different types of thing and different types of stuff interact. The Aristotelian universe contains four basic stuffs: earth, air, fire and water.42 These material elements are not straightforwardly to be identified with the familiar earth, air, fire and water of experience (GC 2.3, 330b21–25). Instead they are best thought of as pairings of one from each of two sets of contraries (hot-cold, wet-dry), those contraries themselves being defined in terms of potentialities. So, for example, heat is defined as the active potentiality to bring together things of the same kind, wet is defined as the passive potentiality of something to be shaped by its solid surroundings.43 Fire is the hot-dry, air the hot-wet, water the cold-wet and earth the cold-dry.44 Aristotle goes

41

Should this seem surprising, given that the potentiality to heat is an example familiar from Metaphysics H? In fact references to the potentiality to heat aren’t that numerous in Metaphysics H. I count four uses: H 1, 1046a26–28; H 2, 1046b6–7, 1046b18; H 3: 1047a5–6. Of those, the final case (H3) isn’t relevant, since it occurs there along with cold and sweet as an example of something perceptible, in an anti-Megarian argument. The first case (H1) isn’t terribly significant either—heat turns up as an example of a (active) capacity being present in something (heat is present in what-can-heat). It is the two H 2 occurrences which carry the weight, and which make the example (the potentiality to heat) seem as if it is a paradigm of the sort of potentiality for change that is the focus of Metaphysics H 1–5. For, first, the capacity to heat is the example of a one way non-rational capacity (1046b6–7). And second, it is also the example used to illustrate the correlative point about non-rational agents at 1046b18–19 (‘the healthy produces only health, and what can heat only heat and what can cool only cold’). 42 GC 2.3, 330a30–330b7. 43 GC 2.3, 329b24–33. The intuitions underlying these definitions are appealing. If I heat up an impure solid mass of stuff then it will tend to separate out into its components (e.g. the dross floats to the top of the pure molten metal, while larger particles of solid contaminant all sink to the bottom as sludge). By contrast, if I throw ash, flakes of wood and bits of egg shell into molten fat and then freeze the lot, I will end up with a single solid mass with these different types of things scattered evenly through it. Compare also Meteorology 4.2, 380a5: ‘the action of heat is to make things more compact, thicker and drier.’ 44 GC 2.3, 330b3–5. According to Meteorology 4.1, 378b10–14 these potentialities are the ‘causes of the elements.’ There is long running (and intractable) scholarly debate on whether the elements are pairs of powers inhering in some underlying prime matter. If they do then the elements will be hot-or-cold and wet-or-dry determinations of some matter which, when thus determined is earth or

64

S. Makin

into considerable detail about how these different potentialities affect one another. The general picture is that there is some mutual interaction along the hot-cold continuum as the differential heat and cold of the bodies involved affect and moderate one another; there is also mutual interaction along the wet-dry continuum as the moistness and dryness of the bodies involved affect one another; and each body acts, in virtue of its position on the hot-cold continuum, on the others involved, which are affected in virtue of their position on the wet-dry continuum.45 The argument of the preceding section was that such potentialities—those for which there is no plausible notion of associated optimal conditions—remain vulnerable to the familiar objection (Sect. 5), because the lack of associated optimal conditions puts such potentialities outside the bounds of Aristotle’s successful response and rebuttal. I have argued that there are no optimal conditions for potentialities which it is plausible to view along the lines of model [B]. As a consequence they remain vulnerable to the familiar objection. But the impact of that consequence for the Aristotelian is lessened to the extent that such potentialities can in some way be subordinated to other potentialities for which model [A] is attractive, and for which there could be optimal conditions. And this is precisely what happens as the notion of a potentiality is used in Aristotelian biology. It is often the case that material level interactions—those for which [B] is attractive—are subsumed under higher level formal processes, by which the material interactions are structured, and for which [A] is attractive—somewhat in the same way that a doctor heals a patient by administering medicine which heals the patient, the healing activity of the medicine being a series of lower level heatings and coolings, moistenings and dryings. A prime example is Aristotelian embryology. A fertilised dog-seed (a zygote) is potentially canine.46 Aristotle views the process by which that fertilised seed gradually develops into something which is actually canine as teleologically structured. It proceeds so as to produce the tissue and organs which are required in order that the resulting animal can lead the life characteristic of the type of animal it is—in the case of a dog, teeth appropriate for eating meat, legs appropriate for running and hunting etc. This process is instantiated throughout in various heatings and coolings, moistening and dryings, through which the requisite tissues and organs are produced. But the lower level heatings and coolings do not determine the air or…, but which is in itself in indeterminate. If they don’t then the elements will simply be pairs of causal powers. It is not necessary to go into that issue for present purposes. 45 This is skeletal. There is considerable detail in Meteorology 4. See for example, Chapter “Potentiality in Aristotle’s Metaphysics” and this chapter distinguish different specific types of heating and cooling (e.g. ripening, broiling, parboiling, rawness); Chapters “Potentiality in Aristotle’s Psychology and Ethics and Potentiality in Classical Arabic Thought” treat of moistening and drying, and Chapters “Potentialities in the Late Middle Ages—The Latin Tradition –From Potentiality to Possibility” discuss the various ways in which different stuffs are melted or solidified, soluble or insoluble, flexible or rigid etc.; Chapter “Potentialities in the Philosophy of Mind” moves on to the homogeneous organic stuffs of which living things are composed, such as flesh, blood and bone. 46 See Metaphysics H 7, 1049a1–3, 14–18.

Potentiality in Aristotle’s Physics and Biology

65

direction of the process, nor do they properly explain it. The heatings and coolings that occur are those that are required for the structures necessary for the animal to live the canine life—which are necessary for the animal to be a dog.47 The following extended quotation gives a typical passage from Aristotle’s Generation of Animals: The homogeneous parts are formed by heat and cold, for some are put together and solidified by the one and some by the other. The difference between these has already been discussed elsewhere, and it has been stated what kinds of things are soluble by liquid and fire, and what are not soluble by liquid and cannot be melted by fire. The nutriment then oozes through the blood vessels and the passages in each of the parts, like water in unbaked pottery, and thus is formed the flesh or its analogues, being solidified by cold, which is why it is also dissolved by fire. But all the particles given off which are too earthy, having but little moisture and heat, cool as the moisture evaporates along with the heat; so they become hard and earthy in character, as nails, horns, hoofs, and beaks, and therefore they are softened by fire but none of them is melted by it, while some of them, as egg-shells, are soluble in liquids. The sinews and bones are formed by the internal heat as the moisture dries, and hence the bones are insoluble by fire like pottery, for like it they have been as it were baked in an oven by the heat in the process of development. But it is not anything whatever that is made into flesh or bone by the heat, but only something naturally fitted for the purpose; nor is it made in any time or place whatever, but only in a place and time naturally so fitted. For neither will that which exists potentially be made except by that moving agent which possesses the actuality, nor will that which possesses the actuality make anything out of anything whatever; the carpenter will not make a box except out of wood, nor will a box be made out of wood without the carpenter. GA 2.6, 743a4–26

What this passage presents is a range of cases in which there is a significant interplay between potentialities for which there are optimal conditions for successful exercise and potentialities for which the notion of optimal conditions does not apply—between potentiality exercises for which [A] is appealing and those for which [B] is appealing. There certainly are optimal conditions for a canine zygote to exercise its potentiality to be an adult dog—the mother needs to be healthy and well nourished, there needs to be an absence of infection, etc.48 So, given that the

47

Aristotle introduces the notion of hypothetical necessity at Physics 2.9. He offers as an illustration the building of a house. Large stones are at the foundations of the walls, lighter timbers are present at the tops of the walls. Still, if someone asked why those stones were at the foundations, it would be unhelpful to say that anything which is large and heavy tends to move downwards. While that is true, the more revealing explanation is that strong heavy materials are required to stand under the walls in order to serve as foundations, and support the walls, so that the walls can in turn support a roof, so that the whole structure can provide shelter. As Aristotle puts it at Physics 2.9, 200a5–9 ‘though the wall does not come to be without these, it is not due to these, except as its material cause: it comes to be for the sake of sheltering and guarding certain things. Similarly in all other things which involve that for the sake of which: the product cannot come to be without things which have a necessary nature, but it is not due to these (except as its material); it comes to be for an end.’ 48 It is a moot question whether those optimal conditions are such that what results from the exercise of the zygote’s potentiality to be an adult dog is an adult male dog which resembles its father—that is to say, whether conditions under which what results is a female dog, or a dog which

66

S. Makin

embryological processes which are the exercise of that potentiality (to be an adult dog) are manifested in, and themselves structure, lower level material heatings and coolings, we could also and derivatively view those lower level material potentialities as sensitive to some degree to the conditions optimal for the higher level process. In effect, while there are no optimal conditions for the exercise of potentialities to heat and to cool simpliciter, there are in a sense optimal conditions for an exercise of heating in the course of canine embryology. Optimal conditions for the heating which solidifies bone out of nutriment are just those which are optimal for the zygote developing into a healthy adult dog. Now this reduces the impact of the continued vulnerability of [B]-like potentialities to the familiar objection—although it comes at a cost. For the main explanatory work in Aristotelian embryology is being done, not by reference to lower level material potentialities, but rather by higher level teleologically ordered developmental potentialities—that is, by potentialities such that it is appealing to think of their exercise in terms of [A] rather than [B]. But there are other types of case for which the subsuming of [B]-like potentialities under those of type [A] doesn’t seem an attractive strategy for the Aristotelian—namely those in which material level interactions are not subsumed under, and guided by, processes which are the exercise of higher level teleologically structured potentialities. Consider Aristotle’s discussion in GC 1.10 of the combination of simple stuffs into homogeneous complex stuffs.49 Aristotle holds two views about a wide range of complex stuffs, including, most importantly, the organic stuffs of which living things are composed. On the one hand those stuffs are homogeneous. Flesh, for example, is flesh all the way down, any portion of flesh contains two sub-portions each of which are themselves flesh, and there is no level at which flesh would turn out to be some other types of stuff (as we might say, carbon, hydrogen, oxygen…). On the other hand, flesh is a combination, a mixture, of simpler material elements—earth, fire etc. How can flesh be a combination of earth, fire etc.—rather than being a new stuff which simply replaces the earth, fire etc. which preceded it—if there is no earth or fire in flesh, but only homogeneous flesh? Aristotle’s answer to this question turns on the distinction between what is present potentially and what is present actually: Since, however, some things are potentially while others are actually, the constituents can be in a sense and yet not-be. The compound may be actually other than the constituents from which it has resulted; nevertheless each of them may still be potentially what it was before they were combined, and both of them may survive undestroyed.

resembles its more distant paternal or maternal ancestors, are revealed thereby to be non-optimal conditions. Aristotle certainly sometimes writes as if this is the case, as in his notorious remark at GA 4.3, 767b6–9 ‘For even he who does not resemble his parents is already in a certain sense a monstrosity; for in these cases nature has in a way departed from the type. The first departure indeed is that the offspring should become female instead of male; this, however, is a natural necessity’—and see GA 4.1–3 more generally on Aristotle’s account of sexual differentiation and inherited characteristics. For a recent discussion which seeks to ameliorate these regrettable features of Aristotle’s biology see Henry (2007). 49 See Sect. 2 for a preliminary characterisation of this phenomenon.

Potentiality in Aristotle’s Physics and Biology

67

GC 1.10, 327b23–26 … a drop of wine does not combine with ten thousand gallons of water; for its form is dissolved, and it is changed so as to merge in the total volume of water. On the other hand, when there is a certain equilibrium between their powers [dunameis, potentialities], then each of them changes out of its own nature towards the dominant; yet neither becomes the other, but both become an intermediate with properties common to both. GC 1.10, 328a26–31

Both the details and the cogency of Aristotle’s answer are keenly disputed.50 The point here, though, is a limited one. It is that there will be some cases of combination which result in complex stuffs, and where those complex stuffs do not go to make up living organisms, and so are not guided by a teleologically structured natural process. There are plenty of heatings and coolings in the world which occur outside the context of teleologically structured biological processes, as in the wide variety of weather conditions51 And there are plenty of cases of combination which, it would seem, are to be understood purely in terms of [B]-like potentialities.52 In such cases it does indeed seem difficult to refer to optimal conditions. While there may be optimal conditions associated with the potentiality to become a dog, that is much less obviously the case with a potentiality such as being malleable.53 Recall one of Aristotle’s examples in the passages quoted immediately above. If we combine quantities of water and wine, such that one is not greatly in excess of the other, then—according to Aristotle—we get a homogeneous product. The combination of water and wine can accommodate different proportions of each ingredient, within limits of comparability. If there is more wine, the resultant stuff will be more intoxicating, if more water then less intoxicating. The core notion for understanding this type of phenomenon is that of a potentiality. The potentialities (dunameis, powers) of each ingredient interact with one another. There are no optimal conditions for the production of the range of watery-wines that can result. The

50

For some recent discussions see Fine (1996), the two discussions by John Cooper, compiled as Cooper (2004), Wood and Weisberg (2004). 51 For an example see Meteorology 1.12, 348b2–10: ‘Now we see that warm and cold react upon one another. Hence in warm weather the lower parts of the earth are cold and in a frost they are warm. The same thing, we must suppose, happens in the upper region, so that in the warmer seasons the cold is concentrated by the surrounding heat and causes the cloud to go over into water suddenly. For this reason rain-drops are much larger on warm days than in winter, and showers more violent.” 52 Meteorology 4.10, 389a8-14: ‘Gold, then, and silver and copper and tin and lead and glass and many nameless stones are of water (hudatos); for they are all melted by heat. Of water (hudatos), too, are some wines and urine and vinegar and lye and whey and serum; for they are all congealed by cold.’ 53 Meteorology 4.9, 386b18–26: ‘Some things are malleable, like copper. Some are not, like stone and wood. Things are malleable when their surface can be made to move (but only in part) both downwards and sideways with one and the same blow: when this is not possible a body is not malleable. All malleable bodies are impressible, but not all impressible bodies are malleable, e.g. wood, though on the whole the two go together. Of squeezable things some are malleable and some not: wax and mud are malleable, wool is not.’

68

S. Makin

potentialities which are in play here are paradigms of those for which model [B] is plausible, and there are no model [A] potentialities in the background. In this context the argument of the preceding sections—that [B]-like potentialities remain vulnerable to the familiar objection—is indeed pessimistic. It is tempting to agree that Aristotle’s appeal to the potentialities of the wine and of the water in the present case does lack for explanatory force, and that I don’t explain much about why the liquid in this flask is more intoxicating than the liquid in that flask by saying that the one is potentially more winey, the other potentially more watery. Was Aristotle insensitive to this tension? Of course not. We can see the direction in which thinking about [B]-like potentialities takes us as Aristotle says more in detail about the combination of simpler stuffs, characterised in terms of their potentialities (i.e. their causal powers), but without making explicit reference to any teleologically structured higher level process under which the combination is subsumed, as in this passage from GC 2.7: Now since there are differences of degree in hot and cold, then although when either is actual without qualification, the other will exist potentially; yet, neither exists in the full completeness of its being, but both by combining destroy one another’s excesses so that there exist instead a hot which (for a hot) is cold and a cold which (for a cold) is hot; then there will exist neither their matter, nor either of the contraries in actuality without qualification, but rather an intermediate; and this intermediate, according as it is potentially more hot than cold or vice versa, will in accordance with that proportion be potentially twice as hot as cold – or three times or whatever. GC 2.7, 334b9–16

But this seems to involve a intriguing development in Aristotle’s way of thinking about potentiality, and to contrast significantly with the notion which was the focus of discussion in such central texts as Metaphysics H and Physics 2. Further consideration of these issues is beyond the scope of this chapter, but I hope that contemporary metaphysicians working in these areas are encouraged to continue to mine Aristotle’s thought.54 Acknowledgements I would like to thank Kristina Engelhard, Rosanna Keefe and Anna Marmodoro for their encouragement with, and their comments on, this chapter.

54

An excellent Aristotelian text to explore is Generation and Corruption. In addition to the translation by Joachim in the Revised Oxford Translation, there is the Clarendon Aristotle volume by Williams (1982). For those interested in an ancient perspective on this text, the whole of the commentary by John Philoponus (approximately 490–570 CE) is now available in translation as part of the Ancient Commentators on Aristotle series: see Williams (1999a, b) and Kupreeva (2005). For a volume of papers from the XVth Symposium Aristotelicum on Generation and Corruption I see De Haas and Mansfeld (2004).

Potentiality in Aristotle’s Physics and Biology

69

References Aristotle. (1984). The complete works of Aristotle. (The revised oxford translation) (two volumes). In J. Barnes (Eds.). Princeton: Princeton University Press. Beere, J. (2009). Doing and being. Oxford: Oxford University Press. Bowin, J. (2011). Aristotle on various types of alteration in De Anima II 5. Phronesis, 56, 138– 161. Burnyeat, M. (2002). De Anima II 5. Phronesis, 47, 28–90. Charlton, W. (1992). Aristotle Physics I and II translation with commentary. Oxford: Clarendon Press. Cooper, J. (2004). Two notes on Aristotle on mixture. In J. Cooper (Ed.), Knowledge, nature and the good (pp. 148–173). Princeton: Princeton University Press. De Haas, F., & Mansfeld, J. (2004). Symposium Aristotelicum: Aristotle’s on generation and corruption I. Oxford: Clarendon Press. Fine, K. (1996). The problem of mixture. In F. A. Lewis & R. Bolton (Eds.), Form, matter and mixture in Aristotle (pp. 82–182). Oxford: Blackwell. Gill, M. L., & Lennox, J. (Eds.). (1994). Self-motion: From Aristotle to Newton. Princeton: Princeton University Press. Graham, D. (1999). Aristotle: Physics, book VIII, translation with commentary. Oxford: Clarendon Press. Heinaman, R. (2007). Actuality, potentiality and De Anima II 5. Phronesis, 52, 139–187. Henry, D. (2007). How sexist is Aristotle’s developmental biology? Phronesis, 52, 1–19. Henry, D. (2008). Organismal natures. Apeiron, 41, 28–74. Hussey, E. (1983). Aristotle: Physics, book III and IV, translation with commentary. Oxford: Clarendon Press. Kupreeva, I. (2005). On Aristotle on coming-to-be and perishing 2.5–11. London: Duckworth. Lear, J. (1988). Aristotle: The desire to understand. Cambridge: Cambridge University Press. Lennox, J. (2001). Aristotle: On the parts of animals, book I–IV, translation with commentary. Oxford: Clarendon Press. Makin, S. (2006). Aristotle: Metaphysics, book H, translation with commentary. Oxford: Clarendon Press. Menn, S. (1994). The origins of Aristotle’s concept of Emeqceia: Emeqceia and Dύmali1. Ancient Philosophy, 14, 73–114. Peramatzis, M. (2011). Priority in Aristotle’s metaphysics. Oxford: Oxford University Press. Shields, C. (2007). Aristotle. London-New York: Routledge. Shields, C. (Ed.). (2013). The Oxford handbook of Aristotle. Oxford: Oxford University Press. Williams, C. J. F. (1982). Aristotle: De Generatione et Corruptione, translation with commentary. Oxford: Clarendon Press. Williams, C. J. F. (1999a), Philoponus: On Aristotle’s on coming-to-be and perishing 1.1–5. Ithaca, New York: Cornell University Press. Williams, C. J. F. (1999b). Philoponus: On Aristotle’s on coming-to-be and perishing 1.6–2.4. Ithaca, New York: Cornell University Press. Wood, R., & Weisberg, M. (2004). Interpreting Aristotle on mixture: Problems about elemental composition from Philoponus to Cooper. Studies in History and Philosophy of Science, 35, 681–706.

70

S. Makin

Author Biography Stephen Makin is Reader in Ancient Philosophy at the University of Sheffield, United Kingdom. He has written books on Pre-Socratic philosophy (“Indifference Arguments”, a study of a characteristic Eleatic and Atomist argument form, published by Blackwell in 1993) and Aristotle (the Clarendon Aristotle volume on “Metaphysics H”, comprising translation and commentary was published by Oxford University Press in 2006). He has also published papers on the Pre-Socratics, Aristotle and Aquinas, as well as non-historical papers on topics in philosophy of religion and contemporary metaphysics.

Potentiality in Aristotle’s Psychology and Ethics Frans A. J. de Haas

The previous chapters have dealt in considerable detail with Aristotle’s notion of potentiality in the context of his metaphysics, and his physics and biology, respectively. Examples from the themes of De anima have already figured prominently in previous chapters, which itself suggests that psychological observations may have served as a source of inspiration for the notion in the first place. Early occurrences of dunamis in the Aristotelian corpus, e.g. in the Protrepticus and in the Eudemian Ethics, indeed show that in Aristotle’s mind the notion of potentiality is closely linked to ethics, and thereby to the road of the fulfillment of human life.1 Hence, while this chapter focuses on the notion of potentiality as it is applied in specific situations in Aristotle’s psychology and ethics we shall have to keep in mind that we are dealing with the context of origin of the notion, and, most likely, with one of the original purposes of Aristotle’s elaboration of the distinction between dunamis and energeia. Furthermore, part of Aristotle’s attack on the Megarian collapse of dunamis into actuality in Metaph. H 3 turned precisely on the need to retain the possibility to talk sensibly about human intellectual development [see Witt (1995, esp. 251–4), (2003)]. After a survey of this area we shall be ready to assess the innovations in the notion of dunamis that Aristotle develops in his De anima and related works, and see how they are made to fit into the framework of his philosophy.

Cf. EE 1218b38–1219a1: “Let this then be assumed, and also that virtue is the best state (diathesis) or condition (hexis) or faculty (dunamis) of all things that have a certain use or work (khrêsis ê ergon).”

1

F. A. J. de Haas (&) Ancient and Medieval Philosophy, Leiden University, Leiden, The Netherlands e-mail: [email protected] © Springer Science+Business Media B.V., part of Springer Nature 2018 K. Engelhard and M. Quante (eds.), Handbook of Potentiality, https://doi.org/10.1007/978-94-024-1287-1_4

71

72

F. A. J. de Haas

1 Plato and Early Aristotle Aristotle’s use of dunamis in the realm of psychology and ethics derives to a large extent from distinctions in Plato. In the Protrepticus, Topics, Magna Moralia and Eudemian Ethics Aristotle does not seem to develop a notion of dunamis that goes beyond Plato’s at all, although he is showing an increasing preference for energeia instead of Plato’s khrêsis to designate the use or exercise of a power, as opposed to merely possessing it.2 From the start, nearly all discussions of potentiality in this area are located in the context of human progress towards a virtuous and (thereby) happy life. Since human beings are rational, the highest virtue required for a happy life is the product (ἔqcom) of the activity of their most valuable part, i.e. the activity of thinking (φqomeῖm, hexqeῖm) that is the virtue of the rational soul or mind (moῦ1). It comes as no surprise, then, that knowledge (ἐpirsήlη) plays a central role in many examples of potentiality and actuality. Aristotle often points out that the use (vqῆri1) or exercise (ἐmέqceia) of virtue or knowledge constitutes happiness rather than the mere possession of it. In the language of the famous analogy of sleeping versus waking: a life spent ‘asleep’ is like a plant’s life, while only a life spent ‘awake’ is the happy human life that Aristotle strives for.3 This relation of priority is also mirrored in language and definition: Whenever each of two different things is called some one and the same thing, and one of these is so-called either through acting (poeῖm), or through undergoing action (pάrveim),4 we shall define the former as expressing the stricter sense of the word; e.g. we shall use the word ‘know’ rather (ἐpίrsarhai lᾶkkom) of him who is using (sὸm vqώlemom) than of him who merely possesses knowledge, and ‘see’ rather of him who is looking at something than of him who merely can do so (soῦ dtmalέmot). (Protr. B81 Düring)

In other words, knowing in the sense of possessing knowledge is knowing in a derivative sense, whereas knowing in the sense of using knowledge is knowing properly speaking. The notion of a dunamis is used here to identify the state of possession of the powers—of both sense perception and knowledge. From Protrepticus B65 it is clear that Aristotle conceives of a human being as not merely a mind, whose proper function would be to grasp the truth, but as a being with several powers. If so, a more refined argument is needed to identify the proper function of a human being which leads to happiness. Only the best and most

2 See Menn (1994, 78ff) for the claim that in his Protrepticus (ed. Düring 1961) Aristotle does not go beyond Plato’s use of dunamis, but differs in his growing preference for ἐmέqceia for Plato’s vqῆri1. Cf. Plato Euth. 277e–278a, 280b5–282a6. This section is indebted to Menn’s study, although I do not share all of his conclusions about Aristotle’s development. 3 Cf. Protr. B80 Düring, MM 1185a10, EE 1095b32, 1216a3, 1219b16–20, Metaph. K 1072b14– 17. Cf. Top. V.2 129b33–34: ‘To perceive’ means several things, one to possess perception, the other the use of perception (sὸ aἰrhάmerhai pkeίx rηlaίmei, ἓm lὲm sὸ aἴrhηrim ἔveim, ἓm dὲ sὸ aἰrhήrei vqῆrhai). 4 Here both infinitives mark the exercise (pace Düring’s translation ad loc.).

Potentiality in Aristotle’s Psychology and Ethics

73

desirable function of that complex being, or of its best part, the soul, will be the function of its best power. Aristotle of course identifies thinking (diάmoia, φqόmηri1) as the best human power. The exercise of that function in the attainment of truth is what wisdom and happiness consist in.5 It is important to keep in mind that the order of the stages of knowledge is not so much a matter of degrees of knowledge of the same kind, but of logical priority between definite and different stages in the process of knowledge acquisition. Potentiality and actuality describe the relation between such stages: For we say ‘more’ (lᾶkkom) about things for which there is only one word, not only in the sense of ‘to a greater degree’ (jah΄ ὑpeqovήm), but also in the sense of [logically] prior and posterior; e.g. we say that health is ‘more good’ than wholesome things, and similarly that which is by its own nature worthy of choice [is more good] in relation to that which is productive of this; yet we observe that the same word [‘good’] is predicated of both, though not in its absolute sense (ὡ1 oὐv ᾓ ἐrsi); for both of useful things and of excellence we say that they are ‘good’. (Protr. B82 Düring)6

At the same time Aristotle draws attention to different aspects of the state of possession. Again picking up on some of Plato’s terminology,7 Aristotle considers a state of possession as a more or less firm disposition that characterizes the soul,8 but also as the state of having acquired something (jsῆri1, jέjsηrhai). To have acquired, e.g., knowledge clearly points to a preceding activity or change that led to the possession of knowledge, i.e. the acquisition of knowledge by learning. This principle gets a wider application in the following text, where, again, disambiguation of terms is the mode of analysis: The word ‘to live’ (sὸ fῆm) seems to be used in two senses, one according to capacity (jasὰ dύmalim), the other according to actuality (jah΄ ἐmέqceiam). For we describe as ‘seeing’ both those animals which have sight and are born capable of seeing (dtmasὰ pέφtjem ἰdeῖm), even if they happen to have their eyes shut, and those which are using this capacity and looking at something. Similarly with knowing (ἐpίrsarhai) and cognizing (cicmώrjeim). We sometimes mean by it using (vqῆrhai) and contemplating (hexqeῖm), sometimes having acquired (jejsῆrhai) the capacity and possessing (ἔveim) the knowledge (ἐpirsήlη). (Protr. B79 Düring)9

5

See also Protr. B67, B91 Düring; cf. EN 1177a19–27, 1100b18–1101a8. Of course the exercise of a naturally good disposition need not be beneficial, see e.g. EE 1248b29–37, and below p. 87. 6 Famously, the examples of ‘good’ and ‘wholesome’ figure prominently in Aristotle’s introduction of homonymy or ‘focal meaning’ to deal with things in an ordered series of things prior and posterior. Cf. EE I.8 with Woods (1982, ad loc) * EN I.6. See further Shields (1999, Chap. 8). 7 Plato in Theaetetus represents pieces of knowledge as birds in an aviary; he denies that knowing is a ἔni1 ἐpirsήlη1, but admits that this jejsῆrhai is in a sense ἔveim (to have caught the birds and keep them in the aviary) and in a sense not (i.e. in the sense in which true mastery consists in catching the right bird when needed). 8 Cf. Cat. 8b27–9a13 which emphasizes that ἔni1 is a more permanent quality that a diάheri1. See further below pp. 85–86. 9 Protr. B79–87 Düring develops this distinction at some length. See Düring o.c. 245–249 for parallels in the corpus. For the homonymy of ‘to live’ see Shields (1999, Chap. 7).

74

F. A. J. de Haas

In Phys. VIII.4 255a30–b13 Aristotle once more distinguishes the twofold meaning of ‘knowing potentially’ (ἐpirsήlxm dtmάlei): it applies both to a learner who does not possess knowledge, and to a knower who possesses knowledge but does not exercize (ἐmeqceῖm) it. But in Phys. VIII.4 this psychological distinction serves as the analogy by which Aristotle explains a feature of his physics. Things can be ‘potentially heavy’ in two ways as well: a light object from which something heavy is produced is potentially heavy, but so are heavy objects that are being prevented from actually moving downwards towards their natural place.10 This is interesting, because it shows that in this context the psychological distinction is considered to be more familiar than the physical one. So far, we have seen that powers of the soul are to be conceived as covering an ordered series, from not-yet-having-but-capable-of-acquiring to having-acquired, and from possessing-but-not-using to using. Everything that falls short of the full exercise at the end of the series deserves the terms of e.g. ‘to see’ or ‘to think’ only derivatively, not in the proper sense of the terms. The ordered series itself is originally suggested, or so I submit, by the context of progress towards happiness, of which the acquisition and use of knowledge are important aspects. Such an ordered series requires a vocabulary to designate the different stages in relation to one another. Aristotle takes up Platonic precedents, and develops the vocabulary according to his own preferences. It is now time to see what the more detailed discussions of potentiality in De anima have to offer in this regard.

2 Potentiality in De anima In this section I shall discuss two sections in De anima which are crucial to Aristotle’s psychology and for which the notion of potentiality is of primary importance: the definitions of the soul in DA II.1, and the famous chapter DA II.5 which contains an elaborate discussion of potentiality (and actuality) in perception.11 Following his usual method Aristotle devotes the first book of DA to an aporetic account of his predecessors’ views of the soul. Right at the start he lays down that one of the questions he wishes to answer is “whether [the soul] is one of the things that are in capacity (sῶm ἐm dtmάlei ὄmsxm) or rather a fulfilment (ἐmsekέveiά si1), for it makes no little difference.” (DA I.1, 402a25–b1). In DA I.3 407b13–26 he complains that his predecessors combined soul with body too easily: they did not set out what causes this combination, or what state the body should be into receive a 10

See p. 85 below for further applications in the context of Aristotle’s ethics. Within the confines of this handbook I cannot begin to do justice to the wealth of secondary literature on these topics. My main sources of inspiration, and sparring partners, in this section are Ackrill (1972–1973), Bowin (2011), Burnyeat (2002), Caston (2002, 2004, 2005, 2006), Everson (1997), Heinaman (2007), Johansen (1997, 2012a, b), Menn (2002), Polansky (2007), Sorabji (1992, 2001), Sisko (1996, 1998, 2004). 11

Potentiality in Aristotle’s Psychology and Ethics

75

particular kind of soul, so that according to the Pythagoreans any soul could end up in any body. As we shall see, Aristotle will address both the problem of causation, and the problem of the unity of soul and body in terms of potentiality.

3 Definitions of Soul and Body At the start of the second book of DA, Aristotle sets out to give a most general description of the soul. A general description is the only possibility, because the souls of plants, animals, and humans differ as members of an ordered series, not as species of a genus, which renders a definition by genus and differentia inadequate. Aristotle immediately introduces distinctions of which he supposes we are familiar from his Physics and Metaphysics. We say that substance is one kind of what is, and that in several senses: in the sense of matter or that which in itself is not a this, and in the sense of form or essence, which is that precisely in virtue of which a thing is called a this, and thirdly in the sense of that which is compounded of both. Now matter is potentiality (dύmali1), form is fulfilment (ἐmsekέveia); and actuality is of two kinds, one as e.g. knowledge (ἐpirsήlη), the other as e.g. contemplating (hexqeῖm). (DA II.1, 412a6–412a11, tr. RevOT modified)

Aristotle quickly surmises that in the compound of a body that possesses (ἔvei) life, the soul cannot be a body, given the assumption that a body more naturally functions as substrate and matter. He then swiftly produces three general descriptions of soul as form of the body, and concludes rather cryptically that the question of the unity of soul and body has now become superfluous. [1] Hence the soul must be a substance in the sense of the form (oὐrίa ὡ1 eἶdo1) of a natural body having life potentially (dtmάlei fxὴm ἔvomso1). But substance is actuality, and thus soul is the actuality of a body as above characterized. Now there are two kinds of actuality corresponding to knowledge and to contemplating. It is obvious that the soul is an actuality like knowledge; for both sleeping and waking presuppose the existence of soul, and of these waking corresponds to contemplating, sleeping to knowledge possessed but not employed, and knowledge of something is temporally prior. [2] That is why the soul is a fulfilment of the first kind (ἐmsekέveia ἡ pqώsη) of a natural body having life potentially (dtmάlei fxὴm ἔvomso1). The body so described is a body which is instrumental. The parts of plants in spite of their extreme simplicity are instruments; e.g. the leaf serves to shelter the pericarp, the pericarp to shelter the fruit, while the roots of plants are analogous to the mouth of animals, both serving for the absorption of food.

76

F. A. J. de Haas [3] If, then, we have to give a general formula applicable to all kinds of soul, we must describe it as an actuality of the first kind (ἐmsekέveia ἡ pqώsη) of a natural instrumental body. That is why we can dismiss as unnecessary the question whether the soul and the body are one: it is as though we were to ask whether the wax and its shape are one, or generally the matter of a thing and that of which it is the matter. Unity has many senses (as many as ‘is’ has), but the proper one is that of actuality. (DA II.1 412a19–b9; tr. RevOT modified)

Armed with the distinctions we surveyed so far, and marshalling our general background in Aristotle’s physics (see Makin’s contribution in this volume), the first definition [1] should be reasonably clear. If a living being consists of soul and body, and body is matter, then soul must be form. What does this mean? Soul as substance is fulfilment (ἐmsekέveia), and fulfilment has two senses, the same two senses that Aristotle elsewhere ascribed to actuality (ἐmέqceia): either as in ‘knowledge’ or as in ‘contemplating’, which are equivalent to ‘sleep’ and ‘waking’. Since to have soul, i.e. to live, includes both sleep and waking, the second definition follows as a paraphrase of the first. For, in order to be alive, a living being need not be awake; also when it is asleep we call it alive. Hence soul is fulfillment of the first kind, i.e. the kind that corresponds to the first member of the analogy, viz. knowledge as mere possession,12 or life as mere possession, viz. sleep.13 This is one of the very few passages in Aristotle in which he actually adds a number to a type of actuality. Later ancient commentators elaborated on this passage by attaching this ‘first actuality’ to a preceding ‘first potentiality’, while seeking to recognize in this first actuality a ‘second potentiality’, followed by its own ‘second actuality’.14 After all, did not Aristotle say that a human being is a potential knower in virtue of her genus, i.e. qua human being, as a first potentiality granted by birth? Does not this first potentiality turn into first actuality when she has actually acquired knowledge? And is not this acquired knowledge itself equivalent to a further, second, potentiality of using the acquired knowledge, with the actual use as second actuality? This may be a tempting enumeration of Aristotle’s distinctions, but it is a kind of systematization that Aristotle did not himself formulate. Aristotle may have had his reasons to avoid such enumeration: what if the process is more complex: can we just add a third or fourth potentiality to the series, or is each of the two still tied to the original possession/use distinction (see p. 72)? And how does this translate into the soul/body relationship?15 While the second definition has shed some light on the type of actuality soul is, we are still in the dark as to what kind of potentiality is involved in ‘a natural body having life potentially’? For if this natural body is the body of which the soul is the

12

It is helpful that Aristotle can write about a natural body possessing life, thus invoking the earlier associations of ἔveim and ἔni1. 13 Cf. Phys. VIII.4 quoted above. 14 For the subsequent history of these distinctions in late antiquity see De Haas (1999, 2000). 15 See Bowin (2011) for a careful distinction and partial overlap of kinds of potentiality and actuality.

Potentiality in Aristotle’s Psychology and Ethics

77

fulfilment, it is alive already, not merely potentially. And a body that is not alive can be called a body only homonymously, as Aristotle insists on numerous occasions.16 The explanation that follows the second definition focuses on the meaning of ‘a natural body having life potentially’. Such a body is capable of serving as an instrument for the soul and its capacities, in the same way parts of plants serve the life of the plant, i.e. its procreation and nutrition. An analogy with an axe follows in order to deepen the notion of instrumentality. If a man uses an axe to chop wood, is he related to the axe as soul to body? No. Soul is to body as the essence of an axe to the material axe. The latter is called ‘axe’ only because it has this essence, the fitness to be used for chopping; the same applies to the living body. The soul is not the essence of an homonymous body, but of a particular kind of natural body that has life potentially. This natural body is a body that already has a principle of motion and rest within itself, and this is … the soul! Hence, not only do the three descriptions of soul, as we saw, mention the instrumental body it needs to exists in; also the definition of the natural body that potentially has life, i.e. the natural instrumental body, refers to the soul as its causal principle. They are so closely related that neither can be defined without mentioning the other.17 A parallel analogy between a living being and one of its parts, sc. an eye, underlines the same point. Suppose the eye represents the living being as a whole; then the soul is like the power of sight (ὄwi1) which is the essence of the eye. The physical eye is the matter of the power of sight. Without the power of sight the eye is an eye merely homonymously, no different from an eye made of stone or drawn in a painting. So the eye that is the matter of the power of sight, is the eye in which the power of sight is present. In short: We must not understand by that which is potentially so that it lives, what has lost the soul it had, but only what still retains it; but seeds and fruits are bodies which are potentially of that sort. Consequently, while waking is fulfilment in a sense corresponding to the cutting [sc. of the axe] and the seeing, the soul is fulfilment in the sense corresponding to sight and the capacity of the tool; the body corresponds to what is in potentiality; as the pupil plus the power of sight constitutes the eye, so the soul plus the body constitutes the animal. From this it is clear that the soul is inseparable from its body, or at any rate that certain parts of it are (if it has parts)—for the fulfilment of some of them is the actuality of the parts themselves. Yet some may be separable because they are not the fulfilments of any body at all.18 (DA II.1, 412b25–413a8)

16 See e.g. Meteor. 389b31–390a2, DA 412b11–22, GA 734b24–27, 735a6–11 (see also p. 85), Pol. 1253a20–25. 17 The priority of actuality over potentiality itself implies that a potentiality can only be conceived with reference to the corresponding actuality. 18 The last sentence is one of the notorious references to the possibility that the mind (moῦ1) might not be tied to the body because the brain was not known to be the organ of the mind as the senses were of sensation; cf. Caston (2000).

78

F. A. J. de Haas

Seeds and fruits are identified as good examples of things that have life potentially. In fact they stretch the notion of potentiality: neither seeds nor fruits can exhibit life without additional changes and appropriate conditions, and in this they are unlike the possession of knowledge.19 Nevertheless, they are more than arbitrary matter, since they grow in or on living things as the product of a living being’s power of procreation. Therefore it makes sense to say they have life potentially, in a way that stones and fire do not. On this strong account of the relation between soul and body as the relation between form and matter, and as the relation between a power and the instrument in which it resides, it is indeed superfluous to demand an account of the unity of soul and body. If so, this relation between soul and body renders transmigration among species impossible. This inference is drawn in the concluding sentences of DA II.1: For this reason those suppose correctly who think that the soul neither is without body nor is some body. For it is not body but something of body, and for this reason it exists in body, and in body of a certain sort, and not in the way our predecessors fitted the soul into a body without specifying in which body and in what sort of body, even though it is clear that not any old thing will receive any other thing. But it happens also in this way according to reason: for the fulfilment of each thing comes by nature to be in what exists in potentiality and in the appropriate matter. It is clear from these considerations that there is some fulfilment and account of something that has the potentiality of being such. (DA II.2, 414a19–28, tr. Johansen 2012b, 15–6)

We are now ready to grasp potentiality in terms of the four modes of physical explanation that Aristotle developed in his Physics: formal, material, efficient, and final [Cf. DA II.4, 415b8–23; cf. Lennox (1995)]. We have seen that the body is cast in the role of dunamis and matter, and soul in the role of substance and form. For a living being, ‘to be’ means ‘to live’, and the soul is responsible for living as the formal principle of the living being. But we have also seen that soul is the principle of motion and rest in the living being (at least in those that can change place, plants and anemones excluded). The natural body also potentially has life as the complement of that principle of motion. Last but not least, the notion of instrumentality brings final causality into play. Natural bodies exist for the sake of the soul, what they are is what they are for, as the axe is for cutting. Their purpose makes it possible to identify the characteristics that make them what they are. The rich notion of causation that Aristotle introduces in this part of the DA shows that the potentiality of the living body is not merely associated with material causation as the counterpart of formal causation, but also with being the co-relative and beneficiary of the efficient and final causation of the soul. Each of these causal relations between soul and body confirms and strengthens their union in the fulfillment that is called living (fῆm).20 In sum: in the first chapters of the De anima Aristotle has worked hard to show that the stages of progress towards happiness that we encountered in the earlier 19

See p. 84 for a similar role of appropriate conditions in the context of ethics. For the use of way of life (bίo1) as a unifying principle of method in Aristotle’s biology see Lennox (2010).

20

Potentiality in Aristotle’s Psychology and Ethics

79

work sit squarely within the much wider genus of biological developments. In a living being, the formal, efficient and final causation of the soul always has its necessary complement in the body that has life potentially, and vice versa. Potentiality here receives a rich content as it answers to the specific needs of each form that is the soul of a particular species of living being.

4 Potentiality and Actuality in Perception So far we have seen that the stages in the acquisition of knowledge (to continue the focus on Aristotle’s dearest example) are an ordered series towards fulfilment (ἐmέqceia or more often ἐmsekέveia). Everything that falls short of complete fulfilment is potentially of this nature in one way or another. We have also seen how Aristotle takes care to specify, e.g. by means of notions of causality, what the precise markers of a given potentiality are. In these contexts Aristotle thus avoids a situation in which potentiality becomes a mere spectrum of possibilities. Each form demands its own characteristic potentiality with which it constitutes a functional unity. In DA II.5, which Aristotle announces as a treatment of sense perception in general (416b32–3), the focus is on the process that gives rise to perception in the sense of seeing a colour, or hearing a sound. Sense perception is identified as a motion and a kind of action and passion between qualities as analysed in the Physics and in more detail in Generation and Corruption I.7–9.21 From these discussions Aristotle can invoke the following distinctions as familiar (DA II.5, 416b33–417a20): 1. the power of sensation (sὸ aἰrhηsijόm) is potentially such as long as there is no matching object around, or when the living being is asleep; hence no sensation occurs of its own account. In contrary conditions the power of sensation is actually such; in this sense perceiving (aἰrhάmerhai) and perception (aἴrhηri1) are said in the two familiar ways; 2. the object of perception (sὸ aἰrhηsόm) is potentially such as long as its perceptible features do not act upon a sense organ; 3. when the necessary conditions are met, the object acts upon the sense, which is initially unlike the object, so as to render it like itself (in other words, the sense receives the relevant form). This is an incomplete motion of the kind described in the physical works. It is also a non-rational motion in the sense that when the circumstances are right, the interaction between agent and patient will occur necessarily.

21

For more details see Makin’s contribution in this volume.

80

F. A. J. de Haas

At this point (417a21) Aristotle inserts a further division of potentiality (dύmali1) and actuality (ἐmέqceia), which, he now claims, he had so far spoken of generally (ἁpkῶ1). In what follows Aristotle introduces the following insights: 1. the analysis of ‘knowing’ now clearly features three distinct stages: (a) someone is ‘knowing’ (ἐpirsήlxm) in the sense that she belongs to the genus of human beings that have the capacity to know and have knowledge; (b) someone is ‘knowing’ in the sense that she possesses, e.g., knowledge of grammar in such a way that she is capable (dtmasό1) of exercising that knowledge when she so wishes, provided nothing interferes; (c) someone is ‘knowing’ in the sense of contemplating and fully knowing, e.g., this particular alpha; 2. Aristotle points to the difference between process P1 which leads from (a) to (b), and process P2 which leads from (b) to (c). Although he initially approaches both processes as plain cases of qualitative action and passion, it is clear that the case of knowledge calls for several distinctions.22 3. At stage (b) someone has experienced qualitative change through learning, and has often changed from the opposite disposition; someone who changes from possessing, but not using, the knowledge of arithmetic or grammar, to using it, suffers a different kind of change. 4. The passion involved in each of these transitions is different: P1 is a kind of corruption by the contrary state; P2 is a kind of preservation of what is potentially (possessing knowledge) by what is in fulfilment (contemplating). However, P1 is not ordinary qualitative change, because the repetitive changes involved in learning do not seem to imply corruption of the contrary state at each instance. P2 is not ordinary qualitative change because it is no qualitative change at all, but rather an increase towards itself, or towards fulfilment. In Aristotle’s words: Hence it is wrong to speak of a wise man as being ‘altered’ when he uses his wisdom, just as it would be absurd to speak of a builder as being altered when he is using his skill in building a house. What in the case of thinking or understanding leads from potentiality to fulfilment ought not to be called teaching but something else. That which starting with the power to know learns or acquires knowledge through the agency of one who actually knows and has the power of teaching (ὑpὸ soῦ ἐmsekeveίᾳ ὄmso1 jaὶ didarjakijoῦ) either ought not to be said ‘to be acted upon’ at all—or else we must recognize two senses of alteration, viz. the change to conditions of privation, and the change to a thing’s dispositions and to its nature (sὴm ἐpὶ sὰ1 ἕnei1 jaὶ sὴm φύrim). (DA II.5, 417b8–16, tr. RevOT)

22

The distinctions are subtle and hotly debated in the secondary literature. The survey provided here states my current understanding of the chapter, which is indebted esp. to the discussion between Burnyeat and Sorabji with corrections by Heinaman, Sisko, and Bowin. My interpretation coincides with neither in every respect—The need for further refinements is clear from e.g. EE 1218b35–6: Of things in the soul some are dispositions or powers, others actualities and motions (sῶm dὲ ἐm wtvῇ sὰ lὲm ἕnei1 ἢ dtmάlei1 eἰrί, sὰ d᾽ ἐmέqceiai jaὶ jimήrei1).

Potentiality in Aristotle’s Psychology and Ethics

81

This passage makes clear that neither P1 nor P2 are ordinary cases of qualitative action and passion because neither are changes to conditions of privation. Both P1 and P2 are changes to a thing’s dispositions and to its nature. This matches the general description of soul we discussed above. The power of thought, as one of the powers of the soul, is the fulfilment of the potentiality of a human body. Such powers, and such potentiality, did not and could not come up for discussion in the study of inanimate nature. Hence the changes from potentiality to actuality involved here are different in kind from those. It would be best to invent new words for P1 and P2, but Aristotle eventually settles for the old words of ‘passion’ and ‘qualitative change’, with a caveat.23 One would almost forget that this entire section on potentiality and actuality of knowledge is part of a chapter on sense perception. Again, Aristotle has developed new distinctions in the case of knowledge as an analogue for the elucidation of a different topic, this time perception. In the case of what is to possess sense, the first transition (lesabokή) is due to the action of the male parent and takes place before birth so that at birth the living thing is, in respect of sensation, at the stage which corresponds to the possession of knowledge. Actual sensation corresponds to the stage of the exercise of knowledge. But between the two cases compared there is a difference; the objects that excite the sensory powers to activity, the seen, the heard, &c., are outside. The ground of this difference is that what actual sensation apprehends is individuals, while what knowledge apprehends is universals, and these are in a sense within the soul itself. That is why a man can think when he wants to but his sensation does not depend upon himself—a sensible object must be there. A similar statement must be made about our knowledge of what is sensible—on the same ground, viz. that the sensible objects are individual and external. (DA II.5, 417b16–28, tr. RevOT)

The first sentence makes clear that at the moment of procreation the male parent (who contributes the form) renders the living thing potentially sentient in the sense of possession. Animals do not have to acquire the power of sensation, they are born with it. Once alive, animals have souls, and thereby sense perception, and bodies with fully developed sense organs to match. At no stage do finished bodies have to wait for their souls: from inception onwards their development is a sign of its presence. Of course, the objects of knowledge (universals in the soul) are different from the objects of sensation (individuals outside) which explains why the changes involved in sensation do not depend on the seer or the hearer, as they do in the case of the knower who can exercise her knowledge at will. In this key chapter of the De anima Aristotle introduces new kinds of transition between the three stages of a development for which the power and acquisition of knowledge serve as the key analogy. To be potentially a knower moves from lack of knowledge, through acquisition by learning, to full exercise. The development is natural, since the human being is a soul/body unity that has a tendency for such

23

See DA II.5, 417b29–418a6. Here the same points are repeated with the example of the (still nameless) difference between the way a boy is capable of leading an army (sὸm paῑda dύmarhai rsqasηceῖm), and the way the same boy is when he has grown into a young man. The latter stage is the stage that compares to the power of perception (sὸ aἰrhηsijόm).

82

F. A. J. de Haas

fulfilment as its final goal. In that sense, the transitions from potentiality to actuality involved in the process are all transitions towards the human being itself, towards her own natural dispositions. Nothing is lost on the way, every transition (in the right direction) is gain. Once knowledge has been acquired, it is continuously available for use, at will, unless amnesia or old age interfere. The case of sensation corresponds to the same model, with some obvious adaptations. To be potentially a perceiver moves through the stages of inception, with lack of full-grown organs, through birth at which the power of sensation has an instrumental body to match, to acts of seeing particular sensible objects, when encountered under the right conditions. Many of Aristotle’s chapters on the individual powers of sensation—vision, hearing, smell, taste, and touch (DA II.7–11)— follow the same patterns. Aristotle goes to great lengths to specify exactly what the potentiality of the natural instrumental body consists in (e.g. the anatomy of the sense organs), and under which conditions the transition to full sensation will occur (e.g. the role of a suitable medium to transport the relevant forms to the sense organs). Only in De anima III.4 does Aristotle discuss the rational power of the soul. Now the analogy works in the other direction. Like the power of sensation, the mind, too, is potentially like its objects, the forms, to which it relates in a similar way (429a16–18, 27–29; b30–31), even though mind can think what it knows whenever it wishes. However, objects of thought are not enmattered, and exist in material things only potentially. Mind (moῦ1) is itself thinkable (moηsόm) in exactly the same way as its objects (moηsά) are. For in the case of objects which involve no matter, what thinks (sὸ mooῦm) and what is thought (sὸ mooύlemom) are identical; for speculative knowledge (ἡ ἐpirsήlη ἡ hexqηsijή) and its object are identical. (Why mind is not always thinking we must consider later.) In the case of those which contain matter each of the objects of thought is only potentially present. It follows that while they will not have mind in them (for mind is a potentiality of them only in so far as they are capable of being disengaged from matter) it belongs to mind to be thinkable. (DA III.4, 430a1–9, RevOT modified)

Although these lines have spawned a long tradition of interpretation, they contain little that is unexpected after the earlier chapters of the De anima. A problem that remains after these lines is how the objects of thought that exist potentially in material objects are disengaged from matter. Perhaps sense perception, that receives forms, does part of the work. Perhaps, too, the notorious chapter DA III.5 that follows is meant to provide, among other things, a solution of this problem. In DA III.5 Aristotle supposes that the difference between matter which is everything potentially, and an active cause is present within mind itself, not just in the relationship between soul and body that we have seen. This supposition yields the distinction between a kind of mind that becomes everything, and one that makes everything, and can be regarded as a disposition, such as light. For light—which Aristotle considers to be the transparency of the medium of vision (air or water)— renders potentially visible colours actually visible.

Potentiality in Aristotle’s Psychology and Ethics

83

Mind (moῦ1) in this sense of it [sc. as disposition] is separable, impassible, unmixed, since it is in its essential nature activity (for always the active is superior to the passive factor, the principle (ἀqvή) to the matter). Actual knowledge is identical with its object: in the individual, potential knowledge is in time prior to actual knowledge, but absolutely it is not prior even in time. It does not sometimes think and sometimes not think. When separated it is alone just what it is, and this above all is immortal and eternal (we do not remember because, while this is impassible, passive mind (ὁ pahηsijὸ1 moῦ1) is perishable); and without this nothing thinks. (DA III.5, 430a17–25, RevOT)

The transmission of the Greek text is full of problems, and commentators have struggled with these lines for centuries.24 I limit my comments here to what seems relevant in the framework of this chapter on potentiality. Mind, too, like perception and life in general, exhibits the distinction between potentiality and actuality. Mind potentially is like a writing tablet without anything written on it. As soon as it is provided with its objects it gains all characterization it has as it becomes identical with its objects. Mind in actuality is essential, perfect, and therefore eternal, pure actuality which is not liable to any interaction in terms of action and passion. At least in that sense it is separate, nothing but a ‘place of forms’.25 If we try to apply the analogies employed so far in De anima, mind in the sense of a disposition and mind in potentiality will have to form a natural functional unity. From this perspective it would be surprising, though not impossible, if this little chapter, in its original form, would have contained the claim that rational soul is immortal as a separately existing entity after human life has ended. It would seem even less plausible that this passage in De anima, a work on animal soul, would here identify a single divine mind that, on its own, somehow causes all human minds to think—as later commentators would have it. But to argue for any more definite interpretation along the lines of the preceding analysis would demand a long article of its own.

5 Potentiality in Ethics The elaboration of types of potentiality that we found in Aristotle’s psychology also received application in his ethics. In his ethical works Aristotle makes a point of criticizing Socrates and Plato for focussing on the definitions of the virtues rather than on how and by what sources virtue arises, even though his own answer to these questions is also arrived at by an investigation into what virtue is (Cf. EE 1216b10– 22, with Woods p. 56). His point against Socratic intellectualism is the following: If something is fine, understanding it is fine also; but still, in the case of virtue, the most valuable thing is not to have knowledge of it, but to know from what sources (ἐj sίmxm) it arises. For what we wish is to be courageous, not to know what courage is; to be just, not to

24

See for recent attempts at an integral interpretation e.g. Frede (1996), Caston (1999). DA 429a27–29: “It was a good idea to call the soul ‘the place of forms’, though this description holds only of the rational soul, and even this is the forms only potentially, not actually.” 25

84

F. A. J. de Haas know what justice is; in the same way as to be healthy rather than to know what being healthy is, and to be in a good state, rather than to know what it is to be in a good state. (EE I.5, 1216b19–25 tr. Woods, my emphasis)26

In EN I.8 Aristotle formulates the same point in by now familiar terms: It makes, perhaps, no small difference whether we place the chief good in possession or in use (ἐm jsήrei ἢ vqήrei), in state or in activity (ἐm ἕnei ἢ ἐmeqceίᾳ). For the state may exist without producing (ἀposekeῖm) any good result, as in a man who is asleep or in some other way quite inactive, but the activity cannot; for one who has the activity will of necessity be acting, and acting well. And as in the Olympic Games it is not the most beautiful and the strongest that are crowned but those who compete (for it is some of these that are victorious), so those who act rightly win the noble and good things in life. (EN I.8, 1098b31–1099a10 tr. RevOT)

We may surmise that Aristotle’s virtues correspond to realized human capacities (first actuality, ktêsis or hexis), which as such represent potentialities for virtuous action (second actuality, khrêsis or energeia). A virtuous, happy person leads a life in which she fully realizes such second potentialities by committing virtuous acts, and continues to do so her entire life (Cf. EN I.11, esp. 1100b12–17). Aristotle specifies that virtuous acts differ from products of art that have a value in themselves (Cf. EN I.3, 1105a26–b7). Virtuous acts, while being good themselves, are virtuous only when the agent is in a specific condition when acting: The agent also must be in a certain condition when he does them; in the first place he must have knowledge, secondly he must choose the acts, and choose them for their own sakes, and thirdly his action must proceed from a firm and unchangeable character. These are not reckoned in as conditions of the possession of the arts, except the bare knowledge; but as a condition of the possession of the virtues, knowledge has little or no weight, while the other conditions count not for a little but for everything, i.e. the very conditions which result from often doing just and temperate acts. (EN 1105a28–b5, RevOT modified)

This passage establishes a close connection between virtuous actions on the one hand, and both choice and character on the other: without choice for its own sake, or without a firm and unchangeable character of the agent, actions are not truly virtuous. In a striking criticism of Socratic intellectualism Aristotle states that knowledge ‘has little or no weight’ as a condition of the possession of the virtues. His famous study of akrasia, or lack of self-control, may serve as the prime example of the meaning of this statement. We shall find not only a reappearance of the application of potentiality in psychology, but a further enhancement of the notion of a potentiality that is fully present, but cannot be actualized because it is impeded by internal or external causes. Aristotle investigates the phenomenon of akrasia in EN VII.1–11: an akratic is a person who possesses knowledge of the good, but does not act accordingly.27 After 26

Cf EN 1103b26–31, 1105b12–18. The literature on the topic is extensive. For a useful entry into akrasia in ancient philosophy at large see Bobonich-Destrée (2007), esp. the contributions by Destrée (who has inspired much of my discussion here), Zingano and Charles on Aristotle; see also Gosling (1993), Moss (2009), Charles (2009a, 2011a, b) 27

Potentiality in Aristotle’s Psychology and Ethics

85

rehearsing the common opinions on the subject (VII.2), he begins the examination of the problems involved with the question: “What kind of right belief is possessed by the man who behaves without self-control?” (EN VII.2, 1045b21–22) Aristotle considers Socrates’ downright rejection of akrasia a view that “contradicts the plain phenomena” (EN VII.2, 1145b27–28), but he goes on to investigate what kind of ignorance might be involved all the same.28 First Aristotle applies the familiar distinction: someone might possess knowledge but fail to use it (1146b31–35). But why does this happen? For our discussion here the sheer number of occasions in which the possession of knowledge does not lead to virtuous action is most relevant as a remarkable illustration of Aristotle’s point that knowledge is of little or no weight when it comes to the possession of virtues—which, after all, come to be from repeated virtuous actions. In these chapters a whole spectrum of possible obstacles unfolds: physiological causes might be in play, as in the case of sleep, madness, and drunkenness, for which only natural scientists know the cure.29 The akratic is like the drunk or the actor in saying the right words, but without having the desire that corresponds with the words, and desire is needed to give consequence to the words by causing action.30 Passion or desire may cause a failure to apply a universal rule in a particular case (EN VII.5, 1146b35–1147a10, 1147a24– b5), or cause one to act in haste so that one does not even take time to think at all. In short, the akratic has knowledge potentially in the sense that she could use it if not prevented from doing so. Such potential knowledge is not only completely inactive, but even incapable of being activated due to the state the agent is in (see esp. Phys. VII.3, 247b13–248a6 in n. 40 above). As Aristotle summarizes elsewhere: But a thing existing potentially may be nearer or further from its realization in actuality, just as a sleeping geometer is further away than one awake and the latter than one actually studying. (GA II.1, 735a9–11)

Let us now return to the conditions of virtuous action in the text quoted above (p. 84) to draw out a last point. In the final sentence of the quoted passage Aristotle suggests that the conditions of choice and a firm character result from often doing just and temperate acts. Aristotle is here concerned with a puzzle that he mentioned at the start of EN II.3: How can we become virtuous by repeatedly committing virtuous acts if acts are only virtuous when they are committed by a person who is 28

See Destrée and Zingano oo.cc. for an analysis of Aristotle’s dialectical method here. EN VII.5, 1147b6–9, cf. Phys. VII.3, 247b13–248a6: people who recover from drunkenness, sleep, or disease do not change to the opposite state, and we do not say they have become knowing all over again—even though the person concerned was previously unable to use her knowledge (b13-16 ὥrpeq ὅsam ἐj soῦ lehύeim ἢ jaheύdeim ἢ moreῖm eἰ1 sἀmamsίa lesarsῇ si1, oὔ φalem ἐpirsήloma cecomέmai pάkim (jaίsoi ἀdύmaso1 ἦm sῇ ἐpirsήlῃ vqῆrhai pqόseqom); MM 2.6.17.3-11: drunk people have not lost their knowledge but their knowledge was overpowered by the drunkenness; in the same way the overruling passion has brought the akratic’s reasoning to a standstill. When she recovers she will be herself again. (ἐpijqasῆram cὰq sὸ pάho1 ἠqeleῖm ἐpoίηre sὸm kocirlόm ὅsam d᾽ ἀpakkacῇ sὸ pάho1 ὥrpeq ἡ lέhη, pάkim ὁ aὐsὸ1 ἐrsίm). 30 EN VII.5, 1147a35–b1, b9–12. For desire as the motive force in action see De anima III.7–11; cf. Charles (2009b), several contributions to Pakaluk-Pearson (2011). 29

86

F. A. J. de Haas

already virtuous? Here we see the background for Aristotle’s discussion with the Megarians on the meaning of dunamis. The Megarians, who denied the existence of potentiality, thereby denied the possibility of acquiring virtue by learning and exercise in precisely the sense Aristotle needs here. We have seen that Aristotle ran into the same problem in the psychology of learning: new knowledge is acquired from learning and observation by often going over the same material. In the same way a state of virtue is made firm and unchangeable by repeatedly committing virtuous acts, guided by e.g. instruction and punishment at the hands of educators, the example of practically wise men, and the laws. At this point we do well to recall Aristotle’s distinction between kinds of virtue. Aristotle distinguishes virtues of character (ἀqesὴ ἠhijή), such as justice, bravery, and temperance, from intellectual virtues (ἀqesὴ diamoηsijή), such as wisdom (roφίa), and practical wisdom (φqomήri1).31 Aristotle makes it clear that virtues of character are states of the non-rational parts of the soul, acquired by habituation (ἔho1), whereas intellectual virtues are states of the rational part of the soul, acquired by learning and observation.32 We receive praise or blame with respect to such states, not because of our passions or our proneness to suffer from them.33 What is more, anger and fear have nothing to do with that other condition of virtuous action, choice, whereas acquired states (whether good or bad) play an important role in choice making.34 Virtues of character are acquired by training and exercise, i.e. training in feeling pleasure and pain—which define the natural tendencies of all animals—at appropriate occasions.35 Intellectual virtues are acquired by teaching and education in accordance with the analysis of knowing discussed above (see pp. 72–74, 80). Let us return to the question of how ‘a firm and unchangeable character’ can be established. Does it come to be from prior potentialities, as the parallels noted 31

In EN VI Aristotle carefully relates these two to three further candidates for intellectual virtues: insight (nous), understanding (epistêmê), and art (tekhnê). 32 EN II.1, 1103a14–18. Aristotle emphasizes that all of these states are acquired to avoid association with innate knowledge of the Platonic kind. Hence the need to investigate how this acquisition takes place. 33 In EN II.5, 1105b16–1106a2 Aristotle argues that virtues of character are neither passions (pathê) like anger, fear, and joy; nor capacities (dunameis), viz. ‘things in virtue of which we are said to be capable of feeling these (pathêtikoi)’; but rather states (hexeis), ‘the things in virtue of which we stand well or badly with reference to the passions, e.g. with reference to anger we stand badly if we feel it violently or too weakly, and well if we feel it moderately; and similarly with reference to the other passions’. 34 It remains unclear how exactly Aristotle believes virtuous states determine action. Here I cannot go into the notorious problem that a ‘firm and unchangeable character’ seems to rule out voluntary choice if one necessarily acts in accordance with it. Against this position, Aristotle holds firmly that we remain responsible for the formation (or correction) of our character, even if at a particular point in time we cannot act but badly on account of it, see esp. EN III.1, III.5 and III. 35 EN II.2, 1104a27–b24; II.4. The desires that prohibit the akratic to exercize her knowledge may well derive from bad habits that make it all too easy for her to give into the wrong representations (φamsarίa aἰrhηsijή instead of φamsarίa kocirsijή, cf. DA III.11, 433b27–30); cf. Destrée o. c., and Kosman (1999), Leighton (2011).

Potentiality in Aristotle’s Psychology and Ethics

87

above suggest? Aristotle’s answer to this question links his biological with his ethical and political works by means of the notion of natural virtue. Natural virtue is a (first) disposition which animals and human children possess from birth, in some degree or other; only adult human beings can bring it to complete fullfilment, due to the presence of practical reason, and provided circumstances are favourable.36 We must therefore consider virtue also once more; for virtue too is similarly related as practical wisdom (phronêsis) is to cleverness (deinotês)—not the same, but like it—so is natural virtue (phusikê aretê) to virtue in the strict sense (kuria aretê). For all men think that each type of character belongs to its possessors in some sense by nature; for from the very moment of birth we are just or fitted for self-control or brave or have the other moral qualities; but yet we seek something else as that which is good in the strict sense—we seek for the presence of such qualities in another way. For both children and brutes have natural dispositions (phusikai hexeis) to these qualities, but without thought these are evidently hurtful. Only we seem to see this much, that, while one may be led astray by them, as a powerful body which moves without sight may well take a ‘powerful’ fall because of its lack of sight, still, if a man once acquires thought that makes a difference in action; and his state, while still like what it was, will then be excellence in the strict sense. Therefore, as in the part of us which forms opinions there are two types, cleverness and practical wisdom, so too in the moral part there are two types, natural virtue and virtue in the strict sense, and of these the latter does not arise without practical wisdom. (EN VI.13, 1144b1–17, RevOT modified, cf. Lennox)

This section establishes an intriguing analogy: natural virtue is like virtue in the strict sense, in the same way as cleverness is like practical wisdom. Earlier in the chapter cleverness was defined as a capacity (dunamis) to act in ways that achieve any assumed goal (EN VI.13, 1144a24–26). Practical wisdom, in contrast, is not a capacity but cannot exist without the capacity of cleverness (1144a28–29). What does this mean? As the quote above explains, cleverness belongs to us from birth, and so do natural virtues of character (justice, bravery, self-control). What is more, such good qualities belong to children and brutes alike.37 Thus it is clear that humans do not have to develop their virtues from scratch. However, children and brutes have in common that they lack practical wisdom, and without the guidance of practical wisdom even good qualities may cause harm.38 It turns out that for all natural virtues of character, the presence of practical wisdom is necessary for the acquisition of true virtues of character. Aristotle defines their relation as follows: virtue makes the goal correct, and practical intelligence makes that which promotes the goal correct (1144a7–9). Although the sequel of this passage provides a much

36

For natural virtue and its development see e.g. Burnyeat (1980), Sherman (1996) 151–164, Lennox (1999), Lawrence (2011), Leunissen (2012). 37 Cf. EN III.2 1111b6–10: children and animals share in what is voluntary (the wider category), but not in choice. Cf. HA VIII.1, 588a18–b3. For the extent of animal phronêsis see Labarrière (2005). On brutes and human beings see Lorenz (2006). 38 Rational capacities are open to opposite results, cf. Metaph. 1046b1ff, 1048a8–12.

88

F. A. J. de Haas

more nuanced picture, in this way practical intelligence relies on the prior capacity of cleverness, but cleverness does not represent genuine virtue without the guidance of practical wisdom. As long as children have not yet developed their own practical wisdom, their actions may be as harmful as they may be beneficial. During that time educators and lawgivers are its substitutes (see e.g. EN II.1, 1103b2–6); they provide examples of proper action, and create the conditions necessary for the training of the growing child in making the right decisions according to reasoned choice, rather than according to unreasoned natural virtue. The rise of practical wisdom in a maturing human being eventually gives humans an edge over animals: hence only humans can develop true virtue. Despite the many similarities Aristotle notes between human virtues and animal qualities, he remains careful in his wording: the analogous character traits in animals are never called ‘virtues’ in any straightforward way. In the context of ethics, then, we have a similar situation that we found in psychology. Human beings possess an innate disposition (natural virtue) that is the capacity for good action, open to a natural development of the living being towards the good. However, this capacity needs practical wisdom to be successful in life. In Aristotle’s biological works character is one of the four biological differentiae by which animals are said to differ from one another, in addition to their ways of life, actions, and parts (Cf. HA IX.3, 610b20–22, IX.44, 629b5–8). They apply to animals and humans alike.39 Character is defined as a natural capacity (phusikê dunamis) of the soul that predisposes the emotions, actions, and cognitive acts related to survival and procreation (HA 588a29–b10). It depends primarily on the living being’s material nature, viz. the particular elemental blend of its body, and the kinds and amounts of food it can digest [see Leunissen (2012)]. Differences in material nature may thus lead to differences of degree in natural character, and hence, in differences of degree regarding the success of development of virtue proper among human beings. Hence the lawgiver of Aristotle’s Politics can be advised to select citizens that are ‘most easily led to virtue’.40 In this respect, of course, Aristotle considers the Greeks as living in the best part of the inhabited world, whereas North-Europeans and Asians come with less favourable dispositions. By way of conclusion we may point out that this chapter has shown how Aristotle’s belief in the scala naturae, the hierarchical continuity of animals and humans,41 relies on the notions of potentiality and actuality as applied in the continuum of psychology, biology, ethics, and politics. Without disregarding the boundaries that separate the species, the notions of potentiality and actuality lead us from primitive 39

HA VIII.1, 588a18–29. For discussion where comparison as to degree stops and analogy starts see Leunissen (2012). 40 Pol. VII.7, 1325b39–1326a5, 1327b18–38, p. 37–38. Cf. PA II.2, 648a2–11 on the relation between different qualities of blood and profitable conditions for courage and practical wisdom. See further Leunissen (2012). 41 Cf. DA II.3; PA II.10, 656a1–13; IV.5, 681a10–15; the scala also pertains to their character traits, see HA VIII.1, 588b4–12.

Potentiality in Aristotle’s Psychology and Ethics

89

animals to the full actualization of human happiness, from innate capacities to further dispositions rich with new capacities that serve animal life, in particular the good life of human beings.

References Ackrill, J. L. (1972–1973). Aristotle’s definitions of Psuchê. Proceedings of the Aristotelian Society, 73, 119–133. Bobonich, C., & Destrée, P. (Eds.). (2007). Akrasia in Greek philosophy. From Socrates to Plotinus. Leiden-Boston: Brill. Bowin, J. (2011). Aristotle on various types of alteration in De Anima II 5. Phronesis, 56, 138– 161. Burnyeat, M. F. (1980). Aristotle on learning to be good. In A. O. Rorty (Ed.), Essays on Aristotle’s ethics (pp. 69–92). Berkeley. Burnyeat, M. F. (2002). De Anima II.5. Phronesis, 47(1), 28–90. Caston, V. (1999). Aristotle’s two intellects: A modest proposal. Phronesis, 44(3), 199–227. Caston, V. (2000). Aristotle’s argument for why the understanding is not compounded with the body. Boston Area Colloquium in Ancient Philosophy, 16, 135–175. Caston, V. (2002). Aristotle on consciousness. Mind, 111, 715–815. Caston, V. (2004). More on Aristotle on consciousness: Reply to Sisko. Mind, 113, 523–533. Caston, V. (2006). Aristotle’s psychology. In M. L. Gill & P. Pellegrin (Eds.), A companion to ancient philosophy (pp. 316–346). Oxford: Blackwell Publishers. Charles, D. (2007). Aristotle’s weak akrates: What does her ignorance consist in? In C. Bobonich & P. Destrée (Eds.), Akrasia in Greek philosophy. From Socrates to Plotinus (pp. 193–214). Leiden-Boston: Brill. Charles, D. (2009a). Nicomachean ethics VII.3: Varieties of akrasia. In C. Natali (Ed.), Aristotle’s “Nicomachean ethics”, book VII. Symposium Aristotelicum (pp. 41–71). Oxford: Oxford University Press. Charles, D. (2009b). Aristotle on desire and action. In D. Frede & B. Reis (Eds.), Body and soul in ancient philosophy (pp. 291–308). Berlin-New York: Walter de Gruyter. Charles, D. (2010). Metaphysics H.7 and 8. Some issues concerning actuality and potentiality. In R. Bolton & J. G. Lennox (Eds.), Being, nature and life in Aristotle (pp. 168–197). Cambridge: Cambridge University Press. Charles, D. (2011a). Akrasia: the rest of the story? In M. Pakaluk & G. Pearson (Eds.), Moral psychology and human action in Aristotle (pp. 187–209). Oxford: Oxford University Press. Charles, D. (2011b). Desire in action: Aristotle’s move. In M. Pakaluk & G. Pearson (Eds.), Moral psychology and human action in Aristotle (pp. 75–93). Oxford: Oxford University Press. De Haas, F. A. J. (1999). Mixture in Philoponus. An encounter with a third kind of potentiality. In J. M. M. H. Thijssen & H. A. G. Braakhuis (Eds.), The commentary tradition on “De generatione et corruptione”: Ancient, Medieval and early modern (pp. 21–46). Turnhout: Brepols. De Haas, F. A. J. (2000). Recollection and potentiality in Philoponus. In M. Kardaun & J. Spruit (Eds.), The winged chariot. Collected essays on plato and platonism in honour of L. M. de Rijk (pp. 165–184). Leiden: Brill. Destrée, P. (2007). Aristotle on the causes of Akrasia. In C. Bobonich & P. Destrée (Eds.), Akrasia in Greek philosophy. From Socrates to Plotinus (pp. 139–165). Leiden-Boston: Brill. Düring, I. (1961). Aristotle’s “Protrepticus”. An attempt at reconstruction. Göteborg: Elanders Boktryckeri Aktiebolag. Everson, S. (1997). Aristotle on perception. Oxford: Oxford University Press.

90

F. A. J. de Haas

Frede, M. (1996). La théorie aristotélicienne de l’intellect agent. In G. Romeyer-Dherbey & C. Viano (Eds.), Corps et âme: sur le De anima d’Aristote (pp. 377–390). Paris: Librairie philosophique J. Vrin. Frede, D., & Reis, B. (Eds.). (2009). Body and soul in ancient philosophy. Berlin-New York: Walter de Gruyter. Gosling, J. C. B. (1993). Mad, drunk, or asleep?—Aristotle’s akratic. Phronesis, 38(1), 98–104. Heinaman, R. (2007). Actuality, potentiality and De Anima II.5. Phronesis, 52, 139–187. Johansen, T. K. (1997). Aristotle on the sense-organs. Cambridge: Cambridge University Press. Johansen, T. K. (2012a). Capacity and potentiality: Aristotle’s Metaphysics H. 6–7 from the perspective of the De Anima. Topoi, 31, 209–220. Johansen, T. K. (2012b). The powers of Aristotle’s soul. Oxford: Oxford University Press. Kosman, L. A. (1999). Being properly affected: Virtues and feelings in Aristotle’s ethics. In N. Sherman (Ed.), Aristotle’s ethics. Critical essays (pp. 261–275). Lanham: Rowman & Littlefield. Labarrière, J.-L. (2005). La condition animale. Études sur Aristote et les Stoïciens. Leuven: Peeters. Lawrence, G. (2011). Acquiring character: Becoming grown-up. In M. Pakaluk & G. Pearson (Eds.), Moral psychology and human action in Aristotle (pp. 233–283). Oxford: Oxford University Press. Leighton, S. (2011). Inappropriate passion. In J. Miller (Ed.), Aristotle’s “Nicomachean ethics”. A critical guide (pp. 211–235). Cambridge: Cambridge University Press. Lennox, J. G. (1995). Material and formal natures in Aristotle’s De Partibus Animalium. Proceedings of the Boston Area Colloquium in Ancient Philosophy, 11, 216–240. Lennox, J. G. (1999). Aristotle on the biological roots of virtue. The natural history of natural virtue. In J. Maienschein & M. Ruse (Eds.), Biology and the foundation of ethics (pp. 10–31). Cambridge: Cambridge University Press. Lennox, J. G. (2010). Bios and explanatory unity in Aristotle’s biology. In D. Charles (Ed.), Definition in Greek philosophy (pp. 329–355). Oxford: Oxford University Press. Leunissen, M. E. M. P. J. (2012). Aristotle on natural character and its implications for moral development. Journal of the History of Philosophy, 50(4), 507–530. Lorenz, H. (2006). The brute within: Appetitive desire in Plato and Aristotle. Oxford: Oxford University Press. Menn, S. (1994). The origins of Aristotle’s concept of ἐmέqceia: ἐmέqceia and dύmali1. Ancient Philosophy, 14, 73–114. Menn, S. (2002). Aristotle’s definition of soul and the programme of the De Anima. Oxford Studies in Ancient Philosophy, 22, 83–139. Moss, J. (2009). Akrasia and perceptual illusion. Archiv für Geschichte der Philosophie, 91, 119– 156. Natali, C. (Ed.) (2009). Aristotle’s Nicomachean ethics, book VII. Symposium Aristotelicum. Oxford: Oxford University Press. Pakaluk, M., & Pearson, G. (Eds.). (2011). Moral psychology and human action in Aristotle. Oxford: Oxford University Press. Polansky, R. (2007). Aristotle’s De Anima. Cambridge: Cambridge University Press. Sherman, N. (1996). The fabric of character. Aristotle’s theory of virtue. Oxford: Clarendon Press. Shields, C. (1999). Order in multiplicity: Homonymy in the philosophy of Aristotle. Oxford: Clarendon Press. Sisko, J. E. (1996). Material alteration and cognitive activity in Aristotle’s De Anima. Phronesis, 41(2), 138–157. Sisko, J. E. (1998). Alteration and quasi-alteration: A critical notice of Stephen Everson, Aristotle on perception. Oxford Studies in Ancient Philosophy, 16, 331–352. Sisko, J. E. (2004). Reflexive awareness does belong to the main function of perception: Reply to Victor Caston. Mind, 113, 513–521.

Potentiality in Aristotle’s Psychology and Ethics

91

Sorabji, R. R. K. (1992). Intentionality and physiologial processes: Aristotle’s theory of sense-perception. In M.C. Nussbaum & A. Oksenberg Rorty (Eds.), Essays on Aristotle’s De Anima (pp. 195–225). Oxford: Clarendon Press. Sorabji, R. R. K. (2001). Aristotle on sensory processes and intentionality: A reply to Myles Burnyeat. In D. Perler (Ed.), Ancient and Medieval theories of intentionality (pp. 49–61). Leiden: Brill. Witt, C. (1995). Powers and possibilities: Aristotle vs. the Megarians. Proceedings of the Boston Area Colloquium in Ancient Philosophy, 11, 249–266. Witt, C. (2003). Ways of being: Potentiality and actuality in Aristotle’s metaphysics. Ithaca, NY: Cornell University Press. Woods, M. (1982). Aristotle’s Eudemian ethics: Books I, II, and VIII. (trans: Woods, M., [from the Greek] with a commentary) Oxford: Clarendon Press. Zingano, M. (2007). Akrasia and the method of ethics. In C. Bobonich & P. Destrée (Eds.), Akrasia in Greek philosophy. From Socrates to Plotinus (pp. 167–191). Leiden-Boston: Brill.

Author Biography Frans A. J. de Haas is Professor of Ancient and Medieval Philosophy at Leiden University, The Netherlands. His research focuses on Aristotle and the reception of Aristotelian philosophy in Late Antiquity, with emphasis on psychology and philosophy of mind. He has published on the concept of matter, on ancient philosophy of science and its relation to mathematics, and on the innovations that Alexander of Aphrodisias introduced into Aristotelian philosophy. Key publications include “John Philoponus’ New Definition of Prime Matter. Aspects of its Background in Neoplatonism and the Ancient Commentary Tradition” (Philosophia Antiqua, Brill 1997), “Aporiai 3–5” [between universal science and first philosophy] (in: M. Crubellier and A. Laks (eds.), Aristotle: Metaphysics Beta. Symposium Aristotelicum, Oxford University Press 2009), “Principles, conversion, and circular proof. The reception of an Academic debate in Proclus and Philoponus” (in: T. Bénatouïl, F. Trabattoni and G. Van Riel. (eds.), Plato, Aristotle or both? Dialogues between Platonism and Aristotelianism in Antiquity, Olms 2011).

Part II

The Concept of Potentiality in the History of Philosophy—Medieval Philosophy

Potentiality in Classical Arabic Thought Taneli Kukkonen

Much of what the major Arabic philosophers have to say about the concept of potentiality is relatively unremarkable. The principal authors in the classical period —al-Kindī, al-Fārābī, Ibn Sīnā (Avicenna), and Ibn Rushd (Averroes)—for the most part reformulate familiar points made in the Metaphysics and Physics and Aristotle’s other treatises in natural philosophy. But the Arabic authors capably synthesize Aristotelian insights and present them in attractive, well-ordered, and pithy ways; what is more, their texts prove particularly perceptive when it comes to discerning what Aristotelian potentiality is for—what the uses are to which the notion is put and what purposes it serves in the larger project of constructing a philosophical account of nature and a general ontology. This is because Avicenna (980–1037 CE) and Averroes (1126–1198), in particular, had to contend with the ontology of kalām theology, which was unabashedly occasionalist in outlook and thus fundamentally inimical to the received picture of Aristotelian potentialities.

1 Early Developments From the late 8th century CE onwards, Aristotelian potentialities formed an essential part of the Muslim philosopher’s toolkit. The term will first have been encountered in full force in an early translation of Aristotle’s Physics, which in the third book defines motion in terms of the ‘actuality of the potential qua potential’ (Phys. III.1, 201a10–11). From here, the prominence of potentiality within the Aristotelian explication of the workings of nature became apparent through an increasing acquaintance with the rest of Aristotle’s natural philosophy (biology even more than elemental physics). The central place afforded to potentialities in T. Kukkonen (&) New York University Abu Dhabi, Abu Dhabi, United Arab Emirates e-mail: [email protected] © Springer Science+Business Media B.V., part of Springer Nature 2018 K. Engelhard and M. Quante (eds.), Handbook of Potentiality, https://doi.org/10.1007/978-94-024-1287-1_5

95

96

T. Kukkonen

Aristotle’s mature ontology, meanwhile, was revealed to the Arabic philosophers once they began to work in earnest on the Metaphysics, in particular books Theta and Lambda. Because the two books were read in conjunction, medieval discussions of potentiality typically came embedded within a grand cosmological framework in which one descends from the pure actuality of the divine on down until one reaches the pure potentiality of prime matter. Not that there was anything specifically Islamic about this: the utilization of the concepts of potentiality and actuality in service of a hierarchical vision of reality was common to all late antique and medieval philosophy.1 For all that potentiality and actuality were central to the enterprise of Arabic philosophy from the start, it is noticeable how two of the most prominent early Muslim philosophers, Abū Yūsuf Ya‘qūb al-Kindī (d. ca. 870) and Abū Naṣr al-Fārābī (d. 950), appear simply to seize upon the concept of potentiality rather than going to the trouble of defining it. To be more precise: although both philosophers, each of them standing at the head of whole interpretive traditions of Aristotelianism, produced treatises in which basic philosophical terminology is explained to newcomers to the field, neither potentiality nor its Aristotelian counterpart, actuality, is singled out for analysis in any of these introductory works. To say, as al-Kindī does in his treatise On First Philosophy, that “potentiality is nothing other than the possibility of that thing existing which is said to be potentially” (Oeuvres 2:30.16–7), is not so much to offer a definition as to state the obvious—that any mention of potentiality appears to evoke some sort of pre-existent possibility. It appears that both al-Kindī and al-Fārābī considered potentiality to be a primitive term of sorts, one whose immediate sense was sufficiently obvious to render it useful in the elucidation of other concepts. Both philosophers do expound on aspects of Aristotelian potentiality in an incidental way —as occasion demands, one might say—typically in contexts where Aristotle himself adds conceptual flourishes and complications. Examples would be the curious case of the potential infinite, which never blossoms into actuality, and the different degrees of potentiality and actuality in the soul. But the extent to which potentiality in Arabic Aristotelianism simply forms part of the assumed backdrop to the wider project of falsafa (classical Arabic philosophy) is striking. A few general features of the Aristotelian analysis of potentiality bear noting insofar as early philosophers working in the Arabic tradition commonly recognized these already. First and most importantly, any given potential on the Aristotelian picture is invariably defined in terms of its corresponding actuality: a potential is

1

The late antique view came to be transmitted to the Arabic milieu through sources such as the pseudo-Aristotelian treatises The Theology of Aristotle (which in truth is an adaptation of Plotinus) and The Book on the Pure Good (the Latin Liber de causis, which constitutes a heavily edited set of propositions from Proclus’s Elements of Theology). Accuracy demands that I acknowledge that in late antique Platonism, the notion reared its head that the very highest and most divine principles should be described in terms of potency rather than actuality, and that this view also found its way into the Arabic adaptation of Neoplatonic materials: see Adamson (2002), 94–102.

Potentiality in Classical Arabic Thought

97

always a potential for some determinate and definable outcome. This means that the actual enjoys both conceptual and ontological priority over the potential—a familiar Aristotelian point (Met. IX. 8) repeated again and again in the Arabic materials. Second, the determinacy of the range of potentialities habitually gets applied to the study of natural kinds rather than individuals. Al-Fārābī, for instance, whom later historians came to honour as the ‘second teacher’ of philosophy after only Aristotle himself, in his work On the Perfect State states his interest in how each species (naw‘) becomes actually existent (III. 5.1, 108.15 Walzer). In no text I have canvassed does any Arabic thinker take a per se interest in, say, Zayd’s potential for becoming a philosopher; at most there is a question of why Zayd became a philosopher while Amr did not, that is, why some members of the human species actualize the innate human capacity for rationality while others fail to do so (assuming the definition of the human being as ‘rational animal’ is correct). The notion of a strictly individual potentiality is close to being a nonsense on the picture adopted by the Arabic Aristotelians; at the very least, there is nothing much to be said about it in the philosophical sense, given that the particular qua particular is unknowable (see Adamson 2005). And because the emphasis is so squarely on specific (i.e., species-level) potentialities, the question about unrealized potentialities also typically assumes the form of asking whether it makes sense to attribute a potentiality to a species or a natural kind if members of that species do not actualize that potentiality always, for the most part, or at all. Third, it was Aristotelian hylomorphism that provided the overall framework for how potentiality and actuality were understood. Thus for instance al-Fārābī in his survey work The Philosophy of Aristotle straightforwardly equates matter with ‘the principle by virtue of which a thing is potentially’, while form is that ‘by which a thing is in act’.2 This of course follows the precedent set in Aristotle’s own Physics, where Aristotle mentions in passing that in addition to the tripartite division between matter, form, and privation, one may also examine the process of becoming in accordance with the broader notions of potentiality and actuality (kata tēn dynamin kai tēn energeian). But where the relevant sentence in Aristotle’s Physics (I.8, 191b27–29) reads like a cross-reference to Metaphysics Theta that has been added at a later date, Arabic thinkers saw in it an acknowledgement of a wider systematic confluence, where not only the larger concerns of Aristotelian metaphysics but also the account of coming-to-be and passing-away could be introduced as part of an integrated system of thought (see De gen. et corr. I.3, 317b13–33 for the language of potentiality and actuality). Telling here is Averroes’s summary regarding the whole point of the first book of the Physics. According to Averroes, potentiality was posited by Aristotle as a third, “different nature, intermediate

2

Philosophy of Aristotle IV.20; likewise On the Perfect State III. 5.1, 108.15–109.4; cp. al-Kindī, On the Five Substances, in Opera, 2:20.12–13, which aims to explicate the same terminology, but seems rather confused, comparatively speaking; and for Aristotle, De an. II.1, 412a9.

98

T. Kukkonen

between pure being and pure nonbeing”, all in an effort to solve the general problem of coming-to-be in the world.3 Fourth, the distinction between active and passive potencies is applied systematically by Arabic philosophers, and the point gets stressed that any genuine change in the world comes about through a confluence of the two powers. Again taking al-Fārābī as our example, the Second Teacher in the immediate follow-up to the aforementioned passage in The Philosophy of Aristotle acknowledges that on the Aristotelian picture no potentiality ever achieves actuality of its own accord, instead requiring an activating principle external to it in order to become actualized. This further principle al-Fārābī calls the ‘agent’ (al-fā‘il), i.e., the efficient or moving cause in Aristotle’s classification of the four causes (op. cit., IV.21). But what he is really bringing into the picture is the Aristotelian binary of the active potency and the passive potency, each of which is needed in order to explain any given instance of change.4 Active and passive potencies are further conflated in the Arabic materials with the partially overlapping notions of acting and being-acted-upon (fi‘l wa-infi‘āl), terms first introduced in Aristotle’s Categories as poiein and paschein. Although the Greek concepts of actuality (energeia) and acting (poiein) are not identical, the move nonetheless is hardly surprising: Aristotle’s language in the Physics allows for it, while the way in which Aristotle expands on the notions of acting and being-acted-upon in On coming-to-be and passing-away explicitly evokes potentiality and actuality. Thus we find Averroes, for instance, in his Paraphrase of Aristotle’s De gen. et corr. I.9 intimating that his readers should by now be familiar with this terminology and its centrality in Aristotle: We posit a principle which we have posited several times, which is that things are composed of matter and form; that some of them are potentially some thing and these are [the things which are] acted upon while some of them are in actuality that precise thing and therefore the agent; and that under these circumstances the cause for something’s being acted upon is nothing other than the patient’s susceptibility to [be acted upon by] the active thing. (Paraphrase 78 [tr. Samuel Kurland])

Averroes’s remarks form part of a larger effort to establish what level of similarity must hold between the external actualizing agent of a change and the thing that undergoes the change, a question whose beginnings the Arabic philosophers would trace via Plato and Aristotle back to Empedocles. Averroes’s purposefully vague choice of term (‘susceptibility’) neatly captures the notion of passive potencies

3

In Phys. I, comm. 78, fol. 4:44G; Commentary IX, comm. 6 has the same formulation in the Latin, but not in the original Arabic; similarly Ibn Bājja, On Coming-to-be and Passing-away, 46– 47. 4 See Aristotle, Phys. III.2, 202a3–12; Phys. VIII.4, 255a30–b3: al-Fārābī goes on to cite the universality of ends and goals in all instances of natural change, thus completing the standard list of four causes (Philosophy of Aristotle IV.22). This, too, ties in with potentiality and actuality: since nature does nothing in vain, potentialities only develop in relation to some determinate end-states that are to be considered perfections for the being whose potentialities they are, i.e., actualities.

Potentiality in Classical Arabic Thought

99

being identified in terms of corresponding active agents and a suitable relation holding between the two, while leaving open the precise nature of that correlation. But what is important for our purposes is how potentialities and actualities, on the picture presented here, become firmly embedded in regular and naturally constituted processes between interlocking types of entities and their respective causal powers. Fifth and finally (following from the previous point), even though Arabic philosophers take an interest in all the contexts in which Aristotelian potentialities are found—as Avicenna would put it, even though questions concerning potentiality arise in contexts as diverse as nature, habit, art, and even coincidence (Healing: Metaphysics IV.2.14)—it is in their study of natural capacities that a continuous thread of discussion, indeed an evolution in thought, is most readily discernible. The systematization of Aristotelian natural philosophy in the light of the guiding concepts of potentiality and actuality begins early and is thoroughgoing. Thus for instance al-Fārābī grounds the cycle of elements in the powers (pl. quwan) which allow them to act on one another.5 Terminologically, this is an advance on Aristotle, who in the treatise On coming-to-be and passing-away speaks merely of differentiae and contrarieties (diaforai kai enantiōseis: De gen. et corr. II.2, 329b17) when it comes to explaining how earth, water, air, and fire act upon one another. Ibn Bājja (d. 1139), an Andalusian philosopher who drew heavily on al-Fārābī for his understanding of the Peripatetic tradition, in his own treatise On Coming-to-be and Passing-away (12, 14ff.) develops this notion into a dynamic understanding of all physical reality. For Ibn Bājja, the four elements are made what they are by their primary qualities, which in turn must be understood in terms of motion-causing powers. Potentialities thus form the very foundation of elemental physics through the way that they define the four simple bodies of earth, water, air, and fire (op. cit., 26–40). For a thoroughgoing treatment of the concepts of potentiality and actuality we turn to the two best-known Arabic philosophers of all time, Avicenna and Averroes, who wrote at opposite ends of the Islamic world in the early 11th and late 12th century, respectively. The principal reason why out of the classical Arabic philosophers, Avicenna and Averroes have the most to say on potentiality has to do primarily with the modalities of literary production. Averroes wrote some forty commentaries, paraphrases, and compendia explicating Aristotle’s philosophy, while Avicenna in his encyclopaedic works sought to rewrite and recast Peripatetic philosophy in his own image, so that it became necessary for him to think through what the role was that each Aristotelian concept played in the whole of Peripatetic learning (and what role it should play in Avicenna’s renewed and revitalized Aristotelianism). In this manner, Averroes and Avicenna both ended up generating

5

Philosophy of Aristotle VII.44; see Anna Marmodoro’s chapter in this book for the Greek provenance of this notion.

100

T. Kukkonen

a great deal of material on potentiality and actuality. One did so in an attempt to expound on Aristotle, while the other did so in an effort to supplant him. Luckily, Avicenna’s and Averroes observations are not only voluminous, but insightful as well. Both Avicenna and Averroes ably summarize and expound on what the Aristotelian notion of potentiality is and what philosophical work it does. One would not hesitate to recommend their treatises as introductions to the subject on a par with any modern treatment. But both Avicenna and Averroes can also be seen to react to an utterly different set of precepts, one that it behoves us now to introduce, even if briefly: the metaphysics of Muslim speculative theology or kalām. The aspect of classical kalām ontology that most interests us in the present context is its staunch denial of any autonomous power to created entities. This follows directly from the desire on the theologians’ part to interpret God’s omnipotence in the most literal manner: all that is issues directly from God’s power (qudra) and his will (irāda), which must stand unopposed by any created thing. Coupled with an atomist understanding of space and time and an overall occasionalist worldview whereby God re-creates the world out of sheer nothingness at each instant, we end up with a construct in which all agency and therefore also all potency resides with God and in which beings and their properties are exactly what they are—indeed, necessarily what they are—at each moment, without any potential for further development in any direction. This could scarcely be further removed from the Aristotelian understanding of passive and active potencies interlocking in a constantly evolving dynamic. As Richard M. Frank puts it, the theological term qudra (or istiṭā‘a) in its proper and native meaning is in every respect opposed to the normal meaning and the content of Greek dynamis, as it is usually used in Greek philosophy… parallel and consistent with this basic understanding, the state of actuality of being for the kalām is, in general, a finite and discrete moment of being, not, as it is for Aristotle, an actuality or reality achieved out of becoming; it is rather the present state of being, perfect and complete in the fullness of its being, discrete and without inherent reference to any past state… for most of the early mutakallimīn, all things are no more than they are and their being is complete and fulfilled at any given moment of their existence. No being has in itself any intrinsic “potentiality” to change and alteration except the knowing, willing, and intending agent; its becoming other is entirely dependent upon and reside in the potentiality of an exterior agent who is capable of effecting the change. (Frank 1966, 18–20)

There was considerable variance among the theologians when it came to the exact workings of divine power within the world (for a survey of views see al-Ash‘arī, Opinions, 548ff.). Nevertheless, the theological mainstream unanimously regarded it as axiomatic that no created entity possesses any capacity other than what God grants it in the moment in which that same capacity finds actualization (see, e.g., Bāqillānī, Introduction, 323–328; Juwaynī, Guidance, 215–225). Such a view eventually spread beyond the confines of formal theology, so that according to the testimony of al-Kalabādhī (d. ca. 995) the Sufis, too,

Potentiality in Classical Arabic Thought

101

agreed that every breath they draw, every glance they make, and every motion they perform, is by virtue of a power which God originates in them, and a capacity which he creates for them at the same time as their actions, neither before them nor after them, and that no action can be performed without these… (Doctrines of the Sufis, ch. 14, tr. Arberry, modified)

Later theologians could self-consciously present this as a tenet held in opposition to the opinions of the philosophers: Abū l-Ma‘ālī l-Juwaynī (d. 1085), for instance, discerns in his master al-Ash‘arī’s teachings on the world’s creation a wholesale repudiation of the party of matter (aṣḥāb hayūlā), “those who say that the world exists in matter potentially, the date palm in the date stone, the man in the drop of sperm”.6 It was thus imperative for Aristotelian philosophers to explicate what philosophical work the notion of potentiality does and why it is an indispensable part of any reasoned approach to the world’s workings.

2 Avicenna: Potentiality as Fitness Avicenna’s most comprehensive account of potentiality is found in book IV, chapter 2 of the section dedicated to Metaphysics in Avicenna’s mid-period philosophical encyclopaedia The Healing.7 The work can, and ultimately should, be read in conjunction with Avicenna’s works in natural philosophy, in which the concept of potentiality is applied across the board along with some refinements. However, as befits the overall standing of first philosophy in relation to the special sciences, the account offered in the Metaphysics is meant to enjoy the greatest level of generality and as such be universally applicable. It therefore forms the appropriate starting point to our investigations as well. Avicenna’s remarks on potentiality proved highly influential, as his refashioning of Peripatetic metaphysics and natural philosophy not only became the dominant philosophical paradigm in the Islamic world: portions of the Healing (pointedly including the Metaphysics) were translated into Latin, too, and became popular among the scholastic philosophers. It can in fact be claimed that Avicenna is the last major philosopher studied in depth in both the European and Middle Eastern branches of Aristotelian philosophy (or Latin and Arabic, or Christian and Islamic). As regards Avicenna’s metaphysical account of potentiality, it is useful first of all to take note of where the lengthy chapter on potentiality and actuality falls within the overall architecture of the Metaphysics. The account is embedded in an account of those things that relate to existence and unity in the manner of properties and concomitant accidents (al-khawwāṣ wa-l-‘awāriḍ al-lāzima). In Aristotelian terms, properties are those characteristics of things which do not directly disclose the essence of the thing or the species to which it belongs, but which nonetheless 6

Comprehensive Book, 132 Klopfer; for this translation Frank (1966), 22. See also Healing: Physics III.1.1, which applies a similar analysis of potentiality to the issue of the eternity of the world. 7

102

T. Kukkonen

are predicated convertibly of it and belong properly to it (Aristotle, Top. I.5, 102a18–30; see Avicenna, Healing: Introduction I.14, 83–85). Avicenna’s own preferred term lāzim or concomitant does much of the same work, in that we are again dealing with an inseparable accident (Pointers and Reminders I.11, 8–9). This does not seem an inappropriate way of classifying the terms potentiality and actuality as they are introduced in the Metaphysics, although Aristotle, true to form, never goes to the trouble of doing so, but instead plunges straight ahead into a discussion of the reality of the two. For Avicenna, then, potentiality and actuality are terms that are inextricably bound up in the characterization and analysis of all manner of beings, even though they do not disclose directly the essence of any one being.8 Second of all, the previous chapter IV.1 in Avicenna’s Metaphysics is dedicated to priority and posteriority, two properties which according to Avicenna similarly structure all of reality. Avicenna, in fact, eases his reader into a discussion of potentiality and actuality by saying that it is their relative priority and posteriority that bears investigating (IV.1.20). The sentiment is echoed in the conclusion: from all that has gone before regarding potentiality and actuality, Avicenna avers, the reader is now supposed to have learned that “in reality, act is prior to potency and, moreover, that it is prior in terms of nobility and perfection” (bi-l-sharf wa-ltamām: IV.2.34). Avicenna, by sandwiching his account of potentiality and actuality between accounts of priority and posteriority on the one hand and the complete, the more than complete, the sufficient, and the deficient on the other, generates the impression that the potential and the actual not only are hierarchically ranked when it comes to the being of each thing, but that all existents may be ranked according to the degree of potency and actuality that they enjoy. Potentiality and actuality thus emerge as structuring principles for establishing a global hierarchy of being.9 Third and perhaps most remarkable is the fact that Avicenna wraps his exposition of potentiality and actuality in the form of an unfolding historical narrative. Instead of setting forth his own views on potentiality, starting from first principles on down, Avicenna builds his account from the outside in, as it were: he starts from what he terms a primitive, everyday understanding of potentiality and from there advances to more abstract considerations. What is more, this progression of thought is at the same time presented as an actual historical development. Avicenna presents his case as though people in the past really would have started from everyday observations and then refined their thinking in stages.

8

This is of course due to their very universality. Averroes follows Avicenna on this point in his early Epitome Aristotle’s Metaphysics, calling potentiality and actuality concomitants of being (Epitome, 2–3, 34, and 79); in the later Commentary (VII, prooemium, 744), however, potentiality and actuality apparently count among the species of being (anwā‘ al-mawjūd). 9 One may compare this with Witt’s (2005) recent reading of Metaphysics Theta, which similarly stresses the normative and hierarchical aspect of Aristotle’s work. The difference, of course, is that Avicenna leans on a Neoplatonic metaphysical hierarchy: see, e.g., Netton (1989) for an overview of such hierarchies in Arabic philosophy, and also n. 1 above.

Potentiality in Classical Arabic Thought

103

Certainly, sufficient Peripatetic precedent existed for this mode of presentation to come naturally to Avicenna. Aristotle himself more than once indicates that raw everyday notions of how things are may gradually give way to a more precise conceptual apparatus; while within Avicenna’s proximate intellectual environment, al-Fārābī in the second part of his Book of Letters—an important precursor to Avicenna’s exposition of metaphysics—had theorized about the way in which a process of conceptual refinement might take place in stages in accordance with a culture’s improving grasp both of the rules of argumentation and of the contours of reality. So it is possible that Avicenna’s story was never meant to be taken entirely in earnest, but that it was instead presented as a mere rhetorical construct, one whose very obviousness telegraphed to the educated reader that it should be taken allegorically. Whatever the provenance of Avicenna’s style of presentation, the fact that Avicenna does introduce his account by making a claim about the way in which people first came by the term potentiality or quwwa establishes the intuitive mode of speaking as the bottom line for conversation about both the extension and intension of the term, much as Aristotle’s allusion to the obvious referents for the term ‘substance’ had done for individual animals, plants, and other natural bodies (see Met. VIII.1, 1041b5–11; De an. II.1, 412a11–13). To be more precise: Avicenna establishes as a baseline for the understanding of potentiality the notion of power (qudra), that which makes it possible for a living creature to undertake acts of a certain nature. In this sense, potentiality, understood as capacity, is primarily defined against incapacity (‘ujz: see Healing: Metaphysics IV.2.1). On top of this, Avicenna adds an odd flourish which is lifted directly from Aristotle’s Metaphysics (V.12, 1019a25–31): by potency one can also mean an entity’s ability to withstand the actions of others, that is, the capacity not to be acted upon (Healing: Metaphysics IV.2.2). The principal point, then, is that potentiality denotes “power itself, which is a state belonging to the animal by means of which it acts or does not act” (IV.2.3). What would have led Avicenna to begin his exposition of potentiality from an account of power understood in such zoo- or even anthropocentric terms? An otherwise oddly placed pair of paragraphs (IV.2.8–9) provides a clue. Avicenna brings up an imaginary objector who claims that on the picture presented so far, no agent that is active eternally and forever can be said to possess the potentiality for doing so. After all, it looks as though such an agent is not in the business (min sha‘ni-hi) of both doing something and not doing it. Although Avicenna couches the issue in general terms, there is no mistaking the theological subtext: the issue ultimately has to do with whether God, who according to the philosophical theory is eternal and changeless and stands in an eternal relationship to an equally eternal universe, can on the parameters laid down be said to exercise power at all (provided the latter is defined as the ability to withhold as well as to grant and as a power towards some thing as much as towards its opposite, as was standard in classical kalām expositions concerning the divine attribute of qudra: see, e.g., al-Bāqillānī, Introduction, 337). Avicenna’s contention is that he can and that consequently, the notion of an eternally and uniformly active potency poses no problem either

104

T. Kukkonen

philosophically or theologically.10 The whole flourish is meant to catch the eye of the reader attuned to such matters without needlessly distracting the more philosophically inclined. Consistent with his practice in other contexts (see Marmura 1991–1992), Avicenna is content to criticize the kalām views he finds objectionable in a manner that is at once fairly oblique, yet at the same time unmistakable to the educated reader. Avicenna tables the notion of potentiality as power only to set it aside, in effect reducing the anthropomorphic theological conception of God as a powerful agent (fa‘‘āl qādir) to the level of humanity’s childish and fumbling first attempts to come to grips with the nature of potentiality. It is a sly and surprisingly savage move, meant to undercut the credibility of the whole kalām mode of analysis. This emerges more clearly once we proceed through the text. From a consideration of what he terms elementary and popular notions of quwwa Avicenna proposes to move to properly philosophical work done on the concept. First, and crucially, the philosophers according to Avicenna expanded their understanding of potentiality to encompass non-volitional acts, to the point where even an inanimate quality such as heat could be considered a potency residing in the hot substance, since it is this that allows one thing to affect another in a determinate fashion (in the example, by heating it: IV.2.4). Avicenna here alludes to the separation of active from passive potencies, with a specific reference to Aristotle’s refutation of self-movers in Physics VIII.4. In any legitimate instance of worldly change, we must recognize the coming together of an active principle and a passive principle, that which does the changing and that which changes, so that even in a supposed instance of self-motion (a doctor healing herself is the classic example), this happens insofar as the agent takes itself as an object of action qua another. The doctor, in other words, does not heal a doctor but a patient, who in this instance just happens to be a doctor as well (and numerically identical with the doctor: see, e.g., Healing: Physics I.5.6, I.10.2; also Salvation, 214). Similarly, to take another example from the Physics of the Healing (I.5.2), a piece of iron that appears to move itself does not in fact do so, but instead is moved by a magnet: nor does it move insofar as it is a body, instead, the motion occurs due to a specific potentiality—in this case, a passive potentiality—which allows the body made of iron to be moved while other bodies around it stand still. A yet more important historical innovation, if we follow Avicenna’s developmental narrative, was the shift in attention from the concrete notion of power to the altogether more abstract concept of possibility (imkān: IV.2.5). This reconceptualization, which represents a further level of abstraction, was—so Avicenna says— triggered when ancient thinkers recognized that it makes sense to refer to

10 Avicenna’s argument, which is somewhat beside the point of our investigation, comes in two parts. First, Avicenna posits that will and choice (irāda wa ikhtiyār) can assume a uniform and unchanging nature without thereby losing their volitional character; second, he appeals to the logical point that hypothetical propositions of the general form “If he does not will, then he will not act” remain true qua hypotheticals even when the antecedent is not realized. See also Healing: Metaphysics IV.2.27–34 for a refutation of the view that the priority of the actual over the potential would mean that God must precede the universe in the temporal sense.

Potentiality in Classical Arabic Thought

105

potentialities as something real even at times when they are not realized. This, too, contains a barb aimed at the kalām theologians, although its significance only becomes apparent later. The theologians, after all, had with very few exceptions denied precisely that a thing’s possibility would precede its actuality. Avicenna connects this view with a certain Ghāriqū among the ancients, which is certainly a reference to the Megarians whom Aristotle criticizes for holding the precise same view in the Metaphysics (IX.3), and with certain later thinkers (mu’akhkhirūn) who go unnamed but must be taken as a reference to the Ash‘arite school (Healing: Metaphysics IV.2.15). Avicenna’s response is to ridicule Ghāriqū, and with him the theologians, for positing a view that is plainly nonsensical: a person on the picture they present, so Avicenna says, remains blind at all times except for those in which she or he happens to be seeing, just as the seated person does not have the ability to rise except for—all of a sudden—those moments in which she or he rises. Yet what could account for a wooden piece’s chiselling into a key if not the disposition to be so chiselled? (Healing: Metaphysics IV.2.15–16) Avicenna’s implicit answer is that —contrary to what the Megarians and theologians had suggested—a propensity or disposition (jibla) towards a given kind of change is precisely what must be assumed to exist prior to the change taking place. This discussion leads to what is presented as the centrepiece of Avicenna’s argument regarding potentiality and actuality. As the subheading of the chapter has it, his account of potentiality is intended “to affirm the existence of matter (Ar. mādda) in all that is generated” (IV.2, 130.8). Every real potentiality resides in a subject—it is some pre-existent thing’s possibility for taking on some trait, for being changed a certain way, or for transforming into something else entirely—and therefore is classifiable as a potentiality pre-existing its actualization. This means that the theologians, as soon as they admit that God’s power does not take as its object that which is impossible but instead must have possible objects towards which it is directed (maqdūr ‘alay-hi), must, if they wish to remain consistent, admit that no possibility arises out of sheer nothingness but instead comes to be actualized from a potentiality. (Healing: Metaphysics IV.2.24–25; see similarly Salvation, 219; Pointers and Reminders, 151.) The vocabulary matters here. Avicenna is not talking exclusively or even primarily about prime matter, hylē/hayūlā, the unknowable stuff that is supposed to underlie all corporeal reality. Instead, his concern is over the acknowledgement of matter as substrate (Ar. maḥall), that which is needed to account for each individual instance of generation and corruption, at every level of ontological complexity. Wherever the real possibility of something’s coming to be is discussed, a subject can be identified as the proximate site in or from which that actuality eventuates, and it must therefore be the case that some specific thing was potentially x before actually coming to be x. (See Salvation, 213–214, 216–218.) We call the possibility of existence the potentiality of existence. And we term the bearer of the potentiality of existence, which has the potentiality of a thing’s existence, “subject”, “hyle”, “matter” and other [names] according to various ways of considering [things]. (Healing: Metaphysics IV.2.26)

106

T. Kukkonen

So matter and potentiality are inextricably bound, and found, together, though one is not the same as the other. At the same time, not everything bears the potentiality for the emergence of just anything. For the concept of potentiality to have any practical usage, one must pay attention to how specific material conditions delimit the possibility of specific kinds of actualities coming to pass. Here the Peripatetic doctrine of prime matter (mādda ūlā) being potentially all things acquires fresh resonance (Healing: Metaphysics IV.2.12). As certain forms are realized in matter, some of its potentialities are shut off, though they may become fresh possibilities again through the divestiture of form. Under certain conditions, then, the possession of a specific form can be seen as an impediment, something to which Avicenna’s vocabulary in the Healing testifies: the water in a pot of tea cannot easily be converted for use in a broth anymore nor, as per Averroes’s later example, can a donkey be taught to play the lute. The nested, indeed hierarchical ordering of potentialities allows for a fairly sophisticated consideration of possibilities in nature. The primary tool for this is Avicenna’s distinction between proximate and remote potentialities. For Avicenna, it does make a degree of sense to talk about the sperm’s potentiality to become a man, even though the sperm cannot do so immediately and proximately, but instead must pass through boyhood first. In this case, the sperm has the remote but not yet the proximate potentiality of developing into a fully grown human being (IV.2.12). Similarly the tree is not yet potentially a key in the proximate sense: it must first be uprooted, sawed, and chiselled, at which point it stands in readiness to receive ‘keyness’ from the effective power (quwwa fā‘iliyya: IV.2.13). In the Physics of the Healing (I.12.4) Avicenna speaks of potential and actual subjects in this connection, as well as remote and proximate subjects (and general and specific, and simple and composite). According to Avicenna’s explanation, semen is potentially a subject for the human form to come into being just as, say, unworked wood is potentially the subject for a particular chair, while the wholly developed human body is the actual subject of the human being and a set of wooden pieces, sawed and sanded just so, taken together forms the actual subject matter for the chair.11 On this account, passive potentiality amounts to a certain kind of fitness or suitability (isti‘dād), a concept Avicenna appropriates from late antique Platonism and grafts back onto a Peripatetic framework.12 The basic message is one that is reiterated in a number of contexts across Avicenna’s works in natural philosophy (see, e.g., Healing: Physics II.3.6; Healing: Animals IX.5, 172–178). A form—any given form, not only substantial form—will only enter matter that has been prepared just so and that stands in just the right kind of relationship to the actuality that it is to receive. This is why it is important that the remote subject be in place first for 11 See similarly Averroes, Epitome, 85–87. Averroes’s preferred example comes from the elements becoming food, which through the process of nutrition becomes blood, which in turn becomes flesh. The elements in a sense are flesh in potentiality, but only in a very remote sense, bordering on the metaphorical (bi-tajawwuz): as in Aristotle’s Parts of Animals (II.3–6), it is the blood that bears the full and proximate potentiality for becoming flesh. 12 For the Neoplatonic background see Sambursky (1962, 104–110), George (2005, 290–293).

Potentiality in Classical Arabic Thought

107

the proximate subject to come into being, for instance, that the male body first generate the sperm, which then becomes the fetus, which then in turn becomes the human child and consequently the full-grown adult. In each case, a precise kind of material fitness exists that allows the generation of the more complex type of form: a drop of sweat would not be fitted to receive the principle of life the way the male seed is, nor again are an infant’s toes capable of becoming the locus of the senses in the way the sense organs are, etc. In practice, fitness on Avicenna’s understanding cashes out in terms of an appropriate elemental mixture: flammability is based on an excess of dryness, for example, while in a more complex case an animal’s receptiveness to the higher cognitive functions is based on a suitable equilibrium (isti‘dāl) or balance between the elements. And although the vegetative soul precedes the animal soul in the order of generation, it is the balanced material mixture that forms the substrate for the animal’s coming to be (Salvation, 198). The balance Avicenna wants to strike is quite precise. In the case of natural potentialities, a complete passive potentiality for the reception of some particular form will of necessity lead to the acquisition of that form once the actualizing agent is present and provided no external impediment interferes with the process. At the same time, we are not led to understand that the completion of the passive potencies would itself constitute the actualization of the form: rather, there is a qualitative leap (ikhtilāja), in that at one moment the form is absent while at another it is suddenly present. The leap occurs instantaneously, in the manner of Aristotelian substantial change, which brings with it all the attendant demarcation problems concerning what the last moment of potentiality is and what the first moment of actuality. But more to the point, the appearance of form is somehow supposed to be irreducible to the material processes whereby the receptivity to form is brought to the state of a completed passive potentiality, a “complete suitability that is nothing other than complete commensuration to a specific thing” (Healing: Metaphysics IX.5.5). Something more is needed in each and every case. This is where Avicenna’s famous Giver of Forms (i.e., the transcendent and immaterial agent intellect) enters the picture, and with it Avicenna’s larger project of arguing for a complex immaterial superstructure to lend corporeal reality its moorings. (See Janssens 2006.) This metaphysical side of the story should certainly not be belittled, and in the absence of evidence to the contrary, I suggest we take Avicenna at his word. An external agent is needed to impose each and every form upon that which does not previously possess it and this, as odd as it may seem to us, is a change whose origins lie beyond the confines of the material universe. (The reasons, very briefly, have to do with the ancient debate concerning ‘complete mixture’ on an elemental level, and whether such mixtures could account for the emergence of more complex entities. Avicenna contends that they cannot, and hence must insist like a good Aristotelian both that [a] form is determinative of the qualities and accidents subsumed under it and [b] substance is irreducible to its material characteristics: for comments see Stone 2008.). For all this, it is certainly remarkable how on the empirical or observational level, the system described by Avicenna would be indistinguishable from a

108

T. Kukkonen

naturalistic explanatory scheme whereby fully-fledged potentialities blossom into actualities in and of themselves once all the antecedent conditions are met (in Avicenna’s terms, upon reaching completion) through suitable movers or agents acting on the patient. What is more, Avicenna contends that all natural processes can be analyzed in like terms, meaning that metaphysical fitness through physical constitution provides the key to explaining all change, pointedly including generation and corruption, and at all levels of complexity. There is nothing whatsoever mysterious about any of this, or, as the matter is put in the section dedicated to Actions and Passions: The attraction of the magnet to the iron is no more wondrous than the attraction of plants to the direction in which they grow, or the senses to what they sense, or the motion to what is moved by volition… The account (qawl) in all these [cases] is one and the same, which is that the composite body through its temperament becomes suited to the reception either of its disposition or its form or the potency specific to it. Each flows into it from the Giver of forms and potentialities, none other. As regards [the Giver’s] influx, this is due to its generosity, and because of this, whatsoever is worthy and suited is not denied said [influx]. As for the specification of just this influence to the exclusion of others, this is due to [the recipient’s] complete fitness, which results from its disposition. All these actions are accomplished due to these things possessing the [appropriate] active potencies, which they have solely due to God’s potency and love. (Healing: Actions and Passions II.1, 256.6–14; for a different reading, Stone 2008, 117–119).

The picture Avicenna paints is one of hierarchically arranged capacities, each one becoming activated by a higher-ranking power and then becoming capable of actualizing ones further down the line. The fact that none of the mundane powers is supposed to be able to produce a form on its own (the gold standard for actuality on the Aristotelian picture) becomes very nearly incidental on the given account. It is, after all, explicitly stated that neither God nor the Giver of Forms—the two, lest we forget, are distinct transcendent principles—ever fails to give of the abundance of their being; furthermore, Avicenna says that the generosity of these divine principles is indiscriminate in the same manner that light falls on everything alike and is everywhere the same. So the only thing that leads to this particular thing taking place, as opposed to that, is a difference in the receptivities of different entities to the divine effluence: and if the coming about of these receptivities cashes out as the working of active potentialities on passive ones so as to produce different material constitutions (‘temperaments’), then the resulting picture is not that different from the standard Aristotelian account after all.13 13

See Healing: Metaphysics IV.2.20–23. One important difference remains between the activities of the Necessary Existent and the Giver of Forms (the Agent Intellect): the activity of the former is fully unitary—it is existence as such that flows from the First Principle on down—whereas in the latter case, we may envision the Giver of Forms as constantly radiating all the forms, a dense field of information, so to say, out of which different things find instantiation in suitably prepared matter. (In line with Neoplatonic metaphysics, the difference stems from the different rungs the two transcendent principles occupy on the ladder of reality, with unity gradually giving way to multiplicity and sameness to difference as one descends.) From the point of view of the recipient,

Potentiality in Classical Arabic Thought

109

Much the same point is made in the Physics of the Healing, where Avicenna takes the 6th-century Greek commentator John Philoponus to task in a long digressive note regarding the concept of ‘nature’ (Gr. physis, Ar. ṭabī‘a), which Aristotle in the original Physics had defined as an internal principle of motion and rest. As Avicenna reads the situation, Philoponus in his commentary on the Physics had taken it upon himself to expand upon Aristotle’s definition of nature for the reason that Aristotle’s definition had addressed only the observable results or ‘acts’ (af‘āl) of nature, not the very nature of nature itself. Probably unbeknownst to Avicenna, similar efforts had formed part of the standard Platonist programme since at least Plotinus: the goal, typically, was one of showing how the immanent explanations of the workings of reality favoured by Aristotle were incomplete in and of themselves, requiring Platonic appeals to the transcendent foundations of things for their completion and proper comprehension. In keeping with this tendency, Philoponus had reinscribed nature as a potency (dynamis/quwwa) permeating bodies, productive of immanent forms and temperaments.14) But Avicenna counters that “nothing more is meant by potency than a principle producing rest and motion. Also, potency is described only with respect to relative association” (alnisba l-iḍāfiyya: Healing: Physics I.5.7, McGinnis’s translation modified). So to make ‘nature’ into some mystified transcendent power is to misunderstand Aristotle’s intention: according to the Aristotelian outlook, natures are principles existing in bodies and moving them towards their perfections and providing for their rest in that state once they have reached it (mabda’ mawjūd fī l-ajsām litaḥrīki-hā ilā kamālāti-hā wa-taskīni-hā ‘alay-hā). A separate inquiry and another discipline (metaphysics rather than physics) is needed, Avicenna concedes, to determine whether nature is enough to produce a form or whether the operations of nature only serve to prepare a body for the reception of form from elsewhere (Healing: Physics I.5.8; see also I.10.3). This is yet another reference to Avicenna’s favoured theory of a Giver of Forms, which in some ways brings Avicenna closer to Platonic territory after all. But what interests us in the present context is what Avicenna says about potentiality (quwwa). Because potentiality is a term that always and invariably invokes correlates—an active and a passive entity, somehow commensurate with one another—it makes no sense to speak of potency in generic and indeterminate ways, as if some one thing could exercise power over all things alike. What Avicenna objects to is Philoponus’s insistence on reifying nature as if it were some sort of transcendent principle—again, a common Neoplatonic conceit (see Martijn 2010). Avicenna correctly discerns that for Aristotle, there is no nature over and above the natures of individual things, except perhaps in terms of a post rem conceptual abstraction (Healing: Physics I.7.2–3). Aristotelian nature is nothing other than a thing’s form,

however, the resulting picture is much the same: the deciding factor in what emerges into actuality is the preparatory process that results in a complete passive potentiality or ‘the actual subject’. 14 Macierowski and Hassing (1988); for further materials, Sorabji (2005, 33–44 and ff.

110

T. Kukkonen

seen from the point of view of the various ways in which it may interact causally with other things, in other words, its active and passive potencies. So the form of water, for instance, is a power that makes the water’s matter to subsist as a species – namely, water. The former [namely, the matter] is imperceptible, but the effects that proceed from it are perceptible – namely, perceptible coolness and weight… So the nature’s action in, for example, the substance of water is either relative to its passive influence and so is coolness; or is relative to its active influence, giving it its shape, and so is wetness; or is relative to its proximate place and so brings about motion; or is relative to its proper place and so brings about rest. (Healing: Physics I.6.2)

The specifics of what Avicenna describes here have to do with Aristotelian elemental physics and need not detain us. What is important is the principle that a thing’s nature is manifested in the specific set and range of active and passive potencies with which it is endowed, namely, that set which flows from its form. This also accounts for Avicenna’s insistence that natural potentialities are those that find actualization always or for the most part (Healing: Physics I.13.6). Following Aristotle’s lead on this point (Phys. II.7–8), Avicenna contends that the outcomes of natural potentialities can be “neither accidental nor compelled”, but instead must be “permanent and prevalent” (Healing: Metaphysics IV.2.20, IV.2.21): they need not manifest every time and everywhere, but the exceptions must stem from sufficiently rare circumstances as to be classified as extraordinary ‘impediments’. Everything that has a given potentiality typically passes into the corresponding actuality (Healing: Physics II.1.2), with a regularity that is dictated both by our epistemic requirements and by the underlying essentialist ontology to which Avicenna subscribes. Conversely, where passage into a given mode of actuality is impossible, there is no potentiality for it (Healing: Physics II.1.2). Nature simply will not have provided for a potentiality that has no chance of ever becoming actualized. Potentiality thus emerges as a principal tool for the Peripatetic philosopher to go about exploring the inherent possibilities in things and the differences between them, with the distinction between proximate and remote potentiality serving the additional purpose of distinguishing between a disposition (hay’a, the tendency to react to a certain kind of stimulus in a determinate manner) and a more broadly considered aptitude. Considering where we began with the Metaphysics of the Healing, it is certainly remarkable where the conversation has taken us. Avicenna, we will remember, started out from an account of potentiality that ostensibly stresses the freestanding agency of the active principle, in a manner akin to the kalām mode of framing divine power or qudra. Avicenna even intimates early on that potency would commonly be evoked in conjunction with those acts that are undertaken only infrequently and not without difficulty (Healing: Metaphysics IV.2.1). Now, however, we have arrived at a conception where the focus is squarely on those changes that happen regularly and as a rule, and on the passive capacities and propensities for things to undergo some specific kind of change, rather than on the power and/or discretion of the actualizing agent. This inversion, I submit, lies at the heart of Avicenna’s intentions when it comes to making the case for the usefulness of potentiality as a philosophical concept. Rather than evoke the notion of divine

Potentiality in Classical Arabic Thought

111

power haphazardly and in anomalous circumstances, Avicenna thinks it best to embed the concept of causal powers in the natural processes whereby representatives of natural kinds actualize their potential. And this, in a word, means putting the notion of passive potentiality—an aptitude, susceptibility, or fitness for a certain kind of change—front and centre. This emphasis on the natural and the regularly occurrent can also be demonstrated through the negative, by looking at what it is set against. Avicenna pledges allegiance to the standard Aristotelian distinction between rational and natural capacities. The difference, he says in the Metaphysics (IV.2.10–11), is that natural potentialities are moved to actuality immediately and automatically upon the coming together of the appropriate active and passive potentialities, whereas a rational capacity lies within the agent’s will to exercise or not. Related to this, every rational capacity is in Avicenna’s words “a power over a thing and its opposite”, whether contrary or contradictory—appetite may prefer its food hot, cold, or anything in between, depending on the circumstance; or a thing may be desired or shunned—whereas natural capacities are always geared towards single outcomes.15 This lends several degrees of added complexity to rational actions: and while their ultimate determinacy may not seriously be in question—indeed, Avicenna hints that the formation of an act of will is reached equally as much through a determinate causal chain—the least that can be said is that such actions cannot be explained solely by appealing to the nature of the agent involved. (To pick an example: human nature itself does not suffice to explain a given human being’s resolve to overeat, rather, one must investigate all the particular perceptions, desires, and calculations that go into an individual decision.) By contrast, natural capacities, precisely because of the limited number of variables involved, remain pre-eminently susceptible to scientific analysis.

3 Averroes: Potentiality, Nature, and Essence Moving from Avicenna to Ibn Rushd or the Latin Averroes, we are immediately faced with an author of a starkly different temperament. Far from cultivating an image as a philosophical iconoclast and trailblazer, the way Avicenna had done, Averroes over the course of a distinguished career came to view his role as guardian and protector to the authentic Peripatetic tradition, as codified in the works of Aristotle and his most authoritative commentators (for materials see Endress et al. 1999). Accordingly, his comments on potentiality also veer towards the conservative; they are essential reading nonetheless.

15

In the Physics of the Healing (I.5.3) there is a four-fold classification of potentialities divided along two axes: the relevant criteria are (A-B) whether volition (irāda) is involved and (1–2) whether the power results in a single outcome or motion or many and varied ones.

112

T. Kukkonen

Consistent with Averroes’ general outlook, his main presentation of potentiality comes in three works that address Aristotle’s Metaphysics directly: an Epitome produced early in Averroes’ career, a Paraphrase stemming from the middle period of his authorship, and a full Commentary that numbers among his final major works. Each of these writings purports to explicate the authentic Aristotelian teaching, although the grain of detail in Averroes’ engagement—and, consequently, the results he gets philosophically—varies considerably. In addition, Averroes’ polemical writings in defence of philosophy rely heavily on the Aristotelian framework of potentiality and actuality. Many of the structuring features in these works are common to Averroes and the larger Arabic philosophical tradition— Averroes subscribes to the distinction between remote and proximate potentiality, for instance; he explains the difference between active and passive potentiality exactly as Avicenna had done; and he accepts the equation between matter and potentiality—but his chosen polemical stance, that of taking Aristotle as the sole source of genuine philosophical teaching, affords him a vantage point from which to assess critically the claims of those he calls ‘the later philosophers’ (almu’akhkhirūn). Pointedly this includes Avicenna, whom Averroes criticizes frequently for innovating needlessly and for tampering and thereby damaging a philosophical edifice that was complete and flawless as it stood. A dramatic example of this comes from a field that has a direct bearing on the topic of potentiality. This is Averroes’ critique of the Avicennan theory of emanation. In his earliest works, including the first version of the Epitome of the Metaphysics, Averroes appears to have accepted the standard emanationist view of divine creation shaped by the major Muslim Peripatetics, al-Fārābī and Avicenna, complete with a Giver of Forms similar to what we encountered in Avicenna earlier in this piece. But by the time Averroes formulated his defence of philosophy in the face of al-Ghazālī’s famous critique—specifically, in writing the Incoherence of the Incoherence in 1180 as a point by point rebuttal of al-Ghazālī’s earlier Incoherence of the Philosophers—Averroes had jettisoned the whole theory of emanation as an un-Aristotelian accretion that left Peripatetic philosophers needlessly vulnerable to attacks from the Ash‘arite theologians (see Kogan 1984). The shift is significant, metaphysically speaking, as it removes from consideration an entire mechanism for explaining the causal relationship between the divine and the mundane, which must then be supplanted by something else. Important from our point of view is that Averroes in patching up his own version of Aristotelianism ends up underlining the importance of potentiality as the basis for explaining any worldly processes of becoming. Consider how Averroes in the late Commentary on Aristotle’s Metaphysics maps out the various explanations given to the phenomenon of generation, i.e., the coming-to-be of that which is not. In keeping with a rhetorical trope popular within the Peripatetic tradition, Aristotle is presented as providing the sensible mean between two unhelpful and extreme positions. One extreme according to Averroes is the doctrine of latency (kumūn), that is, the notion that nothing ever comes to be but that everything instead emerges out of a full, though concealed, pre-existence. The agent on this picture only moves to uncover that which was already there, so to

Potentiality in Classical Arabic Thought

113

speak, so that for instance hot particles are released from under the dominance of cold ones by removing enough of the latter from the proximity of the former. The mechanism of conjoining and separation provides the means by which certain elementary qualities on the microscopic level achieve dominance and become recognizable forces on the macroscopic—an explanatory device that allows Averroes to present Empedocles as having inspired the ancient atomists as well as discredited early kalām theologians of the likes of Ibrāhīm al-Naẓẓām (d. 845 CE).16 The opposite extreme is that of the Ash‘arite theologians and their monotheist coreligionists such as the Christian philosopher John Philoponus: on their view, as Averroes’ memorable formulation goes, “all possibility resides in the agent” (Commentary XII, comm. 18, 1498.4–6; similarly Epitome, 84; Incoherence I, 100–102), with no constraints on the agent’s agency being imposed by something being worked on and providing the locus for a thing’s coming to be. This latter view Averroes considers beneath contempt; we shall have occasion to ponder why later on. For now, let us focus on a still more remarkable set of refinements. The mediating view, put forward by post-Socratic philosophers, is split first into two, then into three distinct doctrines, with the authentic Aristotelian teaching pitted against the Muslim philosophers al-Fārābī and Avicenna. According to Averroes, each of the latter two had evoked the notion of a Giver of Forms in order to explain generation across a range of phenomena—al-Fārābī (following the 4th-century Peripatetic Themistius) in certain special cases, Avicenna generally and across the board. Both doctrines concerning divine agency are still wide of the mark, however, since they evoke un-Aristotelian notions of emanation and in general fail to stay within the parameters of the four established Aristotelian causes. The third doctrine is the one we have borrowed from Aristotle and it is that the agent produces only the compound from matter and form by moving matter and changing it to educe the potentiality it has for the form into actuality. This view is similar to the view of those who think that the agent merely assembles and organizes discrete things… the agent, in Aristotle, does not really unite two things, but makes them pass from potentiality into actuality, putting together, as it were, potentiality and actuality, I mean matter and form, by making potentiality become actuality, without suppressing the substratum which receives the potentiality. (Commentary XII, comm. 18, 1499.2–11; tr. Charles Genequand)

For good measure, Averroes repeats a second time that the correct view, which is Aristotle’s, most closely resembles the theory of latency (1499.14). So the proper philosophical theory does not lie in the middle, after all: Averroes subtly but unmistakably positions it left of centre, as it were, in the second slot out of five on a spectrum that extends from gross materialism to ad hoc creationism. To review the facts once more: according to Averroes, we have five positions all in all. (1) On one end of the spectrum, we have a doctrine of immanence, according to which nothing ever comes to be, instead, pre-existent material substances only

16

Commentary XII, comm. 18, 1497. It is in fact from theological discussions that Averroes adopts the vocabulary of latency: see the Encyclopedia of Islam, Second Edition, s.v. kumūn.

114

T. Kukkonen

emerge one from the other. (5) On the other, every substance is created de novo from nothing in its entirety, as the theologians and John Philoponus had maintained. In between, there are various ways of mediating between the two positions, which amounts to maintaining simultaneously that nothing comes from nothing and that coming-to-be is real. Nevertheless, with (3) al-Fārābī’s view that life-forms must needs be imposed from beyond the material cosmos being discredited and, a fortiori, (4) Avicenna’s notion that all forms must similarly have an extracosmic origin, all that is left is (2) the belief that moving around matter and material substances in suitable ways will allow for the emergence of pre-existent substantial forms from potentiality to actuality such as genuinely did not exist before (in actuality, that is, but only potentially). The Aristotelian theory on Averroes’ view ends up being closer to the immanentist pole than to the dualist one: and it is precisely the doctrine of potentiality that allows for this and that also accounts for the theory’s explanatory superiority over its alternatives. The details of the case Averroes wants to make merit closer scrutiny. Against the reductionist view that operates on the basis merely of aggregating basic stuffs, the properly philosophical theory acknowledges the phenomenon of becoming, i.e., it recognizes that genuinely new existents come to be and that there are emergent levels of complexity to reality that are irreducible to their constituent parts’ material makeup. At the same time, the correct philosophical theory does not indulge the sort of dualism that the Muslim Peripatetics had inherited from late antique Neoplatonism. In Averroes’ analysis, it was in fact the immodesty (ifrāṭ) of the emanationists, with their unfounded assumption of a creation (ikhtirā‘) in toto of forms as an imposition from the outside by a Giver of Forms, that had cleared the conceptual space for the creationist theologians to put forward the absurd notion that things could be created out of nothing, with nothing but the arbitrary will of the Creator to stand as an explanation.17 The true Aristotelian philosopher, by contrast, understands that in the process of generation, the agent merely “brings out the form of its equivalent from potentiality into actuality: it is not an agent in [the sense] that it would introduce into the matter another thing from the outside or something external to it” (Commentary XII, comm. 18, 1499.18–1500.2). Because any concession to Platonic metaphysical dualism (of forms divorced from matter) cedes unnecessary ground to the creationist theologians, with their improvisational appeals to God’s absolute power, the only safe alternative is to march a retreat into an airtight interlocking network of active and passive potentialities, the way Averroes sees the Aristotelian universe as operating. Both active and passive potentialities, furthermore, need to be hemmed in by the natures of things interacting upon one another; potentialities on Averroes’ view are foundational to an Aristotelian ontology of natural kinds. When Averroes avers that the different kinds of potentiality and actuality mirror the ten categories of being, then, 17

Commentary XII, comm. 18, 1503.9–1504.2. This is a remarkably astute reading of how John Philoponus arrived at his creationist view (see de Haas 1997), although it is doubtful that Averroes would have had access to the relevant texts that show this; it is also not a bad approximation of the attraction of the Avicennan view to later kalām theologians.

Potentiality in Classical Arabic Thought

115

this is more than an incidental observation (see Epitome, 26; Commentary IX, comm. 21, 1220). In Averroes’ estimation, the Aristotelian innovation of potentiality, understood suitably technically, affords the only workable explanation of change on a metaphysical level. Across the nine categories of accidents, this happens through assuming a stable substrate that allows for the coming and going of contrary properties, while in the category of substance, the notion of an existent coming into being from that which is not would be unintelligible without philosophers evoking a ‘real possibility’ that would be explanatory of how humans are only ever born of human sperm and tables only made of wood or metal, not vapours or silly putty. In each case, all the makings of a fresh actual substance are found in the substrate: all one needs to introduce is a moving cause of a suitable nature, one that allows the real potentiality to convert into an actuality. Actuality (Ar. fi‘l/Gr. energeia) in turn is understood primarily in terms of activity, as those actions natural and specific to a specific type of being. The principal challenge to this picture in Averroes’ view comes from those who deny antecedent (passive) potentiality as an identifiable ground for change in a thing. In Aristotle’s time, such a challenge had come from the Megarians, while in Averroes’ own age the Megarian challenge had been picked up by the Ash‘arite theologians, who had denied that things would have inherent capacities for exhibiting particular properties before they actually manifest them. The centrality of the Megarian/Ash‘arite challenge to Averroes’ conception of Aristotelian science is shown in an otherwise throwaway passage in the commentary on Aristotle’s Metaphysics II.3. Here, Averroes criticizes the many youths who had started building their worldview from premises provided by kalām theology and who had moreover held these to be self-evident. At the masthead of the follies of the mutakallimūn is “their denial of the existence of natures and potentialities” (alṭabā’i‘ wa-l-qawā: Commentary II, comm. 14, 44.6). Now, the theologians had discussed both natures (quite extensively) and powers (less frequently) in their works, it is true, and with the exception of a few they had indeed denied both to created existents. But one can hardly say that this would have laid at the heart of their theoretical ruminations. It is the foundational role that the concepts of nature and potentiality play in Aristotelian scientific explanation that merits highlighting them in the context of an Aristotelian commentary.18 Accordingly, after discussing the views of the Ghāriqūn, which is to say the Megarians, in light of Aristotle’s report in Metaphysics Theta, Averroes goes on to say that theirs “is a theory accepted by today’s Ash‘arites from among our coreligionists: it is a theory opposed to the nature of the human being, [both] in terms of

18 For natures in classical kalām see Frank (1974). Adamson (2010, 368–369) in an illuminating article has emphasized the similarities between this passage and Averroes’ Decisive Treatise, a work that along with the Disclosure of the Methods of Proving the Tenets of Faith (Kitāb al-kashf) and the Incoherence of the Incoherence takes aim at the Muslim theologians and their reasoning. However, the Decisive Treatise does not raise the issue of powers and natures at all, presumably because the questions it addresses are of a theological provenance; whereas in the parallel passage in the Metaphysics commentary the terminology is front and centre.

116

T. Kukkonen

his beliefs and his actions” (Commentary IX, comm. 5, 1126.12–14). A little further on, Averroes elaborates on the destructive effects of the Ash‘arite doctrine: according to him, “those who deny potentiality deny [the existence of] prime matter, all motions, generation and corruption, and all those arts that act by way of deliberation and choice as well as those that act for the sake of repelling the harmful and acquiring the useful” (Commentary IX, comm. 6, 1132.9–11). In other words, occasionalism and the denial of potentialities results in utter nihilism, since it leads to the decimation of all three branches of Aristotelian knowledge—theoretical, practical, and productive (see further Epitome, 85). Why should this be? Kogan (1985, 100–135) has persuasively shown that the reason why Averroes adopts a hard line when it comes to the issue of potentiality is because at the heart of his mature ontology lie what Kogan calls ‘powerful particulars’. Substances are made what they are (and in a secondary sense can be identified for what they are) by the potentialities they exhibit, that is, by their propensities for acting and being acted upon. This makes potentiality a non-negotiable component in describing every being qua being—which, of course, is why potentiality and actuality assume a central place in Aristotle’s Metaphysics. In this connection, it is worth noting one highly interesting consequence that Averroes draws from Aristotle’s doctrine concerning the priority of actuality over potentiality as well as from his own notion of powerful particulars. This is that remote potentialities and their actualizations can only ever be identified in relation to the ultimate level of actuality, that is, with at least one eye on actual and fully specified real existents—Aristotelian substances as hylomorphic compounds. Since in actuality there are no living beings simpliciter (the living-being-as-such is merely an abstraction: see Aristotle, De an. I.1, 402b7–9) but only beech trees, horses, and humans, the potential for rationality—to pick an example—does not stand independently, nor can it be understood in isolation from the reality it is used to describe: rationality can only be identified as a defining feature of what it is to be one particular kind of animal, viz. the rational animal ‘human’. Likewise the capacity for nutrition only makes sense against the backdrop of what it is actually to live, which in turn requires an appeal to some specific and fully determined kind of living being; etc. The upshot is that all levels of reality identified by the philosopher as brimming with potentiality, from prime matter all the way up to those proximate potentialities which stand in readiness to bring forth actual fully specified entities, must be understood as deriving from the primary reality of a fully actual and active form, one that yokes all the potentialities found at different levels to its own purposes. (Epitome, 87–88; see also Di Giovanni 2004.) On the level of epistemology and philosophy of science, meanwhile, the concept of potentiality cannot be relinquished, lest one simultaneously give up the prospect of classifying entities as belonging to natural kinds—and hence, as having determinate modes of being in the first place: for without stable essences that manifest in recurring patterns of acting and being-acted-upon, “all things would be one, indeed not even one” (Incoherence XVII, 520.14–15). In explicating the meaning of Aristotle’s words that “no thing exists which would not possess its own potency”

Potentiality in Classical Arabic Thought

117

(lā yakūna shay’an lā yaqwā ‘alay-hi), Averroes thus draws a stark contrast with the kalām theologians for whom all power resides with God. The people of our time, however, posit for all the actions of [all] existents a single agent, without anything to mediate between them, this [agent for them] being God. From this it follows that there are no proper acts natural to any beings at all: and if there are no actions proper to beings, then they have no proper essences either, since actions differ only on account of essences being different. And with the abolition of essences, names and definitions are similarly abolished, so that being becomes a single thing. This view is one that is very remote from human nature… (Commentary IX, comm. 7, 1135.15–1136.5)

The reason why denying potentialities is repugnant to human nature is that as rational animals, all humans by nature desire to know (Aristotle, Met. 1.1, 980a21): and since genuine knowledge in the sense of epistēmē is always a knowledge of causes (Aristotle, An. post. 1.2, Met. 1.2), intellection is nothing more—but also nothing less—than the perception of existents in light of their causes (Incoherence XVII, 522.7). To deny causes, therefore, is to deny the very possibility of knowledge and to make it null and void (522.10–11). On the view of the creationist theologians, the way Averroes sees it, anything can follow upon anything, given that everything ultimately issues from nothing or from the unrestrained will of God: but this destroys every prospect of ever knowing anything in a fixed manner (‘ilm thābit: 530.18–531.8). The reference to practical philosophy—that is, ethics and politics—is likewise readily understood. If human beings do not possess the power to act (or, as some Ash‘arites would claim, if they only acquire the power to act in the exact moment in which they act) then one cannot truly regard them as agents, and all sense of deliberation and of personal responsibility fades away. But Averroes’ passing reference to the productive arts is interesting in another way, for it touches on an aspect of potentiality that we have not yet addressed in full. The identification of passive potentialities in, or for, artefacts, differs from the process of identifying natural potentialities in that the latter have their principle of actualization in themselves: potential artefacts, meanwhile, do not, instead, they are framed in terms of the external agent’s intent with regard to them (see Aristotle, Met. IX.7). To pick an example adduced earlier by Avicenna, the locksmith will size up a block of wood with an eye towards its suitability for chiselling keys; a builder with an eye on its usefulness in construction; while to the ordinary person it may be merely a doorstop or a simple source of heat. And in each of these cases, moreover, the wood may retain its own nature while being yoked to another purpose, contrary to what we have found in the cases of natural substances (where the substrate is always subsumed: see Epitome, 69). In such a case, how much credibility does any talk of real potentialities have anymore? Are these not as arbitrary and ad hoc as the theologians’ surmises regarding the will of God? Averroes does not answer this question directly, but a credible answer can be teased from his and Avicenna’s remarks. In a real sense, potentialities for artefact production are indeed in the eye of the beholder: they would not exist at all if not for the purposeful gaze of the artisan, given that potentiality and actuality are

118

T. Kukkonen

always correlative notions. An example derived from Avicenna would be that of the carpenter who uses wood as material (Healing: Physics I.12.8): neither the human being nor the piece of wood need necessarily be conceived of in terms of the causal relationship of carpentry that ties them together as agent and patient, respectively, since neither of their essences as natural substances have anything to do with the act of woodworking. Notwithstanding their relational standing, however, such artisanal passive potentialities—the aptitude of stuffs for moulding in some intentional manner—clearly ride on some natural properties specific to the types of materials being employed for human work: malleability, rigidity, energy density, and the like. Thus the young willow’s potential for becoming a woven basket is just as accidental to the kind of being it is (considered in itself) as is scammony’s capacity for purging excess bile from the human body: yet the willow’s very bendiness, which is what makes it so valuable to the basket-weaver, is also what allows it to survive and thrive in places where blustery winds blow and so does contribute to its own specific mode of being. The same goes for various pharmacological agents and the like: these have not been inserted into natural substances in order to benefit humankind, but neither do they occur at random or without reason. Rather, if they occur naturally, which is to say always or for the most part, then they form part somehow of that substance’s natural life and of its complement of passive and active potencies (though it is of course possible that their occurrence in a particular natural substance is a secondary, accidental side effect rather than a primary characteristic of the species under examination). If this is true, then the development of the arts, too, depends on there being true natural kinds whose nature can be analyzed in terms of layers of passive and active potentialities, to be identified and accessed by the human agent according to need and circumstance.

As steeped as Avicenna’s and Averroes’ ruminations are within the minutiae of the school tradition, and as much as they can be seen to be reacting to particular pressures emanating from the direction of Muslim theology, both philosophers manage to demonstrate quite conclusively how Aristotelian potentialities are best understood in the context of the overall Peripatetic project of explaining nature. Avicenna’s and Averroes’ works effectively connect the dots between Aristotle’s various presentations of potentiality and actuality within the school corpus; they lend Aristotle’s remarks added systematicity and coherence; usefully, they also supply a large storehouse of affecting examples that throw light on how potentiality and actuality are supposed to function at multiple levels of reality and what their explanatory force is. (For the latter, the best I can do is direct the reader to the texts themselves.) What emerges with especial clarity is how potentiality is an important concept for the way it bridges ontology, natural philosophy, and philosophy of science within the Aristotelian philosophical tradition. Potentiality in Avicenna’s and Averroes’ estimation is the principal tool for retaining a certain necessary dynamism within Aristotelian essentialism: potentiality, combined with an essentialist ontology of actuality and activity, is what allows for a philosophical outlook wherein the true natures of things can be mapped out and determined through a

Potentiality in Classical Arabic Thought

119

process of observing and recording what things do—how substances interact with other substances in complex interlocking ways.19

Bibliography Adamson, P. (2002). The Arabic Plotinus. London: Duckworth. Adamson, P. (2005). On knowledge of particulars. Proceedings of the Aristotelian Society, 105, 257–278. Adamson, P. (2010). Yaḥyá Ibn ‘Adī and Averroes on Metaphysics Alpha Elatton. Documenti e studi sulla tradizione filosofica medievale, 21(2010), 343–374. Akl, Z. B. (2016). Des Méqariques aux Ashʿarites le commentaire d’Averroès à Métaph. Theta 3. Rursus, 9, 2–14. Ash‘arī, Abū l-Ḥasan Ismā al. (1963). Maqālāt al-islāmiyyīn. In H. Ritter (Ed.), Wiesbaden: Franz Steiner Verlag GmbH. [Opinions]. Bāqillānī, Muḥammad Ibn al-Ṭayyib al. (1987). Kitāb al-tamhīd al-awā’il wa-talkhīṣ al-dalā’il, Ed. ‘Imād al-Dīn Aḥmad Ḥaydar, Beirut: Mu’assasat al-Kutub al-Thaqāfiyya. [Introduction]. de Haas, F. (1997). John Philoponus’ new definition of prime matter. Leiden: E. J. Brill. Di Giovanni, M. (2004). Averroes on the doctrine of genus as matter. Documenti e studi sulla tradizione filosofica medievale, 15, 255–285. Endress, G., et al. (Eds.). (1999). Averroes and the Aristotelian tradition. Leiden: E. J. Brill. Fārābī, Abū Naṣr al. (1961). Falsafat Arisṭūṭālīs. In M. Mahdi (Ed.), Beirut: Dār Majallat Shi‘r. [Philosophy of Aristotle]. Fārābī, Abū Naṣr al. (1985). On the Perfect State (Mabādi’ ārā’ ahl al-madīnat al-fāḍilah) (R. Walzer, Ed. and Trans.). Oxford: Oxford University Press. Frank, R. M. (1966). The structure of created causality according to al-Aš‘arī. Studia Islamica, 25, 13–75. Frank, R. M. (1974). Notes and remarks on the ṭabā’i‘ in the teaching of al-Māturīdī. In P. Salmon (Ed.), Mélanges d’Islamologie (pp. 137–149). E. J. Brill: Leiden. George, L. (2005). Listening to the voice of fire: Theurgical fitness and Esoteric sensitivity. In: J. Finamore & R. Berchman (Eds.), History of Platonism. Plato Redivivus (pp. 287–303). New Orleans: University Press of the South. Ibn Bājja, Abū Bakr Ibn Ṣā’igh. (1995). Kitāb al-kawn wa-l-fasād/Libro de la generacion y corrupcion (J. P. Montada, Ed. and Trans.). Madrid: Consejo Superior de Investigaciones Científicas. [On Coming-to-be and Passing-away]. Ibn Rushd. (1562̇). Aristotelis in Physicorum auditu libri octo, cum Averrois Cordubensis variis in eosdem commentariis, in Aristotelis opera cum Averrois commentariis (Vol. 4). repr. Frankfurt a.M.: Minerva Gmbh. [In Phys.]. Ibn Rushd, Abū l-Walīd. (1930). Tahāfut al-tahāfut, In M. Bouyges (Ed.), Beirut: Imprimerie Catholique. [Incoherence]. Ibn Rushd, Abū l-Walīd. (1938–1952). Tafsīr mā ba ‘d al-ṭabī‘a, M. Bouyges (Ed.), Beirut: Imprimerie Catholique (Vols. 4). [Commentary].

19

Since the completion of this piece in 2013 a number of important studies have appeared that should be consulted on various issues touched upon: on the role of potentiality in Ibn Bājja's thought we now have Wirmer (2014); on nature in Philoponus and Avicenna, Lammer (2015); on Avicenna and fittedness, Üçer (2015); and on Averroes and the Megarians, Akl (2016). Thanks are due to Jon McGinnis and Peter Adamson for their perceptive comments and to Kristina Engelhard for impressing on me the need for maximal clarity. Any failings remain my own.

120

T. Kukkonen

Ibn Rushd, Abū l-Walīd. (1958). Talkhīṣ [sic] mā ba‘d al-ṭabī‘a. In ‘Uthmān Amīn (Ed.), Cairo: Muṣṭafā l-Bābī l-Ḥalabī. [Epitome]. Ibn Rushd, Abū l-Walīd (1995), Talkhīṣ al-kawn wa-l-fasād. In Jamāl al-Dīn al-‘Alawī (Ed.), Beirut: Dār al-Gharb al-Islāmī. [Paraphrase]. Ibn Sīnā, Abū ‘Alī. (1892). Al-ishārāt wa-l-tanbīhāt. In J. Forget (Ed.), Leiden: E. J. Brill. [Pointers and Reminders]. Ibn Sīnā, Abū ‘Alī. (1938). Al-Najāt, Cairo: Muḥyī l-Dīn al-Ṣabrī l-Kurdī. [Salvation]. Ibn Sīnā, Abū ‘Alī. (1952). Al-Shifā’: al-Madkhal, In G. Anawati et al. (Eds.), Cairo: al-Maṭba‘a l-amīriyya. [Healing: Introduction]. Ibn Sīnā, Abū ‘Alī. (1952). Al-Shifā’: al-Ṭabī‘iyyāt: al-Samā’ wa-l-‘ālam, al-kawn wa-l-fasād, al-af‘āl wa-l-infi‘ālāt, In Maḥmūd Muḥammad Qāsim (Ed.), Cairo: al-Maṭba‘a l-amīriyya, 201–267. [Healing: Actions and Passions]. Ibn Sīnā, Abū ‘Alī. (1970). Al-Shifā’: al-Ṭabī‘iyyāt: Kitāb al-Ḥayawān, In ‘A. Muntaṣir et al. (Eds.), Cairo: al-Maṭba‘a l-amīriyya. [Healing: Animals]. Ibn Sīnā, Abū ‘Alī. (2005). The Metaphysics of the Healing, (M. E. Marmura Ed. and Trans.) Provo: Brigham Young University Press. [Healing: Metaphysics]. Ibn Sīnā, Abū ‘Alī. (2009). The Physics of the Healing (Vols. 2) (J. McGinnis, Ed. and Trans.). Provo: Brigham Young University Press. [Healing: Physics.] Janssens, J. (2006). The Notions of Wāḥib al-Ṣuwar (Giver of Forms) and Wāḥib al-‘Aql (Bestower of Intelligence) in Ibn Sīnā”, in: Intellect et Imagination dans la philosophie médiévale, In M. C. Pacheco & J. F. Meirinhos (Eds.), Turnhout: Brepols, 3 vols., 1:531–562. Juwayni, ‘Abd al-Malik Abū l-Ma‘ālī l. (1950). Kitāb al-irshād, In M. Y. Mūsā & ‘A. ‘Abd al-Mun‘im ‘Abd al-Ḥamīd Kabīr (Eds.), Cairo: Maktaba al-Khānjī. [Guidance]. Juwayni, ‘Abd al-Malik Abū l-Ma‘ālī l. (1959). Shāmil fī uṣul al-dīn, In Helmut Klopfer (Ed.), Cairo: Dār al-‘arab. [Comprehensive Book]. Kalabādhī, Abū Bakr Muḥammad Ibn Ibrāhīm al. (1935). The Doctrines of the Sufis (A. J. Arberry, Trans.). Cambridge: Cambridge University Press. Kindī, Abū Yūsuf Ya‘qūb al. (1950–1953). Rasā’il al-Kindī l-falsafiyya, In Muḥammad ‘Abd al-Hādī Abū Rīda (Ed.), Miṣr: Maṭba‘a l-ittimād. [Opera]. Kindī, Abū Yūsuf Ya‘qūb al. (1998). Oeuvres philosophiques et scientifiques d’al-Kindī. Volume 2. Métaphysique et cosmologie. In Roshdi Rashed & Jean Jolivet (Ed.), Leiden: E. J. Brill. [Oeuvres]. Kogan, B. S. (1984). Averroes and the theory of emanation. Mediaeval Studies, 43, 384–404. Kogan, B. S. (1985). Averroes and the metaphysics of Causation. Albany: The State University of New York Press. Kosman, A. (2013). The activity of being. An essay on Aristotle’s ontology. Cambridge, MA: Harvard University Press. Lammer, A. (2015). Defining nature: from Aristotle to Philoponus to Avicenna. In A. Alwishah, & J. Hayes (Eds.), Aristotle and the Arabic Tradition. Cambridge: Cambridge University Press, 121–142. Macierowski, E. M., & Hassing, R. F. (1988). John Philoponus on Aristotle’s definition of nature: A translation from the Greek with introduction and notes. Ancient Philosophy, 8, 73–100. Marmura, M. E. (1991–1992). “Avicenna and the Kalām”, Zeitschrift für Geschichte der Arabisch-Islamichen Wissenschaften, 7, 172–206. Martijn, M. (2010). Proclus on nature. Philosophy of nature and its methods in Proclus’ commentary on Plato’s “Timaeus”. Leiden: E. J. Brill. Netton, I. R. (1989). Allah transcendent. Studies in the structure and semiotics of Islamic philosophy, theology and cosmology. London: Routledge. Sambursky, S. (1962). The physical world of late antiquity. Princeton: Princeton University Press. Sorabji, R. (Ed.). (2005). The philosophy of the commentators 200–600 AD. A Sourcebook. Volume two: Physics. Ithaca: Cornell University Press. Stone, A. (2008). Avicenna’s theory of primary mixture. Arabic sciences and philosophy, 18, 98–119.

Potentiality in Classical Arabic Thought

121

Üçer, I. H. (2015). Aristotle’s dunamis transformed: On Avicenna’s conception of natural Istiʿdād and Tahayyuʾ. Nazariyat Journal of the History of Islamic Philosophy and Sciences 2/3: 37–76. Wirmer, D. (2014). Vom Denken der Natur zur Natur des Denkens. Berlin: De Gruyter. Witt, C. (2005). Ways of being. Potentiality and actuality in Aristotle’s metaphysics. Ithaca: Cornell University Press.

Author Biography Taneli Kukkonen is Professor of Philosophy at New York University, Abu Dhabi. He is the author of “Ibn Tufayl” (Oneworld 2014) and over thirty research articles on the Aristotelian and Platonic traditions in antiquity and the middle ages, with a particular emphasis on classical Arabic philosophy: for recent contributions, see “Al-Ghazālī on Error” in: G. Tamer (ed.), Islam and Rationality. The Impact of al-Ghazālī, vol. 2, Brill 2016 and “On Aristotle’s World” (in: Oxford Studies in Ancient Philosophy 2014).

Potentialities in the Late Middle Ages—The Latin Tradition Stephan Schmid

The notion of potentiality is central in Aristotle’s philosophy. The assumption that things or substances exhibit certain potentialities figures prominently in his metaphysics, epistemology, his natural philosophy or physics, and in his ethics. In metaphysical respects potentialities are important in order to account for our talk of possibilities and to characterize natural kinds. The perhaps most famous example of this is the Aristotelian definition of a human as an animal that is capable of thinking. In Aristotelian epistemology cognition is explained by cognitive capacities. Natural philosophy relies on potentialities because Aristotle argues (against the Megarians) that one can only account for (natural) change if one concedes that substances are endowed with causal potentialities to elicit and undergo certain changes. Furthermore, potentialities are decisive for his ethics since Aristotle explains the moral goodness of actions with regard to the virtues of their performers—and virtues are nothing but acquired potentialities or habits of acting in a good way. It is not surprising therefore that the notion of potentiality played an equally prominent role in the thinking of medieval authors in the Latin tradition who were strongly influenced by Aristotle. The fascination of medieval philosophers with Aristotle was never greater than in the 13th century when Aristotle’s Physics and Metaphysics as well as the Arabic comments and developments of these texts— most famously by Avicenna and Averroes—had become available in the Latin west. The availability of these texts led to a veritable boom of Aristotelianism in these times which is most evident in the ambitious attempts of Albert the Great and Thomas Aquinas to reconcile the Christian doctrines with Aristotle’s philosophy.1

1

For an overview of the medieval reception of Aristotle see Dod (1982).

S. Schmid (&) University of Hamburg, Hamburg, Germany e-mail: [email protected] © Springer Science+Business Media B.V., part of Springer Nature 2018 K. Engelhard and M. Quante (eds.), Handbook of Potentiality, https://doi.org/10.1007/978-94-024-1287-1_6

123

124

S. Schmid

At this historically late point in time I will start my investigation of potentialities in western medieval philosophy. The late Middle Ages are of particular interest for the history of potentialities since it is mainly via late scholastic authors that the notion of potentiality finds its way into the mechanistic philosophy of the early modern period where it was frequently subject to harsh criticism and rejected altogether (as by Descartes and Malebranche) or fundamentally reconceptualised on the model of conceptual containment (as by Spinoza and Leibniz). An important reason for the early modern discontent with the scholastic notion of potentiality was that scholastic potentialities were conceived as being explanatorily futile and epistemologically inaccessible, obscure entities. As I want to argue in this chapter, these complaints were not completely unfounded. The late medieval project of reconciling Aristotelianism with Christianity often involved a reinterpretation of central Aristotelian notions by which they changed their theoretical significance and their explanatory role. At least, this is, I submit, what happened in the 13th and 14th century to the Aristotelian notion of potentiality with respect to its role in explaining natural teleology and modalities. Instead of explaining these phenomena by appeal to the natural potentialities of things, authors of this period increasingly accounted for natural teleology and modality in terms of the rational potentialities of God, his will and his intellect. It goes without saying that I cannot give a comprehensive account of the complex debates about potentialities in the late medieval times here. Rather, I want to focus on two theoretical contexts in which potentialities played an important role using the example of three select authors. These authors are Thomas Aquinas (1225–1274), John Duns Scotus (1266–1308) and William of Ockham (1288–1348), who influenced philosophers even far beyond the Middle Ages. There were of course many theoretical contexts, in which late medieval authors relied on potentialities. As in Aristotle’s works potentialities played a crucial role in medieval epistemology and theories of cognition, which made crucial use of the assumption of cognitive faculties,2 as well as in medieval ethics, which evolved around a discussion of moral virtues.3 Still, I will restrict my focus to the medieval treatment of potentialities in metaphysics and natural philosophy, which are also subject of the first two chapters of this volume, and thereby confine myself to the following two questions: 1. The metaphysical question of contingency: In virtue of what is it possible for some actual things not to exist? 2. The physical or natural philosophical question of natural teleology: In virtue of what are natural processes directed towards a certain end or purpose?

2

For a detailed exposition of Aquinas’s influential views on this matter see Pasnau (2002, 143– 329). 3 An overview and discussion of the various forms of medieval virtue ethics is to be found in Bejczy (2008).

Potentialities in the Late Middle …

125

As I will try to show in the next two sections in turn, Aquinas answers both questions by appeal to natural potentialities while Scotus and Ockham try to avoid such reference. Rather they refer to the rational potentialities of will and intellect in order to account for contingency and teleology. In this respect they deprive the notion of natural potentiality of some of its explanatory functions. As I will submit by way of conclusion, it is not at least due to this explanatory deprivation of Aristotelian potentialities that early modern authors were encouraged to devise their theories without reference to such potentialities and to ban them as explanatorily idle and epistemologically inaccessible, obscure entities.

1 Potentialities in Metaphysics: The Problem of Contingency For Aristotle potentialities are actual properties of things. They are actual ways of how things are. They explain what kind of changes these things can elicit and which changes they can undergo. In the first case scholastic authors talked about an active potentiality (potentia activa), in the second case about a passive potentiality (potentia passiva). Fire, for instance, has the active potentiality to burn dry wood, while wood has the passive potentiality to be burnt. Similarly, a pianist can play the piano and hence actually possesses the active potentiality to play the piano, whereas the piano possesses the passive potentiality of being played. But although potentialities are actual properties possessed by things they involve a modal element insofar as that they determine what its possessor can do or which changes it can undergo given appropriate circumstances. For this reason potentialities are closely related with possibilities. But in what way exactly? In thinking about this question Thomas Aquinas was strongly influenced by Aristotle. Aristotle gave a dense exposition of his view on possibilities in Metaphysics V.12 where he distinguished various senses of “possible” (dynaton) that Aquinas comments carefully. For our concerns a passage about impossibility is particularly relevant. As Aquinas explains “impossible” is used in two ways according to Aristotle. In “one sense some things are said to be impossible because they have the foregoing incapacity which is opposed to potentiality.” In another sense things are said to be impossible not because of the privation of some potentiality but because of the opposition existing between the terms in propositions. […] Hence the term impossible means that of which the contrary is necessarily true. (InMet. V, l. 14.17f.)

Thus, for Aristotle as well as for Aquinas there are strictly speaking two notions of (im-) possibility. According to the first notion of possibility something is said to be possible with regard to a potentiality or capacity of a thing. Insofar as this kind of possibility is grounded in the capacity of a thing (res), I will follow John Duns Scotus in labelling it real possibility (potentia realis). An example for such a real

126

S. Schmid

possibility is that some people can stem 300 lb. This is true because some humans have the power or capacity to do so. According to a second notion of possibility, which again in accordance to Scotus I call logical possibility, we call something possible because it is not necessarily false or contradictory.4 An example for a logical possibility is the possibility that there are no frogs. This is true because it is not inconsistent to assume (and hence not necessarily false) that frogs do not exist. These two kinds of possibilities are closely related with two ways we apply the word “possible”. Aquinas is quite clear about this difference and explains it as follows: Since potency is referred to being, just as “being” is predicated not only of things that exist in reality but also of the composition of a proposition inasmuch as it contains truth and falsity; so likewise “possible” and “impossible” are predicated not only of real potency and incapacity but also of the truth an falsity of the compositions or divisions in propositions. (InMet. V, l. 14.18)

Since “potencies” designate ways of being—that is ways of how things are—they are subject to the same ambiguities as “being” which can, according to Aquinas, be referred to things as well as to propositions. So, predicating “being” of a thing allows us to say that it exists—as when we say that there is a God. Predicating “being” of a proposition on the other hand, allows us to say that it is true—as when we say that it is the case that snow is white. Now, the same holds for the word “possible”, which can also be applied to things (depending on the capacities of these things) and to propositions (with respect to the compositions of its terms). Employing contemporary terminology one might say that Aquinas points to the fact that “possible” is sometimes used as a predicate modifier (when it is used to express a capacity of a thing) and sometimes as a sentential operator (when it is related to propositions). In the first case, the word “possible” is used in order to attribute a capacity or disposition to a certain thing.5 If we say in this sense for example that it is possible for some humans to stem 300 lb, we say of certain humans that they are potential 300 lb stemmers and thus modify the predicate ‘to stem 300 lb’ that we then attribute to some people. Now, real possibilities rely on such predicate-modifying attributions of “possible”. They obtain if certain things are possible for certain things in virtue of their having certain capacities. This is different with regard to logical possibilities. Statements of such possibilities employ

4

For Scotus’s distinction between real and logical possibilities see his Ordinatio I, d.2/2 q.1-4, §262 (Vat 2, 282f.); an exposition is given by Söder (1999, 94–97). Note that Scotus’s logical possibilities are more akin to what we nowadays call “metaphysical possibilities” (than what we call logical possibilities) since they are constituted by compatibility relations between terms, which are more restrictive than pure logical relations of consistency. So, unlike most of us nowadays, Scotus would take it to be logically impossible that a whole is smaller than its parts. 5 For a discussion of these two senses of “possible” and its sources in Aristotle see Jacobi (2001) and Jansen (2001).

Potentialities in the Late Middle …

127

the notion of possibility in the second sense mentioned above and are true in virtue of the inconsistency or “non-repugnance” of their terms.6 Even though it is important not to conflate these two uses of “possible” (as attributed to things on the one hand and to propositions on the other), they are closely related with one another. Attributions of potentialities immediately entitle us to corresponding statements about real possibilities, which can also be conceived as logical possibilities. That is, whenever we may say that a has the potential to be F so that it is a real possibility that a is F, the terms “a” and “F” cannot be incompatible. Consequently, it is also logically possible that a is F. On account of this, at least some logical possibilities can be conceived as being grounded in capacities of things. This immediately opens up an interesting systematic option for the answering of the metaphysical question as to what possibilities are. At least with respect to some possibilities we can say that they are facts or truths about actual things that have a certain capacity or potentiality. This potentiality-based account of possibilities is attractive for it allows us to tie possibilities back to the world so to speak as that they need not to be regarded as ontologically free-floating entities. What is possible simply depends on the potentialities of actual things, and ultimately, facts about what is possible for certain things determine which possibilities there are.7 The prospect of such a neat metaphysics of possibility is of course appealing so that one is tempted to expand this account for all possibilities.8 But is it feasible? Can we account for all possibilities in terms of potentialities? Aquinas clearly answers this question to the negative.9 His reason is that assuming that possibilities are restricted by the potentialities of things would yield an unduly restriction of God’s omnipotence. By appealing to the aforementioned two senses of “possible” Aquinas argues that God’s omnipotence is only restricted by logical possibilities and not by real possibilities: Now according to the Philosopher in Metaphysics V a thing is said to be possible in two ways. First in relation to some power, thus whatever is subject to human power is said to be possible to man. Secondly absolutely, on account of the relation in which the very terms

6

Note that the (ontologically demanding) distinction between real and logical possibilities is not to equivalent to the nowadays well-known (but merely syntactic) distinction between de re and de dicto modalities. Although real possibilities are de re modalities (i.e. possibilities for certain things), they are more than this: they are de re possibilities that are grounded in a capacity of the thing they are predicated of. Thus, de re possibilities are not simply real possibilities and there can also be logical de re possibilities like ‘It is possible for me to live in a world without frogs’. 7 Note, however, that in the absence of an analysis of potentialities in non-modal terms, such a potentiality-based account of modality will not be reductive. Nonetheless, such an account might be attractive since our practice of attributing potentialities to things is quite wide-spread and well established. 8 An attempt to systematically flesh out such a potentiality-based account of possibility or modality is undertaken by Barbara Vetter in this volume. 9 Knuuttila (1993, 131–134) has argued that Aquinas generally spelled out the notion of possibility in terms of potency. I agree with him only partially. As I will argue, potentialities for Aquinas were not important in order to account for modalities in general, but only in order to account for what is naturally possible and contingent.

128

S. Schmid

stand to each other. Now God cannot be said to be omnipotent through being able to do all things that are possible to created nature; for the divine power extends farther than that. If, however, we were to say that God is omnipotent because he can do all things that are possible to his power, there would be a vicious circle in explaining the nature of his power. For this would be saying nothing else but that God is omnipotent, because he can do all that he is able to do. It remains therefore, that God is called omnipotent because he can do all things that are possible absolutely; which is the second way of saying a thing is possible. For a thing is said to be possible or impossible absolutely, according to the relation in which the very terms stand to one another, possible if the predicate is not incompatible with the subject, as that Socrates sits; and absolutely impossible when the predicate is altogether incompatible with the subject, as, for instance, that a man is a donkey. (STh. I q. 25, art. 3, corp.)

On Aquinas’s view it would be insufficient to answer the question as to what is possible for the almighty God simply by pointing to the potentialities of actual things. The potentialities of actual things only determine what is naturally or physically possible, i.e. those state of affairs which can obtain in virtue of the capacities of actual things. Since God is a super-natural being, he transcends the natural realm. He created the world and he could have created it differently. For instance, God could have created winged horses although there is actually no animal that is capable of giving birth to such a creature. Consequently, God’s omnipotence is not to be characterized in terms of the potentialities of the actually existing things, which only determine the range of natural possibilities. Nor can we, on pain of a vicious circle, characterize the range of God’s omnipotence by his own power. For this reason Aquinas suggests characterizing God’s omnipotence in terms of logical possibilities, which depend on compatibility relations holding between terms. On Aquinas’s account, God’s power is thus restricted by the realm of logical possibilities, which are constituted by conceptual compatibilities—and not by any potentialities. That is, it is possible for God to have created a winged horse, not because there is something (beside God of course) capable to engender a winged horse but solely because the predicate “winged” is compatible with the term “horse” such that the complex expression “winged horse” does not amount to a contradiction. Of course Aquinas’s characterization of God’s power in terms logical possibilities cannot be the end of the story. Rather, it raises the question as to what conceptual compatibility relations are grounded in. Although Aquinas has an answer to this question I will not pursue it here so as not to loose track of my initial question concerning the relationship of potentialities and possibilities in Aquinas. I will however come back to this issue in my discussion of Scotus’s account of modality. So, what about the significance of potentialities in Aquinas’s modal theory? Are they completely irrelevant for this enterprise since a potentiality-based account of possibilities is, on his view, not feasible in general? Such a dismissal seems to be mistaken. Even in the passage quoted above, Aquinas explicitly asserts that “whatever is subject to human power is said to be possible to man” and thus affirms that there are possibilities which are accounted for in terms of potentialities. This holds for all natural or physical possibilities because

Potentialities in the Late Middle …

129

those things, which are of such kind as to be done by inferior causes, are said to be possible in reference to those inferior causes. For it is according to the condition of the proximate cause that the effect has contingency or necessity. (STh. I q. 25, art. 3, ad 4.)

In referring to inferior causes Aquinas draws on the Aristotelian cosmology according to which the universe consists of an inferior sublunar world inhabited by corruptible substances on the one hand and the superior heavens or supralunar world on the other hand where everlasting celestial bodies move in circles.10 Now, while Aquinas is clear that “nothing happens contingently in celestial bodies, but only of necessity”,11 this is different in the changing sublunar world. Here things can also happen contingently. And this contingency is due to the proximate causes of these things, as Aquinas says. We will shortly turn to the question what it means exactly to be contingent for Aquinas and why he thought that the contingency of things is due to their proximate causes. For now it is just important to note that he conceived of the modality of natural affairs as a kind of causal modality and, for this matter, as being grounded in the potentialities of things. This is, and we will see this in greater detail in the next section, because for Aquinas causes produce their effects in virtue of manifesting a certain power or potentiality. Thus, even though Aquinas refuses to account for all kinds of possibilities in terms of potentialities, he is happy to appeal to potentialities or causal powers when it comes to explaining the modality of natural things. Let us now turn to the question what “contingency” (“contingentia”) is for Aquinas and why he thought of the contingency of things as being due to certain causes and their causal powers and thus subject to a potentiality-based account of modality. Aquinas is quite explicit about the meaning of contingency: “contingent is what can be and not be”.12 This characterization of contingency immediately renders intelligible why contingent truths cannot simply be identified with logical possibilities. For logical possibilities—defined as what is not impossible or necessarily false—is a one-sided notion of possibility insofar as it is limited only by the impossible. In contrast “contingency”, as it is introduced by Aquinas, is a two-sided notion of possibility. It is limited by the necessary and the impossible. So, unlike the one-sided notion of logical possibility that applies to everything that is not impossible (be it necessary or not), the two-sided notion of contingency is restricted to what is neither impossible nor necessary. Nonetheless, one could define contingency by appeal to logical possibility simply by adding the restriction that something contingent is logically possible and not logically necessary. Why then does Aquinas prefer to provide a completely different account by explicating contingency in terms of potentialities?

10

For an exposition of the Aristotelian cosmology and its medieval transformation see Grant (1996, especially Chap. 6). For Aquinas’s division in a contingent and necessary realm of nature see Aertsen (1988, 241–248). 11 Thomas Aquinas: InPeriHerm. I, l. 14.9. Cf. also SCG I.85.4. 12 STh. I q. 86, art 3, corp.

130

S. Schmid

As his relegation of contingency to the sublunar world of inferior causes already suggests, Aquinas associates contingency with a form of imperfection or privation. This is confirmed by his holding that “contingency arises from matter”,13 i.e. that constitutive principle of things which is devoid of all forms and perfections.14 As he illustrates by using the example of the growth of plants caused by the sun, contingency arises from imperfection or deficiency: “the effect of a first cause is contingent on account of the secondary cause, from the fact that the effect of the first cause is hindered by deficiency in the second cause, as the sun’s power is hindered by a defect in the plant.”15 As this example shows, an effect is contingent to the extent that it can be prevented from being brought about by its cause.16 So, the growth of a plant is contingent because the plant’s vegetative soul (which figures as its secondary cause) may, due to defects in the plant’s matter, not succeed to cause the plant’s proper development although the sun (as its first cause) infallibly emits its rays. Thus, for Aquinas the phenomenon of contingency, as the possibility to be or not to be, can only occur because there are imperfect things the performances of which are liable to failure and can be prevented by intervening causes. And it is exactly due to this liability to failure that it is possible for the effects of inferior causes to be or not to be. This shows that Aquinas holds a normative conception of contingency according to which the contingency of a state of affairs depends on whether it could have failed to be brought about.17 This also renders intelligible why Aquinas accounts for contingency in terms of causal powers. According to his normative conception there can only be contingency if there are things which can fail to exist because they are products of fallible causes that can be prevented in exerting their specific causal powers. Being contingent simply is to be the possible effect of a fault-prone cause. Aquinas’s normative conception of contingency as a causal modality grounded in the fallible powers of inferior causes is congenial to another strand of his thinking about modality I have not mentioned so far. This is the idea that there is an intimate connection between modality and change.18 Aquinas explains that “everything

13

STh. I q. 86, art 3, corp. Aquinas gives a concise exposition of his conception of matter in his De principiis naturae, [Thomas (1976)] 1.36-39 and 2.53-69. 15 STh. I q. 19, art. 6, corp. 16 In contemporary philosophy we are used to saying that only facts, state of affairs or propositions can be contingent. Therefore it might be puzzling that Aquinas holds that “effects” (i.e. substances or events) can be contingent. This puzzlement can be resolved, though, if one takes Aquinas to be talking about the “obtaining of certain effects”—which is indeed a state of affairs. 17 This normative conception of contingency traces back to Aristotle’s treatment of fate and accidents in his Physics II.5. As I will point out in the next section this normative conception of contingency is closely related with his acceptance of natural teleology. 18 Associating necessity with immutability and possibility with change has a long tradition. Boethius for instance has prominently endorsed such an association in his commentary to Aristotle’s Peri Hermeneias, III.9 (ed. Meisner II, 186f.). Cf. also Söder (1999, 59f.). 14

Potentialities in the Late Middle …

131

contingent is imperfect and mutable”.19 Conversely he holds that “every necessary thing is eternal”20 and thus endorses that necessity entails immutability since the “idea of eternity follows immutability”.21 For Aquinas then what is necessary is immutable and what is contingent (in the sense of being possible but not necessary) is liable to change. It is not surprising, therefore, that Aquinas conceives of contingency as a causal modality that is to be accounted for in terms of causal potentialities. After all, changes can only occur if there are causes that, by virtue of their power, exert a certain causal influence. Moreover, following Aristotle Aquinas maintains that time is the measure of change (InPhys. IV, l. 17). Therefore, Aquinas’s association of modality and change immediately entails an additional association of modality and time. In particular, Aquinas is committed to accepting that things which have happened and thus cannot be changed anymore are necessary. And indeed he explicitly states that “after a contingent being has been brought into existence, it can no longer be prevented”22 and hence no longer be the case and not be the case. For this reason Aquinas accepts the view—already to be found in Aristotle—that “everything that is must be when it is, and everything that is not, necessarily not be when it is not.”23 I will refer to this principle as the principle of the necessity of the present.24 The association of modality and change then, which allows Aquinas to account for contingencies in terms of (fault-prone causal) potentialities or powers, forces him in the same instant to accept an association of modality and time such that all present and past events are to be qualified as necessary since they can no longer be prevented.25 By means of his potentiality-based account of contingency Aquinas establishes a close relationship between contingency and imperfection, mutability and change and he thereby also associates contingency with time. This has the startling

19

STh. I q. 19, art. 3, obj. 4. STh. I q. 10, art. 3, obj. 3. 21 STh. I q. 10, art. 2, corp. 22 De Verit, q. 2, art. 12, corp. 23 InPeriHerm I, l. 15.2. Aristotle states this principle in De Int. IX, 19a23. 24 This principle is also discussed by Knuuttila (1978, 83f.), who takes it as the starting point for his statistical interpretation of modality. According to this reading Aquinas interprets modal statements extensionally being about truth at certain times. That is, ‘possibly p’ means that ‘there is a time t such that p is true at t’. I do not think however that Aquinas interpreted modal statements statistically. In his InMet. IX, l. 3.14, for example, Aquinas explicitly states that “there are some things which nothing will prevent us from designating as capable or possible of coming to be, even though they never will be or ever come to be”. For further critical remarks on Knuuttila’s interpretation see MacDonald (1995). 25 To be clear, Aquinas was not unhappy to accept this principle since it allowed him to give a neat solution of how God—whose knowledge is necessary—can know about contingent facts. Since even preventable and thus contingent facts are due to the principle of the present necessary when they occur, they are appropriate objects of God’s knowledge, who is eternal or outside of time and to whom the whole series of time is present at once. (Cf. STh. I q. 14, art. 13, corp.; De Verit, q. 2, art. 12, corp., and SCG I.67). 20

132

S. Schmid

consequence that only future events can be conceived as genuinely contingent. Hence, Aquinas’s theory of contingency generates a notion that is not applicable to all states of affairs one would like to call contingent in the sense of saying that they are possible to be or not to be. For surely we would like to say of some present or past events, which do and did actually happen, that they could have been otherwise. We would like to say for instance that the attack on the world-trade-centre in September 2001 is and remains contingent although it cannot be prevented anymore. But this might only be a sign that we—unlike Aquinas–are used to employing an untensed notion of contingency and there seems to be nothing inherently wrong in employing a tensed notion of contingency. What is indeed problematic however, is that the normative conception of contingency falls short of covering all two-sided possibilities Aquinas himself does accept. First, there is the phenomenon of free choice. Aquinas thinks that we can perform free choices and takes it to be decisive that they are not necessary. Yet, since contingency is grounded in failure and deficiency, free choices cannot literally be described as contingent because they are performed by the free will that is one of our noblest faculties, which we even share with God,26 who is perfect and hence devoid of any deficiencies.27 Second, according to Aquinas’s normative conception, contingency is only to be found at the level of inferior causes in the deficient sublunar world. This makes it impossible to regard God’s act of creation as contingent. This result is awkward for at least two reasons. On the one hand this stands at odds with Aquinas’s view that God is not necessitated to create a world, let alone this one, but that he freely chooses to do so.28 But once again, by associating contingency with imperfection Aquinas is unable to conceive of God’s choice of our world and thus of the existence of our world as something genuinely contingent. On the other hand, the denial that God’s act of creation is itself contingent since it arises from a perfect being whose powers are infallible seems to eliminate every form of contingency whatsoever. It is this problem John Duns Scotus called attention to only one generation after Aquinas’s death. He writes: If the first cause [that is, God; my addition] necessarily causes and moves its proximate cause and has a necessary relation to it, then that second cause necessarily moves as far as it is moved by the first cause. […] Therefore it is necessary that, if there is contingency in things, the first cause moves either the second cause contingently or the effect contingently, so that contingency proceeds from the action of the first cause. (Lectura I 39, §35 (ed. Vat. 17, 489; Vos Jaczn et al., 90))

26

In STh. I q. 83, art. 2, ad 3, Aquinas even opposes Augustine’s view that we have acquired our free will through the original sin for we can only choose the evil since then. Thus, our power of free will is not due to deficiency even though its range is widened by the original sin. Accordingly, acts of free will cannot be described as contingent, i.e. as relying on a fallible cause. 27 Cf. STh. I q. 19, art. 10. It is not surprising that Aquinas refuses to describe God’s creation as contingent exactly because contingency, on his account, is a form of deficiency; see STh. I q. 19, art. 3, ad. 4. 28 See for instance STh. I, q. 25, art. 5 ad 3, and De Pot. q. 1, art. 3, ad 8; q. 3, art. 17, corp.

Potentialities in the Late Middle …

133

Relegating contingency to the realm of secondary causes is of no help if these secondary causes are by themselves necessitated by a first cause. The necessary operation of the first cause seems simply to annihilate the contingency of the second causes.29 From this Scotus concludes that contingency—even in the sublunar world —can only be saved if we “assign the cause of this contingency to God’s causality, for an effect can only proceed contingently from a second cause if the first cause in that order moves it contingently […].”30 The problems for Aquinas’s normative conception of contingency are serious: He can neither account for the contingency of free choices nor can he describe God’s creation as contingent. Yet, there is a neat solution to both of these problems. If free choices count as genuinely contingent events one has not only solved Aquinas’s first difficulty, but has all one needs in order to tackle the second as well. One can then describe our world or the creation as genuinely contingent insofar as it arises from a free choice of God. It is not surprising therefore, that Scotus abandons Aquinas’s normative conception of contingency and instead tries to account for contingency in terms of free choices that can also be attributed to a perfect and thus immutable God. For this, however, free choices must not involve any change. Scotus seeks to meet this requirement by carefully distinguishing between two moments in the operation of our free will. First, our “will is free in regard to opposite acts (as in relation to willing and not-willing, and loving and hating), and second—mediated by opposite acts—it is free in regard to opposite objects to direct freely to them”.31 As he explains, these two moments differ in their respective perfection: Now there is something imperfect and something perfect in this entire freedom of the will: for having freedom in regard to opposite acts is something imperfect, because according to this it necessarily is receptive, and mutable as a consequence (because it does not have opposite acts at the same time). However, the freedom in regard to opposite objects is something perfect, for the fact that the will can operate as to opposite objects, is not something imperfect, but something perfect, just as the fact that the intellect can know opposite objects is something perfect. (Lectura I 39, §46 (ed. Vat. 17, 493f., Vos Jaczn et al., 110), translation modified)

29

Of course, Scotus draws his conclusion rather quickly. Aquinas could try to counter his objection by arguing that after God has brought his first creatures into existence they bring about their effects on their own. Accordingly, these effects remain contingent since they arise from finite and fallible causes. Still, this way out is problematic for Aquinas since he also thinks that God concurs in every act of a secondary cause by keeping it in existence. To this extent every effect is the product of a secondary cause and the unfailing and necessary first cause such that its effects threaten to be necessary (for this line of criticism see Lectura I 39, §36 (ed. Vat. 17, 489f.)). Moreover, there arises some pressure towards necessitarianism from Aquinas’s theory of providence according to which God ordained the whole creation and has thus already anticipated all causal interferences, which are the source of contingent events. Thus, ultimately nothing really fails, but everything accords to God’s plan (cf. STh. I q. 19, art. 6, corp.). Hence there seems to be no genuine contingency. 30 Scotus: Lectura I 39, §41 (ed. Vat. 17, 492). 31 Scotus: Lectura I 39, §45. (ed. Vat. 17, 493).

134

S. Schmid

In choosing a certain object we align our will with this object. Thereby we perform a certain motion: We change for example from the state of not-willing to have an ice-cream to the state of willing to have an ice-cream; or after having eaten too much ice-cream we can eventually pass on from loving ice-cream to hating it. All these volitions are changes that occur in time and are thus an expression of an imperfection on our will’s part. (If we were perfect we would have chosen the right thing from the outset and would not need to change our will anymore). However, that our acts of will can be directed at different objects, such that we have the real choice of, say, willing to have an ice-cream, a chocolate bar, an apple or an orange or anything else at a given moment is not an imperfection, but a perfection of our will. It is actually what freedom of will consists in. And in this sense even God’s will is free although his will is—due to his perfection—always already actualized and not liable to any change. Like ours, God’s will is free insofar as it can extend to different possibilities in the very moment it is directed at a certain object of will. As a result the possibilities, to which the free will extends, are not determined by any causal potentialities that unfold over time. Rather, they are logical possibilities determined by compatibility relations between terms and as such they exist simultaneously. [T]his logical possibility is not one according to which the will has acts successively, but it has them at the same moment. For at the same moment the will has an act of willing, at the same time and for the same moment it can have an opposite act of willing […] for willing for that moment and at that moment does not belong to the essence of the will itself, nor is it a natural attribute of it, consequently it results from it accidentally. But the opposite of an ‘accidental accident’ is not incompatible with a subject for some moment, and so the will, willing a at this moment and for this moment, can not-will a at and for the same moment. And this logical possibility is there with respect to terms which are not incompatible. (Lectura I 39, §50 (ed. Vat. 17, 495, Vos Jaczn et al., 116))

While Aquinas sticks to the necessity of the present and denies that ‘p’ and ‘possibly not-p’ can be true at the same time,32 Scotus admits (or even requires) that for a free person who wills a it is simultaneously possible to not-will a. Contrary to the principle of the necessity of the present then, Scotus allows for synchronic contingencies established by a free will that extends to all possibilities at once. Moreover, by freely choosing a possibility the will figures as the source of contingent truths. In virtue of a free choice something can obtain, although its opposite is simultaneously possible. Similarly, our world is contingent because it was freely chosen by God from a range of other possible worlds, which are all equally logically possible. From this it becomes clear how there is contingency in an effect: for just as our will can be considered as far as it is prior to its volition [i.e., its act; my addition] […], so at that moment and for that moment at which it has a volition with respect to something, it can not-will it and can have an opposite act – similarly the divine will, although it cannot have

32

Aquinas thought that the principle of the necessity of the present does not rule out contingency because he took it to be sufficient for its being contingent that p if “p” can be false at another point in time.

Potentialities in the Late Middle …

135

opposite acts (because his will is identical with his volition), yet wills in eternity a stone by one single volition and can will in eternity that there is not a stone or can not-will that there is a stone. (Lectura I 39, §54 (ed. Vat. 17, 497, Vos Jaczn et al., 126))

For Scotus, the reason why things could have been otherwise is not that they are the products of fault-prone secondary causes, as Aquinas explained. Rather, our world is contingent because it is the product of God’s free choice thanks to which in this very moment a range of other things are possible. And what is possible or contingent is not owed to certain causal powers of sublunar substances, but is simply determined by logical compatibility relations. Even though Scotus abandons Aquinas’s potentiality-based account of contingency his central idea to explicate possibilities in terms of logical compatibility or non-repugnancy is already to be found in Aquinas who maintained that what is possible for God is not restricted by the potentialities of natural things but only by logical consistencies. However, by employing this logical account of possibilities for characterizing contingencies in general, Scotus can provide a notion of contingency which is—unlike Aquinas’s normative conception of contingency—a genuine alethic modal notion that is to be distinguished from evaluative or deontic modal notions like deficiency and imperfection on the one hand, as well as from temporal and causal notions on the other.33 For Scotus ‘p is contingent’ simply means that ‘p’ and ‘possibly non-p’ are simultaneously true34—and this requires neither that p’s cause is in some way deficient nor that due to some change p may turn into non-p. The contingency of a state of affairs only demands that its negation is logically possible. In this respect Scotus’s abandonment of Aquinas’s potentiality-based account of contingency is surely a progress in conceptual clarity. At the same time however, Scotus’s logical conception of modality generates a new question concerning the ontology of possibilities. Since possibilities are no longer grounded in the potentialities of everyday things, the pressing question arises as to what these possibilities are. Inasmuch as possibilities are determined by compatibility relations holding between terms or concepts, a Platonic answer stands to reason: possibilities are abstract objects beyond space and time. Yet, being an Aristotelian, such a crude Platonic answer is not open to Scotus. Nonetheless, the Christian background allowed scholastic authors to borrow Platonic views without infringing the Aristotelian ban on uninstantiated universals. The divine intellect provided a perfect surrogate for the Platonic heaven since the abstract and detached Platonic ideas

33

However Scotus has not yet completely separated the notion of modality from the notion of time because he still abides by the principle of the necessity of the past (see his Lectura I d. 40, q. unica, §9 (ed. Vat. 17, 512f.) and as a comment Normore (2003, 136)). That Scotus purges the conception of contingency from any associations of imperfection is also emphasized by Söder (1999, 207) and Gelber (2004, 138). 34 Note that Scotus’s notion of contingency is narrower than the one employed by Aquinas. For Scotus contingency implies truth; this is not the case for Aquinas. For this reason commentators have suggested translating Scotus’s talk about contingency as “factuality” (cf. Vos Jaczn et al. 99 and 103).

136

S. Schmid

could simply be interpreted as thoughts of God’s intellect.35 And indeed possibilities for Scotus are ontologically grounded in the divine ideas with their mutual compatibility relations, which are brought about by the divine intellect.36 In this respect, Scotus deviates once more from Aquinas, who holds that God does not produce his ideas or the essences of possible things. On Aquinas’s account the ideas of God are rather included in his essence and God thinks the essences of possible things by reflecting on himself.37 This view has the advantage that it rules out that essences are arbitrarily created by God. This is important for otherwise it would be questionable to what extent essences can constitute necessary truths. Still, Aquinas’s position that God’s essence contains all ideas is in tension with God’s simplicity, which was accepted by all Christian thinkers. How can God, who is an absolutely simple being, have an essence that contains a variety of different ideas?38 It is exactly in response to this problem that Scotus maintained that God’s ideas are not actually contained in his essence but rather produced by his intellect. Unfortunately this view immediately yields the difficulty Aquinas could evade, namely that the essences of possible things, which are represented and simultaneously produced by God’s intellect, seem to be arbitrary and hence inappropriate to constitute necessities.39 Scotus solves this difficulty by insisting that it does not stand at God’s disposal which ideas there are and with which they are compatible. God by necessity only thinks the ideas he does in fact think and he could not think different ones. Moreover, the compatibility relations holding between these ideas depend on these ideas alone.40 Scotus—who was not coincidentally called “doctor subtilis”—substantiates this reply by way of introducing a distinction between the

35

For a discussion of this strategy and its impact on later medieval nominalism see Michon (1994), 475–482. 36 See his Ordinatio I d. 43, q. unica, §14 (ed. Vat. 6, 358f.) and as helpful exposition of Scotus’s metaphysics of possibilities Hoffmann (2002, 192–201). 37 Aquinas expounds this view in STh. I, q. 15. 38 In STh. I, q. 15, art 2, corp. and ad 2, Aquinas responds to this difficulty by arguing that the various ideas in God’s essence are nothing but his single essence, insofar as it is regarded as being imitable by a creature. Scotus rejects this solution by appeal to his theory of relations according to which relata are ontologically prior to the relation they are part of. Against this backdrop Aquinas’s solution becomes untenable: that God’s essence stands in the relation of being imitable by a possible creature already presupposes that there is a well defined essence of this creature which is only yet to be explained. For a detailed exposition of this dispute including references see Hoffmann (2002, 74–89) and Schmidt (2009). 39 Descartes has become famous for denying that this is a real problem. For him God did not make it the case that ‘2  4 = 8’ because this is necessarily true, rather ‘2  4 = 8’ is necessarily true because God has made it the case that ‘2  4 = 8’ (cf. his Sixth Reply, AT VII 436, CMS II 294). For a discussion see Alanen (1985). 40 The reason for this is twofold: First, God’s intellect is a natural faculty which—unlike the will— is determined to produce only one effect (see Lectura I d. 3 p.1 q. 3 §191 (ed. Vat. 16, 302). Second, possibilia have their possibility formaliter ex se, that is, simply in virtue of being the very things they are (see Ordinatio I d. 43 q. un §5 (ed. Vat. 6, 353)).

Potentialities in the Late Middle …

137

validity and the reality of a possible object:41 While the reality of a possible object is fully owed to God, who produces its idea or essence, its validity is independent from God. It rather pertains to a possible object “formaliter ex se”. Relying on this distinction, Scotus can even accept the counterpossible conditional that ‘even if there had been no intellect, it would have been possible that this world exists’, since the concept of the world possesses a sort of validity that is independent from its being thought and thus produced by the divine intellect.42 Scotus’s uniform logical account of modality goes—unlike Aquinas’s potentiality-based account of contingency—to a great extent without any appeal to potentialities. What is possible and what is impossible is simply a matter of logical consistency, and does not depend on the fact that natural substances are endowed with certain powers. Rather than on potentialities, possibilities depend on the general essences of things, which—according to the scholastic theories—figure as meanings of natural kind terms. For Scotus, that a human or a zebra are possible beings while a chimera is not rests on what these things essentially are: while a chimera is incompatible with being (repugnat esse), a human and a zebra is not. At least as far as the validity of possibilities is concerned, and thereby the answer to the question as to which possibilities are genuine and which not, potentialities are irrelevant. The answer to the question of the validity of possibilities solely depends on the essences of (possible) things. However, potentialities or, more precisely, rational faculties like will and intellect are of utmost importance if one turns to the ontological question concerning the reality of possibilities, i.e. the question as to what possibilities really are. With regard to this question Scotus is clear: possibilities consist in terms, ideas or concepts that stand in certain compatibility relations and these ideas are produced by God’s intellect. Hence the reality of possibilities (as opposed to their validity) still depends on a certain potentiality, namely on God’s power of thinking. Similarly contingency depends on the potentiality or power of will by means of which rational beings can perform free choices. That is, which possibilities happen to be actual, ultimately traces back to the fact that certain purely logically characterized possibilities are chosen by a free will, be it the free will of God that is responsible for the contingency of our world or be it the free will of imperfect rational agents like angels or humans.43 To apply the distinction introduced at the beginning of this section then, one can say that for Scotus even logical possibilities are, due to God’s intellect and omnipotence, realized in real possibilities. God’s will extends to everything logically possible and

“Validity” is my translation of “ratitudo” which Scotus ascribes to possibilities independent from their reality. For a thorough discussion see Honnefelder (1990, 45–56). 42 See his Ordinatio I d. 7, q. 1 §27 (ed. Vat. 2, 118f.) and Lectura I 39, §49 (ed. Vat 17, 494). Such counterpossibles are discussed by Knuuttila (1996, 140f.) and by Hoffmann (2002, 210–214). 43 That there are accordingly two sources of contingency for Scotus, is also stressed by Söder (1999, 208f.). 41

138

S. Schmid

consequently God has the potentiality to create everything that is logically possible.44 Despite the fact that the reality of modalities is ultimately grounded in rational potentialities and that due to God’s omnipotence every logical possibility corresponds to a real possibility it is beyond question that on Scotus’s logical account of modalities potentialities have to a great extent lost their pivotal role in accounting for possibilities. After all, the validity of possibilities is grounded in the essences of (possible) things rather than in their potentialities—and what is possible and what is impossible is solely determined (i) by the existing concepts or divine ideas and (ii) by the compatibility relations holding between these ideas. Of course one might doubt, whether this is a systematically coherent or stable position: Insofar as the “validity” of possibilities is not to be identified with their reality (since the validity of divine ideas explains why God’s intellect produces them), there is some pressure to concede that it is ultimately the validity of possible objects which explains their reality.45 And in this case one could legitimately ask whether it would not be more appropriate to fully embrace Platonism (in an undistorted version) in order to account for the ontology of modalities, and to dispense with potentialities in this question altogether.

2 Potentialities in Physics: The Problem of Natural Teleology As it has already become plain with regard to his normative conception of contingency in the last section, Thomas Aquinas thinks that natural processes can literally fail. This is no coincidence but an expression of a general teleological conception of nature that he adopts from Aristotle.46 Aquinas’s teleological conception of nature is general to the extent that it pertains not only to biological processes, but to all natural process—that is, to processes which are performed by natural things that “have in themselves a principle of motion and rest”.47 Natural substances for Aquinas include elements like earth, water, fire and air, which strive to fall to the ground or to move upwards or typically heat or cool certain things, as well as living beings, which grow and procreate. Of all these natural substances Aquinas thinks that they pursue or act for the sake of certain ends. Thus, fire does not only happen to burn flammable things but teleologically strives to do so, and Indeed Scotus explicitly says: “Corresponding with this logical possibility there is a real potency, for every cause is understood prior to its effect. Likewise, at the moment at which it elicits an act of willing, the will is prior by nature in regard to its volition and is freely related to it.” (Lectura I 39, §51 (ed. Vat. 17, 495, Vos Jaczn et al.,118); cf. also §49 (ed. Vat. 17, 494)). 45 Hoffmann (2009, 379) tries to avoid this consequence by simply rejecting the question as to whether possibilia are possible before God has thought them as “ill formed”. 46 Aristotle defends his teleological view of nature most notably in his Physics II.8. 47 Aquinas: InPhys. II, l. 1.2. 44

Potentialities in the Late Middle …

139

earth does not simply fall to the ground but somehow intends to do so. Aquinas expresses his view about the universal teleological make-up of nature by means of the slogan that “every agent acts for the sake of an end”.48 Such a theory of pervasive natural teleology is of course controversial. What are Aquinas’s reasons for this claim? In his Summa contra gentiles he provides the following argument: Every agent acts by its nature or its intellect. Concerning the agents who act by their intellect, though, there can be no doubt that they act for the sake of an end; for they act by conceiving in advance what they pursue in their actions, and it is on the basis of such a pre-conception they act; this namely is acting by the intellect. As there is a total similarity in the pre-conceiving intellect with the effect that is attained by the actions of the intelligent agent, likewise there is a natural similarity in the natural agent with its effect in advance, which is why the action is determined to this effect. So, fire engenders fire and the olive tree an olive tree. Thus, as the intelligent agent tends to a determinate end by its action, so does the natural agent. Therefore, every agent acts for the sake of an end. (SCG III.2.6)

Aquinas argues for his claim that every agent acts for the sake of a certain end by way of analogy. His argument can be reconstructed in three steps: (P1) Rational agents act for the sake of certain ends. (P2) The relation between natural agents (i.e. natural substances such as elements and living beings) and their effects is relevantly analogous to the relation between rational agents and the outcome of their actions. (C) Therefore: non-rational, natural agents or substances act for the sake of certain ends. If we concede that arguments by analogy are legitimate arguments, this argument is surely valid. It remains to examine whether its premises are acceptable. As Aquinas himself says there can be no doubt concerning the first premise (P1). If there is anything that acts (at least sometimes) for certain ends then there are rational agents. The second premise (P2), however, appears to be highly problematic. Why should the performances of elements like earth and fire be in any interesting way analogous to human actions? After all, humans perform their actions intentionally, they know about the effects their actions tend to bring about and most of the times they decide to perform these actions exactly because they want these effects to occur. None of this holds for the behaviour of elements. Earth falls down if it is dropped and fire sets flammable materials on fire if it is sufficiently close to it. These things just happen and they do not involve any acts of thinking and decision, as is the case with regard to human actions. Why then should we believe that the actions of rational and non-rational agents are relevantly analogous? Almost nothing, one might think, could be more different.

“omne agens agit propter finem.” (SCG III.2.6.) Note that Aquinas uses the terms “agens” and “agere” in a very broad sense to designate everything that is causally operative. Unlike contemporary authors, Aquinas does not restrict the term “agere” and its cognates to apply only to rational agents and their doings. Hereafter I will follow Aquinas in this use of terminology and will thus not presuppose that only rational agents can be agents in the proper sense of the term.

48

140

S. Schmid

In the quoted passage Aquinas tries to justify (P2) by describing human actions in such a broad sense that this description also applies to performances of non-rational substances. What characterizes rational actions and what is responsible for their teleological structure is the fact that their actions are guided by a cognitively anticipated end. By virtue of this cognitively anticipated end rational beings exhibit a similarity between the objects of their mind and the effects they tend to bring about. This point is best illustrated with the help of an example: If you want to surprise a friend for birthday, you think about what could make her happy, and consequently you act in such a way that everything looks similar to the way you have imagined it. Thus, the intentionality of our actions is due to the fact that our actions are guided by a cognitively anticipated situation and in the best case (when we succeed) this situation is brought about by our action. Now, according to Aquinas something analogous is the case in natural processes. In such processes the typical effects of natural substances are also anticipated—even though not in a cognitive or immaterial way, but in a material way—insofar as natural substances literally bear the forms of their typical effects. So, although fire does not think about sparking fire and although olive trees do not think about engendering more olive trees, they anticipate their specific effects by their own forms instantiated in their matter. Accordingly, due to their materially instantiated forms, natural substances exhibit a similarity to their effects quite like rational agents, who exhibit this kind of similarity insofar as the forms of their intended effects are immaterially instantiated in their intellect.49 For this reason Aquinas can maintain (P2) that the relation between natural agents and their effects is relevantly analogous to the relation between rational agents and the outcome of their actions. On the assumption (P1) that we are allowed to describe human actions as being performed for the sake of certain ends, Aquinas can now conclude that we should equally be entitled to say that (C) natural substances act for the sake of certain ends. Unfortunately, Aquinas’s justification for his second premise appears not to be very convincing. First, his strategy for arguing that human actions and natural processes are analogous to the extent that they both involve similarity of forms, seems not to be applicable to every natural process. Just think about the earth that falls down if dropped or the rising of fire. Contrary to Aquinas’s examples, these processes do not involve any form of procreation and consequently it is dubious in what respect natural substances anticipate their effects. Second, one wonders whether a mere likeness of cause and effect is sufficient for establishing a teleological relation between them such that a cause is operative for the sake of its effect. In order to tackle the first problem it is worth pointing to a difference between the cases where there is a similarity of forms involved (like in the fire’s burning a piece of wood) and cases in which there is not (like in the fire’s rising upwards). These cases differ to the extent that the former involves an agent as well as a patient

Strictly speaking “similarity” (“similitudo”) is a technical term for Aquinas that designates “a correspondence or convenience according to the form” (STh., I q. 4 art. 3 corp.). So, similarity is based on sharing the same form. Cf. Perler (2004, 55–60).

49

Potentialities in the Late Middle …

141

(the fire and the piece of wood), while the latter does not. That is, in cases where there is a similarity of forms involved an agent acts on a patient so as to induce its materially or cognitively anticipated form to the patient. The sculptor shapes the copper such that it looks the way he wants it to look, and fire ignites the wood so as to make it share its form. In the scholastic technical sense of the term only these cases are genuine instances of efficient causation. And in justifying his principle that every agent acts for the sake of a certain end, Aquinas is only interested in cases of efficient causation where an agent triggers the manifestation of a certain potentiality in a patient distinct from itself. In contrast to this, natural processes that only involve an agent doing something are—again in the scholastic sense of the term—instances of formal causation. They are simply actualizations of potentials things have in virtue of their substantial forms. That is, it belongs to the substantial form of fire to be above such that fire essentially strives to move upwards. Similarly, it belongs to the substantial form of an acorn to be an oak, such that the acorn grows to be a fully developed oak when circumstances permit.50 Now, even if Aquinas is mainly interested in efficient causation when he argues to the effect that every agent acts for the sake of an end, this does not amount to denying that formal causation exhibits a teleological structure as well. Rather, Aquinas holds quite generally that “the act of an active power relates to the object as a term and end, as the object of the power of growth is perfect quantity, which is the end of growth.”51 So, on Aquinas’s account, every exercise of a power can be described as obtaining for the sake of the power’s manifestation; may it involve a patient, like in efficient causation, or not, as it is the case in formal causation. This general view about the teleological structure of processes which not only applies to efficient, but also to formal causation is also reflected at the end of the above quoted passage, where Aquinas more loosely says that intelligent and natural agents are both determined to their specific effects. Sticking with this less rigid formulation one can also account for the teleological structure of natural processes that only involve formal causation like the earth’s falling down or the fire’s rising. What is crucial for there being a teleological relation between agents and their effects or performances is simply that the agents are determined by their forms to produce their specific effects or performances. So much for the first problem. Given that Aquinas takes it to be sufficient for teleological directedness that an agent is by its form determined to produce a certain

In STh. I q. 77, art. 1, ad 3-4, Aquinas states explicitly that “the substantial form is the first principle of action” and that hence every substance “operates by the power which results from the substantial form”. (In this respect the distinction between formal and efficient causation must be handled with care: efficient causation, where an agent operates on a patient, requires the efficacy of forms, by means of which an agent can trigger a certain change in a patient). Moreover, it was an issue of scholastic debate how formal and efficient causation should be distinguished, and the distinction suggested above was by no means accepted by all scholastic authors. (For an alternative distinction between formal and efficient causation see Suárez’s Disputationes Metaphysicae [Suárez (1866)], XV and XVII.) 51 STh. I q. 77, art. 3, corp. 50

142

S. Schmid

effect, the second worry raised above gains a slightly different shape. Now, one wonders why the fact that agents are determined to bring about their effects by their forms entitles us to assume that causes teleologically strive to bring about their effects. This depends on the special sort of determinacy in play here. Aquinas is clear that the fact that a cause is by its form determined to bring about its effect does not yield causal determinism according to which a given cause necessarily produces its effect. It is false, he writes, that when a cause has been posited – even a sufficient one – the effect must be posited, for not every cause (even if it is sufficient) is such that its effect cannot be impeded. For example, fire is a sufficient cause of the combustion of wood, but if water is poured on it the combustion is impeded. (ExPeryerm. I, l. 14.11)

That a certain substance is determined to produce a specific effect does not entail that it achieves to do so. It can always be impeded. A cause only brings about the effect it is determined to bring about if nothing interferes. That is, the kind of determinacy Aquinas has in mind here is a sort of dispositional determinacy: by virtue of its form a cause has the power to bring about its specific effect and is thus disposed to do so.52 But this does not rule out that under certain circumstances a cause can be prevented from manifesting its power. By using contemporary terminology we can say that Aquinas holds a dispositionalist account of causation, according to which causes are substances with certain powers, which enable them to bring about specific effects. And this dispositionalist theory of causation seems to suffice for Aquinas to justify the pan-teleological claim that every agent acts for the sake of an end.53 By their forms substances are endowed with certain powers or causal potentialities in virtue of which they are directed at certain effects. And on account of this they can be described as being operative for the sake of their effects.54 This shows that there is a strand of natural philosophy in Thomas Aquinas according to which the pervasive natural teleology is solely grounded in the fact that things have certain potentialities by virtue of which they strive to bring about their specific effects. However, Aquinas was not only, let alone mainly, interested in natural philosophy. His main-interest was to reconcile the Christian doctrine with Aristotle’s philosophy. And his strategy for doing so was integrating Aristotle’s natural philosophy into a broader Christian framework according to which the In STh. I q. 19, art. 6, corp. Aquinas writes for instance that “an effect is conformed to the agent according to its form”. 53 Cp. Footnote 52. Note however, that I have said nothing about whether a dispositionalist theory of causation does in fact commit one to accept that causes and effects are related in a teleological way as Aquinas seems to think. I explore this issue in Schmid (2014). 54 So, Aquinas’s main idea seems to be that powers exhibit a sort of intentionality (they are about their manifestations) which entitles us to describe causal processes arising from the manifestations of powers as we are allowed to describe human actions as teleological insofar as they emanate from intentional states. That powers do indeed display a sort of intentionality was recently maintained by a range of authors Martin und Pfeifer (1986), Place (1996), Place (1999), Heil (2004) and criticized by others Mumford (1999), Bird (2007, 118–126), Borghini (2009). 52

Potentialities in the Late Middle …

143

universe can be conceived of as a creation of a personal God. An important conceptual tool for integrating Aristotle’s natural philosophy into a superior Christian theory was surely his claim that there can only be genuine teleology or final causation when there is some cognitive intentionality in play. That is, Aquinas explicitly required what I will henceforth refer to as cognition-condition of final causality:55 (C) “Now that a thing is to be done for the sake of an end, requires some cognition of the end.” (STh. I-II, q. 6, art. 1, corp.) This requirement enabled Aquinas to achieve a range of results that perfectly accord with his theoretical aim to integrate Aristotelianism into a Christian framework. To begin with, this condition immediately allowed him to use the observation of natural teleology as a premise for a physico-teleological proof of the existence of a benevolent, omniscient and omnipotent God.56 With the same stroke Aquinas could exploit the natural teleology of Aristotle’s natural philosophy as evidence for his theory of divine providence. According to this theory, everything in the world is ordained by God, who is responsible for the fact that everything serves a certain purpose. Ultimately then, natural things exhibit a teleological striving because God by knowing their ends has directed natural substances towards their specific effects as an archer directs the arrow towards its target.57 Hence the teleological striving of natural things is not primarily due to their potentialities (as his natural philosophical account of teleology suggests), but rather due to God’s intentional directing of the natural substances to their specific ends by endowing them with their specific potentialities.58 Moreover, by means of his cognition-condition Aquinas can even admit more forms of teleology in nature than a pure Aristotelian account of teleology, which is restricted to the potentialities of things, could allow for. With regard to the potentialities of things one can at best establish an internal teleological structure of natural processes; that is, one can only argue that causes strive to bring about their specific effects, which they are intrinsically directed at in virtue of their causal powers. By appeal to his cognition-condition, however, Aquinas can additionally argue for a global teleological structure of the universe, that is, for the fact that the different kinds of substances constitute a teleological order insofar as the Aquinas was not the first to require that an end is to be known in order to qualify as a final cause. The cognition-condition is already to be found in Avicenna’s Metaphysics IV.5, 429. 56 Aquinas presents the physico-teleological proof as the fifth way of proving God’s existence in STh. I q. 2, art. 3, corp. 57 Aquinas explains his theory of divine providence over and over by analogy to the archer who directs his arrow to the target. See for instance STh. I q. 103 art.1 ad 3, De Ver., q. 22, art. 1 corp., and InPhys. II, l. 12.1. 58 Aquinas says that “if a thing has no knowledge of the end, even though it has an intrinsic principle of action or movement, nevertheless the principle of acting or being moved for an end is not in that thing, but in something else […], by which the principle of its action towards an end is imprinted on it.“ (STh. I–II, q. 6, art. 1, corp.) Note that Aquinas’s view that God endows natural things with their ends seems to be in some tension to the fact that these ends are supposed to be intrinsic to them. For more on this problem see Schmid (2011). 55

144

S. Schmid

operation of every species serves the purposes of the species on the hierarchically higher level. In this vein, Aquinas explains that the universe is a teleologically structured whole, in which plants grow in order to provide nutrition for animals, which serve humans, whose main aim is to worship God.59 For the defence of such a global teleology the assumption of the cognition-condition is crucial. It allows Aquinas to argue to the effect that there are also teleological relations holding between operations, which are not linked by potentialities, as long as they are arranged by an intelligent agent who wants these operations to serve each other.60 Even though Aquinas still abides by the Aristotelian doctrine that at least some forms of natural teleology (viz. the forms of internal teleology) rely on the potentialities of natural substances, this is not his ultimate analysis of final causation. If there is any genuine final causation involved in processes then this, due to the cognition-condition, owed to the fact that an intelligent agent knows about their ends and by virtue of this knowledge performs or establishes these processes. This deprives potentialities of their function to ultimately explain natural teleology. Moreover, and we will see this in our discussion of Ockham, making the teleological orientation of natural things dependent on the cognition of an agent makes it hard to defend a form of natural teleology that is intrinsic to the behaviour of natural substances. For now it is important to note that by requiring that teleological structures must ultimately rely on actions of intelligent agents that are performed for the sake of a certain purpose, the cognition-condition slightly shifts the nature of teleological explanations. If final causes are just states that (natural) substances are striving towards by virtue of their essential potentialities, then citing the final cause of a thing helps explaining of what kind of substance a certain thing is and what kind of activities this substance typically engages in. This in turn tells us which kinds of activities we can expect from a certain thing and thus helps us to understand why a given thing does what it does.61 If, however, final causes are known states of outcome that somehow cause or motivate a rational agent to perform a specific action, an appeal to final causes amounts to giving a sort of causal explanation of why a certain action was or is performed. This etiological conception of final causes 59

Aquinas discusses this at great length in SCG III.22. This marks a departure from Aristotle. This is perhaps at no place more explicit than in his discussion of Aristotle’s Physics II.8, 198b17–23, where Aristotle argues that the rain does not fall for the sake of the crop. In commenting this passage, Aquinas explicitly opposes Aristotle by stating that in general the rain does indeed fall for the sake of the crop (cf. InPhys. II, l. 12.5). A great deal of his confidence in his rejection of Aristotle might be due to his quite un-Aristotelian view—even expressed in this commentary (InPhys. II, l. 12.1)—that teleological relations are ultimately established by God’s purposeful arrangement of the world, and not primarily by the essential potentialities of things. 61 On this construal of teleological explanations, ends primarily characterize the kind of things and activities involved in a certain process, and thereby make intelligible why this process occurs. Giving a teleological explanation then, is to be understood as an attempt to answer a why?question by pointing out what the things in question are. For a systematic development of such a construal of teleological explanations see Thompson (2008, 85–119). 60

Potentialities in the Late Middle …

145

of course immediately raises the follow-up question in what sense ends or purposes do cause agents to perform their actions. It is exactly this question concerning the “causation” (causatio) of final causes that William of Ockham took up fifty years after Aquinas’s death. His answer is clear and brief: “the causality of an end is its being efficaciously loved and desired, so that in absence of that love and desire the effect would not be brought about.”62 Final causation for Ockham then is just a sort of counterfactual dependence. His core-idea can be put as follows: (FC) State x is the final cause of agent’s A performing U, if and only if (i) A loves or desires x, (ii) U is the bringing about of x, and (iii) if A had not loved or desired x, then A would not have performed U. Although this is not obvious, Ockham’s definition of final causes presupposes Aquinas’s cognition-condition. This is because (i) requires that a final cause is loved or desired, and thus is the object of certain emotions or “passions” as Ockham would say. But a passion, Ockham is clear, “requires an actual cognition for its actual existence.”63 The reason for this is simply that Ockham takes passions such as love and desire to be essentially intentional phenomena, which consequently involve a cognition or awareness of the thing they are directed at. Yet, by means of his definition (FC) Ockham does more than merely respect Aquinas’s restriction of final causes. He even imposes further restrictions on the application conditions of ascriptions of final causes by requiring that only those processes can be described as being subject to final causes which are performed by agents that love or desire these final causes by themselves. In the strict sense of the term then, only rational agents are liable to final causation and only their actions have final causes by which they are caused. This of course immediately raises the problem that natural or inanimate substances can no longer be seen as being properly subject to final causation. Ockham is quite aware of this difficulty. He writes: For this reason one has to consider in what respect inanimate things have a final cause. Concerning this I am inclined to say […] that if no knower governs or moves inanimate things towards an end, there is no final cause in them because in this case they are mere agents of natural necessity which intend nothing properly. If however they are governed by a knowing and desiring being, one has to assume a final cause in them. This final cause is not properly intended or desired by the inanimate agent but by the governor or the mover. Yet, it is properly said to be the final cause of this action or operation, like when someone shoots at a target – because he directs the arrow or wants to kill someone – the killing is the true final cause of the arrow’s motion even though it is not properly intended by the arrow, but properly intended by the archer. In this sense, properly speaking, such an agent is not said to act for the sake of such an end, but this is properly said of the one who directs or moves that agent. Hence it is not properly said that the arrow moves or acts for the sake of

62

Ockham: Quodlibet IV.1.1; OTh IX, ed. Wey: 293; Freddoso & Kelley 245. See also Summula Philosophiae Naturalis II.4; OPh VI, ed. Brown: 223f. 63 Ockham: Quodlibet II.17.1; OTh IX, ed. Wey: 186; Freddoso & Kelley 156. For an exposition of Ockham’s theory of emotions see Perler (2011, 155–197).

146

S. Schmid

the killing, but rather it is well said that the archer shoots the arrow for the sake of the killing. (Summula Philosophiae Naturalis II.6; OPh VI, ed. Brown: 228f.)

Ockham struggles with the fact that his definition of final causes strictly speaking rules out that the performances of inanimate substances, which are not able to love or desire a certain object, are due to final causes. Yet, the Christian tradition sets some constraints to maintain that all processes in this world serve a certain purpose and are even ultimately directed at God. In order to meet these constraints Ockham suggests distinguishing between intending a final cause and having a final cause. According to this distinction, only those agents can intend final causes (in the strict sense) who perform their actions due to their love or desire of a certain end. And only actions performed out of an intention of a final cause can be properly described as being for the sake of an end. Nonetheless, Ockham wants to allow that processes which are not performed by an agent that intends a final cause can be described has having a final cause. That is, it should be legitimate to say that the arrow’s flight has the final cause of hitting the target, although its flying is not itself subject to final causation, but is efficiently caused by the archer’s shooting the arrow. And this is exactly what Ockham suggests: the flying arrow has a final cause, even though it does not, unlike the shooting archer, intend it.64 On Ockham’s account then it is still possible to ascribe genuine final causes to natural processes as long as one is prepared to assume that they are “governed by a knowing and desiring being”. However, these final causes are only mediated inasmuch as they do not directly apply to natural processes. This marks a departure from Aquinas, who still tried to abide by the claim that God has endowed natural substances with an intrinsic teleological striving by equipping them with potentialities intrinsically directed toward ends. In contrast to this, Ockham denies that natural processes display an intrinsic teleological striving and are thus immediately liable to final causes. At most natural processes have an extrinsic teleological structure insofar as they are directed towards an end by an intelligent agent. Consequently, Ockham takes it to be inappropriate to describe the behaviour of natural substances as literally being performed for the sake of a certain purpose. This can only be said of actions performed by rational agents who exclusively can be sensitive to final causation.65 And so, “someone who is strictly following reason

In line with the distinction of intending and having a final cause, Ockham also distinguishes between proper and improper modes of attributions of final causes. I do not take this to be an inconsistency on his part—as McCord-Adams (1998) has argued—but rather as a legitimate way to do justice to the Aristotelian doctrine of natural teleology. For a further defence of Ockham against the inconsistency objections of McCord-Adams see Pasnau (2001, note 23). 65 This causes Ockham some troubles in explaining the behaviour of higher animals. In order to be able to account for their behaviour in terms of final causation he quickly lowers the standards for being subject to final causation, by no longer requiring that the agent whose actions are to be explained with regard to final causes has to love or desire her end, but only that it has an appetite for this end. In this respect, an animal’s behaviour can also be described as being finally caused by an end if it counterfactually depends on an appetite for this end on the animal’s part (see Summula Philosophiae Naturalis II.6; OPh VI, ed. Brown: 227f.). 64

Potentialities in the Late Middle …

147

would say that the question ‘For the sake of what’ is inappropriate with regard to natural actions for he would maintain that it is no real question to ask for the sake of what fire is generated; rather, this question is only appropriate with regard to voluntary actions.”66 The cognition-condition that requires that only recognized ends can qualify as proper final causes suggests a psychological understanding of teleological explanations according to which teleology is ultimately grounded in the intentions of intelligent agents. So, in his definition of final causes Ockham accommodates this understanding by restricting final causation to rational agents, who can perform their actions out of their love or desire of certain ends. This yields the metaphysical consequence that only actions of rational agents display an intrinsic teleology, whereas the teleology of natural processes is only extrinsic or derivative.67 If natural processes have a final cause at all, their final cause is in fact the end of God that motivated him to create or arrange these processes. Connected with this metaphysical consequence, there is also an epistemological worry about how we can know of the (derivative) final causes of natural processes. Since the ends of the performances of natural substances are no longer just fixed by their essential potentialities, but by the intentions of God who created them, knowing natural ends amounts to knowing God’s intentions or his will. For this reason Ockham writes: [I]t cannot be sufficiently demonstrated or known, either through principles known per se or through experience, that a thing that acts by a necessity of nature acts because of a final cause fixed beforehand by a will. And this is because the action of such an agent never varies without a change either in the agent or in the patient or in something that concurs in the action. Instead, the action always follows in the same way. And so it cannot be proved that such an agent acts because of an end. (Quodlibet IV.2.1; OTh IX, ed. Wey: 302; Freddoso & Kelley 251; translation slightly modified)

Since natural substances have (at best) only derived final causes they do not have any ends belonging to their essences. Therefore, one cannot know per se, that is, one cannot deduce a priori from the nature of natural substances, whether they have final causes. The question whether they have final causes thus becomes an empirical one. Unfortunately though, experience does not help us here. The only evidence for the fact that natural substances have derived final causes, which are imposed on them by a free will, would be that natural substances alter their performances from time to time as it is the case with regard to rational agents, who tend to perform many different actions depending on what they take to be desirable. It is, however, exactly this, which we cannot observe with regard to natural substances “because the action of such an agent never varies without a change either in the agent or in the patient or in something that concurs in the action.” On Ockham’s psychological conception of final causes then, natural teleology must remain an article of faith and cannot be known with certainty by reason. It no

66

Ockham: Quodlibet IV.1.2; OTh IX, ed. Wey: 299; my translation. That Ockham denies that final causes can be directly and literally applied to natural processes is also highlighted by Goddu (1999, 154f.) and Pasnau (2001, 309–315).

67

148

S. Schmid

longer relies on the essential potentialities of things but on God’s intentions by means of which he imposes ends to natural things from without. But since God’s intentions are neither empirically accessible to us nor to be known per se, as God freely chooses them, we can only speculate about the purposes of natural substances and never know them for certain. To this extent, Ockham’s position on natural teleology is not very different from the one defended by Descartes, who, 300 years later, made a case for withdrawing from speculating about natural purposes, for “we should not be so arrogant as to suppose that we can share in God’s plans.”68

3 Potentialities in the Late Middle Ages: Increasing Loss of Relevance? In Thomas Aquinas—or at least in certain strands of his thought—potentialities play a pivotal explanatory role. They figure in his explanation of contingency and provide a natural source of natural teleology. On the one hand Aquinas takes potentialities to be those properties of substances that endow them with a teleological striving for certain effects or operations. It is by virtue of their causal power that substances are directed at specific effects and can be described as acting for the sake of these effects when they exercise their powers. On the other hand the very idea that substances by virtue of their potentialities teleologically tend to bring about their effects allows Aquinas to maintain that something literally goes wrong if substances are prevented from manifesting their powers. And this is crucial for his normative conception of contingency according to which an event is contingent if it is produced by a cause that may literally fail should it be impeded. It is no coincidence then that Aquinas accounts for both, contingency and natural teleology, in terms of potentialities. If causes were not teleologically related to their effects in the sense that they are objectively supposed to bring them about, an interfering event, which prevents a cause from manifesting its potentiality, would not ensure that the disrupted causal process literally goes wrong. And consequently one could not hold that there are contingent events in Aquinas’s normative sense of the term, that is, events that can deviate from the way they are objectively supposed to occur. For this reason, it is crucial for Aquinas’s account of contingency that natural processes arise from potentialities of substances, by which they teleologically strive to bring about their natural effects. It is only on account of this teleological directedness toward their natural effects that natural substances are objectively supposed to bring these effects about. As it has hopefully become clear with respect to John Duns Scotus and William of Ockham, these complementary accounts of natural teleology and contingency, which both rely on a certain understanding of potentialities, radically changed in the

68

Descartes: Principles of Philosophy I.28, AT VIIIa 37, CMS I 202; without the addition of the French translation.

Potentialities in the Late Middle …

149

late 13th and early 14th century. Scotus was no longer willing to regard contingency as a sort of deficiency that is only to be found among inferior causes whose operations are liable to failures. Rather, he required that choices or free acts of will are contingent such that the creation as a whole qualifies as contingent insofar as it owes its existence to a free choice of God. In line with this, the potentialities of natural things were no longer relevant in order to account for contingency. Instead, their explanatory role is adopted by God’s free will. His will extends to all possibilities, which in turn are determined by compatibility relations between the essences of possible things, which are thought and thereby produced by God’s intellect. And it is from the range of clusters of compossible objects that God picks out our world and freely chooses to create it—and hence endows it with its contingent existence. In this same short period of time, potentialities also lost their relevance in accounting for natural teleology. The starting point for this development is already found in Thomas Aquinas, who sought to reduce the teleology of natural processes to God’s intentions in order to integrate Aristotle’s natural philosophy into a broader Christian framework. For this purpose (amongst others),69 Aquinas postulated the cognition-condition that requires that ends must be recognized in order to qualify as final causes that teleological explanations appeal to. According to this condition, there are no longer the potentialities of things, which provide the ultimate source of teleology, but acts of cognition that anticipate the ends of performances. Yet, only subsequent authors like William of Ockham spelled out the fatal consequence of Aquinas’s cognition-condition by arguing that, strictly speaking, only rational actions, which counterfactually depend on recognized and loved ends, display a genuine and intrinsic form of teleology. Natural processes can at best exhibit a sort of derived teleology if they are arranged and governed by an intelligent agent. As a result, having a potentiality by which a thing strives to bring about a certain effect is no longer sufficient for this thing being operative for the sake of this effect on Ockham’s account. He indeed explicitly acknowledges that sometimes “an end or a final cause is conceived as that which results form the operation of another according to the general course of nature, if it is not prevented, in just the same way as if it were foreknown or desired by an agent”. And he correctly notes that this “is how the Philosopher [i.e. Aristotle; my addition] speaks about final causality by the end of Physics II.” But Ockham makes it clear that this is an improper way of speaking.70 To be sure, Scotus and Ockham did not doubt that things possess various potentialities that account for their causal roles and thus enable them to elicit and

69

For further reasons that might have motivated Aquinas to postulate this condition see Schmid (2011, 29–33). 70 Summula Philosophiae Naturalis II.6; OPh VI, ed. Brown: 229f., my addition. See also Pasnau (2001, 310).

150

S. Schmid

underlie a range of changes. Nor did they deny the importance of these potentialities for explaining why things display typical behavioural regularities. But they refused to accept that such potentialities explain more than that. Although potentialities are modal properties to the extent that they determine what can happen with its bearers, they are in general irrelevant for answering the question of what it means to be possible and what possibilities are. And although potentialities endow their bearers with a certain striving or tendency, these inclinations or propensities are no longer conceived as being something genuinely teleological. In view of this it is to be conceded that potentialities have lost their explanatory functions in the later medieval theories of modality and natural teleology—at least in general. But not throughout. There are in particular two kinds of potentialities that became more important in the theories of Scotus and Ockham. These are the rational capacities of will and intellect. In the theories developed by Scotus and Ockham these two rather outstanding potentialities inherit the tasks, which in Aquinas—to a certain extent at least—were performed by the ordinary potentialities of mundane things. According to Scotus and Ockham, both, contingency and teleological directedness, are to be explained in terms of the activities of the intellect and will, which figure as sources of genuine contingency and proper teleological behaviour. With regard to this, it would be wrong to say that in the later Middle Ages potentialities lost their explanatory relevance tout court. Rather, one can notice a transformation or shift of explanatory burden. While the explanatory range of natural potentialities was restricted to physics (i.e. to explaining the various kinds of changes natural substances can be involved in), the non-natural potentialities of will and intellect gained explanatory importance. In light of the late medieval tendency of depriving natural potentialities of their explanatory tasks, the famous early modern critique of potentialities and attempt to do without them, begins to appear much less revolutionary than it is often said to be. In a way early modern mechanists just radicalised a general tendency, which can be already observed among late medieval authors. Moreover, the fact that these late medieval authors tended to shift the explanatory tasks formerly attributed to natural potentialities to the rational potentialities of will and intellect also points to another famous topic commonly related with early modern philosophers. This is the dualistic idea that the mind with its rational faculties is categorically distinct from everything else in nature. This idea gains plausibility if one thinks that rational potentialities can account for radically different things than natural potentialities. But this is a story to be told at another occasion.71

71

I am grateful to Kristina Engelhard, Marko Malink, Dominik Perler, Paolo Rubini and Barbara Vetter for their helpful comments on earlier versions of this chapter. I would also like to thank the editors of this volume, Michael Quante and Kristina Engelhard, for their patience and support and Simone Ungerer as well as Anna May Blundell for carefully proofreading this chapter.

Potentialities in the Late Middle …

151

References Aertsen, J. A. (1988). Nature and creature: Thomas Aquinas’s way of thought. Leiden: Brill. Alanen, L. (1985). Descartes, Duns Scotus and Ockham on Omnipotence and Possibility. Franciscan Studies, 45, 157–188. Aristotle. (1963). De Interpretatione, in idem: Aristotle’s Categories and de Interpretatione (J. Ackrill, Trans. and Ed.). Oxford: Clarendon Press. (cited as De Int.). Aristotle. (1956). Metaphysics (J. Warrington, Trans. and Ed.). London: Dent. (cited as Met.). Aristotle. (1995). Physics, Books I and II (W. Charlton, Trans.). Oxford: Clarendon Press. (cited as Phys.). Avicenna. (1907). Die Metaphysik des Avicenna (M. Horten, Trans. and Comm.). Halle, New York: Verlag von Rudolf Haupt. Bejczy, I. P. (Ed.). (2008). Virtue ethics in the Middle Ages: Commentaries on Aristotle’s “Nicomachean ethics”, 1200–1500. Leiden: Brill. Bird, A. (2007). Nature’s metaphysics: Laws and properties. Oxford, New York: Clarendon Press. Boethius, A. M. S. (1877–1880). In C. Meisner (Ed.), Commentarii in librum Aristotelis Peri Hermeineias. Leipzig. (cited as Meisner). Borghini, A. (2009). Dispositions and their intentions. In G. Damschen, R. Schnepf, & K. Stüber (Eds.), Debating dispositions: Issues in metaphysics, epistemology and philosophy of mind (pp. 204–219). Berlin, New York: De Gruyter. Descartes, R. (1983). In Ch. Adam & P. Tannery (Eds.), Œuvres de Descartes. Paris: Vrin. (cited as AT). Descartes, R. (1984). The Philosophical writings of Descartes (J. Cottingham, R. Stoothoff, & D. Murdoch, Trans.). Cambridge, New York: CUP. (cited as CMS). Dod, B. G. (1982). Aristoteles latinus. In N. Kretzmann, A. Kenny, & J. Pinborg (Eds.), The Cambridge history of later medieval philosophy (pp. 45–79). Cambridge, New York: CUP. Duns Scotus, J. (1950ff). Opera omnia. Commissio Scotistica, Civitas Vaticana: Typis Polyglottis Vaticanis. (cited as Vat.). Duns Scotus, J. (1994). Contingency and freedom, Lectura I 39 (A. Vos Jaczn, H. Veldhuis, A. H. Looman-Graaskamp, E. Dekker & N. W. Den Bok, Introduced, Trans. and Comm.). Dordrecht, Boston: Kluwer. (cited as Vos Jaczn et al.). Gelber, H. G. (2004). It could have been otherwise: Contingency and necessity in Dominican theology at Oxford, 1300–1350. Leiden: Brill. Goddu, A. (1999). Ockham’s philosophy of nature. In P. V. Spade (Ed.), The Cambridge companion to Ockham (pp. 143–167). CUP: Cambridge, New York. Grant, E. (1996). The foundations of modern science in the Middle Ages. Cambridge, New York: CUP. Heil, J. (2004). Natural intentionality. In R. Schantz (Ed.), The externalist challenge. New studies in cognition and intentionality (pp. 287–296). Berlin, New York: De Gruyter. Hoffmann, T. (2002). Creatura intellecta: die Ideen und Possibilien bei Duns Scotus mit Ausblick auf Franz von Mayronis, Poncius und Mastrius. Münster: Aschendorff. Hoffmann, T. (2009). Scotus on the origin of possibles. In S. Brown, T. Dewender, & T. Kobusch (Eds.), Philosophical debates at Paris in the early fourteenth century (pp. 359–379). Leiden: Brill. Honnefelder, L. (1990). Scientia transcendens. Hamburg: Meiner. Jacobi, K. (2001). Das Können und die Möglichkeiten. Potentialität und Possibilität. In T. Buchheim (Ed.), Potentialität und Possibilität: Modalaussagen in der Geschichte der Metaphysik (pp. 9–23). Stuttgart-Bad Cannstatt: Frommann-Holzboog. Jansen, L. (2001. Sind Vermögensprädikationen Modalaussagen? In R. Hüntelmann, U. Meixner, & E. Tegtmeier (Eds.), Metaphysica, Vol. 1: Ontologie der Modalitäten (pp. 179–193). Dettelbach: Röll. Knuuttila, S. (1978). The statistical interpretation of modality in Averroes and Thomas Aquinas. Ajatus, 37, 79–98.

152

S. Schmid

Knuuttila, S. (1993). Modalities in medieval philosophy. London: Routledge. Knuuttila, S. (1996). Duns Scotus and the Foundations of Logical Modalities. In L. Honnefelder, R. Wood, & M. Dreyer (Eds.), John Duns Scotus: Metaphysics and ethics (pp. 127–143). Leiden: Brill. MacDonald, S. (1995). Synchronic contingency, instants of nature, and libertarian freedom: Comments on ‘The Background to Scotus’s Theory of Will’. The Modern Schoolman, 72, 169–174. Martin, Ch B, & Pfeifer, K. (1986). Intentionality and the non-psychological. Philosophy and Phenomenological Research, 46(4), 531–554. McCord-Adams, M. (1998). Ockham on final causality: Muddying the waters. Franciscan Studies, 56, 1–46. Michon, C. (1994). Nominalisme: la théorie de la signification d’Occam. Paris: Vrin. Mumford, S. (1999). Intentionality and the physical: A new theory of disposition ascription. The Philosophical Quarterly, 49(195), 215–225. Normore, C. G. (2003). Duns Scotus’s modal theory. In T. Williams (Ed.), The Cambridge companion to Duns Scotus (pp. 129–160). CUP: Cambridge, New York. Ockham, W. (1967–1988). In G. Gál, et. al. (Eds.), Opera philosophica et theologica (Vol. 17). St. Bonaventure: The Franciscan Institute. (cited as OPh and OTh). Ockham, W. (1991). Quodlibetal Questions (Vols. I & II, Quodlibets 1-7, A. J. Freddoso & F. E. Kelley, Trans.). New Haven, London: Yale University Press. (cited as Freddoso & Kelley). Pasnau, R. (2001). Intentionality and final causes. In D. Perler (Ed.), Ancient and medieval theories of intentionality (pp. 301–323). Brill: Leiden. Pasnau, R. (2002). Thomas Aquinas on human nature. Cambridge, New York: CUP. Perler, D. (2004). Theorien der Intentionalität im Mittelalter. Frankfurt am Main: Klostermann. Perler, D. (2011). Transformationen der Gefühle: Philosophische Emotionstheorien 1270-1670. Frankfurt am Main: Fischer. Place, U. T. (1996). Intentionality as the mark of the dispositional. Dialectica, 50(2), 91–120. Place, U. T. (1999). Intentionality and the physical: A reply to Mumford. Philosophical Quarterly, 49/195, 225–231. Schmid, S. (2011). Teleology and the dispositional theory of causation in Thomas Aquinas. Logical Analysis and the History of Philosophy, 14, 21–39. Schmid, S. (2014). Teleologie und Potentialitäten - Gedanken über eine aristotelische Herausforderung. In A. S. Spann & D. Wehinger (Eds.), Vermögen und Handlung (pp. 63– 92). Münster: Mentis. Schmidt, A. (2009). Power and possibility in Thomas Aquinas. In J. Pietarinen & V. Viljanen (Eds.), The world as active power: Studies in the history of European reason (pp. 113–129). Brill: Leiden. Söder, J. R. (1999). Kontingenz und Wissen: die Lehre von den futura contingentia bei Johannes Duns Scotus. Münster: Aschendorff. Suárez, F. (1866). Disputationes Metaphysicae. In Ch. Berton (Ed.), Opera Omnia (Vol. 25). Paris: Vivès. (Editio Nova). Thomas, A. (1976). De principiis naturae. In Opera omnia. Editio Leonina (Bd. 43). Rom: Editori di San Tommaso. Thomas, A. (1949). In P. M. Pession (Ed.), De Potentia. Turin: Marietti. (cited as De Pot.). Thomas, A. (1952ff). In P. Caramello (Ed.), Summa Theologiae. Turin: Marietti. (cited as STh.). Thomas, A. (1953). In R. M. Spiazzi (Ed.), De Veritate. Turin: Marietti. (cited as De Verit.). Thomas, A. (1955). In P. Fr. Raymund & R. M. Spiazzi (Eds.), In Aristotelis libros Peri Hermeneias exposito. Turin: Marietti. (cited as InPeriHerm). Thomas, A. (1961ff). In P. Marc (Ed.), Summa contra Gentiles (= Liber de veritate Catholicae fidei contra errores infidelium qui dicitur summa contra gentiles). Turin: Marietti. (cited as SCG).

Potentialities in the Late Middle …

153

Thomas, A. (1965). In P. M. Maggiòlo (Ed.), In VII libros Physicorum Aristotelis exposito. Turin: Marietti. (cited as InPhys.). Thomas, A. (1994). In P. Fr. Raymund & R. M. Spiazzi (Eds.), In XII libros Metaphysicorum Aristotelis exposito. Turin: Marietti. (cited as InMet.). Thompson, M. (2008). Life and action: Elementary structures of practice and practical thought. Cambridge, MA: HUP.

Author Biography Stephan Schmid is Professor for the History of Philosophy at the University of Hamburg, Germany. He works on Late Medieval and Early Modern Philosophy, where he mainly focuses on metaphysical and epistemological questions. His publications include a monograph on “Finalursachen in der frühen Neuzeit. Eine Untersuchung der Transformation teleologischer Erklärungen” (de Gruyter 2011), as well as several edited volumes (among these “Dispositionen”, co-edited with Barbara Vetter, Suhrkamp 2013) and articles such as “Early Modern Debates on Faculties“ (in: D. Perler (ed.), Faculties, Oxford University Press 2015).

Part III

The Concept of Potentiality in the History of Philosophy—Early Modern Philosophy

Potentiality in Rationalism Michael-Thomas Liske

1 Introduction Let us start from a definition of ‘potential’ in order to examine to what extent the most important rationalists Descartes, Spinoza and Leibniz are in a position to recognise potentials in a significant and unstinted sense on the basis of their fundamental assumptions. A potential can be conceived of as a capacity, grounded in the inner constitution of a substance, to develop into a new positive quality under suitable external conditions. It is characteristic for a potential, on this view, to provide the enabling basis for a development towards a genuinely new state which springs from the inner constitution of a substance—towards a new state, which is not necessitated by the original constitution of the subject. If this conception of ‘potential’ is appropriate, we have to start from an essentially Aristotelian conception of substance. First, if we want to plausibly contrast the basis of development within the subject with suitable external conditions, we must assume a plurality of autonomous entities, each of which is one in virtue of its essence as an inner principle and distinct from all other entities of this kind. Secondly, if the substance is supposed to continually develop (out of itself), we must assume: The substance can, in virtue of its permanent essence, develop without this development affecting its identity, i.e., it can change in its accidental qualities and still continue to exist as itself. In Descartes, this second criterion for being a substance cannot be satisfied. For, because of his theory of a creatio continua, he atomises the temporal existence of a substance into numerous isolated, that is, unconnected, moments (and hence he also atomises the structure of time which depends on objects existing in time). Thus, the existence and condition of the substance within a given moment cannot just develop from its previous state according to internal laws, but requires permanent recreation. However, if there are no continuants, i.e., no substances M.-T. Liske (&) University of Passau, Passau, Germany e-mail: [email protected] © Springer Science+Business Media B.V., part of Springer Nature 2018 K. Engelhard and M. Quante (eds.), Handbook of Potentiality, https://doi.org/10.1007/978-94-024-1287-1_7

157

158

M.-T. Liske

continually persisting through time, then there can be nothing that continually springs from inside the substance, either. In Spinoza, in contrast, the first criterion for individual objects in the world is not satisfied. Since these are merely modi or unautonomous moments within a necessarily determined course of the world, these modi cannot really be the subjects of any processes that are internally determined, originating from a potential or an inner disposition, and that unfold this inner disposition under appropriate external conditions. This is different in Leibniz. Not only does he allow within the world a plurality of separate individual substances. Since he conceives of a substance as an internally determined succession of states, the notion of a potential as the inner principle of this succession or development plays an important role and appears under the title of a primitive (metaphysical) force. Because the individual substance mirrors the series rerum, the world as a succession of states of affairs, under an individual perspective, the individual is a distinctive succession of perceptual states, each of which successions represents the course of the world from its own perspective. The principle which constitutes the individual in its individuality is therefore the law determining this succession in a way characteristic for this individual. Leibniz conceives of this law of development as a primitive force or (as we might say) as a potential. However, there is a further feature of a significant conception of potential for which Leibniz has no place: the property of providing the basis for an internally grounded development which leads to something genuinely new, that is, to something which is preformed in the potential but not necessarily determined. The reason is that the closely-knit course of the actual world leaves no room for any indeterminate development whose designated goal may be defeated by external conditions. Certainly, potentials are compatible with a kind of determinism if they are (as sketched above) conceived of as inner principles which explain the advancement of its subject. Yet, we may add to this general notion of potential a more pointed understanding by postulating the following: A development which defines a potential (in this significant sense) must not merely unfold what is already preformed in the present state of a subject, but needs to transcend the status quo or the actual towards something new. A condition of a potential thus conceived would be a not just epistemically, but also ontologically understood contingency as the principle of not entirely determined change. In this case, that which is (or the real) is not exhausted by what actually exists or actually is the case (as in Spinoza). This richer ontology would allow for weaker, but still ontologically relevant ways of being in addition to what is actual. On this view, even for the development actually taking place it is ontologically relevant that non-actualized alternatives exist. A potential could, of course, not provide a complete explanation for an indeterminate development of this kind, but only the essential presupposition and basis for this development.

Potentiality in Rationalism

159

2 Descartes 2.1

Is an Atomistic Conception of Time Compatible with Genuine Powers of Bodies?

We want to start from Descartes’ atomistic conception of time, which he develops in the context of his doctrine of creatio continua in Principia Philosophiae (Princ.) I 21 (AT VIII. 1,13) and, parallely, in Meditationes de prima philosophia (Med.) III 31 (AT VII, 48f.). We then want to show that he cannot ascribe to bodies any genuine power as the cause of what happens in the next instant in the realm of bodies. Yet, power or force1 (as we have already indicated) is the notion under which Leibniz in particular discusses potentials. Descartes’ initial assumption is an atomistic view of the nature of time (which, for him, is not absolute time, but linked up with the duration and consistency of objects in time: temporis sive rerum durationis natura). Time consists of innumerable parts which are independent of one another and cannot, therefore, exist simultaneously. The immediate consequence of this view is: The temporal existence of beings such as man is also a succession of unconnected moments. That a human being has existed in the previous moment does not guarantee that he or she will continue to exist in the next as long as he or she is not destroyed. For him or her to actually continue to exist requires a permanent preserving effectiveness of a cause (causa). The next step is to further specify this preserving function. Since what is created does not, in itself, have any basis for continuance, but would instantly be annihilated if left to itself, its preservation is equivalent to a recreation ex nihilo. For it is obvious to anyone who attentively considers the nature of time that the same power (vis) and action are needed to preserve anything at each individual moment of its duration as would be required to create that thing anew if it were not yet in existence. (AT VII 49)2

The consequence of this is: Since only God is capable of creating something ex nihilo, we cannot really ascribe to any inner-worldly being, but only to God the properties of being a cause (e.g., of preservation), possessing power, and displaying activity (actio). Della Rocca (1999, esp. 58–62) tries to deny the conclusion that, according to Descartes, bodies cannot truly be said to possess any powers by which they cause body movements. He is surely right: That God is the prime and real cause of all movement does not rule out the possibility that bodies gain true powers and

1 In translating ‘vis’ I follow the common English translations. In Descartes, ‘power’ is always appropriate, although in decidedly physical contexts ‘force’ could be used. In Spinoza, ‘potentia’ is rendered by ‘power’. In Leibniz, who has an elaborated physical theory of force, ‘vis’ (French: force) is generally translated by ‘force’. 2 For the English translations of quotations from Descartes, Spinoza and Leibniz, see bibliography part II.

160

M.-T. Liske

tendencies in virtue of their relation to God and may hence be the proximate cause of the changes we observe in the realm of bodies. This is also the standard view of the scholastics. However, if a bodily power or, more generally, a dispositional property is to be responsible for an event in a subsequent moment, then this presupposes the identical continuance of its subject. In particular, such an identical continuant existence in time is to be presupposed in the case of the subject of a potential which is the ground for a lasting temporal development.3 For if the development of new properties is supposed to spring from the potential as the inner active principle, then the same potential needs to be effective during the entire development. However, Descartes’ creatio continua can only guarantee that a substance of the same kind with properties of the same kind exists without interruption. It cannot guarantee that the substance and hence its potentials are preserved as identical in the strict sense of the word. The reason is that, strictly speaking, in each new moment a new substance is created and with it numerically different properties which are identical only in the sense of being properties of the same kind. Continuant existence as an identical substance needs to have its basis in the substance itself.

2.2

Descartes’ Talk of Tendencies as a Useful Description of Phenomena

That Descartes does not assume any true capacities (or potentials) as a mode of being different from being actual becomes especially clear where he employs the terminology of ‘tending’ (tendere), as in the context of the second law of nature (Princ. II 39). It is claimed there that a particle of matter, viewed by itself, never tends to continue its movement in a curved line, but always in a straight line (AT VIII. 1,63, L. 20–23). A stone which is moved circularly in a sling, therefore, tends to move away from the centre of the circle in a tangent to the circumference (64, L. 7–27).4 Leibniz reduces this behaviour—that the spun around stone tends to fly away in the tangent—to the stone’s dead force (vis mortua), i.e., to the still dormant force or potential energy. In contrast, Descartes views God’s inmutable preserving activity as the cause of this law, assuming that through this activity God keeps the body in its actual movement which, according to the principle of inertia, needs to be a straight one (63, L. 26–64, L. 2). Descartes does not conceive of tendencies or powers, that is, dispositions of bodies, as modal ways of being, i.e., as non-observable, but—with regard to the explanation of observations—conjectured

3

Supposing a creatio continua suggests an occasionalism. It is a matter of debate in how far Descartes was an occasionalist. Relying on the experience of ourselves in which we are immediately and therefore indubitably aware of our soul’s power to cause bodily movements, Descartes refuses occasionalism concerning mental causation. See Kolesnik-Antoine (2009), esp. 203–205. 4 See Machamer (2008) 76–78.

Potentiality in Rationalism

161

real properties. Rather, they are auxiliary quantities for a useful description of phenomena, by which we can represent the relations between phenomena in an intelligible and coherent manner. The talk of powers, therefore, is not an ontologically significant characterisation. Accordingly, Descartes (Princ. III 56) does not define striving (conatus) by means of underlying reality, for example as a rational or generally intentional directedness—as would be appropriate if he conceived of it as a modal status of being or an actually efficacious momentum. Rather, he identifies striving with the help of that phenomenon by which we can most perspicuously describe and explain the notion of striving. Bodies display a striving towards a certain direction if they are in a position to actually move in that direction if nothing else hinders them (ut revera sint eo versus ituri, si a nulla alia causa impediantur). In Princ. II 64, it becomes quite clear why Descartes cannot ascribe to bodies any powers as something which they by themselves and originally possess. Since he takes the nature of bodily substance to consist in extension, all original qualifications of bodies can be captured by the quantitative geometrical categories of extended space. In so far as they are spatial, bodies are (in principle) arbitrarily divisible (divisibilis), manifoldly figurable (figurabilis) by different spatial boundaries and moveable (mobilis) through a change of their spatial position. An obvious objection against this is that, in order to fully capture the specific nature of matter, we need to rely on dynamic categories as well. Thus, Leibniz in particular defines matter by appealing to passive force and conceives of the latter as the reactive force of resisting (vis resistendi), which expresses itself in impenetrability and inertness. Now, Descartes draws upon such dynamic notions as well in order to describe phenomena of motion. However, in his fundamental assumptions, he cannot allow them any ontological reality in the bodies themselves, but only admit them as useful explanations of phenomena which correspond, in the ontological deep structure, to God’s preserving activity. Yet, that, for Descartes, bodies do not possess a tendency to motion in an ontological sense, and that he rather conceives of God as the general cause of motion (causa generalis omnium motuum, Princ. II 36), does not mean in turn that, by themselves, bodies display a tendency to rest (tendere ad quietem, Princ. II. 37), against which God must maintain motion. This would amount to the kind of aristotelian-scholastic conception, resolutely objected to by Descartes, which common-sensically infers from appearance that bodies have a tendency to discontinue their motion. What God’s moving activity amounts to from an ontological perspective, can—closer to the phenomena and secondarily—be described and explained as a tending of the bodies in question. That the tendencies of bodies relate not to rest,5 but to motion, is proven by the natural sciences.

5

Provided one does not conceive of resistance as a power or tendency towards rest. See Coelho (2002) 48.

162

2.3

M.-T. Liske

Descartes’ Powers as Instantly Realisable Dispositions

It may be contentious whether the passages in which Descartes’ wording seems to attribute powers to bodies disclose the deeper ontological grounds of phenomena or whether the terminology he uses is just a useful mode of presentation which allows us to grasp the law-like relations between the phenomena and to make appropriate predictions. In any case, however, the capacities and powers Descartes discusses are not directed at the development into something new. They are, therefore, not potentials, but dispositional properties in general, that is, capacities for instantly realisable behaviour. For one thing, Descartes discusses tendencies of motion, e.g. Princ. II 40 the power (vis) to continue a movement in a straight line even against resistance, or, in turn, the power to resist and impede such movement of another body. Or, to give another example, what is being described as powers in the letter to Mersenne from the 28th of October 1640 are dispositions to a behaviour in line with the law of inertia (la force de continuer à se mouvoir, AT III 213, L. 4f.). Furthermore, Descartes employs the notion of a capacity in the context of the freedom of the will.6 In his letter to Mesland (2nd of May 1644), he describes the freedom of the will as the actual and positive capacity to determine oneself (puissance réelle et positive de se determiner, AT IV, 116, L. 17)—in contrast with the mere negative freedom from coercion, which we also find in animals. Quite obviously, the act to which this capacity is directed: to (positively) determine oneself for or to choose what is appropriate for oneself, is punctual and not a process of development which is grounded in a potential. The opposite acts which we can (posse) perform in virtue of the freedom of the will, that we may affirm or negate the same thing, consent to it or not, pursue or avoid it, and all this at our discretion (ad arbitrium) and without being externally determined (Princ. I 39; 4. Med. AT VII 57), are not developments towards a new quality, but actual activities. Descartes discusses another important mental faculty (facultas) with regard to innate ideas. As contents of actual thought, these are not permanently present to us (semper obversari). In this sense, there are no ideae innatae. Rather, we only have in us the capacity or disposition to, as it were, elicit these contents from ourselves in a given situation (facultas elicendi) (answer to objectio X. of the third objections raised by Hobbes, AT VII 189). An innate idea is given merely as the immanent and therefore innate faculty to think (cogitandi facultas quae in me est), to grasp through thinking alone (i.e., without any information deriving from the external world or any fictions of the will) true ideas of intelligible entities, such as God (Notae in programma, AT VIII. 2, 358, L. 1–6). This, too, is a clear example of an instantly realisable mental disposition, given the appropriate conditions, not of a potential for development. That potentials for development do not play a role in Descartes is also shown by the fact that, in Princ. II 43 for instance, he relates powers to the tendency to persist: „[…] each thing tends, as far as it is up to it, to persist in the same state in which it 6

See Campbell (1999).

Potentiality in Rationalism

163

(already) is.”7 This tendency to persist, which is also relevant to Spinoza’s notion of conatus, is taken by Descartes to be of such central importance that he reduces to it the power which manifests itself in the interactions of bodies, i.e. “the power of any given body to act on another body or to resist the action of another body”.8 Moreover, in Princ. III 122 he not only speaks of a power, but of an impetus to persist in a given state of motion which metals have to a higher degree9 than other substances, so that they are more solid.10 If the tendency to persist in the status quo according to the metaphysical principle of inertia is so important in Descartes as to provide the explanation for something else, such as the active and reactive powers of bodies in bodily interactions or solidity, then the potential to develop into something new cannot, in turn, be of any particular importance.

2.4

Descartes’ Turn Away from the Scholastic Potency-Act Doctrine

In his later work Les passions de l’âme (Pas.), Descartes is particularly keen to avoid the notion of potency and speaks rather of faculties (facultés), obviously in order to distance himself from the scholastic potency-act doctrine.11 A potency (potentia), according to Descartes’s objection, is realised in an act. There are, however, several receptive potencies: mental faculties, such as perception, as well as bodily capacities, such as to receive motion in the case of a collision. Yet, these are realised by being acted on (passio) (see to Regius Dez. 1641, AT III 454, L. 3– 455, L. 19). This objection surely rests on a misunderstanding of the potency-act relation, which springs from an ambiguity of the scholastic ‘actus’ and the underlying Aristotelian ‘e̓meq ́ ceia’. In its specific use, ‘actus’ describes an activity, as is already suggested by the etymological derivation from ‘agere’ = ‘to do’ or ‘to effect’. Nevertheless, ‘actus’ must not be confused with ‘actio’, which exclusively refers to activity. In contrast, an act (actus) can generally be any form of realisation of what a potency is disposed to. It thus also covers the realisation of a passive capacity by actually being acted on. Yet, even if Descartes’ reservations against the notion of potency thus rest on a misunderstanding, they may still explain why potentials are not important to him. A potential includes the active momentum of an impetus towards a certain development. Among the mental faculties Descartes confines, in accordance with the voluntarist tradition, the active to the will, whereas

“[…] unaquaeque res tendat, quantum in se est, ad permanendum in eodem statu in quo est.” (AT VIII. 1, 66) 8 “[…] vis cujusque corporis ad agendum in aliud vel actioni resistendum” (Ibid.) 9 “[…] majorem agitationem sive majorem vim ad perseverandum in suo motu retinere.” (AT VIII. 1, 172) 10 On the connection between solidity and the law of inertia, see Slowik (1996). 11 See Sherman (1974) 558–561. 7

164

M.-T. Liske

he conceives of all understanding, not just sensory perception, as passive (Pas. 17). On account of this, most mental faculties, being capacities to be acted on, do not provide plausible candidates for a potential. Descartes is also dismissive of potencies for another reason. A plurality of relatively autonomous mental faculties seems to presuppose different parts of the soul, so that each part can bear one of these faculties respectively. However, for Descartes we can conceive different parts with different forms and sizes only of what is bodily extended, and not of the incorporeal mind (to Morus, 5. Febr. 1649, AT V 270, L. 8–12). Only the uniform, partless subject soul in its entirety can be capable of many things. Therefore, manifold faculties (facultés) of the soul as a whole can make their appearance (Pas. 68).

3 Spinoza 3.1

Does the Theory of Modality Allow for True Potentials?

In so far as potentials, at least according to the common understanding, are dispositions towards a non-necessary, that is, contingent development, an asymmetry within Spinoza’s important explication of modal concepts in Eth. I p.33 sch.112 is illuminative. Spinoza confines the ontological modalities, which represent a modal differentiation of the way of being, to necessity and impossibility. Contingency, therefore, is possible only as an epistemic modality, which modally differentiates to what grade of certainty we can—according to our knowledge—expect a certain event or thing to happen or to exist. Of the real ontological modalities, which belong to an object independently of our knowledge, Spinoza assumes, in virtue of two possible causes, two types (here as already in the early Cogitata Metaphysica (Cog. Met.) I 3, G I 240): Either, we have an essential necessity or impossibility, if it necessarily follows from the inner nature of the object itself, as given in the real definition, that it exists or that its existence is ruled out by a contradiction in the supposed nature.13 Or, a causal necessity or impossibility is conferred on the object from without because of the causal course of events, if the determination thereby effected either necessarily brings about the existence of the object or necessarily

12

See Schütt (1985). Spinoza’s formulation “A thing is called impossible […] because its essence or definition involves a contradiction […]” cannot be taken literally. According to II def. 2, essence and thing (res) are dependent on each other: that which belongs to the essence can neither be nor be conceived of without the thing and vice versa. Since, therefore the constitutive principle and the thing constituted are ultimately identical, a contradictory essence and definition are ruled out as is a contradictory thing. Spinoza’s meaning is: If we conceive something of which we find out that its supposed essence includes a contradiction, then this something does not have an essence. Being impossible, it is excluded from existence in the first place.

13

Potentiality in Rationalism

165

rules it out. Of the contingent, in contrast, Spinoza has only the epistemic notion,14 which consists in the following: Because of our incomplete knowledge (respectu defectus nostrae cognitionis) of the true nature and the scheme of determination, we can neither claim the existence of the object as necessary nor definitely rule it out. If, from an ontological perspective, the necessary coincides, therefore, with what is (necessarily) real, and the impossible with what is (necessarily) not-real—which means that the two form a complete disjunction and leave no room for a intermediate modality (the real contingent as neither necessary nor impossible), then there is no ontological modal status distinct from being real, that is, no way of being of the object itself, which precedes its actual occurrence and from which its being real can, but need not, emerge. In Spinoza, potentials as capacities for a development which may fail to appear or remain incomplete do not qualify as real capacities, but only as rational concepts which, given our limited view, represent a legitimate and useful approach to how the properties of an object relate to its possible behaviour or its prospects of development. As a specific type of dispositional properties, potentials are no irreal or counterfactual properties whose realisation is left to a merely possible world (with never realised conditions). Rather, they are hypothetical properties which are realised under conditions that may, but need not, occur. They may, therefore, turn out to be counterfactual. Since Spinoza recognises only one total reality in the strict metaphyiscal sense, he has to exclude all that is counterfactual.15

3.2

Does the Causal Necessity of Individual Beings Allow for Potentials?

Potentials for development presuppose changeability which in turn presupposes contingency. For this reason, Spinoza (as we have seen) cannot recognise potentials in the full sense of the word with all their ontological implications. In what respect we can still speak of potentials within the framework of his metaphysical system appears from the difference between causal and essential necessity. Essential necessity belongs to substance16 which actively presents itself as the (creator-)God and passively as the totality of the created world. Subject to causal necessity are, according to I p. 28, the finite individual objects which exist and operate in a

14

See Laerke (2007) 41–47. On necessitarianism, see, for example, Bennett (1984) Chap. 5, 111–124, Garrett (1991), Curley and Walski (1999), Koistinen (2003), Perler (2006) and Griffin (2008). Carriero (1991) shows how Spinoza revises the medieval theological notion of necessity under the influence of the new natural sciences. 16 This connection is already included in the origin of these concepts, in Aristotle’s notion of oὐría, which in Latin is to be given as both ‘essentia’ and ‘substantia’. 15

166

M.-T. Liske

specific, determined sense only in so far as they—being incorporated in causal chains—are connected with the existence and operation of other finite individual objects. With regard to essential necessity, potentials have no place. Where Spinoza talks of God’s potentia, only one of the two meanings of this term is relevant, namely the meaning of ‘force’ or ‘might’, not the one of potency.17 God cannot possess an Aristotelian active potency to operate thus or otherwise, to become effective or not to become effective. This is because, according to I p. 34, God’s potentia coincides with His essence,18 which since Aristotle is regarded as the thing’s constitutive core that cannot be lost remaining the same throughout a thing’s entire existence. God’s efficacy necessarily makes unfold what is included in His nature (I p. 16),19 without the possibility of something external hindering it, which cannot be given with regard to the whole in the first place. God’s power to act and His actions are one and the same and therefore necessary. Because of the immutability of His creative activity, God could not have produced a different total reality (res) with a different order (ordo) than He has actually produced;20 for whatever is subject to the divine power (potestas) happens by necessity (I p. 35). In the ontological sense, that is, in the nature of those things which together make up the world (in rerum natura), contingency is ruled out, according to I p. 29. Rather, the total reality is determined to its actual way of being and operating through the necessity of God’s nature.21 It does not possess, therefore, any passive capacity of being formed thus or otherwise. How about the causal necessity of the individual beings? Their being part of causal chains (which together constitute an order of the world without any alternatives) rules out an acausal chance and a spontaneous beginning of new chains of events. Nevertheless the active power of an individual being need not remain the same (in contrast to that of the whole), but can gain or lose with respect to the active power of other individuals (in the latter case it is rather passive). Being mere modi, individual beings do not possess an unchangeable nature through which they are determined to their way of being. Even if, in accordance with the conservation of energy, the active power need remain the same within the entire system, variability and thus potentials for development may occur in individual beings, in so far as the causal efficacy can be distributed differently among individuals. Since an individual is only relatively, that is, more or less potent or powerful, comparative statements are appropriate: “There is no singular thing (res singularis) in all of nature that is not surpassed in power or strength (potentior et fortior) by some other.” (IV ax.)

17

Spinoza’s potentia is always activity and does not therefore, unlike Aristotle’s potency,differ from the act nor precede it. See Narbonne (1995), esp. 38. 18 “Dei potentia est ipsa ipsius essentia.” (I p. 34) 19 See Matheron (1991). 20 “Res nullo alio modo neque alio ordine a Deo produci potuerunt, quam productae sunt.” (p. 33) 21 “[…] sed omnia ex necessitate divinae naturae determinata sunt ad certo modo existendum et operandum.” (I p. 29)

Potentiality in Rationalism

167

The elucidation of ‘potentior’ through ‘fortior’ reveals: Even in the case of individual beings, Spinoza originally conceives of ‘potentia’ (and ‘potens’) not as a (contingent) capacity or potential (for development), but—as in the case of the divine whole22—as the active power to be effective (see also I p. 36). However, since this power is only relative in the case of individuals, there can still be changes or development and therefore potentials for one individual’s changing in its active power with regard to another and becoming more or less powerful. In II p. 13 sch., Spinoza further elaborates on the criteria by which we may read off the gain of potency, i.e. the greater or lesser power. To be sure, he speaks here of ‚more apt’ (aptior), but means something like ‚having more potency’ (potentior). Since Spinoza conceives of the bodily and the mental as two parallel ways of being (attributes) of the one total reality (identified with God), they exercise their respective activities or modes of activity in exact analogy with one another— without, however, necessarily causally operating on one another. […] in proportion as a body is more apt than others to do many things at once (aptius ad plura simul agendum), or to be acted on in many ways at once, so its mind is more apt than others to perceive many things at once. And in proportion as the actions of one body depend more on itself alone, and as other bodies concur with it less in acting, so its mind is more apt to understand distinctly.

Quantitatively, the degree of efficacy depends on how far-reaching and comprehensive the specific ways of being effective are. In the case of bodies, these consist in the causal interactions of acting and being acted on: A body is the more powerful, the more it can simultaneously act or the more active power it can passively receive from the operation of another body. Mentally, the efficacy consists in how much a mind can perceive at a time. Besides, the efficacy is to be measured according to a thing’s autonomy, that is, according to whether it is capable of producing something out of or by itself or whether it depends on the assistance of others. In the mental sphere, autonomy is constituted by a distinct grasp. For in the case of distinct ideas discriminating all individual beings mind permeates its matter and rules over it—whereas in the case of unmastered and confused impressions (in the form of affects), it is determined and dominated by external influences. If, according to these criteria, the efficacy can increase or decrease, then an individual can possess the potential to develop towards a greater efficacy—a potential whose unfolding characteristically depends on external conditions and may be supported or hindered by them. This is implied by III p. 11: Whatsoever increases or diminishes, aids or restrains (auget vel minuit, juvat vel coercet) our body’s power of acting (agendi potentiam), the idea of this thing increases or diminishes, aids or restrains our mind’s power of thinking (cogitandi potentiam).

22

Thus, in IV def. 8, Spinoza equates potency (potentia) and virtue (virtus) also with regard to man and conceives of it as the effective power (potestas […] efficiendi) to bring about something that can be explained solely by reference to the laws of his inner nature.

168

M.-T. Liske

Because of his psycho-physical parallelism, Spinoza presupposes here (as in II p. 13 sch.)23: The mind’s power of thinking can increase or decrease, just as the body’s power of acting, and hence possesses potentials for development. If something, or the idea thereof, is supposed to increase or decrease this power of acting or power of thinking then what Spinoza here has in mind could very well be an inner moment of the bodies or the mind. However, it seems that we can only speak of ‘aiding and restraining’ with regard to such factors which externally affect an internally-generated development of an object. Therefore, what we are concerned with here are in fact potentials which are grounded in the internal structure of objects, but whose actual realisation and unfolding depend on external conditions. Now, it is hardly a coincidence that Spinoza does not distinguish between the external and the internal. Since individual beings are mere modi and not substances, they do not represent autonomous entities which are distinct by themselves. Rather, each of them is the result of the causal interdependence of the whole whose moments they are. However, given that the human body like the human mind is something relatively permanent, we can still view them, with Spinoza, as quasi-subjects and in this way distinguish, in a relative sense, their internal structure from the external world. In this qualified sense, we may then also speak of potentials as tendencies for development which are internally constituted, but referred for their realisation to suitable external conditions. That we may only speak of potentials in Spinoza in a qualified sense also follows from another aspect: the indeterminacy or openness with regard to alternative development without which we cannot sensibly assume potentials for development. In view of the essential necessity of the whole (God or world), there is no room for potentials in this regard. The individual objects too are part of an inevitable causal chain which determines their existence and activity (according to I p. 28). Viewed as a whole, these causal chains surely create inevitable determination. The causal necessity of all individual objects and the essential necessity of the all-encompassing substance are merely two interpretations of the same thing (otherwise, Spinoza’s metaphysics would be inconsistent). Accordingly, we find in the definition of impossibility, in the framework of the causal modalities, a typically necessitarian reduction of the modalities to factuality, in this case to the never-being-real: Something is impossible if there is no actual external cause which is determined to produce it (nulla causa externa datur ad talem rem producendam determinata, I p. 33 sch. 1). There are not, therefore, two areas of modality such that in one of them, namely in the area of causal modalities, true potentials occur. Rather, it is only one of two interpretations under which we may sensibly speak of potentialities. For the causal view of individual objects (res singulares) allows us (according to IV def. 4) to confine ourselves to their individual cause (dum ad causas, ex quibus produci debent, attendo) and not to consider the entire scheme of

23

The foundation for this parallelity of the development of bodily and mental potency is already laid in the definition of the affects (III def. 3). See Della Rocca (2008), in particular 29–31.

Potentiality in Rationalism

169

determination (which creates inevitable necessity). From this abstracting view, which limits itself to a part of reality which is accessible for human cognitive capacities, there arises a contingent possibility (possibiles) which includes indetermination. Of course, this is not a real indetermination, but only an epistemic one which amounts to our not being certain whether these causes are determined to their individual effects (nescimus an ipsae determinatae sint ad easdem producendum) or whether these effects are prevented by the universal course of determination. All this shows: Even if we may, within the Spinozean framework, legitimately and sensibly speak of capacities, Spinoza is not in a position to recognise dispositions or potentials in the full sense of the words. For then they would need to count as irreducible modal properties of (real) individuals with an ontological status different from the status of being real. While the latter does not rule out that dispositions or potentials have their basis in the real constitution of their subject, it forbids reducing them to their categorical basis. Such a reduction, however, is suggested by the passage from II p. 13 schol. already discussed. In this passage, potencies are not treated as irreducibly given modalities, but are the result of the actual causal relations between real individuals. Thus the active power consists in whether something is capable of simultaneously producing several things or to bring about an event on its own etc. Capacities differing in strength, therefore, are conceived of as causal roles with a different range. This interpretation directly follows from the basic metaphysical assumption that individual beings are unautonomous modi. What an individual is, what properties and capacities it possesses, is, according to this view, a function of the mutual causal influences, ultimately in the entire universe. Dispositions and potentials in an unqualified ontological sense can only be recognised by one who conceives of individual beings as a plurality of autonomous individual substances in the Aristotelian sense. Dispositional properties or potentials, on this view, have their basis in the inside, that is, the essence of the thing. In their own right, however, they are accidental and hence alterable, which means that they may or may not be realised and develop thus or otherwise under certain external influences. This kind of tension between their being grounded in a constant inner nature on the one hand, which makes them something actual, and the contingency of their realisation being subject to random external factors on the other, which makes up their modal nature, is crucial for dispositions and potentials in a substantial sense of the words. Yet, this tension is only possible in the case of individual substances inside the world which are constituted by their inner nature. It can neither occur in the case of unautonomous moments within the causal scheme nor in the case of the all-encompassing total reality or divine substance (in its essential necessity), which cannot develop towards a alternative state outside of itself. On Spinoza’s view, therefore, modal properties (such as dispositions or potentials) by no means need to be mere fictions. However, he can attribute to them only a derived, secondary reality. The reason is that he does not recognise them in those features which would ensure an ontological status which, though different from the status of being actual, is still real, but that he reduces them to their

170

M.-T. Liske

causal roles within the actual course of determination.24 Only such a reductionist talk of modalities (such as dispositions and potentials) is compatible with the basic assumption of a total reality to which there is no alternative. This essential necessity of the whole does not allow for a modal differentiation of the real ontological status.

3.3

Potentia: Present Control or Potential for Future Development?

In Aristotle, a capacity (dt́mali1) is essentially directed towards what will be the case in the future (cael. I 12, 283b 13f.). As a state which precedes development and is different from it it can be the potential for alternative developments all of which remain unrealised except for the one that is realised. In Spinoza, in contrast, potentia is an active power which is inherent in the development itself and determines it in accordance with a law. Since such an immanent active power is virtually the same as the produced event in which it unfolds itself, it coincides with the actual development and hence allows for no alternative developments—as should be the case with something we may describe as a potential. Something similar is shown by Spinoza’s notion of conatus,25 which is a notion we might expect to come closest to that of a potential. For it does not describe a striving peculiar to living beings who are endowed with consciousness or at least sensation, but a kind of tendency we may find in any being whatsoever:”The tendency (conatus) by which each thing tends to persist in its being (in suo esse perseverare conatur) is nothing but the actual essence of the thing (nihil praeter ipsius rei actualem essentiam).” (III p. 7). Again, Spinoza reduces conatus as a kind of potency, that is, as a modal property, to something actual. Conatus is not a forward-looking striving to develop into something new, but is aligned to the preservation of the status quo. Instead of using ‘conatus’ as a term for a potential for development, Spinoza extends the principle of inertia, by not confining it to the merely physical preservation of a state of motion, but also ontologically relating it to a complex body’s tendency to preserve the constitution characteristic to its kind and thereby to conserve its existence. Conatus coincides with actual essence since it directs itself to the continuance of the current state determined by the essence.

24

This view of the position of modalia can be compared with the opposition of a reductive physicalism, which recognises the mental as far as it is reducible to the physical, and an eliminativism, which wants to rule it out as fictitious. A similar view is developed by Renz (2009) with regard to mental dispositions. 25 On conatus, see, for example, Rotenstreich (1977), Garrett (2002) and Cook (2006). According to Hassing (1980), the notion of conatus central to Spinoza’s determinism cannot be derived from physics.

Potentiality in Rationalism

171

Even where Spinoza speaks of the power of mind over the affects (Mentis in affectus potentia, V p. 20 sch., G II 293L. 6,17f.), as in the doctrine of the affects26 of Eth. III-V, ‘potentia’ does not refer to a forward-looking capacity, to a potential which in the future realises itself in the goal of a development.27 Rather, Spinoza thinks of the control which is to be exercised over the current situation.28 This control is solely constituted by the cognition (sola cognitione, L. 25f.), i.e. the mind permeates the given situation in adequate ideas and is thereby in a position to control it (and also the affects). ‘Impotentia’, in turn, does not so much refer to the incapacity to do something in the future or to develop toward the better. Rather, it refers to the weakness of a mind which is incapable of controlling the affects because it is not mentally in command of the situation (cognitionis privatione, L. 26f.) and hence succumbs to the influence of the affects because of inadequate ideas (27f.), that is, is itself dominated and suffers an influence (pati, 28f., first meaning of ‘passio’) or is determined by the affects or passions (second meaning of ‘passio’). On the power (potentia) of the mind, in the sense of a control over the affects (imperium in affectus, G II 277, L. 17f.), that is, the topic of Eth. V, Spinoza also provides important comments in the preface to V (Praefatio). There, he distances himself from the voluntarist conception of Descartes (G II 278, L. 3f.) which he traces back to the Stoics: The affects absolutely depend on our will (a nostra voluntate absolute pendere)29 and by our will, we can absolutely control them (iis absolute imperare posse) (277, L. 20–22). This means: Simply by truly wanting it, we can replace undesired affects by desired ones. By dismissing such resolutions of the will operating in a causally unconditional manner, which resolutions would represent a kind of creatio ex nihilo, Spinoza renews Aristotle’s much more realistic psychological insight: Since our affects depend on our character, we can only influence them in the long term by forming our character, i.e., by repeated action we can dispose ourselves toward the desired way of feeling and acting on these feelings. Accordingly, Spinoza emphasises that a considerable amount of practise and

26

On Spinoza’s theory of the affects and the emotions, see Renz (2007, 2008), where also the importance of conatus for the doctrine of the affects is pointed out. 27 Spinoza does not acknowledge any potentia for development. Still, the power of acting (potentia agendi) is subject to a development. There is a transition to greater or lesser perfection depending on whether the power of acting is increased or decreased, which results in joy or sadness (III af. def. 3 ex.) See Look (2007), in particular 23–26. 28 Spinoza also discusses the aspect of control as a remedy (remedia) to the affects (G II 293, L. 4); see de Dijn (2006). Since in the passages here discussed, Spinoza consistently deals with the control over the affects in terms of ‘potentia’ the terminological differentiation of Barbone (2002) 102–104 appears exaggerated: ‘Potentia’ is said to describe an immanent constitutive capacity of the individual, its inner nature, tendency (conatus), or force; ‘potestas’ in contrast is said to relationally describe the control or authority over something or somebody (as a function of potentia). 29 On Spinoza’s discussion of Descartes with regard to the relation of intellect and will, see Lloyd (1990).

172

M.-T. Liske

effort is needed in order to bridle and moderate the affects.30 In considering the question of how the mind can gain power or control over the affects, Spinoza again embeds the discussion of potentia in his conception of the causal necessity of the individuals, which in II p. 48, he explicitly relates to the act of willing. This necessity stands in conflict with the assumption that, by way of a spontaneous resolution of the will, we can immediately move our affects into a desired direction. Rather, for Spinoza, the regulation of the affects through the mind must be integrated into the complex causal scheme of the manifold mental elements. Something similar can surely be said, in Spinoza’s sense, about the unfolding of our mental potentials: We cannot immediately controll it by the resolution of the will, but can only indirectly influence it, being aware of how the different mental elements mutually depend on each other.

3.4 3.4.1

Potentials and Teleology Are There Potentials for a Goal-Directed Development into Real Perfections?

A potential may count as a disposition for a positive development towards a new perfection and hence be conceived of as a state underlying a teleological development: a state from which a goal-directed process emerges. The question of whether one can recognise potentials within the framework of Spinoza’s metaphysics, therefore, is closely connected to Spinoza’s position regarding teleology,31 as it is developed in the appendix to Eth. I. As is well-known, Spinoza claims there: Human beings legitimately assume goal-directedness with regard to their own actions, but illegitimately transfer it to nature and thus judge: The natural objects in their entirety act—as humans themselves do—for the sake of some goal (omnes res naturales ut ipsos propter finem agere) and are aligned by God to a specific purpose (I Ap., G II 78, L 3f.). For two reasons in particular, Spinoza rejects such a goal-directedness of the whole reality. One reason is the Neoplatonic doctrine of emanation: “[…] that effect is most perfect which is produced immediately by God, and the more something requires several intermediate causes to produce it, the more imperfect it is.” (80, L. 16–18) The world is understood here as a fall from God. What immediately proceeds from God has comparatively the highest degree of perfection; the more something departs from its cause in the causal chain, i.e., the “[…] usum et studium non parvum requiri ad eosdem coercendum et moderandum.” (G II 277, L. 23f.) 31 The provocative claim of Bennett (1983) and (1984) Chap. 9 213–230 that Spinoza turns against any form of teleology or teleological explanation has given rise to an intensive debate about whether Spinoza at least recognises a teleology in a qualified sense: Rice (1985) and Curley (1990) and the response of Bennett (1990) and Garrett (1999) (critique of Bennetts 314–323), Manning (2002). 30

Potentiality in Rationalism

173

more intermediate stages are required for its emergence, the more it has also departed from divine perfection. Teleology, in contrast, is by its very notion a process towards something more perfect. As already Aristotle holds (NE I 1, 1094a 1–4), the nature of a desired goal is constituted by its being good. A state, therefore, qualifies as a goal only if it includes a higher degree of perfection than those states which lead up to it. Furthermore, the possibility of God acting for a goal is to be ruled out, because the inevitable consequence of this would be that God strives for something which He lacks (appetit, quo caret) (80, L. 22f.). Therefore, whoever accepts the presupposition that God in His absolute perfection cannot lack anything (as does the entire theological tradition), can assume a goal-directedness of the world only by distinguishing it from a transcendent God. Yet, it is this very assumption which Spinoza denies. Whether one has to assume objectively goal-directed powers, then, is intimately related to whether an objective value judgement is possible on the basis of ontologically founded degrees of perfection. Can, for instance, the perfection of an individual be determined with the help of the objective criterion of to what extent it satisfies the norm of its kind as the result of a goal-directed organic development? Since, for Spinoza, everything happens by the necessity of nature, one cannot allow deficits of nature in the sense that an object lacks what it should by nature possess (non quod ipsis aliquid, quod suum sit, deficiat vel quod Natura peccaverit, IV praef., G II 208 L. 3f.), that is, in the sense that an object fails to comply with an objective norm although the compliance with this norm is in principle possible and hence to be postulated. Perfection and imperfection are merely subjective modi of thought (modi cogitandi, 207, L. 18f.), rational notions which arise from the fact that we conceive individuals as members of a species or genus. We do so in order to have a criterion for comparison in the common nature of the species or genus and to grasp, in virtue of the aspect of consideration thus brought to bear, more or less positive content (plus entitatis seu realitatis), that is, perfection, or, in turn, to ascertain something negative, such as limitation and hence imperfection. Perfection and imperfection is only the way in which something affects our mind (mentem afficiunt), that is, appears to us (on the whole, see G II 207, L. 18–208, L. 7). In III af. generalis def. ex. (G II 204, L. 13–17) Spinoza talks of the power of existence (existendi vis), that is, ontological potentials, instead of perfection. Their degree does not arise from the imagined object by objectively comparing its material constitution at different moments in time: rather, it arises from how much positive objective content (realitas) we include in the idea under which we consider it.32 However, without objective perfection there can be no true potential as a goal-directed capacity to develop into a more perfect state.

32

The affectuum generalis definitio itself has to be interpreted accordingly. That an affect or passion is a confused idea is due, not so much to the power of existence of the body represented in the idea, its increase or decrease, but rather to the mode of imagination. See Benardete (1990) 212.

174

3.4.2

M.-T. Liske

Can the Goal-Directedness of Human Action Be Transferred to All Individual Beings?

So far, Spinoza rejects, the goal-directedness of the world as a whole, especially in the form of divine ends for the universe. This fits in with the fact that, according to Spinoza’s view of an essential necessity of the whole, there can be no potentials. This does not necessarily imply, though, that he must reject any kind of teleology in nature. Thus he could very well recognise teleological explanations of individual events or individual beings, provided they are compatible with the throughout causal determination which he assumes with regard to the individual events or beings. In line with this, he does not seem to have any reservations against the most common form of a teleological explanation, namely that of human behaviour, and he himself uses this form of explanation in his ethical passages. Yet, he also shows in IV praef. (G II 207, L. 2–12) in how far such teleological explanations of human behaviour can be reduced to causal explanations. Two objections against the possibility of such a reduction are implied here that he tries to dispel. (1) The final cause lies in the future, the efficient cause, in contrast, must temporally precede the event.33 (2) The true end is an ultimate end which is pursued for its own sake only. The final cause is therefore the prime cause (causa primaria) and cannot be reduced to anything higher. Efficient causes, in contrast, are, especially to the determinist, nothing but moments within a scheme of determination. The first objection can plausibly be rebutted: The event functioning as a final cause is imagined (imaginatus est), that is, anticipated in the mind. This imagination, as far as its content is perfection, produces an appetite (appetitus). This appetite, however, is the cause preceding the human action and is therefore an efficient cause. Since we are not aware of the psychological preconditions of our appetite, we think of it as operating originally or spontaneously (L. 11f.), even though it is integrated into a causal scheme. This resolves the second difficulty, the problem of how something standing within a causal relation can appear as an original cause. In order to provide this explanation of finality compatible with causal determinism, the notion of appetite is crucial. An appetite can intentionally direct itself towards a goal to be achieved in the future and thereby anticipate this goal. In this way, the desired goal can become the temporally preceding cause of the action motivated by this appetite. Now, in III p. 6 und p. 7, Spinoza attributes to every object a striving or tending (conatus), in virtue of which it tends to remain in its being, as far as this is up to the object itself (conatus, quo unaquaeque res in suo esse perseverare conatur). Can, in virtue of this tending, finality be transferred to every individual natural being? Appetite (appetitus) is a particular form of tending (conatus). According to III p. 9 sch., it is one that concerns mind and body together and hence presupposes consciousness or at least imagination. Now, with regard to teleology imagination is crucial because only in imaginations the desired good can be anticipated. Employing a thorough terminology, we should, therefore, describe

For this objection that final causes invert the natural order, see I Ap. (G II 80, L. 11).

33

Potentiality in Rationalism

175

the conatus which is constitutive for all beings only as a tendency and confine appetite and with it (as we shall see) goal-directedness in Spinoza’s sense to beings capable of imagination, men and animals.34 Thus, in Renati des Cartes Principiorum philosophiae pars I et II more geometrico demonstratae III def. 3, Spinoza explicitly conceives of the tendency to motion (conatum ad motum) as something which does not include thinking (cogitatio), but merely means that a material particle is by itself disposed to a certain motion (ad motum inicitata) in virtue of its position and will therefore actually move into the direction in question unless it is hindered by something else. Here Spinoza obviously recognises dispositions or tendencies (namely, the disposition to out of itself behave in a certain way in accordance with the laws of nature in the absence of hindering external factors), which are not to be interpreted teleologically and hence not to be viewed as potentials. For these tendencies or powers are not directed at any development going beyond the actual state of the object, which would be a central aspect of teleology and potentiality. This not-going-beyond the status quo emerges from two aspects of III p. 7: The tendency is a tendency to remain in being and it is nothing other than the actual essence of the object. In Cog. Met. I 6 (G I 248), Spinoza already equates the object itself with its tendency for self-preservation (conatus … ad suum esse conservandum) and argues that the two are distinguished only conceptually or verbally, but not with regard to reality (L. 4-8). Moreover, he explains this equation with the help of the identity of motion with the power of remaining within the state of motion (vis in suo statu perseverandi) (L. 9–18): As the phenomenon of motion simply consists in the fact that a moving body remains in the state of rest or of uniform motion according to the physical law of inertia—unless external forces act upon it—an object consists in (or its nature includes) subsisting in its particular way of being, so that (according to III p. 4) it can only be destroyed by an external cause, that is, by external powers that are more powerful than the object’s own conatus.35 Since these tendencies or dispositions towards preservation merely imply that something does not alter its (essential) state when it is left to itself, they do not presuppose any goal-directed powers striving for a higher perfection, let alone the awareness of goals in all beings. Therefore, they do not represent potentials for further development. In III p. 7 dem., however, the potency or tendency (potentia sive conatus) of an object to by itself or in conjunction with others effect something is generally explicated (hoc est) as potency for self-preservation. According to this, all other potencies to effect something are merely differentiations of this basic potency or basic tendency to remain in being which is identical with the actually given essence (data sive actualis essentia) and hence given with the object itself. Thus, there generally are no potencies which would go beyond the tendency of a being, 34

However, Garrett (2008), in particular 19–25, tries to demonstrate: Mental attributes such as imagination, power of thinking, representional content, consciousness are associated, for Spinoza, with self-preserving behaviour, that is, the conatus, and appear therefore in a rudimentary form in all beings. 35 See Oksenberg Rorty (1990).

176

M.-T. Liske

grounded in its essence, to preserve its existence. In the strict ontological sense, therefore, there is no place for potentials towards a development to something new and higher which would be leading the object beyond its actual state. Even if we encounter tendencies in all objects, there need not be goal-directed powers in all of nature.

4 Leibniz36 4.1

Leibniz’s Notion of Force and the Three Stages of Capacity

Not only does Leibniz’s metaphysics of individual substances provide the ontological basis which in principle allows for talk of potentials in an unrestricted sense of the word. There is even a great affinity of certain philosophical tenets, in particular of the later Leibniz, to the notion of potentiality, even more so than to the notion of a disposition.37 A disposition is not normally realised out of itself, but needs to be prompted from the outside by favourable conditions. In ‘potential’, in contrast, the meaning of ‘potentia’ as ‘power’, ‘might’ is virulent. Therefore, the aspect of self-activeness is more pronounced. Something has power and force in itself to develop into something new unless something external hinders it (and not just, as in Spinoza, to remain in its present state and to thereby preserve its existence). Through the notion of force Leibniz revises the active capacity (potentia activa) of the scholastics.38 This active capacity too is not (as the innate faculty for language) an undeveloped disposition which first needs to be unfolded in a sometimes long-lasting process in order to enable its subject to exercise the activity in question (first or lowest stage of a capacity). Rather, as an already developed capacity it disposes its subject to instantly become active under given circumstances (second stage). It is the proximate or immediate possibility of acting (propinqua agendi possibilitas),39 in so far as it is fully developed for this activity and therefore satisfies all presupppositions by itself. Still, this resting state of being able (facultas) is not activated out of itself. Rather, the capacity needs an external momentum as a stimulus, as it were, which excites its activity (aliena excitatione et velut stimulo indiget), usually a situation which makes the use of the capacity required or desirable. For instance, the exercise of the developed capacity to speak a

36

A more comprehensive version of this chapter can be found in Liske (2009b). On dispositions in Leibniz, see Liske (2009a). 38 See Gaudemar (1992). 39 These and the other phrases quoted in this section can be found in: De primae philosophiae Emendatione et de Notione Substantiae (PPE) (GP IV 469), on which the present discussion is based. 37

Potentiality in Rationalism

177

foreign language usually has to be triggered by a situation in which its possessor needs it. This contradicts Leibniz’s conception of a self-sufficient substance which spontaneously produces from its own inner self the complete succession of its states, which, for Leibniz, are its perceptions. Therefore Leibniz not only assumes a developed capacity (second stage), but a capacity that is enhanced by a striving (third stage). Even though we do not find these exact terms in Leibniz, in effect, he distinguishes in the Nouveaux essais (NE) II 1 § 2 (A VI 6, 110) three stages of modality. The lowest stage is repesented by the mere potencies of the scholastics (les pures puissances). Such a potency is conceived of as the principle of matter that is by itself undetermined and therefore differently determinable, that is, open for becoming this or that. The mere natural disposition to speak, for example, can be related to any language. For Leibniz’s deterministic approach, an undefined, not yet determined generality of this kind is the product of abstract thought, a mere fiction. In reality, we only find developed and hence necessarily specified dispositions to one particular activity rather than another (second stage), such as the capacity to practice one particular language. Beyond the disposition, there must be a tendency or a striving towards activity (outre la disposition il y a une tendance à l’action) (third stage). Since a striving is directed at a specific goal, the capacities that each encompasses such a striving need to be specialised. Furthermore, these tendencies can never be completely ineffective (ces tendances ne sont jamais sans quelque effet). For their very function is the transition to activity. Without any activity, they would be devoid of all function and hence futile. There are then in reality no mere potencies which are not directed towards any activity at all. These two postulates lead to a more significant notion of potential. In order for such a notion to be a fruitful one, one must assume potentials for quite specific developments towards precisely assignable goals. If, further, a potential is to be verifiable, in principle at least, it must not be a mere disposition, but needs to already manifest itself in activity: not in the one which it has as its goal, but at least in the sense that the development towards this goal has already begun. In this way, Leibniz blurs the distinction between an instantly realisable disposition and a potential for a development. This is very well justified by the matter in question. If a capacity can instantly manifest itself, we would speak of a disposition even if the manifestation includes a certain temporal process. Such a disposition therefore continually merges into a potential whose realisation can instantly set off, even though the complete manifestation is accomplished only after a long-lasting development. Leibniz was searching for a way in which this capacity, required by the metaphysical approach, may be described: a capacity which is able, because of the striving it entails, to transform itself into the activity which is the goal of the striving. Here the physical notion of force offers itself to Leibniz. In ‘potentia’, the idea of a capacity enhanced by a striving is already suggested, given the further meaning of ‘strength’ and ‘might’. However, since this term traditionally refers to a mere albeit developed capacity, Leibniz extends it with the help of the notion of force to the overall idea: potency or force (potentia vel vis, GP II 170; force ou puissance, A II 2, 272, GP III 48). From the outset, a capacity is a force which is

178

M.-T. Liske

directed, in virtue of an internal striving, towards a specific realisation. Now, the force is capable of accomplishing the specific achievement of a potential, namely to initiate a development towards a new state. What makes this development explicable is that the force includes a striving (conatus, nisus) or tendency towards a new state.

4.2

Force as Mediation Between a Capacity and Activity

Since a striving is something given in a present state and directed at some future state, it can explain the transition from one state to another. In virtue of this integrated striving the force causes the transition from the state of merely being capable to act to the state of actually acting and, in this sense, is intermediate between the two states: ”But an active force (vis activa) contains a certain act or ἐmsekέveia and is midway between the capacity of acting (facultas agendi) and the action itself and involves a striving (conatus). It is thus carried into action by itself (per se ipsam in operationem fertur) and needs no assistance but only the removal of an impediment.” (PPE, GP IV 469) The notion of force is also sympathetic to Leibniz’s fundamental concerns in that it mediates between qualitative opposites, bridges the gap between them by intermediate grades or even makes them continually merge into one another. The notion of force achieves this kind of mediation (in addition to what is said in the above quote) in at least four areas. It mediates between (1.) a mere capacity and activity, (2.) physical and metaphysical perspective, (3.) activeness and passiveness (and the underlying capacities), (4.) efficient and final causes. These four achievements will provide the structure for the constructive part of our discussion of Leibniz in Sects. 4.2–4.5. Let us turn to the passage quoted above which deals with an intermediate stage between capacity and activity. This stage consists of a goal-directed principle, the entelechy. Even though this is a metaphysical principle of action, the notion of conatus suggests that what Leibniz has in mind is also the motion of bodies in the physical sense. Thus, one important meaning of ‘conatus’ is the initiation of motion which the striving to motion represents in the infinitely short period of time in which motion takes its start. In such an infinitesimale time there can be no fully developed activity, which, according to Leibniz, must be measured by way of the effect it has brought about. In an infinitely short period of time, however, there can be no effect of a verifiable, measurable size. Yet, it is not a case of rest, either, but an intermediate state between rest and motion, from which the full motion or activity emerges and through which it can be explained.40

40

For Leibniz, innate ideas represent another important example of an intermediate stage between mere possibility and actual (intellectual) activity. See Poser (2006).

Potentiality in Rationalism

179

In accordance with this, Leibniz in his treatise on dynamics of 1702 characterizes his extended notion of potency by remarking that it involves actualisation (actum involvit). For Leibniz, this involved actualisation is not confined, as for the Aristotelian tradition, to the actualisation’s first stage (actus primus), in the form of a developed, but resting capacity, but means activity (actio). A capacity includes activity in that it strives for this activity or is directed towards it (ad actionem ad quam tendit). Leibniz is aware of the fact that the capacity does not necessarily fully proceed to the activity it aims at (non semper integre procedat ad actionem), but he thinks that it always does so to some extent.41 The striving, then, is never completely ineffective. It does lead to its goal, the fully developed activity, only if it does not meet obstacles. However, even when there are such obstacles, it leads to some kind of rudimentary activity. This can be taken to primarily refer to an active disposition to instantly unfold itself. Yet, it can also be understood as the potential to develop into a new perfection. If the potency always involves a striving, then it is always activated. However, according to our modern conception, a body can still remain completely inactive. This Leibniz rejects as impossible since the substance is essentially active: “This power of acting (agendi virtus), I say, inheres in every substance and some action always arises from it, and thus even the corporeal substance does not, any more than spiritual substance, ever cease to act (ab agendo cessare umquam).” (PPE, GP IV 470) The true substance, on Leibniz’s view, is spiritual. Yet, a metaphysical reality inheres in the motions of the body as well because of the force causing them —as becomes obvious if one looks for their ultimate cause (ultima ratio motus, 469). Force, however, constitutes, from a metaphysical point of view, substance. Thus, Leibniz takes corporeal substances as subjects of bodily motion. From Leibniz’s fundamental characteristic of a substance, i.e. its autonomy, incessant activity of the substance follows—either as the causal operation of bodies or (mentally) as imagination. For just as a substance cannot externally be bestowed with activity or with the active force which constitutes it, it cannot wholly be deprived of them. Rather, it only externally receives from the collision of bodies (conflictu corporum) the limitation and precise determination of the pre-existing active force inherent in itself and of its striving.42 Activity can only be hindered from the outside, not prevented. By making the assumption that even in a state of apparent rest there is some kind of inhibited or reduced motion or activity, Leibniz can explain how a seemingly resting potential can by itself produce a detectable development under favourable conditions, namely when it is not being hindered from the outside. Since Leibniz views the activity as the fundamental feature (caractère) of the substance, he conceives of force, that is, the principle of such activity, as the constitutive principle of the substance (not merely in the weaker

“Talis enim potentia actum involvit neque in facultate nuda persistit, etsi non semper integre procedat ad actionem ad quam tendit, quoties scilicet objictur impedimentum.” (GP IV 395) 42 “[…] substantiam creatam ab alia substantia creata non ipsam vim agendi, sed praeexistentis jam nisus sui, sive virtutis agendi, limites tantummodo ac determinationem accipere.” (GP IV 470) 41

180

M.-T. Liske

sense that it can produce activity under given circumstances, but in the sense that it always produces activity).43 The assumption of substance as being essentially and hence always active stands in contradiction with the phenomenon that bodies sometimes seem to be fully at rest. In order to make this apparent contradiction disappear, Leibniz draws a distinction within the notion of living force (vis viva), i.e. kinetic energy (Specimen dynamicum (SD) p. 14, L. 182–194,44 GM VI 238f.): The vis respectiva sive propria is peculiar (propria) to the different parts of an aggregate and refers to the inner motion of the parts in relation to one another (respectiva), that they affect one another and thus change their position towards one another. In contrast, the vis directiva seu communis refers to the whole (communis) and makes it operate in a directed manner on the outside. The activity a body strives at can be hindered to the extent that a body shows no verifiable external effects and thus appears devoid of all motion. However, it always possesses an activity which confines itself to an internal motion and can at any point manifest itself in detectable outward motion. In this way, it can be shown how dispositional properties such as elasticity45 (the disposition of deformed bodies to return to their original shape) are in fact no dormant dispositions at all. In accordance with the appearance of motionlessness, Leibniz too speaks of a dead force (vis mortua), which we would nowadays describe as potential energy (SD p. 12, L. 171–173, GM VI 238). Strictly speaking, however, the motion has retreated to the inside. When an elastic object A is being deformed or when another body B collides with A (see GP IV 397), the inner motion of A is disturbed. When this inner motion is restored, the external effect typical of elasticity occurs: the restoration of A’s original shape or the rebounding of B. Accordingly, a potential can be conceived of as a seemingly dormant capacity, which turns out to be a non-observable activity retreated to the inside which under favourable conditions changes into a manifest development. In virtue of the permanent inner activity, a substance is always ready to act outwardly. What is needed is simply a trigger (like the spark that sets off the gunpowder, see GP IV 399) in order for a manifest activity to come about. Leibniz’s discussion of the active capacity conceived of in terms of force not so much refers to potentials for the future acquisition of properties, but rather to dispositions to instantly realisable action. In Théodicée (Théod.) I § 46, he emphasizes against the scotistic-molinistic liberty of indifference: There must always be, according to the beliefs of the philosophers, a disposition to action (dispositon à l’action) which can also rest on the previous inner constitution (constitution) of the agent and which contains a predetermination (prédétermination) to a specific type of action. And beyond that even, a development towards a new property needs to be predetermined by the inner constitution of a subject which we may view as its potential.

“C’est pourquoi je la [sc.la force] considère comme le constitutif de la substance étant le principe de l’action, qui en est la caractère.” (Système noveau (SN), Draft, GP IV 472). 44 W. Leibniz: Specimen Dynamicum, ed. by H.G. Dosch/G.W. Most/E. Rudolph, Hamburg 1982. 45 See Fichant (1995), in particular pp. 78–80. 43

Potentiality in Rationalism

4.3

181

The Reconciliation of the Physical and the Metaphysical Perspective

Striving is, for Leibniz, a mental characterisation. Thus, Leibniz attributes the primitive forces to the true mental substances which he all takes to be capable of perceptions (if often to quite a low degree) as well as of a striving for clearer and more distinct perceptions. However, force is first and foremost a physical notion.46 Leibniz came across this notion through a controversy within mechanistic physics relating to the question of which was the decisive physical quantity to which the laws of conservation refer (Brevis Demonstratio A VI 4, 2027–2030; Discours de métaphysique (DM) § 17). The quantum of motion, which according to Descartes and his followers is conserved, is what actually happens in motion, i.e. that mass of a certain quantity moves with a certain velocity. In contrast, Leibniz argues that the decisive quantity is the force as the cause of the motion which is measured according to the effect it manages to bring about, e.g. which mass it can lift to what height. In turn, the living force a body assumes during fall,47 is measured according to the height from which a body of a certain mass has fallen. According to Galileo’s laws of gravitation, velocity does not increase linearly, but in squared proportion to the height of fall. Only if we take the force, conceived of as the product of mass and squared velocity (mv2), and not the quantum of motion (mv) as that which is being conserved in motion, we arrive at the true laws of motion—according to Leibniz. Metaphysical considerations certainly play a role here, such as the idea that we can only justifiably speak of real, non-relative motion if we relate it to the underlying force as its cause. Yet, the kind of force discussed in this controversy is undoubtedly a physical quantity. How can it then include a mental characterisation such as striving? In the conatus, there is also physically a kind of striving. Its function is to provide for the transition of the state of one moment to that of a future moment and to thereby make it explicable. Although, in a sense, the conatus is constantly efficient in each moment of either the actual or the still inhibited motion, it is something that can already be given in a single moment in time. In contrast, kinetic

46

Garber (2009) 99–179 assumes, in accordance with his view of Leibniz’s development, that force is first to be understood in a physical sense with regard to bodies. On how an originally mechanistic notion of force turned into a metaphysical one, see Boudri (2002) 74–99. However, even the primitive forces are not metaphysical in any strong sense of the word. As the basis of the regularities in natural phenomena, they form an intermediate stage between mechanistic explanations of individual phenomena and the metaphysical theory of origin. See Allen (1984) and Gale (1970, 1984). Here physics and metaphysics mutually influence each other. See Gale (1988) and Hassing (2003). 47 In the vis-viva-controversy, two incompatible but equally legitimate interpretations oppose each other, since as a matter of fact, both the quantum of motion (mv) and the force (mv2) are preserved. Since Leibniz wanted the dynamics to constitute a new science of the forces of motion, he preferred the force being preserved. See for instance, Gale (1973), Freudenthal (2002) and Shimony (2010).

182

M.-T. Liske

energy is acquired only in the course of motion. Since striving causally precedes motion, the conatus in a significant sense is given in a moment before motion begins. A heavy object which, due to gravitation, strives towards the ground is held by a rope, it tightens the rope, but cannot begin its motion. Or a drawn bow strives, in virtue of the elastic force, to retrieve its original shape and to thus let off the arrow, but is unable to do so as long as it is being held by the shooter. For Leibniz, these specific forms of force can mechanically be explained by the motion of the particles of the air (PPE, GP IV 469). While the scholastics postulated capacities which were specifically directed towards the particular kinds of effect to be explained, a good explanation needs to start from broadly recognised, fundamental and general facts which can explain more than just these particular effects. This is true of the mechanistic explanation which explains all natural phenomena in quantitative terms of extension and its modifications size, shape and motion (GP VII 338). However, if we want to give a complete and appropriate explanation of those phenomena, we need notions which allow for a transition to the metaphysical perspective—such as the force without which one cannot speak of true movement, or as the conatus as a still inhibited striving to move. A bridge to metaphysics is also built by the use of ‘conatus’ with regard to non-inhibited motion. Here too conatus is distinguished from the actual or ‘living’ execution of motion (vis viva), conatus being the infinitesimal initiation of motion in the first infinitesimal moment of movement (conatus primus, GP II 154). Just as the duration of the conatus, the movement performed due to it approximates the boundary value zero. Just as the conatus, the derivative force (vis derivativa) underlying physical motion is something instantaneous, namely the current state in so far as it strives towards the following state and thus includes it beforehand, i.e. brings about the transition to it (ipse status praesens dum tendit ad sequentem seu sequentem praeinvolvit, GP II 262). Now, since, according to Leibniz, all that is instantaneous or temporary must be grounded in something permanent, he arrives at the idea of a primitive metaphysical force which determines the entire succession of states and thus represents something permanent, the law of succession (lex seriei, GP II 171 and 262). It is not merely a momentary striving towards the next state, but the permanent fundamental direction or tendency of the striving, and does not govern the physical succession of motion, but the mental succession of perceptions. But primitive forces (vires primitivas), can—as I think is obvious—be nothing but inner tendencies (tendentias internas)48 of simple substances, in virtue of which these substances, according to a definite law of their nature, proceed from perception to perception while agreeing with one another as they reflect the same phenomena of the universe from different perspectives. (GP II 275)

Being an overall tendency, the primitive force confers a characteristic individual imprint to the entire succession of perceptions and determines it in accordance with

48

Thanks to internal tendencies, both the continuity of motion and the conservation of the substance are guaranteed by the object itself without any need for a Cartesian creatio continua. See Quintana (2006) 26.

Potentiality in Rationalism

183

a law not of the general, but of the own individual nature (certa suae naturae lege). In this way, the same phenomena of the universe are represented by these perceptions in a different way (diverso habitu) which is peculiar to the respective individual. For this reason, the primitive force can be conceived of as the potential for an individual development which elicits and regulates the development of the mental states in a manner characteristic for this individual and thus constitutes the substance in its individual nature. The derivative force, as the tendency to a particular motion, is viewed by Leibniz (for example, GP IV 396) as a modification, i.e., as a temporary particular specification and manifestation of the fundamental striving in its permanent law-like directedness.49 In this way, only the primitive force and not the derivative force which is already given in a moment, can be understood as a potential. For the development to which a potential leads represents a succession of phases. However, the principle of such a successively unfolding activity (actionis principium) or the law of such a succession (series) is the primitive force. Moreover, a potential is a (relatively) permanent property of a substance. This is true of the primitive force which Leibniz identifies with the substantial form. With the help of the physical and at the same time metaphysical notion of force Leibniz thus tries to give empirical content to substantial form, a principle postulated by the scholastics that constitutes something as a substance of a particular kind.50 As a principle constituting substance, the primitive force is necessarily conserved in the individual. In contrast, the derivative force as something which is momentary and accidental is not identically conserved in the individual body, but only in the overall system, such as when it is conferred to another body in a collision; in this sense, it is merely a phenomenon.51 The notion of force is, due to its origin in physics (regarding the motion of bodies), intimately related to the idea of a law. Leibniz’s account of a body’s continuing its motion according to its inner forces is that it performs its changes due to the laws which God planted in its nature and which He preserves there.52 A behaviour which surpasses the inner forces, in contrast, is a miracle. In order to be subject to laws, however, a behaviour not only needs to be regular, but also to be explicable by general, fundamental principles. For example, if a body which has been brought into a circular motion by a sling were to continue this circular motion on release, this could be explained (with the scholastics) by a capacity for circular motion that is inherent in the body. However, such an ad hoc explanation in terms of a force specifically directed at this very effect is not a genuine explanation in terms of laws. Behaviour that is not explicable in terms of laws, even if it happens

49

See Phemister (2005) 194–202. Since activity is what essentially constitutes a substance, the force as the inner principle and law of activity is constitutive for the substance. See Nachtomy (2007) 125–131. 51 On whether the derivative physical forces are real or imaginary, see Gueroult (1967) 190–199. 52 “[…] la substance corporelle a la force de continuer ses changements suivant les lois que Dieu a mis dans sa nature et qu’il y conserve.” (To Arnauld A II 2, 179) 50

184

M.-T. Liske

regularly, would amount to a permanent miracle. In our example, only that the body spun in a circle continues its motion linearly in the tangent on release is explicable in terms of a general law, the law of inertia. This can be understood as a potential. A development is the result of a potential as an inner force only if it is subject to laws. Now, the term ‘law’ applies only to those more general principles which are so basic and fundamental that their explanatory power goes beyond the behaviour observed in the present case. With regard to physics, Leibniz (e.g. Animadversiones II ad 25, GP IV 369; to Arnauld A II 2, 272) uses the notion of force to overcome the understanding of motion as something merely relative,53 which is the result of a mechanistic philosophy that conceives of motion purely geometrically54 as a change in the position of bodies relative to each other within the spatial extension.55 Relative to the arbitrary assumption of a particular body as the resting centre, other bodies altering their position with regard to it appear as moving. If motion is not to be a mere phenomenon in this way, but is rather to be a well-founded phenomenon or the appearance of something real, it must be founded, according to Leibniz’s ontology of substances, in an absolute property of a substantial individual, namely the moved body, and hence be caused by force as its inner principle. Since force is the only thing that is real about motion, and since the laws of conservation require that reality as a whole is neither augmented nor diminished, these laws refer to force, not to the quantum of motion. The metaphysical importance of force stems from the fact that it helps to overcome the mechanistic interpretation which explains motion solely in terms of extension. If one were to wholly deprive things of their force or potency (force ou puissance), one would in effect be denying them their status as substances and turn them into Spinozistic modi.56 If there is to be in the world a plurality of individual substances, any one of them must have a striving force as its inner principle from which its characteristic activities, namely its perceptions, emerge and which thus constitutes it as a substance.57 This reveals the metaphysical importance of potentials in Leibniz: Only because there are different potentials, each directed towards peculiar individual developments of perceptions, can there be numerous substantial entities.

53

See Duchesneau (1998), in particular p. 78f. In how far Leibniz can discard absolute space, but consistently assume absolute motion is shown by Roberts (2003). 54 Since the active forces transcend geometry, they cannot, according to Muntean (2007), be grasped by an infinitesimal calculus. 55 See, for instance, Gabbey (1971) (also with regard to Newton) and Garber (2004). 56 “[…] il y a dans la nature quelqu’autre chose que l’étendue et le mouvement, à moins que de refuser aux choses toute la force ou puissance, ce qui serait les changer de substances qu’ils sont en modes; comme fait Spinoza […]” (To Arnauld A II 2, 274). 57 Kulstad (1999) 20f., 32f.

Potentiality in Rationalism

4.4

185

Passive Capacities as Gradually Inferior Active Capacities

Potentials possess a passive aspect in so far as they refer to a development which something is subject to. However, not every development is grounded in a potential: think, for example, of an externally governed development for the worse. In accordance with the connotation of ‘strength’, ‘might, ‘force’ and ‘power’, potentials are directed towards an internally governed progression to something better and, in this sense, are essentially active. This mingling of activeness and passiveness accords with Leibniz’s conception of force. Yet, isn’t force essentially active? Leibniz can (for example, in GP IV 395) justifiably speak of a passive force in so far as he conceives of it as a form of active force, namely as a reactive force which opposes (repugnare) the primarily operating active force. Even though it is not capable of prevailing against the superior active force and is therefore acted on, it is effective and active in so far as it modifies the effect of the superior active force. This limitation of the operation of the active by the passive force can also be found, secondarily, in the physical realm in the form of impenetrability (impenetrabilitas, antitypia) and inertia,58 since, primarily, in the case of mental substances and their perceptions the primitive forces interact in this way. The primitive active forces (equivalent to Aristotle’s first entelechies) strive towards a further development of the activity, that is, towards more and more clear and distinct perceptions. Therefore, they can be viewed as potentials. The primitive passive forces (equivalent to the materia prima) are not independent further moments (i.e. capacities to be acted on) and thus are not to be conceived of as potentials. Rather, they modify this further development by constraining it while not cancelling it. Now, the individuality of a created monade just consists in its not perceiving everything clearly and distinctly, but only a small section (which differs from monade to monade). This section constitutes the monade’s individual perspective from which it conceives everything else. Therefore, the forward-striving primitive active force and the hindering passive force need to be combined with one another in order to jointly constitute a monade in its individuality, i.e. in its peculiar ability to achieve something and its peculiar limitation.59 In this way, the primitive forces, i.e. active and passive forces together, constitute a potential.

58

See Bernstein (1981). “[Entelechia primitiva …] conjungitur cum passiva potentia primitiva quam complet, seu cum qua Monadem constituit.” (GP II 250)

59

186

4.5

M.-T. Liske

The Reconciliation of the Perspectives of Final and Efficient Cause Through the Notion of Force

If, in contrast to Spinoza, we recognise potentials to develop into something new and positive beyond the actual being, then this suggests a teleology in the less problematic60 sense of an immanent teleology: That the present state is directed towards a future result as its goal explains why a disposition regularly leads to a certain positive result. Thus, an individual in an early stage has the potential to become, by way of a goal-directed development, a fully developed exemplar of its species, given that the process is not disturbed. In addition to this metaphysical teleological approach, a causal view is required: For when potentials or forces physically function as causes of motion, they are, according to Leibniz, to be considered mechanically. Yet, the concept of force can reconcile the perspectives of efficient cause and final cause, just as it can reconcile the physical perspective and the metaphysical perspective. In some passages in Leibniz, we find two separate, seemingly unconnected perspectives on the same thing. According to SD I 14 (p. 28, L. 375–388, GM VI 243), everything is doubly explicable: first, by the realm of power (i.e. physical force) or efficient causes which can quantitively be captured by mathematical laws; secondly, by the realm of wisdom or final causes which is qualitatively concerned with an optimum. Obviously, the moving forces of physics can be grasped quantitatively in mechanical terms (size, shape and motion), whereas the primitive forces of metaphysics in their striving for clearer perceptions can be grasped in terms of final causes. This classification is in accordance with the methodological dichotomy we frequently find in Leibniz (e.g. GP VII 343f.): A physical explanation of a single case has to be mechanic. If, however, we are concerned with the metaphysical foundations of these explanations and ask why these laws apply rather than others, final causes are appropriate. That Leibniz’s metaphysical principle is to be conceived of in teleological terms can already be seen from the fact that he commonly describes it with a term borrowed from Aristotle, namely as a first entelechy,61 but significantly reinterprets the latter as a primitive entelechy62 and conceives of it as a primitive force which, in virtue of an inherent striving or tendency, is directed towards a specific activity as its goal. Only under this telelological interpretation, which is suggested by the word component ‘séko1’ (goal), can the requirement— that a potency as the source of action needs to include a striving or tendency towards this activity in addition to aptitude—be elaborated on as Leibniz does in NE II 22 § 11 (A VI 6, 216): “It is in order to express this sense that I appropriate the term ‘entelechy’ [sc. to stand for force].” The derivative forces of physics which cause motion, however, are primarily to be understood in terms of efficient causes. 60

See Gale (1984) 67–71. See Latour (2002) and Jove (2004). 62 On ‘Entelechie primitive’, see, for example, Théod. III § 396 (GP VI 352); on ‘Entelechia primitiva’, see, for example, To de Volder GP II 250–252. 61

Potentiality in Rationalism

187

Thus in SD I 14 (p. 28, L. 386–388, GM VI 243) Leibniz remarks that the task in this work of natural philosophy is to formulate general rules for effective forces (virium effectricium) in order to apply them in specific explanations in terms of efficient causes (in causis specialibus efficientibus explicandis). However, if in a causal explanation of bodily motion any goal-directedness is to be ruled out, a conflict arises with Leibniz’s conception of the relation of the two forces. Thus, in SN the secondary moving force is viewed as a limitation or accidental modification of the primitive force.63 This manner of speaking is only appropriate if the same content as in the primitive force appears in the moving forces of physics in a limited, specific form and emerges through the passage of time in manifold accidental, that is, more or less coincidental appearances in a modified and varying way. The essential content of the primitive mental force is the mind’s striving for a further development of its perceptions. Physically, this striving appears in a limited, momentary form as a conatus,64 for example, as the inhibited striving for motion of a drawn bow. Since it precedes the actual movement, which ensues as soon as the body is released, it can be seen as its efficient cause. In so far as it is a striving, this striving for motion can be viewed as the way in which the primitive mental striving forces appear to us within the realm of bodily motion. Because Leibniz’s notion of force includes the concept of striving, which can equally refer mentally to the further development of perceptions as it can refer physically to motion, he is capable of reconciling the opposition between the teleological perspective of metaphysics and the causal-mechanistical perspective of physics.65

4.6

Do Potentials Lead to a Result Inevitably or Merely Probably?

In the remaining two sections, we want to discuss two regards in which it may appear problematic to speak of potentials in Leibniz. According to the realistic view that we can presuppose in Leibniz, potentials are properties of individual substances. As a mirror of the universe, a Leibnizian individual is world-bound, that is, it cannot exist in any possible world other than that to which it actually belongs. As a result, it is subject to the strict determination which governs the course of the world in question, notably of the real world. As the properties of an individual cannot be altered without altering its identity, the development to which a potential “[…] force mouvante qui est une limitation ou variation accidentelle de la force primitive.” (GP IV 473) 64 According to Carlin (2004), the opposite also holds: Since in the the discovery of optimal means derivative forces such as conatus and appetitus are at work, purposive action is subject to causal determination. 65 For Leibniz, especially the conservation laws, e.g. that of force, have to be justified through a teleological basis. See Moretto (2004), 189f. 63

188

M.-T. Liske

as a modal property leads is strictly determined, which means that it must necessarily attain its goal. If Leibniz repeatedly qualifies that an active force reaches its goal provided it is not hindered by anything else (e.g. GP IV 395, 469, 472), then this is due to an abstract perspective, which is confined to a section of reality, just as we may only sensibly speak of potentials in Spinoza if we confine our view to individuals. From a concrete point of view considering the connectedness of everything with everything, there can be no external impediment at all; rather, the goal to which something is disposed must always be realised completely. The plurality of possible worlds66 would only legitimize the modal assertion de dicto: it is possible that an individual with this potential does not develop into this kind of perfection. Yet, if one conceives of a potential as a de re modal property of an individual which causes in this subject a development towards a certain goal, then this result must occur, given Leibniz’s premises, with the necessity of natural laws. This raises the following issue: Is a potential given only when reaching the goal is inevitable, or doesn’t just such a necessity make the notion of a potential empty since such a potential ultimately results in a categorical property which unfolds itself in an already real because inevitably determined development? What speaks in favour of the first option are the arguments which seek to establish a total possibility as the only true possibility: Something counts as genuinely possible only if it is guaranteed that all single enabling conditions must at some point become realised. The totality of the necessary single conditions (requisita) defines, for Leibniz (e.g. A VI 2, 483 or A VI 3, 587), the sufficient cause which inevitably brings about the result. Accordingly, the totality of single enabling conditions, including the external conditions, is presupposed for a potential as well. However, since a Leibnizean substance is a living mirror of the external world, a potential as an intrinsic modal property of such a substance can encompass all conditions within itself. Epistemically, only after something or somebody has already acquired a new positive quality it can be said with certainty that this something or somebody had possessed the potential to develop towards the quality in question. At this point, however, the attribution has lost its informative value: We refer to potentials in order to predict future developments and are interested in whether something or somebody possesses still undeveloped potentials. In attributing a potential as an internal property we assume that somebody or something satisfies all internal presuppositions for a certain development. From the experience of a sufficient number of completed developments we have learned to distinguish between whether an existing potential only failed because of certain external conditions or whether the internal presuppositions were not given and therefore there was no potential in the first place. In this way, we can also decide with practically sufficient certainty whether a potential is given from which we can expect a certain future 66

By referring to merely possible, but never real courses of the world, Leibniz attempts to, in opposition to Spinoza, ensure never realised alternatives to the actual course of the world and hence a freedom in the unqualified sense which includes contingency as its precondition (see Liske (1993), in particular § 41).

Potentiality in Rationalism

189

development under normal conditions. If we demanded of a potential that it inevitably leads to what it is a potential for, we would not treat it as an inner property of a substance—unless we presuppose Leibniz’s particular conception of a substance as the mirror of the universe—but as a moment within a global scheme of determination. Moreover, a potential would become so complex that we would not be able to (epistemically) ascertain it. In order to see to what extent his deterministic premises allow Leibniz to develop a genuine conception of potential, we can best start from the fundamental tension which is implied in the notion of a potential: According to the common understanding, a potential is a property which is presently given and, in this sense, real and in virtue of which we can, under normal conditions, expect a state or activity in the future that is fundamentally new and hence not yet real. Accordingly, Leibniz demands against Descartes: A force’s absolute quantity has to be measured by its future effect (se doit estimer … par l’ effet futur, GP III 48). To begin with, this concerns physical facts: the work defined as force multiplied by distance. One needs to exert force of a certain intensity over a certain distance in order to lift a body to a certain height against the force of gravitation or to strain a spring against the elastic force (Leibniz’s examples). Here Leibniz perceives important metaphysical implications: “The force or power, but not the future effect, is already at present something real.”67 This tension can plausibly be explained by the striving inherent in the force in so far as it is already effective in the present moment, but anticipates something in the future in virtue of its being directed to its future effect. This supports Leibniz’s anti-cartesian conclusion: If one does not want to completely annul the power to act, one must not reduce bodies to the geometrical categories of size and velocity. For these are fully given in the present without any direction towards the future. Rather, one needs to assume a dynamic momentum directed towards the future, as it is given in the notion of force. In this regard, the notion of a striving helps to plausibly conceive of potentials. On the other hand, difficulties arise with Leibniz’s claim that striving must never be completely ineffective (see 4.2). Force needs thus to be always accompanied by activity (albeit minimal), physical force by a corresponding locomotion (GP III 59), even if this may only be an undetectable inner motion. A potential in the full sense of the word, which is not just to enable a gradual enhancement of something already present but directed as its goal towards something genuinely new that is not yet real, is hereby excluded. A potential even in the everyday sense of the word is a property or state of a subject which creates a (though not inevitable) determination in that it commits the subject to a development which is to be expected and, as far as it is up to the subject (if nothing hinders it), is passed through successfully towards the goal, namely the significant new property. It is remarkable that, in the lists of definitions of De

“[…] la force ou puissance est quelque chose de réel dès à présent, et l’effet futur ne l’est pas.” (GP III 48)

67

190

M.-T. Liske

affectibus,68 Leibniz defines determination in just this way: Determination is a state (status) from which something follows (ex quo quid sequitur), unless something else prevents it (A VI 4, 1426, L. 20). With this definition of 1679 (and hence long before he discovered the notion of force), Leibniz formulates a metaphysical desideratum: What inner state of the subject, if taken on its own, creates in this subject a determination towards a development (a determination that, due to the abstract perspective which considers the subject on its own, is not necessarily to be taken in the strict deterministic sense)? In the science of physics, he found in ‘force’ a notion which includes metaphysical content, thanks to the associated conception of striving, and can thus provide the desired metaphysical explanation and help to clarify the concept of a potential.

4.7

Are There Potentials for Temporal Developments in Leibniz?

The second problem concerning whether Leibniz’s basic assumptions allow for genuine potentials arises from the fact that in his metaphysics Leibniz conceives of the possibilia as being essentially atemporal. Just as the world itself, its mirror, the individual substance is determined by a complete concept which entails in an atemporal manner all properties that an individual ever possesses (DM §§ 8 and 13). Here all relations, even between physical states, are given as atemporal conceptual relations. However, under the dynamic approach to the substance proper to natural philosophy, the primitive force as the law of the entire succession of the perceptive states (4.3) occupies the role which is played by the concept of the individual under the approach of term logic: Both encompass the entire content of individual existence. Are these two perspectives on substance incompatible, or is Leibniz’s primitive force in the same way essentially atemporal as the concept of the individual? If so, the concept of primitive force could not serve as an adequate account of potentials which are essentially directed towards a temporal development. For Leibniz, the dynamic approach of natural philosophy, according to which the principle constituting the substance lies in the primitive active force which makes the substance produce changes in accordance with its own laws,69 is amalgated with the mentalist approach, according to which God has endowed the soul with the power to produce its thoughts (la vertu de produire ses pensées, GP III 464). Yet, for Leibniz the soul is not a special case, but the true model of an individual substance that constitutes itself by producing from its inner self the succession of its changing states in accordance with the laws constituting its individuality. All these

68

See Schneider (2001) 102–104. “[…] la nature de chaque substance consiste dans la force active, c’est à dire dans ce qui la fait produire des changements suivant ses lois.” (GP III 464). 69

Potentiality in Rationalism

191

states are to a lower or higher degree mental states, that is, thoughts or perceptions. How can Leibniz then speak of transformations, which entail time as a necessary condition and therefore are primarily physical events? Now thoughts are timeless only with regard to the objective side of their content; the subjective succession of our thoughts appears to us as a temporal one-after-the-other. The mentalist approach is compatible with the approach of term logic in so far as concepts allow for something analogous to succession, namely logical relations such as the derivability of consequences, the implication of necessary conditions etc. These logical relations appear to us in our subjective thinking as a temporal one-after-the-other of operations. The development towards which a striving is directed then need not be temporal in the strict sense of the word, but can also be a further development of the mental contents themselves (i.e. in their contents). Since there are analogues to successions in the true, i.e. the mental, reality, the successions of bodily motions are no mere phenomena but appearances of something real. Thus Leibniz can acknowledge genuine bodily substances which are constituted by a primitive active force as their inner active principle that turns an aggregate into a substance with a uniform activity, even though it is a phenomenon that they appear as bodies. If a primitive active force is what is ultimately at the core of the temporal processes of the bodily world, then a potential for a development is something real after all.

5 Summary Since Descartes atomises time and temporal existence into independent moments, he cannot recognise any identically persisting substance. Rather, God needs always to create a new substance, which is merely of the same kind as the previous one. Without identical properties in a strict sense of the word, however, there cannot be a potential, which as the same inner principle is the foundation of the entire development, but only dispositions to instantly realisable behaviour. Descartes’s talk of (bodily and mental) capacities, tendencies and powers is then simply a useful description of phenomena. In view of the essential necessity of the whole, in virtue of which there can be nothing for Spinoza beyond the actual way of existence and activity, the overall reality (identical to God) precludes potentials. However, in view of causal necessity, the distribution of active power over individuals can vary and therefore we can legitimately speak of potentials to develop towards greater active power, at least as long as the perspective is confined in an abstract manner to individual causes. In the context of Spinoza, potentials are not so much internally determined tendencies for development; rather they represent causal roles within the interplay of determining moments. In the more pregnant sense of objectively goal-directed powers in virtue of which something develops teleologically towards a real perfection that transcends its present state, Spinoza does not acknowledge potentials. For in his view, perfection is not to be measured against objective norms, but follows from the

192

M.-T. Liske

subjective perspective. A goal-directed development is not only ruled out with regard to overall reality. Even with regard to individual beings, Spinoza only assumes a tending (conatus) to be conserved in one’s essential way of being (metaphysical principle of inertia), not a tending to develop any further. Potentia is thus an active power which does not—in terms of alternative possibilities of development—go beyond the actual course of events. Such a power manifests itself, for instance, as the control over the affects in the present situation. Yet, a teleological explanation of human actions is possible in so far as this explanation can be reduced to a causal one. Leibniz’s ontology of substance is quite sympathetic to potentials as inner principles of further development. Accordingly, he makes an illuminating contribution under the notion of force. As a capacity that is enhanced by a striving towards a specific realisation, force, for Leibniz, by itself proceeds towards activity. Only seemingly it is a resting capacity. Since it has retired to non-observable internal motion, it is permanently ready to become effective again and can externally manifest itself in a development, once impediments are removed. In this way, it stands between potency and act. Moreover, it mediates between physics and metaphysics. For as a present state, in which inheres a striving for a subsequent state and which thus causes the transition to that state (for example, as an inhibited striving to motion), physical force is the temporary modification of the primitive force as the permanent basic tendency of the striving. Only this latter force represents a potential in that it regulates the development of mental states in accordance with principles characteristic of the individual and thus constitutes the individual’s nature. The initially physical notion of force, which, as an inner principle, can render motion real, simultaneously gives empirical content to the metaphysical postulate of a principle constituting substance. The causal-mechanistical perspective of physics and the teleological perspective of metaphysics are reconciled in the notion of striving which is directed mentally towards a further development of perceptions and, at the same time, physically towards locomotion as a goal and also functions as a preceding efficient cause of motion. Activeness and passiveness are reconciled in that the primitive force actively is a potential for further development, which is limited and modified by the passive force. In this way, both taken together represent the potential for the developments characteristic of a substance’s individuality. It may appear problematic whether Leibniz’s determinism allows for potentials in a full sense of the word. Strict determination of substances and their properties in the actual world stands in an ambivalent relation towards potentials. On the one hand, a potential just means that a presently real property determines a future development. If, on the other hand, force consists in always actually being active, the goal of the development can only be a gradual increase, and not something genuinely new. Everything that ever happens to the individual is at any time given as a reality in this individual. This is because, as a living mirror of its universe, an individual cannot truly be affected by something external to it within this universe; neither can the individual belong to another possible world. Although, for Leibniz, atemporal conceptual relations represent the basic structure of the world, the

Potentiality in Rationalism

193

primitive force can represent a potential for temporal developments, since logical relations of inference—given atemporally in the concept of the individual—are mentally experienced as temporal successions. If one deals with the question of in how far the rationalists of early modern philosophy can acknowledge true potentials, one can see to what extent the conception of a potential depends on the presuppositions of a metaphysic of substance. A potential can be understood as the capacity of a substance to develop—in virtue of its inner nature and under suitable conditions—towards a new (normally) positive quality (which itself can be a faculty), a quality to which the subject is not determined by necessity. A potential in the full sense of the word is characterised by a fundamental tension: It is a real categorical property in so far as it is grounded in the persisting inner nature of its subject. As a modal property, on the other hand, it is for its realisation and development dependent on contingent external conditions. This presupposes a plurality of substances which exist by themselves and do not necessarily depend on each other. If one speaks of a potential or its equivalent in the context of Descartes, who does not acknowledge any substances continually persisting by themselves, or in the context of Spinoza, who does not allow for anything existing outside the solely real and all-encompassing substance, the terminology of potential can only be a useful description of phenomena: to advantageously describe what the relation between the observed properties of a thing and its expected behaviour or development is. Nearest to a true potential is Leibniz’s concept of primitive metaphysical force: It is an inner principle governing the whole succession or development of the states of an individual in a way characteristic of it and thus, being the law of the individual development, constituting the individual substance.

Bibliography

I Primary sources with abbreviations AT Oeuvres de Descartes, ed. by C. Adam and P. Tannery, Nouvelle présentation, Paris: Vrin 1974–1986. G Spinoza, Opera, ed. by C. Gebhardt, Heidelberg: Winter 1925, Reprint 1972. Unless noted otherwise, references are to the Ethics where the following abbreviations are used: p. (propositio), sch. (scholium), dem. (demonstratio), ex. (explicatio), ax. (axioma), def. (definitio), af.def. (affectuum definitio(nes)), Ap. (Appendix), praef. (praefatio). A G.W. Leibniz, Sämtliche Schriften und Briefe, ed. by the Preußische Akademie der Wissenschaften zu Berlin. Darmstadt 1923ff., Leipzig 1938ff., Berlin: Akademie-Verlag 1950ff. Cited by Series (Reihe) and Volume (Band). Roman numerals refer to the Series, II: Philosophischer Briefwechsel (Philosophical Letters); VI: Philosophische Schriften (Philosophical Papers). Arabic numerals refer to the Volume. GP Die philosophischen Schriften von G.W. Leibniz, ed. by C. I. Gerhardt, Berlin 1875–1890, Reprint Hildesheim: Olms 1978.

194

M.-T. Liske

GM Leibnizens mathematische Schriften, ed. by C.I. Gerhardt, Berlin/Halle 1849–1863, Reprint Hildesheim: Olms 1962.

II English translations The English translations of quotations from primary texts are based on the following English editions. In many cases, my translations are compilations from several translators. Philosophical Writings of Descartes, 3 volumes, translated by J. Cottingham, R. Stoothoff, D. Murdoch, and A. Kenny, Cambridge: Cambridge University Press 1984–91. A Spinoza Reader. The Ethics and Other Works, edited and translated by E. Curley, Princeton: Princeton University Press 1994. Spinoza, Complete Works, with translations by S. Shirley, edited, with introduction and notes, by M. L. Morgan, Indianapolis/Cambridge: Hackett 2002. Leibniz, G.W., New Essays on Human Understanding, translated and edited by P. Remnant and J. Bennett, Cambridge: Cambridge University Press 1996. Leibniz, G.W., Philosophical Papers and Letters, translated and edited by L. E. Loemker, 2d ed., Dordrecht: Reidel 1969. Leibniz, G.W., Philosophical Texts, translated by R. Franks and R. S. Woolhouse, Oxford/New York: Oxford University Press 1998.

III References Allen, D. (1984). From vis viva to primary force in matter. Studia Leibnitiana Sonderheft, 13, 55– 61. Barbone, S. (2002). What counts as an individual for Spinoza. In O. Koistinen & J. Biro (Eds.), Spinoza: Metaphysical themes (pp. 89–112). New York: Oxford University Press. Benardete, J. A. (1990). Therapeutics and hermeneutics. In E. Curley & P.-F. Moreau (Eds.), Spinoza: Issues and directions (pp. 209–220). Leiden: E. J. Brill. Bennett, J. (1983). Teleology and Spinoza’s conatus. Midwest Studies in Philosophy, 8, 143–160. Bennett, J. (1984). A study of Spinoza’s Ethics. Cambridge: Cambridge University Press. Bennett, J. (1990). Spinoza and teleology: A reply to Curley. In E. Curley & P.-F. Moreau (Eds.), Spinoza: Issues and directions (pp. 53–57). Leiden: E. J. Brill. Bernstein, H. R. (1981). Passivity and inertia in Leibniz’s Dynamics. Studia Leibnitiana, 13, 97– 113. Boudri, J. C. (2002). What was mechanical about mechanics. The concept of force between metaphysics and mechanics from Newton to Lagrange. Dordrecht/Boston/London: Kluwer Academic Publishers. Campbell, J. K. (1999). Descartes on spontaneity, indifference, and alternatives. In J. Gennaro & C. Huenemann (Eds.), New essays on the rationalists (pp. 179–199). New York: Oxford University Press. Carlin, L. (2004). Leibniz on conatus, causation, and freedom. Pacific Philosophical Quarterly, 85, 365–379. Carriero, J. (1991). Spinoza’s views on necessity in historical perspective. Philosophical Topics, 19, 47–96. Coelho, R. L. (2002). Zur Physik von Descartes: Naturgesetze und Stoßregeln. Philosophia Naturalis, 39, 45–60. Cook, T. (2006). Der Conatus: Dreh- und Angelpunkt der Ethik. In M. Hampe & R. Schnepf (Eds.), Baruch de Spinoza: Ethik in geometrischer Ordnung dargestellt (pp. 151–170). Berlin: Akademie-Verlag. Curley, E., & Moreau, P.-F. (Eds.) (1990). Spinoza. Issues and directions. Leiden: E. J. Brill.

Potentiality in Rationalism

195

Curley, E. (1990). On Bennett’s Spinoza: The issue of teleology. In E. Curley & P.-F. Moreau (Eds.), Spinoza: Issues and directions (pp. 39–52). Leiden: E. J. Brill. Curley, E., & Walski, G. (1990): Spinoza’s necessitarianism. In R. J. Gennaro & C. Huenemann (Eds.), New essays on the rationalists (pp. 241–262). New York: Oxford University Press. Damschen, G., Schnepf, R., & Stüber, K. (Eds.) (2009). Debating dispositions. Berlin/New York: de Gruyter. De Dijn, H. (2006). Ethik als Heilkunde des Geistes (5 p 1–5 p 20). In M. Hampe & R. Schnepf (Eds.), Baruch de Spinoza: Ethik in geometrischer Ordnung dargestellt (pp. 267–282). Berlin: Akademie-Verlag. Della Rocca, M. (1999). If a body meet a body: Descartes on body-body causation. In R. J. Gennaro & C. Huenemann (Eds.), New essays on the rationalists (pp. 48–81). New York: Oxford University Press. Della Rocca, M. (2008). Rationalism run amok: Representation and the reality of emotions in Spinoza. In C. Huenemann (Ed.), Interpreting Spinoza (pp. 26–52). Cambridge: Cambridge University Press. de Gaudemar, M. (1992). Quelques questions autour de la notion leibnizienne de puissance. Studia Leibnitiana, 24, 216–220. Duchesneau, F. (1998). Leibniz’s theoretical shift in the Phoranomus and Dynamica de Potentia. Perspectives on Science, 6, 77–109. Fichant, M. (1995). De la puissance à l’action: la singularité stylistique de la Dynamique. Revue de Métaphysique et de Morale, 100, 49–81. Freudenthal, G. (2002). Perpetuum mobile: The Leibniz-Papin controversy. Studies in History and Philosophy of Science, 33, 573–637. Gabbey, A. (1971). Force and inertia in seventeenth century dynamics. Studies in History and Philosophy of Science, 2, 1–67. Gale, G. (1970). The physical theory of Leibniz. Studia Leibnitiana, 2, 114–127. Gale, G. (1973). Leibniz’ dynamical metaphysics and the origins of the vis viva controversy. Systematics, 11, 184–207. Gale, G. (1984). Leibniz’ force: Where physics and metaphysics collide. Studia Leibnitiana Sonderheft, 13, 62–70. Gale, G. (1988). The concept of ‘force’ and its role in the genesis of Leibniz’ dynamical viewpoint. Journal of the History of Philosophy, 26, 45–67. Garber, D. (2004). Leibniz on body, matter, and extension. Proceedings of the Aristotelian Society, 78, 23–40. Garber, D. (2009). Leibniz: Body, substance, monad. Oxford: Oxford University Press. Garrett, D. (1991). Spinoza’s necessitarianism. In Y. Yovel (Ed.), God and nature, Spinoza’s metaphysics (pp. 191–218). Leiden: E. J. Brill. Garrett, D. (1999). Teleology in Spinoza and early modern rationalism. In R. J. Gennaro & C. Huenemann (Eds.), New essays on the rationalists (pp. 310–335). New York: Oxford University Press. Garrett, D. (2002). Spinoza’s conatus argument. In O. Koistinen & J. Biro (Eds.), Spinoza: Metaphysical themes (pp. 127–158) Oxford: Oxford University Press. Garrett, D. (2008). Representation and consciousness in Spinoza’s naturalistic theory of the imagination. In C. Huenemann (Ed.), Interpreting Spinoza: Critical essays (pp. 4–25) Cambridge: Cambridge University Press. Gennaro, R. J., & Huenemann, C. (Eds.) (1999). New essays on the rationalists. New York: Oxford University Press. Griffin, M. V. (2008). Necessitarianism in Spinoza and Leibniz. In C. Huenemann (Ed.), Interpreting Spinoza: Critical essays (pp. 71–93). Cambridge: Cambridge University Press. Gueroult, M. (1967). Leibniz. Dynamique et métaphysique. Paris: Aubier-Montaigne. Hampe, M., & Schnepf, R. (Eds.) (2006). Baruch de Spinoza: Ethik in geometrischer Ordnung dargestellt. Berlin: Akademie-Verlag. Hassing, R. F. (1980). The use and non-use of physics in Spinoza’s Ethics. Southwestern Journal of Philosophy, 11, 41–70.

196

M.-T. Liske

Hassing, R. F. (2003). Leibniz without physics. The Review of Metaphysics, 56(2002/3), 721–761. Huenemann, C. (Ed.) (2008). Interpreting Spinoza: Critical essays. Cambridge: Cambridge University Press. Jove, D. (2004). Aristóteles, Leibniz y la noción de entelequia. Logoi, 7, 135–149. Koistinen, O., & Biro, J. (Eds.) (2002). Spinoza: Metaphysical themes. Oxford: Oxford University Press. Koistinen, O. (2003). Spinoza’s proof of necessitarianism. Philosophy and Phenomenological Research, 67, 283–310. Kolesnik-Antoine, D. (2009). L’homme cartésien. La «force qu’a l’âme de mouvoir le corps». Rennes: Presses Universitaires de Rennes. Kulstad, M. A. (1999). Leibnizian meditations on monism, force and substance, in relation to Descartes, Spinoza and Malebranche. The Leibniz Review, 9, 17–42. Laerke, M. (2007). Contingency, necessity, and the being of possibility. Leibniz’s modal ontology in relation to his refutation of Spinoza. Revue Roumaine de Philosophie, 51, 39–62. Latour, A. (2002). Le concept leibnizien d’entéléchie et sa source aristotélicienne. Revue Philosophique de Louvain, 100, 698–722. Liske, M.-T. (1993). Leibniz’ Freiheitslehre. Die logisch-metaphysischen Voraussetzungen von Leibniz’ Freiheitstheorie. Hamburg: Felix Meiner. Liske, M.-T. (2009a). Harmonizing modern physics with Aristotelian metaphysics. Leibniz’s theory of force. In G. Damschen, R. Schnepf & K. Stüber (Eds.), Debating dispositions: Issues in metaphysics, epistemology and philosophy of mind (pp. 99–126). Walter de Gruyter: Berlin Liske, M.-T. (2009b). Nach Verwirklichung strebende Aktivkräfte versus schlummernde Potenzen: Kann Leibniz’ Vermögensbegriff die Konzeption eines Potentials erhellen? Studia Leibnitiana, 41(2), 203–232. Lloyd, G. (1990). Spinoza on the distinction between intellect and will. In E. Curley & P.-F. Moreau (Eds.), Spinoza: Issues and directions (pp. 113–123). Leiden: E. J. Brill. Look, B. C. (2007). Perfection, Power and the Passions in Spinoza and Leibniz. Revue Roumaine de Philosophie, 51, 21–38. Machamer, P. (2008). The dispositions of Descartes. In G. Damschen, R. Schnepf & K. Stüber (Eds.), Debating dispositions: Issues in metaphysics, epistemology, and philosophy of mind (pp. 71–78). Berlin: De Gruyter. Manning, R. N. (2002). Spinoza, thoughtful teleology, and the causal significance of content. In O. Koistinen & J. Biro (Eds.), Spinoza: Metaphysical themes (pp. 182–209). Oxford: Oxford University Press. Matheron, A. (1991). Essence, existence and power in Ethics I: The foundations of proposition 16. In Y. Yovel (Ed.), God and nature: Spinoza’s metaphysics (pp. 23–34). Leiden: E. J. Brill. Moretto, A. (2004). Physics and metaphysics: The Leibnizian foundation of mechanics. In M. Carrara, A. M. Nunziante & G. Tomasi (Eds.), Individuals, minds and bodies: Themes from Leibniz (pp. 173–191). Stuttgart: Franz Steiner Verlag. Muntean, I.-L. (2007). Leibniz and the Mathematization of Forces. Revue Roumaine de Philosophie, 51, 113–128. Nachtomy, O. (2007). Possibility, agency, and individuality in Leibniz’s metaphysics. Dordrecht: Springer. Narbonne, J.-M. (1995). La notion de puissance dans son rapport à la causa sui chez les stoiciens et dans la philosophie de Spinoza. Archives de Philosophie, 58, 35–53. Oksenberg Rorty, A. (1990). The two faces of Spinoza. In E. Curley & P.-F. Moreau (Eds.), Spinoza. Issues and directions (pp. 196–208). Leiden: E. J. Brill. Quintana, A. P. (2006). Fuerzas, Potencias, Tendencias, Substancias: Fisica y Metafísica en Leibniz. Revista Laguna, 18, 11–34. Perler, D. (2006). Das Problem des Nezessitarismus (I p 28–36). In M. Hampe & R. Schnepf (Eds.), Baruch de Spinoza. Ethik in geometrischer Ordnung dargestellt (= Klassiker Auslegen, Vol. 31) (pp. 59–80). Berlin: Akademie Verlag. Phemister, P. (2005). Leibniz and the natural world: Activity, passivity, and corporeal substances in Leibniz’s philosophy. Dordrecht: Springer.

Potentiality in Rationalism

197

Poser, H. (2006). Leibniz et la potentialité des idées innées: un problème modal. In F. Duchesneau & J. Griard (Eds.), Leibniz selon les Nouveaux essais sur l’entendement humain (pp. 21–33). Montréal/Paris: Bellarmin. Renz, U. (2007). Zwischen ontologischer Notwendigkeit und zufälliger Semantik. Zu Spinozas Theorie der menschlichen Affekte. In H. Landweer (Ed.), Gefühle – Struktur und Funktion (pp. 35–56). Berlin: Akademie-Verlag. Renz, U. (2008). Spinoza: Philosophische Therapeutik der Emotionen. In H. Landweer & U. Renz (Eds.), Klassische Emotionstheorie (pp. 311–330). Berlin/New York: de Gruyter. Renz, U. (2009). Explicable explainers: The problem of mental dispositions in Spinoza’s ethics. In G. Damschen, R. Schnepf & K. Stüber (Eds.), Debating dispositions: Issues in metaphysics, epistemology and philosophy of mind (pp. 79–98) Berlin/New York: de Gruyter. Rice, L. C. (1985). Spinoza, Bennett and teleology. The Southern Journal of Philosophy, 23, 241–253. Roberts, J. T. (2003). Leibniz on force and absolute motion. Philosophy of Science, 70, 553–573. Rotenstreich, N. (1977). Conatus and Amor Dei: The total and partial norms. Revue Internationale de Philosophie, 31, 117–134. Schneider, M. (2001). Leibniz’ Theorie der Aktion im Jahrzehnt vor dem Discours de métaphysique 1677–1686. Studia Leibnitiana, 33, 98–121. Schütt, H.-P. (1985). Spinozas Konzeption der Modalitäten. In R. Bubner, K. Cramer & R. Wiehl (Eds.), Kontingenz. Neue Hefte für Philosophie 24/25 (pp. 165–183). Sherman, K. (1974). Descartes’ change of mind. The Philosophical Forum, 5, 557–571. Shimony, I. (2010). Leibniz and the vis viva controversy. In M. Dascal (Ed.), The practice of reason (pp. 51–73). Amsterdam/Philadelphia: Benjamins. Slowik, E. (1996). Perfect solidity: Natural laws and the problem of matter in Descartes’ universe. History of Philosophy Quarterly, 13, 187–204.

Author Biography Michael-Thomas Liske is Professor of Philosophy at the University of Passau, Germany. He tries to reinterpret classical philosophical texts in the light of contemporary discussions and by contemporary methods especially of analytic philosophy. Publications: “Leibniz’ Freiheitslehre. Die logisch-metaphysischen Voraussetzungen von Leibniz’ Freiheitstheorie“ (Meiner 1993), ”Aristoteles und der aristotelische Essentialismus. Individuum, Art, Gattung“ (Alber 1985).

Potentiality in British Empiricism Katia Saporiti

1 Introduction We attribute properties to objects, and we may regard some of these properties as potentialities or dispositional properties. For some properties (e.g. brittleness, solubility, and flammability) can be said to have conditions of actualization but need not to be actualized at the time at which they are truthfully ascribed to an object. Something is water-soluble if and only if it will dissolve when put into water. But if there is no water around and the thing in question does not dissolve, it may nevertheless count as soluble in water. Some dispositional properties do not have to be actualized at all during the time an object possesses the property. To be sure, some inflammable things can be destroyed without burning. In some cases the relevant actualization conditions of an ascribed dispositional property will never obtain during the time a given object exists. There are, on the other hand, properties (e.g. the location, shape and size of an object) regarding which it seems to make less sense (if any) to speak of their actualization and of conditions of their actualization. If something is in Paris (e.g. the Eiffel Tower), it is just there. Being in Paris is so to speak actualized by everything that can correctly be said to be in Paris. By attributing a certain shape to an object we do not seem do indicate a potentiality. To put it differently, the attribution of some properties (i.e. dispositional properties) seems to involve hypothetical judgements and relying on the truth of counterfactual conditionals in a way in which the attribution of other properties does not. There is a different and somewhat narrower sense in which we often speak of potentials, or so it seems. When we say that somebody has the potential to become an accomplished musician we seem to attribute to the person in question a number of properties which enable her to acquire further properties. But every dispositional property is a property of being such that it is possible for the object that has the K. Saporiti (&) University of Zurich, Zürich, Switzerland e-mail: [email protected] © Springer Science+Business Media B.V., part of Springer Nature 2018 K. Engelhard and M. Quante (eds.), Handbook of Potentiality, https://doi.org/10.1007/978-94-024-1287-1_8

199

200

K. Saporiti

property to undergo or resist change, that is, to acquire or to lose or to keep properties (to a greater or lesser extent, where these properties admit of degrees). A relevant difference seems to lie in a teleological aspect that enters our use of the word “potential” in its narrow sense. One might think of a goal to be reached, an aptitude for something, a predisposition, the makings of e.g. a musician, where the nature of the kind determines what will count as an achievement or a perfection. In this sense, “potential” seems to refer to a range of capabilities something has, or to powers or resources not yet developed. But it is not clear whether, on an ontological level, this amounts to more than the attribution of a number of (dispositional or non-dispositional) properties which, assuming the relevant conditions of actualization to be highly probable, are considered to make it more or less likely that a number of properties will be acquired, retained or lost by the object in question. (I take potentialities or dispositional properties to comprise all kinds of capacities, abilities, liabilities, susceptibilities, inclinations and tendencies of sentient and non-sentient beings). In order to spell out an account of this narrow sense of “potential” one would have to carefully consider the different kinds of dispositional and non-dispositional properties that can be relevant to the attribution of different kinds of potentials. In what follows I will look at some basic ontological claims made by the British empiricists Locke, Berkeley, and Hume and ask how “potentialities” and “potentials” in a wide sense may fare within their respective ontologies.

2 John Locke: Substance and Power John Locke is famous for his way of ideas. Ideas, he thinks, are the material of the understanding. Most, if not all, of our cognitive capacities involve, when exercised, the perception of ideas—where to perceive an idea means to have an idea. Whether we perceive something with our senses or imagine something, whether we rack our brains or let our minds wander, etc. we always have ideas. Lockean ideas exist if and only if they are in some mind. They are mental representations in the following sense: they are ideas of something, and they present to the mind what they are ideas of. An external world may exist independently of the mind. But we have no immediate or direct access to it. We perceive the world, as it were, through our ideas. Thus, all we have got to go on if we want to know something about the world are our ideas. Consequently, when in his Essay Concerning Human Understanding1 Locke wishes to investigate what we can know, and how we come to believe and to know things, he concentrates on ideas. He explains how we come by our ideas and how the mind operates on its ideas. He distinguishes different kinds of ideas and

1

I will refer to this work as the Essay. To refer to sections of the Essay I will use the abbreviation “E” followed by number of book, chapter and section. For example, „E II. viii. 4” refers to Book II, Chap. viii, Sect. 15.

Potentiality in British Empiricism

201

explains their relations to words, on the one hand, and their relations to a world outside our minds, on the other. But Locke does not confine himself to suggesting a theory of ideas. In spite of his criticisms of the terms “substance” and “essence”, or rather of the ways he thought these terms were used by some of his contemporaries and predecessors, Locke himself gives theoretical accounts of the notions of substance, essence, and quality. Considering this, one may well be justified in asking the question whether Locke thought about and how he conceived of potentialities. Locke takes our ideas of corporeal and spiritual substances to be complex ideas. Like all ideas they ultimately derive from experience, i.e. from simple ideas we obtain through sensation and reflection. Ideas of corporeal substances are collections of ideas of sensible qualities which we believe to be united in things we call by whatever name we give to the relevant substance. Given that we are unable to conceive how sensible qualities can exist alone or in one another, we take them to exist in and to be supported by something—a bearer called “substance”. Of this support—of substance in general—we have no clear and distinct idea (E II. xxiii. 4). Something analogous holds for our ideas of spiritual substances. Through reflection we acquire ideas of thinking, reasoning, fearing, knowing, doubting, etc. and, concluding that these operations do not subsist by themselves, we assume that they are actions performed by something we call “spirit”. We distinguish spiritual from bodily substances because we cannot conceive how the observed operations can belong to, or be produced by, bodies. But in truth, as regards the supposed spiritual substances, we are just as incapable of forming a notion of the supporting thing as we are in the case of material substances, “the one being supposed to be (without knowing what it is) the Substratum to those simple Ideas we have from without; and the other supposed (with a like ignorance of what it is) to be the Substratum to those Operations, which we experiment in our selves within” (E II. xxiii. 5). Yet, as Locke points out in one of his letters to Stillingfleet, we conclude with certainty that such a substratum or support exists wherever there is a simple idea of sensation or reflection. Substance in the sense of a substratum or bearer of qualities and ideas cannot simply be discarded.2 So, according to Locke, every idea of a corporeal substance comprises, besides various ideas of sensible qualities, the idea of something these qualities inhere in. Something analogous holds for our ideas of spiritual substances. They too comprise an idea of something which is modified and affected, an idea of something which has the faculties whose exercise we witness in reflection. Note, though, that Locke is not a Cartesian dualist. Our idea of substance in general may be the same in both cases. It can become a component of the complex idea of matter (if we add the idea of solidity to it) as well as a component of the complex idea of spirit (if we add the idea of thinking to it).3 Moreover, Locke notoriously allows for the possibility of 2

W IV. 18 & 445f. (“W” is used to refer to The Works of John Locke, followed by number of volume and page.) 3 In a letter to Stillingfleet (of 1697) Locke writes: “[…] the general Idea of Substance being the same every where, the Modification of Thinking, or the Power of Thinking joined to it, makes it a Spirit […]. As on the other side Substance, that has the Modification of Solidity, is Matter […]”

202

K. Saporiti

God super-adding thought to material substances.4 Though the obscure and confused idea of something (“we know not what”) which supports qualities or is modified or affected in different ways forms part of every idea of an individual substance (concrete being) and of kinds of substances, it would be wrong to conclude that, on Locke’s account, we have no ideas of substances, or that Locke denies the existence of substances. For Locke believes that Sensation convinces us, that there are solid extended Substances; and Reflection, that there are thinking ones: Experience assures us of the Existence of such Beings; and that the one hath a power to move Body by impulse, the other by thought; this we cannot doubt of.5

We may not be able to say much about substance in general, but we can say something about substances, as our ideas of them are complex and comprise more than just the obscure and confused idea of an unknown bearer. Substances, according to Locke, are qualified by their properties, which, in the case of corporeal or material substances he calls “qualities”.6 Qualities or properties of substances are what, in the external world, corresponds to our simple ideas acquired through sensation and reflection. And Locke defines qualities as powers. Whatsoever the Mind perceives in it self, or is the immediate object of Perception, Thought, or Understanding, that I call Idea; and the Power to produce any Idea in the mind, I call Quality of the Subject wherein that power is. Thus a Snow-ball having the power to produce in us the Ideas of White, Cold, and Round, the Powers to produce those Ideas in us, as they are in the Snow-ball, I call Qualities; and as they are Sensations, or Perceptions, in our Understandings, I call them Ideas: which Ideas, if I speak of sometimes, as in the things themselves, I would be understood to mean those Qualities of the Objects which produce them in us. (E II. viii. 8)

Locke proceeds to distinguish three kinds of qualities of material substances: Primary qualities, which are inseparable from body—qualities each particle of a material substance will possess (even if it is too small to be perceived). Here, Locke mentions solidity, extension, figure and mobility (E II. viii. 9). Secondary qualities of substances, which are “but Powers to produce various Sensations in us by their primary Qualities”. Locke mentions colours, sounds and tastes as examples of secondary qualities (E II. viii. 10). A third kind of quality is formed by powers of substances to change the qualities of other substances. Locke mentions the power of

(W IV: 33); cf. Ayers (1977, 91) and McCann (2007, 182), where the passage is read in the same way as is suggested here. 4 E IV. iii. 6. For a helpful discussion of this idea of super-addition of thinking to matter see Downing (2007). 5 E II. xxiii. 29. One of the reasons why experience convinces us this way is the passivity of the mind in perceiving simple ideas of sensation and reflection. For the mind cannot, according to Locke, produce or destroy any simple idea it receives from one of these sources (E II. ii. 2). Locke calls the kind of knowledge we have of the existence of things other than God and ourselves sensitive knowledge (E IV. iii. 21). 6 As regards the properties of immaterial substances, souls or spirits, Locke prefers to speak of „modifications” and „affections”. For an exception see E II. xxiii. 30, where Locke refers to thinking and the power of action as qualities or properties of spirits.

Potentiality in British Empiricism

203

fire to change the colour and texture of wax or clay. Like secondary qualities, qualities of the third kind depend on the primary qualities of the particles of substances (ibid.). In view of the fact that Locke declares secondary qualities to be “nothing in the Objects themselves, but Powers to produce various Sensations in us” (E II. viii. 10), some scholars have argued that Locke does not regard primary qualities as powers.7 But this is unconvincing for several reasons. Locke claims qualities in general to be powers. He does not qualify this claim. By way of illustrating the distinction between qualities (i.e. powers to produce ideas in us) and ideas of these qualities Locke mentions not only secondary qualities (white and cold) but also a primary quality (round). And surely, Locke takes primary qualities as well as secondary qualities to produce ideas in us (E II. viii. 9).8 The wording of the passage in which Locke first draws the distinction (cited above) may well be taken to suggest a difference between qualities that are nothing but powers to produce ideas in us, on the one hand, and qualities that are powers to produce ideas in us and something more, on the other hand.9 Read this way the distinction may have been intended to draw attention to the fact that only primary qualities enter into certain kinds of explanations of the appearances of bodies, whereas secondary qualities are such that there is nothing more to say about them than that substances which possess them will produce specific ideas (sensations) in us. Finally, taking qualities in general to be powers to produce ideas agrees nicely with the empiricist strategy Locke deploys when introducing substances and their qualities into his ontology. For what he takes as his starting point is our experience: our simple ideas of sensation and reflection. Only on account of our observing the co-occurrence of some of these ideas (which cannot be produced or destroyed by the mind itself) do we suppose what they are ideas of (i.e. qualities) to be united in and supported by one bearer. But what is it that our ideas are ideas of? What is it that we thereby suppose to exist in that supporting thing? At this point of the story, Locke’s answer seems to suggest itself as a matter of course: what we are dealing with are powers to produce ideas in us— powers to produce those ideas in us from which we infer the existence of something having these powers. Following our best scientific theory we may attribute qualities to imperceptible parts of substances. But this does not change our notions of the qualities in question. Something triangular will look triangular if it is big enough to be seen. To attribute primary qualities to imperceptibly small parts of substances is not to attribute imperceptible qualities to them but to attribute qualities to them which, were they possessed by something big enough to be perceived, would produce 7

E.g. Cummins (1975, 408–410), Mackie (1976, 11–12) and Stuart (2003, 70). For a recent discussion see Jacovides (2007). 8 Cf. the end of the section in which Locke introduces the expression “primary qualities”: “These I call original or primary Qualities of Body, which I think we may observe to produce simple Ideas in us, viz. Solidity, Extension, Figure, Motion, or Rest, and Number” (E II. viii. 9). 9 See Campbell (1980) for an attempt to spell out what this extra could be which justifies the distinction.

204

K. Saporiti

certain ideas in us. Locke seems to take the corpuscularianism of his day, which he finds in Robert Boyle’s writings, to be the best hypothesis available when it comes to giving an explanation of the qualities of bodies. According to this new mechanistic science macroscopic bodily phenomena can be explained by reference to the motions and impacts of sub-microscopic particles. Corporeal substances are configurations of these particles (corpuscles), which can be characterized entirely in terms of their size, shape, solidity and motion or rest. It is somewhat controversial to what extent Locke endorsed corpuscularianism, or whether he endorsed it at all.10 At any rate, Locke explicitly refers to corpuscularianism as a mere hypothesis. I have here instanced in the corpuscularian Hypothesis, as that which is thought to go farthest in an intelligible Explication of the Qualities of Bodies; and I fear the Weakness of humane Understanding is scarce able to substitute another, which will afford us a fuller and clearer discovery of the necessary Connexion, and Co-existence, of the Powers, which are to be observed united in several sorts of them.11

Locke takes the question “which ever Hypothesis be clearest and truest” concerning the qualities of bodies to lie outside the scope of his investigation, and he apologizes for engaging in “Physical Enquiries a little farther than, perhaps [he] intended”, asking to pardon his “Excursion into Natural Philosophy”.12 Furthermore, he was well aware that corpuscularianism, for all its merits, leaves us unable to explain cohesion, the communication of motion, and mind-body interaction. And he does mention the possibility that the corpuscularian hypothesis might be wrong.13 Given his assessment of the situation, it seems that Locke would have been ill-advised to rest his basic metaphysical claims about the nature of substances and their qualities on corpuscularianism. It is reasonable, therefore, to take the point of the distinction between three kinds of qualities to be the following: our best scientific theory—the one which is most resourceful in explaining the qualities of bodies, even if it leaves much to be desired—yields explanations which take substances at a sub-microscopic level to possess only some of the qualities

10

Alexander (1985, 6–7), Mandelbaum (1964, 1–3) and Yolton (1970, 11) are among the great number of scholars who suggest that Locke assumes Boylean corpuscularianism to be true and rests his philosophy on what he takes to be the best scientific theory of his day. Some passages in the Essay look as if Locke tried to give arguments for corpuscularianism (e.g. E II. viii. 9 or E II. iii. 20). But, on the other hand, there is strong evidence for the view that Locke diagnosed several explanatory gaps and shortcomings of corpuscularianism, cf. Wilson (1979). For a recent attempt to develop a coherent account of Locke’s position see Downing (2007). 11 E IV. iii. 16, see also E IV. iii 11. 12 E II. viii. 22, see also E IV. iii. 16. Right at the beginning of the Essay Locke declared: „I shall not at present meddle with the Physical Considerations oft he Mind; or trouble my self to examine, wherein its Essence consist, or by what Motions of our Spirits, or Alterations of our Bodies, we come to have any Sensation by our Organs, or any Ideas in our Understandings; and whether those Ideas do in their Formation, any, or all of them, depend on Matter, or no. These are Speculations, which, however curious and entertaining, I shall decline, as lying out of my Way, in the Design I am now upon” (E I. i. 2). 13 E IV. iii. 11. Explanatory gaps of the hypothesis seem to be recognized e.g. in E II. xxiii. 23–29 and E IV. iii. 29, see also W IV, 467–8.

Potentiality in British Empiricism

205

which are exhibited by substances on a macroscopic level, and which suppose furthermore that some of the differences between various qualities of the sub-microscopic particles of a substance will not come to the fore in the substance in question but only in other substances. The first supposition gives reasons for drawing a distinction between primary and secondary qualities, the second supposition warrants a distinction between primary and secondary qualities, on the one hand, and qualities of a third kind, on the other. It is somewhat misleading to say that secondary qualities (and, perhaps, qualities of the third kind mentioned by Locke) can be explained in terms of primary qualities. For this suggests a kind of reducibility which Locke would certainly have to deny. It is not as if the temperature of a perceived object were explained by its motion. On a corpuscularian account, the temperature of a corporeal substance depends on the motion of its small parts. It is only by invoking different levels of description (by relating the macroscopic to a sub-microscopic level) that the corpuscularian’s explanations work. At the macroscopic level, primary and secondary qualities are perfectly alike in that they produce ideas in us and can be regarded as powers to do so. Qualities of the third kind too are powers of substances to produce ideas in us. But they will do so only mediately, that is, by changing the qualities of other substances. To be sure, a substance’s power to produce an idea of a secondary quality in us (e.g. the sensation of yellow) is, according to Locke, most likely due to the primary qualities of imperceptible parts of the substance. But this does not mean that the quality in question is a power to produce in us ideas of the primary qualities of parts of the substance. A sensation of yellow is not an idea of a primary quality at all.14 According to Locke, it is an idea of a secondary quality. He regards simple ideas of sensation as sensations. At a pinch, secondary qualities of perceptible objects could, on a Lockean account, be expected to be reducible to the primary qualities of imperceptible parts of the objects (not to primary qualities of the objects themselves). But there is yet a further sense in which secondary qualities are not reducible to primary qualities. For even if we knew much more about the primary qualities of the imperceptible parts of substances than we actually do know, reducibility would presuppose the existence of necessary connections between the primary qualities of the imperceptible parts of a substance and its secondary qualities. But, according to Locke, “there is no conceivable connexion betwixt the one and the other”.15 14

Locke points out that, if we were able to perceive the sub-microscopic parts of material substances, our sensations would differ dramatically from the ones we actually have when perceiving such substances (E II. xxiii. 11). 15 E IV. iii. 13. As Locke takes the real essences of things to be that from which all their qualities spring necessarily, this means that we cannot know the real essences of substances (E IV. iii. 12– 14). Locke defines real essences of things as “the very being of any thing, whereby it is, what it is. And thus the real internal, but generally in Substances, unknown Constitution of Things, whereon their discoverable Qualities depend, may be called their Essence” (E III. iii. 15). Real essences are relative to what Locke calls nominal essences (E III. vi. 6). For nominal essences are the abstract ideas (formed by the mind from simple ideas of sensation and reflection) by which we classify and sort all things. “But it being evident, that Things are ranked under Names into sorts or Species,

206

K. Saporiti

We identify qualities of substances by means of what it is that the possession of these qualities enables the substances to effect in us, that is, by means of the ideas they enable the substances to produce in us (immediately or mediately, as in the case of the third kind of qualities Locke mentions). And we distinguish between substances according to their qualities. It is by exercising their powers that substances make themselves known to us. The powers of a substance enable it to do or to suffer something, to perform an action or to undergo a passion. According to Locke, we acquire the idea of power by observing change. In reflection as well as in sensation we notice a constant change of our ideas. And from having observed a certain systematic co-variation in our ideas we conclude that the same kinds of changes will take place in the same way in the future. And we infer that there exists a common feature in the things that effect the changes as well as in the things that suffer the changes. This common feature is what we take to be responsible for the regularities of the changes we experience. In one thing we locate the possibility to have its qualities, modifications, or affections changed, and in another the possibility of effecting these changes. It is therefore a possibility to cause or to suffer a change—a possibility which we infer to exist in things, and which we call a power. [A]nd concluding from what it [the Mind] has so constantly observed to have been, that the like Changes will for the future be made, in the same things, by like Agents, and by the like ways; considers in one thing the possibility of having any of its simple Ideas changed, and in another the possibility of making that change; and so comes by that Idea which we call Power. (E II. xxi. 1)

Locke distinguishes between active and passive powers. An active power enables a substance “to make […] any change”, a passive power enables a substance “to receive any change” (E II. xxi. 2). It is only from reflection that we gain a clear and distinct idea of an active power, which according to Locke includes the initiation or production of a change exclusively worked by the thing which possesses the power in question. In reflection we experience ideas of various operations of our own mind such as the ideas of willing something and of triggering motion. These are ideas of actions not of powers, but from them we immediately infer the existence of powers to perform these actions, viz., powers co-existing in a substance called spirit. From the changes we experience in material substances (bodies) we get but an imperfect obscure idea of an active power, as the communication of motion by impulse seems to Locke rather to be a continuation of a change suffered by one body in another one than a proper initiation of movement (E II. xxi. 4). Still, Locke is convinced that, starting from a constant change in our ideas, we arrive at the idea of substances (corporeal and spiritual) which are qualified by their properties, that is, by their qualities, modifications, and affections. These, as we have

only as they agree to certain abstract Ideas, to which we have annexed those Names, the Essence of each Genus, or Sort, comes to be nothing but that abstract Idea, which the General, or Sortal […] Name stands for” (E III. iii. 15). For an illuminating discussion of nominal and real essences in Locke see Atherton (2007).

Potentiality in British Empiricism

207

seen, are powers—powers to produce certain ideas in us. The idea of a power is the idea of a possibility to suffer or to cause change, existing in a thing which possesses the power. Even though it involves some kind of relation, the idea of power can, according to Locke, be counted among our simple ideas, which, in turn, form part of our complex ideas of substances. In fact, right at the beginning of his discussion of our idea of power Locke takes the idea of power to “make a principle Ingredient in our complex Ideas of Substances” (E II. xxi. 3). And later, when dealing with our idea of substance, he emphasizes again that ideas of powers form a great part of the ideas contained in our ideas of substances (e.g. E II. xxiii. 8 and 10). Our ideas of material substances are ideas of sensible qualities that we have observed to exist together, combined with an idea of something wherein these qualities exist (E II. xxiii. 6). The qualities are powers of the substance to produce ideas in us by our senses— primary as well as secondary qualities and qualities of the third kind (E II. xxiii. 7 & 9). Our ideas of spiritual substances are taken from simple ideas of various operations of the mind such as thinking, understanding, willing, knowing, and starting a motion, and hence from ideas we acquire in reflection (E II. xxiii. 15). From these we infer the existence of powers to perform these operations as co-existing in one substance (E II. xxi. 4). Thus, our ideas of spiritual substances, like our ideas of material substances, are ideas of powers co-existing in and being supported by something of which we have no distinct idea. We have, as Locke puts it, “no Idea of what it is, but only a confused obscure one of what it does” (E II. xiii. 19). If we take together different strands in Locke’s discussion of the ideas of power and substance, we arrive at a conception of bodies and spirits as beings that are qualified by possibilities to effect or undergo specific changes. Our idea of the existence of such beings amounts to the idea of the co-existence of these possibilities. It is the idea of collections or bundles of possibilities as actual possibilities.16 It is tempting to regard Locke’s explanation of the way we acquire the idea of power (part of which has been cited above) as a hint that Locke took modalities to be more primitive than powers. On the other hand, Locke insisted on the existence of substances as bearers of qualities, i.e. as beings that have powers. The claim that there exists some unknown X which has certain powers might be understood as meaning that certain possibilities exist or are actual, but these would be possibilities for the unknown X to effect or undergo changes. As Locke takes all qualities (affections or modifications) of substances known by us from sensation and reflection to be powers, substances (whether material or spiritual) have but dispositional properties—as far as we can come to know about them through experience (and there is, according to Locke, no other way of coming 16

If something is possible, this implies the actuality of a possibility. But not vice versa. If a substance can cause a certain sensation (e.g. if a snowball can—has the power to—cause a sensation of white) this does not imply that it is possible that the substance causes the sensation. For some of the conditions that make an actualization of the actual possibility possible may not be fulfilled (cf. White 1975, Chap. 1, for a discussion of the actuality of possibilities and the possibility of actualities).

208

K. Saporiti

to know about substances and their properties).17 For these properties (powers) to be actualized or realized (exercised) certain conditions have to obtain. More specifically, the actualization of a power to produce an idea in a mind presupposes the existence of a perceiving or reflecting mind. This leads to the following question regarding Locke’s account: to what extent and in which way do the existence and nature of substances depend on the mind? As the possession of a dispositional property does not presuppose its actualization, a substance may have the properties it does have without there being a mind which perceives it. If to be white and cold means to cause sensations of white and cold in a perceiving mind if there is one and the right conditions obtain, something may in the absence of such a mind or the right conditions be white and cold without being perceived by any mind. (There may be dispositional properties that one cannot possess unless they are actualized at least sometimes, e.g. certain character traits, but none of the qualities of non-sentient beings seem to be of this kind.) Hence, to say that, according to Locke, all qualities of substances that can be known from experience are powers to produce ideas in us, or mere possibilities to suffer or to cause change, does not amount to saying that the existence and nature of substances depend on the substances being perceived by some mind. On Locke’s account there seems to be yet another sense in which the nature of substances depends on perceiving and reflecting minds. The mind forms its complex ideas of substances (collections of qualities co-existing in a thing supporting them) according to its experience. Which qualities it takes to belong to such a collection depends on the ideas it has had in the past. It is based neither on the detection of necessary connections between different qualities (or different ideas) nor on knowledge of the real essences of substances. (Real essences would yield all the qualities spelled out by a nominal essence defining a kind or species.) While Locke supposes the world to exist independently of any mind, the way we carve up nature into different kinds of beings (the way we classify and sort things) depends on our faculties and individual experiences. Our complex ideas of substances are nominal essences only, and they are subject to the shortcomings of human understanding. We take them to represent patterns of necessarily co-occurring qualities in nature. But we can never know them to be correct, and they will always be inadequate. Besides other deficiencies, they are incomplete in that they will never comprise all the qualities that will systematically co-occur in nature. [W]hatever Collection of simple Ideas it [the Mind] makes of any Substance that exists, it cannot be sure, that it exactly answers all that they are in that Substance. Since not having tried all the Operations of all other Substances upon it, and found all the Alterations it would receive from, or cause in other Substances, it cannot have an exact adequate Collection of all its active and passive Capacities; and so not have an adequate complex Idea of the Powers of any Substance, existing, and its Relations, which is that sort of

17

Our knowledge of the existence of God and our knowledge of our own existence are exceptions, see E IV. ix.–xi. “The Knowledge of the Existence of any other thing [but God and ourselves] we can have only by Sensation: […] no particular Man can know the Existence of any other Being, but only when by actual operating upon him, it makes it self perceived by him” (E IV. xi. 1).

Potentiality in British Empiricism

209

complex Idea of Substance we have. […] For since the Powers, or Qualities, that are observable by us, are not the real Essence of that Substance, but depend on it, and flow from it, any Collection whatsoever of these Qualities, cannot be the real Essence of that Thing. Whereby it is plain, that our Ideas of Substances are not adequate; are not what the Mind intends them to be. (E II. xxxi. 12)

Where Locke points out that we distinguish between different substances primarily by relying on their secondary qualities and their qualities of the third kind, he often speaks of these qualities as relational qualities or relations.18 The fact that Locke does not speak in such a way of primary qualities has been taken as evidence against the view defended here that, according to Locke, all qualities are powers. If, as I have argued, primary qualities are powers to produce ideas in us, why does he not claim them to be relational too? Locke (arguing against Aristotelianism and Cartesianism in the philosophy of nature) takes secondary qualities and qualities of the third kind to be more important than primary qualities to the way we classify and sort things. He seems to feel the need to justify this because of the fact that these qualities are relational. All qualities of substances are relational in that they are powers to produce ideas in us. But secondary qualities are nothing else and may thus be thought not to exist in the substance at all (they may be thought not to inhere in the substance), and hence they may be thought not to be fit to form part of our complex idea of a substance—of its nominal essence. Qualities of the third kind are not only relational in the sense of being powers to produce ideas in us. They are powers to do so mediately only—by changing the qualities of other substances or by having their own qualities changed by other substances. Thus, besides being powers to produce ideas in us, they also are powers to change the qualities of other substances or to suffer a change from other substances. Just like secondary qualities, but for different reasons, they may consequently be suspected not to inhere in a substance and, hence, not to form part of the nominal essence of the substance. But, as Locke insists, qualities of the third kind are just as much powers in the substance as are its secondary qualities, although they are relational and will not be regarded as being in a substance “considered barely in it self”.19 The upshot of Locke’s remarks on the qualities of substances is that we do not individuate substances by way of non-relational or intrinsic properties, but by way of their potentials or potentialities to suffer or effect change. (In fact, according to Locke, we distinguish substances from one another primarily by relying on qualities of the third kind.) Different substances are beings with different potentialities. As far as I can see, Locke uses the term potentialities only once in the Essay, and that is in a passage where he tries to defend his taking our ideas of qualities of the third kind to form part of our complex ideas of substances, counting them, as it were, among

18

Cf. the passage just quoted, and, e.g., E II. xxiii. 37. E II. xxiv. 37. “For the power in Fire to produce a new Colour, or consistency in Wax or Clay […] is as much a quality in Fire, as the power it has to produce in me a new Idea or Sensation of warmth or burning” (E II. viii. 10), cf. E II. xxiii 10.

19

210

K. Saporiti

the simple ideas comprised by our ideas of substances. This way of proceeding stands in need of justification on account of the fact that the idea of power, as Locke had noted before,20 involves an idea of some kind of relation and should thus be regarded as a complex, rather than as a simple, idea. For he has the perfectest Idea of any of the particular sorts of Substance, who has gathered, and put together, most of those simple Ideas, which do exist in it, among which are to be reckoned its active Powers, and passive Capacities; which though not simple Ideas, yet, in this respect, for brevity’’ sake, may conveniently enough be reckoned amongst them. […] For all those Powers, that we take Cognizance of, terminating only in the alteration of some sensible Qualities, in those Subjects, on which they operate, and so making them exhibit to us new sensible Ideas, therefore it is that I have reckoned these Powers amongst the simple Ideas, which make the complex ones of the sorts of Substances; though these Powers, considered in themselves, are truly complex Ideas. And in this looser sense, I crave leave to be understood, when I name any of these Potentialities among the simple Ideas, which we recollect in our Minds, when we think of particular Substances. For the Powers that are severally in them, are necessary to be considered, if we will have true distinct Notions of the several sorts of substances. (E II. xxiii. 7)

In this passage Locke is concerned with qualities of the third kind, as can be told from the examples he gives. These are: the power of a load-stone to draw iron and the power of iron to be so drawn, the power of fire to change the colour and consistency of wood, which in its turn has the power to undergo certain changes and to become charcoal. Locke insists that these qualities are “inherent qualities in those subjects” (ibid.). If, as we have argued, all qualities of substances are powers to produce ideas in us and hence relational as well as dispositional properties, why are qualities of the third kind the only ones called “potentialities” by Locke? The significant difference between primary and secondary qualities, on the one hand, and qualities of the third kind, on the other, may well be the following: only qualities of the third kind can be regarded as powers to change powers, as dispositions to acquire or to confer new dispositions. They are, so to speak, dispositions of a second order, powers to gain and to lose powers and to cause other substances to gain or to lose powers. So it seems that qualities of the third kind recognized by Locke are more like potentials in the narrow sense alluded to above, that is, in the sense in which, for example, we hold a person to have the potential to become an accomplished musician. If we look at Locke’s discussion of the idea of power, on the one hand, and his discussion of the idea of substance, on the other, certain tensions seem to come to the surface. Some scholars have taken this to be due to the fact that sometimes

“I confess Power includes in it some kind of relation, (a relation to Action or Change,) as indeed which of our Ideas, of what kind soever, when attentively considered does not?” (E II. xxi. 3) Locke goes on to point out relational aspects in our ideas of extension, duration, number, figure, motion, colours, smells, etc. and concludes: “Our Idea therefore of Power, I think, may well have a place amongst other simple Ideas, and be considered as one of them, being one of those, that make a principal Ingredient in our complex Ideas of Substances […]” (ibid.). 20

Potentiality in British Empiricism

211

Locke argues from a relatively sceptical epistemological point of view while at other times he dogmatically seems to hang on to certain vestiges of a traditional metaphysical doctrine. Others take the conflict simply to result from Locke’s thinking both in terms of bodies and in terms of powers. Be this as it may, I will briefly mention some of these tensions and in some cases hint at ways to loosen or to dissolve them. As we have seen, the idea of power is taken to be a simple idea by Locke. But according to different sets of criteria Locke proposes for the simplicity of ideas, the idea of power is strictly speaking a complex idea. Locke’s explanation seems to be that, in thinking of bodily substances, i.e. collections of qualities co-existing in one thing, powers may for reasons of convenience be reckoned among the simple ideas that form part of our complex ideas of substances.21 In a similar vein, Locke seems to be prepared to attribute active power to material substances in assuring us that “active Powers make so great a part of our complex Ideas of natural Substances”, adding that he does “mention them as such, according to common apprehension” (E II. xxi. 2). Here, as in the case of taking the idea of power to be a simple idea, Locke is well aware of a potential conflict. For he goes on to argue for his contention that, strictly speaking, only the operations of a spirit or a will afford us with a clear and distinct idea of an active power. And while we infer from experience that, in solid extended substances, there doubtlessly exists a power to move a body by impulse (E II. xxiii. 29), this power is ultimately not to be regarded as an active one (E II. xxi. 4). The tension seems to dissolve, as Locke evidently opts for the view that there is no active power in matter. As regards the simplicity of the idea of power, things may be more complicated. For one thing, Locke’s notion of a simple idea is notoriously difficult to assess, as he seems to offer heterogeneous criteria for ideas counting as simple ideas. But none of these would allow us to count the ideas of powers among our simple ideas. Secondly, it has been doubted that our ideas of sensible qualities (e.g. ideas of white, cold, and round) can be counted as simple ideas, owing to the fact that they seem to have a double-faced nature in the Essay, as particular sensations, on the one hand, and as general ideas, on the other. There is no room here to discuss Locke’s distinction between simple and complex ideas and the way it relates to his distinction between particular and general ideas.22 But if we take the idea of power to be complex, this would mean that our ideas of substance are not collections of simple ideas. They would comprise, not only the idea of a substratum in which qualities inhere, but further complex ideas as well. In fact, our ideas of the qualities of substances would have to be regarded as complex ideas. An additional difficulty seems to arise with regard to the perceptibility of qualities. Secondary qualities (e.g. white and cold) are said to be nothing but powers to produce certain ideas in us. At the same time Locke seems to suppose that

21

Locke acknowledges the complexity oft the idea of power in several places in the Essay. Two passages have already been quoted: E II. xxi. 3 and E II. xxiii. 7. 22 For a thorough discussion of Locke’s notion of general ideas see Specht (2011), especially Chap. E4.

212

K. Saporiti

powers are not directly perceivable, as can be gathered from his discussion of the way we come to form the idea of power from the changes we experience in our ideas (E II. xxi. 4). The existence of powers has to be inferred from the outcome of their actualization. This means that we do not directly perceive that a snowball, for example, is white and cold. We do not experience its whiteness or coldness. Rather, we infer from the ideas (sensations) the snowball produces in us that it is white and cold (that it possesses the qualities in question). Locke (and this is a tension in his view that we have mentioned before) declares qualities of material substances to be powers to produce ideas in us. A material substance, a snowball for instance, causes certain ideas in us from which we gather its qualities, e.g. that it is white, cold, and round. Just as we infer from our sensations that the snowball is cold and white, we also infer that it is round. At the same time several passages in the Essay have been taken by scholars to indicate that Locke thought primary qualities to be non-dispositional, categorical or intrinsic properties of substances. Accordingly, Locke is held to acknowledge the fact (if it is one) that dispositional properties presuppose the existence of non-dispositional ones.23 Not all the qualities of a substance, it is claimed, can be dispositional. Not all the qualities of a substance can be powers. Whatever has the power to do or to suffer something has to have non-dispositional properties in virtue of which it has this power—properties that enable the thing in question to do or to suffer whatever the power is a power to do or to suffer. But while it is true, that Locke links secondary qualities of substances (powers to produce certain sensations in us) to primary qualities of their imperceptible parts, this does not imply that primary qualities are non-dispositional qualities. Secondary qualities such as colours, smells and tastes depend on the primary qualities, not of the substance which possesses these secondary qualities, but of its minute parts. According to Locke, it is by the primary qualities of their insensible parts that substances, will produce certain sensations in us (viz., ideas of secondary qualities) if the right conditions obtain.24 To consist of minute particles possessing certain qualities (i.e. powers) is itself a non-dispositional property. It is this property that every substance with a secondary quality will possess, according to the corpuscularian hypothesis. It may be possessed at times and under conditions where none of the dispositional qualities (powers) either of the substance or of its imperceptible parts are actualized. It should also be noted that, on Locke’s account, our sensitive knowledge of the 23

Cf. Chappell (2007, 135). Chappell holds that Locke claims all qualities to be powers. However, he also believes that Locke at the same time took the primary qualities of the imperceptible parts of bodies exhibiting secondary qualities to us to be always and unconditionally present in these bodies. This seems to be right, but it does not imply that primary qualities fail to be dispositional properties. 24 It has to be admitted, though, that Locke is not always careful enough to distinguish between the qualities of substances and the qualities of their minute particles (see e.g. E II. viii. 10). He seems to assume that this lack of precision is not likely to confuse his reader, just as he thinks that no harm will be done by talking of ideas (of sensible qualities) as existing in (material) substances when he is actually referring, not to our ideas of these qualities, but to the qualities themselves (as he does in E II. xxiii. 7; cf. E II. viii. 8; both sections have been quoted above).

Potentiality in British Empiricism

213

existence of material substances possessing powers to produce ideas in us is based on experience, while the attribution of a non-dispositional property to consist of imperceptible parts of matter is based on a scientific hypothesis. It is important to Locke that his metaphysical claims are compatible with this hypothesis, as indeed they are. But Locke’s admiration for the corpuscularian hypothesis as the hypothesis “that goes farthest in explaining the qualities of bodies” does not force him to retract or qualify his definition of qualities as powers to produce ideas in us.

3 George Berkeley: Collections of Ideas Just like Locke, George Berkeley pursues an empiricist strategy in developing his ontology. We experience a constant change of ideas. Ideas exist in the mind. To exist ‘in a mind’ means to be perceived by a mind. For ideas to exist they have to be perceived. Just as thoughts or pains cannot exist without being thought or felt, ideas cannot exist without being had.25 Ideas exist if and only if they are perceived by something: a mind (spirit). Berkeley does not say much about minds or spirits. A spirit is something active—an active principle or a substance. But in his ontology Berkeley, unlike Locke, acknowledges nothing but ideas and minds or spirits. Spirits have ideas and operate on them in various ways. Berkeley notoriously denies the existence of matter and of material objects. The formula esse est percipi vel percipere captures one of his basic tenets: Everything that exists is either something which is perceived or something which perceives. The only kind of thing that can perceive something is a mind (a spiritual substance). The only kind of thing that a mind can perceive is an idea. Thus, this is all there is: ideas and minds. Berkeley identifies perceivable things with ideas. [A] certain colour, taste, smell, figure and consistence having been observed to go together, are accounted one distinct thing, signified by the name apple. Other collections of ideas constitute a stone, a tree, a book, and the like sensible things; which as they are pleasing or disagreeable, excite the passions of love, hatred, joy, grief, and so forth. (P 1) For what are the forementioned objects [houses, mountains, rivers, and in a word all sensible objects] but the things we perceive by sense, and what do we perceive besides our own ideas or sensations; and is it not plainly repugnant that any of these or any combination of them should exist unperceived? (P 4)

According to Berkeley, perceivable objects such as apples, stones, trees and books are collections of ideas. They are collections of perceivable qualities; and perceivable qualities, e.g. a colour, a figure, a taste or a smell, are ideas. It is important to note that Berkeley takes sensible qualities as well as sensible objects to

See Berkeley’s (1710), hereafter referred to as Principles, Sect. 2. From now on I will use “P”, followed by number of section, to refer to sections of the Principles.

25

214

K. Saporiti

be ideas,26 and that he takes sensible qualities as well as sensible objects to be perceived immediately.27 Berkeley, as opposed to Locke, does not distinguish between things we perceive by our senses and things we perceive immediately. On the contrary: the argument Berkeley puts forward in the first of his Three Dialogues turns on identifying sensible things with things immediately perceived by our senses. Right at the beginning of the dialogue, Philonous and his antagonist Hylas agree on calling only those things ‘sensible things’ which are immediately perceived by our senses. It is of some importance to Berkeley that we perceive sensible things immediately. Berkeley hoped that his theory would contribute to combating scepticism, which in his opinion posed a serious threat to science and religion. Representational realism as advanced by Descartes or Locke seems to lie open to the sceptic’s attack. If there exists a world of mind-independent things, which evoke ideas in us without being (directly) perceivable, we can do no more than infer their existence and nature from our ideas. Thus the question will always remain whether our conclusions are justified. Consequently, Berkeley rejects a model of perception according to which mind-independent objects cause ideas in us, which in their turn are the entities we are immediately aware of in perception. In Berkeley’s view, our ideas of sense do not represent anything different from themselves, neither in the sense of the scholastic distinction between the objective existence of things in the mind and their formal or real existence without the mind, which was taken up by Descartes, nor in the sense of Locke’s quasi Epicurean view of simple ideas as signs from which we infer their causes. Berkeley’s ideas of sense are not ideas of something. They are not the effects of things; they themselves are the things perceived. By identifying ideas and things Berkeley intends to lift the veil of perception and to block the sceptical consequences arising from the thesis that the only things we immediately perceive are our own ideas. As the sensible things are the ideas we immediately perceive, there seems to be no question as to their existence; and as the sensible things are nothing but collections of the sensible qualities we perceive, there seems to be no question as to their real nature. Berkeley’s idealism has to be interpreted as an attempt to prevent us from being forced by certain basic ontological assumptions to

26 In the first section of the Principles he calls sensible qualities like smells or colours, etc. ideas; in the fourth section he calls sensible objects like houses, mountains and rivers, ideas. (A sensible quality is a quality which can be perceived by the senses. A sensible object is an object which can be perceived by the senses.) 27 In his Three Dialogues between Hylas and Philonous (1713), hereafter referred to as Three Dialogues, Berkeley writes “And that sensible qualities are objects immediately perceived, no one can deny” (3D III: 237). And: “Wood, stones, fire, water, flesh, iron, and the like things, which I name and discourse of, are things that I know. And I should not have known them, but that I perceived them by my senses; and things perceived by the senses are immediately perceived; and things immediately perceived are ideas; and ideas cannot exist without the mind; their existence therefore consists in being perceived; when therefore they are actually perceived, there can be no doubt of their existence.” (3D III: 230) („3D” is used as an abbreviation for “Three Dialogues”, followed by number of Dialogue and page number in Volume II of the Luce-Jessop edition of Berkeley’s Works.)

Potentiality in British Empiricism

215

defend an epistemologically disastrous position—a position according to which we can at best gain mediated knowledge of the sensible world.28 Like Locke, Berkeley takes us to sort and classify things according to the systematic co-variation of our ideas. It is because we perceive certain qualities constantly to occur together that we take them to constitute one thing. Berkeley also acknowledges that we perceive ideas of sense, as he calls them, involuntarily. But unlike Locke, Berkeley believes that this does not warrant an inference to something existing without the mind. He rejects the existence of material substances as an untenable speculation about the causes of our ideas. According to Berkeley, the notion of a material substance (matter) is either empty or self-contradictory. Nothing in our experience indicates the existence of material substance (which itself is supposed not to be perceivable). If, in our mind, we try to separate from a sensible thing all its sensible qualities, our mind will be empty. We end up with nothing we could use the term “substance” to refer to. Following Locke, Berkeley also points out that the assumed relation between a material substance and its qualities is unintelligible. For what could it mean for a material substance to support its qualities, or for its qualities to inhere in the substance? Where the term “matter” is not empty or obscure, it entails a contradiction. For matter is supposed to be entirely passive and different from mind. Yet it is taken to support perceivable qualities. But nothing that exists in something other than a mind (and hence independently of and without the mind) can be perceivable. For only ideas are perceivable. And ideas exist within a mind. They are perceived by a mind. [I]t seems […] evident that the various sensations or ideas imprinted on the sense, however blended and combined together (that is, whatever objects they compose) cannot exist otherwise than in a mind perceiving them. I think an intuitive knowledge may be obtained of this, by anyone that shall attend to what is meant by the term exist when applied to sensible things. The table I write on, I say exists, that is, I see and feel it; and if I were out of my study I should say it existed, meaning thereby that if I was in my study I might perceive it, or that some other spirit actually does perceive it. There was an odour, that is, it was smelled; there was a sound, that is to say, it was heard; a colour or figure, and it was perceived by sight or touch. (P 3)

Some scholars have been encouraged by this passage to interpret Berkeley as a phenomenalist. The existence of objects we can perceive by the senses would thus come down to the existence of sets of possible sensations. Perceivable objects could be regarded as mere potentials. But it is misleading if not plainly wrong to call Berkeley a phenomenalist, for there is no evidence that Berkeley did not endorse the fundamental anti-scholastic principle that the potential exists only in virtue of the actual. Berkeley neither holds that all statements about sensible objects are equivalent in meaning to statements describing sensations (whose existence and

28

One of his motives for identifying ideas and things, instead of regarding ideas as representations of the things we perceive, lies in his conviction that what we have immediate and complete knowledge of are our own ideas. Cf. e.g. P 87.

216

K. Saporiti

nature depend on the sensing subject) nor does he believe that sensible objects are permanent possibilities of sensations. According to Berkeley, our talk of sensible objects rests on the laws of nature—on lawful regularities revealed by our ideas of sense and sustained by God. In the light of this theory a translation of statements about sensible objects into statements about the sensations and ideas of human beings is not possible without reference to these regularities, which have a source independent of human beings and their sensations and ideas. Furthermore, if Berkeley were a phenomenalist he would not need to invoke God to guarantee the continuous existence of sensible things. If for him a chair were nothing but a set of systematically correlated possible sensations, it could exist without its being perceived. Berkeley does not advocate a phenomenalistic account of perceivable objects. For he not only holds that sensible objects are collections of ideas but also claims that ideas exist if and only if they are perceived. For an idea and for any collection of ideas to exist means to be perceived not just to be perceivable. While Berkeley is well-known for his idealism and immaterialism, he is less so for his realism. Although he abandons the concept of formal existence—which was traditionally supposed to stand in contrast to objective existence—and thereby makes the objective existence of things (their existence in the mind as things perceived or thought of) their only way of being, Berkeley wants to hold on to the real existence of sensible things. He insists on the existence of a world which we can perceive through our senses and which we can come to know through experience—a world that exists independently of any finite spirit. To account for the difference between imagination and reality Berkeley distinguishes between two kinds of ideas: ideas of sense, on the one hand, and ideas of imagination on the other. Our ideas of sense (e.g. tables and smells) are caused by God in a systematic way (obeying, as it were, the laws of nature). Ideas of imagination are produced by us. They are less lively, orderly and stable than our ideas of sense. Berkeley is convinced that the distinction between these two kinds of ideas enables him to set apart reality and imagination (fancies, dreams, hallucinations, etc.). And he insists that he does not want to question the reality of the perceptible world.29 [B]y the principles premised, we are not deprived of any one thing in Nature. Whatever we see, feel, hear, or any wise conceive or understand, remains as secure as ever, and is as real as ever. There is a rerum natura, and the distinction between realities and chimeras retains its full force. (P 34) I do not argue against the existence of any one thing that we can apprehend, either by sense or reflexion. That the things I see with mine eyes and touch with my hands do exist, really exist, I make not the least question. The only thing whose existence we deny, is that which philosophers call matter or corporeal substance. (P 35)

Although sensible things are (collections of) ideas, they do not cease to exist just because no finite being perceives them. According to Berkeley, sensible things exist in the mind of God or are perceived by God. God guarantees the objectivity of the

29

In fact, Berkeley’s theory may be regarded as a kind of direct realism (cf. Saporiti 2006).

Potentiality in British Empiricism

217

perceptible world. He not only ensures the continuous existence of sensible things, but also their intersubjective perceptibility. He does this by producing ideas of sense in us finite beings in such a way that we are able to notice regularities in our ideas of sense. These, according to Berkeley, are the laws of nature. Scientists misinterpret the structure of the system of ideas produced by God in finite beings if they claim the existence of necessary connections between things we perceive by our senses. They misinterpret the regularities we experience if they postulate substances whose nature or inner constitution is causally responsible for the observable qualities of sensible objects and the changes sensible objects undergo. If we look at Berkeley’s notions of causality and power, it becomes clear that he cannot assume potentialities or attribute potentials to perceivable objects (collections of ideas) in the way Locke does. While Berkeley takes the connection between a cause and its effect to be a necessary connection, he denies the existence of any such necessary connections in the realm of perceivable things. The things we tend to regard as causes and effects in nature are, just like signs and things signified, only arbitrarily connected. According to Berkeley, ideas are entirely passive—unable to cause anything. Our idea of causation is exclusively acquired through our experience of the voluntary production of ideas. And in our minds we cannot separate the production of ideas from our ideas of cause and effect. So the concept of an unthinking cause is empty. All our ideas, sensations, or things which we perceive, by whatsoever names they may be distinguished, are visibly inactive, there is nothing of power or agency included in them. So that one idea or object of thought cannot produce, or make any alteration in another. (P 25) I find I can excite ideas in my mind at pleasure, and vary and shift the scene as oft as I think fit. It is no more than willing, and straightway this or that idea arises in my fancy: and by the same power it is obliterated, and makes way for another. This making and unmaking of ideas doth very properly denominate the mind active. Thus much is certain, and grounded on experience. But when we talk of unthinking agents, or of exciting ideas exclusive of volition, we only amuse ourselves with words. (P 28)

Only a thinking substance which is essentially active and has a will can cause something.30 Nothing truly active or productive can be detected in the successions of our ideas of sense, which constitute the course of nature. Perceivable objects, whether at rest or in motion, are essentially inactive, passive, powerless, unable to cause anything. In taking the sun to warm our faces we project causation and power into the things we perceive, where, in fact, there is only a regularity of successions of ideas. We claim to know the cause of something if we know under which conditions and when to expect it. We say that an open flame causes pain when touched, that the sun heats up the stone, that a falling rock makes a noise, etc. Berkeley accounts for our tendency to assume causes in nature in a similar way as Hume does and advises his readers, in these matters, to think with the learned, and

30

See also P 102, 105, 107 and 3D II: 215ff., 3D III: 236.

218

K. Saporiti

speak with the vulgar (P 51). But scientists are to be counted among the learned. And they distort reality if they claim that everything perceivable causes something. Scientists will discover uniformities in the course of nature. But they cannot find any causes and effects. There are no causal connections in nature. Scientific explanations, therefore, cannot be causal explanations. Nothing can be both perceivable and causally potent. According to Berkeley, causality and necessity are notions that concern the metaphysician, but are not an object to be studied by science or the philosophy of nature. In spite of the fact that Berkeley praises the achievements of the new science and regards Newton’s Principia Mathematica as an important and justly admired book (P 110), he is very critical of certain ways of interpreting these achievements (e.g. by Leibniz and Locke). In his treatise on motion, De Motu (1721), Berkeley points out that terms such as “power”, “gravity”, “attraction” are often used inconsistently or merely metaphorically, and that they are sometimes used as abbreviations to denote known phenomena (that allegedly get explained by reference to what these terms are supposed to denote). In other cases they will be empty words without any denotations (this is the case where they are supposed to refer to abstract ideas, which according to Berkeley, however, we are unable to frame). Sometimes they will be used as theoretical terms which gain meaning only within the theories they are embedded in. From Berkeley’s point of view, this happens in mechanics, which in his day formed the greater part of natural philosophy. In giving mechanistic explanations scientists use mathematical concepts and operations to establish general principles and laws. In Berkeley’s view, this is unobjectionable as long as these principles are taken to be mathematical hypotheses—useful fictions that do not reveal anything about the real constitution of nature.31 Berkeley argues against a realistic interpretation of the new doctrine of motion: against the assumption that the postulated powers in bodies are causally efficacious properties of these bodies. He claims that the assumption of causally efficacious powers in bodies is nothing but a postulation of occult qualities of the type found in scholastic theorizing. While we support heavy bodies we feel in ourselves effort, fatigue, and discomfort. We perceive also in heavy bodies falling an accelerated motion towards the centre of the earth; and that is all the senses tell us. By reason, however, we infer that there is some cause or principle of these phenomena, and that is popularly called gravity. But since the cause of the fall of heavy bodies is unseen and unknown, gravity in that usage cannot properly be styled a sensible quality. It is, therefore, an occult quality. But what an occult quality is, or how any quality can act or do anything, we can scarcely conceive—indeed we cannot conceive. (DM 4) Force likewise is attributed to bodies; and that word is used as if it meant a known quality, and one distinct from motion, figure, and every other sensible thing and also from every

31 Berkeley takes Newton’s principle of gravity to be such a hypothesis (DM 17). (“DM” is short for Berkeley’s De Motu, followed by number of section.) He regards Leibniz’s use of the terms “sollicitatio”, “nisus”, or “conatus” as metaphorical and often inappropriate.

Potentiality in British Empiricism

219

affection of the living thing. But examine the matter more carefully and you will agree that such force is nothing but an occult quality.” (DM 5)32

Those who attribute (active) powers to material substances blur the distinction between bodies and spirits,33 they contradict the principle that only a soul endows a body with the power to move, and they question the notion of a creatio continua, i.e. the assumption that God creates all finite beings and keeps them in existence from one moment to the next.34 Berkeley believes that the term “power” is empty when it is allegedly used to denote something that is both existent in bodies and causally efficacious. What could it mean, on a Berkeleyan account, to attribute a dispositional property to a sensible object? And how are we supposed to understand the claim that something has a second-order dispositional property—a disposition to acquire further properties and to lose others? According to Berkeley, sensible objects are collections or bundles of ideas. As becomes clear from his discussions of the continuous existence of sensible things, of their intersubjective perceptibility and of perceptual error (i.e. his account of the difference between appearance and reality),35 Berkeley takes these collections to comprise more than the ideas of a single finite being at a given time. In his attempt to account for the relativity of perception (i.e. the fact that things look different in different conditions) and perceptual error (i.e. the fact that we can be mistaken about which sensible qualities an object we perceive actually has) Berkeley admits that, on his account and strictly speaking, no sensible object can be perceived at different times, or by different persons, or through different senses. If an oar in the water looks crooked, but looks straight the very next moment when pulled up, we are strictly speaking dealing with two different sensible objects. If the oar, still under water, looks crooked but feels straight to the touch, we are again dealing with two different objects—an object of sight and an object of touch. As one has immediate access only to one’s own ideas, no two persons can, strictly speaking, perceive the same sensible thing. Berkeley recaptures the trans-personal, trans-temporal and trans-sensual identity of sensible objects by distinguishing between a strict philosophical and an ordinary way of speaking and by accounting for the difference in terms of the systematic relations that hold between sensible objects in the strict sense. The bundles of ideas that constitute sensible objects in the ordinary non-philosophical sense are so to speak collections of sensible objects in the strict philosophical sense. A sensible object (e.g. a tree) is a collection of ideas, which contains all the ideas of sense which God causes and has caused in every finite being who perceives or has perceived the object in question. The reason why we take all these ideas to be one thing lies in the relations they bear to each other as well as to further ideas within the whole system

32

See also DM 8, 11. See e.g. DM 3, 30–31. 34 See e.g. DM 34. 35 See Saporiti (2006, Chap. VI), for a discussion of Berkeley’s account of perceptual error and the difference between appearance and reality. 33

220

K. Saporiti

of ideas God causes in finite beings. These are but relations of systematic co-variation in successions of ideas. Furthermore, every collection of ideas which forms a sensible object will contain some of God’s ideas. For only God’s perceiving a sensible thing guarantees its continuous existence. To attribute an imperceptible property (e.g. a dispositional property or power) or a quality not yet perceived to such a collection is to make a prediction about future ideas of sense or to affirm the truth of contra-factual conditionals. It means making a judgement about the systematic relations that obtain between the ideas of sense we (finite beings) have had and do have, on the one hand, and ideas of sense we (finite beings) are going to have in the future (or are likely to have in the future) as well as about ideas we would have had or would have under certain circumstances (that is, the occurrence of certain other ideas of sense). Such a prediction will always involve ideas that belong to many more than just a single collection of ideas. To what extent can Berkeley allow for the attribution of unperceived and imperceptible qualities to perceivable objects? As there is no such thing as an unperceived quality in Berkeley’s ontology, it seems as if it were not possible for him to take any such attribution into account. But sensible objects in the ordinary sense do have continuous existence (because God perceives them) and experience justifies our taking objects we perceive at different times as continuous individual objects. Experience will also support predictions about qualities an object would have or will acquire under certain circumstances. As long as we understand attributions of imperceptible and yet unperceived qualities to be such predictions, they can be accounted for even on Berkeleyan terms. What Berkeley would certainly have to deny is the correctness of a claim about the actual existence of anything imperceptible or unperceived in the object in question. The object itself is just a collection of perceived qualities (ideas that have been perceived or are perceived currently). A sensible object contains nothing conceptually or causally entailing its own existence or constitution at any future moment.

4 David Hume: Scepticism About Powers Being an empiricist, Hume, just like Locke and Berkeley, asks himself how we can derive the idea of an active, causally efficacious power from experience. Locke believed that we immediately and with certainty infer the existence of powers in the things which we perceive as undergoing or effecting changes. Though powers (like dispositional qualities or potentialities) are not perceivable by themselves, we witness their being exercised or actualized. Locke, as we have seen, admitted that it is only by reflection (by paying attention to what is going on in our minds) that we get a distinct idea of an active power. Observing changes in bodies will only provide us with a confused idea of an active power. Berkeley went even further and insisted that the idea of an active power in a sensible thing is vacuous. As only the experience of the voluntary production of ideas can afford us with an idea of

Potentiality in British Empiricism

221

causation, Berkeley argued that only spirits have powers and are able to cause something. Hume seems to be even stricter in his empiricist approach and points out that neither sensation nor reflection provides us with anything from which we can derive an idea of an active, causally efficacious power. According to Hume, all our ideas are ultimately derived from impressions either of sense or of reflection. For this reason we have to look for the impressions from which our ideas of “efficacy, force, energy, necessity, connexion, and productive quality” are taken.36 But, as Hume observes, we never witness the actual production of anything. Whether we perceive changes in bodies or changes in our minds, the only things we perceive are successions of single impressions or ideas.37 Hume’s search for an experience of genuine “production, where the operation and efficacy of a cause can be clearly conceiv’d and comprehended by the mind” (T 1.3.14.2, SB158) is of no avail. Whether we consider changes that we perceive in bodies or the voluntary movements of our own body or the voluntary performance of mental actions, all we ever come across are distinct impressions and ideas, none of which contains anything providing a necessary connection between it and other impressions or ideas. We will observe our arm moving when we want it to move, and we will succeed in voluntarily forming an idea, but all we perceive is the wish to move our arm followed by its movement and the wish to form an idea followed by the idea. Neither the volition to move our arm nor our wish to form an idea reveals anything actually bringing about or producing the movement of our arm or the occurrence of a certain idea. Thus, Hume concludes: In short, the actions of the mind are, in this, respect, the same with those of matter. We perceive only their constant conjunction; nor can we ever reason beyond it. No internal impression has an apparent energy, more than external objects have. Since, therefore, matter is confess’d by philosophers to operate by an unknown force, we shou’d in vain hope to attain an idea of force by consulting our own minds. (T 1.3.14.12, SB 633, App.)

As an empiricist Hume acknowledges the fact that experience teaches us how matters stand but can never teach us whether they stand so necessarily or ought to stand as they do. At the same time he believes that all our knowledge of matters of fact that reaches beyond our immediate experience is based on inferences we draw from causes to effects or from effects to causes. If we infer the existence of something we do not perceive, we do so because we take it ultimately to stand in a

Hume declares these terms to be “nearly synonimous” and holds that “therefore ‘tis an absurdity to employ any of them in defining the rest” (T 1.3.14.4, SB 157). “T” is used as an abbreviation for Hume’s A Treatise of Human Nature (1739–1740), followed by number of book, part, section, and paragraph, as numbered in the Clarendon edition. In addition, page numbers of the Selby-Bigge edition (SB) are given. 37 Hume uses the term “perceptions” to denote objects of the understanding (much like Locke used “idea”). Perceptions comprise both impressions and ideas. Ideas, at least all simple ideas, are copies (resemblances) of impressions which enter the mind through sensation or reflection and which are represented by these ideas: “all our simple ideas in their first appearance are deriv’d from simple impressions, which are correspondent to them, and which they exactly represent” (T 1.1.1.7, SB 4). 36

222

K. Saporiti

causal relation to something we do perceive. This is why Hume takes questions “concerning the power and efficacy of causes” (T 1.3.14.2, SB 156) to be of great importance. Like Locke and Berkeley, Hume believes that causation implies some kind of necessitation of the effect. It is therefore of importance to discover the origin of our idea of necessity. In the eyes of these three authors our idea of causation and the necessary connection between causes and their effects is intimately connected with our ideas of powers and forces as inhering in the things that are held to be causes. Hume is concerned with the claim that a causal power is a “quality, which binds the effect to the cause, and renders the one an infallible consequence of the other” (EHU 7.6, SBN 63),38 and with the claim that “the efficacy of causes” lies in “that quality which makes them be follow’d by their effects” (T 1.3.14.3, SB 156). At the same time Hume observes “[t]hat there are no ideas, which occur in metaphysics, more obscure and uncertain, than those of power, force, energy, or necessary connexion” (EHU 7.3, SBN 62). His investigation of our impressions, undertaken to clarify these ideas, yields the following conclusion: All ideas are deriv’d from and represent impressions. We never have any impression, that contains any power or efficacy. We never therefore have any idea of power. (T 1.3.14.11, SB 161)

Thus, according to Hume, the origin of our idea of the necessity connecting causes and their effects does not lie in an impression we receive from the things we take to be causes or effects. There is nothing in these causes and effects as taken by themselves that binds them together. The impression from which we take our idea of a necessary connection between cause and effect is an impression of reflection. It arises once the mind has formed a habit of moving from a given idea to another one. The mind forms such a habit as it becomes accustomed to the experience that ideas similar to the former are followed by ideas similar to the latter. For after a frequent repetition, I find, that upon the appearance of one of the objects, the mind is determin’d by custom to consider its usual attendant, and to consider it in a stronger light upon account of its relation to the first object. ‘Tis this impression, then, or determination, which affords me the idea of necessity. (T 1.3.14.1, SB 156)

In reaction to the repeated experience that events of one type are followed by events of another type the mind develops a tendency to transfer its attention from one idea to another one customarily associated with it. Our imagination is so to speak taught by experience to form certain ideas whenever it entertains certain other ideas. The repeated experience of one type of event being followed or preceded by another type of event leads us to believe that an event of the latter type will occur or

“EHU” is used as an abbreviation for Hume’s Enquiry Concerning Human Understanding (1748), followed by number of section and paragraph as given in the Clarendon edition. In addition, page numbers of the Selby-Bigge edition, revised and with Notes by P. H. Nidditch (SBN) are given. 38

Potentiality in British Empiricism

223

has occurred whenever an event of the former type occurs. Hume calls the habitual movement of the mind from one idea to another a “transition” and takes it to be performed immediately, without any intermediate step.39 The upshot of this explanation of the origin of our idea of necessity is that the inference from causes to their effects and from effects to their causes is a matter, not of reason, but of custom and habit. It seems misleading, however, to say that Hume denies the existence of “real causes and effects”, or that he reduces the relation between cause and effect to a relation of constant conjunction or succession. Contrary to Malebranche, in whose view God is the only real cause, and contrary to Berkeley, who only allows for the existence of spiritual agents, Hume does not deny the existence of causation in a world of finite or material beings. Rather, he believes that we are wrong in supposing, as we often do, that we are able to perceive the actualization or exercise of causal powers.40 According to Hume, causes and effects do stand in a necessary relation to each other. It is only when we mistake our idea of necessity to arise from impressions received from the things taken to be causes or effects themselves that we arrive at a wrong conception of the relation between cause and effect. The idea of a necessary connection between things we know from sensation is confused, and it is impossible to make this idea more intelligible for the simple reason that there is no impression from which it is derived. Hume thinks that there is no clear way of finding out what a necessary connection supposedly holding in virtue of unperceivable qualities (powers) of the things perceived might amount to. To be sure, things may have qualities we do not know of, but these will not be of any help in explaining causation, nor are they what creates the problem, according to Hume. I am, indeed, ready to allow, that there may be several qualities both in material and immaterial objects, with which we are utterly unacquainted; and if we please to call these power or efficacy, ‘twill be of little consequence to the world. But when, instead of meaning these unknown qualities, we make the terms of power and efficacy signify something, of which we have a clear idea, and which is incompatible with those objects, to which we apply it, obscurity and error begin then to take place, and we are led astray by a false philosophy. This is the case when we transfer the determination of the thought to external objects, and suppose any real intelligible connexion betwixt them; that being a quality which can only belong to the mind that considers them. (T 1.3.14.27, SB 168)

The clear idea which we can make the terms “power” and “efficacy” signify is the idea of the determination of the mind to form specific ideas (if it perceives other ideas of a specific kind) and to believe in the existence or occurrence of specific things. It is when we project the fact that the mind is so determined into the things

39

For a lucid account of Hume’s notion of causation and remarks on some of the problems of interpretation that it gives rise to, see Bell (2009). 40 “The generality of mankind never find any difficulty in accounting for the more common and familiar operations of nature; such as the descent of heavy bodies, the growth of plants, the generation of animals, or the nourishment of bodies by food: But suppose, that, in all these cases, they perceive the very force or energy of the cause, by which it is connected with its effect, and is for ever infallible in its operation” (EHU 7.21, SBN 69).

224

K. Saporiti

we take to be causally connected, thereby assuming them to possess qualities or powers which are responsible for their being connected the way they are, that we are led astray. As we have seen, Hume thinks that the idea of a cause (taken as something which is necessarily linked with its effect) is often linked to the idea of a power, force or energy producing the effect.41 What, then, are we to make of powers, potentialities and potentials if we follow Hume in supposing that the necessary connection between causes and effects is established by a perceiving mind rather than by something which inheres in the relevant things themselves? Hume famously defines a cause as “an object, followed by another, and where all the objects, similar to the first, are followed by objects similar to the second […] in other words where, if the first object had not been, the second never had existed” and as “an object followed by another, and whose appearance always conveys the thought to that other”.42 Among the consequences of his account of causation and necessity Hume mentions the claims that there is no difference between a cause and an occasion and that there is no difference between a power and its exercise.43 This has led scholars to conjecture that according to Hume there are no powers or that everything merely has the power to do exactly what it does do. In giving his “definitions” Hume just seems to describe instances of causation in such a way as to identify the impressions from which our idea of causation arises. As Bell points out, “[t]he relations between objects mentioned in the definitions, namely succession, resemblance, and habitual association, are all observable in the sense that they can be perceived or felt” (Bell 2009, 165). In a case where we rightly take something to be a cause there seems to be no room for a distinction between its causing something and the actualization of its causal power. Further, according to Hume’s definition, something is a cause of something else just in case it is always and without any exception followed by the latter—or rather, just in case all objects of the kind the cause belongs to are without exception followed by objects of the kind the effect belongs to. This feature of the Humean account captures the idea of necessity that he takes to be involved in our idea of a cause. But it also seems to make it impossible to distinguish between cause and occasion. For something can only be a cause if its mere occurrence or existence will always prompt its effect. How does this prevent us from attributing powers or dispositional qualities to objects in situations where these powers or qualities are not actualized or exercised?

41

Cf. T 1.3.14.2, SB 156; T 1.3.14.4, SB 171; EHU 7.3, SBN 62. EHU 7.29, SBN 76–77; see also T 1.3.14.31, SB 170. There is an ongoing debate about how these two definitions relate to each other, cf. Bell (2009). I will here assume without further argument that Hume did not offer two different accounts but tried to capture different aspects of his complex account of causation. For an object to qualify as a cause in Hume’s sense it will presumably have to meet both conditions. 43 “[…] we must reject the distinction between cause and occasion, when suppos’d to signify any thing essentially different from each other” (T 1.3.14.32, SB 171). “The distinction, which we often make betwixt power and the exercise of it, is equally without foundation” (T 1.3.14.34, SB 171). 42

Potentiality in British Empiricism

225

By ascribing to an object the power to cause something we presumably want to say that we believe the object to be of the same kind as, or to be similar to, objects which in the past actually did cause the kind of thing the power in question is supposed to cause. The difficulty seems to lie in this question: on what basis may we ascribe a power, or in which respects will an object to which one wishes to attribute a certain power have to resemble certain other objects? No object can belong to the kind of objects referred to in the definition of a cause without actually being a cause, that is, without exercising its causal power. For the definition of this kind involves that the existence of objects belonging to this kind is necessary and sufficient for the occurrence or existence of their effects. Hume’s attempt to tie the sense of necessity connected with our idea of causation to something we can learn from experience seems to lead him into scepticism about powers and potentialities. From Hume’s point of view, there is no way of conceiving or making sense of a power of an individual entity to cause something. Thus upon the whole we may infer, that when we talk of any being, whether of a superior or inferior nature, as endow’d with a power or force, proportion’d to any effect; when we speak of a necessary connexion betwixt objects, and suppose that this connexion depends upon an efficacy or energy, with which any of these objects are endow’d; in all these expressions, so apply’d, we have really no distinct meaning, and make use only of common words, without any clear and determinate ideas. (T 1.3.14.14, SB 162)

References Alexander, P. (1985). Ideas, qualities and corpuscles. Locke and Boyle on the external world. Cambridge: Cambridge University Press. Atherton, M. (2007). Locke on essences and classification. In L. Newman (Ed.), The Cambridge companion to Locke’s ‘Essay concerning human understanding’ (pp. 258–285). Cambridge: Cambridge University Press. Ayers, M. (1977). Ideas of power and substance in Locke’s philosophy. In I. Tipton (Ed.), Locke on human understanding (pp. 77–104). Oxford: Oxford University Press. Bell, M. (2009). Hume on causation. In D. F. Norton & J. Taylor (Eds.), The Cambridge companion to Hume (2nd ed., pp. 147–176). Cambridge: Cambridge University Press. Berkeley, G. (1710). A treatise concerning the principles of human knowledge. Dublin. Berkeley, G. (1713). Three dialogues between Hylas and Philonous. London. Berkeley, G. (1721). De Motu. London. Berkeley, G. (1948–1957). In A. A. Luce & T. E. Jessop (Eds.), The works of George Berkeley, bishop of Cloyne (Vol. 9). Edinburgh: Nelson. Campbell, J. (1980). Locke on qualities. Canadian Journal of Philosophy, 10, 567–585. Chapell, V. (2007). Power in Locke’s Essay. In L. Newman (Ed.), The Cambridge companion to Locke’s ‘Essay concerning human understanding’ (pp. 130–156). Cambridge: Cambridge University Press. Cummins, R. (1975). Two troublesome claims about qualities in Locke’s Essay. Philosophical Review, 84, 401–418. Downing, L. (2007). Locke’s ontology. In L. Newman (Ed.), The Cambridge companion to Locke’s ‘Essay concerning human understanding’ (pp. 352–380). Cambridge: Cambridge University Press.

226

K. Saporiti

Hume, D. (1739–1740). A treatise of human nature (Ed. by D. F. Norton & M. J. Norton 2000. Oxford: Clarendon Press). Hume, D. (1748). An enquiry concerning human understanding (Ed. by T. L. Beauchamp 2000. Oxford: Clarendon Press). Jacovides, M. (2007). Locke’s distinction between primary and secondary qualities. In L. Newman (Ed.), The Cambridge companion to Locke’s ‘Essay concerning human understanding’ (pp. 101–129). Cambridge: Cambridge University Press. Locke, J. (1690). An essay concerning human understanding (Ed. by P. H. Nidditch 1975. Oxford: Clarendon Press). Locke, J. (1823, repr. 1964). The works of John Locke (Vol. 10). New edition, corrected. London. Mackie, J. L. (1976). Problems from Locke. Oxford: Clarendon Press. Mandelbaum, M. (1964). Locke’s realism. In Philosophy, science, and sense perception. Historical and critical studies. Baltimore: Johns Hopkins University Press. McCann, E. (2007). Locke on substance. In L. Newman (Ed.), The Cambridge companion to Locke’s ‘Essay concerning human understanding’ (pp. 157–191). Cambridge: Cambridge University Press. Newman, L. (Ed.). (2007). The Cambridge companion to Locke’s ‘Essay concerning human understanding’. Cambridge: Cambridge University Press. Saporiti, K. (2006). Die Wirklichkeit der Dinge. Eine Untersuchung des Begriffs der Idee in der Philosophie George Berkeleys, Philosophische Abhandlungen, Band 93, Frankfurt am Main: Vittorio Klostermann. Specht, R. (2011). Das Allgemeine bei Locke. Berlin, Boston: Walter de Gruyter. Stuart, M. (2003). Locke’s colors. Philosophical Review, 112, 57–96. White, A. R. (1975). Modal thinking. Oxford: Blackwell. Wilson, M. (1979). Superadded properties: The limits of mechanism in Locke. American Philosophical Quarterly, 16, 143–150. Yolton, J. W. (1970). Locke and the compass of human understanding. Cambridge: Cambridge University Press.

Author Biography Katia Saporiti holds the Chair for Philosophy and the History of Philosophy at the University of Zurich, Switzerland. She works on British Empiricism and problems in epistemology and the philosophy of mind. Her publications include a book on George Berkeley (“Die Wirklichkeit der Dinge. Eine Untersuchung des Begriffs der Idee in der Philosophie George Berkeleys”, Klostermann 2006) and a book on the language of thought (“Die Sprache des Geistes”, De Gruyter 2004).

Part IV

The Metaphysics of Potentials

Real Potential Jennifer McKitrick

There’s a student in my philosophy class who has “real potential.” I might express this thought in any of the following ways: “She is potentially a philosopher”; “She is a potential philosopher”; “She has the potential to be a philosopher.” The first way uses a cognate of “potential” as an adverb to modify “is.” The second ways uses “potential” as an adjective to modify “philosopher.” However, the third way uses “potential” as a noun to refer to something that the student has. What kind of thing is this potential? One worry about even asking this question is that this nominalization of the adjective “potential” suggests a metaphysical picture that is an artifact of language. This is even more strongly suggested by the less ambiguous nominalization “potentiality.” Once we have the term “potentiality,” we have a new kind of entity to countenance, and questions about its nature arise. One might argue, just because we use the word “potentiality,” we should not think that it refers to a “thing” that someone can “have.” There is something disingenuous about such an argument. It proceeds as if the adverbial and adjectival uses of “potential” are unproblematic, and questions only arise with the nominalization. But it is not obvious what it means to potentially be something, or what it means to be a potential something. To say that someone “is potentially” a philosopher is to talk about a way of being that falls short of actuality.1 And a “potential philosopher” is not a kind of philosopher at all. So what is it? Each of the three above formulations is a modal claim. If there is anything philosophical puzzling about a potentiality claim, it is not going to go away by translating it into an equivalent modal claim. 1

This is not to say that a potentiality statement cannot be accounted for in terms of the actual; however, such statements would require some analysis along the lines of other “mere” possibility statements.

J. McKitrick (&) Department of Philosophy, University of Nebraska, Lincoln, USA e-mail: [email protected] © Springer Science+Business Media B.V., part of Springer Nature 2018 K. Engelhard and M. Quante (eds.), Handbook of Potentiality, https://doi.org/10.1007/978-94-024-1287-1_9

229

230

J. McKitrick

In this chapter I defend the existence of potentialities against anti-realist arguments, and make a proposal as to their nature.2 The proposal, in short, is that potentialities are properties, specifically dispositions, though more needs to be said about properties and dispositions. I will do this in Part I. In Part II, I will address two lines of argument against potentialities: that they are reducible, and that they are causally inert.

1 Properties, Dispositions, and Potentialities 1.1

Properties

Properties are ascribed to things by predicates, such as “is red.” We need not assume that there is a one-to-one correspondence between properties and predicates. There can be some predicates that have no corresponding property, like “being non-self-instantiating,” and there are most likely properties that do not have any corresponding predicate, if there are undiscovered properties that we have no terms for. I will consider three competing metaphysical views of properties: sets of possibilia (Lewis 1983), universals (Armstrong 1978), and sets of tropes (Williams 1953). One approach to thinking about properties is to start with particular things—the things that have the properties. We often group things according to their shared properties. So, one way to think about properties is in terms of these groupings. I call this an extensional approach to properties since properties are identified with a set of particulars, or an extension. For example, redness is the set of red things. However, since sets are individuated by their members, formulating the view this way has the implication that redness would be a different property if there were different particular red things. But that seems wrong. So, if redness is a set, it is better to think of it as the set of all the possible red things. This is essentially what David Lewis does. According to Lewis, a property is any class or set of actual or possible entities. One property consists of all the red things in the actual world, and all the possible red things in all of the possible worlds. Another property consists of my pencil, Bill Clinton, and some possible apple in some possible world. On this view, properties are extremely abundant, and there need not be any similarity among their members. However, for any class of entities, there is just one property; triangularity and trilaterality turn out of be the same property on this view. So, this account is inclusive, but also coarse-grained. Entities can be grouped in a large variety of ways. Some of these classes will be arbitrary groupings. Other classes will group entities that are similar in some natural respect. Given the abundance of sets of possibilia, the majority of properties will consist of random, miscellaneous things,

Some of these ideas are discussed in my “Dispositions and Potentialities” (2014).

2

Real Potential

231

and the classes with members that are naturally similar will be an elite minority. Classes of similar entities are importantly different from classes of dissimilar entities, and so there needs to be a way to distinguish these different types of properties. Lewis takes the predicate “—is a natural class” to be primitive, and applicable to some properties, such as the class of red things, but not to others. Implicit in this view is a notion of natural kinds; natural properties correspond to real distinctions in nature. Arbitrary groupings of dissimilar entities, such as the set {my pencil, Bill Clinton, a non-actual apple} are non-natural, disjunctive properties. Between these two extremes, Lewis allows that naturalness can be a matter of degree. For instance, the property of ‘being blue or green’ seems more natural than the property of ‘being blue or square’. While David Armstrong presents a competing account of properties, he can agree with Lewis that there are a great many classes of entities, and that some of these classes are special in that their members are similar to one another. But Armstrong does not call such classes “properties.” He would say that members of natural classes literally “have something in common”—they each instantiate the same universal. Armstrong reserves the expression “property” for these universals. Just as Lewis’ “natural properties” correspond to natural kinds, Armstrong’s universals correspond to the true natural kinds that would be discovered by a perfect science. Since these are an elite minority, Armstrong’s minimalist view is that properties are sparse. Many of our predicates do not correspond to real properties. A substantial difference between Lewis and Armstrong is how to characterize this elite set of classes of entities. According to Lewis, we can characterize the “elite” classes via the primitive predicate “—is a natural class.” According to Armstrong, in order to characterize this set we need to bring in the notion of universals. A third view of properties is that they are sets of tropes (Williams 1953). In the language of universals and extensional theories described above, a trope is a particular instance of a property, or the aspect of a particular in virtue of which it is similar to other things in the same natural class. The page that you are reading right now is white (I suppose). That is a property its shares with every other page in this chapter. However, we can refer to this page’s whiteness, as distinct from the next page’s whiteness, as the white “trope” that this page has. The whiteness that is common to all the pages in the chapter is a set of white tropes. As with Lewis’ view, certain sets of tropes are special because they have some irreducible similarity relation to one another. Some versions of trope theory call these sets “properties.”

1.1.1

Second-Order Properties

In addition to natural properties, another type of property that is important to this discussion is that of a second-order property. A second-order property is the property of having some property or other which meets some specification. The having of a first-order property, by contrast, does not necessarily involve the having of another property. For example, suppose that color properties, such as ‘being red,’ or ‘being blue,’ are first-order properties. A second-order property, then, would be

232

J. McKitrick

‘being colored’—a property that a thing has just in case it has some color property or other. ‘Being red’ or ‘being blue’ are different ways of ‘being colored.’ In other words, ‘being red’ and ‘being blue’ are determinates of the determinable ‘being colored.’ Determinables are a kind of second-order property (Yablo 1992; Funkhauser 2006), a property that a thing has in virtue of having any one among a number of more specific, determinate properties. If having a potentiality is a matter of having certain other properties, then potentialities would be second-order properties, and their ontological status would be linked to that of second-order properties. A so-called second-order property might actually be third-order, or higher-order. Perhaps a color property such as ‘being green’ is not a first-order property at all, but a property that an object has just in case it has some surface-reflectance property or other. Then, the property of being colored would be a third (or higher) order property, that is, the property of having some property. To avoid making assumptions about the basicness of any particular property, the expressions “first-order property” and “second-order property” can be thought of as relative terms. Some potentialities are likely to be higher-order properties. For example, if the potential to be a philosopher is a matter of having other properties such as intelligence and inquisitiveness, but these properties are also possessed in virtue of possessing other properties, then the potential to be a philosopher would be at least third-order. Armstrong uses the expression “second-order properties” in a different sense, to refer to properties which properties have. On the universals account of properties, universals can have things in common. For instance, some universals are relations. So, ‘being a relation‘ is a property that properties can have. These are what Armstrong calls “second-order properties.” Note that this is different from a property that a particular has in virtue of having another property. (While a thing has the second-order property ‘being colored’ in virtue of being red, in Armstrong’s terminology, one might say that redness is a first-order property that has the second-order property of being a color.) On a sparse universals theory such as Armstrong’s, the reality of second-order properties of particulars is called into question. If universals are sparse, we need not posit universals that things have in virtue of having other universals. It would not be in keeping with a minimalist account to say that a red thing instantiates universals for being colored, being red, being a certain shade of red, in addition to being a certain surface reflectance. There might be just one universal that makes all of these predications true—they might all have the same “truth-maker.” A thing’s so-called “first-order” property might be the only truth-maker we need for all of our predications of it. According to trope theory, second-order properties would be sets of tropes. So “being colored” would be the set of all the sets of color tropes. A particular trope would be a color trope in virtue of being a red trope, say. That seems to cash out a sense in which something could have a second-order property, being colored, in virtue of having a first-order property like being red. On an extensional account of properties, a given property is not essentially first, second, or higher order. “Being red, orange, yellow, green, blue, indigo, or violet”

Real Potential

233

seems like a disjunctive first-order property. However, if it has the same extension as the apparently second-order property “having some color property or other,” then it is the same property. Likewise, “the property of having property P” would seem to be a second-order property. However, it would necessarily have the same extension as property P, an apparently first-order property. So “P” and “the property of having P” are but two names for the same property. On this account, it seems that the “order” of the property is a function of the way it is characterized, rather than something about the nature of the property itself. Therefore, it makes more sense to talk about first and second-order characterizations of properties for an extensional account. However, one can still do some justice to the idea of a second-order property by talking about properties whose canonical characterization is second-order. This might be the case if the most salient point of resemblance between members of a class is the possession of various properties which meet a certain specification.

1.1.2

Real Properties

Now that the major metaphysical views about properties are on the table, we can return to questions of existence and reality. What does it mean for a property to exist, or to be real? According to Armstrong, it is a matter of the existence of a universal. But how do we know which universals exist? Armstrong accepts an instantiation condition, so a universal does not exist unless some actual particular instantiates that property. But that does not settle the matter, for it is not obvious which properties are instantiated. A number of things may appear to be jade, but are not instantiating the same universal if some of them are nephrite and others of them are jadeite. The real properties of things are the subject of scientific investigation, not something we can discern from ordinary predicates and folk-descriptions of things. A consequence of this view is that many of our predicates do not correspond to real properties. Given the underlying differences in things that appear to be the same shade of red, it is unlikely that “red,” let alone “the potential to be a philosopher,” picks out a genuine property. However, if there are any fundamental physical potentialities, like electric charge for instance, then a universals account could allow that such potentialities exist. On trope theory, a property exists whenever you have a set of similar tropes. Since a set can be a singleton, the existence of any trope entails the existence of a property. However, trope theory as such does not indicate which tropes exist. If a predicate is aptly applied to something, there is nothing about trope theory per se to tell us there is no corresponding property. Of course, one could have a minimalist trope theory, akin to a minimalist universals theory. But if it is true that one way for something to be is for it to have the potential to be something else, then a potentiality trope, and a potentiality, exist. Lewis’ extensional theory of properties is the most liberal with respect to property existence. On this view, for any set of possible things, there is a property. Furthermore, the things need not actually exist (in this world). This view has no

234

J. McKitrick

effective instantiation condition, which enables us to talk about properties that nothing actually has. The only “properties” that do not exist are sets that have no possible members. The potentiality to be a philosopher is a property because it is possible that there are some people who could be philosophers in the future. Since properties are so easy to come by on an extensional account, perhaps when one asks “are potentialities real?” she is really interested in whether potentialities are “natural,” and whether any of their instances (members) exist in the actual world. The question becomes: Do any actual things have potentialities, and if so, do they share a certain natural respect of similarity? The answer to that might depend on not only on the nature of potentiality and the particular potentiality in question, but also on one’s understanding of what it means to be natural. If being natural is a matter of degree, the equation of “natural” with “real” has the odd implication that reality comes in degrees. To summarize, none of the three metaphysical views about properties considered rule out the existence of potentialities. On Armstrong’s theory of universals, potentialities, like other properties, would be sparse at best. Trope theory would offer the less-than-helpful suggestion that there are potentialities if there are potentiality tropes. On Lewis’ extensional view, potentialities, like other properties, would be abundant, though not necessarily natural. So, anti-realism about potentialities does not follow from major approaches to the metaphysics of properties. But before I examine more targeted arguments against potentialities, I need to say more about them. But since I am giving a dispositional account of potentialities, I need to say more about dispositions first.

1.2

Dispositions

When someone has a disposition, he or she is prone to act in certain ways in certain circumstances. A cowardly person is disposed to flee from danger, for example. A sociable person is disposed to seek the company of others. Physical objects can also have dispositions. Fragile objects are disposed to break when struck. Elastic objects are disposed to stretch when pulled. As with ordinary language, science, too, is rife with dispositions talk. A substance is volatile or reactive if it is disposed to enter into reactions with other substances. A material is conductive if it is disposed to transmit an electric charge. An element is unstable if it is disposed to decompose. An organism is fit for a certain environment if it is disposed to survive in that environment. Numerous synonyms and near-synonyms, such as “power,” “capacity,” “tendency,” and “predisposition” expand the range of dispositional locutions even further. It is widely acknowledged that dispositions-talk is ubiquitous. However, it is a matter of some controversy how this talk is to be understood. On a realist account, a disposition is a property, or a quality that things can have, like redness or solidity. A disposition has a characteristic manifestation. For example, the characteristic manifestation of fragility is shattering, and the characteristic manifestation of

Real Potential

235

cowardliness is avoidance of danger. The manifestation need not occur for the object to have the disposition. A glass can be fragile even if it never shatters. A manifestation of a disposition occurs (at least typically) when the object with the disposition is subject to certain circumstances. The fragile glass shatters when it is struck. The circumstances in which the manifestation occurs are called “the circumstances of manifestation.” They include not only the salient “trigger” for the manifestation (the striking), but also the necessary background conditions, such as ambient temperature and gravitational forces. In some cases, the manifestation will occur at the location of the disposed object itself. In the case of the fragile glass, the shattering occurs where the glass is. However, in other cases, the “locus of manifestation” may be elsewhere. For example, if something is provocative, the manifestation will occur in the thing provoked. Given that a disposition is associated with a manifestation, and with circumstances which trigger the occurrence of this manifestation, there is a natural association between a statement attributing a disposition to a thing and a certain conditional statement: If the conditions were to obtain, the manifestation would occur. For example, the statement “This glass is fragile” bears some important relation to the statement “If this glass were struck, it would shatter.” An attribution of a disposition to some object licenses inferences about what will happen in various circumstances. These inferences may be defeasible, but the ability to make these inferences is what makes dispositions talk so useful, if not indispensable. We frequently have pressing reasons to be concerned about predicting what things will do in various circumstances. It is important to know what is poisonous and what is nutritious, which animals are aggressive, and which situations are dangerous. We are interested to predict the behavior of our fellow human beings, and so describe them as friendly, hostile, irritable, shy, ambitious, trust-worthy, and so on. Disposition ascriptions are an important means of communicating our understanding of what to expect from the things in our environment. Furthermore, when an object is disposed to behave in a certain way in certain circumstances, it is often thought that there is something about the object “in virtue of which” this is so. The fragile glass is disposed to shatter when struck due to some feature of the glass. The silicon bonding in the glass is such that, if an excessive strain is placed on these bonds due to warpage of the glass, some of the bonds will break, starting a chain reaction of bonds breaking. Hence, the glass shatters when struck. This type of molecular bonding is the causal basis of the glass’s fragility. A disposition’s causal basis is a property of the disposed object which is causally efficacious for the disposition’s manifestation, if and when such manifestation occurs. Dispositions and causal bases do not typically have a one-to-one correspondence. Crystal wine glasses and egg shells are both fragile, but have very different structures and constituents. Presumably, these things have different properties in virtue of which they are prone to break when struck. That is to say: the causal basis of fragility in a wine glass is a different property than the causal basis of fragility in an egg shell. Or consider electrical conductivity and thermal conductivity, two different dispositions which have the same causal basis in metals.

236

J. McKitrick

Furthermore, there may be ungrounded dispositions that have no causal bases at all (McKitrick 2003b).

1.2.1

Dispositional and Categorical Predicates

Dispositional predicates are conceptually associated with manifestations, circumstances, and conditionals. Non-dispositional, or categorical predicates do not have these associations. As paradigm examples of categorical predicates, philosophers sometimes offer shape predicates. Arguably, to say something is square is not to say anything about what it would do in particular circumstances; ‘being square’ has no associated manifestation or triggering event. Arguably, categorical predicates do not have the relevant relation to conditionals.3 As Elizabeth Prior notes “dispositional ascription sentences possess a relationship to certain subjunctive conditionals not possessed by categorical ascription sentences” (Prior 1985, 62). Most philosophers acknowledge a distinction between dispositional and categorical predicates. Nevertheless, this characterization is rough, the nature of this conceptual association is left vague, and how to categorize a given predicate may be a matter of dispute. According to Goodman, for example, “almost every predicate commonly thought of as describing a lasting objective characteristic of a thing is as much a dispositional predicate as any other” (Goodman 1983: 41). The predicate “is square” seems categorical, but it is associated with certain conditionals, such as “if you try to put it in a round hole, it will not fit.” Some philosophers claim that the dispositional/categorical distinction ends there, as merely a distinction among predicates, which does not correspond to any interesting metaphysical distinction among properties. Sidney Shoemaker, for example, says “I think that the term ‘dispositional’ is best employed as a predicate of predicates, not of properties” (1980: 211). According to Shoemaker, what determines the identity of a property “is its potential for contributing to the causal powers of the things that have it” (1980: 212). Though Shoemaker would resist, one might want to describe his view by saying that all properties are dispositional properties. At the other extreme, David Armstrong claims that all properties are categorical properties (1996: 16). On Armstrong’s view, these categorical properties stand in Necessitation relations, thereby establishing lawful connections between instantiations of universals, and leaving no role for dispositions. He acknowledges that some properties are picked out by disposition terms, but claims that such terms simply provide us with a useful way of speaking of categorical properties. As Armstrong puts it, the dispositional/categorical distinction is a “verbal distinction that cuts no ontological ice” (1973: 15). 3

Some argue that all predicates are related to conditionals in the relevant way, undermining this way of distinguishing categorical from dispositional properties (Mellor 1982). One could reply by arguing that dispositional and non-dispositional predicates differ in the ways they are related to conditionals. However, I will not pursue such arguments here, for the claim that there are some non-dispositional predicates is not central to my main thesis that potentialities are dispositions.

Real Potential

237

Clearly, Armstrong thinks that the distinction between properties makes sense— he just thinks that one of the categories is empty, and so if a non-trivial distinction is wanted, then it must apply to predicates. Although Shoemaker wants to reserve “dispositional” as a predicate of predicates, he can agree with Armstrong that the distinction between properties makes sense—he just disagrees about which category is empty. Questions about whether the world contains dispositions, categorical properties, or both are metaphysical, not merely linguistic. Categorical properties, then, are properties which do not have the relevant associations with manifestations, circumstances, and conditionals. Along these lines, Stephen Yablo offers as an intuitive characterization, “a property is categorical just in case a thing’s having it is independent of what goes on in non-actual worlds” (1987: 306).4 Naturally, there is room for dispute as to which properties are categorical and which are dispositional. Hugh Mellor argues that even shape properties are dispositions. For example, he says that triangularity is the property of being disposed to be counted as three-angled (1974: 171). Similarly, Goodman says “a cubical object is one capable of fitting try squares and measuring instruments in certain ways” (1983: 41). I am not going to dispute these claims here; I am just trying to explain the distinction between dispositional and categorical properties, without taking a stand as to which (if any) specific properties fall into each category, or if the categories are exhaustive.

1.2.2

Marks of Dispositionality

To summarize the key points of this section so far, a disposition such as fragility can be characterized as follows. A fragile glass will shatter if you strike it hard enough. Fragility is the glass’s disposition, shattering is the manifestation of the disposition, and being struck is the circumstance of manifestation. The underlying cause of the glass’s shattering constitutes the causal basis of the glass’s fragility. The glass can remain fragile even if it never shatters. One can say of the fragile glass, with certain qualifications, that if it were struck, it would shatter. This characterization suggests certain “marks of dispositionality,” according to which a property is a disposition if it: (1) (2) (3) (4)

has a characteristic manifestation; is such that certain circumstances can trigger that manifestation; can be possessed without the manifestation occurring; is instantiated by things of which a conditional of the form “if it were subject to the circumstances, it would exhibit the manifestation” is generally true; and (5) can be accurately characterized with an expression of the form “the disposition to produce the manifestation in the circumstances” (McKitrick 2003a).

4

Yablo goes on to argue that this characterization is inadequate.

238

J. McKitrick

I take it that these conditions are jointly sufficient for dispositionality, but I am not committed to their being individually necessary. The association with conditional statements (my fourth mark) has received extensive discussion in the literature. Later, I will discuss numerous counterexamples to conditional analyses of dispositions. These counter-examples suggest that analyzing a disposition ascription in terms of a conditional statement will always be fraught with difficulties. One may question whether an association with a conditional is a mark of dispositionality at all. However, the fourth mark of dispositionality is carefully hedged. I claim that if a property is a disposition, then it is instantiated by things of which a conditional of the form “if it is subject to the circumstances, it exhibits the manifestation” is generally true, allowing for exceptions. That is to say, a thing might have the disposition in question even if the relevant conditional is not true of it. For example, “if you drop it, it will break” is not true of the carefully packed glass, but nevertheless, the glass is still fragile. The marks of dispositionality apply to property types, while typical counter-examples to conditional analyses feature particular property tokens or tropes. While there may be particular dispositional tropes of which the conditional is not true, that does not show that the general property does not bear the marks. So, even though a particular glass may have its fragility masked by careful packing, the general property ‘fragility’ still: (1) (2) (3) (4)

has a characteristic manifestation—breaking; is such that certain impacts can trigger that manifestation—breaking; can be possessed without breaking; is instantiated by things of which a conditional of the form “if it were struck, it would break” is generally true; and (5) can be accurately characterized with an expression of the form “the disposition to break when struck.” If a property was never instantiated by anything of which the relevant conditional were true, that would not be a clear case of dispositional property. But I do not claim that the truth of the conditional is necessary or sufficient for a property to be a disposition. If a property bears most of the marks, I claim that is some evidence that the property in question is a disposition. As an example of a disposition that lacks one of the other marks, consider stability, which:

(1) has a characteristic manifestation—maintaining structural integrity, or staying intact; (2) has circumstances of manifestation—various stresses to this structural integrity. (Note that “stability” will always be relative to certain kinds of stress—stable in wind, stable in earthquakes, etc.); (4) is instantiated by things of which a conditional of the form “if it were stressed, it would stay intact” is generally true; and (5) can be accurately characterized with an expression of the form “the disposition to stay intact when stressed.”

Real Potential

239

Note that the third mark, the possibility of possession without manifestation, is absent. Stability is a disposition that cannot be possessed without being manifest—a structure that is not manifesting stability is not stable. Others argue that dispositions like radioactivity manifest spontaneously and thus lack stimulus conditions (Vetter 2010). In those cases, the second mark—having circumstances of manifestation— would be absent. Perhaps these counterexamples seem to reveal the shortcomings of the marks of dispositionality. However, these marks do not constitute a decision procedure for determining whether a disposition is instantiated in a particular instance. Instead, their purpose is to help determine whether a general property is dispositional. Here, I use them to argue that potentialities are dispositions. Marks of dispositionality provide evidence that a property is a disposition, but do not constitute a reductive analysis of disposition ascriptions.

1.2.3

Accounts of Dispositions

Various accounts of dispositions have been offered which are consistent with these marks of dispositionality. On some views, dispositions are fundamental, irreducible powers (Molnar 2003). While such powers cannot be analyzed in terms of anything more basic, proponents of such views argue, contrary to Hume, that it is a concept that we acquire by experience, and that we can characterize roughly as outlined above. According to other views, to ascribe a dispositional predicate to something is tantamount to asserting a certain subjunctive conditional. For example, “x is fragile” is said to be true if and only if a conditional such as “if x were dropped, x would break” is true (Gundersen 2002; Choi 2006). According to a simple conditional analysis: x is disposed to exhibit manifestation M in circumstances C iff (if x were in C, x would exhibit M).

The idea that dispositions have causal bases suggests a second-order property view of dispositions, according to which having a disposition is having some property which would be causally efficacious for its manifestation. This can be more formally stated as follows: x has a disposition D to exhibit M in C iff x has some property P which is a causal basis for giving M in C (where property P is a causal basis of D iff P is a property of x which would be causally efficacious for M in C). (Johnston 1992: 230)

Lewis’s revised conditional analysis (1997) is also a second-order property view. The idea, somewhat simplified, can be put as follows: x has a disposition at time t to give M in C iff x has intrinsic property P, and if x were to be in C at time t and retain P, then P and C would cause M (1997: 157).

240

J. McKitrick

Manley and Wasserman’s “PROP” offers a conditional analysis, which relaxes the stringent requirement that a disposed object must manifest its disposition in its circumstances of manifestation, and counts an object as having a disposition as long as the disposed object manifests often enough. More formally: x is disposed to M when C iff x would M in a suitable proportion of C-cases (2008: 76).

It is not my objective to endorse and defend a particular account of dispositions here. I am concerned with these analyses, however, to the extent that they challenge a realist dispositional account of potentialities. Accordingly, I will argue against the idea that any of these accounts facilitate a reduction or elimination of dispositions or potentialities from our ontology. But first, I will make the case that dispositions are potentialities.

1.3

Potentiality

I began with the example of my student who has philosophical potential. Here are a few more examples of potentiality claims: An embryo is a potential person; a patient has the potential to be conscious; a caterpillar is potentially a butterfly; an acorn has the potential to be an oak tree. In general, when one says “x is potentially F,” “x is a potential F,” or “x has the potentiality to be (an) F,” ‘F’ refers to either a property that x can have, or a class or kind of which x can be a member. The occurrence or state of affairs, “x being F,” is the actualization of x’s potential to be F. One way for a thing to become a member of a kind is by acquiring certain properties that are characteristic of that kind. In that case, the two ways of being potentially F (where F is either a property that one can have or a kind of which one can be a member) come to essentially the same thing, slightly complicated by the fact that kind membership may require having more than one property. Another common use of the term “potentiality” that should be kept in mind might be called “epistemic potentiality.” When you think there is some chance that something has a certain property, you might say that it “potentially” has that property. For example, suppose your perfectly healthy friend, John, is in the next room and you are not sure whether he is asleep or awake. You might express your judgment about John by saying “John is potentially conscious.” But John is not like the comma patient, whose capacity for future consciousness is in question. So, presumably, you are not making a claim that you take for granted, that John, who is now sleeping, has the potential to be conscious in the future. Rather, you are saying that, as far as you know, John is conscious right now. This sense of potentiality would seem to have little to do with the relevant potentialities of a student, embryo or patient. However, it may be important to keep it in mind, lest we confuse our uncertainty as to whether something already has a certain property with the judgment that it could possibly acquire that property in the future. Perhaps this conflation is going on in Noonan’s “An Almost Absolute Value in History,” where he compares aborting a fetus to shooting into some bushes where

Real Potential

241

a hunter might be (Noonan 1970). In both cases, you could say “there is a potential person there.” However, in the case of the hunter, the potentiality is epistemic, whereas in the case of the fetus, on the most plausible interpretation of the claim “there is a potential person there,” the potentiality is metaphysical, not merely epistemic.

1.3.1

Potentialities Are Dispositions

Potentiality claims and disposition claims are importantly similar. To talk about what something has the potential or disposition to do is to make a claim about a future possibility—the “threats and promises” that fill the world (Goodman 1983: 41). Potentiality-talk and dispositions-talk are often interchangeable. When x is potentially F, one can say that x is disposed to be F, where “being F” is the manifestation of x’s disposition. Disposition ascriptions can likewise be put in terms of potentiality: If the bomb is explosive, it has the potential to explode; the fragile glass has the potential to break. Shoemaker’s classic paper uses the terms “disposition” and “potentiality” interchangeably (1980: 213). Granted, potentiality-talk and dispositions-talk are not perfectly interchangeable. You might say that an oak tree is potentially a table, but that it is not disposed to be a table. And you might have the potential to be a drug dealer, even if you would not say that you are disposed to be a drug dealer.5 Many disposition ascriptions in ordinary language suggest a stronger tendency, a higher probability of the manifestation occurring, than do the analogous potentiality ascriptions. However, such observations are consistent with potentialities being dispositions nevertheless. The ordinary-language connotation of many disposition claims may be that the manifestation has a high probability of occurring, but this is defeasible. Some particular instantiations of dispositions are unlikely to manifest. For example, a nuclear bomb has a disposition to explode, but let us hope that is unlikely to happen. Furthermore, it is a misnomer to talk of “the” disposition to so-and-so, for the manifestation alone does not uniquely specify a disposition. Two different dispositions could have the same manifestation, but not be the same disposition. Someone who gets red-faced whenever she is complemented is not manifesting the same disposition as he whose ruddy complexion reflects his ample consumption of alcohol. What differentiates these dispositions is that they have different stimuli, or circumstances of manifestation. Granted, there are some cases where the expressions “the potential to be F” and “the disposition to be F” pick out different properties. However, even in those cases, “the potential to be F” still picks out a disposition, since the expression “the disposition to be F” is ambiguous. Though the imperfect interchangeability of dispositions-talk and potentiality-talk is suggestive, a better way to gauge whether potentialities are dispositions is to

5

Thanks to the audience at my presentation of an earlier version of this paper at the University of Nebraska for these examples, especially Reina Hayaki for suggesting the right response.

242

J. McKitrick

consider whether they bear the marks of dispositionality. Consider an embryo’s potential for rationality: (1) It has a manifestation—being rational; (2) This manifestation will occur given certain (albeit very complicated to specify) circumstances of a favorable environment, nurturance, and so on; (3) An embryo can possess the potential to be rational without being rational; (4) A certain conditional is, other things being equal, true of the embryo: If a certain favorable environment and nurturance were to obtain, the embryo would become rational; and (5) It is not inappropriate to call the potential for rationality “the disposition to become rational.” It is also worth noting that, like many dispositions, potentialities often have causal bases. An embryo’s potential to be rational is not a brute, fundamental feature, but is presumably based on its genetic code and other biological factors. Given that potentialities are dispositions, it is not clear whether dispositions and potentialities are coextensive, or whether potentialities are a subset of dispositions. Perhaps potentialities are a distinctive subset because the locus of manifestation is always where the object that had the potentiality is located. Given that the manifestation of x’s potential to be F is “x being F,” it may seem as though the manifestation of x’s potentiality must occur where x is. When x manifests x’s potential to be rational, “being rational” happens where x is. This is not true of dispositions in general. A thing can be disposed to have an effect on something else: roses are disposed to smell sweet, provocative capes make bulls charge, and soporific lullabies put babies to sleep. In those cases, the locus of manifestation is not where the disposed object is. However, a potentiality’s manifestation is not always a matter of the individual who had the potential instantiating a property. One reason for this is because a thing can have the potential to become a different thing, changing its characteristics to such an extent that it is no longer the same individual anymore. Of course, whether this happens will depend on your views of diachronic identity, or what changes an individual can survive over time. Consider the acorn’s potentiality to be a tree. The manifestation of that potentiality is something ‘being a tree.’ But it is not clear whether the individual that is a tree now is numerically identical to the original acorn. Furthermore, you could burn the tree, so it has the potentiality to be a pile of ashes. But if that happens, arguably you do not end up with something that is at once the acorn, the tree, and a pile of ashes. To consider another example, when an unfertilized egg realizes its potential to develop into a rational being, it is not clear that the original ovum and the rational being are the same individual. Or consider a case of fission where, for example, one plant can have the potential to become several plants. If we divide and root several parts of the original plant, arguably, the original plant has ceased to be, and whatever potentialities it had are being realized by its descendants.

Real Potential

243

A way to deal with these sorts of cases is to say that, when a potentiality to be F is realized, either the individual who had the potential or its causal descendant is F. To make good on this proposal, one would have to cash out the notion of a “causal descendant,” which may prove difficult. Surely, one’s children are candidates for being one’s causal descendants. But while parents may try to live vicariously through their children, your daughter becoming a doctor would not realize your potential to be a doctor. Perhaps it is better to relinquish the notion that the locus of manifestation for a potentiality is where the potentiated individual is, for there are other sorts of counter-examples to this claim. For example, when you realize a potential to do something, such as score a goal, the effect that you have does not always happen where you are. Also, some people have the potential to be dangerous, funny, or annoying, but when someone manifests being dangerous, funny, or annoying, it is often someone else that is hurt, laughing, or annoyed. Having the potential to instantiate a relational property is a similar sort of case. Consider the potential to be President of the United States. When that potentiality is actualized, it is the person who had the potential that now has the property of being the President of the United States.6 However, since having that property depends on complicated historical and social relations that go beyond the bounds of the individual, it is not clear that the manifestation is entirely located where the individual who had the potential is. Another idea about how to distinguish potentialities from other dispositions is that potentialities cannot be possessed while manifesting. Generally, when a disposition manifests, the disposed object might continue to instantiate the disposition, or it might not. Sometimes, a thing loses a disposition when it manifests it, or even ceases to exist. The bomb is no longer explosive after it explodes. The match is no longer flammable once it has burned. The disposition, in effect, gets spent, and is no longer possessed. Elastic bands, on the other hand, are still elastic when stretched. Magnets are still magnetic even when they are manifesting their magnetism. What about potentialities? Does a person have the potential to be a person? Does a living thing have the potential for life? Perhaps, when potentialities have been and continue to be actualized, it is often no longer appropriate to say that they are possessed. However, there are exceptions. Someone may have the potential to grow, or to learn, realize those potentialities, and yet still have the potential to grow or learn. So, it is false to say that a potentiality is never both actualized and possessed. Sometimes, saying that an F is potentially F is misleading and inappropriate, but perhaps it is not false. In sum, potentialities are dispositions whose manifestation is a matter of the disposed individual acquiring a property and/or becoming a member of a kind. Typically, the manifestation occurs where the disposed individual is. And perhaps potentialities often cease to be instantiated once they are manifest. However, manifestations of dispositions also involve things acquiring properties. Furthermore, like dispositions, the manifestation of a potentiality can happen either

6

Thanks to Harry Ide for these examples.

244

J. McKitrick

where the potentiated object is or elsewhere, and potentialities can be possessed while their manifestations are occurring. Consequently, potentialities do not seem to have any essential characteristics that distinguish them from dispositions. So there is no clear reason to distinguish them. Lacking any reason to think that potentialities are different from dispositions, we can assume that what is true about dispositions is true about potentialities, and reasons to think that dispositions are real are reasons to think that potentialities are real.

2 Realism Versus Anti-realism About Potentialities 2.1

Reduction

We employ numerous and diverse dispositional and potentiality concepts. What does that tell us about the world? I hope it is not too naïve to think that long entrenched traditions of employing certain concepts with apparent success gives some reason to think that those concepts are related to the world in meaningful ways. Perhaps, if our potentiality ascriptions are true, then the potentialities we ascribe to things exist, in whatever sense properties exist. This assumption stands in opposition to a notable philosophical project of the last century, to semantically reduce disposition ascriptions (Schrenk 2009). That reduction can be applied to potentiality claims as well. If the reductionist project is successful, one could say that potentiality ascriptions are true, but not because potentialities exist, but because the ascriptions are merely ways of asserting something that is consistent with the non-existence of potentialities, such as a conditional, or a claim about other sorts of entities. Just as the claim that “The average American woman has 1.5 children” does not commit one to the existence of the average American woman nor half-children, one may argue that truth of claims such as “Alice is a potential philosopher” and “an acorn is potentially an oak tree” does not commit one to the existence of potentialities. So, one reason to be anti-realist about potentialities is because you think that potentiality claims can be reductively analyzed and eliminated in favor of other sorts of entities. I have suggested that a prima facie reason to think that a property exists is that we consistently apply a certain predicate. But this prima facie reason is defeasible. As Shoemaker writes: …we have a notion of a property … which is such that not every phrase of the form ‘being so and so’ stands for a property which something has just in case the corresponding predicate of the form ‘is so and so’ is true of it, and is such that sometimes a predicate is true of a thing, not because (or only because) of any properties it has, but because something else, perhaps something related to it in certain ways, has certain properties (1980: 209).

Real Potential

245

The assumption that we are referring to a property can be undermined if a reductive analysis of disposition and potentiality claims can succeed. As Troy Cross writes: if, as it was hoped, subjunctive conditionals could be given truth conditions in purely categorical terms, dispositions could be eliminated from fundamental ontology and replaced with purely categorical properties, or perhaps categorical properties together with laws of nature (2012: 115).

2.1.1

Simple Conditional Analyses

A common reductive strategy has been to analyze disposition claims in terms of conditional statements. For example: (A) x is disposed to exhibit manifestation M in circumstance C iff (B) if x were to be subject to C, x would exhibit M. Similarly for potentialities, the analysis would be: (A)P x has the potential to become F in circumstance C iff (B)P if x were in C, x would become F. (If, as argued above, potentialities and dispositions are co-extensive, these come to the same thing. However, in order to demonstrate that what is true of disposition claims is also true of potentiality claims, I will consider each type of claim in turn.) The reductionist would argue that since the analysans in either case not mention dispositions or potentialities, but only particulars, circumstances, and properties, as long as we have an account of what it means for the relevant kind of conditional to be true, we have an account of how the disposition or potentiality statement can be true without positing any dispositions or potentialities. A common response to such reductionist strategies is that conditional analyses are false. One initial problem for a conditional analysis is trying to accurately specify the appropriate conditional. A few moments of reflection is enough to realize that “if you drop it, it will break” is a woefully inadequate and overly simplistic analysis of fragility. Such a conditional will not be true of a fragile glass if you drop it a fraction of an inch, drop it onto a fluffy cushion, or drop it in a low gravity environment. Dropping is not even necessary for a fragile glass to break— you can strike it where it sits. But if you strike it softly with a feather, it will not break. However, a very powerful blow could break even non-fragile things. In order to state a conditional that is true of a thing if and only if it is fragile, you would have to figure out the precise conditions under which fragile and only fragile things break —no easy task. If it is difficult to specifying the conditions under which fragility manifests, it seems virtually impossible to specify in detail the circumstances of manifestation for certain potentialities like potential for intelligence. If the view is that something has a certain potential if and only if a certain conditional is true, then it is fair to

246

J. McKitrick

expect to be told what that conditional statement is. But, in order to say what that conditional statement is, one would have to articulate the precise conditions which are sufficient for a thing to realize its potential. It would be inadvisable to make them too precise, lest you deny that potentiality to other things which may aptly be said to have it. Even if you figure out the right conditional, a conditional analysis is still challenged by a number of counterexamples. One such counterexample is a “mask” (Johnston 1992) or “antidote” (Bird 1998) which challenges the necessity of the analysis. For example, imagine a fragile glass that is packed so that it has internal supports to prevent the glass from warping and therefore from shattering when struck. If you struck the packed glass, it would not shatter. The disposition ascription (A) is true—the glass is fragile—but the conditional claim (B) is false. Surely, a thing’s potentiality can be masked as well. An acorn could be placed in fertile soil, but if it were coated in hard plastic, the seed could not break through and grow. In that case, the acorn might still have the potential (A)P to become an oak tree, but the associated conditional (B)P is false. The general problem is that, even if the specification of the circumstances of manifestation articulates all that must be present in order for the thing to realize its potential, the specification cannot rule out all of the possible factors which may interfere. To add to the analysis “and nothing interferes” would trivialize it. It would be a matter of explicating “x has the potential to be F” as “x will become F, unless it does not.” One counterexample to the sufficiency of a conditional analysis of dispositions is the case of so-called “mimics” (Smith 1977). A mimic is something that lacks a certain disposition, but for idiosyncratic reasons, acts as if it does. Odd circumstances result in a certain conditional being true of something that nevertheless fails to instantiate the requisite disposition. Smith’s example is a wooden block that is brought to Neptune, where something about the atmosphere results in the block shattering when it is dropped. The conditional (B) “if you had dropped it, it would have broken” is true of the block on Neptune, but intuitively, the corresponding disposition ascription (A) is false; the block is not fragile. Interestingly, some of the most talked-about counterexamples to the moral relevance of potentiality appear to be variations on mimicking cases. Consider “super kitten,” a kitten which intuitively does not have the potential to be a person, but which could be injected with a special serum that would turn it into something with the characteristics of a person (Tooley 1972). A kitten receiving such extraordinary treatment is perhaps analogous to a wooden block being taken to Neptune, in the sense that, in both cases, unusual circumstances could lead to unusual results, and this challenges our application of the concepts in normal circumstances. In that case, the potentiality ascription (A)P is false, but the associated conditional (B)P turns out to be true. Another purported counterexample to the conditional analysis is the case of so-called “finkish” dispositions (Martin 1994; Lewis 1997). Once we note that things can acquire or lose dispositions, we can generate counterexamples to a conditional analysis by supposing that these gains and losses can occur when the circumstances of manifestation occurs. An example of a finkish disposition is the

Real Potential

247

fragility of a glass which is protected by a wizard who will immediately render it non-fragile if it is ever struck. A less fantastical example of a finkish disposition is the instability of the DNA molecule. DNA is susceptible to breaking up due to forces, such as radiation and heat. However, forces which would break the molecule also trigger mechanisms within the cell nucleus that maintain the molecule’s structure (Tornaletti and Pfeiffer 1996). An object has a “finkish disposition” if that object has a disposition which it loses in what would otherwise be the circumstances of manifestation. If the disposition D is finkish, the same C that would normally cause x to exhibit M instead causes x to lose D before it can exhibit M. In this case, (A) is true: x does have the disposition. But (B) is false: If x were subject to C, it would not exhibit M. So, the analysis fails to state a necessary condition for x’s having a disposition. A potentiality could also be finkish. Consider again an acorn with the potential to become an oak tree. The circumstances of manifestation of that potentiality include dropping onto fertile soil. But suppose the gardener does not want any more oak trees in the yard, so he crushes any acorn that drops. The circumstances that would normally result in an acorn manifesting its potential lead to its destruction. If the acorn has the potential to be an oak tree, that is a case in which the potentiality claim (A)P is true, but the associated conditional (B)P is false. A similar type of counterexample is called “altering” (Johnston 1992). A glass swan is fragile, but a vigilant monitor equipped with a laser beam will rapidly melt the swan the moment it is struck. The conditional (B) is false, but the swan is fragile (A). Another example is the shy, but intuitive chameleon (Johnston 1992). The chameleon is green and thus disposed to look green, but before anyone can turn on the light and look at it, it blushes red. In both these cases, the conditions of manifestation are such that, if they were realized, the object would “alter” and lose its disposition. A thing can also finkishly lack a disposition. When green, the chameleon does not have the disposition to appear red, but when the circumstances of manifestation occur, the chameleon acquires that disposition. In this case, an object x which does not have disposition D gains D when exposed to circumstance C, and subsequently exhibits manifestation M. Arguably, prior to C occurring, (A) is false: x does not have the disposition. However, (B) is true: if x were to be subject to C, x would exhibit M. So, this kind of case also shows that the analysis fails to state a sufficient condition for x’s having a disposition. Something could finkishly lack a potentiality as well. If some cloning or nanotechnology could turn something non-human into a human fetus, and if that procedure is initiated only if that something is placed in a uterus, then that thing would, at the outset, lack the potential to be a human being. However, if it gets placed in the circumstances of manifestation, it would acquire that potential. Again, the potentiality claim (A)P would be false, but the associated conditional (B)P would be true. I suppose the super-kitten could be recast as something that finkishly lacks the potential to be a person, if those who would provide the transformative agent would do so only on the condition that the kitten gets adopted by a family that intends to raise it as a human child, teaching to it speak and so on. It would be true

248

J. McKitrick

of the kitten “if it is nurtured and educated, it will become a person,” even though we may be disinclined to say that the kitten has the potential to become a person.

2.1.2

The Revised Conditional Analysis

Lewis’ revised conditional analysis (RCA) was specifically designed to overcome finks. Recall the simplified version of Lewis’ analysis states: x has a disposition at time t to give M in C iff x has intrinsic property P, and if x were to be in C at time t and retain P, then P and C would cause M (1997: 157).

According to RCA, the activating conditions and an intrinsic property of the disposed object jointly cause the manifestation of a disposition. In a finkish case, something causes the object to lose the relevant intrinsic property, and subsequently to lose the disposition. The condition that the object retains the intrinsic property is not satisfied by objects with finkish dispositions, and so they pose no counterexample to RCA. The condition that the object must have the intrinsic property is not met when something finkishly lacks a disposition, so that counterexample is defeated as well. While Lewis’ analysis does address cases of finks, it can be contested on at least three grounds: (1) It assumes that all dispositions have causal bases; (2) It does not address the masking counterexamples: and (3) it assumes that dispositions are intrinsic properties. I have argued elsewhere that dispositions do not necessarily have causal bases; there can be ungrounded or ‘bare’ dispositions (2003b). RCA fails to extend to such dispositions. I argue that it neither follows from the concept of a disposition nor from the idea that disposition claims must have truthmakers, that dispositions necessarily have causal bases. Others such as Molnar (2003) and Mumford (2005) attempt to identify a class of ungrounded dispositions in the fundamental properties of subatomic particles. While it is reasonable to think that many of the potentialities I have been talking about (the potential to be a tree, a person, or a philosopher) are grounded, perhaps fundamental particles have potentialities, and these lack causal bases. Molnar argues that the nature of these particles is exhausted by their dispositionality, and extensive experimentation has revealed no deeper structure to serve as the intrinsic properties to ground these dispositions (2003: 131–132). RCA would seem, therefore, to be inapplicable to the most fundamental properties of the physical world. Other second-order property accounts can be rejected for similar reasons. In addition, it is acknowledged that RCA still faces the problem of masking. Consider the glass that is carefully packed. According to RCA, to say the glass is fragile is to say that it has some intrinsic property P, and if it were to be in circumstances of striking at time t and retain P, then P and striking would cause breaking. However, the carefully packed glass retains its intrinsic properties, but its intrinsic properties and the striking do not cause the glass to break.

Real Potential

249

As I have also argued elsewhere, dispositions are not necessarily intrinsic to the objects that have them (McKitrick 2003a). A property is extrinsic if perfect intrinsic duplicates can differ with respect to having it.7 Contrary to Lewis, perfect duplicates could differ with respect to having certain dispositions; a thing can lose or acquire dispositions without changing intrinsically. Weight may be dispositional, but it is not intrinsic. The weight of an object is relative to its gravitational field. According to RCA, weight could be defined as follows: An object weighs one hundred pounds (it has a disposition at time t to exhibit the reading “100 lbs.” in circumstances of standing on a standard scale) if and only if it has an intrinsic property P, and if it were to stand on a standard scale at t and retain P, then P and standing on the scale would cause it to exhibit the reading “100 lbs.”

But if the object stood on the scale on the moon at t, its intrinsic properties plus standing on a scale would not cause a “100 lbs.” reading.

Extrinsic Potentialities As with looking for examples of extrinsic dispositions, the strategy for finding cases of extrinsic potentialities is to consider perfect intrinsic duplicates and see if they could differ with respect to potentiality attributions. So, consider an acorn with the potentiality for “treehood” and a perfect duplicate of that acorn in a different world, all by itself. This acorn is “lonely” in the sense that it is the only object in its world. If the lonely acorn lacks the potentiality for treehood, then that potentiality is extrinsic. So, does the lonely acorn have the potentiality for treehood? If the circumstances of manifestation must occur somewhere in that world in order to have the potentiality, then the lonely acorn lacks the potentiality. The lonely acorn can also lack the potential for treehood if it is in a world where the laws of nature prohibit any circumstance which would enable it to develop into a tree. However, some philosophers think that the relevant sense in which dispositions are intrinsic is that they are “intrinsic, keeping the laws of nature fixed” (Lewis 1997). An interesting and relevant question, at any rate, is whether duplicates in the same kinds of worlds, or different parts of the same world, could differ with respect to a potentiality. It would be helpful here to consider end of life cases, where one makes claims such as: a coma victim is potentially conscious, or a terminal condition is potentially reversible. A patient might not be capable of recovering or becoming conscious, given current medical technology. However, it is possible that some future medical technology could reverse his condition. Do we want to say that the patient currently has the potential to recover? That seems misleading at best. Perhaps it would be more correct to say that they are not now potentially conscious, but if extrinsic factors were different, they would be. Then the patient’s potential for recovery is an extrinsic potentiality. It is plausible that similar considerations apply This is not a reductive analysis of “extrinsic” but is offered as a heuristic, intuitive guide to judgments about extrinsicality.

7

250

J. McKitrick

if the necessary medical technologies exist, but are not practically accessible, if they are very far away or prohibitively expensive, for example. Two patients with the same condition could have different potentialities due to the differing circumstances, and this entails that those potentialities are extrinsic. Similar considerations apply in beginning of life cases. Perfect duplicate embryos in different circumstances could have different potentialities. For instance, consider an embryo outside of a uterus and its perfect duplicate inside of a uterus. If there are no available means for implanting the embryo into a favorable environment, then by parity of reasoning with the end of life cases, we should say that the embryo lacks the potentialities enjoyed by its duplicate inside of a uterus. Therefore, the embryo’s potentiality for rationality is an extrinsic potentiality. Another thing to consider is what counts as “available means” for implanting an embryo. Surely, the medical technology must exist in this case, too. So, if we consider two duplicate frozen embryos, one in a fertility clinic with all the staff and equipment necessary for successful implantation, and the other in a remote location with no such amenities, perhaps we should say that those embryos have different potentialities. And if perfect duplicates can differ with respect to having a certain potentiality, then that potentiality is extrinsic. Typically, those that think that causal bases are essential to dispositions are thinking of those causal bases as intrinsic (Lewis 1997). However, an alternative position is that not only can dispositions be extrinsic, but the causal bases of dispositions can be extrinsic too (Nolan 2005). So, the mistake of those that claim that intrinsic duplicates have the same disposition is not that they are only focusing on the causal basis of that disposition, but rather that they are only focusing on part of the causal basis of the disposition, and not taking into account the properties that are extrinsic to the disposed individual that are part of the causal basis of its disposition. In other words, if an embryo’s potentiality is extrinsic, then the causal basis of its potentiality does not merely consist of the intrinsic properties of the embryo, but also includes properties of its environment. One may argue that, if potentialities can be extrinsic properties, that is all the worse for realism about potentialities. Extrinsic properties are the quintessential “Cambridge properties,” the properties that things acquire when they undergo “Cambridge changes,” which are not genuine changes (Geach 1969; Shoemaker 1980). However, proponents of such a view are committed to rejecting all extrinsic properties and relations, including being married, being taller, happening later, and standing a meter away. Perhaps some relational properties can be reductively explained in terms of intrinsic properties, but reducing all spatial and temporal properties to intrinsic properties promises to be problematic. Taking stock of where we are, the above argument for extrinsic potentialities is presented for the purposes of resisting attempts to reduce disposition claims to conditional claims about intrinsic non-dispositional properties. Ungrounded dispositions and masks are also outstanding problems for such attempts. At this point, aspiring reductionists may want to look elsewhere.

Real Potential

2.1.3

251

Reduction by Other Accounts

Views according to which dispositions are irreducible powers cannot be used to reduce potentialities away. If such views are correct, and if potentialities just are (a type of) disposition, then potentialities are irreducible too. Second-order property accounts will not do the job either. (Recall that a second-order property account is one according to which “x has a disposition D to exhibit M in C iff x has some property P which is a causal basis for giving M in C.”) As mentioned above, a second-order property account assumes that all dispositions have causal bases, a claim which has been contested. But even if we grant that potentialities have causal bases, this second-order property account is not sufficiently reductive. The causal basis is characterized as a property that “would be causally efficacious for M in C”—it would cause the disposition’s manifestation in the circumstances of manifestation. In other words, what makes a property a causal basis of a disposition is its power to have a certain effect. But that is tantamount to saying that the causal basis is or has a dispositional property. Lacking some reductive account of what it means to say that a property is such that “would cause” a certain effect, a second-order property account seems to analyze disposition ascriptions in terms of having dispositional properties. Another account to consider is Manely and Wasserman’s PROP. Recall that PROP says “x is disposed to M when C iff x would M in a suitable proportion of C-cases”—where C-cases are possible circumstances. It matters not that some C-cases include finks or masks, as long as enough of the C-cases are cases where x M’s. Also, it does not matter if there are some mimics; if some y is not disposed to M in C, but nevertheless M’s in some particular circumstances. Presumably, if we are still disinclined to say that y is disposed to M in C, that is because there are still too many relevant possible situations in which y does not M in C. Nor does it matter if the disposition is extrinsic or ungrounded, since the analysans makes no mention of bases, intrinsic or otherwise. So, it seems that we have an analysis that steers clear of many of the pitfalls that plagued other conditional analyses. Does this mean we can make sense of all our disposition and potentiality talk, while denying the existence of dispositions and potentialities? One response is that PROP cannot deal with all of the counterexamples, particularly those that have come to be known as intrinsic finks (Ashwell 2010; Clarke 2008; Everett 2009). Johnston’s shy-but-intuitive-chameleon example is one of an intrinsic fink. It is disposed to look green, but due to its intrinsic properties, it loses that disposition when people look at it. Given that description, it will not look green in the majority of the C-cases. This disposition claim “x is disposed to M when C” is (arguably) true, but the associated conditional “x would M in a suitable proportion of C-cases” is false in the case of this chameleon. Another response is to say that this conditional analysis is neither here nor there with respect to the reality of potentialities. A conditional analysis of potentialities says something about what we mean when we say something has a potentiality. One may argue that it places no restrictions on our metaphysical views about the

252

J. McKitrick

existence of such properties. In fact, Manley and Wasserman do not claim that PROP reduces disposition ascriptions to conditionals. They write: We take PROP to capture a necessary connection between dispositions and conditionals, but we withhold judgment on the priority of either side of this equation (whether metaphysical or conceptual) (2011: 4).

To say that something has a potentiality if and only if a certain conditional is true of it does not go to show that there are no potentialities any more than it goes to show that there are no true conditional propositions. On abundant (non-sparse) views of properties, a thing can have the property of being such that a certain conditional is true of it. On an extensional view of properties, for example, the set of things that a certain conditional is true of constitutes a property. Or perhaps the reason why certain conditionals are true of a thing is because it has certain potentialities. In that case, the order of dependence goes in the other direction. In sum, anti-realism about potentialities has not been established by reductive analyses of potentiality claims.

2.1.4

Subjectivism

Another argument that one could level against realism about potentialities begins with the idea that potentiality claims say more about subjective human psychology than they do about any mind-independent properties. Recall some of the questions considered above: Is super-kitten a potential person? If some therapy might be developed in the future which would save someone’s life were it to exist now, does that patient have the potential to recover now? Whether we answer “yes” or “no,” it may be asked, on what basis do we determine our answer? Perhaps what determines our answers is our own criteria of plausibility or estimation of future possibilities. We could model a person’s thinking when making potentiality judgments as follows: Consider some object and some relevant sample of its possible future stages; then arrange those future possibilities in order of your estimation of their likelihood; then, pick a threshold beyond which the possible outcome is one which you consider too unlikely to take seriously. Every future state below that threshold is one which the object has a potentiality for. For example, consider the acorn. Possible future states might include getting crushed, becoming an oak tree, becoming a table, and becoming a chicken. Depending on your bounds of plausibility, you might say that the acorn has the potential to become an oak tree, but not a table or a chicken. But others may have different orderings of likelihood, and put the threshold in a different place, and so make different judgments as to a thing’s potentialities. In this regard, something having a potentiality is like it being something that inspires expectation, hope, or dread—all descriptions of a thing in terms of the psychological states of observers.

Real Potential

253

If this is the right way to understand our potentiality talk, then whether a thing has a certain potentiality will be relative to individuals, based on their subjective probabilities and standards of plausibility.8 This would, in effect, be another reductionist strategy. It amounts to saying that if a potentiality claim is true, it is because of the psychological states of the person making it, and not because of the existence of any property possessed by the object that they are making the claim about. One can find inspiration for such an argument in Hume, who famously locates the origin of the idea of causation in a feeling of expectation, as well as more recent theories that regard causal claims as implicitly contrastive or value-laden (Hume 1978; Maslen 2004; Kagan 1988). If whether ‘A caused B’ depends on expectations, salience of contrast classes, or moral judgments, and these things differ from person to person, causal powers, and by implication potentialities, cannot be regarded as objective, mind-independent properties. To respond to such an anti-realist argument, note an implication of the view being suggested. If my claim that something has a potentiality is merely a report about my subjective probabilities, then as long as my words accurately reflect my psychological state, I cannot be mistaken. If it seems to me that being a chicken is a possible future state of an acorn, and I do not regard that possibility as too far out, then when I say “an acorn is a potential chicken,” I am saying something true. But it seems that people can disagree about what potentialities things have, and that they are not merely talking past each other, each reporting on their own subjective probabilities. Another consideration to keep in mind is the distinction between epistemic potentiality and metaphysical potentiality. Recall that when one thinks that some claim has some chance of being true, one may express that idea by saying “it is potentially true.” However, that is not necessarily to attribute a potentiality to anything. The idea that attributing a potentiality to something is a report of one’s subjective probability treats all potentiality claims as if they were about epistemic potentiality. But it is not clear that this is so. It is worth pointing out the analogy to chance. Many claims about the chance of some event occurring can be appropriately interpreted as claims about subjective probability, based on one’s epistemic situation. This can, of course, deviate from the objective chance of the event occurring. If I flip a fair coin once and conceal the outcome from you, from your point of view, it has a 50% chance of being heads up, but in terms of objective probability, once the flip is over, it either has a 0% or 100% chance of being heads up. Just as we can interpret “chance-talk” as being about subjective probability, objective probability, or some mixture of the two, so we can understand “potentiality-talk” as sometimes epistemic, sometimes metaphysical, and sometimes a mixture. In other words, if I say “x is potentially F” I might mean: “For all I know, x is F”; “There are circumstances in which x can become F”; or “For all I 8

Thanks to Matthew McKitrick for suggesting this line of argument.

254

J. McKitrick

know, there are circumstances in which x can become F.” Where the threshold of plausibility judgment comes in is how to interpret the “there are circumstances” part of the metaphysical claim. What does it take for it to be true that “there are circumstances in which x can become F”? For example, in order for a fetus to be potentially rational, there must be circumstances in which it would become rational. But in what sense is it the case that “there are” such circumstances? Since it is possible to possess this potential without manifesting it, it cannot be the case that these circumstances must actually obtain for the particular embryo to be potentially rational. However, saying that there are such circumstances, as long as such circumstances are logically or metaphysically possible, would allow too many things to count as potentially rational. Even given a metaphysically possible scenario in which rocks are turned into rational beings, we want to say embryos are potentially rational in the actual world while rocks are not. So, the circumstances which enable an embryo to become rational must be at least physically possible. We might want to place more restrictions on the range of possible circumstances and say that they must not only be possible in this world, but that circumstances of this kind are instantiated somewhere in this world, or that they are not too remote or unlikely to obtain. I think that ordinary uses of “potential” underdetermine how to proceed here. If a philosopher wants to employ a more technical concept, “potentiality,” it would behoove her to clarify what range of possible circumstances are relevant to her potentiality attributions. Given a clarification of the concept, I see no reason to think that whether such a concept applies to a thing is merely a subjective matter.

2.2

The Eleatic Principle and the Inert Dispositions Thesis

Another possible reason for thinking that potentialities are not real is the belief that potentialities run afoul of the Eleatic Principle—to be real is to have causal powers. On the face of it, this seems strange: If potentialities are dispositions and dispositions are powers, who would think that powers are powerless? But proponents of such arguments are not thinking of dispositions as irreducible powers, but are typically thinking of them as second-order properties (Prior et al. 1982; Jackson 1995, 1996; Block 1990; Kim 1993; Rives 2005). If having a disposition is a matter of having some other property, then arguably dispositions are derivative and causally irrelevant. If this “inert dispositions thesis” and the Eleatic Principle are correct, dispositions, and by implication, potentialities, are not real. Two common strategies supporting the Inert Dispositions Thesis are analyticity arguments and exclusion arguments. According to analyticity arguments, there is an analytic or necessary connection between a disposition and its manifestation, and this goes to show that there is no causal connection. According to exclusion arguments, manifestations of dispositions already have sufficient causes, and this excludes dispositions from playing any causal role. However, as I will explain in this section, these arguments do not succeed (McKitrick 2004).

Real Potential

255

I have already raised some problems for second-order property approaches to reductively analyzing dispositions. But though I argue that some dispositions are ungrounded, I have left open the possibility that others have causal bases, so the causal relevance and thus the ontological status of those properties is challenged by these arguments. By the same token, I would be concerned to reply to dispositional realists who claim, for similar reasons, that only fundamental powers are genuine (Bird 2007; Molnar 2003: 27).

2.2.1

Analyticity Arguments

According to an analyticity argument, there is an analytic relation between a disposition and a manifestation, and this is evidence that there is no causal connection between them. Any adequate definition of a disposition will refer to its characteristic manifestation. For example, “fragility” is defined by reference to breaking or shattering. So, there is a definitional, conceptual, or logical connection between a disposition term and an event-type—between ‘fragility’ and breaking, for example. “Fragile objects tend to break when struck” is analytic (if anything is). It is further assumed that causal claims are contingent, and therefore not analytic. It follows that, if a statement is analytic, it is not a causal statement. Consequently, “The glass broke because it was fragile” cannot be a causal claim, and so fragility must be causally inert. By the same reasoning, a potentiality is not the cause of its actualization; the acorn’s potential to become a tree is not a cause of it becoming a tree. Such analyticity arguments have long been disputed (Davidson 1980: 14). Causal connections between events hold independently of our descriptions of them. If “the cause of e caused e” is analytic, that does not go to show that the cause of e did not cause e. A similar point can be made in terms of properties. The analyticity of the claim “The property that was causally efficacious for e was causally efficacious for e” should not lead us to think that the property that was causally efficacious for e was not causally efficacious for e. According to more modest versions of an analyticity argument, our sense that the disposition is relevant to the manifestation is explained by an analytic connection, and this leaves us with no reason to suppose a causal connection holds as well. This is what Block suggests when he writes: The fact that dormitivity is sufficient for sleep is perfectly intelligible in terms of this logical relation. What reason is there to suppose that there must also be a nomological relation between dormitivity and sleep? (1990: 157).

The quick answer to Block’s rhetorical question is that our language can track causal connections. Examples of this are familiar: sunburn is caused by excessive exposure to sunlight; lethal injections and fatal accidents cause death. A related argument for the Inert Dispositions Thesis appeals to Hume’s Principle that there are no necessary connections between distinct existences. According to Frank Jackson, to allow that fragility causes breaking upon dropping

256

J. McKitrick

would be to allow that there are properties that have causal powers essentially: in every world the property of having the property or properties responsible for breaking on dropping in that world is possessed only by objects which are such that were they dropped they would break. There is no way that the second-order property can be instantiated without the relevant causal power being instantiated. So, if we are to respect Hume’s insight, we must deny that fragility itself does the causing of the breaking… (1995: 257).

To assess this argument, one should first consider what it means to “respect Hume’s insight.” As I interpret Hume, his argument against necessary connections between distinct existences goes as follows: (1) If one can conceive of A without B, then there is no necessary connection between A and B. (2) For any distinct existences A and B, one can always conceive of A without B. (3) Therefore, there are no necessary connections between distinct existences (1978: 79). If you were to discover two distinct existences which are such that you cannot conceive one without the other, that would be a counterexample to premise (2). But perhaps Jackson takes himself to have conceived of a case of a necessary connection between distinct existences—the one which would obtain if fragility had the power to cause breaking. But instead of taking this as a counter-example to Hume’s Principle, he assumes the principle in a modus tollens argument against the possibility he has conceived. Jackson’s argument can be put like this: i. There are no necessary connections between distinct existences. ii. ‘Fragility’ and ‘the power to cause breaking’ are distinct existences. iii. If fragility had the power to cause breaking, there would be a necessary connection between ‘fragility’ and ‘the power to cause breaking.’ iv. Therefore, fragility does not have the power to cause breaking. Jackson, then, faces a dilemma with respect to Hume’s argument above. Either Jackson cannot conceive of ‘fragility’ without ‘the power to cause breaking,’ thus contradicting Hume’s premise (2), or he can conceive of ‘fragility’ without ‘the power to cause breaking,’ and given Hume’s premise (1), there is no necessary connection between ‘fragility’ and ‘the power to cause breaking,’ contradicting Jackson’s own premise (iii). The two arguments are not mutually consistent. In short, Jackson employs Hume’s Principle in an argument which is inconsistent with Hume’s own argument for the principle, and Jackson gives us no independent reason to accept it. Even if we accept Hume’s Principle on other grounds, in order for dispositional properties with causal powers to run afoul of it, a disposition and its causal power must be distinct existences. But it is not clear that a thing’s fragility and its power to cause breaking are distinct existences. Furthermore, even if this argument shows that ‘fragility’ is not causally relevant to breaking, there is another disposition mentioned in the argument, namely ‘the power to cause breaking,’ whose causal relevance is never doubted. So, even if Jackson’s argument is successful against

Real Potential

257

fragility, it does not stand as an argument against causally relevant dispositions generally. In sum, neither the straightforward Analyticity Argument, nor Jackson’s appeal to Hume’s Principle, succeed in showing that dispositions are causally inert or unreal. Now, let us turn to a different argumentative strategy for the Inert Dispositions Thesis—exclusion arguments.

2.2.2

Exclusion Arguments

Exclusion arguments are familiar in the literature on mental causation. Consider mental property M and physical property P, which are candidates for being causally efficacious with respect to a brain event with mental property M* and physical property P*. Jaegwon Kim argues that M has no causal powers of its own: P is doing all the causal work, and M’s causation of P*, or of M* turns out to be derivative from P’s causal powers. Thus, M has no causal powers over and beyond those of P… (1993: 353).

Similarly, Prior, Pargetter, and Jackson (hereafter PPJ) support the Inert Dispositions Thesis with an exclusion argument. They write that since the causal basis and the circumstances of manifestation are sufficient for the manifestation, “there is nothing left for any other properties of the object to do” (1982: 255). These exclusion arguments target second-order properties generally. It is claimed that the base properties are causally efficacious for a certain event, and this excludes the second-order property from being causally efficacious as well. If the circumstances of manifestation and the base properties are sufficient for the manifestation, then it is thought that all other properties must be inert. PPJ’s (1982: 255) argument is typical. It can be summarized as follows: a. The Causal Thesis: Every disposition has a causal basis. b. The Distinctness Thesis: Causal bases are distinct from their attendant dispositions. c. When a disposition’s manifestation occurs, its causal basis and its circumstances of manifestation are its jointly sufficient causes. d. The Exclusion Principle: If the instantiation of a set of properties is sufficient to cause an effect, then all other properties are causally inefficacious with respect to that effect. Therefore, e. Dispositions are causally inefficacious. Suppose a particular causal basis for a potentiality and the properties of the circumstances of manifestation are sufficient for the manifestation. It would follow from the Exclusion Principle that the potentiality is causally irrelevant, and by the Eleatic Principle, the potentiality is not real. In support of the Exclusion Principle, Kim says “The general principle of explanatory exclusion states that two or more complete and independent

258

J. McKitrick

explanations of the same phenomenon cannot coexist” (1993: 250). PPJ succinctly echo “a complete causal explanation excludes competitors” (1982: 255). Proponents of the Exclusion Principle say that to deny it is to allow for spurious overdetermination. Denying the Exclusion Principle amounts to saying that both a potentiality and its causal basis are each sufficient for the effect, given the circumstances of manifestation. If this happened every time a potentiality was actualized, we would have, as Block says, “bizarre, systematic over-determination” (1990: 159). Standardly, an event e is causally overdetermined if two or more distinct events occur, each of which is sufficient to cause e. Perhaps there is something wrong with postulating too many coincidences. And if a great many effects systematically had two distinct events that were sufficient for causing them, that would run counter to our understanding of the causal structure of the world. Maybe that would be too high a price to pay for real potentialities. However, one should not overlook the fact that these considerations and intuitions about overdetermination apply to two (or more) events overdetermining an effect. But what we are concerned with in the case of potentialities and causal bases are the properties of a single individual. It is not clear what it means to say an effect is overdetermined by an object’s properties. The Exclusion Principle asks us to single out one special set of properties that are minimally sufficient for an effect, and declare the rest causally inert. This may prove difficult to do, especially when properties are so intimately related, as a potentiality is to its causal basis. Furthermore, it is not clear that we have any compelling reason to suppose that, for any effect, there is only one minimally sufficient set of causally efficacious properties. Consider the following propositions: The cape has surface reflectance property R; the cape is red21; the cape is crimson; the cape is red; the cape is colored (Yablo 1992: 257). Now, suppose that each proposition entails its successor. It follows from the Exclusion Principle that, if one of the above properties is causally efficacious for a certain effect, all of the others are causally inert with respect to that effect. However, it seems reasonable to think, as Funkhouser claims: “Instances of determinables and their determinates do not causally exclude each other” (2006: 3). If that is not right, it is not clear what would determine the level of specificity at which the causal action is going on. One might argue, as Kim does at times, that all of the causal action happens at the most fundamental level (1993). Such an assumption has serious counter-intuitive consequences, for example, that all of the macro-properties we regularly observe are causally impotent. Furthermore, assuming that all causal action happens at the level of fundamental properties may not serve the Inert Dispositions Thesis, since it seems likely that fundamental properties are dispositional.

Real Potential

259

3 Conclusion I have argued that potentialities are properties, specifically dispositions. Metaphysical accounts of properties and of dispositions provide resources for a variety of views about the nature of potentialities, according to which some potentialities are genuine, real, and existent. An anti-realist might argue that potentiality claims reduce to claims about other sorts of entities, or that potentialities cannot be real because they have no causal powers. I have defended realism about potentialities by countering reductionism and the Inert Dispositions Thesis.

References Armstrong, D. M. (1973). Belief, truth and knowledge. Cambridge: Cambridge University Press. Armstrong, D. M. (1978). A theory of universals. Cambridge: Cambridge University Press. Armstrong, D. M. (1996). Dispositions as categorical states. In T. Crane (Ed.), Dispositions: A debate (pp. 15–19). New York: Routledge. Ashwell, L. (2010). Superficial dispositionalism. Australasian Journal of Philosophy, 88, 635– 653. Bird, A. (1998). Dispositions and antidotes. The Philosophical Quarterly, 48, 227–234. Bird, A. (2007). Nature’s metaphysics: Laws and properties. Oxford: Oxford University Press. Block, N. (1990). Can the mind change the world? In Meaning and method: Essays in honor of Hilary Putnam, New York: Cambridge University Press. Choi, S. (2006). The simple vs. reformed conditional analysis of dispositions. Synthese, 148, 369– 379. Clarke, R. (2008). Intrinsic finks. The Philosophical Quarterly, 58, 512–518. Cross, T. (2012) Recent work on dispositions. Analysis Reviews, 72. Davidson, D. (1980). Essays on actions and events. Oxford: Clarendon Press. Everett, A. (2009). Intrinsic finks, masks, and mimics. Erkenntnis, 71, 191–203. Funkhouser, E. (2006). The determinable-determinate relation. Nous, 40(3), 548–569. Geach, P. T. (1969). God and the soul. London: Routledge and Kegan Paul. Goodman, N. (1983). Fact, fiction, and forecast (4th ed.). Cambridge, MA: Harvard University Press. Gundersen, L. (2002). In defense of the conditional account of dispositions. Synthese, 130, 389–411. Hume, D. (1978). A treatise of human nature (2nd ed.). Oxford: Oxford University Press. Jackson, F. (1995). Essentialism, mental properties and causation. Proceedings of the Aristotelian-Society, 95, 253–268. Jackson, F. (1996). Mental causation. Mind, 105, 377–413. Johnston, M. (1992). How to speak of the colors. Philosophical-Studies, 68(3), 221–263. Kagan, S. (1988). Causation and responsibility. American Philosophical Quarterly, 25(4), 293–302. Kim, J. (1993). Supervenience and mind. New York: Cambridge University Press. Lewis, D. (1983). New work for a theory of universals. Australasian Journal of Philosophy, 61, 343–377. Lewis, D. (1997). Finkish dispositions. Philosophical Quarterly, 47(187), 143–158. Manley, D., & Wasserman, R. (2008). On linking dispositions with conditionals. Mind, 117, 59–84. Manley, D., & Wasserman, R. (2011). Dispositions, conditionals, and counterexamples. Mind, 120, 1191–1227. Martin, C. B. (1994). Dispositions and conditionals. The Philosophical Quarterly, 44, 1–8.

260

J. McKitrick

Maslen, C. (2004). Causes, contrasts, and the nontransitivity of causation. In J. Collins, N. Hall, & L. A. Paul (Eds.), Causation and counterfactuals. MIT Press. McKitrick, J. (2003a). A defense of extrinsic dispositions. Australasian Journal of Philosophy, 81, 155–174. McKitrick, J. (2003b). The bare metaphysical possibility of bare dispositions. Philosophy and Phenomenological Research, 66(2), 349–369. McKitrick, J. (2004). A defense of the causal efficacy of dispositions. Sats: Nordic Journal of Philosophy, 5(1), 110–130. McKitrick, J. (2014). Dispositions and potentiality. In J. Lizza (Ed.), Potentiality: Metaphysics and bioethical dimensions (pp. 49–68). Baltimore: Johns Hopkins University Press. Mellor, H. (1974). In defense of dispositions. Philosophical Review, 83, 157–181. Mellor, H. (1982). Counting corners correctly. Analysis, 42, 96–97. Molnar, G. (2003). Powers: A study in metaphysics. Oxford: Oxford University Press. Mumford, S. (2005). The ungrounded argument. Synthese, 149, 471–489. Nolan, D. (2005). David Lewis. Durham: Acumen Publishing. Noonan, J. (1970). An almost absolute value in history. In J. Noonan (Ed.), The morality of abortion: Legal and historical perspectives (pp. 51–59). Cambridge: Harvard University Press. Prior, E. (1985). Dispositions. Aberdeen: Aberdeen University Press. Prior, E. W., Pargetter, R., & Jackson, F. (1982). Three theses about dispositions. American Philosophical Quarterly, 19, 251–257. Rives, B. (2005). Why dispositions are (still) distinct from their bases and causally impotent. American Philosophical Quarterly, 42, 19–31. Schrenk, M. (2009). Hic Rhodos, Hic Salta. From reductionist semantics to a realist ontology of forceful dispositions. In G. Damschen, R. Schnepf, & K. Stüber (Eds.), Debating dispositions: Issues in metaphysics, epistemology and philosophy of mind (pp. 145–167). Berlin: de Gruyter. Shoemaker, S. (1980). Causality and properties. In P. Van Inwagen (Ed.), Time and cause: Essays presented to Richard Taylor (pp. 109–136). Dordrecht: Reidel. Smith, A. D. (1977). Dispositional properties. Mind, 86, 439–445. Tooley, M. (1972). Abortion and infanticide. Philosophy & Public Affairs, 2, 37–65. Tornaletti, S., & Pfeiffer, G. P. (1996). UV damage and repair mechanisms in mammalian cells. BioEssays, 18, 221–228. Vetter, B. (2010). Dispositions without the stimulus. Paper presented at the Dispositions Workshop, Center for the Study of Mind in Nature, Oslo, Norway, March 22–23, 2010. Williams, D. C. (1953). The elements of being. Review of Metaphysics, 7(3–18), 171–192. Yablo, S. (1987). Identity, essence, and indiscernibility. The Journal of Philosophy, 84, 6. Yablo, S. (1992). Mental causation. The Philosophical Review, 101(2), 245–280.

Author Biography Jennifer McKitrick is a Professor in the Department of Philosophy at the University of Nebraska, USA. McKitrick’s research concerns metaphysics, particularly causal powers and social ontology, as well as metaphysics in early modern philosophy. Her most influential publications are “Reid’s Foundation for the Primary Quality/Secondary Quality Distinction” (Philosophical Quarterly 2002), “A Case for Extrinsic Dispositions” (Australasian Journal of Philosophy 2003), “Gender Identity Disorder” (in: Harold Kincaid and Jennifer McKitrick (eds.), Establishing Medical Reality: Essays in the Metaphysics and Epistemology of Biomedical Science, Springer 2007).

Powers and Potentiality Stephen Mumford and Rani Lill Anjum

1 Introduction—The Return to Power What can an ontology of causal powers show us about the nature of potentiality? Can a knowledge of the powers of things tell us what could be, for instance? If so, how? Can the ontology of powers explain how something new can arise: some new property in a thing, for instance, or some new particular thing existing? It might be said that a tomato in the sun turns from green to red, for instance, because it has a power to do so and that a caterpillar has a power to become a butterfly. Furthermore, fertile males and females have a power to procreate another individual of the same species. Might it be that their potentiality of doing so is because they each have individual powers that can come together to produce something jointly? The ontology of causal powers harks back to the metaphysics of Aristotle and Aquinas (see Geach 1961). During the reign of empiricism, powers fell into disrepute, however, as they seemingly were foul of a number of empiricist precepts. Hume argued, for instance, that we could acquire no empirical knowledge of powers and, according to the standard empiricism position, this meant that we could not talk meaningfully of them. Powers-talk was literally meaningless. Another Humean argument was that a power would introduce a necessary connection into nature and imply the absolute impossibility of its effect not occurring. Yet we do know effects can fail to occur or, at the very least can be conceived not to occur (Hume 1739: 161). There are no necessary connections in nature, concluded Hume, and there are therefore no powers. S. Mumford (&) Department of Philosophy, Durham University, Durham, UK e-mail: [email protected] R. L. Anjum School of Economics and Business, Norwegian University of Life Sciences, Ås, Norway e-mail: [email protected] © Springer Science+Business Media B.V., part of Springer Nature 2018 K. Engelhard and M. Quante (eds.), Handbook of Potentiality, https://doi.org/10.1007/978-94-024-1287-1_10

261

262

S. Mumford and R. L. Anjum

The last century, however, has seen a gradual rediscovery of powers as a legitimate part of the furniture of the world. The ontology has been developed gradually, for instance by Armstrong (1968), Harré and Madden (1975), Prior (1985), Ellis and Lierse (1994), Armstrong, Place and Martin (1996), Mumford (1998, 2004), Ellis (2001), Molnar (2003), Bird (2007) and Martin (2008) and there are now some who profess to be out and out realists about powers. One reason for accepting the reality of powers is the ‘work that they can do’ (Molnar 2003: 186) and an explanation of potentiality might be some such work. Molnar (2003: Chap. 12, Pt 2) suggests, for instance, that one virtue of the powers of ontology is what it can tell us about modality. In many cases, the detailed work is yet to be done and this may be because, in terms of the lengthy history of philosophy, it is only very recently that analytic philosophers have started to take the metaphysics of powers seriously. The development of this metaphysics is an ongoing job.

2 Powers as the Grounds of Potentiality Let us return specifically to the issue of how powers relate to this issue of potentiality. The connection is that a power is seen as something that is productive of other phenomena. Although a power is taken to be actual, by those who believe in them, that which they produce need not be. It can be potential only, in the sense that the power disposes towards a certain kind of manifestation but can exist independently of that manifestation. To take a simple, commonplace example, a type of chemical substance may have a power to corrode—we call it corrosive—but it need never actually corrode anything if it is always suitably contained. Given that powers are supposed to be things in the world—perhaps properties or tropes of particular things—they give us a worldly and natural grounding for the potentialities. Understanding powers and potentialities in this way has the virtue of explaining many of the pre-theoretical data. We do accept that there is a potential within certain objects and substances. A sperm has the potential to fertilise an egg, for instance, and a fertilised egg has the potential to grow into a living being. The powers account explains why there are these potentialities. There is a property or trope of the sperm that is empowered in these ways. But the account also explains why some potentialities need never become actualised. The sperm may never meet the egg, for instance, and thus never get its chance to fertilise it. The fertilised egg may fail to embed in the lining of the uterus and not grow into a living being or a pregnancy may be terminated. Such powers only dispose towards the production of a certain outcome and many dispositions go unmanifested. The account thus explains how there are many potentialities not all of which are realised. Indeed, there can be cases of incompatible actualities that nevertheless, understood in terms of powers, can be perfectly real potentialities. A young man, for instance, has the potential to become a full-time and life-long baker but also the potential to become a full-time and life-long butcher. He cannot become both, which means that only one of these potentialities can be actualised. But both are

Powers and Potentiality

263

nevertheless real potentialities for the young man. And we can explain this. The man has the power to become a baker: he can do so if he takes the necessary training and gains employment in that trade. But he also has the power to become a butcher. He can wield a meat-cleaver and carry a side of beef and may do these things for his trade after taking the appropriate training. The powers account crucially, however, explains why there are at least some constraints on potentiality. Not everything is possible and there are some things for which we simply do not have the powers. Our young man does not have the potential to become a poached egg, as far as we know, and we humans cannot potentially jump to the moon. Such constraints on potential can be explained naturalistically because the powers account grounds them in features of the world and its inhabitants. Our legs do not have the power to escape the gravitational pull of the earth: indeed we can jump only about a metre into the air at best. And even if our legs were strong enough, our bodies would not have the power to survive a journey through the vacuum of space, which would be required to get us to the moon. Similarly, we might be able to say that our young man does not have the potential to become a top-class basketball player, because he is too short, or the potential to become a world-leading mathematician, because he is not bright enough. This is not to deny that there may have been a point earlier in his life when he could have had such potentialities. Perhaps in his mother’s womb being tall or intelligent were still possibilities for him, given that he did have dispositions to develop that way. But then perhaps those dispositions were thwarted: his mother smoked during pregnancy, for instance, which might have stunted his physical and mental development. Powers can be gained and lost over time and thus what is potential at one point may cease to be at another. This is also shown by the choice between being a butcher and baker. When our young man chooses to be a full-time baker, his potential of being a life-long full-time butcher ceases. That a potentiality can cease is a common sense view. One can one day potentially take any one of five different routes to work, let us assume. There are five potential routes when one leaves the house. But once one of those routes is taken, the other four cease being potential. We are not denying that there can be tensed attributions of potentiality: we may say that there were, earlier, five potential routes. But, similarly, we can make tensed attributions of powers: something could have had a power until a certain time, and thus have had the potential to act, but then lost that power and thus lost the potentiality. Someone may have had the power of sight, for instance, and thus potentially could have seen their hand in front of them. They had this power even when they were not exercising it, for instance when their eyes were closed in a deep sleep. But then at a later point they become blind and lose the power. At that time they can no longer potentially see their hand in front of them. The powers account thus constrains the potential. We might see this as an important task because we usually want to distinguish real potentialities from mere logical possibilities. It is logically possible that someone jumps to the moon, for instance, but this is only because the constraints on logical possibility are so weak. Something is logically possible just if it contains no contradiction but much more

264

S. Mumford and R. L. Anjum

than that is required for a potentiality. Many things will be logically possible that we do not think of as real potentialities. There is nothing to stop someone having a very broad notion of potentiality but we can question the value of such a notion. If it collapsed into mere logical possibility we may think we have lost a vital distinction, for instance. And the most useful, and used, notion of potentiality seems to be one that is more constrained. By grounding potentialities in the powers of things, we have a principled basis for narrowing the scope of the potential.

3 Naturalising Potentiality It is claimed that the powers that ground potentialities have a theoretical advantage of being naturalistic. What is meant by this? The exact meaning of naturalism is not easy to determine but in the case of powers we can point to two relevant, no doubt connected, features. First, we can say that powers are a part of the natural world and, second, we can say that they are available for empirical investigation. More detail should to be added to both these claims. The claim powers are a real part of the natural world is not easy to defend: nevertheless it could still be done. Claims that some category has real existences in the world may be a very fundamental type of claim, as we get in this case. What we take to be real may depend, for instance, on what we adopt as our criterion of reality and there is unlikely to be any direct argument that tells us what criteria of reality to hold. There is one candidate reality test that is ancient and puts powers at the centre. This is the Eleatic Stranger’s reality test as articulated in Plato’s Sophist: I suggest that anything has real being that is so constituted as to possess any sort of power either to affect anything else or to be affected, in however small a degree, by the most insignificant agent, though it be only once. (Plato Sophist 247d–e)

This tells us that powers are existents par excellence: the very paradigms of real beings. To be real is to be powerful and thus to contain potentiality. The problem, however, is that those who are against powers are simply at liberty to adopt some alternative reality test. The empiricists, for instance, while not explicitly presenting a reality test seem to have one in mind. One should accept as real only that which can be known through the senses. Given that a test of reality is just about the most basic of philosophical commitments one could have, it is hard to find any direct argument to choose between rival reality tests for what could be more basic than the test in terms of which we could justify the test itself? There are, however, less direct ways in which we might choose between different criteria of reality. We might think that one is more coherent than another, for instance, if it leads to greater consistency and if more theoretically productive. Apart from that debate, however, it should be noted that it is also open to those who believe in powers to claim that their acceptance is still justified even on another reality test. The empiricist claim that powers are unknowable might be challenged, after all. This challenge has more going for it than empiricists allow, we will claim shortly.

Powers and Potentiality

265

In what manner do powers exist as part of the natural world? They are frequently discussed as if they are property-like. We need not rule out the possibility of a trope-ontology, in which case there could be power tropes, and there may even be a place for powers in a nominalist metaphysics where they are explained in terms of resemblances among particulars. Setting those alternatives aside, let us stick with their property-like characterisation, for this requires more investigation. We certainly speak of powers as if they are like any other property of a thing. Just as we might say an object is red and spherical, we might also attribute powers or dispositions to it, as when we say it is explosive, elastic or soluble. Does this mean, perhaps, that there are two types of properties that things can instantiate: dispositional properties, which are the powers, and non-dispositional or as some say categorical properties? Some have made this claim (Prior 1985, for instance). Whether it is right to divide reality in this way, which we can call property dualism, is a matter of debate (see Mumford 1998: Chap. 5). But we can note that there is another approach, which sits better with an Eleatic reality test. According to Shoemaker (1980), properties themselves are not basic, fundamental entities but can be explained as clusters of powers. Any real property must, by the lights of the Eleatic reality test, involve the having of a power. If the categorical properties are not powerful, then one might wonder why we should take them to be real. Anything that was not powerful could make no possible difference to the world, it would seem, so why should we bother to posit it? (Rather perversely, in Prior’s dualism it is the dispositions that are impotent as all the work is being done by the categorical causal bases of the dispositions. See also Prior, Pargetter and Jackson (1982). We could then ask the same question of those impotent dispositions: why would we need to posit them?) However, if we look carefully, we can see that even those properties that are offered as supposedly good examples of categorical properties are nevertheless powerful, and thus do deserve their place in our ontology. Shapes are often offered as categorical properties but if we consider sphericity, we can see that anything that is spherical thereby has a cluster of causal powers in virtue of that. Something spherical is disposed to roll in a straight line on a flat surface, for instance. If the Shoemaker theory is right, therefore, there is only one kind of property and all properties are clusters of causal powers. This makes powers central existents in the natural world. Everything that exists bears properties (again allowing that these may be tropes or resemblance classes, if realism about properties is false). Thus everything that exists has powers and, in virtue of that, contains potentialities. A spherical object has a potential to roll in a straight line on a flat surface, for instance, because it has a power to do so, whether it manifests that power or not. And this leads us to a second major point: such powers—the grounds of the world’s potentiality—are empirically accessible to us. We can know of properties through their effects on us, including their effects on our instruments and measuring devices. How we come to acquire knowledge of powers is no simple matter, however. One way is through testing things. When I pull and release a rubber band, I can see that it is elastic: it has a power to stretch and contract. Here I am coming to know the power through its manifestation. This is not to say that the existence of the power is

266

S. Mumford and R. L. Anjum

dependent on it manifesting, however (Wright 1991), as some powers will fail to manifest themselves even when tested. The manifestation is a symptom of the power, just as one may come to suspect there is illness, such as measles, by seeing its symptom of red spots. And like all symptoms, they cannot be depended on to guarantee the presence of a power. It may be possible to mimic the symptom of a power just as red spots may have some cause other than measles. We can, however, sometimes be in a position to know of the presence of a power even without testing it to the point where it manifests itself. I may be able to see on inspection that an object is like others that have manifested their powers (see Quine 1960: 224). I may come to know that my rubber band is elastic, for instance, although I have never stretched it, but simply because it otherwise resembles rubber bands that have been pulled and stretched. This way of knowing of powers is fallible and inductive, of course, like much of our empirical knowledge. But even without this directly inductive argument, it still may be possible to come to know the powers and potentialities of a thing without testing it but simply through an inspection of it. I may know that a rock I find on the ground would be capable of smashing a certain window of my car not because I have seen a rock smash a car window (which I haven’t) but because I see how big the rock is, feel in my hand how hard it is, feel its weight, and so on, and similarly I can see how thin the glass is. I might, on the other hand, see from a simple inspection of a coin that it would have a 50:50 chance of landing heads (and tails) not because I have watched coins being tossed but because I can see that it is a flat disc. Such knowledge is certainly fallible and it is arguably indirectly inductive because it might be thought that I couldn’t know of such powers unless I already had a store of background knowledge, concerning laws of nature and so on. Hume argued along these lines, claiming that Adam, encountering water for the first time and without any background knowledge, could never know that it has a power to quench thirst or drown a man (Hume 1748: 19–20). There is, however, a way of knowing about powers that is perhaps more direct and more significant insofar as it might meet the empiricist challenge that powers are unknowable. We creatures are embodied causal agents and patients. We exercise causal powers though our bodies and have them exercised upon us. This creates the possibility for us of direct, non-inferential knowledge of powers, particularly in the case of agency. This is an old idea, which we can find in Locke (1690: II, i, 2) for instance, and Hume was aware of it and denied it (Hume 1739: Appendix 632– 633, to be inserted Bk I, p. 161, line 12). But there may now be more sophisticated versions of this view that are able to overcome Hume’s challenge. It is no simple matter, however (see Mumford and Anjum 2011: Chap. 9). Hume had alleged that the most we could experience though agency was a constant conjunction between our own volitions and bodily movements and he was thus able to apply his well-known constant conjunction account of causation. But we can certainly challenge the plausibility of such an account in which willing and acting are disunified. It seems more plausible that the volition must occur simultaneously with the act (an issue that we may consider for causation more generally). When one intentionally raises one’s arm, for instance, it is not as if there is a temporal gap between the willing and rising. One may form an intention at some future point to

Powers and Potentiality

267

raise one’s hand but this cannot be what makes the raising of one’s hand intentional for this prior decision might be forgotten or one might have a change of mind. Instead, one must be intending to raise one’s arm throughout the action. It might be possible to lose that intention half way through the process, and lower the arm before it was fully extended. So the intention must remain simultaneously with the act all the way through to completion. This creates the possibility of a reunified account of agency in which a power and its exercise can be known in one and the same experience. In the case of proprioception, for instance, our muscles give us a sense of required effort. It is essential that this works with a feedback mechanism so that we can adjust our efforts as required. When I exercise my power to lift a wineglass, for instance, it is important that I don’t use too much effort as I might throw its contents over my shoulder. And when I lift a suitcase, it is important I exercise enough effort, otherwise I might not get it off the ground. Proprioception is a single sense, providing a single experience in which the power is having its effect. It is therefore a very good candidate for the way in which we come to acquire our notion of power. Through our bodily experience, we can directly know of powers and how they work. Once the concept of power is so acquired, we can apply it analogously to other cases that we know less directly. But even if we knew of only one power directly, that would be enough to meet the empiricist challenge that there is no original experience from which we have acquired the concept and that the notion of power is therefore meaningless. This kind of direct knowledge of powers is no doubt the exception, but because powers are a part of the natural world, they are scientifically accessible and this allows us insight into the potentialities that they support. Hence, we can understand the potentialities within the chemical elements the more that we understand chemistry. We can predict how the chemical elements with behave and interact with each other, for instance. By investigating and understanding the powers that the individual elements have, we understand which potentialities they create. We know, for instance, that oxygen and hydrogen have the potentiality to become water because we understand the powers they have in virtue of their subatomic structure. A virtue of the powers account, therefore, is that potentialities become naturalistic, capable of scientific investigation, explanation and prediction.

4 Production Through Stimulus and Response It has not yet been explained, however, exactly how powers are able to be productive, which is the cornerstone of a theory in which they sustain potentialities. There are two main models for how they do this. One is the stimulus-response mechanism, which we will reject in favour of a mutual manifestation model. The stimulus-response model tells us that powers are able to produce a response when they are appropriately stimulated. This was of understanding production is supported by the datum that typically there is an appropriate stimulation type and manifestation type for each type of power. When something is soluble, for instance,

268

S. Mumford and R. L. Anjum

we can identify the stimulus, which triggers the exercise of the power, to be immersion in water and the response or manifestation to be dissolving. Hence, the disposition or power is to be understood in terms of the ordered triple {in water (stimulus), soluble (power), dissolves (response)}. For almost all powers, we can identify a similar ordered triple, for example {drops, fragile, breaks}, {pulled, elastic, stretches}, {ignited, explosive, explodes}, and so on. We may allow a class of multiply manifested dispositions where the first or third term in the ordered triple is disjunctive. An elastic object may also bounce if dropped, for instance. It may further be alleged that because we can know the appropriate triple for each power-term through conceptual analysis alone, they are analytically true. It is thus analytic that solubility is the power that manifests itself in dissolving upon being placed in liquid. This is still probably the orthodox model. It is assumed by Barker (2009), for instance, as a basis for criticising dispositionalism. There are reasons why we might be reluctant to accept the stimulus-response model for the exercising of powers. The first may be a relatively small matter but while almost all powers have appropriate stimuli, not all do. Some powers we can think of as spontaneous, requiring no stimulus. Radioactive decay is the most common example from physics and from psychology we can note that some people are described as spontaneous. This seems to mean, in both cases, that they are liable to change for no apparent reason, that is, without stimulus. A particle simply decays at some point without any cause other than it is simply in its nature to decay at some point, roughly specified by the half-life for particles of that type. And spontaneous people can simply up and do something for no reason, except that they are changeable as part of their nature, requiring no causal stimulus. Perhaps these could be sidelined as special cases, where the first member of the ordered triple is left blank, but there is a more serious objection to the stimulus-response model. The idea that powers stand in need of stimulation is already a thesis that a thoroughgoing dispositionalist is likely to reject. This is more than just the point above, that some powers manifest spontaneously, but is based more on a rejection of what can be called passivist accounts of nature (see Ellis 2001: 7). A passivist metaphysics is one in which there is no internal principle of change within things. They do not contain potentialities within them, for instance. Rather, they move or change only when they are pushed to do so from outside. Particulars are passive recipients of change rather than movers or instigators of change themselves. If this were the case, we would be wrong to look to powers as the basis of potentialities. Instead, we should be better placed to look to those exterior things that were the genuine producers of change. But what would they be? It looks like they are the true causal powers because it is they that make things happen. There are some metaphysics is which nothing, or virtually nothing, is causally active, such as occasionalism, but such metaphysics are not very attractive. This general concern about passivism can be found in the specific instance of the stimulus-manifestation pair. The power itself is depicted as being on its own passive. It will produce no change unless it is stimulated to do so by the appropriate stimulus. But then this makes the stimulus look like the power: it is the element that has the power to produce change, while the power itself has, on this account, been rendered

Powers and Potentiality

269

impotent. Something seems to have gone wrong with the account, therefore, as it disempowers the very powers it is supposed to explain. The activation of the exercise of a power is ‘explained’ only by allotting responsibility entirely to this stimulus that is itself left unexplained and, furthermore, it is hard to see how it could be explained if the elements in the world are all said to be passive. There is, however, an alternative model for the exercise of a power’s potentiality. This is the mutual manifestation model as developed by Martin (2008).

5 Mutual Manifestation Partnerships Instead of taking the stimulus and power as playing unequal and different roles in the production of a manifestation, the mutual manifestation model depicts the situation as more of an equal partnership. This solves some of the quandaries outlined above. We do not have to say and justify taking the power to be passive and the stimulus to be active. Instead, a manifestation occurs simply when a number of powers that are mutual manifestation partners have come together and produced something jointly that neither could have produced alone. We can see how this works with some of the examples used earlier. Although sugar can potentially be dissolved, this occurs only when it meets the right mutual manifestation partner, namely being in liquid. The final effect, the sweet solution, is produced by both the sugar and liquid together. There seems to be no solid reason why we should privilege one of these two partners and depict it as the active partner—the trigger or stimulus—while relegating the other partner to a passive supporting role. There seems to be no ontologically serious difference between the positions of the sugar and the liquid. There may be a pragmatic difference that explains why we sometimes tend to view one partner as more active than another but it is hard to defend this on a metaphysical level. Perhaps we think of the liquid as the trigger of dissolving because our sugar cube sits there, seemingly doing nothing, while the liquid moves and flows into the sugar cube, breaking it down as it goes. But while we may tend to think of the process in this way, there is nothing to stop us taking the opposite view. Sugar could be seen as flowing into the water and making it sweet when it does so, enforcing a change in the liquid. Similarly, we tend to think of an ice cube as active, cooling the drink in which it is placed. But we could also view this as a single process in which the temperature in the glass reaches an equalised distribution. While the drink cools, the ice cube melts. Another example that was used earlier also illustrates the model of mutual manifestation partnering. We may say that a sperm has the potential to turn into a man. But this is a power that manifests only when the sperm meets the appropriate partner for doing so, namely an egg. Neither the egg nor the sperm can produce the man alone: both are necessary. And nor can we say on any principled basis that one plays an active role while the other is passive. There is a single process involved that starts when they come together and leads to there being a human being.

270

S. Mumford and R. L. Anjum

Suppose we are persuaded by the initial credibility of a mutual manifestation model. This would, in the first place, have an implication for what we say about potentialities. When we say that some particular x is potentially F, what we must mean is that x has some power that has one or more mutual manifestation partners, y, z, …, whose joint manifestation is in the production of F. When F remains an unactualised potentiality, this means that x is not currently partnered with y, z, …, and thus F is not currently produced (mutatis mutandis for tensed attributions of potentiality). How many mutual manifestation partners should we say are involved? The answer is any number. Effects are typically polygenic, meaning that they are produced by many powers acting together. Hence the exact temperature of a room may be determined by many different powers working together: the power of a draft to cool; the power of a radiator to heat; the power of the sunlight to warm the room and similarly from the body heat of any persons present; the heat loss through the ceiling, and so on. While at the lower end of the scale, we have already seen the theoretical possibility of a power able to exercise with no other manifestation partners. Spontaneously manifested powers will be those able to act alone, requiring no other additional powers working in combination. We have no real account of potentiality arising out of powers, however, unless we can tell a plausibly story of how such partnered powers are able to produce change. If powers could produce no actual changes then they would be a poor basis for grounding potential changes. It is vital, therefore, that we provide a credible account of the way in which powers produce their manifestations, typically jointly. Martin, who provided the notion of mutual manifestation, has his own story to tell: You should not think of disposition partners jointly causing the manifestation. Instead, the coming together of the disposition partners is the mutual manifestation: the partnering and the manifestation are identical. This partnering-manifestation identity is seen most clearly with cases such as the following. You have two triangle-shaped slips of paper that, when placed together appropriately, form a square. It is not that the partnering of the triangles causes the manifestation of the square, but rather that the partnering is the manifestation. (Martin 2008: 51)

This is a nice image of how manifestations are produced and we will show that there is something very much right in it. However, we will also say that the example fails if it is supposed to be an adequate analogy for the case of production. The kind of production of change that concerns us, which is the kind of change that regularly occurs in the natural world, would exhibit at least the three following disanalogies with Martin’s two triangles that make a square. First, Martin’s analogy suggests that the triangles form a square immediately, the moment they are brought together whereas partnerships often result in processes that take time before their final product appears. Second, Martin’s example suggests that composition is merely additive. The area of the square that is formed would be equal to the sum of the areas of the two triangles. But we know there are many cases of production that are nonlinear: not merely additive but where the quantity of a thing produced is greater or less than the quantities of the components that produced it. Third, Martin’s example does not allow for any degree of emergence. There seem to be cases where the powers of a whole are not simply the addition of the powers of the component

Powers and Potentiality

271

parts. Martin’s analogy has the two triangles always forming a square but in nature we have novel phenomena that are mundanely produced when powers come together. Component powers may undergo interactions that involve an alteration in certain respects. If this feature were conveyed in an analogy, it would be like saying that when the two triangles came together, they in fact formed a circle or some such shape other than a square. What this suggests is that the model of mutual manifestation partnering that Martin offers is actually a model for mereological composition of powers whereas what we want from an account of the production of manifestations is indeed a causal model, contrary to Martin’s claim. These amendments to Martin’s account will be set out in more detail shortly but first something positive can be said in its defence. The idea that manifestation occurs when partners ‘come together’ is basically sound. What exactly constitutes their coming together is something that needs to be detailed further, however. Perhaps some partners will do their work only if they are brought into physical contact but in other cases it seems they need only to be in the same vicinity. The factors that operate on our room, for instance, and determine its temperature, are all within a region but not necessarily in direct contact. The radiator on the wall heats the room as does a person stood on its opposite side. This is a case where the effect is also spread throughout that region. It may be, then, that it is all well and good for the mutual manifestation partners to be spread within a spatiotemporal region if the effect is also. What would seem problematic would be where a cause was located away from its effect. Causes can travel, of course. The sun warms the room even though it is located outside of it but what really concerns us is rather the power of the heat that has travelled into the room through its windows and walls. One crucial respect in which Martin’s model is right is that we do not need any further element other than the mutual manifestation partners in order for their joint manifestation to be produced. A possible concern about this claim would be that some partners when brought together need some further stimulus, perhaps we could call it a catalyst, to get started on their work. There are some chemicals, for instance, that are capable of interacting but do so only with the addition of a catalyst. And to use a less scientific example, adult pandas of opposite sexes are mutual manifestation partners for the production of baby pandas but are notoriously difficult to get started. They need aphrodisiac catalysts if they are to exhibit their powers. However, further reflection shows us that this concern is not serious. When we spoke of a stimulus earlier, we noted that this was something that had to be powerful, in order to do its stimulatory work, and thus the stimulus should be taken as itself another power. We can now saw the same thing. This further stimulus—the catalyst that gets powers to do their work—should itself be taken as a further power that ought to be included within the mutual manifestation partnership. Thus, where we have chemicals A and B that we say are capable of producing M, but do so only if we add catalyst C, the situation should be understood as one in which A and B are an incomplete partnership for the production of M. More accurately, we should say that it is A, B and C that are the complete partners for the production of M. And similarly for the production of baby pandas. A male panda has the power to produce

272

S. Mumford and R. L. Anjum

a baby but only in partnership with a female panda and, let us assume, the requisite aphrodisiac, whatever that might be. As soon as everything is in place—all the complete partnership with respect to manifestation M—then it seems that they begin their work immediately. If this is a model of causation, then it shows us that effects will be simultaneous with their causes. This would have to happen because it would be otherwise mysterious why there would be a time gap between all the requisite partners being in place and their effect commencing. If there were a time gap, what would lead to its end and the beginning of the effect? One would have to either concede that there was no answer to why the time gap ended, and why it took the time that it did, or allow that there was some extra element that was introduced that led to the start of the effect. But then this extra element is in the same position as the catalyst. This will have been to concede that everything had not after all been in place for the production of M. We had only an incomplete partnership for M. And then we should allow that this further element completes the partnership for M and M commences immediately otherwise we will be back to the same problem that this further element was meant to address. In other words, unless we accept that the mutual manifestation partners produce their joint effect as soon as they are together, we are left with either some inexplicable time gap, that for no reason ends, or we have a potential infinite regress of further elements that are needed to explain the end of the time gap. These comments are made in support of Martin’s notion of mutual manifestation partnerships and such a notion is indeed supported. When we do not support, however, is the account quoted above of the way in which the manifestation partners become the manifestation. We said that Martin’s account looked more like mereological composition whereas we need something more like causation. Let us now look, then, in more detail at how the model needs to be amended.

6 Amendments to Martin’s Notion of Mutual Manifestation The first defect of Martin’s account of manifestation is the suggestion that the full effect appears immediately, as soon as the two triangles come together. They make a square instantly, rather than gradually merge into one. As soon as they touch, appropriately orientated, they form a square. We have suggested above that powers begin to act as soon as they are partnered, and we stand by this, but it does not entail that they will have acted fully at that point. Kant saw this. He accepted the simultaneity of cause and effect, for instance, but he realised that simultaneous causation does not entail instantaneous causation: The great majority of efficient natural causes are simultaneous with their effects, and the sequence in time of the latter is due only to the fact that the cause cannot achieve its complete effect in one moment. But in the very moment in which the effect first comes to be, it is invariably simultaneous with the causality of its cause. (Kant 1781: A203)

Powers and Potentiality

273

Immediately before, Kant gave as an example the heating of a room by a stove. He offered this as a case of simultaneity of cause and effect. As soon as the stove is doing its work, the room warms. But he says, in the passage quoted, that it can take time for the cause to have its full effect. Hence, the stove may have the power to heat the room to 25 °C. But it cannot do so immediately. It takes time. But while ever the cause is working, it has its effect of warming the room. And as soon as the stove is extinguished, it ceases to warm the room. If the room temperature begins at just 10°, however, then clearly it will take an extended period of time before it reaches 25°. Many other cases will similarly take time. With the example of sugar and liquid, for instance, it takes time for the reaction to be fully realised even if it begins as soon as the manifestation partners are together. As soon as the sugar is in the tea, for instance, a process begins whereby the liquid ‘eats into’ the sugar cube or granules, gradually taking it into solution, until the process has run its course and the sugar is no longer in solid form. If one of the partners is withdrawn, the process can be interrupted before it is complete: if the liquid is rapidly evaporated, for instance. But as long as they are together, without any preventer or interferer, the process will be seen through to completion. Martin’s analogy fails in this respect, therefore. It could cover only those cases where power partners are able to produce their joint manifestation instantly. There may be some such cases of full, instantaneous manifestation—perhaps within the realm of fundamental physics—but the commonplace examples with which we are familiar are likely to be ones that take time. It need not be a very long time that is involved. When a massive hard rock with momentum meets a fragile window, it seems very quickly to manifest its effect in breakage. But this too is a process: just a very fast one. On impact, stress appears in the glass, is distributed throughout its area but, when it quickly surpasses its tolerance threshold, results in cracks, then fault lines, then breakage and shattering. This may be too quick for the human eye to see but it is a process nonetheless. Martin’s analogy with triangular slips of paper forming a square does not work for many cases of mutual manifestation, therefore. Powers to do not ground potentialities in precisely that way. More typically, they contain within themselves a principle for uniting in mutual partnerships for the commencement and continuation of a process. They dispose towards the completion of this process but, of course, whether they succeed in doing so will be dependent on the contingencies of whether they are brought together with the appropriate partners in the first place and whether they are allowed to see their work through to completion, unfettered and uninterrupted. Any truths about potentiality that come out of the theory of powers should be understood in this way and it seems indeed that this is how we understand potentiality. A young man is potentially a scholar and can embark on the process of training to be one yet may be diverted into other things before that training is complete. The second way in which Martin’s account of mutual manifestation should be revised is the way in which it is restricted to linear, additive composition. Again, there may be some cases in which powers combine to produce a manifestation in an additive way, but it is also clear that not all do. Hence, usually the potentialities that powers ground are not simply the addition of the contributing powers. In the case of forces, which we may justifiably understand as powers, it seems that these often do

274

S. Mumford and R. L. Anjum

compose in an additive way. A group of men may pull on a rope, for instance, and the total force exerted on the rope may be equal to the sum of the individual forces contributed by each participating man. This means that we can describe the composition of these forces with a linear function, where the total force F = fa + fb + fc + fd, where fa is the force exerted by a, and so on for b, c and d. If powers did always compose in this way, it would certainly leave us with a simple picture of the world. The composition of powers would follow basic mereological principles. Martin’s example suggests this insofar as the area of the square formed by the two triangular slips of paper is equal to the sum of the areas of the two triangles. This would be false for many cases of manifestation, however. Instead, we could have cases where the area of the square was greater than or less than the sum areas of the component triangles. It is not easy to explain in all its detail what exactly is meant by a nonlinear system. We will, however, restrict ourselves to the specific case that interests us of nonlinear composition of powers. Such a nonlinear composition will be one where the extent of the output or result is not merely equal to the sum of the inputs or components. We can give some relatively easy examples. Money has the potential to make people happy. We can understand this in terms of it having a power to do so. But the way in which it does so is nonlinear to the extent that if we plotted the relationship between wealth and happiness on a graph, we would not get a straight line. Additional money tends to produce increased happiness at a certain level but once one’s material needs have been met each unit of additional wealth is likely to have a diminishing positive impact on happiness. Extremes of wealth might even lead to unhappiness. A possible explanation for how we can get nonlinear composition of powers— and there may be different explanations for different cases—is that the composing powers can interact. Martin’s description of the case suggests that the individual contributing powers are unaffected by their partaking of a partnership—the triangles do not change their size or shape when they come together—but there are reasons to think this is often false. As an example, we can consider noise levels at a drinks party. A typical conversation is conducted at around 60 decibels (dB), and the first couple to arrive may speak to each other at that noise level. When a second, third, fourth and fifth couple arrive, however, they need to speak louder to be heard because of the other conversations occurring. The result is that each conversation will be held at a louder level—around 70 dB—than if each of those conversations had been conducted alone and in isolation. Each couple speak loader so they can be heard by their companion, which forces all the other conversations to become louder for the same reason. The sound level of all the conversations compose in a nonlinear fashion, therefore, because being part of a single system affects the individual components of that system. But even if three conversations are being held at 70 dB, this does not mean that the total dB of all the conversations is 210 dB. Noise also composes in a nonlinear way: indeed the loudest possible noise is only 194 dB. There is one other case of nonlinear composition that we should mention. This is of particular relevance to the account of potentiality as it shows how complicated

Powers and Potentiality

275

and specifically non-additive the composition of powers is. We could have two powers that individually disposed in the same direction but when partnered together disposed in the opposite direction. We cannot presuppose, therefore, that if a disposes towards F and b disposes towards F that a + b together will dispose towards F. Indeed, they can dispose towards not-F. This sounds counterintuitive so we need a credible example to persuade us to accept it. A good one comes from pharmacy. There can be cases where there are two drugs that each dispose towards a certain outcome but which, when taken in combination, dispose in the opposite direction. The drug clonidine, for instance, disposes towards the lowering of blood pressure, as do beta-blockers such as propranolol. It was found in at least one study, however (Warren et al. 1979), that when taken in combination the two drugs tended to increase blood pressure. We can call this an antipathetic partnering. We now understand the mechanism that produces this particular antipathetic effect. It shows how we need to view the situation holistically, if we are to understand the potentialities within a situation. We cannot consider just the powers of the individual components but what they are partnered with, how those partners will interact, and how the situation as a whole will tend. This is why a doctor should always check what other medication the patient takes before issuing a new prescription. We are led to the third amendment to Martin’s model of mutual manifestation. Martin assumes that the two triangles that are brought together have to form a square. But we have already seen enough to realise that there can be a transformation of powers when they are brought together into partnership. It might be, to continue with Martin’s analogy, that a triangle brought together with another form jointly a circle rather than a square. We have already seen a case where two powers towards F do not, when joined together, dispose towards F. We may extend this and note that we should not in general assume that where a has power P and b has power Q, that if we bring a and b together we will get something with the joint power {P & Q}. We do not, as Martin’s account suggest, simply add all the powers together. An example that nicely illustrates why this does not occur is the partnering of sodium and chlorine: Sodium is a soft, bright, silvery metal. It can float on water and, when doing so, decomposes with the production of hydrogen and the formation of hydroxide. Sodium may ignite spontaneously on water, depending on the amount of oxide and metal exposed to the water. It normally does not ignite in air at temperatures below 115 °C. Chlorine is a greenish– yellowish ga0073 that is a respiratory irritant. As little as 3.5 p.p.m. (parts per million) can be detected as an odour, and 1000 p.p.m is likely to be fatal after only a few deep breaths. Chlorine is so toxic it was used in gas warfare in 1915. From this information it is impossible to predict that sodium chloride should be the benign compound that makes the oceans salty and is an essential compound for life, not to mention potato chips and margaritas. It is possible that a knowledgeable chemist could make this prediction—not today, but perhaps sometime in the future—but at this point ‘salt’ appears as an emergent property of sodium and chloride. (Rothschild 2006: 153)

Whether it is correct to call this genuine emergence is a matter for debate but what we seem to have is a case where the powers of sodium chloride are completely different from those of sodium and chlorine separately. A novice in chemistry when

276

S. Mumford and R. L. Anjum

hearing what sodium and chlorine can do individually might assume that sodium chloride would be a highly dangerous substance, far from suitable for human consumption. Again, therefore, we must consider the composed powers holistically: considering not just the powers of the components but how they interact and can be transformed by those interactions. Such interactions will be vital if we are to understand what potentialities the powers can ground.

7 Conclusion This chapter has attempted to address the question of how potentiality relates to the ontology of powers. It seems that there are some putative virtues of such an approach. This seems good for the powers ontology because one of the main things to recommend it is that it can explain a host of other metaphysical categories. If we are to summarise the strengths of accounting for potentialities in terms of powers we can say, therefore, that the account shows how potentiality is grounded in the actuality of perfectly naturalistic, property-like existents, which the powers are. We can say that the account explains the constraints on potentiality: how some things, but not all, are genuinely potential. The account can explain how attributions of potentiality can be tensed and what is potential at one point might not be at a later point. Furthermore, by grounding potentialities in powers, we are able to make them empirically accessible, and as much ready for our knowledge, understanding, explanation and prediction as anything else in science. We also said that the powers account was plausible only of it explained the way in which powers grounded potentiality. To do this, we explained how powers were able to be productive, dependent on the contingencies of circumstance. To be a genuine potentiality, it must be possible for it to become actual though not necessary that it do so. We showed how such an account can indeed be given. There were two models for how powers produce their manifestations. One was the stimulus-response model but we argued that the mutual manifestation model was a better description of how powers exercise. We offered amendments to that account, however, so that it fits better with what we know on empirical grounds are some of the facts of causation. It is a further advantage of the powers account that it can cohere with such tricky cases of potentiality.

References Armstrong, D. M. (1968). A materialist theory of the mind (Rev. ed.). London: Routledge (1993). Armstrong, D. M., Place, U. T., & Martin, C. B. (1996). Dispositions: A debate. In T. Crane (Ed.). London: Routledge. Barker, S. (2009). Dispositional monism, relational constitution and quiddities. Analysis, 69, 242–250.

Powers and Potentiality

277

Bird, A. (2007). Nature’s metaphysics. Oxford: Oxford University Press. Ellis, B. (2001). Scientific essentialism. Cambridge: Cambridge University Press. Ellis, B., & Lierse, C. (1994). Dispositional essentialism. Australasian Journal of Philosophy, 72, 27–45. Geach, P. T. (1961). Aquinas. In G. E. M. Anscombe & P. T. Geach (Eds.), Three philosophers (pp. 65–125). Oxford: Blackwell. Harré, R., & Madden, E. H. (1975). Causal powers: A theory of natural necessity. Oxford: Blackwell. Hume, D. (1739). A treatise of human nature. L. A. Selby-Bigge (Ed.). Oxford: Clarendon Press (1888). Hume, D. (1748). An enquiry concerning human understanding. P. Millican (Ed.). Oxford: Oxford University Press (2007). Kant, I. (1781). Critique of pure reason (N. K. Smith, Trans.). London: MacMillan (1929). Locke, J. (1690). An essay concerning human understanding. Oxford: Clarendon Press (1975). Martin, C. B. (2008). The mind in nature. Oxford: Oxford University Press. Molnar, G. (2003). Powers: A study in metaphysics. S. Mumford (Ed.). Oxford: Oxford University Press. Mumford, S. (1998). Dispositions. Oxford: Oxford University Press. Mumford, S. (2004). Laws in nature. London: Routledge. Mumford, S., & Anjum, R. L. (2011). Getting causes from powers. Oxford: Oxford University Press. Plato. Sophist (F. M. Cornford, Trans.). In E. Hamilton & H. Cairns (Eds.), The collected dialogues of Plato (pp. 957–1017). Princeton, NJ: Princeton University Press. Prior, E. (1985). Dispositions. Aberdeen: Aberdeen University Press. Prior, E., Pargetter, R., & Jackson, F. (1982). Three theses about dispositions. American Philosophical Quarterly, 19, 251–257. Quine, W. V. O. (1960). Word and object. Cambridge, MA: MIT Press. Rothschild, L. J. (2006). The role of emergence in biology. In P. Clayton & P. Davies (Eds.), The Re-emergence of emergence (pp. 151–165). Oxford: Oxford University Press. Shoemaker, S. (1980). “Causality and properties”. In S. Shoemaker, Identity, cause and mind (expanded ed., pp. 206–233). Oxford: Oxford University Press, 2003. Warren, S., Ebert, E., Swerdlin, A.-H., Steinberg, S., & Stone, R. (1979). Clonidine and propranolol paradoxical hypertension. Archives of Internal Medicine, 139, 253. Wright, A. (1991). Dispositions, anti-realism and empiricism. In Proceedings of the Aristotelian Society (Vol. 91, pp. 39–59).

Author Biographies Stephen Mumford is Professor of Metaphysics in the Department of Philosophy at Durham University, United Kingdom, and Professor II in the School of Economics and Business at the Norwegian University of Life Sciences (NMBU). He works mainly in metaphysics, especially on causation. His books include “Dispositions” (Oxford University Press 1998), “Laws in Nature” (Routledge 2004), “Getting Causes from Powers” (Oxford University Press, 2011, with Rani Lill Anjum), “What Tends to Be. The Philosophy of Dispositional Modality” (Routledge 2018, with Anjum) and “Causation in Science and the Methods of Scientific Discovery” (Oxford University Press 2018, with Anjum).

278

S. Mumford and R. L. Anjum

Rani Lill Anjum is a Research Fellow at the School of Economics and Business at the Norwegian University of Life Sciences, Norway. Anjum’s current research is on the nature of causation and its relation to scientific methods, in particular in medicine. Other related research areas are dispositions ontology, propensity theory of probability and complexity and emergence. She is the author of “Getting Causes from Powers” (Oxford University Press 2011), “Causation—A Very Short Introduction” (Oxford University Press 2013), “What Tends to Be. The Philosophy of Dispositional Modality” (Routledge 2018) and “Causation in Science and the Methods of Scientific Discovery” (Oxford University Press, 2018), all co-authored with Stephen Mumford.

From Potentiality to Possibility Barbara Vetter

1 Introduction Things have potentials: tea-cups are fragile, sugar is soluble, and if you are reading this, you have the ability to read English. Dispositions such as fragility and solubility, as well as abilities such as your ability to read English, are potentialities or potentials—they are properties that enable their bearers to behave in certain ways. Some potentials have names, such as fragility or solubility, others do not, such as a baby’s potential to grow. All potentials are characterized by what they are potentials to do (or to be, or to become): fragility is the disposition to break, solubility is the disposition to dissolve, and so on. All potentials concern, at least, what a given individual can do. I believe that our intuitive grasp of the properties I mentioned already provides some good pre-theoretical understanding of potentiality. A more theoretical understanding can be gained by looking at the connections between potentiality and possibility. Possibility is global: its being possible that such-and-such is not primarily a fact about any one particular object; it is a fact about how the world could turn out or could have turned out to be. Hence our intuitions about what is possible and what is not can be captured by postulating, for everything that is possible, a world that does turn out to be that way. Potentiality, on the other hand, is local: a potentiality is a property of a particular object. That I have the potential to write this paper is first and foremost a fact about me; it is a property that I possess. The distinction is paralleled by the relation of essence to necessity. Essence, like potentiality, is local: a property is essential to a particular object. Necessity, like possibility, is global: its being necessary that such-and-such is not primarily a fact about one particular object, but a fact about how the world must be. The difference between essence and necessity has been pointed out and studied by Fine (1994). B. Vetter (&) Freie Universität Berlin, Berlin, Germany e-mail: [email protected] © Springer Science+Business Media B.V., part of Springer Nature 2018 K. Engelhard and M. Quante (eds.), Handbook of Potentiality, https://doi.org/10.1007/978-94-024-1287-1_11

279

280

B. Vetter

While essence had been traditionally understood in terms of necessity (a property F is essential to an object x just in case it is necessary that x has F if x exists) Fine argues that such a reduction of essence to necessity yields the wrong results. Being a member of the singleton set {Socrates}, or being such that 2 + 2 = 4, are necessary properties of Socrates, but they should not be thought to constitute his essence. Essence is more discriminative than necessity: not any necessary truth is part of a given object’s essence; only those truths that are constitutive for that object’s identity are allowed into its essence. Contrary to orthodoxy, Fine then goes on to suggest that we should understand necessity in terms of essence, rather than the other way around: Given the insensitivity of the concept of necessity to variations in source, it is hardly surprising that it is incapable of capturing a concept which is sensitive to such variation. Each object, or selection of objects makes its own contribution to the totality of necessary truths; and one can hardly expect to determine from the totality itself what the different contributions are. … Indeed, it seems to me that far from viewing essence as a special case of metaphysical necessity, we should view metaphysical necessity as a case of essence. (Fine 1994, 9)

In this paper, I am going to make (and try to substantiate) a similar suggestion for the case of potentiality and possibility. Understanding objects’ potentialities in terms of possibility yields the wrong results. Like that of essence, the concept of potentiality is ‘sensitive to variations in source’: different things have different potentialities, just as different things have different essences. Like that of necessity, the concept of possibility cannot capture this sensitivity; or so I will argue in Sect. 2. Like Fine, I will then go on to suggest that we should understand possibility in terms of potentiality, rather than the other way around, a claim that I will spell out and defend against objections in the second half of this paper (Sects. 4 and 5). Before I can do so, however, I need to take a closer look at potentiality. For my proposed account of possibility is faced with two main difficulties. One is formal: potentialities are potentialities to … (fill in a predicate), while possibilities are possibilities that … (fill in a sentence). How can we get from a predicate operator ‘… has the potentiality to …’ to a sentence operator ‘It is possible that…’? The second difficulty is more material: it is the worry that there are not enough potentialities around to ground all the possibilities that we think there are. To meet (the various versions of) this objection, I will have to first provide an account of potentiality which is wide enough to provide for all the possibilities that we think there are and which does so without collapsing into possibility. This account will be sketched in Sect. 3. Similar accounts of possibility have been sketched by Borghini and Williams (2008) and Pruss (2002); Jacobs (2010) provides a closely related theory of counterfactual conditionals. The account here offered differs from these in various details, in particular in its emphasis on the formal difficulties and in the broad

From Potentiality to Possibility

281

conception of counterfactuals. (For an overview of these views, see Vetter 2011; for a further development of the version presented here, see Vetter 2015.)1 Throughout this paper, I will be assuming a very liberal conception of properties in general, and of the properties that are potentialities in particular. Many of the potentialities that I will be appealing to will not be among the fundamental elements of reality; they may very well be reducible to other, more fundamental properties. I believe that those fundamental properties, or at least very many of them, are themselves potentialities. But that is not a claim I will be defending (or assuming) in this paper.

2 Motivating the Account How does potentiality relate to possibility? It is tempting to assume that an object x has a potentiality to be F just in case it is possible that x is F; that is, just in case there is some possible world in (or according to) which x is F. A straightforward strategy in arguing against this approach would be to provide counter-examples: true statements of the form ‘It is possible that x is F, but x has no potentiality to be F’, for instance. (This is the strategy that Fine used, mutatis mutandis, against the reduction of essence to necessity.) And it seems certainly true that objects do not have dispositions or abilities to be F whenever it is possible that they be F. However, I have not yet said anything to delineate the extension of ‘potentiality’, and I will shortly argue that the potentialities of objects include much more than just their dispositions and abilities. So for the time being, the straightforward route seems precluded.2 Instead, I am going to focus on a pervasive feature of potentialities that can easily be illustrated by the familiar examples of dispositions and abilities. Potentialities come in degrees: some glasses are more fragile than others, and some of us possess the ability to play the piano to a much greater degree than others. Moreover, those degrees are contingent: a given glass might have taken a little crack and thus have been more fragile than it is; had I practised the piano more when I was a child, I would now possess the ability to play it to a much greater degree. If we were to analyse potentiality ascriptions in terms of possible worlds, we would have to say: an object x has a potentiality to F just in case x F’s in some possible world. That approach would have to be extended to accommodate degrees

1

A note on chronology (added in 2017): this paper was written in 2011 and minimally revised in 2013. My own views have developed further in the meantime and can be found, along with more details, in Vetter (2015). For critical discussion of the view that is proposed here, see Yates (2015) and Wang (2015). 2 The straightforward route is not precluded in principle, but it is narrower than it might appear at first glance. I will say nothing to advance that route in this paper. Doing so would require identifying the limit to the range of an object’s potentialities. I will here only be concerned with pushing that limit and extending the range of objects’ potentialities.

282

B. Vetter

of potentialities; and of course it can be so extended. In particular, possible worlds talk allows us to introduce degrees of possibility. The degree to which an object x has a potentiality to F will then simply be the degree of the possibility that x F’s. There are two ways to understand degrees of possibility on a possible-worlds framework: one is in terms of closeness to the actual world, the other is in terms of a probability-like measure or proportion of possible worlds. Let me look at these two options in turn. First, closeness. According to this conception, to what degree it is possible, at w, that p depends on how close to w the closest p-world is. Closeness of worlds is determined by contextually determined similarity relations between worlds. (Classic statements: Lewis 1973, 1986.) The degree of a possibility, on this account, is contingent like that of a potentiality: which worlds are close differs from world to world; and as time progresses, worlds that were initially close may diverge and others that were initially farther away may converge. So far, all is well. But while closeness does deliver contingency, it delivers the wrong contingency. The problem is known as ‘accidental closeness’ (Manley and Wasserman 2008). According to the present proposal, the degree of a glass’s fragility is just the degree of possibility that the glass breaks, which in turn depends on how close to the actual world the closest world is in which the glass does break (times being held constant). Now take two glasses that are equally fragile, store one of them safely on a shelf above a soft carpet, and place the other one at the edge of a table on a stone floor. On any reasonable closeness relation, there is a close world where the second glass breaks (one careless movement is enough) but no equally close world where the first glass breaks. But by hypothesis the two glasses were equally fragile. Hence the closeness account gave us the wrong degree for at least one of them (or more likely, for both: too low for the first, too high for the second). Second, proportion. On this account, to what degree it is possible (at w) that p is determined by the measure of worlds at which p, or the proportion of p-worlds to non-p-worlds. This is not subject to problems of accidental closeness: any possible circumstance, stone floors as well as soft carpets, will be among the worlds measured, no matter how close to actuality. However, we can see immediately from the characterization above that degrees of possibility on this account are not contingent: there is no relativization to w on the right-hand side (and no space for such relativization). If degrees of potentiality were identical to degrees of possibility on this conception, then worlds where this glass had taken an additional crack would count towards determining its degree of fragility, and worlds where I had violin lessons (which in fact I have not had) would count towards the degree of my ability to play the violin. If closeness yields the wrong contingency, and a measure/proportion account yields none at all, how about combining the two? Let us say that the degree of the possibility that p is determined by the measure of reasonably close worlds where p. Then closeness may take care of the contingency of degrees of potentialities, while the proportion rids us of the problem of accidental closeness. Among the close worlds, there should be those with stone floors and those with soft carpets, and in general enough variation of external circumstances to preclude accidental factors

From Potentiality to Possibility

283

such as the glass’s actual storage from mattering. Not everything must be varied, however. We want a measure of breaking-worlds out of all the worlds where the glass has taken all and only the cracks that it has actually taken, and of violin-playing-worlds out of all the worlds where I have had exactly the amount of violin lessons I have had (that is, none). In general, we want to measure the manifestation-worlds out of the worlds where the potentiality-possessing object is intrinsically just as it is in the actual world (including, if you like, its history). This is the closeness relation that should be at work: closeness is determined by similarity to the actual world with respect to the intrinsic constitution of the object whose potentialities are at issue. The degree of x’s potential to F is determined by the proportion of worlds, out of those where x is intrinsically just like it is in the actual world, where x F’s. (This proposal is close to what Manley and Wasserman (2008) suggest for the case of dispositions.) Yet this is still not fine-grained enough. For the requirement of intrinsic similarity is both too strong and too weak. It is too strong because arguably, there are extrinsic potentialities.3 To determine the degree of my vulnerability you have to consider exactly those worlds where not only am I intrinsically just as I actually am, but where certain features of my surroundings are too. Which features? Well: those which count towards my vulnerability. In general, the relevant worlds have to include those and only those where the potentiality-possessing object and the relevant features of its environment are just like they are in the actual world. Which are the relevant features of the environment? They are those, if any, that possession of the potentiality in question depends on. Moreover, and more importantly, the requirement is too weak. Suppose I have had piano lessons, but have made a firm decision never to play the piano. That decision too is part of my intrinsic make-up. So considering only worlds where I am intrinsically just as I am in actuality (in the supposed scenario) will yield very few if any worlds where I do play the piano, thus misrepresenting the degree of my ability. This is not a far-fetched kind of case: anyone who is irascible but well-bred, disposed to cry but disciplined, or prone to overeat but strictly following a diet will serve as an example. The suggested treatment would have us say that by adhering to a diet, one loses the tendency to overeat—a claim which goes strongly against the phenomenology of dieting. In general, that the potentialities I have mentioned are not lost but only masked in such cases can be seen from the effort that it takes to avoid their manifestation.4 To capture all and only the relevant worlds for an

3 McKitrick (2003) has made a very convincing case for the existence of extrinsic dispositions, and her arguments can be carried over to potentialities in general. Prior to her paper, it was commonly assumed that all dispositions were intrinsic. It may still be that all fundamental dispositions are intrinsic; but as I have noted in Sect. 1, my interest here goes far beyond the fundamental properties. 4 I am here appealing to so-called intrinsic masks or antidotes. On masks and antidotes in general, see Johnston (1992) and Bird (1998). Arguments for the existence of intrinsic masks in particular can be found in Everett (2009) and Ashwell (2010).

284

B. Vetter

object’s potential to F, we have to include worlds with some variations in the object’s intrinsic make-up. But only some variation: in fact, variation in anything but the potentiality itself. Again, the specification of the close worlds cannot help but refer to the very potentiality that is to be reduced—and that cannot constitute a reduction. We thus find ourselves, concerning potentiality and possibility, in a similar situation to that which Fine has pointed out for essence and necessity: the ‘local’ modality (potentiality and essence, respectively) appears not to be straightforwardly reducible to the corresponding ‘global’ one (possibility and necessity, respectively). Fine suggested that we can explain this finding, in the case of essence and necessity, by realizing that the traditional direction of explanation has things backwards: rather than understanding essence in terms of necessity, we should understand necessity in terms of essence. Each object, in the Finean picture, contributes its essence to the totality of necessary truths, in much the same way as each individual singer contributes their line to the performance of a choir. Given the totality, we should not expect to be able to trace back each individual’s contribution; the direction of explanation should go from the individual contributions to the totality, not vice versa. Given the results of my discussion so far, we may now try to mimic the Finean explanation. Perhaps the attempt to fully capture potentiality in terms of possibility failed because it got things backwards. If the potentialities of individual objects each contribute their share and degree to the totality of possibilities and their degrees, we should not expect to be able to trace back the individual contributions when given merely the totality, any more than we can expect to determine the lines of each particular singer by listening to a piece of choral music. The suggestion, then, is an account of possibility that is based on potentiality. Such an account will be proposed in Sect. 4 and defended in Sect. 5. But before I can do either, I need to take a closer look at potentiality.

3 Potentiality This section is an exploration of the variety of potentialities that things have. As such, it should be interesting in its own right; but it also pursues two specific aims that will be crucial for the account of possibility introduced in Sect. 4. The first aim concerns the formulation of my account of possibility. Such an account will normally take the form It is possible that p if and only if … where the blank is replaced by a statement relating p in some way or other to potentiality. The most straightforward way of doing this would be if we could ascribe to objects a potentiality for p to be the case; but this is not the way that potentiality works. As noted in Sect. 1, potentiality is always potentiality to be

From Potentiality to Possibility

285

some way or other, or to do something or other. So one thing that I will aim to do in this section is to make sense of a sentence-modifying potentiality operator, which takes complements of the form ‘that p’. A second aim concerns the defence rather than the motivation of the potentiality-based account of possibility. As we will see in Sect. 5, objections to the account will take the form of challenges to provide enough potentialities to explain all the possibilities that we think there are. The second aim, then, is to demonstrate that there are more potentialities than we might have thought there are.

3.1

Beyond Abilities and Dispositions

Dispositions are potentialities, and so are abilities. But do objects have potentialities that go beyond their dispositions and abilities? I believe they do; and I believe we can see that they do if we look carefully at the examples I have already supplied, and apply a principle of non-arbitrariness. Consider fragility: there are some very fragile things, such as old parchments and long-stemmed champagne glasses, and others that are less so, but still fragile, such as an ordinary tumbler. Among the non-fragile things, there are some that come closer to being fragile, such as a robust plant-pot, and others that are far remote from being fragile, such as my a desk or (even further) a rock. There seems to be a continuous spectrum that leads all the way from the champagne glass to the rock, and somewhere on that spectrum we can find the cut-off point—or perhaps we cannot, since ‘fragile’ is a vague predicate. Even if we could precisely locate it, however, that cut-off point would not be likely to cut nature at its joints; it might have been a little more to one side of the spectrum or the other, thus excluding the tumbler or perhaps including the plant-pot.5 I have said there is a continuous spectrum; but a spectrum of what? The answer seems to be: of possibility-like properties that come in varying degrees of strength; in other words, of potentialities. Next consider abilities (of the agentive kind—or, if you prefer a different term, capacities). It is a difficult question exactly what kind of potentialities qualify as abilities; answering it would, I think, require doing a substantial amount of action theory. But we do not need to answer that question for the present purposes; some intuitive grasp of what qualifies as an ability will do. I am hopeless at sports; I cannot even catch a ball if you throw it to me. That is to say, I have no ability to catch a ball if you throw it at me; having an ability requires some particular kind of control, and I do not have that control. Still, I may be lucky in some cases, and catch the ball thrown at me. Nothing about myself prevents me in principle from

To be more precise, we would have to say: the cut-off point for things to be truly called ‘fragile’ varies from one context to another. This merely reinforces the idea that in any given context, we are not cutting nature at its joint in distinguishing the fragile objects from the non-fragile ones.

5

286

B. Vetter

catching a ball thrown at me, in the way that lacking arms would prevent me from catching a ball, or in the way that a table is in principle prevented from catching a ball. This is a fact about me, a property of mine, just like its potentiality to break is a property of the plant-pot. It seems reasonable to say, then, that I have a potentiality to catch a ball when it is thrown at me, a potentiality which my desk lacks; but that my potentiality to catch a ball, unlike (perhaps) yours, is not an ability. It would seem, then, that we should allow for there to be potentialities that are neither dispositions nor abilities. We can go further. Note, first, that a potentiality’s manifestation may be a logically complex property. A fragile object, by being disposed to break, must be disposed to break into two pieces or break into more than two pieces—although of course it need not be disposed to break into two pieces, or disposed to break into more than two pieces. The ability to do a headstand has as its manifestation a complex activity that involves holding one’s feet up while keeping the balance on one’s head and arms. Anyone who has the ability to do a headstand must therefore have an ability to possess this complex property: to hold up one’s feet while keeping the balance on one’s head and arms. Manifestations of potentialities must, thus, include such complex properties as the disjunctive and conjunctive properties expressed by the italicized predicates. So far, so (relatively) uncontroversial. But now comes a crucial point, one which takes us one step in the direction of producing, from the predicate-modifying ‘potentiality’ operator, a sentence-modifying ‘possibility’ operator.

3.2

Being Such That

I have an ability to walk in the rain, but not an ability to walk in a tornado. That is, I have an ability to walk while it is raining (but not to walk while there is a tornado); or, in vocabulary that is more amenable to formal treatment, I have an ability to walk and be such that it is raining. The manifestation of this ability is conjunctive again; but more importantly, it includes a conjunct that is not, in some natural sense, a property of the individual who possesses the ability. Being such that it is raining is a property of me only in some parasitic and very ‘thin’ sense; if it is a property that I could possess, it is one that I would possess by no doing of my own, purely in virtue of matters that are entirely external to me. Nonetheless, it seems that this property (or whatever else it is) can form part of the manifestation of my abilities; or else it is hard to see how we can truly ascribe to me an ability to walk in the rain. Note one intuitive difference between the ‘conjunctive’ ability that is entailed by the ability to do a headstand, and the ‘conjunctive’ ability to walk-and-besuch-that-it-rains: in the former case, the ability appears to ‘distribute’ over conjunction, while in the second it does not. That is to say, by having the ability to hold up my feet while keeping my balance, I have the ability to hold up my feet and the ability to keep my balance. It is certainly not obvious that, analogously, by having the ability to walk-and-be-such-that-it-rains I have an ability to walk and an ability

From Potentiality to Possibility

287

to be such that it rains. (Or perhaps it is obvious that this is not the case.)6 What about potentiality in general (as opposed to the specific kind of potentiality that is ability)? Do I, by having a potentiality to walk-and-be-such-that-it-rains, a potentiality to walk and a potentiality to be such that it rains? This is a difficult question, the answer to which turns on the general question of how far we want to extend the notion of potentiality: are we willing, in principle, to ascribe to an individual such as myself a potentiality to be such that something else is one way or another? Fortunately I will not have to answer this issue in full for the purposes of this paper. But it is worth taking another look at the kinds of potentialities that this concerns: potentialities to be such that p. The ‘such that’ locution is a powerful device in logic (where it is written as the Greek letter k): it allows us to turn any sentence, open or closed, simple or complex, into a predicate. Thus we can semi-formally write the predicate ‘is walking while it is raining’ as kx.(x walks ^ it is raining)’, etc. Once we have allowed, as I argued that we should, potentiality ascriptions that include a ‘such that’ operator in the specification of the manifestation, there is a plethora of new potentiality ascriptions that appear to be true. By having an ability to walk, it would seem that I also have an ability to be such that I am walking. (The latter ascription may be thought to be merely an elaborate paraphrase of the former, but it is not, or at least not obviously: for my ability to walk is one that I share with you, while my ability to be such that I am walking is, presumably, not.) Now suppose that I exercise that ability. Then it is true that I am walking; it is also true that something is walking, and that I am such that something is walking. Should we therefore ascribe to me an ability (or at any rate, some potentiality) to be such that something is walking? I see no reason why we should not: I make the same contribution to there being (and hence to my being such that there is) something that walks as I do to my walking. Nor can there be a principled objection to ‘such that’ predicates in the scope of ‘potentiality’, given my ability to walk and be such that it is raining. Note that this line of reasoning does not justify ascribing the potentiality to be such that something walks to a stone or a star, or the potentiality to be such that it is raining to me; for all that has been said, it is only because I have a potentiality to walk that I have the potentiality to be such that something walks. Without answering the question whether I might have a potentiality to be such that it is raining, and hence without definitively fixing the extension of ‘potentiality’ (a task that I leave for another time)7, we have seen some legitimate uses of the ‘such that’ locution in ascribing potentialities. As I have said earlier, a potentiality is always a potentiality to be or become some way or other, or to do something or other, while possibilities are possibilities

6

See Schnieder (2008) for discussion. As the question of distribution over conjunction has already indicated, this is a question of great importance for the logic of potentiality; but providing such a logic goes far beyond the scope of this paper. See Vetter (2015) for more details.

7

288

B. Vetter

that something be the case. If being some way or other includes being such that something is the case, then we have found one way of linking potentialities to propositions, as is required for a potentiality-based theory of possibility. Have we, then, found what we were looking for: the link between potentiality and a proposition p that will help us formulate a potentiality-based account of the possibility that p? Not quite. Take the possibility that my great-granddaughter will be a painter. Who or what might have the corresponding potentiality? It will come as no surprise that I do not have a great-granddaughter, and I might never have one. Nonetheless, it is possible that my great-granddaughter will be a painter, and I seem to be our best candidate for possessing the relevant potentiality. Perhaps I do have a potentiality to be such that my great-granddaughter is a painter. But note that this potentiality’s manifestation is subtly different from the content of the envisaged possibility: my being such that my great-granddaughter is a painter requires not only that my great-granddaughter be a painter, but also that I exist. (I could not be such that anything is the case without being in the first place.) The possibility we were after makes no such requirement: its realization is perfectly compatible with my being long dead when my great-granddaughter becomes a painter. We need to continue, then, and look at another important phenomenon: iterated potentiality.

3.3

Iterated Potentialities

Iterated potentialities are simply potentialities to have potentialities—potentialities whose manifestation property is itself a potentiality. We have intuitive examples of such potentialities to have potentialities from the paradigm examples of dispositions and abilities. Here are some. I do not have an ability to play the violin. Nor does my desk. Unlike my desk, however, I possess the ability to learn to play the violin—the ability, that is, to acquire the ability to play the violin. That latter ability distinguishes me from my desk, insofar as playing the violin is concerned. Let me call it an iterated ability to play the violin. We can go further. A violin teacher possesses the ability to teach students how to play the violin. The manifestation of this ability consists in another individual, a student, acquiring an ability: the ability to play the violin. There are further intuitive examples that do not involve abilities. Whether or not colours are secondary qualities, i.e., dispositions to appear in a certain way to a normal observer, coloured objects certainly have these dispositions. Now take an apple about to ripen and turn a bright shade of red. The apple has a potentiality to become red; it thereby has a potentiality to acquire the disposition to look red to a normal observer. Or take a colouring agent, used to turn white textiles red. When its potentiality is manifested, a piece of textile becomes red and thereby acquires the disposition to look red to normal observers.

From Potentiality to Possibility

289

The violin teacher and the textile dye both seem to have an iterated potentiality (a potentiality to produce a potentiality to …) the manifestation of which involves something other than the possessing objects, the teacher and the dyeing agent, themselves. The individuating manifestation of a non-iterated potentiality is a property; if an object manifests or exercises a potentiality that it possesses, it comes to possess that property. The same does not hold of iterated potentialities: their manifestation is not a property of the object possessing them; rather, it has propositional structure. The reason is straightforward: the manifestation of the iterated potentiality may concern something other than the iterated potentiality’s possessor itself. The above-mentioned dyeing agent does not have an iterated potentiality to appear red to normal observers; it has an iterated potentiality for a textile to appear red to normal observers. The iterated potentiality’s manifestation is not, then, appearing red to normal observers; it is that a textile appear red to normal observers. It is not even the property of being such that the textile appear red to normal observers, for that property would still have to be possessed by the dyeing agent when the iterated potentiality is manifested. But the dyeing agent itself may have ceased to exist by the time a dyed textile appears red to anyone; in which case it cannot have the property of being such that the textile appears red to normal observers, yet its iterated potentiality for the textile to appear red is still being manifested. With iterated potentiality, then, we have reached a phenomenon that is best expressed by a propositional operator: as iterated potentialities that something be the case, rather than as iterated potentialities to be some way or other. As the case of the dyeing agent illustrated, the iterated potentiality’s possessor may not be at all when the iterated potentiality is manifested. To understand this case in a little more detail, note first that its manifestation (if it does manifest) happens in several stages. First, the dye would manifest its potentiality to enter into a particular relation—dyeing—with a piece of textile. (This manifestation requires the presence of some particular piece of textile, but the potentiality itself is general: it is a potentiality to dye a piece of textile.) We may think of the dyeing relation as involving a potentiality that is possessed jointly by the dye and the piece of textile. The manifestation of this jointly possessed potentiality consists in the textile’s becoming red, and thereby acquiring a new potentiality: the potentiality to look red to a normal observer. At the final stage of the iterated potentiality’s manifestation, this last potentiality is manifested, and the textile is seen as red by some observer. This final stage no longer requires the continued existence of the dyeing agent; it has done its work in the first few stages. It is the dyeing agent that possesses the iterated potentiality because it is the dyeing agent’s potentiality that sets the chain of manifestations into motion if the iterated potentiality is manifested; but it need not be involved in the subsequent stages. Similarly, I have an iterated potentiality that my great-granddaughter be a painter. That is to say, I have a potentiality to have a child that has the potentiality to have a child that has the potentiality to have a daughter who has the potentiality to be a painter. My existence is required at the first stage of this iterated potentiality’s manifestation (my having a child), but not at the subsequent stages. (I could

290

B. Vetter

even have an iterated potentiality for my great-granddaughter to become a painter after my death.) Hence the possibility that my great-granddaughter be a painter may be grounded in my potentialities without requiring that I exist when it is realized. Thus we have now reached an understanding of potentialities that something be the case which appears wide enough to be used in an account of possibility—an account which I will delineate in the next section.8 In order to have a general way of referring to iterated and non-iterated potentialities alike, I will understand potentiality (simpliciter) as once-iterated potentiality. To fit once-iterated potentiality into the propositional form of iterated potentiality, we can then adopt the following convention: where an object x has a (once-iterated) potentiality to be F, I will say that x has a once-iterated potentiality for x (itself) to be F, or for it to be the case that x is F. Given this convention, we can subsume all potentiality statements under the form ‘x has an iterated potentiality (for it to be the case) that p’, with the understanding that in the case of once-iterated potentiality, or potentiality simpliciter, p must be of the form ‘x is F’.

4 Possibility We have now assembled the materials for a potentiality-based account of possibility. The intuitive idea of the account is this. When we think about possibility, we think about potentiality in abstraction from its bearer; a possibility is a potentiality somewhere or other in the world, no matter where; possessed by something (or some things) or other, no matter which. Of course, the potentialities may be iterated: in possibility claims, we abstract not only from a potentiality’s bearer, but also from the number of steps that it takes the potentiality to reach its manifestation. Progressing from this intuitive idea to a more precise formulation, and using ‘iterated potentiality’ to include once-iterated potentialities (i.e., potentiality simpliciter), I propose the following: Possibility It is possible that p if and only if something has (or some things have) an iterated potentiality for it to be the case that p.

It is necessary that p if and only if it is not possible that not p; i.e., if and only if nothing has (or no things have) an iterated potentiality for it to be the case that not-p.

8

This is to say that the possibility of my great-granddaughter’s being a painter is, in a sense, a fact about me. And indeed this is the basic idea of the account: all possibilities are ultimately rooted in, and thus are in some sense facts about, the objects that there are. What kind of fact about me might be responsible for the possibility in question? As a start, the answer is: my biological ability to give birth to children with intact reproductive capacities of their own. The potential (further down the line of iterations) to give birth to a daughter with the potential to be a painter requires hardly more than the potential to give birth to a daughter with healthy hands and brains, and so on—a potential that most women have.

From Potentiality to Possibility

291

Why should we adopt this account? And how does it compare to its rivals? The proposed account is a realist, actualist and (a special type of) modalist account; let me take up these three in turn. The potentiality-based account is realist: it recognizes modality, that is, possibility and necessity, as real and mind-independent features of reality. Why? Because possibility is grounded in potentiality, and potentialities are real and mind-independent features of reality; they are real, mind-independent properties of real, mind-independent objects. It is open to the potentiality theorist with anti-realist inclinations to make potentiality, and thereby possibility, mind-dependent in some way or other. But that is not the route I have taken so far in examining potentiality, and it would take away much of the account’s attraction.9 Second, the potentiality-based account is actualist: it does not recognize any genuine worlds apart from the actual world. What is more, it does not follow philosophical orthodoxy, which has it that possible worlds, genuine or otherwise, have an explanatory role to play, one way or another, in the metaphysics of modality. (For some, most notably David Lewis, its being possible that p straightforwardly consists in there being a world in which p (or some counterpart-theoretic translation of p) is true. Other theorists pursue a more modest project: Stalnaker (2003) and Plantinga (1974), for instance, take modality to be a primitive that has to be built into any account of possible worlds, but are nonetheless convinced that possible-worlds talk provides the best elucidation of this primitive. Of course, like any other actualist account, mine can admit that there is some interest in talk about possible worlds, only we should not take the label ‘worlds’ too seriously. In fact, we should not take the label ‘possible’ too seriously either. The so-called possible worlds of modal logic are a useful device in producing a formal model. But not only are they not worlds in any but a figurative sense. They also have no more intimate connection to possibility than they do to any of the other phenomena that they can be used to model: knowledge, obligation, vagueness, and a wide variety of others. Why deviate from orthodoxy in this way? Some philosophers have already argued against orthodoxy; roughly, the reasoning goes as follows. If we give an account of modality in terms of possible worlds, those worlds are either concrete, Lewisian worlds (locus classicus: Lewis 1986) or some kind of abstract entity, such as sets of propositions (loci classici: Adams 1974; Plantinga 1974). As to the former, the ‘incredulous stare’ (Lewis 1986) is a strong objection. And even if it were true that there are infinitely many concrete universes, this would seem to be no more than a curious fact about the structure of the one, actual world, unrelated to the nature of modality. With abstract entities, on the other hand, it is hard to give a non-circular account of modality, since possibility comes into the very process of

9

For the anti-realist about modality, potentiality does not seem to be a good starting point. If modality is a matter of what we can conceive [see Bealer (2002) and Peacocke (1999) or, for a radically anti-realist approach, Blackburn (1987)], it had better start with a propositional operator, such as ‘It is possible that…’, since what we conceive seems to have propositional structure.

292

B. Vetter

singling out the abstract entities that are to play the role of possible worlds.10 Some have chosen a non-reductionist approach, treating modality as primitive and possible worlds as only a useful descriptive and heuristic tool (see, for instance, Stalnaker 1976, 2003). But for those who want a more substantial account of metaphysical modality, that approach simply fails to answer the relevant questions. Third, like many actualists (including Plantinga and Stalnaker), the potentiality theorist is a modalist: she thinks that modality is a primitive feature of our world, not to be reduced to something non-modal. However, the potentiality theorist does envisage an intra-modal reduction of the ‘global’ modalities (possibility and necessity) to the ‘local’ modality of potentiality. Most modalists would take the opposite route, but we have seen in Sect. 2 that that route is, at the very least, faced with serious problems. The proposed account of possibility helps us understand the root of those problems. If possibility is just potentiality in abstraction from its possessor(s), then given a possibility there is not always a way of telling which potentiality-possessing object has been ‘abstracted away’. So much for situating the proposed account in the metaphysical landscape. I have already provided one reason for finding this account attractive. Potentiality is part of the modal package, the package of concepts that we are already equipped with when we come to the meta-physics of modality. Approaching modality from the side of possibility and necessity, or of possible worlds, we do not quite get a hold on this part of the modal package (or so I argued in Sect. 2). If, starting with potentiality, it turns out that we do get a better hold on possibility and necessity, this will be one advantage of the potentiality-based account over both a modalist account of the usual variety, and a possible-worlds based account. The potentiality-based account has a further advantage: it provides a realist metaphysics of modality that does not make the epistemology of modality mysterious. Unlike those accounts of possibility and necessity that appeal to possible worlds, a potentiality-based account is firmly rooted in everyday life and thought. Potentialities are ubiquitous in our ordinary thought about, and dealings with, the world: we handle glasses carefully if they are fragile, we handle people carefully if they are irascible or vulnerable, we learn languages in order to acquire the ability to speak them, and practise the piano to improve our skills at it. A central advantage of my account is that it treats modality not as a philosophers’ creature but as something on which we plausibly have a good pre-theoretical grasp.11 To prove its rank as a serious contender, however, my proposed account of possibility has to meet two requirements. First, it has to be formally adequate: possibility defined in this way must conform to the formal structure of possibility as 10

Cf. Williamson (1998) and Jubien (2007) for contemporary versions of this kind of criticism. Of course, there are various lines of response available; see, for instance, Lewis (1986, Chaps. 2.1 and 2.4). For present purposes, the argument I have sketched is to serve merely as a preliminary motivation. My goal is to describe an alternative to orthodoxy, not to refute orthodoxy. 11 For more on the desirability of such an account, see Williamson (2007), whose take on the epistemology of modality is of course a little different. I provide a variant on Williamson’s view that is closer to the metaphysics developed here in Vetter (2016).

From Potentiality to Possibility

293

studied in modal logic. Second, it has to be materially or extensionally adequate: it has to make the right possibility claims come out true (or false, as the case may be). The first requirement would need a more formal (and much longer) argument than is appropriate in this context; suffice it to say that such an argument is available.12 For the rest of this paper, I will be concerned with the second requirement. I will argue that a potentiality-based account can meet the requirement of extensional adequacy, and that it has various options in meeting that requirement. That is to say, I am not going to put forward a particular version of the potentiality-based account. Rather, I am going to point out that there are various solutions available, all of which preserve the spirit of the proposal made above. Extensional adequacy seems relatively easy to come by in the cases of simple predicative possibilities: it is possible that the apple is red because the apple itself has a potentiality to be red; it is possible that something is red because the apple has a potentiality to be red, and hence a potentiality to be such that something is red. It is possible that my granddaughter will be a painter, even though there is no individual yet who is my granddaughter, because I have a three-times iterated potentiality for it to be the case that my granddaughter is a painter: a potentiality to have a child who has the potentiality to have a daughter who has the potentiality to be a painter. But not all cases are as simple as these. In the following, I will discuss two paradigmatic cases of problems for the potentiality-based account: the possibility of alien objects and properties, and the possibility of different laws of nature. I will examine the different general strategies available to the potentiality theorist, and offer various solutions to these two problems. My discussion will be primarily exploratory: trying to understand which options are available to a potentiality-based theory of possibility. The positive conclusion will not be an endorsement of any particular such option. Rather, it will be that there are a great many options, several of which are promising. A potentiality-based account of modality is a fruitful research programme, though it is one that cannot be fully carried out here.

5 Objections and Strategies 5.1

Objections

There is one obvious form of objection to the view I have proposed: counterexamples of the form ‘It is possible that p, but nothing has a potentiality (no matter how many times iterated) for it to be the case that p’. These counterexamples are best understood as a challenge: a challenge to point out where in the world something has an iterated potentiality for it to be the case that p.

12

I have spelled it out in detail in Vetter (2010) and Vetter (2015).

294

B. Vetter

Here are two such examples. Aliens. David Lewis has argued (see Lewis 1986, 159 ff.), and has convinced many, that there could have been properties that are completely alien to our world. For David Lewis, a property is alien to a world if it is never instantiated in that world. Of course my proposal allows for the possibility of alien properties in this sense: some potentialities may never be manifested, and hence their manifestation property never instantiated; so some properties that are never instantiated are nonetheless possibly instantiated. However, there is no guarantee that the potentialities of actually existing objects ‘reach’ all the alien properties there might have been; there may be uninstantiated potentialities for some alien properties but not others. Or so it may be argued.13 Similarly, it is contingent which objects there are; there might have been different objects from the ones there actually are. In fact, it might have been that none of the actually existing objects existed, and that the world consisted of a completely different set of objects. Now, the actually existing objects may have potentialities for there to be a whole range of other objects: people have unmanifested potentialities to have a child of such-and-such a description, to mak/craft/ produce a watch of such-and-such a description, etc.; things jointly have potentialities to constitute (to jointly be all the parts of) infinitely many objects that they never actually constitute; and so on. But, once again, that may not seem sufficiently far-reaching: could there not have been objects that were not produced by any actually existing objects (or by anything that actually existing objects have a potentiality to constitute or produce, or …), nor constituted by any of the actually existing objects (or anything that actually existing objects have a potentiality to produce, or …)? Moreover, which actually existing object(s) should possess a potentiality, even iterated, for it to be the case that none of the actually existing objects ever existed?14 Contingent laws. The world is governed by certain laws, the most fundamental of which are probably the laws of (real) physics. But it might have been governed by a different set of laws; the laws of nature are contingent. Yet how could any of the things that are actually governed by the actual laws have an (iterated) potentiality for those laws to be different? Those are typical objections to the potentiality-based account of possibility. They take the form: surely it is possible that p; potentialities, perhaps, can take us some way towards p, but there is no guarantee that they will take us all the way; the intuition about p is too fundamental for the concept of possibility to be left to (epistemic) chance concerning which potentialities there are; so the potentiality view should be rejected. 13

Lewis uses alien properties to argue against ersatzist views on possible worlds. More recently, Divers and Melia (2002) have argued that Lewisian modal realism itself has difficulties in dealing with certain intuitions concerning alien properties without appealing to modal concepts. So whatever the potentiality view’s standing is with regard to alien properties, it will not face problems alone. 14 See also Cameron (2008) and Contessa (2009) on this problem.

From Potentiality to Possibility

295

In responding to such objections, there are two principal strategies. One is to acknowledge the alleged possibilities and point to the relevant potentialities, thus showing that the potentiality-based account is more resourceful than its opponent might have thought. A second strategy is to bite the bullet and reject the alleged possibility. There are precedents for this kind of strategy: before Kripke, did people not think it was possible that water is not H2O? There are precedents even for some particular such claims: the view that the laws of nature are metaphysically necessary, for instance, has been increasingly popular in the last decade or so. In biting that bullet, the potentiality view will at least not have to stand alone. However, in denying what appeared (at least to the opponent) to be a reliable intuition regarding the possibility of p, this strategy incurs an obligation: to explain how we (or, at any rate, the opponent) came by that intuition. Sydney Shoemaker, in defending one restriction to what is metaphysically possible (viz., only what is nomologically possible), says that ‘the main theoretical obstacle to [recognizing this restriction] has been the deeply rooted conviction that necessary truths should be knowable a priori’, a conviction that metaphysicians after Kripke should long have abandoned (Shoemaker 1998, 74). Of course, we cannot know a priori what the laws of nature are, and we cannot exclude variant laws a priori—but then, we could not know a priori that water is H2O, and hence could not exclude a priori that it was XYZ. In general, the problematic possibilities (problematic, that is, for the potentiality view, which does not accord to them the status of possibilities) are not metaphysical possibilities but only, in some way, epistemic ones. I will illustrate both kinds of strategies in what follows. With alien objects and properties, I will sketch a response that produces more potentialities than the opponent might have thought there are; with contingent laws, I will outline a strategy that bites the bullet. There are different strategies available in both cases, but the following should serve as an illustration.

5.2

Answering Objection 1: Aliens

The world contains a limited number of objects, instantiating a limited number of properties. Which objects there are, and which properties they have, seems to be mostly a contingent matter. There could have been an entirely different set of objects; and there could have been an entirely different set of properties instantiated. (Or rather, almost entirely different: there may be necessary existents, e.g. the number two, and necessarily instantiated properties, such as the property of being self-identical.) Or so the intuition goes. What can the potentiality account say to accommodate (or defuse) these intuitions? Let me begin with the idea that there could have been different objects. This has two sides: there could have been some objects that don’t actually exist, and some of the objects which do actually exist might not have existed. Taking both sides together and to the extreme, it might have been that none of the actual existents

296

B. Vetter

existed (leaving aside necessary existents, if such there are) and that an entirely different set of objects existed instead. It is clearly true, on the proposed account, that there might have been objects other than the ones that there actually are: my parents had the potentiality to have another child, the machines in a car factory have the potentiality to make another car (in addition to the ones they do make), an apple seed has a potentiality to become an apple tree (arguably a distinct object) which bears apples of such-and-such description that never exist if the seed is thrown away with the apple. So we have the possibility of alien objects by the actual objects’ potentiality to produce them. We have, furthermore, the possibility of a great many more alien objects because objects in the world jointly have a potentiality to constitute them: the handle of one knife and the blade of another have a joint potentiality to make up one knife even if they are never put together; the molecules in my body have joint potentialities to make up a great many things of various descriptions that are very different from me, and most of which will never come to exist; the atoms in the world have joint potentialities to make up different molecules, which would then have potentialities to constitute different objects; etc. Going all the way down to the fundamental particles, whatever those are (and if there are any—if Schaffer (Schaffer 2003) is right, we can just keep going down), it will be possible that there are such-and-such objects just in case those particles (or, if Schaffer is right, something somewhere down the infinite hierarchy of levels), or some of them, jointly have a potentiality to constitute an object of such-and-such a nature. That certainly yields a wide range of possible alien objects. It is nonetheless a limited range. It may be said that, intuitively, there could have been objects other than the ones that could have been made up of our particles; there could have been different kinds of particles, or things not made up of any particles at all—immaterial souls, and their like. In general, and most radically, it seems that the basic furniture of the world could have been of a completely different nature from ours; the world could have been made up by objects of a completely alien kind. This problem is usually referred to, following David Lewis (Lewis 1986), as the problem of ‘alien properties’. Alien properties are problematic for most actualist theories of modality. In the case at hand, the problem is simply that it is hard to see how any actually existing object could have a potentiality for there to be objects of such alien kinds, and with such alien properties. We might, of course, bite the bullet. But it is more interesting in this case to see how the possibility at issue can be accommodated. There are at least two ways in which we might do so. First: suppose that there are universals, and suppose that some of them exist despite never being instantiated. Suppose further that universals, and perhaps abstract objects in general, have potentialities. Then we can ascribe to those universals the potentiality to be instantiated (if universals possess any potentialities, then it seems reasonable they all possess this one), thus yielding the possibility for there to be the alien objects that instantiate the universal. Similarly, we can ascribe to those potentialities that are actually instantiated a potentiality to be

From Potentiality to Possibility

297

uninstantiated, thus yielding the possibility that there existed no such objects as those that actually exist. This option is appealing if (and presumably only if) you hold a Platonist theory of universals, that is, if you think that universals exist independently of the objects that instantiate them. Ascribing potentialities to such Platonic universals may appear to be a bold move, because we tend to associate potentiality (in some way) with causation, and universals are generally thought to be causally inert. Given the rather liberal approach to potentiality that I have argued for earlier, I do not think that the association of potentiality with causation is a necessary one. Still, I would like to offer a second solution for the non-Platonist. The second option strongly relies on the notion of iterated potentiality. We can ascribe to the objects that actually exist the potentiality to exist while something instantiates an alien property; in which case we would have extended the realm of possible alien objects to those that some of the actual objects could co-exist with. Suppose, for instance, that there were two kinds of fundamental particles, the Q1s and the Q2s. The Q1s have certain potentialities to interact with Q2s. More precisely: Take any two particulars, a and b, where a is Q1 and b is Q2. a has the potentiality to interact with b in certain ways. a also has the potentiality to interact with any Q2 in a certain way; this is an intrinsic potentiality of a which does not depend on the existence of any particular Q2. In fact, it seems to be a potentiality which a might possess (though not exercise) in the absence of anything that is actually Q2. And if this is so, then it seems only reasonable that a should also have potentialities to interact with things of a certain uninstantiated kind, Q3, even though there really are no Q3s. If a has a potentiality to interact with Q3s, then a must a fortiori have the potentiality to be such that there are Q3s for it to interact with. In fact, a’s potentialities extend even further: a may have a potentiality to interact with, and hence to be such that there are, entities of a kind Q3, where Q3s are in turn such as to have a potentiality to interact with, and hence to be such that there are, entities of a kind Q4, which may in turn … and so forth. a, which is of the kind Q1, co-existing only with entities of kind Q2, may have iterated potentialities for there to be entities of any kind Qn that can be reached via a chain of potential interactions. The second option also provides us with the material to answer the second aspect of the question discussed here: could it not have been that none of the actual objects existed, and an entirely distinct set of things had existed instead? The manifestation of an object’s (more than once) iterated potentiality, unlike the manifestation of an object’s (only once-iterated) potentiality, may involve the very same object’s nonexistence. Thus our particle of type Q1, a, may have a potentiality to be such that there are objects of (the actually uninstantiated) kind Q3, which may in turn have a potentiality to be such that there are objects of (the actually uninstantiated) kind Q4 while there are no objects of kind Q1. Hence the Q1s themselves, or rather, any one of them such as a, can have a more-than-once iterated potentiality for there not to be any Q1s. (The above argument works for any property. But if being Q1 is an essential kind property, then the potentiality mentioned amounts to a potentiality for there not to be any of the actual Q1s.)

298

B. Vetter

Whether any of this is actually the case is, for the philosophers’ part, mere speculation. It is a feature of the potentiality-based account that truths about what is possible and what is not are hostage to the way things actually are, about which we can find out (at best) by empirical inquiry. To a certain degree, that situation is accepted by every (orthodox) metaphysician post-Kripke. The potentiality-based view merely intensifies that dependence of the possible on the actual—but it does so by making the actual deeply modal.

5.3

Answering Objection 2: Contingent Laws

The arguments of the preceding section will help in addressing a further worry: does the potentiality-based account collapse metaphysical possibility into physical (or nomological, or causal) possibility? The worry goes as follows. Surely nothing has a potentiality to do what is contrary to the laws of nature. I, for one, have no potentiality to travel faster than light; the laws of nature constrain my potentialities. But at the same time we want to say that the laws of nature are contingent: there might have been different laws from the ones that actually hold. So it looks as though there are possibilities, namely the possibility of the laws being different, for which no object has a corresponding potentiality. In this case, the potentiality theorist may bite the bullet and say: yes, it may have seemed that the laws of nature were contingent; well, we were wrong and it turns out they are necessary. Here she will not have to stand alone. The so-called dispositional essentialists, including Bird (2007), Ellis (2001), Shoemaker (1980) and Swoyer (1982), have argued that the laws of nature are necessary in the strictest, metaphysical sense. Witness Shoemaker (1998, 74): As is suggested by the way ‘necessity’ is used in ordinary speech, and is defined in dictionaries, causal [i.e., nomological] necessity is, pre-theoretically, the very paradigm of necessity. The main theoretical obstacle to according it that status has been the deeply rooted conviction that necessary truths should be knowable a priori. Once that obstacle has been removed, anyone who holds that causal necessity is not really necessity, or has only a second-class kind of necessity, owes us a reason for thinking this.

From the dispositional essentialists, we can also borrow a strategy that may help alleviate the anti-necessitarian worries. According to dispositional essentialism, the laws are not externally imposed on properties to govern the behaviour of the things which possess those properties. Rather, the laws are grounded in the properties themselves. The fundamental properties, on this view, are essentially dispositional: negative charge, for instance, is the disposition to repel other negative charges and attract positive ones. So it is not that negatively charged objects repel each other because it is a law that they do. Rather, it is a law that negatively charged objects repel each other because that is what being negatively charged consists in. It’s the properties that make the laws.

From Potentiality to Possibility

299

Dispositional essentialism is clearly congenial to the view of possibility that I am defending: just as I say that possibility is grounded in potentiality, so the dispositional essentialists say that laws of nature are grounded in (a certain type of) potentialities. Now, for the dispositional essentialist, it will indeed be necessary that those properties which figure in the laws figure in them in exactly the way they do: if negative charge were not such as to repel other negative charges, it would no longer be negative charge. But it will not be necessary that the properties which are actually instantiated are the only properties instantiated. I have earlier discussed the possibility of alien properties. Alien properties, if they have dispositional essences, would give rise to alien laws. Or more precisely, properties that are not actually ever instantiated, if they have dispositional essences, give rise to laws that do not actually govern anything that actually happens. Whether those ‘alien’ laws hold in some vacuous fashion or not at all is not the material point. The point is that, even if the laws are necessary in that nothing could ever go against them, what actually happens in the world might have been (non-vacuously) governed by a partly or entirely different set of laws involving some or only different properties. Bird (2007, 48f.) discusses this view under the label ‘weak necessitarianism’. It is a view that nominally counts as necessitarianism: the laws of nature are necessary in that nothing could go against them. However, they do not necessarily hold non-vacuously: they might have had nothing to govern. Weak necessitarianism is a view that should be palatable to many critics of necessitarianism: it allows for the possibility of the world being governed by very different sets of laws. Those different laws could not, of course, have contradicted the ones that actually govern the way things happen; they could not require different patterns of interaction between the properties that are actually instantiated (though they might require different patterns of interaction between those properties and other properties that are not actually instantiated). But they could have involved properties that are just like ours, except for being involved in laws of nature that differ from those in which our (actually instantiated) properties figure. Weak necessitarianism thus accounts for the anti-necessitarian intuitions: the world could have been as the anti-necessitarian claims it could in all the phenomenal detail that you may imagine; the only caution we need to take is not to identify the properties that would then have been instantiated with those that actually are.

6 Conclusion I have argued that a potentiality-based account of possibility is both desirable (Sects. 2 and 4) and, prima facie at least, feasible (Sect. 5). En route we have gained a much more detailed and flexible understanding of potentiality (Sect. 3).

300

B. Vetter

The potentiality-based account of possibility has only been sketched, but I hope to have shown that it is promising.15

References Adams, R. M. (1974). Theories of actuality. Noûs, 8, 211–231. Ashwell, L. (2010). Superficial dispositionalism. Australasian Journal of Philosophy, 88, 1–19. Bealer, G. (2002). Modal epistemology and rationalism. In T. Gendler & J. Hawthorne (Eds.), Conceivability and possibility (pp. 71–125). Oxford: Oxford University Press. Bird, A. (1998). Dispositions and antidotes. The Philosophical Quarterly, 48, 227–234. Bird, A. (2007). Nature’s metaphysics. Oxford: Oxford University Press. Blackburn, S. (1987). Morals and modals. In C. Wright & G. Macdonald (Eds.), Fact, science and value. Oxford: Blackwell. Borghini, A., & Williams, N. (2008). A dispositional theory of possibility. Dialectica, 62, 21–41. Cameron, R. (2008). Truthmakers and modality. Synthese, 164, 261–280. Contessa, G. (2009). Modal truthmakers and two varieties of actualism. Synthese, 174, 341–353. Divers, J., & Melia, J. (2002). The analytic limit of genuine modal realism. Mind, 111, 15–36. Ellis, B. (2001). Scientific essentialism. Cambridge: Cambridge University Press. Everett, A. (2009). Intrinsic finks, masks, and mimics. Erkenntnis, 71, 191–203. Fine, K. (1994). Essence and modality. Philosophical Perspectives, 8, 1–16. Jacobs, J. (2010). A powers theory of modality: Or, how I learned to stop worrying and reject possible worlds. Philosophical Studies, 151, 227–248. Johnston, M. (1992). How to speak of the colors. Philosophical Studies, 68, 221–263. Jubien, M. (2007). Analyzing modality. Oxford Studies in Metaphysics, 3, 99–139. Lewis, D. (1973). Counterfactuals. Oxford: Blackwell. Lewis, D. (1986). On the plurality of worlds. Oxford: Blackwell. Manley, D., & Wasserman, R. (2008). On linking dispositions and conditionals. Mind, 117, 59–84. McKitrick, J. (2003). A case for extrinsic dispositions. Australasian Journal of Philosophy, 81, 155–174. Peacocke, C. (1999). Being known. Oxford: Oxford University Press. Plantinga, A. (1974). The nature of necessity. Oxford: Oxford University Press. Pruss, A. R. (2002). The actual and the possible. In R. M. Gale (Ed.), The Blackwell Companion to metaphysics. Oxford: Blackwell. Schaffer, J. (2003). Is there a fundamental level? Noûs, 37, 498–517. Schnieder, B. (2008). On what we can ensure. Synthese, 162, 101–115. Shoemaker, S. (1980). Causality and properties. In S. Shoemaker Identity, cause, and mind: Philosophical essays (pp 206–233). Oxford: Oxford University Press (2003). Shoemaker, S. (1998). Causal and metaphysical necessity. Pacific Philosophical Quarterly, 79, 59–77. Stalnaker, R. (1976). Possible worlds. Noûs, 10, 65–75. Stalnaker, R. (2003). Ways a world might be. Oxford: Oxford University Press.

15

This paper is based on my DPhil thesis at the university of Oxford (Vetter 2010). I am greatly indebted to my supervisors, Timothy Williamson and Antony Eagle. In working on the DPhil thesis, I have received financial support from the Arts and Humanities Council (AHRC), the Studienstiftung des deutschen Volkes, and the DAAD, as well as a Hanfling scholarship from the Faculty of Philosophy at Oxford, for which I would like to thank the Hanfling family. For comments specifically on this paper, I would like to thank Stephan Schmid, Romy Jaster, Sebastian Bender and Kristina Engelhard.

From Potentiality to Possibility

301

Swoyer, C. (1982). The nature of causal laws. Australasian Journal of Philosophy, 60, 203–223. Vetter, B. (2010), Potentiality and possibility (Ph.D. thesis), University of Oxford. Vetter, B. (2011). Recent work: Modality without possible worlds. Analysis, 71, 742–754. Vetter, B. (2015). Potentiality: From dispositions to modality. Oxford: Oxford University Press. Vetter, B. (2016). Williamsonian modal epistemology, possibility-based. Canadian Journal of Philosophy, 46(4-5), 766–795 Wang, J. (2015). The modal limits of dispositionalism. Nous, 49, 454–469. Williamson, T. (1998). Bare possibilia. Erkenntnis, 48, 257–273. Williamson, T. (2007). The philosophy of philosophy. Oxford: Blackwell. Yates, D. (2015). Dispositionalism and the modal operators. Philosophy and Phenomenological Research, 91(2), 411–424

Author Biography Barbara Vetter is Professor of Theoretical Philosophy at Freie Universität Berlin, Germany. She works on all things modal and dispositional; her monograph “Potentiality” (Oxford University Press 2015) develops a theory of metaphysical modality in terms of dispositional properties or potentialities. She has also published on modal semantics and modal epistemology, abilities, counterfactuals, and other related topics. Her articles include “Dispositions without conditionals” (Mind 2014), “Counterpossibles (not only) for dispositionalists ”(Philosophical Studies 2016).

Part V

Potentiality in Specific Fields of Philosophy

Potentialities in the Philosophy of Mind Frank Hofmann

1 Introduction Talk of ‘potentialities’ (or ‘potencies’ or ‘potentials’) is frequent, if not ubiquitous, in the philosophy of mind. Some notions are explicitly about potentialities, some implicitly. Talk of faculties (capacities) and dispositional states is straightforwardly potentialities talk. And speaking of things like ‘accessibility’ (or ‘access consciousness’) or ‘acting for a reason’ (requiring the subject to exercise a rational capacity) or inferential roles or … is at least implicitly about potentials or potentialities. A leading—perhaps the leading—idea here is the thought that ‘the mind is nothing but potential’. (Call this the ‘potentiality thesis’, ‘PT’ for short.) It can be fleshed out in various ways. One natural way is the following one: What distinguishes a mind from a system which is not a mind is just its potential; to have certain potentialities is to have (or to be) a mind. This potentiality thesis is probably very old, and is somehow virulent in Aristotle. Classical, causal-dispositional functionalism is the most recent implementation of this idea—and it still has some currency on the contemporary scene. In addition, there has been a kind of re-habilitation of potentials and modal entities in general metaphysics recently, and in metaphysical theories of properties, laws and causation in particular. Nowadays, many think that potentials and dispositions belong to the fundamental furniture of the world. For example, Cartwright (1989) holds that capacities are an indispensable and irreducible element in describing causation. Thus it is natural to consider the role that capacities and dispositions may play in matters of the mind. If there is no principled objection to such modal entities, then the door is open for trying to understand the mind in terms of them. And perhaps, as the potentiality thesis wants to have it, the mind is essentially nothing but potential. F. Hofmann (&) University of Luxembourg, Esch-Belval, Luxembourg e-mail: [email protected] © Springer Science+Business Media B.V., part of Springer Nature 2018 K. Engelhard and M. Quante (eds.), Handbook of Potentiality, https://doi.org/10.1007/978-94-024-1287-1_12

305

306

F. Hofmann

But doubts can arise from various sides. The mind exhibits certain fundamental features that make it hard to think of it as entirely constituted by potentials. First, considering the topic of normativity, it is by no means clear that functions in the sense of purposes or jobs (what a system or state is supposed to do) can be reduced to, or analysed in terms of, potentialities (causal-dispositional functions). Secondly, it is by no means clear that content externalism can be accommodated by the potentiality thesis. If actual relations to things in the environment, including historical relations, determine what a state is about (its representational content) and content is essential for the mind, then the mind cannot be merely potential. Thirdly, there is the intuition that phenomenal consciousness is not just potential but actual, or manifest; awareness involves a kind of Russellian acquaintance with things. Therefore the mind cannot be understood in merely dispositional terms. Nevertheless, even if the mind is not exhausted by potentialities, potentialities still play an important role. So there is good reason to think that understanding the mind requires an understanding of its potentiality face. The plan for the following is this. In Sect. 2, I will investigate into what a ‘potential’ is. I will identify three preliminary categories of potentials: faculties, dispositions, and pre-dispositions. Then I will argue for a reduction to just one category, the category of dispositions. In Sect. 3, I will introduce the potentiality thesis, PT. According to PT, ‘the mind is nothing but potential’. In Sect. 4, problems for PT will be presented and discussed, to the effect that PT cannot be upheld in the end. Section 5 is a conclusion which will state some important remaining roles for dispositions in the philosophy of mind. Even if potentials are not sufficient, they are necessary for minds like ours and still have an important place in the workings of our minds.

2 Categories of Potentials There are many things that can be said to have or to be potentialities (or potentials): faculties, processes and capacities, tendencies and dispositions, pre-dispositions and innate programs, and so on. This poses the question of what the fundamental categories are in which potentialities come. A plausible suggestion is that everything that goes by the name of ‘potentiality’ belongs to one of the following two categories (or can be constructed from elements of these two categories): faculties and dispositions. A problematic third kind is perhaps the category of pre-dispositions. In the following I will try to sketch these categories, and I will end by discussing the question of reduction. My conclusion in this section will be that reduction to the category of dispositions is successful. In the end, we need only one category of potentials. Let us begin with the very natural view that potentialities know of two different kinds, faculties and dispositions. The fundamental difference between faculties and dispositions is the ontological difference between particulars and states. Faculties are systems or sub-systems, and so they are particulars. Dispositions are states or

Potentialities in the Philosophy of Mind

307

properties, in contrast. A visual perceptual faculty, for example, is a subsystem of an individual being and, therefore, is a particular. It may be integrated in a whole system of cognitive faculties, and may interact with these in very complex ways. It may work in a ‘modular’ way or not. In any case, it is a particular (a complex particular, to be sure). A disposition, on the other hand, is a state or a property. Fragility is a classical example, but according to many philosophers there are very many dispositions, no matter whether they carry their dispositional nature on their linguistic sleeves or not. For example, mass and charge may be dispositions, or bestow dispositions on their bearers. So there are probably many potentialities around in the world—in the state or property sense. Plausibly, to be in a certain mental state comes with having certain dispositions, at least in many cases. For example, beliefs and other propositional attitudes bestow certain dispositions on their bearers.1 The connection between faculties and dispositions is of course that (sub-)systems can have (or possess) dispositions. A faculty can be in a certain state at a time, and then it has a certain set of dispositions. Or, perhaps more precisely, if one focuses on the whole system one could say that the whole system has certain dispositions by having sub-systems which are faculties that are in certain states. Then there is twofold potentiality. First, the system has a potential by having a part which is a faculty (the visual perceptual faculty in question). The system ‘can’ do certain things, in virtue of having this faculty, in one sense of ‘can’. (The states of the faculty form a manifold or state space whose elements it can occupy.) Second, each possible state of the system is itself a potentiality. Correspondingly, we can note that there are two steps of manifestation, or two different manifestations. The state need not be manifested. It may be unmanifested, mere potential. If it is manifested (by some suitable triggering stimulus under appropriate circumstances), this is the second kind of manifestation. But there is already some first manifestation when there is the—unmanifested—state. This first manifestation is the manifestation of the potential belonging to the faculty (consisting of the manifold of its possible states). This is so, at least, whenever the state is the normal output of the relevant faculty. Then the potential of the system that it has in virtue of having the relevant faculty is manifested. The system has done what it ‘can’ do. For example, the visual perceptual faculty has produced a certain visual experience, or it has produced a certain perceptual belief (in cooperation with a certain conceptual, doxastic faculty). The first and the second manifestation are independent of one another. As we have just seen, even if the potential of the faculty is manifested, by the occurrence of a suitable state of the faculty, this state itself could be unmanifested. On the other hand, perhaps the state may occur without having been produced by the relevant faculty in the normal way, but be manifested nevertheless. This would be a case of a 1

Dispositions are not to be confused with competences (or abilities). On a dispositionalist account of competences, competences are dispositions with special epistemic features—the epistemically good dispositions so to speak. For a more detailed proposal about what distinguishes competences see, for example, Hofmann (2013).

308

F. Hofmann

manifestation of the second kind without a manifestation of the first kind. (Perhaps this latter possibility is rather problematic, but the former is certainly quite clear.) Let us now take a look at the various items and entities on our list above in order to see whether they can plausibly be assigned to these two categories, faculties and dispositions. To begin with, processes are changes of particular (sub-)systems. When we think of a process as a potential, we (primarily) think of it as a possible change in state of a certain (sub-)system. So it is the system which has the potential. It ‘can’ undergo the relevant changes, by moving from one state into another. (Remaining in a certain state is a limiting case of a process and can be harmlessly included.) ‘Capacity’, it seems to me, is merely another term for a faculty. Or it stands for the potential that a faculty has or bestows. Tendencies, on the other hand, can be classified as dispositions. For example, to have the tendency to make the judgment that spatial distance is symmetric is to have the disposition to make this judgment. Plausibly, doxastic systems bestow tendencies like these on their bearers. Pre-dispositions and innate programs are perhaps another breed. One could mean different things by ‘innate program’. Innate programs may either be innate faculties, or parts of such innate faculties. Or they may be pre-dispositions. This leads us to the hypothesis that there is a third kind of potential, apart from faculty and disposition, which we have to consider now, the category of pre-dispositions. The paradigmatic kind of example of a pre-disposition is something like the pre-disposition of an embryo to develop into an adult living being with many faculties and dispositions. A human embryo may have the pre-disposition to form a linguistic faculty. It ‘can’ speak certain natural languages, in one sense of the term ‘can’. (In contrast, an embryo of another species may lack this pre-disposition; it ‘cannot’ speak these languages.) Given a certain environment with certain conditions, the embryo will or would develop in a certain way; it will or would grow into a system having many faculties, and thus many potentialities. In one word, we are dealing with a ‘potential for a potential’, a ‘second-order potential’ if you will. Ontologically, it is not easy to classify such a second-order potential. It is not a potential for producing a certain state of the system. It is a potential of the system to develop into another kind of system, with quite a different structure. To take another, simple case, consider a cell. It may have the pre-disposition to split into two ‘new’ cells. It can give rise to two particulars; it can ‘reproduce’, as we say. This poses a question of identity over time, of course. The old system has gone, and two new systems have taken its place. At least, this is the picture if we stick to an ordinary objects ontology. (If we rely on a more scientific ontology, of spatio-temporal regions, fundamental physical properties, and facts, for example, things may look quite differently. But let us stick to an ordinary objects ontology for the present purposes, since this is where talk of pre-dispositions is at home.) Perhaps we can put this problem of identity over time to one side, and focus instead on cases where the individual remains one. Nevertheless, a second-order potential seems to be a potential for changing into a system with a quite different structure. In the beginning, it has one set of faculties (with corresponding state spaces), and after the change it has a quite different set of faculties (with corresponding different state

Potentialities in the Philosophy of Mind

309

spaces). In one word, the second-order potential is a potential for acquiring a whole new profile of potentials. It is a ‘diachronic’ potential. Reduction is the philosopher’s favorite project. How about reducing all potentialities to just one category? The most promising candidate seems to be the category of dispositions. So can we reduce faculties to dispositions, in some interesting sense (and also pre-dispositions)? Can we get along with just one category, the category of dispositions? Certainly, a particular cannot easily be reduced to a state or property. But can we not say that having a faculty is having a part with a certain set of dispositions? What makes a part of a system a faculty is no more and no less than its bestowing certain dispositions on the system. So the way to an interesting reduction seems open. Systems have parts with certain dispositions—these are their faculties. Under appropriate conditions, they produce states as outputs. And these output states are dispositions. A first objection here is the following one. According to the reductive proposal, a faculty is supposed to be a sub-system with certain dispositions. It is disposed to produce certain states. But these states then are dispositions again. So we get dispositions to producing dispositions. And isn’t this quite different from dispositions? Don’t we need then two different kinds of dispositions, those having dispositions as manifestations and those having actualities as manifestations? Which would mean that we do not get any really interesting reduction. In reply to this objection, it must be recognized that in general, the manifestations of dispositions can be instantiations of states or properties, and that these states or properties can be dispositional. A faculty produces the instantiation of a state. Then the system is in that state, and thereby it has certain dispositions. There is no problem as long as we distinguish clearly between the instantiation of a disposition and its manifestation. The manifestation of a disposition may be the instantiation of a further disposition. The objection depends on confusing these two. So far, then, no problem for reduction. How about pre-dispositions then? Pre-disposition could be cited as a second objection, as follows. They do not seem to be ordinary dispositions, since they involve a potential for changing into an altogether different system, a system with a quite different structure, and with different faculties. The embryo does not—yet— have any perceptual faculty, but it is pre-disposed to grow into an adult system with such a faculty. So we cannot simply say that the system (the embryo) is disposed to produce a certain output state; the output is not a state of the system, but of a different system! In response to this second objection it may be helpful to compare the case of a reproducing cell with the case of a paradigmatic disposition, such as the fragility of a window pane. Triggered by a suitable stimulus, the window pane manifests its fragility by breaking into several pieces. The system has thereby vanished, and several new systems have emerged—the remaining pieces of glass. So it seems that we have basically the same structure. One particular (the window pane—the initial cell) goes out of existence and some other particulars come into existence

310

F. Hofmann

(the pieces of glass—the two new cells). So what is the big difference? Don’t we just have dispositions in each case, dispositions of the same kind? I submit that the answer to this question is yes. There is indeed no difference of a principled kind. Pre-dispositions are dispositions. It is just that the manifestations of dispositions are of different sorts. Sometimes the manifestation is simply a state of the same system. For example, if an object is flexible, it can be bent without breaking. If so, it is in that state—being bent. The manifestation is a state of the same object. Sometimes, however, the manifestation of a disposition involves what is called a disposition partner. Then, the manifestation is a certain relation between the two objects. For example, consider the disposition to attract other bodies that a body has in virtue of having a certain mass. The manifestation is a certain relation between the body and some other body, a relation of attracting one another (with a certain strength). This shows that the manifestation of a disposition of a system may involve other systems than the system itself and its parts. In still another kind of case the manifestation will involve the ‘destruction’ of the system itself. Interpreted in one way, this is exemplified by the breaking window pane. The manifestation involves the destruction of the system, and the emergence of one or several new systems (the pieces of glass). The decay of an atom is another example of this kind. The atom decays—it vanishes—and some other atom or atoms remain. Here, again, we are dealing with a case in which the manifestation involves systems different from the original system which was the bearer of the disposition. We can conclude, therefore, that dispositions can have manifestations of different kinds—but are simply dispositions nevertheless. Pre-dispositions are no exception. The following picture emerges from these considerations. We need only one category of potentialities, dispositions. Faculties are particulars that have certain dispositions. A person who has a visual perceptual faculty, for example, has a sub-system with certain dispositions, such as a disposition to produce certain mental states (visual experiences) under appropriate conditions (good lighting etc.) and given certain stimuli (stimulation of the retina). Of course, the particular remains a particular, and not a state or property. But it is a faculty simply in virtue of having certain dispositions. In this sense, no more than dispositions are needed for faculties. Finally, pre-dispositions are nothing but dispositions. Perhaps, these dispositions are called ‘pre-dispositions’ since in paradigmatic cases long-term processes of manifestation are involved. The final manifestation is somehow temporarily removed from the triggering. (Or the triggering comes in several steps, or even is continuous, spread out in time significantly.) These are indeed differences, but no differences that force us to introduce potentials of a different category. So reduction to one category, the category of dispositions, seems successful.

Potentialities in the Philosophy of Mind

311

3 The Potentiality Thesis As mentioned in the introduction, a leading—perhaps the leading—idea in the philosophy of mind is the thought that ‘the mind is nothing but potential’. (Call this the ‘potentiality thesis’, or ‘PT’, for short.) According to PT, a mind is a system with a certain potential. Having this potential is both necessary and sufficient for being a mind. No more and no less is needed. This excludes many things, and we will see that PT must face serious problems. But initially, it does have some intuitive plausibility. And it is interesting to see how far we can get by following PT as persistently as possible. Perhaps, PT can be found in Aristotle. It is not quite clear which position Aristotle takes in De Anima. But PT seems to be an idea that is close enough to Aristotle’s thinking. According to Aristotle, the mind has various parts or faculties, and these parts or faculties are characterized by their abilities, or by what they contribute to the abilities and capacities of the living being to which they belong. Human beings have a special faculty—the ‘nous’—and this is what explains their being able to think, reason, and do science and philosophy. Contemporary causal-dispositional functionalism—‘cd-functionalism’, for short —is a recent position in the philosophy of mind which subscribes to PT. According to cd-functionalism, mental states are functional states, and functional states are ‘defined’ by their dispositions to be caused by input states and inner functional states, and to cause other inner states and output states.2 Functionalism in this sense has been proposed by philosophers such as Hilary Putnam, Jerry Fodor, and David Lewis, and it is still en vogue. There are different versions of it, but the basic idea is common to all of them. The network of causal relations between input states, inner functional states and output states is all that is essential for a mind. It is important to note that here ‘causal’ is to be understood in a dispositional sense. Whether a state qualifies as having the right causal relations to the other states depends (only) on whether it bestows the right dispositions. This is why we can call this kind of view ‘causal-dispositional’ functionalism. (It is to be distinguished clearly from the quite different kind of functionalism to be considered later, i.e., teleo-functionalism.) This fits perfectly well with PT: to be in a mental state is to be in a state with the right kind of potential. How this potential has been brought about drops out as irrelevant. Whether it is innate or acquired, and how it has been acquired, is irrelevant. As long as the potential is there, there is the mental state. Indeed, it does not seem to be an accident that cd-functionalism fits well with PT. For, PT seems to be equivalent to cd-functionalism. At least, this is so if we accept the reduction discussed above. If potentials are (ultimately) identical with, or grounded in, dispositions, then it is hard to see how a proponent of PT could not be a cd-functionalist. In what other way could one adhere to PT, if not by being a cd-functionalist? Mental states are potentialities, and potentialities are dispositions. So mental states are states with a certain dispositional profile. What remains to be 2

Cf. Block (1980).

312

F. Hofmann

added, in order to arrive at cd-functionalism, is merely the further idea that there is a division into input states, inner states, and output states. But that a cognitive system has such a threefold state space structure (at some level of description), seems to be a very basic assumption that is quite uncontroversial. On the other hand, a cd-functionalist subscribes to PT, of course. So PT and cd-functionalism are basically equivalent.3 There are some important objections against cd-functionalism. Ned Block has raised some of them.4 An important one has to do with phenomenal consciousness. A further important objection concerns content (or truth conditions). But the most fundamental objection would be the hypothesis that there are no potentials, or that talk of potentials does not make sense. We can set this objection to one side. It is the radical skeptical objection against potentials in general, and so it is not specific to potentials in the philosophy of mind. (And as already mentioned, dispositions have been re-habilitated in general metaphysics recently, so the radical skeptical objection does not seem to have much to be said in its favor.) But a similar objection goes against the appeal to dispositions in the philosophy of mind, without doubting the existence of potentials. It runs like this. Cd-functionalism requires us to believe that a system with a certain state space which is in one of its states is a mind—merely by these states’ having certain dispositions. None of these dispositions must be manifest at any time; they only have to be instantiated. This may raise some doubts. Intuitively, it is not clear that potential in this sense is sufficient for a mind. There may be such a potential, but is it enough for accounting for the mind? Don’t we need some amount of manifestation, or actualisation? Perhaps not at the time that we are considering. But at least at some earlier times. Call this the ‘mere-potentiality objection’. Perhaps, it is most compelling when we consider states of phenomenal consciousness, the famous ‘qualia’ of the mind. Can we really accept the idea that a state comes with a certain feel or what-it-is likeness by merely having certain dispositions (to bring about other states, and to be brought about by other states)—without any of these being ever manifested? For example, can we really accept the idea that one could feel a pain by merely having certain unmanifested dispositions? Here, the merepotentiality objection seems to have some intuitive bite.5 Similarly, we can doubt the necessity of potentials. For, consider a mind which is disrupted or mad. It works quite differently from a healthy mind. Its mental states seem to have quite different dispositions. If ‘mad pain’ is still pain, it has a quite different dispositional profile.6 It is not clear that there is one set of dispositions which is common to all pains (not even within one species). Perhaps, what makes a

3

Therefore, it is no surprise that some have interpreted Aristotle as a cd-functionalist. This interpretation has been given by Hilary Putnam, for example. Cf. Putnam (1975). 4 Cf. Block (1978), for example. 5 I have heard the objection many times in conversation. A written version can be found in Jacobs (2011), for example. 6 Mad pain has been discussed by Lewis (1980).

Potentialities in the Philosophy of Mind

313

state a state of pain is something actual. Then it is not merely a matter of its dispositions whether it is a pain. Actual causal relations may be essential, and the instantiation of dispositions may be inessential. If so, cd-functionalism and the PT are wrong. Call this the ‘irrelevance-of-potentiality objection’. These two objections are perhaps not all that forceful. It is hard to make up one’s mind about them. To put heavy emphasis on phenomenal consciousness is perhaps rather risky. And with respect to non-phenomenal mental states it is not so clear that the mere-potentiality objection is intuitively very strong. Equally, the irrelevance-of-potentiality objection is somewhat ‘thin’ or ‘abstract’. We would need some more concrete illustration, some story about which actual things are decisive for a mind, in order to come to a considerate judgment. So we have to look for further arguments. The goal of the next section is to present three such arguments.

4 Problems with the Potentiality Thesis Doubts about PT can arise from various directions. A first and very important consideration concerns the normativity of the mental. Two further arguments concern content externalism and phenomenal consciousness. Let us start with normativity. Many philosophers have emphasized that mental states are essentially normative. What exactly this means is quite controversial.7 But it is remarkable that philosophers of very different views have appealed to normativity. In particular, it is both naturalist and anti-naturalist philosophers who have pronounced that the mind is somehow normative. Perhaps, it is no surprise to encounter anti-naturalists who appeal to normativity. But there are also many naturalists who hold the view that mental states are essentially normative. For example, Ruth Millikan has defended quite explicitly the thesis that mental states cannot be reduced to mere dispositions since they are essentially normative, and the normative cannot be captured by mere dispositions. If so, PT is wrong. Normativity poses a problem for PT.8

See, for example, David Copp’s distinction between three kinds of ‘normativity’ in Copp (2007). Of course, the literature on normativity is too vast to be cited even in a representative fashion. Within the discussion raised by Kripkenstein a distinction between mis-functioning and violation of oughts is often made, and only the latter—oughts and their violation—are counted as ‘normative’ (cf., for example, Handfield and Bird (2008). The distinction can be found in Kripke (1982)). Any such view is controversial, however, and I will not rely on it here. These are questions that a complete theory of normativity has to settle. 8 Millikan’s critique of a dispositionalist account is most clearly stated in Millikan (1993), esp. sc. 8. 7

314

F. Hofmann

Millikan herself proposes a teleo-functionalist account of normativity. According to such a view, mental states have ‘functions’ in the sense of purposes, or jobs, or oughts.9 This is teleo-functionalism, to be distinguished very clearly from the kind of functionalism that we have discussed above, causal-dispositional functionalism. Teleo-functionalism is quite controversial, but what is considerably less controversial is the view that mental states have teleo-functions—jobs, purposes, functions in the sense of oughts, not in the causal-dispositional sense. For example, perceptual states have the teleo-function of veridically representing the environment, and beliefs have the teleo-function of truly representing the world.10 But these are mere examples. The threat to PT is general and independent of which specific teleo-functions one thinks a certain mental state has. As long as there are some mental states which are essentially normative, PT is in danger. Let us compare cd-functionalism and teleo-functionalism. The contrast is simple and striking. According to cd-functionalism, a mental state is distinguished by what it does, according to teleo-functionalism it is distinguished by what it is supposed to do. The two can come apart. A mental state may not do what it is supposed to do. For example, it may not correctly represent the world—it may misrepresent—but have the teleo-function of correctly representing the world. A job without fulfillment. Equally, a mental state may do something—correctly represent the world, e.g. —without having the teleo-function of doing so. For example, if one imagines (or conceives) that p, this may be a correct representation of the world—since it is a fact that p—even though this state of imagination does not have the teleo-function of correctly representing the world. (Which teleo-function it has, if any, is a difficult question.) ‘Fulfillment’ without job. Thus, teleo-functions are radically different from causal-dispositional functions. And it cannot be emphasized enough that we have to keep them apart conceptually very strictly. Of course, in many cases cd-function and teleo-function go hand in hand. Whenever a state fulfills its teleo-function, it does what it is supposed to do, it brings about its goal and, thus, manifests a causal potential. This is perhaps why it is often overlooked that there are two quite different sorts of ‘functions’. But once one is aware of the ambiguity, it should not be too hard to keep the two apart.

The ‚ought’ here is not a practical ought, not the ought of a practical reason, as Millikan emphasizes. If a mental state has a certain teleo-function, this does not entail that anyone ought to do anything in order to make it fulfill its function. It is an ‚ought’ in the sense of an ‚is supposed to’, without giving rise to any practical reason. The normativity in question is similar to the normativity of a toaster or some other artifact with a ‚function’. It is a goodness of a certain kind— being good as a toaster, e.g.—that an object has when it fulfills its ‚function’, its job. For an interesting analysis of various different kinds of goodness compare Thomson (1997). 10 That perceptual states have the teleo-function of veridically representing the environment has been emphasized by the—anti-naturalist—Tyler Burge, for example. Cf. Burge (2010). It should be noted that, as I use the term ‘teleo-function’ here, it is not a particular theory or account of the functions in question that I mean. ‘Teleo-function’ is simply the pre-theoretic term I am using to refer to the phenomenon in question. Whether a certain account or theory of teleo-functions, such as Ruth Millikan’s, is right or not is a quite different, further question. 9

Potentialities in the Philosophy of Mind

315

When it became clearer that cd-functionalism had to face serious problems, Sober (1990) announced that we should ‘put the function back into functionalism’. This was the quasi-paradoxical recommendation to bring back teleo-functions into one’s functionalist view. The recommendation makes sense, however, since classical functionalism was cd-functionalism, and cd-functionalism ignored teleofunctions. In effect, what Sober suggested was a turn away from cd-functionalism and toward teleo-functionalism. By switching from cd-functionalism to teleo-functionalism, one acknowledges the essentially normative nature of the mind (independently of whether a naturalistic reduction of teleo-functions is possible or not). Normativity is no longer a threat, as it was to cd-functionalism. Nowadays, many agree with Sober’s recommendation.11 But equally, cd-functionalism still has its adherents.12 At this point, however, one might wonder whether a reduction of teleo-functions to dispositions is possible. Why not account for teleo-functions in terms of dispositions? (Remember that by ‘teleo-functions’ we simply mean purposes or jobs.) Normativity might be respected by appeal to ‘normal conditions’ as the associated manifestation conditions. Let us assume that dispositions are characterized by a stimulus, a manifestation condition, and a manifestation. For example, fragility (of a window pane, for example) has a suitable transfer of momentum as its stimulus, ordinary circumstances like the ones prevalent on the surface of our planet as its manifestation condition, and breaking as its manifestation. Now we could appeal to certain conditions as the ‘normal conditions’ in order to get a kind of normativity. If being in a certain state m makes the system disposed to behave in a way w, and the manifestation conditions are certain circumstances N, then the state exhibits a certain normativity. For, the purpose or goal of state m is to bring about behavior w. If conditions N hold, the behavior will occur. But under different circumstances, w will not occur. Producing behavior w is the purpose, or teleo-function, of state m. And it can fail to produce what it is supposed to, namely, when abnormal conditions occur. So can we not account for normativity and teleo-functions in terms of dispositions along these lines?13 One major worry about this proposal is that it misses the mark of normativity, as it is present in mental states. In general, a system possesses very many dispositions to behave in very many ways corresponding to the many different conditions in which it could be. Under different conditions, it may be stimulated to behave in

11

For example, Nick Zangwill has vigorously argued against cd-functionalism, on the basis of its missing teleo-functions. Cf. Zangwill (2005, 2009). 12 Interestingly, many philosophers who had been cd-functionalists have turned toward representationalism, for example Michael Tye, Sydney Shoemaker, and William Lycan. 13 The proposal is somewhat similar to one considered (and rejected) by Paul Boghossian, in his discussion of dispositionalist accounts of rule-following in Boghossian (1989). Famously, one aspect of the normativity of the mental—representational, or semantic, normativity—has been brought to our attention by Wittgenstein and Kripke. Cf. Wittgenstein (1984) and Kripke (1982). The proposal is also significantly similar to Tye’s proposal (Tye 1995), building on Stampe’s account of representation. Correlation under normal conditions can be taken as a way of behaving.

316

F. Hofmann

various different ways. To call certain conditions ‘normal’ is not to make them special in any interesting way. With respect to the so-called ‘normal’ conditions N, the system is disposed to behave in way w. With respect to some other conditions M, it is disposed to behave in way w’, let us say. And so on. The system has many dispositions, corresponding with many different manifestation conditions. So far there is nothing normative in the picture. There is no basis for saying that the system is supposed to behave in way w rather than in way w’. It does behave differently when in those other conditions M (the ‘abnormal’ ones), but why should this be a case of (systematic) mis-behavior? Unless some further story is added (which may very well be possible, in my view) normativity has not been captured. But how about adding a story about what makes certain conditions special, what makes them really the normal conditions? Well, again, this seems to miss the target. We are not looking for what makes certain conditions normal (for the system in question)—what makes it the case that the system is supposed to be in certain conditions—but rather of what makes it the case that a state of the system is supposed to do a certain thing. And whether the state really shows this behaviour under certain conditions or not is a further and independent question. The system either has a certain disposition or not. There is no room for mis-functioning.14 So we have to conclude that reduction, or at least the straightforward reduction of normativity to dispositions considered so far, is not successful. Whether a more sophisticated story could succeed remains to be seen.15 A dispositionalist proposal which is (perhaps) of a different kind appeals to interfering factors, not to normal conditions. It has been proposed by Heil and Martin (1998), and criticized by Handfield and Bird (2008). According to this proposal, mis-functioning can be explained by reference to some factor which interferes in such a way as to prevent the occurrence of the corresponding disposition. (The interfering factors are of different kinds and go by various names, such as ‘masks’, ‘finks’, and ‘antidotes’.) To take Kripke’s famous example, a subject may have the disposition to add certain natural numbers (37 and 42 for example) correctly (79) but fail to do so on a particular occasion because of the interference of a certain intervening factor, such as emotional distress or fatigue. Mis-functioning is due to an intervening factor.

14

It is not clear that all (mental) representations must be capable of being false. Perhaps, there is a class of representations which cannot be false. (Cf. Millikan 2004.) But this does not affect the important point that many mental states can mis-function, can fail to fulfill their function. Thus, the proposal still fails since it cannot account for any mis-functioning. 15 The following remark may be helpful. The state does not have a conditional purpose (‚it ought to {behave in way w if conditions N hold}’) but an unconditional one (‚it ought to {behave in way w}’). Even when in other, non-N conditions, the state is supposed to behave in way w. This much is presupposed when we call it mis-functioning in non-N conditions. A car, when put on the moon, malfunctions (since oxygen is missing).

Potentialities in the Philosophy of Mind

317

The proposal, however, does not seem to be a new one. For one can treat intervening factors as abnormal conditions. Plausibly, factors like stress and fatigue are abnormal conditions for the operation of mathematical reasoning.16 If so, the criticism mentioned above applies. The subject has one disposition—to add numbers (mathematically) ‘correctly’ (79) when not tired and not emotionally stressed (and so on). And she equally has the dispositions to add ‘incorrectly’ (89, say) when tired or emotionally stressed. But what ground is there for saying that, in the latter case, some mis-functioning takes place, or that the subject is mis-performing? The subject simply exercises one of her dispositions. And that the manifestation conditions associated with the one disposition are ‘normal’ cannot turn the manifestation of the other disposition—the ‘wrong’ calculation—into a case of mis-functioning.17 Rather, the subject does not perform an addition at all in those conditions and, thus, does not mis-calculate the sum of the two numbers. (Whether she is following some other ‘rule’ or no ‘rule’ at all is a further question that we need not answer for the present purposes.) But if no mis-functioning is possible, then it is equally unclear whether any proper functioning—and proper calculation of addition—can take place. The dispositionalist account, therefore, cannot explain normativity. Mere dispositions are not sufficient.18 It should be noted that it has not been argued that there is no way to account for a condition’s or disposition’s being normal. It has merely been argued that the subject’s dispositions alone cannot explain normativity. Something else must be added. Indeed, I suspect that a suitable addition, perhaps in terms of a causal-historical story similar to Millikan’s teleo-functionalism, can be provided. But then dispositions do not carry the main burden any more.19 16

Issues having to do with infinity and very large numbers do not seem to concern the normality of operation conditions. But they are special to the domain of the example, mathematics, and do not pertain to the general problem of mis-functioning. 17 The case is different if we consider a subject which uses a calculation procedure that systematically produces some wrong result. Then it is questionable whether the procedure is one of adding, and whether the person is adding (wrongly) when applying the procedure. (Charity of interpretation may lead us to ascribe an ‚attempt at addition’, but interpretation is an altogether different topic.) In contrast, if the procedure is mathematically correct (for addition) and the subject applies it when stressed or tired, and comes up with the wrong result, we do seem to be dealing with a case of mis-adding. Note that the errors are still systematic errors, not singular ‚freaks’. 18 Mathematics is difficult to understand and poses special problems. As already mentioned, issues of infinity are special to mathematics, for example. The phenomenon of normativity is quite pervasive to the mind (if it exist at all), and it is thus not good advice to concentrate too much on mathematics. Another specialty in mathematics is of course the non-concrete nature of numbers and other mathematical entities which makes it hard to account for mathematical thinking in causal terms. But the case is perhaps not entirely hopeless, since numbers could be analyzed as relations (universals) between physical quantities—as proportions—following Bigelow’s ingenious account. Cf. Bigelow (1988). If so, causal-historical relations to number instances are not entirely out of the question. 19 Decomposition of the system into (functional and/or physical) sub-systems may be helpful. A system’s behavior can be explained by the dispositions of suitable sub-systems of it, where the

318

F. Hofmann

A second objection to PT and cd-functionalism concerns content externalism. Many, if not all, mental states have a representational content. (Some call it ‘intentional content’. Here I take this term to be synonymous to ‘representational content’.) But as is well known from considerations put forward by Hilary Putnam, Fred Dretske, Tyler Burge, Donald Davidson, and many others, representational content allows for externalism—content externalism. Roughly speaking, which content a certain state has can depend on the environment of the subject, including its history. (We need not assume that all content is externalist. It suffices for the present argument that there is some externalist content.) But plausibly, which dispositions a state has supervenes on the intrinsic physical makeup at the relevant time, and so does not depend on any external circumstances.20 A system with the same intrinsic physical makeup shares the entire profile of dispositions. But it can have states with different representational content, according to content externalism. Thus, content externalism clashes with PT and cd-functionalism. If we accept content externalism, PT and cd-functionalism have to go. The point can be made in a slightly different way. If content externalism is right, there are some relations to external matters on which content depends. To take Putnam’s famous example, it matters whether one has been in causal contact with H2O or with XYZ. But the relation to the external stuff is not merely a potential one (if that makes sense at all). It is the actual instantiation of a certain causal relation (to an actual instance of the relevant stuff) which matters. So it is not mere potential which determines content but certain actual affairs. Which stuff a person was actually confronted with and interacting with is essential for content. Thus, content externalism excludes PT and cd-functionalism. Can we reconcile PT and content externalism? It is not easy to see how this should be possible. Two subjects on Earth and twin Earth may be dispositionally exactly alike, yet the one’s mental state refers to H2O and the other’s to XYZ, respectively. Is there room for some kind of ‘externalist dispositions’, dispositions the possession of which depends on external affairs and not just on the intrinsic makeup? One might think so, by considering the context dependence of dispositions. This means that whether an object possesses a certain disposition, fragility, for example, might seem to depend on the context in which the object is situated. Under normal conditions on the surface of Earth, with moderate temperatures, a hammer is not fragile. But put on planet Pluto, with extremely low temperatures, it turns fragile. So its being fragile depends on the context, it seems. An ‘externalist disposition’, if you like. Why not think that content is equally ‘externalist’.21 As a criticism to this proposal the following might serve. It has not been shown that possession of the disposition depends on the context or environment. What has dispositions of the subsystems can differ from the dispositions of the entire system. But it must be ensured that the problem is not merely shifted, as Handfield and Bird (2008) note. 20 This is at least the orthodox view. Cf. Fara (2006). 21 ‘Externalist dispositions’ seem to be what Jennifer McKitrick calls ‘extrinsic dispositions’ (McKitrick 2003). (We deal with dispositions that fail to be necessarily shared by perfect intrinsic duplicates.) If so, the following criticism would apply to McKitrick’s view as well.

Potentialities in the Philosophy of Mind

319

been pointed out is merely that there is a multiplicity of dispositions to break, not just one. All of them are relative to certain manifestation conditions. One such fragility disposition is individuated with respect to normal conditions of planet Earth (‘Earth-fragility’). Another one is individuated with respect to conditions prevalent on planet Pluto (‘Pluto-fragility’). And so on. Our object is not Earth-fragile, but it is Pluto-fragile—and it is so absolutely, independent of its context or environment. An object’s being Earth-fragile is absolute, not dependent on its context or environment. Equally, being Pluto-fragile is not ‘externalist’. So possession of dispositions like these is not externalist. The proposal rests on a confusion about the role of manifestation conditions. They belong to the individuation of the disposition, and they do not play the role of external factors on which content depends.22 Finally, phenomenal consciousness seems to present a third problem for PT. The basic intuition here is that phenomenal states are somehow ‘actual’ or ‘manifest’, and not just dispositional. A phenomenal state is an occurrent, manifest state. If we compare phenomenal states with beliefs, we have to recognize that beliefs come in two kinds, manifest beliefs and dispositional beliefs. When asleep one still believes many things, without manifestly believing them (in particular, without having the corresponding thought). But phenomenal states do not allow for such a duplicity. They are always manifest, occurrent. Perhaps, dispositional belief is only potential, and conforms to PT. But phenomenal states do not. Therefore, PT is wrong. Or so runs the objection. The objection is somehow elusive, I think. It is hard to get a clear grasp on what exactly it trades or rests on. The distinction between dispositional and manifest belief is clear enough, but also misleading to some extent. For if we say that phenomenal states are manifest states, along with manifest beliefs, and that this is the reason why PT fails, then manifest beliefs should count as counterexamples as well. But intuitively it is not clear that manifest beliefs count as counterexamples as well. Phenomenal states seem to present a special difficulty. So the grouping of phenomenal states with manifest beliefs cannot carry the whole burden of the argument. Something else is going on, I suspect. But it is difficult to articulate what it is. Perhaps, the heart of the intuition is that when one is undergoing a phenomenal state, something is ‘present’ to one’s mind in an immediate way. When seeing red, redness seems to be present to one’s mind. And similarly for other phenomenal states. This is meant by calling phenomenal states ‘manifest’: the ‘immediate presence’ of things to one’s mind that one enjoys when undergoing phenomenal consciousness. If so, what can we conclude about PT? The ‘immediate presence’ in phenomenal consciousness seems to involve a relation of acquaintance, of the sort Russell had 22

In general, manifestation conditions do not only include external conditions but also internal ones. Here we focus on the external ones.—Everyday life thinking may be confused about the intrinsicness of dispositions since it may not be sufficiently clear about the status of manifestation conditions.

320

F. Hofmann

in mind. One is acquainted with redness when seeing red (undergoing a ‘red quale’, if you like), for example. This relation of acquaintance is instantiated, by the subject and the color red. But it is hard to see how such a relational fact could be analysed in terms of ‘mere potential’ or dispositions. What disposition involving the two disposition partners (the subject and the color red) could account for this relational fact? Certainly not a disposition to behave in a certain way on the disposition partner.23 One might try to analyse acquaintance in terms of some causal relation. But this will not yield a dispositionalist account in PT’s spirit.24 If (more plausibly) one opts for a representationalist analysis, then acquaintance with the colour red is constituted by a representation of redness.25 If this is the case, we are back to our earlier, second problem with content externalism. If representing redness is a matter of external affairs, as content externalism says, so is acquaintance. One could plead for a special case here, but on what basis? Why should the sort of representation constitutive of acquaintance be exempted from externalism? Some have claimed so. And this yields an objection against externalist representationalism with respect to phenomenal consciousness. Thus, an internalist representationlist would hold that phenomenal states are representational states (of some sort or other) which supervene on the intrinsic physical makeup of the subject.26 However, there is not much force to such a line of thought. It seems considerably ad hoc to assume that representation in general is dependent on external affairs but not in the case of those representations which are constitutive of phenomenal states. In addition, there are even positive reasons for thinking that the kinds of representational states which are phenomenal are externalist. If, for example, we consider visual phenomenal states—states like having an experience of red—then the most plausible representationalist view is that these represent objective physical features of objects (or their surfaces), namely, colours.27 But what makes it the case that a certain mental state represents such a colour, a feature of external objects? Plausibly, some relation to certain instances of it—some actual relation, not merely a potential one. At least, this is the most prominent and plausible story of how

23

I am going to focus on acquaintance with properties here. Of course, there is also acquaintance with particulars, such as cats and tables. Acquaintance with particulars seems to conform even more obviously to what I am going to argue for in the case of acquaintance with properties. 24 Unless one could analyse causal relations in terms of dispositions, which seems out of the question. Bird’s attempt at analysing causal relations in terms of the manifestation of dispositions of course requires not just possession but manifestation. (Cf. Bird 2010.) So it violates PT. 25 Recently, Michael Tye has tried to re-habilitate acquaintance. Cf. Tye (2009). According to Tye, acquaintance with an entity involves more than just a non-conceptual representation of it. It requires (in the case of visual experience) „a good look at the thing“ (Tye 2009, 101). 26 This is a proposal by Terry Horgan and John Tienson. Cf. Horgan and Tienson (2002). 27 Even though somehow ‘anthropocentric’, as Hilbert calls it quite properly, these are still objective. Cf. Hilbert (1987).

Potentialities in the Philosophy of Mind

321

visual perceptual states come to represent certain properties, to be found in works of Fred Dretske, Alex Byrne, Tyler Burge, and many others.28 So when we consider visual phenomenal states, paradigms of phenomenal consciousness, these do seem to exactly be cases of externalist representation, contrary to the internalist representationalist proposal. To be sure, it is not incoherent to think that the properties which are represented by visual experiences are not the external colours that are instantiated by external physical objects (if instantiated at all). But to think so is to introduce a radical gap between the mind and the external world, and it clashes with the famous transparency intuition (going back to Moore and revived by Fred Dretske and Michael Tye more recently).29 More plausibly, what is ‘given’ to the mind in phenomenal consciousness is exactly the way the world is (at least, in cases of veridical perceptual experience). An experience as of red represent exactly the kind of property that chairs, books, and tables can have and do have when they are red. If so, externalist representationalism is vindicated positively.30 Not only can it avoid an ad hoc manoeuver, it is also supported positively by significant arguments concerning what constitutes representational content in the relevant cases. If content externalism applies to phenomenal states, we are indeed confronted with the earlier, second problem of content externalism.31 An attempt at saving dispositionalism might point to the idea that acquaintance, as a kind of knowledge, can be analysed as the potential to acquire de-re knowledge about the thing with which one is acquainted. Acquaintance is knowledge of a thing, non-propositional knowledge. Acquaintance is a two-place relation between subject and object, not a propositional attitude. But knowing a thing can be analysed as being in a position to know various propositions about this thing. So acquaintance is, after all, a potential—a potential to acquire a certain propositional knowledge. Or so runs the dispositionalist’s proposal. The proposal, however, has to face a serious difficulty. It is unclear which propositional knowledge one is supposed to be in a position to know by acquaintance. What is it that I am in a position to know about redness if I am acquainted with it (when undergoing an experience of red)? The answer is not obvious at all. That this is redness (where ‘this’ refers to the property I am acquainted with) will not work, since I might not know that it is redness that I am experiencing (instead of blueness). (Think of Frank Jackson’s Mary who is acquainted with redness when she is experiencing it for the first time but does not immediately know that it is redness.) That this is what I am experiencing now might be something I am in a position to know on the basis of acquaintance. This would 28

What external relation? Here philosophers differ. But causal relations, in the broad sense, are obvious candidates. For the present purposes, we do not have to decide which ones are the right ones. 29 Cf. Moore (1903), Dretske (1995) and Tye (2000, Sect. 3.1). 30 In the case of experiences, the most plausible view is that the content is non-conceptual—as many have argued. But strictly speaking, the question of non-conceptuality is a further and independent one. 31 Phenomenal externalism has been argued for by Michael Tye. Cf. Tye (2009, Chap. 8).

322

F. Hofmann

be a piece of immediate self-knowledge.32 But are we really supposed to think that being acquainted with redness in one’s experience comes down to no more than being in a position to know this indexical-demonstrative piece of self-knowledge? It seems much more plausible to assume that acquaintance is the basis or ground for acquiring this self-knowledge. Being acquainted enables one to acquire this kind of self-knowledge, at least normally or typically. But it itself is not identical with this potential self-knowledge (neither partly nor entirely). Then the dispositionalist proposal fails. It rests on confusing the basis for a certain (potential) self-knowledge with the (potential) self-knowledge itself.33 We can conclude that phenomenal consciousness indeed poses a problem for PT and cd-functionalism. Probably, the source of the problem lies in the representational nature of phenomenal states and the externalist nature of representational content. Acquaintance, being such that it makes a world ‘immediately present’ to the mind, is not a mere dispositional matter. It requires the actual standing in certain relations to external things in the world that is constitutive of the relevant representational content.

5 Remaining Roles for Potentials Even if the mind is not exhausted by dispositions, they still have a place and role in our mental life. They are close enough to being necessary. Many mental phenomena involve dispositions and cannot occur without suitable dispositions. Dispositional beliefs (and other dispositional propositional attitudes) are just one important example. Believing a proposition, in the sense of dispositional belief (and not manifest or occurrent belief), is perhaps “nothing more or less than being disposed to do and experience certain things”, as Eric Schwitzgebel proposes.34 Rational capacities are also constituted by dispositions. An inferential capacity to transit from one belief to another consist of a disposition (or a set of dispositions) to form the second belief, on a suitable stimulus (perhaps, considering) under appropriate conditions (being awake, not too distracted, and so on). Accessibility of content is 32

I have tried to argue that we do have this kind of immediate indexical-demonstrative self-knowledge and that it plays an important role, without any attempt at reducing acquaintance with it. Cf. Hofmann (2009). 33 How about the view that acquaintance is just the unspecific, or indeterminate, potential to have some de-re attitude or other about the relevant object? Again, it seems that the proposal ignores the difference between the basis or ground for a disposition and the disposition itself. Being acquainted with x enables one to have de-re attitudes about x, at least normally or typically. But it is not identical with this disposition. 34 Schwitzgebel (2002), 2. Of course, the account cannot be straightforwardly applied to manifest, occurrent belief.—It is perhaps interesting to note that a recent account of so-called ‚cognitive qualia’ (i.e., phenomenal features essentially involved in cognitive states like beliefs) by Terry Horgan and John Tienson is concerned with manifest beliefs only, and does not apply to dispositional beliefs, as Horgan and Tienson explicitly say. Cf. Horgan and Tienson (2002).

Potentialities in the Philosophy of Mind

323

another example. Access-consciousness, as defined by Block (1995), is the immediate accessibility of a state’s content for certain cognitive systems, such as the reasoning system, the language system, and possibly others. (One could distinguish various forms of accessibility, depending on what system or systems one picks out here.) Poisedness, as proposed by Tye (1995), is the immediate tendency or disposition of a non-conceptual representational state to have an impact on the subject’s cognitive system (beliefs and desires, most importantly). The dispositions of being poised is (partly) constitutive of phenomenal states, according to Tye (1995).35 Another example are emotional character traits, such as stubbornness or jealousy. To have such a character trait requires a disposition to undergo certain emotional experiences under certain conditions. And the list could be continued. So there are many mental phenomena which require certain dispositions. Plausibly, therefore, no mind without dispositions—at least no mind as we have it. Very simple minds could perhaps exist without these dispositions. But they would probably lack any interesting form of rationality, and probably also phenomenal consciousness. So little would remain which would justify the ascription of a mind.36 Potential is thus necessary for a mind, or close enough—even though not sufficient. I conclude with two final remarks on dispositions in epistemology and the explanatory power of dispositions. In epistemology, dispositions play a very prominent role. The recent virtue epistemology of Sosa (2007) can serve as an illustration. According to Sosa, knowledge is the manifestation of a competence, and competences are dispositions. So one only knows if one exercises a suitable disposition. Rational or justified belief can be taken to arise in a similar way. (I have already mentioned rational capacities as examples of dispositional items.) So many epistemic properties have a dispositional nature. This fits well with the idea of weakness of will as a failure to exercise a capacity that one possesses.37 There is room for epistemic akrasia, in other words.38

35

This is entirely consistent with our earlier result that phenomenal states require more than dispositions—due to their representational content. Appeal to some kind of availability or accessibility is far-spread, I believe. Bernard Baars’ talk of a ‚global workspace’, for example, can be straightforwardly understood as a disposition of states to ‚feed’ into a large variety of cognitive and behavioral systems. Cf. Baars (1988). 36 In addition, it is not clear what counts as ‚mental’. If it is a connection to phenomenal consciousness, or to practical agency, then minds require dispositions of certain kinds. For, even practical agency requires that the subject apply her rational capacities—at the minimum, a certain disposition to put together information about the world and goals or preferences in an instrumentally rational way. Plausibly, all other kinds of rationality, such as acting for a reason, also require certain rational capacities or dispositions. The kind of rationality I have in mind here is rather process rationality than state rationality. For the distinction between process and state requirements of rationality see the debate between Broome and Kolodny in Kolodny (2005, 2007) and Broome (2007). 37 An analysis of this kind has been given by Michael Smith. Cf. Smith (2003). 38 For an interesting discussion of epistemic akrasia see Owens (2002).

324

F. Hofmann

One could have the capacity to make one’s belief conform with what one takes to be the appropriate belief but fail to exercise it—similarly to the capacity to make one’s intention conform to what one takes to be the appropriate action. In such cases, a normative belief is the (motivating) reason for one’s belief or action intention. Appeal to dispositions as potentials opens up a question of explanatory power. What explanatory power does it have to explain a phenomenon by the invocation of a disposition? The question is a very general one and is not restricted to the use of dispositions in the philosophy of mind. Therefore, I do not treat it here.39 It must be resolved in general metaphysics and general epistemology. If explanation by dispositions in general is alright (as I believe it is) then it is alright in the philosophy of mind, too.40

References Baars, B. (1988). A cognitive theory of consciousness. Cambridge: Cambridge University Press. Bigelow, J. (1988). The reality of numbers. Oxford: Oxford University Press. Bird, A. (2010). Causation and the manifestation of powers. In A. Marmodoro (Ed.), The metaphysics of powers (pp. 160–168). London: Routledge. Block, N. (1978). Troubles with functionalism. In N. Block (Ed.), Readings in the philosophy of psychology (Vol. 1, pp. 268–305). Cambridge, MA: Harvard University Press. Block, N. (1980). Introduction: What is functionalism? In N. Block (Ed.), Readings in the philosophy of psychology (Vol. 1, pp. 171–184). Cambridge, MA: Harvard University Press. Block, N. (1995). On a confusion about a function of consciousness. Brain and Behavioral Sciences, 18(2), 227–247. Boghossian, P. (1989). The rule-following considerations. Mind, 98, 507–549. Broome, J. (2007). Wide or narrow scope? Mind, 116, 359–370. Burge, T. (2010). Origins of objectivity. Oxford: Oxford University Press. Cartwright, N. (1989). Nature’s capacities and their measurement. Oxford: Clarendon Press. Copp, D. (2007). Moral naturalism and three grades of normativity. In D. Copp. Morality in a natural world. Selected essays in metaethics (pp. 249–282). Cambridge, MA: Cambridge University Press. Dretske, F. (1995). Naturalizing the mind. Cambridge, MA: MIT Press. Fara, M. (2006). Dispositions, Stanford Encyclopedia of Philosophy. http://plato.stanford.edu/ entries/dispositions/. Handfield, T., & Bird, A. (2008). Rules, dispositions and finks. Philosophical Studies, 140, 285–298. Heil, J., & Martin, C.B. (1998). Rules and powers. Philosophical Perspectives, 12, 283–312. Hilbert, D. (1987). Color and color perception. Stanford: CSLI-Publications. Hofmann, F. (2009). Introspective self-knowledge of experience and evidence. Erkenntnis, 71(1), 19–34. Hofmann, F. (2013). Epistemic value and virtues. In T. Henning & D. Schweikard (Eds.), Knowledge, virtue, and action (pp. 119–139). Oxford: Routledge.

One can find some useful insights about explanation and, in particular, explanation by dispositions in Jenkins (2008). 40 Many thanks to Kristina Engelhard, Siegfried Jaag and Thomas Grundmann for helpful comments and discussions. 39

Potentialities in the Philosophy of Mind

325

Horgan, T., & Tienson, J. (2002). The intentionality of phenomenology and the phenomenology of intentionality. In D. Chalmers (Ed.), Philosophy of mind. Oxford: Oxford University Press. Jacobs, J. (2011). Powerful qualities, not pure powers. The Monist, 94(1), 81–102. Jenkins, C. (2008). Romeo, Réné, and the reason why: What explanation is. Proceedings of the Aristotelian Society, 108, 61–84. Kolodny, N. (2005). Why be rational? Mind, 114, 509–563. Kolodny, N. (2007). State or process requirements? Mind, 116, 371–385. Kripke, S. (1982). Wittgenstein on rules and private language. Cambridge, MA: Harvard University Press. Lewis, D. (1980). Mad pain and Martian pain. In D. Lewis. Philosophical papers (Vol. 1, pp. 122–132). Oxford: Oxford University Press. McKitrick, J. (2003). A case for extrinsic dispositions. Australasian Journal of Philosophy, 81(2), 155–174. Millikan, R. G. (1993). Truth rules, hoverflies, and the Kripke-Wittgenstein paradox. In R. G. Millikan (Ed.), White queen psychology and other essays for Alice (pp. 211–239). Cambridge, MA: MIT Press. Millikan, R. G. (2004). Varieties of meaning. Cambridge, MA.: MIT Press. Moore, G. E. (1903). The refutation of idealism. In G. E. Moore. Philosophical studies (pp. 1–30). London: Routledge. Owens, D. (2002). Epistemic acrasia. The Monist, 85(3), 381–397. Putnam, H. (1975), Philosophy and our mental life. In H. Putnam. Philosophical papers vol. 2: Mind, language, and reality (pp. 291–303). Cambridge, MA: Cambridge University Press. Schwitzgebel, E. (2002). A phenomenal, dispositional account of belief. Nous, 36(2), 249–275. Smith, M. (2003). Rational capacities, or how to distinguish recklessness, weakness, and compulsion. In S. Stroud & C. Tapolet (Eds.), Weakness of will and practical irrationality. Oxford: Clarendon Press. Sober, E. (1990). Putting the function back into functionalism. In W. Lycan (Ed.), Mind and cognition (pp. 63–70). Oxford: Blackwell. Sosa, E. (2007). A virtue epistemology (Vol. 1). Oxford: Oxford University Press. Thomson, J. J. (1997). The right and the good. Journal of Philosophy, 94(6), 273–298. Tye, M. (1995). Ten Problems of Consciousness. Cambrigde, MA: MIT Press. Tye, M. (2000). Consciousness, color, and content. Cambridge, MA: MIT Press. Tye, M. (2009). Consciousness revisited. Cambridge, MA: MIT Press. Wittgenstein, L. (1984). Philosophische Untersuchungen. Frankfurt/M.: Suhrkamp. Zangwill, N. (2005). The normativity of the mental. Philosophical Explorations, 8(1), 1–19. Zangwill, N. (2009). Normativity and the metaphysics of mind. Australasian Journal of Philosophy, 88(1), 1–19.

Author Biography Frank Hofmann is Professor of Systematic Philosophy at the University of Luxembourg since 2011. He did his Ph.D. and habilitation at the University of Tübingen, Germany. His areas of specialization include epistemology, the theory of rationality and reasons, philosophy of mind, metaphysics, philosophy of science, and philosophy of language. Ethics, meta-ethics, and action theory belong to his areas of competence. Among his publications are “Die Metaphysik der Tatsachen” (mentis 2007), “Natur und Begriff des Bewußtseins” (mentis 2002), and “Gettier for justification” (in: Episteme 2002).

Potentiality in Bioethics Marco Stier

1 Introduction The concept of ‘potentiality’ plays a central role in bioethics. However, when an ethicist uses this term he has something quite different in mind than a biologist using the same word. The latter gives it a physical1 sense while the former a metaphysical one. Commonly, when potentiality comes into play in bioethics the moral concept of a person is at stake, too. To be sure, there are many more bioethical contexts in which the word “potential” plays some role or another, e.g. the potential outcome of a therapy, the potential risk of an innovative technique, and so forth. The different modes of use of the term have to be understood on two levels, a more general one and a specific one. The specific usage of ‘potentiality’ in the sense of ‘having a potential’ to develop into a viable human being is the subject of the present article. Potentiality gets its importance in bioethics from the normative weight of personhood. Since persons have a specific high moral status that includes aspects such as a right not to be interfered with or a right to life, a developing entity that will be a person in the future already has the same moral status now. This is the central idea. Yet, at this point opinions already differ: does an embryo develop into a person, or is it a person right from the beginning of its existence? While the concept of potentiality is as contested in philosophy as the concept of a person is, their amalgamation in ethical contexts makes the discussion even more problematic and ramified. This is why John P. Lizza assumes that “the disagreement over the 1 There are many different sorts of ‘potentials’ in the natural sciences that cannot be discussed in detail here. It may suffice to point to the general difference between physics and metaphysics.

M. Stier (&) Institute for Medical Ethics, History and Philosophy of Medicine, University of Münster, Münster, Germany e-mail: [email protected] © Springer Science+Business Media B.V., part of Springer Nature 2018 K. Engelhard and M. Quante (eds.), Handbook of Potentiality, https://doi.org/10.1007/978-94-024-1287-1_13

327

328

M. Stier

significance of potentiality may stem from a disagreement over the theory of ‘personhood’ invoked in the discussion” (Lizza 2010, 54). Accordingly, the bioethical debate on potentiality has two focal points which are not always easy to keep apart: one regarding the concept itself, another regarding its normative weight. If two authors go along with each other in terms of the meaning of ‘potentiality’, they may completely disagree when it comes to the normative implications of the term. They may also agree regarding the latter, but disagree over the former. They may even agree on both, but nevertheless come to different conclusions in the light of practical social conditions. The main general context in which potentiality is discussed is the moral status of the human embryo. More specifically, this may relate to abortion as well as artificial reproduction, embryo and stem cell research, reproductive cloning, and even prenatal enhancement. Beyond the embryo protection debate, potentiality is sometimes an issue regarding cognitive disabilities, persistent vegetative state and brain death, and human-animal chimeras. In all these contexts ‘potentiality’ is discussed against the backdrop of being—or having central features of—a person. Because the term potentiality and the corresponding arguments nevertheless are almost exclusively discussed within the embryo protection debate I will concentrate on this issue and only touch upon the other problem areas at the end of this article. To begin with I will outline in what meaning ‘potentiality’ is an issue in bioethics. There are different ways to conceive of ‘potentially’, not all of which are relevant in the present context (2.1). The special metaphysical understanding of potentiality gives rise to one of the most famous—and most contested—arguments in bioethics: the argument from potential (2.2). Here, the notion of a ‘person’ is of utmost importance. In fact, the bioethical debate on potentiality cannot be adequately understood without an idea of personhood. One central question in the embryo protection debate is, consequently, in what relation both terms stand to one another. One reading of the argument from potentiality is that embryos are “potential persons” (3.1). This view is subject to a number of meanwhile classical arguments, some of which regard identity relations. Because identity is to some extent a problem of its own that can be understood as a hinge in the potentiality literature, it will be introduced below in a separate paragraph (3.2). An alternative conception of the argument from potential regards human embryos not as potential persons but as “persons with potential” (3.3). Another aspect of the potentiality discussion is the question whether the term has to be understood as an intrinsic quality, or rather as something that depends at least partly on external conditions (3.4). Although potentiality is, as mentioned, almost exclusively discussed with respect to moral rights of the human embryo, there are a few more bioethical contexts in which the term also plays a role. One such context is that of the potentiality and the respective moral rights of people with severe cognitive disabilities (4.1). A second issue concerns people in persistent vegetative state or with brain death (4.2). The third field in which potentiality is discussed is that of the moral status of human-animal chimeras (4.3). These topics will, however, only be

Potentiality in Bioethics

329

touched upon, since the arguments regarding the concept of potentiality, potentiality ascriptions, and their normative consequences are similar to those in the embryo protection debate.

2 Potentiality and the Argument from Potential 2.1

The Concept of Potentiality in Bioethics

Potentiality in the bioethical sense in question does not just refer to everything an entity may become in the future, may turn or be turned into. We have to distinguish at least three general modes of use of the word “potential”. Here are three corresponding illustrations: (1) A bicyclist is a potential crash victim. (2) A piece of wood is a potential chopstick. (3) An embryo2 is a potential adult. The bicyclist in (1) is—as we sometimes say in everyday language—a potential crash victim since biking can be dangerous. There is a certain probability for the bicyclist to get involved in an accident, and this probability is higher in his case than it is in the case of a woman of 85 years who never leaves her room in the retirement home. In example (2) it is not very likely that the piece of wood will indeed become a chopstick, but because this is well possible after all, it can rightly be seen as a potential chopstick. In (3) the probability that the embryo will turn into an adult is not only much greater than the probabilities involved in (1) and (2), but every embryo will become an adult if all goes well. Hence, it can rightly be seen as a potential adult; it is part of the embryo’s nature to develop into an adult person. A key phrase for understanding the crucial difference between example (3) on the one hand and examples (1) and (2) on the other hand is the conditional phrase “if all goes well”. It would not only be strange but plainly wrong to claim that the bicyclist will become a crash victim or the piece of wood a chopstick “if all goes well”. Yet, not everything an embryo or baby might become lies in her nature: When I was a new-born, I certainly had some “potential” to become a drunkard, athlete, or stamp collector. But if I had become one of these kinds of persons, this would not have been due to my intrinsic properties as an atopoietic system, even though there may have been a certain probability for such a development. It would, again, at least be inappropriate to claim that I would have become a drunkard if all had gone well. Thus, in (1) ‘potential’ refers to a certain statistical probability and therefore necessarily to a corresponding possibility. In (2) ‘potential’ refers mainly to the

For brevity, I will use “embryo” in the following to refer to the so called pre-embryo as well as to the fetus. Relevant differences will be emphasized if needed.

2

330

M. Stier

mere possibility; even if the probability of any piece of wood to becoming a chopstick is very low, this is thereby not impossible. Only in (3) does ‘potential’ refer to something specific: Firstly, it is possible in the sense of not being impossible that an embryo will become an adult. Secondly, the embryo will undergo some kind of prefigured development (if all goes well), this is the core meaning of the bioethical concept of potentiality. Thirdly, irrespective of the fact that the early embryo’s (resp. the zygote’s) probability of in fact becoming an adult may not be very high (much can happen between conception and nidation, during childhood and in adolescence) its “program” directs it to adulthood.3 What ‘potentiality’ means in bioethics is—in a first approximation—well described in the words of John Burgess: The potential to be F is, very roughly, the capacity a thing, x, has, strictly in virtue of its current intrinsic properties, to become F given suitable or standard nurturing. Potential is, if you like, the contribution intrinsic properties of x are capable of making to its tendency to become F. The potential of x to become F is its aptitude to become F. (Burgess 2010, 141)

In the sections below we will come back to the concept of potentiality and the various arguments concerning its application as well as to the debate on the concept itself. Expressions such as “if all goes well”, “autopoietic system”, or “intrinsic properties” point to the distinction between active and passive potentiality, that is critical for those authors who follow Aristotle. According to Aristotle, to whom this distinction goes back, an active potential is present “in the cases in which the source of the becoming is in the very thing which comes to be”, and he continues: “a thing is potentially all those things which it will be of itself if nothing external hinders it”. In contrast, an entity has a mere passive potential if something from outside has to come in addition. The thing in question “needs another motive principle, just as earth is not yet potentially a statue (for it must first change in order to become brass)” (Aristotle 2002, IX, 7, 1049a). Using the examples from above: the piece of wood only has a passive potential to become a chopstick (or a match, a candlestick, or whatever), because the principle of change has to come from outside. The same is true for the bicyclist. The embryo, however, will develop into a toddler, a child, and later into an adult all by itself; “nothing” has to be added to make it an adult except the appropriate environment of a functioning female pregnant body and postnatally nutritive and protective external conditions. The embryo’s own functional principle will make sure that it develops in a certain direction.4 What this means will become clearer in the following, particularly in Sect. 3.3.

3

It is nevertheless possible, even if not very common, to make an argument from probability as Noonan did: “To make a distinction between the rights of spermatozoa and the rights of the fertilized ovum is to respond to an enormous shift in possibilities.” (Noonan 1970, 56) 4 For a detailed discussion of the metaphysics of potentiality see part one and two of this volume.

Potentiality in Bioethics

2.2

331

The Argument from Potential

The reason why potentiality is an issue in bioethics at all is that it functions as a criterion for determining an entity’s moral status. Usually, the kind of entity referred to in this way is the human embryo, but in principle potentiality may also apply to other beings such as great apes, highly developed robots, or angels.5 But the mere fact that an entity has a developmental potential of some sort is—in the eyes of many—not enough, however. What has to come in addition is the notion of ‘personhood’. Therefore, potentiality, personhood and moral status almost always come as a triad in bioethics, at least as far as a full moral status is at stake. Respective arguments appeared together with the rise of academic bioethics as such in the early 1970s. It was John T. Noonan who first referred to potentiality when he characterized the theological position as recognizing “a being […] as man because he had man’s potential” (Noonan 1970). While he at that time wrote about the potential to be human, it was Michael Tooley who first coined the term “potentiality principle” and applied it to certain properties that more liberal oriented authors usually see realized only in persons in the sense of ‘adult human being’.6 Tooley formulated the associated argument in the following way: The conservative’s defense will rest upon the following two claims: first, that there is a property, even if one is unable to specify what it is, that (i) is possessed by adult humans, and (ii) endows any organism possessing it with a serious right to life. Second, that if there are properties which satisfy (i) and (ii) above, at least one of those properties will be such that any organism potentially possessing that property has a serious right to life even now, simply by virtue of that potentiality, where an organism possesses a property potentially if it will come to have that property in the normal course of its development. The second claim— which I shall refer to as the potentiality principle—Is critical to the conservative’s defense. Because of it he is able to defend his position without deciding what properties a thing must possess in order to have a right to life. It is enough to know that adult members of Homo sapiens do have such a right. For then one can conclude that any organism which belongs to the species Homo sapiens, from a zygote on, must also have a right to life by virtue of the potentiality principle. (Tooley 1972, 55 italics original)

It should be noted that the alleged conservative does not have to specify the rights-bestowing property; it is enough that human adults have it (Tooley 1972), or—to put it another way—that there is a group of objects whose members typically have that (unspecified) property and that embryos are sufficiently similar to members of that group (Jokic 1993). From this principle, the “argument from

5

Apart from potentiality, there are various other accounts of moral status. One, for example, is what Robert Streiffer subsumes as anthropocentric views, according to which normal human adults have their particular moral status just because of their membership in the species homo sapiens. A discussion of the notion of “moral status”, its justifications and ramifications is beyond the scope of this article, see instead Jaworska and Tannenbaum (2013), Lee and George (2008), Singer (1993), or Steinbock (2007a). 6 Other animals such as great apes may be persons too. I will spare this issue here.

332

M. Stier

potential” (PA) soon emerged. It is usually cited in a version based on a formulation by Peter Singer in his 1979s Practical Ethics (Singer 1993, 152): (i) All innocent7 persons have a moral right to life. (ii) Since all innocent persons have a moral right to life, all potential innocent persons also have a moral right to life. (iii) The human fetus is a potential innocent person. Conclusion: The human fetus has a moral right to life.8 Formulated this way the argument implies that the embryo does not have the relevant rights-bestowing properties yet, since it is not yet a person, but will or might be in the future. This raises the question of whether embryos in fact are mere potential persons, and it is here that the argument forks into two quite different versions: in the formulation above the embryo is not yet a person but has a natural, intrinsic potential to become one. On the second reading—which is advocated by defenders as well as some critics9 of embryo rights—the embryo already is a person. In this light, two main versions of PA have to be distinguished (cf. Stier and Schöne-Seifert 2013): Potential-Person-Account (PPA): embryos are potential persons/human subjects and therefore to be treated as if they were such. Person-with-Potential-Account (PwPA): embryos are persons/human subjects and therefore to be treated as such. According to PPA embryos are not persons but—because of their potential—have to be treated as if they were such; according to PwPA embryos indeed are persons in the first place—because of their potential. The embryo is “a thinking and conscious being”, even if “not a fully developed one” (Gómez-Lobo 2004, 205). While in both readings of the argument “the capacity to develop morally significant characteristics is itself morally significant” (Buckle 1988, 234) there is a difference in emphasis. In PPA the embryo’s moral rights are deduced from the capacity of having person characteristics later. In contrast, according to PwPA, it is literally the capacity itself that establishes those rights. In other words: PPA transfers the moral status of the embryo from its future “backwards” into the present; whereas in PwPA this is not necessary, because having the capacity now is crucial. In PPA the “main actor” is personhood, and the capacity is a sort of “vehicle”. For the advocate of PwPA the capacity is not just a vehicle but the true main actor. This is why central arguments against the first version do not apply to the second one. The fact that This refers to circumstances where a person is not in a strict sense “innocent”, such as in self-defence or at war. 8 Singer uses the phrases “innocent human being” and “potential innocent human being”. The formulation above is from Manninen (2007). 9 Critics who are willing to accept that embryos are persons, like Thomson (1971) and Wertheimer (1971), usually invoke social arguments in favor of women’s right to abortion, but that is not the topic here. 7

Potentiality in Bioethics

333

opponents of embryo rights most often attack PPA while its proponents in fact hold PwPA has been commented by Burgess with the ironic remark that “[t]he argument, as understood by proponents, and the argument, as understood by critics, have been like ships that pass in the night” (Burgess 2010, 140). Both readings of PA are, nevertheless, being debated up to the present. Getting back to the above cited formulation of PA, premise (i) is acceptable for most authors, whereas premises (ii) and (iii) are regularly put into doubt. According to (ii) the right to life follows from the embryo’s potentiality: but why should that be? Critics of PA bring a bunch of meanwhile classical arguments against this claim to the fore. Premise (iii) may be refuted, too, since even if (ii) is true, it is not self-evident that fetuses really are potential persons, or—to avoid the much-debated term ‘person’—do in fact have the rights-bestowing properties potentially.

3 Potentiality and Personhood 3.1

Potential Persons

Human adults surely have moral rights, and human embryos have the potential to become adults. That tells us nothing about the reason why the embryo should have the adult’s rights. The acceptance of this transfer of rights, so the critics, leads to inconsistencies in similar cases, if not to complete absurdity. There are two main arguments against PA, let’s call the first one the logical point against potentiality and the second one the absurd extension argument. Defenders of PA have an answer to each, as we will see later. The logical point about potentiality: Joel Feinberg coined this phrase to name the problem of how to deduce an actual right from the merely potential qualification to this right. According to him, actual rights can be deduced only from actual qualifications, not from potential ones (Feinberg 1992, 49). Stanley Benn put the same rational in his classic remark that a potential president of the United States is not on that account Commander-in-Chief. If satisfying some condition P is a necessary prerequisite for having rights, and if a being could satisfy P at some time in the future, so Benn adds, then “it only follows, that he will have rights when he has P” (Benn 1973, 102). This argumentative failure has often been illustrated with similar analogies: a prince is the potential king but he does not have the king’s rights yet, and the aspirant for a driver’s license is not yet allowed to drive a car, et cetera. As this argument is mainly directed against PPA, defenders of the principle in form of PwPA can rightly stress, that it does not meet the point, because the embryo already is a person with potential, and not the other way around (see the section after the next). The defender’s second reply to the logical point refers to the concept of a potential as such: potentiality in the rights-bestowing sense means that a being has an intrinsic power to develop. The above phrase was “if all goes well”. Admittedly, one may say that the prince will become king if all goes well, but his potential is

334

M. Stier

nevertheless only a passive one, his source of becoming is not, in Aristotle’s words, “in the very thing which comes to be”. The same is true for the potential president and the aspirant for a driver’s license, but not so for the embryo. Analogies like these just don’t work against PwPA in the eyes of its defenders. Absurd extension arguments: Another line of attack is to take PA for granted, assuming the embryo is indeed a potential person with respective rights, and demonstrate the counterintuitive consequences. The argument goes like this: Let Z be a kind of being with moral rights. Because Y is a potential Z, it has in virtue of this potential even now the same moral rights as Z. This is the assertion. Critics of this claim wonder why to stop here: if X is a potential Y, and Y a potential Z, then X, too, has the same moral rights as Z—you may extend this argument backwards as far as you like. In terms of embryo protection the counterintuitive consequence of this rationale is that a particular sperm and a particular ovum, considered jointly, would have to count as a potential person. They surely have an intrinsic potential to develop; unlike the prince they will develop by means of their own intrinsic properties and therefore already have to be seen and treated as a person (Kuhse and Singer 1982; McGinn 1992; Sumner 1981; Warren 1977).10 But, of course, nobody is willing to bestow sperm and ovum with a moral right to life, so we should not do it in the case of the embryo either. This version of the argument, however, can be countered by the logic of identity relations, as we will see in the next section. While the argument from absurd extension is, in a way, already historic, on the background of actual scientific insights it has recently been revitalized and even been made stronger: It is now possible to derive induced pluripotent stem cells (iPS-cells) from simple cells of the body, the skin for instance. These iPS-cells can in turn be brought to develop into a new organism. If we accept PA, we would have to be willing to give skin cells moral rights (Charo 2001; Baertschi and Mauron 2010; Devolder and Harris 2007; Stier and Schöne-Seifert 2013; Testa et al. 2007). This would even be an absurdest extension of PA.11 Yet, one may take the view that it is not the extension that is absurd but the whole argument. If it were valid, the DNA molecules which combined to form sperm or ovum would be potential persons as well (Annis 1984, 156). Anyhow, Annis’ reply does not necessarily show the absurdity of the extension argument, for the anti-potentialist may now say: “Look, PA is even absurder than we first thought!”

Hare asked as early as 1975 “why [it is] supposed to make a difference that the genetic material which causes the production of the future child and adult is in two different places?” (Hare 1975) While the argument is usually applied to refute the potentiality principle, Hare himself endorses it. According to him contraception just as abortion is prima facie morally wrong even though there are a lot of moral reasons that override the wrongness of preventing the embryo’s (and the gametes’) potential from actualizing. 11 For an extended discussion of this point see Stier and Schöne-Seifert (2013) and respective commentaries in the American Journal of Bioethics vol. 13, issue 1. 10

Potentiality in Bioethics

3.2

335

Problems of Identity

To say of a certain X that it is a potential Y presupposes an appropriate identity relation between X and Y. The reason why this is so can be illustrated with an example by Jim Stone. A house is a potential pile of ashes, he notes, but it must cease to exist to produce them. What we need is identity between X and Y, and the house—for it ceases to exist—is not identical to the later ashes (Stone 1987, 818). Without a restriction of potentiality through an identity criterion, thereby excluding cases in which an entity of one kind would have the “potentiality” to become an entity of another kind, we end up with “the proposition that everything is potentially everything else, and thus the destruction of all utility in the concept of potentiality” (Feinberg 1992, 48). Setting aside for a moment which kind of identity relation must hold between two entities in this context, numerical or personal identity or something else, at least the identity of the zygote with the later adult has been doubted. An objection is frequently made from the fact that the zygote can divide into two or more embryos within the first 14 days of development (twinning): Since identity is a one-to-one relation it is not possible to indicate whether the zygote really is identical to the later adult (DeGrazia 2006; Green 2001, 29; Singer 1993, 157; Sagan and Singer 2007). If it indeed divides, we have a one-to-two (or even one-to-many) relation: If the zygote is identical to the later embryo A and at the same time identical to embryo B (A’s twin), for reasons of logic A and B would have to count as identical, too. Since A and B are clearly not identical because of their spatiotemporal distinctness, the zygote cannot be identical to either A or B. From this it follows, that at least the zygote is not a potential person and, hence, cannot have the same moral rights.12 On the other hand, defenders of PA can use the same kind of argument to refute the above presented extension argument with regard to sperm and ovum. Since they cannot both be identical to the embryo, neither the one nor the other is identical to it, and therefor neither of them has the needed potentiality to be a person.13 At first glance, the defender of potentiality as a morally relevant notion seems to be left with a dilemma: either he accepts the identity criterion and runs the risk of no

12

Two further versions of the identity argument can only be mentioned here: The single cells of an early embryo (blastomeres) up to its eight-cell stage are each totipotent, i.e. they can each develop into a whole organism. Therefore, numerical identity is once more excluded because of the one-many-one relation. A further reason to contest numerical identity at this stage is the fact that the newborn does not develop from the zygote, but from “a part of a part of a part” of it. In the course of blastulation—the formation of the blastocyst (which consists of trophoblast and embryoblast)—and the further development, several fissions happen. Stone refutes this argument with the following claim: “If I fission into twin offshoots I cease to exist. But if I grow a large tumor that weighs as much as the rest of me, I survive.” (Stone 1994, 291) 13 The opponent of PA uses the logic of identity in a fission case (twinning), the proponent uses the same logic for his purposes in a fusion case. The rationale behind the two arguments is the same.

336

M. Stier

longer being able to protect the early embryo, or else he drops the restriction of potentiality through identity and is thereby able to explain the potential of the zygote, but now has to explain why sperm and ovum are no potential person(s). Nevertheless, he can try to escape this dilemma in three ways: He can bite the bullet and admit that the potential is present only after nidation when the embryo can no longer split. Norman Ford argues that “it would be hard to admit the presence of an individual human being in the zygote if, in principle, it could lose its ontological identity whenever twinning occurs in the course of development” (Ford 1988, 112). The other possibility is to refute the logic of the argument, the third way is to base potentiality on an alternative conception of identity. There are several ways to refute the logic of the argument (cf. Lee and George 2008, 123). (a) In the usual course of development no twinning occurs; twinning is a special case that does not count against the potentiality principle. (b) Secondly, one may claim that to identify the zygote with A but not with B (or vice versa) is not ruled out. (c) Even if one admits that the zygote is neither identical to A nor B, that does not mean that the first is not an individual worth protecting for its own sake. (d) The zygote could be seen as an amalgam of the later A and B (Koch-Hershenov 2006). (e) One can also hold with the President’s Council the view that the moral status of the early embryo does not diminish or vanish through twinning but even increases (President’s Council on Bioethics 2002, 176f.). There may be more attempts to cope with that problem. At any rate, most of them come at the cost of weakening the identity relation and thereby the restriction of PA to the embryo. While (a) is certainly true, (b) means simply to ignore the logic of identity. What is more, this “solution” to the identity problem turns out to be a blind lane for the defender of PA: If he accepts that the zygote can be identical to only one of the twins, he must be able to explain why it should not be possible that it is identical to only one of the gametes. Here again, we are faced with a dilemma. Apart from the problem that the respective gamete would have to count as a (potential) person, this would have quite absurd implications: If I am really identical to the respective sperm, then I once penetrated an egg; and if I am identical not to the sperm but to the egg—because it is so much bigger—I was once penetrated by a sperm (Stone 1987). The answer (c) might be true, but if it is, the zygote’s worthiness does not result from its potentiality because this is excluded by the missing identity (see above). There is a possible objection to (d), too: Since we can never be sure whether a zygote is already an amalgam of a pair of twins, and since both these twins would already have a right to life, if it is, we would be obliged to make sure that both twins in fact come into existence. Each and every zygote would have to be treated preventively as twins. Moreover, since we cannot be sure in advance whether the zygote is not an amalgam of triplets, quadruplets and so on, we would be obliged to treat each zygote as triplets, quadruplets (or quintuplets, or …) and possibly even make sure that all these potential persons come into existence. It could be argued— against this outlandish consequence—that it is the potential as such that counts, and

Potentiality in Bioethics

337

not the potential to be a certain number of possible adults. But if the zygote in fact is an amalgam of, say, three individuals, and since the aim of the amalgam-argument is to preserve identity, there would in fact be three individuals and three potentialities. If we have an obligation to support a single “zygote-person” whose coming into existence is in danger to fail, we have an even stronger obligation if there are three or more “zygote-persons” whose coming into existence is in danger. Imagine there is a house burning and you are a firefighter. Would it be appropriate assuming there is only one person to rescue? The liberal’s answer to (e) could be a combination of the replies to (a) to (d).

3.3

Persons with Potential

Counterintuitive thought experiments like those above may be futile if the whole argument have been misunderstood by the critics of PA. PPA my turn out to be nothing but a “straw man” (Burgess 2010, 145f; Stone 1987, 818), since what the potentialist has in mind is something quite different. As we have seen in the beginning, ‘potential’ can have different meanings. In particular, the defender of PA has to exclude cases in which any entity X is “potentially” a very different thing Y, as in Stone’s example of the burned down house. In particular, the defender of PA and resulting embryo rights must be able to refute the extension arguments (gametes or even somatic cells as potential persons) while keeping PA’s force with regard to the embryo. The whole argumentation for embryos as “persons with potentials” (PwPA) can be reconstructed in four steps: The first step is to differentiate between a weak and a strong form of potentiality. With the second step the concept of a sortal is introduced. The third step (which is the most complicated one) consists in coping with the identity problems. Step four reintroduces the concept of a person, thereby completing the “persons with potential account”.14 Step one: One strategy in favor of the potentiality of the embryo and against other “run-of-the-mill” versions of a potential is to differentiate between weak and strong. The core of the weak-strong distinction is the “natural course of things” (Jacquette 2001, 84). According to this understanding, formulated similarly by a number of authors, the zygote is the “beginning of the spatio-temporal-causal chain of the physical object that is a human body” (Wertheimer 1971, 79). According to David Annis the strong potential of the embryo consists in “the thrust of a causal mechanism’s inhering” it (Annis 1984, 157). The same differentiation can be found in Stone’s writing: what grounds the right to life is not the embryo’s future properties, but rather “that the particular entity in question has a nature sufficient to be the primary determinant of its following a path which leads to the adult stage of

14

These four steps are an idealization of different accounts that are quite similar all in all, even if they diverge in detail.

338

M. Stier

members of its kind” (Stone 1987, 819).15 With the reference to the “beginning of the […] human body” and its “nature” both the extension argument and the logical point are meant to be countered. On the one hand it is surely not arbitrary to set the zygote as the beginning of the human body and to connect a particular nature to it. On the other hand it is nonetheless reasonable to ask why specifically the zygote should mark the starting point, because it is not much less plausible to let the human body begin with a particular pair of gametes: The zygote is one single cell that is not very similar to a body or organism, and the pair of gametes is just one cell more and probably not less similar to an organism. What is needed, therefore, is a further argument for the zygote as the human body’s beginning. If it is not completely unreasonable to see the gametes as the causal starting point, it would not be unreasonable either to let the specific intrinsic nature begin with them. At least there is no physical or logical necessity to prefer the one over the other. Step two: At this point the metaphysical concept of a sortal gets introduced and differentiated: Describing the embryo as something that has a potential to become a person in the future (as in PPA) uses ‘person’ as a phase sortal. In contrast, proponents of PwPA see the embryo’s potential as an indicator for the kind of being it already is, namely a person, where the latter is understood as a substance sortal.16 The crucial difference is that an entity can fall under a phase sortal only temporarily, but it will fall under a substance sortal throughout its entire existence. Examples for phase sortal concepts are “newborn”, “child”, “adolescent”, “philosopher”, “drunkard” and many more. I was not born a philosopher, I became one, and if I should ever cease to be a philosopher, I will not cease to exist for this reason. What I have been from the very first moment of my life is a human being, and I will continue to be one until my dying breath. With this in mind, one can—as Michael Burke argues—refute the extension argument: Animals like human beings are essentially animals, and nonanimals are essentially nonanimals. When a human person develops from an ordinary body cell, the body cell does not survive. No entity can turn from one essential kind of being (nonanimal) into another one (animal) without ceasing to exist. The same is true for the gametes. Simply put, the body cell ceases to exist, and no entity can become another entity after having gone out of existence. Above that, the end of the existence of something stops all diachronic numerical identity, and without identity no potentiality (Burke 1996, 505). 15

There are more, similar distinctions on the philosophical market: Brown (2007) distinguishes between a “First Order Potential” (the capacity to acquire a capacity), a “Second Order Potential” (the capacity to acquire First Order Potential), and a “Third Order Potential” (the capacity to acquire Second Order Potential). Covey (1991) speaks of “proximate” and “remote” potentials. Another distinction is made between a “capacity in hand”, e.g. to speak Spanish here and now, and the “natural potentiality” to develop a capacity (cf. Eberl and Ballard 2009, 475). A distinction regarding the consequences of a potential comes from Feinberg: According to the “strict potentiality criterion” a creature that has certain higher characteristics either actually or potentially is a moral person now; the “gradualist potentiality criterion”, by contrast, gives such a being “not a right, but only a claim, to life” (Feinberg 1992, 47ff). 16 For more on sortals see Grandy (2014).

Potentiality in Bioethics

339

Now we are left with “strong potentiality” and the understanding of ‘person’ as a substance sortal. But even that does not seem enough to ensure the zygote’s right to life. If a being cannot survive fission, and the zygote undergoes fission either by twinning or in the morula stage, then it ceases to exist and the later embryo will be a different being. The identity of the zygote with the later embryo, child and adult is still but a claim, without a supporting argument. Step three: The account of the embryos as falling under the substance sortal ‘human being’ (or ‘person’), or as having a particular intrinsic nature—what amounts to quite the same—leaves open whether this already applies to the zygote or not. As we have already seen, this open question results from dissent on whether and in what sense there is an identity relation between the zygote and the later child. Even if the later embryo has a strong potential that is due to its falling under the sortal ‘human being’ (or ‘person’), and even if this means that it has a strong right to life, it is not thereby already guaranteed that the zygote, from which the later embryo developed, has the same strong potential and that it, too, falls under the respective sortal. If not, the zygote probably doesn’t have a strong right to life even though the embryo after 14 days has it. The question whether the very early embryo already has rights is of particular importance for stem cell research. A first reaction to this challenge is to simply deny the identity of both, zygote and later embryo. According to Jim Stone, “[z]ygotes cannot grow up”, because “no substance can survive fissioning into duplicate substances” (Stone 1994, 291). For Stone things change but after the first cleavage about twenty-four hours after conception, when an organism is present (Stone 1987, 819). The follow-up question is, though, whether the embryo in the two, four or eight-cell-stage really constitutes an organism in the narrow sense, or whether this structure should still be seen as a “community” of individual substances (see footnote 12 above).17 In the latter case the entity with the specific nature, falling under the relevant substance sortal, will be present not before the embryo proper begins to develop. The first is what Burke holds, but he goes even further back, contra Stone: already the zygote is viewed as numerically identical to the later child. Stone would be right if the two cells resulting from the first cleavage would develop independently in the way the two offsprings of an amoeba go separate ways. But since all cells stemming from the zygote usually develop together, the zygote is actually a one-cell organism that will develop into a two-cell organism, and so forth (Burke 1996, 502). While Stone holds, together with many others, that zygote and later embryo cannot be numerically identical since both are different substances, Burke’s claim works the other way around: both are numerically identical because they are part of the same substance. According to Stone’s account the zygote can neither be a potential person nor a person with potential for the lack of numerical identity. Burke’s argument allows for both.

Brown (2007, 608) comments this ironically: “I can no more ask whether I am a two-cell, four-cell or eight-cell embryo than I could ask if I am my second grade class in elementary school.”

17

340

M. Stier

If the zygote falls under the same substance sortal as the later adult does, both must share its persistence conditions. Mark T. Brown doubts that they do (Brown 2007). First, it is not obvious that the zygote is a biological individual at all: This doubt does not so much result from the fact that it may split into two individual beings through monozygotic twinning, but rather from the fact, that it can divide. If this is true, the zygote cannot be identical to the later embryo or adult, because on the left-hand side of the equation there is no individual at all to be preserved through identity. It is not the problematic one-to-many relation discussed above that excludes identity here, but rather the very ability to divide. Second, even if this is not sufficient to deny the zygote’s individuality, it nevertheless cannot fall under the same substance sortal because it has different persistence conditions: Human beings cannot divide into two independent living entities, zygotes can. Hence, the persistence conditions of zygotes and adults are different. This implies that they cannot fall under the same substance sortal, because everything that falls under the same substance sortal has identical essential properties, a part of which is the persistence conditions. With this we are confronted with a “constructive dilemma” (Brown 2007, 605ff.): Either the zygote is not an individual, then it cannot have a person-potential,18 or it is an individual, then it has different persistence conditions and cannot have a person-potential either. A way out of this dilemma is proposed by David Oderberg. He tries to show that zygote and later embryo are not numerically identical but nonetheless fall under the same substance sortal, so that already the zygote has the necessary right-bestowing person-potential (Oderberg 1997). The zygote, Oderberg holds, does not fall under a different substance sortal than the later embryo, but it does not constitute a phase sortal either. Instead, he introduces, with reference to David Wiggins, the “constitutive sortal, “a term denoting an object which constitutes another but is not identical with it” (Oderberg 1997, 265). These sortals come in two kinds: constitutive substance sortals and constitutive phase sortals. ‘Lump of bronze’, for instance, is a constitutive substance sortal, since lumps of bronze are substances. ‘Zygote’ in contrast—as well as ‘morula’ and ‘blastocyst’—are constitutive phase sortals. In this sense, the zygote is not the human being itself, but a part of it and therefore not identical to it. A part cannot be identical to the whole it is a part of. Since it is a constitutive phase sortal of the human being (that falls under a substance sortal), it is not completely different either. On the basis of that account Oderberg concludes, that the early immature human being has all the rights a child or an adult has, and “the defense of the zygote and of the embryo must be of paramount importance” (Oderberg 1997, 297). Step four: With all this in mind, we’ve come full circle: The very early embryo already is a human being. Since the reference to species alone can be charged with what Peter Singer termed “speciesism” (Singer 1993, 55ff), the ‘person’ must be added: Following Patrick Lee, the term ‘person’ denotes nothing over and above the

My term “person-potential” is neutral with respect to PPA and PwPA.

18

Potentiality in Bioethics

341

individual human being.19 There is not me on one side and my person on the other. ‘Person’, he explains, refers to human organisms and, therefore, “human persons begin when the physical human organisms begin” (Lee 2004, 258). This is the very reason why one cannot become or cease to be a person at some point within life, for one either is one from the beginning or never. He continues: “The pro-life position is not that unborn human beings are potential persons and therefore have a right to life. Rather, potentiality is important only because it is an indicator of what kind of thing is already present” (Lee 2004, 262). PwPA relies widely on two ideas. The one is the “intrinsic nature” of an entity, the other is its “active potential”. While this seems to be reasonable on the one hand, on the other one should keep in mind what Burgess stresses: “Wrenched from the Aristotelian teleological picture of biology that gives it point and purpose, talk of active potency is nothing more than a metaphor, and it is potentially a highly misleading metaphor.” (Burgess 2010, 141) According to Aristotle, “everything that comes to be moves towards a principle”, and in this teleological sense “man is prior to boy and human being to seed” (Aristotle 2002, IX, 8, 1050a). If one is not prepared to accept that there are entities that not only develop but that are driven by a prefigured final cause, then it is hard to make sense of the notion of an active potential. A further problem of the substance sortal based account has been identified by Jason Morris: Substance based ethics relies on a “strict ontology”. It requires distinguishing between accidental and substance changes. Recent biological insights, however, show that cell types and developmental stages are separated only by subtle changes that do not differ from what strict ontology would classify as mere accidental changes. In the absence of the necessary substantial changes during cell development, strict ontology “relies on a false understanding of human development” and cannot solve the problem of the moral status of the embryo (Morris 2012). Similarly, Ronald Green explains vividly why it is virtually impossible to single out a point or event from which on a new human is present. What we usually call events are “complex and extended processes with many discrete and ever changing components” in biological reality (Green 2002, 21). That’s why we have to decide from which moment in embryological development on we want to see the criteria for a right-bestowing potentiality fulfilled. The defender of embryo potentiality may object that biology is one thing and metaphysics quite another and that it is therefore impossible to argue by reference to the former against the latter. The problem with this reply is, however, that it deprives the potentialist for his part of the possibility to use biology based arguments. He can no longer point to, say, genetic differences between gametes and zygotes or whatever other biological circumstances may pertain.

19 To be sure, it is also possible do regard ‘person’ as a substantial kind of its own (Baker 2005). Such an account does not necessarily lead to a strong rights to life of the early embryo (McMahan 2002).

342

3.4

M. Stier

External Constraints to Potentiality

The debate on potentiality in bioethics has been mainly on the moral status of the embryo against the backdrop of abortion. However, since the advent of artificial reproduction techniques such as in vitro fertilization (IVF) in 1978, not only the potentiality and alleged rights of the embryo in vivo, i.e. in a mother’s womb, are at stake, but also that of the embryo in vitro, i.e. the embryo in the proverbial “petri dish”. One particular ethical problem of IVF is that more embryos can be created than are later implanted. This has the effect that the debate about potentiality and moral status gets extended to fertilized eggs, each of which will have a right to life if one follows the arguments of embryo protectionists. One crucial question, then, is whether a fertilized egg has the same (active) potential in vitro as in vivo. If it has, then the next questions are whether it is at all morally acceptable to create surplus embryos and how to deal with them, once created. The main point—to be sure—is still on embryo rights, but with IVF and stem cell research a number of new questions arose. Is it morally justifiable to perform IVF when it is certain in advance that some embryos will have to be discarded? Is it acceptable to destroy an early human embryo to derive stem cells? If embryos indeed have a strong right to life—or at least a right not to be killed—both questions have to be answered in the negative. Without IVF many couples would not be in the position to have children, and embryo research is associated with high hopes for new therapies for many diseases hitherto untreatable. Against this background, some authors propose to define potentiality extrinsically (cf. Lizza 2010). To give an example, many people consider it morally wrong to throw a flag in the garbage not because the flag gets hurt, but because it is a particularly dignified symbol. Similarly, we show respect to dead human bodies not because there is a danger of hurting them, but rather because they have once been living human organisms and are a symbol of the person who is gone. According to Bonnie Steinbock, embryos deserve respect in the same way— instead of unlimited protection—because “they are a developing form of human life, and also a symbol of human existence” (Steinbock 2007b, 134). Many things and beings get their particular dignity at least partly from such an extrinsic evaluation. One core point of the weak-strong distinction of potentiality is the “natural course of things” as we have seen. If something has the person-potential thanks to its natural developmental features, it may count as a (developing) person. Since the embryo in the laboratory needs a lot of additional external assistance to develop, e.g. implantation into the uterus, its potential is, according to Dawson and Singer, not very different from that of the gametes, considered jointly. Therefore, if sperm and ovum don’t have a right-bestowing potential the embryo in vitro doesn’t have it either (Singer and Dawson 1988). Against this one might argue with the President’s Council that an embryo’s extracorporal creation neither affects its potential nor its moral status (President’s Council on Bioethics 2002, 178). One point of intense debate here is again whether the moral status of a being is to be completely deduced

Potentiality in Bioethics

343

from its intrinsic nature. If this is accepted, external factors can play no role, since the nature of a being remains the same no matter where it happens to be located. A second contested aspect regards the active-passive distinction of potentiality: the opponent of embryo rights sees the embryo’s need for external assistance as evidence for the passivity of its potential, whereas its defender claims that the laboratory embryo just lacks the appropriate environmental conditions to unfold its active potential. Having an active potential, so the latter assumes, does not mean that one is able to develop in each and every circumstance; that would be a misunderstanding of the concept. Steinbock’s account is based on an external evaluation, and Singer and Dawson refer to the natural course of things. A third, partly extrinsic strategy is to take the concept of potentiality seriously but to add an actualizability-condition. This is particularly relevant for IVF and embryo research: No woman would be willing to have all the embryos implanted that could be created in the IVF-process (whether this is possible at all is a different question). Because of this, for the (prospectively) leftover embryos there are only three possibilities: of not being created at all (if IVF were performed or even prohibited), of being discarded, or of being used for research purposes. In none of these cases is their potential actualizable. Ingmar Persson and Julian Savulescu explain the relation of potentiality and actualizability like this: As we conceive of the concept of actualizable potential, the concept of potentiality refers to states internal to the being and the concept of actualizability refers to states external to the being that together cause it to actualize its potential […] Thus, such an embryo in a petri dish lacks actualizable potential to become a person because it is not in an environment that allows it to become one, whereas such an embryo in a normal uterus has this actualizable potential. (Persson and Savulescu 2010)

If the potential—be it even “active” and “intrinsic”—is by no means actualizable, the embryo lacks the relevant moral status. The defender of embryo rights will, to be sure, again counter with the assertion that the moral status does not hinge on any extrinsic conditions because these are irrelevant for the kind of being one is. A quite different external constraint to potentiality is introduced by the creation of embryos with an inbuilt “developmental break”. According to PwPA every intact embryo has a natural capacity to develop due to which it is a person as soon as it comes into existence. Embryos without this capacity, so the idea behind “break” proposing, consequently would not be persons. Because it might seem as morally problematic to deprive an already existing zygote of this potential as it is to kill it, the idea is to “switch off” the ability to further develop beforehand. One method to achieve this is the altered nuclear transfer (ANT). This technique is based on the somatic cell nuclear transfer (SCNT), a cloning method by which the sheep “Dolly” was created in 1997 (Wilmut et al. 1997), and which is meanwhile a standard procedure for cloning farm animals. The crucial difference between SCNT and ANT is that in ANT the nucleus of the somatic cell or the cytoplasm of the oocyte is altered prior to the injection of the nucleus into the egg cell (Hurlbut 2005). Thereby the subsequently created embryo is not able to develop beyond a certain

344

M. Stier

stage. The other constraints mentioned above consist in redefining ‘potentiality’, whereas ANT means in fact altering—better: preventing—the very potential in question. While some embryo protectionists do indeed view this method as a morally acceptable way of making embryos available for research, some critics reject this view. The President’s Council on Bioethics, for instance, warned that the respective embryo can be seen as “destined to become a defective, severely deformed human embryo, the defect and deformity having been deliberately inflicted on it by the scientist” (President’s Council on Bioethics 2005, 39). The crucial question, then, is whether the “product” of ANT is a disabled human embryo or no human embryo at all. According to Hurlbut the entity created through ANT “lacks the integrated unity and potency of a living being”. This is why “pluripotent cells can be produced while not creating an embryonic human being” (Hurlbut 2005, 296). However, others argue that despite its limited developmental capacity the ANT-entity is a human embryo, deserving at least a high degree of respect. Referring to the fact that naturally defective embryos are nevertheless viewed as “embryos”, they insist that “it makes no difference in moral status that an ANT embryo cannot develop to term” (Clausen 2010, 52). In all these cases of external constraints the very concept of ‘potentiality’ is contested. When even an embryo that does not have the slightest chance of developing into an adult is seen as having the relevant rights- or respect-bestowing potential, then the whole argument may turn out to be an argument from species membership, and not one from potentiality.

4 Further Areas of Discussion The term ‘potentiality’ is discussed in bioethics almost exclusively within the scope of the embryo and stem cell debate. But the fact that possible embryo rights are usually not derived from its being human but from its being a (potential) person20 already points to further moral questions. Thus, when it comes to potentiality in bioethics, the concept of a person comes into play one way or the other. With this, all those moral problems show up in which being a person or having capacities like rational self-governance play some role. There have been three areas of debate in which potentiality is at issue. The first regards humans with severe cognitive disabilities; the second is about people in persistent vegetative State (PVS) and brain death; and the third, more extensively discussed compared to the first two issues, deals with human-nonhuman chimeras.

20

See step four in Sect. 3.3.

Potentiality in Bioethics

4.1

345

Cognitive Disabilities and Potentiality

The issue in this case is not so much whether human beings who are severely cognitively disabled from birth in fact have or should have moral rights or not. Rather, the question is why this is so: Being a person is usually conceived of as (potentially) being equipped with certain cognitive resources. Furthermore, moral rights are usually derived from personhood. How, then, can it be that human beings have moral rights, even though they do not have these resources? They may count as persons only in a metaphorical sense. How can it be explained that humans without the typical rational abilities and potentials of persons have rights, while nonhuman animals enjoy hardly any moral or legal protection (McMahan 2008, 2009; see also Wasserman et al. 2013)? Many animals have capacities that are clearly higher than those of the severely cognitively disabled.

4.2

PVS, Brain Death and Potentiality

In contrast to the problem of personhood and moral rights of those who will never develop person capacities there is—regarding the other end of life—a debate on the potential of humans in a persistent vegetative state (PVS) or who are brain-dead. Jason Eberl, for instance, holds with reference to Thomas Aquinas that “PVS patients continue to possess, although they can no longer actualize, an intrinsic, natural, active potentiality for self-conscious rational thought and volition” (Eberl 2011). Alan Shewmon maintains that even the artificially sustained, whole-brain dead organism still has the potential for intellect and will (Shewmon 1997). According to his view, one loses one’s potential only after the whole organism has “deanimated”, where “deanimation” is defined by a cessation of the whole organism that could be reversed not even by any future medical means (Shewmon 2010).

4.3

Human-Animal Chimeras and Potentiality

Since the first successful creation of embryonic chimeras, where cells from a goat and a sheep were fused to what was called a “geep” there has also been a debate on human-animal chimeras. Chimeras already appear in Greek and Egyptian mythology, denoting a being consisting of parts of different species, e.g. lion, snake, and goat. Similarly, a griffin resembles parts of lion and eagle. In actual fact, there are even now lots of human-animal chimeras among us, because as soon as one has some nonhuman material integrated, just think of pig heart valves, one is strictly speaking an interspecies chimera. In contrast, people with a donor kidney, heart, liver or other organs or tissues from another human, are intraspecies chimeras.

346

M. Stier

Regarding human-animal chimeras, one moral question is whether their creation is ethically permissible, in the first place. If this is acceptable, one has to decide whether this includes chimeras that have a person-potential. Should this, too, be morally permissible, then again a third question arises: How are these beings to be treated? Would they have exactly the same moral and legal rights as human persons? If they do, the next question is whether this also includes embryonic animal persons, presupposed that embryos are persons at all (see above). Apart from these more general questions—which have to be skipped here—there has also been a debate on the potentiality and moral status of chimeras.21 On the background of potentiality the question is not in the first place whether human cells or tissues should be engrafted in animal embryos at all. This is a follow-up issue. Rather, the problem is in what regard human-animal chimeras have a person-potential. This question arises in particular when the engrafted material comes from the human brain.22 Because the brain is the biological substrate of our human rational capacities, one worry is that we may create beings that do not fit into any category of species, but yet have those capacities to a morally relevant degree. What does this mean regarding potentiality? Whether human-nonhuman embryos are to be seen as persons can be a relevant question only for those who hold PwPA, the view that human embryos already are persons with a potential. If human embryos are persons due to their intrinsic potentiality, then any other being, whatever its species, has to be seen as a person, too, if only it has this potentiality. From this it follows that “any experimentation that would harm or destroy a human embryo or an a-h [animal-human, MS] chimera of the relevant type is morally problematic or outright impermissible” (Eberl and Ballard 2009, 481). Critics of PA will probably not grant any moral right to chimeric embryos since they do not even regard human embryos as persons. If, however, a chimera is already born and shows person-like capacities, then it would possibly have to have the moral rights of a person. Denying such creatures moral rights could be regarded as a manifestation of the same “species prejudice” as the withholding of moral rights and respect from great apes today (DeGrazia 2007). If one indeed holds such an “anthropocentric account” (cf. Streiffer 2005) no animal can be a person, no matter what its potential may be. At this point, all the problems and quarrels that have been discussed in the major part of this article return: What is the relation of personhood and species membership, of species membership and potentiality, of potentiality and self-consciousness, and so forth? So, we do not need to repeat them. If the argument from potentiality is really—and without qualifications—relevant for the moral status of a being, denying moral rights with regard to any being that has the relevant potential is in need of ethical justification. 21

For the general debate on human-animal chimeras see e.g. DeGrazia (2007), Karpowizc et al. (2005), or Streiffer (2011). 22 It has been discussed whether the origin of the engrafted cells (e.g. human or dolphin) are relevant for the moral classification of chimeras, see for example Piotrowska (2014) and Sagoff (2007).

Potentiality in Bioethics

347

5 Conclusion In bioethics potentiality is important because being a person is important. The major part of the respective debate has been on the moral status of the human embryo. Persons are usually defined as self-conscious, rational beings. The question is, then, in what sense human embryos can be persons if they are neither self-conscious nor rational. The “argument from potential” (PA), which is meant to provide an explanation for this, stands in the focus of a debate that dates back at least to the early 1970s. Yet, it is not only contested whether this argument is valid, but also what it actually means. Critics of PA frequently argue against a reading of PA according to which embryos are “potential persons” (PPA). In contrast, defenders of PA hold that embryos already are “persons with potential”. A further, hotly debated issue in this context is whether and in what sense potentiality presupposes identity. This question is in fact discussed in a great many of bioethical books and articles devoted to normative issues regarding human embryos. The embryo debate comes down to the question of whether the human embryo has a right to life and whether it has to be protected against abortion as well as against being used for research purposes. Beyond that, there are a few less extensively debated bioethical problems. These have, nevertheless, an argumentative core which is similar to that of the embryo protection debate. Here, too, the definitions of ‘potentiality’ and ‘personhood’, and the normative implications following from those definitions, are crucial. All in all, potentiality is, at least indirectly, a through and through normative concept in bioethics.

References Annis, D. B. (1984). Abortion and the potentiality principle. The Southern Journal of Philosophy, 22(2), 155–163. https://doi.org/10.1111/j.2041-6962.1984.tb00332.x. Aristotle, (2002). Metaphysics. available online: http://ebooks.adelaide.edu.au/a/aristotle/ metaphysics/. Baertschi, B., & Mauron, A. (2010). Moral status revisited: The challenge of reversed potency. Bioethics, 24(2), 96–103. https://doi.org/10.1111/j.1467-8519.2008.00686.x. Baker, L. R. (2005). When does a person begin? Social Philosophy and Policy, 22(2), 25–48. Benn, S. (1973). Abortion, infanticide, and respect for persons. In J. Feinberg (Ed.), The problem of abortion (pp. 92–104). Belmont, CA: Wadsworth. Brown, M. T. (2007). The Potential of the human embryo. Journal of Medicine and Philosophy, 32(6), 585–618. https://doi.org/10.1080/03605310701680973. Buckle, S. (1988). Arguing from potential. Bioethics, 2(3), 227–253. https://doi.org/10.1111/j. 1467-8519.1988.tb00050.x. Burgess, J. A. (2010). Potential and foetal value. Journal of Applied Philosophy, 27(2), 140–153. https://doi.org/10.1111/j.1468-5930.2010.00482.x. Burke, M. B. (1996). Sortal essentialism and the potentiality principle. The Review of Metaphysics, 49(3), 491.

348

M. Stier

Charo, R. A. (2001). Every cell is sacred: Logical consequences of the argument from potential in the age of cloning. In P. Lauritzen (Ed.), Cloning and the future of human embryo research (pp. 83–89). New York, Oxford: Oxford University Press. Clausen, J. (2010). Stem cells, nuclear transfer and respect for embryos. Human Reproduction & Genetic Ethics, 16(1), 48–59. https://doi.org/10.1558/hrge.v16i1.48. Covey, E. (1991). Physical possibility and potentiality in ethics. American Philosophical Quarterly, 28(3), 237–244. DeGrazia, D. (2006). Moral status, human identity, and early embryos: A critique of the president’s approach. The Journal of Law, Medicine & Ethics, 34(1), 49–57. https://doi.org/10. 1111/j.1748-720X.2006.00008.x. DeGrazia, D. (2007). Human-animal chimeras: Human dignity, moral status, and species prejudice. Metaphilosophy, 38(2-3), 309–329. Devolder, K., & Harris, J. (2007). The ambiguity of the embryo: Ethical inconsistency in the human embryonic stem cell debate. Metaphilosophy, 38(2–3), 153–169. Eberl, J. T, & Ballard, R. A. (2009). Metaphysical and ethical perspectives on creating animal-human chimeras. Journal of Medicine and Philosophy, 34, 470–86. doi:https://doi.org/ 10.1093/jmp/jhp035. Eberl, J. T. (2011). The unactualized potential of PVS patients. APA Newsletter, 11(1), 14–18. Feinberg, J. (1992). Abortion. In J. Feinberg. Freedom and fulfillment. Philosophical essays (pp. 37–75). Princeton, NJ: Princeton University Press. Ford, N. (1988). When did I begin?: Conception of the human individual in history, Philosophy and Science. Cambridge, New York: Cambridge University Press. Gómez-Lobo, A. (2004). Does respect for embryos entail respect for gametes? Theoretical Medicine and Bioethics, 25(3), 199–208. Grandy, R. E. (2014). Sortals. In E. N. Zalta (Ed.), The Stanford encyclopedia of philosophy (Spring 2014). Available online: http://plato.stanford.edu/archives/spr2014/entries/sortals/. Green, R. M. (2001). The human embryo research debates: Bioethics in the vortex of controversy. Oxford, New York: Oxford University Press. Green, R. M. (2002). Stem cell research: A target article collection: Part III—determining moral status. The American Journal of Bioethics, 2(1), 20–30. https://doi.org/10.1162/15265160 2317267790. Hare, R. M. (1975). Abortion and the golden rule. Philosophy and Public Affairs, 4(3), 201–222. Hurlbut, W. B. (2005). Altered nuclear transfer as a morally acceptable means for the procurement of human embryonic stem cells. Perspectives in Biology and Medicine, 48(2), 211–228. https:// doi.org/10.1353/pbm.2005.0055. Jacquette, D. (2001). Two kinds of potentiality: A critique of McGinn on the ethics of abortion. Journal of Applied Philosophy, 18(1), 79–89. Jaworska, A., & Tannenbaum, J. (2013). The grounds of moral status. In E. N. Zalta (Ed.), The stanford encyclopedia of philosophy (Summer 2013). Available online: http://plato.stanford. edu/archives/sum2013/entries/grounds-moral-status/. Jokic, A. (1993). Why potentiality cannot matter. Journal of Social Philosophy, 24(3), 177–193. Karpowicz, P., Cohen, C. B., & van der Kooy, D. (2005). Developing human-nonhuman chimeras in human stem cell research: Ethical issues and boundaries. Kennedy Institute of Ethics Journal, 15(2), 107–134. Koch-Hershenov, R. (2006). Totipotency, twinning, and ensoulment at fertilization. The Journal of Medicine and Philosophy, 31(2), 139–164. https://doi.org/10.1080/03605310600588673. Kuhse, H., & Singer, P. (1982). The moral status of the embryo. In W. A. W. Walters & P. Singer (Eds.), Test-tube babies: A guide to moral questions, present techniques, and future possibilities (pp. 57–63). Melbourne: Oxford University Press. Lee, P. (2004). The pro-life argument from substantial identity: A defence. Bioethics, 18(3), 249–263. Lee, P., & George, R. P. (2008). Body-self dualism in contemporary ethics and politics. Cambridge, New York: Cambridge University Press. Lizza, J. P. (2010). Potentiality and persons at the margins of life. Diametros, 26, 44–57.

Potentiality in Bioethics

349

Manninen, B. A. (2007). Revisiting the argument from fetal potential. Philosophy, Ethics, and Humanities in Medicine, 2(5), 7. https://doi.org/10.1186/1747-5341-2-7. McGinn, C. (1992). Moral literacy, or, how to do the right thing. Indianapolis: Hackett Publishing. McMahan, J. (2002). The ethics of killing: problems at the margins of life. Oxford University Press. McMahan, J. (2008). Challenges to human equality. The Journal of Ethics, 12(1), 81–104. https:// doi.org/10.1007/s10892-007-9020-9. McMahan, J. (2009). Cognitive disability and cognitive enhancement. Metaphilosophy, 40(3–4), 582–605. Morris, J. (2012). Substance ontology cannot determine the moral status of embryos. The Journal of Medicine and Philosophy, 37(4), 331–350. https://doi.org/10.1093/jmp/jhs026. Noonan, J. T. (1970). An almost absolute value in history. In J. T. Noonan (Ed.), The morality of abortion: Legal and historical perspectives (pp. 51–59). Cambridge: Harvard University Press. Oderberg, D. S. (1997). Modal properties, moral status, and identity. Philosophy & Public Affairs, 26(3), 259–276. Persson, I., & Savulescu, J. (2010). Actualizable potential, reproduction, and embryo research: Bringing embryos into existence for different purposes or not at all. Cambridge Quarterly of Healthcare Ethics, 19(1), 51–60. https://doi.org/10.1017/S0963180109990247. Piotrowska, M. (2014). Transferring morality to human-nonhuman chimeras. The American Journal of Bioethics, 14(2), 4–12. https://doi.org/10.1080/15265161.2013.868951. President’s Council on Bioethics. (2002). Human cloning and human dignity: The Report of the President’s Council on Bioethics. New York: PublicAffairs. President’s Council on Bioethics. (2005). Alternative sources of human pluripotent stem cells. A White Paper of The President’s Council on Bioethics. Washington. D.C. Available online: https://bioethicsarchive.georgetown.edu/pcbe/reports/white_paper/. Sagan, A., & Singer, P. (2007). The moral status of stem cells. Metaphilosophy, 38(2-3), 264–284. https://doi.org/10.1111/j.1467-9973.2007.00482.x. Sagoff, M. (2007). Further thoughts about the human neuron mouse. The American Journal of Bioethics: AJOB, 7(5), 51–52. https://doi.org/10.1080/15265160701290447. Shewmon, D. A. (1997). Recovery from ‘brain death’: A neurologist’s apologia. The Linacre Quarterly, 64(1), 30–96. Shewmon, D. A. (2010). Constructing the death elephant: A synthetic paradigm shift for the definition, criteria, and tests for death. The Journal of Medicine and Philosophy, 35(3), 256– 298. https://doi.org/10.1093/jmp/jhq022. Singer, P. (1993). Practical ethics (2nd ed.). Cambridge, New York: Cambridge University Press. Singer, P., & Dawson, K. (1988). IVF technology and the argument from potential. Philosophy & Public Affairs, 17(2), 87–104. Steinbock, B. (2007a). Moral status, moral value, and human embryos: Implications for stem cell research. In B. Steinbock (Ed.), The oxford handbook of bioethics (pp. 417–440). Oxford, New York: Oxford University Press. Steinbock, B. (2007b). The science, policy, and ethics of stem cell research. Reproductive BioMedicine Online, 14(Supplement 1), 130–136. https://doi.org/10.1016/S1472-6483(10) 60746-8. Stier, M., & Schöne-Seifert, B. (2013). The argument from potentiality in the embryo protection debate: Finally ‘depotentialized’? The American Journal of Bioethics, 13(1), 19–27. https://doi. org/10.1080/15265161.2012.743619. Stone, J. (1987). Why potentiality matters. Canadian Journal of Philosophy, 17(4), 815–829. https://doi.org/10.1080/00455091.1987.10715920. Stone, J. (1994). Why potentiality still matters. Canadian Journal of Philosophy, 24(2), 281–293. https://doi.org/10.1080/00455091.1994.10717370. Streiffer, R. (2005). At the edge of humanity: Human stem cells, chimeras, and moral status. Kennedy Institute of Ethics Journal, 15(4), 347–370.

350

M. Stier

Streiffer, R. (2011). Human/non-human chimeras. In E. N. Zalta (Ed.), The stanford encyclopedia of philosophy (Spring 2011). Available online: http://plato.stanford.edu/archives/spr2011/ entries/chimeras/. Sumner, L. W. (1981). Abortion and moral theory. Princeton, NJ: Princeton University Press. Testa, G., Borghese, L., Steinbeck, J. A., & Brüstle, O. (2007). Breakdown of the potentiality principle and its impact on global stem cell research. Cell Stem Cell, 1(2), 153–156. https://doi. org/10.1016/j.stem.2007.07.006. Thomson, J. J. (1971). A defense of abortion. Philosophy & Public Affairs, 1(1), 47–66. Tooley, M. (1972). Abortion and infanticide. Philosophy & Public Affairs, 2(1), 37–65. Warren, M. A. (1977). Do potential people have moral rights? Canadian Journal of Philosophy, 7(2), 275–289. https://doi.org/10.1080/00455091.1977.10717018. Wasserman, D., Asch, A., Blustein, J., & Putnam, D. (2013). Cognitive disability and moral status. In E. N. Zalta (Ed.), The Stanford encyclopedia of philosophy (Fall 2013). Available online: http://plato.stanford.edu/archives/fall2013/entries/cognitive-disability/. Wertheimer, R. (1971). Understanding the abortion argument. Philosophy & Public Affairs, 1(1), 67–95. Wilmut, I., Schnieke, A. E., McWhir, J., Kind, A. J., & Campbell, K. H. (1997). Viable offspring derived from fetal and adult mammalian cells. Nature, 385(6619), 810–813. https://doi.org/10. 1038/385810a0.

Author Biography Marco Stier works as a Scientific Assistant at the Institute for Medical Ethics, History and Philosophy of Medicine in Münster, Germany. His main fields of research are bioethics in general, neuroethics, philosophy of action and philosophy of mind. Among his publications are two monographs on free will: “Verantwortung und Strafe ohne Freiheit” (Mentis 2011), “Willensfreiheit: Bestimmt mein Gehirn oder bestimme ich?” (Reinhardt 2014) as well as several articles, e.g. “Normative Preconditions for the Assessment of Mental Disorder” (Frontiers in Theoretical and Philosophical Psychology, 2013).

Part VI

Potentiality in the Sciences

Potentiality in Physics Max Kistler

1 Introduction “Potential” is a technical term in physics, “potentiality” is not. Potentiality is a general philosophical concept that can also be applied to physical phenomena. In this chapter, I will concentrate on the concept of potentiality. However, before I ask how the concept of potentiality is or can be used in a philosophical interpretation of physics, I begin with a brief presentation of the physical concept of potential. Potentials can be used to represent physical systems in which all forces are conservative1. In that case, one can represent the system by the potential W, which is a simple mathematical function of spatial location only: contrary to a vector, which has several components, the potential is a scalar quantity, which has only one component. The potential W can be used to retrieve various other physical quantities. Let us consider a system in which only gravitational forces are acting. The potential energy U(x) of an object with mass m, situated at a given point x in space, is the product of m with the value of the potential W(x) at point x. U ¼ mW

1

All fundamental forces are conservative. Non-conservative forces, such as friction, only appear in incomplete representations of physical systems. See Feynman et al. (1963, p. 14–6/7). The presentation of the relations between the different physical concepts in this paragraph follows Feynman et al.’s (1963) exposition.

M. Kistler (&) IHPST, Université Paris 1 Panthéon Sorbonne & CNRS, Paris, France e-mail: [email protected] © Springer Science+Business Media B.V., part of Springer Nature 2018 K. Engelhard and M. Quante (eds.), Handbook of Potentiality, https://doi.org/10.1007/978-94-024-1287-1_14

353

354

M. Kistler

The force F acting on the object in a given direction dx is the partial derivative of U in the direction dx.2 Fx ¼ @U=@x (The force as a vector in three dimensions can be written F ¼ rU, where “r” represents the gradient operator). The force can be considered as a power to accelerate objects that undergo it. From the potential, one can also calculate, at each point of space, the gravitational field C, which is the force acting on a body located at that point, per unit of mass: C = F/m. C ¼ rW ¼ rU=m ¼ F=m: Equivalently, the potential energy of a body with mass m at a given point x0 is equal to the work3 required to bring the body to that point, from infinitely far away, where the strength of gravitational forces is taken to be zero, and where the potential energy is by convention taken to be zero.4 Zx0 Uðx0 Þ ¼ 

Fdx 1

In the same way, the (gravitational) potential of the system at point x0 is equal to the work per mass required to bring a massive body to that point: Zx0 Wðx0 Þ ¼ 

Zx0 ðF=mÞdx ¼ 

1

Cdx 1

As I have just presented its definition and use, the gravitational potential appears as a mathematical construction. Only the field and forces are real and causally efficacious, whereas the gravitational potential is nothing more than a mathematically convenient way of representing the gravitational field and forces. This is indeed how potentials are often interpreted. However, the interpretation of 2

Here is a way to understand the minus sign. The gravitational potential around a massive body M is by convention set to be zero at infinite distance from the body. The value of the potential is increasingly negative with decreasing distance from M. The derivate of the potential with respect to positive distance from M is positive, but the gravitational force is directed in the opposite direction, i.e. towards M. 3 Work, a form of energy, is defined as the integral taken over the path of the body, of the force acting on the body at each point on the path. 4 To understand the minus sign: Positive work (energy) is required to remove a body from the proximity of the source of the potential, i.e. the work is positive if the direction of the movement is directed against the direction of the force.

Potentiality in Physics

355

potentials as mathematical fictions does not seem to be adequate for all potentials. The electromagnetic potential can be defined in a similar way to the gravitational potential. As the gravitational potential, it appears to be a mathematical device from which the physically real electromagnetic field and forces may be derived. However, the electrodynamic potential may exercise causal influence at points at which the field strengths are zero.5 According to the Causal Criterion of Reality6, this is a good reason to regard the potential as real in itself, over and above the field and forces that can be derived from it. The concept of potentiality, though the word is not used within physics, played a role in the philosophical interpretation of superposed states of quantum mechanical systems. Heisenberg (1958, p. 27) suggested that the wave function describing the state of certain quantum systems—those in so-called superposed states7—describes potentialities of the system, in a sense in which potentialities are not actual properties.8 In this context, Heisenberg uses the term “potentia” (Heisenberg 1958, p. 15, 154, 155) that belongs to the Aristotelian tradition. The use of the concepts of potentiality and actuality in the interpretation of quantum mechanics is sometimes accompanied by the thesis that only superposed states of quantum systems are potentialities, but not the states of “classical” physical systems. Some authors who follow Heisenberg in interpreting quantum states as potentialities, judge that “a substantial distinction between potential and actual existence is rendered obsolete in classical mechanics” (Karakostas 2007, p. 281). However, Aristotle introduced the concepts of potentiality and actuality to make sense of change in general. Thus in order to defend the claim9 that only certain quantum mechanical systems have (or are) potentialities, whereas physical objects that can be described in the terms of classical physics do not, the concepts of potentiality and actuality need to be given a new and more specific sense.

5

This is the Aharonov-Bohm effect. Cf. Healey (1997), Belot (1998) According to Causal Criterion of Reality, all and only those entities exist that make a causal difference. Armstrong has called it the “Eleatic principle” by reference to its formulation by the Eleatic Stranger in Plato’s Sophist (247d-e). In Armstrong’s words, “everything that exists makes a difference to the causal powers of something” (Armstrong 1997, p. 41), but it is the converse that is used to justify the postulation of the existence of both particulars and properties: Everything that makes a difference to the causal powers of something exists. The justification of the principle itself rests on the claim that it is a central part of scientific methodology. Jaegwon Kim called it “Alexander’s Dictum” (Kim 1992, p. 134), in honour of Samuel Alexander (1920) who defended it as a metaphysical principle: “To be real is to have causal powers” (Kim 1992, p. 135; Kim’s italics). Cf. Kistler (2002). 7 I will explain shortly what superposed states are. 8 The fact that potentialities, as Heisenberg uses the term, are non-actual properties of physical systems shows that they are very different from potentials, as the term is used in physics and briefly explained above. Potentials are actual and underlie actual fields and forces acting on objects. 9 As we will see in the second part of the chapter, this claim, explicitly made by Karakostas (2007), is part of Heisenberg’s conception of the difference between quantum mechanical systems and classical systems. 6

356

M. Kistler

This chapter will have two parts. In the first part, I present a general interpretation of the conceptual role of potentiality in physics, based on a general thesis on the relation between the concepts of potentiality and dispositionality. In the second part, I will suggest a realist interpretation of quantum mechanical systems in terms of powers and dispositions, which I will develop by starting from a critical analysis of Heisenberg’s thesis that quantum mechanical systems in superposed systems have only potential, but no actual, existence. I will suggest that superposed quantum states are theoretical properties that can be conceived as powers, which are indirectly characterized by a specific set of dispositions.

2 Potentiality as Dispositionality I this first part of the chapter, I argue that the terms “potentiality” and “dispositionality” have the same meaning. A potentiality is a disposition, in the sense this concept has in metaphysics and philosophy of language. The difference between the terms “potential” and “dispositional” is, I would suggest, pragmatic. The words “disposition” and “dispositional”, as well as similar terms, such as “potency” (Bird 2007) are philosophical terms of art and not technical terms of physics. Other terms, such as “capacity” (Cartwright 1989) are technical terms of physics but are also used as philosophical concepts whose domain of application exceeds that of the homonymous technical term. The terms “potential” and “capacity” are used in physics with a meaning that is very similar to that of the philosophical concepts of potentiality and dispositionality. These concepts are the object of an amount of important literature in metaphysics, philosophy of language and philosophy of science. The names of some physical entities, such as “gravitational potential” and “heat capacity”, indicate that they are potentialities or capacities. However, the philosophical concepts of potentiality and disposition also apply more generally to physical entities or properties whose potentiality or dispositionality is not reflected in their name. Such entities or properties are neither called “potentials” nor “capacities” nor “dispositions” nor do their names have a suffix like “-ible”, “-able”, “-uble”, “-ability”, “-ubility” etc. However, they share the characteristics of potentiality and dispositionality with entities that are explicitly called “potentials” or “capacities”, i.e. they are both concepts that characterize entities indirectly, through the manifestations they cause in certain situations, which are called their “triggering” conditions. It is essential to both concepts (dispositionality and potentiality) that any property to which one of them applies is essentially linked to a second property: for a disposition, the second property is its manifestations, for a potentiality, it is its actualization. I suggest that, although the meaning of these concepts is the same, there is a pragmatic difference between their uses, and this is in particular the case with their use in the interpretation of physics. Before I explain this difference, let me first introduce the philosophical concept of a disposition (see Mumford 1998, Kistler and Gnassounou 2007).

Potentiality in Physics

357

There are many common sense dispositions. Window panes are solid, transparent, and fragile, rubber is elastic. Dispositions are not directly observable, but there is usually a quite direct link between them and observable properties, which are their manifestations. The link can only be expressed by counterfactual conditionals. Such “counterfactuals”, as they are also called, link a disposition with one or several pairs of conditions that are called, respectively, the disposition’s “triggering” and “manifestation” conditions. To say that a window pane is fragile implies that, if it were strongly struck in normal or ordinary circumstances it would break. One can express this by saying that the truth condition of the proposition that window pane w is fragile is the same as the truth condition of the counterfactual proposition that if w was struck strongly in normal circumstances, w would break. It is necessary to include a clause like “in normal circumstances”, often called a “ceteris paribus” clause, because the equivalence does not hold without exceptions. Typically, dispositions have what has been metaphorically labelled “antidotes” (Bird 1998). To be toxic is a dispositional property: a substance t is toxic for a given organism if it is the case that if the organism came, in normal circumstances, into contact with t (by ingestion, inhalation, touch etc.), it would be harmed or die. However, the counterfactual analysis of the meaning of the word “toxic” would be incorrect without the qualifier “in normal circumstances”. Indeed, organisms can often be immunized against toxic substances by antidotes. An antidote for a given toxic substance t is a substance such that, if an organism has absorbed this antidote for t, then it is not the case that if the organism came into contact with t (and if circumstances were otherwise normal) it would be harmed. Some dispositions manifest themselves independently of any triggering conditions (Mumford and Anjum 2011, p. 36). Gravitation and radioactivity are cases in point. Massive objects have the disposition to be attracted by other massive objects, and this disposition manifests itself (indirectly) without any triggering. It contributes to the net force acting on the object, which manifests itself in the object’s acceleration. Radioactive decay too needs no triggering. The question of what makes true the attribution of a disposition is controversial. One thesis is that what makes true the attribution of a disposition is always a “categorical” property.10 It is difficult to define the concept of a categorical property otherwise than by opposing it to that of a dispositional property, but categorical properties are usually taken to be structural, and typically microscopic, properties. The categorical properties that are taken to provide the truth-maker for ascriptions of fragility are micro-structural properties, such as properties of the chemical bonds between the molecules of the fragile object. According to the main alternative thesis, dispositional properties, in this context often called “powers”, are properties in their own right, which are real and causally efficacious just as “categorical” or structural properties. The relation between a macroscopic disposition such as fragility and the microscopic structural and chemical properties underlying it is reduction: the microscopic structure may reductively explain the fragility of the

10

Quine (1966, 73), Quine (1971, 10), Prior et al. (1982), Armstrong (1973), (1997).

358

M. Kistler

macroscopic window pane. However, the possibility of micro-reduction is independent from the issue whether a given property is dispositional or categorical. The microscopic properties in the reduction base may themselves be dispositional. Indeed, micro-reducibility does not seem to be necessary for being a disposition because at least some (maybe all) fundamental (i.e. non-reducible) properties, such as mass and charge, are dispositional.11 It does not seem to be sufficient either, for some macroscopic categorical properties, such as structural properties, are reducible: the cubic structure of a crystal is reducible in geometric terms to the geometrical relations between its atomic or molecular components. All this applies to physical dispositions such as electric conductivity or heat capacity. Electrical conductivity (usually represented by the symbol r) is a property of metals (and other materials), which is defined by the fact that if an object is electrically conductive, then if the object is put in an electric field of strength E then electric current of density J = rE will flow through it.12 Placing the object in the electric field is the triggering condition of the disposition and the current flow its manifestation. Physics is full of such dispositional properties. As other dispositions, potentials and capacities in the technical sense of physics can be characterized by specific pairs of triggering conditions and manifestations. Saying that something has a potentiality (or a potential) for some effect means that it has a disposition for that effect. Even though potentials are dispositions and everything that has potentiality also has dispositionality, it does not always seem appropriate to attribute potentiality to dispositional entities or properties. It may be just a contingent fact about language that some dispositional physical properties are called “potentials” or “capacities”, whereas others, such as electric conductivity, are not. But “potentiality” is a philosophical term, which doesn’t appear in physics texts any more than the terms “disposition” and “dispositionality”. So it is a philosophical issue whether the concepts of dispositionality and potentiality are equivalent. My suggestion is that they share their literal meaning but not the pragmatic conditions of their appropriate use. The first pragmatic constraint that distinguishes the use of the terms “potentiality” and “dispositionality” concerns the time at which a disposition manifests itself and the time at which a potentiality is actualized. The difference can be expressed in terms of the possibility of simultaneous attribution: it can be appropriate to attribute a disposition to something at a time at which the disposition manifests itself, but it seems less appropriate to attribute a potentiality at a time at which it is actualized. Let us suppose that a system s has at time t a dispositional property P. This means that if s were subject to the triggering condition (or one of the set of On fundamental, irreducible or “ungrounded” dispositions, see Blackburn (1990), Molnar (2003), McKitrick (2003), Mumford (2006), Kistler (2007). 12 The current density J is defined as the current per unit cross-sectional area of a conductor. For a given field E, the electric current I flowing through a wire depends on E, the wire’s conductivity r and on the wire’s cross-section. 11

Potentiality in Physics

359

triggering conditions if there is more than one) T at time t, it would manifest M at time t* (t* may be equal to t or later than t). Take r to be electrical conductivity and w to be a copper wire that has r. Then for all times t, if w is at t in an electric field E, it will manifest electric current with current density J = rE at t.13 It seems perfectly appropriate to attribute the disposition to w at time t, at which it is triggered, even if it manifests itself at the same moment t by the flow of current. However, it is central to the concept of a disposition that it can also be attributed at times t at which it is not triggered. Therefore, a wire can correctly be said to be conductive during periods in which it is not subject to the triggering condition, even if it is never put in any electric field, and therefore never manifests its conductivity.14 By contrast, and this is the first aspect by which the use of the term “potentiality” differs from the use of the term “disposition”, it is less appropriate to attribute a system s at time t with the potentiality P for some actualization A if P’s actualization A has already occurred at t. (The actualization of a potentiality is analogous to the manifestation of a disposition). A property is more naturally considered to be a potentiality if the manifestation is expected to occur after some finite delay, relative to the ascription of the potentiality.15 There is no such constraint for the use of “disposition”. It can be appropriate to attribute a disposition at a time at which it is manifested. There seems to be no difference with respect to the need for triggering: just as dispositions, some potentials but not all need triggering to be actualized. Just as dispositions, those potentials that require triggering for their actualization can exist untriggered.

13

In case the disposition is attributed at t but triggered only at t* (where t* is later than t), it must be presupposed that the dispositional property P is a permanent property of the system, which persists from t to t*, so that the system still possesses P at the time t* of triggering. This requirement is explicitly expressed in Lewis (1997), one of the most recent and sophisticated analyses of dispositions in terms of counterfactual conditionals. 14 The question of whether it is appropriate to attribute a disposition at a time at which it manifests itself must be distinguished from the question of whether there is a delay between the triggering and the manifestation of a disposition. The concept of a disposition neither requires a delay between triggering and manifestation nor the absence of such a delay. Both delayed and immediate manifestations are conceivable and compatible with dispositionality. In the case of dispositions, there is room for variation along three dimensions. First, a system having disposition P at t may be triggered at any time, at t or later. Second, if the disposition is triggered at t, no conceptual restriction is imposed on the time at which the characteristic manifestation appears: It may be simultaneous to triggering or follow triggering after a certain delay. Third, some dispositions need no triggering at all but always manifest, so to say, spontaneously. 15 What is crucial here is the time at which the potential is attributed or ascribed. This is to be distinguished from the question when the object has acquired the potential. The latter question is independent of the distinction between potentiality and dispositionality that is the target of the present analysis. Say the radioactive nucleus n decays at time t1. At some earlier time t0, when it has not yet decayed, we may truthfully attribute to n the potentiality of decaying. The truth of the attribution of this potentiality to n at time t0 is completely independent of the time at which n has acquired that potentiality, which is the time at which it came into existence.

360

M. Kistler

Take a simple pendulum. When the mass of the pendulum starts its movement at a given height h, its potential energy is a function of h. On its downward movement, this potential energy is gradually transformed into kinetic energy, which is its actualization. When it reaches the lowest point on its trajectory, the potential energy has become completely actualized and entirely been transformed in kinetic energy. At least a part of the potential energy that the pendulum’s mass has when it is above its lowest position, satisfies the condition of appropriateness for potentiality: it is not yet actualized and will only be actualised after some finite interval of time. Another example for potentiality in physics is radioactive decay. There is a delay between the truthful attribution of the potentiality for decay to a nucleus and its actualization. Each type of radioactive nucleus has a characteristic half-life, which is necessarily greater than zero.16 Thus, the actualization condition, which consists in the presence of the decay products, is realized later than the moment at which the potentiality for decaying is attributed to the nucleus. The potentiality of radioactive decay is incompatible with its actualization. Once the decay has occurred, the radioactive nucleus that had the potentiality no longer exists; and thus, the potentiality does not exist at the same time as its actualization, i.e. the decay and its products. The second condition of appropriateness that distinguishes the uses of the words “dispositionality” and “potentiality” is that it seems more natural to speak of potentiality if the actualization is not necessary but only possible (and probable with a probability