Psychology and Value in Plato, Aristotle, and Hellenistic Philosophy: The Ninth Keeling Colloquium in Ancient Philosophy 0192858106, 9780192858108

Citation preview

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

Psychology and Value in Plato, Aristotle, and Hellenistic Philosophy

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

Stanley Victor Keeling

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

Psychology and Value in Plato, Aristotle, and Hellenistic Philosophy The Ninth Keeling Colloquium in Ancient Philosophy Edited by

FIONA LEIGH AND MARGARET HAMPSON

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

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 © Oxford University Press 2022 The moral rights of the authors have been asserted First Edition published in 2022 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: 2022943043 ISBN 978–0–19–285810–8 DOI: 10.1093/oso/9780192858108.001.0001 Printed and bound by CPI Group (UK) Ltd, Croydon, CR0 4YY Links to third party websites are provided by Oxford in good faith and for information only. Oxford disclaims any responsibility for the materials contained in any third party website referenced in this work.

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

In memoriam Sarah Broadie (03 November 1941–08 August 2021)

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

Contents Preface List of Contributors Keeling Colloquia List of Abbreviations

1. Psychology and Value in Ancient Greek Philosophy Margaret Hampson

ix xi xiii xv

1

2. Intellectualism and the Method of Hypothesis in Plato’s Early Dialogues Rachel Barney

21

3. Comments on Rachel Barney, ‘Intellectualism and the Method of Hypothesis in Plato’s Early Dialogues’ Terence Irwin

54

4. Against Bare Urges and Good-Independent Desires: Appetites in Republic IV Jessica Moss

67

5. The Blind Desires of Republic IV Matthew Evans 6. Comments on Matthew Evans, ‘The Blind Desires of Republic IV’ and Jessica Moss, ‘Against Bare Urges and Good-Independent Desires: Appetites in Republic IV’ Mary Margaret McCabe 7. Plato on the Object of Thirst: Comments on Jessica Moss, ‘Against Bare Urges and Good-Independent Desires: Appetites in Republic IV’ A. W. Price 8. Courage and Pleasure in Aristotle’s Ethics Raphael Woolf 9. Comments on Raphael Woolf, ‘Courage and Pleasure in Aristotle’s Ethics’ Sarah Broadie

82

97

105 114

134

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

viii



10. Memory, Anticipation, Pleasure James Warren 11. Modelling the Memory and Anticipation of Pleasure: Comments on James Warren’s ‘Memory, Anticipation, Pleasure’ Margaret Hampson and Katharine R. O’Reilly 12. Three Mistakes about Stoic Ethics Daniel C. Russell

141

170 183

13. Comments on Daniel C. Russell, ‘Three Mistakes about Stoic Ethics’ David Sedley

203

Index of Passages Cited General Index and Index of Names

209 217

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

Preface The chapters in this volume, with one exception, arose out of the ninth Keeling Colloquium in ancient philosophy, entitled ‘Moral Psychology in Ancient Thought’, which took place at University College London (UCL) in 2011. (The exception is chapter 11, the piece responding to James Warren’s paper in chapter 10, ‘Memory, Anticipation, Pleasure’ by Margaret Hampson and Katharine O’Reilly, commissioned after the colloquium.) As a full decade has passed since the colloquium and publication, the authors have revised their papers—in some cases, considerably—with the respondents revisiting their comments accordingly. The colloquium was structured around the paper-response format, which has been preserved in the volume. Most papers are paired with a single response: So, Terence Irwin responds to Rachel Barney’s paper, Sarah Broadie responds to the paper by Raphael Woolf, Margaret Hampson and Katharine O’Reilly respond to James Warren’s paper, and David Sedley responds to the paper by Daniel C. Russell. However, two papers, whose authors, Jessica Moss and Matthew Evans, were both invited to speak on the same topic, that of the good in book IV of Plato’s Republic, are discussed together by the one respondent, M. M. McCabe, while Anthony Price offers a response mainly to Jessica Moss’s paper, but also discusses some aspects of the paper by James Warren. In preparing the volume for press, Margaret Hampson has edited the papers with me, and has also written the introduction. Acknowledgement and sincere thanks are due to the contributors to this volume, particularly for their not inconsiderable patience over the decade it has taken for their arguments and interpretive interventions to appear. I am grateful, as always, to M. M. McCabe for her guidance and encouragement, and to Peter Momtchiloff at OUP for his assistance and valuable advice in establishing the new Keeling Colloquia series, of which this volume represents the second instalment. As co-editors, Margaret and I are grateful to Saloni de Souza for her excellent research assistance in producing this volume, especially the indices, and, on a more personal note, to Jerome and Charlotte for their general forbearance and unfailing support. Sarah Broadie very sadly and quite unexpectedly died just as these papers were going to press. To the contributors for this volume, she was, variously, friend, teacher, colleague, and mentor—razor sharp and quick witted, as well as generous and kind. As the numerous memorials to her now being prepared will attest, Sarah made a huge contribution to the scholarship in ancient philosophy,

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

x



particularly to our understanding of Plato and Aristotle, and will long be remembered as one of the field’s brightest lights. This volume is dedicated to her memory. FL, London 2021

*** Stanley Victor Keeling was a lecturer and reader in the Department of Philosophy at University College London until his retirement in 1954, where, during World War II, he also served as head of department. Upon his death in 1979, his partner having predeceased him, Keeling left his estate to a friend and former student with the wish that, if possible, he would like the establishment of an annual lecture on ancient Greek philosophy given at UCL by a distinguished scholar of international note in the field. This friend (who wished to remain anonymous) generously supplemented Keeling’s estate, making possible not only the annual S.V. Keeling Memorial Lecture in Ancient Philosophy (since 1981), but also a series of Keeling Colloquia in Ancient Philosophy (the first in 1994), and the Keeling Graduate Scholarships (since 2008). In 2013, the anonymous donor passed away and left a further legacy which resulted in an expansion of the Graduate Scholarships programme, and the creation of the Keeling Centre in Ancient Philosophy at UCL in 2016. In addition to the annual memorial lecture, colloquia, and scholarships programme, the Keeling Centre now also hosts a Keeling Scholar in Residence, an annual Graduate Conference in Ancient Philosophy, a Keeling Research Fellow (from time to time), occasional visiting academics, and supports numerous events in ancient philosophy in and around London. Curiously, S.V. Keeling did not himself specialize in the field of ancient philosophy. Educated at Trinity College Cambridge (BA Philosophy), UCL (MA Philosophy), and Toulouse-Montpellier (Doctorat ès lettres), Keeling’s philosophical work was for the most part centred on Descartes and McTaggart. His principle published works were an annotated edition of McTaggart’s work, Philosophical Studies (Edward Arnold, London: 1934), a monograph entitled Descartes (Ernest Benn, London: 1934), and the 1948 annual British Academy Master Mind Lecture, which Keeling gave on Descartes (Proceedings of the British Academy 34 (1948), 57–80). Keeling nonetheless had an abiding affection for, and a firm belief in the central importance of, ancient Greek philosophy. It is said that in Paris, where he moved after retirement and remained until his death, he and his wife often read ancient Greek philosophy to one another in the evening after dinner. This was the period in which he conceived his wish to foster and promote ancient philosophy at UCL, for the benefit of students and academics at UCL, but also in London more generally.

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

List of Contributors Rachel Barney, University of Toronto Sarah Broadie, University of St Andrews Matthew Evans, University of Texas (at Austin) Margaret Hampson, University of St Andrews Terence Irwin, University of Oxford Fiona Leigh, University College London Mary Margaret McCabe, King’s College London Jessica Moss, New York University Katharine R. O’Reilly, Toronto Metropolitan University A. W. Price, Birkbeck University Daniel C. Russell, University of Arizona David Sedley, University of Cambridge James Warren, University of Cambridge Raphael Woolf, King’s College London

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

Keeling Colloquia 1st S.V. Keeling Colloquium, Aristotle and Moral Realism (1994), organized by Robert Heinaman. Published as Aristotle and Moral Realism, ed. R. Heinaman (UCL Press, London: 1995). 2nd S.V. Keeling Colloquium, Whose Aristotle? Whose Aristotelianism? (1998), organized by Robert Sharples. Published as Whose Aristotle? Whose Aristotelianism? ed. R. W. Sharples (Ashgate, Aldershot: 2001). 3rd S.V. Keeling Colloquium, Descartes and Ancient Philosophy (1999), organized by Gerard O’Daly and Martin Stone. 4th S.V. Keeling Colloquium, Plato and Aristotle’s Ethics (2001), organized by Robert Heinaman. Published as Plato and Aristotle’s Ethics, ed. R. Heinaman (Ashgate, Aldershot: 2003). 5th S.V. Keeling Colloquium, Philosophy and the Sciences in Antiquity (2003), organized by Robert Sharples. Published as Philosophy and the Sciences in Antiquity, ed. R. W. Sharples (Ashgate, Aldershot: 2005). 6th S.V. Keeling Colloquium, The Eudemian Ethics (2006), organized by Robert Heinaman. Published as The Eudemian Ethics on the Voluntary, Friendship, and Luck, ed. F. Leigh (Brill, Leiden: 2012). 7th S.V. Keeling Colloquium, Particulars in Greek Philosophy (2007), organized by Robert Sharples. Published as Particulars in Greek Philosophy, ed. R. W. Sharples (Brill, Leiden: 2009). 8th S.V. Keeling Colloquium, Self-Knowledge in Ancient Philosophy (2009), organized by Robert Sharples and Fiona Leigh. Published as Self-knowledge in Ancient Philosophy, ed. F. Leigh (Oxford University Press, Oxford: 2020). 9th S.V. Keeling Colloquium, Moral Psychology in Ancient Thought (2011), organized by Fiona Leigh. Published as Value and Psychology in Plato, Aristotle, and Hellenistic Philosophy, ed. F. Leigh and M. Hampson (Oxford University Press, Oxford: 2021). 10th S.V. Keeling Colloquium, Method in Ancient Philosophy (2013), organized by Jenny Bryan.

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

List of Abbreviations Collections of fragments, doxographia DK DL

Diels and Kranz, Die Fragmente der Vorsokratiker (The Fragments of the Pre-Socratics) Diogenes Laertius, Vitae Philosophorum (Lives of Eminent Philosophers)

Works of Homer Od.

Odyssey

Works of Plato Gorg. Phd. Phileb. Prot. Soph. Theaet. Tim.

Gorgias Phaedo Philebus Protagoras Sophist Theaetetus Timaeus

Works of Aristotle De an. De mem. EE EN Mem. MM NE PA Pol. Rhet.

De Anima (On the Soul) De Memoria (On Memory) Ethica Eudemia (Eudemian Ethics) Ethica Nicomachea (Nicomachean Ethics) De Memoria (On Memory) Magna Moralia Nicomachean Ethics (Ethica Nicomachea) De Partibus Animalium (Parts of Animals) Politics Rhetoric

Works of Epicurus Ep. Men.

Epistula ad. Menoeceum (Letter to Menoeceus)

Works of Virgil Virg. Aen Aeneid

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

xvi

  

Works of Arius Didymus AD

Arius Didymus

Works of Cicero Cic. Fin./Fin. De Finibus (On the Ends) Cic. Tusc./Tusc./Tusculans Tusculanae Disputationes (Tusculan Disputations) TD Tusculanae Disputationes (Tusculan Disputations)

Works of Marcus Aurelius Med.

Meditations

Works of Plutarch Plut. Quaest. Conv.

Quaestiones Convivales (Convival Questions)

Works of Epictetus Diss.

Dissertationes (Discourses)

Works of Sextus Empiricus M

Adversus Mathematicos (Against the Mathematicians)

Works of Augustine August. Serm

Sermons

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

1 Psychology and Value in Ancient Greek Philosophy Margaret Hampson

1. Introduction A familiar starting point for studies of ancient Greek ethics is the thought that the ancients were concerned principally with the kinds of lives that we lead. What a successful human life consists in and how such a life is to be achieved were central preoccupations of the writings of Plato, Aristotle, and Hellenistic thinkers. For Plato, Aristotle, and the Stoics, the best human life is a life of virtue, though their accounts of what virtue consists in or what a virtuous life involves differ in important ways. The Epicureans and the Cyrenaics, meanwhile, famously posit that a successful human life is a life of pleasure, though once again with divergent views as to how such a life is to be achieved. Whilst these schools of thought might endorse different views of the human good and a flourishing human life, insofar as these issues are central to their ethical thought—issues, that is, of the human good and the flourishing human life—each school also exhibits a concern with the nature of the ethical subject and her psychological dispositions. Ancient Greek ethical thought thus sees the birth, in so-called Western philosophy at least, of the study now known as ‘moral psychology’. Precisely what falls under the label of moral psychology may be understood more broadly or more narrowly. In its narrower sense, the study of moral psychology is concerned principally with the nature of motivation and reasons for action in the moral sphere. Understood more broadly, moral psychology encompasses the study of all those aspects of our psychology relevant to our moral lives: desire, the emotions, ethical knowledge, moral perception, moral imagination, and so on. Each of the above schools of ancient thought displays an interest in moral psychology understood in this broad sense. But we can also see these schools of thought, and their corresponding interest in human psychology, as unified in a more specific—albeit still quite general—respect. For each recognizes that the ethical subject, if she is to lead a flourishing life, must be able to identify and respond appropriately to what is determined to be of value. This being so, it is critical that we understand the way, or ways, we may engage with such value, how our capacities to so engage with value might best be developed, Margaret Hampson, Psychology and Value in Ancient Greek Philosophy In: Psychology and Value in Plato, Aristotle, and Hellenistic Philosophy: The Ninth Keeling Colloquium in Ancient Philosophy. Edited by: Fiona Leigh and Margaret Hampson, Oxford University Press. © Oxford University Press 2022. DOI: 10.1093/oso/9780192858108.003.0001

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

2   and how these capacities can be harnessed in such a way that we lead the best human life. What is identified as being of value varies according to each school of thought. Candidates include the good, the fine (kalon), the pleasant, or virtue and what participates in virtue.¹ Various too are the modes through which an ethical subject might engage, psychologically speaking, with such value. The ethical subject’s engagement might be intellectual: she might have knowledge of, or beliefs about, what is of value, and this intellectual grasp of value might itself enable or even constitute her flourishing. Or her engagement with what is valuable might be emotional: she might identify sources of harm or relief, and experience a corresponding emotional response, such as fear or confidence. Or again, she might engage with value—whether real or apparent—by way of her desiderative capacities, perhaps identifying certain objects or activities as valuable in some respect and, in virtue of this, engaging in the pursuit of these things. Moreover, her capacities for memory and anticipation might allow her to make contact, so to speak, with value that is temporally remote, and bring this to bear on her current state. The papers in this volume each deal, in some way, with issues surrounding our identification of and responsiveness to value in the writings of Plato, Aristotle, and the Hellenistic schools. Each paper, too, makes a significant contribution to scholarship on these issues, whether in challenging orthodoxies, making innovations in long-running debates, or mapping the philosophical treatment of a concept in such a way that invites us to approach that concept with fresh eyes. This introductory essay draws out some of the ways in which the issues of our identification of and responsiveness to value figure in these schools of ethical thought, and the contributions made by the papers in this volume to our understanding of the psychology of value in ancient Greek ethics.

2. Intellectualism The explicit concern, in ancient Greek thought, with how we should live is typically traced back to the figure of Socrates, particularly as he is represented in Plato’s dialogues (e.g. Gorgias 500c–d; Republic 352d). These dialogues seemingly identify the good human life with the life of virtue (e.g. Euthydemus 278e–283e; Gorgias 507c, 508a–b; Republic 353e–354a), with many dialogues thus organized around the questions of what virtue consists in and how it is to be acquired.

¹ This list is neither exclusive nor exhaustive. In particular, a given thinker might recognize various forms of value and a variety of objects and activities as valuable. For Aristotle, for example, the human good consists in a life of virtue; the actions that constitute virtuous activity have the quality of being fine, and both fine actions and the virtuous life as a whole are pleasant.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

      

3

In the early, so-called Socratic dialogues,² however, it is typically thought that we find a quite distinctive account of the virtuous life and corresponding psychology of the ethical subject, captured by the label ‘Socratic Intellectualism’. The foremost thesis of Socratic Intellectualism is the claim that virtue is a kind of knowledge. Associated with this is the thesis that becoming virtuous is a purely cognitive process, to be effected through teaching of some form. In addition to these claims are a number of further theses that are thought to flesh out the Intellectualist thesis: Socrates is taken to recognize only desires oriented towards the good and the existence only of a rational soul in human beings; he is taken to deny the importance of emotional, affective, and non-rational factors in both virtue and its acquisition; and finally, in seeming contradiction with the phenomena, he is thought to deny that there are any good-independent desires, that irrational passions can cause action, and that the phenomenon of akrasia ever occurs (e.g. Protagoras 352b–c). Thus, intellectual contact with value—in the form of knowledge of good (and bad) things—on its own determines how well our lives go, and non-rational forms of engagement with value are apparently ruled out as impossible or irrelevant. That we read the early Socratic dialogues in this way is due, to some degree, to the influence of Aristotle. Reporting the views of Socrates³ in his Nicomachean and Eudemian Ethics Aristotle writes that Socrates ‘thought that all the virtues are kinds of wisdom [phronēseis]’ (NE 1144b19–20) or ‘kinds of knowledge [epistēmas]’ (EE 1216b6), stressing that on Socrates’ view, the virtues are thus acquired by coming to know them.⁴ As Aristotle also acknowledges—and as any student of Plato will know—these views appear to be abandoned by Plato in his so-called middle period, most notably in Republic IV with its introduction of the appetitive and spirited parts of the soul. The acknowledgement of the existence, and relevance to ethical discourse, of these parts of the soul already represents a seeming advance on the Intellectualist thesis which acknowledged only the existence of a rational soul. Moreover, not only is the soul constituted by more than the rational part, but Plato also recognizes that the spirited and appetitive parts are important sources of motivation. Perhaps most striking is the fact that Plato argues for the existence and motivational efficacy of the non-rational parts of the soul on the basis of phenomena denied by Socratic Intellectualism, namely

² In this list are typically included the Apology, Euthyphro, Crito, Protagoras, Charmides, Laches, Euthydemus, Hippias Major, Hippias Minor, and Ion. In addition to these, Barney in this volume adds (as perhaps somewhat later or signalling a transition to a new phrase) the Gorgias, Lysis, and Meno. For an overview of the issues surrounding the grouping and classification of the Platonic dialogues, see Irwin (2008). ³ It is unclear whether Aristotle has in mind the historical Socrates or the character of Plato’s dialogues. In discussing ‘Socratic Intellectualism’ Barney uses ‘Socrates’ for Plato’s character, unless otherwise indicated. ⁴ Other Aristotelian evidence of the theses that are thought to constitute Socratic Intellectualism can be found at NE 1145b23–7; MM 1182a17–26.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

4   observations about mental conflict, akrasia, and irrational desires; where Socratic Intellectualism is content to deny the real existence of these phenomena, despite their apparent existence, Republic IV takes the existence of these phenomena as given, and as the basis for his arguments about the constitution of the soul (439c–441b). With the appetitive and spirited parts of the soul now firmly on the table, Republic IV develops a picture of virtue which, in contrast to the view that virtue is knowledge, takes virtue to consist in the harmonious state of the soul, with each part performing its proper function, under the direction of reason (441e). The Republic also presents a picture of moral education which, although intellectualist in some respects,⁵ emphasizes the importance of non-rational training, particularly with regard to pleasures and pains. According to the received reading, then, a dramatic shift occurs in Plato’s dialogues, from the endorsement of Intellectualism in the early Socratic dialogues, to the abandonment of this thesis in favour of a more nuanced psychology and ethical outlook in the Republic. In the first paper of this volume, ‘Intellectualism and the Method of Hypothesis in Plato’s Early Dialogues’, Rachel Barney takes issue with this received interpretation of the early Socratic dialogues and their relation to Republic IV. Barney does not deny the significance of Republic IV in the development of Plato’s ethical and psychological thought, but urges us to resist seeing this as a rebuttal, by Plato, of an Intellectualist view endorsed by the Socrates of his early dialogues. For, Barney contends, many of the claims that are taken to constitute Socratic Intellectualism find no clear statement in the early dialogues, nor are they ever presented by Plato as anything resembling doctrine. Moreover, the traditional, Aristotelian, reading of the early dialogues actually distorts what Intellectualism is, properly understood. For its central thesis, Barney argues, is not Socratic as such, but rather the Protagorean claim that virtue is a kind of craft (technē)—a craft that is sufficient for successful living. This thesis, Barney suggests, is posed in the early dialogues as a hypothesis, with the characters exploring what follows from the hypothesis if true, in order to test for its acceptability. Through an examination of the method of hypothesis as presented in the Meno and Phaedo, Barney identifies a number of key features of this method, all of which are present, she argues, in the early dialogues. Returning to the Intellectualist hypothesis—the hypothesis that virtue is a kind of knowledge or technē—Barney goes on to show that this hypothesis, if true, will have consequences, amongst others, for the teachability of virtue, the content of virtue, the unity of the virtues and their attainability. These consequences, following from the presumed truth of the hypothesis, serve to support and clarify one another and indeed to disarm potential objections to the original hypothesis. The other theses traditionally associated with Intellectualism, meanwhile—the thesis ⁵ Famously, the importance of mathematical training is emphasized in the account of the education of the Guardians 525a–531e.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

      

5

that akrasia is impossible; the thesis that no one errs willingly; and the thesis that all desires are for the good—stand somewhat apart from the consequences that follow more directly, with the thesis that all desires are for the good serving as a kind of first principle and a higher hypothesis which lends support to the Intellectualist thesis. The results of Plato’s investigation of the Intellectualist hypothesis, Barney argues, are manifest in the Republic, though whether we take the Republic to be affirming or denying the Intellectualist hypothesis depends on our understanding of the relation between the Republic’s account of virtue in Book IV and its discussion of knowledge of the good. In his response to Barney, Terence Irwin asks whether (Aristotelian) Socratic Intellectualism is, as Barney had suggested in her original presentation of her paper, such a ‘boring and obvious falsehood’ that we should wish to resist ascribing it to Socrates as doctrine. Appealing to Aristotle’s own treatment of the thesis, its later endorsement by the Stoics, and the strength of the philosophical considerations adduced in support for elements of the thesis in Plato’s own dialogues, Irwin suggests it is not. As to the claim that Plato’s Socrates never defends Intellectualism as doctrine, Irwin asks whether Socrates’ critical and inquiring approach to the thesis that virtue is knowledge is incompatible with his endorsing this thesis as doctrine, and draws on evidence from the Gorgias and the Crito to support the thought that Socrates often maintains a confident but tentative attitude to theses he defends as such. Indeed, Irwin argues, even if Barney is right to say that Intellectualism is often presented as a hypothesis, this is not incompatible with laying that thesis down as doctrine, and in fact—on one understanding of how hypotheses work, derived from the Phaedo—requires this. Finally, Irwin considers whether the contrast Barney identifies between the functionalist, Homeric, conception of virtue, and the moral, Hesiodic, conception accurately captures Socrates’ concerns when he posits that virtue is knowledge. For as Irwin explains, Socrates is not only seeking to reconcile justice and happiness, but—since the demands of justice might often be taken to point in different directions—to say what is required by justice that is sufficient for happiness. And once we correctly identify the true means to happiness, and lose our false beliefs about what justice demands, we will see that Socrates’ choices are both just and conducive to happiness.

3. Appetites Whether the ethical and psychological theory of the Republic should be seen as a denial of a thesis previously endorsed by Socrates, or Plato’s answer to a series of philosophical experiments, it remains the case that, with its explicit introduction of non-rational sources of motivation and incorporation of these into an account of the flourishing life, Republic IV represents a pivotal moment in the history of

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

6   ethics and moral psychology. Yet in allowing for the possibility, and ethical relevance, of non-rational forms of motivation—particularly those issuing from the appetitive part of the soul—we are forced then to reckon with questions concerning the nature of such motivation and the part of the soul from which appetitive desires issue. One question that arises when we turn our focus to the appetitive part of the soul concerns how much, cognitively speaking, can be achieved by this part of the soul. This in turn has implications for how we understand the nature of appetitive desires and whether, in particular, such desires are capable of representing their objects as in any way ‘good’. In her paper ‘Against Bare Urges and Good-Independent Desires: Appetites in Republic IV’, Jessica Moss argues against two commonly held theses concerning appetitive desires, based on what she argues are mistaken assumptions about the appetitive part of the soul. These two theses are, first, that appetitive desires are ‘bare urges’ or ‘simple desires’, and, second, that these desires are goodindependent. In a well-known passage in Republic IV that argues for a division of the soul into three (the ‘Division’ passage), Plato has Socrates claim that thirst itself is, by nature, for drink itself. Certain scholars have thus taken this claim to reveal something about the class of appetitive desires in general: that these desires are not for qualified objects (such as cold drink, or sweet drink) but only for unqualified objects (such as drink simpliciter). This is because the appetitive part of the soul is thought to be too cognitively deficient to grasp qualifications such as ‘cold’ or ‘sweet’. Yet such a reading, Moss points out, conflicts with those passages in which Socrates does describe appetites for complex objects, as well as rendering the appetitive part of the soul motivationally helpless (or at least requiring assistance from reason). As Moss goes on to show, Plato appeals to the existence of unqualified appetitive desires as part of his argument for the division of the soul, but this appeal is only intended to show that appetites can be for unqualified objects, such as drink itself, in order to generate a conflict between appetite and reason. Moreover, that there are some unqualified appetitive desires does not entail that appetitive desires are cognitively limited; a desire for an unqualified object is not necessarily unable to discriminate species of that object, it is rather a desire for something more general or abstract. Finally, she argues that the existence of such unqualified desires is not peculiar to the appetitive part of the soul: it is equally possible to have unqualified rational or spirited desires, such as desires for knowledge or honour in general. Having established that appetitive desires are neither necessarily nor uniquely for unqualified objects, Moss turns her attention to a second common supposition about appetitive desires, namely that such desires are good-independent. One can have an appetitive desire, so the thought goes, without finding the object of desire good (perhaps, again, on the assumption that the appetitive part of the soul is incapable of representing objects as good). If this is so, one might wonder how, if at all, we should understand the role of such desires in an account of a good

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

      

7

human life. Yet this interpretation, Moss argues, does not follow straightforwardly from the text, and moreover is undermined by the reading of the Division passage argued for in the first part of her paper. For as we have seen, in the Division passage Socrates only needs to show that some appetites (or indeed rational or spirited desires) can be for unqualified objects, and it does not follow from this that in desiring such objects, the subject does not find these, in some way, good. Moss argues that in desiring drink, for example, a subject must see the drink as good under some guise—perhaps as pleasant. On this view then, when we desire something appetitively, we must evaluate that object as good in some respect (namely, as pleasant), but it does not follow from this that the object is in fact good, or even that our non-rational evaluation of the object as good must correspond to our considered beliefs about what is in fact good. In his paper ‘The Blind Desires of Republic IV’, Matthew Evans also takes up the issue of how to understand the relation, for Plato, between desire and goodness, particularly in the case of appetitive desires. As will already be apparent from Moss’ discussion, it is well evidenced that Plato is committed to the view that all human motivation is fundamentally linked to the good in some way; that all human motivation is, in some way or other, guided by, fixed upon, directed at, or oriented towards the good. The challenge for commentators is how to fill out this basic thesis concerning the connection between desire and goodness in Plato. Many commentators have supposed that in linking desire to the good in some fundamental sense, Plato is placing a restriction on the sorts of things that humans can genuinely desire: we desire things only insofar as we evaluate them as good. More specifically, it has been suggested that Plato holds the view that for a psychological state to be a desire is for that state to be a perception of some object as good.⁶ Yet this account, argues Evans, comes into conflict with a central thesis of the moral psychology of Republic IV, namely that some desires are, as it were, ‘blind’. Whilst thirst of a certain sort is for drink of a certain sort, Socrates maintains that thirst itself is for drink alone. Indeed, Evans argues that the ‘perception account’ entails commitments that are in direct conflict with Socrates’ line of argument in Republic IV, in which he argues that the soul is constituted by three distinct parts. Thus, Evans argues that Plato’s basic thesis concerning the connection between motivation and the good should be read not as a restriction on the sorts of things that humans could genuinely desire, but rather on the things that humans could correctly desire. Plato is committed to the view that, for all desires, if the content of a desire is good, it is a correct desire, and if the content of the desire is bad, it is an incorrect desire. On this view, the motivational force of a desire is independent of whether the object of desire,

⁶ This view is developed and defended by Moss (2008, 57–60), and is implicit in the account presented in her contribution to this volume.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

8   specified in the content of the desiring state, is good, or perceived as good. Properly speaking, though, we ought only to desire the good.⁷ According to Evans, then, the connection between goodness and motivation endorsed by Plato in the Republic is weaker than scholars have supposed it to be. Plato’s account, however, is by no means trivial. For his account yields an important insight, namely that desires could not possibly serve as the source, or the basis, of the goodness of things,⁸ on the grounds that part of what it is to be a desire is to be normatively responsive to the goodness of things. Plato’s account of desire in the Republic and elsewhere is thus closely linked, Evans concludes, to his profound commitment to the desire-independence of all goodness. In her response to Moss and Evans, M. M. McCabe considers both the points of contact and of divergence between the two papers, and asks what is at stake in interpreting these complex passages. Where Moss and Evans’ views appear to diverge is in regard to the content of desire: Moss takes appetitive desires to be complex and cognitively rich, whilst Evans appears inclined towards the view that appetites do not have complex content.⁹ One thing that is at stake here, McCabe argues, is the issue of development. The psychology of Republic IV is offered not just as an analogue of the tripartite state, but also as part of a discussion of how and why someone might become good; thus, it is a condition of the moral psychology—and so any interpretation of it—that it can account for moral development. Yet it is hard to see how appetite, if it consists in bare urges, could ever be susceptible to development. If we are to think of the whole soul as susceptible to development and orientation to the good, we might think that this requires that each part of the soul is susceptible to reason by virtue of its content, and thus not bare. Where both Evans and Moss agree—as we have seen—is on the view that (for Plato) all motivation is orientated towards the good (whilst denying that the good must figure explicitly in the content of desire). This, they take to be in evidence in Book VI of the Republic, and in the apparent agreement between Socrates and his imagined opponent in the Division passage that ‘everyone after all desires good things’. But it is not clear, argues McCabe, that the Book VI passage does support this view, in the sense that this is agreed to by Socrates in Book IV. Rather, ⁷ Evans acknowledges that an alternative interpretation of Plato’s conception of the connection between desire and motivation is possible, according to which Plato, in the Republic, maintains that for a psychological state to be a desire is for it to be generated by, and to motivate its bearer in accordance with, a belief to the effect that something is good. This view is compatible, Evans points out, with the view that the contents of some desires are for mere objects, and not objects under a guise: one could consistently hold that thirst is always generated by a belief to the effect that drink is good, whilst maintaining that thirst itself is always for mere drink. Evans argues, however, that this ‘belief account’ of the nature of the desire should be read as an extension of his preferred account, rather than a rival to it. ⁸ Evans cites Protagoras and Democritus as possible ancient proponents of this view. Modern proponents might include Lewis (1989), Smith (1994), and Street (2008). ⁹ McCabe asks, in her responses, what is the content of desire on Evans’ view?

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

      

9

McCabe suggests, the elaborate psychological picture that is developed from Book IV to the end of the Republic is Socrates’ very response to the whole psychological picture assumed by his imagined objector. In short, McCabe argues that from Republic IV onwards Socrates shows that: people are in fact psychologically complex and, considered as a whole, are not explicitly motivated only by the good; the orientation towards the good is certainly something that people are willing to make an effort to get right; appetites need not be determined by the good to be complex; and that appetites need not merely mirror the psychological structure of people. In doing so he offers an account of the nature of the soul which allows us not only to explain the individual actions of an agent, but how her very nature might change through education and influence. In a supplementary response to Moss’ discussion of Republic IV,¹⁰ Anthony Price takes up a question raised by the Division passage: whence does thirst learn its object? Price takes as his starting point Moss’s claim that thirst can indeed be understood as ‘desire for drink as a good’ and asks how this claim should be understood if it is to apply plausibly to thirst. For whatever is being claimed must be consistent with the description of appetite in the Republic as unreasoning. It is, in Schiffer’s language, a reason-providing desire, not a reason-following desire, and thus any explanation of its orientation to an object—drink—and specifically, drink as good, must be consistent with the fact that, from the point of view of the soul, thirst arises as a brute fact and not in response to an evaluation. In what sense, then, Price asks, might we take drink to be a good? Price notes that in the Republic, the objects of appetites are pleasures, though it remains an open question whether to desire drink as pleasant is to desire it as good (on Moss’ view, as we have seen, it is; others may not be persuaded). Price suggests that a more promising approach is to turn from the intentional object of desire to its functional role, pointing to the connection, evident in a number of dialogues, between desire and deficiency. The functional role of appetite, he argues, is to prompt a release from physical deficiency, and thirst thus aims at the (human) good as the repletion of deficiency. The question remains, however, from where thirst derives its orientation towards drink. Price points towards a much-examined passage in the Philebus in which Plato raises this question, and appeals to memory in answer. Price offers, here, an excursus on the nature of anticipatory pleasure—drawing on Wollheim’s (1984) analysis of imagining—to present a critique of an influential interpretation, from Verity Harte (2004), of Philebus 39a–40d.¹¹ Returning to the question of the object of thirst, Price concludes that within the Philebus there is ¹⁰ Price’s piece was originally presented at the 2011 Colloquium as a response to James Warren’s paper in Chapter 10 of this volume. Given its close engagement with Moss’ discussion, however, and for reasons of thematic continuity, we have chosen to place it at this point in the volume. ¹¹ This interpretation is adopted by James Warren in his contribution to this volume, and Price’s discussion thus presents a challenge to Warren’s conceptualization of ancient views of the nature of memory and anticipation (discussed below).

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

10

 

indisputable evidence that appetites are viewed as originating from memories of past satisfactions, and expectations of similar satisfactions to come. Noting that the Philebus discussion is couched in terms of pleasure, and not of the good, Price asks what significance this might have for a reading of the Republic. The Republic, he suggests, anticipates an account of pleasures as replenishings and the Philebus can thus be read as completing what, in the Republic, remained unfinished business.

4. Pleasure and Pain The first three papers of this volume and their responses, focused on the writings of Plato, trace a development of sorts in accounts of the psychology of value and a flourishing human life, from a (hypothesized) Intellectualism where our engagement with value—in this case ‘the good’—is centrally intellectual, to an account which makes space for non-rational engagement with forms of value and a richer picture of human flourishing. Our next two papers—by Raphael Woolf and James Warren—extend beyond the writings of Plato, to include those of Aristotle and some of the Hellenistic schools. Both, too, engage with issues concerning the ethical subject’s experience of pleasure, and consider the ways in which a subject’s proper appreciation of the pleasantness of some activity or experience can bear on her flourishing. Before looking at the specific contributions to such questions that each of these papers make, it will be worth reminding ourselves of the accounts of flourishing and, in particular, the relationship between flourishing and pleasure, endorsed by Aristotle and some of the later Hellenistic schools. Like Plato, Aristotle maintains that a flourishing human life consists in a life of virtue, distinguishing between the intellectual virtues which belong exclusively to the rational part of the soul, and the moral or character virtues which correspond to the non-rational, but reason-responsive, part of the soul.¹² The morally virtuous agent not only possesses right reason (orthos logos, NE 1144b26–7), but importantly her desires and emotional responses are aligned with right reason. She hits both the mean in action and the mean in feeling (NE 1106b17–24; 1109a25–30): the courageous agent not only remains at her post (if that is what courage ¹² There is a long-standing disagreement between scholars as to whether moral virtue, for Aristotle, should be understood as a state of the non-rational part of the soul alone (e.g. Burnet 1900, 64; Hutchinson 1986, 12; 1995, 213; Bostock 2000, 35; Taylor 2006, xv, 106; Moss 2012, 163–74), or a state of the rational part also (Sorabji 1973–74; Irwin 1975, 576; Sherman 1989; Lorenz 2009). In support of the former view, scholars cite NE I.13 1103a3–4 and the stronger claim in the Eudemian Ethics that moral virtue ‘belongs’ to the non-rational part of the soul (EE 1220a10–11). In support of the latter view, scholars point to Aristotle’s claim that virtue is a ‘state which decides’ (NE 1106b36), and thus involves reason and thought, as well as Aristotle’s claim that virtue involves right reason (NE 1144b26–7).

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

      

11

demands in a situation) but also fears neither too much nor too little. She is not a coward, fearful of the most trivial things, but nor is she rash, and unaware of the presence of danger. Importantly, she takes pleasure in the virtuous actions that she performs (NE 1099a7–15, 1099a21–4, 1174b14–16), recognizing that such actions are fine or noble (kalon, e.g. NE 1120a23) and delighting in this quality. Indeed, Aristotle emphasizes the alignment of the virtuous agent’s values in NE II.3, where he writes that (1104b30–2): there are three objects of choice and three objects of avoidance: [the] fine, beneficial and pleasant, and their contraries, [the] shameful, harmful and painful. About all these, then, the good person is correct and the bad person is in error, especially about pleasure.

The virtuous agent, in summary, is one whose rational and non-rational capacities are aligned with the appropriate values, who can identify in particular situations what is of value, and who responds correctly to such value in her feelings and actions. Importantly, not only does the virtuous agent correctly identify what is of value, and take pleasure in those activities which realize this, but owing to her realization of virtuous activity, her life as a whole is said by Aristotle to be pleasant too (NE 1099a5). The connection between happiness and pleasure is more pronounced still in the thought of two rival Hellenistic schools: the Cyrenaics and the Epicureans. Where Plato and Aristotle identify a life of virtue as the human good, and maintain that such a life is also a pleasant life, both the Cyrenaics and the Epicureans identify pleasure as the good and promote a life in pursuit of pleasure as the best human life. Both schools, too, appeal to ‘Cradle Arguments’¹³ in support of their conception of the best human life. Yet the Cyrenaics and Epicureans have differing conceptions of the ways in which pleasure ought to be pursued, and of the nature of our experience of pleasure. The Cyrenaics identified pleasure as the only good and something to be pursued positively (DL 2.87–8). The pleasure that is identified as the goal, however, is not pleasure over a long time, but the immediate pleasures of the moment: the ethical subject is advised to ‘enjoy each pleasure as it comes’ (DL 2.91). Importantly, pleasure on the Cyrenaic view is ‘unitemporal’ (Athenaeus 12, 544A–B), it only exists, and can only be enjoyed, in the moment, a view likely to be grounded in a central tenet of Cyrenaic epistemology that pathē (of which pleasure and pains, on the Cyrenaic view, are forms) are constituted by a particular, immediate, interaction between the perceiving subject and the

¹³ Arguments, that is, which take the pre-social behaviour of infants to establish what is natural to human beings (as opposed to what is conventional).

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

12

 

perceived object. In the absence either of perceiver or object, the pathos cannot be experienced.¹⁴ The Epicureans, meanwhile, identified a more long-term conception of pleasure as the goal. More specifically, the highest pleasure, on the Epicurean view, consists in the absence of both mental and physical pain (Ep. Men. 128).¹⁵ Epicurean followers are advised not to pursue every immediate pleasure, since the pursuit of some pleasures may cause greater pain in the long run. Instead, they are advised to cultivate simple and inexpensive desires that can easily be satisfied, in order to avoid the psychological pain produced by frustrated desires (Ep. Men. 131). Importantly—and as James Warren in his contribution to the volume discusses in further detail—the Epicureans place great emphasis on the role of anticipation and recollection of future and past pleasures as a means of counterbalancing present pains and maximizing the pleasantness of one’s life in the long term. In his contribution to this volume, ‘Courage and Pleasure in Aristotle’s Ethics’, Raphael Woolf takes up the question of the extent to which Aristotle’s virtuous agent enjoys a pleasant life. For despite Aristotle’s claim that the happy life—i.e. the virtuous life—is the most pleasant, as Woolf points out, the exercise of the virtues does not always seem particularly pleasant. The exercise of courage, in particular, will involve engaging with situations that are themselves unpleasant. And not only that, but since such activity is likely to be frightening, the exercise of courage will often involve the powerfully unpleasant emotion of fear. Indeed, fear is painful both in respect of its quality (it is a painful emotion) and in respect of its object (things that cause destructive pain). In his paper, Woolf explores the relation between the virtue of courage and the experience of pleasure and pain, with a view to defusing the apparent tension in Aristotle’s conception of a happy life. Woolf begins with an examination of the courageous person’s emotional response to the situations in which he finds himself. Aristotle himself recognizes that the courageous person seems to present something of a puzzle when we consider their emotional response: on the one hand, if the courageous person merely withstands things that are fearful to another person, but not to himself, courage is nothing impressive. On the other, we surely want a way of distinguishing the courageous person and the coward. In response to this difficulty, Aristotle ¹⁴ As Warren thus discusses in his paper, the pleasures of anticipation and recollection, on the Cyrenaic view, are tied to the occasion of recollecting and anticipating: ‘To enjoy anticipating a pleasure is not to receive an advance instalment of some pathos yet to come and, similarly, to enjoy the recollection of a pleasure is not to receive some recovered instalment of a past pleasure’ (p. 155 this vol). The pleasure associated with a particular occasion of anticipating or recollecting is a new pathos, identified with a presently obtaining psychic motion, and distinct from the pathos that is being anticipated or recollected. ¹⁵ Unlike the Cyrenaics, the Epicureans deny the existence of an intermediary state between pleasure and pain.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

      

13

draws a distinction between what is fearful simpliciter (i.e. to human beings as such) and what is fearful to a particular individual. Aristotle explains that, qua human beings, the courageous person fears those things that are fearful simpliciter, but at the same time, qua courageous, is fearless in response to such things. How does the courageous person both fear and not fear these things? Woolf proposes that this seemingly contradictory remark can be interpreted in the following way: the courageous person recognizes as fearful those things that are fearful simpliciter, but he does not himself experience the emotion of fear in response to these things (or experiences it hardly at all). In this way, the courageous person does not experience pain, insofar as it accompanies the emotion of fear, since the courageous person is characteristically fearless. But what, Woolf asks, of the other painful aspect of courage: that couragedemanding situations are often themselves painful? Aristotle, after all, explicitly states that ‘courage brings pain’ (NE 1117a33–4). Yet as Woolf notes, Aristotle also states that there is a pleasurable aspect to courage, namely what pertains to the goal of the activity. Just as boxers endure pain for victory, so too the courageous person endures pain for the sake of the fine. And the pleasure of doing what is fine can seemingly outweigh the pain that he experiences. This is not to say that he experiences no pain, but unlike the coward, he does not experience the pain of fear, and since he acts for the sake of the fine, he will gain more pleasure than either the coward or the reckless person. This approach, Woolf argues, is motivated by the need to reconcile two convictions about the nature of the good life: that it must be virtuous, and that it must be pleasant. Aristotle does so by showing that virtuous activity is inherently pleasant, and it is thus the goodness of the life of virtue that is responsible for its pleasantness. In her response to Woolf, Sarah Broadie emphasizes the importance for Aristotle—particularly against his hedonist rivals—of maintaining the pleasantness of eudaimonia, and acknowledges the putative problem that the virtue of courage thus poses. Against Woolf, however, Broadie argues that we need not read the claim that the virtuous agent is fearful qua human, but fearless qua courageous as a psychological claim, to the effect that he recognizes the fearfulness of what he is faced with, but does not feel the emotion of fear. Rather, we can understand the claim to be that he recognizes the fearfulness of a situation, and that this may indeed manifest in the experience of fear, but that he is not moved by this feeling to act as frightened people do. That the courageous person may feel fear, however, is not incompatible with the thought that courageous action is nevertheless pleasant to the virtuous agent, for ‘pleasantness’ in this situation need mean no more (and perhaps no less) than that the agent is thoroughly identified with his actions, however painful the situation may be. The analogy, pointed to by Woolf, between boxers who endure pain for victory, and the virtuous agent who endures painful situations for the sake of the fine, calls our attention to an intriguing feature of our psychology, namely that we can find

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

14

 

things painful in one respect or at a certain time, and yet take pleasure in the very same thing, either in a different respect or at a different time. This phenomenon is discussed by James Warren in his paper ‘Memory, Anticipation, Pleasure’, in which he takes as his focus the human capacities for memory and anticipation, and the way in which these allow us to make contact, as it were, with sources of value that are temporally remote.¹⁶ For human beings have the ability not only to remember or anticipate events that have taken place or will possibly take place in our lives, but we can also remember or anticipate past or future affective experiences. We can remember or anticipate pleasures and pains, and indeed, we can do this with pleasure or pain. Tracing the treatment of this phenomenon from Plato to the Hellenistic period, Warren identifies two broad ways of thinking about this phenomenon. On one view, memory and anticipation are ways of reaching out to the past or future, and hauling a temporally remote experience from there to the present. Thus conceived, this ability allows a subject to do a number of things, as evidenced in the Protagoras, Republic, and Philebus. First, it allows a subject to compare present affective states with those that are anticipated or remembered, from which a subject may then draw further conclusions. In the Protagoras, for example, Socrates advises the prudent hedonic calculator to set aside the temporal distance of possible pleasures and pains in order to accurately rank these; in this way, the prudent calculator can maximize her pleasures over the course of her life. In the Republic, meanwhile, it is shown that the comparison between present state and past or future states allows us to re-evaluate our current state, by encouraging us to see it as being either not as good as, or not as bad as, we originally assumed. But the ability to carry a temporally remote experience into the present can also be employed directly to improve (or harm) one’s present state, insofar as it allows a subject to re-live or pre-live a pleasurable (or painful) experience. This conception of memory and anticipation emerges, Warren argues, in Plato’s Philebus, and is put to explicit use by the Epicureans, who—as we have seen—argue that present pain can be offset by recalling past pleasures.¹⁷ Warren, however, identifies another—less dominant—mode of thinking about the phenomenon of remembering or anticipating pleasures and pains. Where the Platonic dialogues and Epicurean texts take memories or anticipations of pleasure to be valenced accordingly, Aristotle’s Rhetoric appears to deny that an anticipated or remembered pleasure or pain necessarily has the same valence as the original: we can, for example, recall past painful experiences with pleasure (as evidenced in ¹⁶ As Warren emphasizes, it is our ability to recall past experiences and to anticipate possible future experiences, as Plato argues in the Philebus, that distinguishes a human life from that of a ‘mollusc’ (pleumōn, 21c—or ‘jellyfish’, as O’Reilly 2019, has recently argued). ¹⁷ As noted above, Warren’s discussion of the Philebus is critiqued by A. W. Price, who argues that what Socrates describes at Philebus 40a cannot be understood as an ‘advance instalment’ of pleasure, since Socrates here describes an imagining that is, in Wollheim’s terminology, acentral. If what he were describing could be called an ‘advance instalment’ of pleasure, what Socrates describes would have to share a point of view with that pleasure.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

      

15

Raphael Woolf ’s contribution to this volume too). This, Warren explains, can be (i) because it is pleasant to compare past painful experiences to our comparatively positive present state, (ii) because those painful experiences had pleasurable consequences, or (iii) because we have come to view such events in a different light, as positive rather than negative. On this view, then, our memories (and presumably our anticipations) can be affected in important ways by our present state of mind; a thought that chimes with the view in contemporary psychology that our memories are often shaped by our present state. In response to Warren, Katharine O’Reilly and I question Warren’s characterization of two models of the memory and anticipation of pleasure and their application to the texts he surveys. We begin with the observation that the two models of memory and anticipation are presented by Warren as a contrastive pair, but argue that what are in fact contrasted as two varieties of the same thing are in fact two very different kinds of thing, and are thus seemingly non-comparable. Whilst the first model presents us with an account of the metaphysics of memory and anticipation, when it comes to the second, we are presented instead with a claim about the affective character of acts of memory or recollection. We suggest, however, that we can better appreciate the contrast Warren has in mind if we consider what a particular model of the metaphysics of memory or anticipation implies for the affective character of an act of recollection or anticipation. For if we think of memory as a form of re-living, say, it would seem to follow that the affective character of an act of recollection will be the same as that of the object experience, whilst an alternative account of the metaphysics of memory might allow for the possibility that the affective character of an act of recollection will differ from that of the object experience. Having thus filled in the gaps in the original contrastive pair, we consider the fit between Warren’s models and the ancient examples he discusses, and note that in several cases the models as Warren describes them strain to accommodate the texts. Whilst this might seem to present a serious challenge to Warren’s identification of two models in the texts under consideration, we end with a suggestion as to how the distinction Warren makes in his paper, and its central insight, might still be preserved. Rather than representing competing accounts of the metaphysics of memory and anticipation, from which certain commitments about the affective character of memory may or may not follow, we ask whether the two models Warren identifies in the ancient texts do not, instead, represent two models of how each of us can remember and anticipate—one that is more immediate and one that builds in some psychological distance.

5. Virtue Whilst the previous two papers engaged with the topic of pleasure—examining the nature of our experience of pleasure, the relation between pleasure and value, and

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

16

 

the role that our experience of pleasure can play in our ethical lives—the final paper in this volume returns to a picture of human flourishing as consisting specifically in a life of virtue, namely that endorsed by the Stoics. In Stoic moral theory we see significant emphasis placed on the ethical subject’s ability to identify correctly what is truly of value and a sophisticated account of the form that correct psychological engagement with value takes. Central to any understanding of Stoic ethics are the concepts of the good, the bad, and the class of ‘indifferents’. Goods include wisdom, moderation, justice, and all that is virtue or participates in virtue (DL 7.101–2). What is bad is vice and all that participates in vice (DL 7.102). Only the good (and, conversely, what is bad) contribute to our happiness directly. Indifferents, meanwhile, are things such as life and death, reputation and ill-repute, health and sickness, wealth and poverty, and so on (DL 7.102–4) These indifferents neither add to, nor detract from, our happiness directly, though some may be preferred (such as health, over illness), insofar as they are ‘appropriate’ to us, as human beings (Stobaeus 2.79 18–80; Epictetus Diss. 2.6.9). At the heart of Stoic psychology (and epistemology), meanwhile, are the notions of ‘impressions’ (phantasisai) and ‘assent’ (sunkatathesis). An impression is an alteration of the soul, which provides information about the way the world is (DL 7.49–51). If we take an impression to accurately represent the way the world is, we assent to the impression, and in assenting to an impression we thereby form the belief that such-and-such is the case (Cicero, Academia 2.145; Sextus Empiricus, M 7.151). We can also withhold our assent, however, and maintain that an impression is a ‘mere’ impression, that it does not accurately represent the way the world is. Our ability to assent or withhold assent is entirely under our control (Epictetus, Diss. 1.1), and the Stoics maintain that we should assent only to true impressions—impressions which accurately represent the way the world is— whilst withholding our assent from false impressions. Among the impressions to which we should or should not assent, the most important are those concerning value. Since virtue is the only good, and vice the only bad thing, we should not assent to any impression that an indifferent is itself good or bad. This has implications for desire, since, on the Stoic view, a desire involves an evaluation of an object as good or bad. A desire for wealth or honour, say, involves the evaluation of these objects as good, but since neither wealth nor honour are truly good, such desires should not be cultivated. There are important consequences, too, for the role of emotions in our ethical lives. Emotions (pathē), on the Stoic view, are identified with beliefs, and, like desires, involve an evaluation of an object as good or bad—but since, again, only virtue and vice are good and bad, almost all pathē are false.¹⁸ In order to achieve virtue, and thus live a flourishing life, it is ¹⁸ The only ‘true’ pathē are those that consist in a (correct) evaluation of virtue or vice, though such responses are strictly picked out by the term ‘eupatheiai’.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

      

17

necessary to expel all pathē that contain evaluative judgements about anything other than virtue or vice.¹⁹ Our overall disposition to assent is referred to, by Epictetus, as our prohairesis. The virtuous agent—the Stoic sage—assents only to the impressions that virtue is good and vice is bad. The sage does not desire indifferents such as wealth or pleasure, though he may select indifferents when such selection is recognized to be appropriate and in accordance with nature. The sage suffers no pathē, such as anger or fear, but only experiences the ‘eupatheiai’ of volition, caution, and joy (DL 7.96). His virtuous activity, in short, consists in an appropriate pattern of selection and response, and this virtuous selection of good things—rather than their attainment as such—constitutes human happiness. Indeed, the Stoics maintain that virtue is sufficient for happiness. How should we understand the relation between the thesis that virtue is sufficient for happiness and the thesis that virtue is the only good? According to a number of modern commentators, the Stoics assume that virtue is sufficient for happiness and this explains their commitment to the thesis that virtue is the only good. Yet as Daniel C. Russell observes in his contribution to this volume, in the ancient reconstruction of the Stoic view in Cicero’s De Finibus the direction of explanation is reversed: the Stoics believe that virtue is self-sufficient and that virtue is the only good, and this explains their commitment to the view that virtue is sufficient for happiness. In the final paper of this volume, ‘Three Mistakes about Stoics Ethics’, Russell identifies three mistakes in the modern reconstruction of Stoic thought—(1) it reverses the order of explanation between the thesis that virtue is sufficient for happiness and the thesis that virtue is the only good; (2) it fails to recognize the distinctness of and the distinct contribution played by the thesis that virtue is self-sufficient; and (3) it fails to offer a rationale for the belief that virtue is sufficient for happiness—and shows how, when properly reconstructed, the Stoic view remains one to be reckoned with. That the modern reconstruction of Stoic ethics reverses the order of explanation between the two theses is immediately evident when we compare Cicero’s reconstruction. But it is not the case that the thesis that virtue is the only good entails the sufficiency thesis by itself; this much is evident if we consider that virtuous activity might be vulnerable to fortune and the contingency of the perishable world. Even if virtue is the only good, it might be the case that activity in accordance with virtue is dependent on contingent matters. To arrive at the view that virtuous activity is sufficient for happiness, then, Russell argues that the Stoics must assume a particular view of what constitutes virtuous activity, such that it is independent of fragile connections to the perishable world. On the Stoic view, virtuous activity is the exercise of a faculty that is always under our control, ¹⁹ For an overview of Stoic texts on the pathē, see LS 65. For a comprehensive discussion of the Stoic account of pathē, see Graver (2007). See also Brennan (2005).

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

18

 

and virtue is thus self-sufficient.²⁰ When this idea is combined with the thought that virtue is the only good that the Stoics can conclude that virtue is sufficient for happiness. The question remains not only why modern commentators have supposed that the sufficiency thesis is basic, but also why—if it is not basic, but the conclusion of an argument—it is a thesis that the Stoics accepted, and was not taken as evidence that one of the premises in the argument is false. Russell suggests that some commentators have supposed the sufficiency thesis to be basic on the grounds that the Stoics were concerned primarily with perceived vulnerability to loss; it is for this reason, the thought goes, that the Stoics assert that virtue is sufficient for happiness, and following that, must commit to the thought that virtue is the only good. Yet as Russell notes, not only does this reconstruction run counter to the texts, it leaves the Stoic view unpersuasive. As Russell points out, the passage that is sometimes taken to support this reading²¹—a passage which contains Cicero’s claim that ‘we want the happy man to be safe, impregnable, fenced and fortified’ (TD V.40–1)—does not in fact support it, for the context of the passage reveals that what we wish to be fortified against is not vulnerability to loss, but vulnerability to degradation of character. Beliefs that there are goods other than virtue— beliefs that make us prone to fear the loss of such goods—can obstruct our ability to act virtuously, and lead to the degradation of our characters. We must thus be fortified against such beliefs, and become so by accepting that our good lies entirely in virtuous activity. And, as Russell argues, since virtuous activity is something that we control completely, it is self-sufficient, and so we must therefore accept that virtue is sufficient for happiness. Whether or not we find the Stoic position ultimately convincing, the Stoic argument forces us to reflect on the sense in which staking our happiness on things outside our control can threaten both our autonomy and our virtue. In his response to Russell, David Sedley points to other tacit premises—beyond the premise that virtue is self-sufficient—required to infer the sufficiency of virtue for happiness from the premise that virtue is the only good. These are: the assumption that happiness is humanly attainable, for if it were unattainable, even the possession of all goods would be insufficient to secure it; the assumption of the unity of virtue, for one must be confident that the virtuous life contains no

²⁰ The Stoic conception of virtuous activity, Russell explains, is grounded in the Stoic conception of value and what kinds of objects can participate in virtue. ²¹ Russell cites Brennan (2005, 122) as taking this passage in support. David Sedley, in his response to Russell’s paper does so too, writing that this passage “may well itself be doing just that, invoking the invulnerability of happiness as a philosophical aim, not as the premise of any argument. As Cicero says there, ‘We want the happy man to be safe, impregnable, fenced and fortified, so that he is not just largely unafraid, but completely.’ ” (pp. 206–7 in this volume). Sedley takes it, contra Russell, that it is the vulnerability of happiness that we wish to be safe from. Russell’s claim is that it is in fact vulnerability to degradation of character that Cicero has in mind, and this passage thus does not support the received interpretation.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

      

19

residual moral defects which might deflect it from total harmony with nature; and the assumption that virtues are not hexeis, but diathesieis, not admitting of differing degrees, being states of moral perfection. Sedley thus draws attention to the interconnection of a number of prominent theses of Stoic ethics, and their place in the Stoic ethical project as a whole. Sedley, however, questions Russell’s emphasis on the value of virtuous activity, rather than virtue as such, noting that this emphasis makes the Stoics sound rather more like Aristotle. For the Stoics saw no gap between virtue and virtuous activity, and for this reason they did not need to take the Aristotelian option of pointing to virtuous activity, rather than virtue itself, as the basis for happiness. In the final part of his response, Sedley takes up the thought—rejected by Russell—that the Stoics, in arguing that virtue is the only good, were motivated by the aim of proving the invulnerability of happiness. If— contra Russell’s argument—this was indeed a motivation for the Stoics, what commitments, in turn, would underlie this? Sedley identifies some possible responses and, with these, avenues for future research. Each of the papers in this volume makes a significant contribution to scholarship on ancient moral psychology. In challenging the orthodox interpretation of the ethical stance taken in Plato’s early Socratic dialogues, Rachel Barney invites us to reconsider how we understand the development of Plato’s thought, whilst Matthew Evans and Jessica Moss, in offering fresh interpretations of a muchstudied passage of Republic IV, make important innovations in the long-running debate about the nature of appetitive desire. In his account of the affective states of the virtuous agent, Raphael Woolf offers a radical solution to a familiar tension in Aristotle’s ethical theory, whilst James Warren’s identification of two divergent views of the relation between memory, anticipation, and pleasure in ancient thought invites us to reflect with renewed focus on the very nature of affective experience. Finally, Daniel C. Russell draws attention to a fundamental, but perhaps unacknowledged, tension in Stoic scholarship, offering a textually motivated corrective to the received interpretation of Stoic ethics. Each paper—and the responses offered—advances our understanding of ancient moral psychology, and invites us to think afresh about the ways in which our psychological engagement with value contributes to a flourishing human life.

References Bostock, D. 2000. Aristotle’s Ethics. Oxford: Oxford University Press. Brennan, T. 2005. The Stoic Life. Oxford: Oxford University Press. Burnet, J. 1900. The Ethics of Aristotle. London: Methuen. Graver, M. 2007. Stoicism and Emotion. Chicago, IL: University of Chicago Press.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

20

 

Harte, V. 2004. ‘The Philebus on Pleasure: The Good, the Bad, and the False.’ Proceedings of the Aristotelian Society 104: 111–28. Hutchinson, D. S. 1986. The Virtues of Aristotle. London: Routledge & Kegan Paul. Irwin, T. 1975. ‘Aristotle on Reason, Desire, and Virtue.’ The Journal of Philosophy 72 (17): 567–78. Irwin, T. 2008. ‘The Platonic Corpus.’ In The Oxford Handbook of Plato, ed. Gail Fine, 63–87. Oxford: Oxford University Press. Lewis, D. 1989. ‘Dispositional Theories of Value.’ Proceedings of the Aristotelian Society suppl. 63: 113–37. Lorenz, H. 2009. ‘Virtue of Character in Aristotle’s Nicomachean Ethics.’ Oxford Studies in Ancient Philosophy 37: 177–212. Moss, J. 2008. ‘Appearances and Calculations: Plato’s Division of the Soul.’ Oxford Studies in Ancient Philosophy 34: 35–68. Moss, J. 2012. Aristotle on the Apparent Good. Oxford: Oxford University Press. O’Reilly, K. R. 2019. ‘The Jellyfish’s Pleasures: Philebus 20b–21d.’ Phronesis 64: 277–91. Sherman, N. 1989. The Fabric of Character. Cambridge, MA: Harvard University Press. Smith, M. 1994. The Moral Problem. Oxford: Blackwell. Sorabji, R. 1973–74. ‘Aristotle on the Role of Intellect in Virtue.’ Proceedings of the Aristotelian Society, n.s. 74: 107–29; reprinted in Essays on Aristotle’s Ethics, edited by A. Rorty, 201–20. Berkeley, CA: University of California Press, 1980. Street, S. 2008. ‘Constructivism about Reasons.’ Oxford Studies in Metaethics 3: 207–45. Taylor, C. C. W. 2006. Aristotle, Nicomachean Ethics Books II–IV. Oxford: Oxford University Press. Wollheim, R. 1984. The Thread of Life. Cambridge: Cambridge University Press.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

2 Intellectualism and the Method of Hypothesis in Plato’s Early Dialogues Rachel Barney

Plato’s early dialogues have often been read as expressing his commitment to Socratic intellectualism, a kind of ethical rationalism centring on the thesis that virtue [aretê] consists in knowledge [epistêmê].¹ Socratic intellectualism comprises a whole cluster of theses organized around that one: that all the virtues are one; that virtue is acquired by purely cognitive means, so that affective factors are irrelevant to it; that weakness of will [akrasia] is impossible; that no one does wrong voluntarily; that the human soul is wholly rational; and that all desire is for the good. These views are standardly attributed at once to the Socrates of the early dialogues, to Plato as his author, and to the historical Socrates: the presumption is that the intellectualist package represents Plato’s philosophical inheritance from his teacher. And as far back as the Aristotelian Magna Moralia, histories of ethics have often told a narrative of philosophical progress on which this ensemble of strange views is corrected by Plato in later works. When Plato divides the soul into rational and irrational parts in Republic IV, it is on this reading precisely in order to give non-rational motivations their due, to account for weakness of will, and in general to develop a more empirically adequate moral psychology (Magna Moralia I.1.7, 1182a24). I will refer to this as the Aristotelian reading; and I will argue against it by proposing two claims, one about method and one about content. First, the early dialogues do not propound Socratic intellectualism as doctrine. Rather, they investigate it as a hypothesis, roughly in terms of the method of hypothesis proposed in the Meno and Phaedo.² Second, the Aristotelian reading distorts ¹ This ‘standard intellectualist reading’ has in recent years been subjected to several forceful challenges. (1) Kamtekar 2017, with which my reading has considerable common ground, uses hypothesis to explain what Socrates is up to in dialogues including the Republic and Protagoras. However, she does not use hypothesis as a framework for the kind of overall reading of the early dialogues which I here attempt. (2) Brickhouse and Smith 2010 offer a new reading of Socratic intellectualism which avoids committing him to denial of the reality and importance of non-rational desire. (3) Several recent studies have argued for continuities between the psychology of early/Socratic and middle/Platonic dialogues: see Carone 2001 and the other works cited in n. 69. ² So I take hypothesis to be Plato’s method from the start; the Meno and Phaedo accounts represent a new reflectiveness about method rather than a change from one method to another. For what this implies about the ‘Socratic elenchus’, see the latter part of section 3. Rachel Barney, Intellectualism and the Method of Hypothesis in Plato’s Early Dialogues In: Psychology and Value in Plato, Aristotle, and Hellenistic Philosophy: The Ninth Keeling Colloquium in Ancient Philosophy. Edited by: Fiona Leigh and Margaret Hampson, Oxford University Press. © Oxford University Press 2022. DOI: 10.1093/oso/9780192858108.003.0002

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

22

 

what intellectualism is. Its central thesis is not Socratic but sophistic, originating with Protagoras, and it is more precisely the claim that there is a politikê technê: an art of decision-making sufficient for expert political rule and for success in private life as well. Not all of the propositions traditionally taken as part of Socratic intellectualism form part of a natural package with that thesis, nor does Plato ever think that they do; and there is no reason to assume that its purpose or upshot was ever to downplay the ethical importance of our emotions and irrational desires. It is not a revisionist theory of human nature, but an educational and political programme, aimed at the reliable production of expert statesmen. On the reading I propose, Plato’s early dialogues—most of them, anyway— undertake to investigate this sophistic programme from a number of different angles.³ The strategy by which they do so is sophistic as well: they present arguments both for and against it in the manner of Protagorean antilogia, ‘argument on both sides of the question’.⁴ What gives the investigation a distinctively Platonic twist is that these arguments are further structured by the method of hypothesis. How to understand that method is a notoriously difficult and controversial question. But I think it is safe to say that hypothesis is at once explanatory and, insofar as possible, deductive: a method for tracing the complex ways in which different philosophical theses are connected to each other in relations of grounding or explanation, entailment, and contradiction. In particular, by enabling us to hunt down and exclude propositions from which contradictions follow, the method promises a kind of magical way to get something for nothing, to make epistemic gold out of straw. For (like the other wonder-working method of early Greek philosophy, Gorgias’ method of argument by elimination in the Helen and On Not-Being) it enables us to move from the unknown to the known, to eliminate certain positions as false without having to already know the true state of affairs. The early dialogues depict a single, awe-inspiringly complex inquiry in which Socrates applies this method to the most exciting ethical theory of the day: the Protagorean thesis that virtue is a technê.⁵ It seems fair to presume that Plato

³ By the ‘early dialogues’ I mean a conventional list linked both by thematic and by stylometric considerations, including the works on which Aristotle and the Magna Moralia evidently rely. I take the Meno, Protagoras, Laches, Charmides, Euthyphro, and Hippias Minor to be the core texts in which the intellectualist hypothesis is investigated. Other early dialogues (including the Ion, Hippias Major, Apology, Gorgias, and Lysis) have distinctive agendas which set them somewhat apart. Still, each can be seen as contributing indirectly to the larger project I discuss. For instance, the inquiry into rhapsody in the Ion clarifies the requirements for something to be a craft [technê], which is obviously useful for determining whether virtue could be one; and the Gorgias can be read as interrogating the claims of Gorgianic rhetoric to be the art of successful living postulated by the hypothesis. ⁴ Cf. Diogenes Laertius IX.51 (cf. IX.55) (DK80A1), Seneca Letter 88.43 (DK80A20). ⁵ So I take Socrates to proceed on Plato’s behalf, as his protagonist and philosophical deputy: Plato’s texts are investigations of the intellectualist hypothesis, and they investigate by depicting Socrates’ investigations. Whether there is daylight between the two—whether Plato ever presents his hero as going wrong, and other characters as having insights he lacks—are questions we can here leave open.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

     

23

hopes for a positive result: that he would not bother to work out in such tremendous detail a hypothesis which he did not find philosophically attractive and important. But that is not the same thing as even provisional acceptance of it; and his purpose in writing is to subject it to thorough scrutiny, not to persuade us of its truth. This paper argues for this hypothetical reading in four stages. I begin in section 1 by noting some of the problems with the Aristotelian reading; turning to positive argument, section 2 then proposes that intellectualism is not Socratic but sophistic, and gives an account of what Plato takes it to mean. In section 3 I spell out what I take to be the key features of the method of hypothesis; section 4 shows that many discussions in the early dialogues are organized around intellectualism seen as a hypothesis in this sense. I conclude with some preliminary remarks on the question of what, by the time of the Republic, Plato might have taken his inquiry to have shown.⁶ To do all this will require covering a lot of terrain quickly, and no doubt my argument will be insufficient to be fully convincing. But then the hypothetical reading is itself a hypothesis: my aim here is simply to set it forth, and to point out some of its advantages over the traditional alternative.

1. The Aristotelian Reading According to Aristotle, Socrates claims that all the virtues are ‘kinds of wisdom’ [phronêseis] (EN VI.13, 1144b19–20) or ‘kinds of knowledge’ [epistêmai] (VI.13, 1144b28–30).⁷ As corollaries to this intellectualist thesis, he also holds that we attain the virtues simply by coming to know them (EE I.5, 1216b4–11) and that weakness of will [akrasia] is impossible (EN VII.2, 1145b22–31). Further implications are spelled out in the Magna Moralia, a work included in the Aristotelian

⁶ In holding that the positive upshot of the early dialogues is postponed to the Republic, the hypothetical reading is akin to the very illuminating proleptic reading of Kahn 1996. However, Kahn’s framing sometimes suggests that the answers given in the Republic are already in the back of Plato’s mind as he writes the early dialogues; on my reading, the early dialogues are the vehicle by which Plato figures out for himself what he thinks. ⁷ My concern in this paper is with Plato’s Socrates; if, as seems likely, Aristotle means to be reporting both on that figure and on the historical original, I set aside the question of what his further evidence for the latter might be. As will become clear in the latter part of section 1, a major difficulty with the standard intellectualist reading is that in the early dialogues the discussion of intellectualism sits uneasily alongside what looks like an echt-Socratic stratum of ethical commitments: these include a conception of wisdom as consisting essentially in humility and self-knowledge (Apology), a conception of justice as a kind of psychic health (Crito, Gorgias), and a strong skepticism about whether virtue can be taught (Apology, Meno, Protagoras). If these are the views which Plato’s dialogues mean to depict as characteristic of Socrates, and if their attribution should be extended to the historical figure as well, then the historical Socrates was not an intellectualist either, and Aristotle’s reading could not be derived from any accurate non-Platonic evidence regarding the historical original. But then I doubt that any of the authors of Sokratikoi logoi were primarily concerned to give a historically faithful portrait of Socrates himself.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

24

 

corpus but the authenticity of which is disputed.⁸ The MM emphasizes Socrates’ commitment to the thesis that the virtues are kinds of knowledge [epistêmai] (I.1.7, 1182a15–31, 1183b9–18, I.20.4, 1190b28–32) or reason [logos] (I.34.25, 1198a10–12). It also emphasizes the negative implications of that thesis: Socrates recognizes only a rational soul in human beings, ‘doing away’ with the irrational part; and he denies the importance of ‘passion and character’—the emotional, affective, and non-rational factors in virtue and moral education (I.1.7, 1182a15–31). On this view, any non-rational virtues are otiose, since knowledge of virtue suffices for virtue (I.1.7, 1183b9–18); vice is involuntary (I.9.7, 1187a5–12) and, again, akrasia is impossible (II.6.2, 1200b25–30). The MM spells all this out the better to condemn it. That virtue involves only the rational soul is clearly false; and the denial of akrasia even more so: Because of this sort of argument he did not think that there was akrasia. But he was wrong. For it would be absurd for us, persuaded by this argument, to abolish what indisputably occurs. For people are akratic, and do bad things at the very time at which they themselves know them to be such. (Magna Moralia 1200b25–33)

So Plato sets moral psychology on the right track on the Republic when he divides the soul into rational and non-rational parts, and Aristotle makes further corrections: the MM’s story is one of rapid philosophical progress. Now as presented by Aristotle and the MM, Socratic intellectualism is a deeply puzzling theory. Socrates is supposed to just insist, apparently with no attempt to ‘save the phenomena’, on some claims which seem to be refuted by our most ordinary experience of human action: that no emotional or irrational forces ever determine our agency, that no one ever acts out of weakness of will, and that philosophical discussion magically makes people good. But why would he, or anybody else, say such things? Despite its apparent lack of charity, descendants of the Aristotelian reading remain prominent today, in the form of the ‘standard intellectualist reading’.⁹ So for instance in Plato’s Ethics, Terence Irwin says, ‘the doctrine we have found in the early dialogues corresponds very closely with the views that Aristotle ascribes to the historical Socrates’.¹⁰ Charles Kahn, after a heroic attempt to make sense of intellectualism so understood, ends up describing it as essentially ‘normative’ or ⁸ On the MM, see Simpson 2017. Nothing I say here depends on deciding the much-debated question of its authenticity. ⁹ I do not mean to attribute the whole of the Aristotelian reading to any particular modern scholar: I take it to operate as a kind of shared baseline, subjected in each case to individual modifications. The most impressive account of Socratic intellectualism remains Irwin 1977: but some of Irwin’s most interesting claims are not a standard part of the Aristotelian reading, so I do not focus on it here. Brickhouse and Smith 2010 discuss the recent history of the intellectualist reading in some detail. ¹⁰ Irwin 1995, 242.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

     

25

‘protreptic’ rather than as a genuinely descriptive moral psychology.¹¹ Less sophisticated and sympathetic readers tend to end up suggesting that Socrates was engaging in a kind of projection, misled by the workings of his own psyche. For instance the translator of the Loeb edition of the Magna Moralia says in a footnote: ‘Probably his extraordinary power of self-control made him under-rate the force of passion in others.’¹² This is obviously unsatisfying—as if we could explain the philosophical appeal of Kantian rationalism or Humean sentimentalism (or Cartesian dualism, for that matter) by proposing that it was true, or almost true, of the philosopher himself. More important, anyone turning from these ancient or modern readings to the early dialogues themselves is in for a surprise. For there is remarkably little textual evidence here for some of the key propositions of ‘Socratic intellectualism’. Plato’s early dialogues simply do not discuss either the emotions or the nature of the soul, let alone propound sweepingly radical theories of them.¹³ Nor does Socrates ever present a theory of how virtue is acquired: the only full picture of moral education given in the early dialogues belongs not to him but to Protagoras, and his ‘Great Speech’ emphasizes the importance of early non-rational conditioning (320d–328d).¹⁴ Akrasia is denied in only one place, the Protagoras, and on the basis of a hedonistic hypothesis from which Socrates himself is markedly distanced (353a–358e). Moreover, that virtue is knowledge is firmly tied to the proposition that it is teachable; and Socrates regularly expresses scepticism about the latter, not only in the Protagoras (where it is presented as a standing conviction at 319a9ff.) but in the Meno (89d–96d) and Apology. The case of the Apology seems particularly significant, given its self-presentation as the defence of the historical Socrates and thus its self-appointed standing as a specially authoritative locus for echt-Socratic views. It is an underappreciated basic problem for the Aristotelian reading that in the Apology Socrates not only disavows any teaching of virtue himself, but refers to the sophistic project of doing so in snarky, implicitly condemnatory terms, as involving a claim to a wisdom beyond the human (20c–e).¹⁵ That nobody has any such thing is not only clear from the tone of Socrates’ remarks: it is also presented as the verdict of the Delphic oracle, which

¹¹ Kahn 1996, 226–33, cf. 241–75. ¹² Armstrong in Tredennick and Armstrong 1935. ¹³ This observation is at the root of the important and innovative reading of Brickhouse and Smith 2010, who credit the inspiration of Devereux 1995. Devereux infers that Socrates is in effect a Kantian: well aware of the power of our non-rational motivations, he ignores them because ignoring them, and behaving instead as reason dictates, is precisely what virtue requires (see esp. 406–8). In Barney 2021 I present a reading which reaches the same conclusion by a different route: a Platonic craft imposes unhypothetical imperatives on the genuine practitioner (the ‘doctor qua doctor’), so if virtue is a technê it too must involve a commitment to following the dictates of a disinterested practical reason. ¹⁴ Gorgias 460a–e is sometimes cited as support for the Aristotelian attribution to Socrates of a simplistic intellectualism about moral education: but the argument here is ad hominem against Gorgias, and tells us nothing about what Socrates or Plato himself thinks that learning justice might require. ¹⁵ Given that perhaps the strongest count against taking Plato’s Socrates to be an intellectualist is his skepticism about the teaching of virtue, it is possible that we should read him as simultaneously

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

26

 

would otherwise be making a terrible mistake in his regard (23a–b). All this should give us a serious evidentiary qualm about several key parts of the Aristotelian reading. And this dovetails with the philosophical qualm at which I have already hinted. When Plato introduces the new moral psychology of Republic IV, it is on the basis of utterly unsurprising observations about irrational desires, mental conflict, and akrasia. It is hard to believe that in his earlier work Plato meant simply to deny the reality of these phenomena. If on the other hand he was for all that time reflectively committed to an intellectualist theory which in his considered view had an adequate way to account for them, how could that commitment be so easily dislodged?¹⁶ Finally, the aspects of Socratic intellectualism which do make an appearance in the early dialogues are not discussed in the spirit which the Aristotelian reading would lead one to expect: so there is room for a third, methodological qualm here as well. Though the idea that virtue is knowledge does figure heavily in the early dialogues, there is not a single dialogue in which Socrates explicitly and unequivocally defends it as doctrine.¹⁷ The Protagoras seems to be Aristotle’s principal source;¹⁸ and it does include arguments for the identity of the other virtues with wisdom. But again, the only general and conclusive-looking argument to that effect is ad hominem, dependent on the hypothesis of hedonism and distanced from Socrates himself in a number of ways. Moreover, the Protagoras opens with Socrates presenting an argument against the claim that virtue can be taught, suggesting a presumption on his part that it is not knowledge (319a–320c). Most readers feel that Protagoras refutes this successfully, but it is hardly obvious that in doing so he speaks for Plato—let alone, somehow, for Socrates. Socrates’ own conclusion in the Protagoras is thoroughly aporetic: if the discussion had a voice of its own, he says, it would say that both he and Protagoras were ridiculous for changing sides, and that further investigation is needed (361b–c). The Meno in inclined to believe (a) that virtue is a kind of knowledge; (b) that virtue cannot be taught; and (c) that whatever is knowledge is teachable—so that the urgency of his investigations comes from the awareness that his own views on the topic are in an unacceptable state of incoherence. ¹⁶ In his comments, Terence Irwin refers to my view as being that ‘Aristotelian Socratic intellectualism (i.e. Socratic Intellectualism as expounded by Aristotle), is not only false, but so boringly false that we ought to doubt whether Socrates could have seriously affirmed it.’ And he notes that the position in question is taken very seriously by Aristotle as well as being revived by the Stoics, showing that it is a worthy view for an intelligent philosopher to have held. I do not really disagree, and do not mean to claim that the ‘Socratic’ view is boringly false. What I do claim is that it is prima facie subject to certain obvious empirical objections, against which careful explanation and defence is required. If Socrates and Plato mean to assert the ‘Socratic’ view in the early dialogues, it is surprising that so little defence along these (later, Stoic) lines is provided; and if Plato is for mysterious reasons keeping such a defence in reserve, it is surprising that he would later take the story of Leontius as decisive against it. ¹⁷ Here and in the first part of the next section I rehearse a line of argument now published as sections I and II of Barney 2021. I have tried to avoid redundancy in the framing. ¹⁸ This is of course speculative, but the Protagoras is the only obvious source in Plato for the denial of akrasia, and an equally good source for the claim that all the virtues are phronêseis or epistêmai (cf. 357b–e, 361b).

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

     

27

a way comes closer to fitting the Aristotelian reading; for it does include a clear and complete argument to the effect that virtue is knowledge (86e–89a). However, Socrates then proceeds to argue against that conclusion, recycling arguments from the Protagoras (89d–96d). And the upshot seems to be that at least some virtue is not knowledge, but rather a matter of divine luck and correct belief (99a–b). So neither the Meno nor the Protagoras propounds the intellectualist thesis in the single-sided, unambiguously affirmative way that we standardly take as the marker of authorial doctrine. And it is not that Socrates argues only tentatively and dialectically for what is nonetheless clearly his own view: rather, these dialogues present inconclusive arguments on both sides, and in doing so exemplify a recognized ancient philosophical genre. For they are exercises in sophistic antilogikê, ‘argument on both sides’ of a disputed question. The form of antilogia was associated with Protagoras; and one thing the Protagoras is evidently intended to prove is that Plato can do it better than the master himself. But how to extract positive philosophical morals from Platonic exercises in this sophistic form is surely a tricky question, to be handled with caution.¹⁹ Now in one dialogue, the Euthydemus, Socrates does argue, with no counterargument in sight, that knowledge or wisdom is the cause of happiness: for it is knowledge which reliably causes success in deliberation and in the use of other goods (281b). And this idea, which has echoes in the Meno (88a–89a) and Protagoras (344e–345c, 352b–358c), gives us a helpful sense of how intellectualism might first have been attractively formulated. It suggests that we are to hear the intellectualist hypothesis as the claim that only the person with practical wisdom—a kind of generalized deliberative expertise [technê] analogous to that of the specialized craftsperson (cf. Protagoras 318e5–319a2)—acts with reliable success, and so counts as having virtue. (I will turn in a moment to the question of what that line of argument presupposes about ‘virtue’.) But the Euthydemus argument proceeds by linking knowledge and happiness without using virtue as middle term, so that it can only be rather oblique evidence for the intellectualist hypothesis strictly speaking.²⁰ In the Meno, a similar argument from successful agency is taken as a reason to identify virtue with knowledge—but this is the later argument overturned by the counter-argument that true belief should also count. So the argumentation of the Euthydemus is neither complete nor the final word. And this is unsurprising given that it is explicitly offered as an exercise in protreptic—that is, as an advertisement for philosophy, not an exposition of its final results.

¹⁹ The problem is most obvious in the case of the Cratylus, an exercise in antilogia on the question whether there is a natural correctness of names, the upshot of which is unclear and remains highly controversial. ²⁰ This is all the more puzzling given that Socrates does group wisdom and virtue together just before and after the argument at 278d3 and 283a4 (cf. 283b3 and more generally 273d–275a).

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

28

 

The dialogues which discuss the particular virtues provide even less compelling evidence for taking ‘Socratic intellectualism’ as Platonic doctrine. The relation of virtue to knowledge is certainly a central topic. The Laches and Charmides discuss definitions of courage and sôphrosunê respectively as knowledge or wisdom (Laches 194d–199e, Charmides 165c–175a); and one definition of piety in the Euthyphro is as knowledge of how to sacrifice and pray (14b–15c). Indeed at Laches 194d, Nicias introduces his intellectualist account of courage by saying to Socrates: ‘I have often heard you say that every one of us is good with respect to that in which he is wise and bad in respect to that in which he is ignorant’ (194d).²¹ But all of these definitions are rejected for various reasons: these definitional dialogues end in aporia, with no satisfactory account of the virtue in question. The aporetic form of these dialogues, and their deployment of antilogikê, do not imply that it is impossible to extract any positive philosophical results from them. Nor does the complexity of their interrelations, with the prima facie results of one dialogue reconsidered and sometimes overturned in another, show that no overall story about their doctrinal content can be told. But both factors should make us wary of any assumption that they must be read as straightforward exercises in the exposition of doctrines to which Socrates and his author are themselves committed. On the contrary: the discussions in which Socrates considers the intellectualist hypothesis have none of the conclusiveness, comprehensiveness, and consistency which would provide the appropriate basis for the Aristotelian reading. Were it not for Aristotle, and certain modern assumptions about what philosophical writing must do, we would find it natural to speak of intellectualism not as a theory being affirmed by Socrates, but as a topic of debate: a fascinating proposal being subjected to repeated and apparently inconclusive examination. It isn’t like empiricism in Epicurus or materialism in Hobbes, a dogma to be hammered home; it’s more like forgiveness in Mozart and marriage in Wagner—a thematic obsession, an endlessly fascinating possibility returned to with different results every time.

2. The Intellectualist Hypothesis In this section I will say a bit more about what I take the appeal of the intellectualist hypothesis to be, and how we should understand ‘virtue’ and ‘knowledge’ in its equation of the two. To begin with another negative point: it should already be ²¹ This usually is read as attributing the whole of the account which follows to Socrates, as a matter of standing conviction. But not only does he go on to reject Nicias’ definition in the Laches itself; all that is here ascribed to him is the sort of inductive inference from non-moral kinds of goodness (the person who is good at playing the lyre is the one who wise about it, i.e. has been educated in it, and so forth) which we see in the Euthydemus argument, and which the Meno refutes. Except as otherwise noted, translations of Plato are from the various hands in Cooper (ed.) 1997, sometimes with revisions.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

     

29

clear that the traditional phrase Socratic intellectualism is a misnomer. For, on Plato’s own evidence, that virtue is knowledge is the signature promise of the professional sophists who present themselves as teachers of it—above all, Protagoras in the Protagoras. The advanced version of the politikê technê which he teaches consists, as he says, in the art of ‘good deliberation [euboulia]’, sufficient for success in both private life and politics (318e5–319a2).²² As already noted, both the Apology and the Protagoras represent Socrates himself as deeply skeptical of this advertisement (319a–320b, 361a, Apology 19d–20c with 22a–23b). If anything, Plato’s Socrates is presumably meant to strike us—and to indicate that the historical Socrates struck Plato—as a powerful prima facie counterexample to this intellectualist hypothesis. For he is clearly presented in the early dialogues as an exemplar of virtue (at least of a ‘human’ kind); and a crucial part of that virtue is the epistemic humility expressed in his vehement disclaimer of any teachable moral knowledge himself.²³ It is worth saying a bit more about the kind of knowledge which Protagoras claims to teach. First, as the Great Speech spells out, the concept of knowledge in play here is that of a technê: a skilled, specialized, systematic practice which produces reliable success in action. Shoemaking is a technê; so is horse-training; so is medicine. Since a technê can be taught, each technitês can be seen as having a double function: his proprietary activity, which is always in the service of some end or good, and the production through training of ‘another like himself ’. The claim—or hope—that virtue might thus be teachable was advanced by Protagoras and a range of lesser sophists, who offered themselves as teachers of it. And this promise of the kind of virtue (‘the human and social kind’, as Socrates says in the Apology (20b4)) is the historical seed of the optimistic vision which, appropriated and transformed by Plato, flowers in the Republic: the dream that with the right educational programme, a city might be able to reliably produce practically wise statesmen. It is worth pausing to note what does not follow from the intellectualist hypothesis so understood. For, contra the Magna Moralia, it is wholly compatible with the acceptance of non-rational prerequisites for virtue: almost every acknowledged technê involves non-rational training and practice. It does not immediately imply that all the virtues are the same body of knowledge (it is interesting that the Aristotelian and MM versions of the intellectualist reading, as cited in section 1, speak of epistêmai and phronêseis in the plural). In its original, Protagorean form it is not even a theory about all the virtues: at any rate in Plato’s presentation of him, Protagoras is emphatic that courage is wholly distinct from the kind of

²² In work in progress I argue for taking the Protagoras to present an accurate, terminologically careful picture of the historical Protagoras. ²³ Cf. Forster 2007 for a reading of Socrates which emphasizes the religious dimension of his disavowal of knowledge.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

30

 

social or political virtue he teaches, so that Socrates’ hard-fought identification of wisdom with courage amounts, for him, to the deduction of an unwelcome consequence (Prot. 329e, 349d, 350e–351a).²⁴ Without further substantial argumentative moves, it does not imply any particular conception of human motivation: it certainly does not entail that the emotions are motivationally inert, or that they can be reduced to intellectual judgements. The hypothesis is not really a psychological thesis at all; or at any rate the investigation which it prompts belongs not to psychology but epistemology. The question it poses is: what else can we determine about virtue, if we can determine it to be a kind of knowledge? So much for what ‘knowledge’ means in the intellectualist hypothesis. It is when we turn to the term ‘virtue’ that matters get really complicated. For the early dialogues make clear that several competing conceptions of aretê, virtue or excellence, were current in Socrates’ day. On one side stands what we might call the Homeric or functional conception of virtue: aretê is what makes someone function successfully in his social role.²⁵ So aretê sans phrase is what enables the successful functioning of the best person, the adult male aristocrat-warrior like Achilles or Odysseus: it consists in the practical intelligence and courage which make such a person honoured as a ‘speaker of words and doer of deeds’. But Homer also recognizes other, less self-serving virtues such as piety, reverence, and pity, whose relation to aretê is unclear. And the virtues of justice and moderation are traditionally conceived in very different, more obviously ‘moralistic’ terms; here the locus classicus of the tradition is instead Hesiod’s Works and Days, which preaches the central values of law-abidingness, fairness, and self-control. Now the virtue Protagoras promises is thoroughly functional: an art of successful living through sound deliberation. In Plato’s Protagoras, he explicitly promises to make his students ‘most powerful’ (or ‘most effective’) [dunatôtatos] both in private business and in matters of state, and in the latter respect to be effective in both

²⁴ Thus Plato himself signals that the intellectualist hypothesis as investigated in the early dialogues, as a thesis about virtue in toto, is not identical with the original Protagorean version: we might think of it as ‘intellectualism 2.0’. I take it that Plato’s alterations are intended as, so to speak, internal improvements, designed to clean the hypothesis up and present it in the most viable form available. After all, how could the virtue allegedly acquired by teaching ensure good deliberation everywhere, even on the battlefield; how could it demonstrate the sovereignty of knowledge over emotions and external stresses; how could it make a person steadily better and guarantee their success in life; how, in short, could it possibly have all the powers which Protagoras himself claims for it if it does not include courage? One way to read the Protagoras is as inaugurating Plato’s investigative project by arguing for this swapping out of version 1.0 for 2.0. (For a very different courage-centric reading of the Protagoras cf. Segvic 2008.) The Protagoras also features another plausible candidate for a Platonic alteration in the very ‘scientific’ understanding of knowledge or technê, modelled on epistêmai such as geometry and weighing, which is imported into Socrates’ understanding of the hypothesis in the form of the ‘measuring art’. It seems unlikely that Protagoras’ own art was conceived along such quantitative lines. ²⁵ See e.g. Meno 71e–72a, and Adkins 1960, with its still-helpful contrast between the cooperative and the competitive virtues. Cf. also Finkelberg 1998.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

     

31

action and speech—a promise which hearkens back to the double Homeric ideal.²⁶ So in its original sophistic form the intellectualist hypothesis seems to presuppose an updated version of this ancient conception, given an exciting modern sheen by the framework of technê. However, nobody in the Platonic dialogues, least of all Socrates, is committed to thinking about virtue only in functionalist terms. All show some loyalty to the moralistic or Hesiodic conception as well, particularly as regards the virtues of justice and moderation.²⁷ And in working through the hypothesis and its consequences, Plato seems also to be attempting to work out whether the two conceptions can combined to produce a coherent theory. That is: ‘virtue’ in the hypothesis as discussed in the early dialogues is intended to stand both for the Protagorean art of successful living through deliberative skill and for the ‘otherregarding’ or cooperative norms of Hesiodic justice. Plato’s discussions of the hypothesis embed a presumption that the two conceptions must somehow denote the same thing: that the science of successful living must converge with or incorporate justice, piety, and moderation rightly understood. In Plato’s attempt to work out this bold equation, each conception is transformed, and in the process the bar for the possession of virtue is dramatically raised. For mere cleverness in decision-making does nothing to entail Hesiodic justice, nor does the latter do anything to generate the former: the two must be fused in a practical wisdom which is (as Aristotle’s phronêsis will be) at once moral and intellectual, both other-regarding and the key to one’s own success. And if that synthesis can be worked out, a further equation which lurks within the hypothesis will go through: namely that the virtuous life (understood at least in part in moralistic Hesiodic terms) is also the happy life.²⁸ For some kind of happiness, success, flourishing [eudaimonia] is what the art of successful deliberation promises; if that art runs through justice and moderation, then Hesiod was right to insist that the practice of these virtues is in the agent’s own interest. Thanks to this inbuilt conceptual

²⁶ Strikingly, Protagoras’ phrase ‘most powerful in word and deed’ is also used by Thucydides to describe Pericles in introducing his first speech (I.139). For discussion see Segvic 2008 (17ff.). ²⁷ So for instance Meno advances a debased version of the functionalist conception, cuttingly summed up by Socrates as ‘virtue is the acquisition of gold and silver’ (78d); but then he immediately grants that this acquisition must be accompanied by justice and the other parts of virtue (78e). Socrates’ third, abortive eristic argument in the Protagoras begins by establishing that most people believe that sôphrosunê (here meaning, apparently, a kind of self-serving good sense) can be exercised in unjust actions; but his planned refutation would presumably have made use of intuitions to the opposite effect (333c–d). Note also the dramatic but utterly unacknowledged conflict between the conceptions of justice presented by Cephalus and Polemarchus in Republic I. ²⁸ I do not mean by this anything so precise (or extreme) as that virtue is necessary and sufficient for happiness, so that (as the Stoics will later claim) the wise man is happy under torture. Rather, I take it that the Platonic position involves the looser traditional idea of a ‘path’—a chosen way of life with a determinate result, contrasted with alternative ‘paths’ leading elsewhere. That the path of virtue is the way to happiness is a doctrine common to all the philosophical descendants of Socrates, and also ancient competitors such as the Epicureans. We might fairly think of it as a shared commitment partially constitutive of ‘philosophy as a way of life’ in the ancient Greek world.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

32

 

complexity, in exploring the intellectualist hypothesis, the early dialogues also indirectly explore the viability of the proposal for which the Republic will explicitly argue: that the fully rational agent—the expert deliberator, master of the art of politics—will practice the other-regarding or cooperative virtues as a part of his excellence, and will attain happiness in so doing. If this is the origin and import of the hypothesis, then when Plato undertakes to investigate it—sympathetically, reflectively, and in great detail, with Socrates cast as lead investigator—he is performing a very complex philosophical operation. There is an aspect of reconciliation to it, an attempt to progress dialectically beyond the opposed camps of the earlier generation. His thought, I take it, is that the historical Socrates was quite right to suppose that Protagoras and the other sophists had no hope of making a technê out of virtue and teaching it, not least because they failed to fully grasp what either virtue or knowledge really is. But that does not show it to be an impossible dream. Perhaps a very different kind of philosopher—a Socratic philosopher, indeed—might be able to succeed where Protagoras failed. Such a philosopher is at any rate in a position to give the sophistic proposal the kind of rigorous philosophical development and examination it deserves.

3. The Method of Hypothesis So, in place of the enigmatic little slogan ‘virtue is knowledge’, the proposal on the table in Plato’s early dialogues may now be put as follows: there is an art of politics—more generally, an art of living—along the lines of the deliberative technê which Protagoras advertised but failed to deliver, which incorporates the traditional moral virtues including Hesiodic justice, and which promises its possessor happiness. I now want to turn to questions of method, and to the status of this as a ‘proposal’. More precisely put, the intellectualist thesis is discussed by Plato as a hypothesis—in fact, as the hypothesis around which the early dialogues are organized.²⁹ What this status as a hypothesis means can be gathered from the two pre-Republic passages in which Socrates discusses hypothesis explicitly, the Meno and Phaedo.³⁰ (In addition to the arguments which directly accompany ²⁹ ‘Hypothesis’ is a fortunately close counterpart to the Greek hupothesis: literally something one hupotithêtai, ‘sets down ’. Wolfsdorf 2008a gives a helpful survey of the early usage of hupothesis (2008a, 37–41). He concludes that Plato’s use in the Meno is of ‘a method of reasoning from a postulate’ (161; cf. also Wolfsdorf 2008b). This sounds right to me—but only because I do not hear ‘postulate’ as implying certainty or knowledge, as Wolfsdorf does. Obviously ‘a cognitively secure proposition’ (2008a, 177) would be the best kind of starting point for argument, but certainty is by no means presupposed by that functional role; an overt act of adoption or postulation indeed may draw attention to its absence. ³⁰ After long neglect, the method of hypothesis has recently been the subject of welcome interpretive attention: in addition to Wolfsdorf 2008a and 2008b, see Gentzler 1991, Benson 2015, Palmer 2021, and the more specialized discussions of Bailey 2005, Benson 2003, Byrd 2007, Ebrey 2013, 2016, and

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

     

33

Plato’s two discussions, our sample set includes the discussion of the theory that the soul is a harmony, referred to as a hypothesis at Phaedo 93c10.) Hypothesis is also prominent in the Republic and in the Parmenides, but these later versions are more philosophically technical and specialized; I will set them aside here. The accounts of the Meno and Phaedo raise enough puzzles of their own: they seem to use the concept ‘hypothesis’ in quite different ways, and the relation of Socrates’ programmatic remarks to his practice is not very clear in either case. What follows is an intentionally low-resolution sketch of the kind of procedure which seems to be envisaged and applied in the two texts.³¹ My hope is only to get clear enough about how hypothesis works in these cases to explain the central claim of this paper: that, if we pull the camera back, we can recognize the same kind of pattern, on a grander scale and with vastly greater complexity, structuring the early dialogues as a whole. In the Meno, Meno repeatedly asks Socrates to pronounce on whether virtue can be taught, while Socrates tries to shift the discussion to the more basic question of what virtue is. Socrates finally agrees to investigate Meno’s question ‘by means of a hypothesis’ (86e). Hypothesis as introduced here seems to be related to the geometrical procedure known later as analysis;³² the use of hypothesis is thus part of a broader Platonic aspiration to bring mathematical-style rigour to bear on questions of ethics, also visible in his enduring fascination with the idea of measurement (Euthyphro 7b–d, Protagoras 356a–357d, Statesman 283c–285c). Socrates even gives a geometrical illustration of the method, though this is so confusingly expressed that it only makes matters more obscure.³³ Still, the basic procedure emerges clearly enough. If I want to know whether X has some property Y, I can adopt a biconditional to the effect that all Z’s are Y and all not-Z’s are not-Y. Then I can answer my question by determining whether X is or is not a Z, which may be more tractable than my initial question. In the case at hand, Socrates proposes, something is teachable if and only if it is a kind of knowledge. So if I want to know whether virtue is teachable, I can answer my question by discovering whether or not virtue is knowledge. Socrates then proceeds to argue that, since virtue is always beneficial, and of all the things in the soul only wisdom is reliably so, virtue must be wisdom (87d–89a)—i.e. knowledge, so that it is indeed taught and learned (89b–c). He then reconsiders, however, arguing on a range of empirical grounds that virtue cannot be taught and so must not be knowledge after all (89d–96c, cf. 98e). The way out of the impasse is to see that forthcoming, Franklin 2010, Ionescu 2018, Iwata 2015 and 2016, Landry 2012, Menn 2002, Scolnicov 2018, Scott 2006, Sedley 2021, and Costello (ms). The classic account of Robinson 1953 also continues to be very useful, and what follows is close to it on many points. ³¹ Palmer 2021 and Costello (ms) argue for readings of the Phaedo as deploying the method still more widely (cf. also Byrd 2007, 150–1), which seems to me highly plausible. ³² On the mathematics, see Menn 2002, Iwata 2015, and Wolfsdorf 2008a, 170–7. ³³ Socrates’ geometrical example may be intended to confuse Meno, who has expressed a preference for pretentious definitions (cf. Weiss 2001 and Lloyd 1992). But cf. Menn 2002.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

34

 

true opinion is also beneficial, and that the virtue of respected statesmen is due to divine luck—though the possibility of a virtue constituted by knowledge is carefully left open (97a–100b). The tricky question here is exactly which of these propositions count as hypotheses.³⁴ The claim that virtue is good is explicitly labelled a hypothesis when introduced at 87d3;³⁵ but since the biconditional is the central structural feature which Socrates’ procedure shares with his geometrical example, it seems to me natural to suppose that this is the principal instance of the method being displayed.³⁶ Moreover, Meno goes on to speak of the application of the hypothesis in a way which implies it is the biconditional, remarking at 89c2–4: ‘And it is clear, Socrates, that according to [kata] the hypothesis, if virtue is knowledge, it is teachable’ (cf. 87c2–3). Thus, as a method, ‘hypothesis’ seems to be a strategy for reducing an obscure problem to one which I know how to solve by using both unargued postulates as starting-points and biconditionals so as to substitute coextensive concepts. The more complex use of hypothesis comes in the Phaedo, in the wake of Socrates’ famous account of his youthful enthusiasm for, and disillusionment with, natural science. In describing his own alternative, the so-called second sailing, he refers to it as proceeding by hypotheses: Well then I started in this manner: taking as my hypothesis [hupothemenos] in each case the account [logon] which I judged to be strongest [errômenestaton], I set down [tithêmi] as being true, both about cause and everything else, the things which seem to me to agree [sumphônein] with this, and as not true whatever does not. (100a2–5)

Socrates’ hypotheses here are centred on the Forms; his demonstrandum, and so the rough counterpart here to the Meno problem about the teachability of virtue, is

³⁴ Iwata 2016 gives a helpful scorecard of positions in the recent English-language literature. ³⁵ It is introduced as ‘standing’ or ‘staying in place’ for us [menei]: a suggestive hint as to the norms governing Socrates’ procedure, along with ‘strongest’ [errômenestaton, Phaedo 100a4] and ‘sufficient’ [hikanon, 101e8]. The statues of Daedalus passage later in the Meno (97e) indicates that even a true claim may fail to ‘stand’ for us if we consider it to have been refuted or otherwise lose faith in it, and that we are at risk of this happening so long as we do not understand why it is true. So if these norms are related, and a strong hypothesis is (perhaps among other virtues) one that will ‘stand’ for us, then we should from the start be looking for a hypothesis for which we have an explanatory higher hypothesis ready to hand. ³⁶ This is controversial. A more popular alternative candidate for the hypothesis, preferred by Robinson in his second edition (1953) and followed by many interpreters since, is ‘virtue is knowledge’. This would obviously be useful for my argument that this thesis is the organizing hypothesis of the early dialogues as a whole! Alas, ‘virtue is knowledge’ is here deduced as the conclusion of a complex chain of reasoning at 87d–9a: it is not something hypothesized, ‘set down’, at all. Robinson is thus forced to admit that the passage is ‘not very like Platonic hypothetical method in general as I described it in the previous chapter’ (1953, 121). It also entails a very strained reading of Meno’s remark at 89c2–4; and there is no obvious parallel with the geometrical example.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

     

35

the immortality of the soul. There is again some blurriness as to the identity of the hypothesis itself: so far as I can see it could be any or all of (a) the relevant Form (so that the hypothesis proper is not a proposition at all, but an object, the Beautiful itself), (b) the thesis that the Form exists (and has the predicate which it is: the Beautiful itself is beautiful, for instance), or (c) the causal claim linking Form and explananda (‘beautiful things are beautiful by the Beautiful itself ’). Each of (b)–(c) seems to presuppose and incorporate its simpler predecessor(s), and I suspect that the hypothesis should be thought of as comprising the whole package (a)–(c). Socrates goes on to explain how the hypothesis itself may become the object of explanation and argument. One should hold fast to the hypothesis as a ‘safe’ answer: but you, afraid, as they say, of your own shadow and your inexperience, would cling [echomenos] to the safety of your own hypothesis [hupothesis] and give that answer. And if someone were to take hold of [echoito] your hypothesis itself, you would ignore him and would not answer until you had examined the consequences that follow [ta hormêthenta] from it, to see if they agree [sumphônei] with one another or disagree [diaphônei]. And when you must give an account of your hypothesis itself you would give it in the same way, setting down [hupothemenos] another hypothesis again, whichever one seems to you best of the higher ones until you come to something sufficient [ti hikanon]; but you will not jumble the two as the debaters [antilogikoi] do by discussing the principle [archê] and its consequences [hôrmêmena] at the same time, if you wish to discover any truth. (101c9–e3)³⁷

We do not see either a refutative attack on the hypothesis or regress to a higher one on display in the Phaedo argument itself, which from here on proceeds only ‘downwards’ to the desired conclusion that the soul is immortal. But Socrates concludes with a reminder that more discussion is needed (107b). Moreover, earlier in the Phaedo, the theory that the soul is a harmony of the body (91d–95a) was identified as a hypothesis (94a12–b2); and here the hypothesis is attacked. The soul-harmony hypothesis was shown to be incompatible with ³⁷ I follow Grube’s translation (in Cooper 1997), with modifications, which is especially controversial on two points. Sedley 2021 argues, first, that Socrates intends at 101d2 to distinguish between the safe part of the hypothesis, the ‘ultra-safe virtual tautology’ (53) that it is (for instance) by beauty that beautiful things are beautiful, and the unsafe part, namely the postulation of metaphysically transcendant Forms such as Beauty itself. I am not convinced, however, that Plato would allow that the parts of his causal hypothesis can be separated in this way. Second, Sedley rejects Grube’s translation of echoito as ‘were to attack’, arguing for a new understanding of it as non-hostile (2021), 57. I translate ‘take hold of ’, for to preserve neutrality on that question; but it seems to me that Grube is likely right to suppose that hostile ways of ‘taking hold’ are primarily in view. It seems relevant that ‘safety’ is, unlike the other terminology here in play, quasi-technical as a term of dialectical art: a safe answer is one which an eristic opponent will be unable to refute (cf. 100d8, 100e1, 105b7–8, c1, Protagoras 351d3, Sophist 231a7, Statesman 262b6).

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

36

 

the better-warranted theory of recollection, and also with a set of uncontroversial truths about souls and harmonies. A harmony is determined by its parts; a soul is not. A harmony cannot be disharmonious; a soul can be. A harmony can exist as a matter of degree; not so a soul. A harmony could not rule what it is a harmony of; the soul can rule the body. As even this rough sketch shows, the Meno and the Phaedo references of hypothesis are, at least superficially and in emphasis, quite different from each other; and Plato’s practice is another can of worms again. With those large caveats, Plato’s pre-Republic method of hypothesis seems to me to have six principal features:³⁸ (1) A hypothesis is a thesis literally set down or posited [tithetai], with the hupo suggesting a ‘below’ or ‘underneath’, i.e. a kind of floor or starting-point for argument. It is not itself presented as established by argument; and even if a hypothesis might be chosen as secure or self-evident, to explicitly perform positing something is to flag it as having this functional role rather than any particular epistemic status.³⁹ The identity of some thesis as a hypothesis is thus situational, and hypotheses are unrestricted as to content. How then do we select a hypothesis, and what counts as a good one? In the Phaedo, Socrates says only that we select whatever thesis seems ‘strongest’, which looks like deliberate vagueness on his part. Given the explanatory role of hypothesis (see (2) below), ‘strength’ presumably involves both explanatory power and some independent philosophical attraction.⁴⁰ The hypothesis that the soul is a harmony, for instance, is one which deeply attracts Simmias, and has the merit of being compatible with some of Socrates’ earlier arguments (85e–86d); the hypothesis of Forms as causes is one Socrates himself takes to be explanatory and defensible across the board. (2) A hypothesis is used to derive other propositions: the consequences [hormêthenta and hôrmêmena]. This derivation of consequences is, in the Phaedo, initially phrased in terms of setting down what agrees or harmonizes [sumphônei] with the hypothesis as true, and whatever doesn’t so harmonize [diaphônei] as not true (100a). ‘Consequences’ is a standard but perhaps misleading overtranslation of the plural participle, ‘things coming from’. This is more

³⁸ Cf. Robinson 1953, esp. 105–9, and Benson’s four-stage analysis (2015). ³⁹ Cf. Iwata 2016, ‘a plausible but still suppositional starting-point’ (210). ⁴⁰ A deliberate untechnicality, even for Plato’s day and by his own standards, pervades the Phaedo passages: notably, he does not use sumbainein, to ‘result’. This is the closest thing to a standard term for entailment (as at Phaedo 80b1), though the root idea is the broader one of how something ‘turns out’ to be (e.g. Phaedo 67c5, 68e4, 74a2 (cf. Benson 2015, 144)). And archê, ‘first principle’ or ‘starting-point’ is used for a hypothesis (apparently) only towards the end (101e2). This supports my sense that what most interests Plato about hypothesis is explanation, and that he is groping towards a new vocabulary for explanatory adequacy. He does use sumbainein in the more deductively oriented passage of the Meno (87a6, b1).

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

     

37

commonly used for people physically coming from some place: a deliberately vague ‘what proceeds from’ or ‘comes out of ’ the hypothesis might be better. ‘Harmonizing’ and ‘disharmonizing’ or ‘disagreeing’ (as I will translate) are, as applied to propositions, likewise hard to pin down—though here too, as with hupothesis itself, the English is a luckily close counterpart to the Greek, similarly used for musical chords. It is easy to see why Plato is vague here. For harmonizing, as the norm governing both the hypothesis–consequence relation and the relation of consequences to each other (Phaedo 100a5–7, 101d4–6), is very tricky to make out. It cannot consistently be taken either as mere logical compossibility or as entailment, on pain of absurd results. To understand harmonizing as mere compossibility might work for the desired relation of consequences to each other, but gives hopeless results as a standard for their derivation. If harmonization is restricted to entailment, though, then it seems impossible as a demand on the consequence– consequence relation.⁴¹ The ‘disagreement’ of propositions with a hypothesis, and of consequences with each other, presents similar puzzles.⁴² To make sense of Socrates’ programmatic remarks, as scholars have noted, we may need to allow that harmonizing represents something different in the two relationships; and it must in any case pick out some relation between the extremes of strict entailment and mere compossibility.⁴³ Scholars have also argued that the process of deriving consequences must involve bringing to bear other relevant beliefs and assumptions to combine with our hypothesis, as when a scientist incorporates background assumptions in order to determine the implications of a theory—or when Socrates introduces anodyne-looking auxiliary premises in order to refute some interlocutor’s definition of virtue. Insofar as such auxiliary premises are secure, the hypothesis itself can then be taken to bear full responsibility for any philosophically decisive consequences, whether welcome problem-solving ones or an unacceptable contradiction.⁴⁴ I would note, though, that in the case of the harmony theory, Socrates’ refutation proceeds by a rather precise method of teasing out the opposed necessary properties of souls and harmonies, which show that the two cannot be identified. This suggests a capacious view of the ⁴¹ But see Forster 2006. ⁴² I take it that diaphônia, disharmonizing or disagreement, is the same relation as when a proposition ‘does not agree’ [ou sumphônei] with the hypothesis; and that it is to be understood as a relation of conflict, i.e. contradiction. But if harmonizing must be a tighter relation than mere compossibility, we get the paradoxical-sounding result that the two relations are not exhaustive, even though the one is presented simply as the negation of the other: that is, many propositions will neither agree nor disagree with the hypothesis, and some consequences may neither agree nor disagree with each other. But this is perhaps as it should be; if the agenda of the method is explanatory, many propositions will have no interesting relation to each other. ⁴³ See Gentzler 1991, followed by a number of more recent scholars, for argument that the relation of harmonization should be seen as one of coherence (268–9). Cf. also Bailey 2005, for a corrective emphasis on the important of explanatory relations here. ⁴⁴ Cf. Benson 2015, 173–5, 182; Gentzler 1991, 273ff.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

38

 

soul-harmony hypothesis as involving the postulation of souls conceived as harmonies, pregnant with all the implications of the essential properties implied by both terms. On this conception of what a hypothesis is, the contradictions which Socrates elicits could indeed be seen as coming out of the hypothesis alone.⁴⁵ So perhaps, at least in this kind of definitional case, the derivation of consequences from a hypothesis should be seen as a kind of making-explicit: a matter of spelling out the properties of the definiens and definiendum which the hypothesis of identity transfers from the one to the other.⁴⁶ (3) This procedure of deriving consequences is a multi-purpose tool.⁴⁷ It is open to deployment in at least three kinds of philosophical inquiry: it may be aimed at (i) settling an open question, or (ii) establishing a desired demonstrandum, or (iii) testing the hypothesis itself for acceptability. This is where the Meno and Phaedo present very different pictures. In the Meno, which seems to represent the simpler form of the method, the project at hand is of type (i), founded on the analysis-like use of the biconditional. In the Phaedo, it is (ii): there is a puzzle to be solved, but one side of the question is presumed to be correct, and the task at hand is to deduce that desired result (viz., that the soul is immortal). Task (iii) is the critical version of the method, exemplified in the examination of the thesis that the soul is a harmony. (4) At least in the third kind of deployment, the consequences of the hypothesis are examined for harmonizing or the lack of it, their disharmony or disagreement [diaphônia]—in practice, for any contradictions. And this emphasis on contradiction is worth highlighting as an important clue to the significance of the method. We are not to reject a hypothesis on the basis of a vague intuition that it is implausible, but only if its consequences are such that it can be excluded as impossible. The method is a vehicle for tracing networks of entailments and ⁴⁵ Compare the important account of Socratic refutation given by Forster 2006, defending Robinson’s much-deprecated (but textually well-supported) claim that Socrates in his refutations at least sometimes aims to show the thesis of his interlocutor to be self-contradictory, without the introduction of any genuinely independent auxiliary premises (9). The account of ‘sense-internality’ by which Forster supports this reading comes close to being an account of what a capacious understanding of hypothesis might look like. ⁴⁶ On this reading the method emerges as a kind of missing link between fifth-century eristic and Aristotelian science. In eristic, a respondent might undertake to defend the thesis: ‘The soul is a harmony.’ The questioner would try to refute him by forcing him to assent to the claim that the soul is not a harmony, by the same sorts of steps as we see in the Phaedo refutations. (Socrates’ strategies of attack are all natural candidates for sophistic topoi, applicable to any proposed identity claim: ‘Do both these things come in degrees?’ ‘Do they have the same opposite?’ ‘Does each have the same relation to its parts?’) This eristic method becomes philosophically constructive dialectic whenever the players are less interested in victory than in figuring out what the soul really is, by figuring out what exactly the proposed definition entails and so reaching a verdict on its acceptability. In Aristotle’s scientific (and no longer dialogic) version of the method, a person with knowledge of a certain kind deduces the necessary properties of that kind from its definition—a constructive version of the procedure, starting from a hypothesis known to be correct. ⁴⁷ Robinson 1953 argues plausibly that the Meno version, in which the hypothesis is a tool for the deduction of consequences, is prior, and that the Phaedo version, in which the hypothesis itself becomes the demonstrandum or refutandum, evolves from it (110–13).

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

     

39

conceptual connections, and detecting options which can be eliminated. As Sherlock Holmes says, ‘when you have eliminated the impossible, whatever remains, however improbable, must be the truth.’⁴⁸ (5) The hypothetical method is multi-level and open to regress or ascent, in that a hypothesis can itself, as the Phaedo rather vaguely explains, be supported by relation to a ‘higher hypothesis’. Like being a hypothesis, the status of a higher hypothesis is situational: a higher hypothesis will be any thesis selected as having suitable explanatory power to ground the original hypothesis—and, perhaps, some independent philosophical attraction. In section 4 I will briefly note a possible candidate in the early dialogues for a ‘higher hypothesis’ in relation to the intellectualist hypothesis. In the Republic, the Divided Line presents the Form of the Good as a special limit case: an unhypothetical highest principle for the Forms as a whole (511b–d). However exactly this is supposed to work, it brings out the essential limitation of the method in the earlier dialogues, where there is no such deus ex machina: hypothetical reasoning, because its starting-points are unproven, can never be fully demonstrative.⁴⁹ And this seems worth listing as a final feature: (6) Hypothetical argument is inherently provisional and defeasible, always open to rethinking and further examination—a status ostended by Socrates through his immediate reconsideration in the Meno (96dff.) and warned of in the Phaedo (107a–b). To fully demonstrate a hypothesis, one would have to somehow establish the truth of the associated highest hypothesis and deductively confirm everything said to ‘follow’ from it—thus rendering the whole package unhypothetical. In the absence of this, the method must start from unproven postulates. And so it is perhaps more decisive as a critical tool than a constructive one: it can only clarify the menu of philosophical options, enabling us to exclude some of them as impossible. So much for hypothesis as it appears in the Meno and Phaedo. As I have noted, many difficulties and controversies surround Socrates’ remarks on his method; and they are less than a perfect fit either with each other or with Platonic practice. But I think it should be uncontroversial that the six features I’ve identified do all characterize the dialectic of the early dialogues quite generally. In them, Socratic argument typically (1) starts from posited theses (such as candidate definitions of the virtues) from which (2) consequences are elicited, with (3) a range of argumentative agendas and results, but (4) often so as to elicit a contradiction, (5) sometimes with a regress to the positing of some ⁴⁸ The Sign of Four, ch. 6. Given that the last words of Conan Doyle’s ‘The Final Problem’ allude to those of the Phaedo (118a15–17), the resemblance between Holmes’ methods and Socrates’ seems unlikely to be coincidence. David Ebrey points out to me, however, that Socrates’ method cannot really be meant to exclude all the alternatives in the Holmesian manner. ⁴⁹ I here assume that the Good is a grounding for lower hypotheses in the same way that they ground each other: for a challenge to this conventional reading see Broadie 2021.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

40

 

more explanatorily basic thesis,⁵⁰ and (6) always with defeasible, provisional results—as Socrates habitually reaffirms by emphasizing the need for re-examination and further discussion. Many of these features are obvious and pervasive, and all have been widely recognized. But in recent accounts they have tended to be distorted or obscured by an overly restrictive conception of the method of the early dialogues, framed in terms of the ‘Socratic elenchus’.⁵¹ By the elenchus, commentators mean the kind of refutative argument in which Socrates elicits a target thesis (usually a proposed definition) from his interlocutor; adds some auxiliary premises presumed to be uncontroversial; and, eliciting their joint consequences, derives a contradiction. The target thesis is then rejected as responsible for the contradiction, though all that the method can prove strictly speaking is that some one of the premises must be false—and therefore, importantly, that the interlocutor is not wise. That some of Socrates’ arguments are of this form is indisputable; but we should now be able to see that the elenchus is just a subspecies of the hypothetical method. Elenctic arguments are hypothetical arguments of type (iii) above, also sharing three further specific characteristics: (a) the hypothesis is one advanced by the interlocutor, (b) the discussion proceeds by question and yes–no answer, and (c) the result is a contradiction or otherwise unacceptable result. The broader framework of hypothesis should be preferred as better capturing the texts, for there is no significant subset of the early dialogues in which all of Socrates’ arguments belong to this narrower subspecies. He is also setting down hypotheses of his own and working out their consequences even in echt-Socratic dialogues like the Euthyphro (11e–12e), Crito (46b–49a), and Protagoras (352a–d). (As David Wolfsdorf has noted, Socrates himself proposes eight of the twenty-nine definitions considered in the early dialogues (2008a, 150).) In the Apology, in addition to describing his refutative mission, Socrates affirms the importance of examining the ethical principles guiding one’s own life (38a); and the Crito depicts him doing precisely that (most explicitly at 46b–49e). This is sufficient to show that if Socrates’ arguments in most of the other early dialogues skew towards the refutation of

⁵⁰ This is probably the least obvious of the features listed. I have in mind cases like the shift in the Protagoras from Socrates’ arguments for the piecemeal unity of various virtues (329d–334a) to the argument that they are all wisdom (349a–360e); unlike the first round of arguments, the second would explain why justice and piety are the same, and it in turn is explained by the principle that everyone pursues what seems best to them (358c–d). Cf. Allen 2006 for an account of the difference, and Kamtekar 2017 for a reading of the latter part of the dialogue as structured by hypotheses. The Gorgias seems to me also structured as a regress of hypotheses, here organized for the purposes of refutation (like those of the first part of the Theaetetus, incidentally): Gorgias holds that rhetoric is the finest of crafts, Polus that power is sufficient for happiness (which would explain why rhetoric is finest of the crafts), Callicles that pleasure is the good (which would explain why power is sufficient for happiness). ⁵¹ Cf. Robinson (1953), 106–7; Benson, ‘it must be admitted that the difference between the method of hypothesis and elenchos may appear rather slight’ (2015, 100). For a helpful survey of the elenchus literature (up to 2010), see Wolfsdorf 2013. For a welcome array of heterodox readings see Scott (ed.) 2002.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

     

41

others, this is a function not of his method as such but of the ongoing philosophical project in which he has been enrolled by his author. And the hypothetical reading can explain why his project has this other-directed slant: the intellectualist hypothesis is not Socrates’ own, and the obvious way for him to examine it is by interrogating those who do propound it, or who lay claim to wisdom or virtue in some way which might illuminate it. In the next section I will put the results of this and the preceding section together to sketch how the early dialogues are organized around the intellectualist hypothesis. But first it is worth pausing briefly to say a bit more about the hypothetical method itself, and in particular about what makes it distinctive and exciting. For if the details are murky, we can still see in hypothesis a powerful contrast to the alternatives of the day. That contrast is thematized in the Gorgias, where Polus is simply dismissive of Socratic refutation, on the basis of an appeal to ‘witnesses’, i.e. to what most people would find credible (471e–472c, 473e). Though conflated here, really there are a pair of alternative strategies of argument which Polus simply assumes to be jointly decisive: appeals to intuitive plausibility and to testimony. Both are native to the Athenian courtroom, where arguments typically appeal either to eyewitness testimony or to to eikos—to what is probable, plausible, likely, or reasonable to expect, particularly in relation to human motivations and to patterns of cause and effect.⁵² Plato dramatizes this pair of methods at the start of the Laches, where Laches and Nicias give opposed speeches on whether young men should study fighting in armor (181e–184c). Nicias appeals to to eikos, giving a long list of the beneficial consequences we would expect to result from the training. Laches refutes him with testimony, telling a long unfunny story about a teacher of the training who made a fool of himself in actual combat. When Socrates enters the fray, his strategy is very different from both. He proceeds by identifying a single general point of principle which would settle the question: if this training makes men courageous, it should be practiced. What follows is a regress of hypotheses to determine whether the antecedent is something we have reason to accept.⁵³ The Laches thus enacts the basic proposal involved in the hypothetical method. When faced with a controversial ethical question, rather than trying to assemble a mass of considerations which would tip the balance of probabilities one way or ⁵² Cf. Wohl (ed.) 2014 for a range of aspects of the eikos in Greek thought. ⁵³ Of course, that courage is a kind of knowledge would not be a sufficient condition for determining that it is taught by this form of instruction. Strictly speaking, the investigation undertaken in the Laches could be sufficient neither to prove that fighting in armour should be studied (since even if courage is knowledge, it might not be learned through this discipline) nor to refute it (since even if courage is not knowledge, there might be other reasons to study it). In this it exemplifies the necessarily incomplete and provisional character of any one line of hypothetical argument. On the other hand, the hypothesis which Socrates investigates, as it eventually emerges, is not just the bare thesis ‘courage is knowledge’ but something richer: a conception of courage as a kind of science of goods and evils analogous to the other technai. And that hypothesis not only implies that learning to fight in armour could not contribute to the acquisition of courage, but enables us to see why it could not.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

42

 

another, we should try to identify a single thesis which would be sufficient to settle the question at hand—one with some explanatory power, not just giving a verdict on the question but enabling us to see why it is so. We are then to investigate that thesis, working upwards to more basic principles which would ground it, and downwards to what would follow. Throughout, we are to seek implications, explanations, and contradictions instead of merely gauging the levels of intuitive plausibility which might accrue to each thesis in isolation.

4. Intellectualism: Consequences and Higher Hypotheses We can now put the pieces together to consider my central proposal: that the early dialogues are collectively organized around the hypothesis that virtue is a kind of knowledge or craft. Of the kinds of hypothetical argument I identified in the previous section, type (iii), in which the aim is to test the hypothesis itself, seems to be the pattern applicable to the inquiry as a whole. However, local arguments that look more like cases of the first or second will still fit this reading: for the ability of a hypothesis to (i) successfully solve open questions and (ii) secure important demonstranda are among the main advantages that might commend it. And the grounding of a hypothesis through a higher hypothesis will also obviously be in order here: given the novelty and importance of intellectualism, it is well worth Plato’s while to consider what deeper principle might explain it in turn. The most striking evidence in favour of the hypothetical reading is the way in which the early dialogues showcase discussions, for and against, of propositions which are best understood as ‘consequences’ of the intellectualist hypothesis. The single most prominent consequence is Teachability: if virtue is knowledge, then virtue can be taught. As I argued in section 2, and as the dialectic makes clear in all of the Protagoras, Laches, and Meno, this sophistic aspiration provides the original philosophical impetus for the hypothesis and its investigation as a whole. As noted already, the investigation of this question is complex, and does not yield a simple yes–no verdict: the results of the Protagoras are avowedly confusing and inconclusive (360e–361d), likewise the Meno (86e–100b). Indeed in the Meno it seems that the hypothesis is refuted by the derivation of contradictory consequences. For Socrates argues here as follows: (1) Whatever is teachable is knowledge (87b–c); (2) virtue is knowledge (87d–89a); (3) therefore virtue is teachable (89b–c); (4) but, empirically, virtue seems not to be teachable (89d–96d); (5) so virtue is not knowledge, or at least not always knowledge, after all (96d–99a).⁵⁴ Since the ⁵⁴ Again, we are faced with the question why the contradictory claims here, (2) and (5), would be seen as ‘coming out of ’ the hypothesis, given that (2) and (4) (on which (5) depends) seem to be independent of it. The only explanation I can see is that Socrates takes the biconditional to be pregnant

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

     

43

inquiry shows that neither accepting that application nor rejecting it gives a viable result—for Socrates can refute both ‘virtue is knowledge and is teachable’ and ‘virtue is not knowledge and is not teachable’—then it would be reasonable to conclude that the fault is with the hypothesis itself. At the same time, the case of the Meno shows that such a finding of disharmony may represent only a preliminary result. Socrates goes on to provide what later dialecticians would call a lusis of the problem: both (2) and (4) (and with it (5)) need to be rethought and more carefully put, and when we do so the contradiction is eliminated.⁵⁵ For knowledge is not the only beneficial thing in the soul, and so it need not be the case that all virtue is knowledge: most often it consists in divine luck and right opinion. This still leaves open the possibility that some is knowledge and that this kind could be taught, if only by a rare ‘Teiresias’ figure.⁵⁶ So the Meno shows that even conclusive-looking type (iii) uses of hypothesis are only provisional: when the consequences of a hypothesis are disharmonious, the right response may still not be to reject it, but rather to see whether a more careful rethinking might dissolve the problem. In truth, neither the biconditional nor the opening hypothesis that virtue is good and beneficial is here at fault. The consequences of intellectualism also include a different kind of application of the central hypothesis: namely the claims discussed in the Charmides, Laches, and Euthyphro that the specific virtues or parts of virtue—respectively, sôphrosunê, courage, and piety—are kinds of knowledge or technê.⁵⁷ These Specificationconsequences figure in the dialogues as both an attraction and a locus of problems. The Charmides’ investigation of sôphrosunê increasingly comes to focus on the idea that it is a reflexive kind of knowledge, a knowledge of knowledge itself and of ignorance (165c–176b). Thus it recalls the suggestion made by the Apology that Socrates’ self-knowledge and humility—his awareness of what he does not know— might itself be what human virtue is. By representing such self-awareness as an epistêmê and technê, the Charmides explores the prospects for reconciling this anti-intellectualist-looking echt-Socratic conception of virtue with the intellectualist hypothesis. The Laches and the Protagoras both discuss the hard case of

with all the implications of its application to the case at hand—that is, to virtue. Republic 436eff. confirms that for Plato the consequences [here sumbainonta] of a hypothesized principle include the implications of its application to a particular case, the case in point being the application of the ‘principle of opposition’ to cases of desire and aversion. Cf. also the consequences of the hypothesis deduced in the latter part of the Parmenides [sumbainonta ek tês hupotheseôs, 136a1, 142b4, with talk of examining them in relation to itself and each of the others 136c1]. ⁵⁵ Socrates’ procedure is easier to understand if we bear in mind that hypothesis bears the DNA of eristic as well as of geometry. When eristic arguments reach contradictory conclusions (as for instance at the start of the Euthydemus (275e–278b)), solving them standardly involves some kind of disambiguation—which is what Socrates’ distinction between the virtue of the statesman and that of the Teiresias figure supplies. ⁵⁶ On which see Scott 2006. ⁵⁷ The Hippias Minor in effect does the same for justice, though it does not present as a ‘definitional’ dialogue.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

44

 

intellectualism about courage, with ambiguous results. The Laches finds that it conflicts with the commonsensical presumption that courage is merely a part of virtue, at least if we suppose that the knowledge which is courage consists in the knowledge of goods and evils (199e). The Protagoras vindicates the identity of courage with wisdom only via the higher hypothesis of hedonism, the status of which is notoriously in question. So the results in all these dialogues are largely negative and inconclusive. The Hippias Minor raises a further problem, regarding the specific virtue of justice: if a craft is a power for opposites, then it must be the part of the just person to do wrong voluntarily and with skill (376b). This problem of Bivalence is raised again in Republic I.⁵⁸ These Specification consequences are enmeshed with some of the others. One of these is Content: if virtue consists in knowledge, it must have some determinate subject matter distinct from that of all the other crafts and sciences. (In fact Plato seems to assume that the subjects of real technai cannot even overlap, and that this stipulation can be used to weed out suspiciously broad-looking pseudo-crafts such as rhetoric (Gorgias 453d–459c, and cf. Ion 537d–540b).) We can distinguish as a second-generation consequence Plato’s preferred conception of that content, namely the knowledge of good and evil (KGE)—or, less misleadingly, of good and bad things, i.e. of value in general (Laches 199e, Protagoras 352b–360d, Charmides 174b–175b, cf. Euthydemus 292a–e). KGE is a promising and ingenious way to spell out Content, for it provides the determinacy which is distinctive of a craft while having the universal scope required for virtue conceived as a functional art of living, able to govern all deliberation. But KGE also presents problems of its own. For one thing, it implies that all the parts of virtue are really the same. Moreover, the early dialogues present as echt-Socratic, and independent of intellectualism, the idea that virtue itself is the human good, by being a kind of health of the soul (Crito 47e–48a, Gorgias 506c–509e). And in conjunction with KGE this produces a vicious circularity, in which the good is defined as knowledge of the good. This third-generation consequence Circularity is noted as a worry in the Euthydemus and Republic—and from another angle in the Charmides, which discusses whether there can be a knowledge which knows only itself (165e–171c). A solution is suggested in Republic VI: even if the human good is knowledge of the Form of the Good, since the two are distinct the claim is non-circular (505aff.). ⁵⁸ At Republic 332b–334a the idea of justice as a bivalent technê is again shown to lead to hopeless paradoxes; and at 335a–e it is established, contra Polemarchus, that the function of a just person can only be to benefit, not to harm. This entails not that justice could not be a craft, but only that if it is one, craft must not be bivalent. And sure enough, a new account of craft as unidirectional and inherently beneficent is then sketched in the discussion with Thrasymachus. One sign that we should not take either the HM or the discussion with Polemarchus as simply rejecting the conception of justice as a craft is that the paradox-generating final argument of the HM is presented as applying whether justice is a craft or a power [dunamis] (375d–376c). I take the latter to be the broader genus of which craft is a species—one so broad, indeed, that a virtue will count as one on almost any conception of it. So intellectualism is no worse off than any other theory here; and the revised, unidirectional conception of technê in Republic I shows how it alone can escape the paradox. Cf. Barney 2021.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

     

45

KGE is one of the consequences which seem to particularly occupy Plato, being important to the discussion in more than one dialogue. Another, derived from it in turn, is Unity: that is (as I understand it), that the virtues are not only mutually entailing but identical, consisting in a single psychological power. This perhaps could be viewed as a direct consequence of the intellectualist hypothesis itself; but it would seem to be possible in principle that each virtue might consist in a distinct body of knowledge. It is only once all the virtues are identified with a universal, comprehensive knowledge of goods and evils that there is no room left to differentiate them. Meanwhile Unity is problematic in turn: the Protagoras and Laches both bring out how implausible it (and for that matter Specification itself) is for the case of courage (Prot. 329e, 349d, 350e–351a, Laches 199e, cf. 190d). Several of these consequences point in the same further direction. If the whole of virtue must be present in the same person; if it requires an awe-inspiringly comprehensive grasp of all the goods and evils attainable by action; if to acquire it requires expert instruction (as well as practice, like any proper technê)—then it must be extraordinarily difficult to achieve. And so we must also suppose it to be vanishingly rare; and most of the people we usually count as in some way virtuous can only have attained some lower level of (or surrogate for) the art. I will call this pair of consequences Difficulty and Levels respectively. These are particularly evident in the Meno—and from there must feed in somehow to the ‘higher’ and ‘lower’ kinds of virtue we encounter in the Phaedo (68b–69d) and Republic (430c), though we should not assume that Plato’s distinctions are identical in all these texts. The end of the Meno, as I noted above, seems intended to disambiguate virtue into two levels, one constituted by knowledge and teachable, the other a matter of right opinion and due to divine luck. The former is clearly superior: but it is confined to the Teiresias figure, evoked in a conditional suggestive more of an aspiration or educational programme than an achieved reality. And the same aspirational stance is implicit in the Protagoras, if we are to take Socrates’ ‘art of measurement’ as a serious (not merely ad hominem) proposal for a practical wisdom constitutive of all the virtues. For surely almost no one has ever attained such a comprehensive, scientific, illusion-free grasp of the weight to be accorded to every possible deliberative consideration. Difficulty and Levels are triply important consequences of the intellectualist hypothesis. They are, for Plato, likely to have been independently attractive: not only for their moral elitism, but because they give the right kind of limited endorsement of Socratic virtue.⁵⁹ They also give a new configuration of a ⁵⁹ That Socrates is intended as an exemplary figure in the early dialogues is hard to doubt. But he is also markedly apolitical in any normal sense, and certainly not an obvious candidate for political office, as the possessor of the Protagorean technê would be. It is thus reasonable to suppose that his is not the highest form of social virtue conceivable. This is compatible with his claim that he alone ‘puts his hand to’ the art of politics at Gorgias 521d, which is hardly a resounding boast of expertise or success.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

46

 

powerful traditional Greek idea, that humans are precluded—or, perhaps, almost precluded—from a highest, divine kind of virtue. And third, they show that once the hypothesis is properly understood, some of the most intuitive objections to it miss their mark. For instance, the unity of the virtues strikes interlocutors such as Protagoras as quite obviously empirically false: but the problem vanishes once we see that Unity is meant to hold only for the vanishingly rare virtue which is knowledge. (It’s no surprise if the dispositions we commonly accept as virtuous are found separately—that’s part of what makes them inferior to the real thing.) Likewise with Teaching. The highest virtue must be taught, since it is a technê; but how the second-rate, piecemeal kind we typically encounter is to be acquired is another question altogether. (In the Meno, real learning turns out to consist in recollection: so perhaps ‘teaching’ and ‘learning’ need to be disambiguated into higher and lower varieties as well.) Teaching, Content, KGE, Circularity, Unity, Difficulty, Levels, and the Specifications are the consequences which run from one dialogue to another, discussed over and over from different angles, with different-looking but complementary results—the red threads which bind together the early dialogues, revealing them as a single project of inquiry. And their investigation appears, in a general way, to give results supportive of the hypothesis. All told, these consequences do harmonize, interlocking to support and clarify each other and to disarm potential objections. So far, it seems that under full examination, intellectualism turns out to be defensible. However, this is not the whole story. There are three further important theses which are standardly treated as part of ‘Socratic intellectualism’, but which I have so far set aside because they do not seem to be consequences of the intellectualist hypothesis itself. (My discussion of these will necessarily be brief and dogmatic; my aim is only to sketch how these endlessly controversial doctrines fit into the hypothetical reading.) These are: (i) the denial that akrasia or weakness of will is possible, which so scandalizes the Magna Moralia (No Akrasia, as I will call it); (ii) the notorious ‘Socratic paradox’ that no one does wrong voluntarily (Nemo, from the Latin tag nemo sua sponte peccat); and (iii) the thesis that all desire is for the good (DG). None is in any obvious or direct way a consequence of the Protagorean thesis that political virtue can be taught; but all are discussed by Socrates in the relevant contexts, as somehow bound up with it. To see the status and role of these theses on the hypothetical reading, we need to start by locating them in their dialectical context, and seeing how they are connected. No Akrasia is confined to the Protagoras, which diagnoses akratic experience as a kind of cognitive error, comparable to being taken in by an optical illusion.⁶⁰ The akratic—contrary to the self-interpretation of ‘the Many’—pursues

⁶⁰ On the massively complex and controversial latter part of the Protagoras, see Kamtekar 2017.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

     

47

what in the moment appears best to him, in what looks like a strengthened version of DG. (Strengthened in that simply to say that all desire is for the good is not yet to say that we only ever pursue what seems best to us, all things considered; and it is only the latter claim which rules out akrasia.) The belief which the akratic acts upon is a matter of temporary deception by ‘the power of appearance’—an intermittent false belief, of a kind which the right kind of expertise can block. So the Protagoras is in accord with the Meno: the special power of knowledge is that it renders impossible the kind of lapse into false belief which leads to deliberative mistakes. Nemo has traditionally has been taken as the heart of ‘Socratic intellectualism’. But though it is a recurrent Platonic commitment, it seems not to be theoretically fundamental, and its connection to intellectualism seems adventitious. In the early dialogues it is associated with intellectualism at Gorgias 509e, Protagoras 345e, and implicitly at Hippias Minor 376b; but Plato continues to hold it in late dialogues such as the Timaeus—on quite different grounds, having to do with the limits imposed by incarnation (86b–87b). Interestingly, Nemo in the early dialogues is not as radical as it sounds: at any rate Plato does not take it to entail that wrongdoers should receive forgiveness rather than punishment and blame. In the Timaeus, though, he does seem to infer that blame for the morally defective person is misguided (86d–e, 87b). So evidently Nemo is for him a thesis of enduring independent moral importance: but also one which needs to be derived from some more basic principle or other, with implications which vary according to what that grounding is.⁶¹ In the Gorgias, it seems that Nemo follows from the conjunction of two theses: DG and the echt-Socratic idea which helped to generate Circularity, that virtue is itself the greatest good for us. Together, these imply that unjust action, which confirms and increases injustice in the soul, can only frustrate our desire and so is possible only on condition of ignorance. The Hippias Minor leads us to the complementary suggestion that if justice is a technê (another instance of Specification), it must be the case that the person who has it will understand, as part of his art, that it is never in his interest to do wrong, and so will never be motivated to do it. Thus DG stands as a hypothesis in relation to both No Akrasia and Nemo. In addition to the Protagoras, it figures, by different argumentative routes, in the Meno (77a–78a), Gorgias (466a–468e) and Republic (505a–b). Its explanatory priority is somewhat obscured by the way in which, in the Protagoras, it is introduced as if inferred from No Akrasia at 358b6–c3. This order of appearance is, I would suggest, somewhat misleading: DG is not the result of the argument (which would be invalid if it were), but for Plato represents a kind of first principle [archê]. Here and elsewhere, it is supported not by deduction from anything more

⁶¹ Cf. Kamtekar 2017 for a full and insightful discussion.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

48

 

basic but by explanations of why apparent types of counter-example do not count: the person who seems to desire bad things in the Meno, the person who undergoes medical treatment in the Gorgias, and the akratic agent in the Protagoras. Hence my use of the term archê. According to Aristotle, a true first principle can only be supported dialectically, by disarming any alternatives and objections (cf. Arist. Metaph. IV.3–6); this is, it seems to me, roughly what Plato is doing in these much-debated passages. His purpose in doing so is to propose and defend DG as the strongest available higher hypothesis for intellectualism itself. In effect, he asks: on what condition would it follow that the possession of a deliberative art— the Protagorean technê—would be sufficient to guarantee success in life? It would follow only if our desire is already oriented to what is best for us, so that the only obstacles are recognizing what it consists in and how to obtain it. How exactly we should understand DG is an enduringly tricky question, which I cannot address properly here.⁶² So far as I can see, DG as presented in the Meno and Gorgias claims only that every human desire is prompted by the apprehension of its object as good or beneficial in some way for the possessor.⁶³ Thus it amounts to the claim that all human behaviour is ultimately motivated by self-interest. This thought (also found in Gorgias’ Palamedes and Thucydides passim) should probably be seen as a commonplace of fifth-century thought, essential then as now to explanation in the social sciences—compare the ‘economic man’ of the modern era. Only in the Protagoras do we find the more precise and less intuitive claim that the strength of our various desires tracks the comparative expected utility of their objects: that all our desires, whether based on knowledge or mere opinion, are (to put it in a few different ways) systematically regimented, reasonresponsive, all-things-considered, or good-maximizing. Here Plato seems to mix in with DG some independent claim about the structure of desires which, taken together with it, gives the result that we can never be motivated to pursue a lesser good over a larger one: we only ever act on a desire for what we consider best. This result seems to me worth distinguishing from DG proper: I have already alluded to it as strengthened DG. It is strengthened DG which grounds No Akrasia and Nemo; is it also strengthened DG which he takes to best ground the intellectualist hypothesis? Without it, there would at least be an important loophole in the Protagorean promise. For possession of the deliberative art could not guarantee success in life if the possessor remained vulnerable to desires which are unresponsive to reason, and excessive and irrational in relation to the value of their objects. This seems to be part of the point of Protagoras 352e, where Protagoras agrees that wisdom and knowledge are rulers in our soul, not to be overthrown by emotional ⁶² I do my best in Barney 2010. Note that the difference between DG and strengthened DG which I note here is orthogonal to the distinction between the object of desire understood as the apparent and as the real good. ⁶³ Cf. Gorgias (367c–368e), Meno (77c–78b); also Protagoras (358b–c), Republic (505a–b; for a banal application cf. e.g. 369c), and Gorgias Palamedes (13, 19).

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

     

49

pressures and temptations: tellingly, he says that it would be shameful for him ‘above all people’ to say otherwise. But whether he, Socrates, or their author has thought through what this stance presupposes about the structure of human motivation is another question. In sum, it seems to me that DG should be seen as a higher hypothesis relative to the intellectualist hypothesis, and No Akrasia and Nemo as a kind of collateral ‘branch line’ of consequences, reached via strengthened DG, by which it is to be tested. That all the consequences of a higher hypothesis should be explored, so as to test it in turn in the manner of hypothesis type (iii), is nowhere explicit in Plato’s discussions of the method, but seems a natural extension of it. These two are not themselves first principles, then, nor echt-Socratic paradoxes, but a second set of consequences—a bonus or penalty, as the case may be, attached not to sophistic intellectualism itself but to the best grounding Plato can supply for it. Plato is sufficiently content with Nemo to retain it even in his latest works; but No Akrasia is of course rejected in Republic IV. The verdict on the intellectualist package as a whole, then, appears to be mixed.

5. Conclusions So much by way of a sketch of the intellectualist hypothesis and Plato’s investigations of it in the early dialogues. I have passed over many difficulties and left any number of questions unanswered; but I hope what I have said is sufficient to show that the hypothetical reading doesn’t boil down to the unexciting claim that the early dialogues assert ‘virtue is knowledge’ in merely a provisional or tentative way. I would prefer to say that they do not assert it at all, and that this tells us something important about Platonic philosophical writing. In reading philosophical texts we habitually presuppose that the central operation of the text is doctrinal assertion. We assume, in the absence of explicit instruction to the contrary, that when a work of philosophy advances an argument for a thesis it is because the author endorses that thesis as true and that argument as sound, and that he puts them forward in order to urge us to join him in that endorsement. That package of assumptions is fair enough for most philosophical writing from Aristotle onwards; and it has for centuries made the Aristotelian reading of the early dialogues seem natural. But applied retroactively to Plato’s early works, as to the texts of the sophists who so deeply influenced him, it gives grotesque results. Here, no one argument gives more than one side of the story; in advancing arguments on both sides, Plato merely commends them to our philosophical attention and further investigation. And this is not merely because of a general spirit of provisionality, of the kind which we also find in (for instance) Cicero’s more skeptical works—a rote commitment to only weakened assertion, so as to leave room for future inquiry. It is because Plato’s texts are inquiry: philosophy being done in real time, not a report on its results.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

50

 

To see what those results might be, when the inquiry is at last complete, we can do no better than to turn to the Republic; but the evidence it presents is complex and ambiguous. The tripartite psychology presented there looks like Plato’s solution to at least some of the problems which the early dialogues have disclosed, and especially to the difficulties with strengthened DG. Tripartition explains how an agent may have desires which float free of his considered opinion as to what is best: it can be seen as designed not to reject DG but to rescue it, by showing that is indeed compatible with the phenomena if conjoined with a suitably complex and psychologically realistic account of the agent who believes and desires.⁶⁴ But Plato in the Republic is less interested in spelling out the results of his old inquiry than in diving into a new one, built around just the kind of substantive theorizing in moral psychology from which the early dialogues refrained. In fact, how we see his overall verdict on the hypothesis depends on how we resolve a question which the explicit argument of the Republic leaves open. If the knowledge of the Good possessed by the Guardians according to Books V–VII is just a greatly elaborated version of the euboulia attributed to them in Book IV, then it is distinct from the virtues of courage, sôphrosunê, and justice. So the art of living is at most one of the virtues; and the hypothesis as construed in the early dialogues, where it entailed Unity, has failed. (At the same time, we might see here a reversion to the original Protagorean form of the hypothesis, on which the politikê technê is by no means identical with all the virtues.) Alternatively, if we suppose that the wisdom which flows from the Guardians’ dialectical grasp of the Good replaces in them the ensemble of the Book IV virtues,⁶⁵ the Republic would in effect be reapplying the consequences Levels, Difficulty, and Content so as to distinguish the virtue of the Guardians from even the best lesser kind, as outlined in Book IV. Since that higher virtue would consist in a deliberative technê qualifying them to rule, it would be fair to read the Republic as in the end affirming the intellectualist hypothesis. I see no simple or decisive way to decide between these two readings of the Republic; that being so, Plato’s verdict on the early dialogues likewise seems to me to remain open. That the grand project of investigation presented in the early dialogues does not in the end give a simple and decisive result—that Plato’s answer in the end is apparently, ‘it depends what you mean by virtue’—makes it, it seems to me, not a failure but a model of philosophical inquiry.⁶⁶

⁶⁴ For recent readings bringing out this kind of continuity between ‘Socratic’ psychology and that of the Republic see Carone 2001, Moss 2008, Gerson 2014, Wilburn 2015, Kamtekar 2017, and Butler 2019. ⁶⁵ This would explain why the Auxiliaries have merely ‘civic courage’ (430c): only a courage constituted by knowledge, which the Guardians alone will possess, deserves to be called courage sans phrase. At the same time it is only the Auxiliaries who exercise courage on behalf of the city; so their courage is ‘civic’ in a positive sense as well. ⁶⁶ This first version of this paper was submitted for a graduate school ‘unit’ to a deeply unconvinced Alexander Nehamas in 1992. Since then ancestors of it have been presented at Washington University (St. Louis), Durham University, Oxford University, King’s College London, University of Michigan, Harvard University, the Universidade São Paulo, and other locales, in addition to the Keeling

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

     

51

References Adkins, A. W. H. 1960. Merit and Responsibility: A Study in Greek Values. Oxford: Clarendon Press. Allen, J. V. 2006. ‘Dialectic and Virtue in Plato’s Protagoras.’ In The Virtuous Life in Greek Ethics, ed. B. Reis, 6–31. Cambridge: Cambridge University Press. Bailey, D. T. J. 2005. ‘Logic and Music in Plato’s Phaedo.’ Phronesis 50: 95–115. Barney, R. 2010. ‘Plato on the Desire for the Good.’ In Desire, Practical Reason, and the Good, ed. S. Tenenbaum, 34–64, Oxford: Oxford University Press. Barney, R. 2021. ‘Technê as a Model for Virtue in Plato.’ In Productive Knowledge in Ancient Philosophy, ed. T. Johansen, 62–85. Cambridge: Cambridge University Press. Benson, H. 2003. ‘The Method of Hypothesis in the Meno.’ Proceedings of the Boston Area Colloquium in Ancient Philosophy 18: 95–126. Benson, H. 2015. Clitophon’s Challenge: Dialectic in Plato’s Meno, Phaedo, and Republic. Oxford: Oxford University Press. Brickhouse, T. and Smith, N. D. 2010. Socratic Moral Psychology. Cambridge: Cambridge University Press. Broadie, S. 2021. Plato’s Sun-Like Good: Dialectic in the Republic. Cambridge: Cambridge University Press. Butler, T. 2019. ‘Refining Motivational Intellectualism: Plato’s Protagoras and Phaedo’, Archiv für Geschichte der Philosophie 101(2): 153–76. Byrd, M. N. 2007. ‘Dialectic and Plato’s Method of Hypothesis.’ Apeiron 40 (2): 141–58. Carone, G. R. 2001. ‘Akrasia in the Republic: Does Plato Change His Mind?,’ Oxford Studies in Ancient Philosophy 20: 107–48. Cooper, J. M. (ed.). 1997. Plato: Complete Works. Indianapolis: Hackett. Costello, W. (ms) ‘The Method of Hypothesis in the Phaedo.’ (unpublished). Devereux, D. 1995. ‘Socrates’ Kantian Conception of Virtue.’ Journal of the History of Philosophy 33: 381–408. Ebrey, D. 2013. ‘A New Philosophical Tool in the Meno: 86e–87c.’ Ancient Philosophy 33 (1): 75–96. Ebrey, D. 2016. Review of H. Benson, 2015. Clitophon’s Challenge: Dialectic in Plato’s Meno, Phaedo, and Republic. Oxford: Oxford University Press: https://ndpr.nd.edu/ reviews/clitophons-challenge-dialectic-in-platos-meno-phaedo-and-republic/.

Colloquium at UCL; I am afraid that this almost geological timescale prevents me from thanking all the colleagues who have corrected errors and otherwise helped me with it, but I know that they include Hugh Benson, Tad Brennan, Eric Brown, Rachana Kamtekar, Fiona Leigh, Stephen Menn, Marco Zingano, and of course Alexander Nehamas. I am also grateful to Terence Irwin for his acute but generous comments; and to Merrick Anderson, Julia Annas, Taylor Barinka, David Ebrey, Mark Gatten, Rachana Kamtekar, and David Wolfsdorf for their very helpful comments on the final draft.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

52

 

Ebrey, D. (Forthcoming.) Plato’s Phaedo: Forms, Death, and the Philosophical Life. Cambridge: Cambridge University Press. Finkelberg, M. 1998. ‘Timê and Aretê in Homer.’ Classical Quarterly 48 (1): 14–28. Forster, M. 2006. ‘Socratic Refutation.’ Rhizai 3 (1): 7–57. Forster, M. 2007. ‘Socrates’ Profession of Ignorance’, Oxford Studies in Ancient Philosophy 32: 1–35. Franklin, L. 2010. ‘Investigation from Hypothesis in Plato’s Meno: An Unorthodox Reading.’ Apeiron 43 (4): 87–116. Gentzler, J. 1991. ‘συμφωνεῖν in Plato’s Phaedo.’ Phronesis 36: 265–77. Gerson, L. 2014. ‘The Myth of Plato’s Socratic Period’, Archiv für Geschichte der Philosophie 96 (4): 403–30. Ionescu, C. 2018. ‘Elenchus, Recollection, and the Method of Hypothesis in the Meno’, Plato Journal 17: 9–29. Irwin, T. H. 1977. Plato’s Moral Theory: The Early and Middle Dialogues. Oxford: Clarendon Press. Irwin, T. H. 1995. Plato’s Ethics. Oxford: Oxford University Press. Iwata, N. 2015. ‘Plato on Geometrical Hypothesis in the Meno.’ Apeiron 48: 1–19. Iwata, N. 2016. ‘Plato’s Hypothetical Inquiry in the Meno.’ British Journal for the History of Philosophy 24 (2): 194–214. Kahn, C. 1996. Plato and The Socratic Dialogue. Cambridge: Cambridge University Press. Kamtekar, R. 2017 Plato’s Moral Psychology, Oxford: Oxford University Press. Landry, E. 2012. ‘Recollection and the Mathematician’s Method in Plato’s Meno.’ Philosophia Mathematica 3 (20): 143–69. Lloyd, G. E. R. 1992. ‘The Meno and the Mysteries of Mathematics.’ Phronesis 37: 166–83. Menn, S. 2002. ‘Plato and the Method of Analysis.’ Phronesis 47: 193–223. Moss, J. 2008. ‘Appearances and Calculations: Plato’s Division of the Soul.’ Oxford Studies in Ancient Philosophy 34: 36–68. Palmer, J. 2021. The Method of Hypothesis and the Nature of Soul in Plato’s Phaedo. Cambridge: Cambridge University Press. Robinson, R. 1953. Plato’s Earlier Dialectic, 2nd edn. Oxford: Oxford University Press. Scolnicov, S. 2018. Plato’s Method of Hypothesis in the Middle Dialogues. Edited by Harold Tarrant. Baden-Baden: Academia Verlag. Scott, D. 2006. Plato’s Meno. Cambridge: Cambridge University Press. Scott, G. 2002. Does Socrates Have a Method?: Rethinking the Elenchus in Plato’s Dialogues and Beyond. University Park: Pennsylvania State University Press. Sedley, D. N. 2021. ‘Socrates’ “Second Voyage” (Plato, Phaedo 99D–102A).’ In Themes in Plato, Aristotle, and Hellenistic Philosophy: Keeling Lectures 2011–18, ed. F. Leigh, 47–62. London: Institute of Classical Studies.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

     

53

Segvic, H. 2008. ‘Protagoras’ Political Art.’ In From Protagoras to Aristotle, ed. M. F. Burnyeat. Princeton: Princeton University Press. Simpson, P. L. P. (trans., comm.) 2017. The Great Ethics of Aristotle. London: Routledge. Tredennick, H. and Armstrong, G. C. 1935. (trans., notes) Aristotle: Metaphysics Books X–XIV, Oeconomica and Magna Moralia. Cambridge, MA: Harvard University Press. Weiss, R. 2001. Virtue in the Cave: Moral Inquiry in Plato’s Meno. Oxford: Oxford University Press. Wilburn, J. 2015. ‘Courage and the Spirited part of the Soul in Plato’s Republic’, Philosopher’s Imprint 15.26: 1–21. Wohl, V. (ed.) 2014. Probabilities, Hypotheticals, and Counterfactuals in Ancient Greek Thought. Cambridge: Cambridge University Press. Wolfsdorf, D. 2008a. Trials of Reason: Plato and the Crafting of Philosophy. Oxford: Oxford University Press. Wolfsdorf, D. 2008b. ‘The Method ἐξ ὑποθέσεως at Meno 86e1–87d8.’ Phronesis 53 (1): 35–64. Wolfsdorf, D. 2013. ‘Socratic Philosophizing.’ In The Bloomsbury Companion to Socrates, ed. N. D. Smith and J. Bussanich, 34–67. London: Bloomsbury.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

3 Comments on Rachel Barney, ‘Intellectualism and the Method of Hypothesis in Plato’s Early Dialogues’ Terence Irwin

1. Introduction In the decade that has passed since Rachel Barney presented a version of her paper at a Keeling Colloquim, she has worked out a detailed, careful, and thoughtprovoking account of how the set of claims that she calls Socratic Intellectualism should be understood within the hypothetical method (or methods) that is developed, or at least sketched, in the Meno and the Phaedo. The comments I offered on the decade-old paper would need to be extensively revised and expanded to deal adequately with the case that she develops in impressive detail in the present paper. I will confine myself, however, to raising some of the questions that I originally raised, with revisions that I hope will take account of Barney’s present position. Barney tries to overcome the bad effects of reading Plato’s early dialogues under the shadow of Aristotle’s comments on Socrates. In particular, she believes that Aristotle’s comments mislead us into treating Socratic Intellectualism as Plato’s doctrine. I will begin by discussing Aristotle’s comments. Then I will ask whether Socrates maintains Socratic Intellectualism as his doctrine. I will compare my answer to this question with Barney’s view that Plato treats it as a hypothesis.

2. Aristotle and Socratic Intellectualism In Barney’s view, Aristotelian Socratic Intellectualism (i.e. Socratic Intellectualism as expounded by Aristotle), is not only false, but so boringly false that we ought to doubt whether Socrates could have seriously affirmed it.¹ She says that in Republic IV Plato refutes it by appeal to utterly banal and obvious observations, and she ¹ See Barney n.16, commenting on my previous comment. Terence Irwin, Comments on Rachel Barney, ‘Intellectualism and the Method of Hypothesis in Plato’s Early Dialogues’ In: Psychology and Value in Plato, Aristotle, and Hellenistic Philosophy: The Ninth Keeling Colloquium in Ancient Philosophy. Edited by: Fiona Leigh and Margaret Hampson, Oxford University Press. © Oxford University Press 2022. DOI: 10.1093/oso/9780192858108.003.0003

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

   

55

asks whether Plato in the early dialogues could really have meant to ‘deny the reality’ of obvious phenomena about weak will and conflicting desires. She assumes, therefore, that if our interpretation ascribes an obvious and boring falsehood to a major philosopher as one of his central doctrines, we should reconsider our interpretation. Aristotle, therefore, either read the dialogues carelessly or distorted Socrates for polemical purposes. In either case, he got Socrates wrong. Barney implies that if Aristotelian Socratic Intellectualism were philosophically defensible, she would be more inclined to attribute it to the Platonic Socrates. Is it as bad, then, as she thinks it is? I will approach this question by asking how bad Aristotle thinks it is. Aristotle certainly believes Socrates was wrong to identify the virtues of character with states of the rational part of the soul. The ground for the identification was that it is no good if one does brave and just actions, without knowledge and rational election (Magna Moralia 1198a10–12).² This ground is, in Aristotle’s view, a correct observation about virtue of character. Socrates was not wrong in this observation, but in his inference that therefore virtue is simply reason. Contemporary thinkers are more nearly right when they say that virtue is doing fine actions in accordance with correct reason (1198a12–15). But they are not completely right, because simply acting in accordance with correct reason is not sufficient for being virtuous; for one might act on a non-rational impulse in accordance with correct reason, but without election or knowledge of fine things, and this would not constitute virtue. Virtue requires us to do fine things with reason (1198a15–21). This passage implies, then, that Socrates is wrong on something that contemporaries get right (because virtue is not only reason), but also right on something that contemporaries get wrong (because, as he saw, virtue requires the active participation of reason, and not only conformity to reason). Aristotle asserts that Socrates’ denial of incontinence is incorrect (1200b30), and that it is absurd (atopon) to deny what ‘plausibly comes about’ (pithanôs ginomenon), that people are incontinent. When he says it is plausible that people are incontinent, we might think he means it is obvious that we can choose the option that we believe to be worse, at the very moment when we believe it to be worse. We might take Plato in Republic IV to agree with this description of incontinence. But Aristotle does not agree with it in MM. He believes that when we act incontinently, we have only potential, but not actual, knowledge that what we are doing is worse (1201b24–1202a1). He entirely agrees with Socrates that we cannot actually know that what we are doing is the worse option. Though Socrates is wrong to ignore the role of non-rational desire in depriving us of actual knowledge, he is right to believe that we are deprived of actual knowledge when we choose the option that we previously knew to be worse.

² I take MM to be a genuine work of Aristotle that is earlier than EE and EN.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

56

 

In this case as in the case of virtue and reason, it would not be false to say that Aristotle believes Socrates is quite wrong, but it would be one-sided. Aristotle also believes that Socrates is right on a point that many people get wrong. One might object that Aristotle does not seem to acknowledge that Socrates is right. Even if we think he ought to agree that Socrates is right about some things, is his judgement not negative? If we are inclined to say this about MM, we should look at the corresponding passages in Nicomachean Ethics VI–VII.³ In VI.13 Aristotle says Socrates was right on one point and wrong on one point (1144b18–20), and then presents the same treatment of Socrates and contemporary philosophers as in MM. In EN VII.2 he says, as he said in MM, that Socrates was wrong to reject the possibility of incontinence, but he now adds that if ignorance is the cause, we need to ask what kind of ignorance it is (1145b27–31). That is the question that Aristotle asks and answers. In his answer he says that one of Socrates’ claims turns out to be right; for, just as Socrates thought, we cannot act against full knowledge, and this knowledge is not dragged about by non-rational desire (1147b13–17). The details of Aristotle’s account are open to dispute (both about what he means, and about whether it is at all plausible), but all we need to notice for present purposes is that he affirms that on one point Socrates was right. As in MM, Aristotle does not say that Socrates was wrong to deny that we can choose what we actually know to be the worse of two options, in the moment when we actually know it. It is no doubt difficult to say what exactly Aristotle means when he says he is agreeing with Socrates. But, in whatever respect he agrees, he affirms that Socrates has come close to the truth about some aspects of incontinence, just as he has come close to the truth about some aspects of virtue. If Socrates is right on some important and controversial questions, Aristotelian Socratic Intellectualism is not a position that only an uncharitable reader would attribute to Socrates. Aristotle might fairly reply that he is paying Socrates a compliment in attributing this position to him. It is not only Aristotle who takes Aristotelian Socratic Intellectualism seriously. The Stoics seem to have agreed with him. The doctrine that they attribute to Socrates is no less intellectualist than the one that Aristotle attributes to him. Moreover, they attribute it to Socrates, not to Plato. We do not find any Stoic sources that ascribe the same intellectualist position to Plato, or that represent Plato as simply taking the Socratic position as a hypothesis to be examined. The Stoics go further than Aristotle, since they believe that Aristotelian Socratic Intellectualism as a whole, and not only the part that Aristotle accepts, is true and defensible.⁴ Admittedly, the Stoic defence may make Socratic Intellectualism ³ These passages may well correspond to passages in EE V–VI. ⁴ See e.g. A. A. Long, ‘Socrates in Hellenistic Philosophy’, in Stoic Studies, 33. After citing one of the passages from MM cited by Barney, he says ‘It would be worthwhile to consider how the orthodox Stoics welcomed this interpretation of Socrates, and fully developed its implications in their own account of the mind’s faculties and moral states.’

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

   

57

appear more credible than it appears when Socrates states it. But the fact that a defence can be offered suggests that we should not be surprised if we find that Socrates maintains it as doctrine. But are Aristotle and the Stoics perhaps wrong to take Socratic Intellectualism seriously? To show that it is worth considering, we should notice that Socrates does not simply deny the reality of incontinence without any argument. He also tries to explain why other people do not agree with him. He does this in his discussion of time and measurement in the Protagoras (356c–357b). Temporal proximity makes goods look bigger than they turn out to be when we examine their goodness from a temporally impartial point of view. Even though we recognize that the nearer good is quite small, and the consequential evil is much greater, it remains temporally close, and we remain subject to the appearance that it is big enough to be worth choosing. This fact helps Socrates’ case. We might object to him that the incontinent already recognizes that the nearer good is not good enough to choose. He answers that, even if we recognize that the nearer good is too small to choose, we are liable to vacillate and to oscillate in our beliefs. If we choose the lesser good on one of the occasions when we revert to the belief that it is good enough to choose, we can choose the lesser good quite consistently with Socratic Intellectualism. The many suppose that we choose contrary to our belief about what is better because they do not recognize how much beliefs oscillate. Socratic Intellectualism cannot be refuted, then, simply by an appeal to the sorts of appearances that the many point to in the Protagoras. Moreover, the appeal to oscillation of belief is perfectly Socratic. Socrates is aware of the fact that people who sincerely assent to the conclusions of his arguments are none the less liable to backslide when they encounter something that vividly recalls their former belief. That is what he says to Callicles (Gorgias 513c4–d1). In the Crito he recognizes that present danger may cause people to abandon their previous convictions (49a4–b6). He thinks this is a childish error (49b1), but he sees that Crito is about to commit it. It is easy to see how we might mistake this inconstancy for acting against one’s present conviction about what is best. For these reasons, Socrates, Aristotle, and the Stoics have no reason to be embarrassed by their support of Socratic Intellectualism.

3. What Is Socrates’ Doctrine? To decide whether Barney is right to say that the Platonic Socrates never defends Socratic Intellectualism ‘as doctrine’, we need to say what it would mean for Socrates to defend something ‘as doctrine’. We have noticed that the Crito mentions the danger of abandoning one’s convictions. Socrates says it would be childish to abandon a position that has been reached by repeated agreement

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

58

 

(49a4–7). He says he is convinced by the argument that seems best to him when he reasons (46b3–c6). He does not claim to have shown that his conclusions are immune to refutation, and in this respect we might say that he holds them provisionally. But having found that they are not refuted, he sticks to them and is ready to act on them. We find the same position in the Gorgias (508e–509a). Socrates says he has defended and tied down his thesis that it is worse to do injustice than to suffer it by iron and adamantine arguments. Unless an opponent can loosen his arguments, it is impossible to speak well by saying anything other than he is saying. He does not know how these things are, but he has never met anyone who disagreed with him without being ridiculous. He maintains some claims with conviction on the basis of arguments that seem to him extremely cogent. While he disavows knowledge, and he is ready for further discussion, he believes that the results of past discussion support his thesis. If he treats a thesis in a critical and inquiring spirit, and he is ready to contemplate further objections that need to be considered, that is no reason to deny that he defends the thesis ‘as doctrine’. The dogmatic and the critical sides of Socrates are not only consistent but inseparable. Nor should we doubt that Socrates maintains something as doctrine merely on the ground that he elicits it from the beliefs of his interlocutors, with the help of other assumptions that he takes to be plausible. This procedure is a feature of everything that he defends by the use of the elenchus, including the thesis he claims to have proved in the Gorgias. According to Barney: Were it not for Aristotle, and certain modern assumptions about what philosophical writing must do, we would find it natural to speak of ‘Virtue is knowledge’ not as a theory being affirmed, but as a topic of debate—a fascinating proposal subjected by Plato to repeated and apparently inconclusive discussion.

We have seen, however, that Socrates puts forward some conditions under which he takes a conclusion of his arguments to be true. If the intellectualist thesis meets these conditions, and he affirms it as doctrine, we should not be surprised.

4. Is Intellectualism Socrates’ Doctrine? I have now tried to give an idea of how we might recognize a defence of intellectualism as doctrine. Does Socrates, then, actually defend it as doctrine? A proper answer to this question would require discussion of some familiar disputes. I will simply sketch a possible answer. In the Protagoras Socrates alleges that believers in the possibility of being overcome by pleasures against one’s better judgement hold a ridiculous view (355a6, geloion ton logon gignesthai, b3–4 hôs de tauta geloia esti). This allegation is never contested in the rest of the dialogue. This

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

   

59

is one sign that Socrates’ denial of incontinence has one of the features mentioned in the Gorgias. If, then, our attribution of doctrines to Socrates is guided by his standards, we have a reason to say that he takes it to be true that it is impossible to be overcome by pleasures against one’s better judgement. Barney objects that the sequel in the Protagoras undermines the argument I have just given, because Socrates says the argument itself would ridicule each of them (katagelan, 361a5) for their absurdity (atopoi g’este, a5–6) in changing sides. Does this remark not show that Socrates does not after all affirm as doctrine the thesis that I said he affirms? To assess this objection, we need to ask whether Socrates has changed sides for good or bad reasons. He initially said that virtue could not be taught, but now he says it can be taught. The support for his second view is a thesis that it would be ridiculous to deny. The support for his first view was the assumption that Athenian politicians had virtue and were trying to pass it on to their sons. This assumption is not only not ridiculous to deny, but extremely doubtful in the light of what they have discovered about virtue. Socrates has therefore shown that he has a good reason to change sides. The absurdity in his behaviour does not consist in embracing his later view, but in having embraced his earlier view for such bad reasons, and this is why the personified argument ridicules him. Socrates does not say he has reached an aporia,⁵ or that he has no idea what to say on the questions they have discussed. He says that since they have turned things upside down (361c2–3), he is eager to examine the questions again. As we have already seen in the Crito and the Gorgias, willingness to re-examine a conclusion does not imply that Socrates does not take the conclusion to be true, or that he thinks he has not given convincing arguments for it. He is willing to examine the question again, to see whether someone who denies his conclusion can avoid being ridiculous. The Protagoras, therefore, presents Socrates affirming as doctrine the intellectualist thesis that he says it would be ridiculous to deny.

5. Two Types of Hypothetical Argument Let us now consider Barney’s alternative to the view that Socrates affirms Socratic Intellectualism as doctrine. She argues that Plato treats Socratic Intellectualism as a hypothesis. She explains what it is to treat something as a hypothesis by reference to the discussions of hypotheses in the Meno and Phaedo. She intends her explanation of Socratic Intellectualism as hypothesis to provide a clear alternative to the view that it is asserted as doctrine:

⁵ Here I am questioning Barney’s claim that ‘Socrates’ own conclusion in the Protagoras is thoroughly aporetic’ (she then cites 361b–c).

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

60

  the hypothetical reading doesn’t boil down to the unexciting claim that the early dialogues assert ‘Virtue is knowledge’ in merely a provisional or tentative way. I would say that they do not assert it at all, and that this tells us something important about what kind of writing they are.

I will explore some elements of the hypothetical reading, to see whether it excludes doctrinal assertions. Two types of Platonic hypothesis are involved: 1. We can use a hypothetical method to discuss a question, such as ‘Is virtue teachable?’ Socrates assumes that virtue is beneficial, and argues that if it is beneficial, it is knowledge. But if it is knowledge, it is teachable. Hence it is teachable. But if it is teachable, there are teachers of it. But in fact there are no teachers of it. Therefore it is not teachable. Therefore it is not knowledge. 2. We formulate a hypothesis by finding the account that seems to us to be strongest, and then we affirm its consequences as true. If necessary, we defend the hypothesis by finding a higher hypothesis. The first type of argument from hypothesis is found in the Meno, and also in the Phaedo (e.g. as Barney mentions, in the discussion of whether the soul is an attunement of bodily parts). The second type is found at Phaedo 100. The first type of argument does not by itself prove any of the propositions that we use to draw the various conclusions. At most it proves that we cannot both hold our initial assumption and all the intermediate propositions that we use to derive the contradictory of our initial assumption. It does not tell us what we should give up. This limitation in argument from hypothesis is relevant to the conclusions that might be drawn from the Meno. The objection to the claim that virtue is knowledge rests on the feeble support of the assumption that if something is teachable there are teachers of it. The dialogue shows how feeble this assumption is; how could people who do not know what virtue is, or what the relevant sort of teaching is, ever teach it? The second type of hypothetical argument is quite different. If we set out from the account that we take to be strongest,⁶ and we affirm its consequences as true,⁷ the argument gives us some ground for affirming things as true, once we have derived

⁶ Barney suggests that in speaking of the strongest hypothesis Plato is being deliberately vague. If we consider the reference to arguments of iron and adamant in the Gorgias, we need not suppose he is being all that vague; he simply alludes to Socrates’ procedure for showing that some thesis is strongest—i.e. best defended. For this reason I doubt whether Barney’s example—the thesis that the soul is an attunement—is properly described as this sort of hypothesis. Plato says nothing to show that he takes it to be the strongest. Nor does he affirm its consequences as true. ⁷ Barney’s description of this feature of hypothesis in the Phaedo (100a5) seems to me to underestimate the significance of affirming as true.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

   

61

from the relevant sort of account. We then have some—non-conclusive—reason to reject as false anything that conflicts with the consequences of the hypothesis. The reason is not conclusive, because we might be wrong to affirm our initial hypothesis. If a question arises about the initial hypothesis, we need to defend it by appeal to a higher hypothesis, until we reach something adequate (Phd. 101d3–e3).⁸ Though I have omitted many details and qualifications, I hope that this sketch of the two types of hypothetical arguments will make my next question intelligible. What should we make of Barney’s claim that Plato does not affirm Socratic Intellectualism as doctrine, but treats it as a hypothesis? The answer depends on which of the two types she has in mind. If Socratic Intellectualism appears only in the first type of hypothetical argument, Plato does not affirm it as doctrine. But if it appears in the second type of hypothetical argument instead of (or as well as) the first, he affirms it as doctrine. I will try to show that Barney does not confine herself to the suggestion that Plato treats Socratic Intellectualism as a hypothesis of the first type. She also implies that he treats it as a hypothesis of the second type. For this reason, Barney’s own discussion implies that Socrates affirms Socratic Intellectualism as doctrine.

6. The Role of the Intellectualist Hypothesis Sometimes Barney takes Socratic Intellectualism to be a hypothesis of the first type. Its central thesis, as she understands it,⁹ is that ‘there is a politikê technê: an art of deliberation or decision-making sufficient for expert political rule and for success in private life as well’. As she points out, it is easy to interpret this thesis so that it does not imply Aristotelian Socratic Intellectualism, and Protagoras himself does not seem to interpret his political craft in Aristotelian Socratic Intellectualism terms: It does not immediately imply that all the virtues are the same body of knowledge. In its original, Protagorean form it is not even a theory about all the virtues: at any rate in Plato’s presentation of him, Protagoras is emphatic that courage is wholly distinct from the kind of social or political virtue he teaches, so that Socrates’ hard-fought identification of wisdom with courage amounts, for him, to the deduction of an unwelcome consequence (Prot. 329e, 349d, 350e–351a). Without further substantial argumentative moves, it does not imply any particular conception of human motivation: it certainly does not entail that the emotions are motivationally inert, or that they can be reduced to intellectual judgements.

⁸ Whatever exactly Plato means by ‘higher’ and ‘adequate’ here, they are needed if we are to give an appropriate account of the initial hypothesis to someone who attacks it (echoito, 101d3). ⁹ This is to be distinguished from the ‘Aristotelian Socratic Intellectualism’ that I have discussed so far. Barney distinguishes the two versions of intellectualism in her first few paragraphs.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

62

 

It follows that the intellectualist thesis that Socrates discusses is not Protagoras’ version of the political craft, but a Platonic version: So the intellectualist hypothesis as investigated by Plato, which applies to virtue in toto, is not identical with the original Protagorean version, at least as Plato himself (plausibly) represents it: we might think of it as ‘intellectualism 2.0’. It still seems to me right to speak of the hypothesis as sophistic, and very misleading to speak of it as Socratic (the term ‘appropriation’ would also I think be misleading), since (a) as I argue, that is how Plato himself represents it and (b) the alterations made to it need not be read as expressing an affirmation of it, with a transformation for his own purposes. Rather they seem to be intended as, so to speak, internal improvements, designed to clean the hypothesis up and present it in the most viable form available. How could the virtue acquired by teaching reliably ensure good deliberation everywhere, even on the battlefield; how could it demonstrate the sovereignty of knowledge over emotions and external stresses; how could it make a person steadily better and guarantee their success in life; how, in short, could it possibly have all the powers which Protagoras himself claims for it if it does not include courage?

We might think that Protagoras has an answer to this question, if he appeals to 328ab, at the end of his Great Speech. He does not claim that his technȇ is effective without suitable preparation. In his view, everyone teaches virtue, and he only claims to take his pupils a bit further than the point they have reached through civic moral education. He assumes that they have already learned to be brave, just, and temperate, but he makes them a bit wiser. He does not need to show that the virtue acquired by his teaching includes bravery. Even if it is difficult to defend Socrates’ replacement of the Protagorean position by ‘intellectualism 2.0’, Plato may have intended the argument that Barney attributes to him. But the apparent weakness in this argument may lead us to ask whether this is what Socrates is trying to do at all. One might suspect that he holds that Protagoras’ conception of virtue (as the product of the sort of civic moral education that was described) is deeply mistaken. This is at least a possible conclusion from the rest of the dialogue. So far I have only considered Barney’s explanation of why Socrates brings bravery into the discussion. She also finds a place for the more overtly intellectualist theses, and especially for the thesis she calls DG+: that the strength of those desires tracks the comparative goodness of their objects: that all our desires are (to put it in a few different ways) systematically regimented, or reason-responsive, or good-maximizing.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

   

63

DG+ provides strong support for the intellectualist position that Aristotle attributes to Socrates: That is, Plato here adds . . . some independent claim about the structure of desires which, taken together with it, gives the result that we can never be motivated to pursue a lesser good over a larger one.

Since the previous defences of virtue as a craft have not required any such claim about reason and motivation, we might wonder why DG+ is needed for the task that, in Barney’s view, Plato has undertaken. In her answer to this question she suggests that DG+ is intended to support the other parts of the intellectualist hypothesis, not because it is a consequence of them, but because it provides the strongest defence of them: Is it also DG+ which, on his understanding, best grounds the intellectualist hypothesis? Without it, there would at least be an important loophole in the Protagorean promise: possession of the deliberative art cannot guarantee success in life if the possessor remains vulnerable to desires which are unresponsive to reason, excessive in relation to the value of their objects, and sufficient to motivate bad choices.

As before, I am doubtful about the suggestion that this account of motivation fills some gap in Protagoras’ position, as he understands it. An easy reply would be that ‘desires that are unresponsive to reason . . .’ (etc.) have been eliminated by appropriate civic moral education. For the moment, however, I want to concentrate on Barney’s claim that DG+ best grounds the intellectualist hypothesis, because (as she says) it is it is the strongest higher hypothesis. Here she introduces the second type of hypothetical argument. In ‘best grounds’, she does not simply mean that it provides a sufficient condition for the theses about virtue as a craft. Being a sufficient condition does not necessarily make a hypothesis strong. Various sufficient conditions might be suggested for the theses about virtue as a craft, and these theses themselves might be variously interpreted. DG+ provides the strongest higher hypothesis only if it is true and the others are false. Apparently, then, Barney agrees that Socrates takes DG+ to be true, just as Plato takes his hypothesis about the Forms in the Phaedo to be true. What, then, has become of the claim that Socrates does not maintain Aristotelian Socratic Intellectualism as doctrine? Might one object that DG+ falls short of Aristotelian Socratic Intellectualism? If it is true, incontinence is impossible, and virtue does not require the training of non-rational desires in addition to knowledge of what is best. These are the views that Aristotle attributes to Socrates. If Socrates accepts these consequences

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

64

 

of DG+, he holds the views that Aristotle attributes to him. Moreover, he holds the views that Barney took to be so implausible that a charitable interpreter would not attribute them to Socrates. Might one, then, suggest that though Plato puts forward DG+, it is not his settled view? Barney may suggest this: ‘But whether he [sc. Protagoras], Socrates, or their author has thought through what this stance presupposes about the organization of our desires is another question.’ I am a little puzzled by the claim that Plato has not ‘thought through’ the implications of DG+ for the organization of our desires. One obvious implication is the impossibility of incontinence. If incontinence is possible, it provides a counter-instance to DG+. Hence we might expect someone who believed DG+ to try to show that incontinence is impossible. Socrates tries to do exactly this in the Protagoras. Plato, therefore, seems to have ‘thought through’ some implications of DG+. We might reasonably, therefore, take Plato to regard Socratic Intellectualism as a hypothesis, provided that we understand it as a hypothesis of the second type. But to regard Socratic Intellectualism as a hypothesis of the second type is not only compatible with regarding it as doctrine, but requires him to regard as doctrine. Socratic doctrines, as Socrates describes them in the Gorgias, may readily be expressed as hypotheses; Socrates takes them to be most strongly defended, as a result of previous examinations, and is ready to consider their consequences, in the belief that anyone who rejects these consequences will turn out to be ridiculous. If Socrates maintains Socratic Intellectualism as this sort of hypothesis, Aristotle’s view that it is Socrates’ belief is vindicated.

7. Socratic Virtue It may be worth adding a comment on a different, though not unconnected question. Barney asks about the intended content of virtue in the claim that virtue is knowledge. Her remarks on Homer (towards the end of her Section 2) raise this question. Some aspects of the virtue (aretȇ) displayed by Achilles are functional, in so far as they are relevant to his military role. But the Iliad is interesting partly because this aretȇ belongs to Agamemnon, Achilles, and Hector, who all conspicuously fail in their social roles. Since there is no sign that we should revise our view about their aretȇ as a result of these failures, we might infer that aretȇ is somewhat independent of social role. But this is not the whole story. The Odyssey may include the non-martial aspects of Odysseus in his aretȇ. Penelope speaks of him as displaying ‘every kind of aretȇ’ (pantoiȇn aretȇn, Od. 18.205) and Telemachus calls him ‘noble’ (esthlos) in the passage in which he says that Odysseus was as gentle as a father to the people whom he ruled (2.46–7). None the less Achilles is better than Odysseus, even though Odysseus has more virtues than Achilles. This comparative judgement is the starting point of the

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

   

65

Hippias Minor (363b1–4). As we know from a comparison between the Ajax and the Philoctetes of Sophocles, sharply opposed judgements on Odysseus can be defended. The different aspects of Achilles, Ajax, and Odysseus suggest that Barney’s contrast between the functional and the moralistic elements of virtue may need to be qualified. The moralistic elements are surely not functionally irrelevant. Odysseus asks whether the inhabitants of the place he has arrived at are wild and not just (agrioi oude dikaioi, Od. 9.175) Justice, hospitality, and fear of the gods make a collection of individuals into a society. The Cyclopes are wild and not just; they live in isolated families and have no concern for one another (Od. 9.114–15). More examples can readily be found in the Odyssey. But an excessive emphasis on the social virtues may lead us to agree with Odysseus’ view that the greater social good requires flexibility; against these demands of society, the selfcentred (as it may appear) attitude of Neoptolemus in the Philoctetes may seem an appropriate response. I doubt, then, whether the contrast between functional and moralistic elements is exactly the contrast that is most relevant to Protagoras and Socrates. Odysseus might claim that he satisfies both the demands of society and the demands of justice, because they are the same; the just individual is the good citizen. None the less we have misgivings about his argument. Protagoras may intend to signal these misgivings when he takes not only justice but also shame to be necessary for a city (322c1–d5). But he does not say what they require. Shame moves Neoptolemus to reject Odysseus’ apparently sensible suggestions about Philoctetes, but it is not completely clear why. Socrates is preoccupied with conflicts similar to those in the Philoctetes. He rejects the suggestion that a good citizen should see the need to try the generals quickly and collectively after Arginusae. He denies that a good citizen should support strong and stable government when he refuses to take part in the arrest of Leon. In these cases it is not enough to say that he prefers justice over self-seeking; for the demands of justice might themselves be taken to point in different directions. The familiar reflexions by Thucydides on the Corcyrean civil war (III 82) make the same point. Hence the contrast between the Homeric and the Hesiodic, or the self-seeking and the moral, does not capture Socrates’ main concerns. The dispute between Callicles and Socrates is partly about the opposition between self-seeking and morality. But it is also about what morality requires, and in particular whether Socrates’ outlook makes him a bad and useless citizen, as Callicles alleges. Socrates not only wants to reconcile justice and happiness, but also to say what is required by the justice that is sufficient for happiness. What have these questions to do with Socratic Intellectualism? Socrates believes that if you believe that something is good for you overall and you retain that belief, you will choose that thing. People do not realize that it is better for them to

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

66

 

cultivate the virtues, as he understands them, than to do anything else. They suppose correctly that sometimes we have to violate what we think we owe to individuals in order to keep (as he says) walls, dockyards, and all that other rubbish (Gorgias 519a). But they suppose incorrectly that the things we gain by (e.g.) condemning innocent people are means to happiness. Once they recognize the true means to happiness, they will lose their false beliefs about what a good citizen has to do, and will see that the choices Socrates makes are both just and beneficial.

References Long, A. A. 1988. ‘Socrates in Hellenistic Philosophy.’ Classical Quarterly 38, reprinted in his Stoic Studies. Cambridge: Cambridge University Press, 1996.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

4 Against Bare Urges and Good-Independent Desires Appetites in Republic IV Jessica Moss

1. Bare Urges? Thirst itself will never be an appetite for anything other than what it is for by nature, drink itself. (Republic IV 437e4–5)¹ It is widely held that with this line Plato commits himself to the view that an important class of human desires, our appetites (epithumiai) for food, drink, sex, and the like, have a peculiar property: they are always directed toward totally unqualified objects like drink, and never toward qualified objects like cold drink, sweet drink, much drink, or good drink. They are “bare cravings,” “blind drives,” or “simple desires.”² As these descriptions suggest, our appetites are supposed to be limited to such generic objects because they belong to a part of the soul that is too cognitively simple to grasp the nuances that distinguish, say, cold beer from hot milk. Furthermore, it is this extreme simplicity or blindness that Plato has in mind when he calls this part of the soul and its desires “non-rational.” Is this the right reading? We should hope not, for it makes a mess of Plato’s psychology. First, as is widely noted, Plato in many other places refers to appetites for complex things like looking at corpses (439e) or having sex with one’s mother (571d). Some thus take Plato here to be relying on a simplified conception of appetite inconsistent with later passages, perhaps because the simplified conception helps him in his present purpose: distinguishing the appetitive from the

¹ Translations throughout are mine unless otherwise noted. ² Cornford 1941: 130 (“bare craving”), Penner 1990: 52, 59 (“blind desires,” “blind drives”), Anagnostopoulos 2006: 173 (“simple desires”). Annas (1981: 129) agrees that this is the implication of the argument (“simple cravings”), but takes this to conflict with Plato’s considered treatment of the appetitive part. Jessica Moss, Against Bare Urges and Good-Independent Desires: Appetites in Republic IV In: Psychology and Value in Plato, Aristotle, and Hellenistic Philosophy: The Ninth Keeling Colloquium in Ancient Philosophy. Edited by: Fiona Leigh and Margaret Hampson, Oxford University Press. © Oxford University Press 2022. DOI: 10.1093/oso/9780192858108.003.0004

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

68

 

rational part of the soul.³ But the inconsistency is starker than that: even the lines directly preceding the claim about thirst-itself acknowledge the existence of complex appetites: If some heat is present in addition to thirst, it would further render the appetite one for cold drink, or if some coldness [is present, it would render it an appetite] for hot drink; and if on account of the presence of muchness the thirst were great, it would render the appetite an appetite for much drink. (437d10–e3)

Socrates does not say that the desire for cold drink is something other than an appetite, nor a combination of two appetites, nor an appetite influenced by reason; he simply says that heat turns the appetite for “drink itself ” (439a6) into the appetite for cold drink. Some appetites are simple, but others are complex. Second, this is surely the right thing for him to say. The appetite for unqualified drink is by no means a typical appetite: thirst is usually for cool drink, or sweet drink, or hoppy drink, and so on. It is a very odd notion of appetite that would exclude these cases. In fact, we should go further. The bare urge conception of the appetitive part not only fits poorly with Plato’s characterization of appetites elsewhere and with our own understanding of appetites, but is nonsense. Why should a part of the soul so limited as to be unable to make distinctions between hot and cold have the power to make distinctions between drink and non-drink, and thereby to feel thirst rather than general emptiness? Defenders of the bare urge view in fact accept that it renders the appetitive part motivationally helpless (see Penner 1990: 58–61, and especially Anagnostopoulos 2006): their Plato has to hold that this part can only move us with help from reason. But this is a very steep price to pay. What about the clear implication of the political analogy that each part is independent from the others and can motivate action on its own? And how are we to explain the motivationally effective appetites of dogs and babies, who lack reason altogether? I want to show that we can set these worries aside, for the text does not support the bare urge view at all. In fact, the claim about thirst-itself says nothing that is meant to distinguish appetites from other kinds of desire, nor the appetitive part of the soul from other parts. Plato wanted to show that there can be unqualified desires for unqualified objects, but he could just as well have taken rational desires as his examples. The key to the correct interpretation is to determine precisely what role the claim about thirst plays in the text. Clearly it is meant as an important premise in the argument that distinguishes the rational and appetitive parts of the soul—

³ See e.g. Annas 1981: 129–30.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

    - 

69

henceforth the Division argument.⁴ Socrates’ overall aim in the Division argument is to show that the psychological source of appetites is distinct from the psychological source of rational motivation; his strategy is to present a case in which someone has two motivations, one a paradigmatic appetite and one arising from reflection, which are opposed in a way that precludes them from belonging to a single non-complex subject. We can formalize the argument more or less uncontroversially as follows: P1 Principle of Opposites: The same thing cannot do or undergo opposites in the same respect and in relation to the same thing at the same time. (436b9–10) P2

Desiring and rejecting are opposites. (437b1–5)⁵

P3 Appetites (epithumiai) like hunger and thirst, and also wishing and willing (to ethelein, to boulesthai) are species of desire, and their opposites species of rejection. (437b6–9)⁶ P4 Thirst-itself is an appetite for nothing other than drink-itself—not for hot cold, much, little, or wholesome (chrēstou) drink—and mutatis mutandis for the other appetites. (437d7–e6) P5 Therefore the soul of the thirsting person, insofar as it thirsts, wants nothing other than to drink. (439a9–b1) P6 Therefore, by P1, if anything ever holds that soul back from drinking, it must be something in it other than the thirsting thing. (439b3–6) P7 It does sometimes happen that someone is thirsty but wishes not to drink (and therefore, by P3, has opposite attitudes toward drinking). (439c3–4) Conclusion: Therefore, by P6 and P7, there are (at least) two distinct elements in the soul: the one that orders people to drink in such cases—the non-rational and appetitive part; and the one that hinders them from drinking—the rational part. (439c7–d8) The crucial premises for the bare urge view are P4 and P5. To see if these do indeed support that view we need to understand what Plato means by them, and

⁴ Socrates goes on to distinguish a third element, the spirited part, but my concern in this paper is only with the first stage of the soul-division, between reason and appetite. ⁵ This is my paraphrase: Plato gives examples of various states which we would classify as desires and aversions, but offers no general term—presumably because he does not have one. (It is Aristotle who coins ὄρεξις as a general word for desire; Plato sometimes uses ἐπιθυμία, as we will see below, but in the present context he needs to reserve the term for a particular species of desire, appetite (see next premise).) ⁶ This is needed to establish that an appetite for x is genuinely opposite to a not-wishing for x: one is a want, the other a rejection. Note that the Greek phrases translated as “not wishing” and “not having an appetite” (τὸ ἀβουλεῖν καὶ μή ἐθέλειν μηδ’ ἐπιθυμεῖν, 437c7) connote aversions rather than mere absence of desire; I sometimes translate them as “wishing not [to φ]” or “being appetitively averse [to φ-ing].”

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

70

 

to do this we need to understand what role they play in the argument: why does Socrates need to establish that thirst-itself is not for any particular kind of drink? The answer to this will depend in turn on a fleshing-out of P6 and P7: we need to know what kind of conflict Plato here envisages, for this will tell us why he wants one of the conflicting motivations to be an unqualified desire, thirst-itself. It is widely agreed that the case of conflict Plato has in mind is one in which a thirsty person is averse to drink on the grounds that it is bad for her (perhaps, commentators conjecture, because she has dropsy and has been told that drinking will endanger her health). More specifically, her thought must be that drinking in the circumstances is “unwholesome.” This is clear from a passage that appears between P4 and P5: Therefore let no one catch us off our guard and disturb us by claiming that no one has an appetite for drink but for wholesome (chrēstou) drink, nor for food but for wholesome food. For indeed all people have appetites for good things. If then thirst is an appetite, it would be for wholesome stuff, whether drink or whatever it is an appetite for, and likewise with the other appetites. (438a1–5)

I discuss this passage in detail in section 2; my aim at present is to show how it explains the point of P4–5. In order for his conclusion to go through Socrates needs a case in which the Principle of Opposites applies: a case in which someone actually has conflicting conative attitudes, a desire and an aversion, toward the same thing. In P7 he envisions a case in which someone is thirsty (wants drink), but wishes not to drink (is averse to drink).⁷ In the passage just quoted, he anticipates an objection to his use of that case. How are we to understand the objection? The lines are compressed, but the emphasis on wholesomeness suggests that the objector has in mind the following: Why is the person averse to drink? Because she thinks it unwholesome? Ah, but if she thinks drink unwholesome, she cannot actually have an appetite for it: thirst is the desire for wholesome drink. So you really haven’t given me a case of conflicting attitudes at all—this person has an aversion to drink, but no desire for it—and thus haven’t given me any reason to concede a division in the soul.

The point of P4–5 is to forestall the objection. There is a kind of appetite which is for drink, but not specifically for wholesome drink. Therefore it is possible for someone averse to drink qua unwholesome still to desire it qua drink—that is, to

⁷ For not-wishing as aversion, see my note to P3.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

    - 

71

have an appetite for it.⁸ In other words, the main contribution P4–5 make to the argument is simply: P4–5*

Thirst can be for drink without being specifically for wholesome drink.

To establish this point, it is true, Socrates made a broader claim: there is a kind of thirst, thirst-itself, which is for drink without qualification, rather than for drink of any specific kind. Bare-urgeists take the existence of thirst-itself to reveal a distinctive feature of appetitive desire, even the definitive one, namely its blindness or simplicity. What I want to show is that this is not Plato’s point at all. First, the existence of unqualified appetites is incidental rather than essential to his argument: he uses an example involving an unqualified appetite because that makes his case particularly clear, rather than because such appetites are typical, let alone because they are the only kind of appetites. Second, unqualified appetites are not “blind” in the sense of cognitively impoverished. Third, there is nothing distinctively appetitive about unqualified desires: Plato could just as well have used a case of an unqualified rational desire (desire belonging to the rational part of the soul). As to the first point, we have already seen that Plato’s main aim in P4–5 is to establish P4–5*: that one can have an appetite for drink without thereby having an appetite for wholesome drink. To establish this, he made the claim that insofar as one has an appetite for drink, one’s appetite is only for drink, rather than for drink that is wholesome, hot, cold, or qualified in any other way. Bare-urgeists take his point to be that all appetites are like this, but this cannot be his view: as we saw above, in the course of arguing for P4 he mentions qualified appetites—appetites for qualified objects, like thirst for a cool drink. He does, however, claim that some appetites are unqualified: thirst-itself, hunger-itself, and so on. Why does he choose this kind of appetite to focus on in P6 and P7? Presumably for clarity’s sake: to demonstrate that someone can desire x despite its having a particular quality to which she is averse, it is simplest to show that she can have a totally unqualified desire, one for which any instance of x will do completely regardless of its qualities.

⁸ One might worry that in this case the Principle of Opposites no longer applies, since even if the desire and aversion have the same object (rather than appetite wanting drink and reason being averse to unwholesome drink), they must have it in different respects (appetite wants drink qua drink while reason rejects it qua unwholesome). But all of Plato’s examples of conflict seem vulnerable to this objection: Leontius wants to look at corpses insofar as doing so is pleasant, but is averse to looking at them insofar as doing so is shameful; Odysseus wants to kill the maidservants insofar as doing so is gratifying to his anger, but is averse to killing them insofar as it would be inopportune. The charitable conclusion is that this is simply not the sort of thing he has in mind with the “same respect” (κατὰ ταὐτον) clause in the Principle at 436b, and this is confirmed by the fact that when he repeats the Principle later that phrase is replaced by “with the same [part, aspect] of itself ” (τῷ αὐτῷ ἑαυτοῦ). A desire for x and an aversion to x (drinking, killing maids, etc.) are genuine opposites, even if they are held on different grounds.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

72

 

Now to my second point: there is no reason to take the existence of unqualified appetites to reveal any cognitive limitation of the appetitive part of the soul. What Plato claims in P4–5 is that there is such a thing as thirst-itself, which is for drink without being for any particular kind of drink. We want to know how he is conceiving of this desire: as a blind craving, or in some other way? We should begin with Socrates’ own elaboration of the idea of thirst-itself: For all things that are such as to be of something [i.e., relatives], those that have some quality (ta poia) are of something with a quality, while those that are just themselves (ta auta) will be of things that are only themselves. (438a7–b2)

He goes on to give several examples of relatives (the greater, less, more, fewer, and so on), and develops one example in detail: knowledge-itself is of what-can-belearned-itself (epistēmē men auto mathēmatos autou), while a particular sort of knowledge, qualified knowledge (epistēmē tis kai poia tis), for example, medical or house-building knowledge, is of a particular sort of thing that can be learned. I want to show that this analogy undermines the cognitive impoverishment interpretation of thirst-itself. Knowledge-itself is the determinable state that can be conceptually distinguished from any particular kind of knowledge; it is not blinder or less able to discriminate than specific kinds of knowledge, but simply a more general, abstract state.⁹ I know of three interpretations of thirst-itself that try, with varying degrees of success, to respect the analogy with knowledge-itself. Precisely because they try to follow Plato’s own guidance in using that analogy to illuminate thirst-itself, none of them make thirst-itself a blind desire: Reading 1: In saying that thirst-itself is for drink-itself Plato is making a point about the proper or formal objects of desire, along with other states and relations.¹⁰ A thirsty person may desire cold, much, or wholesome drink, but insofar as she is thirsty it is also true that she desires drink simpliciter, for what it is to be thirsty is to desire drink. Thus the person envisaged in P7 does desire the drink qua drink, even though she is averse to it qua unwholesome. She may have some particular kind of thirst (e.g., for cold drink), but will also thereby have thirstitself, just as knowledge of building is also thereby knowledge, or a red thing is also thereby a colored thing. Reading 2: Plato holds that sometimes one is simply thirsty, without being thirsty for any particular kind of drink.¹¹ Normally of course thirst is not so indiscriminate, as any thirsty person who is offered a hot black coffee or a lukewarm peach Schnapps will know. But sometimes it is, precisely in cases like ⁹ Note especially the claim that knowledge-itself “becomes” knowledge of a particular sort when it acquires a particular kind of object (ἐγένετο, 438d1). ¹⁰ See Carone 2001: 118–21. ¹¹ See Lorenz 2004: 96.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

    - 

73

that at issue in P7 (if what Plato pictures there is indeed a case in which the doctor has ordered someone not drink at all because of dropsy). If you are horribly parched because you’ve had nothing to drink all day, you will be so thirsty that any old drink will do. You just want drink, period. Thus the thirsty person in P7 has thirst-itself, without also having any particular kind of thirst. Reading 3: Plato has in mind the claim, expounded in the lead-up to his definition of the philosopher in Book V, that to desire something is to desire the whole of its kind or species (eidos).¹² Thus “drink-itself ” means the whole class of drinks. Some people on some occasions merely have appetites for cool drink or hot drink, but sometimes someone has an appetite for drink-itself, where this means that she wants drinks of all kinds. Like the lover of wisdom in Book VI who desires the whole of wisdom rather than one part or kind, the person in P7 loves and desires the whole of drink rather than wanting only some particular kind. I will not try to decide between these readings here; each has arguments in its favor, and I see no decisive arguments against any. (Reading 1 succeeds best in preserving the analogy with knowledge-itself, while Reading 3 does worst, since classes are quite unlike determinables; Reading 1, however, needs to work hard to block division within the appetitive part itself: if someone thirsty for cold drink is thereby also thirsty for drink-itself, then if she is appetitively averse to hot drink she will have conflicting appetites.) The crucial point I want to make is that on none of them—nor on any reading that takes its clue from the analogy with knowledge-itself—is thirst-itself a cognitively impoverished desire. On Reading 1 it is on a par with every other mental state: it is for drink in precisely the same definitional sense that wish (boulēsis) is for the good, or erōs for the beautiful.¹³ On Readings 2 and 3 thirst-itself is a liberal desire: the issue is not that one’s appetitive part has any difficulty recognizing different varieties of drink, but rather that in its present state it wants any of them, or all of them. This brings us to my third point against the bare urge reading. If thirst-itself is best understood in any of these three ways, then contrary to the standard interpretation on which P4–5 show that appetites are specially limited desires, these premises do not depend on any special feature distinctive of appetites by contrast with other desires. Just as thirst is defined as desire for drink (Reading 1), so rational wish is defined as desire for good. Just as one can be so thirsty that any old drink will do (Reading 2), so too can one be so desirous of victory ¹² “Must you be reminded or do you remember that if one speaks correctly in saying that someone loves (φιλεῖν) something, he must not love one part of it but not another, but must love all of it (πᾶν στέργοντα)? . . . If we call someone desiderative of something, do we say that he desires (ἐπιθυμεῖν) the whole of that kind (εἴδους), or one part but not another?—The whole” (474c8–10; 475b4–7). Some take the “must you be reminded?” at 474c8 as a reference back to 438a–b; certainly there is no other clear precedent. ¹³ Carone 2001: 118, citing Charmides 167e.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

74

 

(a spirited desire) that any old way of winning will do, or so desirous of wisdom (a rational desire) that any old subject of study will do. If we take Reading 3, the case is even clearer. For in Book V, just after claiming that one who desires something desires the whole of it (in the lines quoted above), Socrates explicitly applies that claim to rational desire: Then won’t we also say that the philosopher is a desirer of (epithumētēn) not one part of wisdom but not another, but rather the whole? (475b8–9)¹⁴

Bare-urgeists take the Division argument to show something distinctive about the appetitive part of the soul; I have argued that they are wrong. All Socrates needed, in P4–6, was to establish that desiderative conflict can occur, because one can desire something even though one believes it to have some quality to which one is averse—for example, one can sometimes desire drink even though one believes it to be unwholesome. If any of the proposed interpretations of thirst-itself are right, however, then Socrates thinks this is a feature of all species of desire, not just appetites. Indeed, he could just as well have reached the Division argument’s conclusion by pitting an unqualified rational desire against an appetite, rather than vice versa. Consider a variation on the argument which begins with the same first three premises but continues as follows: P4’ The desire for wisdom is a desire for nothing other than wisdom-itself—not for arcane, popular, much, little, comfortable, or lucrative wisdom. P5’ Therefore the soul of the wisdom-lover, insofar as it wants wisdom, wants nothing other than to acquire wisdom. P6’ Therefore, by P1, if anything ever holds back a wisdom-wanting soul from seeking wisdom, it must be something in it other than the wisdom-wanting thing. P7’ It does sometimes happen that someone wants wisdom but is appetitively averse to wisdom (and therefore, by P3, has opposite attitudes toward wisdom): for example, if the library is cold and smelly, or if she figures that there is much more money to be made by devoting herself to other pursuits. Conclusion’: Therefore, by P6’ and P7’, there are (at least) two distinct elements in the soul: the one that orders people to seek wisdom in such cases—the rational and wisdom-loving part;¹⁵ and the one that hinders them from seeking wisdom— the pleasure- and money-loving appetitive part.¹⁶ ¹⁴ Plato presumably uses “appetitive” (ἐπιθυμητήν) here in a broad sense—desirous, rather than desiring with the appetitive part of the soul—corresponding to the broad use of ἐπιθυμία at 580d7 ff. as a generic term for the desires of all three parts of the soul. ¹⁵ Plato characterizes the rational part as the part that loves wisdom (τὸ φιλοσοφόν, 411e6). ¹⁶ The appetitive part loves bodily pleasures (436b1–2 and 439d8) and wealth or gain (τὸ φιλοκερδές, 581a7).

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

    - 

75

“Comfortable” and “lucrative”—meant to denote species of the pleasant, and thereby objects of appetitive desire—here play the role of “wholesome” in the original argument. The point of P4’–5’ is to forestall an objection parallel to that imagined at 438a. Here the objector is a psychological hedonist (as Plato himself seems at times to be),¹⁷ and argues as follows: all desire is for the pleasant; therefore someone who believes wisdom not pleasant, insofar as it involves discomfort or poverty, does not really desire it; therefore the kind of case imagined in P7’ could not occur. As the objector would put it: Why is the person averse to acquiring wisdom? Because she thinks it unpleasant? Ah, but if she really thinks it unpleasant, she cannot actually have a wish for it: the wish for wisdom is the wish for pleasant wisdom. So you really haven’t given me a case of conflicting attitudes at all—and thus haven’t given me any reason to concede a division in the soul.

P4’–5’

forestall the objection, because they entail:

P4–5*’

The desire for wisdom need not be a desire for pleasant wisdom.

To establish this point our counterfactual Socrates makes a broader one: there is a kind of desire, the desire for wisdom-itself, which is for wisdom without qualification. Could there be such a desire, or is it only appetites that can be unqualified? As we have seen, if we equate “without qualification” with “the whole rather than only some part” (Reading 3), Socrates explicitly recognizes just this kind of desire in Book V (475b). If we take “without qualification” instead to mean something like “indiscriminate,” along the lines of “any old drink will do” (Reading 2), I see no reason to doubt that the desire for wisdom could be like this too: a philosophical soul who has been deprived of the opportunity to learn, like a thirsty person deprived of drink, doesn’t much care where she starts. If we take Reading 1, the case is simple: the love of wisdom is defined by its proper object, wisdom. Thus, neither the logical structure of the Division argument nor the characterization of thirst-itself depend on any difference between appetitive desires and rational ones. There are indeed many such differences: most centrally, the former are non-rational and the latter of course rational. But it is a mistake to think that Plato is explaining that very distinction here.¹⁸

¹⁷ Socrates experiments with psychological hedonism in the Protagoras, and the Athenian Visitor espouses it in the Laws (see especially 732e–733d). I am not insisting that he holds the view in the Republic, but simply noting that this variation on the argument is one that Plato himself might have found prima facie compelling. ¹⁸ In Moss 2008 I argue that he explains it over the course of the Republic, and makes it particularly clear in the later arguments for the division of the soul, in Book X: a non-rational part of the soul is one

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

76

 

2. Good-Independent Desires? So much for bare urges. In the rest of the paper I attack a subtler and superficially much more plausible interpretation of the argument: one on which it establishes that appetites are “good-independent desires”—desires depending in no way on the agent’s evaluation of their objects as good. The phrase is due to Irwin (see e.g. 1995: 209–11), but the interpretation is widespread.¹⁹ The evidence for this reading comes from a passage which comes between P4 and P5, one we saw briefly above. After asserting that thirst-itself is only for drink, Socrates adds: Therefore let no one catch us off our guard and disturb us by claiming that (a) no one has an appetite for drink but for wholesome (chrēstou) drink, nor for food but for wholesome food. For indeed (b) all people have appetites for good things (pantes gar ara tōn agathōn epithumousin). (c) If then thirst is an appetite, it would be for wholesome stuff, whether drink or whatever it is an appetite for, and likewise with the other appetites. (438a1–5)²⁰

On the standard interpretation, Socrates here rejects the very doctrine he upheld in dialogues like the Meno, Protagoras, and Gorgias: that we only have appetites (and indeed desires of any kind) for good things, where this at least entails, and perhaps is equivalent to: if someone has an appetite for a particular thing, she must believe that that thing is good.²¹ Here, the story goes, Socrates recognizes a class of desires that make no reference to the good at all: one can have an appetite for something without in any way finding it good. In other words, on the standard interpretation of 438a Socrates here rejects not only (a) and (c), but also (b). In fact, this reading of the passage is by no means obvious. It has perhaps seemed so in part because translators typically write the rejection of (b) into the text. To take a prominent example, Grube’s text as revised by Reeve has: Therefore, let no one . . . disturb us by claiming that (a) no one has an appetite for drink but rather good drink nor food but good food, on the grounds that (b) everyone after all has appetite for good things, (c) so that if thirst is an appetite, it will be an appetite for good drink. [emphases mine]

that cannot question or criticize appearances, and so desires what appears good to it, while a rational part is one that can use reasoning or calculation (λογισμός) to get beyond appearances, and so can desire what is genuinely good rather than merely apparently so. ¹⁹ See e.g., Woods 1987; Reeve 1988: 135; Penner 1990: 52. ²⁰ Μή τοί τις . . . ἀσκέπτους ἡμᾶς ὄντας θορυβήσῃ, ὡς οὐδεὶς ποτοῦ ἐπιθυμεῖ ἀλλὰ χρηστοῦ ποτοῦ, καὶ οὐ σίτου ἀλλὰ χρηστοῦ σίτου. πάντες γὰρ ἄρα τῶν ἀγαθῶν ἐπιθυμοῦσιν. εἰ οῦν ἡ δίψα ἐπιθυμία ἐστί, χρηστοῦ ἂν εἴη εἴτε πώματος εἴτε ἄλλου ὅτου ἐστὶν ἐπιθυμία, καὶ αἱ ἄλλαι οὕτω. ²¹ See most explicitly Meno 77c1 ff.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

    - 

77

This obscures the difference between “wholesome” (chrēston) in (a) and (c) and “good” (agathon) in (b).²² By choosing to put a comma rather than period before “gar ara”, and by translating it as “on the grounds that,” it also puts (b) firmly in the objector’s mouth. But (b) can stand grammatically on its own: on a neutral reading it might be a premise attributed solely to the objector, but it might just as well be one shared by Socrates and his interlocutors. And indeed, in favor of the latter reading we have Glaucon’s response: But all the same, the person who says these things does seem to be saying something right (’Isōs gar an dokoi ti legein ho tauta legōn). (438a6)

Glaucon can be read as saying that the objection merely seems right, but he can just as well be read as saying that he thinks there is something to it; Socrates can be read as intending to disabuse Glaucon entirely, but can also be read as agreeing that the objection gets something right. The text alone, I hope to have shown, does not settle the matter. Do we have other reasons, then, to think that the standard reading is right? For reasons I and others have given elsewhere, we should instead hope that it is wrong.²³ Plato seems to assert the good-dependence of all desire at 505d–e (on a straightforward reading), implies the good-dependence of appetites in particular in his description of the democratic city at 562b–c, and shows that the appetitive part is aware of normative considerations (such as what one should (dein) do), at 442d. There are philosophical costs to the standard reading too, although this is not the place to discuss them in detail: for example, Bobonich argues that if the appetitive part does not take the drink to be good it is hard to see how there can be genuine conflict between it and the rational part, and I have argued that the standard reading deprives Plato of a compelling argument that we can otherwise interpret him as making in the Republic, namely that the appetitive part desires pleasure precisely because pleasure appears good to it.²⁴ What I want to show here is that the passage at hand in no way necessitates the standard reading—and therefore that we are under no pressure to think that Plato ever questioned the good-dependence of appetites. The interpretation I propose is very similar to Adam’s (1902), and among other later writers Carone’s (2001), and Weiss’s (2006); I add to the pile because I want to show that their reading of this one passage in the Division argument is strongly supported by, and the standard reading undermined by, the account I gave in section 1 of the whole of that argument. ²² For this complaint see also Weiss 2006: 171. The translation of χρηστοῦ ποτοῦ as “good drink” is widespread (see e.g., Jowett 1888, Shorey 1937, Griffith 2000, and Bloom 1991). ²³ See my 2008 section 5 and 2006, section 4. Compare Lesses 1987, Carone 2001, Price 1995: 49–52; Bobonich 2002: 243–5; and Weiss 2006, chapter 6. ²⁴ Bobonich 2002: 253; Moss 2006 and 2008.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

78

 

We saw above that the only claims Socrates needs to reject, in order for the Division argument to go through, are (a) and its equivalent, (c): he needs to show that appetites are what we might call wholesome-independent, i.e., that one can have an appetite for something without finding it wholesome. On the standard interpretation of 438a he takes (a) to follow from (b)—the claim that all appetites are for good things—and therefore rejects (b) as well. But does he really think that (a) follows from (b)? Surely the imaginary objector does, for the argument anticipated at 438a must be something like this: As you, Socrates, have argued in other dialogues, (b) all appetites are for good things. Good drink is wholesome drink. Therefore, (a) all appetite for drink is appetite for wholesome drink.

Does Socrates have to reject his former view to resist the conclusion? No: all he need resist is the last step, the inference from (b) to (a). (For compelling arguments that this is precisely what he is doing, see Carone 2001: 118–19.) For what follows from (b) is simply this: if someone desires drink, she must believe drink to be good. Or rather—and here we have what I take to be the radical move in this argument, which does signal a break from the doctrine of the Protagoras—that part of her which desires the drink must believe drink to be good.²⁵ That the thirsty person’s appetitive part must believe drink good says nothing about what kind of drink that part believes good. As a matter of fact wholesome drink is good; but her appetitive part might believe that hot drink is good, or cold drink, or sweet drink, or much drink, and so on. If or insofar as the thirst she is feeling is unqualified thirst-itself, however, what her appetitive part believes good is just drink-itself: drink as the object desired insofar as one’s desire is thirst (Reading 1), or any old drink whatsoever (Reading 2), or drink as a whole (Reading 3). In other words, drink can be an appetite for the good without “good”—let alone the correct specification of “good” in the case at hand, namely “wholesome”— figuring into the description under which the object is desired. What the appetitive part must believe is that drink is good, not that good drink is good. And that is why Socrates can later rephrase 438a’s point as follows:

²⁵ For evidence that the non-rational parts have their own beliefs, which can conflict with the rational part’s beliefs, see the reformulation of the Principle of Opposites and the conclusion drawn from it in Book X, at 602e–603a: contrary beliefs must belong to different parts of the soul. If you think belief (δόξα) is something distinctively rational you will want to substitute another term for the lower part’s representations (as Aristotle does in de Anima III.3: phantasia); Plato in the Republic evidently had no such thought. As to the claim that the appetitive part believes drink good, this means that it believes drink worth-going-for, a quality it equates with, and indeed is completely unable to distinguish from, pleasantness. So I argue in the articles cited above.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

    - 

79

Thirst-itself is not for much nor little drink, nor good (agathou) nor bad drink, nor in a word drink of any particular sort, but only drink-itself. (439a4–7)

Probably he here uses “good” merely as a synonym for “wholesome,” but the point is general: thirst-itself need not be for pleasant drink, beneficial drink, ennobling or ambition-gratifying drink, nor for drink qualified by any other species of the good—but this is perfectly consistent with the claim that the thirsty person must see drink as good. (If so, she must see drink as good under some guise or other— presumably, as pleasant. But she need not think that only good drink would be pleasant, nor that only pleasant drink would be pleasant: maybe any old drink would do.) Similar interpretations of 438a have been proposed by others (cited above); I hope here to have strengthened their case by showing what the context does and does not demand. If my analysis of the Division argument is correct, then the point of 438a is merely to clear aside a possible objection to P4–5—a possible objection to the thesis that thirst need not be for wholesome drink. Whether or not all desire is for something the agent finds good, under some guise, and with some part of her soul, is simply irrelevant to this point. Again, we can bring out the point by considering the parallel Division argument proposed above. In order to establish P4–5*’—that the desire for wisdom need not be a desire for wisdom qua pleasant—Socrates might have digressed after P4’ as follows: Therefore let no one catch us off our guard and disturb us by claiming that (a)’ no one has a rational desire for wisdom, but for pleasant wisdom. For indeed (b)’ all people have rational desire for good things. (c)’ If then wisdom-loving is a rational desire, it would be for pleasant wisdom.

The imagined objector is arguing as follows: As you Socrates surely think, (b)’ all wishes are for things qua good. Good wisdom is pleasant—as you will argue in Book IX. Therefore, (a)’ all wish for wisdom is wish for pleasant wisdom.

Socrates certainly accepts (b)’; what he rejects is the inference to (a)’. Someone who desires wisdom must believe—or more precisely, the wisdom-loving, rational part of her soul must believe—that wisdom is good. But this says nothing about what kind of wisdom that part believes good. As a matter of fact (and although there are other ways of identifying good wisdom), pleasant wisdom is good, since coming to know the Forms is the most pleasant thing of all (see Book IX). A wisdom-loving agent’s rational part might believe that arcane wisdom is

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

80

 

good, or mathematical wisdom, or much wisdom, and so on. But insofar as she has unqualified “desire for wisdom-itself,” what her rational part believes good is just wisdom-itself. In other words, wisdom-loving can be a desire for the good without “good”—let alone the correct specification of “good,” which the imagined objector takes to be “pleasant”—figuring into the description under which wisdom is desired. What the rational part must believe is that wisdom is good, not that good wisdom is good. That is not the argument Plato gave. But, I have argued, it would have served his purposes in the Division argument just as well as the argument he did give, and therefore we should conclude that nothing in the characterization of thirst reveals any distinctive feature of the appetitive part of the soul.²⁶

References Adam, J. 1902. The Republic of Plato, 2 vols. Cambridge: Cambridge University Press. Anagnostopoulos, M. 2006. ‘The Divided Soul and the Desire for Good in Plato’s Republic.’ In The Blackwell Guide to Plato’s Republic, ed. G. Santas, 166–88. Oxford: Oxford University Press. Annas, J. 1981. An Introduction to Plato’s Republic. Oxford: Oxford University Press. Bloom, A. 1991. The Republic of Plato. New York, NY: Basic Books. Bobonich, C. 2002. Plato’s Utopia Recast. Oxford: Oxford University Press. Carone, G. R. 2001. ‘Akrasia in the Republic: Does Plato Change his Mind?’ Oxford Studies in Ancient Philosophy 20: 107–48. Cornford, F. M. 1941. The Republic of Plato (translation, introduction, and notes). Oxford: Oxford University Press. Griffith, T. (trans.) and Ferrari, G. R. F. (ed.). 2002. Plato: The Republic, Cambridge Texts in the History of Political Thought. Cambridge: Cambridge University Press. Irwin, T. 1995. Plato’s Ethics. Oxford: Oxford University Press. Jowett, B. 1888. The Republic of Plato, 3rd edition. Oxford: Clarendon Press. Lesses, G. 1987. ‘Weakness, Reason, and the Divided Soul in Plato’s Republic.’ History of Philosophy Quarterly 4: 147–61. Lorenz, H. 2004. ‘Desire and Reason in Plato’s Republic.’ Oxford Studies in Ancient Philosophy 28: 83–116. Moss, J. 2006. ‘Pleasure and Illusion in Plato.’ Philosophy and Phenomenological Research 72: 503–35. Moss, J. 2008. ‘Appearances and Calculations: Plato’s Division of the Soul.’ Oxford Studies in Ancient Philosophy 34: 35–68. ²⁶ I am grateful for comments from M. M. McCabe, Fiona Leigh, Anthony Price, and discussion with other participants in the Keeling Colloquium; I owe special thanks to Cian Dorr.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

    - 

81

Penner, T. 1990. ‘Plato and Davidson: Parts of the Soul and Weakness of Will.’ Canadian Journal of Philosophy Supplementary Volume, 16: 35–74. Price, A. W. 1995. Mental Conflict. London: Routledge. Reeve, C. D. C. 1988. Philosopher-Kings: The Argument of Plato’s Republic. Princeton: Princeton University Press. Shorey, P. 1937. Plato: The Republic, Loeb Classical Library, 2 vols. Cambridge, MA: Harvard University Press. Weiss, R. 2006. The Socratic Paradox and its Enemies. Chicago: Chicago University Press. Woods, M. 1987. ‘Plato’s Division of the Soul.’ Proceedings of the British Academy 73: 23–47.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

5 The Blind Desires of Republic IV Matthew Evans

1. Is there some essential or constitutive connection between desire on the one hand and goodness on the other, so that it is impossible to give an accurate account of the former without referring somehow to the latter? It strikes me as relatively clear that Plato, at least, thinks so. For the textual evidence in favor of such an interpretation is remarkably strong. Unequivocal expressions of something close to this view can be found all over the place, from one end of the corpus to the other. Here are just two especially striking passages:¹ Anyone who recognizes [the good] aims at and hunts for it (pan to gignōskon auto thēreuei kai ephietai), wanting to grasp it and acquire it (ktēsasthai) for his very own, and cares for (phrontizei) nothing else except what is accomplished along with goods. (Philebus 20d7–10) So it is because we pursue the good (to agathon . . . diōkontes) that we walk whenever we walk, thinking that it’s better [to walk] . . . and we stand still, whenever we stand still, for the sake of the same thing (tou autou heneka), the good . . . . And don’t we kill a man, if we kill him, or banish him or confiscate his money, because we think that it is better for us to do these things than not to do them? . . . So it is for the sake of the good that the people who do all these things do them (henek’ ara tou agathou hapanta tauta poiousin hoi poiountes). (Gorgias 468b–c)

Many other passages could be added to this list,² but the one I want to focus on in this paper is taken from Socrates’s famous discussion of the good at the end of Republic VI. Here is what he says there:³

¹ The following translations are drawn, with some modifications, from Frede (1993) and Zeyl (1987), respectively. ² See, for example, Protagoras 358c–d, Meno 77a–78c, Symposium 205a–206b, Lysis 216d–222e, Euthydemus 278e–281e, and Timaeus 86b–87b. ³ All translations of Republic texts will be drawn, with some modifications, from Grube and Reeve (1992). Matthew Evans, The Blind Desires of Republic IV In: Psychology and Value in Plato, Aristotle, and Hellenistic Philosophy: The Ninth Keeling Colloquium in Ancient Philosophy. Edited by: Fiona Leigh and Margaret Hampson, Oxford University Press. © Oxford University Press 2022. DOI: 10.1093/oso/9780192858108.003.0005

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

     

83

Isn’t it obvious that most people are content with things that are believed to be just and beautiful, even if they aren’t really, and [isn’t it obvious] that such people act, acquire, and judge in accordance with these things? Yet [isn’t it also obvious that] no one is satisfied to acquire things that are believed to be good (arkei ta dokounta ktasthai), but everyone seeks what is really good and immediately disdains all belief here? . . . Indeed, every soul pursues [the good], and does everything for the sake of this (ho dē diōkei men hapasa psuchē kai toutou heneka panta prattei). It divines that [the good] is something, but is puzzled and cannot sufficiently grasp what it is. (Republic 505d5–e3)

If this passage is any indication of Plato’s own position in the Republic, then it seems clear that in this work, as in so many others, he is fully committed to the view that all human motivation is linked to the good in some deep and nonaccidental way. Call this the basic thesis: All human motivation is fundamentally linked to the good.

Now the question I want to ask here is not whether Plato holds the basic thesis, because (as I have suggested) it seems obvious to me that he does. Instead I want to ask how he would fill out the basic thesis, given the evidence of the Republic in particular. Apparently he sees some essential connection between human motivation and the good, but what exactly is the nature of that connection, as he sees it?

2. One possible answer is that the basic thesis is best understood as a restriction on the sort of things that we humans could genuinely desire. This answer might be developed in any number of ways, but an especially interesting one is to hold that, according to Plato, each of our wants and desires is an evaluative perception of some special kind—a primitive, quasi-sensory response to the goodness of various objects of our attention.⁴ On this view, for example, the thirsty person must somehow see drink as good, simply insofar as she is thirsty. For, as the friends of Socratic moral psychology might say, if she were not struck in this manner by the goodness of drink, then she would have no more of an impulse to drink than not to.⁵ If we found this line of thought compelling, and we wanted an account of desire that would capture it, we might be drawn to something like the following view, which I will call the perceptual account: ⁴ Here I am inspired by a proposal that has been defended and developed in recent work by Jessica Moss (2008: 57–60). But I am not certain that she would accept my elaboration of her proposal here. ⁵ See, for example, Weiss (2006: 168–81).

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

84

  What it is for a psychological state to be a desire is for it to be a perception of some object as good.

Clearly this account offers an explanation of how desire and the good are fundamentally connected to each other. So if we could find adequate evidence that Plato holds this account in the Republic, then we would be in an excellent position to answer the question I posed earlier about how he wants to fill out the basic thesis.

3. But I worry that the evidence points in exactly the opposite direction. For if the perceptual account is true, then it is difficult to see how the contents of our desires could have the structure Socrates insists they have when he argues, in book IV, that at least some of our souls have exactly three distinct constituents or parts. To see the problem I have in mind here, take a moment to consider what the contents of our perceptual states are actually like. Seeing and tasting, for example, seem to have contents with a logically complex structure: what we see is not mere sky, but sky as blue; what we taste is not mere wine, but wine as sweet. In each case what we perceive is not a mere object, but an object insofar as it is a certain way, or an object under a description, or an object under a guise. If this is right, then it would seem to be a consequence of the perceptual account that the contents of our desires are not mere objects, but objects insofar as they are good, or objects under the description of being good, or objects under the guise of the good.⁶ Although this view has a certain appeal, both as an interpretation of Plato’s view and in its own right, my sense is that it is inconsistent with one of Plato’s most important commitments in the Republic—a commitment that is just as central to his moral psychology as (I believe) the basic thesis is. The following passage should give us at least a preliminary sense of what that other commitment is, and of why one might worry that it is inconsistent with the perceptual account: ⁶ Some interpreters might deny that the contents of our perceptual states really are logically complex in this sense. They might claim, for example, that seeing sky as blue should not be analyzed into seeing (the attitude) on the one hand and sky as blue (the content) on the other; they might claim instead that it should be analyzed into seeing as blue (the attitude) on the one hand and sky (the content) on the other. If we accept this second claim, then we will be able to sidestep my worry about the perceptual account. But I doubt that we should accept this second claim. For doing so would require us to recognize, and to employ in our theories of perception, a vast and ugly menagerie of distinct attitudetypes—one for every perceivable property, in fact. In the case of sight, for example, we would have to accept that seeing as red, seeing as square, seeing as clear, seeing as slow, and more generally seeing as F, for every visible property F, are all fundamentally distinct perceptual attitudes. Better to give up on the perceptual account altogether and develop some similarly motivated alternative account instead. See section 8 below for some discussion of one such alternative account.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

     

85

Then nobody will catch us unawares and upset us with the claim that what one desires is not drink but beneficial (chrēstou) drink, and not food but beneficial food, on the grounds that everyone after all desires good things (tōn agathōn epithumousin), so that if thirst is a desire, then it is a desire for beneficial drink or [beneficial] whatever, and likewise for the other [desires]. . . . [Rather,] thirst of a certain sort is for drink of a certain sort, but thirst itself is not for much or little [drink], good or bad [drink]—in a word, it is not for drink of a certain sort (poiou tinos). But thirst alone, by its very nature, is for drink alone . . . So the thirsty person’s soul, insofar as it thirsts, wants nothing other than to drink, and it both yearns for this and sets out toward this. (438a1–5, 439a4–b1)

Now some interpreters might be inclined to interpret this passage as an implicit rejection of both the perceptual account and the basic thesis.⁷ On this interpretation, the argument of book 4 signals a crucial shift, in Plato’s philosophical psychology, away from the Socratic view that the soul is necessarily unified under the rule of good-directed thought, and toward the anti-Socratic view that the soul is only contingently and only ideally so unified. While there is something tempting about this interpretation, I think we need to appreciate that it would require us to see the argument of book IV as a radical departure from a position that is not only ubiquitous in Plato’s other works, but also (as we have seen) evident in the Republic itself. But can we develop a plausible alternative to it—one that better preserves the integrity of Plato’s overall position? In the rest of this paper I will argue that we can do this, provided that we accept exactly half of the anti-Socratic interpretation. While we must concede that the argument of book IV is inconsistent with the perceptual account, we need not (and should not) concede that it is inconsistent with the basic thesis. For I believe there is an exegetically plausible and philosophically attractive way of fleshing out the basic thesis that is, unlike the perceptual account, fully consistent with the argument of book IV. My recommendation, briefly put, will be that we understand the basic thesis not as a restriction on the sort of things that we could genuinely desire, but as a restriction on the sort of things that we could correctly desire. On this alternative reading, the essential connection Plato sees between desire and the good is fundamentally normative in nature, and is similar to the connection he sees between belief and the true: just as it is in the nature of our beliefs to be made correct by the truth of what we believe, so it is in the nature of our desires to be made correct by the goodness of what we desire. If this is the right way to understand his position, then the account of desire that the remarks in book 6 support is not the perceptual account, but rather what I will call the teleological account:

⁷ See for example Reeve (1988: 131–5), Irwin (1995: 206–11), and Anagnostopoulos (2006).

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

 

86

What it is for a psychological state to be a desire is for it to be correct if its content is good, and incorrect if its content is not good.

In what follows, then, I will defend two claims about how Plato wants to flesh out the basic thesis: first, that he does not endorse the perceptual account; and second, that he does endorse the teleological account. Toward the end I will also try to explain why I think this result is not just a minor tweak in the tradition of commentary on Plato’s works, but is an observation of great importance for our understanding of one of his deepest philosophical commitments.

4. If we expect to grasp the full significance of the passage I quoted earlier from book IV, we cannot just consider it in isolation; we need to figure out how it contributes to the argument to which it belongs, at 436a–441c. Then we will be able to determine with greater authority whether or not it is consistent with the perceptual account. As I have already indicated, my claim will be that it is not. The question that motivates the argument of 436a–441c is made explicit at 436a9–b3: Do we “learn,” “get angry,” and “desire (epithumoumen) the pleasures of food, drink, and sex” with distinct constituents of our souls, or with the whole of our souls? The passage I quoted earlier is taken from the argument’s first major phase, at 436a–439d, which purports to establish that our souls have at least two such constituents. Here are what I take to be the four crucial steps in this phase of the argument: (R1)

It is impossible for the same thing (tauton) to do or undergo opposites (tanantia) in the same way (kata tauton), in relation to the same thing (pros tauton), and at the same time (hama). (436b9–c1; cf. 436e7–437a1)

(R2)

Accepting and rejecting are opposites.⁸ (437b1–4)

(R3)

Insofar as one desires (epithumein) a given thing, something in one’s soul accepts it. (437b6–c5)

(R4)

Sometimes one desires a thing and yet, at the same time, something in one’s soul rejects it. (439a9–c5)

From these four claims Socrates appears to derive two conclusions: first, that there are at least two constituents of the conflicted soul mentioned in (R4)—one by which it accepts the desired thing, and another by which it rejects it (439c6–8); ⁸ The terms Socrates uses for what I call “accepting” are epineuein, ephiesthai labein, and prosagesthai; the terms he uses for what I call “rejecting” are ananeuein, aparneisthai, and apōtheisthai.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

     

87

and second, that the constituent by which it rejects the desired thing is essentially a reasoner (logistikon), whereas the constituent by which it accepts the desired thing is essentially (not a reasoner but) a desirer (epithumêtikon) (439c10–d8). For our purposes here it is especially important to notice that the passage I quoted on page 85 above, together with its development and elaboration, takes up the entirety of the textual space between (R3) and (R4). So what we would like to develop, if we can, is an interpretation that shows how this passage might be thought to shed light on the significance of (R4), given (R1), (R2), and (R3). Now suppose you were asked, right after conceding (R1), (R2), and (R3), how one might—as in (R4)—both desire something and reject something at the same time. You would have the following three answers to choose from: (i) it is not the same thing in the soul that both accepts and rejects the same thing in the same way; (ii) it is not in the same way that the same thing in the soul both accepts and rejects the same thing; or (iii) it is not the same thing that the same thing in the soul both accepts and rejects in the same way. From his later remark at 441c4–e1 it seems clear that, according to Socrates anyway, the correct answer is (i).⁹ But if he had accepted the perceptual account, then I assume he would have held that the correct answer was (iii). For consider a thirsty soul who rejects drink (at a certain time) on the basis of a rational calculation to the effect that drink is (at that time) bad. Certainly this soul does not reject drink under the guise of the good; what it rejects, presumably, is either mere drink or drink under the guise of the bad. So if the thing the thirsty person desires is drink under the guise of the good, as the perceptual account says, then this is not the same thing that the thirsty person rejects. Thus the perceptual account more or less demands answer (iii).¹⁰ Now answer (iii) is exactly what we might expect the Socrates of the early dialogues to go for. Once it is properly fleshed out, it will tell us that the thirsty soul is sensitive to lots of different features of drink, some presenting themselves as worthy of acceptance and others presenting themselves as worthy of rejection.

⁹ On this issue I concur with Bobonich (2002: 223–35) and Lorenz (2006: 18–34) against Price (1995: 48–57). After all, if Socrates had seen answer (ii) as superior (or equivalent) to answer (i), then he would have likened the thirsty soul to the spinning top, which both moves in a rotational way and does not move in a locomotive way (436d4–e5); but instead he likens it to the archer, whose one hand pushes the bow away and whose other hand draws it near (439b3–c1). This controversy is somewhat peripheral to my concerns here, however, since my argument requires only that Socrates rejects answer (iii). To my knowledge there are no interpreters who would deny this. ¹⁰ There is a different version of the perceptual account in the neighborhood here—one according to which the thirsty person desires drink under the guise of the pleasant. But this version of the account will also demand answer (iii), and will do so for roughly the same reasons.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

88

 

In certain cases this ambivalent sensitivity might be strong enough to trigger sudden changes of plan in the thirsty soul, but (according to this story) it will not be strong enough to compromise the rational unity of that soul. After all, one can be simultaneously sensitive to both the acceptance-worthy features and the rejection-worthy features of a single thing without failing to be a single person. A single shipwrecked soul, for example, can be simultaneously sensitive to both the quenching and the toxic features of seawater.¹¹ But it is clear from (R4) and elsewhere that, according to the Socrates of the Republic, there are at least some cases of psychological conflict that run deeper than this; and when it comes to these cases he plainly shuns answer (iii) in favor of answer (i) (439a–d). As we have seen, however, anyone who endorses the perceptual account must give answer (iii) even in these cases. Thus the perceptual account actually constitutes a significant objection to the line of argument Socrates gives here. For if the perceptual account is true, then (R4) loses all of its support, and Socrates’s opponents would be left perfectly free to deny it.¹² That is why I think we must interpret the argument of book IV in such a way that it turns out to be inconsistent with the perceptual account.

5. And yet this interpretation of the argument of book IV stands in some tension with a natural reading of the passage I quoted earlier from book 6.¹³ For in this later passage Socrates seems to be suggesting that, whenever you are motivated to do something, the thing you are motivated to do is not some mere act, but some act insofar as it is good or under the guise of the good. This would directly contradict the argument of book IV, as I interpret it. So the question I want to pursue now is whether there is some viable strategy for reading the book VI passage in such a way that it does not directly contradict the argument of book IV, as I interpret it. One strategy would be to hold that there is a difference in subject matter between the two passages: while the claim in book IV is about desires, the claim in book VI is about actions. Perhaps we could argue, along these lines, that Plato is gesturing toward a sort of proto-Kantian position here: although some of our desires are not endorsed by our rational selves, all of our desires that issue in

¹¹ Even when the deliverances of perception come into direct conflict with each other, as they do when someone sees the same finger as both large and small (523b–4b), the Socrates of the Republic would certainly not insist, for this reason alone, that the rational unity of this person’s soul has been compromised. ¹² Compare Lorenz (2006: 25–31). ¹³ Sustained and useful accounts of this apparent tension, and of various ways to resolve it, can be found in Carone (2001), Anagnostopoulos (2006), Kamtekar (2006), Moss (2008), and Barney (2010).

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

     

89

action are endorsed by our rational selves.¹⁴ On this view, then, we pursue the good in all of our actions, but not in all of our desires. So if Plato holds a view of this kind, then the tension between the two passages can be removed without much difficulty. But while I find this solution intriguing, I am not optimistic that it will work. For, as far as I can tell, the Socrates of the Republic never observes any robust distinction between desires and actions. On the contrary, he repeatedly suggests that desires just are actions of a certain kind. Recall that at the beginning of the book IV argument he claims that learning, getting angry, and desiring are all things “we do” (prattomen, 436a8–9). His later description of the thirsty soul reiterates the point: If something draws [the soul] back when it is thirsting, wouldn’t that be something different in it from whatever thirsts and drives it like a beast to drink? It can’t be, we say, that the same thing, with the same part of itself, in relation to the same thing, at the same time, does opposite things (tanantia prattoi). (439b3–6)

According to Socrates, then, accepting (on the basis of desire) and rejecting (on the basis of calculation) are themselves to be understood as actions of a certain kind. So I find it highly unlikely that he would be willing to draw the sort of robust distinction between desires and actions that the proto-Kantian solution requires him to draw.

6. Is there any other solution available at this point—one that still allows us to interpret the book VI passage as a compact expression of (what I have been calling) the basic thesis?¹⁵ I believe there is; and if it works, I think it has the potential to yield an important insight into Plato’s moral psychology. The key to the solution I want to propose here lies in a subtle but potentially significant difference in characterization between the good-dependence of desire that Socrates denies in book IV and the good-dependence of desire that he affirms in book VI. What he denies in book IV is that all desire is “for the good” (tou agathou); but what he affirms in book VI is that all desire is “for the sake of the good” (tou agathou heneka).¹⁶ This will allow us to solve our exegetical problem, ¹⁴ See Anagnostopoulos (2006: 180–3) for a more detailed development of such an approach. ¹⁵ Here I am deliberately setting aside any solution that does not allow us to interpret the book VI passage as a compact expression of the basic thesis. That includes, in particular, the solution proposed by Irwin (1977: 336, n. 45). ¹⁶ He does not use the same “desire” terms in the book VI passage that he uses in the book IV passage, but I take it that this is not significant. For in both places he is making broad generalizations about the nature and structure of human motivation—generalizations that would surely fail if for some reason they did not reflect the nature and structure of human desire.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

90

 

I think—provided that we follow the general contours of an interpretation of Plato I have developed elsewhere, in my work on the Philebus.¹⁷ On this interpretation, Socrates uses the same “for the sake of ” language at Philebus 53c–55a to express two core views about attitudes of pleasure-taking: first, that by their nature these attitudes have certain content-based correctness conditions (that is, conditions under which they are correct or incorrect, given their contents); and second, that these correctness conditions are fixed or determined by whatever these attitudes are for the sake of. Now it is important to notice that something approximating this very same sense of “for the sake of ” seems to be at work in the closing discussion of Republic VI, where Socrates compares the intellectual practice of the mathematicians to the intellectual practice of the philosophers: although [the mathematicians] use visible figures and make their claims about them, they are not thinking about them; rather [they are thinking] about those other things that [the visible figures] are like (tous logous peri autōn poiountai, ou peri toutōn dianooumenoi, all’ ekeinōn peri hois tauta eoike). They make their claims for the sake of the square itself and the diagonal itself (tou tetragōnou autou heneka tous logous poioumenoi kai diametrou autēs), not [for the sake of] the diagonal they draw, and similarly with the other [shapes]. (510d5–e1)

Here Socrates makes it perfectly clear that in his view the things the mathematicians make their claims “about” (peri) are not the same as the things they make their claims “for the sake of” (heneka). And while this view is puzzling in some respects, for our purposes here it is not puzzling at all: the mathematicians’ claims are in some sense about visible shapes, but the correctness of these claims is not to be assessed by how well they measure up to those visible shapes; rather it is to be assessed by how well they measure up to the shapes themselves—of which those visible shapes are mere images. If this (or something like it) is the sense of heneka that Socrates has in mind when he suggests—about five Stephanus pages earlier—that all desire is “for the sake of” the good, then it would seem that in this earlier passage he need not be interpreted as saying that our desires are always only for good things (or things under the guise of the good); he might rather be interpreted as saying that our desires are always only correct if they are for good things (or things under the guise of the good). That is, we could see him as making a claim about how to determine the correctness of our desires, given what they are for, not about how to determine what they are for. Contrast this with what I believe we must interpret him as saying in the pivotal book IV passage. Here he is certainly not denying that the only things we could

¹⁷ See in particular Evans (2007).

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

     

91

correctly desire are good things; he is (at most) denying that the only things we could actually desire are good things. In fact his positive claim in book IV seems to be that our desires could be for all sorts of things—particularly things that are not, and are not perceived by us to be, good or pleasant or honorable or beautiful or whatever. In my view this contrast entitles us to draw a distinction, on Socrates’s behalf, between a desire’s conditions of correctness on the one hand and a desire’s conditions of satisfaction on the other. His claim seems to be that a desire is correct just in case its content is good, whether or not its content is obtained or achieved (this is the sense in which he holds that all desires are good-dependent); but a desire is satisfied just in case its content is obtained or achieved, whether or not its content is good (this is the sense in which he holds that at least some desires are good-independent—or, to use a more provocative term, blind). On my reading, then, the point of the book IV passage—together with its elaboration at 438a7–e9—is to establish that the motivational force of a desire is determined by its (potentially) good-independent conditions of satisfaction alone: Plato is inviting us to appreciate the difference between what a desire actually does push us toward (or pull us away from) and what a desire properly would push us toward (or pull us away from). Once we have this difference firmly in mind, we will be able to see that for him the blindness of our blind desires consists precisely in this: their power to move us is not sensitive to the goodness or badness of anything. Thus the solution I am recommending here is relatively simple. In the book VI passage Socrates is recommending the teleological account, according to which the only things we could correctly desire are good things, whereas in the book IV passage he is rejecting any account according to which the only things we could genuinely desire are good things. Since these two commitments are perfectly consistent with each other, the apparent tension between the two passages turns out to be merely apparent.

7. Of course this solution is not without its own potential problems. One of them, I take it, is that Socrates, in the book VI passage, seems to be suggesting that a desire’s conditions of satisfaction are just as intimately connected to the good as its conditions of correctness are.¹⁸ Consider in particular his remark at 505d7–9: “no one is satisfied to acquire things that are believed to be good (agatha de oudeni eti arkei ta dokounta ktasthai), but everyone seeks what is really good and

¹⁸ Thanks to Daniel Drucker for drawing my attention to this.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

92

 

immediately disdains all belief here.” This remark might be interpreted as saying that whatever a desire motivates us to obtain or achieve, it is always something that is in fact good. But if this reading is right, then the solution I am recommending here is in serious trouble. For this reading does not allow Socrates to distinguish between satisfaction and correctness in the way my solution requires him to. But I doubt that this reading is right. First of all, the potential subject of satisfaction in this passage—the potentially satisfied thing—is not a desire, but a person. The concept of satisfaction I introduced above, on the other hand, is meant to apply only to desires, and is meant to pick out only what these desires are desires for—that is, what these desires motivate a person to obtain or achieve. If we keep this distinction in mind, then we can see how a person might not be satisfied even if all of her desires are satisfied. This might well be the case, for example, if all of her desires are in fact incorrect.¹⁹ Second, I have trouble seeing how Socrates could be saying that, if some particular thing is not really good, then no one could be motivated to obtain or achieve it. If he were saying this, then he would be claiming, in effect, that our desires cannot go wrong. But how could he be claiming that here? The Socrates of the Republic—unlike the Socrates of the Gorgias, perhaps—clearly thinks that our desires often go wrong, and that their going wrong consists precisely in their being directed at things that are not really good, such as excessive amounts of food, drink, sex, or money. So when he says that all of us always “seek” (d8) and “pursue” (d11) what is really good, he must be implying that we succeed—in the normative sense—only if we obtain or achieve what is really good; he cannot be implying that we never actually do obtain or achieve what is not really good. This is why I think we should avoid attributing to Socrates the view that the real badness of bad things (and the real non-goodness of non-good things) is enough to guarantee that our desires will not motivate us to obtain or achieve them. But in that case it seems that we are no longer under any exegetical pressure, from this direction at least, to abandon the interpretation I am recommending.

8. Even if this particular problem can be solved, however, some readers will have been convinced all along that there is an alternative interpretation in the vicinity that is stronger and more substantive, but no less exegetically attractive, than this one. To appreciate the sort of interpretation I have in mind, it is important to see what the teleological account does and does not tell us about the nature of desire.

¹⁹ Compare Kamtekar (2006: 153–7).

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

     

93

What it does tell us is that every desire is correct if its content is good, and incorrect if its content is not good; what it does not tell us is whether every desire is somehow linked to some belief about the goodness of something. Yet it seems at least possible to read the book VI passage—and in particular the talk of “seeking” and “pursuing” we find there—in such a way that it takes a positive stand on this second question, and so suggests the following view, which I will call the doxastic account: What it is for a psychological state to be a desire is for it to be generated by, and to motivate its bearer in accordance with, a belief to the effect that something or other is good.

Many interpreters consider the doxastic account (or something like it) to be an attractive interpretation of Plato’s overall position, both in the Republic and elsewhere.²⁰ As they see it, this is the account of desire that best reflects or explains his core commitment to what I have been calling the basic thesis—the thesis that all human motivation is fundamentally linked to the good. Does the doxastic account, like the perceptual account, have the same troublesome implication that the thing any given desire is for—the content of that desire—is not a mere object, but an object under a guise? In my view it does not. For the doxastic account is conspicuously silent on the question of what the contents of desires are like. Someone could consistently hold, for example, that although thirst is always generated by a belief to the effect that drink is good, thirst itself is always for mere drink, not for drink under the guise of the good.²¹ Thus the argument I gave earlier against attributing the perceptual account to Socrates will not support the same verdict against the doxastic account. But now notice that the doxastic account, when conjoined with a couple of assumptions that strike me as uncontroversially Platonic, entails (something like) the teleological account. Here are those two assumptions: (1) any belief that generates a desire for a given thing is the belief that this thing is good; and (2) the desire generated by the belief that this thing is good is correct if this thing is good, and incorrect if it is not. Given (1) and (2), the connection between desire and belief that is specified by the doxastic account straightforwardly guarantees the correctness conditions for desire that are specified by the teleological account. The converse is not true, however: someone could consistently hold that the desire for drink is correct just in case drink is good, and yet insist—contrary to the doxastic account—that it is possible for an agent to have the belief without the desire, or the desire without the belief. Thus the doxastic account should be

²⁰ See in particular Lesses (1987), Price (1995: 49–52), Carone (2001), Weiss (2006: 168–81), Moss (2006), and Barney (2010). ²¹ On this point see especially Carone (2001), Weiss (2006: 168–81), and Moss (2006).

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

94

 

understood as an extension of the teleological account, not a negation of it.²² The doxastic account, in other words, is logically stronger than the teleological one. Are there any good reasons to avoid attributing the logically stronger account to the Socrates of the Republic, rather than the weaker one? I think there are. For this would involve attributing to him the claim that intemperate agents like Leontius (439e5–440a4) hold contradictory beliefs about the goodness of various actions. It would also, and for the same reason, involve attributing to him the claim that the beliefs one forms with one’s epithumētikon can contradict the beliefs one forms with one’s logistikon—a possibility that requires one’s epithumētikon and one’s logistikon to have access to the same conceptual resources. These exegetical consequences seem to me to be at least somewhat unattractive, if not prohibitive, and I take it to be a significant advantage of my interpretation that it does not force us to accept them.²³ In this context, at least, the relative weakness of the teleological account turns out to be one of its virtues.

9. But the teleological account might still strike us, or many of us, as too weak to be of any real philosophical interest. Why would anyone—ancient, modern, or contemporary—want to deny it? And if no one would, then why should we care one way or the other whether Plato held it? It is a testament to the enduring power of Plato’s vision, I think, that many of us now see the teleological account as an uninteresting, even trivial doctrine. For it most certainly is neither of these things. Indeed I believe it expresses the core conviction of a long and stubborn line of anti-empiricist and anti-constructivist thought in moral psychology and metaethics—one according to which our desires could not possibly be the source or basis of the goodness of things, precisely because part of what it is to be a desire is to be normatively responsive to the goodness of things: if desires are by their nature bound to track the goodness of things, as the teleological account suggests, then evidently desires cannot serve to ground the goodness of things. Maybe you find this line of thought compelling, even obviously so. If you do, then (as I have already suggested) I think you have a

²² In fact I suspect that the same will end up being true of the perceptual account as well. For we can safely assume that the perception of a particular thing as good will be correct just in case this thing is good. But then the essential relation between desire and perception specified by the perceptual account also straightforwardly underwrites the correctness conditions for desire specified by the teleological account. ²³ The nature and significance of these exegetical consequences have been discussed and debated in great detail over the last several decades—see especially Irwin (1995: ch. 13), Bobonich (2002: ch. 3), and Lorenz (2006: ch. 1)—and of course I do not pretend to have settled any of the relevant issues here. My only point is that the interpretation I am recommending does not have these consequences, and that this should count in its favor.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

     

95

friend in Plato. But you have a lot of enemies as well, and among them you will find some extremely sharp and engaging thinkers, not only in Plato’s time—Protagoras and Democritus spring to mind—but also, and perhaps more importantly, in our own.²⁴ That is why the project of interpreting and evaluating Plato’s resistance to these thinkers’ views strikes me as just about the opposite of philosophically boring. At an earlier point in this paper I claimed that the interpretation I am recommending has the potential to yield an important insight into Plato’s moral psychology. So far I have not said what that insight is, and why it is important; but by now I hope both points will be fairly plain. For if I am right about what Plato’s account of desire is, both in the Republic and elsewhere, then this account is tightly linked to a profound commitment on his part to the desire-independence of all goodness. In my view it is this commitment, more than any other, that inspires some of Plato’s finest work in the philosophical tradition we now know as metaethics. But of course I cannot hope to have established that more sweeping claim here. What I do hope to have established is that the interpretation of the Republic I am recommending in this paper is exegetically plausible, philosophically interesting, and thus worthy of your attention.²⁵

References Anagnostopoulos, M. 2006. ‘The Divided Soul and the Desire for Good in Plato’s Republic.’ In The Blackwell Guide to Plato’s Republic, ed. G. Santas, 166–88. Malden: Blackwell Publishing. Barney, R. 2010. ‘Plato on the Desire for the Good.’ In Desire, Practical Reason, and the Good, ed. S. Tenenbaum, 34–64. Oxford: Oxford University Press. Bobonich, C. 2002. Plato’s Utopia Recast. Oxford: Oxford University Press. Carone, G. R. 2001. ‘Akrasia in the Republic: Does Plato Change his Mind?’ Oxford Studies in Ancient Philosophy 20: 107–48. Evans, M. 2007. ‘Plato’s Anti-Hedonism.’ Proceedings of the Boston Area Colloquium in Ancient Philosophy 22: 121–45. Frede, D. 1993. Plato: Philebus. Indianapolis: Hackett Publishing. Grube, D. M. A. and Reeve, C. D. C. 1992. Plato: Republic. Indianapolis: Hackett Publishing. Irwin, T. 1977. Plato’s Moral Theory. Oxford: Oxford University Press. Irwin, T. 1995. Plato’s Ethics. Oxford: Oxford University Press. ²⁴ See for example Lewis (1989), Smith (1994), and Street (2008). ²⁵ I am grateful for the criticisms and recommendations I received from Daniel Drucker, Fiona Leigh, Jessica Moss, and Iakovos Vasiliou, and from audiences at University College London, Northwestern University, Oakland University, and the University of Michigan at Ann Arbor.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

96

 

Kamtekar, R. 2006. ‘Plato on the Attribution of Conative Attitudes.’ Archiv für Geschichte der Philosophie 88: 127–62. Lesses, G. 1987. ‘Weakness, Reason, and the Divided Soul in Plato’s Republic.’ History of Philosophy Quarterly 4: 147–61. Lewis, D. 1989. ‘Dispositional Theories of Value.’ Proceedings of the Aristotelian Society suppl. 63: 113–37. Lorenz, H. 2006. The Brute Within. Oxford: Oxford University Press. Moss, J. 2006. ‘Pleasure and Illusion in Plato.’ Philosophy and Phenomenological Research 72: 503–35. Moss, J. 2008. ‘Appearances and Calculations: Plato’s Division of the Soul.’ Oxford Studies in Ancient Philosophy 34: 35–68. Price, A. W. 1995. Mental Conflict. London: Routledge. Reeve, C. D. C. 1988. Philosopher-Kings: The Argument of Plato’s Republic. Princeton: Princeton University Press. Smith, M. 1994. The Moral Problem. Oxford: Blackwell Publishing. Street, S. 2008. ‘Constructivism about Reasons.’ Oxford Studies in Metaethics 3: 207–45. Weiss, R. 2006. The Socratic Paradox and its Enemies. Chicago: University of Chicago Press. Zeyl, D. J. 1987. Plato: Gorgias. Indianapolis: Hackett Publishing.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

6 Comments on Matthew Evans, ‘The Blind Desires of Republic IV’ and Jessica Moss, ‘Against Bare Urges and Good-Independent Desires: Appetites in Republic IV’ Mary Margaret McCabe

The moral psychology of the Republic is, perhaps, the first psychology that we might reasonably think of as moral. It is complex—not to say congested—both in its detail and in the way in which we might think it fits into the overall project of the whole. And in both respects, the arguments of the second half of Republic IV may have had more ink thrown at them than at any other difficult passage anywhere in Plato; so it is with considerable anxiety that I approach a commentary on these two subtle papers and with trepidation that I try to explain what I think they mean and why they matter. Both Jessica and Matt focus on the discussion of desire at Republic 437–439, supplemented by some reflections on the discussion of choice and the good at 505. Jessica asks two connected questions about Plato’s account of desire at 437–439, to both of which she answers no. First: is desire here thought of as ‘bare’ or ‘blind’? Second: is desire here thought of as ‘good-independent’? Matt focuses on the second: how far does Plato move away, in the Republic, from supposing that human motivation is ‘oriented towards the good’, by way of thinking about some version of the first: is the content of desire such that it must contain some reference to the good? If not, is there any way to explain desire as good-dependent? The argument to separate parts of the soul turns on the claim that someone may have conflicting or opposing attitudes to some object. This conflict—or so Socrates seems to argue—can only be resolved by supposing that the different attitudes are held by different parts of the soul. If, then, desire is to be separate from the reasoning part, Socrates needs to show how this part in itself is responsible for exactly one of the conflicting attitudes. Perhaps he reduces the soul to its

Mary Margaret McCabe, Comments on Matthew Evans, ‘The Blind Desires of Republic IV’ and Jessica Moss, ‘Against Bare Urges and Good-Independent Desires: Appetites in Republic IV’ In: Psychology and Value in Plato, Aristotle, and Hellenistic Philosophy: The Ninth Keeling Colloquium in Ancient Philosophy. Edited by: Fiona Leigh and Margaret Hampson, Oxford University Press. © Oxford University Press 2022. DOI: 10.1093/oso/9780192858108.003.0006

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

98

    

bare elements, such that a desiring attitude just is the relation of bare desire to some object. This would allow there to be a conflict between, for example, the bare desire for drink, driven by appetite, and some rational desire for healthy drinks, just because ‘healthy’ does not qualify desire as such. But such an account—so Jessica argues—not only fails to take account of cases where Socrates actually allows for complex desires, but it also fails to do what is needed to generate parts of the soul: bare desires, she suggests, are just too general for particular conflicts; we might wonder how bare desires could even have a more general object (such as drink, tout court). But is that not what we are forced to find at 437d–e? No—Plato’s argument, while allowing for there to be general desires—whether appetitive or rational—for some determinable (‘drink’, for example, or ‘wisdom’— desires which can be unqualified) also countenances their being qualified (desire for a cold drink when the day is hot, for example), where the qualified desires do not introduce further conflict with the unqualified ones, and so do not generate more parts in the soul. But the postulate of neither unqualified nor qualified desires tells us that desire itself is bare: only that it may be clothed. But still this argument needs to work to divide the soul. If desire is clothed, does it show up in conflict with reason? In particular, does this show that desire may, or must, be independent of the good of what is desired? No, again, says Jessica—that the desire for a cold drink may be in conflict with the desire for a healthy drink does not imply that desires are good-independent; all it implies is that there is no explicit acknowledgement by the desire that being healthy makes the drink good. Desire may still be a complex attitude even if it does not express any commitment to the good. Some of what is at issue here is whether Plato has changed his mind since the Protagoras or the Gorgias where he seems to have supposed that all our desires and attitudes depend on what we find good, and fail when we get what turns out not to be good after all.¹ Matt, like Jessica, supposes that this—what he calls the basic thesis, ‘all human motivation is fundamentally linked to the good’—still remains in place in the Republic (citing 505d–e, to which I return). But then Plato needs to explain how we could have genuine psychological conflict, not just failures of our lower urges to get things right. Matt offers Plato the following solution. We do not desire things under the description of the good (or perceiving them as good: what he calls the perceptual account, a ‘primitive, quasi-sensory response to the goodness of various objects of our attention’). Instead our achievements may be assessed for their correctness or otherwise in achieving the goal of the good (this is a teleological account). Without the perceptual account but with teleology Plato can retain the basic thesis without supposing that desires have ¹ I suspend judgement here on the vexed question of what we might mean by ‘what Plato supposed’ or what it would be for him to change his mind, without a more elaborate account of how any of the dialogues represent ‘Plato’s mind’ in the first place.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

      

99

the good in their explicit content; this allows (I think Matt’s title means us to think) that the appetites can be blind. This allows the good, then, to be explanatory of everything we do (as Matt takes the Book VI passage to say) without its figuring explicitly in our desires. So we may dispense with evaluative perception altogether, while retaining the basic thesis and its goals, the teleological structure on which it depends. Jessica and Matt agree on something like the basic thesis: that all motivation is orientated towards the good. Should they? Is there an account that will both explain the nature of appetitive desire as a part of the soul and retain this account of motivation? Jessica and Matt disagree on the content of desire, especially of appetitive desires. Jessica supposes they are complex, cognitively rich,² but not that they must have reference to the good in their content. Jessica provides an account in which desires—of any sort—have cognitive content, by virtue of having both a general motivation towards a determinable (drink, knowledge) and in being able to mention its determinant (cold drink, knowledge of triangles). The question then arises whether ‘good’ works as a determinant? If not, why not? If so, does it do so for all desires, or only rational ones? If the latter, why? If the former, what are we left with in the phenomenon of psychological conflict apart from different calculations of the good? Matt proposes that in the tripartition argument appetitive desires are blind and unqualified; but that this is consistent with the teleological account, which as such does not explain the content of desire at all, but rather what it would be for a desire to succeed. So while Jessica and Matt disagree on the content of desire, they agree that the role of the good is not explicit in it, while still explaining both success and failure. What is at stake here? In the wider context of Book IV, and of the Republic as a whole, the discussion of the tripartite psychology performs two roles: the first, which I leave aside, is as an analogue to the tripartite state; the second lies in the discussion of how and why someone—especially a guardian—might become good (see, e.g. Republic 423–424). It is, I take it, a condition of the moral psychology—and so an indirect constraint on our interpretation of it—that it should be able to account for moral development. But it is hard to see how the appetite, if it consists in bare urges or blind desires, could itself develop or change; hard to see, too, how its strength or otherwise could be anything other than a brute fact of nature, at worst out of control of reason, at best merely repressed by it. If the development of virtue is the development of a second nature of the whole complex individual, bare desire, repressed or otherwise, reads like a constant and unchanged piece of first nature,

² I mean by ‘cognitively rich’ simply that they may have elaborate descriptive content. This does not elide appetite with cognition properly so-called (so defeating tripartition); nor does it need to be expressed in terms of propositional content, if that is found an uneasy assumption in this context. But this does capture some of what Matt wants to attribute to the perceptual account.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

100

    

brute appetite. But the moral development of the complex whole, we might think, is not merely a matter of reason overcoming brute appetite. Rather, it is a complex matter for all the soul, where reason—by agreement, persuasion, discussion, and reflection—changes the orientation to the good of the soul as a whole.³ So we might think that Plato’s purposes are best served by having the other parts of the soul all, to some degree, susceptible to reason by virtue of their content, and so not bare. If, as Jessica suggests, Plato needs some account of desire whose content is cognitively rich, that neither implies the perceptual thesis nor refutes what Matt wishes to say about teleology and goals. But it is not at all clear to me exactly what Matt’s view of appetite is. It needs—to meet his objection—to be something that ensures that conflict can only be resolved by postulating different parts. It also needs to allow for what he wants from the basic thesis—that all motivation is explained by the good. For appetite, on Matt’s view, this is de facto: whether or not desires are correct depends on whether, in the event, they turn out for the best. The goals need not then be a part of the content, rather than the terms in which the desires are assessed. What then, in Matt’s view, is the content of a desire? And what is the significance of its failing? Can appetite be reformed? Now turn to the question of the basic thesis, and the passage that both Jessica and Matt take to represent it. Matt gives this: Isn’t it obvious that [1] most people are content with things that are believed to be just and beautiful, even if they aren’t really, and [isn’t it obvious] that such people act, acquire, and judge in accordance with these things? [2] Yet [isn’t it also obvious that] no one is satisfied to acquire things that are believed to be good (arkei ta dokounta ktasthai), but everyone seeks what is really good and immediately disdains all belief here? . . . [3] Indeed, every soul pursues [the good], and does everything for the sake of this (ho dē diōkei men hapasa psuchē kai toutou heneka panta prattei). It divines that [the good] is something, but is puzzled and cannot sufficiently grasp what it is. (Republic 505d5–e3: my enumeration)

He takes the central claim in this passage to come in (3): ‘Every soul pursues the good and does everything for the sake of this’. This he takes to be an inference from what has gone before, and to be the central point Socrates is seeking to make here. But that may not quite do justice to the complexity of the passage, nor to the point that Socrates seeks to make, about the importance of taking pains to come to know the good.

³ This might show the possibility of a rich perceptual thesis to explain that orientation: see McCabe (2016), even if the perceptual thesis does not follow directly from such a view of psychological complexity.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

      

101

Here is the Greek of the whole passage, through to the end of Socrates’ second speech (505d5–506a2): Τί δέ; τόδε οὐ φανερόν, ὡς δίκαια μὲν καὶ καλὰ πολλοὶ ἂν ἕλοιντο τὰ δοκοῦντα, κἂν εἰ μὴ εἴη, ὅμως ταῦτα πράττειν καὶ κεκτῆσθαι καὶ δοκεῖν, ἀγαθὰ δὲ οὐδενὶ ἔτι ἀρκεῖ τὰ δοκοῦντα κτᾶσθαι, ἀλλὰ τὰ ὄντα ζητοῦσιν, τὴν δὲ δόξαν ἐνταῦθα ἤδη πᾶς ἀτιμάζει; Καὶ μάλα, ἔφη. Ὃ δὴ διώκει μὲν ἅπασα ψυχὴ καὶ τούτου ἕνεκα πάντα πράττει, ἀπομαντευομένη τι εἶναι, ἀποροῦσα δὲ καὶ οὐκ ἔχουσα λαβεῖν ἱκανῶς τί ποτ’ ἐστὶν οὐδὲ πίστει χρήσασθαι μονίμῳ οἵᾳ καὶ περὶ τἆλλα, διὰ τοῦτο δὲ ἀποτυγχάνει καὶ τῶν ἄλλων εἴ τι ὄφελος ἦν, περὶ δὴ τὸ τοιοῦτον καὶ τοσοῦτον οὕτω φῶμεν δεῖν ἐσκοτῶσθαι καὶ ἐκείνους τοὺς βελτίστους ἐν τῇ πόλει, οἷς πάντα ἐγχειριοῦμεν;⁴

Here is a slightly different translation: Well . . . 1) Isn’t this evident, that many people would choose as just and beautiful things that seem so, even if they are not, and nonetheless do these things and get them and think accordingly? 2) But it would still suit no one to get the things that seem to be good, but they seek what is so, and in that case everyone disregards seeming? Indeed. 3) Then what every soul pursues and for the sake of which it goes to every length, divining it to be something, about this the soul is at a loss and not able to grasp adequately what it is, nor to rely on a fixed conviction about it as it does with other things, and for this reason it misses any benefit there might be in anything else. Are we then going to say that about such an important thing even those who are the best in the city, and to whom we are going to entrust everything, should be in the dark?

One puzzle here lies in what is happening when people settle for appearances in the case of just and fine things. Socrates contrasts whatever is claimed at (1)—that people pursue those appearances—and what is claimed at (3)—that it is for the sake of the good that people panta prattei. Matt takes this to mean ‘do everything’; but that seems to contradict (1), which tells us about some other things that people do. Instead, the focus of the whole passage seems to be the measures that people take to reach what is really good—as opposed to the rather more casual attitude they take towards what is just or fine. So perhaps we should read panta prattei as

⁴ Following Burnet’s text, and see Adam’s note ad loc.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

102

    

something like ‘make every effort to . . .’, ‘do their all to . . .’ vel sim.⁵ This then both fits the passage into the discussion of moral development and coming to know that is in hand still in Book VI; and picks up the earlier point about how people’s actual moral motivation (towards things other than the good) works. But if the passage does not support the basic thesis (for it suggests here that there is a contrasting motivational set directed at the just and the fine, and while insisting that everyone pursues the good, it does not claim that everyone only pursues the good), it does support teleology. For the pursuit of the good explains the efforts people will make to ensure that they know what they are doing, and what its value will be; this is what determines the correctness of what they do. How does that, in turn, relate to what we should say about the content of people’s desires? And how does that connect with what we should say about the appetitive part of the soul? Now the other passage, 437–439, from which I shall excerpt the section that seems to be about motivation and the good: Μήτοι τις, ἦν δ’ ἐγώ, ἀσκέπτους ἡμᾶς ὄντας θορυβήσῃ, ὡς a) οὐδεὶς ποτοῦ ἐπιθυμεῖ ἀλλὰ χρηστοῦ ποτοῦ, καὶ οὐ σίτου ἀλλὰ χρηστοῦ σίτου. b) πάντες γὰρ ἄρα τῶν ἀγαθῶν ἐπιθυμοῦσιν·c) εἰ οὖν ἡ δίψα ἐπιθυμία ἐστί, χρηστοῦ ἂν εἴη εἴτε πώματος εἴτε ἄλλου ὅτου ἐστὶν ἐπιθυμία, καὶ αἱ ἄλλαι οὕτω. (438a1–5)

Matt has: Then nobody will catch us unawares and upset us with the claim that [a] what one desires is not drink but beneficial (chrêstou) drink, and not food but beneficial food, [b] on the grounds that everyone after all desires good things (tôn agathôn epithumousin), [c] so that if thirst is a desire, then it is a desire for beneficial drink or [beneficial] whatever, and likewise for the other [desires].

He takes the opponent here to be asserting what Socrates wishes to deny—namely the combination of the basic thesis (b) with a claim about the specific content of the desire (represented by (a)). Socrates, he supposes, deals with the opponent by accepting (b) but denying (a) and so denying the implication to (c). This, I think, pushes Matt towards the thought that for Socrates here the appetites do not have complex content.

⁵ A similar usage might be found at Lysias Ergocles 7.5. ἅμα γὰρ πλουτοῦσι καὶ ὑμᾶς μισοῦσι, καὶ οὐκέτι ὡς ἀρξόμενοι παρασκευάζονται ἀλλ’ ὡς ὑμῶν ἄρξοντες, καὶ δεδιότες ὑπὲρ ὧν ὑφῄρηνται ἕτοιμοί εἰσι καὶ χωρία καταλαμβάνειν καὶ ὀλιγαρχίαν καθιστάναι καὶ πάντα πράττειν, ὅπως ὑμεῖς ἐν τοῖς δεινοτάτοις κινδύνοις καθ’ ἑκάστην ἡμέραν ἔσεσθε· οὕτως γὰρ ἡγοῦνται οὐκέτι τοῖς σφετέροις αὐτῶν ἁμαρτήμασι τὸν νοῦν ὑμᾶς προσέξειν, ἀλλ’ ὑπὲρ ὑμῶν αὐτῶν καὶ τῆς πόλεως ὀρρωδοῦντας ἡσυχίαν πρὸς τούτους ἕξειν.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

      

103

Jessica would gloss: As you, Socrates, agree (b) all appetites are for good things. Good drink is wholesome drink. Therefore, (a) all appetite for drink is appetite for wholesome drink.

Like Matt, she takes the opponent to be agreeing the basic thesis, (b), with Socrates; but she takes Socrates to be denying the implication of (a) from (b) because she supposes that although the desire is complex, it need not have reference to ‘good’ in its content—even if it were to have reference to ‘wholesome’. So appetites can still have complex content, but they need not contain de dicto reference to ‘good’. Jessica follows Adam (1902, 1963) in taking it⁶ that gar ara introducing (b) ‘indicates that the objector is quoting another man’s view’.⁷ Glaucon’s reply (‘Perhaps the person who says these things has a point’, 438a6) suggests, however, that the opponent is putting forward all three claims, (a)–(c), effectively to say: (b) we all desire the good; so (a) no one desires some special object simpliciter but that object under some evaluation; and (c) if this is so for people, so it is for their desires—so psychological conflict is ruled out. This allows—as it seems to me—the argument as a whole to belong to the objector; and to allow the long and complex sequence that follows in the rest of the Republic to be some kind of response to it. That response may then involve Socrates in a revision of all three of the suggestions of the interlocutor (that is the point of askeptous and the sense in which we could be caught unawares). Thus: • People are in fact psychologically complex and conflicted; the psychological phenomena deny that people considered as a whole are explicitly motivated only by the good—this is the burden of the rest of Book IV. • The orientation towards the good—whatever its scope—is certainly something that people are willing to make an effort to get right; and it finds its explanation in Socrates’ central claims about the reality of goodness (Book VI). This seems to me to vindicate Matt’s view of goals and teleology; but without determining the question of the basic thesis. • The appetites need not be determined by the good to be complex. This seems to be the burden of the accounts of vice and failure in Books VIII and IX.⁸ ⁶ Adam’s view is supported by his assumption, ad loc., that this is Platonic doctrine. ⁷ Although the appearance of the rare gar ara at Prt. 315d indicates, I think, the reverse—we might there translate ‘you know’, in introducing some new piece of information that in fact the interlocutor is taken not to know, and likewise at Symp. 205b; Adam cites Rep. 358c, but there, it seems to me, the point about the views of others is given by ‘as they say’. In any case here this alienating claim could express the objector’s view, cited by Socrates, rather than the objector’s implicating Socrates in what he says—at the least, I think the text is indeterminate here. ⁸ It does not require Jessica’s suggestion that here chrêstos, ‘wholesome’, is expressly marked off from agathos, good; about this I am not convinced.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

104

    

• Moreover, the appetites need not merely mirror the psychological structure of people (so that the inference from (a) to (c) is blocked, as is the risk of a homunculus regress). This will still allow for the possibility that the appetites might be educated, rather than just suppressed. This, I take it, is the point of the evaluative language, for example, of the return to the cave in Book VII— the returning prisoner feels sorry for her earlier self and her aspirations. It is also the point of this argument right here, embedded in a lengthy discussion of virtue and nature from the beginning of Book IV. That is why this argument finesses Glaucon’s missing of Socrates’ point—it is Glaucon, not Socrates, who at 437e7, is committed to the bare view that thirst is only of its natural object. These much-contested passages are a central element in the argument to explain the nature of the soul and its development and liability to fail. The focus of attention, that is to say, is not merely on how individual actions may be explained and analysed, but more significantly for the purposes of the work as a whole, how the nature of the individual may change both through education and at the mercy of bad influences that cause deterioration. That is why, for example, the role of failure in Matt’s teleological account is so significant: failure is a vital element in the Republic’s account of improvement, as the travails of the escaping prisoner attest. These passages tell us about the structure of the soul as the locus, not only of agency, but also of the dispositions, the virtues and the vices that constitute morality in this work. It is for this reason that this has a claim to being the first psychology that is recognizably moral.

References Adam, J. 1902, 1963. The Republic of Plato, vols I and II. Cambridge: Cambridge University Press. McCabe, M. M. 2016. ‘The Unity of Virtue: Plato’s Models of Philosophy.’ Supplement to the Proceedings of the Aristotelian Society 90 (1): 1–25.

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

7 Plato on the Object of Thirst Comments on Jessica Moss, ‘Against Bare Urges and Good-Independent Desires: Appetites in Republic IV’ A. W. Price

1. I start with Republic IV 438a1–5 in Jessica Moss’s rendering:¹ Therefore let no one catch us off our guard and disturb us by claiming that no one has an appetite for drink but for wholesome drink, nor for food but for wholesome food. For indeed all people have appetites for good things. If then thirst is an appetite, it would be for wholesome stuff, whether drink or whatever it is an appetite for, and likewise with the other appetites.

This is commonly taken to reject a Socratic thesis that all desires are for the good on the basis of a confusion between desiring an F as a good, and desiring a good F. C. C. W. Taylor (2008: 232) infers the following: ‘Appetite is simply focused on its specific, internal, object, e.g. thirst is in itself nothing more than desire for drink, and any considerations about whether drink is good or bad must be supplied by a separate principle, namely the intellect’; hence ‘bodily appetites’ are ‘evaluatively blind animal urges’. Yet, even if these urges may be ‘evaluatively blind’, they are not blind. Cf. a passage in the pseudo-Platonic Axiochus (366d2–7, tr. Hershbell): Doesn’t the baby begin his life in pain, and cry from the first moment of birth? Certainly he lacks no occasion for suffering; hunger and thirst and cold and heat and hard knocks distress him, and he can’t yet say what the problem is; crying is his only way of expressing discomfort. ¹ [Editors’ note:] This piece engages in the main with Jessica Moss’s paper in this volume (chapter 4), ‘Against Bare Urges and Good-Independent Desires: Appetites in Republic IV’, as well as, to a lesser extent, with James Warren’s paper in this volume (chapter 10), ‘Memory, Anticipation, Desire’. A. W. Price, Plato on the Object of Thirst: Comments on Jessica Moss, ‘Against Bare Urges and Good-Independent Desires: Appetites in Republic IV’ In: Psychology and Value in Plato, Aristotle, and Hellenistic Philosophy: The Ninth Keeling Colloquium in Ancient Philosophy. Edited by: Fiona Leigh and Margaret Hampson, Oxford University Press. © Oxford University Press 2022. DOI: 10.1093/oso/9780192858108.003.0007

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

106

. . 

The thirst of the newborn infant is a discomfort, but not yet a desire. Desire to drink when thirsty is not instinctive. Yet the thirst of the Republic has an object and not just a cause. Where does it learn its object from? The Philebus (as James Warren reminds us) offers an answer; and I shall propose that, in relevant respects, the Republic and the Philebus are of a piece.²

2. In her contribution to the present volume, Moss shows that Republic 438a can be read quite differently: perhaps Socrates is distinguishing desire for good (i.e. wholesome) drink from desire for drink as a good, and denying that thirst is the former while reiterating, after the Socrates of the early dialogues, that it is the latter. In reply, I want here and now to press this question: how are we to understand the phrase ‘desire for drink as a good’ if it is to apply plausibly to thirst? Whatever is being claimed must be consistent with a description of the appetitive part that ‘calls that with which it [the soul] loves, hungers, thirsts, and feels the flutter and titillation of other desires, the unreasoning and appetitive element—companion of various repletions and pleasures’ (439d6–8, after Shorey). It fits that the pleasure to which it is subject is ‘brutish and irrational’ (IX 591c6). Thus the thesis is not, and cannot be, that thirst is a rational desire, and for two associated reasons corresponding to two possible connotations of the term: (1)

Thirst does not arise through any reasoning within appetite.

(2)

Thirst does not arise through thinking that it would be good to drink.

In Stephen Schiffer’s distinction, thirst is a paradigm not of a reason-following (or ‘r-f ’) but of a reason-providing (or ‘r-p’) desire. As he writes (1976: 198), ‘One

² Here, like Plato when he is applying his principle of non-contrariety to distinguish reason from appetite (437b6–439e1), I focus exclusively upon thirst as such, and neglect the question how appetites may differ in their object, nature, and origin. It is evident, for instance, that on a hot day I may more particularly have an appetite for a cool drink (Moss well cites Republic IV 437d10–e1); and if I am choosy, it may focus upon a Chablis Premier Cru. Though there can be substitute satisfactions, it may even be that, on some occasion, nothing else will satisfy it. As Moss notes, this risks conflict within appetite, as when a man is thirsty on a hot day, needing to drink and yet recoiling from a hot drink, and only hot coffee is available. One solution could be that, in such a case, the appetite that prevails eclipses the other, so that there is no lasting conflict in relation to any particular object—though, if he doesn’t drink, the man will remain, unfocusedly, a victim of thirst. It may even be that he alternates between attraction towards, and recoil from, what is available—but without any simultaneous appetitive attraction and recoil (which would involve the contradiction of his being impelled, by a single faction within his soul, at the same time both towards and away from one and the same object).

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

   

107

does not have a thirst because one expects to gain pleasure and relief of discomfort from drinking; rather the opposite: one expects to gain pleasure and relief of discomfort from drinking because one has that thirst, that desire to drink.’ Thirst arises not out of a thought, but, as Socrates says, ‘through affections and diseases’ (dia pathēmatōn te kai nosēmatōn, 439d1–2)—affections (I take it) always, diseases when it is pathological. (Once it has arisen it may, of course, be directed by perception or reasoning towards some available drink.) An insensitivity to both these points denies early Platonic dialogues any title to contain a credible account of the intentionality of thirst. In the Lysis, Socrates counts thirst among desires (221a7–b1), yet has a general theory of desire according to which the only thing that is truly dear or philon is a ‘first philon’ (219d4–5). What this is isn’t specified; yet it can only be eudaimonia, which Socrates identifies with acting well. This implies that otherwise desire enjoys free mobility (Freud’s freie Beweglichkeit): it can have no loyalties to anything short of eudaimonia itself. Which is clearly false of thirst. The Gorgias argues of boulēsis that something desired as a means towards an end is desired under the qualification ‘if it is beneficial’ (468c2–7), and supposes that what is unethical cannot be beneficial; when it later permits of epithumia that one can desire things not good (503c6–d3, 505a6–b12), the two thoughts aren’t reconciled in any systematic way. The Meno, which uses boulesthai and epithumein equivalently, allows desires for things in fact bad, so long as they are thought to be good (77d7–e3); bad things are those that make a man unhappy, which no one wishes to be (77e5–78a5). Neither thought, as it stands, applies to thirst.³ Nowhere do we find a theory to do justice to the way in which, from the point of view of the soul, thirst arises as a brute fact, and not in response to an evaluation. A leading revisionist, Gabriela Roxana Carone, combines a recognition of the non-plasticity of thirst with an orientation of it towards the good. She can thus write, ‘It is perfectly consistent to claim that thirst qua thirst is for drink while every time we wish to drink we desire drink as good’ (2001: 120)—though I take the phrase ‘thirst qua thirst’ to be a slip, if she is indeed aware (as 118–21 otherwise indicate) that the point is not the verbal one that we count a desire as ‘thirst’ to the extent that its object is drink, but the substantive one that it is in the nature of thirst (cf. 437e4–5) to enjoy limited mobility. (It can migrate from one drink to another, but can’t even take as its object a hospital drip that will also relieve dehydration.) The question presses: in what sense might thirst take drink to be good? It is at least clear within the Republic that the objects of the appetites are the pleasures of nutrition and generation and the like (436a10–b2). Socrates’ ³ In comparing the Gorgias with the Meno (see Price, 2011: 152–66), one must not confuse desiring a hypothetical object with desiring a means as beneficial, where ‘as beneficial’ characterizes an intentional aspect of the object. Desiring (to φ if it is beneficial) doesn’t entail desiring to φ—whether or not it is beneficial to φ; if I desire to φ under the guise of the beneficial, and it isn’t beneficial to φ, I do desire to φ, though under a misapprehension.

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

108

. . 

insistence that thirst is directed solely at drink (439a4–b1) cannot be denying that its goal, more fully described, is drink as a pleasure. That throws open whether this may not also be described as drink as a good. But what would ‘good’ mean here? Moss has suggested (2008: 62), ‘Plato presents each part of the soul as finding its characteristic object worthy of pursuit.’ Yet this lacks any plausible application to thirst in general: even within a temperate soul, this doesn’t arise from an evaluation of drinking as worthwhile. Indeed, it makes better sense to talk of a thirsty man’s desiring to take a drink because he supposes it to be a good idea to take a drink when one is thirsty; but that is a cumbrous way of capturing a perfectly natural transition, one that falls within thirst itself, from feeling thirsty to desiring to take a drink (this one, say)—and even, when the action is spontaneous, to drinking itself.⁴ No doubt it is because of the limited role here of practical judgement and even of evaluation that Plato selects thirst to exemplify how irrational desire can be.⁵ What, then, can ‘good’ signify as qualifying any object of appetite without evident falsity? A simple possibility is that ‘good’ here simply means ‘pleasant’. A less reductive and more plausible view would be that, while the meaning of ‘good’ leaves its extension indeterminate, appetite applies the term always and only to the pleasant. Yet that leaves undefined what thought about the pleasant appetite is adopting in counting ‘good’ and ‘pleasant’ as equivalent, and how the thought is more than idle (bestowing an epithet of honour upon what is motivating already). A less elusive possibility may be suggested by a proposal of Schiffer’s: One’s desire to φ, one’s desire to gain the pleasure of satisfying one’s desire to φ, one’s desire to relieve the discomfort of one’s desire to φ—these are all one and same desire. An r-p desire is a self-referential desire for its own gratification. (1976: 199)

Thus all appetites have in common an intentional object that is self-referential: their own gratification. Does an appetite pursue an end as ‘good’ in pursuing it as gratifying? I take this last suggestion to yield the revisionists a hollow victory: we can all agree that any appetite seeks its own gratification; they may count this as its pursuing the good in its own way, while traditionalists won’t.⁶

⁴ Cf. Aristotle, De Motu Animalium 7 701a31–3: ‘The place of questioning or thinking is taken by the activity of desire. “Let me drink” (poteon moi), appetite says, “and this is drink”, says perception or imagination or intelligence; one at once drinks.’ ⁵ Note how, at 439c7–8, thirst is prescriptive (keleuein). It doesn’t evaluate drink, but says ‘Drink!’ That is what reason contradicts when it ‘prevents’ drinking (kōluein), presumably by an effective prohibition. There is no need to suppose the presence of a contested predicate (such as ‘good’) in order to identify a contradiction between reason and appetite. ⁶ Revisionists like to cite VI 505e1–2. Suppose that this does state—as they claim, but disputably (cf. Irwin, 1977: 336, n. 45)—that every human action is for the sake of the good. That would leave two possibilities open:

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

   

109

Can we come up with something more substantive? I believe that we can, once we turn from the intentional object of desire to its functional role. Socrates linked desires and deficiency when he reasoned as follows in the Lysis: ‘A thing desires what it lacks . . . and it becomes lacking of what it is deprived of ’ (221d7–e1). And he corrected Agathon in the Symposium by insisting, ‘That which desires desires what it lacks; if it doesn’t lack it, it doesn’t desire it’ (200a9–b1). Focal here is the notion of an objective deficiency (cf. ‘lacking’, endeēs, four times in Lysis 221d7–e2, twice in Symposium 200a9–b1). On this conception, one fundamentally desires what one needs, not in order to satisfy some contingent desire, but as a result of a lack originating from a real loss. I take this to apply to the natural and necessary appetites of the Republic. Socrates’ own focus was upon becoming good, and then (without which being good would still be incomplete) acting well, which is how he conceives of eudaimonia. Within Plato’s world-view, physical desires have their fons et origo not within the body itself, but within the soul that entered a body, and found its desires thereby transplanted and transformed. As I quoted, the appetitive part is a ‘companion of various repletions and pleasures’ (IV 439d8). The functional role of bodily appetite is to prompt a release from physical deficiency. Appetite’s relation to the good need no longer be reduced to its pursuit of pleasure and gratification. The Republic contains an inchoate analysis of pleasure, identifying it with the process—not the end-state—of ‘being filled with what befits one’s nature’ (IX 585d11)—or, in Leontius’ case, with what is against nature (IV 440a4).⁷ All the parts of the incarnate soul stand in need of repletion and restoration; pleasure is the process of becoming replete and restored. Even natural and necessary appetites rank low both in pursuing pleasure, which is a process rather than an end-state, and in the pleasures that they pursue. Thirst aims at the pleasure of drinking, and not at the maximization of pleasure; whence its limited mobility. Human good is the repletion of deficiency; being oriented towards the pleasure of becoming replete, a natural appetite stands in an internal relation to that good, even if it is one that is unideal and indirect. Yet I must end with an irony. If we may use an ‘of ’ to state a functional and not an intentional relation, it turns out to be true that thirst is of wholesome drink (pace 438a1–2), for it is when it so directs us that thirst is beneficial. So I can only offer this as a speculation on the side, and not in defence of a revisionist reading of our text. (a) All human desires derive from a fundamental orientation towards the good. (b) No human desires can suffice to generate action without being either grounded in, or reinforced by, an orientation towards the good. It is (a) that is needed for the Socratic conception of desire that we meet in early Platonic dialogues. I hope to have said enough here to set (a) aside. It is hard to make sense of (b) without (a), unless we import an alien conception of the will as a faculty required for action, and necessarily oriented towards the good. That (b) is contradicted by the diagnosis of Leontius’ akrasia offered in Republic IV is contested in Carone (2001), and reaffirmed in Price (2011: D 1 § III). ⁷ This was anticipated by Callicles at Gorgias 492a8, and echoed by Socrates at d7.

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

110

. . 

3. I have touched on nothing yet that promises to answer my initial question: whence does thirst derive its orientation towards drink? Any underlying orientation towards the good should play a part in answering it, but how? The Republic, of course, has a Form of the Good. Before and after the Republic (in the Phaedo and the Phaedrus), Forms are objects of recollection. That is unmentioned in the Republic, but lightly indicated at certain points (VII 518c4–5, 523a1, X 621a6–b1). If the revisionists are right, and to less nugatory effect than I have conceded, recollection, however inchoate, ought to play a role in explaining the origin of thirst. Yet when Plato does give an explanation, in the Philebus, it is as empiricist as traditionalists would expect, focusing not upon Platonic anamnesis but humdrum memory. In those who are empty, thirst arises as a desire to be filled with drink; yet, ‘If someone is emptied for the first time, is there any way he could be in touch with filling, either through perception or memory, since he has no experience of it, either in the present or ever in the past?’ (34e9–35a9, after Frede). The answer is that the soul makes contact with filling through memory (35b11–c1). How does this work? Socrates’ account is complex: If memory and perceptions concur with other impressions on a particular occasion, then they seem to me to inscribe words in our soul, as it were . . . There is also a painter who follows the scribe and illustrates his words by images in the soul . . . This happens when a person derives his judgments and statements from sight or any other sense-perception and then somehow views within himself images of what he has judged and stated. (39a1–c1)

The images only have point if they add to the words. We seem to have this sequence: present experience prompts memory to make certain judgements; these in turn prompt images that flesh out the verbal content. It appears that both words and images receive assent from the soul. For, when their content points to future pleasures, both constitute hopes or expectations (elpides): There are statements in each of us that we call hopes? . . . But there are also those painted images. And someone often sees himself acquiring unlimited gold, and many pleasures in consequence; and he also sees himself within the picture, enjoying himself hugely. (40a6–12)

These hopes can be true or false: they are false when they are about ‘what is never going to happen’ (d10). Thus falsity can affect pleasures (and other affections, e2–3) as well as judgements. This is a source of being worthless (ponēros) in pleasures as in judgements (e6–10). Truth within pleasure is characteristic of worthy men, falsity of the unworthy (40b2–c2).

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

   

111

No passage in Plato has been more frequently and finely debated, not least by two of our symposiasts (Russell, 2005: 176–93, Evans, 2008). All I can offer here are some questions about the interpretation that Warren takes over from Verity Harte.⁸ Harte (2004: 126) proposes that we view an anticipatory pleasure not as ‘a pleasure in the anticipated pleasure’, but as ‘an advance instalment’ of the anticipated pleasure. A natural way of understanding this is that we should not analyse the present pleasure in terms of the propositional-cum-affective attitude of being pleased that (as one supposes) one is going to enjoy oneself in some way. We might rather think of winning a forthcoming competition in a dream (or daydream) in advance of actually winning it—if one ever does.⁹ Such a ‘pre-living’ (in Warren’s happy term) may still count as false, and in either of two ways: it may be that one will not succeed tomorrow, or that one will not enjoy one’s success. In either case, there is then a mismatch between the picturing that constitutes one’s present pleasure, and the realities of tomorrow. This involves falsity as a defect when it is a condition of the pleasure which one finds in the picturing that one takes the picturing to prefigure the future.¹⁰ However, it may be that neither Harte nor Warren intends her proposal to such effect. For Harte also writes, ‘I agree with those commentators who take Socrates to portray the pleasures he takes to be capable of falsity as propositional attitudes’ (2004: 120).¹¹ Yet what can aptly be described as ‘an advance instalment’ upon, say, the pleasures of spending gold on wine, women, and song is enjoying them in one’s imagination, and not being pleased that, as one supposes, one will enjoy them. There is a further mismatch.¹² Expecting to win tomorrow, I can imagine

⁸ I have two subsidiary cavils. Warren writes, ‘Recollection and anticipation, at least for these purposes, are treated as symmetrical capacities differing in their temporal direction: one looks forward in time while the other looks backward’; and later he presents ‘reliving’ and ‘pre-living’ as analogous. I don’t think he can really mean this. Anticipation derives a complexity from its reliance on memory, whereas memory does not rely on anticipation. Also, when he twice says that the painter ‘depicts’ the logoi, he must mean that he illustrates them. ⁹ Perhaps cf. the dreamers of the Theaetetus, who have ‘false perceptions’ (158a1), and conceive of themselves in the dream as being winged and flying (b4–5); However, Socrates connects the latter to having false opinions (b2, cf. d3–4), and may have in mind under ‘things that appear to the individual’ (a2) such contents as that one is flying. ¹⁰ It is crucial here that the picturing fills out the judging or asserting in ways that may multiply the subject’s mistakes. Hence, pace Protarchus, it is not just the logoi that are false. However, the question then arises why it should be essential to such a pleasure of the imagination that one expects things to turn out as pictured. Surely daydreaming has its pleasures, even if an expectation of fulfilment may magnify them. ¹¹ It is a complication that, as Harte notes (2004: 116 n.) ‘A is pleased that’ is standardly factive: I can’t be pleased that p unless p. However, we have the form of words ‘A is pleased that, as he supposes, p.’ Cf. a case familiar within the philosophy of action: ‘A φ’ed for the reason that, as he supposed, p.’ It is not at all clear how best to make sense of such constructions. ¹² Evans (2008: 98 n.) objects that there can be no advance instalment upon a pleasure that will never materialize. Well, I might draw in advance upon a future salary that I am then denied—which gets me into a debt that I have to repay. The inapplicability of this to pleasure reveals that all that has been supplied is a suggestive metaphor, parading as an analysis, that fails to sustain any argument.

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

112

. . 

now what it will be like to win. I then picture tomorrow to myself from the point of view that I expect then to occupy (say, one offering a view of the stadium from the victor’s podium). Such might indeed be called ‘an advance instalment’ upon a victor’s joy. Yet that isn’t what Socrates describes, which is instead (as I quoted) this: ‘Someone often sees himself acquiring unlimited gold, and many pleasures in consequence; and he also sees himself within the picture, enjoying himself hugely’ (40a10–12). As Richard Wollheim distinguished (1984: 72–6), there is central imagining, within which one assumes the point of view of a subject, and acentral imagining, within which one imagines a scene from no point of view that is occupied within it. What could aptly be called an ‘advance instalment’ of an anticipated pleasure would surely share a point of view with that pleasure: imagining enjoying myself in some way, I already enjoy myself in that way within my imagination. Yet what Socrates describes is seeing oneself within a picture as if from outside it—an imagining that, in Wollheim’s terms, is not central but acentral. This fits the conception of anticipatory pleasure as a propositional attitude: one is pleased that, as one supposes, one is going to enjoy oneself in the way the picture represents. Harte’s talk of an anticipatory pleasure as an ‘advance instalment’ is attractive; yet it appears to be neither thought through, nor faithful to Plato’s text. When one is daydreaming, it can be part of the fantasy that one is thirsty, and relieving one’s thirst. In the cases that Socrates is now considering, one is (e.g.) thirsty, and so can at best be subject to a mixed pleasure (which one takes in anticipating a quenching) and pain (which comes of the present depletion). At worst, one suffers both in reality from the present depletion, and in imagination from its expected continuation (35e7–36c1). When one is thirsty, imagining the pleasure of drinking is only itself pleasant if one expects to drink. How is it that falsity in pleasure is characteristic of the wicked and unworthy? Not, I suggest, through divine providence or intervention; rather because the pleasures to which they look forward typically turn out to be of a type that is excessive, and mixed with pain (cf. 42c5–d3 with Republic IX 586b7–c5). What is clear and not disputable within the Philebus is that appetites are viewed as originating from memories of past satisfactions, and expectations of similar satisfactions to come. There is much talk of pleasure, and none of the good. Is this significant for our understanding of the Republic? I noted that the Republic anticipates a conception of pleasures as replenishings (e.g. IX 585d11). We can read this part of the Philebus as completing what, in the Republic, was unfinished business.¹³ Is it better to read the Republic in the light of what came before, or of

¹³ Equally, we can take it, together with Timaeus 71a3–d1, as spelling out what it is, in a political metaphor, for reason to ‘persuade’ appetite that it ‘had better not’ (Republic IX 554d2). If so, we should not read out of such language that appetite is oriented towards the good, and even towards the better and the best.

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

   

113

what came after? There can be no general answer; yet, if we wish to make Platonic sense of thirst, we have to prefer the second.

References Carone, G. R. 2001. ‘Akrasia in the Republic: Does Plato Change his Mind?’, Oxford Studies in Ancient Philosophy 20: 107–48. Evans, M. 2008. ‘Plato on the Possibility of Hedonic Mistakes’, Oxford Studies in Ancient Philosophy 35: 89–124. Harte, V. 2004. ‘The Philebus on Pleasure: The Good, the Bad, and the False’, Proceedings of the Aristotelian Society 104: 111–28. Irwin, T. H. 1977. Plato’s Ethical Theory. Oxford: Clarendon Press. Moss, J. 2008. ‘Appearances and Calculations: Plato’s Division of the Soul.’ Oxford Studies in Ancient Philosophy 34: 35–68. Price, A. W. 2011. Virtue and Reason in Plato and Aristotle. Oxford: Clarendon Press. Russell, D. C. 2005. Plato on Pleasure and the Good Life. Oxford: Clarendon Press. Schiffer, S. 1976. ‘A Paradox of Desire.’ American Philosophical Quarterly 13: 195–203. Taylor, C. C. W. 2008. ‘Plato on Rationality and Happiness.’ [2003] In his Pleasure, Mind, and Soul: Selected Papers in Ancient Philosophy, 223–39. Oxford: Clarendon Press. Wollheim, R. 1984. The Thread of Life. Cambridge: Cambridge University Press.

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

8 Courage and Pleasure in Aristotle’s Ethics Raphael Woolf

1. According to Aristotle the happy life is the life of virtue, specifically the life in which we actively exercise our virtue. The happy life is also, for Aristotle, a pleasant life. And one might immediately think that this indicates a possible tension in his position, since exercising the virtues does not, at least in some important cases, appear to be a particularly pleasant thing to do. Let us take two of the most prominent of the Aristotelian virtues, temperance and courage. To be temperate is to be disposed to keep away from pleasures of a certain sort; to be courageous is to be disposed to confront certain things that are, on the face of it, frightening. How does the exercise of such virtues square with the idea that the life we are talking about is a pleasant one? Vice in these respects seems to offer more prospect of pleasure than virtue. In this paper I shall focus on courage, since arguably it presents the biggest challenge to the notion that the Aristotelian good life will be a pleasant one. Temperance, at least, involves merely the avoidance of certain pleasures, whereas the exercise of courage makes it probable that one will encounter some pretty unpleasant things. Moreover the courageous person’s activities look to be painful in two distinct ways. Firstly, as just indicated, the situations that the courageous person encounters will often themselves be unpleasant—in particular, since this is a paradigm case for Aristotle, when it comes to what one has to face on the battlefield. But secondly, engaging with these situations rather than fleeing them is likely to be a frightening thing to do—and fear itself is a powerfully unpleasant emotion. Can Aristotle reconcile the apparent unpleasantness of the courageous life with the requirement that the life of virtue be a pleasant one? For Aristotle, courage, like all the virtues, is a certain sort of state or disposition (hexis, diathesis), indicating in the ethical context that by which one’s emotions are either in accordance with or contrary to reason (EE II.2, 1220b18–20; NE II.5, 1105b25–6 with II.6, 1106b36–1107a2). Emotions in turn are picked out as

Raphael Woolf, Courage and Pleasure in Aristotle’s Ethics In: Psychology and Value in Plato, Aristotle, and Hellenistic Philosophy: The Ninth Keeling Colloquium in Ancient Philosophy. Edited by: Fiona Leigh and Margaret Hampson, Oxford University Press. © Oxford University Press 2022. DOI: 10.1093/oso/9780192858108.003.0008

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

     ’  

115

being accompanied by pleasure or pain (NE II.5, 1105b21–3)—accompanied, as the EE puts it more exactly, ‘in their own right by perceptible pleasure or pain’ (II.2, 1220b13–14). The emotion that courage is particularly concerned with, according to the EE, is fear (III.1, 1228a27); or as the NE has it, with both fear and confidence, but more with things that inspire fear than those that inspire confidence (NE III.9, 1117a29–30). Fear specifically, according to the EE, is a kind of ‘painful emotion’ (lupē . . . kai pathos, III.1, 1229a36) whose object is that which causes destructive pain (a34–5).¹ The Rhetoric along similar lines characterizes fear as ‘a kind of pain or disturbance arising from the appearance of coming destructive evil or pain’ (II.5, 1382a21–2). This double aspect of fear’s relation to pain—both its own quality and that of its object—reflects the two distinct ways in which a courageous person’s life might be painful. The NE reports, without explicitly endorsing, a definition of fear as ‘expectation of evil’ (III.6, 1115a9). Whether Aristotle endorses this or not (and its connection with Socratic intellectualism suggests that, at least as a definition, he probably does not),² the status of fear as an emotion is not in question, and this means it has a painful or pleasant quality—evidently the former in the case of fear—whether or not that quality defines it. How, then, do courage and fear relate to each other? I shall argue that Aristotle conceives of the courageous person as feeling little or no fear in the performance of his courageous activities,³ and that these activities, insofar as they are performed because they are fine, even produce a certain pleasure for the agent. In reaching this conclusion I shall examine Aristotle’s conception of courage both in the EE and the NE. While noting some differences between the two works,⁴ I shall maintain that they pull very much in the same direction in their accounts of the pleasures and pains of the courageous life. Both works claim that the courageous person feels little or no fear in relation to his proprietary activities; the NE will add the thought that there is some positive pleasure to be derived from the courageous person’s attainment of what is fine. I shall suggest that this overall conception is best explained by Aristotle’s conviction that the good life must be a pleasant one. Since he holds that the good life consists in the exercise of one’s virtues, his strategy is to present this exercise, with courage as its hardest case, in the most pleasant light that he can.

¹ Pears (1980, 174) reports this passage as claiming that ‘the courageous man fears “things that are painful and destructive” ’. But the text makes no such claim, merely stating that the proper object of fear is what is painful and destructive. I shall argue that Aristotle thinks the courageous person does not in the main feel fear when confronted by such objects. ² See Price (2011, 128–9). ³ Given that Aristotle associates courage paradigmatically with the (in his day) chiefly male preserve of the battlefield (NE III.6, 1115a20–31), I shall generally use masculine pronouns when referring to the Aristotelian courageous agent, without prejudice to Aristotle’s wider views on gender. ⁴ For a detailed comparison see Mills (1980).

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

116

 

2. Like the other virtues, Aristotle conceives of courage as being a certain mean. In the EE it is presented fairly straightforwardly as a mean between two contrary extremes, namely cowardice and recklessness (III.1, 1228b2–3). (The picture in the NE is, as we shall see, slightly more complicated.) Aristotle is keen to emphasize, however, that courage is rather closer to recklessness than it is to cowardice (EE III.5, 1232a23–7, III.7, 1234b11–12, cf. VI. [= NE VII.] 9, 1151b6–8). ‘The courageous person seems to be mostly fearless (aphobos),’ he says (III.1, 1228b4–5), and this raises the question of how much fear the courageous person does feel. Aristotle worries that if the courageous agent merely withstands things that are fearful to another person rather than himself, then courage would be ‘nothing impressive’ (ouden semnon, b11). The way that Aristotle addresses this worry is an interesting one. He distinguishes between what is fearful without qualification (haplōs) and what is fearful for an individual (tini), where the former is what is fearful for ‘most people and human nature’, and the latter what is so for a given individual (EE III.1, 1228b22–6). The courageous person is fearless with regard to what is fearful without qualification (b26–7)—the things most people find fearful, he does not, and that is why the courageous disposition is praised (b29–31). As if this has not quite dispelled the worry about the impressiveness of courage, Aristotle’s claim is further nuanced. He says that ‘in a sense’ (hōs) things that are fearful without qualification are fearful to the courageous person, that sense being insofar as he is a human being; insofar as he is courageous they are not at all fearful or only slightly so (1228b27–9). This is a somewhat strange claim. Does the courageous person find the same things frightening that regular people do or not? Although ‘yes and no’ is a characteristically Aristotelian response to a question, it is difficult to say exactly what it means here. Presumably the courageous person cannot simultaneously fear and not fear the same thing (or fear it both slightly and considerably).⁵ He is, on the other hand, unquestionably both a human being and courageous. Perhaps, then, the courageous person, being human, recognizes that the things he confronts are fearful for a human—they are, in that ⁵ Pears’ response to this problem is to read the courageous person as feeling fear but acting as if he did not, arguing that otherwise ‘there would be nothing for him to face’ (1980, 178). But of course there would be something: the possibly destructive pain of the situation, which there is no reason for him not to recognize as such even if it does not evoke in him the fear that a regular human feels. Heil (1996) argues, along broadly similar lines, that the agent does feel fear but that the fear has no (adverse) motivational force. This is perhaps what Aristotle should have said but I see no evidence for such a contrast in the text; and things being ‘slightly (ērema) or not at all’ fearful for the agent is more suggestive of feeling than of motivational outlook. Sarah Broadie points out in her comments (138) that rendering ērema as (of the agent) ‘quietly’ fearful would be more conducive to a motivational reading; but the context, and in particular the application of the term to fearful things (at e.g. 1228b9, 13, and most notably at b29 itself, paired with ‘not at all’), where it can only mean ‘slightly’ fearful, suggests that the latter is Aristotle’s intended sense.

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

     ’  

117

sense, fearful to him as a human; and yet he doesn’t fear them as a regular human would—thus they are not fearful to him insofar as he is courageous.⁶ The courageous person, for example, understands as much as any of his fellow humans that engaging a dangerous enemy on the battlefield is an activity apt to inspire fear,⁷ but his courage is manifested in his feeling unafraid or hardly afraid, where a regular human would feel considerably afraid. This is what Aristotle may mean when he goes on to wonder whether the courageous person perhaps would not even get afraid (EE III.1, 1228b38–9) and replies that maybe he does so ‘in the way described’ (ton eirēmenon tropon, 1229a1). He then continues by noting that what distinguishes the courageous person from the reckless one (as indeed from the coward) is that courage is in conformity with reason, and reason bids that we choose what is fine (a1–2). One might take this either of two ways. Aristotle might mean that the courageous person feels the regular person’s fear all right, but that his reason recognizes that the goal of the activity in question is a fine one, and so, despite the fear, the courageous person performs the activity. Yet Aristotle has just said that, insofar as he is courageous, fearful things are not fearful at all or only slightly so to him. So instead, being afraid ‘in the way described’, which sounds like a way of flagging some non-standard sense of ‘being afraid’, would mean recognizing the situation as a fearful one for a human, but feeling little or no fear because one realizes that tackling it is a fine thing to do. In this rather stronger sense the courageous person is in conformity with reason—reason’s control means not that one is as afraid as the regular person is but one’s fear fails to deter one from acting; rather, the agent’s understanding that acting is a fine thing to do makes it the case that, despite recognizing the humanly fearful situation as such, one acts with little or no fear. This recognition aspect is highlighted later in EE III.1, where Aristotle picks out the courageous person as the one who thinks that genuinely fearful things are fearful, whilst cowards think that what is not fearful is fearful, and the reckless think that what is fearful is not fearful: the courageous person thinks ‘exactly the truth’ (talēthē malista, 1229b25), and facing danger when ignorant of it does not, Aristotle adds, count as courageous. The courageous person, then, fully recognizes, say, that he is facing a formidable enemy and that this situation brings with it the prospect of pain and death and to that extent is fearful. If we were to say that he then also feels a large dose of fear, that would in one important respect fail to distinguish him cleanly from the coward. For the latter, who thinks that things

⁶ As an analogy, consider someone able to, say, run a mile in less than four minutes. Some athletes may perhaps even do this without much difficulty. They can nonetheless recognize that what they are doing is a humanly difficult thing. ⁷ Thus at NE X 1178b12–13 Aristotle can mock the idea that courage would be a virtue of gods given that the courageous ‘withstand fearful things’. The idea of something’s being fearful—in the sense relevant to courage of threatening life and limb—presumably does not arise for gods, as it very much does for humans (including, of course, courageous humans).

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

118

 

that are not fearful are fearful, presumably also thinks that things that are fearful are fearful! Should we say that Aristotle is prepared to attribute a similar emotional state to both the coward and the courageous person when each is faced with a scenario calling for courage? Aristotle is not explicit on the matter. But at 1228b31–8 he had commended, by analogy with the strong (in relation to toil), and the healthy (in relation to excess), the courageous person as being unaffected (apathēs) by things that would affect most people: the implication appears to be that as a strong person will not be worn out, and a healthy person will not get sick, where most people faced with similar circumstances would, so a courageous person will not get scared, fear being the emotion or, more literally, the ‘affection’ (pathos) relevant in the assessment of courage and cowardice, as exhaustion and disease are in the analogous cases. The difference, then, between the coward and the courageous person will be in the fear that is felt—little or none by the latter, a substantial amount by the former. Aristotle is clear later that courage makes one ‘disdainful’ (kataphronētikos, III.5, 1232a39–b1) of danger, while the NE speaks of courage as acquired in part by one’s being habituated to disdain fearful things (II.2, 1104b1–2); and ‘disdain’ does not sound like an attitude compatible with a significant amount of fear. If so, one needs to ask why Aristotle chooses to adopt this way of picking out courage rather than an alternative. He might, for example, have said instead that the courageous person and the coward not only each recognize as fearful situations that call for courage, but also feel the same amount, or at least each a considerable amount, of fear, the difference being that unlike the coward the courageous person is motivated to stand firm because he realizes that that is the fine thing to do. There is no question that Aristotle thinks it a necessary condition of genuinely courageous action that one act for the sake of what is fine. On the other hand, I see no reason to suppose that he thinks cowards must fail to recognize that engaging in this sort of action is a fine thing. Aristotle indeed famously insists that it is virtue that makes the target (skopos) correct (EN VI. [= EE V.] 12, 1144a7–8, EE II.11, 1227b23–4), so if having virtue is what gets us choosing the right ends, then perhaps the vicious person does not see acting courageously as fine. But Aristotle’s point is surely about the agent’s practical goal. Recognizing that certain actions are fine is not the same as being motivated to do them oneself. That virtue causes the agent to adopt the right goal for herself does not commit Aristotle to the (odd) view that a necessary condition for cowardice is a kind of moral blindness about what counts as fine.⁸ What needs explaining, rather, is why courageous people end up doing what is fine, and cowards do not.

⁸ Aristotle does say at NE III.4 1113a29–34 that the good person and the bad each have their own distinctive (idia) view of what is fine and pleasant, with the good person distinguished by having the correct view. But while he goes on to say (1113b1–2) that the many choose pleasure as a good thing and avoid pain as a bad thing, he does not here commit himself to a view about how the fine figures in such

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

     ’  

119

One might say: courageous people value doing what is fine over avoidance of being injured or killed in battle, whereas the coward values the latter over the former. But it is not clear to me that Aristotle thinks even this. That would encourage the opening up of a space to impute to him additionally a conception of a kind of cowardice-like akrasia to parallel the intemperance-like akrasia that he discusses,⁹ such that a certain sort of person, not a proper coward, would think it best to engage in battle because doing the fine thing is more valuable than avoiding death or injury, but who gets overcome by fear and so doesn’t do it, by contrast with the full-on coward who doesn’t even think that engaging in battle is best since (he supposes) avoiding pain and death is better than doing what is fine. Now in the NE Aristotle does briefly, and in a different context (the discussion of friendship), mention people who fail to do what they think is best for themselves through cowardice (IX.4, 1166b10–11). But note: it is still cowardice, not an akrasia-like counterpart, that he describes as underlying this phenomenon. And that does seem the right way to think about cowardice: a person who doesn’t do a fine thing because he thinks that self-preservation is more important might indeed be a coward, but no less a coward would be a person who doesn’t do a fine thing, rightly perceiving it as the best thing, because he is too afraid to do it. As Aristotle had insightfully remarked earlier in the NE (III.12, 1119a21), intemperance is ‘more like a voluntary matter’ (hekousiō(i) mallon eoiken) than cowardice is. There is room to disjoin intemperance, the wholehearted giving up of oneself to pleasure, from akrasia, the conflicted yielding to it against one’s better judgement. One chooses pleasure, Aristotle says, but one merely avoids pain (a22–3). So it would, for example, be odd to talk of a cowardly agent as the one who races wholeheartedly, as opposed perhaps to slinking reluctantly, from the battlefield when one ought not to.¹⁰ With cowardice, the idea of wholeheartedness doesn’t quite fit in the first place, and thus there is less purchase for a distinction between full-blown cowardice and an akratic correlate. That being so, one needs to ask why the coward doesn’t do what he may well regard as the right thing to do, whereas the courageous person does. An obvious answer—and, I take it, this is at least part of Aristotle’s answer—is that the coward is too afraid to do it and the courageous person is not. Thus, according to Aristotle, ‘the coward . . . fears everything, while the courage person is in the opposite state’ (NE III.7, 1116a2–3). Even where the coward does not think it people’s choices, such that, for example, ‘The intemperate man’s distinctive view recommends the principled pursuit of pleasure to him as fine, and not simply as pleasant’ (Nielsen (2017, 10), her emphasis)—still less to a view about the relation between the coward’s choices and their view of the fine. To be sure, Aristotle notes that the coward calls the courageous person reckless (NE II.8, 1108b24–5). But this sounds like a self-serving justification on the coward’s part rather than a denial that, to the extent that something is courageous, it is fine. ⁹ Cf. Barney (2019, 290). ¹⁰ Barney’s serial draft-dodger (2019, 290) might be described as dedicated in his cowardice, but not, I think, wholehearted.

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

120

 

the right thing to do, we have to explain why; and again, one might naturally appeal to fear: fear distorts the agent’s value system and makes him (wrongly) assess self-preservation as being more important than doing what is fine. It is on account of pain that we hold back from fine things, Aristotle tells us (NE II.3, 1104b10–11); and fear is painful. Thus the coward is afraid even of things one ought not to be, while the reckless person is confident about things one ought not to be (EE III.1, 1229a4–6). The courageous person is afraid and confident of the things one ought to be afraid and confident about, since he has fear and confidence as reason bids; and what reason bids is that one not pursue what is painful and destructive unless doing so is fine (a6–9). Note that Aristotle has not simply said that the courageous person acts as reason bids but that he has fear and confidence as reason bids—as the NE puts it (III.7, 1115b19–20), both how he acts and how he is affected will accord with the merits of the case (kat’axian) and with reason. Since Aristotle connects following what reason bids with choosing for the sake of the fine (EE III.1, 1229a1–2, cf. NE III.7, 1115b12–13), it seems unlikely that the courageous person will fear, to any great extent, fighting the enemy in battle—on the assumption that that is a fine thing to do, it will not be something that merits fear. What the courageous person will presumably fear is what only the reckless person is (wrongly) confident about—dangers undertaken other than for the sake of what is genuinely fine. But those in turn would be things that his rationally attuned fear will in any case warn him away from. It would, after all, says Aristotle, be shameful to withstand what is frightening in cases where it would be crazy (manikon) rather than fine to act thus (EE III.1, 1230a32–3). The things that the courageous person chooses to do, then, he will do with little or no fear. It is noteworthy in this regard that when Aristotle begins his main discussion of courage in EE III he recalls his table of virtues and vices in II.3 by saying that he distinguished recklessness and fear as opposites (III.1, 1228a28–9). What is in fact in the table is recklessness and cowardice, with courage the corresponding virtuous mean (II.3, 1220b39). The slip, if it is one, suggests that for Aristotle here fear is specifically associated with cowardice and that the courageous person is not one who will need to live with his fears when he acts, but rather one who in those situations will have little in the way of fear to live with.

3. The EE describes courage as a mean between two extremes, cowardice and recklessness. The NE paints a slightly more complex picture. Although at one end cowardice remains the single extreme, at the other there are now two: recklessness and excessive fearlessness. The latter vice seems to be pathological, and Aristotle does not give it a name, noting that one who feared nothing would

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

     ’  

121

be mad or immune from pain (III.7, 1115b26–7). The reckless person too is now rather different. Aristotle characterizes him as essentially brazen—someone who talks a good game, but when danger actually needs to be confronted doesn’t hang around. Courageous people, by contrast, are ‘keen’ (oxeis) in confronting actual danger but quiet in advance (1116a8–9). Courage, then, is in rather sharper relief, with respect to its proprietary vices, in the NE than in the EE. True, the EE, like the NE, refers to ‘chickenhawks’ (thrasudeiloi)—people who are bold as brass except when they actually have to face danger—but the reference is well away from the main discussion of courage (EE III.7, 1234b3) and so not directly connected to the characterization there of those who are reckless. In the NE, by contrast, in the middle of his discussion of courage, Aristotle says that ‘most’ reckless people are in fact of that stripe (NE III.7, 1115b32). Similarly the EE more frequently portrays the courageous as somewhat closer to the reckless than they are to the cowardly. In its non-common books this relation is mentioned on two occasions, but only once in a book unique to the NE (II.8), with an additional single mention in a common book (VII. [= EE VI.] 9, 1151b6–8), though this in fact compares the reckless to the confident rather than to the courageous. The NE, but not the EE, looks outside courage’s proprietary vices in noting a resemblance between one who doesn’t fear what it is shameful not to fear (e.g. loss of reputation) and a courageous person, since both are, albeit in markedly different ways, fearless (III.6, 1115a11–16). When it comes to courage and its proper vices, the NE, but not the EE, marks a symmetry of extremes: the courageous person appears reckless by comparison to the coward, cowardly by comparison to the reckless person (II.8, 1108b19–20). Appears, not is, of course. The overall picture suggested by the NE is of a virtue more distant from any of its vices than in the EE, even as those vices swell from two to three. We may, however, regard the discussions in the two works as complementing rather than competing with one another. We surely ourselves have a conception of a type who swaggers around when danger is distant but backs off when it has to be faced; Aristotle’s portrayal here is psychologically acute. Equally we have a conception of a type who is foolhardy or impetuous, drawn to dangers that no sane person would think worth engaging with. Although Aristotle calls each type in the respective works ‘reckless’ they are clearly distinct from one another and each belongs distinctly in a catalogue of vices. In an obvious way too, as Aristotle points out, the foolhardy person, while still being an example of vice, is closer to the virtue of courage than is the swaggerer. The unnamed excessively fearless person in the NE, who at any rate does not get much of a billing, is perhaps supposed to represent, in substance, the reckless person of the EE, or perhaps a more extreme version thereof. It is less clear that Aristotle makes room for someone we might regard, not in a pejorative way as excessively fearless, but as heroically ‘too brave’, facing dangers that are fine but absurdly risky or difficult. Such a person would exemplify what is

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

122

 

sometimes called supererogatory virtue. Aristotle does mention what we might regard as its contrary, namely the brutish or morbid coward who fears even the most innocuous things (NE VII.5, 1149a7–9). When it comes to courage, however, Aristotle is if anything inclined to point in the opposite direction, noting that there are some things that are beyond human capacity to withstand (NE III.7, 1115b7–8; EE III, 1229b17–21; cf. Pol. VII.11, 1330b37–9). Aristotle does have a notion, in general terms, of superhuman virtue, but he discusses it only briefly at the start of NE VII.1 and doesn’t single out courage as an aspect of it. Perhaps Aristotle’s ‘great-souled man’ might be a candidate for supererogatory courage. He is disdainful of things that a regular person would be in awe of or intimidated by, and in the EE’s account this is specifically related to the courageous person’s disdain for danger (III.5, 1232a38–b3) that we discussed above. But ‘disdain’ is not, I think, quite what we are after in capturing the idea of the agent who is heroically or stupendously brave. Let us, then, examine what the NE has to say about the relation between courage and fear. In particular, to what extent, if any, does the courageous person feel fear in performing his courageous actions? Aristotle is clear about this—the courageous person is ‘fearless’ (adeēs, III.6, 1115a33) in face of the dangers he has to encounter, in particular, so long as death would be fine, in confronting the mortal dangers of battle, which Aristotle regards as the paradigmatic form of danger that a courageous person will contend with. He is a fearless person of a kind (aphobos . . . tis, 1115a16), ‘of a kind’, as the context makes clear, because it is certain sorts of object that the courageous person as such is fearless about, namely the perils of war rather than, say, poverty or disease, though Aristotle does add that one may call such an attitude to these latter things courageous ‘by resemblance’ (kath’ homoitēta, a19). At the start of NE III.7 (1115b7–9) Aristotle says that there are things fearful to all (or at least to anyone in their senses)—namely those beyond human capacity to withstand—and then in the human range, things that vary in their degree of fearfulness. He concedes that the courageous person will even be afraid of ‘such things’ (b11–12) but in the light of the earlier description of the courageous person as ‘fearless’ I think we should regard this as a fairly narrow point, especially as Aristotle has also just observed that the courageous person is as ‘unshakeable as a human can be’ (anekplēktos hōs anthrōpos, b10–11).¹¹ Presumably, then, it is in situations at the limit of the human capacity to withstand that the courageous feel fear. These are indeed cases where the courageous person is supposed to engage the danger, not avoid it; we are obviously not talking about situations that only the ¹¹ Taylor’s rendition ‘undisturbed as appropriate to a human being’ (2008, 289) seems to me a little too weak. Although Aristotle’s point is that the courageous can only be expected to operate within human limits, he (as any virtuous person) is someone who while not exceeding the bounds of humanity is nonetheless an outstanding example of it. The courageous person is surely not supposed to be averagely undisturbed.

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

     ’  

123

reckless, if anyone, would take on. What Aristotle says is that the courageous person will withstand them ‘as one ought and in accordance with reason for the sake of what is fine’ (1115b12–13). Hence the courageous person is one who ‘withstands and fears . . . and likewise is confident about’ the right things for the right reason in the right way at the right time (b17–19). So as in the EE, it is only dangers the tackling of which is a fine thing that the courageous person will involve himself with. And it seems that if the dangers are grave, albeit worth tackling, he may feel fear in doing so. On the other hand, Aristotle does not say how much fear the courageous person will feel in such situations; in fact he notes that one can fear these things to a greater or lesser degree (1115b13–14). And it may be that his remark about the unshakeable constitution of the courageous person implies that his fear will not, even at the limit, be great. The EE, after all, was prudent enough to allow that the courageous person may be a little afraid. But in general fearlessness in facing the right sort of danger is the mark of the courageous person. Aristotle states in a later portion of the NE discussion that the brave person withstands things ‘that are and appear frightening to a human’ (III.8, 1117a16–17), which looks like a condensed version of the EE’s contrast between what is frightening without qualification— that is, to most people and to human nature—and what is frightening to an individual. The courageous person, being human, will recognize that he tackles the sort of thing that is frightening for people in general. Aristotle immediately goes on to remark that the courageous person’s withstanding such things for the right reason (because it is fine to do so) explains why it is considered more courageous to be ‘fearless and undisturbed’ (aphobon kai atarachon) in unexpected situations of peril than in foreseen ones, since the lack of preparation means that the reaction is indicative of a genuine settled disposition (1117a17–22). Courage is the disposition in question, and fearlessness its proper accompaniment.

4. I turn now to a different but related aspect of the courageous person’s profile and his relation to pain. Thus far I have been arguing that Aristotle takes a rather minimal view of the fear felt by the courageous person in exercising his courage. To this extent the courageous person has a relatively painless existence. Fear is a painful emotion, and so the fearless person is mostly without pain at least in that regard. But Aristotle can hardly deny—and does not deny—that the situations which the courageous person will have to face are often themselves painful, in a brute physical sense. If fighting on the battlefield does not result in death—and possibly a very painful death—there is the risk of debilitating injury to contend with. ‘Courage brings pain’, as Aristotle laconically puts it (NE III.9, 1117a33–4), and is rightly praised (he adds) for that reason.

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

124

 

But no sooner has he admitted this than he does something rather surprising. He makes the case that there is a pleasurable aspect to courage, namely what pertains to the goal or end of the activity. He compares this to boxers who endure a lot of pain for the sake of the victory and honours that they fight for (1117b2–5). Aristotle says that this end may appear to involve nothing pleasant given all the pain that has to be reckoned with (b5–6). But the implication is that the appearance is misleading. Aristotle does not say that the boxers do what they do even though the reward they might eventually gain has nothing pleasant about it. He says that the goal appears to have no pleasure given all that they have to go through—his point, I take it, being that for someone exerting themselves in the middle of an inevitably painful fight the pleasure to be gained from victory might seem so remote as to be virtually non-existent; but actually ‘the end is pleasant’ (to men telos hēdu, 1117b3), pleasant enough to be worth enduring (and risking) the pain for. What is the goal in the case of courageous action? Not, I think, the honour of having won victory or even of just having fought bravely.¹² The analogy with the boxer is not supposed to be as tight as that. Honour, Aristotle has already told us, is the aim of those who display one of the five deficient forms of courage outlined at NE III.8, the one he calls ‘civic (politikē) courage’ (1116a17–29). While he regards this latter as closest to courage proper given that a desire for honour, which is something fine, motivates it, the courageous person properly so called acts for the sake of what is fine without such intermediate motivation, a point emphasized in III.9 when Aristotle says that the courageous person withstands the sufferings of war ‘because it is fine to do so or shameful not to’ (1117b9). He goes on to conclude that the case of courage is an example of a virtue whose exercise is not pleasant except, he adds, insofar as it ‘touches on’ (or ‘attains’, ephaptetai, 1117b16) the end. This is a strange qualification,¹³ given that as a virtue the exercise of courage is already the doing of what is fine, so cannot, one would have thought, help but touch on the end. There is no question that for Aristotle what is fine and what is pleasant coincide for the virtuous person (I.8, 1099a7–24). So what he seems to be doing here, albeit in a suitably low-key way, is reiterate the pleasantness of the activity despite the pain and suffering that the occasion of the activity may bring with it.¹⁴

¹² See Pangle (2018, 586); contrast Curzer (2002, 152). ¹³ As noted by Garver (1980, 163). ¹⁴ The caution (one might almost say contortion) with which Aristotle here affirms that the exercise of courage is pleasant seems to me to tell against Broadie’s suggestion (comments p. 139) that we read his point about the courageous agent’s pleasure as mainly indicating the free and full expression of the agent’s developed (second) nature (see also Broadie 1991, 318–20). A point about the agent’s thoroughgoing commitment to his courageous action would surely not require the hedging that Aristotle gives it. His approach is better explicated if he is indeed arguing for a certain pleasantness in what the agent does—not, as Broadie rightly says (139), that the agent revels in what he does; but that his action’s being fine is, nonetheless, a proper source of pleasure for him. As Anthony Price pointed out in

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

     ’  

125

Aristotle’s point about the goal appearing (misleadingly) to have nothing pleasant about it would then perhaps be a recognition that in the midst of battle the pleasure of performing what is fine is being accrued, even if not felt (or not felt until later); its source is always in the performing of the action itself. This, if I am right, offers an account of the courageous person’s life that, without for the most part straining the bounds of credulity, seeks to play up as much as possible its pleasantness and downplay as much as possible its painfulness. Aristotle does not ignore—how could he?—the degree of pain and suffering involved in the exercise of courage. But the courageous person will have less fear—and to that extent less pain—than the coward (and the regular human); and he will gain more pleasure, insofar as he acts because doing so is fine, than both the coward and the reckless person. Why does Aristotle adopt this approach? One of his chief motivations, I suggest, is the need to reconcile two convictions about the nature of the good life, that it must be virtuous and that it must be pleasant. Aristotle has little sympathy with views that would seek to disjoin the goodness of a life from its pleasantness, as his arguments in NE VII, among others, attest. And certainly he holds that his thesis of pleasure as unimpeded activity will show why the universal view that the happy life is pleasant is a reasonable one (VII.13, 1153b14–15).¹⁵ Equally, he finds no credibility in an account of the good life that would have it as anything other than a life of virtue. That the good life is pleasant and that it is virtuous are each well-entrenched and widely held opinions.¹⁶ For Aristotle, rejecting either would be pointless grandstanding. Characteristically, he aims to construct a theory that will incorporate both. His main way of doing so is to make the case that happiness is virtuous activity, and that such activity is inherently pleasant. Given his view that the good life is the life of virtue, he can thus show that it is the goodness of such a life, and not any contingent features of it, that is responsible to an important degree for its pleasantness. As he puts it in NE I.8, actions in accordance with virtue would be ‘pleasant in themselves’ (hēdeiai kai kath’hautas, 1099a13–15).¹⁷ The virtuous life is a pleasant life—more particularly, the life of those who do fine and good things

discussion, Aristotle concedes that the courageous agent may be ‘unwilling’ (akonti, 1117b8) to face death but will endure it if doing so is fine, which creates difficulty for an account that sees Aristotle as concerned primarily with the wholeheartedness of the agent’s engagement. ¹⁵ There is of course huge controversy about the details of the interpretation of Aristotle’s account of pleasure in Book VII and its relation to the account given in Book X, but a discussion of these issues is beyond the confines of the present paper. ¹⁶ He does say at NE X.1, 1172a27 that the role of pleasure is subject to much dispute (which, in terms of rival philosophical theories, could hardly be doubted). But his chief concern is to do justice to the common opinion that the good life must be pleasant. ¹⁷ At NE X.3, 1174a4–8 Aristotle notes that possessing (echein) the virtues would be valued even if it were not pleasant; but he does not extend this thought to the exercise of the virtues, suggesting perhaps a reluctance to play down the pleasantness of the happy life whose chief component is virtuous activity.

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

126

 

is intrinsically pleasant (1099a5–7).¹⁸ The case of courage, though, as Aristotle will tacitly admit, does not fit tidily into this picture. The courageous person is likely, and knows he is likely, to experience the most unpleasant ordeals in the exercise of his courage. And this is problematic, given the importance of courage as a virtue— the first Aristotle discusses in his accounts of the individual virtues in both the EE and NE. Aristotle, then, seeks to demonstrate that at least one major source of unpleasantness, namely fear, is largely absent from the courageous person’s life; and furthermore that there is something to be said for the idea that there is pleasantness in performing courageous actions because they are fine. Neither of these positions is obviously untenable. We do, I think, recognize a kind of person who is pretty much fearless in situations that most of us would contemplate with deep alarm, and who does the right thing for the right reason in those situations; and we do call such a person courageous. Likewise it is not straining things too much to argue that there is a satisfaction and sense of achievement to be found in acting courageously that is pleasurable, to be set against the pain and suffering experienced in so acting. What is striking about Aristotle’s position on courage is not that it is implausible, so far as it goes, but that it only goes so far. We recognize that there is a kind of fearlessness we should call courage. But we also recognize that there are people who do the right thing, for the right reason, in spite of being afraid, and we would call such people no less courageous—perhaps in a sense more so—than those who face similar situations without fear.¹⁹ Likewise, although it is arguable that acting courageously may be a source of pleasure for the agent, there seems nothing necessary about this. We would recognize as equally courageous—again, perhaps in a sense more so—an agent who did not gain any pleasurable pride or satisfaction from his activities, but simply valued doing what is fine more highly than preservation of life and limb. There is, it seems to me, no particular pressure to attribute either fearlessness or a pleasant sense of achievement to the courageous agent—unless one has an independent reason for doing so. And this, I think, is what explains Aristotle’s stance. He needs to show that the virtuous life and the pleasant life can be one and the same, and that is why his account of courage takes the form that it does.

¹⁸ Aristotle does envisage lives that can be so affected by adverse fortune that they are no longer happy (NE VII.13, 1153b19–21), or at least no longer ‘blessed’ (I.10, 1100b25–30), with the presence of pain adduced as a key feature of such lives. But that does not amount to a denial that insofar as a life consists in the exercise of virtue it is pleasant. ¹⁹ That said, Curzer (2012, 56) is surely mistaken to assert, in arguing that one should not attribute to Aristotle the view that courageous people are fearless, that ‘There is nothing admirable about not fearing the fearful.’ Curzer similarly (57) misreads Aristotle’s own statement at EE 1228b10–11, that it would be nothing impressive if the courageous agent faced situations that only others found frightening, as a definitive claim rather than as part of the dialectic that sets up the distinction between fearful to oneself qua human being and fearful to oneself qua courageous.

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

     ’  

127

Thus Aristotle does not, it seems to me, recognize a category of person who acts courageously in spite of his fears rather than in the absence, or near absence, of fear. In his discussion of the five defective forms of courage he mentions two sorts of creature who act because they are afraid, namely the soldier who goes into battle for fear of punishment from his commander, and the animal who attacks through fear that he is being threatened (NE III.8, 1116a31–b3, b31–1117a1). These, Aristotle tells us, are not genuine species of courage precisely because they are enacted through fear rather than for the sake of what is fine. What we are looking for, rather, is a picture of a courageous agent who indeed acts because doing so is fine, and in spite of not because of his fear. There is, in the EE, a virtue of ‘endurance’ (karteria, II.3, 1221a9), the person of endurance being the one who puts up with pain when it is best to do so. But this is a virtue distinct from courage, to such an extent that Aristotle can later claim that some people of endurance can even be cowards (III.1, 1229b2). Nor, I think, is there for him an enkratic counterpart of courage to match the enkratic counterpart of temperance²⁰—no agent fighting against his fears in order to do the sort of actions a courageous person does, to match the agent who succeeds in resisting unsuitable pleasures in spite of having a desire for them.²¹ In the NE, admittedly, endurance is no longer a virtue but instead the quality of being able to resist pain, with the enkratic person being the one who can resist pleasure (VII.7, 1150a13–15). But despite appearances, this is not a state relating to courage as enkrateia relates to temperance. Endurance is, together with enkrateia, firmly located by Aristotle within the framework of temperance and intemperance, which is concerned with the pleasures and pains of touch and taste (1150a9–13). Aristotle thus conceives of endurance as being about one’s ability to withstand physical discomfort rather than the mortal perils of the battlefield that are the proper concern of courage.

²⁰ See here Charles (1984, 171), who also notes (n. 8) that Aristotle does not talk of a fear-based version of akrasia. Perhaps the idea of control that underlies the concept of enkrateia has less purchase for Aristotle in the ‘negative’ case of what, as he sees it, deters one from acting (fear) than it does in the ‘positive’ case of what urges one to act (desire). On the relation between courage and enkrateia see also Vigani (2017, 324–7). ²¹ Leighton (1988), while right to deny that Aristotelian courage should itself be regarded as a form of enkrateia, seems to me mistaken to regard it as incorporating a substantial positive role for fear. Much of his evidence comes from the Rhetoric; but that work, of course, is concerned with a general audience rather than specifically with the virtuous, and its claim that ‘fear makes people good at deliberating’ (II.5, 1383a6–7) is unlikely to mean that fear is a necessary condition for deliberating well. Taylor (2008, 292) suggests that ‘the desire to avoid pain and death, even in circumstances when undergoing them is fine, is not a bad desire but an aspect of properly functioning human nature’ so that feeling fear in such circumstances would not for Aristotle entail enkrateia rather than virtue. But surely something’s being an aspect of properly functioning human nature is compatible with its being bad. If I desire to have sex with my neighbour’s wife, then my desire, part of my properly functioning human nature, is made bad by the occasion of its being felt. So too a desire to avoid pain when the situation demands I undergo it is not obviously a correct desire just because aversion to pain reflects a properly functioning nature.

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

128

 

Yet even if Aristotle had extended his conceptual apparatus of enkrateia to include a (non-virtuous) counterpart of courage,²² that itself would be telling. For it is of course a familiar question whether we should regard an agent who does the right thing, because it is the right thing, in spite of impulses to the contrary, as virtuous or not, even perhaps as more virtuous than an agent who acts similarly but without any such contrary impulses. Aristotle is explicit in NE I.8 that the person who does not enjoy his fine actions is not virtuous, citing just and generous actions—though not (here at any rate) courageous or temperate ones—as examples of what is to be enjoyed (1099a15–20). Evidently, then, on this view one cannot be properly virtuous if in performing the appropriate actions the agent is beset by conflict or struggle. That certainly does not seem like the only view one could take of the conditions for virtue, so we need to ask what more, if anything, lies behind Aristotle’s position. There is, perhaps, a terminological point that deserves noting. The Greek word for ‘fine’ (kalon) is also the standard term for ‘beautiful’, and it seems natural to say both that one takes pleasure in what one finds beautiful, and that an agent who has no awareness of his actions as fine is not properly virtuous. If the term one is using for ‘fine’ here is a term that standardly means ‘beautiful’, one might easily conclude that a properly virtuous agent will take pleasure in his fine actions. Should that be an element in Aristotle’s outlook, it seems to me an unfortunate one:²³ the attitude of an agent to his fine actions need not in any straightforward way be parallel to the enjoyment of beauty. If the versatile nature of kalon suggests otherwise to Aristotle, then I think he is misled, though of course much more might be said here than my rather peremptory treatment allows.²⁴ Another, and perhaps better motivated, reason to deny that an agent who does not enjoy doing virtuous actions is virtuous is the following: we would otherwise arrive at a situation in which virtuous people might have largely unpleasant existences. Resisting desire and living with fear are each in their own way significant and unpleasant burdens. For one convinced, as Aristotle is, that the good life must be both the life of virtue and a pleasant life, there is every incentive to portray the virtuous life and the pleasant life as in minimal tension with one another. As we shall now see, it is in fact hard to explain Aristotle’s emphasis on the pleasantness of the virtuous agent’s life purely in terms of a certain conception ²² That he might have contemplated this is suggested by the juxtaposing of courage, as well as temperance, with enkrateia at NE I.13, 1102b26–8. See Pearson (2012, 241). ²³ For a more sympathetic view, in the context of Aristotle’s discussion of courage, see Sanford (2010, 438–9). ²⁴ See Kraut (2013). The Rhetoric even characterizes what is fine as that which, by being good, is pleasant, though in fact it is the alternative characterization given there, in terms of praiseworthiness, that Aristotle uses to explain why virtue is fine (I.9, 1366a33–6). For discussion of the role of the fine and the virtuous agent’s relation to it see e.g. Wielenberg (2000); Richardson Lear (2006); Moss (2012, 206–19); Korsgaard (2014).

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

     ’  

129

of what virtue itself must look like. That is, even if we agreed that there was something unsatisfactory about calling one who performs their virtuous actions with a sense of burden or struggle a virtuous agent, that is not sufficient to account for the role Aristotle has pleasure play in the virtuous life. Its standing, rather, is to be explained by the need to integrate pleasure into his overall picture of the good life.

5. Let us turn, then, to Aristotle’s account of how one becomes virtuous. This is, of course, by ‘habituation’, the practice of the sort of actions that when done from a settled disposition and for the right reason will be those of a virtuous agent. We become courageous by doing courageous things, and so forth (NE II.1, 1103a34– b2). Similarly, says Aristotle, it is by acting in formidable (deinois) situations and being habituated to be afraid or confident therein that one becomes respectively cowardly or courageous (1103b16–17). So the aim of habituation is to produce confidence rather than fear when facing the kinds of situation that the properly courageous person will tackle.²⁵ One becomes courageous by being habituated ‘to disdain and withstand frightening things’ (NE II.2, 1104b1–2). ‘Disdain’ perhaps suggests, as we noted earlier, that one recognizes such things as frightening but is not, oneself, frightened by them. At any rate, Aristotle goes on to say (appealing here to the authority of Plato) that the aim of education is to get the young to enjoy and be pained by the right sorts of things (II.3, 1104b11–13), given that the one who enjoys, or at least (Aristotle concedes) is not pained by frightening things is courageous, whereas the one who is pained by them is a coward (b7–8); the temperate person likewise takes enjoyment in his abstention from bodily pleasures (b5–6). And pleasure, Aristotle adds, is why we do base things, pain (as we saw above) why we refrain from fine things (b9–11). What should we make of this picture? Firstly, it reinforces the idea that the properly courageous person is not afraid when confronted by situations that demand his engagement. And Aristotle thinks this disposition is brought about by getting the young to perform acts that are courageous and avoid those that are cowardly. The EE, while having less to say about habituation than the NE, remarks that ‘the road [to becoming good] is through pleasure’, and that fine actions must be pleasant (EE VII.2, 1237a6–7). Presumably reward and punishment, encouragement and discouragement, play their roles in getting the young to opt for the right things; and rewards and encouragements will in general be pleasant, punishments and discouragements painful. As the EE puts it (II.1, 1220a34–7) in discussing how virtue is attained (there are similar remarks at NE II.3, 1104b13–18): ‘corrections’ ²⁵ For a discussion of the relation between courage and confidence (with a different view from mine on the role of fear), see Pearson (2009).

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

130

 

(kolaseis) are achieved through pleasures and pains, and this Aristotle cites as evidence for the more general thesis that virtue and vice are about pleasures and pains. Evidently pleasure and pain, while each potentially corrupting, can when correctly deployed push us towards what is right and away from what is wrong. Now Aristotle is certainly wary, in the case of training for courage, of erring too far in the direction of pain. In Politics VIII.4 he criticizes the Spartans both for making courage such a central goal of their system of education and for, in any event, going the wrong way about achieving it via a brutal training regime. Aristotelian courage is attained by moulding, not crushing, body and spirit; and Aristotle is evidently sensitive to the idea that overly direct mimicry of the realities of war may not be the best way of producing in one’s trainees the fortitude required to confront those realities. Yet equally, it seems that a training regime more distant from the reality cannot truly prepare one for what may have to be encountered when mere rehearsals are over. The question, then, is this: why does the fact that using the levers of pleasure and pain, as Aristotle thinks one must, to train youngsters to do the right thing mean that doing the right thing will itself be pleasant (or at least not painful) for them? Why should not one say that rewarding and encouraging a youngster to do courageous things will result not necessarily in an adult who doesn’t feel fear in doing what he ought in times of war and the like, but instead that the fear he feels will not prevent his doing what he ought? The pleasures and pains of his training, that is, lead him to see courageous acts as being the right thing to do, without this requiring him to take pleasure in doing them; and cowardly acts as the wrong thing to do, with fear, present though it may be, an inadequate motive to act other than rightly. That seems no less plausible an outcome of the kind of training that Aristotle envisages than one in which fear of the battlefield becomes simply, or even largely, absent. In fact, with regard to the courageous person’s attitude to the character of his actions there seems on Aristotle’s own account to be room for a conception that is not explicated in strictly hedonic terms. When Aristotle revisits the topic of habituation in NE X.9, he notes at one point that its objective is to produce in the trainee an affinity with virtue, such that one loves what is fine and loathes what is shameful (1179b30–1). The verbs Aristotle uses here—stergein and duscherainein— could, if a little blandly, be translated as ‘value’ and ‘disvalue’ or (slightly less blandly) ‘favour’ and ‘disfavour’. But even when rendered ‘love’ and ‘loathe’, these are terms that do not demand to be understood as concerned with pleasure and pain. It might, then, be open to Aristotle to give an account of the courageous person’s attitude that is not tied to such considerations.²⁶

²⁶ The puzzle is well (if perhaps unwittingly) brought out by Burnyeat (1980, 78), who summarizes the goal of habituation by noting quite fairly that the intrinsic value of virtuous actions is not fully grasped by the Aristotelian agent ‘until [he has] learned to value (love) them for it, with the

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

     ’  

131

Moreover, in a passage we have already referred to, namely the discussion in NE III.9 of how there may be pleasure for the courageous person in the fineness of what he does, as there is for boxers in the honours they are fighting for, Aristotle says that the courageous person will indeed be pained (lupēsetai) at the prospect of death precisely because he has such a worthwhile life, and significantly, that he may be all the more courageous for acting as he does in the light of this, choosing what is fine over the goods of life he may be depriving himself of (1117b9–15). Yet this description, which renders a little incongruous the connection the passage draws, however delicately, between pleasure and what is fine, never gets translated into a statement that the courageous person is afraid, even very afraid, of what he may be losing,²⁷ his love of the fine being what keeps him on track. Nor is it clear that such a scenario need be thought of by Aristotle as enkratic (or quasienkratic), since one can be afraid at the prospect of death without experiencing conflict in choosing to face it.²⁸ In the absence of a statement that he is afraid, the courageous agent’s pain at the prospect of death seems to represent an emotion more like rueful regret than the rush of fear that, one might wish to say, is perfectly compatible with proper courage.²⁹ It is not as if Aristotle seems to have thought that an appreciable amount of fear would, as a matter of psychological fact, necessarily inhibit an agent’s actions in the kind of situations that the courageous person is likely to encounter. Although he indeed holds, as we have seen, that pain makes us hold back from fine things, fear is not specifically characterized by Aristotle as inherently inhibitive. Indeed in the Rhetoric (II.5, 1383a6–7; cf. n. 21) he tells us that ‘fear makes people good at consequence that [he takes] pleasure in doing them’. But why must intrinsically valuing (or even loving) virtuous actions mean that one has to take pleasure in doing them? If (for non-extrinsic reasons) I love running marathons, it does not, I think, follow that I take pleasure in running them. ²⁷ Notwithstanding that the basis for the agent’s good life is no doubt those very virtues whose exercise can sometimes threaten its continuation (so that Aristotle can even claim at NE IX.8, 1169a22–4 that the virtuous person regards a life lived finely for a year as preferable to one lived indifferently for many years), these are genuine goods that will be lost. On the relation between courage and happiness (eudaimonia) from this perspective see Rogers (1994). ²⁸ As Jiang (2000, 29) observes, ‘There is no necessary connection between the fear of x and the desire to avoid x.’ I do not think Richardson Lear (2004, 150–1) is right to say that for Aristotle the absence of a desire to avoid death would already mean that the courageous agent were unafraid. She cites Aristotle’s claim in Rhetoric II.5 that fear requires at least a faint hope of escape (1383a5–6); but to lack, in the right circumstances, a desire to avoid death is consistent with not regarding a situation as hopeless. It is even arguable that in a non-hopeless situation a desire to avoid death (with its corollary of fear) can itself be consistent with unimpeachable courage (see here Young (1977), who reads Aristotle himself along these lines). It should, then, be possible for Aristotle to say that a courageous person is afraid without imputing conflict to the agent—so we need another explanation of his reluctance to say that the courageous agent is afraid. ²⁹ So Brady (2005, 202) characterizes the Aristotelian courageous agent as ‘regretting death without wanting to avoid it’, and Duff (1987, 11) as ‘realis[ing], with sadness, that she must sacrifice something of great value.’ One might also read the agent’s pain more strongly, as, say, anguish at death’s prospect without this implying fear. In any event one can have a more intensely negative attitude towards death than prospective regret or sadness and still believe unwaveringly that the right thing to do is face it. It would be open to Aristotle to say that in a given situation one can fear death profoundly without desiring to avoid it. But his terminology here seems to avoid committing himself to that thought.

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

132

 

deliberating (bouleutikous)’ and as so often this is an acute observation; fear can sharpen the wits as much as befuddle them. We cannot, then, look to a supposed general inhibitive quality of fear as the explanation for Aristotle’s reluctance to permit the courageous to be afraid. This reluctance, I contend, is not explained, even on Aristotle’s own terms, by appeal to how the courageous person values things, or to what is needed to habituate him, or to what is required for him to act as appropriate. It is there, I suggest, because being afraid is a miserable thing to be. Given the ordeals that are anyway likely to be in store in the performance of one’s courageous activities, it is imperative that Aristotle emphasize that the agent can nonetheless experience those activities without fear and even find a degree of pleasure in them. In this way Aristotle can attempt to show courageous agency as compatible with the pleasantness that the virtuous person’s life must embody.³⁰

References Barney, R. 2019. ‘Becoming Bad: Aristotle on Vice and Moral Habituation.’ Oxford Studies in Ancient Philosophy 57: 273–307. Brady, M. 2005. ‘The Fearlessness of Courage.’ Southern Journal of Philosophy 43: 189–211. Broadie, S. 1991. Ethics with Aristotle. Oxford: Oxford University Press. Burnyeat, M. 1980. ‘Aristotle on Learning to be Good.’ In Essays on Aristotle’s Ethics, ed. A. Rorty, 69–92. Berkeley: University of California Press. Charles, D. 1984. Aristotle’s Philosophy of Action. London: Duckworth. Curzer, H. 2002. ‘Aristotle’s Painful Path to Virtue.’ Journal of the History of Philosophy 40: 141–62. Curzer, H. 2012. Aristotle and the Virtues. Oxford: Oxford University Press. Duff, A. 1987. ‘Aristotelian Courage.’ Ratio 29: 2–15. Garver, E. 1980. ‘Aristotle on Virtue and Pleasure.’ In The Greeks and the Good Life, ed. D. Depew, 157–76. Fullerton: California State University Press. Heil, J. 1996. ‘Why Is Aristotle’s Brave Man So Frightened? The Paradox of Courage in the Eudemian Ethics.’ Apeiron 29: 47–74. Jiang, X. 2000. ‘Courage and the Aristotelian Unity of Action and Passion.’ Philosophical Inquiry 22: 23–45.

³⁰ My thanks to the audience at the 2011 Keeling Colloquium, and especially Sarah Broadie, to whose memory this essay is dedicated, for helpful comment and discussion. A version of this paper was also read at the 2019 Keeling Graduate Conference in Ancient Philosophy. I would like to thank the audience on that occasion, especially Margaret Hampson and Katharine O’Reilly, for their questions and comments. Finally, thanks to Joachim Aufderheide and Fiona Leigh for discussion of written drafts of the paper.

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

     ’  

133

Korsgaard, C. 2014. ‘From Duty and for the Sake of the Noble: Kant and Aristotle on Morally Good Action.’ In Kant on Emotion and Value, ed. A. Cohen, 33–68. London: Palgrave Macmillan. Kraut, R. 2013. ‘An Aesthetic Reading of Aristotle’s Ethics.’ In ‘Politeia in Greek and Roman Philosophy’, ed. V. Harte and M. Lane, 231–50. Cambridge: Cambridge University Press. Leighton, S. 1988. ‘Aristotle’s Courageous Passions.’ Phronesis 33: 76–99. Mills, M. 1980. ‘The Discussions of Andreia in the Eudemian and Nicomachean Ethics.’ Phronesis 25: 198–218. Moss, J. 2012. Aristotle on the Apparent Good. Oxford: Oxford University Press. Nielsen, K. 2017. ‘Vice in the Nicomachean Ethics.’ Phronesis 62: 1–25. Pangle, L. 2018. ‘The Anatomy of Courage in Aristotle’s Nicomachean Ethics.’ Review of Politics 80: 569–90. Pears, D. 1980. ‘Courage as a Mean.’ In Essays on Aristotle’s Ethics, ed. A. Rorty, 171–87. Berkeley: University of California Press. Pearson, G. 2009. ‘Aristotle on the Role of Confidence in Courage.’ Ancient Philosophy 29: 123–37. Pearson, G. 2012. Aristotle on Desire. Cambridge: Cambridge University Press. Price, A. 2011. Virtue and Reason in Plato and Aristotle. Oxford: Oxford University Press. Richardson Lear, G. 2004. Happy Lives and the Highest Good: An Essay on Aristotle’s Nicomachean Ethics. Princeton: Princeton University Press. Richardson Lear, G. 2006. ‘Aristotle on Moral Virtue and the Fine.’ In The Blackwell Guide to Aristotle’s Nicomachean Ethics, ed. R. Kraut, 116–36. Oxford: Oxford University Press. Rogers, K. 1994. ‘Aristotle on the Motive of Courage.’ Southern Journal of Philosophy 32: 303–13. Sanford, J. 2010. ‘Are You Man Enough? Aristotle and Courage.’ International Philosophical Quarterly 50: 431–45. Taylor, C. 2008. ‘Wisdom and Courage in the Protagoras and the Nicomachean Ethics.’ In Pleasure, Mind, and Soul: Selected Papers in Ancient Philosophy, 281–94. Oxford: Oxford University Press. Vigani, D. 2017. ‘Aristotle’s Account of Courage.’ History of Philosophy Quarterly 34: 313–30. Wielenberg, E. 2000. ‘Pleasure as a Sign of Moral Virtue in the Nicomachean Ethics.’ Journal of Value Enquiry 34: 439–49. Young, C. 1977. ‘Aristotle on Courage.’ In Humanitas: Essays in Honor of Ralph Ross, ed. Q. Howe, 194–203. Claremont: Scripps College Press.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

9 Comments on Raphael Woolf, ‘Courage and Pleasure in Aristotle’s Ethics’ Sarah Broadie

I shall not attempt comprehensive comments on Raphael Woolf ’s interesting discussion: it covers much conceptually intricate ground with detailed attention to both ethical treatises. My remarks will be confined to some of his arguments about the ‘courage problem’, as we may call it.¹ This is the well-known problem of reconciling Aristotle’s general doctrine of eudaimonia with the specific realities involved in courageous action. How does Aristotle seek to resolve this tension, and is his effort successful? These are some major questions of Woolf ’s paper. But before addressing them at closer quarters, it is worth reminding ourselves that the courage problem is hardly peripheral to Aristotle’s ethics. Aristotle holds that human eudaimonia, the highest practicable human good, principally consists in the exercise of human virtue.² He therefore rejects the hedonist view that eudaimonia principally consists in pleasure. But at the same time he recognizes the pull of hedonism: the person whose life incorporates eudaimonia must, Aristotle thinks, be living pleasantly, and this pleasantness must be grounded in the nature of eudaimonia. Thus he is committed to the view that exercising human virtue is, as such, something pleasant. However, human beings, unlike the gods, have to live their lives in many different contexts; they have to respond to many kinds of demands internal and external to the individual. In each of these spheres the human being can do well or not so well; consequently (according to Aristotle) there are distinct virtues—and, of course, also vices—corresponding to the different spheres. Given this diversity, it was perhaps predictable all along that Aristotle’s general theory of human eudaimonia would conform better with some virtues than others: in particular, that our pretheoretical intuitions about different virtues would differentially support or resist the theory’s blend of abstractly aretaic with abstractly hedonic aspects. And so it ¹ The focus will be on EE III.1, 1228b4–1229a11 and Woolf ’s understanding of this passage. I have had to leave on one side for now his treatment of the prizefighting comparison at EN III.9, 1117a35–b6, and his discussion of whether Aristotle ought to recognize a kind of enkrateia corresponding to courage. ² Even if it is allowed that eudaimonia requires more than the sheer exercise of virtue (cf. EN. I, 1098a18–20; 1099a31–1099b7; X, 1177b24–5), I do not see how this helps with the difficulty. Sarah Broadie, Comments on Raphael Woolf, ‘Courage and Pleasure in Aristotle’s Ethics’ In: Psychology and Value in Plato, Aristotle, and Hellenistic Philosophy: The Ninth Keeling Colloquium in Ancient Philosophy. Edited by: Fiona Leigh and Margaret Hampson, Oxford University Press. © Oxford University Press 2022. DOI: 10.1093/oso/9780192858108.003.0009

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

   

135

has turned out. Courage is, par excellence, one of the human virtues, and courage disturbs the perhaps otherwise smooth picture. For Aristotle is surely committed to the view that human eudaimonia may, and no doubt quite often does, consist in the exercise of courage: but courage, to our pre-theoretical understanding, is a quality whose exercise costs the agent pain, anxiety, suffering, even agony. This is not a problem to be easily brushed aside. For instance, Aristotle cannot evade it by simply rejecting the deliverance of pre-theoretical understanding: as well as sitting very badly with his general approach in the Ethics, such a response would foster doubts as to whether he even means the same by ‘courage’ as the audience he hopes to reach, and whether teachings on courage in his sense can contribute to their ethical enlightenment. Nor can he plough ahead on the assumption that it does not really matter if one of the many virtues, namely courage (in the ordinary sense), fails to fit his theory: for courage is so far from being a marginal virtue that significant lack of fit between it and theory must leave theory in the ditch. Nor can he abandon the hedonic aspect of eudaimonia. The full-blown philosophical hedonism that defines eudaimonia as pleasure is an attractive rival theory with some intelligent adherents;³ Aristotle would be playing into its hands if he accepted an account of eudaimonia such that episodes whose hedonic quality is (non-accidentally) neutral or even decidedly negative could count as genuine and representative instances of eudaimonia. He would have to fight hard to defend himself from the charge—surely devastating if it sticks—of misapplying the honorific title ‘eudaimonia’ to something concerning which we can on occasion sensibly wonder whether missing it is a loss to us, and even from the charge of applying it to something perhaps only worth pursuing as a tiresome means to something else—on the pattern of all those well-known necessities we embrace that are devoid of hedonic appeal or positively repel us, such as going to the dentist! We must remember too that if, for some well-recognized type of virtuous activity, Aristotle were officially to sever the link between it and pleasantness he might lose part of his ground for proposing the activity of sophia as the most precious form of eudaimonia. For if the exercise of one paradigm human virtue is, as such, a harsh experience for the agent—the sort of thing one would like to have behind one rather than be in the middle of—then it might be wondered whether sophia, whose operation according to Aristotle is more leisurely, unharassed, continuous, and pleasant than that of any practical virtue,⁴ should really count as a serious human virtue rather than as a sort of foolish, immature, selfindulgently recreational disposition prized by eccentrics or the ethically crippled—that is, people who, to the extent that they prize its exercise, are out ³ See EN I, 1101b27–31 and X, 1172b9–25 on the hedonism of Eudoxus. ⁴ EN X, 1177a21–7; b19–24. Aristotle treads a tightrope here. His position requires that representative or typical exercises of the paradigmatic practical virtues are pleasant (or deserve to be called ‘pleasant’)—as befits instances of eudaimonia—but not as much so as corresponding exercises of sophia.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

136

 

of touch with what really matters in life.⁵ But if sophia were to be denied the status of a human virtue, the activity of it certainly could not count as any form of human eudaimonia.⁶ So a lot depends on finding a good solution to the courage problem. Raphael Woolf begins from the fact that Aristotle’s courageous person is exposed to the unpleasant and the painful in two ways: externally he is likely to face ‘some pretty unpleasant things’, and internally, as a consequence, he has to cope in some way with fear, which according to Aristotle is a painful emotion. Now the texts seem to leave it obscure exactly how Aristotle conceives of the courageous person’s relationship, when in action, with fear. Does this character undergo fear then, or doesn’t he? If fear is completely out of the picture, he is not courageous but stupidly impervious; but if fear is in the picture how does he differ from the coward? Inevitably, Aristotle states that courage is a mean disposition with regard to the continua of fear and confidence. But this does not help us to see exactly how he can coherently assert that the courageous person is ‘without fear’ (EE III, 1228b4; 7;15–17; EN III. 1115a16; 19; 33; cf. b1) even while defining him by reference to certain very fearful things, and (of course) by reference to fear for the right objects, in the right way, on the right occasion, etc. However, at least in the Eudemian Ethics, Aristotle takes steps to show that his account is not selfcontradictory. He sets up thesis and antithesis (1228b14–17), then reconciles them by means of a distinction between what is fearful simpliciter—i.e. fearful to human nature or to human beings as such—and what is fearful to a given kind of person. The upshot is that the fearful things the facing of which is definitive of courage are fearful simpliciter; but they are not fearful-to-the-courageous-person (116). Now, one might get the hasty impression that this just says that the courageous person is not afraid of things which the average human being is afraid of—a claim that is nicely coherent even though one might question it as an account of courage: must the courageous person be literally without fear, or have only minor fear (122), of the wounds and death that scare ordinary people? Woolf presses this question in Section 3 of his paper. In Section 2, however, he highlights the remarkable fact that the EE, far from presenting the courageous as unafraid of things that human beings normally fear, actually insists that the courageous person does find fearful the things that are normally fearful to human beings, even though it is also true that he does not find those things fearful, or hardly at all.⁷ For those things, Aristotle says, are fearful to him ‘insofar as he is a human being, but not fearful, or slightly (ērema) so, if at all, insofar as he is courageous’ ⁵ Cf. EE, 1215b5; EE V = EN VI, 1141b3–8. ⁶ He does not always take it for granted that sophia (as he understands this) would be considered a human virtue; see EN I, 1103a8–10, where he argues the point. ⁷ At 1228b28 Allan supplied phainetai, but Inwood and Woolf reject it (2013), in my view correctly. [Editors’ note: D. J. Allan’s supplementation was suggested to—and accepted by—the editors of the 1991 Eudemian Ethics OCT, R. R. Walzer and Jean Mingay, by way of private correspondence.]

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

   

137

(1228b7 and 29, Inwood and Woolf 2013). If the contrast here were between the courageous man and the plain—in the sense of ordinary, average—man, Aristotle would have no reason at all to ascribe the plain man’s fears to the courageous man since the courageous, morally speaking, are not in that sense plain: on the contrary, they are extraordinary. In fact, however, the contrast is between a human being as such, and one particular type of human being, namely the courageous.⁸ Since courageous Callias is a human being, it is only reasonable to ascribe to him the fear that human beings are subject to as such (in fact, it would be thoroughly unreasonable to withhold the ascription), even though (Aristotle wants to say) it is also true that, as the particular kind of person he is, namely the courageous kind, Callias is free or almost completely free of fear. But isn’t this incoherent after all? Woolf comments: Although ‘yes and no’ is a characteristically Aristotelian response to a question, it is difficult to say exactly what it means here. Presumably the courageous person cannot simultaneously fear and not fear the same thing (or fear it both slightly and considerably). He is, on the other hand, unquestionably both a human being and courageous (116).

Woolf then works to save Aristotle from inconsistency through a dual understanding of the verb ‘to fear’. Aristotle’s courageous person, according to Woolf, recognizes that what he confronts is fearful for human beings, i.e. creatures of his own kind, and so in a recognitional way he registers it as fearful for himself, and in that sense fears it. This point is important, because if it were omitted Aristotle would open himself to the following objection: if the agent’s moral character simply cancels, without qualification, the fearfulness of his situation, there is nothing for him to be courageous about and so his behaviour is not courageous. If his moral character proofed him, without qualification, against fear, he would be either a god or completely reckless. But actually he is neither of those things. Since he is a human being, questions of the fearful naturally matter to him; and since he is courageous he does not ignore those questions but gives them practiceembodied answers of a kind that would not be given by any old human being. That is to say that the fearfulness of his situation is real for the courageous person, but does not affect him as it would the run-of-the mill person. This means, Woolf argues, that the courageous man lacks or almost completely lacks the feeling or emotion of fear. And therefore the activity of Aristotelian courage is free or largely free from the pain of fear—i.e. from precisely the sort of pain that normally pervades our response to danger, and hence to situations that call for courage. ⁸ This point is slightly obscured by the fact that at 1228b31–5, which compares the courageous with the physically healthy or strong, the contrast is between the healthy or strong and ‘most people’.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

138

 

But Woolf rightly says: ‘we . . . recognize that there are people who do the right thing, for the right reason, in spite of being afraid, and we would call such people no less courageous—perhaps in a sense more so—than those who face similar situations without fear.’ Not that the presence of fear is required if the action is to count as courageous; only that its absence or near-absence is not required. Aristotle, however, on Woolf ’s reading, fails to see this or fails to acknowledge it. Why? Woolf thinks it is because of the theoretical pressure he is under to maintain intact the connection between eudaimonia and pleasure. It seems clear that for that connection to hold something must give. However, I do not think we need to attribute to Aristotle such a narrow and ungenerous notion of courage, one that automatically disqualifies those who feel fear, perhaps on several occasions, who are even almost sick with fear as they move into battle, but who nevertheless keep their heads and act entirely as they should. Let us go back to the Eudemian passage where Aristotle scotches the seeming contradiction of the courageous man being both afraid and not afraid (or hardly at all). Aristotle says that the situation is fearful to him qua human being but not to him qua the sort of person he morally is. I believe that this dialectical move is enough—or anyway Aristotle sees it as enough—to dissolve the paradox. Woolf may be right to wonder how someone can be in a condition of simultaneously experiencing fear and not experiencing it about the same thing. But it sounds less strange to say that it is simultaneously true both that the man fears X and that he does not. There is a dialectical prising apart of—as we might put it—levels on which ‘afraid’ does and does not apply to someone who exists, or is alive, on both levels: and that is all. It is unnecessary, and I believe in fact mistaken, to read the distinction as meant to convey a psychological point about what is or is not going on in the person’s mind at that moment. In short, the man can really be feeling fear as well as intellectually recognizing the fearfulness of his position. Neither of these psychological conditions, separately or together, undermines his claim to courage if he does not let either of them get the better of him. If he is afraid, Aristotle says, he is so only ērema, which literally means ‘quietly’. That is: he is not moved off course by fear.⁹ Perhaps it is not surprising that some interpreters bring to bear a distinction between fear as an effective motivator and fear as a feeling. For if we consider an actual, concrete, courageous person behaving courageously, we are happy to describe him as ‘fearless’ in light of the fact that fear does not move him to act as frightened people typically do, and also as ‘fearful’ (of the same things) in light of the fact that he is fully aware of being in an acutely dangerous situation—an awareness that may well take the form of feeling afraid. However, as Woolf observes (footnote 5), even if this contrast between fear-behaviour and fear as an internal feeling is what Aristotle should have relied on to solve the ⁹ Pace Woolf, who takes ‘slightly or not at all’ to suggest absence of feeling rather than absence of any tendency to flee.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

   

139

contradiction, there is no textual evidence that he does rely on it. Instead he appeals to a perhaps irredeemably archaic contrast between ‘fearful-to-aparticular person’ and ‘fearful simpliciter (or: ‘to human beings as such’). That both contrasts perform, in this context, the same function is not a reason to think that Aristotle, in reaching for his contrast, was somehow really reaching for the one that seems helpful and natural to us. Presumably there are two main ways of trying to get round the courageproblem: one is to ascribe to Aristotle an attenuated notion of what is painful or unpleasant about courageous action, and the other is to attenuate the sense in which the doing of virtuous action is supposed to be a pleasure to the virtuous agent. The argument from Woolf which I have discussed is an example of the first approach. The second one takes the form of understanding ‘pleasantness’ in this context to mean no more than that the virtuous agent is thoroughly identified with his actions and choices even when they involve grim and painful situations and feelings. He is completely himself in doing what he does: he is fully and freely expressing the ‘second nature’ that is more deeply and comprehensively rooted in him than all other acquired dispositions and capacities.¹⁰ It seems to me that this second approach by itself provides the least unsatisfactory solution to the courage-problem; also that those who employ the first approach are bound to resort to the second as well, so that taking the second by itself has the advantage of economy. For no one wants to say that Aristotle’s courageous agents—whether or not afflicted by painful feelings of fear—necessarily like, love, and revel in doing the things they typically do—finding them attractive, charming, and delectable. Otherwise, why would they need courage? Finally: it is unclear exactly how, without reaching for hedonic vocabulary, Aristotle would express the thought that an agent is, as we say, thoroughly identified with her or his action or activity, or—again as we would say—is engaged in it without reluctance or compunction. The most natural words for Aristotle to use here would have been hekōn, hekousios, and hekontōs, except that he himself has narrowed the existing meaning of these by restricting them to voluntary agency, so that even reluctant or grudging agents may count as acting hekontōs provided they act uncompelled and knowingly. Perhaps this piece of conceptual engineering has helped box him into the artificiality of claiming that virtuous action is necessarily pleasant. One could, with Woolf, raise a similar question about Aristotle’s dictum that education is all about getting the young to enjoy and be pained by the right sorts of things (II.3, 1104b11–13): need this mean more than ‘getting them to welcome or be keen on the things they should, and freely reject the things they should’?

¹⁰ EN I, 1100a12–19.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

140

 

References Inwood, B. and Woolf, R. (trs) 2013. Aristotle: Eudemian Ethics. Cambridge: Cambridge University Press. Walzer, R. R., and Mingay, J. (eds) 1991. Aristotle Ethica Eudemia, Oxford Classical Texts. Oxford: Clarendon Press.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

10 Memory, Anticipation, Pleasure James Warren

1. We humans are able deliberatively and reflectively to look backwards to recall past experiences and to look forwards to anticipate possible future experiences. This ability allows us to stitch our lives together across time and also to have access in the present to temporally remote parts of our lives. Memory and anticipation in the sense I mean here are to be distinguished from a more general ability to think about the past and the future. Rather, in this particular sense, they are involved in a person’s thinking about his or her own past and his or her own future.¹ By ‘memory’ therefore I mean what is variously called ‘personal memory’, ‘autobiographical memory’, ‘recollective memory’, ‘episodic memory’, ‘experiential memory’, or ‘introversive memory’.² By ‘anticipation’ here I mean just the counterpart of this sense of memory: not the ability to look into the future generally and wonder what might or might not happen, but an agent’s ability to consider, bring to mind, or think over what he or she might do and experience in the future. This might be thought to be a limited activity of a more general ability since memory in this sense is restricted to a person’s thinking of past events in his or her own life. However, memory and anticipation in this sense are also richer than the bare ability to think about the past and future. Since they are ways of looking forwards and backwards within a life and bringing to mind in the present some non-present personal event, they also allow access to a wide range of affective experiences. In very rough and ready terms, they allow us to do things such as My thanks to the editors of this volume, to Victor Caston for comments on an earlier version of this chapter, to Anthony Price for his response, and to all the participants of the 2011 Keeling colloquium. This chapter was written while I was working on what became Warren 2014 and that book presents a more developed version of these thoughts. However, this chapter presents a theme that is scattered through various chapters of that book and therefore I hope it will still be of some interest. I have not revised it in the light of more recent scholarship but I would here like to note some of the important pieces that have appeared in the intervening period, some of which take issue with my account: Meyer 2012, Moss 2012c, O’Reilly 2019a and 2019b, Sedley 2017, Tsouna 2016. See also Warren 2013. ¹ This is what makes memory interesting to people who are trying to offer an account of the criteria for the persistence of a person over time. It also makes it unclear whether memory can serve as such a criterion or, rather, is itself dependent on there being some persistent subject to prior parts of whose life memory then may give access. ² Cf. Annas 1992, 299–300; Bernecker 2010, 11–45. James Warren, Memory, Anticipation, Pleasure In: Psychology and Value in Plato, Aristotle, and Hellenistic Philosophy: The Ninth Keeling Colloquium in Ancient Philosophy. Edited by: Fiona Leigh and Margaret Hampson, Oxford University Press. © Oxford University Press 2022. DOI: 10.1093/oso/9780192858108.003.0010

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

142

 

remember pains and pleasures or anticipate joy and sadness. Our ability to think about our own past and future affective experiences allows us to do some other things too. It allows us to plan and consider how best to maximize our pleasures by thinking in a useful way about different possible future experiences. It allows us to draw on our past experiences to learn and benefit from them. And perhaps most intriguing of all, the ability to look forwards and backwards to our future and past experiences allows us to generate further affective responses in the present. We can remember and anticipate with pleasure or with pain. We can remember our pleasures with pleasure and be pained when we anticipate pains. I have no intention of offering here my own account of what is involved when we say that we remember an experience with pleasure or that we look forwards to an experience with trepidation.³ Nevertheless, that there is such a phenomenon is itself not a trivial observation and it attracted the attention of thoughtful ancient writers too. Here I want briefly to explore some of what they had to say about it. I suggest that we can distinguish in their writings two broad ways of thinking about the fact that we can take pleasure and pain in our memories and in our anticipations. On the first model, anticipating and recalling are thought to be means of reaching out to the past or future and hauling some temporally remote experience from there into our present. Within this model, we can identify further two ideas. The first idea is that this ability to set together a present with a non-present affective state allows an agent to arrange some kind of comparison between the present affective state (pleased, pained, neither pleased nor pained) and the anticipated or recollected state (pleased, pained, neither pleased nor pained). The comparison between the two is then noted and used to draw various further conclusions, for example about the nature of pleasure and pain themselves, or this particular person’s consistency of character and the like. The second idea is that the recollected or anticipated pleasure can be used to help to improve one’s state in the present by allowing us to ‘relive’ or ‘pre-live’ a pleasure. For example, the Epicureans claim that recollecting and thereby reliving a past pleasant experience is useful in counteracting a present pain. The second model is a less common approach and is perhaps best illustrated by contrast with the dominant form. In brief, unlike its counterpart, this model does not assume that an experience that was painful to us in the past will always be painful when we remember it. Sometimes a past painful experience can be recalled with pleasure. What is more, the pleasure we may feel in recalling that past painful experience is not simply because, when placed in comparison with our present situation, that past experience is merely revealed not to have been as bad as we

³ Such an account would need to build a story about affective content into a general account of introversive memory or anticipation. The analysis in Bernecker 2010, ch. 8, offers some helpful steps in this direction.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

, , 

143

once thought. Rather, we can recall even with pleasure something that was genuinely painful at the time. While this picture is less common in the ancient texts, we can detect signs of it in Aristotle’s discussion of pleasure and memory in Rhetoric I.11.

2. Let us first look at a familiar text which highlights the importance of this ability to remember and to anticipate. Indeed, it insists that the presence of these abilities is an essential part of our living a recognizably human life. The text in question is Plato, Philebus 21a–d. Protarchus concedes early in the discussion that the primary focus of his and Socrates’ attention should be on the ‘mixed’ life: a life combining the activities of reason and pleasure (22a). Further, he also accepts that a life without any rational activity at all would not be choiceworthy. This is the famous rejection of the ‘life of a mollusc’ at 21a–d. Socrates sums up such a life at 21c1–8: Indeed, without memory it is impossible to remember that you felt pleasure in the past, since as a pleasure falls away into the past not even the slightest memory of it remains. And without true opinion, then you would not think truly that you are experiencing pleasure when you do. And without reasoning, it is impossible to reason how you will experience pleasure in the future, living not a human life, but the life of a sea-slug or some other of the many shelled sea-creatures. (21c1–8)

At the very beginning of the dialogue Socrates listed a number of things he considered better than pleasure for anything able to share in them. The list at 11b6–c2 included thinking (phronein), understanding (noein), remembering (memnēsthai), correct belief (orthē doxa), and true calculations (alētheis logismoi). Those items—memory (mnēmē), opinion (doxa), reasoning (logismos)—return here at 21c together with understanding (nous) as a list of the capacities missing from a life without phronēsis (cf. 21b6–9) and all are allowed to be present in the other candidate life: the life devoid of pleasure (21d9–e2). Together, as 21c1–8 makes clear, they allow an agent to consider and reflect upon pleasures that are being experienced, have been experienced, and will be experienced in various stages of his life: his past, present, and future. And, in preparation for his argument for false pleasures, Socrates will make a point of reminding Protarchus at 39c–d that pleasures and pains can also apply in all tenses. In the extreme case of the mollusc life, the complete absence of such capacities makes it impossible for a mollusc to form the thought ‘I was pleased that P’. In the present, it prevents a mollusc thinking ‘I am pleased that P’. For the future, the absence of

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

144

 

rational capacities makes it impossible for the mollusc to form the thought ‘I will be pleased that P’. Perhaps molluscs have no shell-fish analogue for the scribe and painter we are asked later in the dialogue to imagine at work in a human soul. This means that a mollusc cannot engage in any of the activities of recollection or prediction which are integral to human desire and our human experience of pleasure. At this very early stage of his discussion with Protarchus, when trying to extract a commitment to the mixed life and the rejection of the mollusc life of pleasure alone, Socrates leaves open the question whether these judgements about one’s experience can themselves be pleasures, sources of pleasure, or means to further pleasures, although that is precisely the issue which he wants to explore in the later discussion of false pleasure. He also leaves open the question whether some of these capacities are present in some other non-human animals. Later he insists that memory, for example, does have a role in non-human animal desire as well as human desire.⁴ For the moment, Protarchus recognizes that a life without these capacities, however pleasant, is certainly not a human life and not a life we would choose to live.⁵ That is enough to allow Socrates to embark on the major task of the dialogue, namely the elaboration of the kind of mixture of pleasure and reason that a good human life ought to be. There is more to be said about the complicated psychological picture which Socrates assembles, including its famous depiction of an internal scribe and an internal painter who have the jobs of composing and then depicting these logoi ⁴ See 35c9–10 and 36b8–9. At 22b3–8 Socrates repeats that neither of the two lives contains the good by itself and neither is choiceworthy for a human (ἡμῶν 22b6). However, at 22b4–6 (cf. a9–b2) he adds that if either life were to contain the good then it would be sufficient, complete, or choice-worthy also for any animal or plant capable of always living a life of that kind. This is sometimes thought to be an odd addition (see Gosling 1975 ad loc.). The proviso: ‘for those things capable of always living such a life’ restates the qualification made twice by Socrates at the beginning of the dialogue when declaring his initial disagreement with Protarchus (11b6–c2). Cf. Frede 1997, 177 n.111; Delcomminette 2006, 178–9. ⁵ Cf. Nussbaum 1995, 98–102. An alternative interpretation makes the absence of rational capacities prevent something like prudential reasoning or the use of a hedonic calculus and therefore the absence of rational capacities diminishes the pleasure of the life being considered. The diminished pleasure in turn leads Protarchus to see the necessity—if only in an instrumental sense—of reason for a good life, even if a good life is understood in hedonist terms. Concentrating on the mollusc’s inability to form beliefs or judgements about the future, this view imagines that such a life is rejected because it cannot accommodate the possibility of forming plans about how to set about attaining some future pleasure. For: μηδ’ εἰς τὸν ἔπειτα χρόνον ὡς χαιρήσεις δυνατὸν εἶναι λογίζεσθαι at 21c5–6 translators offer: ‘you couldn’t even calculate that you would enjoy yourself later on’ (Hackforth 1945); ‘lacking the ability to predict you would be unable to predict your future pleasures’ (Gosling 1975); ‘being unable to calculate, you could not figure out any future pleasures for yourself ’ (Frede 1993). Frede’s ‘figure out’ is perhaps the closest to implying that what is being considered is the capacity to plan for and arrange for future pleasures rather than the simple consideration of what pleasures there might be in the future. Unfortunately, the summary of this argument at the end of the dialogue (60d3–e5) refers only to the recognition of present and recollection of past pleasures. (See also Moore 1903, III §52, who interprets the message of the argument to be that it is the ‘consciousness’ of a pleasure that is valuable rather than the pleasure itself. His translation of 21c5–6 has: ‘you cannot even have the power to reckon that you will be pleased in the future’.) See Evans 2007 for an excellent discussion of this passage.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

, , 

145

respectively, and we will return to it shortly. But for now I want to draw a relatively simple point from this passage. The picture of human psychology which emerges both from the rejection of the ‘mollusc life’ and also from the later extended analysis of the psychology of desire gives a prominent place to the capacities of memory, anticipation, and hope. Importantly, the picture which Socrates assembles depicts memory and hope working in combination by making memory play an important role in future-directed attitudes such as desire in general and—in humans more specifically—hope and reflective aspiration. In brief, living a human life involves the use and combination of attitudes to the past and the future since humans are animals who are aware of their living through time; humans are aware, to put it most concisely, of living a life. This temporal perspective has to be taken into account since the entire dialogue is, after all, meant to determine which ‘state or condition of the soul is able to provide all people with a happy life (eudaimōn bios)’ (11d4–6) and a life is something that extends over time. A good human life, evidently, is one in which the person living the life is able to take into account his own past and future in an appropriate way.⁶

3. The recognition of this ability to recall and anticipate experiences of pleasure and pain is put to use not only in the Philebus but also elsewhere in Plato, particularly in the Republic and the Protagoras. In the Protagoras, Socrates sets about arguing for the necessity of what he calls an ‘art of measurement’ (tekhnē metrētikē): a skill at assessing accurately the size of a future pleasure or pain despite the distortions of perspective that are generated by relative temporal distances from the present. A skilled practitioner of this art will be able to gauge without error whether some potential pleasure available in a week’s time is larger, smaller, or equal to, an alternative potential pleasure available tomorrow, mindful that, just as larger buildings far away may falsely appear to be smaller than what are in fact smaller buildings close at hand, we need to be careful to avoid being misled by mere appearances as to the truth of the respective values of the alternative goods. Here, the conception of the art of measurement is put in the service of a larger project to argue for the unity of the virtues and much of the interpretative frustration of the dialogue is generated by a surprising silence on Socrates’ part as to the nature of pleasure and pain themselves. All manner of concerns might be raised, not only about the

⁶ Cf. Russell 2005, 198: ‘The mistake that worries Plato about this way of valuing pleasure is a mistake about self-conception: what matters in life is the way in which one intelligently constructs a life, a future, and a self by one’s actions and goals, in a way that will fulfil one’s deepest needs as a human, and the view that pleasure makes one’s life happy cannot make sense of that.’

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

146

 

justification for the apparently strong thesis of temporal neutrality for the assessment of goods and evils but also for the plausibility and indeed desirability of the associated claim of global prudential reasoning. Fortunately, having already asserted what I know are a number of controversial claims about this section of the Protagoras I can move on leaving all these questions unexplored, because all that matters for present purposes is that here in the Protagoras we are presented with an analysis of prudential reasoning and anticipation which uses an intentionally simple comparative model of pleasures and pains. Socrates advises the prudent calculator to overlook the respective temporal distance of the possible pleasures and pains; only then can they be accurately and appropriately compared and ranked.⁷ Socrates must assume therefore, although he does not explore the issue in any depth, that his ideal agent will be able accurately and reliably to anticipate and evaluate the pleasures and pains consequent on particular choices. And the correct and artful application of this ability is supposed to be necessary in maximizing an agent’s pleasure over the course of a life. In Republic IX (583b–584b), Socrates uses our ability to compare our present hedonic state with some past or future state as part of an argument for there being an intermediate state of feeling neither pleasure nor pain—a state which we may sometimes mistakenly take to be pleasant or painful. These comparative assessments can therefore be wrong; sometimes we may be misled by, for example, a current state of pain to think that a past state—which was in fact merely a state of neither pleasure nor pain—was extremely pleasant. The details of the argument Socrates mounts are complex but it is important to recognize that he is here interested not in our ability to conceive of future states of pleasure or pain simply as part of a process of prudential reasoning. He is interested instead in the fact that we can look forwards to various possible future states of pleasure and, rather than evaluate them against one another as is envisaged in the Protagoras, we can set them against our current state and draw inferences from the comparison between the two. These inferences may lead us to revise how we evaluate our current state by encouraging us to see that the present state is not as bad or not as good as we assumed.⁸ In the main, these two dialogues deal only with (i) our abilities to plan and remember and (ii) our opportunities for comparing and contrasting our present state with a past or future state. They leave aside perhaps the most intriguing and complicated aspect of our abilities to remember and anticipate, namely our ability to take pleasure or be pained in the present moment of recollection or anticipation ⁷ Since he is considering planning and decision-making, his account considers only future-directed thinking. There is nothing here about comparatively ranking past pleasures and pains, although a very strong form of temporal neutrality might indeed encompass these too. See, for example, the discussion of temporal neutrality and whether we might give weight to our past desires in Parfit 1984, 149ff. and Persson 2005, 211ff. ⁸ For a more detailed account of this passage see Warren 2011.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

, , 

147

by what we are thinking about. For example, the Protagoras considers how we might look ahead and evaluate different possible future pleasures but it does not explicitly consider how the thought of some future pleasure might itself generate pleasure in the present. The Republic considers how thinking about a prior state of health may make us believe that it was in fact pleasant but it does not consider how the memory of being healthy might itself bring pleasure to an ailing patient. For a more sophisticated account of the affective aspect of remembering or anticipating we need to look again at the Philebus. We have already seen that Socrates wins Protarchus’ assent to the notion that the ability to consider temporally remote experiences is a necessary characteristic of a human life. Later in the dialogue Socrates builds a detailed picture of human psychology which gives a central role to memory and anticipation. These two are closely linked in Socrates’ analysis of human psychology and, as we have seen already in our brief discussion of the life of the mollusc, Socrates is concerned to stress both the backwards- and forwards-looking temporal aspects of our mental lives. Indeed, perhaps the intense interest in the example of false anticipated pleasure has led to a comparative neglect of the important role played by retrospection and memory in the Philebus’ account of human psychology. It is clear, nevertheless, that a notable feature of the picture as it develops in the run-up from 33c onwards to the discussion of false pleasures is the prominent role granted to memory even in the analysis of desire in general and hope in particular. Prospective and retrospective faculties are therefore not only both stressed as essential characteristics of human psychology in general; they are also both involved, we are told, in desire— something that might at a cursory glance seem to be an exclusively future-directed attitude. Socrates himself seems preoccupied with arguing for a division between the roles of the body and of the soul in desire, but as he does so he states clearly that he thinks all desires and impulses which initiate a drive for such removal or replenishment involve some sort of memory (35c–d).⁹ The discussion of memory is introduced at 33c5–6. Socrates is already preparing the ground for distinguishing a class of pleasures that belong to the soul alone which will then be used in his argument for the existence of true and false, pure and impure pleasures. These pleasures and pains of anticipation are agreed to belong to the soul alone, without the involvement of the body (32c3–5). Socrates next asserts that such pleasures of the soul alone come about through memory and ⁹ For further discussion and comment see Russell 2005, 177–9; Lorenz 2006, 101–7; Delcomminette 2006, 330–45; and Harte 2014. Cf. Sidgwick 1907, 141–2 (II III §4), who makes this use of memory part of all anticipations of pleasure, including those necessarily for performing a hedonic calculus: ‘In estimating for practical purposes the value of different pleasures open to us, we commonly trust most to our prospective imagination: we project ourselves into the future, and imagine what such and such a pleasure will amount to under hypothetical conditions. This imagination, so far as it involves conscious inference, seems to be chiefly determined by our own experience of past pleasures, which are usually recalled generically, or in large aggregates, though sometimes particular instances of important single pleasures occur to us as definitely remembered.’

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

148

 

then notes that to make clear what he means he will need to say something about memory itself. In the next few lines he wins Protarchus’ agreement to the claim that memory is a kind of preservation of perception (34a10–11) and that we ‘recollect’ when, without the involvement of the body, we retrieve something that the soul originally experienced together with the body (34b6–8). Socrates then turns to the nature of desire and, perhaps surprisingly, it turns out that memory plays an important part in this too. Specifically, the memory involved in desire is a memory of the state opposite to that in which the animal currently finds itself. The desire involved when a person is thirsty, for example, involves not only the bodily painful lack of hydration. It also requires the presence of the memory of the state of not being thirsty. Presumably, the drive to find a drink to remove a thirst involves conjuring from memory some appropriate representation of the proper object of desire or perhaps of the proper state of that desire being fulfilled. Back in 32b9–c2 Socrates had asserted that the anticipation (prosdokēma) in the soul alone of a pathos of pleasure is itself pleasant and that the anticipation of something frightening and painful is itself distressing. Now Socrates combines this with his account of memory to give a full account of the pleasures and pains involved when an animal desires something.¹⁰ Socrates distinguishes two cases involving a person who is in pain but can remember the pleasant things he lacks and who thereby desires whatever will replenish this lack. In the first, the person concerned is in pain but also has a ‘clear hope’ (elpis phanera 36a8) of attaining what he lacks.¹¹ In this case, the ‘hope’ is derived from a memory of the state of being without pain, and Socrates states that it is this memory which provides pleasure while the person is also experiencing pain (36b4–6, cf. 47c5–7). His memory allows him to think about an object of desire that he lacks now but ¹⁰ Note that the analysis of desire is intended to cover in general terms not only humans but also other animals which have the appropriate abilities of perception or memory. See e.g. Phileb. 35c9–10 and 36b8–9; this is emphasized by Lorenz 2006, 102. (The mollusc of 21c which has no capacity for memory at all will presumably not be able to form a desire.) This leaves open whether some aspects of humans’ experience of pleasures and pains from memory and anticipation are specific to them in so far as they also have rational capacities. ¹¹ Note that Socrates puts his point here in terms of a case concerning ‘one of us’ (tis hēmōn), which suggests that this is a phenomenon restricted to human animals, although he later seems to include other animals too (36b8–9). The force of the qualification phanera at 36a8 is unclear (and the qualification is omitted in 36b4). Most likely, it denotes that to the hoper, as it were, the hope is clear and vivid. That clear and vivid character of the hope is what allows it to be a source of pleasure even though the hoper is also in pain; hope is the expectation of a future pleasure. And the clear and vivid character of the hope is independent of whether in fact what is being hoped for is likely to be attained. There can be a very vivid and arresting sort of hope for something extremely unlikely to come to fruition. What matters is only that the hoper is sufficiently convinced that it is. (The same might be said of despair: I might have a ‘clear’ desperation also in cases where I merely think that what I need is unlikely to come my way.) Socrates does not elaborate on this, but gives only a contrast between someone with the ‘clear hope’ of fulfilling a lack (36a8) and someone with no hope at all (36b11). And perhaps that is not a surprise. After all, it is in the next four pages or so that Socrates turns to outline to Protarchus that there is a very important distinction to be made between the pleasures to be had from such hopes that are true and those that are false, although they may all seem pleasant enough to the hoper.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

, , 

149

expects to attain and enjoy in the future. In the second, the subject is in pain but also thinks that there is no hope of replenishment. His suffering is two-fold (36b11–c1): he is not only pained by a present bodily lack but also pained by his despairing (anelpistōs ekhein) of attaining again the remembered state of painlessness; he not only has pain in the present but expects pain in the future too. His memory allows him to think about the object of desire that he not only lacks now but also thinks he will lack in the future. Later in the dialogue Socrates emphasizes the idea that ‘hopes’, like opinions more generally, are logoi—specifically logoi concerning the agent’s future pleasures—and therefore might be evaluated in terms of their truth value (see e.g. 39e6–8, 40a3–7, cf. 12d3). Here, however, he is more interested in showing that ‘hope’ (elpis) is one of the two species of ‘expectation’ and has an affective aspect. Hope is an expectation of a future pleasure and therefore is itself pleasant, while fear is an expectation of a future pain and is therefore itself painful.¹² This section makes two important claims, the combination of which offers an attractively simple model for understanding the pleasures and pains of memory and anticipation.¹³ First, recollection and anticipation, at least for these purposes, are treated as symmetrical capacities differing in their temporal direction: one looks forwards in time while the other looks backward.¹⁴ The second claim is about the affective aspect of anticipation and recollection. Whether an act of anticipating or recollecting some experience is pleasant or painful is primarily dependent on whether the anticipated or recollected experience—call this the ‘object experience’—is itself pleasant or painful. This simple picture is complicated in the text by Socrates’ interest in analysing desire, which combines these forwardlooking and backward-looking capacities, and in giving an account of the affective aspects of desire, making clear the role played by both these capacities. The important nuance introduced by the notion of ‘hope’ and its counterpart ‘despair’ should not be overlooked: Socrates is surely right to insist that while it might in ¹² It is clear from 36c10–11 (cf. 40e2–4) that Socrates thinks there can be true and false fears, as well as true and false opinions and anticipations (prosdokiai: this latter may here stand for what he earlier classified as hopes, which alongside fears are two species of anticipations). We can therefore infer that there are false pains of anticipation as well as false pleasures, although it is only the pleasures that Socrates wants to examine in depth as part of his account of a good life. ¹³ The afterlife of this model is long and complicated; it is obviously connected with the more general history of the notion of mental states as a kind of internal theatre. For a brief account of the presence of this notion in early modern philosophical accounts of memory and its modern critics see Warnock 1987, 15–36. ¹⁴ This supposed symmetry is worth considering more closely and perhaps ought to be qualified. Cf. Persson 2005, 214: ‘But then ‘(imaginatively) looking backward’ to an experience takes the form of imagining being at some time(s) in the past and looking forwards to the experience, seeing it in a future which stretches out from that point. So we do not from the present point spontaneously look backwards to experiences in the past, as we look forwards to them from the present time—though, with great effort, we could accomplish a parallel instance of backward-looking in which events are experienced in reverse order and as stretching out backwards from the present. What we call looking backwards to an experience is normally looking forwards to it from a point in the past that we imaginatively take up.’

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

150

 

general be pleasant to recall a past pleasant experience, in cases in which we lack and desire to return to that recollected state it is important to distinguish between expecting fulfilment (in which case there is a pleasure in thinking ahead) and despairing of fulfilment (in which case there is additional pain in such a thought). The appeal of this general model is bolstered in the Philebus by the picture Socrates assembles of the nature of anticipation and recollection themselves and it is at this point that we might detect a move towards specifically human capacities beyond those shared with other animals. Memory, we noted, is the ‘preservation of a perception’, a notion at work also in other dialogues, but which receives relatively little elaboration here.¹⁵ The nature of anticipation, on the other hand, is explored in more detail. At 40a3–4 Socrates asserts that ‘every human (pas anthrōpos) is full of many hopes’ and these hopes in us humans (hēmōn 40a6) are then quickly agreed to take the form of statements (logoi 40a6) and ‘painted images’ (phantasmata ezōgraphēmena 40a9).¹⁶ This characterization makes use of the idea of there being a pair of craftsmen in the soul: a scribe and a painter (38e12–39c3). The combination of these two allows Socrates to account for what he takes to be two crucial elements of our psychology. The presence of a scribe—a creator of logoi—allows Socrates to restate the idea also found in other dialogues that thinking in general can be understood as a kind of internal speech: when we judge or believe something we are in effect saying something to ourselves.¹⁷ Here we are inscribing things in our soul, which is likened to a book, an elaboration that perhaps allows Socrates also to convey the idea that these logoi are often preserved in some way as well as the idea that there is a propositional content to the items in question (38e12–13). The presence of the painter who depicts these logoi allows Socrates to account for the fact that our thinking often proceeds in terms of images and can be imagined as a kind of internal viewing (39b9–c2); we generate internal pictures that depict the internal logoi and then, as it were, view them.¹⁸ These logoi, written by the scribe and depicted by the painter, can be in any tense (39c10–d5) and therefore will include memories and anticipations too. Finally, we must turn briefly to the notorious example that Socrates uses in his attempt to persuade Protarchus of the existence of ‘false pleasures’. The example is a case of someone anticipating some pleasant future experience and taking pleasure in that anticipation:

¹⁵ See e.g. Theaet. 191c–196c for the suggestion that memory involves the reception and preservation of perceptions and thoughts in the soul just as a wax block may receive and retain the impression of whatever is pressed on to it. Cf. Sorabji 2004, 5–6; Delcomminette 2006, 321–4. ¹⁶ I take this to be another signal that Socrates has moved from the account of desire that covers all animals to the discussion of a capacity present in humans alone and which is particularly relevant for understanding the kinds of pleasures proper for a good human life. ¹⁷ See Theaet. 189d–190a, Soph. 263d–264b, Tim. 37b. For suggestions that the model is also at work in earlier dialogues see Sedley 2004, 129–31. ¹⁸ Cf. Delcomminette 2006, 375–83; Moss 2012a, 265–7.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

, , 

151

And someone often pictures himself coming into the possession of a large amount of gold and many pleasures as a result. And also he contemplates himself in this internal picture taking particular pleasure in the situation. (40a9–12)

The central idea seems to be that in anticipating a pleasant experience we take pleasure in it as a kind of advance instalment of the pleasure to come. It is because the future event—the object experience—will be pleasant that we now take pleasure in anticipating it. Unlike the previous cases of hoping when thirsty for a cold drink, the situation is simpler: there is no simultaneous bodily pain here. But the point Socrates wishes to draw is novel. In cases in which the future event will not in fact turn out to be as we currently anticipate then we might well think there is something suspect about the present pleasure we take in its anticipation. Socrates is inclined to think that in such cases we should say that the present pleasure is false. Regardless of all the familiar difficulties in interpreting and then evaluating just what Socrates is proposing, here we appear to be presented again with a relatively simple model of anticipated pleasures and pains. The pleasure that a person takes in imagining himself enjoying some future windfall is apparently an ‘advance instalment’ of the pleasure of getting that windfall, as if our capacity for anticipating can reach forwards in time and in conjuring up the image of the experience—including, importantly, the image of the person himself enjoying that experience—cause the person to feel in the present what he expects he will feel later. We anticipate the pleasure or the pain and either take pleasure in advance’ (prokhairein) or ‘are pained in advance’ (prolupeisthai 39d1–5).¹⁹ And although Socrates is not interested in spelling out an example of the symmetrical capacity of logismos for recalling past experiences, we can plausibly imagine that he would offer the same story for these too. Our memories, for example, function as store-houses of pleasures and pains. We humans can choose to go back and draw from the pleasures, calling both the past experience and the past pleasure to mind. And when we do so we ‘take pleasure in retrospect’ at bringing those experiences and past pleasures to mind: we picture ourselves enjoying some past

¹⁹ For this interpretation and the notion of an ‘advance instalment’ see Harte 2004, 125–6. The idea of an advance instalment does raise the question whether the present experience—the anticipatory pleasure—necessarily diminishes the future—anticipated—pleasure (as an advance instalment of my wages will diminish what I receive later). It is not clear what Socrates’ view is on this because his emphasis is simply on the way in which memory and anticipation can call up in the present pleasures or pains at the thought of past or future experiences but since it is possible to experience false anticipatory pleasure in a case in which there will be no anticipated pleasure, it seems that anticipatory pleasure cannot be a part of the anticipated pleasure. See Evans 2008 esp. n.16 for his criticism of Harte’s view; he nevertheless notes that the ‘advance instalment’ view ‘gives Plato a way to explain the qualitative similarity between A-pleasures [pleasures the anticipator expects to experience] and B-pleasures [pleasures that the anticipator experiences by expecting to experience pleasures]’. The relationship between the anticipated pleasure and the present anticipatory pleasure is the subject of more interest in the discussion between Epicureans and Cyrenaics we will consider below.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

152

 

experience and take pleasure in that recollection. Presumably, were we to recall a memory of a painful past experience this would cause us pain in retrospect in a similar way. The most important claim built into this way of considering the effects of recalling or anticipating such experiences is that recalling a painful event, for example, is in effect a way of ‘reliving’ that event together with the pain associated with it. We can similarly say that this model will consider anticipating an experience as a way of ‘pre-living’ that experience: taking an advance instalment of the pleasure or pain that we imagine it will bring. This model is used by Socrates to persuade Protarchus that there are such things as false pleasures since Protarchus agrees that there are people who will experience such ‘pleasures in prospect’ at the anticipation of events either that will not in fact take place or—on an alternative and to my mind generally more persuasive interpretation—that when they do occur will not be pleasant as they were imagined to be.²⁰ Socrates is not particularly interested in the other important aspect of this human ability, namely the fact that we can choose whether and when to engage our powers of recollection or anticipation. But we can indeed— perhaps with some constraints—exercise some control over what we recall or anticipate. So we can choose to think back to some pleasant experience in the past, perhaps in the hope of reliving the pleasure we experienced at the time. And we can choose to think of some future pleasure, taking an instalment of that pleasure in advance, as it were, by calling it to mind in the present. Although they do not form part of Socrates’ discussion in the Philebus, these two helpful possibilities are what the Epicureans appear to have in mind when they assert the great power of memory and anticipation to counteract and overcome any unfortunate present pains. And they stand as one of the important points of disagreement between the Epicureans and their hedonist rivals, the Cyrenaics. Their disagreement will highlight in more detail the relationship between the ‘object experience’ and the pleasure or pain experienced in anticipating or recollecting it.

4. The Epicureans made significant use of our ability to recall past pleasures in their account of a good and pleasant life since they asserted that we can in this way offset any present physical pains and guarantee a state of painlessness. Famously, Epicurus himself is supposed to have said on his very last day that despite serious physical distress, he nevertheless was living painlessly because he was able to recall

²⁰ For the former interpretation see e.g. Hackforth 1945, 72–3; Penner 1970; Frede 1985 and cf. Williams 1959; Gosling 1975, 111–12. For the latter see e.g. Lovibond 1989–90; Harte 2004, 120–8; Russell 2005, 179–82 and n.26.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

, , 

153

pleasant philosophical conversations he had previously had with his friends.²¹ What Epicurus manages to do is not merely to distract himself from his current sickness. Rather, he claims to be able to summon up genuine psychic pleasure (to kata psykhēn khairon: DL 10.22; laetitia animi: Cic. Fin. 2.96) which counterbalances the physical distress. Recalling previous pleasant experiences is therefore thought to be a way of generating present pleasure, presumably by reliving in some way those past experiences.²² In large part, the Epicurean emphasis on the power of memory and anticipation appears to have been a response to the charge that, in identifying pleasure as the good, they will have to admit that any goods we experience are merely temporary: pleasure simply ‘flows away’ once it has been enjoyed. If that were true, then it would jeopardize the Epicurean claim that a sage can guarantee stable and lasting happiness given that physical pain is to some extent an unavoidable fact of human life.²³ Recollection, however, is for the Epicureans the capacity that allows us to recall and relive past pleasures, suggesting that these pleasures are never entirely lost and, moreover, can be used to counteract any physical pain we encounter. This claim has never been found particularly plausible. Cicero launches an attack on the idea at Fin. 2.104–6 arguing, first, that memory is not in our power to the extent that the Epicureans claim; some things we would like to forget but cannot and some things we wish we could remember but cannot. Second, he argues that there are some pains that we ought not to forget, either because they were the result of virtuous deeds or because it might even be pleasant to remember past sufferings. Here, Cicero offers his Latin version of a verse from Euripides’ Andromeda that we shall later see used also by Aristotle in his account of memory, pleasure, and pain (below §5): ‘pleasant is the memory of past sufferings’.²⁴ Presumably, the phrase is sufficiently proverbial that we need not think of any direct connection between the Ciceronian and Aristotelian texts but it is possible that Cicero has taken the fragment from a critical discussion of the Epicurean view in an Academic source.²⁵ Another critic—Plutarch—puts his dissatisfaction with the Epicurean view in terms that point to a more serious disagreement: But when you hear their loud protest that the soul is so constituted as to find joy and tranquillity in nothing in the world but pleasures of the body either present or anticipated, and that this is its good, do they not appear to you to be using the ²¹ See DL 10.22 for Epicurus’ Letter to Idomeneus and compare the Latin version at Cic. Fin. 2.96. ²² For a recent discussion see Giovacchini 2007. ²³ See also Cic. Tusc. 5.95; August. Serm. 348.3. ²⁴ suavis laborum est praeteritorum memoria (= Eur. fr. 133 Nauck²). Cicero comments that ‘you all know the line in Greek’ (nostis omnes). Pace Madvig 1876 ad loc., this might mean: ‘All you Epicureans’ know the line, since then Cicero can make a subtle jibe against the Epicureans: they remember this bon mot about memory but fail to recognize it being a counter-example to their theory. ²⁵ The same fragment is cited at Plut. Quaest. Conv. 630E.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

154

 

soul as a decanter of the body, and to imagine that by decanting pleasure, like wine, from a worthless and leaky vessel and leaving it to age in its new container, they are turning it into something more respectable and precious? Yet there is a difference: the new vessel preserves the wine that has settled in the course of time and improves its flavour, whereas in the case of pleasure the soul takes over and preserves the memory of it, as it were the bouquet, and nothing else; for the pleasure effervesces in the flesh and then goes flat, and what is left of it in recollection is faint and greasy, as though a man were to lay away and store up in himself the thoughts of yesterday’s food and drink, resorting to these, we must suppose, when nothing fresh is at hand. (Non Posse 1088E–1089A (trans. Einarson and De Lacy))

Note the insistence that although a good storage jar can preserve and even enhance the substance and flavour of a wine, memory is able to preserve only the mere bouquet, as it were, of the past pleasure. The bouquet of a wine is, of course, related to the full taste of the wine; sniffing a decanter or a recently pulled cork will produce an experience that, despite lacking the depth and richness of the full range of sensations involved, may capture part of the experiencing of drinking the wine itself. But even so this will fall far short of the full experience of drinking the wine. Plutarch’s criticism is well aimed and it is easy to think of other similar examples. Consider the pleasure of sitting outside on a lawn on a warm day. Later in the year, perhaps on a cold November afternoon, I might think back and recall that past experience. It might be pleasant to recall it; perhaps the thought might lift my autumnal gloom a little (unless I also reflect upon how long ago that was and how long it will be before I feel that warmth again). But it certainly will not generate on that cold November afternoon the pleasure of a sunny summer’s afternoon even if, perhaps, I have a ‘clear expectation’ that I might feel such a pleasure soon enough.²⁶ The spirit of Plutarch’s criticism would be warmly endorsed by the Epicureans’ hedonist rivals, the Cyrenaics. The Cyrenaics recommended that we should enjoy each pleasure ‘as it comes’ (DL 2.91), a claim which is taken in the sources to be aimed at the Epicurean claim that pleasure can be experienced in the present as a result of anticipating some future pleasant event or recalling some previous pleasant event. (DL 2.89 makes an explicit contrast with the Epicureans.) The Cyrenaics counter by insisting that pleasure is ‘unitemporal’ (mounochronos: Athenaeus 12, 544A–B; cf. Aelian Varia Historia 14.6). The best interpretation of this claim recalls the foundational assertions of Cyrenaic epistemology.

²⁶ Cf. Cic. Tusc. 5.73–4 for a similar point made from the perspective of someone more used to suffering hot Italian summers: ut si quis aestuans, cum vim caloris non facile patiatur, recordari velit sese aliquando in Arpinati nostro gelidis fluminibus circumfusum fuisse; non enim video quo modo sedare possint mala praesentia praeteritatae voluptates.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

, , 

155

Pleasure and pain are pathē and therefore are constituted by a particular interaction between a perceiver and a given object. A particular pleasure is therefore both private and tied to a specific such interaction: in the absence of either the perceiver in this particular state or the object in this particular state that particular pleasure cannot be experienced. Pathē are therefore unrepeatable. For example, if I take pleasure at the thought of enjoying opening my birthday gifts later this year, I am not taking an advance instalment of the pleasure I will experience on that day since my pleasure when I do open the gift will be a pathos constituted by an interaction at that time between me and the various objects I perceive. Any pleasure I take in the anticipation of that event, on the other hand, will be a distinct pathos, constituted by a different interaction. It is not clear just what the two participants in that interaction are that constitute the pathos of pleasure as I anticipate a future pleasant experience. Perhaps we are invited to think once again in terms of the Philebus’ picture. When someone anticipates feeling pleased, this is an internal viewing of an internal depiction of the event in question and that the event will include a depiction of the person’s affective response at the time of the anticipated event. In that case the new pathos will be generated internally as a result of the consideration of that internal depiction. What I view in anticipating the birthday present is ‘me enjoying opening my present’ and I experience a pathos in considering that depiction of the event. However it is generated, the pathos of pleasure when a person anticipates or recollects occurs only at the time of the anticipation or recollection itself and its identity is dependent on that condition of its occurrence. To enjoy anticipating a pleasure is not to receive an advance instalment of some pathos yet to come and, similarly, to enjoy the recollection of a pleasure is not to receive some recovered instalment of a past pleasure.²⁷ (As reported at DL 2.89, the motion of the soul ‘dissipates’ over time.) Rather, if anticipation or recollection is accompanied by pleasure then this is a new pathos to be identified with some presently obtaining psychic motion: pleasure and pain must always be present pathē generated by something present. The Epicureans insist on the efficacy of recalling past pleasures as a means to mitigate present pains. The Cyrenaics think that this is not likely to be effective even if a ‘recollected pleasure’ is not a revived past pathos but is rather a brand new present pathos. However, there is an important role for anticipation in Cyrenaic psychology and ethics which might at first glance appear to be in tension with their criticism of the Epicureans. The Cyrenaics think that it is possible to ready oneself against likely future pains by a form of ‘pre-rehearsal’ of suffering (praemeditatio mali): by focussing attention on a future pain we might lessen the harm

²⁷ Note that what allows us to be sure that this is a memory at all is an appropriate connection between the internal representation and a past experience. See the remarks in Bernecker 2010, 235–9.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

156

 

when that pain eventually comes.²⁸ In his account of this claim, Cicero ascribes to the Cyrenaics the implausible claim that only unexpected pains are painful (Tusc. 3.28). More likely, and more in keeping with the remainder of their discussion, it appears that by thinking in advance of the supposed pain the Cyrenaic will lessen the pain of the event when it eventually occurs. It cannot be that the future pathos is being experienced in advance and its intensity eased by its thereby being spread more extensively through time since this would appear to violate the Cyrenaics’ insistence that pain must be unitemporal: any pain experienced in advance must be a distinct pathos from the future and expected pathos of pain. And even if the effect of pre-rehearsal were that the single pain is spread out over a longer time (‘suffering in instalments’) this would seem unlikely to amount to a diminution of pain overall. The best explanation is that by thinking in advance of possible harms the Cyrenaic’s soul becomes arranged and prepared such that, should the imagined harm occur, it will generate a less intense pathos of pain. For example, if the Cyrenaic has been constantly thinking that his children are mortal and fragile it will apparently be less painful for him should one of them be injured or die. The Cyrenaic does not conjure up for himself the painful grief in advance. (Aelian Varia Historia 14.6 reports that Aristippus advised against both ‘toiling in retrospect over things past’ and ‘toiling in prospect (prokamnein) over things to come’.) Rather, the Cyrenaic constantly reminds himself that a certain painful event is possible, so ensuring that it will not be a shock should he ever experience it in the future. The pre-rehearsal is either not itself painful at all or, if it is painful to some extent, the combination of the pre-rehearsal and the eventual lessened pain is not as bad as an unexpected and intense pain.

5. Aristotle has a lot to say about memory and a lot to say about the nature of pleasure. He also has something to say about the relationship between pleasure and memory. Sometimes it appears that Aristotle is working very much in the spirit of a model of the kind that is at work in Plato’s Philebus and in the Epicurean use of the therapeutic powers of memory and anticipation. But at other times Aristotle seems to want to qualify or modify that picture in ways that we might find rather attractive and may be closer to the Cyrenaic insistence on the independence of the effect of a recollected experience from the effect of the original experience. In Rhetoric 1.11, for example, Aristotle offers a discussion of pleasure that includes an important account of the relationship between pleasure and memory. ²⁸ See Cicero Tusculan Disputations 3.28–35. Cf. Graver 2002a, and 2002b, 96–101 and 195–201; O’Keefe 2002.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

, , 

157

The chapter begins with the assertion that pleasure is a certain motion or change of the soul—a claim it shares with Physics 7.3 and which notoriously seems at odds with his considered view in the Ethics—and relates pleasure to desire (1369b33–1370a27).²⁹ Aristotle then turns to address the connection between pleasure and perception. He first notes that experiencing pleasure depends on the perception of a certain pathos. Then he adds that imagination (phantasia), being a kind of weak perception of what is remembered or anticipated, always accompanies an act of remembering or anticipating since whenever we recall or anticipate something we must engage our capacity for imagination (1370a27–30). This allows Aristotle to note that all pleasures must come about either directly via perception itself (when the object is present) or via memory or anticipation (when the object is past or future, respectively). In the latter case, therefore, they will come about via phantasia. Much of this general model is relatively familiar, although we need to be cautious in determining the precise commitments of Aristotle’s own theory. The Philebus, as we noted, explains the experience of pleasant anticipation in terms of a painter in the soul generating images on the basis of statements written down by a scribe; those images may include depictions of the anticipator at some time in the future taking pleasure at some imagined object. Aristotle himself in his detailed psychological works insists on both the close connection between the faculty of imagination and perception (although these two are also to be kept distinct) and also between the faculty of imagination and the capacity for the deliberate recall or deliberate anticipation of non-present events and experiences.³⁰ The overall picture seems to be something like the following. Aristotle agrees that we can experience pleasure or pain as a result of the form of weak perception that is involved when we engage our capacity to recall or anticipate our experiences. Furthermore, he is inclined to agree that for the most part—and we shall see that this qualification is important—things that are pleasant when present are also pleasant when anticipated or remembered (1370b9–10). This allows him to explain why it is that people who are suffering from a fever can take pleasure in remembering or looking forward to a cool drink and why people

²⁹ For discussion see Gosling and Taylor 1982, 194–9; Dow 2011, 61–71. For a discussion of the passage in the Physics see Wardy 1990, 220–7. ³⁰ In so far as Aristotle makes phantasia central to the workings of memory, he is happy to assign this capacity to some non-rational animals. See Sorabji 1993, 50–1, 55–8 and Osborne 2000. Cf. Lorenz 2006, 131–2. However, Aristotle is also prepared to distinguish a kind of deliberate recollection as an activity of reasoning (syllogismos: De mem. 2 453a9–14). This suggests that it is still legitimate to talk of a human capacity for the deliberate recollection and anticipation of past and future experiences such as that discussed in Rhet. 1.11. Note that the sense of memory relevant here is the counterpart of ‘hoping’ or ‘anticipating’, elpizein: Rhet. 1370a27–20. While Aristotle does occasionally talk of non-rational animals having forward-looking capacities and is prepared to say that a lion takes pleasure in the appearance that he will eat the nearby ox—NE 1118a20–23—he also says at PA 669a19–21 that only humans have hope (elpis) or expectation for the future (prosdokia); this is why only humans experience their heart ‘jumping’.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

158

 

in love can take pleasure in talking or writing about their beloved since in engaging in this recollection of the beloved they think that they are in a way perceiving him (1370b15–22). Other passages point in a similar direction. For example, Aristotle comments at De memoria 450a29–32 that memory is the retention of something ‘like a sort of picture’ (hoion zōgraphēma ti), and that this imprint left by a perception is similar to those left by signet rings.³¹ And in the De motu animalium he seems prepared to entertain the view that when we conceive of, remember, or anticipate some experience then we also on occasion come to feel as we would were the imagined scenario genuinely happening in the present. For example, he writes: For sense perceptions are at once a kind of alteration, and phantasia and thinking have the power of the actual things. For it turns out that the form conceived of the [warm or cold or] pleasant or fearful is like the actual thing itself. That is why we shudder and are frightened just thinking of something. (701b17–22, trans. M. Nussbaum 1978; Cf. 702a5–7 and 703b18–20)

In some ways the general picture that emerges is recognizably similar to what was found in the Philebus. First, Aristotle is sometimes happy to explain or illustrate the workings of memory and anticipation in terms that involve the internal viewing of representations of some object experience. Certainly, memory involves the use of phantasmata (see De mem. 450a19–25) and Aristotle sometimes describes the role of phantasmata in terms of images of some kind in the soul.³² Despite these comments, it is not certain, however, that these phantasmata should be understood in fact to be internal images. Perhaps it would be better to say that Aristotle often borrows Plato’s means of explaining the workings of phantasia without being committed to the literal truth of a kind of internal theatre of the mind.³³ Second, the phantasmata involved in memory can sometimes produce the same kind of effects—and produce the same pleasures and pains—as simple direct perception of the external world, although he allows that the intensity of these effects may not be quite the same.

³¹ Cf. Sorabji 2004, 2–8. Lorenz 2006, 161 n.34 is right to detect allusions here to both the Philebus and Theaetetus while noting that Aristotle adopts neither dialogue’s account without qualification. Compare also De mem. 449b22–23 for the claim that memory is a statement in the soul: ἀεὶ γὰρ ὅταν ἐνεργῇ κατὰ τὸ μνημονεύειν, οὕτως ἐν τῇ ψυχῇ λέγει, ὅτι πρότερον τοῦτο ἤκουσεν ἢ ᾔσθετο ἢ ἐνόησεν and note the use of a ‘wax block’ analogy for memory at 450a25–b11. Scheiter 2012 develops an interpretation of Aristotle’s account of phantasia in De an. 3.3 that emphasizes this Platonic background (cf. Moss 2012b, 85–7). ³² On De mem. 450a19–25 see also Caston 1998, 257–9 and esp. n.21. ³³ For a brief summary of recent debates see Sorabji 2004, xi–xx and King 2009, 40–62, who comments: (58): ‘Aristotle nowhere says that representations are images but in several places he compares them with pictures or images.’ For examples of such comparisons see De an. 427b21–4 (see below), 432a7–14 and De mem. 450b20–5. See also Caston 1998, 281–4 and Caston 2009, 323–6, which has a concise introduction to recent discussions.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

, , 

159

There are two texts which offer important further perspectives on this familiar picture. First, in his discussion of phantasia in De anima 3.3, Aristotle distinguishes phantasia from other psychic capacities such as perception (aisthēsis), thinking (dianoia), and supposing (hypolēpsis).³⁴ He offers two reasons for distinguishing phantasia from belief that deserve our attention (427b17–24). His first reason points towards the connection between phantasia and memory; he observes that phantasia is ‘up to us’ (eph’ hēmin) in a way that belief is not since we can call up this experience (pathos) at will. Here Aristotle notes that it is possible to use phantasia to bring things ‘before our eyes’ (pro ommatōn), as is shown by the practice of those who use mnemonic techniques that involve the generation of images.³⁵ His second reason is worth more detailed consideration: Whenever we have a belief that something is terrible or frightening we immediately feel the accompanying pathos, and similarly for something encouraging; but in the case of phantasia we are like people looking at a picture of something terrible or alarming.³⁶ (427b21–4)

This comment is evidently concerned with the connection between phantasia and pathē and therefore should shed light on the relationship between memory and pleasure and pain.³⁷ Aristotle seems to say that while believing that something is terrible will immediately generate the feeling of fear and perhaps an associated pain, choosing to call up a phantasma of something terrible may well not generate a similar affective response.³⁸ If this applies also to the kinds of images generated by recalling past experiences—that is to the deliberate use of the capacity of memory to call up such phantasmata—then it will follow that there is no reason to assume that if some past experience was, for example, frightening and painful, then the deliberate recollection of that experience will once again generate the same pathos. Note that, presumably in part to explain the difference between affective reactions to beliefs and recollections, Aristotle compares the latter with what we experience when we are looking at a picture. He is often tempted to explain the working of memory as a kind of deliberate contemplation of a representation of a past experience. By observing that when we call things to mind with phantasia then we stand to those contents as we do when viewing a

³⁴ On this passage see Caston 1996, 43–6. ³⁵ On these techniques see Sorabji 2004, 22–34, and Small 1997, 81–137. On this passage in De anima 3.3 cf. Wedin 1988: 74–5. Cicero Fin. 2.104–5, as noted above, offers reasons to think that our powers of memory are not entirely ‘up to us’. ³⁶ Cf. De an. 432b29–433a1 and 403a23–4: it is possible to be frightened when nothing frightening is present. ³⁷ See also Dow 2009, 164–5. ³⁸ Compare De an. 432b29–433a1: when someone contemplates (θεωρῇ) something there is not necessarily any call to pursuit or avoidance; even when someone thinks (διανοεῖται) about something fearful it is not necessary that the person experience fear. Cf. Caston 1996, 46–52.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

160

 

picture—presumably, seeing what we call to mind in some way as a depiction—he can allow the possibility of such a state of mind having a different effect from the direct perception of whatever it is that is depicted in this way. When we choose to cast our minds back to some past experience and thereby conjure up a phantasma in this act of remembering there is no necessary connection between any affective response that phantasma might generate in us and how we would normally—perhaps necessarily—feel were we directly faced with experiencing in the present what the phantasma represents. So, while I will feel terror, break out into a cold sweat, find my heart racing, and so on if I am confronted by a tiger, there is no sufficient reason to think that I will experience those same responses when later on I choose to recall that incident. And even if I choose to recall my being terrified when I saw the tiger, there is again no sufficient reason to expect that that act of recollection will itself generate the same affective responses as were involved in that object experience. For these reasons, it should be clear that there is no necessary connection between the painfulness of a past experience, for example, and the affective response that a person will feel when he or she chooses to recall that past experience.³⁹ Second, a similar qualification can be found in Rhetoric 1.11, this time with explicit reference to the pleasures and pains of memory in particular. Aristotle observes that while it is usually the case that we take pleasure in remembering things that were pleasant when originally experienced and we are pained by remembering things that were painful when originally experienced, this is not always the case. Sometimes we recall with pleasure a past pain.⁴⁰ There are therefore circumstances in which the affective response to something recollected may be different from the response that the recollected events originally provoked. For example, there is no simple and direct connection between what it is pleasant to experience now in the present and what it might be pleasant to remember having experienced: But things that are pleasant when remembered are not only those that were pleasant when they were present. But sometimes also things that were not pleasant [sc. when present are pleasant when remembered], provided that what comes after this was fine and good. (1370a35–b3)

The thought is characteristically compressed and provokes a number of immediate questions which turn on the proper understanding of the important qualification ³⁹ Hamlyn 1968, 131 ad loc. comments briefly and approvingly: ‘What Aristotle says in the latter half of the passage seems quite correct.’ ⁴⁰ Compare Sidgwick 1907, 144 (II III §5): ‘To this case it seems due that past hardships, toils, and anxieties often appear pleasurable when we look back upon them, after some interval; for the excitement, the heightened sense of life that accompanies the painful struggle, would have been pleasurable if taken by itself; and it is that that we recall rather than the pain.’ This denies that we recall the pain itself, perhaps because Sidgwick finds odd precisely what Aristotle seems to claim, namely that recalling a past pain can indeed itself sometimes be pleasant.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

, , 

161

‘provided that what comes after this (to meta touto) was fine and good’. For example, does Aristotle mean that we can recall with pleasure some painful experiences provided that they are followed by something fine and good? Alternatively, does Aristotle mean just that we can recall with pleasure some painful experiences provided that their consequences were pleasant? In that case, is he claiming that we can recall the prior pain with pleasure because in doing so we also bring to mind, as it were, the causal connection between that prior pleasure and our current positive state? We need to determine the precise relationship between the memory and pleasure. Why is it that what was originally a painful experience can be pleasant when recalled later? There are various options: a. It is now pleasant to recall the event because we recall the pleasant excitement and intensity of the experience rather than the pain and struggle. b. It is now pleasant to recall the event because, although painful at the time, it is now over. When we recall the event our attention is drawn to its past-ness and that is a pleasant thought. c. It is now pleasant to recall the event because, although it was painful at the time, when we recall the event our attention is drawn to our comparatively positive state now and that is a pleasant thought. d. It is now pleasant to recall the event because, although painful at the time, it had beneficial consequences which are pleasant. e. It is now pleasant to recall the event because, although painful at the time, in retrospect we see the event itself—not just its consequences—as something positive and beneficial. Some of these possibilities can be ruled out rather swiftly. Option (a) is inspired by a remark by Sidgwick. He writes: To this case it seems due that past hardships, toils, and anxieties often appear pleasurable when we look back upon them, after some interval; for the excitement, the heightened sense of life that accompanies the painful struggle, would have been pleasurable if taken by itself; and it is that that we recall rather than the pain.⁴¹

This seems to be a rather implausible claim in itself but it reveals nevertheless how Sidgwick is evidently tempted to look for some pleasant aspect in the original experience itself which is now being recalled with pleasure in the present. His intuition seems to be that, if it is pleasant to recall some past experience, there must originally have been some pleasant aspect of that experience which is responsible now for the pleasure of the recollection. This is not, I think, what Aristotle has in

⁴¹ Sidgwick 1907, 144.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

162

 

mind. Rather, he is envisaging a case in which we recall with pleasure an experience which, when it was originally present, was in no way pleasant. Option (b) might put us in mind of A. N. Prior’s claim that the fact that we can reasonably say ‘Thank goodness, that’s over!’ about some past event suggests that there is indeed some genuine property of past-ness to which we are now referring and which can be the cause of relief or even positive pleasure. Prior goes on to claim, notoriously, that we cannot give a complete tenseless equivalent for such statements and that this in turn suggests that time does indeed flow.⁴² The metaphysics of time that this kind of locution may or may not imply can be left aside on this occasion, provided we notice only that we do indeed seem to be able to take pleasure in an event’s past-ness, however we choose to analyse what ‘pastness’ amounts to. In other words, what we take pleasure in or express relief at is not so much the content of the past experience as the fact of it being a past experience. However, it is not likely that this is what is on Aristotle’s mind either. The examples that follow his comments in Rhetoric 1.11, drawn from epic poetry and tragedy, shed some more light on what he means. It seems to me that they might fit options (c), (d), and (e) above, although Aristotle does not make their precise sense explicit. For his first example, Aristotle cites a familiar line from Euripides’ Andromeda (fr. 133 Nauck²): ‘Pleasant it is when rescued to remember toils (ponoi).’⁴³ We have already noted the use made of this line in Cicero’s discussion of Epicureanism. Here in Aristotle it is a succinct example of an interesting exception to the general rule that things that are pleasant when remembered were pleasant when first experienced. Sometimes, because of some intervening event we might remember with pleasure something—here, the toils—that was painful when present.⁴⁴ The other example comes from Homer Odyssey 15.400–401; Eumaeus the swineherd is addressing the disguised Odysseus and is about to tell the story of how he was taken from his birthplace as a young boy by some Phoenicians and later bought as a slave by Laërtes, Odysseus’ father. He says to Odysseus, ‘Let us cheer one another by both recalling our sad cares,’ and then in the two lines which Aristotle cites, Eumaeus explains: ‘for a man is cheered even by sufferings when he recalls how much he has suffered and how far he has wandered.’⁴⁵ ⁴² See Prior 1959. Cf. Mellor 1981, which provoked various other pieces helpfully collected in Oaklander and Smith 1994, Part III. ⁴³ ἀλλ’ ἡδύ τοι σωθέντα μεμνῆσθαι πόνων. ⁴⁴ The plot of the lost Andromeda is not certain, but it is possible that the line was spoken by Perseus, perhaps recounting his relief at having escaped his encounter with the Gorgon. See Webster 1967, 192–9, esp. 195, and Wright 2005, 121–33. ⁴⁵ Hom. Od. 15.399–401: κήδεσιν ἀλλήλων τερπώμεθα λευγαλέοισι,|μνωομένω· μετὰ γάρ τε καὶ ἄλγεσι τέρπεται ἀνήρ,|ὅς τις δὴ μάλα πολλὰ πάθῃ καὶ πόλλ᾽ ἐπαληθῇ. It is also possible to use this idea as a means of consolation during periods of suffering since in the future perhaps the recollection of even these sufferings may bring pleasure: cf. Virg. Aen. 1.202: forsan et haec olim meminisse iuvabit. Plotin. Enn. 4.3.28 offers as evidence that the memory of having a desire is not retained in the desiring part of the soul the fact that something which was originally pleasant when experienced is not always pleasant to recall. See King 2009, 167–8.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

, , 

163

A closer look at the Odyssey will also help to explain why Aristotle may have thought this an interesting illustration of his point since the situation to which Aristotle alludes by citing these lines is complicated not only by the narrative of disguise, sincerity, and concealment that is at work at this point of the epic but also by the characters themselves drawing attention to that very complexity. The broader context of Eumaeus’ claim suggests an interest on the part of the epic poet in the effects that stories of past sufferings—both truthful and deceitful—may have on the audience and on the speaker. Remember that Eumaeus not only asserts that it can be pleasant to recall what were painful experiences but also that he imagines this taking place within a context of mutual exchange: ‘let us cheer one another by sharing our sad tales.’⁴⁶ But the exchange is not quite the fair and sincere one that Eumaeus expects. In telling his story, Eumaeus is answering the Cretan story concocted by the disguised Odysseus and recounted in the previous book (14.192ff.): a false story that nevertheless managed to generate an emotional response in Eumaeus (14.361–2). We might wonder whether, given that this was a false story, it can have had the same sort of cheering effect on Odysseus himself as he told it. When his turn comes to tell a story in return, Eumaeus hopes that recounting his own story will have such an effect on both Odysseus and himself; he hopes that the mutual sharing of stories will be positive for speaker and audience in turn. However, Eumaeus’ sincere recollection and Odysseus’ artifice, although perhaps indistinguishable to their respective audiences, will presumably differ in the effects each has on its own speaker. Eumaeus begins his account with the lines cited by Aristotle, setting this hope of some positive emotional benefit as a frame for his tale. When he concludes his tale at 15.484 with his first arrival on Ithaca, he makes no further comment on whether he has in fact experienced any present emotional impact in recalling those events so we are left perhaps simply to remember his opening statement and infer that this has indeed been a positive experience. Odysseus, on the other hand, replies by noting that Eumaeus’ story has moved him, more or less repeating the pair of lines with which Eumaeus had responded to Odysseus’ first—and deceitful—tale (15.486–7, cf. 14.361–2). Odysseus also comments that although Eumaeus’ story—like his own—included many sufferings, it ended well with Eumaeus’ arrival at the home of a good master and a long and happy life.⁴⁷ The situation in the epic is perfect for Aristotle to use as an example for a very specific point and a very specific qualification of his general thesis. Aristotle does not appear to cast any doubt on the truth of the memory. By the ‘truth of the ⁴⁶ Aristotle also comments at Rhet. 1390a6–11 that the elderly are more likely to spend time engaged in—and enjoying—such reminiscences. The young are more inclined to dwell on the future and for the most part live ‘in anticipation’ (elpidi) (1389a20–4). ⁴⁷ Homer Odyssey 15.486–7: Εὔμαι᾽, ἦ μάλα δή μοι ἐνὶ φρεσὶ θυμὸν ὄρινας|ταῦτα ἕκαστα λέγων, ὅσα δὴ πάθες ἄλγεα θυμῷ.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

164

 

memory’ I mean just that the situation recalled did happen in just the way in which it is remembered. In Eumaeus’ case this will amount to the claim that Eumaeus remembers truly the events of his being taken and enslaved and brought to Ithaca. Furthermore, Aristotle presumably will also allow that Eumaeus’ memory might be authentic in the sense that he recalls correctly that he was afraid and distressed at the time by those events.⁴⁸ Nevertheless, something about Eumaeus’ affective response to recalling these events and these attitudes is distinctive and that is why his situation interests Aristotle. For Eumaeus reports that he remembers being enslaved, remembers how it felt at the time but nevertheless now is pleased when he thinks back to those events. There must be some important characteristic of this situation that will make sense of this unusual aspect. It is likely that Eumaeus’ original claim that such reminiscences can bring pleasure is licensed not only by the fact that the sufferings in question have ended but also by the fact that they have ended well. For Eumaeus, we might say that it is pleasant to recall his prior sufferings because they resulted in his arrival in Ithaca and the good life he was able to live there. What is important for Aristotle is that this allows Eumaeus to recall the prior sufferings themselves and to take pleasure in that recollection since this offers an exception to the general rule that we are pained at the recollection of past pains. Eumaeus can, with hindsight, now see those sufferings as leading to a later good whereas obviously this was not possible at the time he was originally experiencing the toils. And it is because he now recalls the toils as leading to a later good that he can now take pleasure in them and not just in their later good results. Eumaeus offers an example of option (e) above and perhaps also of option (d): the painful events led to good consequences and therefore can themselves now be considered in hindsight to be good. It is therefore pleasant now to remember them. In the case of the line from Euripides’ Andromeda we are hampered by a lack of the original context. Perhaps here too the character in question is expressing an appreciation for the positive results that came from previous perils. But the emphasis seems different: it is the simple fact of now being safe that matters rather than some additional positive result. To be sure, in some way or other it is necessary to be in peril in order later to be rescued and in this way the rescue and the pleasure of being rescued may well be thought of as resulting from being first imperilled. Just as it was necessary for Eumaeus to be snatched from his home for him to arrive in Ithaca so too it was necessary for Perseus, let us say, to undergo various perils before he could be rescued from them. And in this sense we can say that the good of being saved is a consequence of the prior dangers. We might even go so far as to identify a particular pleasure in being saved from danger at the very last moment: a euphoric feeling of unexpected salvation. But it would be a little

⁴⁸ See Bernecker 2010, 215–7.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

, , 

165

perverse to imagine that prior peril is a necessary precondition of enjoying being safe and sound. Rather, Perseus can look back and recall the various toils which he has undergone and which are now in his past. As a result, he is able to take special pleasure in his current safety since he realizes that things might not have turned out as they did. Perseus therefore might be an example of option (c) above. In both cases, Aristotle shows that he is not prepared to be shackled by a simple account according to which phantasia always brings along with it the affective response occasioned by the original experience upon which it is based. And he is right to do so. He can also offer a reasonable account of how this can happen in terms of his own moral psychology since it is likely that he considered human perceptions and our consequent ability to ‘envisage’ both prospects and also our own pasts to be conceptually rich. It is therefore correct to say that now Eumaeus ‘sees’ the events in his past differently from the way he originally saw them (or they ‘appear to him’ differently) and hence he now experiences pleasures as he reenvisages them while he originally experienced pains. He now sees and understands those events as being beneficial and good—hence pleasant—whereas he originally saw them as harmful and painful.

6. The picture suggested by Aristotle’s brief discussion of such examples as the memories of Eumaeus represents an important and plausible addition to the understanding of the pleasures of anticipation and recollection found in Epicureanism and implied by the Platonic texts we have considered: when someone recalls a past experience, that memory and the person’s emotional reaction to it are both affected in important ways by his or her present state of mind. Furthermore, this interest in the context of the act of recollection appears to chime well with some modern psychological accounts of the nature and value of memory. Studies show that memories and the nature of our recollection of them are often importantly shaped by social context and by our present state.⁴⁹ It is not incorrect, of course, also to notice that our memories, for example, sometimes are tied to strong affective experiences such that our memories do indeed generate a certain response in us whatever the current circumstance in which we recall them.⁵⁰ However, the extent to which our ancient sources are inclined to consider memory and anticipation as a way of re-living or pre-living our experiences steers them firmly in the direction of thinking that this direct transmission of affect from the object experience to the experience of remembering or anticipating it is either always the case or, at the least, the general rule. If ⁴⁹ For a brief introduction see Engel 1999. See also Bernecker 2010, 217–39. ⁵⁰ See Margalit 2002.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

166

 

there is a deficiency in the Philebus’ account of these matters, or in the Epicureans’ confidence in the power of memory, then it perhaps stems from a particular conception they share of the act of remembering itself. The greater the temptation to consider memory as a form of conjuring and reviving in the present some past experience then the greater the temptation to think that the affective response attached to that act of memory must be the same as the object experience being recalled. We have seen the Cyrenaics and Aristotle offering reasons to doubt that account of what is involved in an act of recollection and hence reasons to think that a more complex account is needed of the pleasures and pains of memory and anticipation. Perhaps the Cyrenaics—consistent with their more general epistemological preferences—went furthest towards offering an analysis of the affections of memory and anticipation in terms that stress the context of the act of recollection itself rather than the object experience. But there are signs already in Aristotle of the recognition that this is a point of view worth taking seriously.

References Annas, J. 1992, ‘Aristotle on Memory and the Self.’ In Essays on Aristotle’s De Anima, ed. M. C. Nussbaum and A. O. Rorty, 297–31. Oxford: Oxford University Press. Bernecker, S. 2010. Memory: A Philosophical Study. Oxford: Oxford University Press. Caston, V. 1996. ‘Why Aristotle Needs Imagination.’ Phronesis 41: 20–55. Caston, V. 1998. ‘Aristotle and the Problem of Intentionality.’ Philosophy and Phenomenological Research 58: 249–98. Caston, V. 2009. ‘Phantasia and Thought.’ In A Companion to Aristotle, ed. G. Anagnotopoulos, 322–34. Oxford: Wiley-Blackwell. Delcomminette, S. 2006. Le Philèbe de Platon: introduction à l’agathologie platonicienne, Leiden: E. J. Brill. Dow, J. 2009. ‘Feeling Fantastic? Emotions and Appearances in Aristotle.’ Oxford Studies in Ancient Philosophy 37: 143–75. Dow, J. 2011. ‘Aristotle’s Theory of the Emotions: Emotions as Pleasures and Pains.’ In Moral Psychology and Human Action in Aristotle, ed. M. Pakaluk and G. Pearson, 47–74. Cambridge: Cambridge University Press. Engel, S. 1999. Context Is Everything: The Nature of Memory. New York: Palgrave. Evans, M. 2007. ‘Plato’s Rejection of Thoughtless and Pleasureless Lives.’ Phronesis 52: 337–63. Evans, M. 2008. ‘Plato on the Possibility of Hedonic Mistakes.’ Oxford Studies in Ancient Philosophy 35: 89–124. Frede, D. 1985. ‘Rumpelstiltskins’s Pleasures: True and False Pleasures in the Philebus.’ Phronesis 30: 151–80, reprinted in G. Fine (ed.) 1999. Plato II. Oxford: Oxford University Press.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

, , 

167

Frede, D. 1993. Plato: Philebus. Indianapolis: Hackett. Frede, D. 1997. Platon: Philebos. Übersetzung und Kommentar. Göttingen: Vandenhoeck & Ruprecht. Giovacchini, J. 2007. ‘Le souvenir des plaisirs: le rôle de la mémoire dans la thérapeutique épicurienne.’ In Hédonismes: penser et dire le plaisir dans l’Antiquité et à la Renaissance, ed. L. Boulègue and C. Lévy, 69–83. Lille: Presses Universitaires du Septentrion. Gosling, J. C. B. 1975. Plato: Philebus. Oxford: Oxford University Press. Gosling, J. C. B. and Taylor, C. C. W. 1982. The Greeks on Pleasure. Oxford: Oxford University Press. Graver, M. 2002a. ‘Managing Mental Pain: Epicurus vs. Aristippus on the Prerehearsal of Future Ills.’ Proceedings of the Boston Area Colloquium in Ancient Philosophy 17: 155–77; with commentary by G. Striker: 178–84. Graver, M. 2002b. Cicero on the Emotions: Tusculan Disputations 3 and 4. Chicago: University of Chicago Press. Hackforth, R. 1945. Plato’s Examination of Pleasure. Cambridge: Cambridge University Press. Hamlyn, D. W. 1968. Aristotle’s De anima Books II and III (with certain passages from Book I). Oxford: Oxford University Press. Harte, V. 2004. ‘The Philebus on Pleasure: The Good, the Bad and the False.’ Proceedings of the Aristotelian Society 104: 111–28. Harte, V. 2014. ‘Desire, Memory, and the Authority of Soul: Philebus 35cd.’ Oxford Studies in Ancient Philosophy 46: 34–72. King, R. A. H. 2009. Aristotle and Plotinus on Memory. Berlin: De Gruyter. Lorenz, H. 2006. The Brute Within: Appetitive Desire in Plato and Aristotle. Oxford: Oxford University Press. Lovibond, S. 1989–90. ‘True and False Pleasures.’ Proceedings of the Aristotelian Society 90: 213–30. Madvig, N. 1876. M. Tulli Ciceronis De finibus bonorun et malorum libri V (3rd edition), Copenhagen: Hegel. Margalit, A. 2002. The Ethics of Memory. Cambridge, MA: Harvard University Press. Mellor, D. 1981. ‘Thank Goodness That’s Over.’ Ratio 23: 20–30. Meyer, S. Sauvé 2012. ‘Pleasure, Pain, and “Anticipation” in Plato’s Laws Book 1.’ In Presocratics and Plato: Fetschrift at Delphi in Honor of Charles Kahn, ed. L. Patterson, V. Karasmanis, and A. Hermann, 311–28. Las Vegas: Parmenides Publishing. Moore, G. E. 1903. Principia Ethica. Cambridge: Cambridge University Press. Moss, J. 2012a. ‘Pictures and Passions in the Timaeus and Philebus.’ In Plato and the Divided Self, ed. R. Barney, T. Brennan, and C. Brittain, 259–80. Cambridge: Cambridge University Press.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

168

 

Moss, J. 2012b. Aristotle on the Apparent Good: Perception, Phantasia, Thought, and Desire. Oxford: Oxford University Press. Moss, J. 2012c. ‘Pictures and Passions in the Timaeus and Philebus.’ In Plato and the Divided Self, ed. R. Barney, T. Brennan, and C. Brittain, 259–80. Cambridge: Cambridge University Press. Nussbaum, M. 1978. Aristotle’s De motu animalium. Princeton: Princeton University Press. Nussbaum, M. 1995. ‘Aristotle on Human Nature and the Foundations of Ethics.’ In World, Mind, and Ethics: Essays on the Ethical Philosophy of Bernard Williams, ed. J. E. J. Altham and R. Harrison, 86–131. Cambridge: Cambridge University Press. Oaklander, L. N. and Smith, Q. (eds) 1994. The New Theory of Time. New Haven: Yale University Press. O’Keefe, T. 2002. ‘The Cyrenaics on Pleasure, Happiness, and Future-concern.’ Phronesis 47: 395–416. O’Reilly, K. R. 2019a. ‘Cicero Reading the Cyrenaics on the Anticipation of Future Harms.’ Epoché: A Journal for the History of Philosophy 23: 431–43. O’Reilly, K. R. 2019b. ‘The Jellyfish’s Pleasures: Philebus 20b–21d.’ Phronesis 64: 277–91. Osborne, O. 2000. ‘Aristotle on the Fantastic Abilities of Animals.’ Oxford Studies in Ancient Philosophy 19: 253–85. Parfit, D. 1984. Reasons and Persons. Oxford: Oxford University Press. Penner, T. 1970. ‘False Anticipatory Pleasures: Philebus 36a3–41a6.’ Phronesis 15: 166–78. Persson, I. 2005. The Retreat of Reason. Oxford: Oxford University Press. Prior, A. N. 1959. ‘Thank Goodness That’s Over.’ Philosophy 34: 12–17. Russell, D. 2005. Plato on Pleasure and the Good Life. Oxford: Oxford University Press. Scheiter, K. M. 2012. ‘Images, Appearances, and Phantasia in Aristotle.’ Phronesis 57: 251–78. Sedley, D. N. 2004. The Midwife of Platonism: Text and Subtext in Plato’s Theaetetus. Oxford: Oxford University Press. Sedley, D. N. 2017. ‘Epicurean versus Cyrenaic Happiness.’ In Selfhood and the Soul: Essays on Ancient Thought and Literature in Honour of Christopher Gill, ed. R. Seaford, J. Wilkins, and M. Wright, 89–106. Oxford: Oxford University Press. Sidgwick, H. 1907. The Methods of Ethics (7th edition). London. Small, J. P. 1997. Wax Tablets of the Mind: Cognitive Studies of Memory and Literacy in Classical Antiquity. London: Routledge. Sorabji, R. R. K. 1993. Animal Minds and Human Morals: The Origins of the Western Debate. London: Duckworth. Sorabji, R. R. K. 2004. Aristotle on Memory (2nd edition). Bristol: Bristol University Press.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

, , 

169

Tsouna, V. 2016. ‘Cyrenaics and Epicureans on Pleasure and the Good Life: The Original Debate and its Later Revivals.’ In Strategies of Polemics in Greek and Roman Philosophy, ed. S. Weisser and N. Thaler, 113–49. Leiden: E. J. Brill. Wardy, R. B. B. 1990. The Chain of Change. Cambridge: Cambridge University Press. Warnock, M. 1987. Memory. London: Faber & Faber. Warren, J. 2011. ‘Socrates and the Patients: Republic IX 583c–585a.’ Phronesis 56: 113–37. Warren, J. 2013. ‘Epicureans and Cyrenaics on Pleasure as a Pathos.’ In Épicurisme et scepticisme, ed. S. Marchand and F. Verde, 127–44. Rome: Sapienza Università Editrice. Warren, J. 2014. The Pleasures of Reason in Plato, Aristotle, and the Hellenistic Hedonists. Cambridge: Cambridge University Press. Webster, T. B. L. 1967. The Tragedies of Euripides. London: Faber & Faber. Wedin, M. V. 1988. Mind and Imagination in Aristotle. New Haven: Yale University Press. Williams, B. 1959. ‘Pleasure and Belief.’ Proceedings of the Aristotelian Society, suppl. vol. 33: 57–72, repr. in his 2006 Philosophy as a Humanistic Discipline (ed. A. W. Moore), Princeton: Princeton University Press: 34–46. Wright, M. 2005. Euripides’ Escape Tragedies: A Study of Helen, Andromeda, and Iphigenaia among the Taurians. Oxford: Oxford University Press.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

11 Modelling the Memory and Anticipation of Pleasure Comments on James Warren’s ‘Memory, Anticipation, Pleasure’ Margaret Hampson and Katharine R. O’Reilly

1. Since its presentation at the 2011 Keeling Colloquium,¹ and the publication in 2014 of his monograph reflecting and developing many of the same themes, Warren’s ‘Memory, Anticipation, Pleasure’ has—directly or indirectly—made a significant impact on research into the psychology of memory and anticipation in ancient philosophy. Its range and macro-level approach invite us to take up many of the questions the paper raises. One of its contributions is the provision of two models for how ancient philosophers understood the memory and anticipation of pleasure and pain. In this response, we take these tools Warren has given us and probe further into some of the arguments he is interested in. In doing so, we aim to make some progress on questions of interest for further research, without expecting to have the final word. We therefore position our response as a kind of interim piece: progressing the discussion Warren began, while refreshing the invitation to take up these questions in more detail. The two models Warren describes as ways ancient philosophers conceived of the memory and anticipation of pleasure are as follows: on the first model, anticipation and recollection are regarded as a means of reaching out and hauling into the present a temporally remote experience of pleasure or pain.² On this model, affective states of different times can be compared, and this comparison ¹ The original respondent to Warren’s paper at the 2011 Keeling Colloquium was Anthony Price, whose discussion of Warren and Moss is included earlier in this volume. We direct the reader to Price’s piece for critical discussion of Warren’s interpretation of the Philebus, as well as a number of other insightful remarks. We thank Fiona Leigh for giving us the opportunity to respond to this paper— which we heard at the Keeling Colloquium as graduate students—and for her help in clarifying several points. ² Warren (2011) p. 142. By the conclusion this is also described as a ‘conjuring and reviving’ (165–6). Margaret Hampson and Katharine R. O’Reilly, Modelling the Memory and Anticipation of Pleasure: Comments on James Warren’s ‘Memory, Anticipation, Pleasure’ In: Psychology and Value in Plato, Aristotle, and Hellenistic Philosophy: The Ninth Keeling Colloquium in Ancient Philosophy. Edited by: Fiona Leigh and Margaret Hampson, Oxford University Press. © Oxford University Press 2022. DOI: 10.1093/oso/9780192858108.003.0011

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

   

171

can facilitate further inferences about the nature of affections, and the character of the agent undergoing them. The agent ‘relives’ or ‘pre-lives’ temporally remote affections, and these experiences can in turn ameliorate their current state. Warren uses this model to understand the Platonic texts he analyses (Protagoras, Republic IX, Philebus) and the Epicurean material. The second model is one in which an experience at t₁ might have a different and opposite affective character to the anticipation of that experience at t-₁, or the recollection of it at t₂. Warren attributes this model to Aristotle in the Rhetoric. He also discusses the Cyrenaics, though for reasons of economy we will not consider them here. There are two sorts of questions we can ask about these models. One sort concerns the characterization of the models themselves: we might ask, for example, whether these models are mutually exclusive and/or competing; if so, what are their major points of disagreement? If not, what motivates Warren to contrast them? The other sort concerns the fit between the models and the texts Warren surveys: do these models accurately capture the views expressed in these texts? And if we strain to fit the texts into the models, what does that reveal? We will first consider the contrast Warren draws between the two models, and then consider each model discretely and examine the internal coherence of Warren’s analysis of each. We will conclude with a tentative suggestion as to how, in the face of the issues we raise, the distinction between Warren’s two models—and with it, the many insights his paper offers—might be preserved.

2. We want to begin by considering the first cluster of questions, since we believe that there is some unclarity in Warren’s characterization of the two models, and indeed something potentially misleading about the way in which the original contrast between the two models is set up. For they are characterized by Warren as a contrastive pair, but what are in fact contrasted appear to be two different phenomena. The first model, as it is initially characterized by Warren, is a model of the nature of memory and anticipation,³ according to which these are modes of reaching out and hauling to the present some temporally remote experience, and forms of re- or pre-living. This could be appropriately contrasted with other models of the nature of memory and anticipation. The second model, though, is not a model of the nature of memory and anticipation. Instead, it is first characterized in terms of the affective character of acts of recollection and ³ ‘On the first model, anticipating and recalling are thought to be means of reaching out to the past or future and hauling some temporally remote experience from there into our present’ (142).

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

172

    .  ’ 

anticipation,⁴ where the affective character of an act of recollection or anticipation can differ from that of the ‘object experience’.⁵ The two models (as characterized) don’t obviously demonstrate two contrasting ways ancient philosophers have conceived of the memory and anticipation of pleasure but, rather, two different types of phenomena related to the memory and anticipation of pleasure that ancient philosophers have been concerned with: philosophers might be interested in the nature of memory and anticipation and what happens when we remember or anticipate some past or future affective experience, or they might be interested in the affective character of an act of remembering or anticipating some affective experience, in what it feels like to remember or anticipate a pleasant or painful experience. To contrast these as two models of the same phenomena is misleading if they pick out seemingly different and non-commensurate phenomena, and ought not, as such, to be directly compared. Yet Warren evidently takes these two models to form a contrastive pair, so how might they be contrasted as Warren suggests? What we think Warren has in mind can be seen most clearly when we ask the following question: what determines the affective character of an act of recollection or anticipation? If we consider what a certain conception of the nature of memory and anticipation implies regarding the affective character of an act of recollection or anticipation, we can thereby start to fill in the gaps and contrast two more complete models of both the nature and affective character of these acts. As Warren indicates in his closing remarks (165–6), if we think of recollection or anticipation as forms of re- or pre-living, then it would seem to follow that the affective character, and in particular the valence of the act of recollection or anticipation, will be the same as that of the object experience. This is so as long as we assume that the affective character of the object experience is part of what is remembered and thus re-lived, or anticipated and thus pre-lived. On this view, an act of recollecting or anticipating appears to directly inherit its affective character from that of the object experience. An alternative view of how memory and anticipation work, meanwhile, opens the possibility that the character of the act of recollection or anticipation may differ from that of the object experience. Contrasting the various texts under consideration in terms of their adherence to one or other of two models—one which holds that memory and anticipation just are forms of re- or pre-living and thus assumes that the affective character of an act of recollection or anticipation will be the same as that of the object experience, and one that denies this—may, then, help us to appreciate an important difference in how ancient thinkers understood the workings of memory and anticipation, and the implications of these differing views for thinking about the affective ⁴ ‘Sometimes a past painful experience can be recalled with pleasure’ (142–3). This expression most naturally reads as picking up on Warren’s preceding observation that ‘the ability to look forward and backward to our future and past experiences allows us to generate further affective responses in the present. We can remember and anticipate with pleasure or with pain. We can remember our pleasures with pleasure and be pained when we anticipate pains’ (141–2). ⁵ As Warren puts it, e.g. on pp. 149–50 and as we adopt here.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

   

173

character of acts of recollecting or anticipating. But this insight is subject to two postulates: that the phenomena picked out by Warren’s two models can be filledin and contrasted in the way we have hypothesized, and that they accurately describe what we find in the texts. In the following two sections we call into question the second of these postulates, while also problematizing the first.

3. A major assumption built into the first model is that the affective character of the object experience is retained or prefigured by the memories or anticipations of that experience. How strong is this claim? Warren says of the second model that ‘unlike its counterpart, this model does not assume that an experience that was painful to us in the past will always be painful when we remember it’ (142–3). As he goes on to explain, with reference to Aristotle especially, the second model is one in which we can remember with pleasure experiences which were painful at the time. Is the first model then, by contrast, committed to the thought that an experience that was pleasurable in the past will always be pleasurable in the future, or was always pleasurable when anticipated in the past? And is the emphasis on the thought that memory or anticipation of a pleasurable experience will always produce a pleasant affective response (where the contrast case is one in which the act of recollecting or anticipating is accompanied by no affective response), or that memory or anticipation of a pleasurable experience will always produce a pleasant affective response (where the contrast case is one in which the act of recollecting or anticipating is painful)? Warren’s choice of examples often suggests that he is interested in the valence of the response that accompanies acts of recollection or anticipation, but (as we will see) other texts would seem more germane to the issue of whether an affective response will accompany the act at all and, if it does, how intense such a response will be. In either case, we can ask why the memory or anticipation of a pleasurable experience will always be pleasant. In respect of ‘reliving’ and ‘pre-living’, we might wonder whether an agent is to be understood as having the same experience again (or in advance). At another point Warren talks of the affect being ‘transmitted’ (165–6) from the object experience to the experience of recollecting, and we wonder whether these (re-living and the affect being transmitted) are equivalent, and whether we are to understand the object experience as distinct from its affective valence, or not, Let’s consider these questions, first, with respect to two of the Platonic texts Warren discusses: the Protagoras and Republic IX, both of which focus especially on anticipation.⁶ In the first, Warren concerns himself with the argument against ⁶ For discussion of the Philebus, we direct the reader to Anthony Price’s contribution to this volume (Price, Ch. 7, pp. 110–12). Price challenges Warren’s interpretation of the Philebus as offering a re- or pre-living view of memory and anticipation.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

174

    .  ’ 

akrasia at Protagoras 354a ff., in which Socrates describes the art of measurement which helps people to avoid choosing poorly due to the distorting effect of temporal distance. Here someone with a commitment to maximizing pleasure and minimizing pain in the long term is described as making choices based on calculations that are analogized to weighing and measuring (356b1–e4). On Warren’s description, ‘Socrates must assume . . . that his ideal agent will be able accurately and reliably to anticipate and evaluate the pleasures and pains consequent on particular choices’ (145–6). Applying model one, we should then be able to understand what this person does as a version of reaching out and hauling to the present a temporally remote pleasure. What does this act of reaching and hauling look like, if this passage is an example of it? With respect to this passage, the objects which go into the pans of measurement to be psychologically counterbalanced are described as pleasant things (τὰ ἡδέα) and painful things (τὰ λυπηρά, at 356b1 ff.), which, if anything, suggests objects, i.e. the objects which cause pleasure or pain (a piece of cake, the dentist’s drill, etc.). However, we also know that what goes in the pans is subject to degrees of size and scale (356a5–6), can be attached to actions in the sense of actions bringing these affections about (356b8), and are subject to distortion (356c6), which can be corrected by measurement. Additionally, Socrates denies that, within the confines of this weighing exercise, actions can differ in any respect other than the pleasure and pain they yield (356a8–b1). This suggests that ἡδονή and λύπη—pleasures and pains themselves (i.e. the pleasure of eating the piece of cake, the pain of the dentist’s drill)—are what go in the pans, since only they meet these criteria. And these could only be the anticipated pleasures and pains that would follow from possible courses of action. In this passage, then, the sense in which an agent could be described as dragging a future pleasure into the present is through this proactive process of estimating the expected pleasure or pain of a future experience in a time-neutral way. The agent is described as doing this hypothetically, and there is no hint of them experiencing the anticipated affect when they engage in this prudential reasoning. In anticipating the dentist’s drill we may feel fear, but we don’t feel pain—or, at least, we don’t feel the pain of a drill in our teeth. This example demonstrates that there is some way, on Plato’s view, in which we can anticipate future pleasures and pains without feeling those very pleasures and pains in the present. Compare Republic IX, 583b–584b.⁷ Here Socrates invokes the phenomenological experience of patients (‘τῶν καμνόντων’, 583c10) suffering illness as part of an argument in defence of there being an intermediate, affectively neutral state between pleasure and pain. This state is often mistaken for a pleasant or painful one because of the contrast these agents experience when they recover from

⁷ Discussed by Warren on pp. 146–7.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

   

175

illness, for instance. They may anticipate their recovery as something pleasant, compared to the pain they suffer while ill, when in fact it will be merely neutral when it comes about. Warren recognizes that this way of conceiving of possible future states of pleasure differs from the prudential reasoning process described in the Protagoras because here in the Republic ‘we can set them against our current state and draw inferences from the comparison between the two’ (146). Indeed, these inferences ‘may lead us to revise how we evaluate our current state’ (146).⁸ However, inference and evaluation do not accurately describe what the patients experience, in Socrates’ example. Rather than evaluating and inferring, the patients’ experiences are altered in a way that they may be passive to. Plato seems here to invoke the example of a phenomenological experience we are all familiar with, where from a position of illness we might look to our future recovery with anticipated relief and pleasure. The comparison between our current state and that recovered state certainly might result in us affectively overvaluing the latter, but describing how we might do so as inferring and evaluating implies something quite cognitively robust. A patient could deliberately dwell on their expected or hoped-for recovery to help them through the pain of illness. But it is plausible that some patients, in some situations, lack self-awareness about how and why their experience is altered as a result of this comparison, such that it could only be described as inference and evaluation in a cognitively weak sense. Here, then, the sense in which pleasures and pains of the future are dragged into the present and contrasted with one’s opposite state can be passive, and the comparison with Protagoras helps to reveal that. The technē metrētikē of the latter involves a deliberate and active sense of dragging. In Republic IX, by contrast, the dragging or conjuring, if it can be called that, looks markedly different—in some cases, at least, it can be more inert, less deliberate, less self-conscious, and less cognitively demanding. Moreover, Plato’s recognition of the possibility of overvaluation is problematic for model one in another sense. If the patient is able to overvalue their anticipated pleasure, then they can hardly be said to be dragging the affect from another temporal position, at least if we understand dragging as a wholesale grasping and transferring of the affect. The conditions for dragging seem to rule out cases where the anticipated affect is not identical to the actual affect when it eventually occurs. They also rule out cases where the overvaluation itself causes the present relief. Taken together, these two passages suggest that the sense of ‘reaching out and hauling’ in model one, if right, would have to cover both the possibility of more passive and non-reflective cases like those imagined in Republic IX, and the more proactive and hypothetical ones described in the Protagoras, including those which generate overappraisals. And perhaps it can: part of the function of a ⁸ Also taken up, in more detail, in Warren (2011). In the latter, Warren is much more explicit about the patients’ commitments, though we will not take up that view here.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

176

    .  ’ 

model is to describe the phenomena at a high level, grouping together different examples under one broad conceptual umbrella. But since the minimal description of the first model doesn’t capture the breadth of the phenomena under consideration in even these two passages, some of the nuance is lost. The difficulty of fitting these two examples of ‘dragging’ under one description reveals that Plato has a range of different possibilities in mind for how we might interact with remembered or anticipated affections.⁹ This indicates either that the first model, when conceptualized in the broad sense just discussed, becomes too general to be able adequately to capture the distinct and psychologically salient features of the different cases it is intended to model, particularly with respect to affect, or that first model, as Warren himself characterizes it, adequately describes only a subset of the cases identified as model one examples (and, importantly, fails to describe some cases, such as the overvaluation case). More work is thus needed to specify the sense in which the model applies in any given case, and the sense in which an agent’s occurrent experience might be altered by the act of remembering or anticipating as the model describes. Warren also positions the Epicurean treatment of the memory and anticipation of pleasure under model one. They certainly seem good candidates for thinkers who conceive of memory and anticipation in a strong model one sense, as a kind of re- or pre-living. In describing their account of the Sage’s ability to re-live a past pleasure, Warren highlights that for them ‘these past pleasures are never entirely lost’ (153). This seems right, and indeed the Epicurean Sage is described as going to some length to ensure that certain particular memories of pleasure are vividly maintained.¹⁰ Yet it is not clear whether, in recalling a past pleasure, an Epicurean is to be understood as tapping into that same pleasure again, or rather as generating a new, occurrent pleasant affection by thinking again of whatever object experience caused the first occasion of pleasure. The language of re-living suggests the former—that they think of themselves as somehow having that first experience again when they recall it, such that it is that original pleasure that is being dragged into the present and re-lived. Warren’s talk of past pleasures not being entirely lost, while ambiguous, suggests that the emphasis is on the pleasure.

⁹ We might here notice how odd ‘dragging’ looks, if taken in a strong sense, as a description of the recollection process and where the object experience is brought forward. That object experience seems to include elements of the object itself, i.e. in the case of eating a certain food, the taste of that food. Where we experience a change of taste, it seems we can remember with fondness eating the thing we used to like and took pleasure in. Were we to drag that object itself, or the specific pleasure attached to it, into the present, it is odd to think that we would experience pleasure at all, let alone that same pleasure. Examples such as this, and many others we are familiar with, strain model one’s ability to describe our pleasure recollection experiences. Perhaps Warren doesn’t intend ‘dragging’ in this literal sense, since the term itself is vague. But its vagueness seems designed to cover a range of different cases which are importantly different, and this loose language obscures that. ¹⁰ See, for instance, Plutarch, Non Posse 1089A.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

   

177

But in the examples Warren highlights, including Epicurus’ deathbed account of having recollected a past pleasure in order to counterbalance his extreme pain, the target of memory is described as dialogismoi—‘conversations’.¹¹ Elsewhere Cicero quotes Epicurus describing them as rationes inventaque—‘my theories and discoveries’, and scripta et inventa—‘writings and discoveries’.¹² So the texts rather suggest that the target of the recollection process—its intentional object—is what generated past pleasures, rather than those past pleasures themselves. Recalling the theories/conversations/discoveries suggests bringing their content to mind again, which could presumably generate pleasure in the present. Since the Epicureans consider the absence of pain and distress to be the greatest pleasure (Ep. Men. 131), we might think that what is recalled are arguments against the fear of death, pain, the gods, etc. It’s at least conceivable, then, that in a moment of need (e.g. while suffering on one’s deathbed) an argument against fear of death or pain could be recalled in order to alleviate present distress and generate occurrent pleasure. Of course, much would have to be said to fill in the picture, but this seems at least a plausible account of the Epicurean strategy. And while these recollections seem to inherit or reduplicate the affective character of the original object experience—which is itself an interesting outcome—the affections felt in the present are not thereby identical with those that belong to the object experience. The recollections of past discussions seem, instead, to generate new affections. Here then, still under the guise of model one, we have yet another account of what dragging pleasures and pains of the past and future into the present means. In looking closely at just a few examples of model one processes, we can already see the model straining to accommodate such different experiences. It’s not clear that it can, or that it should. These strike us as cases where model one either fails to accurately capture the phenomena being described in the text, or fails to accommodate the full range of phenomena described across texts. We have seen examples where the phenomena are either not appropriately described as cases of dragging or re-living, or where the affective character of the anticipated or recollected pleasure is not identical to that of the object experience. The attempt to fit them under one umbrella reveals how interestingly different the cases treated by ancient philosophers are, each serving to add a new aspect to the picture of memory and/or anticipation. It also suggests that how one conceives of pleasure, objects of pleasure, and the psychology of memory cannot be avoided as part of this discussion if we want to devise and evaluate models to capture the phenomena at stake, and to understand the views on offer in these texts, and what might unite them.

¹¹ Diogenes Laertius Lives 10.22: Letter to Idomeneus [Usener 138]. ¹² Cicero Fin. 2.96–7, quoting the Letter to Hermarchus, trans. Rackham (1931).

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

178

    .  ’ 

4. What then if we turn our attention to the second of Warren’s models? This model was initially characterized by Warren in terms of the affective character of an act of memory or anticipation—where the affective character can differ from that of the object experience—so, if our reconstruction of Warren’s account is correct, this model should then deny that memory and anticipation are forms of re- or preliving. Yet whilst Aristotle’s Rhetoric provides a clear example of this second model (as characterized in terms of the affective character of an act of memory or anticipation), it is less clear, in his other psychological works, whether Aristotle endorses or denies the view that memory and anticipation are forms of re- or preliving. Indeed, Warren himself acknowledges that at times Aristotle appears to be ‘working very much in the spirit of the model of the kind that is at work in Plato’s Philebus’ (156)—though Warren ultimately wants to distance Aristotle from the re- or pre-living model he here attributes to Plato. On Warren’s account, Aristotle can explain how acts of recollecting or anticipating tend to produce an affective response thanks to his account of phantasia and its role in recollection and anticipation. As he explains, the experience of pleasure and pain depends ‘on the perception of a certain pathos’ (156–7), and since phantasia—which is involved in acts of memory and anticipation—is ‘a form of weak perception’ (αἴσθησις τις σθενής, Rhet. 1370a27–8), this allows Aristotle to explain how acts of memory and anticipation can generate occurrent pleasures and pains. Moreover, we might suppose that if a certain sort of perceptual content produces a certain sort of response, then the recollection or anticipation of that content will tend to have a similar affective character to the actual perception of that same content; hence why things that are pleasant when present tend also to be pleasant when anticipated or remembered (1370b9–10). Warren insists, however, that there is no necessary connection between the affective character of the object experience and that of the act of recollection or anticipation. Other scholars, though, are inclined to see a closer connection between Aristotle’s account of memory and anticipation and the re- or pre-living model supposedly on offer in the Philebus.¹³ For according to one interpretation of Aristotle’s account of phantasia, and, in particular, the relation between phantasia and perception, phantasia is not merely a weak form of perception, but involves the preservation of perception. And not only does it preserve the narrowly representational content of perception, but it preserves its affective component too.¹⁴ Yet if phantasia involves the preservation of perception, including its affective component, then this suggests that when a subject recollects an experience she will effectively re-live ¹³ For example, Moss (2012), pp. 58–9. ¹⁴ This view is endorsed by Moss (2012). See esp. pp. 57–9.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

   

179

that experience—including its affective component—which would align Aristotle’s psychological works with the first and not the second of Warren’s two models. Here, then, a more detailed examination of Aristotle’s account of phantasia is required in order to settle whether Aristotle’s psychological works are more closely aligned with model one or model two (though this is not something we will undertake in this context). At the same time, Warren’s paper helps us to better appreciate some of the issues that are at stake when trying to determine the relation between phantasia and perception, and might even inform such investigations. For insofar as Warren’s paper draws attention to those occasions when the affective character of an act of recollection or anticipation clearly differs from that of the object experience, it might offer reasons to resist or at least modify an account of phantasia as involving the preservation of perception in a strong sense (where that includes the preservation of its affective component, say). Still, we would like to know how, on Warren’s view, we should interpret those passages in which Aristotle speaks of ‘imprinting’ (e.g. Mem. 450a27–32) if this is not to be understood as a form of preservation (particularly of an affection) or to entail some sort of re-living. Setting this issue to one side, what is it, then, that explains those occasions when the affective character of (say) an act of recollection differs from that of the original experience? And is Aristotle’s explanation the same in all cases? One explanation that Warren offers on Aristotle’s behalf appeals to Aristotle’s claim that when we experience phantasia ‘we are like people looking at a picture’ (Mem. 427b24). For this implies that when we remember or anticipate (or at least when we do so deliberately) we ‘stand’ towards that content as we do to a picture when looking at it, ‘presumably, seeing what we call to mind in some way as a depiction’ (159–60). Hence, Aristotle can allow ‘the possibility of such a state of mind having a different effect from the direct perception of whatever it is that is depicted this way’ (159–60).¹⁵ ¹⁵ But in what sense is the ‘effect different’ from the direct perception? Returning to the question we raised in §3, it is not clear whether Aristotle’s point is that seeing something as a depiction influences the intensity of the affective response (so that in seeing something ‘as a depiction’ we will experience little or no affect in comparison to direct perception—or, by analogy the original experience) or whether he is claiming that seeing something as a depiction also influences the valence of the affective response (so that, in seeing something ‘as a depiction’ we will experience pleasure, when the direct perception of that scene would ordinarily produce pain, cf. Poetics IV 1148b15–17). The Greek does not obviously determine the issue: Aristotle contrasts our experience of phantasia with cases in which we have a belief that something is terrible or frightful, when we are ‘correspondingly affected right away’ (εὐθὺς συμπάσχομεν). The ‘εὐθὺς’ might push us towards reading the contrast as between a case in which we are affected vs one in which we are not. Alternatively, the ‘συμ’ prefix of συμπάσχομεν might be taken to emphasize the sense in which the affective response corresponds to its terrible or frightful nature (that is, the experience is painful), suggesting that the contrast case is where it does not have the same valence. Perhaps all Warren needs the example to show is that ‘that there is no necessary connection between the painfulness of a past experience, for example, and the affective response that a person will feel when he or she chooses to recall that past experience’ (160). But since it is unclear whether the texts surveyed under model one are committed to there being such a necessary connection, we might once again wonder where the contrast between the two models then lies. We will return to this thought in §5 below.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

180

    .  ’ 

Warren also draws attention, however, to Aristotle’s claim in the Rhetoric, that we can recall with pleasure an event that was once painful, provided that ‘what comes after this was fine and good’ (1370b3). Of the various ways of interpreting this claim, the most compelling (to our minds) is the suggestion that ‘in retrospect we see the event itself—not just its consequences—as something positive and beneficial’ (161), which Warren unpacks by appeal to Aristotle’s example of Eumaeus in the Odyssey. As Warren explains: ‘Eumaeus “sees” the events in his past differently from the way he originally saw them (or they “appear to him” differently) and hence he now experiences pleasures as he re-envisages them while he originally experienced pains’ (165). We find this thought particularly resonant given Aristotle’s interest in moral development and the plausible interpretation of this process as one in which the developing subject comes to see the world, including her own experiences, differently. But how, if at all, are these explanations related to one another? Whilst Warren uses the language of ‘seeing as’ in both cases—seeing something ‘as a depiction’, seeing something ‘as fine and good’—what this language of ‘seeing as’ picks out in either case seems to be importantly different—and perhaps that is what Warren intends. In the second case, the agent’s seeing something ‘as fine and good’ indicates that there has been a reevaluation of the content that is recollected (or, indeed, on a rich view of perceptual and imaginative content, a change in the content of what is recollected itself ).¹⁶ In the first case, however, we wonder if the thought is, rather, that when an agent deliberately recollects something, this involves a certain sort of psychological distance—perhaps because the deliberate recollection of an experience involves an awareness of the agent as looking at whatever is depicted; an awareness of the higher-order nature of the looking, as it were—and that this distance is what allows for the possibility that the recollection will have a different effect from the direct perception of what that content corresponds to. Are these, then, simply independent explanations of how the affective character of an act of recollection can differ from that of its object experience, such that the second model, when its positive commitments are identified, is ultimately disjunctive? Or is there some relation of dependency between the two? To put it more concretely, could a re-evaluation of some content that is recollected take place without the psychological distance afforded by deliberate recollection, or do we need that psychological distance, afforded by deliberate recollection, in order to re-evaluate the content? Likewise, is the psychological distance afforded by deliberate recollection sufficient to effect a change in the valence of the affective response,¹⁷ or do we need there to be a ¹⁶ It’s worth considering whether this is necessarily at odds with the interpretation of the Epicurean texts hypothesized in §3, where the point might be that, so long as an agent’s evaluation of some content remains the same (as it would for the Sage), recollection of that content will produce the same sort of response as the original experience. ¹⁷ See our comments in n. 15 above.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

   

181

re-evaluation of the content to account for this further change? Unlike our criticism of the first model, these questions are raised not as a challenge to the internal coherence of this second model, but rather in the spirit of further inquiry. Given the role that memory and reflection on past experiences surely plays in the process of moral development, an answer to these questions might shed important light on this process, in various ways.

5. So where do things now stand with regard to Warren’s two models? In this response we offered a reconstruction of the contrast between the two models we think Warren has in mind and tested it against the texts that he discusses, and in which these models are supposed to originate. When we did so, however, we found that the models, as reconstructed, struggled to fit many of the texts and that the distinction between the two models threatened to collapse. Model one looks particularly problematic, and one might even wonder—given the availability and plausibility of model two—why a thinker (such as Plato or Epicurus) would ultimately endorse the more vexed model one. But if the result of this analysis is that the distinction between these models threatens to collapse, or one ends up striking us as simply implausible, then this would be extremely disappointing. Because—taking a step back—we do think that Warren has put his finger on something important in identifying these two models. We want to end, therefore, by offering a (tentative) suggestion, in the hope that it might help to preserve something of the distinction Warren identifies, and with it at least one of the important insights of his paper. Whilst aspects of Warren’s discussion might lead us to think that these two models represent competing accounts of the nature of memory and anticipation— from which certain commitments about the affective character of memory and anticipation follow—we wonder whether the two models Warren identifies in the ancient texts are not, instead, two models of how each of us can remember and anticipate.¹⁸ That is, there is one way in which we remember or anticipate that is, in a sense, immediate, and which has a certain affective power. When we remember or anticipate in this way, the act of memory or anticipation seems to inherit the affective character of the object experience, both in terms of its valence and, to some extent, its intensity. We do, in a sense, re- or pre-live the experience. But there is another way in which we can remember or anticipate which involves a certain sort of psychological distance, and when we recollect or anticipate in this way, there is no necessary connection (as Warren puts it) between the affective

¹⁸ Or ‘modes’ of remembering and anticipating, if you will.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

182

    .  ’ 

character of the act of recollection or anticipation and that of the object experience. If this is correct, then this might explain why we find a given thinker talking in one place of memory and anticipation as though these are forms of re- or pre-living, whilst in another place talking about remembering or anticipating as something more detached, and which can differ in affective character from its object experience. The texts in which the memory of pleasure and pain are discussed need not be taken to reflect the commitments of an author’s antecedent metaphysics of memory (though some may, of course); rather, the particular contexts of these discussions may simply lead an author to focus more on one way in which we remember and anticipate, or another—since a given agent can do both, there is no need for an author to ‘choose’ between one model or the other. This suggestion is, of course, provisional and leaves much unsaid. What philosophical contexts lend themselves to discussion of one model or the other? How do these different ways of thinking about memory and anticipation relate to a more general account of the metaphysics of memory and anticipation? What sort of influence does the context of recollection or anticipation (to which Warren rightly draws attention) have on either one of these models? These are all issues for further investigation. But we wonder if our suggestion is one that Warren might welcome, and which might help to preserve the many insights we believe his analysis offers.

References Cicero. 1931. De Finibus Bonorum Et Malorum (Fin.). Translated by. H. Rackham. Loeb Classical Library, Cambridge MA: Harvard University Press. Moss, J. 2012. Aristotle and the Apparent Good. Oxford: Oxford University Press. Warren, J. 2011. ‘Socrates and the Patients: Republic IX 583c–585a.’ Phronesis 56: 113–37. Warren, J. 2014. The Pleasures of Reason in Plato, Aristotle, and the Hellenistic Hedonists. Cambridge: Cambridge University Press.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

12 Three Mistakes about Stoic Ethics Daniel C. Russell

According to the ancient Stoics, living virtuously is all one needs for living well— for what the Roman Stoics called vita beata and the Greek Stoics eudaimonia, which today is usually rendered “happiness.” Cicero reconstructs the Stoics’ argument in compressed form: [1] We do not believe that wisdom is like navigation or medicine, but instead like acting or dancing, inasmuch as the goal—namely, the performance of the skill— is contained within it, not sought externally. [2] However, no one can have the virtues without resolving that there is nothing that makes a difference or differentiates one thing from another, except for virtue and vice. [3] Now let’s see how clearly these things follow from what I have put down: inasmuch as the end is to live in agreement and unity with nature, it follows necessarily that all wise persons always live happily, freely, prosperously, obstructed by nothing, held back by nothing, in need of nothing. (de Finibus [Fin.] III.24–6, abridged)¹

Cicero presents two Stoic theses about virtue: (1) that the exercise of virtue requires no special circumstances outside one’s control—virtue is “self-sufficient,” we might say—and (2) that virtue and vice are the only things that aren’t “indifferent”; that is, that only virtue is good, and only vice is bad. And (3) from these two theses about virtue, he says, follows the Stoics’ thesis about happiness. That is: (1) The Stoics believe that virtue is self-sufficient, (2) and that virtue is the only good, (3) and that explains why they believe that virtue is sufficient for happiness. It is therefore surprising, when we turn to modern reconstructions of the Stoic view, to find how sharply they contrast with Cicero’s. For instance, in her influential 1994 book, The Therapy of Desire, Martha Nussbaum proposed that the Stoics maintained that virtue is the only good, because otherwise virtue could not guarantee happiness:

¹ All translations are my own, except as noted. Daniel C. Russell, Three Mistakes about Stoic Ethics In: Psychology and Value in Plato, Aristotle, and Hellenistic Philosophy: The Ninth Keeling Colloquium in Ancient Philosophy. Edited by: Fiona Leigh and Margaret Hampson, Oxford University Press. © Oxford University Press 2022. DOI: 10.1093/oso/9780192858108.003.0012

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

184

 . 

Virtue by itself is self-sufficient, sufficient for eudaimonia. But the Stoics, like Aristotle, also hold that eudaimonia is, by definition, inclusive of everything that has intrinsic value, everything that is choiceworthy for its own sake. Putting these claims together, we are forced to conclude, what a large number of texts in fact assert, that external goods, all goods other than virtue, have no intrinsic value at all. (Nussbaum 1994, 361)

That is: The Stoics believe that virtue is sufficient for happiness, and that explains why they believe that virtue is the only good.

Tad Brennan concurs: Only virtue is good, only vice is bad; the rest are all indifferents of various sorts. How do the Stoics argue for this claim, and why should we believe it? The Stoics employed many strategies . . . . [Some of these] take as their premiss the claim that virtue must be sufficient for happiness, or that the happiness of the Sage must be completely immune from the vicissitudes of fortune, and go on to say that if wealth or health or the other indifferents were good, then the Sage’s virtue would not guarantee happiness. (Brennan 2005, 121–2)

Likewise Terence Irwin: If we regard other things as goods, we must regard them as parts of happiness. In that case, we have to ask whether we ought ever to prefer some combination of these goods to virtue. If we raise that question, we do not live wholly in agreement with reason, as virtuous people do. If we do not live in agreement with reason, we cannot achieve happiness. Hence nothing except virtue is good. (Irwin 2020, 65)

And John Sellars: [P]erhaps the most important argument [for the thesis that only virtue is good] is that the possession of externals cannot guarantee us happiness, but the possession of virtue can, for the Stoics. (Sellars 2006, 111)

And William Stephens: Virtue, [the Stoics] insisted, is the only good because it alone is necessary and sufficient for happiness. (Stephens 2020, 35)

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

    

185

Notice three differences between the modern reconstruction and Cicero’s—and, I argue, the modern reconstruction’s three mistakes. First, the modern reconstruction reverses the order of explanation between the thesis that virtue is sufficient for happiness and the thesis that virtue is the only good, taking as premise what Cicero takes as conclusion. Second, the modern reconstruction mishandles the thesis that virtue is self-sufficient. On Cicero’s reconstruction, that thesis is distinct from the other two. By contrast, Nussbaum’s reconstruction runs the self-sufficiency of virtue—that virtuous activity is “unaffected by worldly contingency”²—together with the sufficiency of virtue for happiness; others leave out self-sufficiency altogether. And third, the ancient reconstruction offers a logical rationale for the Stoics’ belief that virtue is sufficient for happiness, whereas the modern reconstruction leaves us puzzling over that very belief. It is these three features of the ancient reconstruction that I discuss in the three sections of this essay. The reconstruction of the Stoic argument is important for our understanding not only of Stoic ethics, but also of the trajectory of ethical thought in the ancient world—and, I think, its significance for us. On the modern reconstruction, the Stoic view assumes prior commitment to the sufficiency of virtue for happiness, but then for anyone with no such prior commitment the argument is utterly unconvincing—and therefore the enormous influence of Stoicism on ancient ethics is very difficult to fathom. Our ancient sources point another way: it is staking our happiness on anything but virtue alone that is the source of degradation and wrongdoing. That is why the Stoics believed that the only good is virtue, and that virtue is something self-sufficient. And on this ancient reconstruction, it becomes clear both why the Stoics believed that virtue is sufficient for happiness, and why the Stoic argument was—and remains—one to be reckoned with.

1. The Order of Explanation The first point is quick to make, as Cicero clearly reconstructs the Stoics’ argument as moving from their understanding of virtue to their thesis about happiness: “Reason has shown that virtue is the only good. This being so, the wise person must always be happy” (Fin. III.75). This order of explanation is typical in ancient reconstructions: [The Stoics] say that every fine, good man is complete since he lacks no virtue; . . . For this reason, among people the good ones are always completely happy. (Arius Didymus [AD] in Stobaeus, Eclogae II.11g) ² Nussbaum 1994, 363.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

186

 . 

If the Stoics have delineated the good correctly, the matter is settled: a wise person must always be blessed. (Cicero, Tusculanae Disputationes [TD] V.83)

And, of course: Inasmuch as the end is to live in agreement and unity with nature, it follows necessarily that all wise persons always live happily, freely, prosperously, obstructed by nothing, held back by nothing, in need of nothing. (Fin. III.26)

This last passage is especially noteworthy, since it does not just say that the Stoic understanding of virtue explains the Stoic view of happiness. It also suggests why the explanation runs in that direction: the greatest good is to live in agreement with nature, and there is nothing that can keep a virtuous person from living that way. This latter thesis is what I called the self-sufficiency of virtue, above, and Cicero presents it more carefully elsewhere: You don’t know, you madman! You don’t know what power virtue has; you just use the word “virtue”; you’re unaware what strength the thing itself has. No one can fail to be completely happy if he depends only on himself and places everything that is his within himself alone. (Paradoxa Stoicorum 17)

This passage is extraordinary. It comes on the heels of Cicero’s discussion of the familiar, paradoxical Stoic thesis that virtue is the only good (§§6–15). In our passage, Cicero has moved to a second paradox, that virtue is sufficient for happiness. Again, it is significant that Cicero sets up his discussion of the sufficiency thesis by first discussing the unique goodness of virtue, and not the other way around. But it is also significant that Cicero does not say that settling the first paradox simultaneously settles the second.³ Rather, he now introduces into the discussion the further Stoic thesis that virtue is self-sufficient, not dependent on any special external circumstances.⁴ Simply put, Cicero makes it explicit that the Stoic thesis that virtue is sufficient for happiness depends on a pair of ideas about virtue: one about what kind of value virtue has, and one about what kind of thing virtue is. So, then, what exactly is the self-sufficiency of virtue, and why would the Stoics commit to it?

2. The Self-sufficiency of Virtue 2.1 What is self-sufficiency? The Stoic thesis that virtue is the only good cannot by itself establish that virtue— that is, living virtuously—is sufficient for living a happy life. The reason is this: ³ See also TD V.18–19, 33–4.

⁴ See Irwin 1986, 228–34; Lesses 1989, 100.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

    

187

even if living virtuously is the only good thing, so that happiness consists in virtuous activity alone, it is a further question whether living virtuously might depend on circumstances we cannot control. So, for one thing, in making virtue self-sufficient, the Stoics rejected the thought that if worldly circumstances keep us from reaching the goals that we, being virtuous, pursue through our actions, then those actions will not count as virtuous in the first place.⁵ However, even with that thought set aside, a further issue remains for the Stoics, one that concerns the kind of activity that virtuous activity is. Imagine, for example, a woman who is also a mother, and virtuous as a mother: how is that activity—that activity of raising this child of hers—related to her happy life? One view that might seem attractive is that virtuous activity is “indexical,” so to speak: focusing on her motherhood, we might say that her living happily is identical to her living virtuously within this relationship with this unique child. To say that her activity is virtuous is to observe the wisdom with which she lives; to say that it is one of the activities that makes up her happy life, though, is to identify it in all of its particularity for her. So, for instance, if this child should die, she would lose that activity—and therefore that happy life. On such a view, living virtuously is inseparable from one’s particular circumstances, including circumstances beyond one’s control.⁶ By contrast, a more austere view would make virtuous activity purely “adverbial”: it is to act wisely within circumstances as one finds them, whatever they may be. Changes in those circumstances may be terrible, but because virtuous activity is not inseparable from any particular circumstances, it is not interrupted by such changes, and so neither is the happy life. Notice that, on the “indexical” view of virtuous activity, from the thesis that living virtuously is the only good it does not follow that living virtuously is sufficient for happiness. For a woman who is a mother, this relationship with this child is the very form and mode of this activity that is “this specifically maternal happiness” of hers, in C. S. Lewis’ phrase.⁷ But in that case, her happiness will be vulnerable to loss precisely because that activity is vulnerable—even if, please note, virtuous activity is the only good, and even if its counting as virtuous does not depend on outcomes we cannot control. The issue is not whether the loss of those relationships would make one less virtuous, but that such a loss would terminate those activities that are the same thing as that happy life one is now living. In other words, the issue is not how to understand the virtuousness of virtuous activity, or even the unique goodness of virtuous activity. The issue is how to understand the kind of activity that virtuous activity is, insofar as virtuous activity can be the same thing as a happy life. So, this “indexical” way of thinking about virtuous activity is one that the Stoics must rule out before they are entitled to claim that virtue is sufficient for happiness. Instead, they would need to understand virtuous activity as wise ⁵ On this thought in Aristotle’s ethics, see Irwin 1990. ⁷ Lewis 1961, 38f.

⁶ See Sherman 1987; Russell 2012.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

188

 . 

selection and response with respect to the world around one, as one finds it, so that such activity is uninterrupted even when worldly circumstances change. On this purely “adverbial” view, the loss of a particular other may be the loss of something significant, but what it cannot be is the interruption of the activity in which one’s happiness consists. That activity just is the practically wise and emotionally sound exercise of the faculty of choice, and it takes no particular object in order to do that. Put another way, virtuous activity always takes some object, but there is no object that it always takes. Therefore, the Stoic argument for the sufficiency of virtue for happiness must include both the premise that virtuous activity is the only good and also the distinct premise that virtuous activity is independent of such fragile connections to the perishable world, on the grounds that virtuous activity lies entirely in how one exercises choice. And that latter premise is what we have called the selfsufficiency of virtue.

2.2 What “participates” in virtue We can see that the self-sufficiency of virtue contributes a distinct point in another way as well. According to the Stoics, only virtue is good, only vice is bad, and everything else is what they call “indifferent,” neither good nor bad. Now, if living virtuously is to be understood as including such things as relationships in all their particularity, then those relationships will be goods, and part of living well. But for the Stoics, such relationships are indifferent; it is possible for them to take on the wrong role in one’s life, and that isn’t possible for good things. So, if virtue is the only good, doesn’t it follow directly that virtue is sufficient for happiness? Actually, no, and for a surprising reason: for the Stoics, it is possible for an indifferent thing to become part of virtue. If such indifferent things as relationships do not become part of virtue, then the Stoics must explain why not. That is to say, the Stoics must explain why virtue does not include such things but is instead self-sufficient. Let me begin by explaining how it is for the Stoics that indifferent things become parts of virtue at all, before explaining why such indifferent things as relationships do not (§2.3). To use the Stoic vocabulary, our first question concerns what things might “participate in” virtuous activity; to understand “participation,” we must begin with the Stoic theory of value. The Stoics identify three value categories: what is good, what is bad, and what is indifferent. According to Arius Didymus, by “what is good” the Stoics mean virtue and whatever “participates” in virtue (AD II.5a–5b).⁸ ⁸ “Participates in virtue”: metechon aretēs. See also Epictetus, Discourses (Diss.) II.9.15; II.19.13; Sextus Empiricus, Adversus Mathematicos (M) XI.22–3.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

    

189

Arius Didymus (II.5b) says that by “virtue” the Stoics mean both the cardinal virtues (wisdom, justice, temperance, and courage), which are forms of knowledge or skill, as well as certain attributes and capacities of a person with the virtues (or the “supervenient byproducts” of virtue), such as the health, soundness, strength, and beauty of a virtuous soul (II.5b4). The other sort of good is what “participates in” virtue, which they identify as various forms of emotional soundness, that is, the eupatheiai: joy, good spirits, confidence, and wish (II.5b).⁹ All of these things count as good because virtue is good, and these are its parts and attributes. Similarly, the Stoics say that the bad is vice and what participates in vice (II.5a–5b1). The vices include both the “cardinal” vices (folly, intemperance, injustice, and cowardice) and the attributes of a person with the cardinal vices (pusillanimity, weakness, powerlessness; II.5b). What “participates in” vice are unsound emotions, or what they call the pathē: pleasure, distress, longing, and anxiety (II.5b, 10b). The third and final value category belongs to everything that can be put to either good or bad use, what the Stoics call “indifferent” things. These are everything besides virtue, vice, and what participates in them: health, wealth, reputation, pleasure and pain, and even life and death (AD II.5a; DL VII.102–7). None of these things differentiates a good person from a bad one (AD II.5b9),¹⁰ and all of them require virtue to give them an appropriate place in one’s life. Or, as Cicero puts it, only virtue or vice “makes a difference or differentiates one thing from another” (Fin. III.25). Indifferent things, we might say, are “undifferentiated.” Now, my present point about these categories is this: the line between indifferent things and good things is permeable. I am not aware of this point having been made before, but the point is plain enough from the fact that the line between “good” and “indifferent” is in fact permeated, in the case of the emotional dimension of the virtues. The Stoics say that a person’s emotional life can be a good thing, because some emotions “participate in virtue” (AD II.5b). But of course, no emotional life ever starts out that way. Every emotional life, by its nature, starts out as an indifferent thing: just having an emotional life does not separate good people from bad, and can be given either a good or a bad direction (“used well or badly,” as the Stoics say).¹¹ Now, as an emotional life takes on a good direction, it becomes a different kind of emotional life from one with a bad direction; the Stoics make it clear that the eupatheiai are not just redirected pathē.¹² In other words, the line between good things and bad things is not permeable. However, as one’s “raw” emotional life—an indifferent thing—is ⁹ See also Diogenes Laertius (DL), Lives of Eminent Philosophers VII.96. ¹⁰ Cf. Sextus Empiricus, M XI.200–1, 207. ¹¹ I agree with Rist (1969, 49) that pleasure when treated as an indifferent (e.g., DL VII.102) is pleasure in this generic sense, of which the pathos and the eupatheia (joy) are species. ¹² See esp. Seneca, On Anger I; Cicero, TD IV.38–42, 48–55, 77–8. For discussion, see Frede 1986; Graver 2007, chaps 2–3.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

190

 . 

cultivated correctly, it is transformed into a part of virtue itself—it “participates” in virtue—and thus it becomes differentiated as one of the good things. In fact, Arius Didymus says that the Stoics classify being joyful (chairein) and in good spirits (euphrainesthai) among the morally perfect actions (katorthōmata; AD II.11e). It is for this reason, I think, that Arius Didymus understands “caution” both as one of the eupatheiai, along with joy and good spirits, and as a form of the virtue of courage (II.5b and 5b2, respectively).¹³ Likewise, the Stoics also say that the eupatheiai are parts of happiness: as Arius Didymus puts it, joy, good spirits, and confidence are goods that are “fulfilling” of happiness (II.5g).¹⁴ To be sure, they are not parts of happiness in just the same way as the virtues, which are both “fulfilling” and “productive” of happiness. But this is what we should expect: the eupatheiai are parts of happiness because they are an emotional life under the direction of something else, namely right reason, and are parts of virtue itself, whereas virtue is such as to be part of happiness just by its very nature. There is an important lesson here: indifferent things can become good things. At any rate, this is what happens in the case of the eupatheiai. Insofar as something indifferent in its own right can become part of—“participate in”—virtuous activity itself, it can become a good thing and part of happiness. Now, to say that both virtuous activity and the eupatheiai are good things is not to locate happiness in some pair of things. Rather, the addition of things like joy serves to amplify the nature of virtue itself. The eupatheiai are necessary for happiness, then, not in addition to virtue, but because they constitute an emotional life that has become part of virtue itself.¹⁵ Therefore, even if virtuous activity is the only good, when we consider what happiness consists in the question becomes what sorts of things can become parts of virtuous activity. We might agree with the Stoics that it is only insofar as things can become transformed into virtuous activity that we should think of them as potential parts of happiness in the first place. After all, that is how the eupatheiai come to be parts of happiness. But then, in principle the very same point could apply to other things too, like a mother’s relationship with this child of hers. It all depends on what things can be understood as internal to virtuous activity. And that is why the Stoic premise of self-sufficiency—that the virtuous person keeps “everything that is his within himself alone”—is so crucial to the Stoics’ case for the sufficiency thesis. That premise adds the distinct idea that nothing outside oneself—and so, out of one’s control—can become part of virtue and, thereby, part of happiness. ¹³ See also Tsekourakis 1974, 91. ¹⁴ See also Cicero, Fin. III.55. ¹⁵ It is for this reason I think that Arius Didymus (II.6d) says that for the Stoics, joy and good spirits are not “necessary” for happiness. He certainly cannot mean that it is possible to have the virtues without having the eupatheiai; rather, he means that they are not things that have to be added on to virtue for the sake of happiness, as if virtue lacked something necessary. But con. Tsekourakis 1974, 96–7. I would understand Seneca, De Vita Beata 15.2–3 in a similar way (but con. Lesses 1989, 106).

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

    

191

2.3 Epictetus and self-sufficiency For the Stoics, only those indifferent things that are within oneself—one’s emotions—can become parts of (“participate in”) virtue. But then, why are the Stoics so restrictive about which indifferent things are so eligible? Or, why do they so restrict what can be “within oneself ”? The reason, according to Epictetus, is that being any less restrictive would compromise virtue itself. Epictetus’ Discourses opens with the observation that there is only one faculty or skill that concerns not merely how to do things (as the skills of music and grammar do), but indeed whether, when, and what to do in the first place (I.1.1–6). Epictetus describes this faculty as “the correct use of impressions” (I.1.7), and he identifies it with the human capacity for choice (prohairesis, II.23.9), which he sometimes calls simply our “assent” (sunkatathesis; e.g., IV.1.72).¹⁶ According to Epictetus, this faculty of choice is the only thing that is ours to control (I.1.7–13); nothing outside this faculty, he says, is up to us.¹⁷ So, for instance, when Epictetus says that choice is up to us, he does not mean that it is up to us to go for a walk (say) when we so choose, but only that it is up to us to assent to the impression that walking is the thing to do (IV.1.69, 72–3). Put another way, what is up to us is not our engaging the world in certain ways, but how we make our choices (e.g., I.6.40; III.10.9).¹⁸ Now, Epictetus says not only that the use of impressions is all that is mine to control. He also says that the power of using impressions is all that is me. Epictetus imagines a tyrant who threatens to put him in fetters if he does not betray a secret, and he replies that the tyrant may fetter his leg but he cannot fetter him—that would be to fetter his faculty of choice, and not even Zeus himself can do that (Diss. I.1.22–5; cf. I.18.17).¹⁹ This is because, as Epictetus says, “you are not flesh and hair, but choice” (III.1.39). This is not a metaphysical claim; Epictetus does not mean that he cannot be fettered because he is something immaterial—no Stoic believes any such thing. His point is not psychological, either; he is aware that people often do take themselves and their happiness to include things like their bodies, their property, their friends, and their loved ones (cf. I.22.9–11).²⁰ Rather, his point is about practical wisdom: we have a choice about the things we make

¹⁶ As Long says, for Epictetus “choice” (prohairesis) is “the human mind . . . in just those respects that are dependent on nothing that we cannot immediately judge, decide, and will, entirely for and by ourselves” (Long 2002, 212). See also Rist 1969, 299; Dobbin 1991. ¹⁷ See also Diss. II.19.32. Epictetus holds that something is “up to us” (eph’ hēmin) just in case it is something that we can do and that nothing could keep us from doing. See Sorabji 2000, 215, 331–2; 2006, 188–91; Stephens 2007, chap. 1. ¹⁸ We find the same idea in Marcus Aurelius, Meditations (Med.) II.2 and V.26; see Gill 2008, 362 for discussion. Indeed, for Epictetus the soul is not only a gift of a god, but is itself a kind of god; see Dyson 2009 for discussion. ¹⁹ See Sorabji 2006, 47; Stephens 2007, 23–5. ²⁰ Cf. Stephens 1996, 200; Sorabji 2000, 245; 2006, 181–2.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

192

 . 

our own, for better or worse (see Handbook 1). For Epictetus, I am my faculty of choice insofar as this is the wise way to think about myself and my happiness. For our purposes, the crucial point is that this—this demand of wisdom—is why Epictetus denies that anything outside the faculty of choice could become good. To be sure, he also depicts virtue as a magic wand that turns everything it touches into a good, instead of into gold (Diss. III.20.12). However, he makes it clear that what virtue actually makes good is not the thing that the wand touches; rather, virtue makes its possessor good where that thing is concerned (III.20.11–15). For instance, an external thing like a relationship with a brother or father can become “good”—that is, I can become good in the choices I make with respect to my brother or father, even a rotten one (I.15; III.3.7–10; Handbook 30). This perspective stems from the Stoic belief that virtue is a skill or craft: “as the carpenter’s material is wood,” Epictetus says, “and the sculptor’s is bronze, so the material of the skill of living is each person’s own life” (Diss. I.15.2; cf. III.3.1; DL VII.86).²¹ So, let’s summarize what the Stoics say about the line between indifferent things and good things. First of all, an indifferent thing can itself become a good thing, but only insofar as it can “participate in” virtuous activity (§2.2). However, in order for us to be wise, something can participate in virtuous activity only insofar as it is itself a kind of exercise of the faculty of choice. That is why the eupatheiai can participate in virtue and things like health—or relationships—cannot. For indifferent things of the latter sort, then, the only sense in which they can be “good” and thus important for happiness is that one can be good in the choices one makes where such things are concerned. Therefore, on the Stoic view the virtuous activity in which happiness consists is strictly the virtuous exercise of choice. That is why they believe that virtue is sufficient for happiness—not just because virtuous activity is the only good, but because virtuous activity is also just the exercise of a faculty that is always within our control. But now we have another question: why would the Stoics understand virtue as they do, if their understanding has the radical implication that virtue is sufficient for happiness? Why accept the sufficiency of virtue for happiness as a conclusion, ²¹ See also AD II.5e, 5m, 11q; DL VII.86, 95. This seems to be the point of Arius Didymus’ distinction between “mixed” and “unmixed” goods (II.5m): the “unmixed” goods are the virtues, understood as forms of knowledge or expertise (II.5b–5b2, 5b5), and they are called “unmixed” because they are not the mixture of indifferent materials and a differentiating force applied to them; they are themselves that very differentiating force. The “mixed” goods, on the other hand, are indifferent things—such as having children, being old, being alive—that one has “used” well, that is, in relation to which one is wise and virtuous. But of course, in saying that such things become “goods,” Arius Didymus makes it clear that it is not the things that become good, but the virtuous person in relation to them. Con. Reydams-Schils (2005, chap. 2), who takes passages like AD II.5e to say that certain relationships are goods in the straightforward sense. However, she does not consider passages like AD II.5m, which understand that goodness strictly as one’s own virtuous conduct with respect to those relationships.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

    

193

instead of taking it as a reason to question the premises? Here we move to our third and final point of reconstruction, concerning philosophical motivations for the Stoics’ view.

3. Why the Stoics Make Virtue Sufficient for Happiness 3.1 The modern reconstruction The reconstruction of the Stoic position that I have developed here departs sharply from the seminal reconstruction we find in Nussbaum: It is particularly important to understand that when the Stoics deny all value to items other than virtue they are including here all the items to whose presence or absence the contingencies of the external world can make a difference . . . . Not only traditional “external goods” like wealth and honor, not only “relational goods” like having children, having friends, having political rights and privileges, but also individual forms of virtuous activity, such as acting courageously, justly, and moderately, are held to be, strictly speaking, worthless, on the grounds that they can, as Aristotle has argued and as anyone knows, be cut off or impeded by accidents beyond our control. But the wise man must be self-sufficient; his life is always eudaimōn, no matter what happens. (Nussbaum 1994, 362–3)²²

I have argued that the modern reconstruction of the Stoic argument reverses the order of explanation between the nature of virtue and virtue’s sufficiency for happiness (§1), and that it gives short shrift, at best, to the thesis that virtue is self-sufficient (§2). But perhaps the most serious mistake lies in accepting what follows after making the other two: the Stoics turn out to have nothing to say to anyone who does not already agree that virtue must be sufficient for happiness. If the Stoics conclude that virtue is the only good from the premise that virtue is sufficient for happiness, then the obvious question is why the Stoics would start out by making such an audacious claim. Brennan offers an answer: the Stoics make virtue sufficient for happiness to stave off the very “fear of vulnerability” (2005, 122). Likewise, Nicholas White says that the focus of Stoic ethics is “the management of disappointment”: because disappointment is so distressful, the Stoics urge us to avoid disappointment by adopting the view that the things that normally cause disappointment are actually indifferent and do not matter (2006, 93–7).²³

²² In support Nussbaum cites Cicero, TD V.83. But see above, §1. ²³ See also Sellars 2006, 114.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

194

 . 

I find this suggestion alarming. To anyone who does not believe that virtue is the only good, the Stoics, we are told, point out that virtue has to be the only good, because otherwise virtue would not be sufficient for happiness. To one who then replied, “But virtue isn’t sufficient for happiness,” we are told that the Stoics come back: “It must be—otherwise happiness would be vulnerable to fortune.” But what kind of answer is that, to anyone who is not already convinced that happiness must be invulnerable to fortune? I would suppose that I had misunderstood the modern reconstruction on this point, if not for the fact—even more alarming—that its proponents go on to bring precisely this objection against the Stoics. Brennan puts the objection particularly clearly: The argument’s validity is not in question, but its ability to persuade anyone is. Those of us who think that a modicum of health is an important part of complete happiness have long reconciled ourselves to the fact that virtue is not sufficient for happiness, and that no one’s happiness is immune from trauma and disease. To say that we must purge anything vulnerable from happiness lest the fear of vulnerability should undermine our happiness is to follow the route of a chef who begins by seeking only the freshest fruits, but from an increasing concern to avoid over-ripeness decides to purchase only plastic fruit instead. Some goods are perishable, and cannot be replaced with imperishable simulacra—that is a common enough view. Perhaps the Stoics can change our minds about that, but not merely by threatening us with loss or the fear of loss.²⁴

But surely the very obviousness of the objection is precisely its weakness: surely this line of thought cannot have been what kept the Stoics in the hunt in ancient debates in ethics for a few hundred years—debates in which, by all accounts, they were doing rather well.²⁵ The modern reconstruction goes against Cicero’s reconstruction only to end up making the Stoic position startlingly feeble, and thus the trajectory of ancient ethics very hard to understand.

3.2 Cicero’s testimony But perhaps some of the ancient testimony points in this direction too. In fact, Brennan cites Cicero’s Tusculan Disputations in support of this reconstruction: In my opinion, virtuous men are also supremely happy. For if a man is confident of the goods that he has, what does he lack for living happily? Or how can ²⁴ Brennan 2005, 122. See also Nussbaum 1994, 399–401; N. White 2006, 97. ²⁵ Better, in fact, than their Aristotelian rivals; see Annas 1993, chaps 18–21; S. White 2002; Russell 2012, chap. 5.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

    

195

someone who lacks confidence be happy? Yet a man who adopts the [Aristotelians’] threefold division of goods inevitably lacks confidence.²⁶ For how will he be able to be confident of bodily strength or secure fortune? Yet no one can be happy without a good which is secure, stable, and lasting . . . The man who would fear losing any of these things cannot be happy. We want the happy man to be safe, impregnable, fenced, and fortified, so that he is not just largely unafraid, but completely.²⁷

To be sure, Cicero does take as a premise that the happy person must be “safe, impregnable, fenced, and fortified,” but the question is what the happy person must be fortified against. Nussbaum, Brennan, and White assume that the happy person must be fortified against the loss of any good that would lead to disappointment. But what Cicero spent the previous two books of TD arguing was that the happy person must be fortified against degradation of character: those beliefs that make us prone to fearing loss also obstruct our ability to act virtuously, so the happy person must be fortified against those degrading beliefs. For instance, he says, if one is susceptible to distress (aegritudo), then one will not be disdainful of bad circumstances but will instead be susceptible to excessive fear, dejection, and despair. However, each of these is incompatible with courage, and practical wisdom (sapientia) requires courage; so practical wisdom is incompatible with susceptibility to distress (TD III.14–15). Cicero goes on to present no fewer than six more Stoic arguments for the incompatibility of wisdom and distress (III.15–22), and he concludes that susceptibility to distress makes one prone to vice (IV.75), diseased (III.22, 74), degraded, hideous, and craven (IV.35, 64, 68). Furthermore, it is in this vein that Cicero continues the very passage quoted above: the virtuous person must be fortified against fear because it is incompatible with the virtue of courage (fortitudo, V.41–2). Even in this alleged proof-text, the modern reconstruction still goes against Cicero’s.²⁸ Cicero’s discussion is revealing of the basic structure of the Stoic position: an outlook on our happiness that would make us susceptible to distress—such as the belief that there are goods besides virtuous choice—is an outlook we cannot live with, because such an outlook is ethically perilous.²⁹ So, since our only viable option is committing to the ideas that our good lies entirely in virtuous activity, and that virtuous activity is something that we control completely, we must therefore accept that virtuous activity is sufficient for happiness. ²⁶ This “threefold division of goods” refers to the Aristotelian view that happiness requires mental goods, bodily goods, and external goods (see NE I.8). ²⁷ TD V.40–1, trans. Long and Sedley (1987, 63L), cited by Brennan 2005, 132 n. 8. See also TD V.28. ²⁸ Nussbaum 1994, 396–8 discusses liability to degradation as a distinct point from Cicero’s in TD. ²⁹ See also Sorabji 2000, 169–84, who argues that although the Stoics make virtue the only good in order to guarantee the tranquility of the virtuous person, this is because the loss of that tranquility is ethically perilous, as e.g., when love leads to jealousy and hate in one who fails to regard the beloved as indifferent with respect to happiness.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

196

 . 

But perhaps the Stoic view is feeble on this reconstruction too. Even if it doesn’t expect us to agree already that happiness should be invulnerable to fortune, it does presuppose that susceptibility to distress is ethically disastrous. Why think so? To see why, we turn once more to Epictetus.

3.3 Epictetus’ argument A constant theme in Epictetus’ teaching—I would say, its unifying theme—is that by staking our happiness on anything but virtuous activity, we surrender the very freedom that is our humanity. Anyone who longs for or runs away from things that aren’t up to him can be neither steadfast nor free; he himself must shift and be blown around with them, and he must even subject himself to others who can procure or keep those things away. (Diss. I.4.19) If I’m enamored with my mere body, I have handed myself over as a slave; with my mere possessions, again a slave. For in spite of myself, I show straightaway to whom I might become a captive. (I.25.23–4) Once you have handed over what is your own to things outside you, be a slave from then on, and don’t pull back and be a slave when you want to and not when you don’t, but full stop and with all your heart . . . If your jaw drops over things outside yourself, you’ll inevitably be tossed up and down at your master’s desire. And who is your master? Anyone who has power over the things you strive for or pull away from. (II.2.12–13, 25–6) Don’t fail to recognize your masters, but as long as you hold this grip on your body, follow everyone who is stronger than you are. (II.13.23) [A person’s master is] whoever has power to procure or take away the things that someone wants. “Is that how many masters we have, then?” Yes, because prior to them, we have things for masters—and they are many! Consequently, those who have power over any of these things must also be our masters . . . . When we love or hate or fear these things, those who have power over them must be our masters. (IV.1.59, 60) High regard for any external thing whatsoever makes one subject to another. (IV.4.1)³⁰

There is in Epictetus’ Discourses an extended argument for accepting that virtue is self-sufficient. I shall reconstruct that argument, in broad strokes, as follows:

³⁰ On enslavement in Epictetus, see Stephens 2007, 14–15.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

    

197

1. Humans do what they judge to be their best option for preserving their happiness. So, as a corollary, 2. Humans can be manipulated by whatever can dispose of things they take to be important for their happiness which they cannot control themselves. But since 3. Human happiness requires freedom from such manipulation, it follows that 4. Happiness requires that we not take things we cannot control to be determinants of our happiness. Let me discuss each of these premises more closely.³¹ The first premise for Epictetus is the thesis that people do what they judge to be their best option, “best” here being best for the sake of their happiness.³² This thesis encompasses a couple of ideas. First, part of the very form of practical rationality is choosing to do what one judges to be one’s best option. Notoriously, Epictetus argues that even Medea is rational in this sense: even her outrageous act of infanticide was a case of her doing what she thought was her best option—and correctly so, given the priorities she had.³³ Furthermore, second, doing what one judges one’s best option is utterly typical of creatures like us. In fact, he argues that there is no such thing as akrasia (e.g., Diss. I.28.6), since the soul is a single power—the judgment of appearances—whereas akrasia would require another power in the soul to oppose that judgment.³⁴ The second premise is a corollary of the first. Since humans do what they judge to be their best option for preserving their happiness, humans can be manipulated by whatever can dispose of things they take to be determinants of their happiness which they cannot control themselves. Epictetus believes that, ironically, it is our very autonomy and rationality that get us into such traps. It is up to no one but oneself to decide what sort of importance to attach to things like property and position; as Epictetus asks, rhetorically, “Who can force you to believe something when you don’t want to?” (Diss. II.6.21) And because we are rational, we can appreciate that we must do whatever it takes to preserve whatever we give such importance. In such cases, Epictetus says that “choice forces choice”: Can anyone keep you from agreeing with the truth? No one. Can anyone force you to accept a falsehood? No one. Do you see that in this respect you have choice that cannot be kept back or forced or interfered with? Well, then, is it any ³¹ Although this is not the only argument that Epictetus offers (see Stephens 2007, 10–15 for discussion of others), I do believe that it is the argument on which his others ultimately depend. However, I shall not argue for this strong claim here. ³² This is a point he takes himself to share with Socrates; see e.g., Diss. II.1.22–8, rehearsing the argument of Plato, Gorgias 466d–468e. ³³ Diss. I.28 and II.17. Con. Galen, Doctrines 3.3.13–24; see Gill 1983 for discussion. ³⁴ See esp. Diss. I.3.4; I.18.2; I.28.1–10; III.3.3–4; III.7.5, 33; Handbook 42. See also Marcus Aurelius, Med. VIII.14.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

198

 . 

different with desire and impulse? And what can overcome one impulse but another impulse? What can oppose one desire or aversion but another desire or aversion? “But if someone puts the fear of death in me,” he says, “he forces me.” It’s not what’s put in you [sc. that forces you], but the fact that it seems better to you to do some such thing than to die. So, again, it’s your own belief that has forced you—that is, choice has forced choice. (I.17.22–6)³⁵

Consequently, the choice to bind one’s happiness to something outside one’s control ultimately leaves one trapped into doing whatever one must do to keep hold of it.³⁶ Now, this means that a person who binds his happiness only to the use of right reason is also forced to do whatever he must do to keep hold of it, but of course what he is thereby “forced” to do is only to act in accordance with right reason. Such a person cannot become anyone’s implement—and that is the point. Epictetus’ third premise is that autonomy is crucial for our humanity and therefore our happiness. Epictetus constantly reminds his students that to be human is to be capable of giving and withholding “assent” to appearances in accordance with norms we recognize, norms for believing and norms for acting.³⁷ He describes this distinctly human process as “making use of” our appearances. For instance, one may have the appearance now that such and such a thing is the case, or that such and such an act is the thing to do. Such appearances we may share with animals, Epictetus says, but what is distinctive of humanity is that we have the option—indeed, the burden—of choosing whether or not to agree that such and such really is the case or really is the thing to do. Even our emotions, Epictetus argues, are not merely things that happen to us, but a use of our appearances, that is, an agreement that a way it appears fitting to feel really is fitting. For instance, he says, it is not inevitable that one should fear even the sword of a tyrant’s soldier: deadly things by themselves do not frighten us—we don’t go around flinching at roof tiles, but they are deadly too. Rather, they frighten us only if we agree that they and death itself are things worth fearing (Diss. IV.7.1–4, 16, 25–8; cf. Handbook 5, 16). Epictetus’ point is that the very form of human life is acting by way of responding to what reasons to act one takes oneself to have. To abandon that autonomy is therefore to abandon that very form of life that makes our life distinctly human, and as such something awesome and precious. This sense of awesomeness and preciousness is conveyed by Epictetus’ constant references to reason as a gift from Zeus that is so liberating that Zeus himself cannot even take it back (e.g., Diss. I.6.40).

³⁵ See also I.19.16; I.25.4; I.29.12; II.23.19; III.3.11–13; III.19.2; IV.4.23; IV.10.19; Handbook 9. ³⁶ As Long (2002, 217) puts the point, in such cases “we have hindered and constrained ourselves by our own voluntary abrogation of autonomy.” ³⁷ See esp. Diss. I.6.10–17; I.1.7; I.20.5; II.1.4; II.8.20; III.1.25–6.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

    

199

The conclusion that Epictetus draws from all of this is that happiness requires that we not take things we cannot control to be determinants of our happiness: By the gods, stop being enamored with stuff! Stop making yourselves slaves to things, in the first place, and then on account of them to the people too who can give them or take them away. (Diss. III.20.8) “So, what is yours?” [Speaking in persona of Diogenes the Cynic:] “The use of appearances. This [sc. use], [my former master] showed me, is mine: that I possess what cannot be kept back, cannot be forced; no one can hinder, no one can compel me to use [sc. appearances] other than as I want to. So, then, who has power over me? Is it Philip or Alexander or Perdiccas or the great [Persian] king? How could they have it? Whoever is going to be dominated by a person must be dominated by things long before that. So, whoever is not bested by pleasure, or drink, or reputation, or riches, but can spit his whole body at someone and take his leave, whenever this seems right to him—whose slave is that? To whom is he subjected?” (III.24.69–71) The person who cannot be kept back is free, to whom things are at hand as he desires them. But someone it’s possible to keep back, or force, or interfere with, or push unwilling into something—that one is a slave. And who can’t be kept back? The one who goes in for nothing that belongs to others. And what belongs to others? Those things that aren’t up to us, either to have or not to have them, or to have them of a certain variety, or to have them in a certain manner. (IV.1.128–9)³⁸

Since the only thing that is ultimately within our control is the faculty of choice, human happiness must consist in the good exercise of that faculty and nothing else. That is why, for the Stoics, only virtuous activity is good, and virtuous activity must be self-sufficient, separable from anything outside our control. The only way to free ourselves from traps, Epictetus says, is to undo our ensnaring choices about what our happiness requires and be prepared instead to lose anything we cannot control. That is the only way to be free, and that is why the happy person must be fortified against grief over losses of other things.

3.4 The significance of Epictetus’ argument One question remains: how seriously should anyone take the argument that we find in Epictetus? We might note that his first premise draws on the Stoic belief in the impossibility of akrasia, and if we do not share that belief then we might reject the first ³⁸ See also I.9.21; I.18.17, 21; I.29.5–8; II.8.27–9; III.24.56; III.26.38–9; IV.4.38–40; Handbook 1. On what is “your own,” see Diss. I.22.9–11; II.15.1; II.19.32; III.1.40; III.24.3, 69; IV.1.77; Handbook 1.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

200

 . 

premise on those grounds. However, Epictetus’ worry does not require anything as robust as the impossibility of akrasia and the necessity of doing what one judges one’s best option. What worries Epictetus are cases in which we do do what we judge our best option. In the end, Epictetus’ worry does not turn on whether or not practical rationality might fail to function normally. His worry is about precisely those cases in which practical rationality succeeds. Likewise, Epictetus connects his third premise to Stoic compatibilism about freedom of choice, and that idea is not beyond question either. Again, though, we need not go that far to appreciate the force of Epictetus’ point. The point is that to abandon whatever autonomy we have is to abandon that very form of life that makes our life distinctly human. What I think is really the most likely thing to bother us about Epictetus’ argument is the sense that he is simply too extreme in his thinking about the dangers we face in staking our happiness on things besides the use we make of our appearances. Surely we could have agreed if Epictetus had said to beware the line between being appropriately vulnerable to circumstances and being so vulnerable or fragile as to become rationally or emotionally impaired, or even capable of murder. But the worry is that Epictetus seems to have denied any such line at all between some appropriate sort of vulnerability and serious ethical peril.³⁹ However, I think there is a lesson to take from Epictetus that is much more subtle than that. Even granting that there is such a line, surely it matters how good in fact we are at keeping our balance along that line. This question is all the more challenging when we remind ourselves that staying on the right side of that line is a matter of nothing less than holding on to our wisdom, our emotional balance, and our freedom when what is at stake are things to which we are attached as if our happiness depends on them. The point, I think, can be put not as the absolute certainty that from such attachments it follows that one must fall necessarily into the gravest ethical peril, but as a point about human psychology that such attachments expose us to risks of ethical peril that just are where our self-inflicted suffering, impairment, and unfreedom actually do come from. Put another way, we may not agree with Epictetus about what guarantees wrongdoing, but there still remains a question about where wrongdoing comes from—and how easily it can come. And here, I think, Cicero and Epictetus do have a point. It takes virtuous activity and human autonomy to live a happy life, but staking one’s happiness on things outside one’s control can threaten virtue and autonomy alike. We cannot take it for granted that that way of staking our happiness is one that we can afford. Now, there is a point on the other side as well, since happiness that is staked on relationships and circumstances that we cannot control is the sort of happiness that humans usually think is the only sort

³⁹ Cf. Badhwar 2014, 217.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

    

201

of happiness worthy of the name. Perhaps Aristotle was right about that. But my point is that we cannot afford to be glib about that, as we will be as long as we underestimate how serious the alternative is that the Stoics present. And yet most modern philosophers do dismiss that alternative out of hand, a reaction that the modern reconstruction only reinforces.

4. Conclusion Make no mistake: the Stoic view is counter-intuitive. But the Stoic view also demands to be reckoned with. Whether we accept their view in the end or not, what the Stoics certainly show is that in a human life there is no avoiding the tension between the vulnerability of our attachments to the things that make up our lives and the steadfastness of wisdom we need to live our lives well. In an ideal world there would be no such tradeoff to make. The Stoics point out that that cannot be our world—and on that point anyway they were exactly right. Whatever we say about the Stoic position, what we absolutely cannot do is to ignore it.⁴⁰

References Annas, J. 1993. The Morality of Happiness. New York: Oxford University Press. Badhwar, N. K. 2014. Well-Being: Happiness in a Worthwhile Life. New York: Oxford University Press. Brennan, T. 2005. The Stoic Life. Oxford: Oxford University Press. Dobbin, R. 1991. ‘Προαίρεσις in Epictetus.’ Ancient Philosophy 11: 111–35. Dyson, H. 2009. ‘The God Within: The Normative Self in Epictetus.’ History of Philosophy Quarterly 26: 235–53. Frede, M. 1986. ‘The Stoic Theory of the Affections of the Soul.’ In The Norms of Nature, ed. M. Schofield and G. Striker, 93–110. Cambridge: Cambridge University Press. Gill, C. 1983. ‘Did Chrysippus Understand Medea?’ Phronesis 28: 136–49. Gill, C. 2008. ‘The Self and Hellenistic-Roman Philosophical Therapy.’ In Vom SelbstVerstandnis in Antike und Neuzeit: Notions of the Self in Antiquity in Beyond, A. Arweiler and M. Moller, 359–80. Berlin: De Gruyter. Graver, M. R. 2007. Stoicism and Emotion. Chicago: Chicago University Press.

⁴⁰ My thanks to David Sedley for his insightful comments on an earlier draft, to the participants at the Keeling Colloquium, and to Fiona Leigh for inviting me to participate and for her generosity both as host and as editor.

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

202

 . 

Irwin, T. 1986. ‘Stoic and Aristotelian Conceptions of Happiness.’ In The Norms of Nature, ed. M. Schofield and G. Striker, 205–44. Cambridge: Cambridge University Press. Irwin, T. 1990. ‘Virtue, Praise, and Success: Stoic Responses to Aristotle.’ Monist 73: 59–79. Irwin, T. 2020. Ethics through History: An Introduction. Oxford: Oxford University Press. Lesses, G. 1989. ‘Virtue and the Goods of Fortune in Stoic Moral Theory.’ Oxford Studies in Ancient Philosophy 7: 95–127. Lewis, C. S. 1961. A Grief Observed. London: Faber and Faber. Long, A. A. 2002. Epictetus: A Stoic and Socratic Guide to Life. Oxford: Oxford University Press. Long, A. A., and Sedley, D. N. 1987. The Hellenistic Philosophers, vol. 1. Cambridge: Cambridge University Press. Nussbaum, M. 1994. The Therapy of Desire. Princeton: Princeton University Press. Reydams-Schils, G. 2005. The Roman Stoics. Chicago: University of Chicago Press. Rist, J. M. 1969. Stoic Philosophy. Cambridge: Cambridge University Press. Russell, D. 2012. Happiness for Humans. Oxford: Oxford University Press. Sellars, J. 2006. Stoicism. New York: Routledge. Sherman, N. 1987. ‘Aristotle on Friendship and the Shared Life.’ Philosophy and Phenomenological Research 47: 589–613. Sorabji, R. 2000. Emotion and Peace of Mind. Oxford: Oxford University Press. Sorabji, R. 2006. Self. Chicago: University of Chicago Press. Stephens, W. O. 1996. ‘Epictetus on How the Stoic Sage Loves.’ Oxford Studies in Ancient Philosophy 14: 192–210. Stephens, W. O. 2007. Stoic Ethics: Epictetus and Happiness as Freedom. London: Bloomsbury. Stephens, W. O. 2020. ‘The Stoics and their Philosophical System.’ In The Routledge Handbook of Hellenistic Philosophy, ed. K. Arenson, 22–34. London: Routledge. Tsekourakis, D. 1974. Studies in the Terminology of Early Stoic Ethics, Hermes Einzelschriften 32. Wiesbaden: Franz Steiner. White, N. 2006. A Brief History of Happiness. Oxford: Blackwell. White, S. 2002. ‘Happiness in the Hellenistic Lyceum.’ Apeiron 35 (supp.): 69–93.

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

13 Comments on Daniel C. Russell, ‘Three Mistakes about Stoic Ethics’ David Sedley

Daniel Russell’s paper is a philosophically rich one, full of new historical perspectives. I will focus on just a few points at which it seems to me that the argument invites a challenge. Russell starts with the question how the sufficiency of virtue for happiness could follow from its being the only good, and ends with the question what makes Stoic eudaimonia invulnerable. I particularly savoured the sensitive readings of Epictetus, called in to support Russell’s own answer to these questions. Even if doubts remain as to how far the Epictetan notions can safely be traced back into classical Stoicism, he has performed an important service by sketching so effectively how one might set out to do so. My comments will be clustered largely around the opening issue, how the sufficiency of virtue for happiness is inferentially related to the thesis that virtue is the only good. If one looks at the running battle between Stoicism on the one side and the Peripatetic tradition revived by Antiochus on the other, it is in a sense obvious that Russell must be right that virtue’s being the only good cannot follow from its being sufficient for happiness, because the two schools agree on the sufficiency thesis yet disagree as to whether virtue is the only good. On the other hand, once the two schools are allowed their own supplementary premises, it seems to me, their inferences do go through. According to the Peripatetic tradition, there are degrees of happiness, so that the happiness that arises from virtue alone might be enhanced by the addition of bodily and/or external goods. According to the Stoics, by contrast, there are no degrees of happiness, and therefore no supplementary goods that could further enhance one’s happiness, once one has achieved virtue. Hence virtue is the only good. The need for additional tacit premises need not detract from the force of either the Stoic or the Peripatetic inference. What that need shows is that the inferences themselves are not, and should never have been expected to be, context-free and self-explanatory, but are on the contrary embedded in complex ethical systems, any part of which might have to be invoked in their support. Since we are not

David Sedley, Comments on Daniel C. Russell, ‘Three Mistakes about Stoic Ethics’ In: Psychology and Value in Plato, Aristotle, and Hellenistic Philosophy: The Ninth Keeling Colloquium in Ancient Philosophy. Edited by: Fiona Leigh and Margaret Hampson, Oxford University Press. © Oxford University Press 2022. DOI: 10.1093/oso/9780192858108.003.0013

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

204

 

dealing with a formal syllogism, the need to supply one or more additional premises is not as such a defect. Similarly when it comes to the reverse inference, there is probably not much difference between Russell’s approach, which is to deny that the sufficiency thesis follows from virtue’s being the only good, on the ground that an additional premise concerning the good person’s own self-sufficiency is needed, and the more friendly alternative of saying that the sufficiency thesis does indeed follow from virtue’s being the only good, provided that we take careful note of any premises that are being tacitly presupposed. In support of the friendlier response, let me suggest that there are, inevitably, other tacit premises in the inference, beyond the one Russell has himself pointed to. One of these extra premises is the assumption that happiness is humanly attainable. If happiness were unattainable, then even possession of all the available goods—be these virtue alone, or virtue plus bodily and external goods—would have been insufficient to secure it. It is only if we assume that happiness is attainable in the first place that we can hope to justify the further calculation that a life containing all possible goods must possess happiness. Now the Stoics came nearer than any other ancient thinkers to declaring happiness unattainable, notoriously setting the standard so high that they had some difficulty in identifying anyone who had ever reached it. The Stoic sage, who alone is happy, is said to be as rare as the phoenix. Hence their continuing faith that happiness is attainable is unlikely to have rested on any kind of appeal to the obvious. Probably it relied in part on a prior commitment to the world’s providential structure: it would be unimaginable that a beneficent god, in creating this anthropocentrically structured world, did not offer mankind even the possibility of a happy life in it. The role of Stoic teleology in the broad ethical picture should not be underestimated. Thus the attainability of happiness is, if I am right here, one tacit premise of the argument from virtue’s being the only good to its sufficiency for happiness. There appear to be at the very least two further premises underlying the same argument. Tacit premise no. 2 is the unity of virtue. The originally Socratic thesis that the virtues are inseparable had gone largely unchallenged by philosophers in the intervening century. But in the first three generations of the Stoic school we witness an extraordinarily intense debate, involving Zeno, Aristo, Cleanthes, and Chrysippus, as to how best to explicate and defend this thesis. That running debate may well reflect the doctrine’s role in ensuring the sufficiency of virtue for happiness. In order to be assured that a virtuous life guarantees happiness one has to be confident that it could contain no residual moral defects which might deflect it from total harmony with nature. The thesis of the unity of virtue appears necessary to any such assurance. Similarly necessary to that assurance is a third tacit premise, the Stoic thesis that virtues are not hexeis but diatheseis, that is to say that they do not admit of differing degrees, being states of moral perfection. This thesis brought in its wake

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

   . 

205

the notoriously paradoxical consequence that the transition from vice to virtue is instantaneous. We might better understand the Stoics’ dogged commitment to it if we were to reflect on its role as another tacit premise in the inference from ‘Virtue is the only good’ to ‘Virtue is sufficient for happiness’. Only if virtue is a state of moral perfection can it be certified free of any residual moral weaknesses that might impede the life in complete conformity to nature with which Stoic happiness is equated. I turn now to Russell’s own candidate for a missing premise. The full version of the argument, according to Russell as I understand him, will have run roughly as follows: Virtuous activity is the only good.

Therefore, virtuous activity is sufficient for happiness. This is of course still far from being a formally valid syllogism, so we have to ask why the added premise might help to make it persuasive. Russell’s answer is, I think, that if virtuous activity could be curtailed by factors outside the agent’s control then it would not after all be sufficient to guarantee happiness. Now one question that this proposal raises is, what is the role of activity in the Stoic conception of the good? Is Russell’s emphasis on the value of virtuous activity, rather than on virtue as such, justified? It is scarcely deniable that moral activity plays some part in the Stoic argument, because Stoic happiness is not a mere state or disposition but a way of living. Nor is it deniable that virtuous activity, as one of the things that ‘participate in’ virtue, is itself derivatively called a good in Stoic theory. But if Russell is to be understood as making happiness dependent on virtuous activity rather than on mere possession of the virtuous disposition, the shift of emphasis sounds suspiciously Aristotelian.¹ It can hardly be an accident that no Stoic text (none that I can think of, at any rate) resembles NE I.8 in shifting the emphasis from virtue itself to virtuous activity. Russell’s understanding of Stoic virtuous activity is lucidly explained with the following sketch (187–8): They [the Stoics] would need to understand virtuous activity as wise selection and response with respect to the world around one, as one finds it, so that such activity is uninterrupted even when the details of one’s worldly circumstances change. On this . . . view, the loss of a particular other may be the loss of something significant, but what it cannot be is the interruption of the virtuous ¹ Aristotle famously argues in NE I.8 that happiness must lie in the use, rather than mere possession, of virtue, just as at the Olympic Games the prizes go to those who compete successfully, not those who merely display the best physique.

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

206

 

activity in which one’s happiness consists. That activity just is the practically wise and emotionally sound exercise of the faculty of choice, and it takes no particular object in order to do that. Put another way, virtuous activity always takes some object, but there is no object that it always takes.

That this diagnosis depends on privileging virtuous activity over virtue as such is confirmed by Russell’s earlier words (187): The issue is not whether the loss of those relationships would make one less virtuous, but that such a loss would terminate those activities that are the same thing as that happy life one is now living.

And in section 2.3 Russell goes on, relying on Epictetus in particular, to reconstruct a Stoic theory of moral autonomy in the exercise of choice that will ensure that virtuous activity can indeed continue unabated, regardless of external circumstances. My question is whether that celebrated Epictetan theory, for all the light it sheds, is actually needed as an extra premise in order to safeguard the classical Stoic inference with which we are concerned here. According to classical Stoicism, quite independently of the Epictetan perspective, whatever a virtuous person does is automatically a virtuous action (katorthōma). Hence there is no danger of a gap opening up between virtue and virtuous activity: whoever is virtuous is, thanks merely to being alive and hence active, constantly engaged in virtuous activity, and indeed cannot act at all other than virtuously. This intimate interdependence between virtue and virtuous activity may in turn help us understand why the Stoics did not need to take the Aristotelian option of pointing to virtuous activity, rather than virtue itself, as the basis of happiness. Since (for the reason just given) virtue is sufficient for virtuous activity, and virtuous activity is sufficient for happiness, by a sound principle of transitivity virtue itself is sufficient for happiness. I turn finally to section 3.1 of the paper, where Russell takes a stand against a sizeable section of the modern interpretative literature. His objection is to those who say that the Stoics adopted the sufficiency thesis in order to ensure that happiness is invulnerable. As he points out in reply, the Stoics are reported as making the invulnerability of happiness their conclusion, not as invoking it as a premise, and reasonably so, given that it is a premise that few would initially accept as plausible. This does seem to me a promising reply to some of the scholarship he cites. However, others whom he also cites² might more charitably be read as taking the invulnerability thesis not as a premise invoked by the Stoics, but as the aim

² Nussbaum, and perhaps White (as cited in Russell, this volume).

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

   . 

207

that motivated them: they made virtue the only good in order to establish the invulnerability of happiness. And to view it that way seems to me entirely reasonable. In fact the passage that Russell (194–5), following Brennan, cites from Tusculans 5.40–1 may well itself be doing just that, invoking the invulnerability of happiness as a philosophical aim, not as the premise of any argument. As Cicero says there, ‘We want the happy man to be safe, impregnable, fenced, and fortified, so that he is not just largely unafraid, but completely.’ Compare cognitive certainty. The Stoics would have looked silly if, when arguing against the Academic sceptics, they had invoked the premise that certainty is attainable. But it is entirely reasonable for us to say by way of historical explanation that the aim of proving that certainty is attainable, and hence that philosophical truth can be discovered, is what motivated the Stoics to argue as they did against the Academics. Similarly, I am suggesting, it is reasonable to conjecture that the aim of proving the invulnerability of happiness is what motivated the Stoics’ insistence that virtue is the only good. Any such conjecture will then lead on to further questions. Did this goal of proving the invulnerability of happiness arise in its own turn from the Stoics’ providentialist worldview, according to which godlikeness is a realistic human aspiration? Or did it owe more to institutional or pragmatic considerations, such as the need to outbid the rival Epicurean recipe for invulnerability? Those must remain questions for another occasion.

OUP CORRECTED PROOF – FINAL, 18/11/2022, SPi

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

Index of Passages Cited For the benefit of digital users, indexed terms that span two pages (e.g., 52–53) may, on occasion, appear on only one of those pages. AELIAN Varia Historia 14.6 154–5 ARISTOTLE De Anima (On the Soul) 427b17–24 159 427b21–4 158n.33, 159 432b29–433a1 159nn.36, 38 De Memoria (On Memory) 427b24 179–80 449b22–23 158n.31 450a19–25 158 450a27–32 179 450a29–32 158 450b20–25 158n.33 453a9–14 157n.30 De Motu Animalium (Movement of Animals) 701a31–3 108n.4 701b17–22 158 702a5–7 158 703b18–20 158 De Partibus Animalium (Parts of Animals) 669a19–21 157n.30 Eudemian Ethics 1144a7–8 118 1151b6–8 116, 121 1215b5 136n.5 1216b4–11 23–4 1216b6 3–4 1220a10–11 10n.12 1220a34–7 129–30 1220b13–14 114–15 1220b18–20 114–15 1221a9 127 1227b23–4 118 1228a27 114–15 1228a28–29 120 1228b2–3 116 1228b4 116, 136 1228b4–1229a11 134n.1 1228b4–5 116 1228b7 136–7

1228b9 116n.5 1228b10–11 126n.19 1228b11 116 1228b13 116n.5 1228b14–17 136 1228b22–6 116 1228b26–7 116 1228b27–9 116–17 1228b28 136n.7 1228b29 116n.5 1228b29–31 116 1228b31–35 137n.8 1228b31–38 118 1228b38–9 137n.8 1229a1 117 1229a1–2 117, 120 1229a4–6 119–20 1229a6–9 119–20 1229a34–35 115 1229a36 115 1229b2 127 1229b17–21 121–2 1229b25 117–18 1230a32–33 120 1232a23–27 116 1232a38–b3 121–2 1232a39–b1 118 1234b3 121 1234b11–12 116 1237a6–7 129–30 Magna Moralia 1182a15–31 33–4 1182a24 21 1183b9–18 23–4 1187a5–12 23–4 1190b28–32 23–4 1198a10–12 23–4, 55 1198a12–15 55 1198a15–21 55 1200b25–30 23–4 1200b25–33 24 1200b30 55 1201b24–1202a1 55

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

210

   

ARISTOTLE (cont.) Nicomachean Ethics 1098a18–20 134n.2 1099a5 11 1099a5–7 125–6 1099a7–15 10–11 1099a7–24 124 1099a13–15 125–6 1099a15–20 128 1099a21–24 10–11 1099a31–b7 134n.2 1100a12–19 139n.10 1100b25–30 126n.18 1101b27–31 135n.3 1102b26–28 128n.22 1103a3–4 10n.12 1103a8–10 136n.6 1103a34–b2 129 1103b16–17 129 1104b1–2 118, 129 1104b5–6 129 1104b7–8 129 1104b9–11 129 1104b10–11 119–20 1104b11–13 129, 139 1104b13–18 129–30 1104b30–32 10–11 1105b21–23 114–15 1105b25–26 114–15 1106b17–24 10–11 1106b36 10n.12 1106b36–1107a2 114–15 1108b19–20 121 1108b24–25 118n.8 1109a25–30 10–11 1113a29–34 118n.8 1113b1–2 118n.8 1115a9 115 1115a11–16 121 1115a16 122, 136 1115a20–31 115n.3 1115a33 122 1115b7–8 121–2 1115b7–9 122–3 1115b10–11 122–3 1115b11–12 122–3 1115b12–13 120, 122–3 1115b13–14 123 1115b17–19 122–3 1115b19–20 120 1115b26–27 120–1 1115b32 121 1116a2–3 119–20 1116a8–9 120–1 1116a17–29 124 1116a31–b3 127

1116b31–1117a1 127 1117a16–17 123 1117a17–22 123 1117a29–30 114–15 1117a33–34 13 1117a35–b6 134n.1 1117b3 124 1117b8 124n.14 1117b9 124 1117b9–15 131 1117b16 124 1118a20–23 157–8 1119a21 119 1119a22–23 119 1120a23 10–11 1141b3–8 136n.5 1144b18–20 56 1144b19–20 3–4, 23–4 1144b26–27 10–11, 10n.12 1144b28–30 23–4 1145b22–31 23–4 1145b27–31 56 1147b13–17 56 1149a7–9 121–2 1150a9–13 127 1150a13–15 127 1151b6–8 116, 121 1153b14–15 125 1153b19–21 126n.18 1166b10–11 119 1169a22–4 131n.27 1172a27 125n.16 1172b9–25 135n.3 1174a4–8 125n.17 1174b14–16 10–11 1177a21–27 135n.4 1177b19–24 135n.4 1177b24–5 134n.2 1178b12–13 117n.7 1179b30–31 130 Poetics 1148b15–17 179n.15 Politics 1330b37–39 121–2 Rhetoric 1366a33–36 128n.24 1382a21–22 115 1383a5–6 131n.28 1383a6–7 127n.21 ARIUS DIDYMUS 2.5a 189 2.5a–5b1 188–9 2.5a–5b 188–9 2.5b 186, 188–9 2.5b2 189–90

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

    2. 5b9 189 2.5b4 188–9 2.5e 192n.21 2.5g 189–90 2.5m 192n.21 2.6d 190n.15 2.10b 188–9 2.11e 189–90 2.11q 192n.21 ATHENAEUS 12, 544A–B 11–12, 154–5 AUGUSTINE Sermons 348.3 153n.23 CICERO Academia 2.145 16–17 De Finibus (On Ends) 2.96 152–3 2.96–97 177n.12 2.104–105 159n.35 2.104–106 153 3.24–6 183 3.25 189 3.26 187 3.55 190n.14 3.75 185 Tusculanae Disputationes (Tusculan Disputations) 3.14–15 195 3.15–22 195 3.22 195 3.28 155–6 3.28–35 156n.28 3.74 195 4.35 195 4.38–42 189n.12 4.64 195 4.68 195 4.75 195 5.18–19 186n.3 5.28 195n.27 5.40–41 195, 207 5.41–42 195 5.73–74 154n.26 5.83 186, 193n.22 5.95 153n.23 Paradoxa Stoicorum (Stoic Paradoxes) 17 186 DIOGNES LAERTIUS Vitae Philosophorum (Lives of Eminent Philosophers) 2.87–88 11–12 2.89 154–5 2.91 11–12

7.49–51 16–17 7.96 17, 189n.9 7.101–102 16 7.102 16 7.102–104 16 7.102–7 189 9.51 22n.4 10.22 177n.11 EPICTETUS Discourses I.1 16–17 I.1.1–6 191 I.1.22–25 191–2 I.1.7 191, 198n.35 I.1.7–13 191 I.3.4 197n.34 I.4.19 196 I.6.10–17 198n.37 I.6.40 191, 198 I.9.21 199n.38 I.15 192 I.15.2 192 I.17.22–26 197–8 I.18.2 197n.34 I.18.17 191–2, 199n.38 I.18.21 199n.38 I.19.16 198n.35 I.20.5 198n.35 I.22.9–11 191–2, 199n.38 I.25.4 198n.35 I.25.23–24 196 I.28.1–10 197n.34 I.28 197n.33 I.28.6 197 I.29.5–8 199n.38 I.29.12 198n.35 II.1.4 198n.35 II.1.22–28 197n.32 II.2.12–13 196 II.2.25–26 196 II.6.9 16 II.6.21 197 II.8.20 198n.35 II.8.27–29 199n.38 II.9.15 188n.8 II.13.23 196 II.15.1 199n.38 II.17 197n.33 II.19.13 188n.8 II.19.15 188n.8 II.19.32 191n.17, 199n.38 II.23.9 191 II.23.19 198n.35 III.1.25–26 198n.35 III.1.39 191–2

211

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

212

   

EPICTETUS (cont.) III.1.40 199n.38 III.3.1 192 III.3.3–4 197n.34 III.3.7–10 192 III.3.11–13 198n.35 III.7.5 197n.34 III.10.9 191 III.19.2 198n.35 III.20.8 199 III.20.11–15 192 III.20.12 192 III.24.3 199n.38 III.24.56 199n.38 III.24.69 199n.38 III.24.69–71 199 III.26.38–39 199n.38 IV.1.59 196 IV.1.60 196 IV.1.69 191 IV.1.72 191 IV.1.72–3 191 IV.1.77 199n.38 IV.1.128–129 199 IV.4.1 196 IV.4.23 198n.35 IV.4.38–40 199n.38 IV.7.1–4 198 IV.7.16 198 IV.7.25–28 198 IV.10.19 198n.35 Handbook 1 191–2, 199n.38 5 198 9 198n.35 16 198 30 192 42 197n.34 EPICURUS Letter to Idomeneus Usener 138 177 Epistula ad Menoeceum (Letter to Menoeceus) 128 12 131 177 EURIPIDES Andromeda 133 153 GALEN Doctrines 3.3.13–24 197n.33 GORGIAS Palamedes 13 48n.63 19 48n.63

HOMER Odyssey 2.46–47 45–6 9.114–15 65 9.175 65 14.192 163 14.361–362 163 15.399–401 162n.45 15.400–401 162 15.484 163 15.486–487 163 18.205 64–5 LYSIAS Ergocles 7.5 102n.5 MARCUS AURELIUS Meditations II.2 191n.18 V.26 191n.18 VIII.14 197n.34 PLATO Apology 19–20c 28–9 20b4 29 22a–23b 28–9 38a 40–1 Charmides 165c–175a 28 165c–176b 43–4 165e–171c 44 167e 73n.13 174b–175b 44 Crito 46b–49a 40–1 46b–49e 40–1 46b3–c6 57–8 47e–48a 44 49a4–7 57–8 49a4–b6 57 49b1 57 Euthydemus 273d–275a 27n.20 275e–278b 43n.55 278d3 27 278e–281e 82n.2 278e–283e 2–3 281b 27 283a4 27n.20 283b3 27n.20 292a–e 44 Euthyphro 7b–d 33–4 11e–12e 40–1 14b–15c 28

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

    Gorgias 367c–368e 48n.63 453d–459c 44 460a–e 25n.14 466a–468e 47–8 466d–468e 197n.31 468b–c 82 468c2–7 107 471e–472c 41 473e 41 492a8 109n.7 492d7 109n.7 500c–d 2–3 503c6–d3 107 505a6–b12 107 506c–509e 44 507c 2–3 508a-b 2–3 508e–509a 58 509e 47 513c4–d1 57 519a 65–6 521d 45n.59 Hippias Minor 363b1–4 64–5 376b 43–4, 47 Ion 537d–540b 44 Laches 181e–184c 41 190d 45 194d 28 194d–199e 28 199e 45 Laws 732e–733d 75n.17 Lysis 216d–222e 82n.2 219d4–5 107 221a7–b1 107 221d7–e1 109 221d7–e2 109 Meno 71e–72a 30n.25 77a–78a 47–8 77a–78c 82n.2 77c1 76n.21 77c–78b 48n.63 77d7–e3 107 77e5–78a5 107 78d 31n.27 78e 31n.27 86e 26–7, 33–4 86e–89a 184 86e–100b 42–3

87a6 36n.40 87b1 36n.40 87b–c 42–3 87c2–3 34 87d–89a 33–4, 42–3 87d3 34 88a–89a 27 89b–c 33–4, 42–3 89c2–4 34 89d–96c 33–4 89d–96d 25–7, 42–3 96d 39 96d–99a 42–3 97a–100b 33–4 97e 34n.35 98e 33–4 99a–b 26–7 Parmenides 136a1 42n.54 136c1 42n.54 142b4 42n.54 Phaedo 67c5 36n.40 68b–69d 45 68e4 36n.40 74a2 36n.40 80b1 36n.40 85e–86d 36 87d–89a 34n.36 91d–95a 35–6 93c10 32–3 94a12–b2 35–6 100a 36–7 100a2–a5 34 100a4 34n.35 100a5 60n.7 100a5–7 37–8 100d8 35n.37 100e1 35n.37 101e2 36n.40 101c9–e3 35 101d3 61n.8 101d3–e3 61 101d4–6 37–8 101e2 36n.40 101e8 34n.35 105b7–8 35n.37 105c1 35n.37 107a–b 39 107b 35–6 118a15–17 39n.48 Philebus 11b6–c2 143–4, 144n.4 11d4–6 144–5 12d3 148–9

213

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

214

   

PLATO (cont.) 20d7–10 82 21a–d 143 21b6–9 143–4 21c 14n.16, 143–4, 148n.10 21c1–8 143–4 21c5–6 144n.5 21d9–e2 148–9 22a 143 22a9–b2 144n.4 22b3–8 144n.4 22b4–6 144n.4 22b622b6 144n.4 32b9–c2 148–9 32c3–5 147–8 33c 147 34a10–11 147–8 34b6–8 147–8 34e9–35a9 110 35b11–c1 110 35c–d 147 33c5–6 147–8 35c9–10 144n.4, 148n.10 36a8 148–9 36b4 148n.11 36b4–6 148–9 36b8–9 148n.10, 172n.4 36b11 148n.11 36b11–c1 148–9 36c10–11 149n.12 39a1–c1 110 39a–40d 9–10 39c–d 143–4 39e6–8 148–9 40a 14n.17 40a3–7 148–9 40a6–12 110 40a9–12 151 40e2–4 149n.12 42c5–d3 112 47c5–7 148–9 53c–55a 89–90 60d3–e5 144n.5 Protagoras 318e5–319a2 27 319a–320b 28–9 319a–320c 26–7 322c1–d5 65 328ab 62 329d–334a 40n.50 329e 29–30, 45, 61 333c–d 31n.27 344e–345c 27

345e 47 349a–360e 40n.50 349d 29–30, 45, 61 350e–351a 29–30, 45, 61 351d3 35n.37 352a–d 40–1 352b–c 2–3 352b–358c 27 352b–360d 44 352e 48–9 354a 173–4 355a6 58–9 355b3–4 58–9 356a–357d 33–4 356a5–6 174 356b1 174 356b1–e4 173–4 356b8 174 356c6 174 356c–357b 57 358b–c 48n.63 358b6–c3 47–8 358c–d 40n.50, 82n.2 360e–361d 42–3 361a 28–9 361a5 59 361a5–6 59 361b–c 26–7, 59n.5 361c2–3 59 Republic 332b–334a 44n.58 335a–e 44n.58 352d 2–3 353e–354a 2–3 354a–c 2–3 358e 25–6 369c 48n.63 375d–376c 44n.58 411e6 74n.15 423–424 99–100 430c 45, 50n.65 436a–439d 86 436a–441c 86 436a8–9 93 436a9–b3 86 436a10–b2 107–8 436b 71n.8 436b1–2 74n.16 436b9–c1 86 436b9–10 69 436d4–e5 87n.9 436e 42n.54 436e7–437a1 86

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

    437b1–4 86 437b1–5 69 437b6–c5 86 437b6–439e1 106n.2 437b6–9 69 437c7 69 437d–e 97–8 437d7–e6 69 437d10–e1 106n.2 437d10–e3 68 437e4–5 67, 107–8 437e7 104 438a 75, 78–9, 106 438a–b 73n.12 438a1–2 109 438a1–5 70, 85 438a6 77 438a1–5 76, 102, 105 438a6 77, 103 438a7–b2 72 438a7–e9 91 438d1 72n.9 439a–d 88 439a4–7 79 439a4–b1 85, 107–8 439a6 68 439a9–b1 69 439a9–c5 86 439b3–6 69, 89 439b3–c1 87n.9 439c–441b 3–4 439c3–4 69 439c6–8 86–7 439c7–8 108n.4 439c7–d8 69 439c10–d8 86–7 439d1–2 106–7 439d6–8 106 439d8 74n.16, 109 439e 67–8 439e5–440a4 94 440a4 109 441c4–e1 87 441e 3–4 442d 77 474c8 73n.12 474c8–10 73n.12 475b 75 475b4–7 73n.12 475b8–9 74 505a 44 505a–b 47–8, 48n.63 505d–e 77, 98–9

505d5–e3 83, 100 505d7–9 91–2 505d8 92 505d11 92 505e1–2 108n.6 510d5–e1 90 518c4–5 110 523a1 110 523b–524b 88n.11 525a–531e 4n.5 562b–c 77 571d 67–8 580d7 74n.14 581a7 74n.16 583b–584b 146 583b–4b 174–5 585d11 109, 112–13 586b7–c5 112 602e–603a 78n.25 621a6–b1 110 Sophist 231a7 35n.37 263d–264b 150n.17 Statesman 262b6 35n.37 283c–285c 33–4 Symposium 200a9–b1 109 205a–206b 82n.2 Theaetetus 158a1 111n.9 158a2 111n.9 158b2 111n.9 158b4–5 111n.9 158d3-4 111n.9 189d–190a 150n.17 191c–196c 150n.15 Timaeus 86b–87b 47, 82n.2 86d–e 47 87b 47 PLOTINUS Enneads 4.3.28 162n.45 PLUTARCH Quaestiones Convivales (Convivial Questions) 630e 153n.25 Non Posse 1088E–1089A 153–4 1089A 176n.10

215

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

216

   

PSEUDO-PLATO Axiochus 366d2–7 105 SENECA Letter(s) 88.43 (=DK80A20) 22n.4 De Vita Beata 15.2–3 190n.15 SEXTUS EMPIRICUS Adversus Mathematicos (Against the Mathematicians) 7.151 16–17 11.22–23 16–17

11.200–201 188n.8, 189n.10 11.207 189n.10 STOBAEUS Eclogae 2.11g 185 2.79 18–80 16 THUCYDIDES I.139 31n.26 III.82 65 VIRGIL Aeneid 1.202 162n.45

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

General Index and Index of Names For the benefit of digital users, indexed terms that span two pages (e.g., 52–53) may, on occasion, appear on only one of those pages. activity, action 1–3, 9, 24, 30–1, 45, 47, 68, 79, 88–9, 104, 107–9, 111n.11, 116n.5, 137, 141–4, 145n.6, 157n.30, 174, 198 object of, end of 13, 117–18, 124–5 practically wise 29, 118 virtuous 10–13, 17–19, 55, 107, 115, 117–18, 122, 124–32, 134–6, 138–9, 185–90, 192–3, 195–6, 199, 205–6 Adkins, A. W. H. 30n.25 affect, affective response; see also emotion; pleasure 2–3, 13–15, 19, 21, 23–4, 106–7, 110–11, 118, 120, 137, 141–2, 147–56, 159–60, 163–6, 170–82, 185 agathon (good, the good)—see good, the good agency 8–14, 17, 19, 24, 27, 31–2, 47–8, 50, 76, 79–80, 93–4, 104, 115–22, 124n.14, 126–9, 130n.26, 131–2, 134–7, 139, 141–6, 148–9, 170–1, 173–6, 180–2, 205 aisthēsis (perception)—see perception akrasia (weakness of will, incontinence)—see weakness of will alēthēs (true)—see truth, true Allan, D. J. 136n.7 Allen, J. V. 40n.50 Anagnostopoulos, M. 67n.2, 68, 85n.7, 88n.13, 89n.14 animal 105, 144–5, 147–9 non-human 127, 144, 150, 157n.30, 198 Annas, J. 50n.66, 67n.2, 68n.3, 141n.2, 194n.25 anticipation (expectation, prosdokia) 2, 9–10, 12–15, 12n.14, 19, 48–9, 70, 106–7, 110–13, 115, 122–3, 141–58, 160, 163n.46, 165–6, 170–82, 189–90 anti-intellectualism 43–4 antilogia (argument on both sides)—see argument aporia (impasse, puzzlement, resourcelessness) 28, 59 appetite (epithumia), appetites (epithumiai); see also desire 3–10, 19, 67–80, 97–100, 102–9, 106n.2, 112–13 apprehension; see also knowledge; perception 48–9, 107n.3

archē (first principle, principle, rule, starting point) 35, 36n.40, 47–8 aretē (virtue)—see virtue argument by elimination 22–3 on both sides (antilogia) 21–3, 27, 49 Aristotelian Socratic intellectualism—see intellectualism Aristotle 1–3, 5, 10–15, 18–19, 22n.3, 23–8, 31–2, 38n.46, 47–9, 54–8, 63–4, 69n.5, 78n.25, 108n.4, 114–32, 134–9, 142–3, 153, 156–66, 170–1, 173, 178–80, 184, 187n.5, 193, 200–1, 205n.1 Armstrong, G. C. 25n.12 art (craft, technē)—see craft assent 16–17, 38n.46, 57, 110, 147, 191, 198 autonomy 18, 197–8, 200–1, 206 Badhwar, N. K. 200n.39 Bailey, D. T. J. 32n.30, 37n.43 Barney, R. 3nn.2,3, 4–5, 19, 25n.13, 26n.17, 44n.58, 48n.62, 88n.13, 93n.20, 119nn.9,10 beautiful 34–5, 35n.37, 73, 83, 90–1, 100–1, 128–9, 188–9 belief (doxa); see also judgement 2, 5–7, 8n.7, 16–18, 21, 25n.15, 26, 31n.27, 37–8, 46–7, 50, 55–8, 64–6, 74–6, 78–80, 83, 91–5, 100, 131n.29, 143–4, 144n.5, 146–7, 150, 159–60, 179n.15, 183–5, 192, 194–5, 197–200 correct, true 26–7, 85, 143–4 Benson, H. 32n.30, 36nn.38, 40, 37n.44, 40n.51 Bernecker, S. 141n.2, 142n.3, 155n.27, 164n.48, 165n.49 bivalence 43–4 Bobonich, C. 77, 87n.9, 94n.23 body 29–30, 35–6, 45, 60–1, 109, 129–30, 147–9, 151, 153–4, 194–6, 199, 203–4 Bostock, D. 10n.12 boulesthai (to desire), boulēsis (wish, desire)—see desire Brady, M. 131n.29 Brennan, T. 17n.19, 18, 184, 193–5, 206–7

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

218

     

Broadie, S. 13, 39n.49, 116n.5, 124n.14 Brickhouse, T. 21n.1, 24n.9, 25n.13 Burnet, J. 10n.12, 101n.4 Burnyeat, M. 130n.26 Byrd, M. N. 32n.30, 33n.31 Carone, G. R. 21n.1, 50n.64, 72, 73n.13, 77, 77n.23, 78, 88n.13, 93nn.20, 21, 107–8, 108n.6 Cartesian dualism 5–6 Caston, V. 158nn.32, 33, 159nn.34, 38 cave 104 character 3–4, 10–11, 15, 23–4, 55, 120–1, 130, 136, 142, 148n.11, 170–3, 177–82 degradation of 18, 195 Charles, D. 127n.20 choice (prohairesis) 5, 11, 17, 55–7, 63, 65–6, 97, 101, 117, 118n.8, 119–20, 139, 143–6, 151–2, 160, 173–4, 179n.15, 184, 187–8, 191–2, 195, 197–200, 205–6 Cicero 16–18, 49, 153, 155–6, 159n.35, 162, 177, 183, 185–6, 189, 189n.12, 190n.14, 193n.22, 194–5, 200–1, 206–7 circularity 44, 46–7 confidence 2, 5, 18–19, 114–15, 119–23, 129, 136, 165–6, 188–90, 194–5, 204 contradiction 2–3, 12–13, 22–3, 37–43, 60, 88, 94, 101–2, 106n.2, 108nn.5, 6, 136, 138–9 Cooper, J. M. 28n.21, 35n.37 Cornford, F. M. 67n.2 Costello, W. 32n.30, 33n.31 courage 10–13, 28–30, 41, 43–5, 50, 61–2, 114–32, 134–9, 188–90, 193, 195 civic courage 50n.65, 124 cradle arguments 11 craft (art, technē) 4–5, 22–3, 25–7, 29–33, 40n.50, 42–8, 62–3, 150, 192 of measurement (technē metrētikē) 175 of politics (politikē technē) 21–2, 28–9, 50, 61–2 Curzer, H. 124n.12, 126n.19 Cyrenaics 1, 11–12, 12n.15, 152, 154–7 danger 10–11, 57–8, 70, 116–18, 120–3, 137–9, 164–5, 200, 206 death 16, 117–19, 122–3, 124n.14, 127n.21, 131, 136–7, 155–6, 177, 187, 189, 197–8 definition 28, 33–4, 36–41, 43n.57, 72–3, 115, 184 Delcomminette, S. 144n.4, 147n.9, 150nn.15, 18 deliberation 27–33, 36–7, 44–5, 47–50, 61, 127n.21, 141, 157–60, 175, 179–81 good (euboulia) 28–9, 50, 62–3, 131–2 demonstration 34–5, 38–9, 42, 62, 71, 126, 171–2, 174

desire (boulesthai, boulēsis) 1–3, 8–12, 16–17, 42n.54, 47–50, 54–5, 63–4, 69–73, 78, 82–4, 86–9, 92–3, 95, 102–3, 106–10, 124, 127–9, 131nn.28, 29, 143–50, 146n.7, 147–50, 150n.16, 156–7, 162n.45, 196–9 appetitive (epithumia) 5–9, 19, 67–71, 73–5, 77–8, 99, 106, 109 blind 7–8, 72, 91 content of 92–3, 99–100, 102 correct 7–8, 85–94, 100–2 fulfilment 91–2 function of 109 (for the) good 4–5, 8–9, 21, 46–7, 73–4, 78–80, 85, 87, 89–90, 93, 103, 105 good-independent 2–3, 76–80 irrational/ non-rational (see also desire, blind) 21–2, 21n.1, 26, 48–9, 55–6, 63–4, 67, 97–103, 107–8 object of 48–9, 62, 72, 75, 78, 83–5, 93, 103, 105, 107–8, 147–9 rational 68, 71, 73–4, 79–80, 99, 106 reason-following vs. reason-providing 9, 106–7 spirited 73–4 unqualified 71, 74–5, 79–80 despair 148–50, 195 deus ex machina 39 development—see moral development Devereux, D. 25n.13 dialectic 27, 32, 35n.37, 38n.46, 39–40, 42–3, 46–8, 50, 126n.19, 138 dialogue 9 dianoia (thought)—see thought diaphonia (disharmony, disagreement)—see disagreement; disharmony diathesis (disposition)—see disposition disagreement (diaphonia) 35–9, 58, 144n.4, 152–3, 171, 203 disharmony (diaphonia) 36–9, 37n.42, 42–3 disposition (diathesis) 1, 17–19, 45–6, 104, 114–16, 123, 129–30, 134–6, 139, 204–5 Divided Line 39 divine, divinity; see also god 26–7, 33–4, 42–3, 45–6, 112 Dobbin, R. 191 Doctrine 4–5, 21–2, 26–8, 31n.28, 46, 49, 54–61, 63–4, 76, 78, 94–5, 103n.6, 134, 204 Dow, J. 157n.29, 159n.37 doxa (belief)—see belief Duff, A. 131n.29 dualism, Cartesian 24–5 dunamis (power)—see power Dyson, H. 191n.18

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

      Ebrey, D. 32n.30, 39n.48 education 3–4, 8–9, 21–6, 28n.21, 29, 33–4, 41n.53, 45–6, 62–3, 104, 129–30, 139 eidos (kind, species) 72 eikos (likely, plausible, reasonable)—see likely elenchus, elenctic 21n.2, 40–1, 58 elpis (hope)—see hope emotion; see also pathos, pathē 1–3, 10–13, 16–17, 21–6, 29–30, 48–9, 61–2, 114–15, 117–18, 123, 131, 136–7, 163, 165, 187–91, 198, 200, 205–6 empiricism 28 end (goal, telos) 29, 36n.40, 98–100, 103, 107–8, 117–18, 124–5, 130, 130n.26, 145n.6, 183, 186 endurance 13–14, 33–4, 47, 124, 124n.14, 127 Engel, S. 165n.49 enkrateia (self-control)—see self-control Epictetus 16–17, 188n.8, 191–2, 196–201, 203, 206 Epicurus 28, 152–3, 177, 181 epistēmē (knowledge)—see knowledge epistemology 29–30, 154–5, 165–6 epithumia (appetite, appetitive desire)—see appetite; desire euboulia (good deliberation)—see deliberation eudaimonia (happiness, flourishing)—see happiness Eudoxus 135n.3 Eumaeus 162–5, 179–80 eupatheiai (good feelings) 16n.18, 17, 188–90, 192 Evans, M. 19, 90n.17 evil, evils (see also vice, vices, vicious) 41n.53, 43–5, 57, 115, 145–6 expectation (anticipation, prosdokia)—see anticipation false 5, 16–18, 22–4, 26n.16, 40–1, 45–7, 54–6, 60–1, 63, 65–6, 107–8, 110–11, 145–6, 163, 197–8 anticipation 147, 149n.12, 151n.19 pains 149n.12 pleasures 111–12, 143–4, 147–8, 148n.11, 149n.12, 150–2 fear 2, 10–13, 17–18, 35, 65, 114–32, 136–9, 149–50, 149n.12, 158–60, 163–4, 174, 177, 193–8, 206–7 fine, the fine (kalon, to kalon) 2, 10–11, 13–14, 55, 101–2, 115, 117–32, 160–1, 179–81, 185 Finkelberg, M. 30n.25 flourishing (eudaimonia, happiness)—see happiness Form of the Good 39, 44, 50, 110 Forster, M. 28–9, 37n.41, 38n.45

219

Franklin, L. 32n.30 Frede, D. 82n.1, 110, 144nn.4, 5, 152n.20 Frede, M. 189n.12 Freud, S. 107 friendship 119, 152–3, 191–3 Garver, E. 124n.13 Gentzler, J. 32n.30, 37n.42, 37n.44 Gill, C. 191n.18, 197n.33 Giovacchini, J. 153n.22 goal (end, telos)—see end god; see also divine, divinity, 117n.7, 134–7, 177, 191n.18, 204, 207 good, goods (agathon) 1–12, 14, 16–19, 24, 27, 34, 40n.50, 41n.53, 43–5, 47–9, 57, 62–3, 65–6, 70, 76–80, 82–95, 97–103, 105–10, 112–13, 118n.8, 125–6, 131, 134–6, 144–6, 153–4, 160–1, 164–5, 179–81, 183–95, 199, 203–7 bodily 203–4 desire for (see desire) 4–5, 7–9, 21, 46–7, 73–4, 84–5, 102–3, 105–8 division of 194–5 external 203–4 Form of the Good / The Good – see Form of the Good normative conception of 85 relational 193 telos (end, goal) and 29, 98–9 Gorgias 22–3 Gosling, J. C. B. 144n.4, 152n.20, 157n.29 Graver, M. R. 17n.19, 156n.28, 189n.12 grief 155–6, 199 Grube, G. M. A 35n.37, 76, 82n.3 habituation—see moral development Hackforth, R. 144n.5, 152n.20 Hamlyn, D. W. 160n.39 happiness (eudaimonia); see also vita beauta 1–2, 5–6, 10–13, 15–19, 27, 31–2, 40n.50, 65–6, 107, 114, 125–6, 131n.27, 134–6, 138, 144–5, 153, 163, 183–201, 203–7 sufficiency of virtue for happiness—see virtue harmony, harmonious 3–4, 18–19, 32–3, 35–9, 42–3, 46, 204 Harte, V. 9–10, 111–12, 147n.9, 151n.19, 152n.20 health; see also body 16, 23n.7, 44, 70, 97–8, 118, 137n.8, 146–7, 184, 188–9, 192, 194 hēdonē (pleasure)—see pleasure hedonism; see also pleasure 13, 25–7, 43–4, 75, 134–6, 144n.5, 152, 154–5 Heil, J. 116n.5 Hesiod, Hesiodic 5, 30–3, 65

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

220

     

hexis (state, disposition of character) 114–15 Hobbes, T. 28 Homer, Homeric 5, 30–1, 64–5, 162, 163n.47 hope (elpis) 110, 131, 144–5, 147–50, 152, 157n.30 Hume, D. 24–5 humility 23n.7, 28–9, 43–4 Hutchinson, D. S. 10n.12 hypolēpsis (supposition)—see supposition hypothesis, hypothetical; see also doctrine; method, methodology; theory 4–5, 34–6, 39, 45–50 method of 4, 21–3, 32–42, 49 higher hypothesis 4–5, 35–6, 39, 42–4, 47–9 ignorance 28, 43–4, 47, 56, 117–18 illness 16, 174–5 images 90, 110, 150–2, 157–60 imagination (phantasia) 1–2, 111–12, 147n.9, 156–8 immaterial 191–2 impression (phantasia)—see phantasia incontinence (akrasia, weakness of will)—see weakness of will indifferents, (what is) indifferent (in Stoic thought) 16–17, 183–4, 188–93, 192n.21, 195n.29 intellectualism 2–5, 10, 23–32, 39–50, 54, 58–9, 62–3 Aristotelian Socratic 54–7, 63–4 Socratic 2–4, 21–2, 24–9, 46–7, 57–9, 64–6, 115 intentional object 9, 108–9, 177 inquiry 5, 22–3, 38, 42–3, 46, 49–50, 58, 180–1 intuition, intuitive 31n.27, 38–9, 41–2, 45–6, 48–9, 134–6, 161–2, 201 Inwood, B. 136–7 Ionescu, C. 32n.30 Irwin, T. 3n.2, 10n.12, 24–5, 26n.16, 76, 85n.7, 89n.15, 94n.23, 108n.6, 184, 186n.4, 187n.5 Iwata, N. 32n.30, 33n.32, 34n.34, 200n.39 Jiang, X. 131n.28 judgement; see also belief 16–17, 29–30, 34, 56, 58–9, 61, 64–5, 83, 100, 107–8, 110, 111n.10, 119, 144, 150, 191n.16, 197, 199–200 justice; see also virtue 5, 16, 23n.7, 25n.15, 30–3, 40n.50, 43–4, 47, 50, 55, 62, 65–6, 72, 83, 100–2, 128, 188–9, 193 Kahn, C. 23n.6, 24–5 kalon, to kalon (fine, the fine)—see fine, the fine Kamtekar, R. 21n.1, 40n.50, 46n.60, 47n.61, 50n.64, 88n.13, 92n.19 Kant, I. 24–5, 25n.13, 88–9 King, R. A. H. 158n.33, 162n.45

knowledge (epistēmē) 1–6, 21–30, 32–4, 38n.46, 41n.53, 42–50, 55–6, 58, 60–5, 72–3, 79–80, 99–102, 103n.7, 125–6, 139, 186, 188–9, 192n.21 of good and evil 2–3, 43–5 reflexive 43–4 self-knowledge 23n.7, 43–4, 102 Korsgaard, C. 128n.24 Landry, E. 32n.30 learning 9, 25–6, 33–4, 41n.53, 45–6, 72, 75, 86, 89, 106, 130n.26, 141–2 Leighton, S. 127n.21 Leontius 26n.16, 71n.8, 94, 108n.6, 109 Lesses, G. 77n.23, 93n.20, 186n.4, 190n.15 Lewis, C. S. 187 Lewis, D. 8n.8, 95n.24 likely (eikos, plausible, reasonable) 41 logismos (reasoning, calculating) 75n.18, 86–7, 143–4, 151–2 logos (account, argument, reason, statement, word) 10–11, 23–4, 34 Long, A. A. 56n.4, 191n.16, 195n.27, 198n.36 Lorenz, H. 10n.12, 72n.11, 87n.9, 88n.12, 94n.23, 147n.9, 148n.10, 157n.30, 158n.31 Lovibond, S. 152n.20 luck 26–7, 33–4, 42–3, 45 lupē (pain)—see pain Madvig, N. 153n.24 Margalit, A. 165n.50 materialism 28 Mellor, D. 162n.42 memory (mnēmē) 2, 9–10, 13–15, 19, 110, 111n.8, 141–54, 155n.27, 156–61, 162n.45, 163–6, 170–82 Menn, S. 32n.30, 33nn.32, 33 Meyer, S. 141 methodology, method; see also doctrine; hypothesis, hypothetical; theory 4, 21–3, 26–7, 32–4, 36–42, 49, 54, 60 Mills, M. 115n.4 mnēmē (memory)—see memory model 15, 50, 142–3, 145–6, 149–52, 156–8, 170–82 moderation (sophrosunē, temperance)—see temperance mollusc 14n.16, 143–5, 147, 148n.10 Moore, G. E. 144n.5 moral development 1–2, 8, 99–102, 104, 124n.14, 129–30, 179–81 moral psychology 1–2, 5–8, 10, 15–16, 19, 21–2, 24–6, 29–30, 50, 83, 89, 94–5, 97, 99–100, 104, 165

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

      moral value—see value Moss, J. 6–9, 10n.12, 19, 50n.64, 75n.18, 77n.24, 83n.4, 88n.13, 93nn.20, 21, 107–8, 128n.24, 141, 150n.18, 158n.31, 178nn.13, 14 motivation; see also desire 1–9, 18–19, 21–2, 25n.13, 29–30, 47–9, 61, 63, 68–9, 83, 88, 89n.16, 91–3, 97–103, 108, 116n.5, 118, 124–5, 138–9, 192–3, 206–7 nature 1–2, 5–6, 8–11, 13, 15–19, 25–6, 67, 85, 89n.16, 92–5, 99–100, 104, 106n.2, 107–9, 125, 127n.21, 128, 134–6, 142, 145–8, 150, 156, 165, 170–2, 179n.15, 180–3, 186, 189–90, 193, 204–5 human nature 11n.13, 21–2, 116, 123, 127n.21, 136 second nature 99–100, 124n.14, 139 Nehamas, A. 50n.66 Nielsen, K. 118n.8 non-rational desire; see desire normative, normativity; see also good, the good 8, 24–5, 77, 85, 92, 94–5 nous (understanding)—see understanding Nussbaum, M. 144n.5, 158, 183–5, 193, 194n.24, 195, 206n.2 Oaklander, L. N. 162n.42 O’Keefe, T. 156n.28 opposite, opposition 31n.27, 37–8, 42n.54, 43–4, 65, 69–70, 71n.8, 74, 78n.25, 84, 86, 89, 94–5 O’Reilly, K. 14n.16, 15, 141 orthos logos (right reason); see also logos 10–11 Osborne, O. 157n.30 pain (lupē) 3–4, 11–15, 105, 112, 114–15, 116n.5, 117–18, 118n.8, 120–1, 123–7, 129–32, 134–7, 139, 141–66, 170–5, 177–82, 184, 189 painlessness 123, 151–3 painter 110, 111n.8, 143–5, 150, 157–8 Palmer, J. 32n.30, 33n.31 Pangle, L. 124n.12 paradox—see Socratic Parfit, D. 146n.7 participation 2, 16, 18n.20, 55, 188–92, 205 parts 3–8, 10–11, 21, 23–4, 28–9, 31–4, 31n.27, 45–7, 55–7, 67–9, 71–5, 71n.8, 75n.18, 77–80, 89, 94–5, 97–100, 102, 106–9, 112, 118, 126n.19, 127n.21, 143, 146–8, 151n.19, 154, 162n.45, 172, 174–6, 197, 203–4 of virtue 28–9, 43–4, 188–90 of happiness; see also happiness; vita beauta 149n.12, 188–90, 194

221

passion; see also pathos, pathē 2–3, 23–5 pathos, pathē (affection, emotion, experience, passion, undergoing) 11–12, 16–17, 115, 118, 148–9, 154–7, 159–60, 178–9, 188–90 Pears, D. 115n.1, 116n.5 Pearson, G. 128n.22, 129n.25 Penner, T. 67n.2, 68, 76n.19, 152n.20 perception (aisthēsis) 1–2, 7–8, 83–8, 93, 94n.22, 98–100, 106–7, 108n.4, 110, 111n.10, 114–15, 147–8, 148n.10, 150, 156–60, 165, 178–81 of good, goodness 7–8, 83–4, 94n.22 perfection 18–19, 39–40, 57, 189–90, 204–5 perishable 17–18, 188, 194 Persson, I. 146n.7, 149n.14 phantasia (imagination, impression) 78n.25, 156–60, 165, 178–80 phronein (to think) 143–4 phronēsis (practical wisdom)—see practical wisdom picture 110–12, 151, 158–60, 179–80 piety 28, 30–2, 40n.50, 43–4 Plato 1–11, 13–15, 19, 21–50, 129, 196 plausible (eikos, likely, reasonable)—see likely pleasure (hēdonē); see also hedonism 1, 3–4, 9–17, 40n.50, 58–9, 74, 77, 86, 89–90, 106–15, 118n.8, 119, 124–32, 134–6, 138–9, 141–3, 165–6, 170–82, 188–9, 199 advance instalment of 12n.14, 14n.17, 111–12, 151–2, 154–6 false 143–4, 147, 149n.12, 150, 152 true 147–8, 149n.12 unitemporal (tied to a single moment) 11–12, 154–6 weighing of 13, 146n.7, 173–4 ponoi (toils)—see toils power (dunamis) 24–5, 30n.24, 36, 39, 40n.50, 41–7, 62, 68, 91, 152–3, 156, 158, 159n.35, 165–6, 181–2, 191–2, 196–7 practical wisdom (phronēsis, sapienta) 3–4, 16, 23–4, 25n.13, 27, 29, 31–2, 45, 107–8, 143–4, 187–8, 191–2, 195, 197, 199, 205–6 pre-live, pre-living; see also anticipation 14, 111, 111n.8, 142, 151–2, 165–6, 170–3, 176, 178–9, 181–2 preservation 119–20, 126, 147–8, 150, 153–4, 178–9, 197 Price, A.W. 9–10, 14n.17, 77n.23, 87n.9, 93n.20, 107n.3, 115n.2, 124n.14, 173n.6 Principle of Opposites 69–70, 71n.8, 78n.25, 86 Prior, A. N. 162 prohairesis (choice)—see choice prosdokia (anticipation, expectation)—see anticipation

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

222

     

Protagoras 8n.8, 21–2, 27, 30–3, 30n.24, 94–5 protreptic 24–5, 27 punishment 47, 127, 129–30 reason: see also logos 3–4, 6, 8–11, 18, 23–4, 25n.13, 32n.29, 34n.36, 39, 48–9, 55–9, 62–3, 68, 69n.4, 71n.8, 75, 86–7, 98–100, 106–7, 108n.5, 112n.13, 114–15, 117, 119–20, 122–3, 125–6, 143–6, 157n.30, 174–5, 184–5, 189–90, 198 reasonable (eikos, likely, plausible)—see likely recklessness 13, 116–17, 118n.8, 119–23, 125, 137 recollecting; see also memory 12, 15, 35–6, 45–6, 110, 111n.8, 141–4, 146–66, 162n.45, 170–3, 176n.9, 177–82 Reeve, C. D. C. 75n.18, 76, 82n.3, 85n.7 relationships; see also friendship 136, 187–8, 190, 192, 200–1, 206 re-live, re-living; see also memory, pleasure, recollection 14–15, 111n.8, 142, 151–3, 165–6, 170–2, 176–9, 181–2 Reydams-Schils, G. 192n.21 Richardson Lear, G. 128n.24, 131n.28 Rist, J. M. 189–90, 191n.16 Robinson, R. 32n.30, 34n.36, 36n.38, 38nn.45, 47, 40n.51 Rogers, K. 131n.27 Russell, D. 17–19, 111, 145n.6, 147n.9, 152n.20, 187n.6, 194n.25, 203–7 Sage 17, 153, 176, 180n.16, 184, 204 Sanford, J. 128n.23 sapientia (practical wisdom)—see practical wisdom Scheiter, K. M. 158n.31 Schiffer, S. 9, 106–8 Scolnicov, S. 32n.30 Scott, D. 32n.30, 40n.51, 43n.56 Scribe 110, 143–5, 150, 157–8 Second Sailing 34 Sedley, D. 18–19, 18n.21, 32n.30, 35n.37, 150n.17, 195n.27 Segvic, H. 30n.24, 31n.26 self-awareness; see also knowledge 25n.13, 25n.15, 43–4, 128, 138–9, 144–5, 175, 180–1 self-control (enkrateia) 35, 126–8, 131, 134n.1 self-knowledge—see knowledge Sellars, J. 184, 193n.23 Sherman, N. 10n.12, 187 Shorey, P. 77n.22, 106 sickness—see illness Sidgwick, H. 147n.9, 160n.40, 161–2 Simpson, P. L. P. 24n.8 Small, J. P. 159n.35 Smith, M. 8n.8, 95n.24

Smith, N.D. 21n.1, 24n.9, 25n.13 Smith, Q. 162n.42 Socrates 2–9, 14, 21–46, 48–9, 59–66, 60n.6, 68–80, 82, 84, 86–94, 86n.8, 97–8, 100–4, 106–12, 109n.7, 111n.9, 143–52, 173–5, 197n.32 Socratic dialogues/ early dialogues of Plato 2–4, 19, 21–33, 23n.7, 26n.16, 34n.36, 39–44, 45n.59, 46–50, 54–5, 60, 87–8, 106 Intellectualism—see intellectualism Paradox 46, 49 refutation 38n.45, 41 sophia (wisdom)—see wisdom sophist, sophistic, sophistry 21–3, 25–32, 38n.46, 42–3, 49, 62 sophrosunē (moderation, temperance)—see temperance Sorabji, R. R. K. 10n.12, 150n.15, 157n.30, 158nn.31, 33, 159n.35, 191nn.17,19, 20, 195n.29 soul (psūche); see also immaterial 8–9, 16–17, 25–6, 32–8, 42–4, 47–9, 57, 60, 60n.6, 67, 75, 83, 85, 89, 100–1, 104, 106–7, 109–10, 121–2, 143–5, 147–50, 153–8, 162n.45, 188–9, 191n.18, 197 division of 6, 21, 69–70, 75, 75n.18, 86–8, 97–8 irrational 21, 23–4 non-rational 8, 10–11, 67 rational 2–4, 21, 23–4, 55 tripartite 3–8, 10–11, 21, 24, 67–9, 71–2, 74, 75n.18, 78n.25, 79–80, 85–8, 97–100, 102, 106n.2, 107–8 statesman 43n.55 Stephens, W. O. 184, 191nn.17, 19, 20, 196n.30, 197n.31 Stoics, Stoicism 1, 5, 15–19, 26n.16, 31n.28, 56–7, 183–201, 203–7 Street, S. 8n.8, 94–5 sufficiency thesis—see virtue, sufficiency sunkatathesis (assent)—see assent supposition (hypolēpsis) 159 Taylor, C. C. W. 10n.12, 105, 122n.11, 127n.21, 157n.29 teaching 2–5, 21, 25–6, 28–30, 32–5, 41–3, 45–6, 60–2, 134–6, 196 technē (art, craft)—see craft teleology 85–6, 91–5, 98–100, 102–4, 204 telos (end, goal)—see end temperance (sophrosunē, moderation) 16, 28, 30–2, 31n.27, 43–4, 50, 62, 94, 107–8, 114, 119, 127–9, 188–9, 193 temporal neutrality 145–6, 174

OUP CORRECTED PROOF – FINAL, 11/11/2022, SPi

      testimony 41, 194 theory; see also doctrine 5–6, 15–16, 19, 21–6, 28–33, 35–8, 44n.58, 47, 50, 58, 61, 84n.6, 107, 125, 125n.16, 134–6, 138, 153n.24, 157–8, 177, 188–9, 205–6 thirst 6–10, 8n.7, 67–76, 78–80, 83, 85, 87–9, 93, 102, 104–10, 112–13, 147–8, 151 thought (dianoia) 159 Thucydides 30–1, 48–9, 65 toils (ponoi) 155–6, 160n.40, 161–2, 164–5 tranquillity 153–4, 195n.29 Tredennick, H. 25n.12 true, truth (alēthēs) 4–5, 22–5, 34–7, 39, 47–9, 56–60, 63–6, 72, 84–5, 88, 93–4, 109–10, 117–18, 136–7, 145–9, 158, 163, 207 anticipation 110 belief, opinion; see also belief; judgement 27, 34, 85, 143–4, 197–8 fears 149n.12 memory 163–4 Tsekourakis, D. 190nn.13, 15 Tsouna, V. 141 tyrant 191–2, 198 understanding (nous) 143–4 valence 14–15, 172–3, 179n.15, 180–2 value 1–3, 10–11, 13–19, 18n.20, 30, 44, 48–9, 63, 102, 119–20, 125n.17, 126, 130, 130n.26, 131n.29, 132, 145–6, 147n.9, 148–9, 165, 175, 184, 186, 188–9, 193, 205 vice, vices, vicious; see also evil, evils; injustice; recklessness 16–17, 23–4, 103–4, 114, 120–1, 129–30, 134–6, 183–4, 188–9, 195, 204–5, cardinal 188–9 cowardice 10–13, 116–22, 125, 127, 129–30, 136, 189 recklessness 13, 116–23, 118n.8, 125, 137 personal 99–100, 104 Vigani, D. 127n.20 virtue, virtues, virtuous; see also courage, justice, practical wisdom, temperance, wisdom 1–5, 10–16, 23–34, 37–44, 46–7, 50, 55–6, 62–6, 99–100, 104, 114–16, 118, 120–1, 122n.11, 124–30, 131n.27, 132, 134–6, 183–6, 188–90, 192–4, 200–1, 203–7 action 2, 10–14, 16–19, 27, 31–2, 55, 118, 128–9, 130n.26, 139, 153, 185–8, 190, 192, 195–6, 199–201, 205–6 content of 2–3, 45, 55, 64–5, 193 (as a) craft (technē) 4–5, 22–3, 29–30, 32, 42–6, 50, 63 divine 45–6, 117n.6, 121–2

223

exercise of 31–2, 114–15, 124, 126n.18, 131n.27, 134–6, 183 higher 45–6 (as) knowledge / wisdom / reason 2–5, 21, 23–30, 32–4, 42–6, 49, 58, 60–1 lower 45 parts of 18–19, 21, 23–4, 28–30, 31n.27, 40n.50, 43–6, 50, 61, 64–5, 125–6, 145–6, 188, 190–2, 195, 204–5 (as a) power (dunamis) 44n.58 self-sufficiency, sufficiency of/for happiness 17–19, 183–8, 192–6, 199, 203–6 teachability of 23n.7, 25–7, 29, 33–5, 42–3, 45, 59–60, 62 vita beata (happiness, flourishing); see also happiness 183 voluntary 119, 139, 198n.36 Walzer, R. R. 136–7 Wardy, R. B. B. 157n.29 Warnock, M. 149n.13 Warren, J. 9nn.10,11, 10, 12–15, 19, 105n.1, 106, 111–12, 141, 146n.8, 170–82 weakness of will (akrasia, incontinence) 2–5, 21, 23–6, 46–9, 54–9, 63–4, 108n.6, 119, 127, 173–4, 197, 199–200 Webster, T. B. L. 162n.44 Wedin, M. V. 159n.35 Weiss, R. 33n.33, 77, 77nn.22, 23, 83n.5, 93nn.20, 21 White, N. 193, 194n.24 White, S. 194n.25, 195 wholehearted 119, 124n.14 Wielenberg, E. 128n.24 Williams, B. 152n.20 wisdom (sophia) 23n.7, 25–30, 28n.21, 31n.28, 33–4, 40–1, 40n.50, 43–4, 48–50, 61–2, 72–5, 79–80, 97–8, 134–6, 183, 185–9, 191–3, 192n.21, 200–1, 205–6 practical (phronēsis, sapienta)—see practical wisdom Wohl, V. 41n.52 Wolfsdorf, D. 32nn.29, 30, 33n.32, 40–1 Wollheim, R. 9–10, 14n.17, 111–12 Woods, M. 76n.19 Woolf, R. 10, 12–15, 19, 136–7 Wright, M. 162n.44 Young, C. 131n.28 Zeyl, D. J. 82n.1