The Undivided Self
Aristotle and the ‘Mind–Body Problem’
DAVID CHARLES
1
Great Clarendon Street, Oxford, OX2 6DP, United Kingdom
Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries
© David Charles 2021
The moral rights of the author have been asserted
First Edition published in 2021
Impression: 1
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by licence or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above
You must not circulate this work in any other form and you must impose this same condition on any acquirer
Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America
British Library Cataloguing in Publication Data Data available
Library of Congress Control Number: 2020941059
ISBN 978–0–19–886956–6
DOI: 10.1093/oso/9780198869566.001.0001
Printed and bound by CPI Group (UK) Ltd, Croydon, CR0 4YY
To Sunny and Farooq Haroon
Preface
Aristotle undertook a systematic investigation of perception, the emotions, memory, desire, and action, developing his own account of these phenomena and their interconnection. My aim is to gain a philosophical understanding of his views and to examine how far they withstand critical scrutiny.
Aristotle’s approach calls into question the way in which our, post-Cartesian, mind–body problem has been formulated. He was guided throughout by a conception of the psychological and of the material that was rejected by those who originally set up and subsequently sought to address our problem. His views constitute an alternative to basic elements of our conventional thinking about psychological phenomena and their place in a material world. They offer, in effect, the resources to dissolve, rather than solve, the mind–body problem we have inherited. Or so I shall argue.
Aristotle’s account is not, as is sometimes suggested, a version of a standard post-Cartesian theory: dualism, materialism (whether reductionist or nonreductionist), functionalism, or neutral monism. But nor does it rest, as is sometimes claimed, on assumptions that we can no longer take seriously. Properly understood, it is a philosophically live alternative to standard thinking in this area of philosophy.
It is the task of historians of philosophy to grasp, and articulate, what is distinctive about the authors they study in a detailed, historically sensitive way. Doing so is, of course, a good in itself. But does twenty-first-century philosophy have anything to gain from their studies? Shouldn’t contemporary philosophers simply tackle today’s problems without considering the views of their predecessors?
One major benefit of the study of philosophy’s history is that it provides the opportunity to look critically at the way in which contemporary problems have been formulated. Perhaps the manner in which this has been done sometimes leads us astray. It would, after all, be a piece of exceptional good fortune if they all had been ideally set up at just the time we came to study them. Consideration of Aristotle’s views may suggest a way to ‘deconstruct’ major, seemingly irresoluble, questions in what we, but not he, would call the philosophy of mind.
Recognition of diversity is, of course, a good. So too is the possibility of experimenting with different ways of thinking. But it is also important for historians of Aristotle’s philosophy to examine the extent to which his ideas, properly understood, withstand scrutiny, where they should be rejected and where retained and built on. Doing so may reveal ways to avoid apparently unsatisfactory orthodoxies in current thinking. Historians of philosophy need
not, and perhaps should not, be narrow specialists confined to their own sub-branch of enquiry. Contemporary philosophy still has things to learn from a detailed study of its history.
The initial idea for this book emerged from discussions in a long-running reading group on De Anima held at Oriel College, Oxford. When we decided in 2004, after more than fifteen years spent on the second and third books, to read De Anima A.1, much fell into place. I vividly recall our lively meetings at that time in which Paolo Crivelli, Michael Frede, Edward Hussey, Benjamin Morison, Catherine Rowett, Annamaria Schiaparelli, and Cecilia Trifogli were actively engaged. As a result, I began to question and finally reject my earlier, nonreductionist materialist, interpretation of Aristotle’s views. However, our work in 2004 was based on many earlier discussions, dating back to 1989, to which Jonathan Beere, Suzanne Bobzien, Victor Caston, Stephen Everson, Katerina Ierodiakonou, Thomas Johansen, Lindsay Judson, Jean Louis Labarrière, Hendrik Lorenz, Michail Peramatzis, Frisbee Sheffield, Christopher Shields, and many others had contributed.
I have gained from the opportunity to give seminars on this material in Berlin, Brown, Oxford, Taipei, Tokyo, and Yale and from detailed discussions with, and comments from, Andreas Anagnostopoulos, Udit Bery, David Bronstein, Jason Carter, Kei Chiba, Alan Code, Daniel Devereux, Guus Eelink, Verity Harte, Devin Henry, Brad Inwood, Terry Irwin, Vassilis Karasmanis, Emily Katz, Emily Kress, James Lennox, Frank Lewis, Sam Meister, Giulia Mingucci, Scott O’Connor, Mika Perala, Oliver Primavesi, Diana Quarantotto, Bryan Reece, Jake Rohde, Richard Sorabji, Andrew Werner, Jennifer Whiting, Stephen Williams, and Chin-Mu Yang. I have been particularly fortunate to receive thoughtful and friendly criticism and advice from several experts with different views on these topics: Victor Caston, Klaus Corcilius, Mary Louise Gill, Thomas Johansen, Mark Johnstone, Marko Malink, Christof Rapp, and Christopher Shields. Our discussions have been a model of collaborative work: incisive and challenging but unfailingly constructive and good-humoured. I am greatly indebted to a formidable array of colleagues in ancient philosophy, only some of whom are mentioned in this paragraph, for their criticisms, suggestions, and encouragement.
I was helped in thinking about these issues by a second equally long-lasting group which also met in Oriel College. I owe most, in my present investigation, to those of its members who worked in the philosophy of mind and action, including at various times Bill Brewer, Justin Broackes, John Campbell, Quassim Cassam, Bill Child, Adrian Cussins, Naomi Eilan, Jennifer Hornsby, Michael Martin, Paul Snowdon, Helen Steward, and Rowland Stout. Our discussions were acute, enjoyable, and enlightening. On occasion, I found myself wondering how many comparable meetings of specialists in this area of philosophy there could have been since Aristotle and his friends discussed them long ago, in Assos, Lesbos,
and Athens. Justin Broackes, Paul Snowdon, Rowland Stout, and Jonathan Westphal kindly read and commented on parts of the final version.
I am greatly indebted to Peter Momtchiloff for his support and advice, and to Lindsay Judson, Michail Peramatzis, and Howard Robinson for reading the penultimate draft, mixing criticism with encouragement in appropriate measure. Lindsay Judson has once again proved an exemplary critic and editor. Michail Peramatzis and I have discussed these topics in detail over the past fifteen years and I gratefully acknowledge his major role in their development. While I have sought to record his individual contributions on a point-by-point basis, my debt to him far exceeds these. I should also like to thank Paul Schilling for his generous help in preparing the final version for publication.
There are several people whose response to this book I would have valued and will greatly miss: John Ackrill, Myles Burnyeat, Michael Frede, Allen Gotthelf, and David Pears. Each of them discussed earlier versions of some of its ideas and contributed to their development. I am heavily indebted to all of them.
My immediate family, Zafra, Ilan, and Huw, has been an ever-present source of support, encouragement, and criticism. So familiar have they become with some of this book’s philosophical and exegetical ‘moves’ that they have turned them into family jokes. I am very grateful to them for standing by me in the long years it has taken to complete this project.
It is a pleasure to dedicate this book to Sunny and Farooq Haroon, to whose constant friendship over more than fifty years I owe so much.
David Charles
New Haven, Connecticut 18 April 2020
Introduction
Aristotle developed a way of understanding psychological phenomena, such as the emotions, desire, and perception, that differs in important ways from those favoured by nearly all post-Cartesian philosophers. He did not accept the assumptions that we make in setting up and addressing our mind–body problem. If his viewpoint is defensible, it is Descartes’ problem—and not Aristotle’s account— that should be taken off the philosophical agenda.
Our mind–body problem can be expressed, at first approximation, as follows: how can the physical, defined without any explicit reference in its definition to the psychological, give rise to the psychological with its distinctive features, where the psychological is defined (in part or in whole) without any explicit reference in its definition to the physical?
In articulating this problem, the psychological is defined in terms of features such as consciousness, rationality, or certain phenomenal properties, themselves defined (in whole or in part) without reference (in their definition) to any physical properties. Psychological events and states are ones defined in terms of their possession of these distinctive properties. The physical, in turn, is defined in terms of features of events or states, such as being spatio-temporal or being studied by physics, themselves defined without any reference to the psychological. Our mind–body problem is: how does the physical, so defined, give rise to the psychological, so defined? How, for example, does our conscious experience of colour, the painfulness of our grief, what it is like to experience the pleasures of touch or taste and our commitment to principles of rationality arise out of the states, events, and properties studied by physics?
For some, like Descartes himself, this problem, once stated, is insoluble: the psychological, when properly understood, cannot arise from the physical because physical and psychological features belong to fundamentally different types of subject. Subjects of conscious experience are not, and cannot emerge from, purely physical phenomena. The world is made up of two fundamentally different types of substance. Later philosophers, dissatisfied with Descartes’ substance dualism, have searched for ways to solve his problem. Here, in barest outline, are five of their suggestions:
(a) Reductionist materialism: the psychological, properly understood, can be fully explained in terms of the physical, so defined. Psychological
events, states and their properties are reducible to physical events, states, and properties.
(b) Non-reductionist materialism: the psychological, so defined, ‘arises out’ of the physical but is not fully explicable in terms of it. Even if particular psychological events are identical with physical events, their psychological properties are not reducible to physical properties but rest on, or emerge out of, them. (Non-reductionist materialism.)1
(c) The psychological itself, phenomenal consciousness and rational commitments alike, can be fully defined in terms of their causal roles (as in various versions of functionalism) or dispositions to behaviour. So understood, all psychological phenomena can be unproblematically realized in physical events or states. (Hostile critics sometimes refer to this view as ‘consciousness denial’.)
(d) Pan-psychism or spiritualism: the physical is understood in terms which make its connection with the psychological readily intelligible. It, or some parts of it, is redefined as alive with consciousness or proto-consciousness, primitively disposed to have conscious experience.
(e) Neutral monism: the physical and the psychological, defined as above, are each to be understood as emerging from a more basic type of stuff which is neither physical nor psychological but neutral between them.
Aristotle, I shall suggest, did not adopt any of these approaches. Instead, he developed a way of thinking about psychological and physical phenomena which, once properly set out, dissolves the mind–body problem that these proposals are designed to address. More specifically, he did not accept:
(1) the definitions of the psychological or of the physical later used to formulate it or
(2) the ontology of events and states generally presupposed in doing so.
Our perception of colour, experiences of pain and pleasure, desires and the rest are, in his view, inextricably psycho-physical activities whose essential properties are inextricably psycho-physical.2 Neither these activities nor their properties can be adequately defined without reference in their definition to types of physical
1 Some non-reductionists (like Colin McGinn 1990) argue that while the psychological arises out of the physical, we cannot understand or render intelligible to ourselves how this happens. Most seek to render these connections intelligible.
2 Aristotle, as we shall see, carefully distinguished processes and activities of various types. The formulations in this chapter are a first approximation to a fuller interpretation of the relevant ontology.
activities, capacities, and properties. Nor, in his view, can the latter be adequately defined without essential reference in their definition to psychological activities, such as perception or conscious experience of the world. The phenomena at issue cannot be defined by decomposition into two definitionally separable components, one purely psychological (defined without explicit reference to the physical), the other purely physical (defined without reference to the psychological). From his perspective, the two assumptions (1) and (2), which drive nearly all post-Cartesian philosophy of mind, spring from mistaken ways of thinking about the psychological and the physical. Nor does his approach rest on analysing out, still less denying, the problematic features of the psychological or on redefining the physical in the radical way pan-psychists require.
Aristotle’s views have been the subject of scholarly debate. He has, in fact, regularly been interpreted as an advocate of one or other of the options just outlined. For some, he was a dualist.3 For others, a precursor of a version of functionalism or of non-reductionist materialism, options that have become fashionable in recent years.4 For still others, he avoided Descartes’ challenge but only at the cost of embracing an account of the physical which it is difficult, if not impossible, for anyone today to take seriously. This is because, in their view, he adopted a ‘spiritualist’ view of matter as ‘pregnant with consciousness’ or ‘essentially alive and capable of awareness’, a form of pan-psychism which was, many think, rightly ‘junked’ in the seventeenth century.5
Aristotle’s views are not, I shall argue, accurately captured by these interpretations. Their proponents are, or so I shall suggest, led astray by shared, deeply rooted, post-Cartesian assumptions. His account, in fact, offers an interesting and defensible alternative precisely because it does not share the preconceptions that we, and most since Descartes, unreflectingly use in thinking about this problem. It merits and repays philosophical scrutiny even though,
3 See, for example, Robert Heinaman (1990: 83–102).
4 For one example of a functionalist interpretation: see Hilary Putnam and Martha Nussbaum (1995: 27–76), for another, see Christopher Shields (1990: 19–33). For an example of a nonreductionist materialist account, see my earlier attempts to understand these matters (1984: 197–250 and 1988: 1–53), for another see Victor Caston (2005: 245–320). In the present study, I shall discuss functionalist and non-reductive materialist interpretations separately. For some non-reductionist interpreters, the psychological is defined in terms of goal-directed rationality which, to use Davidson’s phrase, ‘finds no echo’ in physical theory. Functionalist interpreters, by contrast, define psychological states in terms of their physical causes and effects without essential reference to specific types of internal physical states. This is true of materialist functionalists: while, in their view, the states that realize psychological states must be (generically) physical, there is no specific type of internal physical state referred to in defining them. To do so would be to violate the constraints of generality and autonomy from (internal) physical states of this type which they see as constitutive of the psychological.
5 Myles Burnyeat and Thomas Johansen developed this line of interpretation. I shall sometimes label their view ‘spiritualist’, following Burnyeat’s use of the term ‘spiritual change’ to capture what he intended (see Myles Burnyeat 1992: 15–26, 1995: 422–34, 2001: 129–53 and 2002: 28–90 and Thomas Johansen 1998). In recent years, pan-psychism, a close relative of the view Burnyeat thought should have been ‘junked’ in the seventeenth century, has undergone a cautious revival in the writings of David Chalmers (1995: 200–19, 1996) and Galen Strawson (2006).
indeed in no small measure because, its assumptions are fundamentally different from those with which we are now familiar.
To make the issues somewhat more determinate, it may be helpful to focus on Aristotle’s account of visual perception. For non-reductive materialist interpreters, this runs (in outline) as follows: in any case of visual perception, there is a particular event of a given physical type which ‘underlies’ or ‘grounds’ the subject’s (particular) conscious awareness of, for example, red.6 Psychological (or phenomenal) awareness is itself defined without reference to any underlying, internal, specific type of physical event or property and is not reducible to such events and property. What makes the underlying (or grounding) type of event physical is that it is of the same general type as occurs in objects which lack awareness or perception.
There is a lively debate among non-reductionist interpreters as to (i) how ‘underlying’ (or ‘grounding’) is best understood and (ii) what physical event is involved.7 But most agree that Aristotle understood particular acts of perception as constituted by (i) a particular physical event (of a type defined without essential reference to anything psychological) and (ii) some psychological feature, such as phenomenal awareness, defined without essential reference to any specific type of internal physical event, where the physical event and its physical properties (in some way or other) ‘underlie’ or (in some way) ‘ground’ the relevant psychological features. Aristotle, so interpreted, engages directly with our mind–body question: how do these purely psychological features arise out of purely physical ones?
According to the spiritualist interpreters, by contrast, Aristotle did not accept that there is a distinct particular physical event, as just defined, essentially present in perception. All that is essential to seeing red (in their view) is a subject’s becoming aware of, or attending to, redness: an activity that does not itself essentially involve any underlying or ‘grounding’ physical event (of the type just specified) in the sense organ (even if it may require the presence of certain physical necessary conditions). Aristotle, in their view, thought of perceptual awareness as a purely psychological phenomenon because he took the underlying matter in the sense organ to be primitively endowed with capacities for phenomenal awareness.8 He did not, so
6 There is debate about the relevant type of physical process: for some, the eye jelly literally turns red, for others the type of physical change involved is exemplified by (e.g.) water turning red when red light is played on it, for others the process is one of being affected by the ratio of the relevant colour. The first view was set out by Richard Sorabji (1974: 63–89) and Stephen Everson (1997), the second by Sorabji (2001: 49–61). The third view was developed by Victor Caston (2005: 267ff.).
7 Some take Aristotle to hold a token identity theory in which psychological properties ‘supervene on’ or ‘emerge from’ physical ones. Others have denied that he held a token identity thesis.
8 Two further points of clarification are required. (i) Burnyeat noted that a material organ’s act of awareness could itself be described (by Aristotle) as a physical change (since it is the act of a physical body with material preconditions) even if it involved no change in its matter (and so would not have been counted as a material change). This is a different use of the term ‘physical’ from the one used in the debate between modern physicalists and their opponents. I shall use the term ‘physical’ in the way
understood, engage with the Cartesian question of how an essentially physical event can be related to purely psychological features because in his account there is (incredibly as it now seems) no physical event essentially involved in perception.
These opposed schools of interpreters share two assumptions. They agree that, for Aristotle, individual cases of perception are instances of some purely psychological type, such as awareness of redness, which can be fully defined without explicit reference to any internal type of physical event defined as such.9 They also both accept that the only types of event we can seriously entertain (as contemporary philosophers) are either purely psychological (in the sense just defined) or purely physical (defined without reference to the psychological) or else a combination of one purely physical and another purely psychological type.10 Aristotle, I shall argue, did not share these assumptions. More specifically, in his account:
[A] The psychological activities involved in emotions, desire and perception (and their essential properties) are defined as inextricably psycho-physical, not definable by decomposition into two separately defined types of phenomena, one purely psychological, the other purely physical. Being psycho-physical is (in a way to be explicated) an essential aspect of their nature.11
[B] The relevant specific type of physical activity cannot be defined without explicit reference in their definition to some psycho-physical activity. It too is an essentially psycho-physical activity, even though the relevant matter is not itself primitively endowed with consciousness.
Both schools of interpreters err in claiming to find in Aristotle’s discussions a purely psychological feature or activity essential to, for example, perceiving or desiring. They also go astray in assuming that his views can only make philosophical sense today if they contain a purely physical event which ‘grounds’ the relevant purely psychological feature. Non-reductionist materialist interpreters seek to find, or make room for, a determinate purely physical grounding event of this type in his writings. Spiritualists, finding little sign of, or space for, any such thing,
defined above (in note 4) and reserve the term ‘physikal’ (if needed) for the sense Burnyeat notes. (ii) Burnyeat allows (2001) that concomitant processes (of a type congenial to the physicalist interpreter) may occur when we perceive (in Aristotle’s account) but denies that these processes are essential elements in perceiving (or stand to perceiving as matter to form).
9 I discuss further aspects of functionalist interpretations in Chapter 1, Section 6 and Chapter 7, Section 4.
10 ‘Combination’, as I use the term, allows for the possibility of psychological events or their properties supervening on, or being in some other way determined by, physical events and their properties.
11 I use the terms ‘psychological’ and ‘mental’ to indicate (in general terms to the modern reader) the type of process at issue. For Aristotle, it should be noted, nutrition is also an inextricably psychophysical activity of the soul (psuchē).
conclude that his view must be rejected by anyone who seriously accepts the view of matter which drove the scientific revolution of the seventeenth century.
Aristotle, by contrast, in committing himself to [A] and [B], presents the relevant phenomena as, in their very essence, inextricably psycho-physical. It is not just that we choose to describe them in inextricably psycho-physical terms. It is because they are psycho-physical in this way that they are the very phenomena they are. In what follows, I shall call this suggestion the ‘inextricabilist’ interpretation.
[A] and [B] are independent exegetical claims: Aristotle might have accepted one and not the other. In holding [A], he sets himself apart from what might be described as the ‘philosophical idea of the psychological as purely psychological’. Emotions, desire, and perception are, in his view, inextricably psycho-physical. However, he could have held [A] while thinking that these states rest on, or emerge from, purely physical states. Conversely, he could have accepted [B] but nonetheless understood emotions and the rest as defined solely in purely psychological terms.
[A] will be, for many scholars, the more controversial exegetical claim. Indeed, [B] has, in some version, been widely (if not universally) accepted in recent years. A major part of my interpretative task is to argue that Aristotle did not accept ‘the philosophical idea of the psychological’ and had good reasons, based on his more general views concerning enmattered essences (or forms), not to do so. However, for many philosophers, [B] will be the more unsettling claim, apparently committing Aristotle to a disturbingly anti-modern conception of matter. In response, I shall suggest that he held an intelligible, if distinctive, metaphysical view of the relation between matter and the psycho-physical in natural organisms. Rightly understood, his viewpoint did not commit him (or his supporters) to rejecting the view of matter which took hold in the seventeenth century. That said, when in the final chapter I assess both aspects of Aristotle’s account, [B] may well remain the more controversial.
Aristotle’s account of fear and anger provides a model for his treatment of other ‘passions common to body and soul’, such as perception and desire. The type of desire for revenge which defines anger is, in his view, an essentially embodied, ‘hot’ type of desire, defined in terms which explicitly refer to its being a specific type of bodily activity. Its form, in Aristotle’s terminology, is captured in this definition. One cannot define its form simply as the desire for revenge without referring to it as an embodied-in-heat type of desire. The type of desire for revenge which defines anger, and constitutes its form, is, in his account, an inextricably psycho-physical, enmattered, activity with essential properties of the same type. Its form itself is, in Aristotelian terms, definitionally enmattered.12 This is his way of expressing claim [A].
12 In the Introduction, the terms ‘embodied’ and ‘enmattered’ are used interchangeably. This is a first approximation to a fuller interpretation, developed later, in which these terms are distinguished. Forms can be enmattered other than in bodies.
We can grasp somewhat more precisely what is involved in Aristotle’s claim [A] by considering his discussion of snubness, which he defined, I shall suggest, as nasal-concavity: a type of concavity which cannot be defined without essential reference to the nose. Snubness is not, in his view, a type of concavity, defined independently of noses, which is realized by or related (in some way) to noses. It is, instead, an essential (de re) aspect of the nature of the relevant type of concavity that it is nasal-concavity. This type of concavity is, we might say, in itself or intrinsically, nasal. The form in question contains, in Aristotle’s own terminology, being nasal ‘as a part’. While these terms may require further elucidation, snubness is to be defined in ways which explicitly refer in its definition to a distinctively nasal way of being concave. It is because snubness is, in its nature, this specific type of concavity that it can only be realized in noses.
In the case of anger: the distinctive type of desire for revenge (that defines anger) is, if like snubness, in its nature a type of essentially bodily phenomenon. It is not that the relevant type of desire for revenge, defined independently of a specific type of bodily activity (such as boiling of the blood), is only a case of anger if it is accompanied by, realized by or otherwise related to that very type of bodily process. On the contrary, to be the relevant type of desire is to be in itself an embodiedin-heat type of desire, one whose nature is defined as a specifically embodied way of desiring revenge. One cannot define this type of desire, or capture its form, without explicit reference in its definition to specific types of bodily capacities or activities. The relevant type of desire must be, in its own nature, inextricably psycho-physical in this specific way to play the role required of it.
In accepting [B], Aristotle commits himself to the view that the relevant specific type of physical activity cannot be defined without explicit reference in its definition to some psycho-physical activity. [B], however, immediately strikes us as problematic. We think that the very same type of physical activity—such as boiling of the blood—could occur in a variety of different cases, not just when we are angry. It might occur, for example, when we are in a fever or simply disgusted or perhaps when blood is heated by some mechanical means without any psychological state being present. Indeed, when we think in terms of events of physical heating, this seems correct. However, Aristotle employed a different ontology, focusing on the unfolding activity (or process) at issue, understood as a type of phenomenon which develops and stops in the way it does in, for example, an angry person, when confronted with reason, pity, changed circumstances, or the recognition of some previously unnoticed fact (e.g. that one has not been insulted). This is why he defines the relevant type of heating as one which results from the psycho-physical capacity to desire revenge. It is because the heating in question is the realization of this goal-directed capacity that it develops and wanes as it does, why it is responsive to some considerations and not others. Remove reference to the relevant psycho-physical capacity and one will have, in Aristotle’s view, no account of the way the specific kind of heating unfolds as it does in cases
of anger. In his model, there will be as many types of heating as there are distinct psycho-physical capacities exercised. These types will differ even if they are all cases of heating at the same temperature in the same area of the body, provided that they realize different capacities of the organism in question. While, at a given moment, two distinct types of heating may appear indistinguishable, they will differ (in type) if they are the realizations of different psycho-physical capacities. There is, in Aristotle’s view, just one essentially psycho-physical capacity realized in anger. There is no need to postulate the existence of a further separate determinate purely physical capacity which underlies it. Indeed, were there to be such a capacity, given its lack of sensitivity to reason or pity, there would be no guarantee (short of a successful reduction—of which Aristotle despaired) that the resulting physical activity would ‘march in step’ with that of the psycho-physical capacity for anger. In that eventuality, the organism in question, Achilles (to take the case of a famously angry person) would not be a properly unified subject, endowed with an integrated capacity to respond appropriately when wronged. In Aristotle’s view, by contrast, Achilles’ capacity for anger is an inextricably psychophysical capacity of an inextricably psycho-physical subject, an essentially integrated organism with its own unified teleological goals. A capacity of this specific type is required to generate the one unified activity that ensued before the walls of Troy.
Another example may serve to make Aristotle’s idea more vivid. Consider weaving and the skill of Penelope the weaver. In Aristotle’s account, one cannot define her activity of weaving without essential reference to her hands, the loom, and the wool on which she operates. But equally, one cannot define which physical movements are essential to weaving without reference to her skill and know-how. It is the latter that controls the developing movements that are essential to the weaving. Further, since the weaver’s skill is defined as that which guides skilled bodily movements of this type, it too must be an inextricably psycho-physical capacity: a type of inextricably embodied phenomenon which can be the per se cause of our skilled bodily activity.13
My investigation has three stages. At first, in Chapters 1, 3, 4, and 5, I focus on Aristotle’s discussion in De Anima and De Motu Animalium of emotions, desire, and perception. I suggest that they are all, in his account, inextricably psychophysical. They are, as I argue in Chapter 6, together with imagination and certain
13 One might also, consistently with Aristotle’s viewpoint, succeed in capturing the causal profile of the state that realizes Penelope’s skill in purely physical terms. Perhaps, with great ingenuity, one might even isolate a physical property P* such that whatever has P* has some purely psychological feature (Q). But, if one were to succeed, the ground for the claim that whatever has P* has Q would be, in Aristotle’s view, the inextricably psycho-physical skill—from which P* and Q would be abstractions. There is, in his picture, one such psycho-physical capacity, the unified cause of the resulting bodily movements. This is why the latter are not over-determined by two distinct properties: P* and Q.
aspects of human thought, the activities of inextricably psycho-physical subjects. The resulting picture exemplifies (what I shall call) his ‘simple theory’ of the relevant phenomena. At the second stage, in Chapter 2 and parts of Chapter 7, I shall consider those features of Aristotle’s general theory of hylomorphism which parallel his account of these psychological activities. My aim is to show that his discussions in the Metaphysics and Physics of form, matter, definition, and causation provide the background for his distinctive account of essentially embodied psychological activities in De Anima. These chapters contain a progressive elucidation of his key concepts: form and matter. The forms of natural substances, I argue in Chapter 2, cannot be defined without reference to matter in the way indicated because they are (at first approximation) inextricably material (or bodily) capacities. Subsequently, in Chapter 7, I suggest that Aristotle’s forms are best understood not strictly as capacities but as distinctive ‘activities’ (energeiai): those of being capable of acting or suffering in a variety of ways. Being capable of acting is not to be identified either with the capacity to act or with the acting that results. It might be better described as the way an object is when it possesses the capacity to act. In these sections of Chapters 6 and 7 my aim is to formulate, defend and motivate what is sometimes called ‘an impure form’ interpretation: the relevant forms cannot be defined without explicit reference in their definition to matter because they are inextricably material ways of being capable of varying actions. At the third stage, beginning in Chapter 7 and continuing into Chapter 8, Aristotle’s own account, as interpreted in Chapters 1–6, is contrasted with some contemporary theories and tested by considering criticisms that might be thought to refute it. If his distinctive viewpoint withstands these objections, as I suggest it does, it offers a credible way to dissolve the mind–body problem that we have inherited from Descartes.
The first two stages of enquiry call for careful reading of complex and difficult texts. Detailed exegesis of this type will appeal more to some than to others. But there is, it seems, no other way to secure a firm, historically sensitive, grasp on what is distinctive of Aristotle’s approach. In developing my interpretation, I have aimed to understand continuous passages in his argument (in, for example, De Anima A.1 and Metaphysics Z.10–11), not to take smaller pieces of texts in isolation from their extended context. I have also attempted to show the extent to which he employs the general approach sketched in these and other passages in his detailed discussion of specific psychological phenomena. Further, I have sought to articulate the philosophical motivations which led him to hold these views: to explain why he held them. All three ingredients are required if one is seriously to challenge now standard interpretations, sometimes based on a few isolated texts (or even fragments of such texts) studied in isolation from the overall flow of Aristotle’s argument in those passages with little concern for the theoretical considerations about form, matter, and causation that underlie them. However, it must be ceded, the resulting exegetical journey is long, arduous, and
in constant danger of being overwhelmed by interesting if peripheral detail.14 It may be helpful for those faced by this somewhat daunting prospect, if I offer a brief sketch of the interpretative arguments developed in Chapters 1–6. Those interested in the philosophical ‘pay-off’ may choose to rest content with this overview and Chapter 1, skip the exegesis and move directly to Chapters 7 and 8.
In Chapter 1, I shall suggest that Aristotle’s account of ‘passions common to body and soul’, such as the emotions and perception, in De Anima A.1. (i) is consistent with the inextricabilist interpretation just sketched and (ii) on balance supports it over its closest competitors. Since the form of anger is defined as an embodied desire for revenge, it is (what I shall call) an impure, essentially enmattered, form: one which cannot be defined without explicit reference to its being material. Aristotle’s discussion of anger and fear provides an intuitive introduction to his ‘simple theory’ of a wider range of psychological phenomena. However, partly because De Anima A.1 is introductory, it does not provide conclusive evidence for the inextricabilist interpretation of these emotions or the other passions at issue. Perhaps, even as a whole, it can be read, albeit with some difficulty, consistently with other interpretations. Or perhaps it does not give Aristotle’s final view of these or other passions, such as desire or perception. This chapter leaves important questions unresolved.
In Chapter 2, I suggest that Aristotle understood the forms of natural substances in the Metaphysics and elsewhere as impure, inextricably enmattered, forms (on the model of snubness already mentioned). I argue that he was motivated to do so because he saw them as efficient and teleological causes. Forms of this type are required as the direct and specific (per se) causes of bodily or, more generally, material phenomena. They are, at least at first approximation, inextricably material capacities. His general account of hylomorphic substances, based on his views of definition and of causation, both efficient and teleological, provides the theoretical background for his views of passions ‘common to body and soul’ in De Anima A.1. In the final sections of Chapter 2, I outline the view of matter which guides this account and compare my interpretation of Aristotle’s views of form and matter with some influential alternatives.
In Chapter 3, in considering Aristotle’s account in De Anima and De Motu of how desire leads to action, I offer further considerations in favour of the inextricabilist interpretation of the relevant phenomena. More specifically, I suggest that this account
14 I am painfully aware of many important detailed interpretative issues which will be not pursued (or not pursued fully) in this investigation. However, my aim is a narrow one: to formulate, and argue for, a distinctive, inextricabilist, overview of Aristotle’s account and to engage in questions of exegetical detail only to the extent that they are required to achieve this goal. If the resulting interpretation is correct, it will need to be supplemented, and refined, by careful analysis of many specific points and of other areas of his thought, especially his biological thought. Conversely, if it is mistaken, it cannot be rescued by the accumulation of further scholarly detail.
(a) offers a simple and complete account of how desire, in Aristotle’s view, leads to action, while its competitors are confronted with an unnerving silence at points where a theory is most needed; and
(b) enables us to understand what was, in Aristotle’s view, mistaken in the other accounts, including the ‘harmony’ theory which, in certain crucial respects, resembles that offered by the non-reductionist materialist.
Both arguments rest on a further claim:
(c) In the case of anger, desire for revenge, the relevant form, is, for Aristotle, the direct and specific (per se) efficient cause of action and, as such, has to be an enmattered, impure, form. Since the form of such desires (defined in part in terms of what they are for: revenge in the case of anger) is the specific cause of bodily movement, it must be an enmattered or impure form. This is, I suggest, an application to the case of desire of his general theory of hylomorphism and causation developed for natural substances, sketched in Chapter 2.
In Chapters 4 and 5, I argue that Aristotle’s discussion of perception of flavours, odours, sounds, and colours follows, in general outline, his account of anger and desire. In the case of taste and sight, for example, the relevant types of awareness and responsiveness are defined as inextricably embodied, the realization of distinct enmattered goal-directed capacities. Capacities of this kind allow us to discriminate flavours and colours, defined as enmattered properties of water or light, which impact on our senses in water- and light-based ways. The connections that Aristotle envisaged between enmattered forms and per se efficient causation, outlined in Chapter 2, are at work in this account of the distinctive capacities, processes, and activities essential to perception. Inextricably embodied capacities and activities are required as the relevant effects of essentially enmattered causal processes originating in appropriately enmattered causes. His relatively simple and intuitive theories of tasting, seeing, and the rest are underwritten by a more general understanding of definition and causation and of the ontology of forms, activities and processes they require. As in the case of desire, the inextricabilist interpretation offers a simple, complete, and theoretically based account of Aristotle’s views at points where other interpreters see (and struggle to fill) unnerving gaps and silences.
In Chapter 6, in considering other aspects of Aristotle’s discussions of perception, desire, and action, I suggest that the inextricabilist interpretation offers a convincing account of how these are integrated as the activities of essentially and inextricably embodied and unified subjects. I also consider, somewhat more briefly, the extent to which this style of interpretation is applicable to Aristotle’s view of human thinking and of the soul more generally. My main aim in this chapter, as in Chapters 1–5, is to formulate and defend an inextricabilist interpretation of
Aristotle’s view of the emotions, desire, perception, and imagination, ‘passions common to body and soul’, and to show its basis in his account of the essentially enmattered, ‘impure’, forms of hylomorphic natural substances. This interpretation offers a textually secure and properly grounded reading of Aristotle’s discussion of the relevant psychological phenomena which enjoys major advantages over its main competitors, spiritualist, functionalist, non-reductionist alike. Or so I argue. Chapters 1–6 provide the exegetical basis for my attempt in the final two chapters to capture, in somewhat more general terms, what is distinctive about Aristotle’s account and to test the extent to which it is still philosophically defensible.
In Chapter 7, I seek to confirm that Aristotle’s viewpoint is not a version, or notational variant, of two options with which it has recently been associated: functionalism (in its various guises) and non-reductionist materialism. Further, I suggest that it differs, in somewhat parallel respects, from several prominent neoAristotelian versions of hylomorphism. In commenting on these differences, I further develop the interpretation of the form of natural substances set out in Chapter 2, focusing on its intrinsic nature as an inextricably material way of being capable of various activities, on the type of unity it possesses, and on the way in which it is prior to matter. My aim in this chapter is not to argue for the superiority of Aristotle’s conception of form or of the psychological but to isolate its distinctive commitments.
In the final chapter, I examine some considerations that challenge, and might be taken to refute, Aristotle’s account, as interpreted in Chapters 1–6. In doing so, I shall focus on those aspects of our psychological lives that seem to exemplify a purely psychological domain, defined without any reference to physical processes or properties. I also consider, and reject, the suggestion that Aristotle’s viewpoint is committed to a wholly unacceptable view of matter. My aim, building on the detailed exegetical work undertaken in the earlier chapters, is to offer an overview of his position and to seek, in some measure, to defend it.
What are the philosophical attractions of Aristotle’s account, as interpreted here? First, the descriptions it offers of anger and other emotions have intuitive appeal. When angry, we are tense, pained, anxious, and excited. We become agitated and distressed. The pain and tension influence how we speak and distort our facial expressions. Anger, as we experience it, is not just any type of desire for revenge, but one which, in its nature, is permeated with physical and psycho-physical features. One cannot, it seems, define the tension, anxiety, or excitement essential to this type of desire without reference to specific bodily states. It is, in its nature, an embodied response to what befalls us.15 It was, I suspect, because he shared this intuition that Aristotle began his discussion in De Anima A.1 by focusing on fear and anger, which he saw as clear cases of inextricably psycho-physical phenomena.
15 It is instructive to compare Aristotle’s account with Damasio’s description of the role of the body in emotional responses. See, for example, his (2000: chapter 9).
Aristotle, however, had a more theoretical motivation for his position. He wished to account for the causal role, especially the efficient causal role, of the relevant type of desire for revenge. It is because it is an embodied type of desire that it affects our bodies and bodily movements in the way it does. Thus, Achilles’ anger led him not merely to kill Hector but to desecrate his corpse and leave it to rot outside the walls of Troy. The bodily pain and tension he experienced at the death of his friend Patroclus, modified, and in some measure controlled, his subsequent actions. Anger has to be a distinctive bodily type of desire for revenge, itself possessed of an impure form, to account for these consequences. In fact, as he suggested in De Anima A.1, it is because it is defined as a bodily type of desire of this type that anger sometimes arises and sometimes does not. If one is calm, relaxed, at ease, he argued, one may not be provoked by phenomena which would spark anger if one was initially tense, excited, or agitated. Sometimes, in the latter states, one can become angry when there is nothing to merit it. Indeed, in some cases, one’s pathology alone, in the absence of any external stimulus, explains the onset of anger. It is because, Aristotle thought, the type of desire involved is, in its nature, bodily that anger arises in the way it does. It must be defined as an essentially bodily type of desire to have as its specific direct causes (and effects) bodily states.
Aristotle’s line of thought rests on two specific features of his account of efficient causation.16 First, the cause and the effect in question have to be of the same general type, possessed of matter of the same general type. Only something which is in its nature bodily can be the direct and specific (per se) cause of bodily changes, one which is adequate to bring them about. Such causes are not only necessary for their effects but adequate to account for their occurrence. If the effect is bodily, the relevant cause must also be and, similarly, if the cause is bodily, so too must be the effect. More generally, if either the cause or the effect is material, the other must be also. There has to be, at some generic level, a type of matter which both share. It is only in virtue of being enmattered in this way that the type of desire for revenge which defines anger can be a per se cause of certain bodily changes and the per se effect of others. Its form, the basic cause, must be inextricably psycho-physical to be a per se cause and per se effect of this type.
Second, to be a per se cause the type of desire in question has not only to initiate the activities that ensue but also to guide and control them. Thus, it was because Achilles desired revenge in this way that he not only fought Hector but killed him in the brutal way he did. This was why he was not easily swayed by pity or by Priam’s tears. It was in virtue of the specific nature of his desire that he acted throughout as he did, relenting only when it was satisfied or modified by entreaty from others or his own sense of justice. His action was an activity directed by, as
16 For further discussion of Aristotle’s activity/process-based view of per se causation, see Chapter 2, Section 6.
well as arising from, this specific type of desire. It is not just that had he not been angry, he would not have killed Hector. Instead, his anger made a difference to how his actions unfolded, determining their order, many of their properties and when they ended. Achilles’ anger was—in the ways described—the controlling cause of what he did, when and how he did it, and of when he stopped doing it. Per se causes, the forms at issue, are not only the starting points of the processes or activities they initiate: they are also, in the ways indicated, their controllers.
Before the walls of Troy, the per se cause of the unfolding drama was the realization, or actualization, of Achilles’ relevant capacity to desire revenge. This capacity must be an essentially material, embodied, capacity to generate bodily results of the type just specified. Its actualization is also sensitive, as the story develops, to reason and pity. Indeed, it is because the capacity is responsive to reason that the resulting bodily activities unfold as they do. This goal-directed capacity is, in its nature, inextricably psycho-physical. It has to be such to be a per se cause of bodily activity which is responsive to reason and pity in the ways indicated.
Aristotle’s account has several distinguishing features. First, what is caused is an activity (or process) that stretches through time, developing and being modified as it develops. The effect is not an event, such as the movement of one’s hand (or the crossing of the road or even the killing of Hector), but rather an activity, such as Achilles’ moving his hand (road-crossing or killing Hector), which continues through time and (if all goes to plan) achieves its goal. Correspondingly, the per se cause is understood as the embodied capacity to be angry: for it is this capacity that, when Achilles decided to act, guided his developing action. The per se cause was not, for example, simply his decision to strike Hector, which initiated the event or chain of events in question. The per se cause, in Aristotle’s account, is that in virtue of which the resulting activity began, continued and ended in the way it did.
The per se causal role of anger will be lost if one takes it to be a purely psychological phenomenon, defined without reference to the physical, which happens to be realized in a purely physical state or capacity that is the per se cause of the resulting bodily movement. In the latter model, the resulting action would not follow in virtue of the presence of a desire for revenge, sensitive to reason and pity in the way just noted. Equally, a purely material or physical capacity, defined without reference to these factors, could not account for why Achilles’ anger developed or ceased in the way it did. A purely material capacity of this type, when actualized, would not be, in the sense explained, the cause of the specific type of action that resulted. The cause needs to be an inextricably psycho-physical, embodied, capacity to account for what unfolded in the way it did. It is this view of causation that underpins, and motivates, Aristotle’s discussion. If the form is to be identified with being capable of acting in these ways, it too must be essentially embodied if it is to play the required causal role. If the capacity for action is essentially embodied, so too will be being capable of so acting.
Aristotle had a further motivation. The emotions, desire, and perception are defined as the inextricably psycho-physical activities (and capacities) of unified,
essentially psycho-physical subjects. This is one part of his organism-based account of a wider range of biological and (in his sense) physical phenomena. In his model, the matter involved is defined as the matter of specific psycho-physical organisms, not in terms of a general science of matter, which accounts for all change in terms of universal laws that apply to all matter, whether in animate or inanimate substances. It is the latter, with the distinctive view of the physical it entails, that helped to generate Descartes’ mind–body problem. How do the states defined by a universal and complete physics of this type give rise to psychological activities?
There is a major shift in perspective from Aristotle’s organism-based account to one defined by a physics of the type just described. But are they fundamentally incompatible? Did the development of a modern view of the physical render Aristotle’s viewpoint obsolete? Or can they be seen as consistent, indeed complementary?
One might characterize parts of our post-Cartesian perspective, in rough outline, as follows:
1) There is, or will hopefully eventually be, a general and complete physical theory taking as its starting point matter defined without reference to sentience (or life more generally).
2) The general theory is expressed in terms of events and the physical laws covering such events.
3) There are,above the most fundamental physical level,a series of other levels, including the biological and the psychological, which—in some way—rest on, or are grounded in, it.
Is the Aristotelian viewpoint, in which inextricably psycho-physical processes and capacities are taken as basic, inconsistent with this perspective? Do we have to choose between them?
Although I shall discuss these questions in the final chapter, it may be helpful to offer a preliminary overview of the issue at stake. If one begins with inextricably psycho-physical activities and capacities, it is possible to view these (in a given way) as physical events or states with purely physical features, as it is possible to see snubness as a case of geometrical concavity (not as nasal-concavity). We can achieve this viewpoint by isolating certain aspects of the relevant activities and characterizing them in a purely physical way. (Aristotle describes this as a type of ‘abstraction’.) We may then be able to construct, on this basis, a general theory of matter applicable to all physical events (so understood), wherever they occur. In this way, we can redescribe the psycho-physical activities in terms of purely physical events and properties and seek laws that apply to them, wherever they are found.
The perspective just sketched becomes available if we are able, in good philosophical conscience, to take as basic the inextricabilist viewpoint and see physical theory as—in a sense to be explained—an abstraction. If we do so, we might even
accept as a constraint on an adequate universal physics that it be consistent with the organism-based laws and regularities that Aristotle took as his starting point. Problems arise only when we reverse this order of explanation and take the basic physical level thus abstracted (whatever it may turn out in the end to be) as the starting point for our account of everything psychological or biological. When we do so, we are faced with some well-known problems:
(a) The danger of unexplained gaps between the different levels (of which our mind–body problem is a striking example),
(b) The difficulty of accommodating the causal role of the entities at the higher level, and
(c) The loss of explanatory power provided by talk of organisms, their capacities and the goods they are set up to achieve.
These problems arise when we take what is physically, or compositionally, basic (whatever that turns out to be) as metaphysically basic and attempt to account for all else in terms of it. They can be avoided if we can construe inextricably psychophysical activities and capacities as the metaphysically basic starting point and separate what is metaphysically basic from what is compositionally basic. From this viewpoint, it is not the scientific picture, as defined by (1)–(3) above, that needs to be rejected, only a form of ‘scientism’ that requires us to take what is compositionally basic (whatever that may be) as metaphysically basic. It is the order of explanation that is at issue.
Aristotle’s perspective has three complementary aspects: (i) his simple theory of the relevant psychological phenomena and their role as causes and effects of enmattered processes, (ii) his general view of the matter and form of natural substances (and their analogues for natural processes) and their central role in definition and causation, and (iii) his conception of what should be taken as basic in a good metaphysical account of natural substances and their processes and activities. (ii) enabled him to integrate his simple theory of the emotions, desire, and perception as inextricably psycho-physical with his understanding of form, definition, and causation. (iii) provided the resources to justify taking as metaphysically basic aspects of his simple theory, now supported by and expressed in terms of the considerations assembled in (ii).
Aristotle’s viewpoint, if adopted, has its advantages.
(a) It avoids the seemingly insoluble problems of post-Cartesian picture: the attempt to account for the psychological on the basis of the purely physical. In recent years, as the reductionist project has stalled, we have been left with brute, unexplained supervenience claims linking the various levels. Aristotle himself compared a similar picture to a badly constructed, episodic, tragedy!
