Skill in Ancient Ethics: The Legacy of China, Greece and Rome 9781350104327, 9781350104334, 9781350104341, 1350104329

Illustrating the centrality of skill within ancient ethics, including Socrates' search for expertise in virtue, the

135 67 3MB

English Pages 392 [393] Year 2021

Report DMCA / Copyright

DOWNLOAD PDF FILE

Table of contents :
Cover page
Halftitle page
Series page
Title page
Copyright page
Contents
Contributors
Introduction
1. Skill in Greek and Roman Ethics
2. Skill in Chinese Ethics
References
Part One Skill in Plato’s Ethics
1 Socrates: Apprentice at Politics
Introduction
1. Is Socrates not the first?
2. Only Socrates
3. Being a craft sman and performing the functions of the craft
4. How Socrates performs the craft of politics
5. Summary and conclusion
References
2 The Question is not ‘Can Virtue be Taught?’ but ‘Can Virtue be Learned?’
1. The Meno : is virtue teachable?
2. Protagoras (320c–328d)
3. Protagoras in the Protagoras
4. Teaching a language vs teaching flute-playing
5. A defence of Protagoras’ evidence over Socrates
6. The nature of a skill according to Socrates (and Plato)
7. Protagoras’ evidence fi ts Socrates’ assumptions regarding the nature of a skill
8. Conclusion
References
3 The Contest Between Philosophy and Rhetoric in Plato’s Gorgias
Introduction
1. Orders and ends
2. The architectonic craft
3. The objects of rhetoric and philosophy
4. A Gorgian inconsistency
5. Review of the analogy and disanalogy
6. Philosopher kings in the Gorgias
7. The use of rhetoric
References
4 A Dewian Conception of Skill as Clue to the Analogy Between Craft and Virtue in the Platonic Dialogues
1. Specifying an art’s end
2. A false analogy: one skill, one end
3. Scholarly consensus
4. A Dewian conception of technê
5. Applying the Dewian model to the Republic
6. Conclusions
References
Part Two Skill in Aristotle’s Ethics
5 Steering Against the Bad: An Aristotelian Account of Virtue as Two-way1
1. Limiting the powers disanalogy between skill and virtue
2. Acting knowingly and for the good
3. Constraining the scope of possibilities for action
4. The two-wayness of virtue’s rational order
5. Steering against the bad as a way of getting it right
6. Conclusion
References
6 Virtue Cultivation and the Skill of Emotion Regulation
Introduction
1. The virtue-skill analogy
2. The Orektikon and character virtue
3. Phantasia as non-rational evaluative cognition
4. Appearance-based emotions
5. Phantastic habituation and emotion regulation
References
7 Aristotle on Techne: Two Theses in Search of a Synthesis?
Introduction
1. The ‘official’ conception of techne: Cognitive-theoretical emphasis
2. Techne as practico-experiential
References
Part Three Skill in Stoic Ethics
8 The Craft sman of Impulse: Chrysippus on Expertise and Moral Development1
1. Preliminaries: impulse and right reason
2. Second senses: expertise and concept formation
3. The craft sman of impulse
4. Conclusion
References
9 Being and Becoming Good: Seneca’s Two Moral Conceptions of Ars
Introduction
1. Wisdom and craft s or skills
2. Wisdom as a craft or skill
3. Becoming good as an ars
4. The ars of the proficiens and the ars of the wise with regard to the blows of fate
5. Doing good deeds as an activity of the proficiens and an activity of the wise
6. Conclusion
References
Part Four Skill in Confucian Ethics
10 Cultivating Goodness or Manifesting Goodness: Two Interpretations of the Mencius
1. The question
2. The Cultivation Model
3. The Manifestation Model
3a. The Manifestation-Will Model
3b. The Manifestation-Mixed Xin Model
4. Concluding remarks
References
11 Ritual as a Skill: Ethical Cultivation and the Skill Model in the Xunzi
1. The skill model of virtue
2. Xunzi’s ethics and the skill analogy
3. Ritual learning as skill acquisition
4. Ritual creation and skills
References
Part Five Skill and Ethics in the Zhuangzi
12 A Path with No End: Skill and Dao in Mozi and Zhuangzi
Introduction
1. Dao, models and skill in the Mozi
2. Cook Ding on dao versus skill
3. Proficiency in the patterns
4. The ends of dao
5. An ethics of dao and de
6. Conclusion
References
13 Skilfulness and Uselessness in the Zhuangzi
1. A tension in the text
2. Uselessness and timeliness
3. Performance and non- attachment
4. Between attachment and detachment
5. Uselessness as the basis of usefulness
6. Uselessness and politics
7. Conclusion
References
14 Dao and Agency: What do the Zhuangzi’s Skill Stories Tell Us about Life?
Introduction, and some remarks on ‘skill’ in the Zhuangzi
1. Dao and ‘skill’ in the stories
2. Do you have (a) dao?
3. Dao, agency and life
References
Part Six Comparative Perspectives on Skill in Ethics
15 Can One Become Wise by Learning to Catch Cicadas? Analogies between Craft s and Wisdom in Daoism and Stoicism
Introduction and texts
1. Why is excellence in skills paradigmatic for the art of living?
2. The differences in the Stoic and Daoist craft analogies
3. The common denominator behind the differences
4. Why the specialist crafts can (or cannot) be conducive to wisdom
References
16 Gendered Skill: Chinese and Greek Skill-Knowledge Analogies from Archery and Weaving
1. Chinese archery metaphors
2. Greek archery metaphors
3. Chinese analogies from weaving
4. Greek analogies from weaving
5. Conclusions
References
17 The Skilful Wanderer: On the Risks and Rewards of Travel in Plato and Zhuangzi
Introduction
1. The a(nti)teliology of wandering
2. Wandering inside the world
3. Wandering as transformative: Between ‘inside’ and ‘outside’
4. The privations of the wanderer in Ancient Greece
5. Wandering in mere eikos: Sophistry in Phaedrus
6. Toward an ethics of wandering: Zhuangzi and Plato compared
Works Cited
Index
Recommend Papers

Skill in Ancient Ethics: The Legacy of China, Greece and Rome
 9781350104327, 9781350104334, 9781350104341, 1350104329

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

Skill in Ancient Ethics

i

Also available from Bloomsbury Chinese and Buddhist Philosophy in Early Twentieth-Century German Thought, by Eric S. Nelson Confucian Ethics in Western Discourse, by Wai-ying Wong Cultivating a Good Life in Early Chinese and Ancient Greek Philosophy, edited by Karyn Lai, Rick Benitez and Hyun Jin Kim Ethics: The Key Thinkers, edited by Tom Angier Health and Hedonism in Plato and Epicurus, by Kelly Arenson Transcendence and Non-Naturalism in Early Chinese Thought, by Alexus McLeod and Joshua R. Brown

ii

Skill in Ancient Ethics The Legacy of China, Greece and Rome Edited by Tom Angier and Lisa Raphals

iii

BLOOMSBURY ACADEMIC Bloomsbury Publishing Plc 50 Bedford Square, London, WC1B 3DP, UK 1385 Broadway, New York, NY 10018, USA 29 Earlsfort Terrace, Dublin 2, Ireland BLOOMSBURY, BLOOMSBURY ACADEMIC and the Diana logo are trademarks of Bloomsbury Publishing Plc First published in Great Britain 2022 Copyright © Tom Angier, Lisa Raphals and Contributors, 2022 Tom Angier and Lisa A. Raphals have asserted their right under the Copyright, Designs and Patents Act, 1988, to be identified as Editors of this work. Cover design: Louise Dugdale Cover images: Top, Tang Dynasty fresco, 4X-image. Bottom, Odysseus shoots a man with an arrow, clu, both via iStock. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage or retrieval system, without prior permission in writing from the publishers. Bloomsbury Publishing Plc does not have any control over, or responsibility for, any third-party websites referred to or in this book. All internet addresses given in this book were correct at the time of going to press. The author and publisher regret any inconvenience caused if addresses have changed or sites have ceased to exist, but can accept no responsibility for any such changes. A catalogue record for this book is available from the British Library. Library of Congress Cataloging-in-Publication Data Names: Angier, Tom P. S., editor. | Raphals, Lisa Ann, 1951– editor. Title: Skill in ancient ethics : the legacy of China, Greece and Rome / edited by Tom Angier and Lisa A. Raphals. Description: London ; New York : Bloomsbury Academic, 2021. | Includes bibliographical references and index. Identifiers: LCCN 2021015213 (print) | LCCN 2021015214 (ebook) | ISBN 9781350104327 (hb) | ISBN 9781350104334 (epdf) | ISBN 9781350104341 (ebook) Subjects: LCSH: Philosophy, Ancient. | Techne (Philosophy) Classification: LCC B111 .S55 2021 (print) | LCC B111 (ebook) | DDC 170.938—dc23 LC record available at https://lccn.loc.gov/2021015213 LC ebook record available at https://lccn.loc.gov/2021015214 ISBN:

HB: ePDF: eBook:

978-1-3501-0432-7 978-1-3501-0433-4 978-1-3501-0434-1

Typeset by RefineCatch Ltd, Bungay, Suffolk NR35 1EF, UK To find out more about our authors and books visit www.bloomsbury.com and sign up for our newsletters.

iv

Contents List of Contributors

vii

Introduction Tom Angier and Lisa Raphals

1

Part 1 Skill in Plato’s Ethics

1

Socrates: Apprentice at Politics

2

The Question is not ‘Can Virtue be Taught?’ but ‘Can Virtue be Learned?’ Naomi Reshotko

31

The Contest Between Philosophy and Rhetoric in Plato’s Gorgias Laurence Bloom

51

A Dewian Conception of Skill as Clue to the Analogy Between Craft and Virtue in the Platonic Dialogues Robin Weiss

75

3 4

Nicholas D. Smith

13

Part 2 Skill in Aristotle’s Ethics

5 6 7

Steering Against the Bad: An Aristotelian Account of Virtue as Two-way Jennifer Rothschild

97

Virtue Cultivation and the Skill of Emotion Regulation Paul Carron

115

Aristotle on Techne: Two Theses in Search of a Synthesis? Joseph Dunne

141

Part 3 Skill in Stoic Ethics

8 9

The Craftsman of Impulse: Chrysippus on Expertise and Moral Development Tue Søvsø

165

Being and Becoming Good: Seneca’s Two Moral Conceptions of Ars Stefan Rōttig

185

v

vi

Contents

Part 4 Skill in Confucian Ethics

10 Cultivating Goodness or Manifesting Goodness: Two Interpretations of the Mencius Winnie Sung

203

11 Ritual as a Skill: Ethical Cultivation and the Skill Model in the Xunzi Siufu Tang

221

Part 5 Skill and Ethics in the Zhuangzi

12 A Path with No End: Skill and Dao in Mozi and Zhuangzi Chris Fraser

241

13 Skilfulness and Uselessness in the Zhuangzi Wai Wai Chiu

261

14 Dao and Agency: What do the Zhuangzi’s Skill Stories Tell Us about Life? Karyn Lai

281

Part 6 Comparative Perspectives on Skill in Ethics

15 Can One Become Wise by Learning to Catch Cicadas? Analogies between Crafts and Wisdom in Daoism and Stoicism David Machek

307

16 Gendered Skill: Chinese and Greek Skill-Knowledge Analogies from Archery and Weaving Lisa Raphals

329

17 The Skilful Wanderer: On the Risks and Rewards of Travel in Plato and Zhuangzi Rohan Sikri

351

Index

375

Contributors Laurence Bloom is a Senior Lecturer in Philosophy at Rhodes University. He publishes primarily on Plato, including his recent book The Principle of Noncontradiction in Plato’s Republic: An Argument for Form (2017). He is currently working on Plato’s account of the nature in the Timaeus. Paul Carron is Associate Professor of Philosophy in the Baylor Interdisciplinary Core at Baylor University. His research focuses on ethics, moral psychology and Ancient philosophy. He explores the role of the emotions in the moral life, the cognitive and affective grounds of moral responsibility ascriptions, and the differences between the emotional and moral lives of human and non-human animals. Wai Wai Chiu is Associate Professor at Lingnan University, Hong Kong. His interests include pre-Qin Daoism, Mohism and Wei-Jin philosophy, specifically the relationship between knowledge, action and self-cultivation. He is a co-editor (with Karyn Lai) of Skill and Mastery: Philosophical Stories from the Zhuangzi(2019). Joseph Dunne is Cregan Professor Emeritus of Philosophy and Education at Dublin City University. He taught for forty years at St Patrick’s College Dublin, and has also taught at Duke University, the University of Oslo, and the University of British Columbia. He is author of Back to the Rough Ground: ‘Phronesis’ and Techne’ in Modern Philosophy and in Aristotle (1993) and has co-edited Questioning Ireland: Debates in Political Philosophy and Public Policy; Childhood and its Discontents: The First Seamus Heaney Lectures (2000); and Education and Practice: Upholding the Integrity of Teaching and Learning (2004). His Persons in Practice: Essays between Education and Philosophy is forthcoming from Wiley. Chris Fraser is Professor and Chair in the Department of Philosophy at the University of Hong Kong. He has published widely on early Chinese thought, covering texts associated with Daoism, Mohism, and Confucianism and topics in ethics, action theory, epistemology, philosophy of language and logic, and political theory. Karyn Lai is Professor of Philosophy at the University of New South Wales, Sydney. She specializes in comparative Chinese-western philosophical research, drawing vii

viii

Contributors

insights from Chinese philosophies to engage in debates in areas including moral philosophy, environmental ethics, reasoning and argumentation, and epistemology. Her books include Introduction to Chinese Philosophy (2nd ed., 2017) and Skill and Mastery: Philosophical Stories from the Zhuangzi (edited with Wai Wai Chiu, 2019). Karyn is section editor of Philosophy Compass (Chinese comparative philosophy), section co-editor of the Stanford Encyclopedia of Philosophy (Chinese philosophy), and Associate Editor of the Australasian Journal of Philosophy. David Machek is a Senior Postdoctoral Research Fellow at the Institute of Philosophy, University of Berne. He has been working on the history of ancient Greek and early Chinese philosophy, both from the comparative and noncomparative perspective, focusing on issues in moral psychology and ethics. His work appeared in journals such as Philosophy East and West, Journal of the History of Philosophy, or Archiv für die Geschichte der Philosophie. Lisa Raphals is Professor of Comparative Literature at the University of California Riverside, and Chair of the Program in Classical Studies and Program in Comparative Ancient Civilizations. She is the author of Knowing Words: Wisdom and Cunning in the Classical Traditions of China and Greece (1992), Sharing the Light: Representations of Women and Virtue in Early China (1998), and Divination and Prediction in Early China and Ancient Greece (2013); coeditor of Old Society, New Belief: Religious transformation of China and Rome, ca. 1st–6th Centuries (2017), as well as many shorter studies. Naomi Reshotko is a Professor of Philosophy at the University of Denver. She is the author of Socratic Virtue: Making the Best of the Neither Good nor Bad (2006) and has published widely on Socratic Ethics and Plato’s Metaphysics and Epistemology. Jennifer Rothschild is an Assistant Professor of Philosophy at the University of Florida. Her primary area of research is ethics, both contemporary and historical, with a particular focus in Aristotelian virtue theory. Stefan Röttig is a Research Fellow in the Department of Philosophy at the University of Würzburg. He completed his Ph.D. in July 2020 with a thesis on Seneca’s philosophy of emotion. Currently, he is working on a new edition, translation and commentary of Cicero’s ‘Tusculan Disputations’, Book 5. Rohan Sikri is an assistant professor of philosophy at The University of Georgia in Athens. His research focuses on Chinese and ancient Greek philosophy, in addition to developing interests in Indian and Buddhist traditions. He primarily

Contributors

ix

utilizes comparative methods to read classical texts, engaging with philosophical problems from an intercultural perspective. In recent scholarship, he has written on therapeutic traditions of philosophy in antiquity, on the intersections of medical and philosophical methods in Zhuangzi, and on the philosophy of language in early China. Nicholas D. Smith is the James F. Miller Professor of Humanities in the Departments of Classics and Philosophy at Lewis & Clark College in Portland, Oregon. He has published work in several areas of Classics and Philosophy, including especially on Aristophanes and Plato. Recent books include What the Ancients Offer to Contemporary Epistemology (co-edited with Stephen Hetherington; 2019), Summoning Knowledge in Plato’s Republic (2019), and Socrates on Self-Improvement: Knowledge, Virtue, and Happiness (2021). Tue Emil Öhler Søvsø studied Latin and Greek at the University of Copenhagen and currently holds a research fellowship in ancient philosophy at the Humboldt University and Free University, Berlin. His Ph.D. dissertation is on Stoic ethics in Cicero’s philosophical writings. He is the author of a Danish translation of Cicero’s De officiis (forthcoming) and co-author of a Danish adaptation of Apicius’ De re coquinaria. Winnie Sung’s main research interests are in Xunzi’s thought, pre-Qin Confucian ethics, moral psychology, and self-knowledge. She is Associate Professor at Nanyang Technological University in Singapore. Her recent publications are on Xunzi’s ethical views and early Confucian conceptions of hypocrisy, loyalty, trustworthiness, sympathy and resentment. Siufu Tang is Associate Professor in the School of Chinese at the University of Hong Kong. His research interest mainly focuses on the moral and political philosophy of Confucianism, in particular that of Xunzi, and their modern relevance. He has published papers on Xunzi in journals such as Dao: A Journal of Comparative Philosophy, Frontiers of Philosophy in China, NTU Philosophical Review, Monthly Review of Philosophy and Culture. He is the author of Learning for One’s Own Self in the Xunzi lj㥰ᆀⲴ⛪ᐡѻᆨNJ (2015) and Self-Realization through Confucian Learning: A Contemporary Reconstruction of Xunzi’s Ethics (2016). Robin Weiss is Assistant Professor at the American University in Cairo. She specializes in the practical intellect and its relation to technical expertise in Roman Stoicism. Her recent work focuses in particular on the political philosophy of the Stoics.

x

Introduction Tom Angier and Lisa Raphals

Skills are an important embodiment of human (and arguably also animal) intelligence. Ranging from artistic, musical, linguistic, social, scientific and technical expertise to athletics, economics and warfare, they cover an impressive array of human virtuosity. Skills are also an important focus of ethical thought, yet the recognition of skill as central to Ancient Ethics – while widespread – remains piecemeal. The topic has received slightly more attention in Chinese philosophy, but in limited contexts related to the Zhuangzi. Nonetheless, things are changing. Skill has, notably, become a topic of major philosophical interest over the last few decades. In ethics, epistemology, philosophy of mind, philosophy of action and cognitive science, the notions of skill, expertise and ‘know how’ are attracting unprecedented attention.1 This volume examines skill in Ancient Ethics in the two very different contexts of Greece and Rome and China, as well as in several comparative contexts.

1. Skill in Greek and Roman Ethics Why skill? And why in Ancient Greek and Roman Ethics? The reason is that skill – or expertise, craft, art (technē in Greek, ars in Latin) – was a, if not the, central heuristic among Greek and Roman philosophers for understanding the moral life. This is undoubtedly so for the first systematic philosopher in the Western tradition: namely, Plato. As Terence Irwin highlights in Plato’s Moral Theory,2 analogies with the crafts are particularly prevalent in the early dialogues, and the

1 2

See Bengson and Moffett 2011, Montero 2016 and Fridland and Pavese 2020. See Irwin 1977. Cf. Angier 2010: ch. 1.

1

2

Skill in Ancient Ethics

central topic of Plato’s Republic – viz. justice – is referred to as ‘the technē of justice’. Notably, in the Protagoras, Socrates pursues the hypothesis that virtue can be reduced to a craft of measuring pleasures and pains, thereby anticipating the classical utilitarian hope for a hedonistic moral ‘calculus’. And even in the late dialogue, The Statesman, we find Socrates exploring in great detail the notion of a ‘kingly’ expertise, which would see the entire apparatus of rule in the hands of one, supremely wise individual. When it comes to the greatest systematizer of all, namely, Aristotle, the theme of skill in ethics is somewhat occluded, since he spends a significant (though brief) chapter – viz. Nicomachean Ethics VI.5 – arguing that moral virtue is not a skill, but rather a sui generis disposition to ‘hit’ the mean in passion and action. Nevertheless, this repudiation of the analogy between virtue and skill is misleading, since – as Tom Angier argues in Technē in Aristotle’s Ethics (Angier 2010) – technē remains a key heuristic in Aristotle’s moral metaphysics and overall conceptual scheme. Skill-based tropes are vital to his elaboration not only of the ethical ‘mean’ (mesotēs), but also of the highest virtue (viz. intellectual contemplation) and of ethical habituation, that is, the mode of inculcation of the character virtues. Although the contributors to Part 2 explore different aspects of skill in Aristotle’s ethical theory, they too are convinced that technē can throw much light on the latter’s structure and purport. The Stoics bring skill once more to explicit prominence, since they advocate what they call a technē tou biou, or ars vitae, i.e. an ‘art of life’. This is partly a reflection of their return to Socratic sources, but also a function of their ambitious desire for a maximally capacious, fully rational grasp of life’s contingencies, which will enable the moral agent to evade sheer bad luck. Armed with the art of life, the Stoic ‘sage’ will, it is claimed, be able to anticipate most of life’s vicissitudes, and, where this is impossible, at least have the wherewithal to control his own responses to them. In this way, Stoicism offers a vision of impregnable virtue, since virtue itself has been reduced to a rational expertise, one fortified against the depredations of the world. While some find this autarkic vision chilling, others find it bracing and attractive, and indeed, the theory behind the widely practised ‘cognitive behavioural therapy’ (CBT) was itself shaped by a reading of Stoic philosophy. Although an exhaustive treatment of skill in Ancient Greek and Roman ethics would tackle also the pre-Socratics, Epicureans and Sceptics, there can be no doubt concerning their relative marginality to our theme. Plato, Aristotle and the Stoics lie at its heart, so they form the focus of Parts 1–3. Before turning to a summary of each chapter, it is worth remarking that the theme of skill in ethics

Introduction

3

is not confined to ancient philosophers, but has made a resounding comeback in our own time. Two key markers of this development are Julia Annas’s Intelligent Virtue,3 which builds on ancient materials to argue that virtue is a form of skill, and Matthew Stichter’s The Skill of Virtue,4 which engages the psychological literature extensively to argue the same. We hope that, by filling in the historical background in scholarly and rigorous fashion, this collection of papers will shed significant light on both the merits and demerits of their arguments. Nicholas Smith opens Part 1 on Plato, with his chapter on ‘Socrates: Apprentice at Politics’. Smith emphasizes that modern epistemology’s assumption that knowledge is propositional, purely cognitive and non-gradable (i.e. doesn’t come in degrees) is not shared by the Platonic Socrates. Rather, knowledge for Socrates is craft-like: it is developed by and embodied in practice, and, crucially, does come in degrees. This explains how Socrates can achieve a degree of moral and political knowledge without any teachers, and how, moreover, this partial knowledge subtends a degree of happiness. Another innovation sponsored by the early Socratic dialogues is that the politikē technē, or art of politics, can be practised by those not involved in public affairs: precisely, that is, by politically ‘marginal’ figures like Socrates himself. The true political craft consists in trying to become as wise as possible. If we, following Socrates, also seek this, we too may attain at least a modicum of happiness. Naomi Reshotko entitles her paper ‘The Question is not “Can Virtue be Taught?” but “Can Virtue be Learned?” ’ (Chapter 2). It is natural, Reshotko maintains, to assume that crafts are specialized in their subject-matter, and, moreover, that to be learned they must be taught. But both these assumptions are false when it comes to the Socratic craft, viz. the craft of virtue. This is hinted at in Plato’s Meno by the fact that the slave-boy can learn geometrical truths without actually being taught them. And it is gestured at when Socrates notes that virtuous parents are not guaranteed to pass their virtue on to their offspring. The craft of virtue, Reshotko argues, is akin to linguistic skill: it is ‘taught’ by everyone to everyone, so long as qua learners we are receptive to the skill being (inexplicitly) imparted. One key lesson here, then, is that Socrates is an expert learner, not teacher, of virtue, and that it is only through imitating his expertise that we can make moral progress. Lawrence Bloom’s paper is entitled ‘The Contest Between Philosophy and Rhetoric in Plato’s Gorgias’ (Chapter 3). Ambitious and challenging, his argument

3 4

See Annas 2011. See Stichter 2019.

4

Skill in Ancient Ethics

is that justice – although in Plato’s eyes a genuine technē – is in several respects unique. Unlike other crafts, it is never ethically ‘ambivalent’: that is, it always aims at good (contrast medicine, which can be used to heal or harm). Justice, unlike other technai, is, moreover, always motivating, precisely because its aim is goodness. Justice is also unique in virtue of its architectonic role, i.e. it is never subordinate to another craft, since nothing is higher than the good. And justice has no external product: true, it brings into good order anything that practises justice, but this is, in effect, merely to make the thing practising justice more itself. By contrast, rhetoric is ethically ambivalent and not necessarily motivating, hence unfit to play the architectonic role of justice. Indeed, shorn of justice, rhetoric becomes the servant of disorder and even tyranny. Robin Weiss entitles her paper ‘A Dewian Conception of Skill as Clue to the Analogy between Craft and Virtue in the Platonic Dialogues’ (Chapter 4). Her core idea is that if each craft has a strongly delimited and unique domain, the claim that virtue is a craft must end in aporia (or perplexity). Virtue’s domain is global, rather, and it is shared, therefore, with other crafts. To escape aporia, Weiss suggests that we adopt another model of craft, viz. John Dewey’s. According to Dewey, a skill can operate in multiple domains, and need not have a unique, exclusive end. Indeed, it is more fruitful to conceive of skills as habits, whose competence is adverbial: they are ways of doing things, and hence transferable between different spheres (cf. the modern notion of a ‘transferable skill’). Once we recalibrate our notion of skill in this Dewian fashion, Weiss holds, we can make sense of virtue as a skill: it is simply accomplishing the ends of the other crafts justly, temperately, generously, etc. Part 2 is on Aristotle, and opens with Jennifer Rothschild on ‘Steering against the Bad: An Aristotelian Account of Virtue as Two-Way’ (Chapter 5). Key to understanding her title is Aristotle’s notion that skills are ‘two-way’, bipolar or ambivalent. That is, they enable one to accomplish opposite ends. We have seen this already in the case of medicine, which, although properly directed at health and healing, can be used to induce also illness or physical harm. By contrast, Aristotle argues, virtue is a one-way, unipolar or unambivalent power: that is, it always seeks the good. Rothschild’s intriguing claim is that this official story is, however, only half the story. Internal to Aristotle’s account is the recognition that virtue involves ‘steering against the bad’, or avoiding the vicious, and hence is at least cognitively more ‘two-way’ than the above account allows. This brings virtue closer to craft than Aristotle explicitly concedes, and also makes sense of deliberation as a consideration of better and worse options.

Introduction

5

Paul Carron takes as his topic ‘Virtue Cultivation and the Skill of Emotion Regulation’ (Chapter 6). Central here is the crux concerning whether it is practical wisdom – an intellectual virtue, according to Aristotle – that sets our moral ends, or habituated character virtue. Aristotle appears to endorse the latter view, assigning practical wisdom a purely means-end role (it discovers the means to the ends set by habit). This famously opens him to the charge of Humeanism, namely, that our moral ends are non-rational, because they are determined by mere habit. Carron objects, arguing that, on closer inspection, it is phantasia or moral imagination that – with habituated virtue and our emotions as its vehicle – determines our moral ends. And crucially, since phantasia is both cognitive and proto-rational, it does not collapse into Humean habit. With all this in place, Carron argues that cultivating virtue is essentially a skill of emotion regulation, anticipating modern therapies such as CBT. Joseph Dunne’s paper is entitled ‘Aristotle on Techne: Two Theses in Search of a Synthesis?’ (chapter 7). Dunne argues there are two key aspects to technē for Aristotle: cognitive-theoretical and practico-experiential. The former predominates in his analysis, yet to ill effect, since craft is essentially craftwork, i.e. activity grounded in acquaintance with material particulars. But despite this undue separation of knowing from making, there are resources in Aristotle, Dunne maintains, to challenge his ‘strikingly theoretical’ construal of technē. To give a flavour of this challenge: phronēsis or practical wisdom is developed by ‘going to work’ on particulars, and so, Aristotle should acknowledge, is technē; the technitēs or craftsman does not merely impose antecedent forms on matter, but completes his productive activity in and through matter; and technē is ‘in the hand’, no less than ‘in the mind’, since the Aristotelian psuchē or soul incorporates both the physical and the mental. Part 3 is on the Stoics, and opens with Tue Søvsø on ‘The Craftsman of Impulse: Chrysippus on Expertise and Moral Development’ (Chapter 8). This historically rich and detailed chapter provides an overview of the Greek Stoic Chrysippus’s ethics, especially his conception of the ‘art of life’. This art or skill involves care of the self, to which end the Stoic sage is enjoined to seek what is ‘appropriate’ (kathēkon). This is achieved, essentially, by reason, whose job it is to craft our variegated impulses in line with human nature. This requires, in turn, not only conceptual refinement, but also rational or expert perception of particulars. No wonder, then, that the Stoic ideal of ethical wisdom is ‘breathtakingly demanding’, with very few likely to attain the expertise it embodies. That said, there is, according to Chrysippus, the possibility of an

6

Skill in Ancient Ethics

imperfect realization of such ethical skill: this contrasts with scientific knowledge proper, which is an all-or-nothing affair. Stefan Röttig rounds off the section on Greek and Roman ethics by exploring ‘Being and Becoming Good: Seneca’s Two Moral Conceptions of Ars’ (Chapter 9). Focusing on Seneca, a Roman Stoic, Röttig argues that his writings reveal a more nuanced picture of the ars vitae than we might expect. Instead of one skill, viz. the ‘art of life’, we find two: namely, the fully developed art of the wise, and the art of becoming wise, which is practised by the proficiens or moral learner. Wisdom qua artifex vitae demonstrates a ‘perfected and optimally developed mind’, whose moral knowledge and motivations are flawless. The sage therefore does good deeds habitually, with no need for moral struggle. By contrast, the mere proficiens finds the road to moral judgement and action strenuous. He must apply therapeutic techniques to liberate himself from the undue influence of the passions, as well as hone his (as yet imperfect) ‘premeditative’ knowledge of what fate has in store for him.

2. Skill in Chinese Ethics If technē was the major Graeco-Roman heuristic for understanding the moral life, skill was equally important in Chinese ethics, but for slightly different reasons. In both primary sources and commentarial traditions, the fundamental virtue of wisdom (zhi Ც) was widely understood as a kind of skill, rather than as propositional knowledge: ‘knowing-how’ to be good. The nature of virtue as a kind of skill first receives sustained attention in the explicitly Confucian tradition in the writings of Mencius and Xunzi. The most sustained treatment of skill in epistemology and ethics is in the Zhuangzi. The Zhuangzi addresses the question of skill in several ways. Zhuangzian skill stories illustrate the importance of being attuned to natural and cosmological processes. In this context, skill is an antidote to acculturation – including the acculturation of Confucian values – and to ‘small’ perspectives, centred on power, government, or even the human. It is also linked to the important Zhuangist theme of ‘nurturing life’ (yangsheng 伺⭏). But the skill stories target official life especially, including the aims of laws and government, and even the value of texts and teaching lineages. The skill stories also focus on the lives and practices of ordinary people earning livelihoods by performing skilled crafts. The Zhuangzi seems to advocate detachment from official life, and uses the daos of its skill masters as exemplars of how to find a balance between

Introduction

7

engagement and detachment. The skill stories also speak to virtuosity in the ‘skill’ of life itself. Can the Chinese ‘skill’ literature be compared to Greek technē? Three comparative chapters provide three very differently grounded attempts to do just this. Part 4 turns to two Chinese Confucian perspectives on skill and ethics. In Chapter 10, ‘Cultivating Goodness or Manifesting Goodness: Two Interpretations of the Mencius’, Winnie Sung addresses a problem in Mencius’s view of moral development. If the ‘heart/mind’ (xin ᗳ) has inherent moral predispositions and naturally delights in the virtue of propriety (yi 㗙), why is it so difficult for it to attain a state of being fixated on propriety? Mencius himself seems to suggest that the xin itself might not always orient towards propriety; how can this occur? Sung discusses two different models that address this problem. In the widely accepted ‘Cultivation Model’, human beings are born with good but weak moral predispositions, and the xin can only ‘establish’ (li ・) them when it achieves a state of being ‘unmoved’. On an alternative view, the “Manifestation Model,” more characteristic of Neo-Confucian thinkers than of modern Confucian thought, Mencius considers that the inherent predispositions of the xin are sufficient for a person to be good. They do not need further strengthening, but the key to achieving an ‘unmoved xin’ is the ‘skill’ of putting them to use. Siufu Tang turns to Xunzi in ‘Ritual as a Skill: Ethical Cultivation and the Skill Model in the Xunzi’ (Chapter 11). When talking about the necessity of ritual transformation of the instinctual and vicious nature of human beings, Xunzi suggests that it is a similar process to the steaming and straightening of crooked wood so that it becomes straight, and the honing and grinding of blunt metal so that it becomes sharp. Xunzi employs skill analogies to explain the creation of ritual by the sages, pointing out that it is just like how a potter making vessels from clay and a carpenter making utensils from wood. This chapter investigates Xunzi’s skill model of virtues, particularly in relation to the two questions of ritual learning and ritual creation. The chapter first shows the relevance of Xunzi’s ethics to some key issues of contemporary skill models of virtue. It also addresses Xunzi’s understanding of ethical cultivation from the perspective of skill analogies, in order to explain how a petty person becomes virtuous through ritual learning, and how ritual is created in the first place. The chapter aims to show that a skill model of virtues is an apt description of Xunzi’s ethics, and that such a model sheds new light on aspects of Xunzi’s ethics. Part 5 begins with a comparative treatment of skill in the Mozi and the Zhuangzi. In ‘A Path with No End: Skill and Dao in Mozi and Zhuangzi’ (Chapter

8

Skill in Ancient Ethics

12), Chris Fraser examines two early Chinese ‘masters’ anthologies that make prominent use of craft metaphors to examine the question of how skill relates to dao 䚃, the ethically apt path and its performance. These two texts, he argues, offer profoundly contrasting answers to this question. The Mohists explicitly compare ethical practice to performing a skill. Like a skill, dao has clear objects, which explicit ‘models’ can guide us in pursuing, as the compass and square guide artisans in making round wheels and square corners. Mohist ethical life can be described as the process of ‘mastering’ the skills of reliably distinguishing and undertaking what is benevolent and righteous. Skill is also a central dimension of one prominent ethical vision presented in the Zhuangzi, where skills can exemplify how to act in accord with dao, but dao itself goes beyond skill. Two chapters address aspects of skill in the Zhuangzi. In ‘Skilfulness and Uselessness in the Zhuangzi’ (Chapter 13), Wai Wai Chiu turns to an apparent contradiction in its depiction of skilled action. For all the intricate stories of skill masters in the Zhuangzi that seem to put skill at the centre of philosophical reflection, and seem to suggest that being skilful in some sense reveals that one is in accordance with dao, the Zhuangzi also emphasizes uselessness (wu yong ❑⭘). It seems to advocate seeing life as it is instead of judging it by human convention and technique. Chiu attempts to reconcile these two themes. He begins by differentiating uselessness from idleness, then points out that uselessness in the Zhuangzi can be understood in three ways. First, that one’s skill cultivation need not be subjected to material success or social hierarchies, and in this sense one is not ‘used’. Second, in order not to be used by others, one needs to develop abilities for responding to others without being swayed or manipulated. In this sense, being ‘useless’ requires skill. Third, uselessness can be the foundation rather than the opposite of usefulness, hence under certain conditions skills can reflect, rather than contradict, uselessness. In ‘Dao and Agency: What do the Zhuangzi’s Skill Stories Tell Us about Life?’ (Chapter 14), Karyn Lai examines how the Zhuangzi showcases stories in which skilful and satisfying achievements are possible in a range of ordinary activities such as making wheels, ferrying a boat and butchering. How, she asks, can we use these stories to enrich our reflections on living well? The plurality of models offered in the Zhuangzi stand in contrast to the singular and elevated paradigms of the Confucian junzi (exemplary person) or the Nietzschean Übermensch. These latter two accounts of authoritative life each offer a single picture, embodied in one set of ideals, of the life that should be lived. Which Zhuangzi stories, she asks, can show us how best to live? We cannot all be expert bell-stand

Introduction

9

makers, swimmers, scribes, cicada catchers and more. Moreover, we are told in some of these stories that these levels of skill take years to cultivate. How then, should we read these stories? She speculates that we might take them as a coherent set of stories, in order to most fruitfully learn from them. Part 6 presents three comparative approaches to Chinese and Greek materials. In ‘Can One Become Wise by Learning to Catch Cicadas? Analogies between Crafts and Wisdom in Daoism and Stoicism’ (Chapter 15), David Machek argues that for both Stoic and, primarily Zhuangist Daoist thinkers, human excellence is an art, skill or craft that enables one to navigate infallibly various situations of one’s life. For both, this excellence structures and vivifies a sage’s mind in ways that match the structuring and vigour of the universe. But there are important differences as to how and whether this skill can be learned, as well as what kind of epistemic state it is. These differences show most clearly in the relationship between the art of living and ordinary arts and crafts such as butchering, engraving, etc. While the Daoist texts espouse the view that acquiring a virtuosity in these ordinary crafts is a or even the way to acquire virtuosity in the art of living, the Stoics believed that these two learning processes are separate. This difference enables us to identify two different accounts of the art of living, as well as to access deeper underlying assumptions characteristic of each philosophical tradition. In ‘Gendered Skill: Chinese and Greek Skill-Knowledge Analogies from Archery and Weaving’ (Chapter 16), Lisa Raphals compares ‘skill-knowledge analogies’ based on two strongly gendered skills: archery and weaving. Skill analogies based on both occur repeatedly in both Chinese and Greek accounts of skill and ethics. She examines metaphors and narratives that liken archery and weaving to various aspects of ethics, wisdom and government, with particular interest in how or whether the account of the skill reflects the experience of the gender of its typical expert. Finally, in ‘The Skilful Wanderer on the Risks and Rewards of travel in Plato and Zhuangzi’ (Chapter 17), Rohan Sikri considers the ‘skills’ inherent in a special kind of travel. To be skilled at roaming far and wide, with no final destination, was a topic of interest to philosophers in Ancient Greece and early China. We can identify several important trends in their approaches to this concept. Some Greek philosophers drew on cultural assumptions that disparaged ‘wanderers’, and focused on the harmful effects of a nomadic lifestyle. They saw wanderers as skilled shape-shifters, deceivers, and wielders of mere opinion, for example, Plato’s criticism of the sophists as ‘wandering’. Some early Chinese philosophical texts are equally critical, but others describe ‘sages’ as adept at the

10

Skill in Ancient Ethics

skill of roaming (you 䙺), the Zhuangzi especially. Sikri enquires into these two diverging accounts of the skill of roaming and contextualizes each within a larger set of assumptions regarding philosophers and the exemplary skills of movement that they should embody. He argues that between accounts that praise unfettered movement and those that uphold rest (or circumscribed movement), we are ultimately presented with a set of dialectical skills that diverge in significant ways.

References Angier, T. (2010), Technē in Aristotle’s Ethics: Crafting the Moral Life, London: Continuum Publishing. Annas, J. (2011), Intelligent Virtue, Oxford: Oxford University Press. Bengson, J., and M. Moffett (2011), Knowing How: Essays on Knowledge, Mind, and Action, Oxford: Oxford University Press. Fridland, E., and C. Pavese (2020), The Routledge Handbook of Skill and Expertise, London: Routledge. Irwin, T. (1977), Plato’s Moral Theory, Oxford: Clarendon Press. Montero, B.G. (2016), Thought in Action: Expertise and the Conscious Mind, Oxford: Oxford University Press. Stichter, M. (2019), The Skill of Virtue, Oxford: Oxford University Press.

Part One

Skill in Plato’s Ethics

11

12

1

Socrates: Apprentice at Politics Nicholas D. Smith

Introduction At Gorgias 521d6–e1, Socrates makes a startling claim for someone famous for always professing his own ignorance.1 I believe that I’m one of a few Athenians – so as not to say I’m the only one, but the only one among our contemporaries – to take up the true political craft and practise the true politics. This is because the speeches I make on each occasion do not aim at gratification but at what’s best. They don’t aim at what’s most pleasant.2

In this paper, I give an interpretation of Socrates’ claim here – an interpretation in which he means precisely what he says. In other words, though I do accept that the claim appears within a significantly charged dialectical context between Socrates and Callicles, I do not think Socrates’ claim needs to be understood as conditioned in any way by that context. Rather, in my interpretation, there is nothing that Socrates says in the claim that he does not mean literally, or that must be understood in terms of what is happening between the two discussants. In the interpretation I provide, Socrates’ claim is a significant one that tells us a great deal about his own assessment of his ethical achievement. In what follows, then, I will go step-by-step through Socrates’ claim and compare what we find to other things he says elsewhere.

1

2

Examples may be found at Apology 20c1–3, 21d2–7, 23b2–4; Charmides 165b4–c2, 166c7–d6; Euthyphro 5a7–c5, 15c12, 15e5–16a4; Laches 186b8–c5, 186d8–e3, 200e2–5; Lysis 212a4–7, 223b4–8; Hippias Major 286c8–e2, 304d4–e5; Gorgias 509a4–6; Republic I.337e4–5. Unless otherwise noted, all translations are those given in Cooper 1997.

13

14

Skill in Ancient Ethics

1. Is Socrates not the first? Let us begin by reviewing very carefully each of the specific elements of Socrates’ claim. He begins by saying that (1) Only a few Athenians in the past have ever taken up the true political craft or practised the true politics.

Given the ways in which the ‘true political craft’ has been described earlier in the dialogue (for example, that it consists in two parts, legislation and justice – see 464c2–3), one might wonder whether Socrates really does believe that anyone has ever tried to practise this craft, and certainly when Socrates makes this claim, he has not anywhere earlier in the dialogue held anyone up as deserving such recognition. Rather, he has shown only a very critical attitude towards those who have been mentioned so far in the discussion. Many scholars have claimed that Socrates was an ideological enemy of the Athenian democracy, aligned with the radical oligarchic faction in Athens. In some accounts, this was the real reason – or at least one important reason – why Socrates was put on trial, convicted and put to death.3 In fact, however, Socrates seems no less critical of famous members of the oligarchic faction than he is when he evaluates the political performances of those among the democrats.4 It is true that he has hardly anything good to say about Pericles, for example – perhaps the most famous of the democrats from Athenian history.5 Other famous democrats are also subjected to similar criticisms, often in the same passages.6 But Socrates turns out to be completely even-handed in his scorn for famous politicians from the past. Again, in many of the same passages in which he shows contempt for the political skills of democratic politicians, he also includes famous members of the oligarchy as exemplars of the same sorts of failings.7

3

4 5

6

7

See, especially, Stone 1988. Although Stone’s characterization is perhaps the most vehement, this ‘political’ understanding of the trial shows up in most accounts of the motives for the trial and execution. The only dissenters from this understanding, to my knowledge, are Vlastos 1991, Brickhouse and Smith 1994 and McPherran 1996, who argue that the trial had a religious motivation. In what follows, I repeat a discussion given in more detail in Brickhouse and Smith 1994: section 5.3. See, for example, Gorgias 472a5–b3 (false witness); Meno 92d7–94e2 (bad father to his sons); Gorgias 514c4–517a6 (made Athenians wilder and less controlled). Nicias and Aristocrates: Gorgias 472a5–b3; Themistocles, Aristeides, and Thucydides (son of Melesias – not the historian): Meno 92d7–94e2; Themistocles: Gorgias 515c4–517a6. So for example, Cimon and his father Miltiades: Gorgias 515c4–517a6; members of The Thirty: Apology 32c3–d4.

Socrates: Apprentice at Politics

15

Given Socrates’ mission of refutation in Athens, it is not surprising that those with whom he is associated in Plato’s dialogues are revealed to lack the wisdom they are reputed to have (see Apology 21c3–e1). But in at least a couple of places, we actually find Socrates offering praise of past politicians. Just a few Stephanus pages after the passage at 521d6–e1 that is my focus here, Socrates singles out one politician from Athenian history who stands out for having been just in spite of having had ample opportunity to be unjust: Aristeides, son of Lysimachus (Gorgias 526a3–5) – who is also known as ‘Aristeides the Just’. We learn elsewhere (at Meno 92d7–94e2) that whatever virtue Aristeides had managed to achieve was not passed on to his son, who is elsewhere characterized as undistinguished in spite of his father’s much more noteworthy achievements (see Laches 179c2– d2). Accordingly, we can safely conclude that even if Aristeides was Socrates’ precursor (the only one I have found in Plato’s texts) in ‘taking up’ the true craft of politics, we may reasonably doubt that he mastered that craft – since one of the things Socrates says about true craftsmen is that they are able to teach their craft to others (Gorgias 514a5–b3; Laches 185b1–4, e4–6; Protagoras 319b5–c8), by explaining to others the nature of that craft’s objects and their causes (Gorgias 465a2–6, 500e4–501b1; Laches 189e3–190b1). Socrates notes that Cimon and Themistocles made the Athenians wilder and less controlled, and cites the fact that both were ostracized as evidence (516d6–8). Aristeides is not included in this list (which Socrates took from Callicles – see 503c1–3), but could have been: Aristeides, too, was ostracized in 482 bce , though he was recalled within two years to command the Athenian victory at Salamis in 480 (see Nails 2002: 48). From the Socratic point of view, then, while distinguished, Aristeides’ political and paternal achievements would have failed as exemplars of fully realized virtue. Even so, the praise Socrates eventually does give to Aristeides, at the close of the Gorgias, seems at least to moderate his earlier claim that ‘we don’t know any man who has proved to be good at politics in this city’ (516e9–517a2). Perhaps Socrates would have been ready to name others who had at least ‘taken up the true political craft’ in Athens, but what he has to say about Aristeides may be enough to show that he really did think that there were at least some who had done so in earlier generations.

2. Only Socrates Whatever may have happened in former generations, Socrates seems convinced that

16

Skill in Ancient Ethics (2) Socrates is the only one of the current generation who has ‘taken up the true political craft’.

Socrates and Callicles seem prepared to agree that among the current generation, there are no true political craftsmen to be found. Socrates takes this one step further, moreover, in claiming to be the only one among his cohort even to have taken up the craft. But what can Socrates mean by this part of the claim? For some help in discerning Socrates’ meaning, we might be inclined to look to Irwin’s commentary in the Clarendon Plato Series: Socrates does not say that he has this craft, but that he ‘undertakes’ (or ‘attempts’, epicheirein) it, looking for its principles; and so this remark need not conflict with his previous disavowal of knowledge. Socrates is never clear about how this ideal of a moral and political craft is to be realised; for he offers no clear account of happiness showing how it requires justice; and without such a clear account of its goal, in ‘undisputed’ terms (cf. 451d), the political craft cannot begin. Irwin 1979: 240–41

One might respond to Irwin with a certain puzzlement: how can one ‘undertake’ or ‘attempt’ something, but do so in a way that is so deficient that, as a matter of principle, the undertaking ‘cannot begin’? (Imagine trying to explain to your partner that you really did ‘attempt’ to buy bread from the shop, even though there is still no bread in the house, and you haven’t made a single step in the direction of the shop!) Perhaps lurking behind Irwin’s puzzling analysis is an assumption often made by those who have written about Socratic virtue intellectualism. First, for those who may not be familiar with the term, by ‘Socratic virtue intellectualism’ I mean the view generally attributed to Socrates8 that holds virtue to be a kind of knowledge. In contemporary epistemology, knowledge is generally treated as a cognitive state, with propositional (or at least propositionalizable) content, and functions as a ‘threshold’ term – which is to say that an agent either knows or does not know. If one satisfies the conditions for knowledge (however these are conceived of in a given theory), then one knows; unless and until one satisfies those conditions, one does not know. In other words, in most contemporary treatments of knowledge, the achievement of knowledge is not gradable, that is,

8

Even in ancient times – see, for examples, Xenophon Memorabilia III.9.4–5, IV.6.11; Aristotle Nicomachean Ethics Γ.8.1116b3–5, Ζ.13.1144b14–21, 28–30; Eudemian Ethics A.5.1216b2–10, Γ.1.1229a12, 14–16, Γ.1.1230a7–10; Magna Moralia A.1.1182a15–23, A.1.1183b8–11, A.20.1190b27–29, A.34.1198a10–13.

Socrates: Apprentice at Politics

17

it does not come in degrees.9 I suspect that because this is what might now be called ‘Epistemology 101’, scholars considering Socratic virtue intellectualism apply our familiar conception of knowledge to the sort of knowledge that Socrates has in mind when he talks about virtue. But this is a serious mistake. Scholars have also had a great deal to say about what is typically called ‘the craft analogy’, according to which Socrates compares virtue to a technical skill, expertise or craft.10 According to the craft analogy, virtue is like one of the crafts. Elsewhere, I have argued that better attention to the craft analogy can resolve the kinds of puzzles scholars have encountered over what has come to be known as the ‘unity of virtue’ doctrine – Socrates’ claim in the Protagoras (349b6–c1) that virtue (and, indeed, all of the virtues) turn out to be just one thing. Were this the only text on the topic, there would not have been the volume of scholarship that has now been produced on it.11 The problem that has energized so much scholarship comes when we find other texts in which Socrates makes clear that he also believes that the various virtues are (proper) parts of the whole of virtue (so see Laches 190a1–199e7, Meno 78d7–e2) and another text where Socrates makes one of the virtues (piety) a proper part of another (justice – see Euthyphro 12d2).12 I have claimed the craft analogy actually helps to resolve the problem, on the ground that the very same knowledge that grounds a craft (the example I initially used was the knowledge of triangulation, as it applies to both land surveying and also navigation)13 might be precisely the same in each instance, but different in its applications (so, no one would say that land surveying just is navigation or the other way around, even though the same knowledge of triangulation is used for both). In brief, then, I argued that the unity of the virtues should be understood in terms of the knowledge used commonly by 9

10

11

12

13

The view I am representing here has come to be known as ‘invariantism’ among contemporary epistemologists, and has been opposed by certain theorists called ‘contextualists’. (For discussion and explication of the varieties of views on this topic, see Evans and Smith 2012, ch. 3.) To my knowledge, all applications of contemporary epistemological assumptions to the Socratic view have been by invariantists; no one has attempted to characterize Socrates as a contextualist about knowledge. Though I will go on to insist that the sort of knowledge Socrates has in mind is gradable, I do not at all intend to characterize him as a contextualist about knowledge. Rather, I contend that the sort of knowledge Socrates has in mind has all along been regarded by epistemologists as gradable, and is, hence, not the sort of concept epistemologists generally work on in the theory of knowledge. A contemporary account of knowledge that is closer to the one I am attributing to Socrates may be found in Hetherington 2011. Important recent studies include Reeve 1989: 37–45, Woodruff 1990, Brickhouse and Smith 1994: 37–8, and A. Smith 1998. Senn 2005 dissents. For citations and review of at least many of the options that have been offered, see Brickhouse and Smith 1997 and 2010. Note that in the Gorgias, Socrates says that there is a single craft of care for the body that has two parts: gymnastics and medicine, which correspond to the two parts of the political craft, legislation and justice (Gorgias 464b–c). So it should be no surprise that the general craft of virtue (or one of the virtue-crafts) should turn out to have parts, as Socrates indicates. So see Brickhouse and Smith 1997.

18

Skill in Ancient Ethics

all of them (moral knowledge, or, if you like, knowledge of good and bad), but applied differently in each (what Brickhouse and I called the ‘ergon condition’ that would complete the definition of a given virtue). But what never came up in this earlier debate about the unity of virtue is another aspect of the craft analogy, which I think will help us figure out not only what Socrates has in mind in saying that he has taken up the true political craft, but also what has gone wrong in Irwin’s puzzling response to that claim. Briefly, unlike the kind of knowledge that has mostly preoccupied modern epistemology, craft-knowledge is not a threshold achievement; instead, craft-knowledge is gradable – it comes in degrees. Even contemporary epistemologists count ‘knowhow’, which is a much better approximation of the sort of knowledge I claim Socrates has in mind, as gradable.14 So, for example, I know how to dance the cha-cha; but anyone seriously trained in ballroom dancing would find my level of skill execrable – some real expert watching me ‘doing the cha-cha’ might even assess what was happening as ‘not the cha-cha’ or even ‘not dancing’. In their eyes, I would surely not ‘make the grade’. Even so, I think it is strictly correct to say that I do (perhaps only barely) know how to dance the cha-cha. In my most recent work on this topic,15 I claimed that understanding Socratic virtue intellectualism as involving know-how that was gradable would allow us to avoid a very uncharitable assessment of Socrates’ own success in life. Briefly, scholars have noted that various passages in the early dialogues (perhaps the best example appears at Euthydemus 281a1–b6) commit Socrates to the necessity of virtue for happiness. But this view seems to have a very grim effect on our assessment of just how happy Socrates thinks he is, and also just how happy he thinks any of us is likely to be. The problem, in brief, lies in the famous Socratic disavowal of knowledge (and wisdom). According to Socratic virtue intellectualism, virtue consists in some kind of knowledge, but Socrates consistently disavows having such knowledge. But then, if virtue is necessary for happiness, Socrates cannot be happy. Moreover, since, according to the oracle to Chaerephon (Apology 21a2–6), no one is wiser than Socrates, no one is any happier than Socrates. To quote from the scholar who made the most recent (and very powerful) defence of this unhappy view, Socrates’ position ‘puts a ceiling on how happy we can take Socrates to take himself to be: that ceiling is just short of minimally happy’ (Jones 2013: 83). In my recent response to this view, I argued that we do not have to suppose that Socrates would take such a dim view of the human condition, or, for

14 15

So see Ryle 1949: 59. N. Smith 2016.

Socrates: Apprentice at Politics

19

that matter, of his own achievement of happiness, partly because the gradability of virtue as craft-knowledge might allow even someone who has not yet achieved mastery of virtue to have achieved at least some success in virtue, and thus some measure of the happiness for which virtue is a necessary condition. But recognizing virtue as gradable is obviously not, by itself, enough to reach this conclusion, since it is still possible that Socrates assumes that a very high degree of skill in virtue – higher than he has yet achieved – would be required to be even minimally happy.16 Whatever we say about this specific issue, I want now to show why the craft analogy helps explain what Socrates has in mind when he claims to have taken up the true political craft. My claim is that we should understand Socrates as having committed himself to becoming as adept at this craft as he can. As Irwin proposed, when Socrates says that he has taken up the true craft of politics, he is not claiming to have mastered the craft, and, given his continued disavowal of knowledge, we may also suppose that he doesn’t take himself to be even close to such a goal. But even so, for Socrates to claim to have taken up the craft allows for a number of insights about what he thinks he has achieved. First, Socrates’ apprenticeship, as I call it in my title, is one that is without benefit of a master-craftsman in virtue. As a result, what Socrates is attempting to do is bootstrap the craft of virtue on his own. Plainly, that is not the best way to gain a craft, but it is the only way that is open to Socrates: there are no masters of virtue for Socrates to study under – if there were, the oracle that claimed no one is wiser than Socrates (Apology 21a2–6) would turn out to be false. But we know Socrates believes one can learn a craft even without a master to teach it to one (see Laches 185e7–9, 186b1–5, 187a1), and so we can assume that this project is what Socrates has ‘taken up’. But secondly, and much more importantly, Socrates regards himself as in a position to ascertain17 that the true craft of politics requires one to speak in ways 16

17

See Jones 2016, who grants my claim about the gradability of virtue, but continues to resist my claim that Socrates regarded himself as happy. In my paper, I also claim that some passages make clear that Socrates regarded himself as happy (even if not optimally so), and as believing that he had the power to make others happy, if they would only follow his advice (see Apology 36d9–10 and Gorgias 527c4– 6). Jones contends that these passages do not show that Socrates had the more optimistic view of the human condition for which I argue. I purposefully use a somewhat ambiguous term to describe Socrates’ cognitive condition here. Some may object that Socrates can’t know this (see, for example, Benson 2013, who claims that without definitional knowledge of virtue, Socrates can’t know anything about virtue). I have argued elsewhere that Socrates can, and does, know at least some things of moral significance (Brickhouse and Smith 1994, ch. 2), and would note here that what Socrates ‘ascertains’ is something to which he has devoted his life. So whatever we may wish to call it, Socrates’ cognitive condition certainly satisfies a norm of assertion, and even the highest norms of performance – he is plainly willing to ‘put his money where his mouth is’. (On how we should understand the latter issues, in relation to Socrates’ disavowals of knowledge, see McPartland 2013.)

20

Skill in Ancient Ethics

that do not aim at gratification, but at what’s best. Indeed, this consideration – this mandate – of the true craft of politics is so important to that craft, that he can explain his having taken up the craft simply on the ground that he has taken it upon himself to act in accordance with this specific mandate. At least as he explains himself to Callicles, it is in virtue of his always acting in accordance with this mandate that he counts himself as having taken up the craft. It is worth pausing to consider the strength of Socrates’ claim. Does he suppose that, when he speaks, he always succeeds in achieving what’s best? If we focus, again, on his disavowals of knowledge, we might well assume that he cannot mean to claim so much success. On the other hand, we also find Socrates at least intimating that he regards himself as having been remarkably successful in achieving the aim of the political craft. At Apology 37a6–7, he claims he has never wronged anyone willingly – which might be supposed an insignificant boast, given the famous Socratic view that no one ever goes wrong willingly – except that he goes on, only a few lines later, to claim that ‘I am convinced that I wrong no one’ (Apology 37b2–3), without the qualification. He makes or implies the same claim several times in the Gorgias: at 511b4–5, it is at least implied; at 521b6, where Socrates says that anyone who prosecuted him would be a wicked man prosecuting a good one; at 521d2, where he describes himself as ‘one who is not a wrongdoer’; and at 522b9–c1, where he says that everything he says and does is in the interest of right conduct. But he goes on to characterize himself as someone very much to be admired, even though he acknowledges that he would not do well trying to win a trial before a jury of Athenians: Callicles Do you think, Socrates, that a man in such a position in his city, a man who’s unable to protect himself, is to be admired? Socrates Yes, Callicles, as long as he has that one thing that you’ve often agreed he should have: as long as he has protected himself against having spoken or done anything unjust relating to either men or gods. For this is the selfprotection that you and I have often have agreed avails the most. Now if someone were to refute me and prove that I am unable to provide this protection for myself or for anyone else, I would feel shame at being refuted, whether this happened in the presence of many or of a few, or just between the two of us; and if I were to be put to death for lack of this ability, I really would be upset. Gorgias 522b4–d7

It is difficult to see how Socrates could make such claims unless he thought he had significant success in living up to the main mandate of the true craft of politics.

Socrates: Apprentice at Politics

21

It may be, of course, that Socrates has managed to do as well as he has because he has had a great deal of help along the way. At Apology 40a4–7, he tells his jurors that his daimonion has frequently opposed him when he was about to do something wrong, and so the fact he experienced no such opposition when making his defence speech counts as powerful evidence (mega tekmērion – 40c2) that he has done well in court. We may make two inferences here. First, Socrates has not been so successful in achieving the aim of the craft of politics, because his judgement of what he should and shouldn’t do has become almost flawless – and were that the case, his daimonion needn’t have been so active over the years. Accordingly, the daimonion’s high levels of activity underscore how far Socrates has yet to go in his mastery of the true craft of politics. But, secondly, even if the daimonion cannot be relied upon, absolutely, to keep Socrates from error, its high levels of activity – which extend to fairly trivial affairs (see smikrois at 40a7) – show it to be highly reliable. Again, we may assume that between Socrates’ own resources and those provided by the daimonion, we have enough to explain how he can be convinced he has wronged no one. On the ‘inside’ of Socrates’ apprenticeship in virtue, we may confidently suppose he regarded his achievements as quite modest, which is why he continues to make his disavowals of knowledge, and also why his daimonion has had to be so active. But at the level of practice (the ‘outside’, if you will, of Socrates’ apprenticeship), we have reason to think he has done remarkably well. In his attempts to follow the main mandate of the true craft of politics, Socrates is unaware of any specific failures. It may be that he would insist he could do so much better if he could significantly improve his inner condition (and thus rely less on a much less active daimonion). After all, although the daimonion’s opposition is a clear enough signal that he is about to do something wrong, Socrates cannot rely on it to show him the best thing to do in any given circumstance. So he may both reasonably suppose that he has not wronged anyone, but also suspect that he has often fallen far short of ‘what’s best’, which is how he characterizes the ultimate aim of the political craft. This conclusion thus supports Socrates’ claim to have taken up the political craft, without further indicating that he believes he is anything like a master of it. But even if we thus limit Socrates’ actual achievements, we should nonetheless firmly reject Irwin’s assessment that Socrates has not even begun in craftsmanship.

22

Skill in Ancient Ethics

3. Being a craftsman and performing the functions of the craft In the last section, I made a distinction between the ‘inside’ and the ‘outside’ of Socrates’ taking up the craft of politics. Another way to put this distinction is to think of it in terms of having a particular craft-skill, as opposed to engaging in the practices and producing the products of that craft. These two – the skill and the practices and achievements with which it is associated – are very closely connected, on Socrates’ view. [Socrates speaking] In working and using wood there is surely nothing else that brings about the right use except the knowledge of carpentry, is there? [Clinias speaking] Certainly not. And, again, I suppose that in making utensils, it is knowledge that produces the right method. He agreed. And also, I said, with regard to using the goods we mentioned first – wealth and health and beauty – was it knowledge that ruled and directed our conduct in relation to the right use of all such things as these, or some other thing? It was knowledge. The knowledge seems to provide men not only with success but also with welldoing, in every case of possession or action. He agreed. Then in heaven’s name, I said, is there any advantage in other possessions without good sense or wisdom? Euthydemus 281a1–b618

But Socrates’ daimonion seems to open up another alternative here, or at least requires a nuance. The covariance of knowledge/wisdom with the right practices can still be maintained if we (reasonably) understand the daimonion as a manifestation of knowledge/wisdom that Socrates does not have himself – inside, that is. Instead, when Socrates acts on the basis of a daimonic alarm, he receives divine guidance which allows him to avoid the errors he would otherwise have made. Socrates’ discussion with Cleinias in the Euthydemus is presumably restricted to the control individuals can exert over their actions from the inside. 18

Translation slightly modified. The Greek term translated here as ‘success’ is often translated as ‘good luck’ or ‘good fortune’, but the examples Socrates actually uses to demonstrate what he has in mind are all clear cases of what we would characterize as success in some endeavour. This translation is one of those recognized as a possibility in the LSJ (s.v. εὐτυχία). For further discussion, see Brickhouse and Smith 2010: 168–172.

Socrates: Apprentice at Politics

23

Even so, we might wonder which of the two aspects carries the most weight for Socrates – inside or outside. Indications of an answer are not difficult to find. The discussion in the Euthydemus begins with the assertion, taken by all parties as obvious, that everyone wishes to do well (eu prattein; 278e3). During the course of the argument, ‘doing well’ is treated as interchangeable with ‘being happy’ (see eudaimonein kai eu prattein at 280b6). The argument then turns to the question of how to do well, and considers that it is perhaps through having good things (279a2–3). After considering what such good things might be, Socrates gets Cleinias to see that, in fact, the only good thing is wisdom, since wisdom is what makes people succeed (280a6–8). But Socrates then goes on to say that it is not the possession of good things that matters, since there would be no advantage in simply possessing good things without using them (280b8–d7). So ‘doing well’ or ‘being happy’ come, according to Socrates, not just from the possession of something of value, but from doing what that valuable thing allows the agent to do. This emphasis on activity allows us to recalibrate upwards our assessment of how well Socrates takes himself to have done in his apprenticeship in virtue. He remains steadfastly modest about what he has on the inside with respect to virtue, but has positive things to say about what he manages on the outside – that is, with how well he has actually done in the activities associated with the operation of the craft. But since the latter – the outer activities, rather than the inner condition – are what bring real advantage, we are now in a position to regard Socrates as doing rather well, as opposed to someone who has, as Irwin put it, not even begun. This is not to say that he could not wish – or even long – for better, both on the inside and the outside. But neither should we suppose that the best Socrates could say for himself, by way of ‘doing well’ or ‘being happy’, is that he remains, as Jones has put it, ‘short of minimally happy’ (Jones 2013: 83) – unless, of course, we think that someone who manages never to wrong anyone continues to fall short of ‘doing minimally well’. At this point, however, we should remind ourselves that Socrates also believes that people who are deficient on the inside should avoid trying to accomplish anything on the outside: Would a man with no sense profit more if he possessed and did much or if he possessed and did little? Look at it this way: if he did less, would he not make fewer mistakes; and if he made fewer mistakes, would he not do less badly, and if he did less badly, would he not be less miserable? Euthydemus 281b6–c3

We might be tempted to take this passage, together with Socrates’ own epistemic modesty, to show that he regards himself as doing well only insofar as

24

Skill in Ancient Ethics

he does as little as possible, and has achievements that extricate him from misery. This conclusion would be an error, however, and the gradability of virtue as a craft allows us, once again, to understand that Socrates is talking only about a theoretical possibility. Learning crafts takes practice, and his actual advice is the same as that given to Cleinias: ‘every man should prepare himself by every means to become as wise as possible’ (Euthydemus 282a5–6; see also Apology 39d8). We should all seek to practise virtue, so that we become more adept at it. The kind of case Socrates seems to have in mind, where someone would be better off not even attempting to practise virtue, does not apply to any of Socrates’ interlocutors – not even Callicles, who is about as far from virtue as any of them,19 is told that he would do better not even to try to act virtuously.

4. How Socrates performs the craft of politics How does Socrates think he performs as an apprentice political craftsman? Recall that he has already described the political craft as having two distinct parts, which he terms legislation and justice. The legislative part corresponds to gymnastics, a sub-craft of body-care, inasmuch as both these sub-crafts aim to create the best ‘health’ of their objects, soul and body, respectively. Both bodycare and soul-care also have sub-crafts that aim at correction – medicine, for the body, and justice, for the soul. Note that even if Socrates manages to engage in only one of the two parts of soul-care, he will, to that extent, qualify as having taken up the true political craft. But I think we can see, already, that Socrates has taken up both the sub-crafts of politics. On the legislation side, we immediately confront a problem: in the Apology, Socrates proclaims that he has actually avoided engaging in politics in the way we would ordinarily expect from someone seeking to improve his political skill (Apology 23b7–c1, 31d4–32a3; see also 32e–33a4, quoted below). But there is, even so, a way to understand his avoidance of normal kinds of political engagement that is consistent with his making a serious effort to work hard as an apprentice at the political craft. After all, he understands his daimonion to have opposed any attempt to engage in normal political activity, on the ground, as he puts it, that ‘a man who really fights for justice must lead a private, not a public life, if he is to survive for even a short time’ (Apology 32a1–3). It follows that

19

Another who seems at great risk is Thrasymachus. But I have argued elsewhere that Socrates’ interactions with him also invite optimism – see Hoesly and Smith 2013.

Socrates: Apprentice at Politics

25

Socrates believes that normal political activity actually provides a very poor environment in which to practise the true craft of politics, which, in aiming at what is best, would certainly require a serious apprentice to ‘fight for justice’. So, Socrates’ abstention from normal political activity, at least in his understanding of it, is not at all in conflict with his apprenticeship in the true political craft. Legislation is that part of the political craft in which one seeks to exhort, guide and support others in the pursuit of what is best. Although he does not do this sort of thing in the customary political venues, Socrates does characterize his own activities in a way that seems to indicate that he does make regular efforts to promote what is best. In the Apology, for example, he claims to his jurors that ‘I was always concerned with you, approaching each one of you like a father or an elder brother to persuade you to care for virtue’ (31b3–5). And even if it is true that Socrates has avoided most political activity of the ordinary sort, he did serve when his name was selected by lot among other members of his tribe to sit on the Boule, and did ‘fight for justice’ when people were demanding a mass trial of the Arginusae generals, which Socrates opposed as illegal. As he recalls, he was the only one on the presiding committee who held out against the illegal procedure, running a great risk in doing so (Apology 32b1–c2). The fact that Socrates was the only one who stood for justice on that day lends some support to his claim that he is the only one among his contemporaries who has taken up the true political craft. At least on that day, he was alone in doing what the political craft prescribed. He also recalls disobeying the Thirty when they ordered him to bring in Leon of Salamis for execution – an affair he describes as indicative of the Thirty’s habit of trying to ‘implicate as many as possible in their guilt’ (Apology 32c3–8). Although he presumably had no hope of preventing such an offence against justice, Socrates was not implicated in it. In these events, and in a number of other places where he indicates how he has lived his life and what has mattered most to him, we find Socrates aligned with what he takes the true craft of politics to require: Do you think I would have survived all these years if I were engaged in public affairs and, acting as a good man must, came to the aid of justice and considered this the most important thing? Far from it, gentlemen of the jury, nor would any other man. Throughout my life, in any public activity I may have engaged in, I am the same man as I am in private life. I have never come to an agreement with anyone to act unjustly. Apology 32e2–33a4

Socrates’ activities in the part of politics he calls ‘justice’ – where correction is the aim – are easier to see, since they appear on virtually every page of Plato’s

26

Skill in Ancient Ethics

early dialogues. Here is not the place to mount a full interpretation of the Socratic elenchos, but it is worth countering one understanding of the so-called ‘method’ that makes it only a thin, poor example of the true craft of politics. Scholars interested in the elenchos have divided into two camps. ‘Constructivists’ have argued that Socrates does positive philosophical work through elenctic argumentation, thinking he can make progress in figuring out the topics in which he engages his interlocutors. ‘Anti-constructivists’ have argued, on the contrary, that all Socrates ever sought to accomplish with his elenctic arguments was to refute his interlocutors, revealing their pretence to wisdom.20 But the literature on this topic has focused almost entirely on the logical and epistemic features of Socratic argumentation – that is, on which inferences can and cannot be made from its premises or conclusions and on what sorts of evidential weight such arguments can generate or sustain. Certainly, Socrates is not uninterested in such matters, but it is not primarily in logical or epistemic terms that he describes his own activities. He acknowledges, of course, that when he finds someone who thinks he is wise, but is not, he will ‘come to the assistance of the god and show him that he is not wise’ (Apology 23b6–7). We should pause to consider why Socrates thinks the god has any investment in such a practice. Is it just that the god is offended, for example, by a non sequitur? Why, in other words, is the god so concerned with human epistemic fallibility? Socrates reveals his own view of the matter when he characterizes what he does more broadly: For I go around doing nothing but persuading both young and old among you not to care for your body or your wealth in preference to or as strongly as for the best possible state of your soul. Apology 30a7–b2

The god is concerned with human epistemic failure because he understands that the kinds of failure to which Socrates calls our attention tend to make us care for and dedicate ourselves to the wrong sorts of things. This, then, is why Socrates says: I was attached to this city by the god – though it seems a ridiculous thing to say – as upon a great and noble horse which was somewhat sluggish because of its size and needed to be stirred up by a kind of gadfly. It is to fulfil some such function that I believe the god has placed me in the city. I never cease to rouse each and every one of you, to persuade and reproach you all day long and everywhere I find myself in your company. Apology 30e2–31a2

20

For a full review of the recent literature on the ‘Socratic method’, see Wolfsdorf 2013.

Socrates: Apprentice at Politics

27

For Socrates, the most important aspect of what he does – while it may have logical and epistemic aspects – is to ‘persuade and reproach’ his fellow Athenians, and thus to engage in both parts of the true craft of politics. In the Gorgias, too, we find Socrates elucidating this approach. At first, he struggles to get Callicles to acknowledge that when a soul has become corrupted by vice, it must be disciplined, instead of gratified. Callicles then baulks, but Socrates demonstrates that their current conversation is precisely of the kind they have been discussing: Socrates So to be disciplined is better for the soul than lack of discipline, which is what you yourself were thinking just now. Callicles I don’t know what in the world you mean, Socrates. Ask somebody else. Socrates This fellow won’t put up with being benefited and with his undergoing the very thing the discussion is about, with being disciplined (kolazomenos). Gorgias 505b11–c4

The best way to understand Socratic argumentation, accordingly, is to focus on how and why Socrates thinks it has a corrective aspect – how, that is, he thinks he can benefit his interlocutors by speaking with them. Socrates thinks of the benefits of elenchos as being not just logical and epistemic, but also, and perhaps most importantly, ethical, by addressing the ways in which his interlocutors generate and sustain their practical beliefs. But an argument for this hypothesis is best left for another day.21 I hope this is enough to secure my contention that Socrates’ activities are well understood as sincere attempts to engage in the true craft of politics, both as persuasion and as correction – that is, as the legislative and justice applications of the craft.

5. Summary and conclusion In this chapter, I have argued that when Socrates describes himself as having taken up the true craft of politics, we should believe him. I have given several indications that the claims Socrates makes to Callicles are entirely consonant with the things he does and says in Plato’s (other) early dialogues. For Socrates,

21

My proposal here is not entirely novel, but its focus diverges significantly from that found in most of the literature on Socratic argumentation. For anticipations of my view, see, for example, Hoesly and Smith 2013; Moss 2007; Levy 2013. For an extended discussion of the moral psychology underlying the Socratic discipline and correction of the soul, see Brickhouse and Smith 2010.

28

Skill in Ancient Ethics

taking up the true craft of politics is the same thing as attempting to become as wise as possible. It is not clear how wise Socrates thinks mere human beings can become. He devotes himself to that effort, yet seems to view himself as falling far short – at least relative to the god, who he says is ‘really wise’ (Apology 23a5–6). But he is one who has taken up the true political craft, and thinks the rest of us can and should do the same. Even if we never outgrow apprenticeship, there is a great deal we can accomplish, in terms of doing well, if we put in the effort. Socrates has help from his daimonion, but lest those of us who are not so blessed give way to despair, we should remember that ‘the affairs [of a good man] are not neglected by the gods’ (Apology 41d2–3). It is plain that Socrates regards himself as such a man, but by telling the jurors who voted in his favour that they ‘too should be of good hope as regards death’ (41c7–8), he seems ready to include others in that category. In brief, were anyone to join Socrates in taking up the true craft of politics, by seeking to ‘become as wise as possible’, he too could hope to become good, even if never surmounting apprenticeship in the craft of virtue. That may not be as good as it gets; but it is good enough, perhaps, to be minimally happy.

References Benson, H.H. (2013), ‘The Priority of Definition’, in J. Bussanich and N.D. Smith, The Bloomsbury Companion to Socrates, 135–155, London: Bloomsbury. Brickhouse, T.C., and N.D. Smith (1994), Plato’s Socrates, Oxford: Oxford University Press. Brickhouse, T.C., and N.D. Smith (1997), ‘Socrates and the Unity of the Virtues’, The Journal of Ethics 1: 311–323. Brickhouse, T.C., and N.D. Smith (2010), Socratic Moral Psychology, Cambridge: Cambridge University Press. Bussanich, J., and N.D. Smith, eds (2013), The Bloomsbury Companion to Socrates, London: Bloomsbury. Cooper, J., ed. (1997), Plato: Complete Works, Indianapolis: Hackett Publishing. Evans, I., and N.D. Smith (2012), Knowledge, Cambridge: Polity Press. Hetherington, S. (2011), How to Know: A Practicalist Conception of Knowledge, Chichester: Wiley–Blackwell. Hoesly, D., and N.D. Smith (2013), ‘Thrasymachus: Diagnosis and Treatment’, in N. Notomi and L. Brisson (eds), Dialogues on Plato’s Politeia (Republic): Selected Papers from the Ninth Symposium Platonicum, 60–65, Sankt Augustin: Academia Verlag. Jones, R.E. (2013), ‘Felix Socrates?’ Philosophia 43: 77–98.

Socrates: Apprentice at Politics

29

Jones, R.E. (2016), ‘Socrates’ Bleak View of the Human Condition’, Ancient Philosophy 36: 97–105. Levy, D. (2013), ‘Socrates vs. Callicles: Examination & Ridicule in Plato’s Gorgias’, Plato Journal 13: 27–36. McPartland, K. (2013), ‘Socratic Ignorance and Types of Knowledge’, in J. Bussanich and N.D. Smith, The Bloomsbury Companion to Socrates, 94–135, London: Bloomsbury. McPherran, M.I. (1996), The Religion of Socrates, University Park, PA: Penn State Press. Moss, J. (2007), ‘The Doctor and the Pastry Chef: Pleasure and Persuasion in Plato’s Gorgias’, Ancient Philosophy 27: 137–170. Nails, D. (2002), The People of Plato, Indianapolis: Hackett Publishing. Reeve, C.D.C. (1989), Socrates in the Apology, Indianapolis: Hackett Publishing. Ryle, G. (1949), The Concept of Mind, New York: Barnes and Noble. Senn, S. (2005), ‘Virtue as the Sole Intrinsic Good in Plato’s Early Dialogues’, Oxford Studies in Ancient Philosophy 28: 1–21. Smith, A. (1998), ‘Knowledge and Expertise in the Early Platonic Dialogues’, Archiv für Geschichte der Philosophie 80: 129–161. Smith, N.D. (2016), ‘Socrates on the Human Condition’, Ancient Philosophy 36: 81–95. Stone, I.F. (1988), The Trial of Socrates, New York: Little, Brown and Company. Vlastos, G. (1991), Socrates: Ironist and Moral Philosopher, Cambridge: Cambridge University Press. Wolfsdorf, D. (2013), ‘Socratic Philosophizing’, in J. Bussanich and N.D. Smith, The Bloomsbury Companion to Socrates, 34–67, London: Bloomsbury. Woodruff, P. (1990), ‘Plato’s Early Theory of Knowledge’, in S. Everson (ed.), Companions to Ancient Thought I: Epistemology, 60–84, Cambridge: Cambridge University Press.

30

2

The Question is not ‘Can Virtue be Taught?’ but ‘Can Virtue be Learned?’ Naomi Reshotko

Plato often has Socrates claim that, if something is knowledge (episteme) or a skill (techne), then it can be taught. This is the test to which Socrates puts virtue in several dialogues. Two significant cases are in the Meno (86c–98c) and the Protagoras (320c–328d). I will use the Meno discussion and a few pieces of other dialogues where Protagoras’ views concerning language are discussed (Cratylus 385d–389b and Theaetetus 151a–181e) to inform an interpretation of the Protagoras passage. Plato’s portrayal of Protagoras’ assumptions about language in these various dialogues and the claims that he has Socrates make in other places (Gorgias 465a–466e) about what a skill, art or science is, lead me to conclude that Plato is not championing either of the models for teaching virtue that are discussed in the Protagoras passage. Rather, Plato is, in that dialogue, using the tension between the positions of Socrates and Protagoras to make an important point: virtue is a techne and improvement in this area can come from the excellence of virtue’s students as inquirers rather than its teachers’ knowledge of virtue and expertise in the art of its instruction. I argue that this is generalizable to all activities that fall under techne or episteme in Plato’s dialogues. In fact, it could be that the best and only teacher of any techne or episteme is a student who is excellent at inquiry and will conduct an inquiry along with us. Further, it could be that Plato is leading us to conclude that this is what makes Socrates the best teacher of virtue that we have available to us. So it is not because Socrates has knowledge of virtue that he is the best ‘teacher’ of it. It is the fact that he is an excellent inquirer when it comes to virtue that makes him the closest thing to a teacher of it and the best colleague with whom to investigate it. This might be why so many people mistake him for a teacher of virtue. Still, being a good inquirer wouldn’t be of any value to Socrates, and Socrates’ expertise at inquiry wouldn’t be of any value to his colleagues, if virtue itself were 31

32

Skill in Ancient Ethics

not the kind of thing that could be approached via the sort of deliberation and truth-seeking to which things that are sciences and skills lend themselves. It is the fact that virtue is itself a skill that makes having an excellent inquirer as an escort in its pursuit the next best thing to having someone who knows what it is as an escort in its pursuit. In addition to noting Socrates’ suitability for this task, Plato wants us to see that Protagoras cannot be as good an inquirer concerning virtue as Socrates is. It is not that one knows what virtue is and the other doesn’t; both Protagoras and Socrates deny that they have expertise in virtue (for Protagoras’ denial see Pr. 328b). It is because, according to Plato, Protagoras has the wrong preconceptions about what kind of thing virtue is and therefore he cannot be a good inquirer about it. Thus if the best teacher is the best fellow inquirer, Protagoras will not inquire well and will not fulfil this role as well as Socrates does. In discussing the thesis that Plato uses Socrates to champion the identification of virtue as a techne (generally translated ‘skill’ or ‘art’), it is my intention to focus only on the criterion for being a techne that Socrates asserts at Gorgias 465a3–5: the practice of a skill requires a commitment to the thesis that there is some one truth about the way things are (this will be discussed in more detail at the appropriate moment) and is not compatible with the Protagorean notion that things are relative to individuals. Whether Plato or Socrates can also be seen committing themselves to other litmus tests for what makes something a skill (that it requires practice, for instance) are outside the scope of my argument.1 My goal is to show that virtue can be taught by everyone to everyone and still be founded on the metaphysics that Plato, but not Protagoras, appears to think language must be founded on: virtue can be taught by everyone to everyone even in a world where things have independent essences of their own. Thus virtue can be taught by everyone to everyone and still be a techne.

1. The Meno: is virtue teachable? In both the Meno and the Protagoras, Socrates gets himself into a bit of a pickle. He argues that virtue is knowledge. If virtue is knowledge, then it ought to be

1

See Annas (2011) for a thorough discussion of what else might be assumed in the assertion that virtue is a skill. I find that, at least in the ‘Socratic’ dialogues, the evidence that virtue meets all the necessary and sufficient requirements for what it is to be skill on these sorts of accounts is less than robust or conclusive. I thank my audience at the conference on Virtue, Skill, and Practical Reason (Cape Town, South Africa, August 2017) for drawing this to my attention.

‘Can Virtue be Learned?’

33

possible for it to be taught. But Socrates finds lots of evidence that virtue cannot be taught. A central feature of the Meno is a dispute over whether we can discover something we do not already know. It is not clear how this dispute is resolved: Socrates offers recollection as a resolution of the apparent paradox of inquiry, but it is unclear whether recollection provides a fully comprehensible and plausible mechanism for how we come to know that about which we are ignorant.2 Still, Socrates demonstrates that a slave can come to know something about which he was once ignorant (the diagonal), and that at least suffices as a counterexample to claims that we can’t come to know that about which we are (or were) ignorant.3 So Socrates and Meno decide to assume that it is possible to come to know what one doesn’t already know, and they decide to try to find out what virtue is even though they are ignorant about it (86c3–6). Meno is most interested in figuring out whether virtue can be taught (86c7–d2). Socrates and Meno agree that, if something is teachable, it ought to have both teachers and students. All the other technai have both, but virtue appears to have no teachers. After disparaging the Sophists as teachers of virtue, Anytus offers the same suggestion that Socrates will in the Protagoras (319b4–e1): any Athenian gentleman can teach virtue better than the Sophists can (92e3–6). After noting how rarely good men end up having good sons, and after witnessing the confusion that both the Athenians and the poet Theognis have over whether virtue is teachable, Socrates and Meno conclude that virtue has no teachers and, most importantly, use this very same evidence to conclude that it has no students (96a6–c6). They give no direct argument for the conclusion that virtue has no students. It is never mentioned that people like Themistocles, Aristides and Pericles must be examples of such students, and Socrates – with some irony it seems – goes on to say that as he and Meno have been poorly educated by their teachers (which would appear to count as evidence that they are students, but have bad teachers) ‘the knowledge of how good men come to be escapes [them]’ (96e4–5). Socrates and Meno note that true belief can also benefit people (98c8–d2), and they say that this must be how statesmen follow the right course of action for their cities. But leaving that observation in limbo, they conclude that virtue is not taught, but that it comes to be in men through divine dispensation (100b2–c1). To summarize these observations from the Meno: Socrates and Meno assume they can come to know things because the young slave was able to come to know

2

3

Fine (2003: 64), Bostock (1986: 109) and Scott (1995: 56–57) are some examples of scholars who find recollection wanting. Fine makes this point (2003: 56).

34

Skill in Ancient Ethics

the diagonal. Thus even though they are themselves ignorant concerning virtue they pursue an inquiry concerning it. They ask whether virtue can be taught. They hypothesize that virtue is knowledge and therefore can be taught. They find no teachers of virtue and for that reason decide that there are also no students (ignoring what might count as evidence that there are students, including the fact that they regard themselves as students who were taught poorly by their own teachers). They leave the question of whether true belief can be taught or learned unasked and unanswered, and they conclude that virtue is acquired through divine dispensation. Now let us look at how some of the same topics are discussed in the Protagoras.

2. Protagoras (320c–328d) At Protagoras 320c–328d, Plato uses the tensions between Protagoras’ and Socrates’ positions to rethink the way this question was answered in the Meno. Protagoras diagnoses the two positions: Socrates seems to assume – on the model of a skill like flute-playing – that virtue is taught by a few expert teachers to a larger number of students (327a4–c4). Protagoras offers the contrasting model that virtue might be taught in a manner analogous to the way Greek speakers are taught their native language; it could be the kind of thing that can be taught by everyone to everyone (327e3–328a1). I will not conclude that Plato thought that Socrates was right and Protagoras was wrong; rather, I propose that Plato is trying to get us to see that the best question to ask, when it comes to virtue, is whether it can be learned. Do not ask whether virtue has, or can have, any teachers, but whether virtue has, or can have, any students. I will argue that Plato is trying to champion a model for learning virtue that fits the (more convincing) evidence presented by Protagoras and the outward appearance that virtue is taught by everyone to everyone, but that also fits the metaphysical assumptions that Plato would champion, and allows virtue to be an episteme and a techne in the manner discussed by Socrates in dialogues like the Gorgias and Cratylus. In this way, Plato demonstrates indirectly that virtue is a skill, because it can be investigated and learned through inquiry. Plato has Socrates and Protagoras espouse views that allow the reader to construct a view from their critiques of one another. He thereby shows virtue to be learnable in the absence of teachers. The suspicion that this is the model that Plato is championing is further confirmed when we consider the claims, at Meno 80–86, that nothing is ever

‘Can Virtue be Learned?’

35

taught and that all learning is recollection, alongside Socrates’ insistence that if virtue is knowledge, then it must have teachers (which is always followed by his futile search for them).4 It is further confirmed by how ‘fellow inquirer’ rather than ‘teacher’ describes the behaviour of the character, Socrates, throughout Plato’s dialogues.

3. Protagoras in the Protagoras There is a certain amount of disagreement concerning exactly what global view Protagoras is supporting in his eponymous dialogue.5 For my purposes, it matters only that what Protagoras says about virtue and his model for how it is taught can be squared with the metaphysical claims that ‘as things appear to a person, so they are for that person’ (Tht. 151a–b), and ‘things do not have permanent essences of their own’ (Cr. 385e4–386e4). These are the two claims that are attributed to Protagoras in passages in other dialogues where the nature of language is also under discussion. In the Cratylus passage, I will argue that Socrates is proposing that if someone is guaranteed that any name they give will be equally correct, then they live in a world in which things do not have permanent essences of their own (‘the things differ as the names differ’ [385e4–6]). In the Theaetetus passage, Socrates argues that Protagoras’ ‘man is the measure’ doctrine will make language impossible if it is taken to its logical conclusion (183a10–b5). Thus my only assumption is that the Protagoras passage also depends for its interpretation on Protagoras’ conviction that Greek (or any language) – and, by analogy, virtue – is a constructed artefact, rather than a

4 5

This declaration and further search is repeated in the Protagoras (319a–320c). Kerferd (1953) argues that Protagoras consistently holds that virtue is both universally shared and a product of education. He also argues that these views are consistent with Protagoras’ relativism in the Theaetetus. Maguire (1977) has Plato artificially attributing to Protagoras the view that ‘custom law’ is the climax of technical learning (118), and also finds this view consistent with the view (artificially) attributed to Protagoras in the Theaetetus (122). Vlastos (1988) thinks Protagoras is portrayed as assuming moral subjectivism, but does not engage Socrates in a discussion of ontological relativism, since it would be too involved, and would hold no interest for Socrates (xvii). Rowe (1983) says that Protagoras’ position appears largely Socratic, but is unsocratic in that it implies that the citizens themselves and the laws by which they choose to live are the ultimate authorities on moral issues (418). Taylor (1991) thinks that Protagoras’ view that Athenians are right to consult everyone equally on matters of public policy is a logical consequence of Protagorean subjectivism (83, 101). Prior (2002) seems to be the exception in claiming that the account of justice defended in the great speech is not relativistic (319). Perhaps Prior is being sensitive to the internal incoherence that we see in Protagoras’ position throughout the dialogues: you can’t claim both that each person is an expert on reality and that you are the best interpreter of that supposed fact (see also Taylor 102–103 for a similar observation).

36

Skill in Ancient Ethics

natural phenomenon, and that Protagoras therefore thinks that both language and virtue do not rely on independent truths or essences. In contrast, Plato appears to share at least some of Socrates’ commitments with respect to virtue: the criterion that measures whether a word is useful or an action or a person is virtuous is – at least in part – independent of the personal opinions of any individual (or any collection of individuals), and further, that part of knowing virtue is having an appreciation for how the actions of any particular agent correspond to that criterion.6 The question of whether virtue is knowable, or is a science, or is learnable, is fundamentally the question of whether there is such an external and independent criterion. The question of whether virtue is teachable or – more to the point – has any teachers, is the question of whether there is anyone who knows this criterion and can convey it. These are two independent theses, and Plato is trying to get us to focus on the first by showing Socrates’ short-sightedness in focusing on the second. Protagoras assumes that there are many teachers of virtue (with varying degrees of expertise), teaching many pupils (who achieve varying degrees of competence); this is modelled by the teaching and learning of one’s first language. Socrates is assuming one expert teacher of virtue with many diligent and, ultimately, somewhat competent students (modelled by the teaching and learning of the flute). Plato is comparing these assumptions in order to allow us, as readers, to separate the question of whether virtue is knowable from the question of whether anyone knows it and teaches it to others. An affirmative answer to the question of whether virtue is knowable is consistent with the evidence that virtue is taught by everyone to everyone (for better or for worse) – the fact that it is knowable does not entail that anyone actually knows it or that anyone can teach it. But the second question – ‘Does anyone actually know virtue and is anyone, therefore, qualified to teach it?’ – requires a negative answer in light of this same evidence. Plato wants us to put together virtue’s being a kind of knowledge and skill with its being learnable through rigorous inquiry, not with its already being known and having an expert teacher. Socrates worries that the

6

The first third of the Theaetetus is, importantly, an argument against Protagoras. At 178b4–7, we see Protagoras’ view summarized as the notion that the criterion for ‘white, heavy and light and all of that kind of thing without exception’ is within us (rather than independent of us), ‘so that when [each person] thinks that things are as he experiences them, he thinks what is true and what really is for him’ (translation follows Levett). It appears that Maguire also thinks that the distinction sought by Plato in the Protagoras is that between a ‘conventional morality’ that has no objective standard, and that which has an absolute standard (122). Vlastos also posits that the ‘power of appearance’ doctrine (356de) is an indirect reference to the appearance-is-reality doctrine in the Cratylus and Theaetetus (xviii). Vlastos (1988) interprets Protagoras’ great speech as implying that, in a moral disagreement, the more influential brings the opinions of the less influential in line with his own (xix).

‘Can Virtue be Learned?’

37

evidence shows that virtue is neither teachable nor has teachers. But Plato is leading us to the conclusion that, while virtue is teachable (it is knowledge), it has no teachers (no one has this knowledge). While this seems like an unfortunate situation, it is not as bad as it appears. Although there is no one to teach virtue, the fact that it is knowable means that it is still, to an important degree, learnable, even in the absence of an expert teacher.7

4. Teaching a language vs teaching flute-playing In the Protagoras, Socrates expresses concerns about the teachability of virtue that are similar to those he expresses in the Meno: when Athenians meet together in the assembly, they will listen only to experts talk about each of the technai, such as shipbuilding and carpentry. However, when the question is about how to run a city-state, they are willing to listen to anyone – carpenter, tinker, rich and poor – even though none of them have learned and been taught the art of politics. So the Athenians must all agree that such a thing cannot be taught (319b4–e1). Also, the best and wisest citizens fail to impart their political wisdom to others, and particularly to their own offspring (319e1–320b2). Protagoras first replies with a myth. The upshot of the myth is that cities cannot exist if only a few of their citizens share in justice, so justice is distributed to all, and anyone who has no share in justice should be put to death (322c5–d2). Therefore all Athenians regard each person as being equally able to testify about justice. In addition, it justifies the claim that it makes sense for a person to have no compunction about confessing their inability to play the flute, while it would be imprudent for them to be honest about an inability to be just. All men agree that those who have no justice in any degree ought not to be in human society (323b5–c1). Protagoras also shows that there is considerable evidence that Athenians believe that virtue is taught. When a person acts unjustly, Athenians consider it his own fault, and they rebuke or even punish him. This would not happen if 7

Gregory Vlastos emphasized the respect Plato had for Protagoras as a philosopher. Vlastos argues that some of the good ideas in the Protagoras come out of Protagoras’ mouth, and that Plato likely attributes them to Protagoras because he believed that they originally came from him. Vlastos (1991) focuses on the distinction drawn by Protagoras between punishment and retaliation, and on Protagoras’ rejection of retaliation, which is ‘two and a half millennia ahead of his time’ (189). We should add this one to Vlastos’ list: Protagoras is credited for noting that there is a model for the learning of a techne even in the case where there are no experts to be its teachers. See Taylor (1991: 78–79) for a summary of the evidence that the argument Plato attributes to Protagoras in the Great Speech originated in Protagoras’ own writings.

38

Skill in Ancient Ethics

they considered injustice a simple lack of natural endowment – as they do with ugliness or feebleness (323d1–324a2). Also, Athenians use punishment not for retribution but for instruction, expecting it to prevent the wrongdoer from wrongdoing in the future. Thus they think that they teach virtue through punishment (324b5–c4).8 Protagoras leaves the myth behind and turns to answer Socrates’ objection regarding the failure of good citizens to teach good citizenship to their sons (324d2–6). He repeats the claim that human virtue (324d6–e2) is the quality that all city-dwellers must have, and he says it would be incredible if good men taught their sons things other than virtue (325b1–3). Protagoras then reviews the list of all who teach children virtue from infancy: mother, nurse, father, tutor, teachers of reading and music (via exposure to the poets and to harmony), masters of gymnastics, and the laws of the state (325c4–326e4). It is during his explanation for why the sons of good fathers often turn out bad that Protagoras diagnoses Socrates’ assumptions about the teaching and learning of virtue. Socrates has assumed that if virtue is teachable, then – just as with other technai (like flute-playing) – there will be few expert teachers and many inexpert students. Further, the teachers and students will be non-identical. Protagoras points out that if we use learning a first language, rather than learning how to play the flute, as our model, we come up with an alternative set of suppositions. First, everyone will be an expert (a teacher) and a student at the same time.9 Second, listening to anyone will be instructive. Third, since everyone has equal access to equally good teachers, people’s consequent abilities will depend not upon the expertise of their teachers, but upon their own innate capacities (327a–328b). The model that Protagoras proposes is a model for something that is teachable, and about which everyone is, more or less equally, an expert. However, our earlier considerations from the Meno illuminate something that Plato has placed in the background as food for thought: Protagoras’ example of language-learning also serves as a model for something that is teachable but concerning which no one comes close to being an expert. Let us lay out the distinct models with which each character is operating, together with their respective evidence and assumptions. Socrates has only one 8

9

This does not indicate that Plato or Socrates (or even Protagoras) agree that virtue can be taught through punishment. It just indicates that the Athenians think virtue can be taught through punishment, so the evidence is that the Athenians think it is teachable. See my response (2012: 351 n. 44) to Brickhouse and Smith (2010: 103–104). That is, everyone will be an example from which others can learn, and everyone will learn from the example of others. I make no commitment here about whether we learn from ourselves (qua ourselves as opposed to qua practitioners that we observe in addition to others).

‘Can Virtue be Learned?’

39

model for what can be taught, that of how one is taught to play the flute. His evidence for it and the assumptions about the natures of sciences and skills that lie behind it can be summarized as follows: (PFS) Playing flute – Socrates Virtue (like flute playing) is taught by one expert teacher to many students, who will come to play better or worse than each other owing to their native talents and other factors (Prt. 327a4–c4). Socrates’ evidence for PFS: SE1 If it is not known there will be no teachers (Mn. 96b6–8, Prt.319a–320c). SE2 If there are no teachers there will be no students (Mn. 96a6–c6). Socrates’ assumptions about the nature of sciences and skills: SA1 Skills require an account of the nature of things and can state the cause of each thing (Gg. 465a3–5). SA2 Skills assume that things have a permanent essence of their own (Cr. 385e–386e4).

I will discuss shortly the evidence that this is what Socrates (and Plato) assume about the natures with which skills and sciences are concerned. Protagoras does not reject Socrates’ model, but he lays out a further model that is also plausible and that Socrates does not appear to consider. It is modelled on how a person who lives in a linguistic community learns their native language. Protagoras’ evidence for this model and the assumptions about the natures of science and skill that lie behind it can be summarized as follows: (NLP) Native language – Protagoras Virtue (like one’s native language) is taught by everyone to everyone. People all get the same amount of instruction, but their abilities vary with their native talents (327e3–328a1). Protagoras’ evidence for NLP: PE1 We hold people responsible for their degree of expertise in virtue (323e7– 324d1). PE2 Virtue is not natural or spontaneous (323c5–7). PE3 Virtue and language are artefacts that are necessary only if we live communally (327a7–b4). PE4 Because virtue is necessary for communal living, we must profess to care about it if we want to live in the community (323b5–c2). PE5 Even the least able communal Greek speaker (practitioner of virtue) knows more about virtue than someone who is language-less (norm-less), and could thus teach him/her something (328a5–b1).

40

Skill in Ancient Ethics

Socrates and Protagoras come up with a different appreciation of how to model teaching a science or a skill, largely because they disagree about the natures of those things with which sciences and skills are concerned. NLP does not conform to Protagoras’ assumptions about the nature of sciences and skills: PA1 There are no independent natures of things, and things do not have permanent essences of their own (applied to language at Cr. 385e4–386a4 vs 385e4–386a4). PA2 [PA1] is vindicated by ‘as things appear to a person, so they are for that person’ (Tht. 153c1–3).

I will discuss shortly the evidence that this is what Protagoras assumes about the natures of the things with which sciences and skills are concerned. I think there is reason to suppose that, where it conflicts with Socrates’ evidence for PFS, Plato finds Protagoras’ evidence for NLP to be more plausible than Socrates’, and that Plato thinks his readers will find this as well. Not that Plato agrees with Protagoras. Plato’s sympathies lie with Socrates’ assumptions about the nature of science and skill. I will argue that Plato is trying to get us to construct a third model that cherry-picks from these two, using the evidence from NLP and assumptions concerning skill from PFS: Plato suppresses a third model of the acquisition of a native language that he is hoping his reader will see and construct. Here is a summary of the evidence for it, and the assumptions about the natures of sciences and skills that lie behind it: (NLS) Native language suppressed – Plato Virtue, like one’s native language, is taught by everyone to everyone. People all get the same amount of instruction, but their abilities vary with their native talents. Evidence for NLS: PE1 We hold people responsible for their degree of expertise in virtue. PE2 Virtue is not natural or spontaneous. PE3 Virtue and language are artefacts that are necessary only if we live communally. PE4 Because virtue is necessary for communal living, we must profess to care about it if we want to live in the community. PE5 Even the least able communal Greek speaker (practitioner of virtue) knows more about virtue than someone who is language-less (norm-less), and could thus teach him/her something. Assumptions about the nature of sciences and skills:

‘Can Virtue be Learned?’

41

SA1 Skills require an account of the nature of things and can state the cause of each thing (Gg. 465a3–5). SA2 Skills assume that things have a permanent essence of their own (Cr. 385e–386e4).

Importantly, NLS conforms to the evidence that Protagoras gave for NLP (PE1– PE5), but it makes the same assumptions about the natures of sciences and skills that Socrates does in PFS (SA1, SA2).

5. A defence of Protagoras’ evidence over Socrates’ Socrates has only two unique pieces of evidence for his view, SE1 and SE2. SE1 [If x is not known it will have no teachers] seems like a safe assumption, and conforms to some evidence discussed in the Meno and Protagoras. Not only can anyone teach virtue better than those who profess to be its teachers (Mn. 92e3– 6), but when Socrates and his companions look around, it is hard for them to identify any teachers. To make matters worse, the most virtuous people seem not to have passed virtue on to their children – something they most certainly would have chosen to do if they could. In addition, Protagoras and Socrates seem to agree that the Athenians are willing to listen to anyone when it comes to virtue, and this is very different from the way they regard lectures on other skills, where they let only a select few, who have demonstrated expertise, speak before the assembly (Prt. 319b4–e1). But we have already noted some ambivalence, within the text of the Meno itself, regarding SE2 [If x has no teachers it will have no students]. It looks as if Socrates harbours the thought that there are students of virtue and that even he and Meno might be students of virtue, albeit ones with bad teachers. Furthermore, even the demonstration with the slave invites us to think about de novo learning: how did Pythagoras come to discover (learn about) the diagonal? How did the first flautist come to play the flute? Furthermore, the moral of the demonstration with the slave is that there is no such thing as teaching, only causing people to recollect (Mn. 80d–86b). In the Phaedo (74b7–c5) Socrates asserts that mere, inanimate, sticks and stones can cause people to recollect; teachers are superfluous or useless. Even Plato seems to hint at the fact that SE1 does not seem to justify the leap to SE2. Protagoras’ evidence, by contrast, appears plausible and conforms to the beliefs of the common person in his day and in ours. We do hold people responsible for their lack of virtue (PE1) in a way that we do not hold them

42

Skill in Ancient Ethics

responsible for some of their other personal features, like their eyesight or beauty. Thus we must think that if they bothered to take pains, they could be more virtuous than they are. Therefore we regard virtue as something that everyone needs to try to learn. This would not be the case if virtue were natural or spontaneous, so it seems easy to sympathize with PE2 [virtue is not natural or spontaneous] as well. Our problems with the lack of civility apparent in hermits, or humans who have been raised by animals in the wild, occur only when we encounter them – just as people do not need to know the traffic and parking laws in our city if they never come to it. So virtue could be just as much of an artefact as are our traffic laws. Just as speaking Greek appears to be an artefact of being born into a community, or visiting one that uses Greek to communicate. Thus PE3 [Virtue and language are artefacts that are necessary only for those who live in a community] looks convincing. If someone does choose to live among us, they will need to figure out some way to communicate with us, and if we are all hell-bent on speaking Greek and nothing else, then they will need to learn Greek as well. If they want to drive here, they will be held responsible for driving according to our laws. We do not want people living among us who do not care whether they can communicate with us and we do not want people driving among us who have no interest in following our traffic laws. Thus PE4 [we must profess to care about virtue only if we want to live communally] looks good, too. PE5 [even the least competent member of a community knows more about the language and norms of that community than someone who has never lived in any community] is such a modest claim that it is easy to agree with: children can teach some of their native language to other children, or even to adults who are ignorant of it. Even someone who is vicious might point to out to a child raised by wolves or to a hermit that if you steal things, you end up in jail (if you get caught) and if you grab food from someone’s plate without permission, they might slap you. So it is easy to be sympathetic to Protagoras’ evidence for the claim that virtue is taught and that it is taught by everyone to everyone.

6. The nature of a skill according to Socrates (and Plato) Now let us look at what assumptions Socrates (and Plato) make about skills and sciences. We have to put together evidence from their views in a few different dialogues in order to fill out their support for SA1 [Skills/sciences require things that have external and independent causes and, ideally, their practitioners can

‘Can Virtue be Learned?’

43

state the cause of each thing] and SA2 [Skills/sciences assume that things have a permanent essence of their own]. At Gorgias 465a3, Socrates characterizes rhetoric as a ‘knack’ (ἐμπειρία) rather than a ‘skill’ (τέχνη) and explains the difference: a knack or form of ‘flattery’ (κολᾶκεία) ‘guesses at what is pleasant with no consideration for what is best’ (465a2) and also ‘has no account of the nature of whatever things it applies according to which it applies them, so that it is unable to state the cause of each thing’ (465a3–5). A science or skill (or art), on the other hand, ‘has an account of the nature of things and can state the cause of each thing’ (Gg. 465a3–5). Accordingly, Socrates considers rhetoric, cookery, cosmetology and sophistry each to be a knack or a form of flattery. He deems gymnastics, medicine, legislation and justice to be crafts or sciences. He does this because actual experts in the crafts and sciences become experts by accumulating true information about their particular subject. Horse trainers must possess a great deal of information about what is actually beneficial and harmful to horses. Doctors must know how to benefit and harm the human body. Navigators must know how to read the stars and how to manage a ship in open seas. The extent to which an expert in these areas is successful will be the extent to which her beliefs about horses, bodies, ships, etc., correspond to the way these things actually are. In contrast, in those disciplines that are classified as knacks or forms of flattery, the capacity necessary in order to be an expert is quite different. Practitioners of these disciplines pride themselves on being able to get their desired result while remaining ignorant concerning how things actually are with their subjects. Gorgias and Polus boast that they can convince someone of a truth and then convince that same person of that same truth’s negation. So clearly the actual truth of that concerning which they are trying to persuade their audience is immaterial to their project. Cookery is the art of making food look and taste good without paying attention to whether it actually is good (nutritious). As far as practitioners of knacks and forms of flattery are concerned, there is no need for any connection between their thinking and the way things actually stand in the world.10

10

Of course, this is likely a misconception on the part of those practising knacks and forms of flattery; they do rely on their being some truths in order to get the effect that they are after, even if they don’t recognise this. Understanding certain psychological facts about people will allow them to persuade them more easily (just ask advertisers), and filling food with fat, salt and sugar will get a lot of people to eat it (just ask fast-food chains). What these practitioners really seem committed to is the absence of any fact of the matter about what is good or best for the person upon whom they practise their form of flattery or knack. I think Plato would argue that this is not a coherent position to hold – you can’t be a relativist about the good, but a scientist about how to get people to behave in certain way – this being implied by his final argument against Protagoras at Tht. 179–181 (see my 1994).

44

Skill in Ancient Ethics

This distinction between skills and knacks is underwritten by the kind of metaphysical assumptions that Socrates champions against Protagoras in the Cratylus (385d7–e3). There, Socrates asks his interlocutor, Hermogenes, whether he sees any correctness to names. Hermogenes replies that there is no correctness to names and that ‘anyone can give a name’ equally well. Socrates then asks Hermogenes whether he thinks that the ‘things differ as the names differ’ and are ‘relative to individuals’ as Protagoras says. According to Socrates, those who disagree with Protagoras must think that things have a ‘permanent essence of their own’ (Cratylus, 385e4–386a4). I think we can permit ourselves to look to the Cratylus to fill out the metaphysical assumptions behind Protagoras’ view in the Protagoras (according to Plato), because it not only discusses Protagoras, but also shows what Plato thinks Protagoras must assume about language and about Greek, which is the background that NLP (Protagoras’ model for learning our native language) in the Protagoras would have to assume. When Hermogenes admits to being occasionally tempted by Protagoreanism, Socrates gives a very quick argument against Protagoras: if Protagoras were right that things are as they appear to each person, then no one could be wiser than anyone else as each person would always be correct; on Protagoras’ view, thinking that something is true makes it be true (386c2–d1). So we see here that Socrates attributes to Protagoras a view in which there are no permanent, independent essences, that is, no natures or independent criteria to which knowledge or skill would have to conform in order to be valuable. Although both Hermogenes and Socrates act as if Protagoras has been laid to rest at this point, Socrates goes on to make an argument that is at least as much an argument against Protagoras as it is an argument against Hermogenes. While Hermogenes has supposedly adopted a less radical type of conventionalism than that proposed by Protagoras – one in which things do not differ as their names differ – Socrates’ argument against this weaker view presents as much or more of a problem for Protagoras’ more radical one. Socrates points out that words are the tools by which we speak, just as knives are the tools by which we cut. I cannot simply pick up any old object, decide that it is a knife (call it a ‘knife’) and then cut something with it. I will only succeed in cutting if I use the correct tool – calling something a knife does not make it a knife. I will cut most effectively if I use a tool that was fashioned by someone who understands the nature of cutting – someone who knows what kinds of materials can be used to cleave what other kinds of materials. I need a knife that is designed by someone who looks to what the world (metal, stone, wool, etc.) is

‘Can Virtue be Learned?’

45

like, not someone who just takes whatever seems like it will cut. Likewise, if I wish to inform someone using speech, I will find that words that are well chosen by someone who understands the things I am trying to talk about will be more effective. It will be better, for example, if the word for ‘help’ is relatively short and distinct from other words that denote ‘look at that interesting thing over there’ or ‘leave me alone’. This is not to say that there is only one word that can be used for ‘help’, just as there is not only one material out of which a knife can be made. But it is definitely the case that it is possible for some words to be better than others, and that the more understanding of reality that the legislator of names has gleaned from the dialectician, the better names he will craft (Cratylus, 390c–e). To apply this to a skill like flute-playing: if neither flutes nor music had permanent essences of their own, there would be no basis for the claim that I play or teach the flute worse than the first chair flautist in the Cleveland Orchestra. In the Cratylus, Protagoras seems to think that language does not rely on anything ‘natural’ to make it what it is, and he is, in the Protagoras, assimilating virtue to this same model – it is a human artefact, with rules that are stipulated by a community of practitioners, more or less arbitrarily. Nothing about virtue is natural and requires us to uncover independent truths about reality. This makes both language and virtue susceptible to NLP in Protagoras’ view. It makes them the kinds of things that, like the ‘norms’ of Facebook, we learn from the people who invented it and from the way we see others behave on Facebook. Facebook is an artificial environment invented and constructed by human beings. Everyone learns how to be a good ‘citizen’ on Facebook (if they do learn) from every other Facebook user, either by following their example, being told something explicitly, or not doing what others have suffered unwanted consequences for doing. Further, it all has to fit into the constraints that were built into Facebook by its engineers. Protagoras is treating both virtue in one’s society and one’s native language as this same kind of construction. But I will argue that SA1 and SA2 (Socrates’ metaphysical assumptions) are consistent with a model of language learning that is equally described by what Protagoras promotes in the Protagoras. Thus NLS (which is the model of learning one’s native language that Plato is trying to get us to think about) can presuppose that both virtue and language carry the metaphysical assumptions that underwrite Socrates’ conception of them in Gorgias and the Cratylus. They can both be taught on the model of NLS, even if they are constrained by natures that are discovered by, and not made up by, the people who try to use them, teach them and learn them.

46

Skill in Ancient Ethics

7. Protagoras’ evidence fits Socrates’ assumptions regarding the nature of a skill Let us revisit Protagoras’ evidence yet another time. In doing so, we will see how it can be re-described so that it can also be underwritten by the assumption that virtue is a skill. This is important to the claim that virtue is not constructed, even given Protagoras’ assumptions. As a result, even according to Protagoras, virtue could (theoretically) have a small number of expert practitioners. However, it doesn’t, because it is so hard to learn. We might think that PE1 [We hold people responsible for their degree of expertise in virtue] is plausible on Protagoras’ assumption that virtue is just an artefact, because we can then hold people responsible for getting to know what the social constructs of their society are. Of course, we could also counter Protagoras’ assumption that people can, be held responsible for their own virtue by deploying some of the traditional arguments used against relativism. If virtue has no basis other than the agreement of its local practitioners, what is to stop me from identifying myself as a member of a different society, and making up my own rules, which will be correct no matter what they are, because they don’t have to conform to any objective, natural criteria? Now I am not responsible for learning the norms others have invented, just for inventing and practising my own. The view championed by Socrates in the Cratylus, on the other hand, fits very well with the notion that I should be chastened if I do not make rigorous efforts to figure out what the objectively best (most virtuous) ways to behave are, and attempt to embody them. So Socrates’ assumptions regarding the nature of a skill (SA1 and SA2) fit just as well, or better, with PE1, than Protagoras’ rejection of them does (PA1 and PA2). In all his dialogues, Plato offers us evidence that, while virtue conforms to nature (it must conform to the way the natural world is, and the way human beings are naturally suited for flourishing), the natural and spontaneous motivation that people have to become virtuous can be less than apparent. Socrates has to convince many of his interlocutors that they do, already, care about virtue and, until he uncovers that motivation for them, it is less than obvious, both to them and to their community, that they want to become virtuous. Thus it makes sense to fault people’s own ignorance (concerning their own actual benefit, and their own motivation to seek it) if they do not appear to be motivated to become as virtuous as possible. Socrates claims that it is, or should be, obvious to them that they wish to do well. From there the path to helping them take responsibility for their own virtue is to make it clear to them

‘Can Virtue be Learned?’

47

that the only way to intentionally increase the likelihood that they will do well is by increasing their knowledge (which increases their virtue). In addition, even once they are motivated to be virtuous, no one becomes virtuous naturally or spontaneously, because figuring out how to do so will require a great deal of skilful inquiry. Thus Socrates’ metaphysical assumptions concerning the nature of a skill certainly fit with Protagoras’ claim that people need some prodding and instruction. In Plato’s case, this prodding complements our natural desire for well-being, but still, seeking and attaining virtue is far from spontaneous. By contrast, when Protagoras claims that virtue-seeking is neither natural nor spontaneous, he means that it cannot have these features, because it is entirely artificial. One might think that this is true because I can’t deduce the rules of Greek logically from first principles, or from the state of nature, and since, on Protagoras’ view, any community’s norms are made up, just as their language appears to be, no one can deduce what anyone holds to be virtuous from first principles either. However, for Plato, this is exactly in line with language and virtue being highly dependent on the nature of the world: a useful language and code of conduct will be influenced and constrained by natural features of our environment, ones that must be uncovered empirically. No one can just stipulate that, in their language, people will communicate, despite the fact that all proper nouns are the same word, or that, in their community, people will be able to become good citizens by following the norms, one of which is to honour one’s parents by setting a bowl of water on fire once a week. In contrast, the appearance that PE2 [virtue is not natural or spontaneous] is true can be accounted for by NLS and squared with the assumption that virtue is not made up, but needs to conform to the actual natures of actual things that are independent of how we think about them. The same will go for PE3 [Virtue and language are artefacts that are necessary only in community]; what the untamed human or the hermit do naturally in the wild or in isolation will look nothing like what will be necessary in order for them to live virtuously with others. Of course the hermit or wild child might still do well to inquire into the objective natures of things in order to survive and thrive in the absence of a community, and for Plato, this will count as an inquiry into how to live well – into virtue. Still, virtue will look completely different in a community than it will for an individual who is trying to survive in a state of nature – so PE3 will look true on NLS. Socrates vindicates PE4 [we must profess to care about virtue if we want to live in community] in the Apology and other dialogues; we do not want people among us who do not care about being virtuous. However, Socrates would assert

48

Skill in Ancient Ethics

(and Plato would agree) that deep down, everyone does care about being virtuous (because no one wants to suffer and be miserable [Mn.77e5–78a8] and you can’t do well by harming others [Ap. 25c8–e5]). So unlike Protagoras, Plato need not advocate putting those who are not virtuous to death. Instead, we can instruct them through inquiry (Ap. 26a2–5) and they will come to acknowledge that they care about virtue. As was defended earlier, PE5 [even the least competent member of a community knows more about the language and norms of that community than someone who has never lived in any community] is such a weak and modest claim that it would be hard to disagree with, and it is well suited to the assumption that things have permanent essences of their own.

8. Conclusion Plato has given us the tools to vindicate NLS. He has shown us that virtue is a domain where the best ‘teacher’ and the best student can be the same person.11 No one knows what virtue is. Everyone learns what virtue is from observing and inquiring about everyone else. The best inquirer will be the most virtuous person, and will also be the best escort on the road to virtue. This can be true because, like most skills, inquiry into and development of virtue requires the assumption that the things with which one is interacting have permanent essences of their own. It is because Protagoras does not hold this assumption that he cannot hope to be the best – or even a good – inquirer, and we should not invite him to be our escort. Treating goodness and virtue as if they were completely artificial, and could be developed on a whim, will no more allow us to thrive than assuming that every conceivable name for the word ‘help’ will allow us to stay safe. There are many skills that must be gained through appropriate experiment and discovery, rather than didactically. Plato shows us that virtue is chief among them. When it comes to these kinds of skills, the best teacher is a fellow student who, despite not knowing what virtue is, is an expert at inquiring. This is true for two reasons. First, some of the most important skills and sciences are so difficult to master that they have no experts, and so there are no teachers. if we assume

11

‘Teacher’ here is in inverted commas, as it does not assume the model of an expert teacher (like a flute teacher), but of an excellent fellow inquirer. Also, this need not imply that the best teacher and student teaches themselves (although it could – perhaps in the case of Socrates where he is his own best student because he asks good questions and conducts good inquiries for himself). It could be that he who learns best from others is also the best at teaching others (this could also be said about Socrates as portrayed by Plato).

‘Can Virtue be Learned?’

49

that teachers are the kinds of experts that flute teachers are. There is no one who can convey virtue in all its complexity all by themselves. Observations of the world, and the infinitely various ways that the people in it do and do not behave virtuously, is the best thing on which the most capable student can reflect in order to become more virtuous. Certainly, we do not want to think we should give up learning skills that seem important to our surviving and thriving just because no one else has figured out how to practise them yet. Second, when it comes to a skill, the best way to learn is to be forced to practise it in an environment that contains sufficiently high stakes that one is motivated to learn it, yet where the stakes are not so high that one’s curiosity is muted by fear of the consequences of poor performance. Because there is no one with sufficient expertise to set up an artificial environment for the study of virtue – there is no analogue to, for example, the closed driving course here – virtue must be studied in a genuine, rather than an artificial, environment. Thus the high stakes are mitigated only by the presence of a fellow student, who excels at being a student. In fact, it could be that this is not limited to virtue: the appropriate engagement with every skill or science is dependent on being an expert at learning, and this is why some people teach themselves amazing things (like how to play a musical instrument) even in the absence of an expert teacher.12 Socrates is only an expert student and learner, not an expert teacher. This might allow him to be mistaken for an expert teacher, as he is a better colleague for learning virtue than those who suppose themselves to have expertise, and present themselves as teachers, or those other than Socrates, who are mistaken for teachers even if they don’t profess to be such. One must gain expertise for the study of, and increase in, virtue through expert inquiry – another skill that is learned more than it is taught. It is good that the study of any skill, including one so important as virtue, is more dependent on the existence of good students than of good teachers, since expert teachers of virtue – if there are any – are few and far between. In fact, this might be true for more skills and sciences than we care to admit.13

12

13

Plato indicates in several places that having a master teacher is not necessary for becoming proficient at a skill. At Laches 186b and Gorgias 514b, Socrates argues that anyone who cannot offer evidence that they had an expert teacher can instead offer evidence that they have produced expert outcomes. The evidence offered in the interchange between Protagoras and Socrates in the Protagoras suggests that an expert teacher is neither necessary nor sufficient. Whether or not a person had a master teacher, the proof of their expertise must be in the outcomes they produce. I am grateful to the anonymous reviewer who brought these passages to my attention. In addition to my audience at the conference on Virtue, Skill, and Practical Reason, I am grateful to Tom Angier and Jordan Watson for their helpful feedback as I prepared this chapter for publication.

50

Skill in Ancient Ethics

References Annas, J. (2011), Intelligent Virtue, Oxford: Oxford University Press. Bostock, D. (1986). Plato’s Phaedo, Oxford: Oxford University Press. Brickhouse, T., and N. Smith. (2010), Socratic Moral Theory, Cambridge: Cambridge University Press. Fine, G. (2003), Plato on Knowledge and the Forms, Oxford: Oxford University Press. Kerferd, G. (1953), ‘Protagoras’ Doctrine of Justice in the “Protagoras” of Plato’, Journal of Hellenic Studies 73: 42–45. Maguire, J. (1977), ‘Protagoras of Plato? II. The Protagoras’, Phronesis 22 (2): 103–122. Prior, W. (2002), ‘Protagoras’ Great Speech and Plato’s Defence of Athenian Democracy’, in V. Caston and D. Graham (eds), Presocratic Philosophy: Essays in honour of Alexander Mourelatos, 313–326, Abingdon: Routledge. Reshotko, N. (1994), ‘Heracleitean Flux in Plato’s Theaetetus,’ History of Philosophy Quarterly 11 (2): 139–161. Reshotko, N. (2012), ‘Socratic Eudaimonism’, in J. Bussanich and N. Smith (eds), A Bloomsbury Companion to Socrates, London: Bloomsbury. Rowe, C. (1983), ‘Plato on the Sophists as Teachers of Virtue,’ History of Political Thought, 4 (3): 400–427. Scott, D. (1995), Recollection and Experience, Cambridge: Cambridge University Press. Taylor, C. (1991), Plato Protagoras, Oxford: Oxford University Press. Vlastos, G. (1988), Plato Protagoras, New York: Macmillan. Vlastos, G. (1991), Socrates: Ironist and Moral Philosopher, Ithaca, NY: Cornell University Press.

3

The Contest Between Philosophy and Rhetoric in Plato’s Gorgias Laurence Bloom

Introduction Plato refers to justice as a craft in the Gorgias in an argument that, somewhat dubiously, uses justice’s alleged craft-like nature to conclude that one who knows what is just cannot be unjust. Perhaps the most striking aspect of Plato’s analogy is that it is used to conclude something about justice – that knowledge of it entails acting on that knowledge – that cannot be said of ordinary crafts. That one has knowledge of a craft, shoemaking for example, does not entail that one is incapable of refraining from practising that craft. That is, the shoemaker can quite easily refrain from making shoes. In fact, Plato, though he uses the analogy to crafts, seems keen on highlighting this unique aspect of justice. I will argue that Plato does indeed hold that the one who knows what is just cannot be unjust and that the craft analogy leads us to this conclusion as much by contrast as by similarity. It should come as no surprise that the analogy between knowledge of justice and craft knowledge breaks down. All analogies break down at some point. If they did not, they would be identities not analogies. What is perhaps surprising about the craft analogy in the Gorgias is that Plato uses the point of breakdown as effectively as he does the point of similarity, or so I will suggest. The breaking point of the analogy, in this instance, is used to reveal that justice, unlike crafts, has a certain unambivalent character. This difference between justice and crafts, although it weakens or limits the analogy, actually strengthens the account. It does so by establishing the place of justice, not as a craft among crafts, but as the master or architectonic craft. Who, Socrates the philosopher or Gorgias the rhetorician, practises this master craft – the craft of justice – is being contested in the opening sections of the Gorgias. To fully understand this contest, and 51

52

Skill in Ancient Ethics

philosophy’s eventual victory, both the analogy to crafts and the disanalogy are essential. In short, it is the ambivalent nature of ordinary crafts – that, in their case, knowing does not entail acting on said knowledge – which motivates the need for this most un-craftlike of crafts. Terence Irwin has argued against the strength of the craft analogy and, in large part as a consequence of this, has concluded that Plato’s view in the Gorgias is inconsistent (Irwin 1977). On the other hand, David Roochnik, contra Irwin, argues that Plato intended the analogy as a sort of reductio ad absurdum to stymie those who would claim an understanding of the equation between virtue and knowledge (Roochnik 1996). The two philosophers form bookends: Irwin takes Plato to be committed to the truth of the analogy (at least in the Gorgias) while Roochnik takes him to be using the analogy to highlight the difference between knowledge of virtue and craft knowledge. My position is in between: there is a strong analogy between justice and crafts, but it has its limits; both the analogy and the limits are essential for Plato’s account. The linchpin, it seems to me, is the relation between a craft and its end or product. Irwin takes the knowledge of the end to be included in the craft, while Roochnik denies even that crafts necessarily have external ends. As a consequence of misunderstanding this relation, they both miss the centrality of the architectonic structure in the text. I will argue that the ends or purposes of crafts – of all the crafts of the sort Plato is interested in in the Gorgias – are external to, and unknown to, those crafts. This externality will necessitate the structure and the subordination of crafts to one another and, ultimately, to the ‘craft’ that is for its own sake: justice.

1. Orders and ends To understand his analogy and its limits we need to begin with Socrates’ understanding of crafts. In the Gorgias, a craft is set over or directed at some object. The craft’s role in relation to that object is either to create the object (for example, pottery or wall-building) or repair and perfect the object (bicycle repair, medicine, gymnastics).1 In addition, the text is clear about the nature of

1

Many of Socrates’ examples, such as wall-building and ship-building (Gorg. 455b7–8), are examples of crafts of making. The crafts of repairing and perfecting are distinguished in the Gorgias in the discussion with Polus (464b2–c5). Their differences need not concern us here. The important point is that crafts are understood and distinguished in terms of what their object is and what they do for it. Justice is classed with the repairing crafts.

Philosophy vs Rhetoric in Plato’s Gorgias

53

the object that the craft is set over: each craft is set over some order or organising structure.2 So, for example, the shoemaker takes the materials of leather and cork and gives them a certain order. The order he gives the materials is the shoe – so much so that, if one were to completely take away that order, the shoe would cease to be and we would be left with nothing but the leather and cork. So too the doctor takes the sick or disordered body and repairs it by returning it to its correct order. Indeed, all crafts can be understood in terms of their ability to impart, repair, or perfect the order or structure of some given material.3 Order as the object of craftsmanship is central for Plato’s account. I do not intend to venture into the thorny issue of whether or not order in the Gorgias is an early prototype for the Platonic forms of the middle dialogues. The point here is that, as far as the object of craft is concerned, the order is both what makes it be what it is as well as what makes it good. This is implicit in the example of crafts: for a shoe to be at all is for it to have at least some of the structure that allows it to fulfil the function that makes it a shoe – for it to serve the purpose of protecting the foot. Indeed, this is what it means to be a shoe. Further, for it to be a good shoe is for it to have the order or structure that allows it to do so well. There is both a metaphysical and an ethical component to this view. Neither the metaphysical nor the ethical components are as intuitive to us as they are to Socrates, or, with some unpacking, to his interlocutors. It is, however, quite reasonable to think that what makes the shoe be, even more than its material, is the unity that the latter comes to embody when it is ordered to do that which it does. Without the order, there would be no shoe as such, but only a pile of leather and cork. The ethical component also has intuitive appeal. We can appreciate it by reminding ourselves that the central Greek ethical concept of aretē, translated as either ‘virtue’ or ‘excellence’, has a wider scope than that of ‘virtue’. So much so, in fact, that any would-be analogy between crafts and virtue is, for Plato, less an analogy than a point of identity. The good order or structure of the shoe, to continue with the same example, is not just analogous to virtue, it is the virtue of the shoe. The shoe’s being well ordered makes it a good shoe, as well as a shoe that

2

3

The importance of the ideas of order (κόσμος) and organising (συντάσσω), and their various synonyms, comes out emphatically later in the dialogue (see 503e7–504a5; 505d1–4; 506 d3–e8). The passages are clear and decisive in claiming that crafts instil or perfect order and organisation in their respective objects. Later in the text, Socrates asks whether activities such as singing, flute or lyre playing, composing dithyrambs and tragedy, and the like, are crafts or mere ‘knacks’ (501d8–502b10). They are crafts if they benefit, and knacks if they merely gratify. In context, it is clear that the benefit he has in mind is the ordering of the soul. Thus even in the case of these sorts of activities, their being crafts entails a relation to some order over which they are set, further confirming the relationship between crafts and ordering in the text.

54

Skill in Ancient Ethics

is able to do whatever it is that a shoe does well; indeed, the two mean the same thing. Virtue, aretē, is an expression for a thing’s fully being what it is. That the object of craftsmanship has its own virtue or excellence is relevant to Socrates’ analogy: it cannot be in the above sense that crafts are analogous to virtue, as in this sense they – or their objects – have their own virtue or excellence. Rather, for Socrates, crafts are analogous to justice, and the analogy comes from seeing that justice is human virtue. The analogy at work in the Gorgias is that justice should do for the human soul what the cobbler’s craft does for the shoe. If the analogy is correct, then just as shoe repair orders the shoe, justice, or the craft of instilling justice, orders the human soul. In doing so, it makes the soul exist to a higher degree, and, more to our point, be more excellent. In other words, the craft of instilling in the soul its own proper virtue – justice – is an activity that is, in at least this one sense, analogous to that of any other craft instilling the appropriate virtue or excellence in its proper object. Thus it makes some sense for Socrates to describe justice as doing for the soul what medicine does for the body (464c5).4 If it is the case that human virtue – namely, justice – is analogous to the virtues of other things, it follows that being just for a human being is to exist more fully, as well as to be able to do what humans do well. Of course, to be able to determine whether we are doing whatever it is that we do well will depend on an understanding of what it is that we should be doing – an understanding of our end or purpose. This brings us to the limit of the analogy. Indeed, if the analogy to crafts revolves around the craft’s relationship to the ordering of the product, the disanalogy is closely connected. The disanalogy follows from the relationship between the order of the object and the end that the object is to serve. The order of the shoe, or of any object of craftsmanship, is determined by how best to use the materials of the shoe to accomplish some pre-determined end – protecting one’s feet, for example. Thus the disanalogy: justice, the order of the soul, is not for the sake of any external end. Rather, unlike the other crafts, justice is valuable for its own sake. Justice’s purpose is internal to itself. Thus to understand justice as order in the same sense as the objects of other crafts is not possible. To put this another way, even if justice is the repairing of the order of the soul, because the order of the soul cannot be understood in the same way as the order of the objects of other crafts, the analogy fails. Before we continue, I want to address a possible concern: the reader has likely noticed an equivocation here. On the one hand, justice is a craft that is (at least 4

Unless otherwise specified, all Stephanus references are to the Gorgias.

Philosophy vs Rhetoric in Plato’s Gorgias

55

somewhat) analogous to other crafts in that it is set over an order in some material. On the other hand, it would seem that justice is not analogous to the crafts so much as it is to their object – viz. the order they impart or perfect.5 I suggest that the equivocation functions as an intentional and positive feature of the account. It is the very equivocation – the simultaneous analogousness and disanalogousness – that will be required of the architectonic craft. Every craft, save this highest one, is for the sake of and ruled by the craft that the order it creates belongs to. In other words, just as the tanner is ruled by the cobbler for whom he makes the leather (and this just because the leather will be part of the order of the shoe), each craft is ruled by, and subordinate to, the craft that is its end. The one that is self-ruling then – as the highest, architectonic craft has to be – can be ordered only in relation to itself. In other words, in the case of this one, unique craft, the ordering is for the sake of itself; to be self-ruling, the highest craft must be its own end. That the strange, simultaneously analogous and disanalogous nature of justice, as the ruling craft, expresses itself also as an ambiguity between justice as the craft acting on an object and justice as the quality instilled in the object is, therefore, actually a support for the claim that justice is the architectonic craft. In fact, the central issue of justice – or of anything that could serve as the architectonic craft – is the same, whether we take justice to be analogous to the craft or to its object. This is because the highest craft must be disanalogous to ordinary crafts by being a craft that is its own object. As we will see even more clearly in the next section, this is exactly what the highest craft necessitates. Thus the equivocation is an instructive one: justice is both the craft and the object of the craft because it is the craft that is set over itself. In this way, the account I am offering takes both the analogous and disanalogous elements of the comparison to crafts seriously as intentional – and necessary – features of Plato’s account. In short, justice must be craft-like, because it is an ordering and, as we will see, because it rules over the other crafts in a way that only a craft can. On the other hand, it must be un-craft-like, because it cannot be for the sake of anything outside itself, and, therefore, cannot be ordered to an external end in the way other crafts are. This latter, disanalogous, element 5

Early in his discussion with Polus, Socrates defines justice as a craft that does for the soul what medicine does for the body (464b2–c6). Yet later in that discussion, he describes injustice as the deformed state (πονηρίαν) of the soul that just punishment repairs (see 477a8 ff.) If the bad state of the soul is injustice, the good must be justice. Later still, in his discussion with Callicles, he speaks of the orderly, organized soul as one that is appropriate to gods and men, which means, according to Socrates, that it is pious and just (507a2–b4). Thus the Gorgias seems to use justice both to refer to the craft that instils and the embodiment of order in the soul. This can be seen simply by contrasting the central metaphor – that justice does for the soul what medicine does for the body – with the claim that justice is human virtue.

56

Skill in Ancient Ethics

requires that we find an internal end for justice. (I will try to do so in section 6 of this chapter.) This end will connect to a particular understanding of philosophy as Socratic inquiry. Before getting to this, we will need to understand better the nature of the architectonic craft (section 2); see that Socrates and Gorgias are competing for nothing other than the role of architectonic craftsman – which they both agree belongs to the instiller of justice in souls, and the determiner of the just use of crafts (sections 3 and 4); and take stock of both the analogous and the disanalogous elements (section 5). The concluding section (7) will assess the positive and negative aspects of rhetoric as they have arisen from the account.

2. The architectonic craft In his well-known and influential account, Terence Irwin makes an apparently similar observation about craft-knowledge in the Gorgias. For Irwin, knowledge of a craft includes knowledge of the end the craft is for.6 So, for example, the craft of shoemaking includes knowledge of the end or purpose for which the shoe is made. Irwin’s claim is not without support. Indeed, as we have just noted, the shoemaker makes or orders the shoe for the end the shoe is designed to fulfil. The end, that is, is closely connected to the order. (In fact, since the end dictates the order, perhaps it would be more precise to say the order is connected to the end.) This entails that the shoemaker must have some access to the end the shoe is designed to serve, as without it, he could not order the shoe correctly. Thus far I am in agreement with Irwin. Yet despite this significant similarity, I submit that Plato’s picture is more complicated than Irwin’s reading suggests. The complication is that the shoemaker cannot have the same sort of knowledge of the end of the shoe as he does of the leather and cork, at least not in so far as he is a shoemaker. As a shoemaker, he knows the purpose of the leather and cork essentially – their purpose is to fit together in such a way as to form the shoe, which is to say, their purpose is the craft of shoemaking. Thus his knowledge of the craft of shoemaking includes within it – we might say encompasses and subordinates – the knowledge of the purposes of its materials insofar as they are its materials. That is, even if he is not

6

More to the point, Irwin claims that crafts, unlike virtue, have a determinate end (Irwin 1977: 84). I will argue that Irwin has it backwards: it is the ambivalent relationship of crafts to their ends that requires justice – a ‘craft’ with an internal end. Irwin also claims that justice does not have an internal end for the Socrates of the Gorgias (ibid., 8), and that the elenchos and the craft analogy represent different ethical doctrines (ibid., 94–96). I will disagree with both these claims below.

Philosophy vs Rhetoric in Plato’s Gorgias

57

himself a tanner and couldn’t prepare (i.e. order) the materials himself, he – in virtue of his craft-knowledge – knows the materials’ purpose.7 Indeed, their purpose – the use of the materials – is his craft. However, his knowledge of the end or purpose of the shoe cannot be of the same type. He does not have this latter knowledge any more than the tanner has the knowledge of shoemaking – at least, he does not have it insofar as he is a shoemaker. Rather, the knowledge belongs to the one who uses the shoe for her craft, just as he uses the leather of the tanner. The knowledge belongs to, perhaps, the runner. Of course, the shoemaker could very well also be a user of shoes, more than likely he is. As such, he would have knowledge of the purpose of the shoe insofar as he is a user (not a maker) of shoes. Many of Plato’s typical examples – the lyre-maker and the lyre-player (Cra. 390b3–5), the rudder-maker and the ship’s captain (ibid., 390d1), the bridlemaker and the horse-rider (La. 185d1–3) – are carefully chosen to illustrate just this point. So, for example, the bridle-maker does not so much have knowledge of the end or purpose of the bridle, as he has that purpose dictated to him by the horse-rider, the one who uses the bridle. So too when Socrates asks Glaucon, ‘Then doesn’t the one who knows give instructions about good and bad flutes, and doesn’t the other rely on him in making them?’ His point, as he immediately makes clear, is that, ‘Therefore, a maker – through associating with and having to listen to the one who knows – has right opinion about whether something he makes is fine or bad, but the one who knows is the user’ (Rep. 601e2–602a1, italics added). Indeed, the examples are well chosen to allude to an architectonic structure in which the higher crafts subordinate the lower crafts as their material and, by specifying the purpose of the lower crafts, rule them all but entirely. So much so, and so consistently, that, through his very use of the craft analogy, Plato alludes to this architectonic structure. Seeing the structure operating in the Gorgias will show how Plato is struggling to give a criterion for determining the just use of crafts,8 and with some success. In short, the higher crafts in the structure, by using the lower ones, determine the justness of those lower crafts, as we will see. There is still another consequence to the separation of the knowledge of the end and of the order of crafts which relates to the need for a craft with an internal

7

8

The ‘insofar as’ disclaimer is necessary here because the shoemaker need only know his materials ‘insofar as’ they are the materials of his craft. There are numerous other uses of leather that the shoemaker need not know, nor need he know how to tan or prepare the leather. This latter is the knowledge proper to another craft entirely: that of the tanner. For a different perspective on ‘just use’ in the Gorgias, see Barney 2010. On this, see note 14.

58

Skill in Ancient Ethics

end. It follows from what we might call the ambivalence of crafts. Craft-knowledge is a knowledge of contraries. The doctor, for example, in virtue of the same knowledge that allows her to heal the sick, is able to harm the healthy (Rep. 333e2–8). Thus there is no reason to be certain that the doctor, simply because she has knowledge of how to heal, will inevitably practise her craft as called upon to do so. Rather, insofar as the craft of medicine is concerned, either outcome could very well be a use of the craft. That is why we need the craft to be ruled; we need someone who is able to dictate to the doctor what end to pursue, and when to pursue it, just as the ship’s captain does for the rudder-maker. Ultimately, this is perhaps the main reason why we don’t want simply to be treated by a doctor, we want to be treated by a just doctor. Yet why, then, should knowledge of the ‘craft’ of justice entail that one cannot be unjust? Irwin’s answer is that it does not. Thus even as he sees some type of architectonic structure functioning in the text, he sees Plato’s account in the Gorgias as so lacking that Plato would have changed his mind about it.9 I suggest that the architectonic structure is connected to this fundamental ambivalence of crafts in relation to their ends. In short, because each craft – due to its ambivalence – is ruled by the craft above it in the structure, the whole structure depends on a craft that is unambivalent at the top. This entails that the craft not be separate from its end – that it be for itself – as we have seen Plato takes justice to be. Thus my disagreement with Irwin’s claim that the intrinsic value of justice is absent in the account of the Gorgias is fundamental. My view is that justice’s intrinsic value is central, and that asking after the craftsman responsible for putting it into souls amounts to debating who occupies the top position in the architectonic structure. Indeed, it is clear that this highest craft must be disanalogous in just the way already noticed about the craft of instilling justice in souls: the highest, ruling craft, if it is to be at the top of the structure, cannot be for the sake of anything other than itself. If it were, it would, by necessity, not be the highest. Thus the structure itself necessitates a ‘craft’ or craftlike activity of just the sort instilling justice seems to be as the highest craft. As we will see, this element of the said ‘craft’ will also erase its ambivalence. What would such a craft be set over? In one sense the answer is indicated by what we have already noted: the architectonic craft rules the other crafts by judging them, based on whether or not they are good for the soul. However, we should keep in view the fact that the

9

For the first point, see Irwin 1977: 76–77, for the second, ibid., 131.

Philosophy vs Rhetoric in Plato’s Gorgias

59

nature of this highest craft – even if we label it as the ordering of the soul – is prima facie unintelligible. At least, it cannot be understood in the same way as the orders of the other crafts. The soul, unlike the objects of the other crafts, is not primarily for the sake of anything other than itself. The very need for something to sit atop the structure requires that it not be craft-like in the relevant way. Thus if the craft of instilling justice in the soul – or whatever we are to call this highest craft – is also treated as a craft, we must ask what it is set over. Yet if it is set over some order, that order would need to be organized around another – and higher – end. Thus the problem of justice’s craft-like nature, I suggest, is best understood as the problem of the architectonic or ruling craft: a craft that is unlike other crafts in a fundamental way.10

3. The objects of rhetoric and philosophy The opening discussion of the Gorgias11 begins with Socrates’ asking Gorgias what craft it is that he practises and teaches (447c2). So central to the nature of a craft is the relation to its object, that this opening question is formulated by Socrates in terms of what Gorgias’ craft can accomplish, and then, more emphatically, as ‘[w]ith which of the things there are’ his craft – rhetoric – is concerned (449d2; d9). Indeed, Socrates’ continued attempt to get Gorgias to identify the peculiar object over which his craft is set is the motor that drives their discussion. When Gorgias answers that rhetoric is concerned with speeches (451a8; d3), Socrates presses him for what the proper object of the rhetorician’s speeches is that differentiates them from the speeches of the other craftsmen. Then, as if to emphasize the importance of the craft’s object, he repeats the question yet again in terms of ‘which of the things that are’ concern the craft of the rhetorician (451d4–7). When Gorgias suggests that his are speeches that produce persuasion (453a6), Socrates, continuing to focus on the object, asks after the thing about which Gorgias’ speeches produce persuasion (454a9). It is here, finally, that Socrates gets an answer in the form that he is looking for: the proper object of the persuasive speeches of the rhetorician is justice (454b7); the rhetorician produces speeches which, in turn, produce persuasion about the just

10

11

Nor can we avoid this circularity by making justice the craft of ordering the state. Rather, the same circle returns: the state is ordered for the sake of the citizens’ virtue, yet the just citizen is so by benefiting the state (see Laws 630e2 and especially 962a10–964d8). The discussion, which ends in Socrates’ elenctic refutation of Gorgias’ claim to teach justice, runs from the beginning of the dialogue to 461b3.

60

Skill in Ancient Ethics

and the unjust (435e9–455a2). There are, as always with a Platonic dialogue, numerous intricacies to the discussion, and it is quite typical of Socrates to drive an interlocutor to a more determinate position. Yet neither of these points should blind us to the fact that, in this case, the driving question is that of the object of the alleged craft. Two central claims emerge from Socrates’ attempt to get Gorgias to identify the object of his alleged craft. The first, into which Gorgias is forced by Socrates’ questioning, is the claim to be concerned with justice. The second, which Gorgias proudly agrees to, is the fact that his craft, rhetoric, is something ‘supernatural in scope’ (456a7), including within its domain all the objects of all the other crafts. In Gorgias’ unrestrained words: ‘Oh yes, Socrates, if you only knew all of it, that it encompasses and subordinates to itself (συλλαβοῦσα ὑφ’ αὑτῇ, literally: “gathers together under itself ”) just about everything that can be accomplished’ (456a9– b1, italics added).12 Gorgias’ claim to supernatural subordination is a sign that we are talking about a contender for the architectonic craft, the craft that gathers together the other crafts under itself. That it comes out of the same exchange as the issue of the just and the unjust, and that it is Socrates, not Gorgias, who leads the way into it (as we will see in a moment), suggests more than a coincidence or empty braggadocio on Gorgias’ part. Indeed, it is immediately after formulating the definition of rhetoric as the producer of conviction-persuasion about the just and the unjust that Socrates begins what is a strange and easily overlooked exchange about wall-building. He says: Well now, let’s see what we’re really saying about oratory. For, mind you, even I myself can’t get clear yet about what I’m saying. When the city holds a meeting to appoint doctors or shipbuilders or some other variety of craftsmen, that’s surely not the time when the orator will give advice, is it? For obviously it’s the most accomplished craftsman who should be appointed in each case. Nor will the orator be the one to give advice at a meeting that concerns the building of walls or the equipping of harbours or dockyards, but the master builders will be the ones. And when there is a deliberation about the appointment of generals or an arrangement of troops against the enemy or an occupation of territory, it’s not the orators but the generals who’ll give advice then. What do you say about such cases, Gorgias? 455a9–c1

12

All translated passages from the Gorgias are from Zeyl 1987. Zeyl uses ‘oratory’ and ‘orator’ where I use ‘rhetoric’ and ‘rhetorician’.

Philosophy vs Rhetoric in Plato’s Gorgias

61

Of course, Socrates is baiting Gorgias. The wall-builders don’t decide if Athens should build a wall around the city; their role is not to decide on the reason or end for which the wall should be built. Yet this is exactly what the one who decides whether or not to build the wall will be reflecting upon. The wallbuilder’s role is to put the bricks and mortar together as well as possible, so that the wall can fill the function dictated by the one who knows the end. Who is it that determines that end? Perhaps even more to the point, what is it that he would have to know to determine it? Note that one way of understanding the question, What would he have to know?, is – What is his craft? As Socrates knows, this is exactly the craft that Gorgias takes himself to practise. Gorgias takes the bait: Well, Socrates, I’ll try to reveal to you clearly everything oratory can accomplish. You yourself led the way nicely, for you do know, don’t you, that these dockyards and walls of the Athenians and the equipping of the harbour came about through the advice of Themistocles and in some cases through that of Pericles, but not through that of the craftsmen? 455d6–e4

Themistocles and Pericles, rhetoricians of high quality, are being offered here as paradigms of what Gorgias has just claimed rhetoric capable of. In advising on the building of the city’s walls, they have exactly the knowledge he claims to teach: that of persuading the many to act for the sake of one’s own ends – the knowledge that ‘encompasses and subordinates’ the other forms of knowledge. The resonance of this particular example for ancient Athenians is strong, though, like much else in the dialogue, it is also ambivalent. The building of the walls around Athens, and especially the middle wall (which Socrates heard Pericles advise on himself, see 455e8–9), was both crucial for the defending of the city against the Spartan land invasion, and largely to blame for the devastating plague that killed a huge portion of the city’s population. Thus the question of whether or not the walls should have been built – whether or not it was a good decision – was, one imagines, still very much a live question in Plato’s Athens. The very importance, and ambivalence, of the issue suits Plato’s purposes perfectly. What would someone need to know to decide whether or not she should build a wall? What is the nature of this knowledge? We should be able to say, with certainty, that it is not at all the same as that of how to build a wall; it is not simply the knowledge of the craftsman. The wall-builder might know how to build a wall, but his knowledge is contingent upon – and subordinate to – the prior knowledge of when and why to build a wall.

62

Skill in Ancient Ethics

If we allow ourselves to ask what Themistocles and Pericles had to know, we can make out two possibilities: they need, on the one hand, to be able to bend others to their will. Socrates and Gorgias are not in dispute over who it is that has this knowledge: this is the skill of the rhetorician. On the other hand, they should have the knowledge of when it is or is not good to build a wall. I would suggest that Socrates and Gorgias also agree on what this latter knowledge is a knowledge of, even if not about who, if anyone, has it. Though it might sound awkward to us, the knowledge in question is the knowledge of justice. To know if it is a good idea to build a wall is to know if doing so is just.13 Justice, or the craft of instilling it, is the architectonic craft. It is the craft Gorgias claims he teaches. Whether or not he does so is the subject of the dispute between the two men. There is still another way to see that Plato’s concern is the nature of the architectonic craft: Socrates’ questions for Gorgias about the object of his craft could just as easily be applied to Socrates’ own craft, i.e. philosophy. What is the object over which philosophy is set? What could we answer to such a question? Perhaps we might wish to claim that philosophy is set over truth, or wisdom. Of course, if we were to do so, we would leave ourselves open to the Socratic criticism that all crafts are concerned with truth and wisdom: shoemaking is concerned with the truth and wisdom about shoes, medicine that of health, and wall-building that of walls. What, then, is the truth or wisdom with which philosophy is concerned? To put this point together with the previous one, both rhetoric and philosophy seem to be confronted by the same problem. Both seem to be ‘crafts’, or ‘knowledges’, without an object (or with an object that cannot be understood in the same way as other objects: in terms of its end). This, however, makes them decidedly un-craft-like. Later in the text, we find that rhetoric is not actually a craft at all, although it does have an object (in a way). It is a knack, and its ‘object’, although it does not care for that object in the way a craft would, is the same as that of justice (see 465c4). We should not let that distract us from the issue here, though, which is that rhetoric and philosophy are the two claimants for the architectonic craft. Do either of them have a justification for their claim? This, I submit, is the central question of the Gorgias, and certainly of its first section. How is the question to be answered?

13

This point is clear if we equate justice with the order of the soul. When Pericles advised on the building of the middle wall, he was doing so, on Socrates’ account, as a means to the end of ruling, or of ordering the souls of his citizens. Thus, for Socrates, he should be judged by how well he did this (see 503c1ff.)

Philosophy vs Rhetoric in Plato’s Gorgias

63

4. A Gorgian inconsistency By the end of their discussion, Socrates has shown Gorgias to be contradicting himself. In short, Gorgias claims that rhetoricians – like any other teachers of competitive arts – should not be blamed if their students use those arts unjustly (456c9–457c3).14 At the same time, however, he claims that rhetoricians must have knowledge of justice – a knowledge that is so necessary that if a student comes to learn rhetoric from him, and does not yet have this knowledge, he will teach it to them (460a3–5). I turn now to the argument with which Socrates shows these claims to be mutually inconsistent. The formal validity of the argument is less my concern than what it reveals about the natures of justice and any possible architectonic craft. One of Gorgias’ reasons for his praise of rhetoric is its ability to be persuasive about a subject without knowledge of that subject (459b4–c7). However, Gorgias’ craft does not stand in this same relation of non-knowing to the subject of the just and the unjust as it does to these other subjects. Rather, it seems that Gorgian rhetoricians necessarily possess knowledge of the just and unjust. With this claim in hand, Socrates’ argument proceeds quickly: a man, it is agreed, who has learned a particular craft is the sort of person that his knowledge makes him (460b6).15 It is also agreed, albeit implicitly, that justice is a craft. Thus it is agreed that one who has learned justice is a just man (460b8). This is only the first or intermediate conclusion. We might note in passing here that the claim that justice is a craft or is craft-like, the very analogy in question, is curiously unstated in the text. At any rate, it is striking that the knowledge of justice, the knowledge of when to use a particular craft, is its own, separate craft or craft-like knowledge. It seems that Socrates and Gorgias agree that one can have knowledge of any other craft and not know when it is just or appropriate to use that knowledge. That is, they agree that the knowledge of the craft does not include knowledge of the end the craft is designed around. The argument continues by combining this conclusion – that one who knows justice is just – with the claim that the just man (i.e. the one who is just) does only just things (460c3; 460e1–2; 461a8). If this latter is the case, and Gorgias offers no resistance, it follows that a person who has learned justice does only just things. 14

15

For a defence of Gorgias’ ‘just use’ defence of rhetoric, see Barney 2010. Though the just use defence could very well suffice in many cases, as Barney argues, the problem with using it here is that Gorgias is arguing for the place of rhetoric as the determiner of just use. Thus this is the one case in which such a defence is inappropriate. Although the word techne is not used exclusively in the statement of this premise, it is clear from the examples used (see 460b1–6) that the sorts of subjects in question are crafts.

64

Skill in Ancient Ethics

The ‘only’ is necessary to expose the contradiction in Gorgias’ account. Indeed, Socrates reiterates, and Gorgias agrees, a rhetorician ‘would never have done what’s unjust’ (οὐκ ἄν ποτε ἀδικήσας, 460e1). Thus no claim to ‘just use’ (457b4) could be relevant or necessary in this particular case. It seems, according to the argument in the text, that the unjust use of Gorgian rhetoric is impossible. If we accept this conclusion, we are left, indeed, with a devastating contradiction in Gorgias’ account. He teaches rhetoric and, in doing so, teaches justice. Yet he himself concedes that, sometimes, his students use their power of persuasion unjustly. This should never occur. No rhetorician taught by Gorgias should ever act unjustly. Gorgias has no way to wriggle out of the contradiction, and he makes no attempt to do so. We might note, as well, that Socrates has still not said what the knowledge of justice consists in. All we know is that it necessarily conditions the use of other skills, like whether or not one should build a wall. Every premise of Socrates’ argument is problematic.16 However, even if we were to justify all these problematic premises, we would still be left with a contradiction in the reasoning itself: the argument, which rests on an analogy or similarity between the knowledge of justice and that of crafts, ends up concluding something about justice that is manifestly not the case for ordinary crafts. To put it simply, one who has knowledge of how to perform a craft does not always and invariably perform that craft. The shoemaker, for example, does not, as a result of having the knowledge of shoemaking, always and only perform shoemaking actions. Thus the very success of the argument – which concludes that knowledge of justice entails performance of just actions – renders its central (and most controversial) premise false. Justice, it seems, is fundamentally unlike crafts. The specific way it is unlike crafts – according to the argument – is that it includes its own imperative to action. Unlike knowledge of shoemaking, knowledge of justice entails doing what it knows; it entails doing just actions. We might go further: one of the interesting aspects of craft-knowledge, as we have already noticed, is that such knowledge is ambivalent, that it includes its opposite. So, for example, the doctor, by virtue of the same knowledge that gives her the power to heal, is able to cause sickness. This ambivalence is a sign of craft-knowledge’s disanalogous nature with respect to knowledge of justice: an ordinary craft, like medicine, can be ambivalent because it does not include any imperative to act – neither to heal nor to make sick. What we need to make the 16

There is the strange equivocation between being someone with a certain sort of knowledge and being someone who is a certain sort of person. The claim that a just person does only just things is also debatable, at best. And, of course, not the least of these problematic premises is Socrates’ use of the analogy with craft-knowledge.

Philosophy vs Rhetoric in Plato’s Gorgias

65

craft worthwhile is a knowledge that can dictate that imperative. Ultimately, this knowledge must be unlike craft-knowledge in precisely the way Plato is asking us to notice: this knowledge must include its own imperative, as the knowledge of justice must do to be the architectonic craft. Again, one interpretative strategy here would be to claim that Plato did not recognize the disanalogy when writing the Gorgias. This seems unlikely. The way in which justice is disanalogous is essential. It is the very disanalogy, the ambivalent nature of crafts, that motivates our interest in justice in the first place: this lack in craft-knowledge is exactly what has to be part of the architectonic craft. As we have seen, unlike the other crafts, the architectonic craft does not and cannot have a purpose that is outside itself. It cannot require the knowledge of a user. Indeed, that is exactly the requirement for the highest craft – that it not be for the sake of any other craft. Thus the reason knowing entails doing in the case of this highest craft is because, in this one case, unlike that of the lower crafts, the craft-knowledge does include the purpose. Whatever we are to conclude about Plato’s use of a strangely and pointedly flawed argument, my affirmative reading of the text motivates the need for a disanalogous activity. I suggest that the obvious and unavoidable flaw in the argument, as well as the fact that Gorgias has missed the point it is signalling, is telling. Plato, it seems, is pushing us to think about the nature of this peculiar craft, while simultaneously indicating, dramatically, that Gorgias does not have knowledge of it.

5. Review of the analogy and disanalogy It will be helpful to review both the analogous and disanalogous qualities of the architectonic craft. With these qualities in hand, we can then compare the activities of the rhetorician and the philosopher. Doing so will enable us to see that philosophy does, and rhetoric does not, fulfil the criteria for this highest ‘craft’. Both the analogous and disanalogous elements are necessary for the argument in favour of philosophy and are, I suggest, intentional on Plato’s part. On the one hand, the highest craft must be analogous to other crafts. This is because the highest or ruling activity must require the actions of the regular crafts. What I mean by this is that the architectonic craft, if it is to rule over the other crafts, must serve as a reason or purpose for their existence – it must be their end. The relation between a craft and its end can be understood by looking at, for example, horse-riding and bridle-making. Just as the horse-rider ‘rules over’ the bridle-maker by dictating to him the end for which the bridle is to be

66

Skill in Ancient Ethics

made – including both whether the bridle is needed at all, and whether it is good – the ruler of the state dictates to the wall-builder whether, what type of, and whereabouts a wall is needed, as well as whether or not the wall built is good enough for its required purpose. This knowledge supplied by the user is not included in the knowledge of how to build a wall. Rather, the latter is subordinate to former. It is subordinate in just the way any craft is subordinate to the activity that uses its product. In other words, the ruling craft must use the products of all the other crafts, either directly, or through some series of subordinated crafts.17 If, then, the ruling craft requires the other crafts in a way that is analogous to the way the horse-rider requires the bridle-maker, it must require them – use them – to perform its function, or to do so well. It is in this way that the activity in question (i.e. the highest or architectonic activity) must be a craft, or be at least analogous to one: the highest end or good uses the crafts immediately below it in the same way that those crafts use the ones below them.18 On the other hand, the architectonic activity must also be disanalogous to that of the crafts, and this also in a particular and determinate way. We have seen this several times already. In short, it follows from the fact that the activity in question is the highest, and final, end or purpose. This entails that it not be used by some other activity as, if it were, that activity would be higher and we would have to ask of it if it is the highest activity or if it is used and so ruled by a still higher activity. In other words, this activity or knowledge has to possess within itself its own imperative to action – its own answer to the questions of when, of why, and of whether it is done well. Which is to say, it must be intrinsically useful and good and not simply useful or good for some external end. To bring this point home, the activity in question is fundamentally disanalogous to the activities of the crafts in that it is not for the sake of some end outside itself, and is not choiceworthy because of that end. Rather, it is choiceworthy in virtue of itself. It is significant, as well, that this entails that it be choiceworthy – i.e. good – irrespective of any circumstances, or that it always be good; the highest craft, unlike the other crafts, cannot be ambivalent. Thus knowledge of it includes knowledge of its being choiceworthy, or, we might say, knowledge of it automatically entails choosing it, just as Socrates suggests in his refutation (460b7–9). It is on this very point – that knowledge of justice entails just action – that Gorgias’ account stumbles.

17 18

The ruler need not use the bridle if she is using the horse-rider’s craft or that of the general. The end that horse-riding is for bridle-making, for example – the way in which horse-riding rules bridle-making by using its product – is the paradigm for the relation between the final highest end or good and those below it. (Cf. Aristotle, Nicomachean Ethics I.1 and I.7 for an analogous, though importantly distinct, account.)

Philosophy vs Rhetoric in Plato’s Gorgias

67

Significantly, the analogousness is essentially linked to the disanalogousness. That is, the activity must be craft-like and un-craft-like for aspects of the same reason: its being the highest or final craft. In addition, the ways in which it is analogous and disanalogous are determinate. It is analogous to crafts because it fulfils the same use relation: it uses the other crafts in just the way that they use each other. It is disanalogous because it fulfils that relation in a more complete, self-sufficient and fundamental way: it is the final end (and not an end that is itself a means). Thus Plato doesn’t get rid of the ambiguity, he incorporates it. Significantly, the ambiguity is that of the final craft – which is an order and an ordering. Or we might say it is the ambiguity of its object, the soul – which is an order with no external end, an order ordered around itself. With this account in hand, we can turn now to the final point: the way the account deals with the contest between philosophy and rhetoric for the title of architectonic craft. My focus here will be illuminating what the account tells us about the nature of each contender.

6. Philosopher kings in the Gorgias It is intuitive, perhaps, that rhetoric does not qualify as the highest good, craft or activity. Still, it is to Plato’s credit that he is not satisfied with this intuition and supplies the reason for it: namely, that rhetoric, especially as understood by Gorgias’ students Polus and Callicles – as, roughly, the bending of others to one’s will – is easily seen to be for the sake of, and subordinate to, another type of knowledge. In short, we still need to know when and why to bend the will of another to our own – we still need a purpose for the bending of the will of another. This purpose is exactly what Socrates is searching for in the text. Further, this knowledge is more important than that of how to bend the will of another, at least for Plato. As in the case of crafts, the knowledge of the purpose is the knowledge that grants value to the action. Plato’s way of labelling this knowledge is as the knowledge of justice. Thus rhetoric requires knowledge of justice, which is exactly what Socrates suggests all along. In this way, Gorgias was right to agree with Socrates that he would teach his students knowledge of justice if they did not have it already, even at the expense of contradicting his own earlier statements.19

19

Or, rather, he was right to agree but inconsistent with himself, which is exactly what Socrates suggests: compare 460a5 (‘Hold it there. You’re right to say so’) with 461a2.

68

Skill in Ancient Ethics

The claim that philosophy should occupy the top of the hierarchy is not as easily justified as the claim that rhetoric should not. The argument for it, even the claim itself, is never given explicitly in the Gorgias. On my reading, it surfaces only when the reader actively thinks through the implications of the flaws in the argument, like the one described above, and thereby comes to see the architectonic structure as a guiding principle of the account. Indeed, even the architectonic structure itself is left implicit in the text (albeit clearly alluded to). In addition, it is striking how seamlessly the philosophical life, as it is conceived of in the Apology – as the life of inquiry into virtue, and the highest good for a human being (Apo. 38a2–4) – satisfies the criteria for the highest end. I don’t want to get off topic discussing Plato’s account of and argument for the life of inquiry as the highest good. That is its own topic. What I want to look at here, rather, are the characteristics such a life would need to fulfil the criteria for the architectonic craft, and the ways in which the life of active inquiry for its own sake possesses them. There are at least three relevant characteristics: inquiry is an end in itself, it is not ambivalent and not teachable. All are controversial, and this section is unlikely to be persuasive on its own. Yet if what we have seen concerning the text’s structure is correct, there must be an activity like Socratic inquiry to solve the problem of the highest good. That the life of inquiry solves it is itself an argument in its favour. To call the activity of inquiry an end in itself is both debatable and trite. Without wading into the long and worthwhile scholarly debate over the issue, it will suffice for our purposes to note that the inquiry in question is into virtue and the highest good for a human being. As such, the inquiry, if it is for the sake of anything, is for the sake of the very knowledge we are seeking here: that of the highest good for a human being, human virtue or the order of the soul. This entails that the end of inquiry, if it has one, be justice. Yet – and here is the dialectic – justice, to be an ordering like any other, must be an ordering in relation to some end. What is the end in relation to which the just person orders her soul? I submit that Socrates’ answer to the question is inquiry. Just as the shoemaker orders the shoe with an eye towards enabling it to fulfil whatever function it is to serve, the just person orders her soul so that it can fulfil its function of inquiring into the order of the soul. Note here that the essential point is not that this inquiring arrives at the said knowledge. What is essential is that the soul remain committed to the pursuit of such knowledge as that in relation to which it orders itself. Thus it is not so much inquiry alone that is an end in itself, but inquiry and justice together, as if Plato were striving to give us a picture of the structure of a thing that is self-related: the inquiring is into the order that

Philosophy vs Rhetoric in Plato’s Gorgias

69

is justice, and the order that is justice is an ordering that is for the sake of inquiring. The two together form a whole that is for its own sake,20 and it is essential that the end be also for the sake of the order, as inquiry into virtue is. One might ask why, if it is the order and not the end that we are seeking, and the said order requires some end around which to be ordered, could we not simply choose any end and order ourselves around it? One answer has already been given: that the end must be for its own sake. Another answer brings us to the second criterion: unambivalence. To be unambivalent, inquiry, unlike craftknowledge, must contain its own imperative. We have now seen that it does so. It is striking, in addition, that this internal imperative manifests in the way inquiry includes its own negation. In other words, its unambivalence manifests even in the dialectical argument for inquiry as an end. The argument is simple enough, though the details remain difficult.21 In short, if we acknowledge the need for an end to order ourselves around, and recognize that we do not know that end and, further, that pursuing the wrong end can result in self-harm,22 the only end we can choose safely is inquiry. If we choose inquiry into the human good as end, we inquire; if we do not choose inquiry as end, and wish not to harm ourselves, we are impelled to inquire into the nature of the human good – into the end we are to pursue. Thus the virtue of Plato’s argument for inquiry is that it produces conviction even if we are not persuaded. Because we all want what is best for ourselves, if we remain unpersuaded that inquiry is what is best, we are still just as motivated to inquire into what is best. It is in this sense, also, that inquiry is unambiguous. That is, inquiry, due to its non-ambivalence, is different from the other crafts in just the way that Socrates’ strange, and flawed, argument suggests: although a shoemaker can refrain from making shoes, a person committed to inquiry will always inquire.23

20

21 22

23

Irwin sets the issue out well but misses Plato’s dialectical solution. He holds what he calls ‘psychic harmony’ (which is what I would call justice) and ‘knowledge of justice’ (which I would replace with inquiry into virtue) apart (Irwin 1977: 128). Plato, I suggest, is putting them together. Indeed, this is the crux of both the analogy and disanalogy with crafts: justice is like a craft in being set over some order. It is unlike a craft in that the thing ordered is identical with the thing doing the ordering; both are the soul of the just person. In other words, as the argument in the text affirms, one who learns what is just is a just person (460b7). A more complete argument for this point is provided in Halper 2004. That nobody willingly harms herself is a familiar Socratic view (in the Gorgias, see 468b2–3). Because avoiding harm requires knowing what one wants (466e8), the ability to truly benefit oneself rests on our ability to identify what is truly beneficial for us. We might mention, in passing, that any other end we choose, any end we assume as the highest good, could prove to be harmful and disordering in two ways: we could either achieve this end, become satisfied and no longer have an end around which to order ourselves, or this end could be pleonexic and intrinsically disordering. The former is the problem with Polus’ form of rhetoric, the latter with Callicles’.

70

Skill in Ancient Ethics

Unlike the first two, the third criterion – that inquiry is not teachable – has not been discussed in this chapter. Although Socrates asks Gorgias whether he will teach his students about justice, we know that Socrates never makes such a claim himself. Indeed, he is known to suggest that justice is not teachable by anyone. I suggest that this fact too – justice’s unteachability – strengthens the claim of inquiry to be the highest good. The end around which the soul is to order itself is an activity that the one who is ordered must engage in herself for it to be of value. One can show another that this activity is necessary and beneficial, but one person cannot engage in the inquiring for another.24 In other words, there can be no positive doctrine that Socrates could teach someone in order to make them just. All he can do is show them the simultaneous need for the knowledge in question (knowledge of the highest human good) and that they do not have this knowledge. In doing so, he will be impelling them to inquire for themselves. In other words, Socrates’ doctrine is to know that one does not know these things of value, and his ‘teaching’ is the elenchus itself. Thus I am, on this point as well, in disagreement with those who, like Irwin, take the elenchus to represent a different doctrine from the craft analogy. When, for example, Socrates says about Callicles that ‘[t]his fellow won’t put up with being benefited and with his undergoing the very thing the discussion’s about’ (505c2–4), he is referring to Callicles’ refusal to accept the elenchus and, thus, his refusal to commit himself to inquiry and the ordering of his own soul. So, too, when he encourages Polus, in the middle of his refutation of him, to submit himself ‘nobly to argument as [he] would to a doctor, and answer me’ (475d8–9). In both cases, Socrates is spurring the two men towards inquiry into the good as the end around which to order themselves. He is not suggesting to them a better, static, answer to the question of goodness, but offering an activity that itself will play this role. Irwin’s concern that there is no replacement in the Gorgias for the role that pleasure plays in the Protagoras (Irwin 1977: 131) – the role of an end around which to order the soul – is answered not by Socrates’ words, but by his deeds: the replacement is active inquiry into the end as an end in itself.

24

The issue is more complicated than depicted here. There is a sense in which the ruler is inquiring for craftsmen lower than her. The sense is that she – along with her life of inquiry – is the end for which their crafts are of value. Thus as members of a just state with a philosopher ruler, the craftsmen do, in a way, have someone else engage in the life of inquiry for them. This issue is developed in the Republic. I will turn to it, albeit briefly, in the next and final section below. It will point us beyond the scope of this chapter (see Bloom 2017: ch. 4).

Philosophy vs Rhetoric in Plato’s Gorgias

71

7. The use of rhetoric With this account of the role of inquiry as highest good in hand, we can now see another, deeper, argument against rhetoric, while also appreciating its ‘supernatural scope’. This argument against rhetoric is connected to the very nature of persuasion. In short, to be persuaded about something entails believing that one no longer needs to inquire into that thing. Thus if we are persuaded that we already have knowledge of the highest things, a knowledge that Socrates consistently denies is possible for human beings,25 we will not inquire into those things. Rhetoric, if taken as the highest or ruling activity, undermines the philosophical life as Socrates conceives of it. Yet why, then, we might ask, is Socrates (or Plato) interested in rhetoric at all?26 Is it simply that he is concerned to show that philosophy is superior? Although there is some truth to this, there is also a deeper reason having to do with rhetoric’s scope and, again, with the architectonic structure. The reason arises from the fact that the craft of justice – active inquiry – is not set primarily over the soul of another, but of oneself, the inquirer. That is, my life of inquiry is, in the first instance, the activity around which I order my own soul. The philosopher, the inquirer, uses the state as a whole so that he can sit around and think about virtue and such things without being disturbed by having to build walls, make bridles or grow crops, for example. Thus he orders his own soul around the activity of inquiring. The benefit for others, on the other hand, comes from their each having a place in the architectonic structure: a place in which they can perform their own function in a way that is also, ultimately, ruled by a principle that is non-ambivalent, by the life of inquiry. Thus their functions too can have the character of not being harmful to them and of being the ends around which they order their own souls.27

25

26

27

This claim is so central to Plato that there are too many places in the dialogues to reference. Two that are of special significance for my account are Apol. 23b1–3 and Rep. 505a4–6. The claim entails that inquiry must be valuable for its own sake (and not only because it may lead to knowledge), as it is on my account. This question echoes one of the opening questions of the dialogue. Callicles asks Chaerephon, with surprise: ‘What’s this? Is Socrates eager to hear Gorgias?’ (447b3–4). Richard Parry is right to call attention to whether the justice in question is for the sake of the self or for ruling others (see Parry 1996: 3–5 et al.) I suggest that the justice in question here manages to be for both. In ordering herself around the life of inquiry, the ruler is able to benefit the state through ordering it around her end as well; that is, she benefits the state by using it for her end. In doing so, she protects everyone in the architectonic structure from inadvertent self-harm. To do so, she will need the aid of rhetoric.

72

Skill in Ancient Ethics

This is, of course, a much longer story.28 The point I wish to make here is that Socrates requires a way of committing those around him to remain in the structure and to perform their own crafts. Such a performance provided the crafts are done for the right end, is beneficial for them; it is the function around which they order their own souls. Thus their acting for the sake of the ruler’s inquiry is what makes their own crafts beneficial for themselves. To enable this self-benefit for them, he, or someone who listens to him, is to persuade them, for their own sake, to be committed to their crafts and to the architectonic structure in general. In other words, the ruler will need rhetoric. Allusions to such a use of rhetoric are peppered throughout the Gorgias. In fact, the very passage in which Gorgias praises rhetoric’s ability to encompass and subordinate contains one of the most significant of these allusions: Oh yes, Socrates, if only you knew all of it, that it encompasses and subordinates to itself just about everything that can be accomplished. And I’ll give you ample proof. Many a time I’ve gone with my brother or with other doctors to call on some sick person who refuses to take his medicine or allow the doctor to perform surgery or cauterisation on him. And when the doctor failed to persuade him, I succeeded, by means of no other craft than oratory. 456a9–b5

Although Gorgias goes on to laud rhetoric inappropriately in the very same passage, this does not undermine the value of rhetoric that he has stumbled upon, and, apparently, embodies to some degree (although he himself is unclear about it, and hence fails to teach it to his students Polus and Callicles). Rhetoric is valuable when it is accompanied by, or rather when it accompanies and is subordinate to, knowledge. This is true when the rhetorician accompanies and subordinates himself to the knowledge of a doctor, and it is true when he accompanies and subordinates himself to a just ruler.29 In other words, Gorgias should either make sure his student knows what justice is or accompany (and subordinate himself to) someone else who does. Of course, this analogy too has its limit. The limit is instructive: the philosopher does not have knowledge in the definitive way in which the doctor does – he does not have a definite, discursive, knowledge of the good for a human being. Yet as we have seen, he has a worthwhile replacement. The replacement is a commitment to inquiry into the highest good or end, which will serve as a sort of surrogate for that good or end. 28 29

Again, it is a story told in the Republic. Of course, the accompaniment does not have to be in the form of another person. The just ruler, like the doctor, could complement her own craft with a knack for rhetoric and accompany herself.

Philosophy vs Rhetoric in Plato’s Gorgias

73

Thus philosophy and rhetoric, the two activities with a universal scope, are distinguished. The former is entirely non-ambivalent, and attaches to every craft as its rightful and beneficial director by setting its ultimate end. The latter is entirely ambivalent, so much so that its value cannot be consistently estimated, except to say that it is not what Gorgias, Polus and Callicles take it to be: the master craft and highest good.30 It does indeed attach to every craft. It does so never, though, as its rightful ruler, setting the end or purpose, but always merely as a tool for attaching the craftsman in question to a structure that will allow him to be of benefit, both to others and to himself. There is, perhaps, something distastefully unmodern about Plato’s account, which may strike us as tyrannical. Indeed the valuing of anything, even virtue, over freedom is unmodern, and in that sense, Plato is guilty as charged. However, the claim of tyranny (or even totalitarianism)31 is deeply misplaced. Two reasons spring directly from what we have seen here. In the first place, the ruler is responsible not for imposing some ideal of morality on the population; indeed, she does not even impose justice. Rather, she functions as a sort of lodestar: as an end toward which the state can direct itself, so that each citizen, through his own actions, can order his soul safely around the practice of his own (inquiry-like) craft. This is the meaning of justice in the text. In the second place, tyranny generally entails the tyrant having some ideal, even if it turns out to be an entirely self-serving one, which he imposes on his people – that is, tyranny entails knowing what ideals to subordinate the state to. Yet the very worth of our ruler entails her being free of the claim to know any such ideals. Indeed, her defining characteristic is her recognition that she does not know (her ‘Socratic ignorance’), which is the very thing that entails her living a life of inquiry. The potential for a tyrannical, totalitarian, element comes with the role of rhetoric – the secondary role of the ruler (her brother who goes around with her, as it were). This is the very element of the picture that looks totalitarian: the one that subordinates the wills of others. That is perhaps the reason why Plato goes through this account in the Gorgias – one that in many, central respects matches that of the Republic – with a focus squarely on the dangers and tyrannical possibilities inherent in rhetoric; dangers that arise when it takes control of setting the end.

30

31

Thus the alleged difference in the appraisals of rhetoric in the Gorgias and Phaedrus is, on this account, less a difference in appraisal than in focus. The concern in the Gorgias is to show that rhetoric is not the master craft. The Phaedrus takes that conclusion for granted, and focuses on rhetoric’s other, more ambivalent qualities. Famously made in Popper 1995.

74

Skill in Ancient Ethics

References Barney, R. (2010), ‘Gorgias’ defense: Plato and his opponents on rhetoric and the good’, The Southern Journal of Philosophy 48: 95–121. Bloom, L. (2017), The Principle of Non-Contradiction in Plato’s Republic: An Argument for Form, Lanham, MD : Lexington Books. Cooper, J.M., and D.S. Hutchinson. (1997), Plato: Complete Works, Indianapolis: Hackett Publishing. Dodds, E.R. (1959), Plato: Gorgias, Oxford: Oxford University Press. Halper, E. (2004), ‘Plato’s case for philosophy as the examined life’, in K. Boudouris (ed.), Conceptions of Philosophy. Ancient and Modern, 133–150, Athens: Ionia Publications. Irwin, T. (1977), Plato’s Moral Theory, Oxford: Clarendon Press. Parry, R. (1996), Plato’s Craft of Justice, Albany, NY: SUNY Press. Popper, K. (1995), The Open Society and Its Enemies, London: Routledge. Roochnik, D. (1996), Of Art and Wisdom: Plato’s Understanding of Techne, University Park, PA : Pennsylvania State University Press. Zeyl, D. (1987), Gorgias, Indianapolis: Hackett Publishing.

4

A Dewian Conception of Skill as Clue to the Analogy Between Craft and Virtue in the Platonic Dialogues Robin Weiss

It has long been scholars’ task to understand why Plato’s Socrates suggests an analogy between virtue and technê, or ‘skill’. In what respects is virtue similar enough to skill for this analogy to be invoked by Socrates, and in what respects, if any, is the analogy inapt? The problem – so we argue – is that scholars assume a particular conception of technê, such that they identify as distinctive of technê characteristics, which not only have a doubtful claim to be distinctive of all technai, but which also make it difficult to find appropriate points of comparison with virtue. At the root of their view lies an assumption about technê as such. A technê, in short, is commonly taken by these scholars to have the following two characteristics: 1. 2.

it has one, clearly delimited end; this end differs from those of other technai.

These are the two characteristics of technê in what we shall call the ‘popular view’ of it. The virtues compared to technai are therefore assumed to share the same characteristics: each virtue has a single and unique end. However, when scholars apply this conception of technê to virtue, it makes it difficult to understand how Socrates and his interlocutors can possibly avoid the aporiai into which they stumble. It is no wonder that, faced with this definition of virtue and the decision to either assert or deny that virtue is a technê, some scholars assert that, for Plato’s Socrates, virtue is a technê, forcing a strained comparison between the two, while others deny that it is, abandoning the skill analogy altogether. These two extremes of interpretation can be avoided, and more light shed on the problem, if one realizes that the above assumptions can be challenged. To this end, however, we need a conception of technê that questions precisely what 75

76

Skill in Ancient Ethics

the popular view assumes. This conception of technê is that advanced by American pragmatist, John Dewey, a philosopher who has advocated the analogy between skill and virtue in modern times. Dewey proposes to define technê broadly as any adaptive activity that brings the external environment into a state of usefulness for human ends (1922: 15). As such, a Dewian technê has two characteristics: 1. 2.

it may be exercised toward multiple ends at once; none of these ends need be exclusive to it.

This conception of technê, as we can see, is diametrically opposed to the ‘popular view’. We shall argue that this conception of technê has significant implications for our reading of the Platonic dialogues and our final assessment of whether Plato’s Socrates endorses the view that virtue is a technê. How to proceed? We will turn to Socrates’ conversation with Polemarchus in Book I of the Republic. For these passages (at 333e1–334b2) provide a particularly useful example of how Socrates’ every attempt to understand virtue as skill ends in aporia. Although we shall focus on these passages in particular, asides and footnotes point out patterns that recur in the early dialogues. To begin, we shall examine these passages in order to show why, when the popular view of technê is assumed, they generate unsolvable aporiai (Sec. 1–2). Thereafter, we shall use Dewey’s conception of technê to interpret these passages. First, Dewey’s conception throws into particularly stark relief some of the assumptions that are made about technê by scholars – and by Socrates’ interlocutors (Sec. 3). Second, once these assumptions have been questioned, it allows us to understand the aporiai the popular view generates (Sec. 4). Third, it points the way out of the aporiai generated by the popular view (Sec. 5). Finally, whereas the opposing view makes it difficult to see how the virtues can be compared to skills with profit, it is much easier to find appropriate points of comparison between skill and virtue if skill is first conceived on a Dewian model.

1. Specifying an art’s end The most basic question we can ask about an art, and the first question Socrates asks Polemarchus about justice is, ‘What is its use?’ This question might be answered in any number of ways. One way to specify the benefit of an art is by citing (a) its sphere of expertise. For example, we can say that medicine is useful for working in the sphere of things medical, which is to say, with the body and

A Dewian Conception of Skill in the Platonic Dialogues

77

medical instruments, since these are the things medicine both knows about and knows how to use. Another way is by specifying (b) the art’s function, or the set of activities in which it allows a person to engage. For example, medicine is useful for performing surgery or administering medicine. Yet another way is by specifying (c) the end the art achieves, where this is distinct from its function. For example, medicine achieves health. At the start of their conversation, Polemarchus is relatively sure about the ‘benefit that justice gives’: ‘it gives benefits to friends and does harm to enemies’ (332d4). But as Socrates points out, doctors can also benefit their friends. It is therefore necessary to ask whether the just person performs this function with respect to a particular sphere of expertise. Hence the question: In what actions and with a view toward what work (en tini praxei kai pros ti ergon) is he most capable of benefiting friends and harming enemies? 332e2, trans. mod.1

The doctor has a particular realm of expertise, so that in medical affairs he can achieve the end of benefiting friends – or harming enemies, as the case may be. As Socrates puts it, a doctor is most capable of treating friends well and enemies badly ‘with respect to matters of disease and health’ (pros noson kai hugieian: 332d9). And the just person? This question brings us to the central problematic of the conversation. Socrates, it seems, will not be satisfied until Polemarchus can identify a function that justice alone performs, or that it performs better than other skills. But for any ergon or ‘function’ that a virtue claims to perform, there is a technê that performs the same ergon. Cases like these, which are common in the early dialogues, are usually interpreted simply to mean that the definition of the virtue in question is too broad, and that additional differentiae are needed.2 But the definition is too broad because a given virtue is assigned an ergon, and a craft or skill can be found that performs the very same ergon – often more efficiently than the virtue in question. The virtue thus loses its claim to specificity because it does not have an ergon which it performs uniquely well. Now, Socrates and Polemarchus, having so far failed to identify an ergon that justice performs uniquely well, redouble their efforts to identify the sphere in

1

2

Translation modified. Except where otherwise noted, all translations are taken from Grube and Reeve 1992. In the Charmides, once temperance is defined as ‘the doing of good things’ (163e4), Socrates refutes the definition by pointing out that doctors also do good (164a2–3). See also: Charm. 173d–e; Laches 192d10, 193a1–b9.

78

Skill in Ancient Ethics

which justice performs this ergon. For even if justice does not have a unique end, it will at least have a unique sphere in which it achieves this end, just as carpentry, although not unique in producing furniture, remains unique in doing so in the sphere of, or with respect to, wood. If the just person is not alone in benefiting friends, they reason, then perhaps he’s able to do so particularly well in particular situations. Polemarchus hazards a guess: ‘It’s in wars and alliances, I suppose’ (332e4). The problem is that this implies that justice is useful only in a restricted context, and that it is useless in other contexts: And to people who aren’t at war, a just man is useless? No, I don’t think that at all. Justice is also useful in peacetime, then? It is. 332e5–333a1

Polemarchus immediately recognizes the error of restricting justice’s field of expertise to war, and expands it to peace as well, agreeing that is ‘useful in peace’ (chrêsimon. . .en eirênêi: 332e9). And so it goes on. The problem is not that Polemarchus cannot identify an end or function for justice – he identifies several – but that he cannot identify a single end or function that justice alone achieves or performs, whether in general or with respect to a particular sphere of expertise. The conversation carries on in this way, until a set of assumptions are made about skill.

2. A false analogy: one skill, one end It is not exactly clear when or by whom these assumptions are first made, since they are not stated explicitly. It can be argued that Socrates himself makes these assumptions, since later in Book I, he asserts that (1) every skill has (at least) one ‘peculiar benefit’ (ôphelian . . . idian 346c2), and at another point, that (2) the function of each object can be identified with ‘what it alone can do or what it does better than anything else’ (353a7). However, several provisos attach to these claims. For instance, a skill can plausibly have a unique ‘benefit’ without having a unique function or end. (Perhaps some skills’ unique benefit is to perform several non-unique functions.) Moreover, Socrates has not explicitly excluded the possibility that sometimes a skill’s – as opposed to an object’s – function is what it does no better than other skills. Or, if a skills’ function is, by definition, what it does better than other skills, Socrates does not exclude the possibility that a skill

A Dewian Conception of Skill in the Platonic Dialogues

79

can have multiple functions (i.e. tasks it performs better than other skills).3 In any case, Socrates says nothing of the kind in this conversation, so Polemarchus reaches this conclusion with minimal prompting from Socrates. Let us see how: In what kind of partnership, then, is a just person a better partner than a lyreplayer, in the way a lyre-player is better than a just person at hitting the right notes? In money matters, I think. Except, perhaps, Polemarchus, in using money, for whenever one needs to buy a horse jointly, I think a horse breeder is [more useful as a partner], isn’t he? Apparently. And when one needs to buy a boat, it’s a boatbuilder or a ship’s captain? Probably. In what joint use of silver or gold, then, is a person more useful [as a partner] than the others? When it must be deposited for safe-keeping, Socrates. 333b5–c5, trans. mod.

Polemarchus has just agreed that the end of justice is partnership (koinônia, sumbolaia). But this end is attained also by other people, with other skills. So, as if challenging Polemarchus to justify the claim that justice has this function by proving that it masters a corresponding field of expertise, Socrates asks after the context in which the just person achieves the aim of partnership, viz. the specific kinds of businesses and partnerships (eis tina de koinônian) in which he does so (333b7). Polemarchus’ answer is ‘in money matters’ (eis arguriou [sc. koinônian]). So now the just person appears to be an expert in using money. But Socrates points out that, depending on the specific context, a horse-breeder, a boatbuilder or a ship’s captain, each with his mastery of a particular expertise, will in many cases be better able to use money for business purposes. It will be suggested that Polemarchus could have avoided taking a wrong turn here, if only he had recognized that justice has no clearly-defined domain.4 It will

3

4

The second claim is intended to apply only to objects such as pruning shears and eyes, leaving unclear whether it applies also to skills. In any case, if what applies to objects applies also to skills, neither assertion excludes the possibility that a skill has multiple functions. Socrates clearly thinks things can have multiple functions. The soul, for example, seems to have several: two of these being ‘deliberation’ and ‘living’ (353d7). This interpretation is advanced by advocates of the view that justice is a second-order skill, notably Sprague 1976. Sprague argues that Socrates interlocutors’ mistake is to agree that virtues such as justice have a specific and restricted domain or subject-matter (29). In this way, she suggests, we can jettison the idea that justice has a specific domain, while retaining the idea that justice has a single function or end, such as ‘the Good’ (91). (On other occasions, defenders of this view seem to deny that second-order skills have specific ends, because they lack an identifiable product (xvi).)

80

Skill in Ancient Ethics

then be asked why Polemarchus does not simply insist that justice’s unique ergon is promoting partnership, without conceding to Socrates that it achieves this end only within a circumscribed domain. In other words, Polemarchus can affirm that justice has a unique end and still deny that it has a clearly-defined realm of expertise. But this view fails to see that he cannot simply win the argument by rejecting the idea that all skills have clearly-defined realms of expertise if he is still committed to this view by another assumption he makes. Indeed, this idea is more the result than the cause of the real problem. The real problem is Polemarchus’ belief that that every skill has a unique ergon. If a skill has a unique ergon, then to have that skill is to have gained unique knowledge and mastery of the field of objects one must know and master in order to perform that ergon. For example, a wood-carver has a unique mastery of wood and its properties as they relate to carving, as well as the tools of the trade.5 Having a unique function thus implies having a unique realm of expertise. This is why, whenever Polemarchus suggests that a skill has a unique function, Socrates rightly asks Polemarchus to justify his claim by proving that it has unique mastery of a corresponding realm of expertise. It’s here that, exasperated, Polemarchus finally gives his most definitive reply: the just person is most useful when possessions ‘must be deposited for safekeeping’ (parakatathesthai kai sôn einai: 333c6). Before offering this response, Polemarchus seems to have made two major assumptions, the first being as follows: 1.

Every art has a single, unified end. He affirms the claim that, in each individual case, it is possible to identify the end an art achieves best among ends, and to clearly delineate this end as the end of the art in question. In other words, he is not willing to entertain the possibility that justice has too diffuse, or too ill-defined, a set of ends for it to be identified with a single, unified end.

Note that it does not follow, however, that – if a skill such as justice has a main end (say, the end that it is best at achieving) – this main end is also its unique end (the end it alone attains or attains better than other skills), although Polemarchus seems to reason this way. The second of Polemarchus’s assumptions, then, is that skill not only has an end, but one that differs from that of other skills. That Polemarchus makes this

5

Advocates of second-order skills are thus similarly committed to the view that a second-order skill has a unique realm of expertise: for example, the arts and their products, insofar as they can be organized.

A Dewian Conception of Skill in the Platonic Dialogues

81

assumption is evident from the fact that, while searching for justice’s end, he eliminates from contention any end that is more efficiently attained by another art. To Polemarchus’ ear, Socrates seems to ask, ‘What is the one end that is not better achieved by the bricklayer, the harpist, or any other person with any kind of skill – but by the just person alone?’ By the time Polemarchus has finished eliminating all the ends that other arts promote as possibilities for the end of justice, only one possibility remains. This is to keep things safe for the time when they will be put to use. Here Polemarchus makes his second assumption: 2.

Every art has a unique end which it can achieve better than the other arts. He is not willing to accept the possibility that justice has an end that is shared with other arts, or for that matter, the possibility that it has no end that is not more efficiently achieved by other arts.

When Polemarchus finally suggests justice’s end is safekeeping, this appears to put a stop to Socrates’ insistent questioning: it supplies justice with a unique end – an end that does not set it in competition with the other arts (with their various areas of expertise). Arguably, however, this end is a ‘trivial’ one.6 If any doubt remains that Polemarchus makes these two assumptions, then it is put to rest when he infers that one skill is useful when one is pursuing one end, and another skill is useful when one is pursuing another end. ‘You’ll agree, then’, Socrates asks, ‘that when one needs to keep a shield or a lyre safe and not to use them, justice is a useful thing, but when you need to use them, it is soldiery or musicianship that’s useful?’ (333d5–7). Polemarchus now interprets his own definition of justice in the most literal sense possible. He interprets it to mean that when one wants to keep something safe, be it a scythe, a shield, or a lyre, justice is useful for this end. But as soon as one wants to employ these possessions for another end, then justice ceases to have any use, and it is rather the vinedresser’s art, military skill, or musicianship that allows one to attain one’s end. This conclusion follows, however, only if one accepts the claim, as Polemarchus clearly does, that each skill is adapted to the pursuit of a specific end of its own, one not shared with other skills. This brings us to a third claim that, as we have seen, either follows directly from the two previous claims, or is at least strongly suggested by them. The two previous claims state that different skills have different ends. To this extent, they

6

According to Julia Annas, ‘Polemarchus is implicitly thinking of justice as being like a skill with a given end – but with the defect that in this case there is no important given end that justice helps us attain’ (1981: 27–28).

82

Skill in Ancient Ethics

give us reason to suspect that, just as there is a one-to-one correspondence between each art and its end, there is a one-to-one correspondence between each art and the realm of expertise necessary to reach that end. Hence the following conclusion: 3.

The realms of expertise of different arts are mutually exclusive. He affirms that only one skill is best adapted to the attainment of a certain end and has knowledge and mastery of that which is necessary to attain that end.

For Polemarchus, this means that justice must retreat from, and cannot stake a claim to, any domain that another art has already occupied. The just man’s movements are then tightly restricted to a single sphere of life. Indeed, Polemarchus watches as the times and places in which justice finds its use shrink from times of war, to contractual matters, to matters concerned with the joint use of money, to times when something needs to be kept safe for later use. The result is ironic and paradoxical because it turns out that justice is useful when other arts are not in use, and thus, when nothing is being put to use, and no work is being done. This result is pointed out by Socrates himself, when he quips: justice is useless in the use of each thing, but useful in useless things. (hê hekastou en men chrêsei achrêstos, en de achrêstiai chrêsimos: 333d8, trans. mod.)

He also wryly remarks that justice isn’t worth much, if it only happens to be useful for useless matters (ei pros ta achrêsta chrêsimon tugchanei: 333e2, trans. mod.)

This is the aporia with which the conversation ends: justice, which at first seemed very useful, now appears completely useless.7 How could this aporia have been avoided?

3. Scholarly consensus Where did Socrates and Polemarchus go wrong? Most scholars seem to think that in order to understand the misstep that leads to this false conclusion, we have to go back to the very first assumption on which the whole conversation is

7

In other dialogues, Socrates and his interlocutors arrive at the conclusion that a virtue is either useless (Charm. 175d), or harmful (Laches 192d2).

A Dewian Conception of Skill in the Platonic Dialogues

83

premised: the assumption that justice is a skill. According to many scholars, this is why it ends the way it does: in aporia. Once it is wrongly accepted that justice is a skill, many scholars say, then the conclusion Polemarchus draws follows inevitably, and Polemarchus could not have avoided it except by rejecting the analogy between virtue and skill altogether (Annas 1981: 24–25; Pappas 1995: 24–25; Dorter 2005: 27; cf. Irwin 1977: 177–191).8 In the same way, it could be argued, a similar assumption about the skill-character of other virtues leads to aporia in many early dialogues (Roochnik 1996). Could Polemarchus not have retained the skill analogy while avoiding the aporia? Most scholars take for granted that the assumptions Polemarchus makes about skill are correct, and that the concept of skill itself cannot be divorced from the claims that lead him into aporia. Indeed, although scholars blame Polemarchus for accepting the technê analogy, most find nothing to fault in the logic that leads Polemarchus from the proposition that justice is a skill to his conclusion (Page 1990: 253), nor in the logic of interlocutors from other dialogues who reason to similar conclusions (Gould 1987: 268; Foley 2009: 218; Vlastos 1991: 109–117). As for scholars who take a more favourable view of the analogy, although they differ from those who think the analogy must be rejected, they still share the same conception of skill: the popular conception. Hence the case for comparing virtue to skill appears, to them, as for their opponents, to stand or fall with the ability to show that virtue has a single and unique end. Either we succeed in identifying a single and unique end for justice, in which case the analogy stands, or the analogy with skill will have to be abandoned entirely. Virtue’s end, to start with the first assumption, has been identified by scholars eager to maintain the analogy either with happiness (eudaimonia), as in the case of Irwin (1995: 65–77), or with the Good (to agathon), as in the case of Penner (2006: 166, 176–177). Sometimes, even when it is admitted that virtue has no single ergon or end, it is maintained that it nonetheless has an object, so that at least one remaining point of comparison to be drawn between virtue and skill is that virtue, like skill, has its own object or subject matter (Benson 2000: 220).9 As for the second assumption, that every technê has a unique end, this is often held

8

9

Pappas (1995: 24, 25–26) specifically claims that the analogy with skill must be rejected because the virtue of justice has no specific and unique end. Benson argues that, for Socrates, knowledge is a ‘power or capacity (dunamis) to make judgments . . . involving a particular object or subject matter’ (2000: 220). Roochnik, although he denies that every technê has its own product, affirms that it does have its own object (1996: 115). Roochnik rejects the analogy, however.

84

Skill in Ancient Ethics

implicitly. Although we rarely find having a unique end listed among technê’s characteristics, we often find this assumption implicit in scholarship. Indeed, Irwin goes so far as to assume that any art that does not have a unique end, and whose only ends are those better and more efficiently achieved by other arts, can be only a pseudo-art, ‘which trespasses on the area of specialized crafts but is incompetent’ (1977: 73–74). Is there no alternative? A third position is proposed by those who argue that virtue constitutes a special type of atypical skill. Proponents of this view assert that virtues such as justice are unlike other arts because they have a special relationship to them, for example, that of a second-order art. The risk involved in this solution, however, is that it stretches the analogy so thin that it ceases to be informative. Sprague, for instance, not only denies that justice has a clearly demarcated field of expertise – it lacks a ‘specific scope’ (1976: 29) – but comes close to admitting that justice has no clearly demarcated end, without saying in what respect virtue is like typical skills.10 Annas (1981:26) articulates a similar view when she suggests that Polemarchus ‘has collapsed too quickly’ in conceding that justice has one, clearly delimited end like other skills, yet (according to Annas) a relatively trivial one. First, she speculates that Polemarchus ‘could have said that justice is a skill aiming at an end, but that this end is not trivial – it is happiness or a satisfactory life’. Then she speculates that it is ‘a very general aim which we try to achieve in exercising more specific skills, and which therefore includes the exercise of the specific skills rather than being pushed out by them’ (1981: 26–27). But it is not clear if this makes virtue something other than skill. Perhaps this is why Annas abruptly suspends this promising line of inquiry, and concludes that, since every skill has a specific end, and since the only such end that can be found for justice is a trivial one – safekeeping – the simplest course would have been for Polemarchus simply to reject the very idea that virtues are skills in the first place. Hence scholars’ dilemma: if one wishes to assert a close analogy between skill and virtue, one must apparently assert that all skills, and hence all virtues, have specific and unique ends, however difficult to identify. Otherwise, one must abandon the analogy with skill altogether. Is there is no alternative? Perhaps there is, if skills do not necessarily have single and unique ends.

10

Sprague does say that justice is directed at the Good (91). But this appears tantamount to denying that it has a specific end at all, or specific and articulable content. In addition, it is not clear how this resolves several problems associated with second-order skills: for according to Sprague herself, ‘they have no content, they are useless, and they generate regresses’ (1976: 91). Sprague’s view has been attacked on these grounds by Roochnik (1996: 122–123).

A Dewian Conception of Skill in the Platonic Dialogues

85

4. A Dewian conception of technê Scholars typically assume that technê involves reasoning to an end, and therefore reasoning to a single and unique end. But these two things should not be conflated. It is possible to affirm that technê involves means-ends reasoning, and is purposefully directed at an end, without also affirming that every technê is therefore directed at a single and unique end.11 Socrates often makes the former claim,12 and he and Polemarchus are committed to it from the moment Socrates points out that a doctor does what is ‘appropriate’ for human bodies, and gives them what is ‘due and appropriate’ (ôpheliomenon kai prosêkon: 332b9–332d4). But the latter? Socrates certainly wants Polemarchus to view the just person as similar to the craftsman, in that he acts in a purposeful way, and does so by bringing the object upon which his action is directed into a state of serviceability to human ends, so that it is useful or beneficial. (That, he suggests, is why we consider the craftsman’s action ‘appropriate’ to its object.) Incidentally, the claim made here answers a debate about whether skills have external ends: all technai have an external end or product,13 not necessarily in the sense of an object that outlasts the exercise of the art, but in the sense that, in any given instance in which an art is exercised, it is directed upon an object, which is conceptually distinct from the art,14 and which it leaves in a better state than that in which it was found.15 However, nothing is said between Polemarchus and Socrates that compels them to assert that a skill’s end is single and unique. Let us proceed then, first, by explaining briefly what would happen were we to refrain from making this set of assumptions. Will we avoid the aporia? 1.

11

12

13

14

15

First, if we reject the proposition that each skill must have a single end, then we will assert that a given skill can have more than one end. The same Two other scholars who maintain, like myself, that virtues are technai, while denying that every technê has a specific and unique end, are Balaban (2007: 7–30) and Sprague (1976: xvi–xvii). Socrates often asks his interlocutors to agree that virtue is aimed at the good or happiness (Charm. 176a, cf. Gorg. 470e, Cri. 48b). In the Gorgias, he says that every technê aims at the good (Gorg. 463a–465a). However, Vlastos (1991: 200–232), Nussbaum (1986: 97–98) and Roochnik (1996: 34–36) have all argued, contra Irwin (1977: 71–75), that not even among the Greeks were all technai considered to have an end, in the sense of a product over and above the activities constituting their exercise. Here we should proceed from the assumption that, since the word technê originally referred to productive arts, and was only later applied to the non-productive (lyre-playing, singing, prophecy and medicine), it is unlikely that Plato’s Socrates would have compared virtue to a technê, only completely to divorce the word from its original association with productive arts. On the word’s gradual application to non-productive arts, see especially Roochnik (1996: 24–26, 34–36). This idea is compatible with the notion that some skills are ends in themselves, contra Aristotle’s sharp distinction between praxis and poêisis (Nic. Eth. 1140b7). Dancing, for example, can take the body as its object, and still have intrinsic worth.

86

2.

3.

Skill in Ancient Ethics

will apply to virtue. We will not have to accept the proposition that a virtue such as justice must have one end; instead, we can plausibly maintain that a virtue such as justice has more than one end. Second, if we reject the proposition that each skill has an end that differs from the ends of other skills, then we will assert, by contrast, that a skill’s end, or ends, can be identical to the end, or ends, of other skills. The same will apply to virtue. We will not have to accept the assertion that the end of a virtue such as justice must be identified with the one end that it pursues that is not pursued by other skills, nor even with the one end that it pursues more efficiently than they; instead, we can plausibly maintain that a virtue such as justice has many ends, the majority, nay, all of which it shares in common with other skills. Third, if we reject what follows from these two premises, namely, that when a given skill is useful, other skills are not useful, we will assert, by contrast, that a skill is useful when other skills are in use. Indeed, perhaps it is most useful when other skills are in use. The same will apply to virtue.

We may now proceed to a more detailed explanation of how these conjectures can be accommodated within a Dewian model of skill. In order to understand Dewey’s conception of skill, it is important first to understand his conception of habit, since for him, skills are nothing but habits, and conversely, ‘habits are arts’ (1922: 15). Usually of course, habits are only potentially in use, since a habit merely entails the ‘readiness to act in a specific fashion whenever an opportunity is presented’ (41). However, what is most distinctive of habits is that they are analysable into a series of steps or stages, and each step ‘marks progress in dealing with materials and tools, advance in converting material to active use’ (15). A habit is thus defined as ‘that kind of human activity which is influenced by previous activity, which is in that sense acquired; which contains within itself a certain ordering or systemization of minor elements of action’ (41). This brings us to the first major difference between Dewey and Polemarchus. For Polemarchus, each art must have its end. For Dewey, the same art can be applied toward multiple ends. For example, astronomy can be used for the purposes of navigation, astrology or farming. Even ballet has many ends: it can be deployed toward the end of exercise, achieving balance, choreographing, or performing. A habit or skill may therefore be deployed and prove useful in various situations, for attaining various ends. For example, if he needed to, a ballet dancer could draw upon his technique to leap over a puddle. Dewey thus sees no reason to associate particular habits or skills with particular ends.

A Dewian Conception of Skill in the Platonic Dialogues

87

It might be thought that Dewey would at least admit that every skill has a specifiable end, insofar as it has a specifiable sphere of expertise. Thus, it may be argued, the carpenter always works with wood, so his end is always ‘woodworking.’ However, nothing prohibits the carpenter from potentially applying the same techniques to other, non-conventional materials, especially those that possess properties similar to wood. Indeed, for Dewey, skills are always transferable across contexts (1922: 38, quoted below). (Dewey would not deny, of course, that some skills are non-transferable. Since habits use and transform materials from the external environment, individual habits sometimes have to be narrowly adapted to a limited field.) His point is to deny that it is an essential feature of skills as such that they always have a specific and clearly defined material or sphere of operation, or that they are always about a specific and clearly defined field. Nor do skills have a unique field. For instance, wood is not a material exclusive to carpentry, since it is used by other arts such as whittling, canoe-making and house-building. And although ballet has the body as its material, this material is shared by contemporary dance, step-dancing and taekwondo. Plato himself points out that medicine and gymnastics share the same material.16 So an art is not by its nature restricted to, or defined by, the sphere in which it is usually practised. Nor is an art’s sphere in all cases unique to itself and sufficient to differentiate it from other arts. What, then, differentiates a skill? For Dewey, an art is defined not by its end or its material, but by the way it proceeds. Ballet and taekwondo constitute radically different skills, differentiated not by their material but the way they use it, applying to it an ordered sequence of steps for bringing about its transformation. This is why it never occurs to Dewey to define a skill in terms of its end, or to insist that the activity which is characteristic of a skill always terminates in the attainment of the same ends.17 In short, it is not an essential feature of skill to have a single, unified end. A further feature of skill, for Dewey, is that skills do not necessarily have unique ends. This has to do with the fact that a given habit or skill is rarely sufficient to 16

17

Medicine and gymnastics, justice and legislation, ‘are concerned with the same thing’ (Gorg. 464c1– 5). This explains why sophists and orators are confused – because they ‘work in the same area and concern themselves with the same things’ (Gorg. 466c5). Dewey states this explicitly, writing that ‘habit is an acquired disposition to certain ways or modes of response, not to particular acts except as, under special conditions, these express a way of behaving’ (1922: 42). Dewey is not denying that people with a given skill react to situations in similar ways, or engage in similar act-types, but insists that a given skill does not fully predetermine the ends and purposes toward which a person acts in individual cases when they are applying the skill. This point is further explained below.

88

Skill in Ancient Ethics

attain the end at which it is directed. Although a particular skill may be perfectly adapted to the attainment of a given end, there are nonetheless many instances in which, due to a change in circumstances, obstacles arise to block the path of the agent, or the end simply becomes out of reach (1922: 34). For Dewey, the solution to these obstacles lies in the fact, which has significant implications for his theory, that one habit or skill comes to another’s assistance whenever it falters or encounters obstacles. For example, a woodworker may need to resort to using his skill in chemistry to bend a particularly unpliable piece of wood into shape, or to using his skill in metalworking to reinforce a wooden structure with metal parts. Indeed, some arts, such as navigation, seem to exist solely for the purpose of helping other arts to achieve their ends. Others, such as writing, on Aristotle’s account, come to the assistance of several other arts, finding use in business, running a household, political activity and learning (Pol. 1338a15). For Dewey, it would not be unusual for some arts to lack a fixed end of their own, and to exist primarily for the purposes of supplementing other skills. Such arts – including perhaps arithmetic, measuring and writing – may lack a unique end, or, if they have one, it may be a ‘trivial’ one that has no value in itself apart from the further ends it serves in particular contexts. Therefore, if an art lacked a unique end, or if it had only a trivial one, it would not be an exceptional, but a fairly typical, art in Dewey’s view. If we fail to recognize the extent to which skills supplement one another, Dewey would say, it is because of the way we group activities together as belonging to a single skill. Dewey, as we have seen, does not make the assumption that an art is unified and defined by a single end. Once freed from this assumption, it does not follow for him that what we commonly call ‘cooking’ is a single skill, unified by the end of ‘preparing food’, and not in fact a set of different skills, which are grouped together by convention because of their perceived relation to a common end. (Dewey would see the appeal to a common end as insufficient in many cases to provide clear criteria for distinguishing among different skills.) Dewey would therefore potentially view ‘cooking’ as the combination of several skills – including chopping, pickling, peeling and seasoning – that are used in tandem to build upon and support each other. Hence what we typically think of as a case of one person applying one skill is actually a case of one person applying several different skills in succession.18 The skill of 18

This accounts for a significant point of contrast between Dewey and Plato’s Socrates. While Plato’s Socrates tends to view each art as constituted by several ‘tricks of the trade’, and the knowledge of how to combine them to achieve the end, Dewey is more likely to consider a ‘trick of the trade’ as itself art. In the Phaedrus, for example, Socrates distinguishes between the various tricks used by doctors: being able to raise and lower someone’s temperature, move the bowels or induce vomiting. A person who knows these ‘tricks of the trade’ knows the preliminaries of medicine, but not medicine itself (268a5–65).

A Dewian Conception of Skill in the Platonic Dialogues

89

sauce-making tells the chef how to make a sauce with the ingredients to hand, and just when the dish begins to dry out, her knowledge of moisture-retention tells her how to correct the problem. Another skill ‘takes over’. Stated this way, only one skill seems to be active at a time: one skill seems to be ‘interrupted’ by a second skill. The picture that Dewey presents is, however, somewhat more fluid, in that the first skill does not cease before the second skill begins to make its presence felt (1922: 180). (This is why it is always a somewhat arbitrary affair to determine where one skill ends and another begins.) While acting on one habit, we feel other habits ‘coming to the surface’ (183). When the habit in use is in danger of failing to achieve an intended end, then other habits assert themselves. This brings us to a further reason Dewey would reject the idea that skills have unique ends: since a second skill must often be called in to finish the work that another skill has started, it cannot be asserted unproblematically that the first skill has a unique ability to achieve the end in question. In fact, we could not even say with certainty which of the two skills should be credited with the power to achieve a given end. Is it the chef ’s culinary training that makes the meal – or her skill at pairing colours and textures, which she perhaps acquired in her former life as an artist? Is it the lawyer’s legal expertise that wins the case – or his skill at winning the jury’s trust? What’s more, it becomes possible to conceive of an end that various skills can strive toward in their own way, about which one could say that they all reach the end in equal measure. Homeopathy, Chinese medicine and pharmacy all aim at the end of physical health, just as psychiatry, psychoanalysis, behavioural therapy and meditation aim at the end of mental health. Even granting that not all these skills achieve the end in question equally well, not every skill among them has an end it is better at attaining than the other skills. So just as a skill doesn’t necessarily have some end in particular, a skill does not necessarily have an end it is particularly good at achieving as compared with other arts. Finally, we can see now why it would be mistaken to think of a given skill as useful only when one is not employing another skill. Many skills, as we have seen, might be useful in helping other arts attain their own ends, and thus may be most useful precisely when other skills are in use. Indeed, Dewey explicitly warns against the danger of assuming that skills and habits are exercised in isolation from one another. If we were to assume that each skill pursues only its own end, by its own means, this would lead us to believe that different habits and skills are confined to different spheres of life. Dewey writes:

90

Skill in Ancient Ethics If each habit existed in an insulated compartment and operated without affecting or being affected by others . . . conduct would lack unity being only a juxtaposition of disconnected reactions to separated situations. But since environments overlap, since situations are continuous and those remote from one another contain like elements, a continuous modification of habits by one another is constantly going on. 1922: 38

A logical implication of the view that Dewey criticizes is that ‘matters of right and wrong, good and evil, obligation and responsibility, form a separate domain’ or realm of knowledge (1922: 88; cf. 1976b: 19–26; 1981: 88–89). Polemarchus, as we have seen, appears to subscribe to a version of this idea. A version of this idea also survives in the Platonic-Aristotelian claim that virtue is a kind of knowledge with some content – the Good – that is marked off from that of other forms of skill, a conclusion that Dewey resists.19

5. Applying the Dewian model to the Republic Applying Dewey’s conception of skill to the Republic, 333e1–334b2, a new perspective on the passage comes into view. Imagine that I am a bricklayer: as I put my skill to work, in order to achieve the aims of my art, it begins to dawn on me that I may fail to attain the ends of the business I have founded in partnership with another bricklayer. This may happen if I shirk my responsibilities and leave more work for my partner, or if I deal dishonestly with her. At this point, I ask myself whether there is a skill or habit of action that would allow me to attain my ends. The just course of action should now appear the most desirable course of action, because it is the best way to achieve my aims. Polemarchus should not therefore deny that the just person is ‘a better and more useful partner than a builder’; often, a just person may be ‘a better and more useful partner’ than the experienced bricklayer (333b3). So it first appears logical for Polemarchus to assume that, when I am using one skill to reach an end, the other arts are of less use or no use. But this assumption is not strictly correct, and it leads Polemarchus to the false conclusion

19

Plato and Aristotle embrace two different strands of thought on this issue. On the one hand, they recognize nominally that virtue is exercised in the same times and places as skills, insofar as they view virtue as ‘superordinate’ to skill. On the other hand, they also assert that virtue and skill occupy distinct and mutually exclusive domains of knowledge: medicine knows the body and health, ‘and not anything that lies outside of it’ (Charm. 171b2), whereas virtue knows the Good (174b10).

A Dewian Conception of Skill in the Platonic Dialogues

91

that when I am working toward the ends promoted by special skills, I need only those skills and not justice. Polemarchus would have arrived at an entirely different conclusion had he held a view of skills as mutually supplementary, which would allow him to better accommodate skills that typically come into use when other skills are being used. (We do not yet say, as advocates of second-order skills do, that these skills represent a special ‘class’ of skills, since our point has been precisely that it is not a necessary characteristic of skills to have single and unique ends.)20 Hence Polemarchus would have recognized, too, that the virtues are particularly clear examples of such skills, and that in consequence of their status as arts that usually ‘supervene’ on other arts, arts upon which they remain parasitic, they seem to lack, more than their host arts, an end of their own.21 First, it is fruitless to ask about the specialized end of such an art, since the instrumental value of the art can be articulated only with reference to the ends of the host art. Second, if the art has a specialized end, then the likelihood is that this end will not have independent value, but value only in relation to the ends of other arts, in the context of whose use its value first appears. So contra what Socrates states, it is not when other arts are not being used, that the virtues are; the virtues are used precisely when one is busy with the materials and instruments that belong to other arts, ‘when it is necessary to use them’ (hotan de chrêsthai: 333d6). It might be argued that there is a disanalogy between virtues and skills, because skills allow one to reason only about the means to a given end, whereas virtues allow one to reason about the value of particular ends. But for Dewey, this distinction is artificial because habits and skills allow one to contemplate the value of certain ends in particular contexts and make deliberation possible: ‘habits are the means of knowledge’ (1922: 176). This is owing to the fact that a skill, once possessed, is such that one can imagine what will happen when it is put into active use. Indeed, if one begins to imagine what will happen when a given skill is put to use, one is already in a certain sense ‘using’ that skill, if only in one’s imagination. This is especially true of the skills we call virtues, and it explains why virtues are likely to lack single and unique ends. 20

21

Contrast this with Sprague’s view. Virtues such as justice do not constitute, as she argues, an exceptional class of skills (because they lack an identifiable product), but typical, even paradigmatic, examples of skill. The difference between these views is important because, while Sprague’s exaggerates the discontinuities between virtue and skill, the view defended here assimilates virtue to skill, thus making the analogy with skill more, rather than less, plausible (1976: xvi–xvii). Here the distinction between host and parasitical arts suggests comparison with the distinction between first- and second-order crafts drawn by Sprague. The former, such as cobbling and carpentry, have an identifiable product, while the latter, such as sophistry and statecraft, have no identifiable product (1976: xvi). However, unlike the distinction between first- and second-order crafts, the distinction between host and parasitical arts is not fixed, nor is it ‘hierarchical in nature’ (xvi).

92

Skill in Ancient Ethics

Dewey asks us to imagine the situation from the perspective of the agent who contemplates putting a given skill to use. The less immediate effects of her action are of course incalculable – ‘vast, immeasurable, unthinkable’ (1922: 263) – but the wider view she takes of the possible consequences of her action, the more likely we are to say that she is ‘moral’. When, in such cases, the agent senses that her actions promote a great number of ends, which she is nonetheless unable to calculate with exactitude, we say that the course of action she contemplates participates in an ‘ideal’, such as the ideal of justice or of courage. Precisely for this reason, Dewey stresses that, technically speaking, ‘This ideal is not a goal to be attained. It is a significance to be felt, appreciated’ (263). Accordingly, when a skill or habit of action is put into practice, which participates in the ideal, it may be considered, not just a skill, but a ‘virtue’ (1976b: 267). The only significant difference between a skill and a virtue, for Dewey, is that an agent employs a virtue whenever she puts into use a skill or habit of action that allows her to attain her present end as well as other, more remote ends – including, of course, ends pursued by other arts. Hence, for Dewey, it is a contradiction in terms to suppose that a virtue has a single end: by definition, a virtue has more than one end. For Dewey, this would explain the fact that Socrates’ interlocutors are unable to narrow down to one end the many ends promoted by a particular virtue. Here, it is important to note the adverbial sense in which we speak of the virtues. The virtues tell me, not simply how to reach the ends of more basic, menial skills, but how to achieve them in a way that promotes other, sometimes more important ends. They tell me how to reach them in a just way, or justly. Invoking an analogy between justice and health that will be familiar to readers of the Republic, Dewey puts it as follows: ‘To say that man seeks health or justice is only to say that he seeks to live healthily or justly’ – that is, with a view to the aggregate of the goods he pursues over the course of his life (1957: 167). In other words, although we may refer to a good like that pursued by justice or medicine as though it were single and unified, it is in fact multiple and diffuse. And although we may refer to it as though it were distinct from other goods, it ‘cannot be set up as a separate and independent good’, especially not a good defined negatively relative to the goods pursued by other skills (1922: 168). Hence, if we picture, for a moment, the road not taken, we can then imagine Polemarchus responding differently to Socrates’ query, when he asks exactly when and where justice is most useful: we can imagine him responding that justice is useful whenever one pursues any end. For justice promotes many ends. Yes, many arts and skills promote the same ends, in some cases more efficiently, but justice achieves whatever end it pursues in a way that promotes many other ends as well.

A Dewian Conception of Skill in the Platonic Dialogues

93

6. Conclusions The foregoing has shown that the skill analogy does not have to be rejected. It is possible to retain it without the negative consequences it would seem to imply. Specifically, we can retain the idea that skill is purposeful and involves meansends reasoning, while rejecting the idea that a given skill is always directed at one and the same unique end. Indeed, when this false set of premises is rejected, the result is a Dewian conception of technê. The aporia is resolved. And virtue appears to be both inseparable from the exercise of various ‘lower-level’ technai, and itself technê-like. At least, there appears to be no obvious disanalogy between virtue and technê. The foregoing does not, of course, address the fraught question of whether Socrates, or Plato, ultimately agrees with some version of the popular account of technê. That question lies outside the scope of the present chapter. However, it may be suggested that any attempt to answer this question must take into account the fact that, in Republic Book I, the popular view is both Socrates’ sin and his salvation. True, it leads him and Polemarchus to an impasse. But Socrates also refutes Thrasymachus by appealing to the claim that every art has a specific and unique end, insofar as every art has an object that it is uniquely qualified to order, so that the object—for example, the city or the soul—performs its essential functions (342e7). This suggests that Socrates attempts to maintain some version of the popular view. After all, it allows him to argue that justice constitutes the means by which souls and cities perform their functions, or in his words, that constitutes a dunamis, or ‘power’, with the single and unique function of ‘produc[ing] men and cities of the sort described’ (443b3).

References Annas, J. (1981), An Introduction to Plato’s Republic, New York: Oxford University Press. Balaban, O. (2007), ‘The meaning of craft (τεχνη) in Plato’s early philosophy’, Archiv für Begriffsgeschichte 49: 7–30. Benson, H. (2000), Socratic Wisdom, Oxford: Oxford University Press. Dewey, J. (1922), Human Nature and Conduct, New York: Henry Holt. Dewey, J. (1957), Reconstruction in Philosophy, Boston: Beacon Press. Dewey, J. (1967), The Early Works, 1882–1898, edited by J.A. Boydston, Carbondale: Southern Illinois University Press. Dewey, J. (1976), The Middle Works, 1899–1924, edited by J.A. Boydston, Carbondale: Southern Illinois University Press.

94

Skill in Ancient Ethics

Dewey, J. (1976a), ‘Valuation and experimental knowledge’, in J.A. Boydston (ed.), The Middle Works, 1899–1924, Carbondale: Southern Illinois University Press. Dewey, J. (1976b), ‘Moral principles in education’, in J.A. Boydston (ed.), The Middle Works, 1899–1924, Carbondale: Southern Illinois University Press. Dewey, J. (1981), The Later Works, 1925–1953, edited by J.A. Boydston, Carbondale: Southern Illinois University Press. Dorter, K. (2005), The Transformation of Plato’s Republic, Lanham, MD : Lexington. Foley, R. (2009), ‘The better part of valor: The role of wisdom in Plato’s Laches’, History of Philosophy Quarterly 26 (3): 213–233. Gould, C.S. (1987), ‘Socratic intellectualism and the problem of courage: an interpretation of Plato’s Laches’, History of Philosophy Quarterly 4 (3): 265–279. Grube, G.M.A., and C.D.C. Reeve (1992), The Republic, Indianapolis: Hackett Publishing. Irwin, T. (1977), Plato’s Moral Theory: The Early and Middle Dialogues, Oxford: Clarendon Press. Irwin, T. (1995), Plato’s Ethics, Oxford: Oxford University Press. Nussbaum, M. (1986), The Fragility of Goodness, Cambridge: Cambridge University Press. Page, C. (1990), ‘The unjust treatment of Polemarchus’, History of Philosophy Quarterly 7 (3): 243–267. Pappas, N. (1995), Routledge Guidebook to Plato’s Republic, London: Routledge. Penner, T. (2006), ‘The forms and the sciences in Socrates and Plato’, in H. Benson (ed.), A Companion to Plato, 165–183, Oxford: Blackwell. Plato (1914) Euthyphro, Apology, Crito, Phaedo, Phaedrus, Vol. 1. of Plato, trans. H.N. Fowler, Cambridge, MA : Harvard University Press. Plato (1924), Laches, Protagoras, Meno, Euthydemus, Vol. 2 of Plato, trans. W.R.M. Lamb, Cambridge, MA : Harvard University Press. Plato (1925), Lysis, Symposium, Gorgias, Vol. 3 of Plato, trans. W.R.M. Lamb, Cambridge, MA : Harvard University Press. Plato (1927), Charmides, Alcibiades I & 2, Hipparchus, The Lovers, Theages, Minos, Epinomis, Vol. 11 of Plato, trans. W.R.M. Lamb, Cambridge MA : Harvard University Press. Plato (1981), Five Dialogues: Euthyphro, Apology, Crito, Meno, Phaedo, trans. M.A. Grube, Indianapolis, IN : Hackett Publishing. Reeve, C.D.C (1988), Philosopher Kings: The Argument of Plato’s Republic, Princeton: Princeton University Press. Roochnik, D. (1996), Of Art and Wisdom: Plato’s Understanding of Techne, University Park, PA : Pennsylvania State University Press. Schmid, W.T. (1998), Plato’s Charmides and the Socratic Ideal of Rationality, Albany, NY: State University of New York Press. Sprague, R.K. (1976), Plato’s Philosopher-King, Columbia, SC : University of South Carolina Press. Vlastos, G. (1991), ‘Happiness and virtue in Socrates’ moral theory’, in Socrates, Ironist and Moral Philosopher, 200–232, Ithaca, NY: Cornell University Press.

Part Two

Skill in Aristotle’s Ethics

95

96

5

Steering Against the Bad: An Aristotelian Account of Virtue as Two-way1 Jennifer Rothschild

One concern for philosophers who find interpretative appeal in Aristotle’s virtue-skill analogy is understanding the analogy’s limits. Aristotle mentions several ways in which skill and virtue are disanalagous, among them his claim about the kind of power exercised in the excellence of each. He argues that skill is a two-way power, meaning the person of expertise can deploy her knowledge either toward the constitutive end of the skill or away from it. Virtue, by contrast, does not work in the same way; the virtuous person as such is aimed directly at good ends. Often the powers disanalogy is taken as evidence that virtue is divorced from the two-wayness of skill, and is instead a one-way power. In this chapter I argue that though the powers disanalogy is importantly right, it is right in a limited way: virtue is not a two-way power in all the ways skill is. I do argue, however, that virtue shares features of the rational order of skill that are best understood as exhibiting two-wayness. Here I take two-wayness to be an irreducible feature of a set of possibilities for action organized against the bad and toward the good. Accepting the two-wayness of virtue, I aim to show, is not only plausible but necessarily Aristotelian: it strengthens the virtue-skill analogy at its core, and unites the analogy with Aristotle’s account of virtue, practical reasoning and practical knowledge more broadly. On my reading, the virtuous person acts virtuously by steering against the bad, and this means that virtue is compatible with, and even depends upon, the thought that the virtuous person both knows the bad and also makes use of that knowledge when she deliberates with a view to acting well.

1

Earlier versions of this paper were presented to audiences at the University of Cape Town (2017), the University of Florida (2018) and the Annual Meeting of the Florida Philosophical Association (2018). My thanks to all involved for the valuable feedback.

97

98

Skill in Ancient Ethics

1. Limiting the powers disanalogy between skill and virtue According to Aristotle’s Nicomachean Ethics, practical virtue and skill are analogous, so that much of what it means to develop and have practical virtue is visible in the person of skilled expertise. For example, the virtuous person and the skilled person both learn to be good at what they do by doing it, gradually and through practice (Aristotle 2002: 111).2 They both aim for the mean, avoiding excess and deficiency (Aristotle 2002: 117).3 In both skill and virtue, an agent’s progress and achievement depend on her tracking the limited ways of getting it right against the very many ways of getting it wrong (Aristotle 2002: 117).4 Both those with virtue and those with expertise count as knowers of what they do well (Aristotle 2002: 115).5 The fruitfulness of the virtue-skill analogy has been well recognized in recent neo-Aristotelian scholarship, provoked in particular by Julia Annas’s Intelligent Virtue (2011) and responses to it. Of course, as Aristotle himself recognizes, the virtue-skill analogy has its limits, and crucial features of the two are not shared. Skilled action aims at an end of production, whereas virtuous action contains its own end (Aristotle 2002: 180).6 Practical virtue is much more demanding than skill in scope and structure: the intellectual virtue of practical wisdom depends upon unity with other, substantially non-rational virtues of character (Aristotle 2002: 188–189);7 practical knowledge of expertise does not depend on desire in the same way (Aristotle 2002: 114–115).8 Further, while skilled expertise is exercised in some domain of knowledge and action, the practically wise person has a more holistic practical knowledge and acts well across domains; such a person can deliberate about ‘what sorts of things conduce to the good life in general’ (Aristotle 2002: 180).9 For the purposes of this chapter, I accept the above points of analogy and disanalogy between virtue and skill, as well as most of what could be added to those lists to fill them out. Where I wish to raise concern is with a further point of disanalogy, one that is important to both Aristotle and those working in his virtue ethical tradition. The concerning point of disanalogy is the understanding 2 3 4 5 6

7 8 9

NE II.1 1103b1 ff. NE II.6 1106b11–16. NE II.6 1106b31. NE II.4 1105a31. NE VI.5 1140b6–8. Strictly speaking, practical virtue contains its end in a more limited way than does contemplation. See NE X.7 1177b1 and b20 (Aristotle 2002: 251). NE VI.12 1144b1 and VI.13 1144b32 ff. NE II.4 1105a30–b5. NE VI.5 1140a27–29.

An Aristotelian Account of Virtue as Two-way

99

of skill as a two-way power, by contrast with virtue, which might be cast alternatively as a one-way power.10 This point is concerning because, if we are not careful with it, it risks undermining the analogy between skill and virtue at just those points worth preserving, and with it elements of Aristotle’s core understanding of virtue. Aristotle addresses the core of the powers disanalogy in Nicomachean Ethics VI.5. He writes: ‘[T]here is such a thing as excellence in technical expertise, not in [practical] wisdom; and with technical expertise it is more desirable if someone voluntarily gets something wrong, whereas with wisdom, as with the [character] excellences, it is less so’ (Aristotle 2002: 180).11 Here Aristotle argues that the excellence of skill can be exercised in two ways: its bearer can count as expert while using her skill to act purposefully either for or against the constitutive end of the skill. For example, the expert doctor has a developed ability to promote the good of health. This means that as a bearer of the skill she has, she can knowingly act for the health of her patient or against the health of her patient, toward the good at which her skill aims or against that good. Acting against that good, NE VI.5 says, will not undermine her having acted skilfully so long as she does it on purpose. By contrast, if the agent makes a mistake in action, this would count against her status as expert, since acting mistakenly is a mark of not knowing what one is doing. Practical virtue does not share this two-wayness of skill. No agent can be virtue’s bearer while exercising it, alternately, both for its end and purposely against it. Rather, practical wisdom is not something one can have and deploy independently of pursuing the good end. The good end is built into the having of practical wisdom itself, meaning the person of practical wisdom just will act well (and not badly). Another way of putting this is that a person’s knowingly acting against virtue always diminishes her status as virtuous. Unlike in skill, which is better preserved by the purposeful pursuit of bad ends than by mistakes in execution, making a mistake is more compatible with virtue than is acting badly voluntarily. Accidentally saying something unkind to someone is a less significant stain on one’s virtue than is doing it on purpose. The above presentation of the powers disanalogy from NE VI.5 is not, I think, concerning so long as we are conservative in our assessment of its scope. It is worth highlighting the limited sense in which skill’s two-wayness is disanalogous 10

11

A more complete discussion of one- and two-way powers would rely on a treatment of Aristotle’s Metaphysics, and is beyond the scope of this chapter (see Aristotle 1998 and Angier 2010: ch. 2). NE VI.5 1140b21–24. Aristotle also contrasts practical wisdom with the two-wayness of cleverness at NE VII.12 1144a24–30 (Aristotle 2002: 187–8).

100

Skill in Ancient Ethics

with virtue. Practical wisdom cannot, as skilled expertise can, be intentionally exercised against doing what is good and remain intact as practical wisdom. I call this reading of the disanalogy ‘limited’, though, because it insists only that some aspect of the two-wayness of skill is not available to virtue. Nothing so far has shown that the entirety of skill’s two-wayness is unavailable to virtue if it should turn out, as I will argue, that two-wayness is best understood as broader than the possibility of maintaining excellence while acting toward its contrary. Suppose we accept virtue does not share the full two-wayness of skill. One option is to consider practical wisdom a one-way as opposed to a two-way power. A one-way power is a power, which, when a subject has it, that subject just will behave in such-and-such a way under the right conditions. Will Small highlights three categories of one-way powers (see Small 2021). The first is non-vital powers, such as the fragility of glass. This fragility makes it the case that glass will break, for example, in a normal case of swinging a hammer at it. The second category is merely vital powers, such as the Arctic poppy’s power to turn toward the sun, which will happen if the sun shines and nothing interferes. The third category of one-way powers he calls merely volitional powers, which he defines as ‘distinguished from merely vital powers by the fact that the desire or choice of the subject of the powers enters into the explanation of its manifestation’ (Small 2021): as, for example, when a horse is motivated by its thirst to drink. The three categories share the feature of doing the thing in question under the right conditions. If the hammer swings, the glass breaks. If the sun shines, the poppy will turn. If the water is there and the horse is thirsty, the horse will drink the water. Practical wisdom looks, in a non-trivial way, like these one-way powers. It shares the feature of reliably issuing in particular outcomes – in this case, intentional actions – when the right circumstances obtain. The good person will speak against injustice when the situation calls for it, will give generously on the right occasions, and will display her courage when facing her fear is the appropriate thing to do. If an agent is truly practically wise, she just will act well and not badly. Accepting this is accepting the other face of (either version of) the powers disanalogy: crediting an agent with practical wisdom is to credit her with having achieved a condition already in active pursuit of the good. If she has it, she acts well. We need to be careful here, however. The three categories of one-way powers, increasingly sophisticated though they are, are all non-rational responses. None of them rise to the level of excellent rational activity – deliberation and decision

An Aristotelian Account of Virtue as Two-way

101

and knowledge about how to do what one chooses to do. Acting virtuously is, however, a kind of practically rational activity: it is necessarily directed by the rational human agent, and it is necessarily directed well by the knowledgeable human agent. This gives us some reason to think that there is a break in kind between the rational power at work in virtue and other one-way powers. By contrast with glass and poppies and horses, there is something different at work in virtue, where the rational operations of the power do not just explain, but also organize, the carrying-off of the action as the action it is. So though it may be true that practical wisdom itself is only deployable one-way (for the good), it might also be true that excellent practically rational activity – the sort of knowledge-involving activity undergirding practical wisdom – cannot be adequately captured by thinking of it alongside one-way powers. We may do well to consider whether, if we approach virtue as exhausted by one-wayness, we might just be obscuring the rich, rational, structure that explains and issues in acting well. One possible route forward from here is to stick to characterizing practical virtue as a one-way power, but to try to carve out a fourth category of one-way powers, those which are rational. Here I will leave this possibility aside and instead follow Small’s insight that two-wayness itself highlights the structure of practical rationality (Small 2021), though where he explores this thought as it applies to skill, I will be more focused on how it illuminates virtue. So far I have been concerned to show that we should not rest comfortably with a strong reading of the powers disanalogy between skill and virtue, one that insists that virtue’s power is exhausted by its one-wayness in direct contrast with skill’s two-wayness. To be sure, the powers disanalogy holds within limits: virtue does not have the flexibility to pursue contraries without diminishing the agent’s accomplishment as virtuous, and is not in that sense two-way; also, virtue looks like other one-way powers in that it reliably generates ends in the right circumstances. I have suggested, however, and I will argue this further below, that important features of the rational structure of virtuous activity go missing if the disanalogy becomes too expansive and we read virtue as strongly one-way. A limited reading of the powers disanalogy leaves room to consider a version of two-wayness involved in knowingly acting virtuously. My aim in what follows is to show that two-wayness is visible not only in the potential pursuit of contraries, but also in the very rational order virtue displays analogously with skill, the very order that allows us to think of virtue as we ought to, as a matter of knowledge, deliberation and choice.

102

Skill in Ancient Ethics

2. Acting knowingly and for the good In section 1, I argued that we should accept a limited version of the claim that virtue is prohibited from the operations of two-way powers. I further suggested that our reason for limiting the disanalogy in this way ultimately has to do with opening the door to seeing the two-wayness involved in navigating virtue’s rational order. I will explore this suggestion beginning in section 3. First, however, we need to attend to the threat posed to Aristotelian deliberation, choice and practical knowledge by denying two-wayness to virtue. In this section, I argue that one-wayness does not give us a clear path to properly crediting the virtuous person with deliberating well and knowingly choosing what is good. Whereas two-wayness treats the possibilities for action as a rationally related set to be navigated by the practical knower, one-wayness merely highlights the thing done as having been the likely possibility to appear in action. Aristotle says in NE VI.5 that one hallmark of practical rationality is making prescriptions concerning ‘what is in the sphere of action and can be otherwise’ (Aristotle 2002: 180).12 A few lines earlier, he makes part of this same point in an even stronger form: ‘Now nobody deliberates about things that cannot be otherwise’ (Aristotle 2002: 180).13 Here he divides the truths of practical reason from truths of another kind, truths which may be appropriate objects of theoretical reason but are not appropriate for deliberation. Whereas one can contemplate an eternal truth, it makes no sense to deliberate with a view to bringing it about. The objects of deliberation, by contrast – those that appear in the sphere of action and decision – are multiple and variable, as are the ways of pursuing them. Some of those possibilities make it to action, because an agent decides to pursue them, and some do not.14 Of the many ways the pursuit of an end might proceed, deliberation sorts the possibilities, decision settles on one, and the agent acts. On the model of virtue as a one-way power, deliberation and decision about the possibilities for action are at best obscure, and at worst absent. On this understanding of the virtuous person, decision in favour of one possibility – the true, rational prescription – predictably emerges and issues in action. And though there may be many possibilities for how someone might act differently, it 12 13 14

NE VI.5 1140b3. NE VI.5 1140a33. Aristotelians disagree about exactly what role practical reason plays in setting the end of action (see, for example, Callard 2020; Irwin 1975; Moss 2011). But Aristotle is clear, at least, that deliberation is responsible for figuring out how to pursue an end. See NE III.3 1112b12 and VI.12 1144a7 (Aristotle 2002: 128, 187).

An Aristotelian Account of Virtue as Two-way

103

is not clear that the virtuous person engages any of these alternatives, or has taken any of them under consideration as part of her decision. At most, the possibilities not chosen feature in the virtuous person as possibilities left in the background. Moreover, on this picture, we have no sense that the various possibilities, the chosen and the not-chosen, have anything to do with one another – no sense that considering them together might illuminate them as a related set, that working through them or deliberating about them might help one decide how to act, or that the discarded possibilities do any work for the good agent in supporting the eventual decision. The discarded here are silent, left to float free of both one another and the one chosen. This picture leaves mysterious why the virtuous person chooses as she does, or why one possibility emerges over any other in the action of the agent who is getting something right. This is unsatisfying, because we cannot see why we should give our virtuous person credit for acting virtuously. We can stipulate that she does it, but we cannot give her rational credit for her virtue unless we can count her as a knower. An agent’s having knowledge in a domain of action means she has substantial understanding of what needs to be done, and how to do it in order to pull off purposeful, deliberate action in that domain. She can settle on appropriate ends of action and knows how reliably to aim for her target.15 This is to say that when we say an agent acts knowingly, we credit her with having understood what she was up to in her action. Notice, too, that whenever we credit a person with substantial practical knowledge, we are already saying that an agent is acting in service of some good. Whereas someone with less understanding of action might have gone for a worse alternative, the person who acts well goes for what is better. The expert baker does not finish with something that looks and tastes roughly like a cake, but knowingly makes an excellent cake, both attractive and delicious. The generous person does not just give a gift, but gives a good one, and she gives it well. Crediting an agent with practical excellence just is crediting her with having knowingly acted for what she took to be the best end, (which in the case of skill could be for or against the end of the skill).16 The excellent agent must be a knower, one who chooses the possibility she does because it is good, 15

16

We might want to mention all sorts of caveats here: while practical knowledge is fundamentally tied to doing, in cases of skill, it can be worth crediting in the teacher who can help others do what she cannot. The good doing of the action (and in some cases, as with virtue, the perfection of the knowledge itself) can depend on all sorts of things beyond the operations of knowledge: e.g. particular bodily movements, or configurations of desires; practical knowledge is worth crediting in all sorts of cases in which the desired outcome never appears; and so on. Note that this does not necessarily mean that acting against the constitutive end of the skill is to act viciously, by contrast with virtue, in which acting against is always acting viciously. I am indebted to Nathan Rothschild for conversation on this point.

104

Skill in Ancient Ethics

and because she knows how to act for the good. In addition, this is visible by degree in those who are only somewhat knowledgeable. On the unsatisfying picture above, where many possibilities for action rest isolated from one another and one just emerges in the virtuous person’s choice, our person of virtue looks less like a knower and more like someone engaging in a little bit of magic.17 Why does she choose as she does? If the possibilities are discrete, why are some better, or best? How does the way she reasons count as excellent deliberation if the process endorses one possibility from the start? We can insist she chooses one because she just knows, arguing that it only looks like magic and not knowledge from the perspective of the person who is not virtuous. But surely this does not resolve the dissatisfaction. Moreover, even if we nonvirtuous people cannot deliberate as the virtuous person does, surely we can expect that what she does should be roughly recognizable to us as rather like what we do in our own cases, cases in which we do pretty well but fall short of excellence. In order to make sense of the virtuous person as knower and agent (who is like us, as far as we can see, only better), we need to be able to look to the virtuous person and say something about the kind of reasoning she must be engaged in as part of being the knower and agent she is, so that that very reasoning helps explain why she is getting it right. Much of the apparent mystery here can be avoided if we push beyond virtue’s one-wayness and treat virtuous action not as a likely outcome but as a rational process. This means starting to think about the genuine possibilities for action not as discrete, but as having something to do with one another. It means treating the work of deliberation as actively involving multiple possibilities, including ones which will ultimately be discarded by decision. With the possibilities so related, reasoning comes into view as a choosing of this one over these, as a process of rejecting these two in favour of this other, as a process which, when performed excellently, supports calling some of its possibilities better than others, as well as calling the one who gets it right a knower. In other words, if we open space beyond one-wayness, we can come to see not only that the virtuous person must somehow have chosen some possibility which is good; we can also animate her ability to approach and navigate the field of possibilities such that the chosen one was chosen well. The virtuous person knows what she is doing 17

Characterizations of the virtuous person as the mysterious and opaque ‘knower’ appear in all kinds of criticisms of Aristotelianism, e.g. situationist critiques; misreadings of ‘silencing’ from McDowell’s ‘Virtue and Reason’ (1979); conversations organized around the thought that the deliberating agent should ask herself what the phronimos would do; and views in which the continent person looks appealing by comparison to the virtuous person, since at least we can see her struggling to do what is right.

An Aristotelian Account of Virtue as Two-way

105

when her understanding is supported by a rational order of better and worse possibilities that she has understood as a set and sharpened into a choice.

3. Constraining the scope of possibilities for action I turn now to the question of what it means to say that the possibilities for action are constituted as a rational order of better and worse. In the remaining sections of the chapter I take up this challenge. This section considers how an agent’s reasoning organizes her possibilities for action, constraining the good possibilities away from the bad. Section 4 argues that the good and the bad should be understood together in a two-way rational order. The only way to make sense of reasoning toward the good is to see the good as emerging against determinate bads, and this will turn out to mean that virtue involves a great deal of knowing about acting badly. Section 5 tracks the way in which the virtuous agent’s deliberations actively involve consideration of what will turn out to be bad possibilities. In all three sections, the operations of virtue look analogous to those of skill. Virtuous and skilled agents use what they know to rationally organize and constrain their possibilities for action. In this section I outline three kinds of constraint. In all three, constraints are anchored in the world, since each is getting onto something true that imposes itself on what it would be good to do. But the more important issue is that each of the constraints works because of the place it occupies in the reasoning of the agent. Constraints highlight the way the agent activates what she knows in order to limit the possibilities to those worth considering. The kinds of constraint I will consider here are the following. Constraints of circumstance limit the available actions based on the agent’s recognition of the circumstances she is in. Consider a case of skill and also one of virtue. Suppose I am a baker, and I know that whenever I am using an oven that cooks hotter than the temperature gauge indicates, I need to either set a lower temperature than prescribed or take my cake out of the oven earlier than planned. In the right circumstances, I would need to bring this knowledge of times and temperatures to bear on my actions. But I also know that the oven I am currently using to make my cake is properly calibrated, so my knowledge of how to adjust for a misaligned gauge will not now be a relevant consideration in my action. Or suppose instead that I am trying to do something generous for a friend, and I would like to take her out for her birthday. My town, however, is quite closed up

106

Skill in Ancient Ethics

at the moment due to a pandemic, and many of the things we would normally do together are not available in the usual way. If a joint activity is to be an option, I will need to be creative about how to construct the possibilities around the circumstances. Domain constraints limit the possibilities in cases of the exercise of skill. If I am a baker, I may also be quite good at gardening and may know how to drive a bulldozer, but neither gardening nor bulldozing are appropriate in pursuit of cake-baking. These other actions simply do not belong to the baker to know, and so are irrelevant to my knowingly acting as a baker (as any baker would know). Domain constraints do not apply in the same way in the case of virtue, since the domain of virtue is the good life as a whole. A third kind of constraint worth mentioning is constraints of virtue. Virtuous actions are constrained by all the virtues, even those virtues that are not advertised in the lead considerations of the action at hand. Suppose I am working from a recipe to bake a cake, and I reach a point at which the recipe calls for an ingredient I do not have. I will not break into my neighbour’s house and steal ingredients from her refrigerator to finish my cake, not because of the kind of baker I aspire to be, but because of the kind of person I aspire to be. I will not, using generosity as my excuse, try to take my friend on an adventure prohibited by the current quarantine ordinances. To do either would be to act against justice, and unjust actions are unavailable ways of going forward and acting virtuously. My acting generously is, in other words, constrained by my acting justly. Insofar as I know virtue, I will be able to use what I know of justice to limit the possibilities for knowingly acting generously.18 The constraints above are not meant to be an exhaustive list.19 They are, rather, an illustration of something important about how the set of good possibilities for action is constituted. The agent is actively, rationally involved in the work of constraining, and through constraining, the possibilities for action become available relative to the worse options eliminated. The better the agent is – the more deserving she is of credit for skill or virtue – the better she will be at constraint. Of course some cases are easy ones, demanding little of the agent: obviously it does not take deep knowledge of baking to see that bulldozing should not enter into the deliberations of cake makers. But moderately robust

18

19

Constraints of virtue in the way I mean them depend upon Aristotle’s claim that the virtues are unified, best seen in NE VI.12 and 13 (Aristotle 2002: 188–9). Further, the constraints are not meant to articulate the explicit thoughts of the agent. Practical reasoning is a process of which an agent may have some awareness at times, though awareness is not necessary, and even when it happens, it is likely incomplete.

An Aristotelian Account of Virtue as Two-way

107

practical knowledge is necessary in order to constrain the possibilities at the centre of the action. It may not be clear to me, mediocre baker that I am, exactly the circumstances I am in, whether I am blocked from making my cake or if can just carry on with my recipe without the ingredient I need. The more knowledgeable baker, however, will be in a better position to judge whether there are good ways of making up for the ingredient left out that will mitigate its loss. Or she may think the constraints of the case mean she should consider changing course and making an entirely different cake than the one she planned. We can expect hard cases in the work of constraint as well, in which even the best agents operate at the margins of what they know. If our agent is a virtuous person but acting in imperfect or unfamiliar circumstances, it may be unclear whether justice prohibits some possibility, or she may have a sense that justice is at issue but be unclear on exactly how it should shape her considerations. In such a case, the virtuous person may need to cast a wider net in her efforts to constrain the possibilities for action. The upshot here is that the good agent’s practical knowledge is now connected to at least one way she uses what she knows to rationally organize the possibilities for action. She constitutes the set of possibilities for good action by navigating obstacles, conflicts, considerations of relevance and so on, actively constraining away the bad options from those which, at least for now, appear to be well aimed at her end. She deliberates in this way whether she is a knower of virtue or of skill, and whether she is an expert or just an agent who is pretty good at what she does, but not quite excellent.

4. The two-wayness of virtue’s rational order Section 3 argued that in both virtue and skill, practical reasoning picks out the set of possibilities for action by sorting those with potential from those worth discarding. In section 4 I aim to show that even this initial rational sorting is already indicative of a two-wayness in practical knowledge as such, including in virtue. (This two-wayness will take on further developments outlined in section five.) Tracking the two-wayness of virtue starts from the thought that the rational sorting above aims at dividing the possibilities into two: the good and the bad, or perhaps the better and the worse, the potentially worthy and the unworthy. Importantly, the good and bad here are not only separated but also joined together in a rational order. The separation between them emerges by way of and

108

Skill in Ancient Ethics

is held together by contrast: it is not just the good and the bad, but the good rather than the bad. The better comes into view and stays in view in comparison with the worse (and vice versa). Rational choice of the best option requires such an order, for it is not choice at all if it could have gone no other way, and it is not choice of the better or best if it is not chosen in such a way as to demote the worthiness of the not-chosen. Here we have in plain view what I have been elaborating in sections 2 and 3: the possibilities for action are not free-standing, but related to one another, and the relation at work is a kind of rational order that both stands behind our ability to credit an agent as knower and is to a large extent constituted by the operations of the knower’s excellent practical reasoning. Among what the person of virtue understands in understanding how to act well, then, is how to organize the possibilities for action toward the good, steering away from the bad. This means, as the limited reading of the powers analogy suggests, excellence moves in one direction. But the rational order itself, and our virtuous person’s efforts at navigating it, exhibit a kind of two-wayness reminiscent of skill: practical reason engages in understanding both bad and good, each as a way of understanding the other. In virtue, the operation may be asymmetrical in its orientation toward the good (in skill, toward the constitutive end of the skill), but it is still importantly two-way. This two-wayness becomes clearer still when we acknowledge the following: in order to operate the rational order well, the virtuous person will need to know quite a lot about acting badly. I think we should be willing to go so far as to say that the virtuous person may, in some sense, know more about doing bad than does the bad agent herself. Consider a case of Socrates and a torturer, where Socrates stands in as virtuous agent and the torturer is very good at torturing. If we compare Socrates’ knowledge of torture to the torturer’s knowledge of it, there is an important way in which Socrates knows more than the torturer. Of course this is not to say that Socrates has experience with torturing or that he would consider torturing, or even that he has spent any time thinking specifically about torture. Of course, too, there are things the torturer knows because they belong to the domain of torturing, such as how to use various instruments and procedures of torture to inflict maximum pain.20 Socrates probably has no knowledge of such things. But he does know how to live well as a human being, and he knows this far better than the torturer ever will. With his knowledge, we

20

Here I do not suggest that torture is a skill in the strictest sense of techne, though it does enjoy many of the features of one.

An Aristotelian Account of Virtue as Two-way

109

might say Socrates has far better knowledge of what truly harms a person – what hurts us, brings us to the brink, destroys our souls. Moreover, Socrates’ knowledge of this harm is vast, broader than the torturer’s knowledge of torture: Socrates knows better how human beings can be harmed by physical pain or psychic disruption or by threats to friendships or other social goods. His sense of how to live well is set off against all the ways in which one could fail to do so. Two further points about the virtuous person’s knowledge of the bad are worth making here. First, it is important to see Socrates’ knowledge as practical (as opposed to theoretical) knowledge. What Socrates knows about how to live badly is relevant to action. What he knows about torture would be useful to the torturer: if the torturer could somehow force Socrates to give advice about what would most harm a person, she could become a better torturer. Second, it is worth highlighting that Socrates has some very determinate practical knowledge of the bad.21 When we say that reasoning against the bad is active in his deliberations, we do not mean merely that he is actively concerned not to do what is the opposite of good – not to act unkindly or unjustly in general, for example – but that he has practical, determinate knowledge of what not to do. We mean that he knows this way of doing something is unkind or unjust. It is his determinate understanding of what not to do that does most to constrain the possibilities for going forward. (In cases in which the good is particularly hard to see, we may even find that what agents know about the bad has crisper edges, may be more determinate, than what they know about the good!) This section has been concerned to show that though the virtuous person’s knowledge of how to live will not be directed toward living badly, having and deploying knowledge of how to live well depends upon her having fairly robust, determinate, knowledge of living badly. Her determinate knowledge of the bad is always active in her deliberative route, specifically shaping how she understands the possibilities for action in pursuit of her own living well. This is because practical reason involves understanding the bad alongside and against the good; the pursuit of either depends on knowing both. This is the two-wayness of virtue: the direction of aim is only pursuable as part of an order it shares with what it steers against, and practical knowledge stands over not just the aim but the order itself.

21

I am grateful to Will Small for conversation about several of the points in this chapter, but this one especially. We can say Socrates has determinate knowledge of the bad without introducing the worry that Socrates acts badly, or is on the brink of doing so, for his understanding of the bad is active in his reasoning only as a way of steering toward the good. (He has no desire to act badly.)

110

Skill in Ancient Ethics

5. Steering against the bad as a way of getting it right In sections 3 and 4 I have been opening space to think about the two-wayness of practical rationality in virtue. Section 3 argued that virtue is shaped by the virtuous person’s knowledge of how to act well, knowledge that helps to constrain the possibilities for action by sorting the good possibilities from the bad. Section 4 took this thought further, arguing that these good and bad possibilities are best seen as organized into a two-way rational order by a knower of both. In the following section I consider what I take to be the strongest ground for virtue’s two-wayness. Here I will focus on a particular set of cases, cases in which the work of constraint has identified one or more seemingly good possibilities for action, but in which a decision has not yet been made. So the focus here, if we were to illustrate it sequentially,22 is on that part of deliberation between constraint and decision.23 If we concentrate on these cases, particularly those in which what looks quite good to start with is ultimately discarded as worse (or even bad) in favour of what is better, then new and interesting features of virtue’s two-wayness come into view. I will point to three: first, not only can a virtuous agent move against the bad on the way to the good, but she can entertain doing the bad on the way to the good. Second, entertaining the bad is not only possible on the way to acting well but may in some cases be an integral part of it. In other words, it is entertaining the bad that, in some cases, prompts the agent toward the good. And third, the rational order of virtue is organized such that entertaining the partially good and the partially bad come together; doing the former will be always, in some sense, doing the latter. Each of these three features takes us a little further toward seeing virtue reasoning and knowledge as two-way. If all three are both right and compatible with virtue, then our practical knower is possible only through a continuing and deep engagement with both the bad and the good. To see the sort of case I have in mind, let us return to our example involving the virtue of generosity. (We could run similar cases for skill, though I will not 22

23

The temporal language of this section is not important here, but is merely for ease of illustration. The description of the process is not meant to suggest an exact sequence or expected passage of time, nor, for that matter, is it meant to suggest the deliberating agent must actively reflect. This may seem like familiar territory, covered by our earlier arguments about constraint. The same work of constraint already considered above can move us all the way to decision. So we might think the right thing to say here is that, even if we are zeroing in on the final stretch of deliberation, moving against the worse toward the better is nothing but further constraint. There may be a difference in how much one has to know to navigate the final stretch well, depending on the complexity of the case, but otherwise, one might say, this is nothing new. I do not deny constraint can take us all the way to decision, or that the one who knows more will have a better shot at using it to do so. But I still maintain, as I hope will become clear, that there are new features in view when we concentrate ‘between’ constraint and decision.

An Aristotelian Account of Virtue as Two-way

111

do so here.) Suppose a fairly virtuous person would like to do something generous for a friend, like get her a gift. What would be a good gift, and a good act of giving? The virtuous person will know a lot about this. She will know how to aim as a good ‘moral archer’ should: the possibilities for giving that appeal to her will be those that are neither too much nor too little, that come at the right time, in the right manner, and so on (Aristotle 2002: 117),24 properly calibrated to and constrained by virtues other than generosity. She knows how to constrain her options in light of the circumstances: how much time she has, available funds, and the kinds of shops in her town. She knows something of what is relevant to gift-giving through practice and learning: giving well means giving something that speaks to the friend, that calls up some time spent together or that communicates to the friend that the giver has been listening, that expresses the friend’s value to her. The good gift involves and is a kind of attention. Even knowing all this, the virtuous person confronts the very ordinary question of what to give this time, how to give it, when, and so on, in order to give well. She needs to get from here to decision. What should she give? What most of us would do in such a case is start by using what we know to come up with an idea that sounds good (or perhaps a few ideas). We would sit with that idea, test it a bit, consider it, and sort through whether it might be good after all. If it holds up, we go for it. If it fails, we reopen the case and think anew. For example, I might consider getting my friend some things for her kitchen. She loves cooking, and I have fond memories of her food at gatherings she has hosted or that we have attended together. I have noticed she is in need of a certain tool for her kitchen, or has mentioned some particular style of cooking she would like to explore and could use a book on, or similar. This seems good, and so a nicely placed cooking item looks like a real contender to be a thoughtful gift. But on second thought, I have noticed recently that my friend has developed some resentment over cooking. It seems the daily grind of doing the cooking in her household is wearing her down, as is her friends’ expectation that she will always bring good food to the party. While she thinks her colleagues’ recognition of her as an excellent cook is nice, I can tell she would rather be valued as smart or interesting. In light of this, cookware starts to look like a bad gift for her, and I cross it off the list. This is a case in which robust practical knowing supports one possibility for action (getting the cookware), which looks like a good one before it doesn’t. This

24

NE II.6 1106b20.

112

Skill in Ancient Ethics

possibility for action, which is going quite well and has a lot that speaks in favour of it, undergoes a moment of reversal. Following the reversal, the possibility looks worse than it did, or even a little bit bad, and is discarded as a possible action. This is clearest from the point of view of the eventual action: from the perspective of whatever the agent chooses in the end, the possibility that was reversed is judged to be worse. That a reversal happens, however, does not mean we should deny the agent genuinely entertains giving this gift before discarding the possibility. True, she never comes to the all-things-considered judgement that she should do it. She makes no decision, and there is no gift given. Yet what comes to be bad is, at one point, an active possibility for her. We might say that it plays a more active role in her deliberation than some other possibilities, or that her deliberation about this option is much closer to action than it is with other options. Saying that I entertain the idea of buying the cookware means that I engage in rationally backing that option as a desirable one, and that I do this as part of sorting out how to go on. In such a case, the agent does not merely use what she knows to avoid bad possibilities while steering toward the good, she partly invests in a possibility that is bad on the way to doing good. What is more, she may even invest in a possibility that is bad as a way of getting toward the good. This might happen: I cross the cookware off the list in a moment of understanding that my friend needs a different kind of attention, the attention of someone who finds her intellectually interesting. In doing so, a new possibility occurs to me. This friend and I like to get together to drink tea and enjoy long conversations. A nice teapot or some delicious tea will be just the thing to remind my friend I find her engaging, that I like to spend time talking to her. What has happened here is that my very investment in the possibility I ultimately reject allows me to refine my sense of what is good in this case. ‘Investment’ reflects the fact that I have to go quite far with my commitment to the cookware before the need for something like the teapot comes into view. My aiming for the good action not only passes through the bad, or moves against it; entertaining it is important as a way of prompting me to do better. It seems clear we do not want to insist examples like these are relevant only to the person of poor virtue, as if it would go differently for the agent if only she were more generous. If she were only more virtuous, such a thought might go, the one-wayness of virtue would set her up to travel almost magically from generous desire to teapot. But even the person of full virtue encounters these everyday, low-level hard cases, such as what to give, and what it is to handle these situations well need not, but often will, involve entertaining the bad on the way

An Aristotelian Account of Virtue as Two-way

113

to the good or even as a way of pursing the good. This is true for anyone who is striving to be virtuous, up to and including the virtuous person herself. Reasoning aimed at virtue just is deeply engaged with reasoning about the bad, and with knowing the bad as part of what it is to act well in the world. This version of twowayness does not undermine the limited powers disanalogy, but it does suggest that the limited version of it is the right one to adopt. One final feature of the cases in this section is worth highlighting. In the cases with ‘reversals’, there is a way in which the good of the initial possibility persists through the reversal. This is because the partial good of that possibility was real: relative to most of the options out there, the cookware was a pretty good gift, one that someone else close to the friend might have given. That gift was good enough to survive the constraints, and good enough to become a real contender near deliberation’s end. When the reversal happens, the agent uses that option to move closer to what is better, but does not close down the partial goodness of the possibility. What happens, instead, is that she gains a sense of its goodness as partial, coming to see how what formerly seemed wholly good is also partially bad. This is because, in the rational order of virtue, the good and bad form a system, hanging together in a way that is organized toward the good. This means it will always be the case that moving toward the best depends on engaging the bad.25 It remains true that we cannot pursue the bad as bad and remain virtuous, but we do pursue the bad as part of what it is to pursue the good.

6. Conclusion At the beginning of this chapter, I mentioned several core features of the virtueskill analogy readily accepted by Aristotelians. First, we see how learners of skill and virtue, in doing their best, get better over time. Second, both skill and virtue aim for the mean, and aim to avoid recognizable ways of missing the mark. Third, in both skill and virtue, an agent’s progress and achievement depend on her tracking the limited ways of getting it right against the very many ways of getting it wrong. Fourth, both those with virtue and those with expertise count as knowers of what they do well. We are now in a position to see why the very features of the analogy that make it so powerfully appealing depend directly on what I have been calling the two-wayness of virtue-knowledge. Of these four 25

This will be true in all cases with imperfect agents (as it happens, all of us), or with imperfectly experienced agents (human beings as such), or agents with imperfect possibilities for action (anyone acting in this thoroughly flawed world).

114

Skill in Ancient Ethics

points of analogy, the final three have been at the centre of my argument. Each of them is, I argue, an indication that the rational order of virtue is analogous to that of skill, including the virtuous person’s counting as a knower of the bad, as well as a user of the bad, on the way to the good (or more precisely, of the worse on the way to the better). The first of the four points – the point about the learner of virtue and skill – has not been part of the discussion, but it is in some ways the point on which it is easiest to see that doing the good involves steering against the bad. It would be impossible to tell the story of developing skill or virtue without allowing for the building of temporary understandings, only to move past them to better ones. What my argument adds here is that this feature of the learner of virtue extends to the adult striving to be virtuous. The fairly virtuous person, or even the virtuous person acting in imperfect or unfamiliar circumstances, would be impossible to understand if she could not use her reasoning to hypothesize, test and discard, or if she could not be active in learning from her mistakes, or operate at the edge of her thought, using whatever she knows about what is bad in order to point to something better, as a person of skill might do. The twowayness of virtue-rationality is thus not only not incompatible with Aristotle’s account of skill and virtue, it also looks to animate that analogy at its foundation.

References Angier, T. (2010), Technē in Aristotle’s Ethics: Crafting the Moral Life, London: Continuum Publishing. Annas, J. (2011), Intelligent Virtue, Oxford: Oxford University Press. Aristotle. (1998), ‘Metaphysics’, in J. Barnes (ed) and W. D. Ross (trans.), The Complete Works of Aristotle: The Revised Oxford Translation, 1552–1728, Princeton: Princeton University Press. Aristotle (2002), Aristotle: Nicomachean Ethics, trans. C. J. Rowe, commentator S. Broadie, Oxford: Oxford University Press. Callard, A. (2020), ‘Aristotle on Deliberation’, in R. Chang and K. Sylvan (eds), the Routledge Handbook of Practical Reason, 126–140, London: Routledge. Irwin, T.H. (1975), ‘Aristotle on Reason, Desire, and Virtue’, The Journal of Philosophy 72 (17): 567–78. McDowell, J. (1979), ‘Virtue and Reason’, The Monist 62 (3): 331–50. Moss, J. (2011), ‘ “Virtue Makes the Goal Right”: Virtue and Phronesis in Aristotle’s Ethics’, Phronesis 56 (3): 204–61. Small, W. (2021), ‘The Intelligence of Virtue and Skill’, Journal of Value Inquiry, https:// doi.org/10.1007/s10790-021-09813-1 1 3.

6

Virtue Cultivation and the Skill of Emotion Regulation Paul Carron

Introduction At Nicomachean Ethics [NE ] II.1, 1103a30–1103b10), Aristotle draws a now familiar analogy between aretai (virtues) and technai (skills). The apparent basis of this comparison is that both virtue and skill are developed through practice and repetition, specifically by the learner performing the same kinds of actions as the expert: in other words, we become virtuous by performing virtuous actions.1 His claim that ‘like states arise from like activities’ (NE 1103b20) has led some philosophers to challenge the virtue-skill analogy. In particular, Aristotle’s skill analogy is sometimes dismissed because of the role that practical wisdom or phronēsis purportedly plays in character virtue. John Hacker-Wright notes that while certain aspects of virtue might be acquired through repetition and practice, on the Aristotelian view ‘there is a component of practical wisdom that creates a crucial distinction between practical wisdom and skill in that it requires a correct conception of worthwhile ends’ (2015: 984, my emphasis). Thus, some intellectualist (or doxastic)2

1 2

For a defence of taking Aristotle’s claim at face value, see Jimenez 2016. I am adapting Jessica Moss’s (2012: 70–84) distinction between doxastic and phantastic accounts, more commonly referred to as the intellectualist and anti-intellectualist accounts. Whereas she applies this contrast to accounts of pathé, I use it to refer to what Julia Annas et al. (2011: 25) calls intellectualist and subrational accounts of virtue. Moss’s taxonomy has distinct advantages. First – and most important – it allows there to be cognitive components in both accounts, whereas Annas’s taxonomy clearly implies that there is no role for cognition in the subrational account. Furthermore, whereas Annas’s taxonomy clearly privileges the intellectualist account, Moss’s puts the two on a more even playing field, allowing the reader to see the finer distinctions between the two accounts. Some of Annas’s earlier writings illuminate her position on Aristotle: ‘it is significant that Aristotle is a lone voice here. The ancient virtue ethics tradition followed Plato and the Stoics in holding that virtue is a skill. That is, it is a kind of skill, there being other kinds as well; virtue is, as the Stoics put it, the skill of living. The claim that we should follow the ancient tradition rather than Aristotle may at first sound rather academic, but this issue of whether virtue is or is not a skill is not merely of historical interest: it raises philosophically crucial issues about the intellectual structure of virtue’ (2003: 16–17).

115

116

Skill in Ancient Ethics

interpreters of Aristotle conclude that character virtue is not, at its core, a skill, because practical wisdom is categorically different from technai.3 I argue that this intellectualist interpretation of Aristotle is mistaken. Aristotle distinguishes the rational and non-rational parts of the soul. Character virtue stems primarily from the non-rational part of the soul (to orektikon), and is developed through the habituation of passionate elements, primarily phantasia and pathé (the Politics refers to the orektikon as the παθητικὸν μόριον, the passionate part of the soul (Aristotle, Politics, 1254b8)).4 Practical wisdom, on the other hand, is an intellectual virtue stemming from the rational part of the soul. As Angier states, ‘In terms of textual sequence, his treatment of ethismos in NE II.1 and II.4 is clearly and strongly discontinuous with his investigation of practical reason (phronēsis) in NE VI. And this discontinuity of treatment is paralleled by a further, theoretical contrast he makes between the irrational and the rational soul’ (2010: 110). Though practical wisdom is necessary for strict or complete virtue (aretè kuria), it is not necessary for the habituated virtue that Aristotle refers to in book II (NE 1104a10–1104b1).5 Character virtue is a skill that is cultivated through the non-rational habituation of the orektikon. Therefore, a phantastic account of moral virtue can better account for the virtue-skill analogy than the intellectualist accounts. This chapter proceeds as follows. I briefly discuss the skill analogy, which is primarily located in book II of the Nicomachean Ethics, a section on character virtue. The location of the discussion reveals that Aristotle’s analogy is with something non-rational, because there are no rational elements in the orektikon. Character virtue is a non-rational capacity that is developed through the habituation of passionate elements, i.e. the emotions. Character virtue also determines how the end appears to the agent, and therefore provides us with our goals through the faculty of phantasia. Phantasia is a form of non-rational evaluative cognition that allows the agent to see actions or events as good or bad, and is the basis of human emotions. Phantasiai are pleasurable perceptions of situations through memory, expectation or ‘imaginative embellishment of a

3

4 5

From Aristotle’s perspective, practical wisdom is not a skill, which is not in debate, since Aristotle makes the point clear at NE 1140b1–5. Moss stresses this point and most often refers to the orektikon as the passionate part (2012: 71–72). Both Moss and Jimenez make similar distinctions. Moss distinguishes ‘habituated’ from ‘strict virtue’ (2011: 239), while Jimenez discusses how a not-yet-virtuous agent can perform some virtuous actions if they have the right motivation (2016: 7). What the not-yet-virtuous agent lacks is the stable disposition. Moss goes further, arguing that an agent can have a stable disposition – i.e. an enduring desire for the good – while lacking practical reason, which allows the agent consistently to hit the target that he already desires, suggesting a tri-level distinction. Bostock explicitly affirms this trilevel account of virtue: natural, trained and full (2000: 86).

Virtue Cultivation and the Skill of Emotion Regulation

117

situation (e.g. visualization)’ (Moss 2012: 70). Anger, courage, generosity, pride and many other moral virtues are painful or pleasurable perceptions of valueladen situations. A virtuous person consistently sees situations in a virtuous way, i.e. the things that should appear good to the agent do in fact appear good. Grasping situations as good requires habituated, moderate emotions. The crux of my argument is that the skill analogy refers precisely to the non-rational habituation of phantasia-based emotions. Habitual repetition alters the emotional responses the agent experiences by changing the agent’s construal (or appearance – phantasia) of value-laden situations. I conclude by briefly mentioning two contemporary forms of emotion regulation – cognitive reappraisal and cognitivebehavioural therapy – that lend support from empirical psychology to Aristotle’s claim that emotions can be habituated. Character virtue is indeed a skill; it is – at least in part – the skill of emotion regulation.

1. The virtue-skill analogy In Nicomachean Ethics II.1, Aristotle draws a now familiar analogy between aretai (‘virtues’) and technai (‘skills’). The comparison yields several parallels between aretai and technai. First, ‘We acquire virtues by first exercising them. The same is true of skills, since what we need to learn before doing, we learn by doing . . .’ (NE 1103a32–34). Both aretai and technai are ‘learned’ or acquired by practice or repetition of the same kinds of activities that the expert will perform. This ‘learning by practice’ claim implies that the non-virtuous agent can perform actions that have – at the least – important similarities to the actions that the virtuous agent performs. Second, Again, as in the case of a skill, the origin and means of the development of each virtue are the same as those of its corruption: it is from playing the lyre that people become good and bad lyre players. And it is analogous in the case of builders and all the rest, since from building well, people will be good builders, from building badly, bad builders . . . NE 1103b8–11

The skilful repetition of these actions will result in a skilled person – a person with a steady habit of performing such actions well. As Jimenez puts it, these habitual dispositions require ‘ “having exercised” (ἐνεργήσαντες) and “doing” (πράττοντες) virtuous actions before possessing the corresponding dispositions’ (2016: 3–4). One main aspect of the process of habituation is repeating the

118

Skill in Ancient Ethics

activities, as in skill acquisition. The importance of activity leads Aristotle to state: ‘For by acting as we do in our dealings with other men, some of us become just, others unjust; and by acting as we do in the face of danger, and by becoming habituated in feelings fear or confidence, some of us become courageous, others cowardly’ (NE 1103b14–16). Repeating or practising the activity leads the agent consistently to perform certain types of actions and have the corresponding feelings: virtue is concerned with both praxis and pathos. Aristotle concludes: ‘like states arise from like activities (ἐκ τῶν ὁμοίων ἐνεργειῶν αἱ ἕξεις γίνονται)’ (NE 1103b20). Aretai are like technai because both are habitual dispositions developed through practice. Aristotle’s comparison between skill development and virtue development leads to a puzzle: if like activities produce like states, then how will the person who is not yet virtuous (and therefore seemingly cannot perform virtuous activities) become virtuous? With skills such as flute playing the answer seems simple: the student must emulate the teacher, and the teacher helps the student correct mistakes and learn to play more precisely. The more consistently well the student plays, the better that student becomes. Excellent practice leads to an excellent habit (ethos) or skill (techné). But Aristotle indicates that here we find a limit to the skill analogy: ‘the products of the skills have their worth within themselves, so it is enough for them to be turned out with a certain quality (τὸ εὖ ἔχει ἐν αὑτοῖς)’; conversely, ‘actions in accordance with virtues are done in a just or temperate way . . . if the agent acts in a certain state (αὐτά πως ἔχῃ)’ (NE 1105a26–30, my emphasis). This passage seems to commit Aristotle to the view that the goodness of techné depends entirely on the quality of the product, whereas virtues are not virtues simply in light of the type of action (praxis) that takes place: e.g. a person could perform what appears to be a courageous action without having the virtue of courage (e.g. if she is motivated by fame or money and not by honour or nobility). Rather, an action is virtuous only ‘if the agent acts out of a sound mind (σωφρόνως)’, namely, based on prohairesis, and ‘from a firm and unshakeable state’ (βεβαίως καὶ ἀμετακινήτως ἔχων πράττῃ) (NE 1104b5a35).6 Here we see the skill analogy break down: éthiké areté is like a techné because both are developed and completed by habit. However, éthiké areté 6

Note that this contrast between techné and arête appears to contradict Aristotle’s use of the virtueskill analogy in other texts. In particular, in Metaphysics I.1, Aristotle states that while artisans (cheirotechnitai) perform their craft through habit, master craftsmen (architektones) ‘possess a theory and know the causes’ (981b5) of their craft. The difference in the Metaphysics isn’t the outcome of the craft but the agent’s state, i.e. his theoretical knowledge of his craft. But Aristotle does not make this point in the ethical writings. For a discussion of Aristotle’s various formulations of the virtue-skill analogy, see Angier (2010: ch. 5).

Virtue Cultivation and the Skill of Emotion Regulation

119

is not like techné because whereas with techné the external quality of the activity or the product is all that matters – technically proficient and aesthetically pleasing flute playing or a beautiful, well-carved statue – for character to be excellent the actions must have the right internal motivation. In other words, Aristotle ‘now concedes that there is a difference between the domain of skills and the domain of virtue, the difference being that something inward is necessary for our being completely satisfied with the goodness of virtuous action, namely, some internal conditions of the moral agent’ (Echenique 2018: 208). Conversely, ‘We judge whether or not someone possesses a skill by looking at the quality of the product of the skill’ (Hughes 2001: 55). If a building is well designed, then the architect was skilled. Hughes goes on to point out that with virtue, we need to know how the agent saw or understood what he was doing to know whether the action counts as virtuous. There are numerous responses to this puzzle in the literature, but there are two main approaches – intellectualist and deflationist.7 The deflationist approach can be further subdivided into strong and moderate. The primary example of the strong deflationist view is what Nancy Sherman calls the ‘mechanical view’ (Sherman 1989: 157). Just as all one needs to do to become a runner is to run every day, so repeatedly performing acts of kindness will result in the character trait of kindness – the agent will be disposed to feeling and acting kindly. A problematic aspect of strong deflationist accounts is that they make virtue look less like a skill and more like a knack (empeiria), because virtue is acquired in ‘a purely unintellectual way, without understanding what it is he is doing and why’ (Annas 2011: 20). While few interpreters seriously consider this view, there are exceptions.8 A more common response to the virtue-skill puzzle is the moderate deflationary view. The moderate deflationary view admits that there is a gap between the kinds of actions that learners perform that the kinds that virtuous agents perform. As Jimenez states, ‘these commentators assume that Aristotle’s view is that the actions performed by the learners are virtuous in that they are the right actions in the circumstances, i.e. the kinds of actions characteristically performed by virtuous people, but that they differ from the activities of virtuous people not only in not being performed from a stable disposition of character, but also in that they lack virtuous motivation’ (2016: 18–19). According to this 7 8

I borrow the term ‘deflationary’ from Jimenez (2016: 6). Jimenez notes that defenders include Grant (1885), Stewart (1892) and Joachim (1951). She notes that there are more recent defenders, but their views are more moderate. See, for example, EngbergPedersen (1983) and Curzer (2012).

120

Skill in Ancient Ethics

view, virtue is like a skill in that the learner becomes virtuous by performing actions that are like the actions of the virtuous agent (the external feature), but yet the learner lacks the crucial aspect of virtuous action, namely, that it be done for the right reason or end (the internal feature). Furthermore, this view takes very seriously Aristotle’s emphasis on the disanalogy between technai and aretai: aretai require that the action be performed for the right reason and out of a stable and enduring disposition, while in the case of technai ‘the products of the skills have their worth within themselves’, i.e. the agent’s intentions bear no necessary connection to the quality of the product (NE 1105a26). Therefore focusing on the disanalogy suggests that virtue is ultimately not a skill, even by Aristotle’s own account. Thus moderate deflationary accounts argue that virtue both is and is not like a skill. The problem with this view is that it does not take seriously Aristotle’s claim – which he makes several times – that like states result from like activities. If the actions that the non-virtuous agent performs are fundamentally different from the actions of the virtuous agent, then either Aristotle misspoke or did not really mean for us to take the ‘like states result from like activities’ claim seriously. Another approach to reconciling Aristotle’s virtue-skill analogy is the intellectualist view of virtue, which essentially rejects the analogy. According to this view, the most important aspect of areté is its rational core. Moral virtue requires understanding and rational insight into morally laden situations. For instance, Mark Rowlands states that morally virtuous action requires that ‘the agent must understand what a virtue is, and be motivated by this understanding to perform a certain action because it would be virtuous . . .’ (2012: 102). Others go even further, claiming that virtuous action requires that the agent grasp ends as ends, and this is a job for practical wisdom. Consider Hacker-Wright again: ‘there is a component of practical wisdom that creates a crucial distinction between practical wisdom and skill in that it requires a correct conception of worthwhile ends’ (2015: 984). In fact, practical wisdom just is ‘good deliberation guided by a correct conception of worthwhile ends’ (2015: 984).9 Crucially, Hacker-Wright argues that in order to achieve practical wisdom, ‘we must engage in reflection on our life considered as a whole, and develop a view of what it is to live well that includes deciding which activities are worth pursuing’ (2015: 984). On this view, there is some affinity between practical wisdom and skill – since deliberating well takes practice – but there is a crucial difference: practical

9

Sherman notes that ‘virtue and practical wisdom require a reflective grasp (hupolēpsis) of the right ends’ (1989: 27).

Virtue Cultivation and the Skill of Emotion Regulation

121

wisdom involves deliberation about what kind of life we should live, and what activities will contribute to that form of life. This task is both reflective and quasitheoretical, and as such is fundamentally different from skills, even complex skills like fire-fighting or neurosurgery. Even though complex skills require deliberation and reflection, they are compartmentalized and therefore do not require the kind of global reflection that Hacker-Wright argues moral virtue requires. Consequently, on this view, moral virtue is categorically different from skills. In my view, these views are misguided. Once we are clear about the nature of character virtue and the orektikon from which it stems, we can take Aristotle to mean just what he says: virtue is like skill in that both require that the learner perform the same kinds of actions as the expert, and both are acquired through habitual practice. Furthermore, a proper understanding of the orektikon and character virtue allows my view to include many of the claims of the intellectualists, because phantasia – of which pathos is a subset – is a form of cognitive evaluation. I will now focus on the orektikon and the nature of character virtue.

2. The Orektikon and character virtue When Aristotle favourably compares virtue to a skill, it is mainly in the context of describing character virtue or excellence. Therefore it is relevant to keep in mind Aristotle’s understanding of character virtue and the part of the soul from which it stems. Aristotle divides the soul into two main parts, rational and nonrational, and then draws a distinction between character and intellectual virtue. He states that ‘one element of the soul has reason, while another lacks it’ (NE 1102a28–30).10 Rackham’s translation is helpful: ‘the soul consists of two parts, one irrational and the other capable of reason’. The word Rackham translates as ‘irrational’ is alogos, which literally means ‘without reason’ (1934: 63). Aristotle puts the orektikon in the alogos part of the soul. The orektikon is ‘lacking reason, but (is) nevertheless . . . partaking in it’ (NE 1102b12–13). It can ‘obey reason’ or be in ‘total harmony with reason’ (NE 1102b25). The orektikon is the part of the soul ‘consisting in appetite and desire in general’ (τὸ δ᾽ ἐπιθυμητικὸν καὶ ὅλως ὀρεκτικὸν μετέχει πως) (NE 1102b29–30); a literal translation of ὅλως ὀρεκτικὸν

10

Unless otherwise noted, all translations of the Nicomachean Ethics are from the Crisp translation (Aristotle 2000). Also, unless otherwise noted, textual citations come from the Nicomachean Ethics.

122

Skill in Ancient Ethics

would be ‘the whole of desire’, which cannot be what Aristotle means, since boulêsis is a form of rational desire. So he must mean that the orektikon contains the entire range of non-rational orexis, chiefly pathé. He gives examples of the various pathé at NE 1105b21–23: ‘epithumia, anger, fear, confidence, envy, joy, love, hate, longing, emulation, pity, and in general things accompanied by pleasure and pain’. Lastly, he tells us that ‘virtue is distinguished along the same lines (presumably as the soul)’ (NE 1103a3–4). Therefore we can conclude that character virtue inhabits the non-rational part of the soul and is primarily related to pathé. Character virtues are ‘states (from ἕξις – Latin ‘habitus’, related to ‘hold’ or ‘have’) [that] arise from like activities (ὁμοίων ἐνεργειῶν)’ (NE 1103b21), and are ‘concerned with feelings and actions’ (πάθη καὶ πράξεις) (NE 1106b17, 25). So character virtues are something the agent possesses – a hexis or disposition. Furthermore, ‘virtue of character (éthos) is a result of habituation (ethos) . . . nature gives us the capacity to acquire them, and completion comes through habituation’ (NE 1103a26). Aristotle also notes that lawmakers make citizens good by habituating them (NE 1103b4). As Jessica Moss puts it, there is significant textual evidence ‘that virtue is the product of habituation in actions and passions . . .’ (2011: 208); Hughes simply states that character virtues are ‘habitual dispositions’ (2001: 55). Aristotle states several times that character virtue is about actions and feelings, but his discussion indicates that the most crucial aspect of character virtue is pathé. The description of courage in the Nicomachean Ethics makes it clear that the deciding factor for whether an activity is virtuous is how the agent understands or sees his actions. Prima facie, courage turns on the action the agent performs. Yet Aristotle defines courage as ‘a mean concerning feelings of fear and confidence’ (NE 1115a6–7). The courageous person ‘feels and acts (πάσχει καὶ πράττει) in accordance with the merits of the case, and as reason requires’ (NE 1115b20–21).11 And how does the agent understand what something is worth? Put differently, what is the virtuous agent’s goal? The courageous agent acts ‘for the sake of what is noble (kalos)’ (NE 1115b12–13). When the courageous man (for Aristotle it is always a man) is faced with danger, his ethical character determines how he understands or sees his goals. The

11

The phrase ‘merits of the case’ translates κατ᾽ ἀξίαν γάρ. Irwin renders it ‘what something is worth’ (2019: 49). The final clause – ‘as reason requires’ – complicates my argument but does not prove fatal. I agree with Moss’s claim that what reason prescribes is the mean and the particulars of the situation, not the value of the end, since the value of the end is discerned through non-rational perception, as Aristotle repeatedly states.

Virtue Cultivation and the Skill of Emotion Regulation

123

virtuous agent brings to the life-threatening situation a conception or schema of nobility.12 Nobility requires, among other things, that the virtuous man be willing to die in battle defending his country (for this is one of the finest acts a human can perform, if done for the right reason). Thus the courageous man sees the threatening situation as an opportunity to be courageous, and desires to be so. Consider two soldiers: both are ordered to lead a surge into dangerous enemy territory, and both follow orders. Both act in a courageous manner. But soldier A feels intense fear and trepidation. Like the disguised Odysseus, he must steel his heart and talk himself into doing what he believes he should. Soldier B is so focused on his duty as a soldier and the honour that comes from risking one’s life for the sake of one’s country that he proceeds without hesitation.13 Soldier A is enkratic or weak-willed, not virtuous, because he construes or immediately perceives that situation in life-threatening terms, rather than honour-conferring terms, and therefore experiences significantly more fear than soldier B. The virtuous agent – soldier B – perceives the situation in the right way and therefore experiences the proper emotion (or at least less of the improper – viz. fear). So even though character virtue regulates both actions and emotions, and is often most expressed in action, ‘an action, in Aristotle’s technical sense, is defined in terms of how the agent sees what they are doing’ (Hughes 2001: 56). An agent who has achieved character virtue will have the habitual disposition to ‘respond to situations by having appropriate or inappropriate feelings’ (Hughes 2001: 57).

3. Phantasia as non-rational evaluative cognition Let’s review: the virtue-skill analogy is located primarily in the section on character virtues. Character virtues inhabit the non-rational part of the soul, specifically the orektikon. The orektikon produces all non-rational desires, including various appetites and emotions. Character virtue has primarily to do with one form of non-rational desire, pathé, and determines how the agent nonrationally sees or perceives ends. This last point needs further unpacking. Many Aristotelians claim that practical reason is the mode through which ethical agents perceive their ends. Recall Hacker-Wright’s claim that it is practical

12

13

Ross and Nisbett 2011 define an interpretative schema as ‘a knowledge structure that summarizes generic knowledge and previous experience with respect to a given class of stimuli and events and, at the same time, gives meaning and guides anticipation with respect to similar stimuli and events in the future’ (2011: 12). Cf. Hughes 2001 for a similar thought experiment.

124

Skill in Ancient Ethics

wisdom that provides us with ‘a correct conception of worthwhile ends’. This claim is a problem for the virtue-skill analogy, because it does not seem that practical wisdom is developed in the same way as a skill. However, the claim that practical wisdom gives the agent his ends flies in the face of robust textual evidence. Here I review a few examples, and argue that, for Aristotle, character virtue sets our ends or ‘makes the goal right’, not practical wisdom. Aristotle stresses that character virtue ‘sets the end’ and provides the agent with the ‘correct belief about the starting point’ (NE 1144a4; NE 1151a17–18). In the Eudemian Ethics, Aristotle asks ‘Does virtue make the goal correct, or what lies on the way to it? We assert that it is the goal, because of this there is no inference or reasoning. Let us, then, accept this as a starting point (ἀρχὴ)’ (11227b23–25).14 Character virtue works much like intellectual intuition or nous, ‘which gives us the true view of the premises from which our theoretical demonstrations follow’ (Moss 2011: 222). Character virtue provides us with our desired ends, and practical wisdom works out how to achieve those ends.15 So character virtue provides us with our ends. Our character tells us what ends are desirable, or what we view as the good or the apparent good. This point is further confirmed by a passage from the Nicomachean Ethics that Sherman translates as follows: ‘Someone might say that everyone aims at the apparent good, but does not control its appearance; but the end appears to each person in a way that corresponds to his character. For if each person is somehow responsible for his own state of character, he will also be himself somehow responsible for its [viz. the end’s] appearance [φαντασίας] (1114b1–3, cf. 1114b17)’ (Sherman 1989: 32). A second, crucial, point about character virtue is highlighted by this passage: ‘the end appears to each person in a way that corresponds to his character.’ Not only does character virtue literally provide us with our ends, but how the end appears to the agent – that is, whether it appears good or bad, desirable or undesirable – is a function of the agent’s phantastic capacity. Aristotle’s use of phantasia and the related phaínō in the passage Sherman cites is not accidental. Phantasia is not just a word for appearances, or even imaginings. It is a faculty or capacity, often referred to as the faculty of the 14 15

For a discussion of these passages, see Moss 2014: 222–225. The perceptive reader might raise an objection at this point: if non-rational character provides our practical ends in toto, then what is the difference between Aristotle and Hume? Moss suggests that Aristotle’s practical reason has two functions in the moral life that are very un-Humean: ‘First, it is reason’s job to grasp what one’s character has fixed as a goal, and also to recognize it as a goal. Second, the reasoning that goes into figuring out how to achieve the virtuous person’s goal requires an ethical sensitivity totally lacking in mere instrumental efficiency’ (2014: 223). One might also note the purpose of boulêsis in the process. Once a practical end – provided by the agent’s moral character – is grasped as an end via phronēsis, the agent can then form a rational wish for that end.

Virtue Cultivation and the Skill of Emotion Regulation

125

imagination. But even this is an inadequate definition. As Dorothea Frede points out, in Aristotle the concept of ‘phantasia does triple duty. It designates the capacity, the activity or process, and the product or result’ (1992: 279). Sorabji emphasizes that in the De Anima, phantasia refers to both perceptual and postperceptual appearance, examples of the latter being imagination, dreams and memory, all resulting from prior perceptions (1993: 197). Thus phantasia is a cognitive capacity. But it is also a non-rational capacity, something that stems from the orektikon in human beings. Do we have reason for thinking that phantasia is both cognitive and non-rational? Yes, based on passages in the De Motu Animalium and the De Anima. The De Motu provides the following insight: ‘We see that the things which move the animal are thought and phantasia and decision and wish and appetite. And all these can be reduced to intellect (nous) and desire (orexis). For both phantasia and perception (aísthēsis) hold the same place as intellect, for they are all cognitive (kritika)’ (700b17–21).16 Aristotle places nous on one side, and orexis on the other, as the main motivating capacities of animals. And under orexis he places phantasia and aísthēsis (perception). Nous – intellect or thought – is clearly a rational capacity for Aristotle. Orexis is generally a non-rational desire and it is certainly non-rational in this passage, since Aristotle is telling the reader what capacities motivate both human and non-human animals (and the latter lack nous). Crucially, this passage emphasizes that both perception and phantasia are cognitive: they are both discriminating capacities. Nussbaum’s translation essentially agrees with Moss’s: ‘For both phantasia and choice and wish hold the same place as thought, since all are concerned with making distinctions . . .’ (1978: 38). What is more, Nussbaum’s translation brings out an important aspect of kritika, from krinein, in this context. Although krinein has commonly been translated ‘to judge’, Ebert provides ample evidence that the more accurate translation, given the historical context, is to discern or discriminate, thus, to make distinctions (1983: 190). So all these capacities – thought, sense perception and phantasia – are discriminating or distinction-making capacities. Based on these passages, phantasia is a cognitive (discriminating) and motivating capacity, but we might still wonder whether it is non-rational. Aristotle answers that question in the Eudemian Ethics, where he states: The object of desire and the object of wish is either the good or the apparent good. And this is why the pleasant is an object of desire, for it is an apparent

16

Translation from Moss 2012: 9–10.

126

Skill in Ancient Ethics

good; for some believe it is [good], and to some it appears [good] although they do not believe it so. For phantasia and belief are not in the same part of the soul. EE 1235b26–29, my emphasis17

We can deduce from this passage that, in humans, phantasia is in the nonrational part of the soul, the orektikon. It is motivating, discriminating and distinct from the intellectual or rational part of the soul.18 One final passage from the De Anima confirms this psychology: ‘Now understanding (nous) is always correct, whereas desire and imagination (phantasia) are both correct and incorrect. That is why what causes movement in every case is the object of desire, which is either the good or the apparent good’ (De Anima 433a26–29). A rational desire is always for the actual good; non-rational desire is based on phantasia and can be for the good or the apparent good (Moss 2012: 139). In sum, I have argued in this section that it is character, and not practical wisdom, that supplies us with the content of our practical goals. This is possible because character virtue determines how the end appears to the agent. Specifically, the faculty that provides the agent with these appearances is phantasia. Phantasia is a cognitive, motivating capacity that stems from the non-rational part of the soul and allows for non-rational evaluative cognition. Phantasia allows the agent to cognize ends as goods or apparent goods. And in straightforward cases, agents with good ethical character will perceive actual goods as good, since they have been habituated to do so.19 The claim that character provides us with our ends is crucial to my argument, because the intellectualist defeaters to the virtue-skill analogy are based on the claim that only practical wisdom – a rational capacity 17 18

19

Translation from Moss 2012: 48. One might object here that this description of discrimination makes it sound like an intellectual capacity. Yet there are many examples of discrimination in non-rational perception. We discriminate the difference between chocolate and vanilla, or loud and soft sounds, unconsciously: the discrimination is intuitive and immediate. And discrimination can sometimes take effort (to really focus and pay close attention to discern fine distinctions), and it may sometimes involve rationality, as when thinking through the taste profiles of two different wines, or the brush styles of two different drummers, to discern which is which. But this high level of cognition is not always involved in perceptual discrimination. It does seem that in difficult cases, where the actual good is not clear, the intellectual capacities (most likely phronēsis) will be necessary to discern the good. But, even in these cases, it is one’s ethical character that limits the field of options. A bigger challenge results from Aristotle’s discussion of the akratic individual, who ‘tends to be carried away contrary to correct reason because of the ways he is affected’, yet the archē is preserved in him (NE 1151a20–25). Jay Elliott 2018 argues that the akratic person has not received a virtuous upbringing, yet still rationally desires a good end. Aristotle does not tell us how the akratic acquired that archē, but Elliott argues that it must have been through some rational process. Elliott concludes that Aristotle’s is a ‘character pluralism’, according to which his moral psychology does not involve ‘a single conception of psychic structure or moral development that is equally applicable to all character types’ (Elliott 2018: 448). Since this chapter is primarily concerned with how the virtuous person’s character provides him with his practical ends, I do not consider how akrasia might affect my argument.

Virtue Cultivation and the Skill of Emotion Regulation

127

from the rational part of the soul – can provide us with our ends. However, Aristotle is clear that phantasia provides us with our ends. If phantasia is something that is subject to non-rational habituation, then virtue can indeed be like a skill. Now I turn to the argument that pathé are based on phantasiai.

4. Appearance-based emotions It has become common among Aristotle scholars to argue that Aristotle holds a cognitive theory of the emotions. More specifically, these scholars argue that for Aristotle, emotions are constituted by beliefs (Fortenbaugh 2006: 113), or even judgements (Nussbaum 1986: 83). However, this is strange since Aristotle distinguishes between the rational and non-rational parts of the soul, attributing the emotions to the part that contains phantasia, and perception and beliefs to the rational part. The straightforward conclusion is that emotions are based on a non-rational form of cognition, but which form? Aristotle provides us with the answer. He defines emotions in general as ‘all those affections that change men so as to influence their judgments (κρίσεις), and are accompanied by pleasure and pain; such are anger, pity, fear and all similar emotions and their contraries’ (Rhetoric 1378a8–9). This makes clear that emotions are discriminating (they cause a change in judgement) and are pleasurable or painful. But it does not speak of any direct relation to phantasia. Fortunately, the rest of the Rhetoric is helpful: of the twelve emotions described therein, seven contain direct references to phantasia or its verb form (Sihvola 1996: 116). Moss notes that ‘A close look at the Rhetoric’s characterizations of the passions shows that they involve phantasia in particular: they all depend on memory, expectation, or imaginative embellishment of a situation (e.g. visualization) – which are paradigm exercises of phantasia’ (2012: 70). Fear illustrates the phantastic basis of emotions: Aristotle defines fear as ‘a certain kind of pain and disturbance out of the appearance of an impending destructive or painful bad thing’ (Rhet II 5, 1382a21–3), adds that these bad things must ‘appear to be close and not far-off ’ (Rhet II 5, 1382a24–5), and finally remarks that ‘it is necessary that those things are fearful that appear to have a great power to destroy or cause harms that lead to great pain (Rhet II 5, 1382a28–30)’. Sihvola 1996: 116, my emphasis

Fear illustrates the three essential features that all emotions have in the Rhetoric. First, they are pleasurable or painful, second, they involve ‘evaluative

128

Skill in Ancient Ethics

representations’, and third, these emotions are representations stemming from the phantastic capacity (Moss 2012: 75). Fear is a perception of something as impending and threatening, and this perception causes pain or disturbance in the perceiver. But if it is a perception and is accompanied by pleasure and pain, why must it be an instance of phantasia? After all, hearing a very loud, highpitched noise is a job for sense perception, and it is also painful. However, emotions are not simply pleasurable or painful perceptions, they are evaluative perceptions. To hear a noise as loud and painful is not necessarily to value it in any way. The noise does not offend you or threaten you (of course, if it does threaten you, then you have evaluated it, but this is not a necessary condition of the sense perception of a loud noise). Emotions, on the other hand, accompany (or rather, are constituted by) evaluative perceptions. Another reason that emotions are not simply sense perceptions but instead rely on phantasia is that emotions are not just about immediately perceivable objects (although they sometimes are). In fact, Aristotle’s characterization of the various emotions in the Rhetoric indicate that, more often, emotions involve memories of past events and imaginative embellishment of future events. Consider the following passage from the Rhetoric: Since to be pleased lies in perceiving a certain affection (ἐστὶν τὸ ἥδεσθαι ἐν τῷ αἰσθάνεσθαί τινος πάθους), and phantasia is a kind of weak perception (αἴσθησίς τις σθενής), [and] some phantasia of what a person remembers or expects [hopes – ἐλπίζει] would always attend in remembering or expecting – if this is the case, it is clear that pleasures come simultaneously to those who are remembering and expecting, since there is perception there, too. Thus it is necessary that all pleasant things are either present in perceiving or past in remembering or future in expecting; for people perceive the present, remember the past, and expect the future’. Rhetoric 1370a28–3520

This quotation makes three things clear. First, phantasia is a kind of perception, but not the same capacity as aísthēsis. Second, phantasia is an affective perception, a perception of a pleasant or painful thing. Finally – and another quality that distinguishes phantasia from aísthēsis – phantasiai are not only about present perceptions (although they can be) but are also about memories and expectations. Rhetoric I.11 further supports this claim by describing winning as pleasurable on account of a mental image (phantasia) of superiority, and honour and reputation

20

Translation from Moss 2012: 78.

Virtue Cultivation and the Skill of Emotion Regulation

129

as being pleasant based on a mental image of being like the excellent person (1370b32–4, 1371a8–9).21 This last point is crucial when discussing emotion regulation and especially cognitive-behavioural therapy. The reader might object at this point, and claim that the texts that I have been examining to support my argument about phantasia (e.g. De Motu Animalium, De Anima, On Rhetoric) have a different goal than the ethical works (none of them are about human virtue or flourishing per se), and therefore contain slight but important differences in psychology. So how do the ethical writings conceive of emotions? In the Ethics, virtues are characteristics defined primarily by the emotions they regulate, and the feelings the agent experiences in a virtueeliciting situation both guide the agent’s response and are indicators of whether the response is virtuous. Furthermore, Aristotle explicitly defines pathé in the Nicomachean Ethics as ‘epithumia, anger, fear, confidence, envy, joy, love, hate, longing, emulation, pity, and in general things accompanied by pleasure and pain’ (1105b21–23, my emphasis). A person’s state of character is primarily revealed by their emotions, and those emotions are always accompanied by pleasure and pain. When Aristotle is discussing what psychological power virtue is, he declares that virtue is a hexis or state: ‘And by hexeis I mean those things in respect of which we are well or badly disposed in relation to pathé. If, for example, in relation to anger, we feel it too much or too little, we are badly disposed; but if we are between the two, then well disposed’ (NE 1105a26–27). Virtues are dispositions to appropriate actions, but actions are virtuous only if motivated by the proper pathé. A chief indicator of the agent’s moral state is the agent’s emotions or feelings. The virtuous response cannot exist apart from the proper feelings. But Aristotle goes even further, asserting that virtues have to do with situations where the agent is affected or acted upon, ‘and pleasure and pain follow from every action and situation of being affected, then this is another reason why virtue will be concerned with pleasures and pains’ (NE 1104b14–17). The emotions that constitute virtues are affective mental states, a point driven home once again by the example of courage: the person who enjoys facing up to danger, or at least does not find it painful to do so, is courageous, while he who does find it painful is a coward. For virtue of character is concerned with pleasure and pains; it is because of pleasure that we do bad actions, and pain that we abstain from noble ones. NE 1104b8–11, my emphasis

21

See Moss 2012: 84.

130

Skill in Ancient Ethics

When a soldier sees the enemy approaching, his immediate perceptual response is either one of intense fear – which is immediately painful – or courage – which is probably pleasurable, or at least not painful.22 The soldier has an affective mental state, and although the Ethics does not make this clear, this mental state is surely tied to previous experiences (and the painful or pleasurable mental states tied to those experiences), and based in part on the anticipation of the impending attack bringing either pleasure or pain. Aristotle repeatedly makes a clear connection between virtue, emotion and feeling (pain and pleasure). In a summary passage, Aristotle declares: ‘we regulate our actions by pleasure and pain. Our whole inquiry, then, must be concerned with them, because whether we feel enjoyment and pain in a good or bad way has great influence on our actions’ (NE 1105a5–8). The Ethics stresses that emotions are at the heart of virtue, that virtue regulates emotions, and that those emotions (and hence the virtues) are accompanied by pleasure and pain. When fear is accompanied by pleasure or pain, it is most likely because it is an affective mental state tied to the memory or anticipation of similar feelings. But the question that remains is whether the Ethics confirms that emotions are pleasurable or affective perceptions. Consider a passage from the end of book II, where Aristotle is asking how we discern the mean: ‘But how far and to what extent someone must deviate (from the mean) before becoming blameworthy it is not easy to determine by reason (λόγῳ), because nothing perceived by our senses is easily determined; such things are particulars, and judgement about them lies in perception (αἰσθήσει ἡ κρίσις)’ (NE 1109b20–24, my emphasis). The mean is discovered through the faculty of perception (aísthēsis), a discriminating faculty. And the mean concerning the disposition or state of courage – discussed above – makes us ‘well or badly disposed in relation to feelings’ (NE 1105a26). And if we are courageous at the proper time, it is because of feelings: ‘it is because of pleasure that we do bad actions, and pain that we abstain from noble ones’ (NE 1104b11). The courageous soldier perceives the impending attack as a situation calling for a courageous response, a perception immediately accompanied by pleasure (or at least the lack of significant pain). In other words, the soldier does not construe the situation as one calling for courage without the accompanying feeling. We can conclude that courage is a mean state of emotion determined by perception and accompanied by pleasure or pain. When we perceive a situation as one calling for courage, our perception is guided by pleasure or

22

These are not the only two options, of course, but are used for the sake of clarity and simplicity.

Virtue Cultivation and the Skill of Emotion Regulation

131

pain. Although in the Ethics the case is not as clear that emotions are pleasurable perceptions as it is in the psychological writings, it is clear that virtue concerns situations of being affected; furthermore, virtue is determined by perception, concerned primarily with emotions, and guided by pleasure and pain. Put together, these aspects strongly suggest that emotions are affective perceptions, and are thus appearances stemming from the phantastic capacity, placing them clearly in the non-rational part of the soul and involving non-rational capacities.

5. Phantastic habituation and emotion regulation I have argued that character virtue is non-rational, provides us with our ends, and is primarily concerned with phantasia-based pathé. Emotions are phantasiai that are accompanied by pleasure and pain and cause a change in the agent’s judgement or construal of the object or situation. Thus emotions are cognitive because they are ways of construing. We now come back full circle, to Aristotle’s virtue-skill analogy, and his crucial claim that ‘like activities produce like states’. If the states in question are primarily concerned with phantasiai accompanied by pleasure, then surely the activities that need to be practised and repeated to produce those states will be activities focused on pleasurable phantasiai. At this point another doxastic objection arises. A puzzle remains, because if ‘like states’ is supposed to mean ‘fully or completely virtuous states’, then the activities flowing from such states are not possible for the non-virtuous person, because doxastic and phantastic interpreters agree that full moral virtue is assumed to include prohairesis as a key component (e.g. Irwin 1980; Sherman 1989; Bostock 2000; Moss 2011).23 Aristotle clearly says as much in NE VI, stating ‘it is impossible to be good in the full sense of the word without practical wisdom (κυρίως ἄνευ φρονήσεως)’ (1144b30).24 Bostock notes that Ross’s rendition, ‘virtue in the strict sense’, comes closest to conveying Aristotle’s meaning (2000: 86). A person cannot be virtuous in the strict sense without phronēsis. But why does Aristotle modify agathos with kuria here? Bostock notes that while Aristotle draws a clear distinction between ‘natural virtue’ and ‘full virtue’ (aretè kuria), a third level is necessary to reconcile Aristotle’s claims

23 24

Thanks to Frans Svensson for pressing me on this point in personal correspondence. See also: ‘Now virtue in the full sense cannot be attained without practical reason (phronēsis)’ (NE 1144b17) and ‘Virtue or excellence is not only a characteristic which is guided by right reason, but also a characteristic which is united with right reason; and right reason in moral matters is practical reason (phronēsis)’ (NE 1144b25).

132

Skill in Ancient Ethics

concerning moral virtue in book II (2000: 86). Bostock calls the third, intermediary, level ‘trained virtue’ (2000: 86) while, Moss calls it ‘habituated virtue’ (2011: 212). What justification do we have for this intermediate level? First, Aristotle stresses that ‘nature gives us the capacity to acquire them (moral virtues), and completion comes through habituation’ (NE 1103a). He ends that chapter stating that ‘it is not unimportant how we are habituated from our early days; indeed, it makes a huge difference – or rather all the difference’ (NE 1103b). Furthermore, the goal passages in book VI clearly state that moral virtue provides us with our ends, and phronēsis with the means to achieve those ends. Once you have acquired habituated virtue, you will want to be virtuous; virtue will be your aim. You are disposed to virtuous emotions and actions before you possess prohairesis. However, there may be times where the virtuous action is not obvious, and conscious reasoning is necessary to achieve your virtuous goal. Finally, Aristotle is very adamant at the end of the Nicomachean Ethics that education and argument will do little or nothing to help the person with bad character traits. ‘For the natural tendency of most people is to be swayed not by a sense of shame but by fear, and to refrain from acting basely not because it is disgraceful, but because of the punishment it brings . . . (therefore) [t]o change by argument what has long been ingrained in a character is impossible or, at least, not easy’ (NE 1179b10–18, my emphasis). This passage makes it clear that the disposition to act nobly or shamefully is formed long before a person can understand complex arguments. A few lines latter Aristotle further illuminates why habituation is so important: Argument and teaching, I am afraid, are not effective in all cases: the soul of the listener must first have been conditioned by habits to the right kinds of likes and dislikes, just as land [must be cultivated before it is able] to foster the seed. For a man whose life is guided by emotion will not listen to an argument that dissuades him, nor will he understand it. How can we possibly persuade a man to change his ways? And in general it seems that emotions do not yield to argument but only to force. Therefore, there must first be a character that somehow has an affinity for excellence or virtue, a character that loves what is noble and feels disgust at what is base. NE 1179b24–30, my emphasis

Habituation during a person’s upbringing develops the disposition to love what is noble and despise what is base. It is this person who is amenable to education and argument. The person with habituated virtue will already consistently have the right affective perceptions and can now develop the intellectual virtue of phronēsis, necessary for aretè kuria. This distinction between habituated and full

Virtue Cultivation and the Skill of Emotion Regulation

133

virtue is why Aristotle can claim that like activities produce like states, because the activities in question do not have to be activities exactly as the person possessing virtue in the full sense would perform them (accompanied by phronēsis), but as the person with habituated virtue would. A question remains: do we have any contemporary evidence supporting Aristotle’s developmental account of moral virtue? A key aspect of his theory is that emotions and feelings must be appropriate and moderate, but other than simply giving a child chocolate every time they have the right emotion (which would surely be counterproductive), how might emotions be habituated? Sherman’s notion of critical practice provides some insight. Sherman states that ‘The model I ascribe to Aristotle is thus that of a chain of activities which increase in discriminated complexity as well as in derivative pleasures’ (Sherman 1989: 184). This critical practice is possible because of the child’s cognitive yet nonrational capacities, and it is pleasurable because as the child becomes better at the practice, he comes to appreciate more and more the goods internal to that practice, and find the good performance of that practice pleasurable. Furthermore, ‘Cultivating the dispositional capacities to feel fear, anger, goodwill, compassion, or pity appropriately will be bound up with learning how to discern the circumstances that warrant these responses’ (Sherman 1989: 166–67). The child will need to be taught how to perceive rightly, how to discriminate between different features of his environment. Learning perceptual discrimination will require instruction and example. The child’s parent or teacher will point out to the child the particular aspects of certain situations that make certain emotions proper and others improper (Sherman 1989: 171). There will also be ongoing conversation and continual attempts at refinement and improvement of the child’s discriminatory capacities. Sherman stresses that though the child lacks the capacity for deliberation (bouleutikon), he possesses the capacities of affect, discrimination, perception and instrumental reasoning (1989: 161). And it is these capacities that make critical practice possible. Sherman’s account is helpful, but still highly speculative. She claims that the child somehow learns how to discern the circumstances that warrant the proper emotional responses, but how? Long before a child can properly ‘compose the scene’, they need to be able to regulate their emotions. Contemporary research suggests that Aristotle was correct in his claim that emotions can be habituated, and that the process of habituation can start at a very early age.25 The most basic 25

Angier notes that ‘there is empirical evidence that moral dispositions are attainments characteristically developed over long periods of time, attainments that require repeated involvement in corresponding activities’, but does not indicate what that empirical evidence might be (2010: 108).

134

Skill in Ancient Ethics

form is calming your body through physical activities such as the STAR (stop, take a breath, and relax), pretzel (giving oneself a hug), water faucet (moving one’s arms and making sounds like a faucet), etc. But there are more complex techniques that older (primary school) children can begin to employ that can be fruitfully utilised by adults as well. Psychologist James J. Gross and colleagues have researched various emotionregulation strategies. They have been conducting experiments for over twenty years that focus on both basic and more advanced cognitive strategies that humans can engage in to influence what emotions they have, how they express them and what behaviours follow from them.26 They state, ‘specific emotion-regulation strategies can be differentiated along the timeline of the unfolding emotional response’ (Gross and John 2003: 348). Most of these strategies are antecedent focused – they take place before the emotion has been generated.27 The most relevant antecedentfocused strategy for virtue cultivation is cognitive reappraisal, ‘a form of cognitive change that involves construing a potentially emotion-eliciting situation to change its emotional impact’ (Gross and John 2004: 1304). With cognitive reappraisal, the agent intentionally construes the perceptual event in different terms. Someone with the vice of excessive anger is going to construe many innocuous actions or situations as offensive by interpreting the ‘offending’ person’s actions as an intentional spite; if they wish to decrease their anger, they need to reconstrue the situation by giving the ‘offending’ person’s actions a different interpretation. In class, I once brought up the example of getting angry at another driver who cuts you off in traffic. One of my students responded that she always assumes that the ‘offender’ really needed to use to restroom, and this keeps her from getting angry. She is reconstruing the situation in different terms and this reconstrual changes her emotional reaction as a result. Agents often use cognitive reappraisal to decrease emotional experience, but can also use it to enhance emotional experience or to experience what is deemed the proper or beneficial emotional experience. And one might do this ahead of time through visualization or imagination. When knowing that one needs to have an uncomfortable conversation with a friend, one might imagine the conversation as resulting in a deeper bond between friends. Numerous experiments validate the efficacy of cognitive reappraisal. One experiment involving cognitive reappraisal asks subjects to view a sad film clip

26

27

The reader should note that these experiments on emotion regulation are not ‘one-off ’ studies. The evidence is well documented and robust. For instance, see Gross 2007. For a related overview, see Beauregard 2007. One strategy is response focused – it takes place after the emotion is generated. Gross and John call it ‘expressive suppression’ (2003: 349).

Virtue Cultivation and the Skill of Emotion Regulation

135

but to think about the film so it does not make them sad (i.e. as a film critic might). In this experiment, ‘reappraisal decreased both the experience and the behavioural expression of negative emotion’ (Gross and John 2004: 1306). More importantly, cognitive reappraisal can have lasting effects. For example, in one experiment, subjects viewed a sad film clip. One group of subjects reappraised during the viewing, and the other watched as they normally would. Researchers then moved the subjects to another room, where subjects read three moral dilemmas. Subjects then related the emotions aroused by the dilemma and their strength and ranked the moral wrongness of the action taken in the dilemma. Researchers concluded, ‘Participants in the reappraisal condition subsequently rated the behavior of the targets in the scenarios as significantly less immoral .|.|. (and) reported experiencing significantly less intense emotion than did participants in the control condition’ (Feinberg et al. 2012: 792). By engaging in the emotion-regulation strategy of cognitive reappraisal, agents intentionally change both their construals of stimuli and their emotional impact and can affect how they encounter future stimuli. But these are still mostly synchronic examples, not instances that bring about deep and lasting change. This is where cognitive behavioural therapy comes in. Jeffrey M. Schwartz’s research on cognitive behavioural therapy (CBT) aimed at patients suffering from obsessive-compulsive disorder (OCD) exemplifies a long-term antecedent-focused strategy that includes cognitive reappraisal as well as other antecedent-focused strategies (and of course practices aimed at behavioural modification). Schwartz and other researchers believe that OCD patients can actually change their behaviours (and indeed their neurocircuitry), and, to this end, developed a multi-step cognitive behavioural method (1999: 122).28 These steps include reappraisal by relabelling obsessive thoughts, attentional deployment by turning one’s attention to something else besides the obsessive thoughts, and behavioural changes. For instance, one can work in the garden instead of washing one’s hands yet again. Or a patient might be taught to

28

CBT is the most widely studied and successfully implemented form of therapy in modern psychology. As Greg Lukianoff and Jonathan Haidt state in their article ‘The Coddling of the American Mind’, ‘Cognitive behavioural therapy is a modern embodiment of this ancient wisdom. It is the most extensively studied nonpharmaceutical treatment of mental illness, and is used widely to treat depression, anxiety disorders, eating disorders, and addiction. It can even be of help to schizophrenics. No other form of psychotherapy has been shown to work for a broader range of problems. Studies have generally found that it is as effective as antidepressant drugs (such as Prozac) in the treatment of anxiety and depression. The therapy is relatively quick and easy to learn; after a few months of training, many patients can do it on their own. Unlike drugs, cognitive behavioural therapy keeps working long after treatment is stopped, because it teaches thinking skills that people can continue to use’ (2015: 46). See also Teasdale 1997 and Steketee et al. 1998.

136

Skill in Ancient Ethics

reconstrue a dirty cloth that normally induces a compulsion to wash their hands, and instead remind themselves that it is just axle grease, and their hands are not dirty and do not need be to washed (Schwartz 1999: 129). Twelve out of eighteen patients who engaged in this nine-week programme experienced both objective and subjective change. They reported significant reduction in their OCD behaviour, and their brain activity differed significantly (based on scans before and after the programme) (1999: 124).29 The cognitive reappraisal strategies used to combat OCD symptoms is notable in relation to Aristotle’s account of virtue habituation. An OCD sufferer who constantly washes their hands has a perception of their hands as dirty, accompanied by a painful or unpleasant sensation, and experiences anxiety or stress. In other words, the OCD sufferer has an affective perception that motivates an action, the action of handwashing. Both the perception and the action are habitual qualities. OCD sufferers are disposed to see their hands as dirty, feel disgust and stress and wash to stop those feelings. When OCD sufferers engage in diachronic cognitive behavioural strategies, they are habituating their affective perceptions so that first they will not feel as strong emotions aimed at their ‘dirty’ hands, and eventually will no longer have the impression that their hands are dirty at all. They practise discriminating between true and false impressions of dirtiness, and this practice regulates their emotions, which in turns generates different perceptions or impressions. Thus, emotion regulation can start at the most basic level and gradually get more advanced. For children, it need not be nearly as complex as for an OCD sufferer. Simple practices such as calming their bodies, focusing or re-focusing their attention, and being rewarded for the appropriate behaviour (or punished for the inappropriate) will slowly habituate appropriate emotions. The example of cognitive reappraisal as seen in CBT reinforces the cognitive nature of emotions; it also shows that emotions are not directly affected by the rational assent to a proposition or judgement. We cannot simply decide that a person or event should not make us angry. Rather, we must focus on different aspects of the situation or construe it in different terms. Furthermore, as Aristotle indicates in his discussion of honour and reputation, we can imagine how we want things

29

Schwartz notes that ‘What this accomplishes is a change in perspective away from automatic responses (exactly the sort of activity the basal ganglia is wired by many millennia of evolution to perform) . . . and toward a more precise, considered, and consciously goal-directed interpretation of the present moment’s experience – which is, of course, a much more cortically directed activity’ (1999: 127).

Virtue Cultivation and the Skill of Emotion Regulation

137

to turn out. Doing this repeatedly (as in the case of CBT) makes it easier and can generate lasting change. The virtuous person will have habituated character virtue and that person’s character will control how he sees or construes the ends and thus what emotions he experiences. The comparison with cognitive emotionregulation strategies also helps us see that although Aristotle’s own view of the habituation of moral virtue is not intellectualist, it is not subrational either. Emotions are non-rational but cognitive and affected through cognitive efforts. And most of all it helps us see that like activities – regulatory activities – will produce like states – that is, a person with well-regulated emotions or habituated virtue. Character virtue is indeed like a skill, the skill of emotion regulation.

References Angier, T. (2010), Technē in Aristotle’s Ethics: Crafting the Moral Life, New York: Continuum. Annas, J. (2011), Intelligent Virtue, Oxford: Oxford University Press. Annas, J., M. DePaul and L. Zagzebski (2003), ‘The Structure of Virtue’, in M. DePaul and L. Zagzebski (eds), Intellectual Virtue, 15–33, Oxford: Clarendon Press. Aristotle (1933), Aristotle: Metaphysics, trans. H. Tredennick, Cambridge, MA : Harvard University Press. Aristotle (1934), Aristotle: The Nicomachean Ethics, trans. H. Rackham, Cambridge, MA : Harvard University Press. Aristotle (1957), Aristotle: On the Soul. Parva Naturalia. On Breath, trans. W.H. Hett, Cambridge, MA : Harvard University Press. Aristotle (1975), Aristotle: The Art of Rhetoric, trans. J.H. Freese, Cambridge, MA : Harvard University Press. Aristotle (1999), Nicomachean Ethics, trans. T. Irwin, Indianapolis, IN : Hackett Publishing. Aristotle (2000), Nicomachean Ethics, trans. R. Crisp, Cambridge: Cambridge University Press (Cambridge Texts in the History of Philosophy). Beauregard, M. (2007), ‘Mind does really matter: evidence from neuroimaging studies of emotional self-regulation, psychotherapy, and placebo effect’, Progress in Neurobiology 81 (4): 218–236. Bostock, D. (2000), Aristotle’s Ethics, Oxford: Oxford University Press. Curzer, H. (2012), Aristotle and the Virtues, Oxford: Oxford University Press. Ebert, T. (1983), ‘Aristotle on what is done in perceiving’, Zeitschrift für Philosophische Forschung 37 (2): 181–198. Echenique, J. (2018), ‘Another Dissimilarity between Moral Virtue and Skills: An Interpretation of Nicomachean Ethics II 4’, in M. Boeri, Y.Y. Kanayama and

138

Skill in Ancient Ethics

J. Mittelmann (eds), Soul and Mind in Greek Thought. Psychological Issues in Plato and Aristotle, 199–215, Cham: Springer International Publishing (Studies in the History of Philosophy of Mind),. Elliott, J.R. (2018), ‘Aristotle on the archai of practical thought’, The Southern Journal of Philosophy 56 (4): 448–68. Engberg-Pedersen, T. (1983), Aristotle’s Theory of Moral Insight, Oxford: Oxford University Press. Feinberg, M. et al. (2012), ‘Liberating reason from the passions: overriding intuitionist moral judgments through emotion reappraisal’, Psychological Science 23 (7): 788–95. Fortenbaugh, W.W. (2006), Aristotle’s Practical Side: On His Psychology, Ethics, Politics and Rhetoric, Leiden: Brill. Frede, D. (1992), ‘The Cognitive Role of Phantasia in Aristotle’, in M.C. Nussbaum and A. Rorty, (eds) Essays on Aristotle’s De Anima, 279–295, Oxford: Oxford University Press. Grant, A. (1885), The Ethics of Aristotle (2 vols), London: Longman. Gross, J.J. (2007), Handbook of Emotion Regulation, New York: Guilford Press. Gross, J. J., and O.P. John (2003), ‘Individual differences in two emotion regulation processes: implications for affect, relationships, and well-being’, Journal of Personality and Social Psychology 85 (2): 348–62. Gross, J. J., and O.P. John (2004), ‘Healthy and unhealthy emotion regulation: Personality processes, individual differences, and life span development’, Journal of Personality 72 (6): 1301–33. Hacker-Wright, J. (2015) ‘Skill, practical wisdom, and ethical naturalism’, Ethical Theory and Moral Practice 18 (5): 983–93. Hughes, G. (2001), Routledge Philosophy Guidebook to Aristotle on Ethics, London: Routledge. Irwin, T.H. (1980), ‘Reason and Responsibility in Aristotle,’ in Amélie Oksenberg Rorty (ed.), Essays on Aristotle’s Ethics, Berkeley : University of California Press. Jimenez, M. (2016), ‘Aristotle on becoming virtuous by doing virtuous actions’, Phronesis 61 (1): 3–32. Joachim, H.H. (1951), Aristotle: The Nicomachean Ethics, Oxford: Oxford University Press. Lukianoff, G., and J. Haidt (2015), ‘The coddling of the American mind’, Atlantic Monthly 316 (2): 42–53. Moss, J.D. (2011), ‘ “Virtue makes the goal right”: virtue and phronesis in Aristotle’s Ethics’, Phronesis 56 (3): 204–61. Moss, J.D. (2012), Aristotle on the Apparent Good: Perception, Phantasia, Thought, and Desire, Oxford: Oxford University Press. Moss, J.D. (2014), ‘Was Aristotle a Humean? A Partisan Guide to the Debate’, in R. Polansky (ed.), The Cambridge Companion to Aristotle’s Nicomachean Ethics, 221–241, New York: Cambridge University Press.

Virtue Cultivation and the Skill of Emotion Regulation

139

Nussbaum, M.C. (1978), Aristotle’s De Motu Animalium: Text with Translation, Commentary, and Interpretive Essays, Princeton, NJ : Princeton University Press. Nussbaum, M.C. (1986), The Fragility of Goodness: Luck and Ethics in Greek Tragedy and Philosophy, New York: Cambridge University Press. Ross, L., and R.E. Nisbett (2011), The Person and the Situation: Perspectives of Social Psychology, Philadelphia: Temple University Press. Rowlands, M. (2012), Can Animals Be Moral? Oxford: Oxford University Press. Schwartz, J.M. (1999), ‘A role for volition and attention in the generation of new brain circuitry’, Journal of Consciousness Studies 6 (8–9): 115–42. Sherman, N. (1989), The Fabric of Character: Aristotle’s Theory of Virtue, Oxford: Clarendon Press. Sihvola, J. (1996), ‘Emotional animals: do Aristotelian emotions require beliefs?’, Apeiron 29 (2): 105–44. Sorabji, R. (1993), Animal Minds and Human Morals: The Origins of the Western Debate, Ithaca, NY: Cornell University Press. Steketee, G. S. et al. (1998), ‘Cognitive Theory and Treatment of Obsessive-Compulsive Disorder’, in M.A. Jenike, L. Baer and W.E. Minichiello (eds), Obsessive-Compulsive Disorders: Practical Management, 3rd ed., 368–399, St. Louis, MO : Mosby. Stewart, J.A. (1892), Notes on the Nicomachean Ethics of Aristotle, Oxford: Clarendon Press. Teasdale, J.D. (1997), ‘The Relationship between Cognition and Emotion: The Mind-inPlace in Mood Disorders’, in D.M. Clark and C.G. Fairburn (eds), Science and Practice of Cognitive Behaviour Therapy, 67–93, Oxford: Oxford University Press.

140

7

Aristotle on Techne: Two Theses in Search of a Synthesis? Joseph Dunne

Introduction By distinguishing techne (craft) so firmly from phronesis (practical wisdom) in Book 6 of the Nicomachean Ethics – assigning governance of successful production (poiēsis) to the former and of virtuous action (praxis) to the latter – Aristotle might seem to close off any appeal to techne for explanatory or illustrative purposes in ethics. Undeterred by this distinction, however, in several other places in the Ethics he casually elucidates ethical conduct by emphasising strong similarities, not differences, between it and the exercise of various technai. My concern here is not to analyse the validity of this recourse to techne as an analogue in clarifying virtuous praxis but rather in acting well but rather – an anterior task – to analyse techne itself. For, as I hope will become apparent, Aristotle’s treatment of it reveals ‘techne’ to be both complicated and crucial – so that it needs and deserves elucidation in its own right. What follows, then, reverses the usual direction of analysis: rather than foregrounding issues of ethical virtue – with techne/poiēsis in the background, so to speak, as a ready-tohand reference-point (the merits or demerits of which are up for discussion) – I will foreground Aristotle’s account of techne/poiēsis. And, while it is often argued that phronesis-informed praxis is distorted by analogies between it and techne-informed poiēsis, I will travel in the opposite direction to argue that techne-poiēsis is distorted by much in Aristotle’s account of it – but that it emerges in a truer light when his attention is drawn to analogies between it and phronesis-praxis. As symbolized in the Promethean myth, techne enshrines all the controlling capability that human beings can interpose between themselves as otherwise

141

142

Skill in Ancient Ethics

nakedly exposed animals (‘unshod, unclad, unarmed’)1 and implacable elements or potentially overpowering forces in their environment. Combining ingenuity and artifice, techne was virtually definitive of human beings, essential to their very survival no less than to many of their highest and most characteristic achievements. Already established in disparate enclaves of expertise – e.g. housebuilding, weaving, cobbling, metallurgy, medicine – and popularly esteemed for the manifest and varied benefits it delivered to the life-world, it perhaps inevitably attracted the philosophical attention of Plato and Aristotle.2 They fixed on it as a pivot between knowledge and activity (or, in Aristotle’s stricter usage, the kind of activity – poiēsis – that terminates successfully in an intended product or state-of-affairs). Techne was the source of, and was exhibited in, activity that produced well-wrought outcomes, grounding the activity in knowledge that directs and orders it and can proffer an account of how and why it has been successful. Undaunted by the plurality and heterogeneity of the domains of its exemplification, these Socratic thinkers forged ‘techne’ into a paradigm of practical effectiveness and analytical rigour; and they were strongly inclined to argue that the former was due to and underwritten by the latter. It is this dual aspect of Aristotle’s conception of techne that I want to examine: on the one hand, his keen awareness that production of its specific outcomes is the defining telos of a techne, to which all its activity is geared and by reference to which it is to be evaluated; and on the other hand, his insistence that any maker is a genuine technitēs only if he possesses authoritative knowledge which informs and justifies his movements – and the structure of which, following Plato, he seeks to articulate and vindicate in the concept of techne. I will show how far-reachingly Aristotle envisaged techne as a kind of knowledge – to the point even of bringing it very close to the theoretical mode of knowledge that he most valued. I will also show how techne remains irretrievably practical, engaged with materials and processes through which changes eventuate in intended, previously non-existing outcomes. Moreover, I will try to demonstrate that in Aristotle’s account of techne these two divergent aspects – what might

1

2

De Partibus Animalium (‘On the Parts of Animals’; abbreviated to De Part. Anim. in subsequent footnotes) 4.10, 687a24. Jaako Hintikka suggests that since ‘the activity of the craftsman is likely to have the most concrete and clearly definable product’, the Socratic philosophers were tempted ‘to discuss all and sundry phenomena in terms of their outcomes’, thus leading to the dominance of ‘the paradigm of the craftsman’ (1974: 41–42). Earlier, R.G. Collingwood had made a similar point: ‘Once the Socratic school had laid down the main lines of a theory of craft, they were bound to look for instances of craft in all kinds of likely and unlikely places’ (1938: 17).

Aristotle on Techne: Two Theses in Search of a Synthesis?

143

conveniently, if not altogether happily, be characterized as the ‘cognitivetheoretical’ and the ‘practico-experiential’– are not satisfactorily integrated. The first is evident in what I call his ‘official’ conception of techne – because it is delineated clearly in the passages that contain his most explicit, ex professo treatments of it3 – while the second can be reconstructed from more penumbral remarks, where techne is not itself the object of analysis, but is appealed to as a powerful pedagogical device. This critical re-reading of Aristotle’s texts, I should say, does not provide a superior vantage-point from which to judge him. Rather, the task of understanding the nature of accomplished performance and skilful execution in a whole range of practical endeavours – which was thrown badly off-course by much modern philosophy – still remains a pressing challenge. And work in contemporary philosophy that is best up to meeting this challenge finds a strong resonance in, and has much to learn from, Aristotle.

1. The ‘official’ conception of techne: Cognitive-theoretical emphasis 1.1. Techne’s proximity to theory In Nicomachean Ethics 6.4, the only chapter in the Aristotelian corpus devoted wholly to techne, Aristotle defines it as a ‘hexis meta logou poiētikē’. The hexis is a skilful and knowledgeable disposition in the maker, which is the generative source (archē) of the things he makes. Like phronesis, it is identified as an ‘intellectual excellence’ (aretē dianoētikē) – though, unlike phronesis, it is also a ‘power of opposites’ (a dunamis tōn enantiōn), a capacity to execute bad as well as good results (a doctor’s techne, for example, enables him all the better to poison as well as to heal a sick person, whereas a just phronimos is not similarly capable of unjust actions). The things produced by techne are distinguishable from the necessary and changeless objects of theoretical wisdom (sophia), which are ungenerated (e.g. the heavenly bodies and mathematical entities), and from the changing things of nature (phusis), whose generative source lies in themselves. The technitēs is the efficient cause of made things (poētika), but his hexis is meta logou only insofar as he apprehends the purpose (telos) of these things, and,

3

Especially Ethica Nicomachea (‘Nicomachean Ethics’) [E. N.] 4.6; Metaphysica (‘Metaphysics’) [Meta.] 1. 1 and 2; Meta. 7. 8.

144

Skill in Ancient Ethics

determined by that purpose, the form (eidos) that he is to induce in the right kind of material (hulē) in order to produce them. To possess a techne is to be capable of bringing these causal or explanatory factors (aitiai) together under one’s rational control, in order to generate products. Techne is, then, the source of a maker’s mastery of his trade, that is to say, his ability not only to accomplish a successful result, but, in doing so, to give an account of his procedures, an account that is rational insofar as it can trace the product back to the aitiai to which it owes its being (‘technitai are wiser . . . in virtue . . . of having the logos for themselves and knowing the aitiai’).4 I will examine more closely techne’s concern with forms and explanatory principles, and argue that – despite its designation as practical, or, more strictly, productive knowledge – this concern makes it appear strikingly theoretical. Before doing so, however, I note two other respects in which techne approximates to ‘theory’. First, while Aristotle’s notion of phronesis departs substantially from Plato’s use of that term, his notion of techne remains very close to Plato’s; and, far from regarding it as merely practical in any sense that might imply inferiority, Plato saw techne as a leading paradigm or prefiguring of the kind of true epistēmē (knowledge) which the philosopher himself seeks. This proximity to epistēmē recurs in Aristotle’s texts, as for instance when he says that ‘if a man does wish to become a master of a techne or epistēmē, he must go to the universal and come to know it as well as possible’.5 If one prescinds from the ontological status of their respective objects (necessary/ungenerated and contingent/generated) and from the cultural privileging of leisurely over productive pursuits (related, in the first chapter of the Metaphysics, to the birth of philosophy itself), it is hard to see a significant difference between ‘epistēmē’ and ‘techne’. Indeed Aristotle often uses the terms interchangeably. Medicine, one of his most frequently cited examples of a techne, is sometimes discussed as an epistēmē.6 Conversely, mathematics and especially geometry, though paradigm epistēmai, are sometimes regarded as technai.7 And one can find medicine and geometry regarded as both technai and epistēmai.8 Moreover, Aristotle does not scruple to speak of ‘productive epistēmai’ and ‘theoretical technai’. The latter expression gestures not to the point I have just been making – that techne resembles epistēmē as a form of knowledge – but rather to the fact that even within epistēmē itself there is a

4 5 6 7 8

Meta. 1.1, 981b5–6. E.N. 10.9, 1180b20–22. E.N. 1143a3–4. Meta. 981b23–4; Topics 170a30. Rhetoric [Rhet.] 1355b28–31.

Aristotle on Techne: Two Theses in Search of a Synthesis?

145

techne at work, i.e. logic, which governs reasoning and produces valid arguments. And we should remember that when Aristotle surveys, classifies, and systematizes in eminently theoretical fashion all the elements, kinds, excellences and distortions of speech-making, he names his treatise on the subject Technē Rhētorikē. The second respect in which techne lies in the neighbourhood of theory arises from the fact that Aristotle conceives of it not only in its own immediate domain as the rationally causative source of production but also as the primary model in terms of which we can grasp the intelligibility and purposiveness already existing in the natural world. Nor is nature the only explanandum in relation to which Aristotle turns to techne. More encompassingly, techne is the original matrix from which he derives his overall conceptual schema, core elements of which it exhibits more perspicuously than anything else: the ‘hylomorphic’ distinction and relation between form and matter as elements within the wider explanatory frame of ‘the four causes’; the distinction and relation between potency (dunamis) and actuality (energeia); and the distinction between activities that are ends in themselves (entelecheiai) and processes that are instrumental to other ends (kinēseis that are ateleis). Carrying all this conceptual freight, then, techne is pervasively present as an explanatory device across the range of Aristotle’s writings, not only in ethics but also in physics, biology, psychology and metaphysics. While its deployment is most prominent in the case of phusis (nature), we find it also in accounts of, for example, various kinds of motion or change, animal morphology, sexual reproduction and life or soul (psychē).

1.2. Techne as architectonic I return now to consider techne at closer quarters in its distinctive role as poiētikē, i.e. concerned with making or production. Aristotle distinguishes between genuine technitai and mere handymen who can produce things while relying only on experience, and so ‘know that the thing is so, but do not know . . . the why and the cause’ (their production is without [aneu] rather than with [meta] logos, viz. a rational account). This distinction can also be understood to divide those who can from those cannot reach ‘one universal judgement about similar objects’. For, as he explains, ‘to have a judgement that when Callias was ill of this disease this did him good, and similarly in the case of Socrates and in any individual case, is a matter of experience; but to judge that it has done good to all persons of a certain constitution, marked off in one class . . . this is a matter of

146

Skill in Ancient Ethics

techne’.9 Moreover, knowledge of ‘the why and the cause’, and of universal categorizations rather than particular instances, enables technitēs – but not the mere handyman – to teach his techne to others. As we have seen, techne enables the accomplishment of a successful result, and the proffering of a rational account (each corresponding, respectively, to poietikē and meta logou in the definition of techne in E.N. 6.4). But Aristotle allows that these two elements can come apart and, more significantly, he seems to concede that, when they do, it is the account rather than the result that is more essential to techne: With a view to action, experience seems in no respect inferior to techne, and we even see men of experience succeeding more than those who have theory without experience. (The reason is that experience is knowledge of individuals, techne of universals, and actions and productions are all concerned with the individual; for the physician does not cure a man, except in an incidental way, but Callias or Socrates or some other called by some such individual name, who happens to be a man. If then a man has theory without experience, and knows the universal but does not know the individual included in this, he will often fail to cure; for it is the individual that is to be cured).10

Aristotle acknowledges here that production is always of individuals, while at the same time asserting that techne is knowledge of universals. He thus leaves us with two elements, both of which, by his own account, are unsatisfactory: mere theory without experience, and mere experience without a grasp of supervening, and thereby explanatory, universals. Both elements are unsatisfactory, just because they remain separate from each other; and Aristotle does not present a conception of techne in which they are properly conjoined. It is telling, then, that in E.N 6 we find a discussion of phronēsis which, while repeating a comparable separation with the aid of an example, affirms the need for precisely the kind of joining of universal and particular that we have just found wanting in the account of techne in Metaphysics 1.1. Having affirmed ‘that those who have experience’ are ‘more practical than others who know’, Aristotle goes on: [F]or if a man knew that light foods are digestible and wholesome, but did not know which sorts of meat are light, he would not produce health, but the man who knows that chicken is wholesome is more likely to produce health. Now

9 10

Meta. 1.1, 981a5–12. Meta. 1.1, 981a12–24.

Aristotle on Techne: Two Theses in Search of a Synthesis?

147

phronēsis is concerned with action; therefore one needs both kinds of knowledge, but particularly the latter’.11

It is worth spelling out the import of Aristotle’s example in this passage. Mere theory is arid because it apprehends universals without retaining and being able to retrieve the many particulars included under them. Mere experience retains particulars, but only as aggregated and retrievable one by one, without the benefit of a higher-order apprehension that would relate them to each other through a conceptual grasp of what they have in common. But there is a third possibility, actuated by phronesis: one might know not only that light foods are wholesome, but know this in the experiential and inclusive way that ensures that one also thereby knows that chicken is a light food and therefore wholesome. (Close to Aristotle’s own usage, the issue might be put in syllogistic terms: theory apprehends the major premise, experience apprehends the conclusion – though not qua conclusion – while phronesis provides the power to connect both of these via an apprehension of the minor premise.)12 Reflecting on the last sentence in the above passage about phronesis – and despite the irony of Aristotle’s treating as a phronetic example what is in fact a medical, i.e. techne-based one – we do not find any comparable sentence about techne (‘now techne is concerned with making; therefore one needs both kinds of knowledge, but particularly the latter’). Aristotle not only fails to offer us any such thought, but, on the contrary, offers a contrast that explicitly deprives techne-informed poiēsis of the rich experiential background that he unhesitatingly grants to phronesis-informed praxis. In E.N. 2.4, he writes: the products of the technai have their goodness in themselves, so that it is enough that they should have a certain character, but if the acts that are in accordance with the virtues have themselves a certain character it does not follow that they are done justly or temperately. The agent must also be in a certain condition when he does them: in the first place he must have knowledge, secondly he must choose the acts and choose them for their own sakes, and thirdly his actions must proceed from a firm and unchangeable character. These are not reckoned in as conditions of the technai except the bare knowledge (plēn auto to eidenai).13

11 12

13

E.N. 1141b16–22. Kant’s famous dictum also comes to mind here: ‘concepts without percepts are empty; percepts without concepts are blind.’ E.N. 2.4, 1105a27–b5 (emphasis added).

148

Skill in Ancient Ethics

Why is techne reduced to ‘bare knowledge’, disconnected from any dispositional ‘set’ in the technitēs that would correspond to the set so readily recognized in the virtuous agent? To be sure, the perfection which is the telos of a techne resides in the thing made (whereas virtue is an immanent perfection of the agent himself). Still, apart from the order of being that prevails after the productive process has run its course, why should Aristotle’s consideration of the exercise of a techne in this process itself – in the order of becoming – focus only on the knowledge of the technitēs, while discounting that hinterland to knowledge, by way of purposiveness, fixity of disposition and secure accomplishment, that has been gained through experience and repeated practice?14 We will be better placed to answer this question if we analyze more closely Aristotle’s wider approach to genesis and becoming – and, as a prelude to this, his concern with form as the crucial aitia or explanatory principle in his conception of techne: From techne proceed the things of which the form is in the soul . . . e.g. . . . health is the formula and the knowledge in the soul. The healthy subject, then, is produced as the result of the following train of thought; since this is health, if the subject is to be healthy this must first be present, e.g., a uniform state of the body, and if this is to be present there must be heat; and the physician goes on thinking thus until he brings the matter to a final step which he himself can take. Then the process from this point onward, i.e., the process towards health, is called a ‘making’. Therefore it follows that, in a sense, health comes from health and house from house, that with matter from that without matter; for the medical techne and the building techne are the form of health and of the house . . . Of productions and movements one part is called thinking and the other making, – that which proceeds from the starting point and the form is thinking, and that which proceeds from the final step of the thinking is making . . .The active principle then and the starting point for the process of becoming healthy is, if it happens by techne, the form in the soul.15

In this passage, Aristotle identifies techne with form, and claims that implicit in knowledge of this form is a whole sequence of inferential steps that a technitēs can explicate, starting from the most generic and leading ultimately to a 14

15

Aquinas’s Aristotle-inspired discussion of prudentia (phronēsis) and ars (techne) makes this point explicit. Aquinas speaks of inquiry (consiliari) and judgement (judicare) as acts common to prudentia and ars – and to theory too (speculativa ratio). But, he adds, prudentia proceeds to commanding (praecipere) which brings about execution (applicatio), while ars does not so proceed: its perfection ‘lies in judging and not in commanding’, so that it is – as prudentia is not – ‘sola consideratio’ or ‘solum cum ratione’. See Summa Theologiae 2a 2ae, q.47, a.1 and a.8. Meta. 7.7, 1032a32–b23.

Aristotle on Techne: Two Theses in Search of a Synthesis?

149

specification of the most immediate action on his part that will, in turn, lead to a series of further actions (in reverse order to the prior series of inferential steps). This issues finally in the made thing, which will now instantiate materially the form that had first to be present immaterially in the soul of the technitēs. Aristotle thinks of this framework as applying to end as well as form (for in an important respect the end is the form): ‘The end aimed at is, then, the starting-point of our thought, the end of our thought the starting-point of action’.16 This is a truly architectonic conception of techne; through his knowledge of the form (design, blueprint), the maker possesses the archē which can then guide the whole process from beginning to end. Or rather there are two processes: first, noēsis, which is an analytic process that transpires in the technitēs (in his ‘mind’, as it must seem), unfolding a whole plan that specifies exhaustively and in right order the action-steps that are then to be implemented in the second, subsequent and quite separate process of poiēsis or actual making. What is striking here is the comprehensive and secure purchase on the making process that is already inscribed in techne as a knowledge-process, and the linearity and unidirectionality of the relationship between both processes. There is no suggestion that something might arise in the making that has not been anticipated in the prior analysis, rendering the latter susceptible to feedback that could revise or redirect it. There is no place for experimental activity that, though indeed thinking-informed, is designed to inform further thinking. A clean separation of ‘end’ and ‘means’ is assumed, the end being already available as ‘starting point’, so that the only issue is one of working out the means towards its effective achievement (‘we deliberate not about ends but about means’).17 There is no hint of a bottom-up kind of deliberation in which the end itself is at stake, it being yet to be determined what in this case or in these circumstances would count as an optimal or acceptable or least worst outcome. What we find here is all of a piece with Aristotle’s belief that nothing which arises in the process of generation, whether in nature or by techne, can be credited with a substantial explanatory role. He insists that explanatory sources always already pre-exist the process of generation itself: ‘in housebuilding . . . these things come about because the form of the house is such and such, rather than its being the case that the house is such and such because it comes about thus’. He criticizes Empedocles for supposing that ‘many of the characters presented by animals were merely the results of incidental occurrences during their development (for instance that the backbone is as it is because it happened to be

16 17

Ethica Eudemia (‘Eudemian Ethics’) [E.E.] 2.2, 1227b32–3. E.N. 3.3, 1112b34–13a1.

150

Skill in Ancient Ethics

broken owing to the turning of the foetus in the womb)’. Against this supposition he looks for explanation to ‘a creative seed endowed with certain powers’ and ‘the parent animal [who] pre-exists. For man is generated from man’. And he goes on: the ‘same statement holds good also . . . for the products of techne [epi tōn techastōn to poiētikon] . . . [where] the agent [to poiētikōn] is pre-existent, such as the statuary’s techne’.18 Excluded here in principle is any emergence or creativity, any element of novelty that was not already fully transparent to the technitēs prior to actual engagement with his materials. Aristotle’s aversion to granting any significant status, in the case of animal evolution, to ‘incidental occurrences during their development [genesis]’ is continuous with his reluctance to acknowledge that there might be a genuinely experimental element in the process of poiēsis. From his perspective, to bring an irreducibly experiential component of techne more into focus (as I wish to do) is to introduce just that kind of haphazardness which he criticizes in Empedocles – where the disposition of a thing (in Empedocles’ case, a backbone, in the case of a techne, let us say a watercolour painting) is ‘indebted to’ events that occur unpredictably in its gestation (in the former case in the womb, in the latter case in the meetings of canvas, paints, brush, and a hand charged with perceptiveness and imagination). I have been quarrelling with Aristotle’s conception of techne as too formal (in the literal sense), as enshrining knowledge that is both too disembedded from a whole experiential background in the technitēs and too aloof from the nuts and bolts of the actual productive process. This conception, one might say, erases history: unanticipated actions or reactions entering into any discrete process of production; many accumulated experiences somehow coalescing in the biography of the individual qua accomplished technitēs; and, for good measure, the evolution of any techne as a communal repository or archive of much cumulative, transgenerational, learning. There is indeed a logos to all this; but it has a narrative structure that one could scarcely guess from Aristotle’s account of techne.

2. Techne as practico-experiential 2.1. Deviant technai: Dealing with chance (tuchē) and opportunity (kairos) I want to go on now to identify some elements that hover at the edges, so to speak, of Aristotle’s analysis of techne, remaining unassimilated by his ‘official’ 18

Quotations here are from De Part. Anim. 1.1, 640a19–31.

Aristotle on Techne: Two Theses in Search of a Synthesis?

151

conception. A helpful way to begin is by attending to the diverse range of examples in his treatment of ‘techne’ and noting the awkwardness they betray in relation to techne’s alignment with the process of making or production (poiēsis). The paradigm element in ‘poiēsis’, distinguishing it from praxis (with which phronēsis is aligned), is its termination in a separate thing, subsequent and external to it as an activity or process. This is the classic mould of fabrication, but it does not fit, for instance, activities that Aristotle routinely countenances as technai, even though they are performative rather than productive, transpiring in the very doing, while leaving no durable product behind – for example, fluteplaying or gymnastics. I leave these aside, however, in order to focus on another set of activities that deviate differently from the mould of fabrication. These other activities – for example, military strategy, navigation, rhetoric and, not least, medicine – do indeed aim at and terminate in an end-result. But this result is more a state-of-affairs than a substantial product and, rather than having disposable materials on which he can impress a pre-conceived form, the technitēs is more readily thought of as intervening in a field of forces, or as immersing himself in a medium, in which he seeks to accomplish a propitious end. While these activities do not attract Aristotle’s analytic attention as a distinctly identifiable sub-class, from some passing remarks about them we may take it that their most characteristic feature is a close relationship – not to be ascribed to the straightforwardly productive technai – with the opportune (ho kairos) and with chance or luck (tuchē). Chance might be placed at the opposite pole from techne – as it is, for example, by one of the Hippocratic authors: ‘they did not want to look into the naked face of chance, so they turned themselves over to techne’.19 Or, as Aristotle himself puts it, ‘those occupations are most truly technai in which there is the least element of chance’.20 This seems to imply that some craft-activities, because they involve a greater element of chance, are less truly technai. Such indeed seems to be Aristotle’s thought when he says that even foolish people are often successful in matters controlled by chance, a fact illustrated even ‘in matters involving techne [but into which] chance largely enters, e.g. strategy and navigation’.21 What emerges, then, is not a simple polarity between techne and tuchē but rather a kind of sliding-scale along which different technai are more or less exposed to chance. Some technai gain their ends not by overcoming chance but rather by confining their operations to spheres in which it can gain no foothold. That the 19 20 21

Quoted in Nussbaum 1986: 89. Politics 1.11, 1258b35. E.E. 1247a5 –7.

152

Skill in Ancient Ethics

technai of fabrication, apparently congenial to the account of techne in E.N. 6.4, are of such a kind, is suggested by a remark in the Analytics: ‘Among the products of thought, some never occur spontaneously – e.g. a house or a statue . . . but with some aim; but others occur by chance too – e.g. health and preservation’.22 Here we find grounds for saying that some technai, by setting for themselves an end which can be attained through a circumscribed process within their direct control, make themselves invulnerable to chance; others, however, are involved in areas where, because they are confined by no fixed limit (peras), the play of chance is ineliminable. Being subject to chance, these technai cannot aspire to the same kind of mastery that obtains in the others. Success is achieved in them not so much by keeping one’s gaze fixed on the preconceived form to be imposed on the material, as by a flexible kind of responsiveness to the dynamism of the material itself. Sensitivity and attunement, rather than mastery or domination, would seem to be the key requirements; actions may have to be quick and decisive, but they arise within a pattern of receptivity. This is the meaning of grasping the kairos: one’s active intervention has skilfully waited until one’s volatile, polyvalent materials – be they the wind and waves at play on one’s boat, the swaying emotions of a crowd, or the changing humours in a sick body – are at their most propitious, i.e., are most able to help, or least liable to hinder, the accomplishment of one’s end – a safe journey, a convinced audience, a recovered patient. If chance is one’s enemy in these technai, then one succeeds not by defeating her but rather by winning her over – a point suggested by Aristotle’s quotation from Agathon: ‘techne loves tuchē and tuchē loves techne’.23 Tuchē favours the technitēs who pays court to her by responding craftily to the more or less capricious motion of materials that lack the stability of stone or wood, while compensating for his inability to provide an accurate prespecification of the route to his end (the pros to telos) by an alert and agile intelligence, capable of improvisation and a ready embrace of opportunity. We seem to be close here to the zone of mētis or cunning intelligence, ‘excavated’ by M. Detienne and J.-P. Vernant as a zone that does not figure in ‘what Greek intelligence has to say about itself when it composed theoretical treatises on its own nature’, but must rather ‘be tracked down . . . in areas which the philosopher usually passes over in silence’. Mētis combines ‘flair, wisdom . . . subtlety of mind, resourcefulness, vigilance, opportunism . . . [and is] applied to situations which are transient, shifting, disconcerting and ambiguous, situations

22 23

Posterior Analytics 2.11, 95a3–6. E.N. 6.4, 1140a20.

Aristotle on Techne: Two Theses in Search of a Synthesis?

153

which do not lend themselves to precise measurement, exact calculation or rigorous logic’.24 (The ‘wily’ Odysseus, described by Homer as polumētis – i.e. equipped with many stratagems for dealing with unforeseen situations – might be taken as the prototypical figure here.) Although Detienne and Vernant claim that mētis is generally suppressed or ignored by the philosophers, they contend – without trying to demonstrate – that it comes closest to showing up in Aristotle’s analysis of phronesis. While supportive of this contention about phronesis, I want to extend it to the philosophically orphaned technai under consideration here: The accounts we demand must be in accordance with the subject-matter; matters concerned with conduct and questions of what is good for us have no fixity any more than matters of health. The general account being of this nature, the account of particular cases is yet more lacking in exactness; for they do not fall under any techne or set of precepts but the agents themselves must in each case consider what is appropriate to the occasion, as happens also in the techne of medicine or navigation.25

In laying it down here as a methodological principle that the matter (hulē) determines the kind of account (logos) that can be given, Aristotle tells us that, because praxis supplies the matter of ethics, the logos of an ethicist (Aristotle himself) can be given only roughly and in outline (tupōi kai ouk akribōs). So multiform and heterogeneous are the materials and circumstances of praxis that no systematic body of knowledge can adequately comprehend them and, when he later comes to consider the orthos logos that regulates them, the best he can do, faced with the impossibility of giving it a determinate content, is to define it obliquely as that which the phronimos would determine. Having declared in the passage above that matters of praxis fall under no techne, Aristotle immediately points out the affinity between the regulation of these matters and a medic’s or steersman’s way of dealing with their respective matters. This affinity is close enough to suggest that the latter do not possess a techne at all. For, no more than in ethics, there are no general rules that would exempt producers from the ever-recurring requirement that they themselves think out what is appropriate to the occasion or, literally, consider that which conduces to the kairos (ta pros to kairon skopein). Aristotle insists that only general knowledge of classes of things comes within the remit of techne (i.e. is entechnon), and tells us – referring specifically to

24 25

Detienne and Vernant 1978: 3–4. E.N. 2.1, 1104a2–10.

154

Skill in Ancient Ethics

rhetoric and medicine -– that ‘[n]one of the technai considers individual cases . . . [because] individual cases are so infinitely various that no knowledge of them is possible [they are apeiron kai ouk epistēton]’.26 Here he seems to limit techne to the ideal type of a domain where exhaustive guidance is available from already formulated rules, an example of which he finds elsewhere in the established, unappealable conventions of a language: ‘in the case of exact and self-contained sciences there is no deliberation, e.g. about the letters of the alphabet (for we have no doubt about how they should be written); but the things that are brought about by our own efforts, but not always in the same way, are the things about which we deliberate, e.g. questions of medical treatment or of money-making. And we do so more in the case of navigation than in that of gymnastics, inasmuch as it has been less exactly worked out (diēkribōtai)’.27 ‘[N]ot always in the same way’: a techne that deals with such variable matters is more exposed to the apeiron kai ouk epistēton. And technai vary to the extent to which they are thus exposed – money-making, navigation, and medicine being conspicuously among the most exposed – and, for that reason, noticeably running athwart the kind of formal deliberation that I criticized in the previous section.

2.2. Techne-poiēsis: An unassimilated remainder I have just considered a group of non-fabricating technai that are out of sync with Aristotle’s official conception of techne. And I had already argued that this conception fails to capture much of what actually goes on – especially by way of an experiential background to their exercise – even in those technai that are poiētikai in the strict sense that they involve fabrication of durable products (e.g. housebuilding, cobbling, sculpting). In what follows, however, I want to suggest that, even though this background slips through the mesh of his official conception, we can still find recognitions of it, in unlikely places, in Aristotle’s texts. A sticking point up to now has been the claim that ‘in a sense health comes from health and house from house, that with matter from that without matter; for the medical techne and the building techne are the form of health and of the house’.28 But consider his different emphasis when, apropos a bronze circle as a poiēton, Aristotle asks: ‘is the matter an element even in the logos?’ and answers affirmatively: ‘The bronze circle has its matter in its logos.’29 If this is so, must the 26 27 28 29

Rhet. 1.2, 1356b30–34. E.N. 3.3, 1112a33–b6. Meta 7.7, 1032b11–14 (emphasis added). Meta. 7.7, 1033a1–5 (emphasis added).

Aristotle on Techne: Two Theses in Search of a Synthesis?

155

matter not also be present in the logos of the techne as archē of the poiēton, i.e. the logos which appears in the definition of techne as a hexis meta logou poiētikē? In a passage in the Eudemian Ethics, Aristotle grants that Socrates was consistent in asking ‘what excellence is, not how or from what it arises’, because for him ‘to know justice and to be just came simultaneously’. But for Aristotle himself, as he goes on to make clear, this reliance on knowledge cannot hold regarding ethical conduct or productive activity: with regard to both of these, ‘not to know what it is but to know out of what it arises is most precious’.30 That Aristotle, in speaking of ‘how or out of what it arises’, is pointing to the material seems to be confirmed by a passage in the Physics where he distinguishes technai that use and commission products (e.g. sailing and horsemanship) from those that make them (e.g. boat-making and saddle-making). The former ‘is in a sense directive; but it differs in that it knows the form, whereas the techne which is directive as being concerned with production knows the matter . . . from what wood it should be made and by means of what operations’.31 The point made here – and the sense in which ‘the matter is in the logos’ – is illuminated in one of Aristotle’s biological treatises. A hand has been carved in wood, and the question is asked: ’What are the forces by which the hand . . . was fashioned into its shape? The woodcarver will perhaps say, by the axe or the augur’. While commending this answer, Aristotle goes on: ‘it is not enough for him to say that by the stroke of his tool this part was formed into a concavity, that into a flat surface; but he must state the reasons why he struck his blow in such a way as to effect this, and for the sake of what he did so; namely, that the piece of wood should develop eventually into this or that shape’.32 This passage is especially illuminating because it brings us unusually close to the nitty gritty of the actual making process, and because it sets the logos of techne within an interlocutory context where the technitēs enunciates this logos in response to questions put to him. And clearly, such questions as ‘why he struck his blow in such a way’ can be put only by someone who is present with him, closely observing his moves as he goes about his task. Here we are reminded of the passage on equity in E.N. 5.10, which asserts that the logos of the aretē (virtue) of justice cannot be captured in the written law, which is always universal and therefore in need of a kind of correction that can be made only by someone who ‘is present there’ in the very situation of action;

30 31 32

E.E. 1.5, 1216b7–23. Physics 2.2, 194b1–7. De Part. Anim. 1.1, 641a8–14.

156

Skill in Ancient Ethics

and this person can formulate the equitable (i.e. the ‘correction of law where law is defective on account of its generality’) only in a special ordinance, a logos that can reckon with the indefiniteness (to aoriston) which belongs to the situation itself. (It is here, fittingly, that Aristotle invokes the image of the ruler used by builders on Lesbos which, being made of lead rather than wood, had the pliability to measure irregularly shaped stones.)33 Similarly, I am suggesting, the logos of techne as an intellectual virtue lies not so much in general formulae as in specific accounts always measured to particular acts of production; it is the ability reliably to produce such accounts that is the hexis meta logou of techne. Moreover, it is telling that the factor singled out in E.N. 5.10 as the cause of the defect in universalized law – and hence of the need for equity as its correction – resides ’neither in the law itself, nor in the lawmaker’, but rather in ‘the material of action’ (hē tōn praktōn hulē).34 The analogy to which I am drawing attention here is affirmed by Aristotle himself just before his discussion of equity. He is arguing against the view that acting justly ‘lies easily within people’s power’ (in the sense that a just man could effortlessly ‘lie with his neighbour’s wife . . . or slip someone a bribe’). What this view ignores, Aristotle thinks, is precisely the stable disposition that lies behind our actions, gives them their moral stamp, and is not so easily within our power. To show this, he reaches easily for a medical, i.e. techne-based, analogy: one might as well say that ‘to know [eidenai] what honey, wine and helibore, actuary and surgery are’ is easy. Indeed, such knowing is easy, but for the task of being a doctor it is not enough; one is not a doctor unless one knows ‘how, to whom, and when to administer [honey etc.] with a view to producing health’.35 Aristotle’s phrase here for the factor which is over and above knowledge of medicine is to hōdi echonta, translated as ‘a certain disposition of mind’ (Rackham) or ‘a certain state of character’ (Ross); it seems to be just that experiential hinterland to knowledge which was credited to virtue but not to techne in the passage from E.N. 2.4 that we met earlier (and which confined techne to mere knowledge [eidenai]) but is here attributed both to the virtuous person and to the good technitēs. In fact, a well-known passage in E.N. 2.1 contains an implicit acknowledgement of the experiential background that I am trying to disclose as essential to techne; unsurprisingly, it is a passage where Aristotle affirms a close affinity between

33 34 35

E.N. 5.10, 1137b29–32. See E.E. 5.10, 1137b12–27. E.E. 5.9, 1137a5–24. The formula ‘how, to whom and when’, used here in reference to the medical techne, echoes what is said in an ethical context about the difficulty of getting angry with ‘the right person, to the right extent, at the right time . . . and in the right way’ (E.N. 2.9, 1109a27–9).

Aristotle on Techne: Two Theses in Search of a Synthesis?

157

techne and character-aretē. Contrasting both of these with a sense faculty, he points out that – while we already possess (e.g.) sight, and then exercise it – we do not possess a techne (or a virtue) until we first learn it, and learn it precisely by doing it and by doing it well: ‘men will be good or bad builders as a result of building well or badly’. And when he goes on to say that ‘if this were not so, there would have been no need of a teacher’,36 it seems clear that what teaching involves here is not transmission of a body of pre-formulated knowledge but rather supervision of an apprentice engaged in graduated tasks, learning the hard way, by dint of repeated practice, as his mistakes are corrected and he develops the ability to self-correct, thereby internalizing the standards proper to his new trade. Such a view of teaching was indeed already hinted at in the opening contention of Book 2 that intellectual virtue – which is how techne will later be classified – ‘is for the most part both produced and increased by instruction, and therefore requires experience and time’.37

2.3 Naturalizing techne Aristotle famously says that ‘techne imitates nature’ (technē mimeitai tēn phusin). While this dictum grants ontological primacy to nature, from Aristotle’s whole handling of the relationship between them it is clear that in the order of explanation techne takes the lead – it provides the conceptual resources in terms of which nature itself is to be understood.38 One may still wonder, however, whether the direction of explanation runs completely one way – or might nature, and the fact that it remains in some sense the primordial model, tell us something about techne that has not come to light so far in our discussion? In Physics 2.8, Aristotle suggests that ‘[i]f a house, e.g., had been a thing made by nature, it would have been made in the same way as it is now by techne; and if things made by nature were made not only by nature but also by techne, they would come to be in the same way as by nature’.39 And a little later, apropos the immanent teleology of nature, he remarks: ‘It is absurd to suppose that purpose is not present because we do not observe the agent deliberating. Techne does not deliberate. If the ship-building techne were in the wood, it would produce the 36 37 38

39

E.N. 2.1, 1103b6–13. E.N. 2.1, 1103a14–16. This fact is scarcely surprising if one understands Aristotle to have transposed the demiurge of Plato’s Timaeus into nature, thus rendering the latter intrinsically demiurgic. Or, as Solmsen points out: since ‘Aristotle’s entirely new concept of nature . . . has been fathered . . . by craft it is not astonishing that [he] . . . persuades himself that craft follows the pattern of nature’ (1963: 488). Physics 2.8, 199a12–15.

158

Skill in Ancient Ethics

same results by nature.’40 These sentences encourage us not to be so impressed by the intervention of the technitēs as to suppose that he inaugurates a new realm that is entirely different from phusis. Mindful that ‘the deficiencies of nature are what techne . . . seek[s] to fill up’,41 we might rather envisage this intervention as but a strategic detour through which nature goes in order to bring about a new class of things (artefacts) which it cannot produce or reproduce through its own channels. And Aristotle goes some distance to assure us that the terrain (of techne) through which this detour passes is not a very alien one when he remarks – contrary to the whole thrust of the ‘official’ conception – that ‘even techne does not deliberate’. When the idea of techne as imitating nature is taken seriously, the heavy emphasis on planning in the mind of the technitēs prior to any engagement with the material (‘techne indeed consists in the account of the product without its matter’)42 recedes from view. What is suggested instead is a certain ‘natural’ fluency in accomplishing results through a process in which noēsis and poiēsis are not discrete, linear sequences but are rather interwoven in one process which is badly understood in terms of any separation of mind and materials. In fact, in his account of change or process (kinēsis), Aristotle prioritizes the material over the mind of the technitēs as the proper site of technazein. To be sure, when men build, they actualize their own dunamis qua technitai; but this actualization properly resides in the thing built and not in themselves.43 Whereas in the case of sight ‘the ultimate thing is seeing, and no other product besides this results’, ‘from the techne of building there results a house as well as the act of building’.44 But ‘as well as’ (para) here is misleading; it does not capture the internality, indeed the identity, of act and product: ‘for the act of building is the thing that is being built, and comes to be – and is – at the same time as the house. Where, then, the result is something apart from the exercise, the actuality is in the thing that is being made (hē energeia en tōi poioumenōi estin), e.g. the act of building is in the thing that is being built . . . and similarly in all other cases, and in general the movement is in the thing that is being moved’.45 40 41 42 43

44 45

Physics 2.8, 199b27–9. Politics 1337a1–2; see also Physics 2.8, 199a16–17. De Part. Anim. 1.1, 640a31–2. This is indeed a basic point about technai, for Plato as for Aristotle: that they exist for the sake of the product and not of the producer. Indeed, it is precisely because they fulfil well their productive function that they also thereby fulfil the technitēs – a point nicely made by Aristotle in the unlikely context of his discussion of friendship: ‘existence [to einai] is to all men a thing to be chosen and loved .|.|. and . . .we exist by virtue of activity [energeia] . . . and . . .the handiwork is, in a sense, the producer in activity; he loves his handiwork, therefore, because he loves existence . . . for what he is in potentiality, his handiwork manifests in activity’ (E.N. 9.7, 1168a6–10) . Meta. 9.8, 1050a24–6. Meta. 9.8, 1050a23–34.

Aristotle on Techne: Two Theses in Search of a Synthesis?

159

Aristotle finds an ingenious way of dealing with the conundrum that seems to follow from his position here – e.g. that the act of teaching and the act of learning are the same (so that a teacher would always have to learn in the very act of teaching). This does not follow ‘any more than it follows from the fact that there is one distance between two things which are at a distance from each other, that being here at a distance from there and being there at a distance from here are one and the same. To generalize, teaching is not the same as learning, or agency as patiency, in the full sense, though they belong to the same subject, the motion (kinēsis)’.46 Sarah Waterlow has offered a bold way of understanding what Aristotle is driving at here. Although the producer and material exist separately before and after the event of change, she suggests that in this event we have no reason to suppose them separate – ‘any more than the fact that different simple bodies are yielded up when an organism decomposes would be, for Aristotle, a ground for saying that they were actually present as their distinct selves during the life of the creature’. Waterlow suggests that the cause of our resistance to conceiving of change in this way is our bewitchment by ‘the structure of the concepts we use to describe our own practical activities’. We need to suspend precisely ‘the point of view of the voluntary agent’ – which so easily separates an in-the-mind planning from an in-the-material implementation – and to adopt instead an organic, i.e. natural, parallel. Echoing my own discussion a little earlier, Waterlow writes: ‘although Aristotle’s analogies . . . between natural and artificial change are intended to illustrate the former by the latter, why should we not reverse the analogy and regard the artifex and his material as forming, in the change, a concrete organic unity, as if the material were an extension of his own body?’47 There is of course good reason to suppose that, as an originative source (archē), explanatory principle (aitia), active power (dunamis) and habitual propensity (hexis), techne should be seen to exist precisely before and not in any actual process or event of production. And this fact is closely related to Aristotle’s emphasis on the definitive role of the form in the soul of the technitēs and to his insistence that, for example, ‘the techne of healing produces health without itself being acted upon in any way by that which is being healed’, because it is one of those ‘active powers . . . whose forms are not embodied in matter’.48 Going with this emphasis and insistence is a one-way, top-down conception of a techne’s

46 47 48

Physics 3.3, 202b16–21. Quotations here are from Waterlow 1982: 201–203. De Generatione et Corruptione (‘On Generation and Corruption’) 1.7, 324a35–b5.

160

Skill in Ancient Ethics

modus operandi, akin to what we have already seen: ‘it is his hands that move his tools, his tools that move the material; it is his knowledge of his techne, and his soul, in which is the form, that move his hands’.49 While soul (and immaterial form) seems the main basis for the mentalist conception of techne with which I am quarrelling, in fact it is exactly by examining more closely his conception of soul that we can most effectively undermine this mentalism and disclose resources at Aristotle’s disposal for a more adequate conception of techne.50

2.4. Souled body and embodied techne In De Anima 1.3, Aristotle dismisses earlier attempts to theorize soul independently of its receiving body (dexomenou sōmatos), insisting that there must be some community of nature (koinōnia) between the two; not to comprehend this – that ‘each body seems to have a form and shape of its own’ – ‘is as absurd as to say that carpentry could embody itself in flutes; each techne must use its tools, each soul its body’.51 Just as carpentry cannot conceivably be realized without tools fitted to carry out the functions of cutting, hammering, planing etc., so the soul depends intrinsically for its actualization on a body of a special character. The soul is to the body, as its proper functioning is to a sense organ: ‘Suppose that the eye were an animal, sight would have been its soul, for sight is the substance [ousia] of the eye which corresponds to the account [logos], the eye being merely the matter of seeing; when seeing is removed the eye is no longer an eye, except in name.’52 Similarly, the soul of an axe, were it to have one, would be its power of chopping. And so there is a double analogy: ‘as sight and the power of the tool are [respectively to eye and axe], so is the soul [to body].’53 Soul, then, is essentially embodied; it is that through which the ‘body’ is endowed with all its modulated capacities for movement and action. It is present in the whole body and in each of its parts; making each integral part capable of its specific functioning, it is to the unified human body what sight is to the eye. If, as Aristotle claims, techne is in the soul, then it is (e.g.) in the hand; without the presence of soul, instead of a hand there would be an inert piece of flesh – or 49 50

51 52 53

De Generatione Animalium (‘On the Generation of Animals’) 1.22, 730b16–18. My argument here parallels Gilbert Ryle’s argument, contra ‘the intellectualist legend’, that ‘the exercise of intelligence in practice cannot be analysed into a tandem operation of first considering prescriptions and then executing them’. Rather, ‘[w]hen I do something intelligently, i.e. thinking what I am doing, I am doing one thing and not two. My performance has a special procedure or manner, not special antecedents’. See Ryle 1990: 40; 32. De Anima (‘On the Soul’) 1.3, 407b23–5. De Anima 2.1, 412b18–22. De Anima 2.1, 413a1.

Aristotle on Techne: Two Theses in Search of a Synthesis?

161

rather the rotting remains of erstwhile flesh. The very potentiality for dexterity and coordinated action that makes it a human hand is already due to this presence of soul. And when this raw potentiality is gradually disciplined into actual capacities for specific types of skilled activity, techne is present, we may say, as much in the hand as in the soul; it is what makes the hand of a technitēs different from the hand of an unskilled person. So, when Aristotle says, as we saw above, that ‘it is his knowledge of his techne, and his soul, in which is the form, that move his hands’, we need not think of the soul as an extrinsic agent directing, through its knowledge (contained in the ‘mind’), soulless, knowledgeless hands. Or when he counts technai as ‘active powers whose forms are not embodied in matter’, we may deem him to be speaking only half the truth. For, even if not embodied in a particular material artefact (with whose disintegration it too would cease to exist), techne is still embodied in the souled matter of the human body, enabling it, through repeated actualizations, to go on producing particular, specifically characterized artefacts. And an embodied techne, aligned with an embodied soul, is at the same time an entooled techne – so that we could not even conceive of technai such as surgery, sailing or snaring if we had no concepts of scalpels, sails and nets; as soul is the principle of the body’s functioning, so techne is the principle of the tools’ functioning. To argue for techne as embodied is not to deny the role of knowledge in techne; it is only to clarify the kind of knowledge that is involved and the conditions to which it is subject. And here one further parallel with phronesis shows up. For Aristotle makes it very clear that phronesis is not just a capacity for calculative knowledge (such as mere cleverness [deinotēta])54 but is the hexis meta logou only of a person whose character has been formed in a specific direction (while it, in turn, comes to inform this character). If the knowledge of the phronimos is inseparable from a certain ordering of his passional life, may we not say that, similarly, the knowledge of the technitēs is inseparable from a certain development and attunement of his sensory-motor capacities – and indeed of his passional life too insofar as this must be disciplined by the demands of his techne? Here I must conclude this sortie into the complexities of Aristotle’s treatment of techne. I do so with a pointer to the wider philosophical import of what has been primarily an exercise in textual exegesis. For I hope to have shown that ‘techne’ may be fairly seen as laying down the prototype of the purposive rationality (the ‘instrumental reason’ or ‘technocratic attitude’) that has been so 54

See E.N. 6.12, 1144a24–35.

162

Skill in Ancient Ethics

definitive of the Western mindset. And while taking issue with what I take to be the one-sidedness of this prototype, I have wanted also to suggest that Aristotle’s texts open out in interesting and fruitful ways to contemporary work on embodied, ‘enactivist’, or ‘extended’ mind. But to make good on this suggestion would take us too far from my intended task in the present chapter.55

References Aquinas, St Thomas (1950), Summa Theologiae, cura et studio, P. Caramello. Rome: Marietti. Aristotle (1984), The Complete Works of Aristotle, rev. Oxford translation, in 2 vols., ed. J. Barnes, Princeton: Princeton University Press (Bollingen Series). Collingwood, R.G. (1938), The Principles of Art, Oxford: Oxford University Press. Detienne, M., and J.-P. Vernant (1978), Cunning Intelligence in Greek Culture and Society, trans. J. Lloyd, Brighton: Harvester Press. Dunne, J. (1993), Back to the Rough Ground: Phronesis and Techne in Modern Philosophy and in Aristotle, Notre Dame, IN : University of Notre Dame Press. Hintikka, J. (1974), Knowledge and the Known: Historical Perspectives on Epistemology, Dordrecht: Kluwer. Nussbaum, M.C. (1986), The Fragility of Goodness: Luck and Tragedy in Greek Ethics and Philosophy, Cambridge: Cambridge University Press. Ryle, G. (1990), The Concept of Mind, London: Penguin. Solmsen, F. (1963), ‘Nature as Craftsman in Greek Thought’, Journal of the History of Ideas 24: 473–496. Staten, H. (2019), Techne Theory: A New Language for Art, London: Bloomsbury Academic. Waterlow, S. (1982), Nature, Change and Agency in Aristotle’s Physics: A Philosophical Study, Oxford: Clarendon Press.

55

For a remarkable example of the kind of work I point to here, see Techne Theory: A New Language for Art (Staten 2019), in which Henry Staten elaborates an anti-Romantic aesthetics, engaging Aristotle as indispensable interlocutor in a far-reaching reconstruction of techne. I should perhaps declare an interest, in that Staten acknowledges indebtedness to my Back to the Rough Ground: Phronesis and Techne in Modern Philosophy and in Aristotle (Dunne 1993) – on which I have drawn here in presenting an abridged version of one of its main arguments.

Part Three

Skill in Stoic Ethics

163

164

8

The Craftsman of Impulse: Chrysippus on Expertise and Moral Development1 Tue Søvsø

The Stoic ideal of virtue and wisdom is breathtakingly demanding. The virtuous person never errs and to some extent her mind even mirrors the divine rationality that pervades and governs the universe. Nonetheless, the Stoics insist that this state of right reason (orthos logos) is in principle accessible to every human being thanks, in essence, to the capacity as a rational being to acquire expertise (technē).2 The present paper examines how Chrysippus, the third head of the Stoa and great systematizer of Stoic philosophy, developed these astonishing claims. In particular, I focus on the moral psychology and epistemology behind his confidence in the human capability to reach perfection. Chrysippus, I argue, propounded a theory of moral progress (prokopē) that turns on his conception of reason as a collection of concepts and makes improvement in our patterns of thought and behaviour dependent on the ordering and articulation of these concepts. Such order and articulation are the hallmarks of expertise, and the transition from our initial, naturally emerging preconceptions (prolēpseis) to the infallible conceptions (ennoiai) of the virtuous can therefore be conceived of as a gradual acquisition of expertise.3 1

2

3

This paper has profited greatly from discussions at the conference on ‘Virtue, Skill and Practical Reason’ at the University of Cape Town in 2017, on various occasions at my home universities (Humboldt and Freie Universität, Berlin), and during a research stay at the University of Toronto. I am particularly grateful to Charles Brittain, Brad Inwood, James Allen, René Brouwer and Jula Wildberger for their generous comments on earlier versions of the paper. All remaining errors are my sole responsibility. I translate technē as ‘expertise’. Technikos and similar derivations are translated as ‘technical’ or ‘skilled’ depending on the context. On the role of expertise in Stoic philosophy, see also Long [1983] 1996a: 160–72, Striker 1991: 24–35, Menn 1995, Sellars 2009: 55–59, Brouwer 2014: 41–9, Tsouna 2021. Here and throughout I use ‘concept’ to refer to the genus ‘preconception’, and ‘conception’ for its two species (see further section two below). In discussions of Stoic ontology, ‘concept’ is often used to translate ennoēma, the representational object of a generic concept, while ‘conception’ covers both ennoia and thoughts (noēseis) more generally. To avoid confusion, I translate ennoēma as ‘conceptual object’ and noēsis as ‘thought’.

165

166

Skill in Ancient Ethics

My examination of this process falls into three parts. I start out in section one with a discussion of the so-called theory of oikeiōsis, which sets out the basic psychological processes that govern animal behaviour, arguing that expertise here appears as the distinctively rational means of governing our impulses. In section two, I examine how this administration plays out by looking at the Stoic account of concept formation and the largely neglected evidence relating to the role of expertise in articulating rational conceptions. Here I argue that expertise systematizes our previous experience (empeiria) and thereby articulates and orders our concepts towards a single end. Finally, I apply this account to the description of the natural inclinations (aphormai) that lead to virtue, in order to bring out how expertise informs the articulation of ethical concepts and reason thereby ‘crafts’ our impulses to achieve virtue, ‘the expertise concerning all of life’ (Ar. Did. Stoic. 2.7.5b10).

1. Preliminaries: impulse and right reason In a condensed passage that draws heavily on Chrysippus, Diogenes Laërtius presents the Stoic views on impulse and the goal of life (henceforth simply the telos) in terms of a theory known as the Stoic theory of oikeiōsis.4 Diogenes’ presentation of the Stoic view begins by establishing that, immediately after birth, all animals feel attachment towards themselves and their own constitution. This attachment is a consequence of the animal perceiving itself, a point that the later Stoics Seneca and Hierocles discuss at length, and their discussions suggest that the Stoics attributed to all animals a continuous sense of their own constitution roughly comparable to modern intero- and proprioception.5 The attachment that follows from this self-perception results in a primary impulse (prōtē hormē) to take care of oneself (tērein heauto), which is expressed in impulses to pursue those things that are proper (oikeion) to oneself and to avoid their contraries. Those actions which follow from such impulses are collectively known as appropriate (kathēkon).6 As Jacob Klein has argued convincingly, the primary impulse to care for oneself is best understood as ‘an animal’s disposition to carry out the kinds of activities implicit in its physical constitution’, i.e. a ‘higher order’ impulse that 4 5

6

Diog. Laert. 7.85–86. See Sen. Ep. 121 and Hierocles El. Eth, with Long [1993] 1996b, Brittain 2002: 266–69 and Klein 2016: 172–74. On these actions as the basic ‘appropriate acts’, see Cic. Fin. 3.20.

Chrysippus on Expertise and Moral Development

167

determines the animal’s quite diverse first-order impulses. The animal’s constitution, its primary object of attachment (prōton oikeion), thus provides the ‘template for each of its first-order impulses: a specification of the patterns of behaviour appropriate to a creature of its kind’.7 The primary impulse and the impulses it produces therefore reflect the animal’s awareness of its own constitution, and we correspondingly find Plutarch describing attachment as ‘a perception (aisthēsis) and grasp (antilēpsis) of what is proper’ (Plut. Stoic. repugn. 1038c).8 Diogenes goes on to develop the implications of this natural order for determining the telos for different kinds of living beings. Climbing the Stoic scala naturae, Diogenes observes that plants get along without impulses and perception, while this is what guides animals to seek what is fit for them. The natural life for animals is therefore a life according to impulse. For humans, however, the natural life is a life according to reason (logos) since human reason, as he puts it, ‘emerges as a craftsman of impulse’ (Diog. Laert. 7.86). The most natural way to read this, I think, is to understand that a life according to reason amounts to a life according to impulses that have somehow been shaped or ‘crafted’ by reason.9 In other words: a) all living beings should live in accordance with nature; b) as a species of living beings who are naturally capable of having impulses, animals must therefore live according to impulse; and c) as a rational species of animals, humans should live according to rational impulses.10 Based on this review of the natural order, Diogenes introduces the Stoic view on the telos: ‘Therefore Zeno, in his On the nature of man was the first to say that the goal is to live in agreement with nature, which in turn is to live according to virtue, for nature leads us to virtue’ (Diog. Laert. 7.87).11 Presumably, the way in which ‘nature leads us to virtue’ has something to do with the way in which reason ‘supervenes as a craftsman of impulse’. However, with the exception of these cryptic remarks and a brief reference a few paragraphs later to a set of sound inclinations (aphormai, Diog. Laert. 7.89) given by nature, Diogenes tells us very little about how we acquire virtue, or how this relies on our living according to reason/crafted impulses. We are merely given the rough outlines of 7

8 9

10 11

Klein 2016: 161. See also Inwood 1985: 189–94 and Brittain 2002: 269–71. This natural ‘template’ is the expression of our spermatikoi logoi which ensure that every part of the cosmos develops according to nature/fate, see Menn 1995: 26–28. Cf. Klein 2016: 150. See, similarly, LS 2: 344, Inwood 1985: 202ff. Contra Cooper 2013: 172, note 41, who takes the rational impulses to replace or suppress ‘the old instinct-based desires’. Cf. the analysis of this argument in Long [1970/71] 1996: 145–49. For a good survey of the longstanding debate about how to interpret the shift from merely appropriate to fully virtuous action described in this passage, see Klein 2016: 153–63.

168

Skill in Ancient Ethics

a psychology of moral progress, but by fleshing it out a bit, we can at least make a qualified guess about what the Stoics had in mind. On the Stoic view, impulses follow upon perception, more specifically, upon impressions (phantasiai) of the special ‘hormetic’ kind that present a given course of action as appropriate. As Diogenes points out in our passage above, these two faculties are characteristic of the animal mind (hēgemonikon), but humans, as rational animals, eventually acquire two further powers: reason and assent (synkatathesis). These powers allow us to spell out our impressions in verbal form and thus to scrutinize, assent to, or reject them. Assenting to an impression means accepting it as true and, if the impression was hormetic, acting upon it. In other words, when we acquire reason we become able to consider our impressions and to choose whether to adopt and act upon the beliefs they suggest to us.12 To the extent that modern interpreters have commented directly on the expression ‘craftsman of impulse’, they have taken it to refer very generally to this ability to scrutinize and reject/assent to our impressions. Long and Sedley, for instance, note: ‘As the craftsman of impulse, reason shapes a way of living correctly’, and add in a parenthesis: ‘Note the description of the virtues – dispositions of reason – as τέχναι’ (LS 2: 344).13 This is certainly right, but, as I shall argue, there is a deeper connection between possessing expertise and having the right impulses. According to Galen, Chrysippus defined reason as ‘a collection of conceptions and preconceptions’ (Galen PHP 5.3). These concepts are stored impressions of general characteristics that allow us to recognize an individual object as the member of a more general class and call it by the name of that class. The conceptualization of our perceptual intake represents a fundamental change in the nature of our impressions and it appears as a distinguishing mark of the rational mind.14 As we saw above, the distinction between rational and nonrational minds was fundamental to Diogenes’ account and one way of describing the difference between the two, I submit, is to say that the former work with expertise and the latter without it. On the most plausible reading of the evidence relating to rational perception, the Stoics were thus committed to perceptual penetration, i.e. the view that our

12

13 14

See Frede 1994 and, more comprehensively, Inwood 1985: 42–101 and Brennan 2003. Here and throughout I use ‘belief ’ as a collective term to refer indiscriminately to antilēpsis, doxa, katalēpsis and epistēmē. See, similarly, Inwood 1985: 202 ff. See e.g. Diog. Laert. 7.49 and Sext. Emp. Math. 8.275–76 with Frede 1994.

Chrysippus on Expertise and Moral Development

169

perceptual intake is not subsequently interpreted by our rational mind, but actively shaped by the concepts we happen to possess. In other words, a trained musician, thanks to her specialized musical concepts, will not only interpret her impressions of a piece of music differently from an amateur, she will actually hear the music differently.15 Diogenes Laërtius makes this point in his report of Stoic epistemology: ‘the former are technical (technikai), the latter non-technical (atechnoi), for a craftsman also views a picture differently than the unskilled’ (Diog. Laert. 7.51).16 Similarly, Cicero includes the following consideration at the beginning of a speech defending the Stoic theory of knowledge in his Academica: ‘But if you add the exercise (exercitatio) and expertise (ars) that allow one’s eyes to dwell on paintings or one’s ears on songs, can anyone fail to see the power of the senses?’ (Cic. Acad. pr. 2.20).17 Most interpreters read the comments of Cicero and Diogenes concerning technical impressions as being limited to the case of experts in a given field, but Cicero is making a general point about the power of our senses, and Diogenes’ Greek strongly suggests that he is talking of rational impressions in general, not a subclass available to experts only.18 I therefore suggest that they are drawing a general connection between expertise and the increasing precision of rational perception, achieved through our concepts, when compared to non-rational perception. A little later in the Academica, Cicero sums up the Stoic theory of knowledge in the following way, providing what I take to be the basic outline of Chrysippus’ moral epistemology: So since the human mind is wholly adapted for knowledge of the world and for constancy in life, it welcomes cognition beyond all else and loves katalēpsis (which, as I said, translates literally as ‘apprehension’) both on its own account – nothing is 15 16

17 18

See the detailed and cogent argument for this interpretation in Shogry 2019. Cf. Phld. Mus. 115.26–116.5 and the descriptions of the precision of cataleptic impressions in terms of expertise in Sext. Emp. Math. 7.248 and 252 with Hankinson 2003: 61 and Shogry 2018. Cf. Cic. Nat. D. 2.145–46 and Arr. Epict. diss. 3.6.8. In the preceding lines, Diogenes uses partitive genitives (tōn de phantasiōn . . . tōn de aisthētikōn . . . eti tōn phantasiōn . . .) when he introduces new distinctions or subdivisions, while the hai men . . . hai de . . . constructions refer to the groups that are singled out by those distinctions. Assuming that Diogenes continues this pattern here, the paragraph states (i) that the Stoics distinguished between rational and non-rational impressions, (ii) that the former were also called thoughts, and (iii) that they were in some sense considered technical. This is contra Frede 1994: 52, Brittain 2005: 181 and Shogry 2019: note 17, who all read the passage as introducing a distinction among rational impressions, and therefore take technical impressions as constituting a separate subclass of rational impressions. Although all rational impressions thus appear to be technical in some basic sense, their degree of expertise (i.e. their degree of articulation and precision) will, of course, differ. See e.g. Cic. Acad. pr. 2.57. I am grateful to Brad Inwood for prompting me to clarify my view.

170

Skill in Ancient Ethics

dearer to the mind than the light of truth – and for its use. Hence it uses the senses, produces expertise as almost second senses, and strengthens philosophy to such a pitch that it creates virtue, the one thing that makes our whole lives coherent. Cic. Acad. pr. 2.31, transl. Brittain, with slight modifications

The speaker here is Lucullus, a representative of Antiochus’ “Old Academy”, and the material is therefore not straightforwardly Stoic. But throughout the work, Antiochus is presented as a defender of Stoic epistemology and, as Cicero later puts it, he ‘never strays a foot away from Chrysippus’ (ibid., 2.143) in logic/ dialectics. We can therefore take his speech to be at least broadly orientated toward Chrysippean epistemology.19 Expertise appears here under the guise of ‘almost second senses’. Presented as a consequence of the mind’s natural aptness for knowledge and constancy, we are given a sketch of human cognitive development divided into the cumulative sequence sensus – ars – philosophia – virtus, where the four stages represent an increasingly discriminating reliance on apprehension.20 On this picture, expertise ‘strengthens philosophy’ and helps ‘create virtue’. I will argue that expertise does this by restructuring our concepts and impressions according to their systematic and goal-directed structure. The central point behind describing expertise as second senses is thus that having rational and technical impressions will not only enable you to actively consider your impressions before acting on them, they will also present you with a more accurate picture of what you can and should do – assuming your concepts of value develop in the right way. It is this relation between rational conceptualization and expertise, I take it, that Chrysippus alludes to when he calls reason the craftsman of impulse. The following section will make this case by bringing out how expertise characterizes the workings of the rational mind as a ‘collection of concepts’ that emerge through experience, and develop through the technical articulation and systematization offered, most importantly, by philosophy. Having laid out these general aspects of Chrysippus’ epistemology, the final section focuses more narrowly on his moral epistemology, and investigates the role of expertise in bridging the gap between the instinctual grasp of the appropriate that is furnished by our sound inclinations, and the unerring and infallible grasp of the good, appropriate and just that is characteristic of the virtues.

19 20

See, similarly, (with some important qualifications) Striker 1997: 258 and Perin 2005: 387. See Cic. Acad. pr. 2.19–24, which describes the relation between apprehension and the senses (19– 20), expertise (22), philosophy (22) and virtue (23–24).

Chrysippus on Expertise and Moral Development

171

2. Second senses: expertise and concept formation In a puzzling fragment from what is probably one of Chrysippus’ works, we find the following description of philosophy:21 For first of all philosophy is a cultivation (epitēdeusis) of the rightness of reason or knowledge (epistēmē) consisting in the specific investigation of reason. And being acquainted with the parts of reason and their configuration (syntaksis) we will use it with experience (empeirōs). By reason, I mean the one present in all rational beings according to nature. PHerc. 1020, col. 108.12–2422

The exact meaning of this text is elusive, but it seems to envision philosophy as offering an insight into reason and a means of achieving right reason.23 However, before the parts of reason and their configuration achieve this state of perfection, and we thus become fully experienced, a long process of concept formation and refinement must take place. This section will outline the steps of this development and try to bring out how the Stoics envisioned the mental processes involved, in particular experience (empeiria) and expertise. Our most detailed report on how this process gets started is given by the firstcentury ce doxographer Aëtius: The Stoics say that when humans are born the controlling part of their soul is like a sheet of papyrus well-suited for inscription. On to this every single conception is written. The first mode of inscription is the one through sense perceptions. For when you sense something white, for instance, you have a memory of it when it is gone. And when many memories of the same type have appeared, we say that you have experience, for experience is a mass of similar memories. Of the conceptions, some are formed naturally in the ways described and without any kind of expertise, others through our instruction and careful attention. The latter are only called conceptions; the former are also called preconceptions. The reason according to which we are called rational is assembled out of the preconceptions during the first seven years . Aëtius 4.11.1–424 21 22

23

24

On the text’s authorship, see Alessandrelli & Rannochia 2017: 8–17. Cf. Clem. Al. Paed. 1.13 and Isid. Pel. Ep. 5.558 (FDS 2B). My translation follows the text of Alessandrelli & Rannochia 2017. See, similarly, Alessandrelli & Rannochia 2017 ad loc., with further references. This is contra Bronowski 2019: 39–40, who construes the passage as rejecting an Aristotelian definition of philosophy as knowledge. The text is preserved in Ps.-Plutarch, Placita philosophorum, but probably derives from Aëtius. See Mansfeld 2016, sections 1–2. Cf. Ps.-Galen Hist. philos. 92 and Cic. Acad. pr. 2.30. Diels, following Ps.-Galen, supplied phantasiōn in the definition of empeiria, but from the immediate context mnēmōn seems more natural, and its omission could be explained by simple haplography.

172

Skill in Ancient Ethics

This report presents the two types of concepts identified by Chrysippus as the constituents of reason, preconceptions (prolēpseis) and conceptions (ennoiai proper), and explains how they are formed. Preconceptions arise naturally (physikōs), i.e. pre-reflectively, as opposed to the conceptions, which require the distinctly rational processes of instruction (didaskalia) and careful attention (epimeleia).25 The former can therefore plausibly be interpreted as a preliminary grasp of the shared characteristics of a class of things, while the latter would appear to be more precise and carefully spelled out.26 Aëtius lists four mental operations that go into the formation of a concept (his ‘first mode of inscription’): the perception of something, the memory (mnēmē) it leaves, its combination with other similar memories to form experience, and, finally, the formation of a preconception.27 These different operations relate to quite different capacities. Memories and concepts thus account for two, distinct, recognitional capacities. Memories enable us to recognize particulars that we have already encountered, while the encapsulation of their general characteristics in a concept allows us to recognize individual things as instances of a more general class.28 Experience, on the other hand, seems to imply a capacity to recognize the similarities between different objects and to store the impressions of them together. This amassing of similar impressions explains, it would seem, how the human mind extracts the shared characteristics from individual impressions of particular objects in order to form preconceptions. In the modern literature, experience is often construed as strictly empirical.29 There is, however, nothing in Aëtius’ text to motivate such a restricted reading and, as we shall see, interpreting experience as an amassment of both perceptual and non-perceptual impressions allows us to make better sense of the overall process of concept formation. I take the basic idea to be as follows: the clean sheet (chartion) that is our mind at birth is made in a special way (it is ‘well suited for inscription’) which ensures that our impressions are sorted according 25 26

27

28 29

On this contrast, see Brittain 2002: 261–64. See, similarly, Frede 1994, Jackson-McCabe 2004: 327–29, Brittain 2005: 170–4, Inwood 2005: 280– 81 and Dyson 2009: 91–95. For concepts as general, see Diog. Laert. 7.54 with Brittain 2005: 171–72 and Wildberger 2006: 86–93. It is controversial whether the Stoics recognized a further type of preconception that is not formed but ‘inborn’ (emphytos). See e.g. Hadot 2014: 25–29, who posits a Chrysippean and an Aëtian meaning of the word prolēpsis. For a good survey of this century-old debate, and the use and meanings of the terms physikē ennoia and emphytos prolēpsis, see Jackson-McCabe 2004: 325–33. These inborn concepts will be discussed in section three, below. See Brittain 2002: 286–87. See e.g. Sandbach [1930] 1971: 26, Jackson-McCabe 2004: 329, Inwood 2005: esp. 274–75 and Hadot 2014: 28–29.

Chrysippus on Expertise and Moral Development

173

to similarity, and these automatically recognized similarities are what become encapsulated in our preconceptions (here, the impression of whiteness, as opposed to something white (leukou tinos)). This automatic ordering of our memories into experience appears to be a distinctive mark of the human mind and, if this is the right way to read Aëtius’ brief account, it constitutes a wonderfully simple way of setting us apart from the non-rational animals. Because they do not register similarities, animals never go beyond the point of recognizing already memorized objects (and perhaps their near identicals), while humans, thanks to their ability to gather experience, go on to form preconceptions and eventually reason.30 Aëtius’ brief remarks can be fleshed out by drawing on Cicero’s parallel description of the process, in which we find several examples of increasingly complex conceptualized impressions.31 Cicero’s presentation suggests a gradual process according to which concepts and experience mutually inform each other in order to produce ever more precise and detailed impressions. First, he says, we start putting our impressions of basic perceptual qualities into verbal form, producing impressions like ‘This is white’. Only after this initial stage do we form impressions of natural kinds like ‘That is a horse’.32 The latter, slightly more complex, concepts could plausibly rely on the former in something like the following way. In order to recognize something as a horse, for instance, you first need to conceptualize the horses you encounter in terms of characteristics such as big, strong and fast. Applying these basic concepts enables you to recognize different horses as similar to each other, and as distinct from other specific kinds of animals (such as dogs), automatically grouping the impressions of them

30

31

32

Contra Dyson 2009: 115–19, who identifies experience as the basic recognitional capacity in animals and humans alike. However, the primary examples of animal recognitional capacities, namely the grasp of their own constitution and ability to recognise things as oikeion or allotrion, are innate, and therefore cannot be the result of experience (see Sen. Ep. 121.19–24). (These are therefore better explained as the result of non-rational ‘quasi-concepts’. See Sorabji 1993: 20–28 and Brittain 2002.) To my knowledge, no Stoic sources attribute experience to animals (Brittain 2002: 265, note 37 allows for the possibility, but quotes no text that suggests as much) and Stoic discussions of animal memory suggest that it is limited to individual memories, which allow animals to recognize the particulars that gave rise to them (see Sen. Ep. 124.16 and August. Qu. an. 50 and 54, with Brittain 2002: 286–7). Given the role of experience in forming concepts, which are clearly distinctive of humans, this capacity is a likely candidate for distinguishing pre-rational from non-rational cognition. As Brittain 2005: 173 points out, transition (metabasis) is another likely candidate, but the two suggestions are not mutually exclusive. Cic. Acad. pr. 2.21. As Cicero emphasizes: ‘This, already, is grasped by the mind rather than the senses’. The point, I take it, is that every act of predication involves employing preconceptions, i.e. the (pre-)rational mind (cf. Diog. Laert. 7.49 and 52). See, similarly, Sorabji 1993: 22–25 and, more cogently, Brittain 2002: 256 ff. Cic. Acad. pr. 2.21. Note the adverbs iam . . . deinceps . . ., which mark the temporal progression between the two stages.

174

Skill in Ancient Ethics

together into an experience of horses, and eventually forming a preconception of a generic horse (the preconceptions big, strong and fast -> experience of horses -> the preconception ‘horse’). On this reading, the emergence of preconceptions gradually refines our impressions, giving rise to more discerningly structured experience, and thereby allowing us to form more advanced concepts. As Aëtius informs us, it is then out of these preconceptions that human reason is eventually assembled.33 Cicero’s final example of a conceptualized impression is ‘If something is human it is a mortal animal partaking in reason’ (Cic. Acad. pr. 2.21). With this technical definition we seem to have transcended the ‘naturally formed’ preconceptions and arrived a conception proper.34 Aëtius contrasts these with the preconceptions on the basis that the latter are formed anepitechnētōs – implying that conceptions are, by contrast, formed with expertise. This lines up with Diogenes’ remarks that reason makes our impressions technikai and ‘emerges as the technitēs of impulse’, and provides further support for the thesis that expertise is somehow characteristic of the development of rational perception. But how might expertise figure in this process? At the most basic level, expertise is ‘a system of apprehensions that have been exercised together (syngegymnasmenōn) for one of the useful goals (telos) in life’ or ‘a disposition that proceeds methodically by means of impressions’ (Olymp. in Plat. Gorg. 7.1).35 Our sources attribute these two definitions to Zeno, Cleanthes and Chrysippus in various constellations, and they appear to complement each other rather than represent competing views.36 The first definition thus focuses on the ‘cognitional’ aspect, the nature and structure of the beliefs involved in the possession of expertise, while the second definition focuses on the motivational aspect of these beliefs, construing expertise as a disposition (hexis) that guides the actions of the expert.37 The word for belief here is katalēpsis, apprehension, the technical Stoic term for a

33 34

35

36 37

On the formation of reason, see Frede 1994 and Brittain 2002. The example takes the form of an indefinite conditional, recognized by Chrysippus as the ideal form of definition (see Sext. Emp. Math. 11.8–11 with Sedley 1985, Caston 1999: 195–99 and Dyson 2009: 99–102). This must represent a culmination in the process of concept formation. Cicero thus presents it as a ‘filled-out apprehension’, from which we form a conception (notitia, giving the Greek term ennoia only, as opposed to ibid., II.30, where he gives ennoia and prolēpsis as the Greek equivalents). I follow von Arnim’s emendation of the MSS’s syngegymnasmenon to syngegymnasmenōn, cf. Mansfeld 2003: 57. See, similarly, LS 1: 263 and Tsouna 2021: 169–170. Cf. Liu 2008: 263, n. 20 and Brouwer 2014: 32–33, speaking of the parallel case of scientific knowledge (epistēmē). See also Sparshott 1978: 281. On hexis as a disposition of the material mind, see Inwood 1985: 38–41.

Chrysippus on Expertise and Moral Development

175

particularly clear grasp of a given fact about the world, i.e. the assent to an accurate and reliably true impression – a so-called cataleptic impression (phantasia katalēptikē) – which therefore always results in a true belief.38 These beliefs are worked together into a system or disposition which, presumably due to its goal-directed structure, can then guide our actions methodically. In technical Stoic terms, this disposition must thus be ‘hormetic’, i.e. a disposition that determines how we react to certain stimuli.39 The Stoics therefore seem to have thought of expertise as an internally coherent set of beliefs that is able to guide our actions in a methodical way.40 In a sense, then, expertise is just a collection of apprehensions, true impressions that have been accepted and committed to memory, just like experience was defined as an amassment of memories.41 One important aspect of acquiring expertise, therefore, is that we start combining these apprehensions in a new way: we co-exercise (syngymnazō) them into a system (systēma), and this coexercise, just like the more rudimentary process of collection involved in prerational experience, results in gradually more refined collections of similar impressions that underlie the formation of new and more precise conceptions.42 Through technical co-exercise, our experience becomes more structured and refined and our concepts and expertise evolve with it. That expertise characteristically develops our conceptions, and itself develops with them, is also consistent with its place within the wider framework of Stoic epistemology. Expertise thus shares many characteristics with the scientific knowledge (epistēmē) involved in perfect virtue – both are defined as systems of beliefs and dispositions.43 But in contrast to scientific knowledge, expertise does not imply an infallible grasp of the world, and we know that the Stoics acknowledged expertise in varying degrees of perfection.44 A further difference

38

39 40 41

42

43

44

See e.g. Diog. Laert. 7.52 and Cic. Acad. pr. 2.17. The exact nature of cataleptic impressions is much debated. For a good survey, see Hankinson 2003 and Shogry 2018. Ar. Did. Stoic. 2.7.9 with Inwood 1985: 189–90. See, similarly, Sparshott 1978 and Tsouna 2021. The parallel reliance of experience and expertise on memorized apprehensions is pointed out at Cic. Acad. pr. 2.22; cf. Sext. Emp. Math. 7.373. See e.g. the connection between inarticulate, imprecise, impressions and lack of training (agymnastos) suggested in Hierocles El. Eth. 7.50–8.30; cf. Sen. Ep. 121.12–13. Compare the definitions of epistēmē in Ar. Did. Stoic. 2.7.5l, where the apprehensions involved are ‘unshakeable and unchangeable by reasoning’. For a good discussion of this difference between technē and epistēmē, see Menn 1995: 18–22. More generally on the definitions of knowledge, see Vogt 2008: 120–6, Liu 2008 and Brouwer 2014: 29–33. On the Stoic notion of systēma and its wider importance in Stoic philosophy, see Bronowski 2019: 52–80. See Ar. Did. Stoic. 2.7.5k+7b, and the material on ‘intermediary expertise’ in Sext. Emp. Math. 9.207, Simpl. in Arist. Cat. p. 224, 30–31 (FDS 865.15–16), and pp. 284, 32–85,1 (FDS 363). See similarly, Menn 1995, esp. 9–10, contra Sellars 2009: 69, 82–84.

176

Skill in Ancient Ethics

between the two is that scientific knowledge, viewed as a disposition, is ‘receptive of impressions’, whereas expertise ‘proceeds by means of impressions’. Scientific knowledge is thus a disposition to understand something in a certain way, whereas expertise is a disposition to react in certain ways, although one and the same disposition can, of course, be both scientific knowledge and expertise, as the virtues are.45 This leads us to the final distinctive feature of expertise, its goaldirectedness.46 Just like the virtues, which are directed towards the telos, the individual apprehensions or theorems (theorēmata), as they are often called, involved in expertise are ‘exercised together for one of the useful goals in life’.47 To sum up, expertise is a prominent characteristic of the perfected human mind, but in contrast to scientific knowledge and virtue, it also comes in less perfect varieties that are accessible to someone making moral progress.48 Based on the remarks of Diogenes, Aëtius and Cicero about the technical nature of rational perception, I have suggested that this reflects a view of expertise as building on our natural capacity to gather impressions into experience, but bringing with it more advanced ways of structuring these impressions, thereby resulting in more refined concepts. Let us turn briefly to the textual evidence for this suggestion. Several later ancient commentators who quote the Stoic definition of expertise specify that the co-exercise involved in expertise happens through empeiria, while other more reliably Stoic sources also connect expertise to experience.49 One might, of course, assume that this notion of experience is unrelated to Aëtius’ use of the term, and it is certainly more complex and fully developed – appearing even in Chrysippus’ description of right reason above.50 On the other hand, acquiring ‘a system of apprehensions’ would certainly involve some

45

46 47

48 49

50

Ar. Did. Stoic. 2.7.5b. The expertise involved in virtue is thus, strictly speaking, epistēmē technikē; see the third definition of epistēmē, ibid., 2.7.5l. Compare the fourth definition of epistēmē, ibid., 2.7.5l. On the goal-directedness of the virtues, see Arius’ remark about their shared goal and theorems, Stoic. 2.7.5b5. Cf. ibid., 2.7.5b3, quoted below. Cf. the more detailed discussion in Menn 1995: 9–26. See e.g. Dauid, Prol. 14 p. 43, 30–44, 17 (FDS 393a), Schol. in Dion. Thrax, 108, 31–33 (FDS 397) and Marcellinus, Proleg. in Hermog. Stat., p.262, 1–23 Rabe (FDS 400). These commentators are working within a broadly Neoplatonic rhetorical tradition and can therefore not be used straightforwardly as sources of Stoic doctrine. However, they generally seem well-informed about the Stoic definition and do not use empeiria in its standard Platonic sense (where it is contrasted with technē), but in the more traditional sense of something that has been observed, tried out, or tested and is therefore known (see Allen 2020). David thus glosses empeiria as ‘a testing through numerous trials (polypeirias)’. This kind of testing seems to have figured in Stoic views about expertise as well; see Sext. Emp. Pyr. 3.251. A more general connection between experience and expertise is made in, e.g., Philo De sobrietate 34 (SVF 3.244), Arr. Epict. diss. 2.24.1–10 and Cic. Acad. pr. 2.57. This, I take it, is the standard interpretation. See, e.g., Brittain 2005: n. 20.

Chrysippus on Expertise and Moral Development

177

‘amassment of similar impressions’. It therefore seems unnecessary to posit distinct meanings of the term ‘experience’, as opposed to seeing co-exercise as resulting in distinctly technical types of experience. In addition, the terms didaskalia and epimeleia, which Aëtius uses to describe the formation of conceptions, are standardly associated with expertise: didaskalia often designates instruction in a discipline or field of expertise, whereas epimeleia is used more generally of the care or effort one expends on something, often in close parallel with the term technē.51 This corroborates the assumption that the development of conceptions is crucially bound up with the acquisition of expertise. In particular, the expertise needed in order to develop correct conceptions is the one involved in philosophy. Philosophy thus provides the primary means for articulating our conceptions both through the formal methods of definition and through philosophical investigations more broadly.52 For the later Stoic Epictetus, this stress on the importance of definitions takes the shape of a general demand to expend ‘epimeleia on the articulation (diarthrōsis) of our preconceptions’ (Arr. Epict. diss. 2.17.13) and a conception of philosophy as an expertise aimed at ‘reason being in its right state’ (ibid., 4.8.12).53 Aëtius – possibly echoing Chrysippus’ definition in terms of a ‘cultivation (epitēdeusis) of the rightness of reason’ quoted at the opening of this section – instead defines philosophy as ‘training (askēsis) in the fitting (epitēdeios) expertise’ (Aët. 1.0.2).54 To which extent these different definitions reflect Chrysippean views is uncertain.55 It does, however, seem safe to assume that he shared the underlying view of philosophy as an activity which imparts the right structure to our rational mind, potentially bringing it to the stage of perfection that is characteristic of right reason and the cosmos as a whole.56

51

52

53

54

55 56

On didaskalia, see e.g. Galen, PHP 4.397, 4–10, Sext. Emp. Pyr. 2.13 and 250–65. On epimeleia, see e.g. Philo De ebritate 88 (SVF 3.301), Arr. Epict. diss. 2.5.3–4, and M. Aur. Med. 6.16.2–3. Whether epimeleia and didaskalia ever acquired a technical, Stoic sense or they are simply adopting the vocabulary traditionally associated with expertise, remains unclear. Plutarch’s echoing of Aëtius’ phrase ‘through education and careful attention’ at De soll. an. 962C, and the listing of didaskalia (alongside mathēsis, prolēpsis, noēsis, hormē and synkatathesis) as one of the things that depends on the existence of lekta at Adv. Col. 1119F–1120A, suggest that these two terms were at least stable parts of Stoic terminology. On the forms and function of definitions in the Stoic theory of language and reason, see Brittain 2005: 179 ff. Cf. ibid., 1.15.3, defining philosophy as ‘the expertise concerning life’, and 1.20, 3.3 and 3.8, on expertise in dealing with impressions. Cf. ps.-Galen Hist. philos. 5. The term askēsis is not attested for Chrysippus. It was central to the philosophy of Epictetus and Marcus Aurelius, and it is not unlikely that they were developing older ideas. Sellars [2003] 2009: ch. 5–7 offers a good discussion of the evidence, but his interpretation of it as reflecting a distinctly practical conception of philosophy is problematic. See Inwood 2004. See Mansfeld 2003 for a detailed discussion. See, similarly, Mansfeld 2003 and Brouwer 2014: 41–49.

178

Skill in Ancient Ethics

This evidence on the relationship between expertise and conceptions can be construed in either of two ways. The weak interpretation limits the scope of expertise and technical conceptions to specialized disciplines (most importantly those involved in Stoic philosophy), and thus sees technical perception as exclusive to these fields.57 The stronger reading interprets expertise as a constitutive aspect of reason, as conceived by the Stoics, and therefore takes rational perception in general to be technical, claiming that all of a rational animal’s concepts are, in some minimal sense, technical. I lean towards the strong interpretation, but I do not think that the evidence allows us to decide conclusively between the two.58 On both of these interpretations, however, the acquisition of expertise and the way this informs the formation and articulation of concepts is a crucial aspect of the development towards rational perfection. Based on the preceding discussion, we can understand this role in terms of the way an experience informed by instruction, careful attention and the formal methods of definition, refines our concepts and ‘co-exercises’ them into a system – thereby perfecting ‘the parts of reason and their configuration (syntaksis)’.

3. The craftsman of impulse Having examined the role of experience and expertise in enabling humans to form and articulate concepts, we now have the resources to spell out the nature of the technical change in our impulses that is implied by Chrysippus’ description of reason as ‘the craftsman of impulse’. We can start by looking at the impulse that reason is most directly supposed to craft, the primary impulse posited as the psychological basis of all animal behaviour. The primary impulse, as Jacob Klein suggests, is probably best understood as a hormetic disposition that co-ordinates an animal’s appropriate behaviour. As such, it describes the motivational side of self-perception.59 To put it another way, it reflects the motivational force of the beliefs that an animal has about its own constitution and what is proper to it. In non-rational animals, these take the

57 58

59

This is the interpretation of expertise adopted by Shogry 2019. This interpretation is suggested by Diogenes’ claim that rational impressions tout court are technical (see note 18). I would speculate that the distinctively rational ability to recognize the relations between different propositions/concepts mentioned at Cic. Off. 1.11 and Sext. Emp. Math. 8.275–6 reflects some degree of proto-technical systematization. On these features of Stoic reason, see Brittain 2002: 256–59. See also Scade 2017 on the technical and rational, but non-verbalized, structure of music and its effect on the soul. Klein 2016: 275, expanding on the interpretation of Inwood 1985: 189–90.

Chrysippus on Expertise and Moral Development

179

form of initial grasps (antilēpseis), but one of the things that happens when humans become rational, I take it, is that these grasps are gradually refined into expertise.60 As we noted above, expertise, just like the primary impulse, is also a hormetic disposition, but one that is based on systematized and goal-directed apprehensions. It is, I suggest, by transforming the primary impulse into a disposition based on a set of technically structured and articulated concepts that reason works as the craftsman of impulse. How would this play out? As we noted above, Diogenes talks of nature leading us to virtue and giving us a set of sound inclinations. These inclinations are part of any human’s constitution and, as such, they co-determine our behaviour: having an inclination for something means that we recognize the relevant behaviour as proper/appropriate to us; due to our primary impulse to perform such actions, we act accordingly.61 Arius Didymus’ Epitome of Stoic Ethics lists four such inclinations as part of his discussion of the four primary virtues: The goal of all these virtues is to live consistently with nature. Each virtue through its individual properties enables the human being to achieve this. For from nature a human being has inclinations for the discovery of what is appropriate (tēn tou kathēkontos heuresin), for steadiness in the impulses, for endurance, and for distribution. Ar. Did. Stoic. 2.7.5b3, transl. Pomeroy, slightly modified

According to this passage, humans will naturally practise steadiness, endurance and just distribution, but they will also engage in the more reflective activity of discovering what is appropriate (as opposed to simply doing it, which is already ensured by the primary impulse). This means that humans are naturally inclined to think about how they act, and such self-reflection, as we shall see, has a major impact on their development. At the most basic level, discovering the appropriate must simply amount to forming a preconception of it – that would seem a necessary condition for discovering it in a more technical sense.62 As noted above, any action that is proper to our natural constitution counts as ‘appropriate’, including our naturally induced steadiness, endurance and just distribution. The fact that these behavioural patterns are integral parts of our inborn, human constitution

60 61 62

I am grateful to Jula Wildberger for suggesting this way of framing the development. On the nature of this instinctive recognition, see Brittain 2002: 261–71. See e.g. Diog. Laert. 7.42: ‘The part on definitions also aims towards the determination of truth; for it is through the concepts that the things are grasped.’ Technically, heuresis is the culmination of philosophical investigation and demonstration. See Clem. Al. Strom. 6.14 and Cic. Acad. pr. 2.26

180

Skill in Ancient Ethics

ensures that we will have first-hand experience of such behaviour and therefore form preconceptions of it.63 This makes ethical preconceptions ‘natural’ in a double sense: like any other preconception, their formation happens ‘naturally and without expertise’; but unlike the preconception of e.g. a llama, we are naturally bound to form a preconception of appropriate action, because we and everybody around us naturally exhibit such behaviour.64 This, I take it, is what Diogenes Laërtius has in mind when he singles out such fundamental ethical concepts as natural.65 In other words, our initial grasp of what is proper to us develops through experience into our preconceptions of the ‘appropriate’ (together with its sub-species, such as the ‘moderate’, ‘brave’ and ‘just’).66 This awareness of our behavioural patterns is achieved when reason supervenes on our impulses, this being the earliest precursor of prudence.67 In his exposition of the theory of oikeiōsis, Cicero similarly explains how we develop an ‘appropriate selection’ (cum officio selectio) once we have discovered (inventa) the selective patterns underlying our appropriate behaviour: ‘Then such selection becomes continuous, and, finally, stable and in agreement with nature. At this point that which can truly be said to be good first appears and is recognised for what it is’ (Cic. Fin. 3.20). The reflective inclination in basic human nature thus ensures that our initial grasp of the proper and the ensuing impulses to act accordingly develop into conscious moral choices or ‘selection’ (eklogē).68 This, in turn, leads to the formation of the conception of the truly good and a life in agreement with nature.69 What it means for reason to craft our impulses in this process is that it gradually brings our ethical concepts to the degree of articulation and systematization characteristic of a fully developed expertise. That Chrysippus regarded such systematization as important to ethics is clear from a section of his

63 64 65

66

67 68

69

See, similarly, Inwood 2005: 277–81 and Klein 2016: 152. Cf. Sen. Ep. 120.1–12, stressing the importance of observing such behaviour in others. Diog. Laert. 7.53. Cf. Plutarch, Comm. Not. 1070C, speaking of these concepts as ‘having their intrinsic generation in the principles within us’ with Jackson-McCabe 2004. Note the parallel between Diogenes’ natural thought of ‘something (ti) just and good’ and Aëtius’ example of sensing ‘something white (tinos leukou)’, suggesting that these natural thoughts, like external impressions, feed into our experience. This is contra Jackson-McCabe 2004: 331 and 339, who reads this as positing a ‘hazy πρόληψις of something good’. See similarly, Jackson-McCabe 2004: 339: ‘The Stoics, then, understood ethical preconceptions to be a unique category of preconception inasmuch as they arise as the result of an innate evaluative disposition that every animal brings to its empirical experience’. By contrast, Hadot 2014: 23–29 denies that experience plays any role in forming these natural concepts. Similarly Fisher 2015: 24–25. See, similarly, Frede 1999: 73–75. On selection, see Inwood 1985: 201–5, who offers an illuminating discussion of how this is based on the primary impulse. Cic. Fin. 3.21 with Jackson-McCabe 2004: 335–36.

Chrysippus on Expertise and Moral Development

181

ethical works called ‘On articulation (diarthrōsis) of the ethical concepts’, which comprises 22 works on topics like definitions, division, genus and species.70 We also find this interest in technical definitions and systematic structure reflected in the long lists of virtues reported by Arius and elsewhere. Articulating our ethical concepts and working out the correct genus-species relation between them – their syntaksis, as Chrysippus phrases it in his definition of philosophy – thus appears to be a crucial aspect of moral progress as he conceived of it. The development of ethical virtue therefore involves, I suggest, three gradual stages: (i) with the emergence of reason, we discover the appropriate, and (ii) start working out the correct definitions and relations of the types of behaviour that constitute the appropriate. This eventually gives us a systematic understanding of the appropriate, and a general expertise in appropriate selection. The final step towards virtue turns on (iii), the full grasp of a further concept, namely that of the truly good. This ‘evaluative revolution’ – as Tad Brennan has aptly called it71 – transforms our appropriate selection into the fullblown virtue of prudence: a ‘knowledge of what one is to do and not to do and what is neither; or the knowledge in a naturally social and rational animal of good things, bad things, and what is neither’ (Ar. Did. Stoic. 2.7.5b1, trans. Inwood and Gerson).72 The concept of the good and its impact on our ethical outlook raises its own particular set of questions, which we cannot consider further here.73 We may simply note that prior to, and probably informing its full articulation, lies a process of refining our grasp of specifically human, appropriate behaviour through experience, co-exercise and articulation. This eventually brings the behavioural patterns encapsulated in our inborn constitution into complete conformity with the rational order of nature, until the point where we suddenly realize that this conformity to nature’s plan constitutes the true telos of our actions.74 Chrysippus famously reformulated this goal in terms of ‘living according to an experience of the things that happen by nature’ (Diog. Laert. 7.87).75 In this

70 71 72

73 74 75

Diog. Laert. 7.199. Brennan 2003: 272. Prudence appears to be the over-arching ethical virtue (a general knowledge of what to do, cf. Striker 1991: 42, n. 25), but I leave it open exactly how to spell out the relation between this and the other primary, ethical virtues. On the interconnectedness and mutually informing character of these virtues and the virtues pertaining to the other branches of philosophy, see Mansfeld 2003 and Annas 2007. See Frede 1999, Jackson-McCabe 2004, Inwood 2005 and Hadot 2014. On how appropriate selection informs virtuous choice, see Klein 2015: 258 ff. Cf. Ar. Did. Stoic. 2.7.6a.

182

Skill in Ancient Ethics

section, I have focused on the ethical aspects of these things, namely the behavioural patterns arising on the basis of our natural constitution and inclinations. I have argued that a first-hand experience of these behavioural patterns is what enables us to become aware of our behaviour and develop the expertise in appropriate selection, which ultimately enables us to live in complete accordance with nature.76

4. Conclusion In this chapter, I have argued that thinking about moral progress in terms of the acquisition of expertise, in the technical Stoic sense, allows us to solve some central questions regarding the relation between impulse, reason, concepts and virtue. We started out by diagnosing two parallel gaps in our sources concerning how the primary impulse and our preconceptions evolve towards rational perfection, and by noting that the scant remarks about the intermediate stages in these processes suggest that expertise plays a role in bridging that gap. Here, I have attempted to outline this technical transformation of the human mind. The main elements of my account – stretching from the theory of a basic attachment to our constitution, and the definition of reason as a collection of concepts, through the definition of philosophy, and on to the description of the telos as relying on experience – can quite safely be attributed to Chrysippus. Taken together, this evidence reflects a conception of virtue as an expertise that emerges through experience and develops through the articulation of our concepts, thus grounding virtue firmly in our natural inclinations to see the world and act in specific ways.

References Alessandrelli, M., and G. Ranocchia (2017), Scrittore stoico anonimo opera incerta : pherc. 1020, coll. 104–112, Rome: ILIESI – CNR . Allen, J. (2020), ‘Reason, Experience and Art: the Gorgias and On Ancient Medicine’, in L. Taub (ed.), Cambridge Companion to Ancient Greek and Roman Science, 39–57, Cambridge: Cambridge University Press. Annas, J. (2007), ‘Ethics in Stoic Philosophy’, Phronesis 52 (1): 58–87. 76

Cf. Alex. Aphr. De an. 2.161.5–6 and Sext. Emp. Math. 11.170. See similarly Long [1983] 1996a: 170–71.

Chrysippus on Expertise and Moral Development

183

Bonhöffer, A. (1890), Epictet und die Stoa: Untersuchungen zur stoischen Philosophie, Stuttgart: Enke. Brennan, T. (2003), ‘Stoic Moral Psychology’, in B. Inwood (ed.), The Cambridge Companion to the Stoics, 257–294, Cambridge: Cambridge University Press. Brittain, C. (2002), ‘Non-Rational Perception in the Stoics and Augustine’, Oxford Studies in Ancient Philosophy 22: 253–308. Brittain, C. (2005), ‘Common Sense: Concepts, Definition and Meaning in and out of the Stoa’, in D. Frede and B. Inwood (eds), Language and Learning: Philosophy of Language in the Hellenistic Age, 164–209, Cambridge: Cambridge University Press. Bronowski, A. (2019), The Stoics on Lekta: All There Is to Say, Oxford: Oxford University Press. Brouwer, R. (2014), The Stoic Sage: The Early Stoics on Wisdom, Sagehood and Socrates, Cambridge: Cambridge University Press. Caston, V. (1999), ‘Something and Nothing: The Stoics on Concepts and Universals’, Oxford Studies in Ancient Philosophy 17: 145–213. Dyson, H. (2009), Prolepsis and Ennoia in the Early Stoa, Berlin: De Gruyter. Fisher, J. (2015), ‘Epictetus on the Epistemology of the Art of Living’, Apeiron 48 (1): 20–44. Frede, M. (1994), ‘The Stoic Conception of Reason’, in K. Boudouris (ed.), Hellenistic Philosophy 2: 50–63, Athens: International Center for Greek Philosophy and Culture. Frede, M. (1999), ‘On the Stoic Conception of the Good’, in K. Ierodiakonou, Topics in Stoic Philosophy, 71–94, Oxford: Clarendon Press. Hadot, I. (2014), ‘Getting to Goodness: Reflections on Chapter 10 of Brad Inwood, Reading Seneca’, in J. Wildberger and M.L. Colish (eds), Seneca Philosophus, 9–42, Berlin: De Gruyter. Hankinson, R.J. (2003), ‘Stoic Epistemology’, in B. Inwood (ed.), The Cambridge Companion to the Stoics, 59–84, Cambridge: Cambridge University Press. Inwood, B. (1985), Ethics and Human Action in Early Stoicism, Oxford: Oxford University Press. Inwood, B. (2004), ‘Review of John Sellars, The Art of Living: The Stoics on the Nature and Function of Philosophy’, Notre Dame Philosophical Reviews, https://doi. org/10.1007/s10790-021-09813-1 1 3 Inwood, B. (2005), Reading Seneca: Stoic Philosophy at Rome, Oxford: Clarendon Press. Jackson-McCabe, M. (2004), ‘The Stoic Theory of Implanted Preconceptions’, Phronesis 49 (4): 323–47. Klein, J. (2015), ‘Making Sense of Stoic Indifferents’, Oxford Studies in Ancient Philosophy 49: 227–281. Klein, J. (2016), ‘The Stoic Argument From Oikeisis’, Oxford Studies in Ancient Philosophy 50: 143–200. Liu, I. (2008), ‘Nature and Knowledge in Stoicism: On the Ordinariness of the Stoic Sage’, Apeiron 41 (4): 247–276. Long, A.A. [1970/71] (1996), ‘The Logical Basis of Stoic Ethics’, in A.A. Long (ed.), Stoic Studies, 134–155, Cambridge: Cambridge University Press.

184

Skill in Ancient Ethics

Long, A.A. [1983] (1996a), ‘Greek Ethics After MacIntyre and The Stoic Community of Reason’, in A. A. Long (ed.), Stoic Studies 156–178, Cambridge: Cambridge University Press. Long, A.A. [1993] (1996b), ‘Hierocles on Oikeiōsis and Self-Perception’, in A. A. Long (ed.), Stoic Studies, 250–263, Cambridge: Cambridge University Press. Long, A.A., and D.A. Sedley ([1987] 2012). The Hellenistic Philosophers. Vol. 2. Cambridge: Cambridge University Press. Mansfeld, J. (2003), ‘Zeno on the Unity of Philosophy’, Phronesis 48 (2): 116–131. Mansfeld, J. (2016), ‘Doxography of Ancient Philosophy’, in E. N. Zalta (ed.), The Stanford Encyclopedia of Philosophy. https://plato.stanford.edu/archives/win2016/ entries/doxography-ancient/ (accessed 27 January 2020). Menn, S. (1995), ‘Physics as a Virtue’, Proceedings of the Boston Area Colloquium of Ancient Philosophy 11: 1–34. Perin, C. (2005), ‘Stoic Epistemology and the Limits of Externalism’, Ancient Philosophy 25 (2): 383–401. Sandbach, F. H. (1971), ‘Ennoia and Prolepsis in the Stoic Theory of Knowledge’, in A.A. Long (ed.), Problems in Stoicism, 22–37, London: Athlone Press. Scade, P. (2017), ‘Music and the Soul in Stoicism’, in R. Seaford, J. Wilkins and M. Wright (eds), Selfhood and the Soul: Essays on Ancient Thought and Literature in Honour of Christopher Gill, 197–218, Oxford: Oxford University Press. Sedley, D. (1985), ‘The Stoic Theory of Universals’, Southern Journal of Philosophy 23: 87–92. Sellars, J. [2003] (2009), The Art of Living: The Stoics on the Nature and Function of Philosophy, London: Bloomsbury Academic. Shogry, S. (2018), ‘Creating a Mind Fit for Truth’, Ancient Philosophy 38 (2): 357–381. Shogry, S. (2019), ‘What Do Our Impressions Say? The Stoic Theory of Perceptual Content and Belief Formation’, Apeiron 52 (1): 29–63. Sorabji, R. (1993), Animal Minds and Human Morals: The Origins of the Western Debate, Ithaca: Cornell University Press. Sparshott, F. (1978), ‘Zeno on Art: Anatomy of a Definition’, in J.M. Rist (ed.), The Stoics, 273–290, Berkeley : University of California Press. Striker, G. (1991), ‘Following Nature: A Study in Stoic Ethics’, Oxford Studies in Ancient Philosophy 9: 1–73. Striker, G. (1997) ‘Academics Fighting Academics’, in B. Inwood and J. Mansfeld (eds), Assent and Argument: Studies in Cicero’s Academic Books. Proceedings of the 7th Symposium Hellenisticum, Philosophia Antiqua 76, Leiden: Brill. Tsouna, V. (2021), ‘The Stoics on Technē and the Technai’, in T.K. Johansen (ed.), Productive Knowledge in Ancient Philosophy: The Concept of Technê, 166–190, Cambridge: Cambridge University Press. Wildberger, J. (2006), Seneca Und Die Stoa: Der Platz Des Menschen in Der Welt, Berlin: De Gruyter.

9

Being and Becoming Good: Seneca’s Two Moral Conceptions of Ars Stefan Röttig

Introduction In this chapter, I will explore Seneca’s characterization of becoming and being good, wise or virtuous, which for a Stoic always amount to the same thing. There is one passage in which Seneca says it is an ars (i.e. skill, art) to become good; in another he says wisdom is an ars, namely an ars vitae. If one bears in mind that wisdom in Stoic philosophy stands for the best possible moral state of character a human being can develop, Seneca’s remarks cannot but attract our attention: it is an ars to become good and an ars to be good. Since it is natural to regard becoming and being (good) as two different things, it is natural to assume that the ars of the person who is striving after virtue (the proficiens) cannot be identical to the ars of the person who already is virtuous. I want to ask what the difference is. What kind of ars does the proficiens have? And how does Seneca understand the ars of the wise? I claim that, on his view, the ars of the proficiens is to be understood as the continuous endeavour to become good, wise or virtuous by handling and applying several therapeutic techniques and doing good deeds, whereas the ars of the wise is to be interpreted as a special kind of skill (or τέχνη) one acquires if one is, over a period of time, artistic in the first sense. The wise person’s skill is such that she no longer needs to apply therapeutic techniques and is doing good deeds habitually. I will provide more context for the key passages in which Seneca’s two moral conceptions of ars become apparent, and then work out the basic aspects he ascribes to the ars of the wise and the ars of the proficiens. I will start off by expounding in what respects wisdom is and is not to be separated from the crafts according to Seneca. Then I will try to show why he regards becoming good as an ars. Thereafter, I will present one crucial therapeutic technique he adduces, 185

186

Skill in Ancient Ethics

most notably in the Epistles, and try to gauge the extent to which its use by the proficiens is artistic. In doing so, the wise will serve as a comparison. Another activity that is characteristic of the proficiens’s ars is the performance of morally good acts. Again, I will ask what ars could mean here, and compare it to the ars of the wise. Finally, I will sum up the main results and evaluate the importance of Seneca’s two moral conceptions of ars in the history of philosophy.

1. Wisdom and crafts or skills In letter 90, Seneca deals with Posidonius’s view of the role the wise played in the golden age (aureum saeculum).1 He agrees with him that the government back then was under their jurisdiction, and that they invented the laws when more and more people became vicious.2 But he refuses to admit ‘[. . .] that philosophy invented the arts of which life makes use in its daily round [. . .]’ (artes quidem a philosophia inventas quibus in cotidiano vita utitur non cesserim).3 In the course of the letter, we get to know which artes cotidianae Posidonius had in mind: architecture,4 machine tool building,5 mining,6 weaving,7 farming,8 breadmaking9 and cobbling (although he seemed to have been less clear whether the latter also belongs here).10 All these crafts or skills, Posidonius opines, were invented and first practised by philosophical experts, namely the wise, who then delegated their exercise to other people.11 Seneca is arguing, however, that they did not, in fact, withdraw themselves from those arts after they had invented and practised them for a while – rather, they did not take them up at all, or even invent them.12 It is reasonable to ask why Seneca wants to separate those arts from wisdom. After all, it is common among ancient philosophers to draw a connection 1

2 3 4 5 6 7 8 9 10

11 12

The ‘golden age’ motif was frequently used by Latin authors, as Jeffrey Henderson in his edition of the Epistles notes (p. 396, b). We already find it in Hesiod’s Works and Days (cf. Hes. erga 109–120) together with four more ages. Cf. Sen. epist. 90.5 f. Cf. ibid. 90.7, trans. Richard M. Gummere. Cf. Sen. epist. 90.7. Cf. ibid. 90.11. Cf. ibid. 90.12. Cf. ibid. 90.20. Cf. ibid. 90.21. Cf. ibid. 90.22 f. Cf. ibid. 90.23 (trans. Richard M. Gummere): ‘Posidonius came very near declaring that even the cobbler’s trade was the discovery of wise men.’ Cf. Sen. epist. 90.30. Cf. ibid. 90.30.

Seneca’s Two Moral Conceptions of Ars

187

between the two. Many think that virtue is, or is importantly like, a craft or skill. Julia Annas mentions the Socrates of the early Platonic dialogues as one of the chief protagonists of the idea that virtue is a skill.13 Aristotle, as she writes, though rejecting this position, at least accepts that there are ‘considerable analogies between virtues and skills’.14 Annas adds that the Socratic stance is taken up again in Hellenistic philosophy, particularly by the Stoics.15 So why does the Roman Stoic Seneca want to separate wisdom from the crafts or skills, instead of bringing them together as other philosophers – even Stoics like Posidonius – did before him? There are at least three notable reasons: an epistemological, an action-theoretical and a moral one. I will explore these in turn.

1.1. The epistemological reason for the disconnection between wisdom and crafts or skills The wise could not have invented and first practised the crafts qua their wisdom because wisdom alone could not have made them understand how to build houses, make mechanical tools or look for iron and copper; wisdom alone could not have made them understand how to weave, farm, make bread or repair and make shoes. That is why Posidonius’s example of the wise man Anacharsis, a Scythian prince and friend of Solon the lawmaker, who supposedly invented the potter’s wheel,16 loses its force. Even if he was the creator of this wheel (which Seneca doubts), he did not invent it by virtue of his being wise (sed non tamquam sapiens);17 he invented it by virtue of his ingenuity (sagacitas).18 For Seneca, what makes the difference between wisdom and ingenuity is not that the one has its root in reason and the other not: ‘Reason [ratio] did devise all these things, but it was not right reason [recta ratio].’19 The difference lies in the content of the knowledge they provide. Seneca thinks that wisdom has a specific scope: Would you know what wisdom [sapientia] has brought to light, what she has accomplished? It is not the graceful poses of the body, or the varied notes produced by horn and flute, whereby the breath is received and, as it passes out 13 14

15 16 17 18

19

Cf. Annas 2008: 228. Cf. also van Ackeren 2017. Cf. Annas 2008: 228. Angier 2010: ch. 2 analyses the passages in which Aristotle draws parallels between virtues and skills. Moss 2011: 257 marks phronesis out as the closest analogue of τέχνη. Cf. Annas 2008: 228. Cf. Sen. epist. 90.31. Cf. ibid. 90.31. Seneca is not mentioning sagacitas in the context of the Anacharsis story, but in the context of the invention of mechanical tools (cf. ibid. 90.11 f.) Nevertheless, it is apt to speak of it here since the potter’s wheel is a mechanical tool. Ibid. 90.24, trans. Richard M. Gummere.

188

Skill in Ancient Ethics

or through, is transformed into voice. It is not wisdom that contrives arms, or walls, or instruments useful in war [. . .] It is not she, I maintain, who is the artisan of our indispensable implements of daily use [non est, inquam, instrumentorum ad usus necessarios opifex]. Why do you assign to her such petty things? You see an artisan of life [artificem vides vitae]!20

Seneca differentiates here between the artisan of indispensable implements of daily use and the artisan of life. The opifex (ingenuity) is set against the artifex (wisdom). The artisan of life, as he continues in the text, has all the other arts under her dominion (alias quidem artes sub dominio habet) because she organizes and regulates life as a whole – life serves her, and so all of its ornaments serve her, too.21 Among those ornaments are the power of ingenuity and the crafts. Seneca does not describe how, exactly, wisdom exercises power over them, but his following statement gives us a hint: ‘[. . .] wisdom’s course is toward the state of happiness [ad beatum statum tendit]; there she guides us, there she opens the way for us’.22 Wisdom makes us happy no matter what we invent or what craft activity we take up; she nurtures life as a whole with happiness, and in this sense she influences and affects every part and area of it. In this sense she has dominion. Unlike ingenuity and the crafts, wisdom is able to make us happy due to the knowledge she provides. First, she shows what things are good or bad and what things are only seemingly good or bad.23 In Stoic thought, it is virtue that is good and vice that is bad. Occasionally, we find the addition that what participates in virtue is good and what participates in vice is bad.24 Avoiding getting angry would be an example of something that participates in virtue, becoming angry would be an example of something that participates in vice – the one paves the

20 21 22 23 24

Cf. Sen. epist. 90.26 f., trans. Richard M. Gummere (adapted). Cf. Sen. epist. 90.27. Cf. ibid. 90.27 f., trans. Richard M. Gummere. Cf. Sen. epist. 90.28. Cf. Stob. anthol. 2.57.19 ff. and DL 7.101–103. Seneca believes the same, even though he is not explicitly using the language of participation. Cf. Sen. epist. 76.17 (trans. Richard M. Gummere, slightly adapted): ‘[. . .] whatever strengthens, uplifts, and enlarges the mind, is a good [. . .] that alone is good which will make the soul better’. Likewise, his remarks on selection (electio/ἐκλογή) reveal that there is something that participates in virtue, insofar as it contributes to its acquisition: ‘[. . .] the selection of neat attire, and not neat attire in itself, is a good; since the good is not in the thing selected, but in the quality of the selection’ (cf. Sen. epist. 92.12, trans. Richard M. Gummere, slightly adapted). The good thing that participates in virtue can, moreover, refer to the action that results from virtue: ‘We do indeed say that those things also are goods which are furthered and brought together by virtue – that is, all the works of virtue [opera eius omnia] [. . .]’ (cf. Sen. epist. 76.16, trans. Richard M. Gummere). What participates in vice can be interpreted in the same manner, but also in the sense that there is something that contributes to the acquisition of the vices (see next sentence and footnote).

Seneca’s Two Moral Conceptions of Ars

189

way for the acquisition of wisdom, the other contributes to the development of irascibility or even bestiality.25 Second, wisdom equips us with the knowledge of the whole of nature and of her own nature (totius naturae notitiam ac suae tradit).26 Knowing her own nature must mean that we understand what wisdom is and what it is like to be wise. Knowing the whole of nature amounts to knowing the essential features of cosmic nature: we start seeing the divine in it,27 we understand what the gods are and of what sort they are28 – things that cannot be seen with one’s bare eyes;29 the first causes are disclosed to us, among which Seneca seems to reckon allpervading eternal reason (aeterna ratio toti indita), the reason that is in seeds and enables them to produce beings of their kind.30 Finally, wisdom entails knowledge of the mind (animus): where it comes from, where it exists, how long it lasts and in how many parts it is divided, as well as of corporeal and incorporeal things, truth and its proofs, and the knowledge of ambiguities in life or language.31 But if one knows all that, one does not immediately know how to build a house, weave or make bread.

1.2. The action-theoretical reason for the disconnection between wisdom and crafts or skills Furthermore, in an action-theoretical respect, wisdom and ingenuity perform different functions for their possessor: ‘She [sapientia] bestows upon us a greatness which is substantial [dat magnitudinem solidam], but represses the 25

26 27

28 29 30 31

Cf. Sen. epist. 75.12, 85.10 and 85.15 for the view that passions have vices as their consequences if they occur regularly. That anger can lead in the long run to the gruesome vice feritas or crudelitas is a position Seneca takes up in Sen. dial. 4.5.3 (= De ira 2.5.3). Cf. Sen. epist. 90.28. Cf. ibid. 90.34 (trans. Richard M. Gummere, adapted): ‘Do you ask what, then, the wise has found out and what he has brought to light? First of all there is truth, and nature; and nature he has not followed as the other animals do, with eyes too dull to perceive the divine in it.’ Cf. Sen. epist. 90.28. Cf. ibid. 90.29 (trans. Richard M. Gummere): ‘For the vision of our eyes is too dull for sights so great’. Cf. Sen. epist. 90.29. Cf. ibid. 90.29. What makes letter 90.29 very hard to interpret is Seneca’s unexplained change from verbs that clearly stand for knowledge (ostendit, nec ignorari sinit, notitiam tradit, declarat [cf. ibid. 90.28]) to verbs that are dynamic in nature (redit, coepit inquere, transtulit, excussit [cf. ibid. 90.29]). But I cannot see why ‘being wise’ for a Stoic should mean ‘beginning to inquire into’ (the first causes, the mind, corporeal and incorporeal things, truth, proofs of truth and ambiguities in life or language). Inquiring is what philosophers and not what wise people do – they know. For the distinction between philosophy and wisdom, cf. ibid. 89.2 and 4–8. Seneca agrees with the view that wisdom is the knowledge of things divine and things human (divinorum et humanorum scientia) (cf. ibid. 89, 5), whereas philosophy’s function is to discover the truth about things divine and things human (huius opus unum est de divinis humanisque verum invenire) (cf. ibid. 90.3).

190

Skill in Ancient Ethics

greatness which is inflated and showy but filled with emptiness [. . .]’32 In De ira, Seneca defines the magnitudo animi as an unshakable, balanced and stable attitude.33 It is an attitude that makes us keep on track if we are subject to the blows of fate. We can face unpleasant experiences without being defeated by them. Ingenuity does nothing like this; it does not help us in any way to deal with fate. It merely helped the men of the golden age to invent the crafts, and it is the artisan of our indispensable implements of daily use. Another practical dimension of wisdom that is peculiar to it can be taken from Seneca’s definition of wisdom in letter 20. There he says that wisdom means always wanting and refusing the same things (semper idem velle atque idem nolle).34 He thinks it is redundant to add that it is the right thing that is always wanted, since nothing would always satisfy if it were not right.35 But it would have been helpful if he had expressed himself redundantly on this point. Seneca seems to hold that the wise always want the right thing. And if there is a symmetry between wanting and refusing in the case of the wise, as Seneca apparently assumes, then the wise also always refuse what is wrong because nothing else is always dissatisfying. How strong the ‘always’ is in the formula that wisdom means always wanting and refusing the same things is shown by a passage in Seneca’s De beneficiis: ‘For a good man it is impossible not to do what he does; for he will not be a good man unless he does it [. . .] (vir bonus non potest non facere, quod facit; non enim erit bonus, nisi fecerit)’.36 This statement implies that a wise person does not even have the ability to want what is wrong and refuse what is right. She is causally determined to want the right and refuse the wrong. The things ingenuity makes us want or refuse have, by contrast, nothing to do with right or wrong. These things belong to the sphere of the indifferents, which never always satisfy or dissatisfy. Ingenuity may make us want to plant vegetables at a certain time. But in adverse weather conditions, this can be very dissatisfying.

1.3. The moral reason for the disconnection between wisdom and crafts or skills The third reason Seneca wants to keep the crafts separated from wisdom is the contribution they make to the development of the vices. His line of argument

32 33 34 35 36

Cf. ibid. 90.28, trans. Richard M. Gummere (slightly modified). Cf. Sen. dial. 3.20.6 (= De ira 1.20.6) and the translation of Robert A. Kaster and Martha C. Nussbaum. Cf. Sen. epist. 20.5. Cf. ibid. 20.5. Sen. benef. 6.21.2, trans. Miriam Griffin and Brad Inwood.

Seneca’s Two Moral Conceptions of Ars

191

anticipates Jean-Jacques Rousseau’s idea that society corrupts the originally innocent individual. In the golden age, Seneca believes, there was enough of everything and nothing was difficult to obtain; anyone could find a natural retreat and sufficient material to build a roof: The men of that day, who had found in some dense grove protection against the sun, and security against the severity of winter or rain in their mean hidingplaces, spent their lives under the branches of the trees and passed tranquil nights without a sigh.37

With the invention of the crafts, people realized that they could live more comfortably. Consequently, materialistic demands increased, up to an ethically unacceptable level. Architects started to build shrewdly contrived houses, one higher than the other – some even as big as cities.38 Their fragility gave rise to fear, so much so that Seneca states: ‘In these days our houses [. . .] constitute a large portion of our dread.’39 Still, people wanted more and more, until luxury became a common vice.40 Wisdom would never have such consequences. On the contrary, it would make people live in accordance with nature, in the sense that they would be satisfied with what they got, not longing for more. If the people in the golden age had been wise, they would probably never have left it. They would have stayed in it, even though their ingenuity made them see the possibility of inventing and practising the crafts. Ingenuity without wisdom led the people of the golden age into the dark age of civilization.

2. Wisdom as a craft or skill Notwithstanding that Seneca tries to keep the artes cotidianae apart from wisdom, he regards wisdom as an ars. This is already indicated in letter 90, where he calls wisdom an ‘artisan of life’ (artifex vitae).41 In letter 117, he characterizes it in a similar way. There he deals with Lucilius’s question whether the Stoics are right in claiming that only wisdom, as opposed to being wise, is a good

37 38 39 40

41

Cf. Sen. epist. 90, 41, trans. Richard M. Gummere. Cf. Sen. epist. 90.7 and 43. Cf. ibid. 90.43, trans. Richard M. Gummere. For its three developmental stages, cf. Sen. epist. 90.19 (trans. Richard M. Gummere, slightly adapted): ‘At first, luxury began to lust for what nature regarded as superfluous, then for that which was contrary to nature; and finally she made the mind a bondsman to the body, and bade it be an utter slave to the body’s lusts.’ Cf. p. 188.

192

Skill in Ancient Ethics

(sapientiam bonum esse, sapere bonum non esse).42 At one point, he gives an explanation of what they had in mind when they defended such a view, a view he treats critically: A field is one thing and the possession of the field another thing [aliud est aeger, aliud aegrum habere]; of course, because ‘possessing the field’ refers to the possessor rather than to the field itself. Similarly, wisdom is one thing and being wise another. You will grant, I suppose, that these two are separate ideas – the possessed and the possessor: wisdom being that which one possesses, and he who is wise its possessor.43

Afterwards Seneca defines wisdom as ‘the perfected and optimally developed mind’ (mens perfecta vel ad summum optimumque perducta) and adds: ‘For it is the art of life’ (ars enim vitae est).44 ‘Being wise’, however, ‘[. . .] falls to the lot of the person who possesses a mind perfected’ (id quod contigit perfectam mentem habenti).45 Even though this distinction makes sense, according to Seneca, he does not agree that wisdom is good while being wise is not46 – they are both good (later he doubts whether the whole question is of any importance).47 I do not want to delve into his criticism here. Instead, I would like to focus on the formulation that wisdom is an ars vitae, which is very similar to the characterization of wisdom as an artifex vitae. What sense of ars does Seneca envisage when he characterizes wisdom in this way? Unfortunately, he does not answer this question. All he does in letter 117 is tell us something about the scope of wisdom. It has, as he states, large and spacious ‘retreats’ (secessus),48 by which he means its subjects. He mentions the nature of the gods, the nourishment (alimentum) of the heavenly bodies, the various courses of the stars, the relationship between the movements of the heavenly bodies and our own movements, the causes of our impulses and the strict laws of providence.49 Leaving aside the fact that this list of wisdom’s subjects differs from the list in letter 90, it is remarkable that the exact link between those subjects and wisdom is missing. If one takes into account that wisdom for Seneca is usually knowledge (scientia), one is inclined to say that wisdom is the 42 43 44 45 46 47 48 49

Cf. Sen. epist. 117.1. Cf. ibid. 117.12, trans. Richard M. Gummere. Cf. Sen. epist. 117.12. I am using Richard M. Gummere’s translation again, with my own emphasis. Cf. Sen. epist. 117.12, trans. Richard M. Gummere. Cf. Sen. epist. 117.6, 8–10 and 15–18. Cf. ibid. 117.19–33. Cf. ibid. 117.19. Cf. ibid. 117.19. Seneca adduces all these topics from the perspective of someone who has not yet acquired wisdom ([. . .] de deorum natura quaeramus [. . .]), despite having started as though he wanted to talk about the subjects of wisdom themselves. I tried to filter them out.

Seneca’s Two Moral Conceptions of Ars

193

knowledge of the gods, the nourishment of the heavenly bodies, etc. A wise person knows all that; she has, in Julia Annas’s words, a unified grasp of her field.50 In this sense, she is skilled, but also in another, more practical sense: the wise person knows that her life is subordinate to cosmic nature and that she can live in harmony with it only if she adapts her behaviour to its peculiarities – and she always adapts her behaviour to its peculiarities. She has a unified grasp of her field and a motivation to act according to the contents of that field. This could be the explanation for why Seneca calls wisdom an ars, a skill. Though it is indistinguishable in its basic structure from every other skill, it is still very different from every other skill. Wisdom is unique with respect to the content of the knowledge it provides, its motivational aspect and in terms of its moral perfection.

3. Becoming good as an ars Seneca attacks Posidonius’s thesis that the artes cotidianae were invented and first practised by wise people also from another angle. At the end of letter 90, he points to the different natures of wisdom and ingenuity. The title of ‘wisdom’, as he argues there, is reserved for an achievement, which he calls the highest achievement (in opere maximo nomen est).51 That means that no one is wise by nature. Ingenuity, however, looks rather like a gift from the gods, like a talent with which one is born. When he describes the golden age in the same passage, he says: ‘[The men of the golden age] were not endowed with a character of highest perfection [non erant ingenia omnibus consummata], though their nature was more powerful than ours and more fitted for toil [omnibus indoles fortior fuit et ad labores paratior]’.52 The term indoles could refer here to human nature’s mental and physical features. The men of the golden age were from birth mentally more capable, for example, more ingenious in comparison to us. At the same time, we could not have competed with their natural physical robustness (ad labores paratior). They could not have been wise, Seneca goes on, because [. . .] nature does not bestow virtue; it is an ars to become good (non enim dat natura virtutem; ars est bonum fieri).53

50 51 52 53

Cf. Annas 2008: 231 f. Cf. Sen. epist. 90.44. Cf. ibid. 90.44, trans. Richard M. Gummere (adapted). Cf. Sen. epist. 90.44 f., trans. Richard M. Gummere.

194

Skill in Ancient Ethics

Apparently, Seneca holds that the discovery of architecture, machine tool building, mining, etc., has its origin in a mental capacity that is not acquired. Wisdom, on the other hand, is an acquired capacity. This is apparent from the last quotation and is expressed in greater detail a little bit later: ‘Virtue is not vouchsafed to a mind unless that mind has been trained and taught, and by unremitting practice brought to perfection [virtus non contigit animo nisi instituto et edocto et ad summum adsidua exercitatione perducto].’54 Therefore, wisdom could not have invented the aforementioned crafts. One might wonder, of course, how convincing this argument is and whether every invention really is merely the product of ingenuity. But one detail is remarkable: becoming good (bonum fieri) is characterized as an ars. What does this mean? Why should becoming good be itself an ars? I think it is necessary to look at the training process in moral matters to answer this question. Seneca occasionally describes moral training as laborious, especially at the beginning: a weak mind fears what is unfamiliar, and if it wants to become good, it has to face what is unfamiliar.55 So the difficulty is taking the first step; each further step is not as difficult. It is similar to being sick and taking medication: first, the medication is bitter, then the healing process starts and the bitterness fades.56 But that the way to wisdom is a difficult one, especially at the beginning, can hardly be the whole explanation for why taking it is an ars. There must be more. Making progress in moral matters seems to require a certain ability to apply what one has learned in varying situations. From the perspective of a Stoic like Seneca, this means, first of all, being able to apply techniques to avoid developing the passions and those that help us liberate ourselves from occurrent passions.57 The proficiens has to know in which situations to apply which techniques and apply them in order to make moral progress. Since she is lacking wisdom, she will make mistakes, but she will try and try again if she really wants to become good; she will try until she attains the skill of the wise. Having basic knowledge when to apply those techniques and moving towards wisdom by applying them (as best one can) is what the ars of the proficiens consists in.

54 55 56 57

Cf. Sen. epist. 90.46, trans. Richard M. Gummere. Cf. Sen. epist. 50.9. Cf. ibid. 50.9. One of the main projects Seneca pursues in De ira is to show how to avoid anger and how to cease when angered (cf. Sen. dial. 4.18.1 and 5.5.3 [= De ira 2.18.1 and 3.5.3]).

Seneca’s Two Moral Conceptions of Ars

195

4. The ars of the proficiens and the ars of the wise with regard to the blows of fate The wise and the proficiens differ with regard to their levels of proficiency. The former is already skilled while the latter is en route to acquiring the skill. As a result, they act differently when fate has struck them a blow. A wise person, Seneca holds, will always say ‘I knew’ (sciebam) after fate has struck her a blow, in contrast to the proficiens, who will say ‘I knew that this was in store for me’ (sciebam hoc mihi restare).58 These utterances suggest that both knew more or less what will happen to them. But how could they know? Where did they get their knowledge from? The proficiens made use of the presumption of future evils (praemeditatio futurorum malorum) (from now on PFE). The PFE is a therapeutic technique by which a rational adult imagines that a ‘bad’ thing59 in the future – let’s say an earthquake – is happening now, with the aim of avoiding or weakening the shocking moment if the bad thing really does take place. The thought is: the weaker a shock is, the less one is inclined to succumb to a passion. The wise person must also have made use of the PFE. But why, then, do they not make the same utterances? Why does the wise person say ‘I knew’ and the proficiens ‘I knew that this was in store for me’? It is not far-fetched to assume that the wise person ‘premeditated’ when she was not wise and stopped premeditating when she gained wisdom. That does not mean that she is not prepared for possible future events. She premeditated so much that she has every possible autobiographical change present in her mind, for example, that an earthquake could happen in the next minute. As a result, nothing surprising will ever happen to her again. If the imagined earthquake really does take place, she may feel a mental jolt,60 but she won’t fall into a passion. The wise person’s premeditative knowledge and its practical effect is what her ars consists in. The proficiens has thought about possible personal calamities, too. But there is something lacking in her premeditative knowledge. Even though she has thought about what could happen to her in the future (for example, that she will witness an earthquake in the next minute), she might still be surprised if it really does

58 59 60

Cf. Sen. epist. 76.35, trans. Richard M. Gummere. ‘Bad’ not in a Stoic, but in a common-sense sense. For the possibility that even the wise suffers mental jolts, cf. Sen. dial. 3.16.7 and 4.2.2 (= De ira 1.16.7 and 2.2.2) and Sen. epist. 11.1 and 6.

196

Skill in Ancient Ethics

happen. Why is it likely that she is surprised when the event occurs, whereas the wise person is not? The difference is that the proficiens has not completed her premeditating activity. She knows what to do to make moral progress and she does it as best she can. That means in the case of the PFE that she is already asking which dispreferred events could happen to her, and she responds, for example: ‘It is possible that an earthquake is going to happen in the next minute.’ But unlike the wise, she does not yet strongly believe that an earthquake could happen in the next minute. She has not yet reached the level of fully developed premeditative knowledge. If she wants to gain that knowledge, she has to keep on practising the ars vitae. Then she will eventually say not only ‘I knew’, but also react appropriately.

5. Doing good deeds as an activity of the proficiens and an activity of the wise Applying therapeutic techniques is, in Seneca’s view, not the only way for the proficiens to make moral progress. She can also give, receive and return benefits (beneficia). A benefit is a ‘[. . .] well-intentioned action [benivola actio] that confers joy and in so doing derives joy, inclined towards and willingly prepared for doing what it does [. . .] [it] consists not in what is done or given, but rather in the intention [in animo] of the giver or agent’.61 A benefit is, as one might say, a ‘matter of attitude’, and it is to be distinguished from the matter that is conferred by it, the materia beneficii,62 which is corporeal and thus subject to fate. Helping a friend out with money can, accordingly, be a benefit or merely a useful action, depending on the attitude with which it is given. If it is a benefit, it will last even if the conferred money gets lost.63 The attitude is crucial for the receipt (beneficium accipere) and return of benefits (beneficium reddere/gratiam referre) as well: Some people don’t just give benefits in an arrogant manner; they even receive them in that spirit, and that is an offence one should never commit.64 Returning the favour while envisaging a second gift amounts to ingratitude; it means hoping while repaying. Ungrateful is what I call a man who sits by a sick 61

62 63

64

Cf. Sen. benef. 1.6.1. I use here and in what follows the translation of Miriam Griffin and Brad Inwood. Cf. Sen. benef. 1.5.2 and 6.2.2. Cf. ibid. 6.2.2: ‘[. . .] when you snatch away ‹from someone the material of the benefit that you gave, you do not also snatch away the benefit›’. Cf. ibid. 2.18.1.

Seneca’s Two Moral Conceptions of Ars

197

man because he is about to make his will and finds time to think about an inheritance or legacy. Let him do everything that a good friend conscious of his obligation ought to do: if the hope of gain is present to his mind, he is fishing for legacies and dropping the hook.65

Part of the right attitude in giving, receiving and returning benefits is the consideration of circumstances.66 One has to pay attention to the when (quando), the where (ubi), the how (quemadmodum), the who (qui), the whom or fromwhom (cui/a quo), the why (quare) and the what (quid/materia beneficii).67 On closer inspection, Seneca seems to draw an implicit difference between factual and normative circumstances: it is important to know the actual time and place, in what manner and to what kind of person (as a king or Cynic philosopher) one is about to give, receive or return a benefit, on whom one is conferring or returning and from whom one is receiving it, what one’s reason is and what one is giving, receiving or returning. But it is equally important to know the right time and place, whether one really should give, receive or return in the manner one is about to give, receive or return, whether one should do it as the kind of person one is, whether one should give or return a benefit to this specific person or receive one from her, whether one should do it for this reason and whether one should give, receive or return this. For Seneca, circumstances of the same kind can fit or not fit together. He thinks with regard to the when, for example, that a benefit should be given before we are asked.68 So when we give it before we are asked, we give it at the right time; if we give it after we have been asked, the actual when would not fit the normative when.69 Apparently, Seneca also assumes that circumstances of a different kind can fit or not fit together: it would be right to send summer clothes at midsummer and winter clothes in winter, but foolish to send summer clothes in winter and winter clothes at midsummer.70 In this case, the actual ‘what’ would(n’t) fit the normative ‘when’. A proficiens will do her best to take all the factual and normative circumstances into account and adapt her action to them. But since she has not yet fully incorporated the ars vitae, she will make mistakes: ‘[. . .] the fool [stultus] [. . .]

65 66

67

68 69 70

Cf. ibid. 4.20.3. Seneca has no significant term for ‘circumstances’. The Latin term circumstantia becomes first prominent in Quintilian, who translates περίστασις in this way (cf. Quint. inst. orat. 5.10.104). Cf. Sen. epist. 89.15 for the quando, ubi and quemadmodum, cf. Sen. ben. 2.16.1 for the qui, cui and quare (the quando and ubi are also mentioned here), and cf. ibid. 1.15.4 for the a quo and quid. Cf. ibid. 2.1.3. Cf. ibid. 2.2.1: ‘[. . .] it is too late when you have given upon request’. Cf. ibid. 1.12.3.

198

Skill in Ancient Ethics

no matter how good his intentions may be either pays less than he owes, or pays it at the wrong time or the wrong place’.71 Seneca attributes these failures to a lack of knowledge (scientia), not to a lack of will (voluntas).72 What he means by that becomes clearer if we look at his description of the wise person’s behaviour in returning benefits. The wise person ‘[. . .] knows exactly what value should be put upon everything’.73 She first inquires (examinabit) how much she has received, from whom, when, where, how, etc.,74 and then compares all things with one another (omnia inter se comparabit).75 I understand this comparison as a practical deliberation in which the wise person figures out whether a factual circumstance fits a normative one – if it does not fit, she will wait until the right time has come, she has found the right gift, and so on. The proficiens thus makes the mistake either on the factual level (for example, she miscalculates how much she has received) or on the normative level (despite a correct assessment of how much she has received, she fails to repay as much as she should). Both the factual and the normative levels are tricky: the former requires a high degree of attentiveness and good memory, the latter context-sensitivity and flexibility in creating the action-guiding rules (sometimes a thousand denarii are more than the most lavish gifts).76 The ars of the proficiens consists in internalising the scheme of the different circumstances and filling it with content (as best she can).

6. Conclusion For Seneca, virtue or wisdom is a skill, in the sense that it provides its possessor with comprehensive knowledge and a disposition to act according to that knowledge. But it differs from other skills in terms of what it is a knowledge of, how strongly it motivates to act according to that knowledge and its moral perfection. The wise person has a complete grasp of how the world is, what her place is within it and what the actual and normative circumstances are in giving, receiving and returning benefits – and wisdom makes her live accordingly (she would never rebel against a blow of fate, nor make mistakes in the field of benefits). The proficiens wants to have that skill and does everything she can to 71 72 73 74 75 76

Cf. Sen. epist. 81.8. Cf. ibid. 81.13. Cf. ibid. 81.8: Uni sapienti notum est quanti res quaeque taxanda sit. Cf. ibid. 81.10. Cf. ibid. 81.14. Cf. ibid. 81.14.

Seneca’s Two Moral Conceptions of Ars

199

acquire it. Her constant endeavour to do the right thing, to apply therapeutic techniques and give, receive and return benefits appropriately, are regarded by Seneca as an ars, meaning a rudimentary moral skill. Even if he is not the first philosopher to distinguish between being in possession of virtue and becoming virtuous,77 he seems to be the first one to call both stages of moral development an ars.78

References Angier, T. (2010), Technē in Aristotle’s Ethics: Crafting the Moral Life, London: Continuum. Annas, J. (2008), ‘Virtue as a Skill’, International Journal of Philosophical Studies 2 (3): 227–43. Diogenes Laërtius (2013), Lives of Eminent Philosophers, ed. with introduction by Tiziano Dorandi (Cambridge classical texts and commentaries, Vol. 50), Cambridge: Cambridge University Press. Hesiod (2010), Theogony. Works and Days. Testimonia, ed. and trans. by Glenn W. Most (Loeb classical library, Vol. 57), Cambridge, MA : Harvard University Press. Moss, J. (2011), ‘ “Virtue Makes the Goal Right”: Virtue and Phronesis in Aristotle’s Ethics’, Phronesis 56: 204–61. Plato (1997), Complete Works, ed. with Introduction and Notes, by John M. Cooper, Associate Editor D.S. Hutchinson, Indianapolis, IN : Hackett Publishing. Quintilian (2002), The Orator’s Education, Vol. II: Books 3–5, ed. and trans. Donald A. Russell (Loeb classical library, Vol. 125), Cambridge, MA : Harvard University Press. Seneca, L.A. (1965), L. Annaei Senecae Ad Lvcilivm epistvlae morales, ed. Leighton D. Reynolds (Oxford classical texts), Oxford: Oxford University Press. Seneca, L.A. [1920] (1991), Epistles 66–92, trans. Richard M. Gummere (The Loeb classical library, Vol. 76), Cambridge, MA : Harvard University Press. Seneca, L.A. [1961] (2003), Des bienfaits, ed. and trans. François Préchac, 2 vols (Collection des Universités de France), Paris: Les Belles Lettres. Seneca, L.A. [1977] (2008), L. Annaei Senecae dialogorvm libri dvodecim, ed. Leighton D. Reynolds (Oxford Classical Texts), Oxford: Oxford University Press.

77 78

Cf. Plat. Symp. 203b–204c. I would like to thank Ondřej Krása, Robert Roreitner and Máté Veres for giving me the opportunity to present this paper at the 2nd Central European Graduate Conference in Ancient Philosophy in Prague. Special thanks go to my respondent István Bodnár for his helpful comments and to the participants who initiated a lively and fruitful discussion. I also benefited from the discussion of this paper in the philosophical colloquium at my home department in Würzburg and from remarks by Jon Bornholdt on matters concerning the English language. Last but not least, I would like to thank Tom Angier for his comments and proof-reading services.

200

Skill in Ancient Ethics

Seneca, L.A. (2010), Anger, Mercy, Revenge, trans. Robert A. Kaster and Martha C. Nussbaum (The complete works of Lucius Annaeus Seneca), Chicago, IL : The University of Chicago Press. Seneca, L.A. (2011), On Benefits (The complete works of Lucius Annaeus Seneca), trans. Miriam Griffin and Brad Inwood, Chicago, IL : The University of Chicago Press. Stobaeus (1884–1912), Ioannis Stobaei Anthologium, ed. Curt Wachsmuth/Otto Hense, 5 vols, Berlin: Olms-Weidmann. van Ackeren, M. (2017), ‘›Technê‹-Analogie’, in Christoph Horn/Jörn Müller/Joachim R. Söder (eds), Platon-Handbuch. Leben – Werk – Wirkung, 342–47, Stuttgart: J. B. Metzler.

Part Four

Skill in Confucian Ethics

201

202

10

Cultivating Goodness or Manifesting Goodness: Two Interpretations of the Mencius Winnie Sung

1. The question This chapter explores Mencius’s view on the kind of effort that is needed to attain a state in which one is fixated on propriety. In Mencius 2A:2, Mencius claims that he attained the unmoved xin ᗳ (heart/mind) at the age of forty. Very roughly, we may understand the unmoved xin as a state in which xin is fixated on propriety.1 That Mencius attained the unmoved xin only when he was forty suggests that xin’s fixation on propriety is not a given state. In the following, I will discuss why, in Mencius’s view, xin is not by default in the unmoved state. Once we have a better understanding of what the obstacle is, we can better understand what needs to be done to attain the unmoved xin. The basic picture of xin that most scholars can agree on is this: in 2A:6, Mencius claims that human beings have four ‘duan’ ㄟ, which has the literal meaning of a ‘tip’ (Allan 1997). By extension, scholars have taken ‘duan’ to mean ‘clues’ (Zhu Xi), ‘principles’ (Legge 1895), ‘germs’ (Lau 1970), ‘sprouts’ (Van Norden 2008) or ‘beginnings’ (Angle and Tiwald 2020). The four duan are: the xin of sorrow and pain for others (ce yin ᜫ䳡), the xin of shame and dislike (xiu wei 㗎ᜑ), the xin of declining and yielding (ci rang 䗝䇃) and the xin of right and wrong (shi fei ᱟ䶎). Corresponding to the four duan are, respectively, the moral attributes of benevolence (ren ӱ), propriety (yi 㗙), observance of ritual (li ⿞) and wisdom (zhi Ც).

1

This is a roughly drawn characterization but should suffice for our purpose. I will only be focusing on the state of unmoved xin that Mencius has attained. We can bracket why Mencius thinks that Gaozi’s unmoved xin is inferior.

203

204

Skill in Ancient Ethics

Mencius also thinks that xin naturally delights in what is morally good. In 6A:7, Mencius says that human beings all have common preferences. Just like everyone’s palate prefers certain tastes, ears prefer certain sounds and eyes prefer certain sights, xin also prefers certain things. According to Mencius, propriety (li yi ⨶㗙) delights xin just like meat delights the mouth.2 In 6A:8, it is said that qi ≓ in the night (ye qi ཌ≓) has a restorative effect. After one night of restoration, human beings’ likes and dislikes are quite similar in the early morning (6A:8). The basic picture we may infer from these passages is that xin has predispositions that tend towards propriety. The ambiguous part of the picture that divides interpretations is this: if xin is constituted in such a way that it has moral predispositions and naturally delights in propriety, it is unclear why it seems rather difficult for xin to attain the state wherein it fixates on propriety. The issue here is not how the environment might not cooperate and thwart one from acting morally. Rather, Mencius seems to be suggesting that xin itself might not always orient towards propriety. How, in Mencius’s view, is it possible for xin not to be in a state that fixates on propriety? In the following, I will outline two ways of interpreting why it is possible for xin not to fixate on propriety. It is difficult to place an existing account neatly under a model, for many of the existing discussions tend to slide between the models. Therefore, I will not attribute any of the interpretations to a particular scholar. When relevant, I will highlight numbers of passages in bold to indicate that readings of those passages vary depending on the interpretations we adopt.

2. The Cultivation Model On this interpretation, xin is not by default unmoved because it starts off only with moral sprouts that are susceptible to hostile forces and could be damaged. Even though xin is endowed with moral predispositions, it is only when xin is cultivated and strengthened that it can become ‘established’ (li ・) and unmoved. I will group accounts that interpret Mencius as saying that xin begins with weak moral predispositions that need to be strengthened under the heading ‘Cultivation Model’. 2

It is difficult to understand what li ⨶ means here. There is little evidence in the text that suggests li is also a moral ideal like yi 㗙. 6A:7 is the only passage that we see the two terms appear together. Aside from this passage, Mencius has not made any comments on the concept of li yi ⨶㗙 elsewhere in the text. 5B:1 and 7B:19 are the only two other passages in which the term li appears, though not highlighted as the topic of discussion. Hence, it is unlikely that li has the same moral status as yi for Mencius.

Cultivating Goodness or Manifesting Goodness

205

There are different ways to capture how the weak moral predispositions are to be developed. One might think of the predispositions as a spark that can start a fire, a depleted air sac that can be filled with air, or a sprout that that can be nurtured and grown. The choice of metaphor bears on what we take the nature of such predispositions to be. At a general level, what is relevant for our consideration of this interpretation is that the moral predispositions of xin need to be strengthened. Since the metaphor that is often discussed in the literature is the sprout model, I will be discussing this interpretation by using the sprout metaphor. The Cultivation Model is widely accepted in modern literature. Its popularity is probably due in part to the assumption that duan ㄟ means ‘emergent shoot’ (Graham 1989) or ‘sprouts’ (e.g. Xu 1969; Yearly 1990; Nivison 1996; Ivanhoe 2000; Van Norden 2000; Wong 2015). When duan is translated as ‘sprouts’, 2A:6 reads: The xin of sorrow and pain for others is the sprout of ren. The xin of shame and dislike is the sprout of yi. The xin of declining and yielding is the sprout of li. The xin of right and wrong is the sprout of zhi.

Translated in this way, one might interpret Mencius as saying that xin is endowed with good sprouts, which under normal circumstances, will develop in the direction of propriety and grow into moral attributes. However, precisely because they are sprouts, it takes time for them to grow and expand into moral attributes. What then could thwart the growth of the sprouts? Mencius speaks of the difference between xin and the senses in terms of a big part opposing a small part in 6A:15: [Gongduzi asked,] ‘Why do some people go with the great body (da ti བྷ億) whereas some go with the small body (xiao ti ሿ億) even though they are all human beings?’ [Mencius replied,] ‘The organs of hearing and sight are unable to think (si ᙍ) and can be blinded by external things. When material objects [senses] interact with material objects [objects of desires], they are led away by it. But the organ of xin thinks. If xin thinks, then it obtains it. If it does not think, it does not obtain. This is what Heaven has given me. By first establishing (li ・) the great part (da zhe བྷ㘵), the small part (xiao zhe ሿ㘵) cannot seize its authority. This is being a great person.’

The ‘great body’ and the ‘small body’ here refer respectively to xin and the senses. According to Mencius, it is only when xin is ‘established’ that the small

206

Skill in Ancient Ethics

part cannot ‘seize’ its authority. ‘Seizing authority’ here may mean sensory desires end up motivating the agent and thwarting the growth of an unestablished xin. The worry that xin is not established pertains to the process of growing the sprouts. During this process, xin is not in a stable condition to respond properly to propriety. Even though it motivates one to perform a good act, such as saving a child, it is not an act that stems from ren because the sprout has not yet developed into ren. Sometimes, one’s spontaneous feeling/thought might be too weak to motivate one to act. For example, even though one might feel ashamed when one takes things that are given to them with insult, one still accepts them because they satisfy sensory desires (6A:10). On the Cultivation Model, there are at least two necessary conditions for the cultivation of xin: si ᙍ (reflection) and a good environment. It is said in 6A:15 that si is what enables xin to be the greater part. Since the sense organs do not have the capacity to si, they will be blinded and led astray by external objects. By contrast, xin can si. In virtue of si, xin can ‘obtain’ (de ᗇ) goodness and establish itself.3 Si can be translated as ‘think’, ‘reflect’, ‘focusing attention’, ‘observation’, and ‘deliberation’. On the Cultivation Model, si in 6A:15 will mean something like reflection. However, since Mencius thinks that xin’s preference for propriety is analogous to the mouth’s preference for meat, si probably does not mean reflectively deliberate about what is morally good. When xin encounters moral goodness, it naturally enjoys it just as the mouth naturally enjoys meat. Moreover, Mencius also thinks that human beings will have certain feelings/thoughts spontaneously aroused. For example, when one sees a child who is about to fall into a well, one will spontaneously feel alarmed and pained (2A:6); when one sees one’s parents’ corpses being abandoned in the gullies and eaten by wild animals, one will spontaneously sweat and feel unable to look at the horrible sight (3A:5). These suggest that si is not so much about reflecting on the situation nor is it about the right way to react to a situation. It is more plausible that si means reflecting one one’s spontaneous feelings and thoughts (e.g. Van Norden 2000; Wong 2015). As David Wong puts it, one has ‘to reflect in an emotionally active way on a particular manifestation of his sprout’ (2015: 170). After someone has performed a good act of saving the child, she will experience pleasure, for her xin naturally delights in propriety. If she reflects on this pleasant feeling/thought, she will be motivated to extend these natural 3

In 6A:6, Mencius draws a connection between si and obtaining (de) ren, yi, li, and zhi. This suggests that the moral attributes are the objects of de. It suffices to say that xin, by si, obtains what is morally good. See Shun 1997: 149–150 and Chan 2016 for further discussions of scholarly debates on the concept si in the Mencius.

Cultivating Goodness or Manifesting Goodness

207

feelings/thoughts (tui ᧘ 1A:7) to more situations and in doing so, nurturing the sprouts. Another necessary condition for moral cultivation is a good environment. It is crucial to ensure that the spontaneous feelings/thoughts that are congenial to morality do not get stifled in a hostile environment. A hostile environment might prevent one from reflecting on one’s spontaneous feeling/thought. Imagine someone who grows up in an environment where she has to constantly perform morally bad actions. When she sees a child who is about to fall into a well, she spontaneously feels alarmed and pained. She immediately runs to save the child. However, after saving the child, she has to quickly get back to work and perform a series of morally bad actions. She does not have time to reflect on the good feelings/thoughts derived from saving the child. In this case, even though she experiences the feeling/thought of compassion, her sprout of ren is not developed. Or imagine an environment that encourages desensitization: one might initially feel pained by others’ suffering but since it is normal that people suffer in the environment, over time one might become desensitized to the suffering of others. In such hostile environment, one can no longer feel the way xin is predisposed to feel. This could be what Mencius means by one has forgotten or lost one’s original xin (2A:2, 6A:6, 6A:10). By contrast, a good environment is one that encourages the growth of and reflection on spontaneous feelings/thoughts. For example, someone might naturally have the urge to offer her seat to an elderly person. In a society that has established ritual practice of offering seat to the elderly, one can fully express her xin of declining and yielding (ci rang) by following the ritual practice. Along with other ritual practices that encourage declining and yielding, one will have plenty of occasions to reflect on her feeling/thought of declining and yielding. Overtime, one’s natural feeling/thought grow and expand to moral attributes so that xin is fully inclined to act morally (zhi ᘇ) instead of performing acts that happen to be in accordance with morality (2A:2).4 The Cultivation Model can explain most of the passages in the Mencius and is reinforced by discussions of xin in relation to cultivation and nourishment. For example, Mencius discusses how the barren Mount Niu, like our xin, was once full of trees but the hostile environment damaged it (6A:8). He also gives the example of a foolish farmer who pulls at the seedlings in the hope of helping them grow but ends up shrivelling and damaging them, to illustrate the danger 4

See Shun 1997: 154 for discussion of ji yi 䳶㗙 and yi xi 㗙㾢 in 2A:2. The second interpretation discussed by Shun, which takes the distinction to be between ‘fully inclined to so act’ and ‘forcing oneself to act in accordance with yi against one’s inclinations’ is compatible with the Cultivation model.

208

Skill in Ancient Ethics

of forcing moral cultivation (2A:2). There are also references to the importance of nourishing (yang 伺) xin (6A:8; 7B:35). The Cultivation model can also explain Mencius’s worry about forcing moral cultivation. Doing so is bad because one might get into a routine of performing moral acts without the appropriate underlying feeling/thought. For example, if a child is always forced by her parents to offer her seat to elderly before the feeling/thought of ci rang arises in her, the child will get used to offering seat without feeling/thinking about ci rang. It might get to a point that she will offer her seat on her own accord. However, her act of offering is not due to her feeling of ci rang, but to her being conditioned to yield. Mencius worries that this way of forcing one to do moral acts when one does not have the appropriate underlying feeling/thought will ‘shrivel’ the sprout of ritual. There are a few considerations, though not decisive, against the Cultivation Model. First, we should also keep in mind Mencius’s claims that: Ren, yi, li, zhi are not externally welded to me. I originally have it. (6A:6)

And, All the ten thousand things are there in me. 7A:4; trans. Lau

These passages suggest that, in Mencius’s view, xin already has ren, yi, li, zhi, not just the sprouts. Second, in 1A:7, King Xuan asks Mencius whether he is capable of protecting his people and being a true king. Mencius assured the king that he can be a true king, citing an incident when the king spared an ox for sacrifice because he could not bear to see the ox shrinking with fear, just like an innocent person who was about to be executed. Mencius says that the king’s xin which reacts compassionately to the suffering of the ox is sufficient (zu 䏣) for him being a true king. If the king fails to care for his people, it is simply the case that he does not act (bu wei н⛪), not because he is unable (bu neng н㜭) to act. Mencius then compares the king’s claim that he cannot care for people but can care for the ox to those who claim that they can lift hundreds of kilograms but cannot lift a feather and those who claim that they can see the tip of a fine hair but cannot see a cartload of firewood. If they have the strength to lift hundreds of kilograms, they have the strength to lift a feather. If they claim that they cannot lift a feather, it simply is the case that they do not lift the feather.

Cultivating Goodness or Manifesting Goodness

209

It is worth noting that Mencius has not attributed the king’s failure in caring for his people to the king’s xin being injured. If Mencius held the view that the king’s xin has sprouts that need to be cultivated, we would expect him to tell the king to nurture the sprouts . Instead, all Mencius has suggested is that the king is not using his xin. Mencius clearly thinks that the king’s xin is sufficient to enable him to care for his people. It is difficult for the Cultivation Model to make sense of why Mencius has not advised the king to tend to the sprouts. The explanation cannot be that Mencius thinks that the king’s xin is already established but he somehow is distracted by his selfish desires. If the king’s xin is already established, according to 6A:15, his selfish desires could not have prevented him from caring for his people. Furthermore, it is questionable whether Mencius thinks that the moral predispositions of xin can be weakened or lost. A few references in the Mencius do suggest that xin can be ‘wrong’ (4A:20), ‘injured’ (7A:27), or ‘blocked’ (7B:21). However, Mencius also says the problem with many people is that they ‘abandon’ (6A:8) or ‘put aside’ (fang ᭮) their xin (6A:11), not knowing that they have to seek their lost xin. And Mencius seems very optimistic that as soon as one seeks xin, one will obtain it (6A:6; 6A:15; 7A:3). This could also suggest that xin itself is still good; it is just that people sometimes do not use their xin.

3. The Manifestation Model Let us turn to consider an alternative interpretation that does not take Mencius to be saying that human beings are endowed with good but weak moral predispositions. A second possible interpretation is that the good predispositions of xin are sufficient to enable one to be good. These predispositions do not need further strengthening. The key to attaining unmoved xin lies in putting the predispositions to use. This interpretation seems to be more widely accepted in the Neo-Confucian period than in modern scholarship (e.g. Zhu Xi 1130–1200 and Wang Yangming 1472–1592). I will label interpretations that take Mencius to be saying that our endowed good predispositions are sufficiently strong as the ‘Manifestation Model’. Another way to understand duan is to understand it as the tip of a certain thing that indicates to one the presence of something more. Zhu Xi, for example, takes the word to mean ‘clues’. On the Manifestation Model, 2A:6 is read as:

210

Skill in Ancient Ethics

The xin of sorrow and pain for others is the clue about ren. The xin of shame and dislike is the clue about yi. The xin of declining and yielding is the clue about li. The xin of right and wrong is the clue about zhi.

On this reading, there is no suggestion that the endowed good predispositions need to be grown or tended. These good predispositions give us clues about the moral attributes that we already have. For example, the alarm and pain that one feels when one sees a child in danger reveals that one already has ren (6A:6; 7A:4). Under ideal conditions, there is no hindrance to the manifestation of xin’s good predispositions. However, when there is hindrance, xin’s manifestation of its good predispositions is blocked. What could have prevented xin from manifesting its good predispositions? The simple explanation is that sensory desires have hindered xin’s manifestation of its good predisposition. But this explanation is unsatisfactory. Recall that Mencius also thinks that if xin is established, the senses are in no position to compete with it (6A:15). If xin is already good and established, the senses are not in a position to hinder it. Something else must explain why xin is not established in the first place. In the following, I will consider two possible explanations for why xin is not established.

3a. The Manifestation-Will Model One possible explanation for why xin is not established is that the person does not act on xin. Although xin itself is endowed with good predispositions and is good in its entirety, the person has to use xin. In 6A:8, Mencius quotes Confucius: Confucius said, ‘Hold on to it and it will remain; let go of it and it will disappear. One never knows the time it comes or goes, neither does one know the direction.’ It is perhaps to [xin] this refers. trans. Lau

We can take Mencius to be saying that the person (ren Ӫ) should hold on to xin. And if one lets go of xin, it disappears. On this reading, what determines the manifestation of xin’s goodness is the person’s using xin. Under ideal conditions, the person’s use of xin and xin’s goodness is manifested in full view. However, due to the presence of sensory desires, the person could be ‘led away’ by objects of desires. When distracted, the person neglects to use xin. On this reading, si means one’s directing attention to and seeking propriety. 6A:15 therefore reads:

Cultivating Goodness or Manifesting Goodness

211

[Gongduzi asked,] ‘Why do some people go with the great body (da ti) whereas some go with the small body (xiao ti) even though they are all human beings?’ [Mencius replied,] ‘One does not seek goodness (si)5 when one goes with the senses and can be blinded by external things. When material objects [the senses] interact with material objects [objects of desires], they are led away by them. But the organ of xin seeks goodness. If one seeks goodness, then one obtains it. If one does not seek goodness, one does not obtain. This is what Heaven has given me. By first establishing (li ・) the great part (da zhe བྷ㘵), the small part (xiao zhe ሿ㘵) cannot seize its authority. This is being a great person’.

This reading takes Mencius to be following up on Gongduzi’s question about why some people go with the great body. Mencius replies that if people use their xin (or go with their xin), they will seek goodness. Once they seek goodness, they will obtain it, presumably because xin already has good predispositions that enable one to be good. The grammatical subject of si in the line ‘si ze de zhi’ ᙍࡷ ᗇѻ is taken to be the person, not xin. ‘Going with xin’ (cong ᗎ) means going in the direction of xin. It could but does not necessarily mean that one who cong self-consciously brings oneself to follow certain directives. To avoid the suggestion that one who cong her xin must self-consciously follow her xin, I refer to the person’s exercising the capacities of xin as ‘using xin’ and assume that using xin will necessarily lead to seeking goodness. However, Mencius has not quite answered Gongduzi’s question. Gongduzi is asking why some people use xin but some do not, even though they are all human beings. As noted above, suppose one walks by a child who is about to fall into the well, one will immediately feel alarmed and pained. This suggests that at least in the very moment, one uses one’s xin. Then why do some people get distracted later but some do not? 2A:2 provides part of the answer: If zhi is concerted or concentrated (yi ༩), then it moves qi (vital energies). If qi is concerted, then it moves zhi. Now stumbling and hurrying is qi, but it turns back to move [xin]. trans. Chan 2002:476

‘Zhi’ is usually translated as the ‘will’ (Lau 1970; Legge 1990; Van Norden 2008). Some scholars take zhi in the Mencius to mean goal or direction of xin (Shun 1996; Chan 2002). We may leave ‘zhi’ untranslated for now. Mencius’s 5

6

Shun notes that Arthur Waley (1964) takes si to mean ‘focusing attention on something’. Nivison 1996 and Shun 1997 also take si to mean focusing attention on something. This reading of si is compatible with taking si to mean seeking goodness on the Manifestation-Will Model. For detailed analyses of 2A:2, see Riegel 1980, Nivison 1996, Shun 1997 and Chan 2002.

212

Skill in Ancient Ethics

remark in 2A:2 suggests that qi (vital energies) and zhi stand in a reciprocal relation. Let us first focus on the qi moving zhi part. There are many possible ways for qi to be concentrated and it might affect zhi in bad or good ways. For example, one might naturally have strong sensory desires, live in an environment that incites the indulgence of sensory desires, or have bad habits such as sleeping late or eating unhealthy food. In these instances, one’s qi will be concentrated and move zhi in a problematic way. Qi can also affect xin in a good way. Mencius, for example, said that he was good at nourishing his ‘flood-like qi (hao ran zhi qi ⎙❦ѻ≓)’. Presumably, flood-like qi is a kind of qi that positively affects xin. Mencius describes floodlike qi as follows: This is a qi which is, in the highest degree, vast and unyielding. Nourish it with [forthrightness and do not harm it] and it will fill the space between Heaven and Earth. It is a qi which unites yi and the Way. Deprive it of these and it will starve. It is born of deriving yi from xin (ji yi 䳶㗙), not from sources independent of xin (yi xi 㗙㾢).7 Whenever one acts in a way that falls below the standard set in one’s [xin], it will starve. 2A:2; trans. Lau, modified8

This passage suggests that flood-like qi can be nourished by accumulating acts of propriety. For example, someone might be in a good environment that constantly encourages her to perform good acts. In constantly performing good acts, one also finds pleasure in xin because xin naturally delights in propriety (6A:7). When xin is delighted, it generates good qi. Over time, good qi accumulates and becomes flood-like. The flood-like qi can in turn sustain one in the state of using xin. This could be what Mencius means by if xin is established, the senses cannot compete with it (6A:15). It also explains Mencius’s worry about forcing the process of self-improvement. If one performs moral acts when one’s attention is not on xin, then her acts will further generate qi that feed sensory desires. For example, one might act in accordance with propriety to get a reward. These moral acts are in effect further damaging and distracting the person from using xin. This is also probably why Mencius is adamant that the source of one’s moral

7

8

See Shun 1997: 154 for discussion of ji yi and yi xi. The first interpretation discussed by Shun, which takes the distinction to be between ‘deriving yi from the heart/mind’ and ‘acquiring yi form sources independent of the heart/mind’, is compatible with the Manifestation-Will model. On this reading, the line about not forcing moral effort is punctuated as ‘wu zheng xin, wu wang, wang zhu zhang ye’ य↓ᗳयᘈयࣙ䮧ҏ. Mencius is saying one should not correct xin. One should not forget about using xin but one should also not be too eager in the effort of nourishing hao ran zhi qi.

Cultivating Goodness or Manifesting Goodness

213

act has to be something in us, just like how the source of finding certain food delicious lies in our tasting the food, not something external in the food (6A:4). What about a petty person, whose qi is concentrated in a way that moves zhi in a problematic way? One might be born into a systemically bad environment and from the beginning of one’s birth, one’s qi is concentrated in a bad way. The more qi concentrated, the more distracted one is from using xin. How, then, can a petty person ever turn her attention to xin when her attention is completely occupied by objects of desires? One plausible explanation is that Mencius has something like an agent’s will involved in his ethical picture. However, we should not be quick to think that ‘zhi’ belongs to a distinctively mental realm. Since qi can also move zhi, it is possible that zhi also physical aspects. Here, by ‘will’, I simply mean the control one has over what wants to do. This brings us to the zhi moving qi part.9 Mencius also says in 2A:2 that zhi is the commander of qi: [Zhi] is commander (shuai ᑕ) over the qi while the qi is that which fills the body. Wherever [zhi] goes qi follows. Hence it is said, ‘Take hold of your [zhi] and do not [desiccate] your qi. (trans. Lau, modified)10

This suggests that even though zhi and qi stand in a reciprocal relation, zhi is in a superior position. It can command qi and determine where qi should go. Therefore, even if one is born with strong desires and grows up in a systemically bad environment, it still is possible for one to use xin as long as one is determined to seek goodness. As long as one seeks goodness, one will obtain goodness. There is no additional skill that one needs to learn or cultivate because xin is predisposed everything that enables one to be good. Hence, on this interpretation, zhi means something akin to the ‘will’.11 There is one more gap in the explanation. Even though one can be good when one seeks goodness, one does not necessarily seek goodness. To complete the explanation, Mencius must have assumed that human beings are beings who naturally use xin. If we set aside environmental influences, human beings’ spontaneous reactions reveal that they naturally use xin (1A:7; 3A:5; 6A:8;

9

10 11

It should be noted that even if zhi is something akin to will, it does not mean that weakness of the will must be possible on Mencius’s account. Mencius could think that it is only when one uses xin that one truly believes that the act is right. In modifying this translation, I consulted Riegel 1980. Nivison 1996 suggests there is some voluntaristic element in Mencian ethical picture (Chapters 5–6). Cua 2002 also suggests that ‘lack of will’ could be a reason for moral failure. However, some other reasons Cua suggests, such as lack of means to support a constant xin, are not necessarily compatible with the Manifestation-Will Model.

214

Skill in Ancient Ethics

6A:10). That human beings naturally use xin suggests that one’s zhi, when functioning properly, has goodness as its goal. If this right, then the obstacle to one’s using xin does not lie in the person not knowing how to use xin nor where to direct zhi. The obstacle to using lies in whether one’s zhi is exercised. Just as, if there is a smell, one naturally uses the nose to smell it. Assuming that the nose itself is functioning properly, if someone cannot detect the odour, then it can only be the person not exercising her capacity to smell. Similarly, the reason one who naturally uses xin to seek goodness does not do so is that she is just not exercising her capacity to seek goodness. Returning to our earlier question, suppose one’s zhi is weakened and one’s qi is concentrated in a problematic way, how can zhi be strengthened so that it commands qi again? One possible answer is that one simply has to will to use xin. Even if one’s qi is concentrated in a problematic way, one can on one’s own volition use xin. As long as one wills to use xin, one will seek goodness. Those who do not seek goodness are said to be robbing themselves of something valuable (2A:6), doing violence to themselves and abandoning themselves (4A:10) because one can easily seek goodness as long as one will to do so. This can explain Mencius’s comment that chess is not a difficult game to learn but if a student does not direct zhi – that is, does not pay attention to the activity – that student will never be able to play it well (6A:9). The crucial part of this picture is that one has to will to seek goodness. It is possible that the person herself cannot tell whether she is seeking goodness. She might sincerely believe that she is seeking goodness, but in hindsight, could realize that she was not in fact seeking goodness. And assuming that selfconsciously seeking goodness does not necessarily result in one’s seeking goodness, it is possible that when she self-consciously tries to seek goodness, she sincerely believes that she is seeking goodness when in fact she is not doing so. But in sincerely believing that she is seeking goodness, she no longer wills to seek goodness. This suggests that one’s self-consciously trying to seek goodness might end up preventing one from actually seeking goodness.12 The Manifestation-Will model offers a good explanation for why Mencius thinks that if one seeks propriety, one will obtain propriety (6A:6, 7A:3), why Mencius thinks that it is a mistake to say that one is not unable to be good (1A:7, 2A:6), and why being reminded that one has a good xin helps to jolt one to use xin (1A:7). It can also accommodate all the passages that discuss xin being

12

This interprets the metaphor of forcing the growth of sprout in 2A:2 as saying one should not selfconsciously seek goodness (see Shun 1996: 154–55 for discussion of this line).

Cultivating Goodness or Manifesting Goodness

215

abandoned, forgotten, and lost (e.g. 2A:2, 6A:8, 6A:11) and passages that urge one to exercise and keep one’s xin (e.g. 6A:8, 7A:12). We can take these verbs to be referring to the agent’s way of handling xin. The difficulty with this interpretation is that it does not provide satisfactory explanation of the passages that suggest that xin needs to be nourished (e.g. 7B:35) and corrected (e.g. 5A:20). Moreover, it is theoretically more burdensome to posit a person that is beyond the senses and xin. One might think that a person just is the combination of xin and the senses (e.g. Van Norden 2000). Some scholars would also dispute that zhi is a faculty or capacity independent of xin.13

3b. The Manifestation-Mixed Xin Model A different version of the Manifestation Model holds that xin is mixed. It is possible that Mencius uses the term in different senses. There is an aspect of xin that has moral predispositions (e.g. 1A:7, 2A:6) that are sufficient to enable one to be good (6A:6). There is another non-moral aspect of xin that has morally neutral capacities such as the capacity to understand, deliberate and feel. The moral and non-moral aspects together constitute the entirety of xin. When xin operates as a whole, it might not always orient towards propriety. One might deliberate poorly and form false beliefs. One’s deliberation might also be affected by one’s mood or physical conditions.14 On this reading, when Mencius discusses si, he is referring to xin’s capacity and mean something akin to deliberation. But si is more than just any kind of deliberation. If it is possible for xin to deliberate and make a bad moral judgement, then it can’t be the case that any act of deliberation leads to the obtainment of propriety. Since Mencius thinks that one can obtain goodness if one carries out si, si probably means the kind of deliberation that is guided by the moral predispositions of xin (6A:15).15

13

14

15

Camus 2019 points out that there is textual evidence that suggests zhi is determined by the individual and also textual evidence that suggests zhi is xin or the direction of xin. Neo-Confucian thinkers invoked the distinction between ‘moral xin’ and ‘human xin’. My discussion of the moral and non-moral aspects of xin here does not make reference to Neo-Confucian thought, though it will have bearing on our understanding of the development of Mencius’s thought in the Neo-Confucian period. Donald Munro also suggests that Mencius is using the term ‘xin’ to refer to ‘two essentially different sets of things’ (2001: 74–75). One set is about ‘potentiality or behavioral tendency’, the other is about ‘covert evaluative activity’ (74–77). Chan (2016: 1173), for example, argues that si refers more often to cognitive activity that involves ‘evaluative judging’.

216

Skill in Ancient Ethics

On this reading, the obstacle to the state of unmoved xin is that the nonmoral aspects are not necessarily guided by the moral aspects and might make wrong judgements. For example, one’s xin of shame and dislike would naturally refuse things that are offered with insult. However, when one has a strong desire to live in a beautiful mansion, one might mistakenly judge that living in such a place is more valuable than using the xin of shame and dislike (6A:10). The effort of attaining unmoved xin lies in ordering the non-moral aspects in such a way that they align with the moral aspect so that xin as a whole orients towards propriety. If this is right, then learning has to be crucial in the Mencian picture. If one is in a good environment in which guidance is provided by good teachers, xin learns to deliberate and judge in accordance with propriety. Given that the non-moral aspects of xin lack moral propensity, one might think that, even if they learn about propriety, they will not necessarily operate in accordance with propriety. The problem with this line of thought is that it assumes that the non-moral and moral aspects of xin are cleanly divided. It is important to bear in mind that, on this reading, we are considering xin in its entirety. We are not assuming that one side of xin is moral and the other immoral. That would be, in effect, another way of saying that there is a third faculty involved. On this reading, xin itself is holistic. It has moral aspects and non-moral aspects, and the two aspects are mixed together. Insofar as xin has moral aspects, it has the propensity to be in accordance with propriety. But that does not mean that xin cannot be derailed. To use an analogy, a tourist might intend to go to destination D, but that does not mean that she will get there. She might make some misjudgements along the way and head off in the opposite direction. The moral effort therefore lies in learning what propriety is, just like the tourist learning where D is. As soon as xin learns about propriety it aligns itself with propriety by virtue of its inherent propensity to do so. In this way, the moral and non-moral aspects are aligned, and xin orients towards propriety. Zhi is more plausibly understood as the direction of xin on this model. This is different from the Manifestation-Will Model, which understands zhi as the will of the person. Mencius’s discussion in 2A:2 can be interpreted as saying that in order to attain the unmoved xin, xin has to orient towards propriety. This can be achieved by making sure that qi moves xin in the right direction. For example, by maintaining a healthy daily regimen and spending more time on activities that promote the learning of propriety, one’s qi will also move xin in the direction of its proper goal. But if one does the opposite, one’s qi will obstruct xin and derail it from its proper goal. By reducing sensory desires, qi will less likely distort the judgement of xin (7B:35).

Cultivating Goodness or Manifesting Goodness

217

When the non-moral aspects align with the moral aspects, zhi is directed at propriety. The qi that is commanded by zhi that has propriety as its goal is a special kind of qi, namely, the flood-like qi. The flood-like qi is generated by xin that directs itself at propriety. If one’s xin is not directed at propriety, even if xin occasionally makes a good judgement and act in accordance with yi, it will not generate the flood-like qi.16 This is probably why Mencius thinks that once xin is established, that is, when it is directed at propriety, sensory desires can no longer pose a problem to xin (6A:15) because the qi that is interacting with zhi is floodlike qi rather than the qi generated by senses. This also explains why even if the superior person is in dire circumstances where his basic desires are not satisfied, he is still unfazed and will not do anything that is contrary to propriety (3B:2). In the picture outlined above, it is crucial for xin to learn about propriety; and a good environment facilitates learning. Again, what if one is in a hostile environment that provides no opportunities for moral education, is it possible for one’s xin in its entirety to orient towards yi? Mencius might think it still is possible for one to learn about propriety in a bad environment. One might witness injustices and come to reflect and know that it is wrong. Although it is possible, much will depend upon one’s natural allotment: how reflective one is, the strength of one’s sensory desires, and one’s intelligence and ability to teach oneself. Such reflection is much more difficult without the help of a good environment. This reading avoids the difficulty of explaining how xin or the person who has to use xin can seek goodness if xin itself is too weak or the person is distracted. On this reading, the propensity is always present to jolt xin in the right way as long as xin learns. It also makes sense of passages that suggest xin could be harmed (7A:27) and that some rulers’ xin need to be corrected (5A:20). It is possible for xin’s judgement to be distorted and derail from its proper trajectory. Hence, it is important to correct, and not mislead xin. One consideration against this reading is Mencius’s dialogue with King Xuan. If it is the case that xin’s judgement can be distorted, then King Xuan could genuinely have difficulty caring for his people even though he can care for an ox. Perhaps his desires for power and fame are so strong that they cloud his judgement. He will need to first correct his mistaken beliefs before he can act in accordance with propriety. If he does not correct his judgement, he will not be

16

See Shun 1997: 154 for discussion of ji yi and yi xi. The third interpretation discussed by Shun, which takes the distinction to be between ‘regularly and persistently acting in accordance with yi’ and ‘sporadically acting in accordance with yi’ is compatible with the Manifestation-Mixed Xin model.

218

Skill in Ancient Ethics

able to care for his people. We can get around this problem by reading the expression ‘bu neng’ н㜭 in a loose sense to mean impossibility. Then, the king would be saying that it is impossible for him to care for his people and Mencius could be telling the king that it is possible for him to care for his people. We can read the rest of the passage as Mencius trying to correct the king’s judgement and teach him ways of governance that are in accordance with propriety (1A:7). Another consideration against this reading is that Mencius has not put much stress on the importance of learning. If this reading is correct, then the effort to orient xin towards propriety relies heavily on one’s acquiring knowledge of propriety. But we have not seen much emphasis on learning in the text. Nonetheless, it is possible that the importance of learning was so widely assumed in Mencius’s cultural context and by his audience that he did not see the need to highlight it. It is also possible that some of the dialogues about learning were not recorded in the text. One key difference between the Manifestation-Will Model and the Manifestation-Mixed Xin Model lies in whether moral knowledge is acquired. The former assumes that moral knowledge is already within oneself. One only needs to seek goodness and one will get it. The latter emphasizes the acquisition of moral knowledge. Even though xin has a moral propensity, it still has to be taught what goodness is. Compared with the Manifestation-Mixed Xin model, the moral effort involved on the Manifestation-Will model is simpler in the sense that one only needs to will to be good. However, it offers no direct strategy to move the will to turn to xin. While indirect strategies might prompt the will, their effectiveness might vary for individuals. At the end of the day, it is up to the individual to develop a resolve to use xin. The Manifestation-Mixed Xin Model, by contrast, can spell out a programme of instruction that is likely to be effective in teaching people about propriety. With enough instruction, one will most likely learn about propriety and follow it.

4. Concluding remarks In this study, I have only managed to sketch two plausible lines of interpretations of Mencius’s view on moral development. The textual evidence is not decisive enough to allow us to adjudicate between these interpretations. Which interpretation we adopt, however, will have implications for our reading of certain key terms (e.g. duan ㄟ, zhi ᘇ, zhi ᙍ, ji yi 䳶㗙, yi xi 㗙㾢) and passages (e.g. 1A:7, 2A:2, 2A:6, 6A:8, 6A:15, 7B:35). Even if we cannot decide which

Cultivating Goodness or Manifesting Goodness

219

interpretation best represents Mencius’s view, a deeper understanding of the theoretical commitments that each interpretation carries will put us in a better position to analyse, develop, or evaluate Mencius’s thought. On the surface, these two interpretations have considerable overlap. Both agree that human beings have moral predispositions, that good environment is important, that sensory desires pose obstacles to the cultivation or manifestation of xin, that the attainment of unmoved xin in practice is a gradual process, and that one’s moral actions will be sporadic before the attainment of unmoved xin. However, if we delve deeper into the theoretical differences, we come to see the different assumptions about moral motivation, agency, and moral knowledge. On the Cultivation Model, in order to be moral, one needs to strengthen moral motivation. On the Manifestation Model, in order to be moral, one needs to put one’s moral predispositions to use, by exercising one’s agency in the right way or by gaining moral knowledge. I hope this study provides a rough theoretical framework for us to develop and critically engage with Mencius’s view. If one thinks, for example, that Mencius adopts the Manifestation-Will Model, then to critique Mencius, we can focus on the question whether xin is the sole source of agency. If one thinks Mencius adopts the Cultivation Model or Manifestation-Mixed Xin Model, then one can focus on the question whether xin has moral predispositions. This study also raises new questions, such as how susceptible xin is to external influences after it has been cultivated and whether manifestation of xin comes in degree, that can be further examined in future studies.

References Allan, S. (1997), The Way of Water and Sprouts of Virtue, Albany : State University of New York Press. Angle, S., and J. Tiwald (2020), ‘Moral Psychology: Heartmind (Xin), Nature (Xing), and Emotions (Qing)’, in K.-C. Ng and Y. Huang (eds), Dao Companion to Zhu Xi’s Philosophy, Cham, Switzerland: Springer Nature. doi.org/10.1007/978-3-030-29175-4 Camus, R.M. (2019), ‘Zhi ᘇ in Mencius: a Chinese notion of moral agency’, Asian Philosophy, 29 (1): 20–33. Chan, A.K.L. (2002), ‘A Matter of Taste: Qi (Vital Energy) and the Tending of the Heart (Xin) in Mencius 2A2’, in A.K.L. Chan (ed), Mencius: Contexts and Interpretations (pp. 42–71), Honolulu: University of Hawai’i Press. Chan, S.Y. (2016), ‘Evaluative Desire (Yu Ⅲ) in the Mencius’, Philosophy East and West 66 (4): 1168–1195.

220

Skill in Ancient Ethics

Cua, A.S. (2002), ‘Xin and Moral Failure: Notes on an Aspect of Mencius’ Moral Psychology’, in A.K.L. Chan (ed), Mencius: Contexts and Interpretations (pp. 126– 150), Honolulu: University of Hawai’i Press. Graham, A.C. (1989), Disputers of the Tao: Philosophical Argument in Ancient China, La Salle, IL : Open Court. Ivanhoe, P.J. (2000), Confucian Moral Self Cultivation, 2nd ed., Indianapolis: Hackett Publishing. Lau, D.C., trans. (1970), Mencius, London: Penguin Books. Legge, J., trans. ([1985] 1990), The Works of Mencius, New York: Dover Publications. Munro, D.J. ([1969] 2001), The Concept of Man in Early China, Michigan: Center for Chinese Studies, The University of Michigan. Nivison, D.S. (1996), The Ways of Confucianism: Investigations in Chinese Philosophy, ed. B.W. Van Norden, Chicago: Open Court. Shun, K.L. (1997), Mencius and Early Chinese Thought, Stanford: Stanford University Press. Van Norden, B.W. (2000), ‘Mengzi and Xunzi: Two Views of Human Agency’, in T.C. Kline III and P.J. Ivanhoe (eds), Virtue, Nature, and Moral Agency in the Xunzi, Indianapolis: Hackett Publishing. Van Norden, B.W., trans. (2008), Mengzi: With Selections from Traditional Commentaries, Indianapolis: Hackett Publishing. Waley, A. ([1938] 1964), The Analects of Confucius, London: George Allen & Unwin Ltd. Wang, F.Z. ⦻ཛѻ (1975), Du Sishu Daquan Shuo 䆰ഋᴨབྷ‫ޘ‬䃚, Beijing: Zhonghua Shuju ѝ㨟ᴨተ. Wang Y.M. ⦻䲭᰾ (1963), in W, T. Chan (trans.), Instructions for Practical Living, and Other Neo-Confucian Writings of Wang Yang-ming, New York, Columbia University Press. Wong, D.B. (2015), ‘Early Confucian Philosophy and the Development of Compassion’, Dao: A Journal of Comparative Philosophy 14: 157–194. Xu, F.G. ᗀᗙ㿰 (1969), Zhongguo renxinglun shi ѝ഻Ӫᙗ䄆ਢ‫〖ݸ‬ㇷ, Taipei: Taiwan shangwu yinshuguan. Yearley, L.H. (1990), Mencius and Aquinas: Theories of Virtue and Conceptions of Courage, Albany : State University of New York Press. Zhu, X. ᵡ⟩ (2002), Sishu Zhangju Jizhu ഋᴨㄐਕ䳶⌘ in Complete Works of Master Zhu ᵡᆀ‫ޘ‬ᴨ. Shanghai and Hefei: Shanghai Guji chubanshe к⎧ਔ㉽ࠪ⡸⽮ and Anhui Jiaoyu chubanshe ᆹᗭᮉ㛢ࠪ⡸⽮.

11

Ritual as a Skill: Ethical Cultivation and the Skill Model in the Xunzi Siufu Tang

Xunzi (㥰ᆀ, born around 310 bce ), the last great Confucian thinker of the preQin period, has famously claimed that human beings are born with a bad nature (xing ᙗ) and thus there is a necessity for the sage-kings (shengwang 㚆⦻), and ritual (li ⿞) and appropriateness (yi 㗙).1 For without the governance of the sage-kings and the transformation of ritual and appropriateness, people will necessarily fight among themselves and bring about chaos and disaster (Xunzi 23.115.2–8)2. Thus people must be transformed by ritual and appropriateness so that they become good, just like crooked wood must be steamed and straightened so that it becomes straight and blunt metal must be honed and ground so that it becomes sharp (Xunzi 23.113.9–10). However, given that all people start with a bad nature, and that the sage-kings have the same bad nature as everyone else (Xunzi 23.114.15–16 and 23.115.22–23), how is it possible that the sage-kings achieve this status in the first place and also develop ritual and appropriateness? Some see this as a fundamental flaw with Xunzi’s ethics and suggest that he has difficulty with explaining the possibility of ethical transformation (Lao 1984: 335; Yu 2005: 29). Xunzi is not unaware of the seemingly difficulty of his position. In his eponymous text, the Xunzi, he records two challenges against his position and expounds his explanation again in terms of skill analogies. The first challenge is that if people are born with a bad nature, then where do ritual and appropriateness come from? (Xunzi 23.114.8) For presumably ritual and appropriateness, which

1

2

The usual translation of yi 㗙 is ‘righteousness’, but I prefer ‘appropriateness’ because although yi does usually carry the meaning of being right, the basic reason of its being right is because of its being proper and appropriate to the circumstances. All citations of the Xunzi refer to the ICS Concordance Series Edition (Xunzi 1996). Citations will be in the form of chapter/page/line numbers.

221

222

Skill in Ancient Ethics

are good, cannot be derived from a bad nature, and there must be an alternative explanation of the source of ethical values.3 Xunzi replies that ritual and appropriateness are created by the sages’ artifice (wei ‫ )ڭ‬and are not originally born in human nature.4 He then gives the examples of a potter who creates utensils from clay and a carpenter who creates utensils from wood. Similarly a sage accumulates thoughts and deliberations, habituates himself to artifice and practices so that he can create ritual and appropriateness (Xunzi 23.114.8–11). Then there comes the second challenge: the basic materials of ritual and appropriateness certainly come from human nature, and that is why the sages are able to create them through artifice (Xunzi 23.115.20).5 In response Xunzi points out that when a potter produces earthenware from clay, or when a carpenter creates wares from wood, in both cases neither the product nor the materials belong to the human nature of the potter or the carpenter. Xunzi underlines that the same is true with a sage’s creation of ritual and appropriateness. In this latter case neither the end product (ritual and appropriateness) nor the materials (accumulation (ji ぽ) and artifice) belong to the human nature of the sage (Xunzi 23.115.20–22). Both are skill analogies, albeit not detailed. Now that we have seen several examples of Xunzi’s employment of skill analogies, it should be apparent that Xunzi takes seriously the analogy between the sages’ creation of ritual and appropriateness and the craftsmen’s skilful production of beneficial utensils. Also, since the sages began their life with a bad nature just like everyone else, the very process via which the sages’ master of the skilful creation of ritual and appropriateness is actually the same as their ethical transformation from petty men into sages.6 Thus for Xunzi, ethical cultivation can also be seen as a process of skill acquisition. While a number of scholars have already pointed out the relevance of skills to understanding Xunzi’s ethics (for example, Lai 2003; Stalnaker 2010; Hutton 2016: 77–78), none has explicitly affirmed or studied the skill analogy in detail as a model for a better appreciation

3

4

5

6

Xunzi is arguing against an influential idea of Chinese philosophy that ethical values are somehow based in human nature. Graham gives an excellent explication of the idea through his detailed study of the concept of xing in early China (Graham 1986). In the texts of the Xunzi the sage-kings (shenwang 㚆⦻) and the sages (shen 㚆 or shenren 㚆Ӫ) are used more or less interchangeably, even though the sage-kings are more frequently mentioned in the contexts of political governance. In this chapter, I do not draw a distinction between the two terms, and use the relevant terms in accordance with the original texts. A literal translation of the challenge is as follows: ritual and righteousness, accumulation and artifice are people’s nature; that is why the sage can produce them. The meaning of such a rendering is obscure by itself. However judged by what Xunzi explains in the immediate following sentences, the intended meaning of the challenge is that the materials and consequently the end product of a craft production are nonetheless derived from human nature. Xunzi does say explicitly that everyone starts his life as a petty man (Xunzi 4.15.14–15).

Ritual as a Skill

223

of the distinctive structure and features of ethical cultivation in Xunzi’s thought. It is to this task that this chapter shall proceed. In the following I first explore the relevance of skill models of virtues to Xunzi’s ethics. Then I explain Xunzi’s understanding of ethical cultivation on the basis of the skill analogy, using ritual learning as a focal point. As part of that explanation I attempt to respond to criticisms of Xunzi’s model of ethical cultivation, particularly regarding the question of how a petty person with an originally bad nature can be transformed ethically by ritual, and the question of that ritual’s origin.

1. The skill model of virtue It has been pointed out that in Ancient Greece there was an overwhelming consensus that virtues are skills, with Aristotle as a lone dissenter (Bloomfield 2000: 23–24). Stichter adds that, although Aristotle disagrees with the Socratic model of virtue as a skill, he nonetheless embraces an alternative understanding of virtue as a skill (Stichter 2007: 184, 189–190). In recent years, a number of scholars have revived skill models of ethical expertise so as to better develop virtue ethics and respond to criticisms directed at ethical virtues (notably Dreyfus and Dreyfus 1986, 1991, 2004; Annas 1995, 2011a, 2011b; but also Bloomfield 2000; Stichter 2007; Hasselberger 2017). Bloomfield points out that a skill model of virtues can help to illustrate moral knowledge because it is simply a species of a general kind of knowledge, that is, practical knowledge of skill (Bloomfield 2000: 23). Annas also suggests that the developmental nature of learning to be virtuous can shed light on the question of action guidance of virtues. We acquire knowledge of what a virtuous action is, or who a virtuous agent is, precisely through such a developmental and learning process. As a child we are told by parents and teachers to follow virtue rules like ‘be honest’. In the beginning we understand honesty only in particular circumstances, such as not taking toys from shops without paying for them. Yet with further learning and experiences we are exposed to multiple facets and the complicated nature of honesty, which both challenge our existing understanding and grant opportunities for a clearer and deeper understanding (Annas 2011: 36–40). Such a developmental nature of virtues can also respond to the charge of parochialism and conservatism (Annas 2011: 52–63). Although we necessarily learn to be virtuous from teachers and within certain traditions and communities, through development we gain a deeper understanding of the rationale, relevant considerations and criteria of virtues. We thus gain the capacity to be critically

224

Skill in Ancient Ethics

reflective of existing practices, and also of the possibility of renewing or transcending such practices. Just as true mastery of a skill means not blindly following the existing rules, but going beyond them as required by circumstances, often in innovative ways, a virtuous agent is not bound by existing social norms and traditions, but can transcend or reconfigure them when needed. A more recent objection to virtue ethics comes from the claims of ‘situationist’ social psychology: namely, that empirical studies show that there are no robust character traits across situations among adult population, and consequently, no possibility of virtues as suggested by virtue ethics (Doris 1998, 2002; Harman 1999). One line of response to such situationist challenges is that these critics simply take for granted that character traits are dispositions to behave in certain ways. Such critics, however, neglect to note that there can be stable cognitiveaffective capacities which make it possible for people to appreciate factors, conditions and reasons involved in different circumstances, and to act in accord with them. Thus, even though there might not be consistent and reliable actions across situations, there can nonetheless be reliable social intelligence or practical intelligence (Russel 2009; Snow 2010). Although such accounts of practical intelligence see practical wisdom as central to ethical expertise, they fall short of affirming a skill model of virtue. As Annas points out, taking the skill analogy seriously helps to investigate and evaluate practical reasoning, particularly in relation to the various pitfalls we might encounter. It also clarifies the difference between learners and experts in terms of development. This is a promising, but so far understudied direction of both empirical research and philosophical reflection (Annas 2011: 174). There are also internal debates among proponents of skill models of virtue. One prominent area of tension is the role and extent of intellectual articulation in practical expertise, including ethical virtues. Stichter casts doubt on Annas’ intellectual view of skills, and suggests that an empiricist understanding more accurately describes practical skills and their acquisition. While acknowledging the necessity of giving articulation of and justification for one’s actions in moral discourse, Stichter nonetheless thinks that this is a separate demand from the achievement of ethical expertise (Stichter 2007). Kurth also thinks that Annas’ account of virtuous agency is over-intellectualized in its picture of the acquisition of virtues but under-intellectualized in its explanation of the role of deliberation after virtues are acquired (Kurth 2018: 304–307). Hasselberger expresses dissatisfaction with both Annas’s intellectualist account and the antiintellectualist account of Dreyfus and Dreyfus, because both have difficulty answering the puzzle of non-deliberative expertise. This puzzle is the combination

Ritual as a Skill

225

of two independently compelling but seemingly incompatible ideas. The first is that action performed without deliberation can exemplify practical excellence, in part by virtue of its non-deliberative character. The second is that skilful and virtuous action is a distinctively human form of agency, and is reflective of an agent’s reasons for action and her practical intelligence. Hasselberger (2017) offers an alternative account of practical perception inspired by Wittgenstein. The debate between intellectualism and anti-intellectualism in virtuous agency is far from conclusive. It can be seen as an extension of the corresponding debate over the nature of knowledge, mind, and action, particularly in relation to knowledge how.7

2. Xunzi’s ethics and the skill analogy If we use ‘virtue’ in a broad sense to refer to whatever tendencies or dispositions that are good, desirable, and praiseworthy, then it is clear that virtues and virtuous persons play a prominent role in Xunzi’s ethics (Hutton 2016: 70n11, cf. Hutton 2015: 114). Unlike Hutton, who defends Xunzi as a virtue ethicist, the present chapter leaves this question open. It defends the position that the skill model of virtue sheds important light on Xunzi’s ethics. This purpose is served by the more modest claim that the central dispositions of the noble person (junzi ੋᆀ) and the sages, including humaneness (ren ӱ), appropriateness, ritual or ritual propriety,8 and wisdom (zhi Ც), are virtues. Given such an understanding as the background, in the following I sketch how the various aspects of virtues discussed above are also relevant to Xunzi’s ethics. Xunzi stresses that any discourse must be validated and verified by experience; its reasonableness must be judged not only in words, but also in practice. So when a person propounds a theory when seated, he must also be able to stand up, establish it in reality, and unfold it in practice (Xunzi 23.115.10–12). Although Xunzi makes this claim primarily in support of the thesis that human nature is bad, which he believes is borne out by empirical evidence, it is also clear that he takes such requirements seriously and think them applicable to any reasonable discourse. Thus Xunzi also says that learning must culminate in practice, for it is

7

8

A helpful account of the controversy between intellectualism and anti-intellectualism over knowledge how is Bengson and Moffett 2011. But also see Pavese 2016a and 2016b. For Xunzi, the term li ⿞ refers not only to rituals as social and ethical norms, but also to one’s faithfulness to rituals which is best expressed in one’s skilful mastery of the rituals. See also Hutton 2015: 121.

226

Skill in Ancient Ethics

through practice that one gains full understanding; and the reason a sage is a sage is precisely because he has successfully put into practice humaneness and appropriateness, and all the standards of right and wrong (Xunzi 8.33.11–13). Seen in this light, it is reasonable to suggest that when Xunzi employs a skill analogy to describe the sages’ creation of ritual and appropriateness, he must believe that the latter process has no mystery and could be both understood and appropriated just like learning a skill. Indeed Xunzi repeatedly points out that an ordinary person can also become a sage simply through accumulation of goodness, just as a person becomes a carpenter through accumulation of experience at chopping and carving wood (Xunzi 8.34.5–7; cf. 1.2.9 and 23.116.13–15). Xunzi also points out that an emphasis on accumulation goes hand in hand with valuing teachers and models (shifa ᑛ⌅) (Xunzi 8.33.18–20). For Xunzi, a person who does not follow teachers and models but instead uses his own judgement is as foolish as someone relying on a visually impaired person to distinguish colours, or relying on someone who is hard of hearing to distinguish sounds (Xunzi 2.8.3–4). The sage-kings embody the ultimate standards, so that learners must follow them as teachers and take their regulations as models (Xunzi 21.106.21–23). Does such an emphasis on teachers and models, and particularly taking the sage-kings as teachers and models, reveal that Xunzi actually embraces ethical and political authoritarianism, as some have suggested? (Lao 1984: 339–343) However, Xunzi also underlines the importance of understanding and being reflective of the rational standards of actions. Indeed he stresses that one should follow the Way (Dao 䚃) and not one’s lord, and one should follow appropriateness and not one’s father. And such doings actually represent the highest standards of filial piety (xiao ᆍ) in particular and human conduct in general (Xunzi 29.141.19–142.5). Such seemingly disparate attitudes towards authoritative figures can be happily accommodated by the developmental nature of virtue learning. Like acquiring a new skill from scratch, in the initial learning stage one must closely follow the instructions of teachers. With further progress in learning the skill and a gradual grasp of it, one can attempt to practise and improve by oneself, seeking advice from teachers only when encountering insurmountable difficulty. After one has gained a certain level of mastery of the skill, one also acquires the ability to judge independently whether an action or a certain move is indeed right or skilful, and similarly whether a judgement by an otherwise well-respected teacher is indeed correct or uncontestable. Some true masters of the skill might alter or even revolutionize the practices of the skill and be able to point out the inadequacies of established practices. Thus when Xunzi

Ritual as a Skill

227

suggests that people should follow teachers and models, he is likely referring to the early learning stage of virtue acquisition. And this is partly validated by the contexts of the three passages that I refer to above, where Xunzi is concerned with explaining the necessity of ritual for transformation, or for preventing a person from doing bad. Yet when Xunzi talks about the cases of disagreeing with the orders of one’s lord or father, he is clearly referring to persons who have successfully acquired the Confucian virtues and that is also why he stresses that such is great conduct (daxing བྷ㹼), as compared to the small (xiaoxing ሿ㹼) or medium conduct (zhongxing ѝ㹼). Actually, the significance of Xunzi’s classification of human conduct into three categories can be better appreciated from the perspective of skill learning. From that perspective, differences among these categories need not be merely synchronic disparity between different persons; they can also be diachronic changes of the same person on a path of virtue cultivation. Indeed Xunzi himself hints at this developmental aspect of virtue learning by noting that when one’s intentions are at ease with ritual and one’s words in accordance with proper classes (lei 于) of things, the Confucian path of cultivation is finally completed, and even a sage-king like Shun 㡌 cannot improve on it at all (Xunzi 29.141.20). Besides being at ease with ritual, another characteristic that Xunzi frequently associates with virtuous agency is being responsive and adaptive. For example, Xunzi once described the noble person as someone without a fixed course of action. Sometimes, he praises others’ excellent qualities, but sometimes makes a direct accusation of others’ faults. Sometimes he asserts his own merits, but sometimes yields like a reed. Xunzi concludes that this is because the noble person bends or straightens, and is flexible and adaptive as appropriate (Xunzi 3.9.19-10.4). It is important to note that virtuous agency is not simply being flexible and adaptive, but rather, being flexible and adaptive as required by circumstances, and more importantly, in accordance with the virtue of appropriateness. So although a Xunzian noble person might not exhibit regular behaviour across situations, or even within similar situation types, it is not the case that his behaviour is dictated by the situation alone. On the contrary, the noble person’s responsiveness to situational factors is actually guided by practical reasoning, expressed in the virtue of appropriateness. Indeed Xunzi suggests that if a person is without models, he will be at a loss with what to do. Yet if he adheres to the models without setting his intentions on the appropriateness embodied in them, he will act too rigidly. It is only when a person abides by the models, but also has a deep understanding of the proper classes of things upon which appropriateness is based, that he can act comfortably (Xunzi 2.7.18–19). In a

228

Skill in Ancient Ethics

similar vein, his comments on the great conduct of a virtuous person also highlight the importance of appropriateness and the proper classes of things. It is clear that for Xunzi, an essential aspect of virtue cultivation is a firm grasp of appropriateness as responsiveness to proper salient features of different circumstances. In the same way, an essential aspect of true mastery of a skill is the ability to act skilfully in different situations. As Mower has forcibly shown in her paper, Xunzi’s ethics can not only respond to situationist critiques, but can actually provide a better account of moral development and ethical norms that upholds robust character traits of practical reasoning while attending to situational features (Mower 2011). Even critics of Xunzi do not deny that the highest stage of the Xunzian ethical cultivation, sagehood, is manifested by effortless, spontaneous yet virtuous responses to circumstances (Slingerland 2003: 246–252; Fraser 2006: 532; Singh 2014: 213–214). The challenge, however, is whether such virtuoso responses can indeed be attained through the regime of rational and deliberative cultivation conceived and advocated by Xunzi. For example, Slingerland points out that, given Xunzi’s insistence that human nature is innately bad, Xunzian ethical cultivation according to external standards is confronted with a Xunzian version of ‘Meno’s paradox’. Without some internal resources as guide, anyone starting out in ethical cultivation would have difficulty in discerning proper ethical standards from improper ones. This problem is particularly acute because, compared to quantitative standards such as a balance-scale, ethical standards are intangible and hard to assess. On the other hand, Slingerland also thinks that the skill analogy breaks down when it comes to the question of internal motivation for virtuous acts. The problem, Slingerland suggests, is that internal motivation is not necessary for the performance of a technical skill; skilful performance stands on its own merits regardless of the internal state of the performer. By contrast, internal motivation is constitutive of virtuous acts, for without the corresponding internal motivation, actions resemble virtuous acts only in appearance but are not virtuous. However, given Xunzi’s presumption that human nature is inherently bad, it is a mystery how learners can begin with and grow in genuine ethical cultivation (Slingerland 2003: 252–264). Singh similarly criticizes Xunzi’s account of ethical cultivation for its discontinuity with human nature. Unlike Slingerland, who bases his criticisms of Xunzi on the dissimilarity between virtue cultivation and skill acquisition, Singh challenges Xunzi’s ethical cultivation over its failure to truly resemble skill learning. Drawing on Dreyfus’ theory of skill acquisition, Singh suggests that, despite Xunzi’s views to the contrary, formulas and conceptual thinking do not

Ritual as a Skill

229

help fully perfected individuals develop in their self-cultivation. Taking John Elster and Mihaly Csikszentmihalyi’s theories of action as evidence, particularly the flow experience suggested by Csikszentmihalyi, Singh points out that – again contrary to Xunzi’s suggestion – conscious deliberation is also unable to help develop fully perfected individuals (Signh 2014). Two issues stand out in both of Slingerland and Singh’s criticism of Xunzi. First, to what extent is Xunzi’s account of ethical cultivation similar or dissimilar to skill learning? Second, what role is played by conceptual thinking and conscious deliberation in ethical cultivation, particularly in reference to agents’ natural motivational set? It is clear that both issues bear a close affinity to relevant debates over intellectualism and antiintellectualism between different skill models of virtues that are briefly discussed in the last section. In order to address these two issues, we need to have a comprehensive picture as well as more details of Xunzi’s account of virtue as a skill. It is to this task that we now turn.

3. Ritual learning as skill acquisition For the time being, we shall leave to one side the question of how the sages, who initially have the same bad nature as others, create ethical standards and virtues like ritual and appropriateness from the very beginning. Instead, we shall first see how Xunzi describes the process through which an initially petty person learns to be good and gradually becomes a noble person, given that ethical standards have already been established. Xunzi describes what people are like before they are transformed by ethical cultivation: People are originally petty people by birth. If without teachers and models they will only attend to benefits. People are originally petty people by birth. If they further encounter a chaotic age and acquire chaotic customs, then this is to double pettiness with pettiness and to get chaos from chaos. Xunzi 4.15.14–159

Judging from what Xunzi says, it is reasonable to infer that the circumstances one is born into and the customs one is immersed in, as well as the availability of teachers and models, are important factors for determining whether someone

9

All translations from classical Chinese are my own, though I have consulted translations by Knoblock 1988–94 and Hutton 2014.

230

Skill in Ancient Ethics

can become good. In, general, of course, we can do little to select the age we are born into, even if we might try to bring about changes in our own lifetime. However a person does have some choice of environment, such as neighbourhood and congenial people and friends. Indeed Xunzi suggests that the noble person is cautious in selecting the place he lives or his companions in travel, because the gradual effects of such associations on a person become transformative over time (Xunzi 1.2.1). Elsewhere, Xunzi points out that, just as great bows and superior swords must be carefully crafted and outstanding horses must be properly trained, if people associate with worthy teachers and valuable friends, by imbuing what they see and hear, they will improve in the virtues of humaneness and appropriateness without even noticing it. But if people associate with bad people, they will similarly be severely corrupted by imbuing cheating words and bad behaviour (Xunzi 23.117.12–19). The type of cultivation that Xunzi describes above, with his emphasis on associating with others and imbuing their qualities, along with the effects of what one sees and hears, resembles immersion training and experiential learning. This is not the place to evaluate how and why immersion and experiential learning are effective as learning methods, though it is generally agreed that they do help people improve skill acquisition, for example as demonstrated in a recent study of experiential learning and outdoor education (Parry and Allison 2020). The question for Xunzi, however, as Slingerland points out by citing the same passage, is how an originally petty person can recognize worthy teachers and valuable friends in the first place (Slingerland 2003: 262–263). Yet Xunzi does seem to think that even a petty person is able to distinguish between an ethical life and an unethical one. Xunzi draws an analogy between food and ethics. He points out that if a person has never seen fine food like the meat of pasture-raised and grain-fed animals and fine rice, then he might think that his own daily coarse food of greens, dregs and husks are already the best. Yet once he has a chance to try fine food, he will certainly prefer it over his typical humble fare. Xunzi suggests that the same is true of the ethical way of life of the sage-kings. Some people dwell in an unethical way of life simply because they are uninformed of better alternatives (Xunzi 4.15.17–22). Slingerland mentions the same passage of the Xunzi and comments that Xunzi’s employment of an internalist moral taste metaphor is a revelation of the inadequacy of his otherwise externalist stance (Slingerland 2003: 258–259). However, even if it might be natural to read Xunzi as assuming that people have a natural moral sense, it is not the only alternative. A more sympathetic reading suggests that the relevant distinguishing ability can be non-ethical. That is, when a petty person judges that

Ritual as a Skill

231

an ethical life is superior to an unethical life, it need not be an ethical evaluation, but can also be a practical one. It is noteworthy that Xunzi describes the first encounter of a person with fine food as follows: ‘He smells it and it is pleasing to his nose, he tastes it and it is sweet in his mouth, he eats it and it is comforting to his body’ (Xunzi 4.15.19). It is clear that Xunzi actually describes not a singular bodily sense, but several senses which constitute the wholesale experience. It is reasonable to infer that when Xunzi thinks that even a petty person can easily judge an ethical life superior, he need not have a natural moral sense in mind but is rather referring to the wholesale practical experience. In a previous passage Xunzi does compare the way of life of the petty person and the noble person. Xunzi points out that the petty person and the noble person have the same nature and capabilities; they both like honour and dislike disgrace, and they both like benefit and dislike harm. It is just that they pursue what they like by different means. The noble person practises the ethical way of life, treating others well and is in general treated well by others. His way of life is relatively easy in terms of deliberation, practice and maintenance, in the sense of not requiring complicated calculations, manipulative manoeuvres and the covering-up of tactics. The petty person does the reverse. He treats others badly but hopes that others will treat him well. He tries various ways to gain benefit at the expense of others. As he lives his life, he has a hard time deliberating what to do, he find it difficult to be at ease, and even harder to establish and maintain the form of his life. When the petty person fails in his endeavours, he looks up to the noble person, thinking that the noble person was born with the outstanding abilities necessary to become a worthy person (Xunzi 4.14.17–23). Now it is possible to criticize Xunzi for being too optimistic about the impacts and effects of the ethical way of life. But his portrayal of the general experience of the noble person, that an ethical way of life is generally rewarded with kind reciprocity, is not unrealistic in a well-ordered society. Whether a well-ordered society is possible for people with an originally bad nature is precisely the question under debate, and is particularly related to how an ethical order is established in the first place. For our purposes, the important point is that even a petty person can appreciate the goodness of an ethical way of life, and he does this not from an ethical point of view but from a practical one. Alternatively, it can be said that an ethical way of life can be perceived and appreciated from at least two different perspectives, one ethical and the other practical. Of course a practical perspective cannot perceive and appreciate all the elements of an ethical way of life, for presumably thick ethical properties like various virtues are not perceived, but such a perspective can perceive and appreciate enough

232

Skill in Ancient Ethics

elements of an ethical way of life to determine that it is practically better than an unethical way of life.10 An analogy to skill might be helpful. Take carpentry as an example. As an outsider who knows little about the art, I might not be able to appreciate how a skilful carpenter works, and it might not be easy for me to distinguish between an apprentice and a master. Nonetheless, there are some practical and external clues available that can help me to make a rough judgement. For example, I might observe that the master feels at ease with his work, that he cuts or carves smoothly and without hesitation, or how quickly and efficiently he completes his tasks. And of course I can also judge the carpenter by the items he creates, and whether his work is useful or durable. While ethical acts usually do not bring about an independent product, it seems the actions themselves and the overall behaviour can nonetheless be judged practically. I might observe whether the ethical acts lead to positive consequences, or whether such acts can be performed consistently or in a straightforward manner. And there is also the possibility that when I ask for advice from potential teachers and I am given answers that I find instructive and useful, then I can regard them as teachers indeed. Admittedly ethical acts are often much more intangible and complicated to judge practically, but the differences between ethical acts and other skills or kinds of behaviour do not seem to be a matter of kind but a matter of degree. A blatant denial of a practical judgement of ethical acts simply begs the question because the controversy lies in precisely whether ethical acts can be perceived and understood, however superficially, from a point of view that is external to ethics. Given that it is at least plausible that a petty person can practically perceive and recognize that the ethical way of life is a better option,11 it opens up the possibility of ethical transformation by teachers and models. Xunzi describes how ritual and teachers help us to rectify ourselves: Ritual is that by which to rectify your person. The teacher is whom by which to rectify the ritual. Without ritual, how is your person rectified? Without the teacher, how do you know if ritual is correct? Ritual is such and you are also such, then your dispositions are at ease with ritual. The teacher says so and you also say so, then your understanding is like that of the teacher. Having disposition 10

11

Xunzi seems to think that the petty person is able to perceive thin ethical properties, such as the property of being a good person. Whether this creates problems for Xunzi’s account and whether Xunzi’s ethics has resources to address these problems are complicated questions that cannot be pursued here. In reality, a petty person might fail to perceive and recognize this fact because of various factors, for example, that he is not sufficiently exposed to the ethical way of life, that he does not attend to its differences from his existing life, or that he simply does not care.

Ritual as a Skill

233

at ease with ritual and understanding like that of the teacher is to be a sage. Thus, to oppose ritual is to be without the model, and to oppose the teacher is to be without a teacher. Xunzi 2.8.1–3

According to Xunzi, ritual is the model by which we transform and rectify ourselves, supposedly by laying out how we should behave and act. The aim of cultivation is not simply to behave as ritual prescribes, but also to identify with the ritual requirements and feel at ease with them. On the other hand, understanding how ritual functions is also important. When Xunzi suggests that the teacher is the person by whom ritual is rectified, he underlines the need for interpretation and comprehension of ritual. A responsible teacher needs to explain to students, at the right moments, how and why ritual applies in different circumstances. It is by fully grasping the appropriateness of ritual and its underlying rationale that one becomes a sage. Indeed one of the distinctive features of being a sage is understanding the commanding ideas (tong ㎡) and proper classes (lei 于) of virtues, in other words the unifying thread (guan 䋛) of the Way, so that he knows how to act virtuously even in unfamiliar circumstances or when confronted with novel changes (Xunzi 8.33.2–5, 9.37.19–20, 17.82.20–21, 23.117.2). Xunzi believes that through accumulation of ritual and appropriateness one becomes the noble person, just like through accumulation of cutting and carving one becomes a carpenter (Xunzi 8.34.6–7). We have seen that such accumulation involves not only unconscious or subconscious association with others and imbuing their qualities, but also more conscious learning from teachers and models, including understanding the underlying rationale of proper behaviour. Xunzi also summarizes these aspects and processes as accommodation and settlement (zhucuo ⌘䥟) and habits and customs (xisu 㘂؇) (Xunzi 4.15.1, 8.34.8–9). But how does a petty person, who only cares for benefit and personal interests, come to care and even delight in ritual, and gradually transform into a noble person? The general consensus is that initially the petty person is attracted to ritual by its prudential and practical benefits, but with further exposure to ritual he comes to discover new values and new sources of satisfaction in it. When he comes to delight in these ethical values, he becomes a noble person (Ivanhoe 2000: 238–239; Kline 2000: 164–165; Stalnaker 2010: 415–416). MacIntyre provides a similar understanding of the goods internal to practices that are only achievable by virtues by giving an example of playing chess, where a child is first lured to chess by external rewards but comes to enjoy goods internal to chess playing (MacIntyre 2007: 188).

234

Skill in Ancient Ethics

However, as we have seen above, Slingerland thinks that the analogy between skills and virtues breaks down when it comes to the question of internal motivation. And he actually directs his criticism both at MacIntyre’s chess analogy and Xunzi’s skill metaphor (Slingerland 2003: 260). But it seems Slingerland conflates two aspects of skilful agency and virtuous agency. First, there is the aspect of internal mental state when attending to a skill or a virtuous act. Second, there is the aspect of motivation for engaging in a skill or a virtuous act. Slingerland is right to point out that internal motivation is not necessary for a skilful performance, particularly regarding a technical skill. The fact that a master carpenter produces wooden wares merely to earn a living need not diminish his skilful performance. Slingerland is also right to point out that internal motivation is necessary for virtuous agency and that the internal mental state of performing a virtuous act is partly constituted by internal motivation for virtues. Unlike a master carpenter, a person who instrumentally practises seemingly virtuous actions for external aims simply fails to act virtuously. However, even if a master carpenter uses his technical skill instrumentally, he must nonetheless be in the right internal mental state for carpentry, for otherwise it is impossible for him to produce wooden goods. Although a petty person initially does not have internal motivation for virtues and correspondingly cannot be in the right internal mental states for virtues, it does not thereby mean that a petty person cannot be introduced into the internal mental state of virtuous acts. Just like a student of carpentry can learn to be in the right internal mental state by practising his skill, a person can also learn to be in the right internal mental state by engaging in actions for virtues. When the person learns more about the internal mental state of virtuous agency, he might find new values, new satisfaction, and new perceptive cum affective stances and be gradually transformed. Although skilful agency and virtuous agency might differ in some important aspects in their mature forms, in terms of learning the analogy between them might nonetheless hold.

4. Ritual creation and skills Given the above illustration of ritual learning in the Xunzi, we actually have sufficient resources to answer the question of ritual creation by the sages. Granted that a petty person can be brought into the path of ethical cultivation by practical reasons, the future sage who is initially a petty person just like others can similarly embark on the path of ritual creation on the basis of practical reasons.

Ritual as a Skill

235

Indeed Xunzi suggests that the origin of ritual lies in the practical challenge of satisfying human desires in an orderly fashion. However, through participation in rituals people find meanings and values beyond satisfaction of desires (Xunzi 19.90.3–11). Xunzi calls these meanings and values differentiation (bei ࡕ) or distinction (bian 䗘). Xunzi thinks that these meanings and values are constitutive of our humanity and are what distinguish humans from beasts. As examples he gives the affection between parents and children that expresses but also transcends the biological tie, and the respective treatment of husband and wife that expresses, but also transcends, sexual and gender relationship (Xunzi 5.18.13–17). As Berkson argues, these meanings and values are symbolic significances that are ‘fabricated’ through ritual but nonetheless represent nonmaterial reality of human life (Berkson 2016). Skills are created by humans, presumably not by a single hand but modified and improved across the generations. There is no reason to think that virtues and ethical standards, such as ritual and appropriateness, cannot be similarly created by humans, again presumably not by a single sage, but by accumulated wisdom over time (Nivison 1996: 238–239; Kline 2000: 163–165). Xunzi suggests it is a process of accumulation of thoughts (si ᙍ), deliberation (lű ឞ), and the habituation of artifice and conventions (gu ᭵). And it is also a process through which our original nature is transformed (Xunzi 23.114.10–15). Xunzi also points out that deliberation is concerned with making choices regarding our natural feelings and desires, and artifice is the expression of such choices in action (Xunzi 22.107.23–24). In other words, although ritual creation is initially motivated by the practical need of living together in an orderly way, it is at the same time a transformation and reformation of our nature. This is also a task that Xunzi thinks is necessary of understanding, which is to guide and direct our feelings and desires (Xunzi 22.111.15). Thus I think Singh is wrong to think that Xunzi’s ethical cultivation is discontinuous with and external to human nature (Singh 2014: 213–216). For Xunzi, human nature is not incorrigibly bad. It is bad because of the inherent inclination of natural feelings and desires to realize themselves in chaotic ways.12 Human nature needs to be combined with artifice to give rise to a flourishing life, but artifice will have no role to play if without human nature (Xunzi 19.95.1–2). Now we are in a position to briefly provide Xunzi’s answer to the puzzle of non-deliberation expertise. Similar to the Wittgensteinian answer provided by 12

For a more detailed explication of the relation between human nature and artifice, and why human nature is bad for Xunzi, see Tang 2016 and Tang 2020.

236

Skill in Ancient Ethics

Hasselberger, Xunzi also suggests that understanding is used in the first place to structure our feelings, desires and perceptions so as to give rise to perceptual knowledge embodied in skilful practices. One prominent example is the funeral ritual. Xunzi points out that the arrangements of the funeral ritual are the cultural forms (wen ᮷) we use to structure our understanding of life and death after weighing our feelings and dispositions. For example, when a person dies, his body is bathed, his hair washed and bound up, his nails trimmed and food put into his mouth. All these arrangements are used to represent the person as if still alive, but also to prepare for his departure (Xunzi 19.95.6–96.8). These funeral arrangements are precisely meant to help participants perceive salient features of the funeral. As for the sage who is fully virtuous, he simply follows his desires and embraces all his dispositions, and yet all are well regulated with proper patterns (li ⨶) (Xunzi 21.105.18–19). This is because the desires and dispositions of the sage are fully transformed and structured by virtues, and his actions are guided by perfected perception of the circumstances. The sage acts spontaneously, and although he usually does not (and need not) provide discursive and articulable reasons, his acts are nonetheless guided by reasonresponsive understanding. It seems reasonable to suggest that the Confucian sage possesses a kind of practical intelligence that is similar to the one articulated by Russell and Snow. We do not have the space to look into all the details of the Xunzi’s skill model of virtues. I hope the chapter has conclusively demonstrated that a skill model of virtues is an apt description of Xunzi’s ethical thought, and that such an understanding sheds new light on otherwise familiar aspects of Xunzi’s understanding of ethical cultivation. It can also be seen that Xunzi’s ethical thought shares a lot of common concerns with the contemporary skill models of virtues. Hopefully further studies of Xunzi’s skill model of virtues will make important contributions to the current debates.13

References Annas, J. (1995), ‘Virtue as a Skill’, International Journal of Philosophical Studies 3 (2): 227–243. Annas, J. (2011a), Intelligent Virtue, New York: Oxford University Press.

13

This paper greatly benefits from Lisa Raphals’ comments and suggestions. Any remaining errors are my own responsibility.

Ritual as a Skill

237

Annas, J. (2011b), ‘Practical Expertise’, in J. Bengson and M.A. Moffett (eds), Knowing How: Essays on Knowledge, Mind, and Action, New York: Oxford University Press. 101–112. Bengson, J., and M.A. Moffett (2011), ‘Two Conceptions of Mind and Action: Knowing How and the Philosophical Theory of Intelligence’, in J. Bengson and M.A. Moffett (eds), Knowing How: Essays on Knowledge, Mind, and Action, Oxford: Oxford University Press. 3–55. Berkson, M. (2016), ‘Xunzi as a Theorist and Defender of Ritual’, in Eric L. Hutton (ed), The Dao Companion to the Philosophy of Xunzi, 229–267, Dordrecht: Springer. Bloomfield, P. (2000), ‘Virtue Epistemology and the Epistemology of Virtue’, Philosophy and Phenomenological Research 60 (1): 23–43. Doris, J.M. (1998), ‘Persons, Situations and Virtue Ethics’, Noûs 32 (4): 504–530. Doris, J.M. (2002), Lack of Character, Cambridge: Cambridge University Press. Dreyfus, H.L., and S.E. Dreyfus (1986), Mind over Machine: The Power of Human Intuition and Expertise in the Era of the Computer, Oxford: Basil Blackwell. Dreyfus, H.L., and S.E. Dreyfus (1991), ‘Towards a Phenomenology of Ethical Expertise’, Human Studies 14 (4): 229–250. Dreyfus, H.L., and S.E. Dreyfus (2004), ‘The Ethical Implications of the Five-Stage Skill-Acquisition Model’, Bulletin of Science, Technology and Societ, 24 (3): 251–264. Fraser, C. (2006), ‘Zhuangzi, Xunzi, and the Paradoxical Nature of Education’, The Journal of Chinese Philosophy, 33 (4): 529–542. Graham, A.C. (1986), ‘The Background of the Mencian Theory of Human Nature’, in Studies in Chinese Philosophy and Philosophical Literature, 7–66, Singapore: Institute of East Asian Philosophies. Harman, G. (1999), ‘Moral Philosophy Meets Social Psychology: Virtue Ethics and the Fundamental Attribution Error’, Proceedings of the Aristotelian Society, New Series, 99: 315–331. Hasselberger, William (2017), ‘Knowing More than We Can Tell: Virtue, Perception, and Practical Skill’, Social Theory and Practice, 43 (4): 775–803. Hutton, Eric L. trans. (2014), Xunzi: The Complete Text, Princeton: Princeton University Press. Hutton, E.L. (2015), ‘Xunzi and Virtue Ethics’, in Lorraine Besser-Jones and Michale Slote (eds), The Routledge Companion to Virtue Ethics, 113–125, New York: Routledge. Hutton, E.L. (2016), ‘Ethics in the Xunzi’, in Eric L. Hutton (ed), The Dao Companion to the Philosophy of Xunzi, 67–93, Dordrecht: Springer. Ivanhoe, P.J. (2000), ‘Human Nature and Moral Understanding in the Xunzi’, in T.C. Kline III and P.J. Ivanhoe (eds), Virtue, Nature, and Moral Agency in the Xunzi, 237–249, Indianapolis: Hackett Publishing. Kline, T.C. III (2000), ‘Moral Agency and Motivation in the Xunzi’, in T.C. Kline III and P.J. Ivanhoe (eds), Virtue, Nature, and Moral Agency in the Xunzi, 155–175, Indianapolis: Hackett Publishing.

238

Skill in Ancient Ethics

Knoblock, J. (1988–94), Xunzi: A Translation and Study of the Complete Works, 3 vols. Stanford: Stanford University Press. Kurth, C. (2018), ‘Emotion, Deliberation, and the Skill Model of Virtuous Agency’, Mind and Language 33 (3): 299–317. Lai, K. (2003), ‘Confucian Morla Cultivation: Some Parallels with Musical Training’, in Kim-Chong Chong, Sor-hoon Tan and C.L. Tan (eds), The Moral Circle and the Self: Chinese and Western Approaches, 107–139, Chicago and La Salle, IL : Open Court. Lao, S. ऎᙍ‫( ݹ‬1984), The History of Chinese Philosophy New Edition ᯠ㐘ѝ഻ଢᆨਢ, Vol. 1, Taibei: Sanmin. MacIntyre, A. (2007), After Virtue: A Study in Moral Theory, 3rd ed., Notre Dame: University of Notre Dame Press. Mover, D.S. (2011), ‘Situationism and Confucian Virtue Ethics’, Ethical Theory and Moral Practice 16 (1): 113–137. Nivison, D.S. (1996), ‘Replies and Comments’, in Philip J. Ivanhoe (ed.), Chinese Language, Thought, and Culture: Nivison and His Critics, 267–341, Chicago and La Salle, IL : Open Court. Parry, J., and P. Allison, eds (2020), Experiential Learning and Outdoor Education: Traditions of Practice and Philosophical Perspectives, London and New York: Routledge. Pavese, C. (2016a), ‘Skill in Epistemology I: Skill and Knowledge’, Philosophy Compass, 11 (11): 642–649. Pavese, C. (2016b), ‘Skill in Epistemology II: Skill and Know How’, Philosophy Compass, 11 (11): 650–660. Russel, D.C. (2009), Practical Intelligence and the Virtues, New York: Oxford University Press. Singh, D. (2014), ‘Zhuangzi, Wuwei, and the Necessity of Living Naturally: A Reply to Xunzi’s Objection’, Asian Philosophy 24 (3): 212–226. Slingerland, E. (2003), Effortless Action: Wu-wei As Conceptual Metaphor and Spiritual Ideal in Early China, New York: Oxford University Press. Snow, N. (2010), Virtue as Social Intelligence: An Empirically Grounded Theory, New York: Routledge. Stalnaker, A. (2010), ‘Virtue as Mastery in Early Confucianism’, The Journal of Religious Ethics 38 (3): 404–428. Stichter, M. (2007), ‘Ethical Expertise: The Skill Model of Virtue’, Ethical Theory and Moral Practice 10 (1): 183–194. Tang, S. (2016), Self-Realization through Confucian Learning: A Contemporary Reconstruction of Xunzi’s Ethics, New York: SUNY Press. Tang, S. (2020), ‘Responses to Hutton, Kim, and Loy’, Dao: A Journal of Comparative Philosophy 19 (1): 139–144. Yu, J. (2005), ‘Human Nature and Virtue in Mencius and Xunzi: An Aristotelian Inter-pretation’, Dao: A Journal of Comparative Philosophy 5 (1): 11–30. Xunzi (1996), A Concordance to the Xunzi, ed. D.C. Lau, Hong Kong: The Commercial Press.

Part Five

Skill and Ethics in the Zhuangzi

239

240

12

A Path with No End: Skill and Dao in Mozi and Zhuangzi Chris Fraser

Introduction How does skill relate to dao 䚃, the ethically apt path and its performance? Two anthologies by early Chinese ‘masters’ that make prominent use of craft metaphors imply profoundly contrasting answers to this question. For the Mozi ໘ᆀ, a key to following dao is to set forth explicit models or standards for guiding and checking performance. By learning to apply the right standards consistently, we can develop the skill needed to follow the dao of the sage-kings reliably, just as a carpenter uses a set square to produce square corners or a wheelwright uses a compass to fashion round wheels. Following dao – and thus the ethical life – is strongly analogous to the performance of skills. Like an artisan’s craft, dao has a fixed end that can be articulated explicitly. A sharply contrasting stance is implied by the renowned skill exemplar Cook Ding in the Zhuangzi 㦺ᆀ. Praised for how his every movement in carving up cattle is perfectly attuned, yielding a display of skill matching that of an exquisite dance or musical performance, Ding responds that what he actually cares about is dao, which is ‘advanced beyond skill’. Intriguingly, most of what Ding says in explaining this point pertains to how he developed his craft, performs his work, and overcomes new challenges. The implication is that the process of acquiring, performing, and extending skills exemplifies dao, yet there is something more to dao than skill. What is this something more? A skill is the ability to perform a task with a specified end in a competent manner. In the Zhuangzi, a key difference between skill and dao, I will suggest, is that dao is unlike skill in having no fixed, predetermined ends. Dao is a general, open-ended process, one that is continually shifting and transforming. We can never fully master dao, nor even know exactly 241

242

Skill in Ancient Ethics

where it will lead, as the nature of dao is such that we must regularly find creative ways of extending it as we proceed along it. Nevertheless, discussions in the Zhuangzi make clear that a distinction obtains between adept and poor performance of dao. Accordingly, there must be factors by which to distinguish more from less fitting paths or ends to pursue and more from less adroit ways of pursuing them. Here again, dao is closely intertwined with skill, as evaluations of dao performance resemble assessments of skill, couched in terms of how effective, successful, competent or adroit one’s activity is. If dao has no fixed ends, however, by what criteria can we assess whether some course of activity amounts to an effective or adroit performance of dao? A plausible Zhuangist answer, I propose, is that particular contexts themselves yield provisional grounds for such evaluations. These grounds can then be revised or replaced in response to developing circumstances and continuing performance of dao. The resulting approach to understanding and living the good life, I will suggest, can informatively be labelled an ethics of dao and de ᗧ (virtue), referring to the path we follow and the capacities by which we follow it. The upshot for the relation between skill and ethics is that in both Mohist and Zhuangist thought, a comparison with skills is helpful to understanding the theory and practice of ethics. For the Mohists, the ethical dao is strongly analogous to a skill. In the Zhuangzi, skilled performances may exemplify dao, but the flourishing practice of dao goes beyond skill, because unlike skills, dao has no fixed, determinate ends or boundaries.

1. Dao, models and skill in the Mozi The Mozi is a collection of writings by anonymous hands presenting the thought of Mo Di ໘㘏 (fl. c. 430 bc ) and his followers, who formed one of early China’s most prominent social and philosophical movements. The Mohists presented China’s first systematic ethical and political theories, which are grounded in their distinctive brand of communitarian consequentialism. As Mohist and other pre-Han texts use the word, ‘dao’ functions much like the English ‘way’. ‘Dao’ can refer to an actual method or manner of doing something, as when the Mozi speaks of the various daos by which inept rulers decrease the population (20/15).1 It can also refer to a normatively competent or 1

References to Mozi cite chapter and line numbers in Hung (1966).

Skill and Dao in Mozi and Zhuangzi

243

appropriate method of doing something, as when the texts speak of the dao of making clothing (20/4) or boats and carts (20/8) and ‘the dao by which to bring order to the people and unify the masses’ (12/61). By extension, it can refer more generally to a normatively apt set of practices or way of life. The Mohists claim, for example that their ethical norm of inclusive care is ‘the dao of the sage-kings’ (16/83). In such contexts, dao overlaps what we think of as morality. For example, referring to obvious moral values, Mohist writers speak of ‘the dao of benevolence and righteousness’ (25/80). Dao is primarily practical, not verbal. Mohist writings regularly pair it with statements or verbal teachings (yan 䀰), distinguishing explicit expressions or formulations of dao from dao itself. Statements can function as guidelines to direct us in following dao, and the activity of promulgating explicit teachings may be part of dao. But dao itself is the normatively apt conduct or practice (xing 㹼), not merely the verbal teaching that directs us toward that practice (10/27, 25/25). Nevertheless, the Mohists see verbal formulations and other explicit standards as crucial in clarifying and guiding the practice of dao.2 They refer to such guidelines as ‘models’ or ‘standards’ (fa ⌅). The Mohist ethical norm of allinclusive care for everyone is considered a ‘model’ of ‘the dao of the sage-kings’, for example (15/11, 16/83). The core Mohist ethical standard of promoting the benefit of all is another prominent model (32/1). One Mohist text urges that no affair of any kind can be undertaken successfully without the use of models (4/1). In labelling guidelines such as inclusive care ‘models’, the Mohists expressly assimilate them to tools that artisans use to guide and evaluate the performance of crafts, since in Classical Chinese such tools are also called ‘models’ (fa). Prominent examples are the wheelwright’s compass and the carpenter’s set square, used to produce round wheels and square corners and to check whether a particular wheel or corner is up to standard. As one Mohist text explains, all artisans, whether expert or not, use models to ‘measure’ their work (4/4). Models do not ensure perfect performance, but they reliably help the unskilled to improve. ‘The skilled can match the models exactly. The unskilled, although they cannot match them, by relying on them in their work still surpass [what they can do by] themselves’ (4/3–4).

2

Accordingly, the Mohists hold that dao can be explicitly articulated and that it can taught, two assumptions that are rejected in the Zhuangzi, as Lisa Raphals discusses in a recent anthology on skill in Zhuangzi (2019: 136–37).

244

Skill in Ancient Ethics

The craft analogies imply that practising dao is a matter of performing a skill.3 As in carpentry or wheel-making, the practice of dao can be clarified, guided, checked and improved by reference to explicit models that the performer seeks to emulate. The models are not identical to dao, nor does merely applying the right model ensure that one follows dao successfully. But models facilitate the normatively apt, skilled performance that is dao. The right model can articulate dao so clearly that using it to gauge what does or doesn’t conform to dao is ‘like differentiating black from white’ (27/72). The process by which models guide action underscores the skill-like character of dao performance. Models function as exemplars to be emulated or matched in carrying out a skilled task. In the paradigmatic cases, the carpenter uses the square to draw a line and then saws an edge that copies the line. The wheelwright uses the compass to mark the rim and then shaves the rim or adjusts the spokes until the rim matches the mark. In more abstract cases, the model might be a norm such as ‘benefiting people’. If a course of conduct conforms to the norm, we perform it; if not, we avoid it (32/1). Conceptually, the Mohists understand a model such as ‘benefit people’ or ‘inclusively care for each other, in interaction benefiting each other’ (15/10–11) as a practical standard to emulate in our course of conduct, rather than a theoretical general principle from which to draw inferences about what to do. A crucial presupposition drives the Mohist conception of dao as a practical performance, analogous to a skill, which can be guided by reference to models. The craft analogies imply that, like making wheels or houses, dao has fixed, specifiable ends. The compass and square function as clear, useful models only given the ends of producing round wheels and right-angled corners. Mohist norms such as all-inclusive care function as models of dao only given an understanding of dao as directed at the end of promoting the benefit of all, which the Mohists take to be the basic good that explains the moral values of benevolence and righteousness. By contrast, the Zhuangzi typically rejects benevolence and righteousness as values and questions whether dao has any fixed or specific ends.4 A dao with no 3

4

A difference between dao and skill is that we can refrain from performing a skill without thereby losing it, whereas we cannot refrain from conforming to dao without thereby being unethical. Someone who rarely rides a bicycle does not thereby lose the skill of being able to cycle, but someone who rarely follows dao loses any claim to ethical virtue. I suggest the Mohist explanation of this difference would be that dao is skill-like, but since it is a comprehensive way of life, we cannot refrain from engaging in it. On rejecting benevolence and righteousness, see, for example, Zhuangzi 6/84. (References to Zhuangzi cite chapter and line numbers in Hung [1956].) On a dao without fixed ends, see, for example, the passage quoted in section 6 below.

Skill and Dao in Mozi and Zhuangzi

245

specifiable end is a dao for which no models can be established – a dao that transcends any one skill.

2. Cook Ding on dao versus skill The Zhuangzi is an anthology of brief texts from roughly the late fourth to the mid-second century bc presenting a range of views that later archivists grouped together under the label of ‘Daoism’. The collection is named after a figure called Zhuang Zhou 㦺ઘ, about whom little is known, but it is unclear what portion of the material, if any, can plausibly be attributed to him. The texts offer a rich medley of criss-crossing ideas and viewpoints sharing various family resemblances; they are probably best read as a compendium of intersecting and overlapping conversations or social media posts than as a systematic attempt to present a unified, integral doctrinal stance. My discussion will treat selected passages as contributions to a discourse touching on dao and its relation to skill, exploring how various ideas in this discourse relate to each other and drawing out some of their implications. In some places, the discussion will reconstruct or elaborate views in the texts; in others, I may be assembling positions that are only latent in the source material. The most well-known discourse on skill in the Zhuangzi – Ding the butcher’s explanation of his craft – begins by distinguishing dao from skill and explaining that its chief concern is with dao, not skill. Yet the remarks that follow are almost entirely about Ding’s craft. What, then, does his discussion imply about the relation between skill and dao? Carving up an ox for Lord Wen Hui, his master, Ding steps, leans and slices as gracefully as if performing an elegant ritual dance to the rhythm of a symphony. Amazed, Wen Hui asks how skill (ji ᢰ) could reach such heights. Ding responds that what he cares about is dao, which is ‘more advanced than’ or ‘advanced beyond’ skill (3/5). He sketches how he developed his craft and performs it now, how he overcomes challenges, and the satisfaction his work brings. Wen Hui exclaims that from Ding’s remarks he has learned how to ‘nurture life’, or provide what is crucial to living well (3/12). The text thus frames the discussion as concerned with dao and living well. Ding’s remarks are presented as doubly significant, applying to the specific skill of meat carving and also to the general concern of living well. Here, then, is a first respect in which dao advances beyond skill: dao applies not merely to any one skill, such as Ding’s, but to living well in general. It

246

Skill in Ancient Ethics

generalizes in a way that skills do not. It is not a particular skill, but skilled work such as Ding’s can exemplify it – even bloody, filthy work slaughtering oxen and carving meat. As Guo Xiang says in his commentary, Ding’s concern is with the ‘patterns of dao’ (dao li 䚃⨶), which he ‘lodges’ (ji ᇴ) in his skill (Guo 1961: 119). Reflecting this concern with dao, rather than skill, Ding talks less about what he does in his work than how he does it. His focus is on the process, not the concrete technique (how he holds the knife, for example). This focus reflects the general nature of dao as understood in this context: it primarily concerns how we do what we do – the approach and manner – and only secondarily what we do: the substance or content. As we will see, however, the how shapes the what: applying the approach to dao that Ding illustrates leads us toward certain sorts of actions and away from others. Ding’s remarks can be divided roughly into three parts (3/5–11). First, he describes how he acquired his skill. As a beginner, he saw only whole oxen; as an intermediate, after three years of practice, he saw them as if already sliced up, directly perceiving the joints and seams where the knife would cut. Now, as an advanced expert, he says, ‘I meet it with my spirit, not looking with my eyes’. He acts without a reflectively self-conscious process of observing the ox and deciding how to cut, instead moving directly on the promptings of ‘spirit’, a more fundamental, automatic mode of agency.5 Next, Ding describes how he works. He carves smoothly and gracefully because he ‘complies with natural patterns (li ⨶)’ – cutting along the major gaps, being guided by the main seams, and thus working in accordance with (yin ഐ) what’s ‘inherently so’ in the structure of the ox. This way, he says, he never directly hits a ligament or tendon, let alone a bone. Through this approach, Ding boasts, he avoids even normal wear-and-tear on his knife, having gone a whole nineteenyear calendrical cycle without needing to sharpen it. There is space between the joints, and the knife edge has no thickness, he says, so ‘there’s always plenty of room for the blade to wander about’. Ding then goes on to explain how, despite his high level of skill, he regularly encounters intricate, knotty situations in which the grain and gaps are difficult to find. He cautiously prepares, slows down, focuses his vision, moves the knife subtly and the meat suddenly comes apart. Having resolved the difficulty, he steps back and looks around in relaxed satisfaction before wiping and sheathing the knife. 5

My interpretation here follows the Cui Zhuan and Xiang Xiu commentaries (Guo 1961: 120). For more on the role of ‘spirit’, see Fraser (2019).

Skill and Dao in Mozi and Zhuangzi

247

What aspects of Ding’s remarks generalize, informing us about dao rather than only meat-carving? What stands out, I suggest, is that dao amounts to a process of acquiring and exercising competence in detecting and responding to ‘patterns’, finding one’s way through ‘gaps’, and regularly working through difficulties. Ding’s experience implies that this process can be deeply satisfying. Looking more closely at his descriptions, we can expand on this characterization. Undertaking dao involves a lengthy process of learning and internalising a way of perceiving and interacting with things. Once we have acquired some dao, acting on it becomes spontaneous and automatic, something we do regularly without reflective self-consciousness. The adroit practice of dao lies in avoiding obstacles by working with the inherent grain or pattern (li) of things, according with (yin) the solid, unmovable facts about them, and so finding space through which to move. Moreover, if we can make ourselves ‘thinner’, figuratively speaking, reducing the extent to which we collide against things, we will more easily fit through the gaps and can avoid being worn down. Crucially, dao is not simply a matter of applying practised expertise to familiar, routine cases. It also includes regularly extending existing skills by working our way through unfamiliar, difficult situations, without knowing in advance exactly how we will do so. Dao refers not only to a well-formed path and a manner of following it; it is also the spontaneous, creative process by which we find a way forward when the path is difficult or blocked.6 The recurrent experience of being pulled from our comfort zone and forced to navigate knotty, unsure terrain is another respect in which we might say that dao advances beyond skill.7 Presumably, this feature marks a sense in which dao cannot be taught or handed down to others, since each of us must find a way for ourselves. It also helps explain why Ding nowhere mentions the models (fa) that are so prominent in the Mohist conception of dao as clearly articulated and signposted by explicit, determinate standards. Since dao as Ding presents it regularly leads beyond familiar, routine cases, it draws on an uncodifiable, creative capacity for adaptation that cannot be captured by fixed, predetermined standards.8

6 7

8

Dan Robins (2011) emphasizes this point, which I will develop below. This is Robins’ suggestion (2011). Arguably, however, skills do include a capacity to adapt to new challenges, albeit within the circumscribed field of the skill. In this regard, consider the contrast between the Mohist figure of the wheelwright guiding and checking his work against the compass and the wheelwright depicted in the Zhuangzi (13/68–74), who claims that mastery in cutting wheels rests on an inexpressible, unteachable art of chipping the wood not too slowly or hastily. The two visions are not necessarily contradictory; cutting wheels could be a subtle art with results that nonetheless need to be checked against the compass. But the difference in emphasis is informative as to the two contrasting conceptions of skill and dao. For further discussion of these conceptions of skill, see Raphals (2019: 133–35).

248

Skill in Ancient Ethics

Cook Ding’s comments thus depict dao as a process or an approach—how to learn and proceed adeptly in some field of activity while regularly overcoming challenges. Ding’s own skill exemplifies this process, but the process itself could apply to any activity. This generality reflects another respect in which dao is distinct from and advances beyond skill. A skill is not just a process: skills are partly defined by their ends. Ding is a skilled meat-carver not simply because of how he approaches his work but because he produces cleanly sliced cuts of meat. Skills such as meat-carving come with inherent ends by which to evaluate performance. Ding knows he has successfully handled the knotty sections when the meat suddenly comes apart—falling, he says, ‘like a clump of earth to the ground’. The carpenter and wheelwright perform their skills well when their work matches the square or compass and they produce houses that stand firmly upright or wheels that roll smoothly. Indeed, the process of performing a skill is intelligible as such only in the context of the defining end to which the skill is directed. By contrast, if my suggestion above is plausible, dao in this context is a general process without any fixed end. What factors, then, determine whether we are indeed pursuing dao aptly or poorly? Lord Wen Hui declares that Ding’s remarks show how to ‘nurture life’. Perhaps a criterion of apt dao-following is whether a person stays alive and healthy – indeed, perhaps my suggestion is mistaken, and the end of dao is life. However, numerous Zhuangzi passages indicate that the adept see life and death as two aspects of a single process, without privileging life over death (e.g., 5/30, 6/46). It seems that the Zhuangist practitioner seeks to stay healthy and alive when doing so is the obvious, fitting course but has no hesitation about accepting death when it too seems fitting. So death need not reflect a failure to follow dao well. The point of Wen Hui’s remark is not that dao aims at preserving life; it is that Ding’s descriptions illustrate the process of how to live a flourishing life. Dao is advanced beyond skill in applying not just to the pursuit of one or another end but to any end. Here an obvious question arises. Does the conception of dao as a process with no specific end cover any activity we might undertake, without constraint?9 Some ends are deplorable. Can a vicious but meticulous and skilful serial killer justifiably claim to follow dao well? Perhaps guidelines

9

Steven Coutinho raises a pair of related questions (2019: 97). Does Zhuangist thought have the resources to distinguish skilled performances that reflect sageliness from those that do not? Has it grounds by which to disapprove of, for example, actions that harm others? As the discussion below explains, I think the Zhuangzi can resolve both issues.

Skill and Dao in Mozi and Zhuangzi

249

like the Mohists’ fixed, determinate ‘models’ are needed as constraints to exclude the serial killer from claiming mastery of dao. The Cook Ding story implies that there are indeed constraints. Ding distinguishes a mediocre cook, who changes his knife monthly, from a good cook, who changes it annually, from Ding himself, who has gone nineteen years without dulling a blade. Clearly, there are better and worse ways of following dao. Arguably, the serial killer’s approach to dao is deeply defective. No matter how skilful his actions may seem, they will most likely induce retribution eventually, rendering his path a repugnant failure. But the relevant constraints, I suggest, are neither general nor fixed. They arise from the process itself, from how well, in a particular context, we conform to the natural patterns, follow the seams and gaps, accord with what is inherently so, and avoid obstacles. What, then, distinguishes how well a particular agent fulfills these criteria? This is a crucial ethical question for Daoist thought. How do we distinguish apt dao from injudicious dao or adept from inept dao-following in the absence of fixed standards or ends?

3. Proficiency in the patterns The question of what marks competence in dao arises again when we look at Zhuangzi passages that focus on dao, rather than skill. The texts describe dao itself and adept performance of dao in terms familiar from descriptions of skill, yet they give no specific, determinate criteria to mark the competent practice of dao. The ‘Autumn Waters’ dialogue, for example, discusses how dao can guide conduct, explaining dao as Cook Ding does by appeal to the patterns (li ⨶) of things (17/46). The practical conditions we encounter are constantly changing, the text holds, with the result that concrete, action-guiding distinctions that seem appropriate in one context may need to be reversed in another. Adept performance of dao requires flexible, adaptive responses, not commitment to fixed norms, as the Mohists advocate. ‘Don’t constrain your intent, or you’ll be hobbled in following dao . . . don’t proceed by a single [standard], or you’ll be at odds with dao’ (17/42–43). We are to ‘embrace the myriad things’ and proceed without limits, boundaries, biases, favouritism, or any fixed direction, seeking to accord with how things ‘transform of themselves’ (17/41–47). ‘Knowing dao’ is equated with ‘attaining proficiency (da 䚄) in the patterns’ and thus understanding how to adapt our conduct to changing situations (17/48). This emphasis on adaptive, contextual responses to patterns resonates with the account of how

250

Skill in Ancient Ethics

Cook Ding conforms to the inherent features of the ox and with descriptions of other Zhuangzi skill exemplars, such as the whitewater swimmer who flows with the dao of the water (19/52–53), the artisan whose fingers transform along with things (19/62), or the woodcarver who lays his hand to work only when he can ‘see’ the finished carving latent in the inherent nature of the wood (19/58). The prominence of the concept of ‘pattern’ or ‘grain’ (li) in ‘Autumn Waters’ and the Cook Ding story reflects the important role of this notion across the preHan literature. ‘Pattern’ refers to facts about how things are structured or organized, how they relate to or interact with each other, and how they develop, proceed, or transform. Conforming to patterns is thus a prerequisite for proper or successful action. Misunderstanding or overlooking them is likely to lead to error or failure. Accordingly, following dao well lies in finding a path that aligns with the patterns; following it badly is struggling against or conflicting with the patterns.10 A distinctive feature of the Zhuangist approach is that the patterns are understood to be deeply contextual and continually transforming. How do abstract descriptions of the ideal of adaptively responding to patterns translate into practice? Two Zhuangzi stories concerning social interactions illustrate such responses well. The first is a story of failed interaction, the second of a successful response. Failure is illustrated by the story of a seabird that serendipitously landed near the capital of the landlocked state of Lu (18/33–39). Delighted by the bird’s visit – an auspicious omen – the Lord of Lu had it escorted to the ancestral temple, where he honoured it with a ritual feast and musical performance. Sadly, the bird became upset, refused to eat or drink, and soon died. Despite his good intentions, the lord only harmed the auspicious guest, because he followed customary norms for honouring a human dignitary, neglecting the reality that music and crowds would only disturb a bird, who instead needed to fly about, roost in trees, and dine on grubs and bugs. The lesson of the story is that ‘names stop at reality; what’s right is determined by what fits’ (18/39). To proceed adeptly, we must set aside labels or titles (‘names’, ming ਽) associated with conventional conceptions of right and wrong and instead attend to the ‘reality’ of the situation (shi ሖ) and what best ‘fits’ it (shi 䚙), adapting our actions to the facts (such as a bird’s normal diet), rather than rigidly following codified standards (such as the ritual protocol for hosting an honoured guest). To do so, the text explains, is to 10

Ontologically, then, dao might be regarded as inherent in the patterns. In some strands of Daoist thought, it may be regarded as the source of the patterns, referring to the way in which they issue forth. These fascinating aspects of dao deserve careful treatment but are beyond the scope of my discussion here, which focuses on dao as an ethical path.

Skill and Dao in Mozi and Zhuangzi

251

‘attain proficiency (da) in the patterns (tiao ọ, a synonym of li) and preserve welfare’ (18/39). In failing to deal with the bird appropriately, the Lord of Lu ineptly disregarded the relevant patterns, with tragic results. A contrasting, successful interaction with others occurs in the story of a monkey-keeper who accommodated his charges by adjusting their daily menu (2/38–40). The keeper announced that the monkeys would get three nuts in the morning and four in the evening. Preferring a larger breakfast, the monkeys were angry. So the keeper reversed the allocation: everyone would have four in the morning, three in the evening. The monkeys were delighted. The keeper ‘harmonized’ (he ઼) the situation by adjusting the nut distribution in a way that defused the monkeys’ anger at no cost to himself, as the total allocation remained seven nuts per day (2/39–40). This adjustment illustrates what the text calls ‘according-shi’ (yin shi ഐᱟ), or adaptively and provisionally ‘affirming’ things or taking them to be ‘right’ (shi ᱟ) ‘in accordance with’ (yin ഐ) particular circumstances (2/37) – just as Cook Ding proceeds ‘in accordance with’ (yin) what is ‘inherently so’ in the grain of the meat.11 The discussion leading up the monkey story contends that action-guiding distinctions between ‘this/right’ and ‘not/wrong’ or between ‘so’ and ‘not-so’ are determined by the dao we carry out, which ‘completes’ one among multiple potential ways of proceeding (2/33).12 Apart from the dao we practise, nothing is inherently right or wrong, ‘so’ or ‘not-so’. Hence if a path we undertake runs into difficulty, we are free to change direction. Grasping these points, the adept refrain from ‘imposing-shi’ (wei shi ⛪ᱟ), or deeming things this or that on the basis of fixed standards applied without regard for particular contexts. Instead, they ‘accommodate things in the ordinary’ (2/36). The ‘ordinary’ (yong ᓨ) is what is useful or effective (yong ⭘), successful (de ᗇ), and connects or proceeds in a proficient, free-flowing manner (tong 䙊) (2/36–37). To accommodate things successfully in some context is to apply ‘according-shi’. Indeed, the adroit practice of dao lies in acting on such provisional, ‘according-shi’ attitudes without knowing one is doing so – without knowing what the appropriate responses will be, since they are discovered in the course of our activity (2/37).13 An implication

11

12 13

I follow A.C. Graham (1969/70) in taking yin shi ഐᱟ to be a set phrase because the two graphs occur together in the text four times apparently referring to the same idea. As Graham proposed, the phrase yin shi seems to contrast with wei shi ⛪ᱟ, which also appears four times, referring to insistently imposing some shi judgment on things. The next several paragraphs draw on the discussion in Fraser (forthcoming), section 3.3. At 2/37, I follow Wang Shumin in taking ഐᱟᐢ to be equivalent to ഐᱟҏ. I also follow Wang Yinzhi ⦻ᕅѻ in reading ᐢ㘼н⸕ަ❦ as equivalent to ↔㘼н⸕ަ❦ (1988: 64). In addition, see Chen (2007: 72).

252

Skill in Ancient Ethics

is that, as discussed in section 3, dao is indeed exemplified by the process of working through the knotty parts without knowing in advance exactly how we will do so. The monkey-keeper’s compromise with his wards thus illustrates a general conception of dao as an adaptive response to circumstances that facilitates ongoing ‘ordinary coping’, or useful, successful, free-flowing movement along a path presented by the situation. Instead of acting on fixed standards of right and wrong, as the Mohists advocate, the idea is to respond adaptively to the ‘patterns’ operative in the context. On this view, the criteria of appropriate action are a cluster of notions such as doing what ‘fits’ or ‘accords with’ the ‘reality’ of the particular situation or what ‘attains proficiency’ (da) in the ‘patterns’ and ‘harmonizes’ with things. An apt course of action will seem ‘ordinary’ while being ‘effective’, ‘successful’, and ‘free-flowing’ or ‘proficiently connecting through’. The overlap between this cluster of terms and descriptions of skill is striking. In effect, conduct that accords with dao is conceptualized along the lines of adept skill performance, using terms implying facility, competence and proficiency. These terms again raise the question of what criteria distinguish apt from inapt dao-following. What makes some course of conduct ‘fitting’, ‘harmonious’, ‘successful’, or ‘free-flowing’?

4. The ends of dao How do we evaluate how well some course of action ‘fits’ (shi 䚙) or ‘flows’ (tong 䙊) in a concrete context? Since dao in general has no fixed ends, a plausible Zhuangist answer is that particular contexts themselves provide provisional or pro tanto grounds for such evaluations. These grounds can then be revised or replaced as we go along, in response to developing circumstances and continuing performance of dao.14 Think of our present dao as encompassing our values, interests and ends, our capacities, and our current path and manner of activity. In any given situation, we find ourselves proceeding according to some dao. As we do so, we interact with our environment and with other agents whose dao intersects or converges with ours, both of which we must take into account if we are to proceed in a fitting, free-flowing way. Our initial dao and our relation to the context, including other agents, jointly provide starting criteria by which to evaluate how well various 14

This section draws on the discussion in Fraser (forthcoming), section 4.

Skill and Dao in Mozi and Zhuangzi

253

ways of continuing forward fit the situation, flow freely, attain harmony, preserve welfare, and so forth. A fitting path forward will accommodate our ends – and those of others we interact with – as we proceed with our dao. Our preliminary conception of good ‘fit’ (shi) or ‘success’ (de) may simply be to continue following our initial dao smoothly, using our existing capacities to fulfill the provisional values and ends we began with. As we proceed through concrete, changing circumstances, however, our dao may need to be modified, and with it our ends and criteria of ‘fit’. Perhaps the dao we are pursuing leads to obstacles, generates conflict, or creates frustration – all indications of a failure to accord with the patterns at hand and of poor fit between our conduct and the context. Perhaps we discover that some of the values or ends we take for granted clash with each other, such that we must modify or forgo some in order to proceed smoothly with others. Perhaps we adopt new values or ends as we extend our path. In all of these cases, we may need to refine or revise our dao, including the internal criteria by which we assess it as ‘fitting’, ‘successful’ or ‘free-flowing’. Cases such as the monkeys and the seabird illustrate this contextual approach to evaluating ‘fitting’ or ‘free-flowing’ activity. The monkey-keeper’s initial dao involved caring for the monkeys and feeding them seven nuts per day, divided among meals in a certain way. The monkeys protested, creating an obstacle to this dao and thereby indicating a failure to ‘fit’ or ‘flow freely’. For the keeper, the criteria of a fitting, successful way forward were ends such as preserving the monkeys’ welfare, calming them, and meeting the overall nut budget of seven per day. So he adjusted the allocation per meal – a relatively marginal feature of his original dao – to accommodate his charges’ preferences while still proceeding in a way consistent with other, more important features. By criteria such as allocating seven nuts per day, keeping the monkeys happy and achieving harmony with them, his modified dao proved ‘competently free-flowing’ (tong). The Lord of Lu responded to the seabird’s visit according to an elaborate religious and cultural dao involving cosmological beliefs about auspicious omens and ritual norms concerning how to honour visiting dignitaries. Obtusely acting on this dao without regard for the ‘reality’ of the situation or the ‘patterns’ of avian welfare led to a disastrous outcome for both sides, as in seeking to honour the bird the lord only harmed it. By both the lord’s and the bird’s lights, a more ‘fitting’, competent path would have been for the lord to modify his original dao in a way that fulfilled his aim of celebrating the rare visit while also successfully nurturing the bird. Perhaps, for example, he could have moved the bird to a wetland sanctuary, affixed a commemorative marker to the tree where it landed, and conducted a ritual at the ancestral shrine.

254

Skill in Ancient Ethics

Zhuangzi writings offer no specific method for finding such ‘fitting’ courses of action beyond proceeding with an open, ‘empty’ (xu 㲋) mind, seeking to adapt to the ‘patterns’ and ‘what’s inherently so’ without rigidly imposing any preconceived way of going on.15 This continual, subtle process of adjustment again reflects how dao is intertwined with skill. The apt dao is discovered by working our way through intricate, difficult spots in a skill-like manner, much as Cook Ding delicately feels his way through the knotty parts. In doing so, we draw on the tacit capacities that have brought us where we are without knowing beforehand just how we will proceed next. Moreover, given that dao has no determinate end, in many cases there is unlikely to be any uniquely correct way to go on. Perhaps, reappraising his nut budget, the monkey-keeper could instead have offered four nuts at both meals. Perhaps, contemplating the monkeys’ happiness, he could have set them free in a forest to find their own nuts. Perhaps, reconsidering the bird’s likely preferences, the Lord of Lu could have simply let it fly off and offered a prayer of thanks. Or perhaps, on seeing the bird, he might have realized it was simply lost and dropped his cosmic-religious interpretation of the event. The aim is not necessarily to follow the ‘correct’ dao, for there may be no such thing. It is to find a path that provisionally allows us to move on, balancing various ends and addressing obstacles or frustrations as best we can. The complex, dynamic nature of the ‘patterns’ of things means that any such path must be undertaken provisionally, with humility. Our initial moves through the ‘knotty’ parts may fail to fully ‘fit’ all of the relevant patterns or may ‘harmonize’ the factors in play partly but not wholly. New patterns may emerge as we go. Our actions themselves may change the situation, such that further responses are needed. A judicious dao-follower will remain perpetually open to adjustments in the path, aware that, insofar as dao can be said to have a rough, general aim, it just is the ameliorative process of continually seeking out what ‘fits’ or ‘flows’ in evolving circumstances, by standards of ‘fit’ or ‘flow’ that may themselves also evolve. So a further respect in which dao is like skill yet goes beyond it is that we must regularly apply our existing competence to reassess and reform our dao in response to changing circumstances that may jar with aspects of it as developed so far. In performing his skill, Cook Ding finds his way through the knotty parts by making fine adjustments to his knife-work. By extension, in navigating through the ‘patterns’ we encounter, we all must make fine adjustments to our 15

The notion of an open, ‘empty’ (xu) mind is pivotal to Zhuangist moral psychology. For more discussion, see Fraser (2014b).

Skill and Dao in Mozi and Zhuangzi

255

dao to find the most fitting path we can. The result may sometimes be a rather different dao from what we started with.16

5. An ethics of dao and de The conception of dao I have been describing entails continually adjusting our provisional ends as we navigate through shifting circumstances. For an example of this view of dao, consider the story of the mountain tree, which criticizes the idea, prominent in some Zhuangzi passages, that being deemed useless by others may be useful to oneself (see, e.g., 1/42–47, 4/64–75). The story contrasts a mountain tree that grows large because woodcutters find it useless for timber with a goose that is butchered because it cannot honk (20/1–9). The tree’s apparent uselessness led to health and longevity, but the goose’s led to premature death. So, a student asks a fictional Zhuangzi, where should we settle? To live well and avoid ‘entanglement’ with things, should we present ourselves to others as worthy or worthless? Zhuangzi suggests that neither worthiness nor worthlessness is an appropriate aim. Nor indeed is a moderate course between them, seeking to avoid either worthiness or worthlessness. Instead, he suggests: Wander about by riding on dao and de . . . without praise or criticism, now a dragon soaring in the sky, now a snake slithering along the ground, transforming together with circumstances, never committing to acting only one way, now above, now below, taking harmony as your measure. Wandering about with the source of the myriad things, letting things be things without letting them treat you as any [fixed] thing. . . . Students, remember this! Make your home only in dao and de!17 20/6–9

Instead of following any set norm or committing to any one path or end – being useless, useful or somewhere in-between, for instance – the text advocates ‘riding along’ with dao and de, allowing ourselves to be drawn along fitting paths (dao) through our inherent virtue or capacity (de) for finding and following them. To ‘ride dao and de’ is to wander about with the flow of natural processes – the mysterious ‘source’ of things – pursuing no particular direction or end by which 16

17

Echoing these observations, Wai Wai Chiu suggests that key features of skillful performance in Zhuangzi include fluency, fine-tuning, and freedom from previously instilled ways of proceeding (2019: 7). An alternative interpretation, reading 䜹 as ੁ, is ‘Take only dao and de as your direction!’ An alternative interpretation of the preceding line is to avoid becoming enslaved to things.

256

Skill in Ancient Ethics

to praise or blame, proceeding now in this manner, now in that, undertaking different modes of activity without committing exclusively to any. We are to shift along with circumstances, seeking only to maintain harmony (he ઼) with things as we flow along, adjusting fluidly without becoming a determinate ‘thing’ that others could aim to use. No fixed norms or ends capture the adaptive, responsive ‘transformation’ (hua ॆ) characteristic of employing our inward de to follow dao as we encounter it. In light of passages such as this, I suggest we can informatively describe the Zhuangzi as presenting an ethics of dao and de.18 The central concepts in this ethics are not right and wrong or moral virtues and vices, but apt or fitting paths of conduct and the virtue, or capacity for agency, by which we follow such paths. The focus is on the path we pursue and how we pursue it – whether the path ‘flows’ or ‘connects through’ (tong 䙊), ‘fits’ (shi), and yields ‘harmony’ and whether our manner of activity displays the flexibility, resilience and creative responsiveness needed to find and follow such a path. Rather than applying familiar moral norms, this ethics assesses conduct and character in terms similar to how we assess the performance of skills: by how responsive we are to particular situations, how proficient we are at proceeding along a sustainable course of activity, how resilient and adaptable we are in dealing with change, challenges, and misfortune.19 To perform dao well is to find our way through a field of ‘patterns’ (li) freely and smoothly, with harmony and ease, while avoiding hindrance or obstruction, in a manner akin to the competent performance of an art or a skill. The good life lies in manifesting de in such performances and thereby ‘wandering’ (you 䙺) through our circumstances, much as Cook Ding’s knife is said to ‘wander’ through the spaces between the joints of the ox (3/9).20 The metaphor of dao as a process of ‘wandering’ is telling: dao is not a path toward a specific end or destination, but a meandering course toward nowhere in particular. This conception of apt conduct and the good life helps to explain the intense interest in skills in the Zhuangzi, as skilled performances offer examples of fluid, adaptive responses to changing circumstances, albeit within the fixed scope of a particular skill, with a particular end. An instructive way of understanding one salient Zhuangist ethical vision, then, is that the crux of a flourishing life lies in ‘advancing beyond skill’ by extending to life as a whole the open, ready 18

19 20

I elaborate on this proposal in Fraser (forthcoming), section 4, from which some remarks in this section are drawn. For a discussion of the moral psychology implied by these remarks, see Fraser (2014b) and (2019). For this interpretation of the Zhuangist good life, see Fraser (2011) and (2014a).

Skill and Dao in Mozi and Zhuangzi

257

responsiveness familiar from the performance of skills.21 The ideal – or at least the version of a Zhuangist ideal implied by whoever compiled the collection’s first chapter, ‘Wandering About Freely’ – is to roam without fixed ends or bounds, relying on nothing in particular and thus following no fixed standards or norms (1/21). Consider a story from ‘Wandering About Freely’ that contrasts living adeptly with its opposite, clumsiness, while extending the metaphor of dao-following as carefree wandering. Zhuangzi’s friend Huizi was given the seeds of a large bottle gourd, which grew to produce gigantic, 500-litre fruits. Gourds were typically used as containers or ladles. Huizi tried using the giant gourds as containers, but when he filled them with water, they collapsed, unable to hold up their own weight. He tried cutting them open to make ladles, but they proved too large to dip into anything. Frustrated, he declared the gourds useless, he tells Zhuangzi, and smashed them (1/36–37). Zhuangzi berates Huizi for being clumsy (zhuo ᤉ) at using big things. Instead of worrying that the gourds were useless as containers or ladles, he says, why not make them into floats, climb astride, and go drifting about – that is, wandering – on rivers and lakes (1/42)? Nothing is useful or useless in itself; usefulness is always relative to some norm or end. Huizi has locked himself into trying to use the gourds according to fixed ends inapplicable to rare, gigantic fruit. His narrowminded competence in conventional uses of regular-sized gourds renders him clumsy in finding creative uses for unusually large gourds. His mind is overgrown with brambles, remarks Zhuangzi (1/42) – obstructed, closed off to novelty, and thus inept at finding new ends to fit unusual situations. Those who ‘ride along with dao and de’, by contrast, develop a open-minded readiness to go beyond familiar ends to find creative ways to engage with their circumstances.

6. Conclusion The Mohists explicitly compare ethical practice to performing a skill. Like a skill, dao has a determinate end, which explicit ‘models’ can guide us in pursuing, much as the compass and square guide artisans in producing round wheels and square corners. To a large extent, the ethical life for the Mohists is a matter of mastering the skill of reliably distinguishing and undertaking what is benevolent and righteous. 21

As Coutinho suggests, in a Zhuangist philosophy of skill, ‘living well’ and ‘cosmic wisdom’ are ‘manifested in practical adeptness in negotiating our environments’ (2019: 87).

258

Skill in Ancient Ethics

For a prominent strand of thought in the Zhuangzi, on the other hand, the performance of skills can exemplify how to proceed with dao, but dao itself goes beyond skill, extending to life as a whole the adaptive learning, competent performance, and problem-solving acumen displayed in skilled activity. Unlike skill, dao is a general, unbounded process with no fixed aims or purpose. In the course of this process, we find ourselves acting on contextually specified models, norms, values, and methods, but these are continually open to revision or replacement. Circumstances regularly present us with ‘knotty’ stretches, prompting us to find creative, adaptive ways of going forward that draw on but go beyond the capacities and path we have developed so far. Indeed, I suggest, the crux of dao lies in working our way through such stretches, extending, redirecting, and refashioning our path. Unlike any skill, or indeed any activity with specified boundaries and goals, dao is a path with no end – one on which we continually make and remake our way forward.

References Chen, G. 䲣啃៹ (2007), Zhuangzi: Contemporary Notes and Paraphrase 㦺ᆀӺ䁫Ӻ䆟, 2 vols. (rev. ed.), Beijing: Commercial Press. Chiu, W. (2019), ‘Skilful Performances and the Zhuangzi’s Lessons on Orientation’, in K. Lai and W. Chiu (eds), Skill and Mastery: Philosophical Stories from the Zhuangzi, 3–14, London: Rowman & Littlefield. Coutinho, S. (2019), ‘Skill and Embodied Engagement’, in K. Lai and W. Chiu (eds), Skill and Mastery: Philosophical Stories from the Zhuangzi, 85–99, London: Rowman & Littlefield. Fraser, C. (2011), ‘Emotion and Agency in Zhuangzi’, Asian Philosophy 21: 97–121. Fraser, C. (2014a), ‘Wandering the Way: A Eudaimonistic Approach to the Zhuangzi’, Dao: A Journal of Comparative Philosophy 13: 541–565. Fraser, C. (2014b), ‘Heart-Fasting, Forgetting, and Using the Heart Like a Mirror: Applied Emptiness in the Zhuangzi’, in J. Liu and D. Berger (eds), Nothingness in Asian Philosophy, 197–212, New York: Routledge. Fraser, C. (2019), ‘The Ferryman: Forget the Deeps and Row!’, in K. Lai and W. Chiu (eds), Skill and Mastery: Philosophical Stories from the Zhuangzi, 163–181, London: Rowman & Littlefield. Fraser, C. (forthcoming), ‘Finding a Way Together: Interpersonal Ethics in the Zhuangzi’, in K. Chong and K. Cheng (eds), Dao Companion to the Philosophy of the Zhuangzi, Dordrecht: Springer. Graham, A. (1969/70) ‘Chuang-tzu’s Essay on Seeing Things as Equal’, History of Religions 9: 137–159.

Skill and Dao in Mozi and Zhuangzi

259

Guo, Q. 䜝ឦ㰙 (1961), Collected Explications on Zhuangzi 㦺ᆀ䳶䟻, Beijing: Zhonghua Shuju. Hung, W. (1956), A Concordance to Chuang Tzu, Harvard-Yenching Institute Sinological Index Series, Supplement no. 20, Cambridge: Harvard University Press. Hung, W. (1966), A Concordance to Mo Tzu, Harvard-Yenching Institute Sinological Index Series, Supplement no. 21, Cambridge: Harvard University Press. Raphals, L. (2019), ‘Wheelwright Bian: A Difficult Dao’, in K. Lai and W. Chiu (eds), Skill and Mastery: Philosophical Stories from the Zhuangzi, 129–142, London: Rowman & Littlefield. Robins, D. (2011), ‘ “It Goes Beyond Skill” ’, in C. Fraser, et al. (eds), Ethics in Early China, 105–124, Hong Kong: University of Hong Kong Press. Wang, S. ⦻਄ዧ (1988), Collated Interpretations of Zhuangzi 㦺ᆀṑ䂞, 3 vols., Taipei: Academia Sinica Institute of History and Philology.

260

13

Skilfulness and Uselessness in the Zhuangzi Wai Wai Chiu

1. A tension in the text Among pre-Han philosophical texts, the Zhuangzi is most famous for its extensive discussion and analysis of skill.1 Stories about such characters as Cook Ding (7/3/2–8/3/12)2, who can carve an ox without hacking and smashing, the swimmer (50/19/49–54), who can swim even in choppy waters, and Woodworker Qing (50/19/54–59), who can make bell-stands look as if they were the work of spirits, have attracted scholarly attention because of their literary and philosophical value. One might even suggest that skill cultivation helps one to embody dao 䚃, as the term ‘dao’ appears in several skill stories, including that of Cook Ding and the cicada-catcher (48/19/17–21). This emphasis on skill, however, seems to be in tension with another theme in the text, namely the emphasis on uselessness (wuyong ❑⭘). In all skill stories, being skilful is being good at something. Masters of a particular skill either create useful products, provide useful services, engage in activities that are useful for human beings, or at least can be reasonably expected to serve one of these uses in the future. In contrast, stories of uselessness, such as the oak tree that teaches a lesson to a carpenter in a dream (11/4/64–75), suggest that being useful is dangerous because it invites exploitation and entanglement (lei ㍟). This is especially so when society is not at peace and the government is keen on gaining power and fame rather than protecting people’s lives. In the story of Wheelwright 1

2

I take the majority view among scholars of the Zhuangzi that the Inner Chapters contain the most important philosophical materials in the text. This does not mean that the Inner Chapters are written by the historical Zhuang Zhou 㦺ઘ or that the passages in the Inner Chapters are all among the earliest materials. For a recent discussion of the Zhuangzi’s authorship, see Klein 2010. All references to the Chinese text of the Zhuangzi are from Zhuangzi Yinde (Hong 1986). All of the translations in this paper are my own, although I rely on the following commentaries and translations: Zhuang Zi Ji Shi (Guo 1961), Chuang-Tzu (Graham 1981), Wandering on the Way (Mair 1994), and Zhuang Zi Jin Zhu Jin Yi (Chen 2001). The order of the chapters and sentences follows Chen’s book.

261

262

Skill in Ancient Ethics

Bian (36/13/68–74), we see a craftsman being threatened with death while serving a king, which shows clearly the danger of being involved in politics – a danger about which Zhuangzi repeatedly warns (10/4/53–11/4/64, 45/17/81– 84). If one strives to become skilful, one risks being ‘used’ by others and may end up in peril. The tension between skilfulness and uselessness is more prominent in the Zhuangzi than in its predecessor, the Daodejing, because the latter does not tell any skill stories or give a positive formulation of skilfulness. Of course, one may agree with Schwitzgebel that Zhuangzi actually ‘does not see skill as particularly worth aiming for, and he does not particularly admire people who display skilful mastery of their arts’ (Schwitzgebel 2019: 101). After all, as Schwitzgebel states, there are many passages in the Inner Chapters that criticize skilfulness, and only one story that appears to praise it (Schwitzgebel 2019: 102–104). Nevertheless, even if readers take the Inner Chapters as the core of the whole Zhuangzi, they may still have two reasons to wonder whether the theme of skilfulness can be reconciled to the theme of uselessness. First, an interpretation that can reconcile those two themes would have more theoretical virtue according to the principle of humanity. Second, reconciliation would let us better appreciate Zhuangzi’s philosophy of engagement and detachment, with skilfulness as an instance of the former and uselessness as an instance of the latter. Such a reconciliation is the work of this chapter. I begin by differentiating uselessness from idleness, then point out that the relationship between skilfulness and uselessness in the Zhuangzi can be understood in three ways. First, skill cultivation need not be subjected to material success or social hierarchies, and in this sense the cultivator is not ‘used’. Second, to avoid being used by others, one needs to know how to respond to others without being swayed or manipulated. In this sense, being ‘useless’ requires skill. Third, uselessness can be the foundation rather than the opposite of usefulness, and there is a sense in which uselessness complements skilfulness. I conclude with the suggestion that the relationship between uselessness and usefulness is also manifest in Zhuangzi’s view of politics.

2. Uselessness and timeliness Chapter 20 of the Zhuangzi (Shan Mu ኡᵘ) contains a story about uselessness with a dialogue between Zhuangzi and his disciples (51/20/1–9). One day, Zhuangzi says to his disciples that a tree can survive for a long time because it is

Skilfulness and Uselessness in the Zhuangzi

263

good for nothing (bucai нᶀ), echoing the story of the oak tree in Chapter 4 (Ren jian shi Ӫ䯃ц). However, when he lodges in a friend’s house, his friend treats him by killing a goose that cannot honk. His disciples then ask him how he positions himself, given that uselessness helps the oak tree to live a long life but invites a quick death for the goose. Zhuangzi’s response is quite long and somewhat obscure, but the gist is to ‘drift and wander by riding dao 䚃 and de ᗧ’, ‘transform together with the times’ and ‘take harmony as one’s measure’. These remarks, I believe, acknowledge that it is futile to be one-sided: the way of navigating one’s life always changes, and a person is advised not to stick to either usefulness or uselessness. Here, my interpretation of the story of the unfortunate goose and the oak tree is slightly different from that of Major (1975: 271–273). Major states that the goose is doomed because it is still useful to humans, whereas the oak tree in Chapter 4 is really good for nothing and thus represents Zhuangzi’s ideal. This statement overlooks two points. The first point is that, in the story, the oak tree is not totally useless, because it serves as a shrine. Presumably many people visit the tree, so it is useful as a place for pilgrimage or religious assembly. The carpenter, who blames the oak tree for being useless only to be taught a lesson by it in a dream, comments that it acting as a shrine is a kind of ‘pretext’ (ji ᇴ). This implies that the tree does not take its post seriously, but only regards it as an expedient way to avoid unnecessary harm. It is thus not like people who are too attached to their career, status or social role. These kinds of attachment fail to recognize that ‘observed in the light of dao, things are neither noble nor base; observed in the light of things, they see themselves as noble and the other as base; observed in the light of convention, a thing’s nobility or baseness does not reside in itself ’ (43/17/29–30). For Zhuangzi, conventional values are subject to change, and one who embodies dao treats conventional value judgements lightly. This does not mean that sages simply refuse to make value judgements. Various characters in the Zhuangzi, even when they seem to represent the text’s notion of ideal personhood, do make judgements and respond to other people; however, they ‘respond but do not retain’ (21/7/33); in other words, they respond but without asserting that their way of responding must be right.3 Once the obsession with being right (Wong 2005: 98) is dispelled, one’s responses can be adjusted according to the context. Therefore, the oak tree acting as a shrine not only shows that it is free from being used, but also that it knows how to use being a shrine tree as a pretext. This, I suggest, is a crucial

3

One might say that they no longer respond with their ‘completed heart’ (cheng xin ᡀᗳ). For discussions of this idea, refer to Chen 2007: 325–331 and Chong 2011.

264

Skill in Ancient Ethics

difference between the oak tree and the unfortunate goose. Uselessness in the Zhuangzi is not idleness in the conventional sense, but neither is it ignoring the world. It involves an ability to find a lodging place in the ever-changing world. Another point that Major’s statement overlooks is that, if one’s horizon is broad enough, there is nothing that cannot be used. As Robins states, ‘nothing, perhaps, is completely useless to one who is sufficiently flexible’ (Robins 2011a: 105). Robins’ statement is based on a story in Chapter 1 of the Zhuangzi (2/1/35– 3/1/42) that shows Zhuangzi’s ingenious use of Huizi’s big gourds.4 Furthermore, in Chapter 2 of the Zhuangzi, which involves a play of perspectives, it is said that ‘there is nothing that is not “that” and nothing that is not “this” ’ (4/2/27), implying that the basis of our language and rules – naming – is indeterminate.5 Shortly after this expression the text adds that ‘there is nothing that is not so, there is nothing that is not affirmable. Thus, whether it be a stalk or a pillar, a hag or beautiful Xi Shi, things however peculiar or incongruous, dao pervades them and deems them one’ (4/2/34–35). If any peculiar or incongruous thing can be affirmed, then whether something is useful is determined not by the thing itself, but by its relationship with other things and the evaluating perspective of agents. If sages are those who can be attuned to the dao and thus deem things one, then they should be able to turn anything into something useful at a suitable time. The oak tree can then still be good for something, not totally useless. Zhuangzi’s message through the interplay of different stories about uselessness is that a person’s position in the world needs to constantly change even if one has only simple desires, such as avoiding harm. In this light, uselessness acknowledges the need to, or even the inevitability of, responding to the world’s transformation. Major is quite right in noting that ‘vegetative, unenlightened idleness . . . is not the goal sought in the Chuang-tzu . . . useless is not an accidental, passive quality, but something to be actively cultivated’ (Major 1975: 274). Perhaps Zhuangzi would question whether one can clearly distinguish activeness from passivity, given his emphasis on spontaneity (ziran 㠚❦).6 Nonetheless, uselessness is not simply a lack of utility, just as wuwei ❑⛪ is not simply a lack of action. This echoes the point that uselessness in the Zhuangzi is not idleness in the conventional sense. Here, we can say that ‘uselessness’ has a technical meaning in 4

5 6

This story also reveals that uselessness, conventionally conceived as an absence, failure or shortage, is not a guarantee of long life: while one can argue that the unfortunate goose is killed because its master thinks that it is still useful (as food), Huizi’s big gourds are smashed because Huizi thinks that they are useless. For discussions of Zhuangzi’s relativism and perspectivism, see Hansen 2003 and Connolly 2011. One reason for spontaneity casting doubt on the distinction between active and passive is that spontaneous acts are neither clearly intentional nor clearly unintentional (Chiu 2019: 398).

Skilfulness and Uselessness in the Zhuangzi

265

the Zhuangzi: that something is ‘useless’ implies that it is not restricted to, or entangled by, conventional use. Instead, it can constantly re-position itself to respond to the world in a timely manner. In other words, uselessness implies timeliness (shi ᱲ). How is the preceding discussion related to skilfulness? Because uselessness implies timeliness and uselessness needs to be cultivated, we can say that the process of cultivating uselessness is a process of skill cultivation, in the sense that one needs to acquire an art of navigating the world. Chapter 4 of the Zhuangzi records a song sung by a mad man, Jie Yu ᧕䕯, when he met Confucius and lamented his misplaced insistence on promoting virtue. The last few lines of that song read: Thistle, thistle, Don’t wound me as I walk! I turn around and go crooked, Don’t wound my feet!

I assume Jie Yu here is describing virtuous people, with Confucius as a typical example. He is saying that virtue (de ᗧ), as a form of interpersonal power, may incur danger when it is displayed, either intentionally or unintentionally, in a disordered time. This is not to suggest that possessing virtue is bad, but one needs to know how not to show it when the place and time are not right. Even for virtuous people, therefore, life is like walking in a land full of thistles. This is why one needs an art of navigating so as not to be entangled. If this is so, then uselessness means staying away from convention (avoiding thistles), but also recognizing and responding to convention (continuing to walk). In the following two sections I examine these two aspects in more detail.

3. Performance and non-attachment From a conventional perspective, it is natural to say that the skill of a master is useful for society, as mentioned in Section 1. However, one who has mastery of a skill may downplay, overturn or criticize conventional judgements of usefulness. In the story of Woodworker Qing, we are told that he prepares for at least seven days before carving a bell-stand. When others are astonished by his bell-stands because they look like the work of otherworldly creatures, he redirects their attention from the stands to the process of creating them. This arduous process, however, does not even guarantee that a product will be made, as Qing says that he will stop if it turns

266

Skill in Ancient Ethics

out that no suitable wood is found. When he prepares for work, he forgets praise, rewards, his role in society, his body, even his own skilfulness – all bases for evaluating whether something is useful temporarily vanish. In the story of Wheelwright Bian, he not only stresses that his skill cannot be put into words, but also comments that books, which are highly valued by his king, are overrated. He therefore questions the usefulness of conventional learning, including listening to teachers’ words or reading their texts. In the story of the forger (60/22/68–70), it is said that ‘something being put to use relies on something not being put to use’, implying that usefulness depends on uselessness.7 In the story of Cook Ding, while he states that his skill surpasses that of good cooks and mediocre ones, the difference between them is not evaluated according to their efficiency in carrying out the task. In fact, in contrast with the story of Woodworker Qing, the result of Ding’s work is not even mentioned. The king is fascinated by Ding’s elegant performance, which presumably is not the reason why he hires a cook in the first place. Distancing themselves from conventional judgements of usefulness, skill masters do not seek material success or social status. They may derive satisfaction from their skill, but they are not keen on showing it to others, unlike Zhao Wen, Shi Kuang and Huizi who display their skills to ‘enlighten’ others (5/2/44–45). Here we can see that Zhuangzi does not simply exalt skilfulness as such, if by ‘skilfulness’ we mean being well trained in a profession with a high level of knowhow. Being a supreme artisan, swimmer or cicada-catcher does not thereby bring one closer to Zhuangzi’s ideal personhood, represented by the sage (shengren 㚆Ӫ), the daemonic person (shenren ⾎Ӫ) or the authentic person (zhenren ⵏӪ).8 Skilfulness can bring harm to the self and others if one is inclined to preach the value of one’s profession or exchange it for profit or fame, becoming conceited and creating pressure. This is what Ding, Qing and other skill masters avoid. In this way, they ‘do not treat things as commodities’ (14/5/53). If one does not treat things as commodities, Zhuangzi asks, ‘what use has he for peddling’ (14/5/53)? Contrary to the general practice of other intellectuals of his time, Zhuangzi is not interested in offering advice or service to those in power (45/17/81–84). He does not ‘peddle’ his knowledge and is not used by society. As Møllgaard states, certain thinkers in the Warring States period, such as Xunzi and Hanfeizi, prioritize technical action over self-emergence (Møllgaard 2007: 36–38).9 They

7 8

9

This theme is further discussed in section 6. The term ‘sage’ is sometimes used pejoratively in the Outer Chapters (e.g. 23/9/14, 24/10/14), but has generally positive connotation in the Inner Chapters. For Møllgaard, technical action serves the human drive for completion, which is the urge to humanize the world and deny the inevitability of death (Møllgaard 2007: 15–17).

Skilfulness and Uselessness in the Zhuangzi

267

intend to use different techniques to guarantee the wealth and power of the state, but from Zhuangzi’s point of view this prioritization amounts to controlling life by what is created by life, and thus violating its spontaneity. In other words, when human life is used in human construction, entanglement arises. One reason Jie Yu’s song describes life as walking in a land full of thistles is that society, a product of human construction, can at certain times threaten the health, peace or even survival of human beings. Even a virtuous person such as Confucius can become entangled. Zhuangzi’s skill masters and Zhuangzi himself may not be sages, but they are similar in that they do not peddle their knowledge or skills.10 They cultivate their skill not because it is useful for an economic or political agenda but because they see skill as a spontaneous way of navigating one’s life in the world, to which we now turn.

4. Between attachment and detachment While skill masters do not become attached to conventional standards, neither do they remove themselves from the human world. Chapter 4 of the Zhuangzi, which bears the title of ‘The Human World’, tells us that ‘there are two things in the world to which one should show greatest caution. One is decree (ming ભ) and another is duty (yi 㗙)’ (10/4/39–40). ‘Decree’ here refers to political relationships. While Analects 2.21 says that political activities are not restricted to government affairs (Watson 2007: 22), the Zhuangzi acknowledges this observation but laments that it creates a burden on everyone: the scope of political power is often larger than expected, to the extent that ‘one cannot escape in heaven or on earth’ (10/4/41). Zhuangzi’s point certainly makes sense in a modern globalized society, where it is almost impossible to live without being influenced by local and international enterprises, media and government policies. Even in ancient times, Zhuangzi may very well agree with Xunzi’s statement that human beings are social animals (Hutton 2014: 76), and point out that this is precisely why social pressure is inevitable. Combining this with the point raised in Section 2, that one needs to acquire an art of navigating the world because it is full of thistles, we can see that Zhuangzi suggests finding a lodging 10

In saying that skill masters do not peddle, I do not mean that they do not earn livings as a skill practitioner. For example, Cook Ding and Wheelwright Bian work in the court, so one can suppose that they are paid for their work. Zhuangzi himself was once a minor official. Still, they have no intention to display their understanding and discourse of dao for fame or profit. In this regard they are unlike Hui Shi and Confucius, who travel around with ‘flapping lips and pounding tongue’ (80/29/12–13), that is, “selling” their doctrines.

268

Skill in Ancient Ethics

place between attachment and detachment. We can call the continuous search for such a place, never binding oneself to either side, as ‘wandering’ (you 䙺), recalling that in the story of unfortunate goose, Zhuangzi tells his disciples to ‘drift and wander by riding dao and de’ (cheng dao de er fu you ҈䚃ᗧ㘼⎞⑨). ‘Wandering’ does not simply refer to an easy and carefree attitude, as suggested by the translation of the title of the Zhuangzi’s opening chapter, ‘Xiaoyao You 䘽䚉䙺’ given by Watson (2013: 1) and Mair (1994: 3), and by Wu Kuang Ming’s comment that you is a self-satisfying movement congenial to the imagination (Wu 1990: 85). We must take into account Chen Guying’s saying that Zhuangzi’s ease is a pretext for his anguish (ji chen tong yu you xian ᇴ⊹Ⰻᯬ‫ݚ‬䯂), because he, like Jie Yu, laments that it is too easy for one to be used by power and end up enslaved (Chen 2007: 154, 345). Wandering, then, is carried out among life’s thistles. It is inevitable to assume certain conventional norms when one lives and acts, but as soon as one takes them too seriously one is ‘confined by things’ (66/24/34). The only way out is to wander, to constantly find elusive and subtle ways to get things done without being swayed by the corresponding norms. As Qing says, he works only when he forgets about conventional norms, like rewards or the order of the court. Skill masters are not only skilful at their profession, they are also skilful at wandering. To further illustrate the relationship between wandering and uselessness, let us examine the story of Ding in context. The story is placed after a short passage at the beginning of Chapter 3 (Yangsheng Zhu 伺⭏ѫ), which tells us a general life lesson, namely ‘going along the central vein’ (7/3/2) and avoiding doing good or bad. We can say that Ding’s avoidance of tendons and bones is a metaphor for one’s skilful navigation through life’s difficulties. His knife has been used for nineteen years but it remains sharp, as if unused. The blade ‘wanders’ (you ren 䙺࠳) through an ox’s inner space. Hence, one may say paradoxically that his knife is used in a way that keeps it from being used. Similarly, by detaching from conventional norms yet still fulfilling what they expect one to do, Ding exercises his skill without being used by the society. This is his wandering. Not surprisingly, scholars often note that Ding’s performance is a lesson for life. For example, Eno notes that Ding is similar to great athletes and artists in that ‘the actor experiences a sense of unmediated interaction with the environment in which the tasks are performed’ (Eno 1996: 136). Chen notes that Ding is inspiring us to have a humble, cautious, focused and artistic way of dealing with the world (Chen 2007: 170). Chong notes that Ding’s ability is a metaphor for finding a non-rigid and spontaneous way through life (Chong 2016: 59). What should be added to these comments is the connection between the story of Cook Ding and the story of

Skilfulness and Uselessness in the Zhuangzi

269

the oak tree. As mentioned in Section 2, the oak tree does not exclude itself from the human realm. By acting as a shrine, it is still engaging in society. Why not move to another place where it can cast off all social roles altogether? One may say, obviously, that trees cannot move. A more sophisticated answer might be that even if they could, when the human population increases or societies seek to explore more land for resources, they would still risk being used by humans. In other words, either due to their physical limitations or the expansion of human civilization, they will have to face humans someday as long as they live on the earth. Thus, they also need to wander in the human world, to find a way that can respond to human needs without being used. This is why the oak tree takes up the role of shrine. In line with the discussion above, Cook Ding also finds a way to deal with his social needs (earn a living, serve the king, etc.) without being used. He uses the ‘pretext’ by being a royal cook. That his knife and his spirit remain as if they are unused, despite Ding having to carry out his king’s orders, can be taken as a response to Jie Yu’s description of the danger in the world: finding rest and peace in a dangerous zone is not easy, but it can be done if one is skilful. It is said that the historical Zhuang Zhou was a minor official (Mair 1994: xxxi–xxxii). Like the oak tree and Cook Ding, he does not simply live as a recluse: he also makes a pretext. Developing Chong’s point that Ding’s way is non-rigid, it should be emphasized that wandering is not to be pursued by any fixed rule, thus the performance and pretext shown by skill masters is not to be taken as a model for imitation or memorization. As the story of the unfortunate goose states, ‘[situating oneself] in-between being good for something and good for nothing seems to be the point but it is not, so one cannot thereby avoid entanglement’ (51/20/5). This is not because ‘in-between’ is not the correct point, but because the very idea of correctness is put into question.11 Once one tries to specify the right way of being useless, one risks being entangled by the rightness and thus defies wandering. When reading the story of the goose, one is told that being useless like the goose will bring misfortune, and being useful like a normal tree will also bring misfortune. Those who believe that there is a fixed answer to the question ‘how should I position myself ’ may be tempted to think that one should be minimally useful in doing one’s job, such that one can fulfil basic social 11

The word for ‘correctness’ here is ‘shi’ ᱟ, which can mean ‘this’, ‘right’, ‘approve’. Therefore, Zhuangzi is saying that ‘in-between’ seems to the right answer, only to be later overridden by the idea of ‘drifting and wandering’. Commenting on Zhuangzi’s refusal to draw rigid distinctions, Graham says: ‘Not only does it depend on viewpoint which of alternatives you pick as “this”, it is a matter of convenience where you draw the line and whether you draw it at all’ (Graham 1989: 179). ‘Wandering’, therefore, implies a refusal of drawing a rigid line between ‘good for something’ and ‘good for nothing’.

270

Skill in Ancient Ethics

requirements without sacrificing oneself. Applying this rationale at work, one might try to do things up to the level that fulfils the job contract while avoiding doing anything beyond that, for overworking brings exhaustion. While not denying that this working style makes sense in certain situations, it is not necessarily free of entanglement. If one takes this working style as a fixed model, one compromises the ability to fine-tune one’s performance according to the particularity of each scenario. For instance, in a company, sometimes the boss forces employees to compete. As a result, doing nothing more than contractual duties may provoke the anger of the boss or other colleagues. A person may then respond by becoming more diligent, changing the boss’s mind, somehow bypassing the competition or leaving the company. They might either stop bothering about being minimally useful, or realize that being ‘minimally useful’ can take many different forms. Adopting a particular response such as leaving the company might be appropriate in one situation but this does not mean that it should be repeated in the future. This is what it means to ‘respond but do not retain’. As Connolly notes, skill stories remind us of ‘knowledge of particulars’ (Connolly 2019: 53) rather than principles and formulas. Presumably, Cook Ding is sensitive to the unique challenge of each ox he encounters. He says that he no longer sees a whole ox, which I take to imply that he can identify the exact spot on each ox for starting his work. The starting point and the proceeding path vary across different oxen, so it is not surprising that sometimes even he faces difficulties (8/3/10). Nevertheless, Ding still manages to pay attention to the particular structure of the present ox he encounters. Connolly also notes that ‘a person at a similar level of skill as Cook Ding could carve the ox in a different way’ (Connolly 2019: 53). This differentiates Zhuangzi from other pre-Han thinkers who emphasize that ideal action is not to be captured by rules.12 Although this is not explicitly mentioned in Chapter 3 of the Zhuangzi, a clue can be found by returning to the difference between Zhuangzi’s skill masters and skill practitioners like Zhao Wen, Shi Kuang and Huizi. Whether it is the ferryman who notes that his skill can be learned under certain conditions

12

That we need knowledge of particulars to handle a situation smoothly is perhaps not limited to Zhuangzi but can be found in other Chinese philosophical thought as well. Mencius, for example, claims that Confucius is a sage of timeliness (Lau 2003: 218–219). Eno even suggests that there is a connection ‘between Cook Ding’s mastery and the skilled virtue in Confucianism as timeliness’ (Eno 1996: 137). However, Zhuangzi is sceptical of the certainty and correctness of knowledge, regardless of whether it is of particulars. For him, two people facing the same situation can give very different responses without one being clearly ‘better’ than the other. I take this as implicit in the ‘pipes of heaven’ dialogue (3/2/1–9): the pipes of heaven are not particular sounds separated from other sounds; rather, every sound can be the pipes of heaven if played spontaneously.

Skilfulness and Uselessness in the Zhuangzi

271

(48/19/22–49/19/23) or other skill masters who remain silent about whether this is possible, all are reluctant to be role models. They allow others to be different, to heed their advice or to ignore it. Indeed, this is implied by the ability of wandering, because those who force others to imitate them become the very oppressive power they want to detach from in the first place.

5. Uselessness as the basis of usefulness To further elaborate on the difference between ‘wandering’ skill masters such as Cook Ding and ‘peddling’ skill practitioners like Zhao Wen, Shi Kuang and Huizi, consider a saying in Chapter 2 of the Zhuangzi: ‘that there are completion and deficiency is Zhao Wen’s playing the lute. That there are no completion and deficiency is Zhao Wen’s not playing the lute’ (5/2/43). Every performance generates some values, while omitting, bypassing or excluding others. Whatever we do, there are always many different ways to do it, including the very evaluation of these ways. As Fraser notes, Zhuangzi ‘recognizes the possibility of there being a plurality of different kinds of value that in some contexts cannot be jointly satisfied, but must be traded off against each other’ (Fraser 2009: 449). This may serve as a further reason for Connolly’s claim that it is possible for someone to carve an ox in a different way than Ding even though they possess a similar level of skill. Zhao Wen, however, does not seem to recognize this interdependence of completion and deficiency; he does not realize that no matter how skilful he is, he cannot exhaust all possible ways of playing the zither simultaneously. Worse still, he is keen to show that his own way of playing the lute is the correct one by ‘enlightening’ others. By insisting on one way of playing as the correct one, his skill in playing in that particular way may improve, but at the same time he is entangled by it. This is not just to say that one who regards themselves as the best will stop improving, but also that one who reinforces their rigid habits fails to appreciate alternative ways of actualising the potential of themselves and others. When one is at rest, one’s potential retains many possibilities for being played out. When one acts, only one such possibility is realized and the others are suspended. If one becomes obsessed with this one possibility, the other possibilities are not just suspended but suppressed, and the scope of the original potential shrinks. This shrinking of potential compromises the ability to fine-tune one’s performance according to the particularity of each scenario. To alleviate this problem, one is advised to pay attention not just to performance but also to potential, the collection of possibilities not currently

272

Skill in Ancient Ethics

actualized.13 Now, to turn something’s potential into actuality is to use it. Sages, however, can put anything into use when the time comes, as mentioned in Section 2. This implies that a sage’s potential retains all possibilities and does not shrink; it is like a reservoir from which one can keep ‘dipping from it but it never exhausts’ (5/2/61), which in turn gives the sage the ability to respond to rights and wrongs endlessly (ying wu qiong ៹❑マ, 4/2/31). This idea is implicit in one of Zhuangzi’s responses to Huizi’s criticism that his words are as useless as a great tree with a knobbly trunk and crooked branches. The response is that the great tree is good for someone who roams by its side and sleeps beneath it (3/1/47); in other words, it is for playing and resting. Here, the great tree is not even acting as a shrine, being planted in the ‘country of Nothingwhatever’ (wu he you zhi xiang ❑օᴹѻ䜹). This does not mean that uselessness is simply a void. As we saw in previous sections, uselessness involves responding to the world. Every response involves turning potential into actuality. That the sages’ potential does not shrink implies that they can respond to the world as if they were at rest. Given this formulation, we can regard the relationship between the tree planted in the country of Nothingwhatever and the oak tree acting as a shrine as an analogy for the relationship between potentiality and actuality. The oak tree’s acting as a shrine is putting itself to use, but in doing so it does not become attached to this use. Because this non-attachment is not part of the shrine’s current function, judged from a conventional perspective it is ‘nothing whatsoever’, a useless feature. However, this non-attachment is crucial in preserving one’s potential. One’s potential qua potential is useless, but it is the basis of actuality. Shrinking the scope of one’s potential means being restricted and eventually hampering one’s actuality, or fails to appreciate other people’s different paths to actualization and thus damages their actuality. Reading the skill stories in this light, Møllgaard says that Cook Ding’s knife ‘cuts with its pure potential to not cut, which, in so far as it can be actualized, can never be dulled’ (Møllgaard 2007: 55).14 Ding’s knife represents pure potential in actuality. The swimmer, who follows the dao of water without imposing himself on it, says that he has no dao of

13

14

On the one hand, the independence between completion and deficiency is not a problem to be solved because, as Fraser states, ‘This is an inevitable consequence of thought and action, not a fault or weakness that could be rectified’ (Fraser 2009: 449). On the other hand, those who ignore this interdependence often fall into obsessions, which blinds them to new situations, methods and perspectives. Here, one may wonder whether skill masters themselves are sages in Zhuangzi’s sense, or whether one can become a sage in real life. Nevertheless, even if answers to these two questions are negative, wandering skill masters are still closer to sages than peddling skill practitioners.

Skilfulness and Uselessness in the Zhuangzi

273

swimming.15 Having no dao of his own, however, means he is able to adapt to any change in the force and direction of the turbulence. Woodworker Qing, as mentioned in Section 3, goes to work only after he sets aside all considerations of usefulness. The naked draughtsman, before displaying any skill, is said to be a real draughtsman (56/41/27). In each case there is emphasis on that which is not yet actualized or that which actualizes without affirming any particular way of actualization. Skill masters use themselves and other things they encounter in a way that preserves the potential of both as much as possible, to make themselves and the things they encounter useful in a way that does not remove their uselessness. This, rather than their actual performance, is what differentiates them from peddling skill practitioners.

6. Uselessness and politics Given Zhuangzi’s non-attachment to politics, it is unsurprising that the story of Cook Ding is sometimes interpreted as hinting at a method for surviving in the dangerous political arena while still doing one’s job (Eno 1996: 137), or for how to be a successful minister with a cautious attitude (Connolly 2019: 57). Stories about uselessness, needless to say, are explicitly about liberation from the political realm, and if skilful performances and uselessness are related, it seems natural to say that uselessness is a technique for survival (Major 1975: 275). While avoiding premature death and skilfully dealing with one’s social and political duties are certainly important effects of being useless, it remains a question whether the theme of uselessness has anything more than that to do with government and society as a whole. Building on the discussion above, I believe we can construct two points about Zhuangzi’s view on politics. The first point is that Zhuangzi not only detaches from but also criticizes political ideologies and practices that rely on rigid standards and models. Whereas practitioners who peddle their skills are firmly attached to conventions, wandering skill masters defy the norms set by the ruling class and social elites. Lai Hsi-San, for example, offers an interpretation of the story of Cook Ding that takes it as an example of subversion and a critique of power (Lai 2013: 28–33). This critique can also be found in other skill stories. Sometimes it is implicit, as 15

This is different from Ding, who says he is fond of dao. For a discussion of their difference, see Robins 2011b: 108–112.

274

Skill in Ancient Ethics

in Woodworker Qing saying that when he works he has no thought of the court, or the swimmer’s casual walking in front of Confucius and his disciples, satirizing the social etiquette.16 However, the critique can also be explicit, as in Wheelwright Bian’s blunt comment that one who reads sage-kings’ writings gets in touch with only dregs of the ancients. Given that those writings symbolize the basis of political authority, Bian’s comment can be read as questioning whether authority can be grasped by taking sage-kings, or any particular conception of sagehood, as models. An obsession for modelling oneself on sages moves one further away from sagehood and creates conflict and suppression, as displayed immediately by the threat of death made by Bian’s king when he hears Bian’s comment. Bian’s bluntness and provocation seem to contrast with passages that recommend avoiding direct remonstration (8/4/4–9/4/14, 10/4/56–11/4/64), and if we remind ourselves of Jie Yu’s song, Bian appears to step on thistles rather than travelling inbetween them. Nevertheless, Zhuangzi himself also displays this bluntness towards the king of Wei (53/20/45–50), so it is not that he wants all people to avoid direct conflict at all costs. To advise identifying with superiors regardless of the circumstances would be suppressive. Sometimes the most appropriate way to proceed may be to remove the thistles, to smash ideologies and throw off models (24/10/15–26). Direct confrontation is not a universal rule, but neither is conformism. Stories about uselessness can also be read under the theme of political criticism. In fact, when Huizi comments that Zhuangzi’s words are ‘useless’ (3/1/44, 74/26/31), he may intend to convey something more specific than merely ‘impractical’: Huizi is situated in the central place of government as a prime minister (45/17/84), and the term ‘use’ has the specific sense of being employed by the government. Confucius in the Analects, for example, sometimes seeks to be ‘used’ in this sense (Watson 2007: 90, 121). Thus, Huizi is saying that Zhuangzi’s words cannot be used by social authorities. This, of course, is welcomed by Zhuangzi. Sima Qian, author of the Records of the Grand Historian, notes that the words of the historical Zhuang Zhou are ‘shimmering, billowing and indulging to please himself ’, such that ‘kings, dukes and great men could not utilize him’ (Mair 1994: xxxii). The term ‘utilize’ (qi ಘ) can also mean ‘use as a utensil’, as the Analects 2.12 indicates that gentlemen are not utensils (Watson 2007: 21). Arguably, utensils are made for a particular use (e.g., a knife is made 16

As said in the Mencius, ‘One who walks slowly, keeping behind his elders, is considered a well-mannered younger brother. One who walks quickly, overtaking his elders, is considered an ill-mannered younger brother. Walking slowly is surely not beyond the ability of any man. It is simply a matter of his not taking the effort’ (Lau 2003: 264–265).

Skilfulness and Uselessness in the Zhuangzi

275

for cutting), so the Analects is saying that gentlemen should not regard themselves as useful only for certain tasks or professions. Instead, they should strive to make the dao prevail in all under heaven, in all aspects of life. Zhuangzi welcomes the idea of not being used as a utensil, but at the same time he questions whether Confucians can be free from being used or from using others. If they cannot reform society, their teachings may end up helping robbers and tyrants to decorate themselves (23/10/1–24/10/14). Even if they were to reform society, their new social order will rely on a particular conception of correctness and hierarchy and thus generate antagonism with other schools (4/2/26, 60/22/79– 80, 66/24/44–45). They may then seek to convert people to take up their conception of correctness and, intentionally or unintentionally, marginalize those who do not wish to do so – people who remain ‘useless’. Being useless, then, is not just mocking people who seek power like Huizi, but also a critique of Confucianism based on an idea agreed on by Confucians: that a prevalence of dao requires one not to be restricted. Hence, this is not merely a retreat from the outer realm to the inner, as Wang Bo suggests (Wang 2004: 39–44). Like other pre-Han thinkers, Zhuangzi sees the prevalence of dao as not confined to the personal realm, but rather as involving a transformation of the social order. Taking the transformation of social order into consideration leads us to the second point about how uselessness is related to politics. Uselessness is the basis of a form of governance that does not create entanglement. Zhuangzi’s suggestion of this form of governance again shows that he is not a conformist. Van Norden interprets Zhuangzi’s response to Huizi’s comment that his words are ‘useless’, along with the story in which Yan Hui is going to meet the king in Wei (8/4/1– 9/4/34), as advice to avoid confrontation with authorities and the status quo when political engagement is unavoidable (Van Norden 2016: 13). However, judging from the stories of skill masters and references to uselessness, far from claiming that the status quo cannot or should not be changed, Zhuangzi shows us how power can be exercised skilfully and without subjugation.17 Such an exercise of power has two features. The first feature is that the functioning of power depends on those who are not currently holding power. In his rebuttal to Huizi’s criticism, Zhuangzi says that although one needs only a tiny space in the earth for use, if all other space is dug away then it is questionable whether that 17

Even the story of Yan Hui mentioned by Van Norden does not fit well with the advice to ‘avoid confrontation with authorities and the status quo’. In that story, Confucius does not stipulate whether Yan Hui should go to Wei, he only tells Yan Hui to be prepared. If Yan Hui succeeds in his mission, the king will be transformed, and so will the status quo.

276

Skill in Ancient Ethics

tiny space will still be useful (74/26/32–33). The message is that usefulness depends on uselessness. To elaborate, recall that the term ‘use’ has a specific political sense. We can say that the power of government depends on its unimpeded connection with all other social sectors, as well as its acceptance that people in these other sectors have the potential to become the leading figure. As Zhuangzi says, ‘although one may be in the relationship of ruler and minister, this is only for a particular moment of time. When the age change, they have no means to look down on each other’ (74/26/35). A butcher can become a teacher for the king; a ferryman may inspire Confucius and his disciples; and a woodworker may have the ability to create products that are comparable to those made by otherworldly creatures. If the government looks down on these people and regards them only as utensils, or if it treats people who criticize the government as useless, then it is as if it is digging away all places except the throne. Thus, it is by not turning other sectors into utensils that one can preserve the potentiality of society, and by not attaching power to a particular sector that one can preserve power. Just as a person who can preserve their potential when acting knows how to wander, a society that does not insist on putting one sector over another is a society that allows people to wander. Even the leader only ‘wanders in non-existence’ (you yu wu you 䙺ᯬ❑ᴹ, 20/7/15). The second feature of power complements the first: the power that can accommodate all can also make use of all. This echoes the forger’s saying that ‘something being put to use relies on something not being put to use’. I have argued elsewhere that the story of the forger hints that ideal governance involves non-interference and sets up standards only for particular situations rather than using them as formulas (Chiu 2019: 272–273). Even Confucian rule by virtue is regarded as akin to ‘the grass huts of the former kings in which one can only stay one night but not settle in them for long’ (38/14/50–51). One can rule by virtue as long as virtues are not fossilized. When they are not fossilized, they take no determinate form and do not depend on rules. In other words, governance at a deep level remains an art rather than a set of rules. This does not imply that there should be no rules. Rather, it means that the use of rules is also an art, which involves skills that cannot be transparently transmitted, as Wheelwright Bian points out. While the Analects states that politics is about ‘rectifying’ or ‘correcting’ names (zheng ming ↓਽) (Watson 2007: 88), Zhuangzi, in the same passage that regards virtue as akin to grass huts, states that the ‘utensils of rectification’ (zheng zhi qi ↓ѻಘ) – resentment and kindness, taking and giving, admonishing and instructing, sparing and killing – can only be put into use by people who can ‘comply with the greatest change without being clogged’

Skilfulness and Uselessness in the Zhuangzi

277

(38/14/54–55).18 These people are sages, for whom all strategies and policies of governance are put into use by acknowledging their possibility of being modified, suspended or withdrawn. As in Zhuangzi’s use of the big gourd, uselessness can be turned to usefulness; as in Ding’s use of his knife, usefulness can come from uselessness. This interplay of use and non-use has no pre-determined, let alone correct, pattern. Different social sectors, strategies and things can ‘take turns as each other’s lord and ministers’ (4/2/17). Combining the two features of Zhuangzi’s ideal governance, we can say that it is of a form that preserves society’s potentiality. It extends the theme of balancing attachment and detachment, and sees use as coming from non-use. When one does not insist on taking anyone or anything (or nothing, even, if this is meant as firm resistance to carrying out any action) as a ruler, one simultaneously does not insist on taking anyone or anything as the ruled. This non-insistence leaves room for everyone to find their lodging places, which means allowing them to be useless. In this way, enlightened rulers ‘have their transforming influence extend to ten thousand things but the people do not rely on them’ (20/7/15).

7. Conclusion Zhuangzi’s thought is expressed through self-criticism. When the conventional obsession with making oneself useful is criticized by a story about big, useless tree, this criticism is in turn challenged by pointing out that uselessness may bring one great trouble in a social network. However, instead of retreating to the conventional, Zhuangzi complements uselessness with timeliness and skill. For Zhuangzi, there is no contradiction between being skilful and being useless. In accomplishing a particular task, one turns potentiality to actuality. If the accomplishment does not shrink the potential, one can be said to use something while leaving it unused, like Ding’s knife. In navigating life, finding a way in and through the changes in the world without being entangled requires the skill of wandering. By comparing wandering skill masters such as Cook Ding and the swimmer, to peddling skill practitioners, such as Zhao Wen and Huizi, Zhuangzi hints that they are separated not by what they accomplish, but by how they

18

This passage appears in Chapter 14 of the Zhuangzi, which is often associated with the Huang-Lao School rather than being regarded as consistent with the philosophy of the Inner Chapters (Liu 1988: 299–304). The Huang-Lao School may justify the use of these ‘utensils of rectification’ differently. Nevertheless, I believe the passage can still be read together with the theme of uselessness and consequently treated as consistent with the Inner Chapters.

278

Skill in Ancient Ethics

accomplish. Speaking playfully, but with a clear realization that the world is full of thistles, stories about skill masters and uselessness echo Zhuangzi’s view on politics, which is not only a matter of the survival and non-attachment of individuals, but also of a form of governance that uses but does not insist on rules, and that puts society in order without ordering it so. Ultimately, people who can handle politics skilfully are like the craftsman Ch’ui, who can ‘draft swiftly in freehand better than when using a compass or an L-square’ (50/19/62).

References Chen, G.Y. 䲣啃៹ ([1990] 2001), Contemporary Notes and Translations of the Zhuangzi (㦺ᆀӺ⌘Ӻ䆟), Hong Kong: Zhonghua Shuju. Chen, G.Y. 䲣啃៹ (2007), New Discourse on Lao-Zhuang (㘱㦺ᯠ䄆), Taipei: Wu-nan Books. Chiu, W. W. (2019), ‘Spontaneity, Perspectivism and Anti-intellectualism in the Zhuangzi’, Dao: A Journal of Comparative Philosophy 18 (3): 393–409. Chiu, W.W. (2019), ‘The Forger: The Use of Things’, in K. Lai and W. W. Chiu (eds), Skill and Mastery: Philosophical Stories from the Zhuangzi, 259–277, London: Rowman & Littlefield International. Chong, K.C. (2011), ‘Zhuangzi’s Cheng Xin and its Implications for Virtue and Perspectives’, Dao: A Journal of Comparative Philosophy 10 (4): 427–443. Chong, K.C. (2016), Zhuangzi’s Critique of the Confucians: Blinded by the Human, Albany : State University of New York Press. Connolly, T. (2011), ‘Perspectivism as a Way of Knowing in the Zhuangzi’, Dao: A Journal of Comparative Philosophy 10 (4): 487–505. Connolly, T. (2019), ‘Zhuangzi’s Politics from the Perspective of Skill’, in K. Lai and W.W. Chiu (eds), Skill and Mastery: Philosophical Stories from the Zhuangzi, 45–59, London: Rowman & Littlefield International. Eno, R. (1996), ‘Cook Ding’s Dao and the Limits of Philosophy’, in P. Kjellberg and P. J. Ivanhoe (eds), Essays on Skepticism, Relativism and Ethics in the Zhuangzi, 127–151, Albany : State University of New York Press. Fraser, C. (2009), ‘Skepticism and Value in the Zhuāngzi’, International Philosophical Quarterly, 49 (4): 439–457. Guo, Q.F. 䜝ឦ㰙. ([1961] 2004), Collective Commentaries on the Zhuangzi (㦺ᆀ䳶䟻), Beijing: Zhonghua Shuju. Graham, A.C. ([1981] 2001), Chuang-Tzu: The Seven Inner Chapters and Other Writings from the Book Chuang-tzu, London: George Allen and Unwin Publishers Ltd. Graham, A.C. (1989), Disputers of the Tao, Indianapolis: Open Court.

Skilfulness and Uselessness in the Zhuangzi

279

Hansen, C. (2003), ‘Guru or Skeptic? Relativistic Skepticism in the Zhuangzi’, in S. Cook (ed), Hiding the World in the World: Uneven Discourses on the Zhuangzi, 128–162, Albany : State University of New York Press. Hong, Y. ⍚ᾝ, ed (1986), A Concordance to the Zhuangzi (㦺ᆀᕅᗇ), Shanghai: Shanghai Classics Publishing House. Hutton, E.L., trans. (2014), Xunzi: The Complete Text, Princeton: Princeton University Press. Lai, H.S. 䌤䥛й (2013), On Taoist Intellectuals: The Critique of Power and Cultural Revitalization in Zhuangzi (䚃ᇦරⲴ⸕䆈࠶ᆀ䄆˖㦺ᆀⲴ℺ᢩ࣋ࡔ㠷᮷ॆᴤᯠ), Taipei: National Taiwan University Press. Lau, D.C., trans. ([1979] 2003), Mencius, Hong Kong: Chinese University Press. Liu, X.G. ࢹㅁᮒ (1988), The Philosophy of Zhuangzi and Its Transformation (㦺ᆀଢᆨ ৺ަ╄䆺), Beijing: China Social Science Press. Klein, E. (2010), ‘Were the “Inner Chapters” in the Warring States? A New Examination of Evidence about the Zhuangzi’, T’oung Pao 96 (4): 299–369. Mair, V.H. (1994), Wandering on the Way: Early Taoist Tales and Parables of Chuang Tzu, New York: Bantam Books. Major, J.S. (1975), ‘The Efficacy of Uselessness: A Chuang-tzu Motif ’, Philosophy East and West 25 (3): 265–279. Møllgaard, E. (2007), An Introduction to Daoist Thought: Action, Language, and Ethics in Zhuangzi, Abingdon: Routledge. Robins, D. (2011a), ‘The Theme of Uselessness in the Jwāngdž’, Warring States Papers 2: 105–108. Robins, D. (2011b), ‘It Goes Beyond Skill’, in C. Fraser, D. Robins and T. O’Leary (eds), Ethics in Early China: An Anthology, 105–123, Hong Kong: Hong Kong University Press. Schwitzgebel, E. (2019), ‘The Unskilled Zhuangzi: Big and Useless and Not So Good at Catching Rats’, in K. Lai and W.W. Chiu (eds), Skill and Mastery: Philosophical Stories from the Zhuangzi, 101–108, London: Rowman & Littlefield International. Wang, B. ⦻ঊ (2004), The Philosophy of Zhuangzi (㦺ᆀଢᆨ), Beijing: Peking University Press. Watson, B., trans. (2007), The Analects of Confucius, New York: Columbia University Press. Watson, B., trans. (2013), The Complete Works of Zhuangzi, New York: Columbia University Press. Wong, D. (2005), ‘Zhuangzi and the Obsession with Being Right’, History of Philosophy Quarterly 22 (2): 91–107. Wu, K.M. (1990), The Butterfly as Companion: Meditations on the First Three Chapters of the Chuang Tzu, Albany : State University of New York Press.

280

14

Dao and Agency: What do the Zhuangzi’s Skill Stories Tell Us about Life? Karyn Lai

The Zhuangzi showcases stories in which skilful and satisfying achievements are possible in a range of activities, ordinary activities such as making wheels, sailing a boat, and butchering.1 Yet, the distinguished audiences in some of these stories are no less captivated by the displays of skill. How can we use these stories to enrich our reflections on living well? The plurality of models offered in the Zhuangzi stand in contrast to the singular and elevated paradigms offered in other traditions, such as the Confucian junzi (ੋᆀ; exemplary person). The Confucian account of authoritative life offers a picture of life embodied in a defined set of ideals, for one and all, albeit with different levels of achievement articulated within its scheme. But, in the Zhuangzi, no one story can show us how best to live. For a start, we cannot all be expert bell-stand makers, swimmers, scribes and cicada-catchers, and more. Moreover, we are told in some of the stories that these levels of skill take years to cultivate. And life is just too short for us to become experts in all those fields. How can we fruitfully learn from these stories? My aim is to explore what some of the Zhuangzi’s skill stories may reveal about the idea of agency, especially in light of how to live well.2 In doing so, I 1

2

What counts as a ‘skill story’? We typically include those that discuss expert or elegant performances – such as the bell-stand maker. Yet, it is less clear that other stories, such as the trainer of fighting cocks, would count. The butcher story is the only one in the ‘Inner Chapters’ section of the text (Zhuangzi 3). Those commonly included as skill stories in other sections of the Zhuangzi include: Wheelwright Bian (13), the cicada-catcher (19), the ferryman (19); the master who trained the cocks for cock-fighting (19), the swimmer (19), Engraver Qing (19), the scribe (20) and the buckle-forger (22). Contested skill stories include: robbery (10); the shooting of not shooting (20), using an axe to cut away mud on the rose (24), and the skill of dragon-slaying (32). I thank Wai Wai Chiu for sharing this assemblage of the stories. An investigation of the skill stories and their philosophical insights is available in Lai and Chiu 2019. My approach takes these skill stories not simply as prescriptions for readers. Rather, I engage with them as examples. This interpretive strategy, whereby we take the text’s words not at face value but rather as paradigms of how we engage, is not unusual. See, for example, Jean François Billeter’s comments on how the text requires readers to reflect on its words to enlighten their understanding. The Zhuangzi is not to be read as a book of instructions on how to handle affairs in any particular way (2011: 3–9).

281

282

Skill in Ancient Ethics

investigate, in particular, how the stories portray skilful performances and how the performances may relate to the term ‘dao’. I propose that the Zhuangzi offers a way of understanding ‘daos’ qua exemplary ways of encountering and engaging with the world, and that we find some of these insights especially in the stories. Is there evidence of this? In the stories, we read about skilful performances, how skill is learnt, as well as the context of the task, a person’s orientation to the task, and his absorption in the activity. I suggest that these stories offer vignettes and narratives that help expand our vocabulary and enrich our imagination on the many possible ways in which life can be (more) rewarding. Rather than focusing on norms and targets as primary considerations, I believe that the Zhuangzi prompts readers to consider possible ways of doing something well to enhance how we live. This interpretation of how these stories illuminate our thinking about a life lived well, is fascinating and promising not least because it reveals an aspect of agency embedded in the text. In the Introduction that follows, I set out the various layers of argument that lend support to this proposal.

Introduction, and some remarks on ‘skill’ in the Zhuangzi The Zhuangzi’s skill stories are both fascinating and bewildering. Not only do the masters’ actions, products or the impacts of their actions manifest a level of achievement that exceed our understanding of the reach of human capabilities but, remarkably, many of the stories speak of the seeming ease or equanimity with which the masters execute these tasks. Even an initial encounter with these stories would prompt readers to ask about the place of their themes in the Zhuangzi. On the other hand, familiarity with the stories does not make it easier to understand their point (if there is one). There are so many possible focal points. We may attend to the products of mastery, such as the marvellous bell-stands, or to the impact of mastery, as exemplified in the scribe’s nonconformism with courtly etiquette. Alternatively, we may focus on the expert execution of a task, for example, in how the wheelwright feels with his hands and his heart-mind (xin ᗳ). There are also unique dispositions exercised by the masters, including: attentiveness, such as that of the cicada-catcher, who both sequesters his attention on everything in his environment and concentrates only on cicada wings; nonresponsiveness, in the fighting cocks that were so still they frightened their opponents away; and the lack of anxiety of the boatman in helming his boat in rough waters.

What do the Zhuangzi’s Skill Stories Tell Us about Life?

283

Is there a thread running through the stories? Or are they speaking to different audiences on different matters? Either way, how can such plurality be actionguiding? The approach I take here is to look across the stories to investigate how relevant notions of skill reveal facets of agency and of a life lived well. This by no means assumes that the text holds a coherent and unified view. Indeed, as my analysis below will show, there are quite marked differences in the way some of the masters respond to the question whether they have (a) dao. Rather than dwell on questions of textual coherence, I employ the nuances afforded by textual incongruities. I seek to demonstrate that the different stories, taken together, suggest there is no single dao. I propose that the stories introduce possibilities of doing things well, even expertly. On this reading, the stories are examples of how readers, in their own undertakings, may be appropriately responsive to situations. And, instead of channelling us all toward a single, human goal, the stories should stimulate our imagination about a variety of ways of achieving excellence, given our individual circumstances. This, I contend, is how the text conveys its notion of agential initiative. To support this position, the second section considers the connections between ‘skill’ and dao in a number of the stories, revealing some ways in which the relation between these notions may be conceived. The third section takes the discussion further by examining how some of the masters respond to the question, ‘Do you have dao?’ As we will see, there are both affirmative and negative answers to the question. I take these differences to indicate that these stories offer personalized accounts of mastery. Finally, the fourth section brings together the discussions of dao and skill within the context of the skill stories. It proposes that the varying conceptions of dao and the different models of skill in the stories provide a fascinating array of options for a good life. The models are distinctive because each master is actively engaged in developing the skills he needs for undertaking a particular activity well. Active engagement in what one learns, and how one learns is, I suggest, the Zhuangzi’s distinctive contribution to a notion of agency in classical Chinese thought. Before I proceed to the second section, I articulate two related caveats concerning the use of the term ‘skill’ in relation to the Zhuangzi. First, what does ‘skill’ mean in the text? The common translation of a cluster of terms in the Zhuangzi’s stories – qiao ᐗ, ji ᢰ, and shu 㺃 – as ‘skill’ or ‘technique’ raises many questions. Does each pick out subtly different aspects of skill? Do they distinguish skill from ability? And do they work in tandem with relevant dispositions? Here, I offer a brief overview of the three terms:

284

1.

2.

3.

Skill in Ancient Ethics

Qiao (ᐗ) is associated with the character for ‘work’ (gong ᐕ), which has multiple meanings associated with labour and production, arising from its different contexts of use.3 Ji (ᢰ) obtains its meaning partly through its left component, the hand (shou ᡻), and partly from its right component (zhi ᭟), which may refer to the demanding task of removing branches from the main stalk of the bamboo.4 If we follow this gloss in the Shuowen, adeptness is required especially as bamboo stalks often have sharp, fine hairs that can cause severe reactions. Shu (㺃) takes its meaning partly from the character xing (㹼), meaning to walk or proceed. The Shuowen notes its association with dao (䚃), suggesting, in the first instance, a literal reading as ‘path within the city’.5 Alternatively, because dao may also mean ‘teaching’ or ‘doctrine’, shu may also refer to a person’s adherence to the guidance provided by the teaching.

There are important differences and inflections across the three characters. However, they seem to share the following features: (1) a method of engagement, which (2) involves some kind of manipulation, and which (3) is manifest in performance. Can we look to the Zhuangzi for clearer and more defined uses of these terms? I suggest not; and this is the second, related, caveat. Each term cuts across pre-Han doctrinal debates, and even a most thorough exploration of their occurrences in extant texts would not provide clear answers to the questions above, given that the terms sometimes were used interchangeably, while also acquiring meanings from specific contexts of use. Moreover, within the Zhuangzi, there are many angles on these skill-related terms based on their place in the text’s complex, varied and often-cryptic imagery and stories. For example, are references to skill related to the text’s criticism of official life?

3

4

5

In the Gu wenzi gulin (ਔ᮷ᆇ䁱᷇) – see Li Pu in the references – the character ᐕ, as it appears in ancient texts, has numerous associations, including with the use of the carpenter’s square, jadepolishing, wood and battle axes, and a beam for building walls (Vol. 4, 742–754). Summarily, there is a cluster of context-related meanings that focus on performance and production. (I thank Wai Wai Chiu for drawing my attention to these aspects of ᐕ.) The Shuowen – see Xu Shen in the references – associates qiao with ji (ᢰ; Shuowen, ধ‫ޝ‬, ᐕ䜘, 3014). It also associates ᐕ with the use of guiju, compass and square (ᐕ˖ᐗ伮ҏDŽ䊑Ӫᴹ㾿Ὲ ҏDŽ㠷ᐛ਼᜿DŽࠑᐕѻኜⲶӾᐕDŽ (Shuowen, ধ‫ޝ‬, ᐕ䜘, 3012). The phrase ‘guiji’ refers to two measuring instruments, compass and square. In pre-Qin discussions, it can be used metaphorically to refer to the application of measuring standards in other areas of life. ᢰ˖ᐗҏDŽӾ᡻᭟㚢DŽ(Shuowen, ধॱй᡻䜘, 7982). ᭟˖৫ㄩѻ᷍ҏDŽӾ᡻ᤱॺㄩDŽࠑ᭟ ѻኜⲶӾ᭟DŽ (Shuowen, ধഋ᭟䜘, 1925). 㺃˖䛁ѝ䚃ҏDŽӾ㹼ᵟ㚢DŽ (Shuowen, ধй 㹼䜘, 1266). Shu may have the connotation of a profession (and a purpose), marked by a person’s paths within the city. Here, again, I am grateful to Wai Wai Chiu for discussing with me the scope of 㺃.

What do the Zhuangzi’s Skill Stories Tell Us about Life?

285

Should we interpret the emphasis on performance in light of the Zhuangzi’s attention to ‘wordless teaching’? And so forth. My discussion does not aim to present a comprehensive account of skill across the texts, or even in the Zhuangzi. Rather, it is to demonstrate the complexity of the terms of reference, as we will see below.

1. Dao and ‘skill’ in the stories This section suggests three ways to understand the connection between dao and ‘skill’ in the stories that mention both. I begin with some preliminary remarks on the conception of dao generally in pre-Han usage, to help contextualize the discussion. The various connotations of the term dao 䚃 derive from both its components, chuo (辶) and shou (꯱俆) (Shuowen ধй, 䗥䜘, 1204). Chuo 辶means ‘to walk’ while shou 俆 refers to ‘leading’ (as in ‘first’). Facets of its meaning in preHan texts include the following: 1.

2.

3.

Dao as ‘path’, referring literally to a road. Metaphorically it may refer to a doctrine or teaching that provides guidance (in the same way a path does) for conduct, actions, or life more generally.6 Dao as a destination or goal. In the texts aligned with the Daoist tradition, in particular, dao may refer to the attainment of certain capabilities, including that of detachment. It may also refer to (having) unique insights, or unparalleled equanimity, or an idealized state of being. The end, and attaining it, is the emphasis here. Dao may refer to the activity of walking or travelling. Metaphorically, it may be used in contradistinction to the rigidity of a strictly-prescribed path, or to the platitudinous nature of a pre-determined state.7

These three senses of dao comprise a selective summary overview of its possible meanings that will help frame my discussion. In each case, whether as guidance, goal or activity, dao may be recommended with greater or lesser stringency, allowing individuals more or less discretion in their undertakings.

6

7

Examples of the metaphorical use include ‘humanity’s dao’ (ren dao Ӫ䚃; see e.g. 11/72–74; 12/41– 45), or ‘Confucian doctrine’ (Rudao; ݂䚃). The character you (䙺), to wander, appears in the title of the Zhuangzi’s opening chapter (‘Uncharted wanderings’ ‘Xiaoyaoyou’ 䘽䚉䙺); it speaks to the importance of a person’s encounters and responses to the world.

286

Skill in Ancient Ethics

To investigate the connection between skill and dao, I examine two stories that link dao specifically to qiao (discussed above and often translated as ‘skill’). In the cicada-catcher (19/17–21) and the buckle-forger (22/68–70) stories, qiao and dao are paired so as to suggest some connection between them. The stories ask two questions in quick succession: ‘Does the master have skill? Is there dao?’ In each, the interlocutor seeks to understand how the master’s action or handiwork is exceptional. Below are the relevant sections of the two stories. When Confucius was on his way to Chu, he passed through a forest where he saw a hunchback catching cicadas with a sticky pole as easily as though he were grabbing them with his hand. Confucius said, ‘Does the master have skill? Is there dao?’ (zi qiao hu? You dao xie? ᆀᐗѾ˛ᴹ䚃䛚˛) 19/17–18. Trans. Lai, adapted from Watson 1968: 1998 The Grand Marshal’s buckle maker was eighty years old, yet he had not lost as much as a hair in his adroitness [for making buckles]. The Grand Marshal asked, ‘Does the master have skill? Is there dao?’ (zi ziao yu: you dao yu? ᆀᐗ㠷˛ ᴹ䚃㠷˛) 22/68. Trans. Lai, adapted from Watson 1968: 244.

There are versions of these stories in two other pre-Han texts, the Liezi’s cicadacatcher,9 and the Huainanzi’s buckle-forger.10 In all versions, the first question appears rhetorical; it seems not to be eliciting information concerning the master’s skill but rather to be expressing amazement at that level of skill. Second, the Liezi’s cicada-catcher story is expressed with an exclamation mark rather than a question mark (zi qiao fu! ᆀᐗѾ!). Third, in all versions the first question begins with the character zi ᆀ – ‘Master’ – in contrast to the second, which begins with ‘Is there . . . ?’ It seems that the interlocutor is so fascinated by what he sees that the two questions are posed in one breath, without waiting for a response to the first. Without stopping for an answer (if it was ever a question), Confucius and the Minister move immediately to pose their actual question, ‘Is there (a) dao?’, or ‘Does the master have dao?’. Following the discussion above, I suggest a reading of the two questions as ‘What skill you have! Is there dao?’ In what follows, I propose three broad ways to think about the connection between the terms qiao and dao.

8

9 10

Note that my preferred translation of ‘ᆀᐗѾ˛ᴹ䚃䛚˛’ is ‘What skill you have! Is there dao?’. The translation here presents a more neutral, literal translation of the text to begin the discussion. Below, I present arguments for my preferred translation. In the ‘Huang Di’ chapter. The Liezi is cited by the name of its compiler, Zhang Zhan. In the ‘Dao ying xun’ chapter. The Huananzi is cited by the name of its author, Liu An.

What do the Zhuangzi’s Skill Stories Tell Us about Life?

287

The first view sees no connection between qiao and dao. On this view, the focus is on how each of qiao and dao are learnt and manifest, and measured. I believe this is unlikely given how the questions are delivered together. I will provide more reasons below. The second position sees a connection between qiao and dao, and views qiao (the first question) as the focus of this pair of questions. This means that the second question follows up on the first. On this reading, the interlocutor is interested primarily in learning about the master’s qiao. In this case, the second question concerning dao elicits information on how the master has acquired such skill. We may then parse the questions this way: ‘What skill you have! What guidance have you received?’ Can we justify this reading, of dao as instrumental to qiao? There may be support for this view from the cicada-catcher’s answer to Confucius. He says, ‘I have dao!’, then immediately provides a detailed description of how he learnt to catch cicadas skilfully. It is possible that dao is the process for acquiring qiao. However, keeping in mind dao’s place within the Daoist tradition and noting especially the relatively frequent use of ‘dao’ to characterize sagely dispositions or insights, we need to be cautious about this reading. This view of dao as a means to the acquisition of skill also sits uncomfortably with the tone in many of the skill stories and, indeed, in the text as a whole, about the significance of dao. The third position also sustains a connection between qiao and dao but places more weight on dao: qiao is a lower level of attainment and may even be considered a prerequisite for (the attainment of) dao. According to this view, qiao is itself an impressive capability, though qualitatively inferior to dao. The structure of the paired questions lends support to this view, with the interlocutor building on the first question, to ask about dao. This view is especially supported by the proposal that the first question is rhetorical, with the question on dao being the focal point of the utterance. Apart from dao being simply a higher level of achievement or technical expertise than qiao, it may refer to a sage-like, enlightened state, one that is qualitatively superior to ordinary human experience. This latter interpretation is supported elsewhere in the Zhuangzi. For example, the ferryman’s actions are so skilful he is ‘like a spirit’ (ruo shen 㤕⾎; 19/22–26).11 Translation of the questions 11

The calibre of mastery is partly a result of the intensity of the masters’ absorption in some of the stories. We are reminded as well of the text’s many references to ‘forgetting’ (wang ᘈ; e.g. 3/14–19) as if the ordinary or conventional is distracting and has to be transcended, in order to attain this enlightened state (6/60–74). The most notable instance of forgetting is the case of ‘sitting and forgetting’ (zuowang ඀ᘈ), where Yan Hui, Confucius’ favoured disciple, claims to have left behind his material form and discarded his knowledge in order to be aligned with the Great Pervader (da tong བྷ䙊) (6/89–93).

288

Skill in Ancient Ethics

aligned with this reading is: ‘What skill you have! Do you also have dao?’ or ‘Do you (just) have skill? Do you have dao too?’ The reasons provided in support of this third view, that dao is a more vital level of attainment than qiao, also speak against the earlier two views on the connection between qiao and dao. I move next to examine four stories about mastery that refer to dao.

2. Do you have (a) dao? The discussion in this section examines four stories. I begin with the two discussed so far. I then assess the story of Butcher Ding who, without prompting, says to Lord Wen Hui that he has dao. Finally, the fourth story, which is quite intriguing, involves the swimmer in dangerous waters at the foot of cascades who, though obviously adept at swimming, claims that he has no dao. I will first discuss the buckle-forger’s response, as it seems to have a more restricted notion of dao; I explain what I mean by this below.

The buckle-forger: ‘I have that which I abide by’ When asked by the Grand Marshal concerning whether he has dao, the forger replies: I have that which I abide by. From the time I was twenty, I was fond of forging buckles. I look at nothing else; not scrutinizing anything but buckles. I have consistently used this approach. Even for one who does not use it, over time, he may come to acquire it; it will be even more so for one who does! Will all things that rely on this not succeed? 22/68–70. Trans. Lai.

The forger answers the Grand Marshal, saying ‘I have that which I abide by’ (chen you shou ye 㠓ᴹᆸҏ). What does he follow? The term shou ᆸ has the implication of defending or abiding by (existing standards or practices). It may also mean that a particular approach has integrity.12 On the most conservative

12

The Shuowen Jiezi glosses shou in terms of shou guan ᆸᇈ, an official whose role is ‘to defend the court’, that is, the court’s values and practices. The term shou consists of two symbols, ᆰ (mian) and ረ (cun). Mian refers to a roof – in this case, an institution, while cun refers to an inch, a standard of measure. The Shuowen entry for shou is as follows: ᆸ˖ᆸᇈҏDŽӾᆰӾረDŽሪᓌѻһ㘵DŽӾ ረDŽረˈ⌅ᓖҏDŽ (ধ‫ޛ‬ᆰ䜘, 4573). The idea of someone’s abiding by a standard of measure is palpable in this character.

What do the Zhuangzi’s Skill Stories Tell Us about Life?

289

reading of this passage, the forger is following an established practice, presumably set out by expert forgers before him. Due to the semantic associations of shou, we have reason to think that the forger’s response – and the focus of this story – is a conservative one, one that seeks to preserve an existing set of practices.13 On this reading, the forger outlines how he adheres to the customs of forging buckles. Having a passion for bucklemaking, he watches how the practice is undertaken, attending to nothing else, and practices it diligently.14 Indeed, he has done so for sixty years! Which aspect of expertise does the story emphasize? Is it intended to drive home the point that persistent practice will get one to an equivalent level of expertise as a forger?15 Why is the forger’s skill so worthy of the Grand Marshal’s admiration, given that buckle-making is not a particularly respected skill? Quite unlike some of the other mastery stories that highlight personal initiative, this story emphasizes persistence. From the (conservative) reading I offer here, it follows that the sense of dao in use relates closely to practical technique or process, rather than to acumen or the attainment of spiritual or epistemological insight. This is especially so if we take it that the buckle-forger’s answer to dao is to abide by tradition or customary practice.

The cicada-catcher: ‘I have dao’ ‘I have dao’, said the hunchback. ‘For the first five or six months I practise balancing two balls on top of each other on the end of the pole and, if they don’t fall off, I will lose very few cicadas. Then I balance three balls and, if they don’t fall off, I will lose only one cicada in ten. Then I balance five balls and, if they don’t fall off, I know it will be as easy as grabbing them with my hand. I position my body like a stiff tree trunk and hold my arm like an old dry limb. No matter how expansive heaven and earth are, or how numerous the ten thousand things, I’m aware of16 nothing but cicada wings. Not wavering, not tipping, not letting any of the other ten thousand things take the place of those cicada wings – how can I help but succeed in taking them?’ 19/17–21. Trans. Lai; adapted from Watson 1968: 199–200. 13

14

15

16

I have selected the most conservative interpretation to allow more nuance in the term dao. The story’s reference to shou, to abide by, is at best unclear, as it does not provide a definite referent. Alternatively, on a more open-ended reading, the forger is claiming that his plan has integrity. He could mean that he has developed his own practice of attending closely to buckles over the years, and has learnt to forge buckles from his own observations. Recall the various conceptions of dao discussed above, that are compatible with greater or lesser scope for individual initiative. We also detect a means–end orientation where, over time, constancy of practice will allow one to acquire such expertise. The character zhi ⸕ in the phrase wei tiao yi zhi zhi ୟ㵙㘬ѻ⸕ may be translated ‘to be aware of ’ but it may also be translated as ‘to comprehend’.

290

Skill in Ancient Ethics

The cicada-catcher says, ‘I have dao!’, which is immediately followed by his detailed description of how he learnt to catch cicadas skilfully. His response is different from the forger’s in that he emphatically affirms that he has dao. Yet, like the forger, he describes an arduous cultivation process for what would seem to be an ordinary undertaking. In these stories, the masters’ responses centre on their unusually adept ways of executing a task. However, they do not discuss their abilities or capabilities as such but rather provide discursive accounts of how they have developed their capabilities. If their answers are intended to respond to the question ‘Do you have dao?’, the masters neither speak about what dao is, nor how it is related to their specific skill. Rather, they describe a procedure or a process of learning. In the case of the forger, buckle-making expertise is both transmitted from the past, and generalizable. Is dao simply a particular way – a method – of executing a task and, for the forger, did this involve tapping into a set of existing practices for forging buckles?17 The forger seems to be following the procedures of others before him, as if he was a keeper of the tradition of forging. It is not unheard of in craftsmanship for skilled masters to be deemed (by themselves and by peers) to be custodians of a tradition and its practices. By contrast, although the cicadacatcher also describes the procedures he uses to develop his capabilities, the

17

There was contemporaneous usage of this restricted sense of dao. There is a particular coupling of dao with a term shu (㺃), discussed in the first section. Shu points to someone walking a path within a city or, metaphorically, to being guided by what has gone before. The use of the compound phrase, daoshu 䚃㺃, appears in the swimmer story in the Zhuangzi (19/49–54), as well as in versions of this story in two Han period (206 bce –220 ce ) Confucian texts. In the Shuo yuan and Kongzi jiayu, there is the same pair of questions on skill and dao. By contrast, the Zhuangzi’s swimmer story has no preceding question about skill, but only a question on dao, the specific dao of treading water. Here are the questions: Zhuangzi ‘May I ask if you have a way (dao) of treading water?’ (19/51–52; trans. Lai). Shuo yuan ‘Do you have skill? Is there dao shu?’ (䴌䀰, ICS 17.24/145/19–24; trans. Lai). Kongzi jiayu ‘Do you have skill? Do you have dao shu?’ (㠤ᙍ 14; trans. Lai). Unlike the Zhuangzi version, the Shuo yuan and Kongzi jiayu’s swimmers are asked about a ‘technique of dao’ (daoshu). This question has the effect of restricting the meaning of dao to a technique. By contrast, to ask whether a person ‘has dao’ in the Zhuangzi leaves the idea of dao open to interpretation, albeit a dao of treading water. Does the representation of this story in the two Han Confucian texts constrict the scope of dao by replacing it with daoshu? I believe so. In both texts, the focus of the swimmer story is on understanding the nature of water, a Confucian symbol for loyalty and trustworthiness (zhong xin ᘐؑ). The point of the swimmer story in the Zhuangzi is quite different, with no mention of either zhong or xin. Interestingly, the Liezi, a text associated with the Daoist tradition, also has the same story, with the questions posed in a similar way as in the Shuo yuan and the Kongzi jiayu. See Liezi ࡇᆀ䃚ㅖ11: ‘⴨լ⇥㩭’. References to the Shuo yuan are cited under the name of its compiler and commentator, Liu Xiang. References to the Kongzi jiayu are cited under the name of its compiler, Wang Su. The Liezi is cited by the name of its compiler, Zhang Zhan.

What do the Zhuangzi’s Skill Stories Tell Us about Life?

291

story points to his resourcefulness in developing a training programme that suits him, the hunchback. It may well be that this programme is entirely unsuitable for Confucius, who ironically is seeking to learn from the hunchback; indeed, it seems that Confucius misses what the cicada-catcher is trying to convey. Whereas the cicada-catcher has developed a programme for his own purposes, Confucius distils the cicada-catcher’s message into a single proposition for his followers: ‘He keeps his will undivided and concentrates his spirit – is this not what we say about the venerable hunchback?’ (19/21; trans. Lai; adapted from Watson 1968: 199–200). The cicada-catcher, who claims he has dao, presents a fulsome account of his training programme which, given his successes in cicada catching, speaks to his acumen in devising a method to handle the case at hand. This goes beyond the idea of following a pattern or established practice, to having the resources himself to develop a unique or superior method. There is a marked sense of agency in the cicada-catcher story, aligned with the notion of dao as activity, rather than pre-determined outcome. I revisit the theme of agency in the final section.

The butcher: ‘What I am devoted to is dao’ Lord Wen Hui had just observed Butcher Ding carve an ox with the most magnificent and elegant movements, as if he were engaged in a dance. The Lord remarked, ‘Ah! How magnificent! Your dexterity has attained such heights!’ (3/4; trans. Lai). To which Butcher Ding replied: ‘What I am devoted to is dao, which exceeds dexterity (ji ᢰ). When I began carving oxen, all I looked at was oxen. Yet, even after three years, I could not see the entire ox. But now, I encounter it with my spirit; not seeing it with my eyes. My understanding ceases, and I am led as my spirit desires. I follow the natural patterns so that my knife drives into the large crevices and is guided into the large hollows; I proceed according to what is naturally there . . . ’ 3/4–11. Trans. Lai.

The butcher was not asked specifically about dao but about ji, another term often translated as ‘skill’. As noted in the first section, the idea of ji derives from removing leaves of bamboo from its stalk, a task that requires dexterity. Butcher Ding is emphatic that ‘dexterity’ does not adequately capture his expertise. This turns the tables in the relative positions of the butcher and his Lord, where there is a pronounced lack in the Lord’s understanding of the

292

Skill in Ancient Ethics

butcher’s actions, mistaking dao for dexterity. As the story unfolds, the Lord recognizes that the butcher has enlightened him on how to nourish life (yang sheng 伺⭏). What is dao? According to Butcher Ding, dao exceeds dexterity. It is not merely the adept handling of a task but rather the quality of its execution: all butchers use knives but Butcher Ding’s has never needed to be sharpened. Others cut, or hack, but his blade moves through the spaces; his use of the instrument is unique and remarkable. Unlike the forger, the butcher is not saying that anyone who practises butchering for an extended time will succeed. The butcher’s dancelike movements emerge from deeper insights: he has grasped the importance of understanding the patterns of nature and is appropriately responsive to it. While the forger might be guided by tradition, the butcher is not held back by human standards. Where we are taught to see oxen joints as challenges to carving, Butcher Ding has ceased looking at oxen in this way. Rather, he appreciates with his spirit that there are spaces in the joints that the sharp blade can slide through. Indeed, particularly in difficult situations, it is critical that he treads cautiously and unhurriedly, to more fully apprehend the physical nature of oxen. By analogy, this is the way the Lord should approach the natural patterns of the world. I suggest it is in being attuned to the world, and in responding appropriately, that the notion of dao lies (this is the active sense of dao – walking the journey). Many epistemological questions arise from this story, as do questions about how best to live one’s life. In the notion of responsiveness lies an important aspect of Zhuangzian agency: it is not constrained by humanity’s plans, and is able to respond as situationally appropriate. Particularly in contrast to the story of the forger, the idea of dao qua attainment or insight is palpable in this story.18 And, indeed, Lord Wen Hui has much to learn from the butcher.

The swimmer: ‘I have no dao’ In this story, Confucius is unsettled by a swimmer in the dangerous cascades, thinking he surely would have lost his life.19 When Confucius asks, ‘May I ask

18

19

Anglophone scholarly discussions of the Zhuangzi focus overwhelmingly on the distinctiveness of the enlightened person’s or the sage’s capabilities, that enable a unique way of executing a task. Rarely are the discussions simply about dao as process. Here is a sample of discussions in the literature that develop the theme of enlightened life in the Zhuangzi: Moeller and D’Ambrosio 2017; Kohn (ed.) 2015; Chong 2011; Huang 2010; Møllgaard 2007; Coutinho 2004; Ames (ed.) 1998. For a contrasting view of skill in the Zhuangzi – indeed, one that understands dao as being applicable to the passivity of the knife, rather than the skill of the butcher – see Schwitzgebel 2019. Refer to footnote 17, where I discuss versions of this story in two Confucian texts.

What do the Zhuangzi’s Skill Stories Tell Us about Life?

293

whether, in treading water, you have dao?’ (19/51–52; trans. Lai), the swimmer replies: ‘I have no dao. I began with what was originally there, developed what was natural (to me), and reached completion in keeping with fate. I go under with the swirls and come out with the eddies, following the dao of the waters without thinking about the self. That’s how I can tread water.’ 19/52–53. Adapted from Watson 1968: 204–05.

Confucius then sought to be elucidated on the swimmer’s first statement, and the swimmer replied: ‘I was born on dry land and felt safe on dry land – that was what was originally there. I grew up with the water and felt safe in the water – that was what was natural to me. I do not know how or why20 I do what I do – that’s fate.’ 19/53–54. Adapted from Watson 1968: 204–205.

Why does the swimmer say he has no dao, and then elaborate at length on how he has learnt to swim in dangerous waters? Like the cicada-catcher, and to some extent the butcher, he has exercised agential initiative in developing plans for handling the waters. The three occurrences of dao in this passage have different meanings. First, in Confucius’ question, dao refers to the actions of treading water (dao shui you dao 䑸≤ᴹ䚃). This is an open-ended question and it could refer either to dao as a particular way of treading water especially where currents are strong, or to particular capabilities the swimmer might have. Second, in the swimmer’s response, he says, ‘I have no dao’ (wu wu dao ੮❑䚃). What could this mean? That he uses no particular method, or that he does not have particular abilities? Both seem unlikely, given the details provided in his responses to Confucius; not only is he able to articulate a developmental trajectory and clearly portray his actions in the water, he also elucidates his own comments when Confucius asks him to do so. In this story, the swimmer says that he follows the water’s dao (cong shui zhi dao ᗎ≤ѻ䚃). What does it mean to say the water ‘has’ dao? Does this suggest that the water manifests agency, or at least some semblance of it? If so, we have to ask whether the swimmer’s decision to follow the water’s dao is not already itself a dao. In other words, is the decision to not have dao itself an exercise or manifestation of dao? 20

Not knowing ‘how or why’ is a fitting translation I have taken from Ziporyn, in his translation of the phrase suo yi ran ᡰԕ❦ (in the statement ‘н⸕੮ᡰԕ❦㘼❦’) (2009: 81), as ‘н⸕ . . . ᡰԕ’ may refer to lack of knowledge either in providing reasons for some incident, or in describing the means whereby that incident occurred.

294

Skill in Ancient Ethics

We note, first, that the swimmer may not have answered Confucius’ question(s). Whereas Confucius asks about a specific dao, the dao of treading water, the swimmer says he has no dao. Was he sidestepping Confucius’ question because he in fact possesses the dao of treading water?21 If we take it that the swimmer possess the dao of treading water, that is, that he is able to execute a particular style of swimming appropriate to these waters, why does he say he does not have dao? The clue to this puzzle lies in the swimmer’s answer to Confucius, where he points to the water’s dao: he follows the water’s dao, ‘without thinking about the self ’. The idea of not thinking about the self (bu wei shi н⛪⿱) may be understood as the swimmer’s not relying on his personal plans for how to encounter the water on each occasion. In the difficult circumstances of swimming in the waters at the foot of the cascades, it is necessary for him to subject himself to the currents, that is, to follow the water’s dao. Swimming in these waters would be like being caught in a rip current, where the most appropriate action would be for the swimmer to yield to the current rather than to struggle against it. In this case, for the simmer to swim according to particular strokes he is familiar with, or in a specific direction, would effectively endanger his life. Is the swimmer ‘in control’? This is an important question if we believe that having dao entails manifesting agential initiative. The swimmer notes that, while swimming, he goes with the swirls and eddies. More importantly, he takes this action because he understands the nature of water, and this is because he has been attentively aware of the water’s dao. As he says, the dry land was his first context and he felt safe in it; and the water, a second context that he learnt to navigate comfortably in.22 This response to Confucius’ further question shows that the swimmer has not missed Confucius’ point about his dao for treading water. The swimmer is indeed aware, more importantly, of whether he should have a dao to tread water, and whether and how he should follow the water’s dao, in these waters. The swimmer’s alignment with his natural environment has led him to understand that, while in the water, it is indeed important for him to have no dao in the relevant sense. I propose that this indicates that the swimmer was in fact answering Confucius’ question: to have a dao of treading these dangerous 21

22

For it is clear that he has successfully navigated the waters; following his swim, he gets up on the bank, sauntering with his wet hair, nonchalant and unaware that Confucius the observer had feared for his life. He also registers that he cannot explain, nor does he understand, why or how he has attained such alignment with his environment, using the vocabulary of ‘fate’ (ming ભ). I take ming to mean the conditions of his life, that were part of his ‘given’ environment, one that he subsequently becomes comfortable navigating within.

What do the Zhuangzi’s Skill Stories Tell Us about Life?

295

cascades is to understand the water’s dao and, in doing so – being in control – he has made an informed decision to have no dao. We have reviewed four stories with quite different conceptions of dao. The forger adheres to the practices or tradition of forging buckles (‘I have that which I abide by’). The cicada-catcher’s self-crafted learning programme enables him to use his pole as if it were his arm (‘I have dao’). The butcher’s spirited carving of oxen is enabled by his attentiveness to the ox rather than by guided by his trained capabilities (‘What I am devoted to is dao’). And the swimmer actively decides to follow the water’s dao by having no dao (‘I have no dao’). In the final section of this chapter, I discuss how we might bring together these complex and varied conceptions of dao to reflect on how the Zhuangzi’s skill stories may contribute to our thinking about agency and life.

3. Dao, agency and life We have now encountered four conceptions of dao in as many skill stories that refer to the term. Together with the other stories about skill in the text, they introduce us to a variety of different models of mastery. But what are these masters models of? Are they models of craftsmanship specific to each (mostly ordinary) vocation? Are they promoting alternative ways of handling tasks? Or do they offer more, in terms of options about how we think about life? In this concluding section of the chapter, I discuss three pictures of dao and agency, corresponding roughly to the three questions posed above about modelling dao. The first examines dao in terms of learning about and carrying out established practices in particular domains. The second suggests that dao offers an alternative mode of conduct that stands in contrast to conventional models of behaviour. The third – the view I propose – offers an approach inspired by the models of craftsmanship but which is nevertheless not locked into any one of them. As we will see later, this an account of dao that highlights the centrality of agency.

Dao and standards The buckle-forger credits his skills to his abiding by a practice. Is this a plausible account of dao and its connection with mastery? I propose we should resist it given that it is a view not shared by most other stories. Importantly, many of the stories – including the buckle-forger story – are not about the masters instructing others in the profession. Instead, in most of them, the masters instruct a range of

296

Skill in Ancient Ethics

people, including Confucius or his disciples, dukes, lords, or general audiences who are keen to find out more about the enthralling performances. Often, too, these interlocutors draw from the performance, and sometimes from what the master says, extracting their insights to enlighten some aspect of life.23 What would their audiences, typically men of higher status than the workmen engaged in ordinary trades, hope to learn from them? The quick answer is straightforward: not swimming, not butchering, and not woodcarving. A more complex answer will be set out in the following two subsections. For now, we can say more against holding this conception of dao, qua standards, as a prominent one in the Zhuangzi. The text uses a raft of terms from the vocabulary of its rivals, in explicit mentions and in stories and imagery, to caricature the kind of life driven and shaped by prescribed standards. These include conceptions of use and functionality (yong ⭘; 4/64–91), of ritual propriety, rightness, models and measures (li, yi, fa, du ⿞㗙⌅ᓖ; 38 14/404), and approvals and disapprovals (shifei ᱟ䶎; 5/7–8; 23/63–66). Perhaps most revealing are the metaphorical references to the use of compass and square (guiju 㾿⸙; 8/13–14; 19/62–64). In these passages, the text highlights the tendency for people to structure their lives in predictable ways through the application of established benchmarks. Contesting this model, the Zhuangzi maintains that life lived according to prescribed standards is impoverished, characterizing the pursuit of fame (ming wen ਽㚎) as ‘fetters’ (zhigu Ṿế; 5/29–31).

Dao as an alternative model The wheelwright Lun Bian does not use a compass and square, as we would normally expect. He veers from standard practices, decidedly rejecting the action-guiding standards provided by these tools.24 Instead, he uses only a mallet and chisel, and is led by what he feels:

23

24

For example, Confucius instructs his followers about the undivided will of the cicada-catcher, and how he concentrates his spirit (yong zhi bu fen, nai ning yu shen ⭘ᘇн࠶ˈѳࠍᯬ⾎; 19/21). Confucius is impressed by the ferryman’s ability not to be distracted by external conditions, stating that those who place weight on such external circumstances become clumsy with what is internal (fan wai zhong zhe nei zhuo ࠑཆ䟽㘵‫ޗ‬ᤉ; 19/22–26). The wheelwright, enlightened with his expertise in wheelmaking, instructs Duke Huan on the limitations of learning and acting from the words of those in ages past (13/68–74). Furthermore, we are taken aback that the wheelwright and the scribe (21/45–47) take their lives into their own hands by denigrating the activities of their superiors, in the case of the wheelwright, and deliberately flouting officially-sanctioned practices, in the case of the scribe. The wheelwright’s non-reliance on standard measuring tools such as the compass and square was first noted by Lisa Raphals (2019).

What do the Zhuangzi’s Skill Stories Tell Us about Life?

297

When I chisel a wheel, if the blows of the mallet are too gentle, the chisel slides and won’t take hold. But if they’re too hard, it bites in and won’t budge. Not too gentle, not too hard – you can feel it in your hand and respond [ying ៹] to it in your heart-mind [xin ᗳ]. 13/71–72. Adapted from Watson 1968: 152–153.

The vocabulary of the Zhuangzi is attuned astutely to the debates of its day. Tellingly, it takes issue with the term xin ᗳ, often translated ‘heart-mind’, a term that figures centrally in the ideas of Mengzi (ᆏᆀ; c. 385 bce –c. 312 bce ) and Xunzi (㥰ᆀ; c. 310 bce– c. 219 bce ), advocates of Ru ݂ (Confucian) ideals and practices. The Zhuangzi uses xin to expose what it saw as faulty foundations of Ru thought. From the Zhuangzi’s point of view, their advocacy for a polished, completed, mind-heart25 (cheng xin ᡀᗳ; 2/21–27) was a misdirected pursuit. Their project, to prepare and shape the heart-mind, was effectively an indoctrination of a fixed value-orientation. These values were affirmed ‘Aye!’ (shi ᱟ) and disagreement with them denied ‘Nay!’ (fei 䶎), through the imposition of a weighted vocabulary comprising terms such as li (ritual propriety), yi (rightness), and others mentioned previously. The Zhuangzi was wary of how officials with these objectives conferred upon themselves the prerogative of determining how all others should live. In addressing its concerns, it weaves xin into its discourse in a number of enlightening ways, including in: 1.

2.

3.

25

rejecting how the polished or completed heart-mind was being used as a teacher (cheng xin er shi ᡀᗳ㘼ᑛ), in such a way that made proponents defensive about their views and quick to deny others’ views (2/21–27); upholding the model of the ‘true person’ (zhen ren ⵏӪ). The true person did not use what was humanly-designed to alter what was beyond its reach (bu yi ren zhu tian нԕӪࣙཙ) (6/7–14). It is fascinating that the use of the humanly-constructed heart-mind is characterized as an attempt to interfere with the natural patterns of the world; and representing the heart-mind as a ‘machine’. Zigong (ᆀ䋒), a Confucian disciple, encounters a farmer (12/52–58). Bothered by the laborious and apparently inefficient methods used by the farmer, Zigong offers him a water pulley that will greatly facilitate the task of irrigation. To Zigong’s

Not all the Zhuangzi’s references to xin are used in a critical manner, to reject Confucian ideas. The term is also used in a more neutral sense, to denote a moral-cognitive capacity that is the seat of human intention, and which is not necessarily aligned with any one set of values (see, e.g. 4/34–53; 13/71–72).

298

Skill in Ancient Ethics

surprise, the farmer rejects the gadget. And to Zigong’s dismay, the farmer sets up the imagery of a ‘machine heart’ (ji xin ₏ᗳ), saying, ‘If there is artifice, there will also be stratagems; if there are stratagems, there will also be scheming hearts.’ 12/56. Trans. Lai.26 Xin is the central and unifying capacity in Ru-ist cultivation. In its cultivation, a person’s sensory capacities and sensibilities are so attuned by and to the cultivated and completed xin, in such a way that impedes their responsiveness to the world. Therefore, the use of eyes and ears – in the way these sensory capacities have been educated to see or hear things a certain way – is challenged in the Zhuangzi (e.g. 5/7–8); the butcher and the forger, for example, explicitly note that they have ceased relying on sight. This prompts us to ask what the Zhuangzi’s alternative proposal actually offers. There is a range of possibilities across the stories, and more broadly in the text, of how the Daoist sage might handle situations, if they are not to be guided by their heart-mind. The wheelwright feels it in his hand, the cicada-catcher and the ferryman’s operations involve the ‘spirit’, and the bell-stand maker takes care not to squander his qi ≓ and fasts his heart-mind (19/54–59). Perhaps there are just too many possibilities offered in the text to construct a neat account of the alternatives offered by the Zhuangzi, if it is not to be found in (the application of) standards and measures. Are the text’s pictures too unwieldy to be collated into a more systematic account of dao? Or is this picture, the via negativa, all that we have? In the following subsection, I discuss a positive approach to life that the Zhuangzi’s dao offers.

Dao that does not make tracks Why was the Zhuangzi so concerned about the imposition of standards and measures? After all, this would have been a widely-used approach for quelling the unrest of the Warring States period (Zhanguo ᡈഭ; 475–221 bce ). From the perspective of the Zhuangzi, this approach fails humanity. Life is rich and diverse, and coloured with many different encounters and circumstances. Those who seek to impose standards have not understood the pitfalls of (using) standards. First, standards do not fully capture the plurality and diversity in the world.

26

Note that the term ji xie ₏Ỡ, translated ‘artifice’ here, also has the meaning of inflexibility, as the second term, xie, refers to ‘fetters’.

What do the Zhuangzi’s Skill Stories Tell Us about Life?

299

Second, measures of human behaviour must surely adapt to changing circumstances; unfortunately, standards are often applied inflexibly.27 The Zhuangzi addresses this issue in the imagery of a monkey, ill-fittingly dressed in the robes of the Duke of Zhou. Understandably, it wildly bites and tears at them. We are told that the differences between the standards in antiquity and the (then) present are no less significant than those between the monkey and the Duke (14/40–44).28 Not to recognize the limitations of a fundamentally standards-driven approach, and to continue to use it, is likened to the armwaving of a praying mantis, who believed that its arm-waving could stop a carriage (12/48–49). Chapter 5 of the Zhuangzi, entitled ‘The Signs of the Fulness of Potency’ (‘De Chong Fu’ ᗧ‫ݵ‬ㅖ), speaks about unfashionable people: the ugly, deformed, and outcasts. They are people who do not meet the ‘normal’ standards but, in this text, they speak Zhuangzian wisdoms or they are upheld as paragons of Zhuangzian life. Here, the Zhuangzi not only challenges conventional measures of success; it turns the tables on these measures, pointedly demonstrating that these figures have exceeded the merely conventional.29 There is a man called ‘Cripple Lipless with the crooked legs’ and another with a distended neck, ‘Pitcherneck with the big goitre’ (5/49–55; trans. Graham 2001: 80). The person with the distended neck excelled at advising Duke Huan such that the necks of other, ‘normal’ men looked scrawny to the Duke! But these figures in the Zhuangzi are deformed only in the conventional sense, for they are used to interrogate conceptions of normality.30 Prevailing conceptions of normalcy are both too narrow and too rigid to honour the plurality in the world. The positive account of dao arising from the Zhuangzi arises from a fundamental feature shared by its many examples and stories: the masters’ responsiveness (ying ៹) to the circumstances, in executing their tasks. I suggest that responsiveness is the crucial element in Zhuangzian agency. It is marked by an individual’s sensibility to salient considerations, together with her capabilities aptly to handle situations or formulate solutions. By contrast, standards and

27

28 29

30

In a chapter entitled ‘The Cycles of Heaven’ (‘Tian Yun’ ཙ䙻), the text notes that ‘propriety, rightness, models and measures (must) responsively change with the seasons and times’ (14/40–44; trans. by author). Alas, this is also a passage that highlights the shortcomings of Confucius’ proposal! There is a story of an ugly, gnarled tree, which smelt bad and whose leaves tasted so bad it could not be eaten. So ugly it is that nobody cared to use its wood. Because of its ugliness, however, it was left alone and now has a canopy so vast it provides wonderful shelter (4/75–83). Perkins (2014: 158–165) and Galvany (2019) provide two fascinating and insightful accounts of these figures.

300

Skill in Ancient Ethics

measures curtail a person’s responsiveness to situations by narrowing the alternatives, allowing only those approved by the cultivated heart-mind. It is not the place to provide a full examination of the notion ying here; suffice to say that the text upholds a model of a person whose responses are not bound by convention (yi ying wu qiong ԕ៹❑マ; 2/27–31). To lack responsiveness is symptomatic of a failure to grasp of dao, Confucius is told (31/1–43). In the Zhuangzi, responsiveness is aligned with a notion of dao as active, agentiallydriven engagement. Let me relate the notion of responsiveness to the idea of track-making and track-following. The Zhuangzi makes metaphorical references to the idea of tracks (ji 䐑), as in the tracks left by a person walking along a path. I propose that the idea of not leaving tracks is the conceptual framing for the Zhuangzi’s skill stories, and which also undergirds its conception of agency. When we walk, we inevitably leave tracks; whether these are materially identifiable is not the point here. The important question here is whether we intend for others to follow our tracks. Tracks made with this intent are of course standards for others to comply with. The idea of tracks is conceptually intertwined with the notion dao, walking a path. In this context, we are prompted to ask, ‘Whose path are you walking?’ This question helps us focus on whether and how we have exercised agential initiative in designing a programme for developing our skills according to our capabilities – just as the masters have. We expect the answer to fall somewhere along a continuum, combining elements of following others’ paths and creating our own, rather than an outright ‘mine’ or ‘theirs’. At one extreme, though, little seems to be left up to the individual. In a passage that accuses Confucius of setting up footsteps, the text expressly states that it is inappropriate to enforce tracks but advocates that a person should instead encourage others to consider the shoes that made those footsteps (14/74–78; see also 31/26–32). Now, the implications of not leaving tracks emerges: when one deliberately sets up footsteps for others to follow, it denies others the opportunity of being responsive and of walking their own paths: their own daos (22/30–33; 22/37–39).31 While there is no specific term for ‘agency’ in the Zhuangzi, the distinction between a person’s being told to follow in someone else’s tracks, or being encouraged to devise her own, is clear. For this reason, the consummate person 31

A fisherman who criticized Confucius for lacking in responsiveness advises him to dwell on what is essential: ‘In being loyal, service is crucial; in drinking, delight is crucial; in mourning, grief is crucial; in filial duty, comfort is crucial. In undertaking deeds elegantly, there is not (just) one set of tracks’ (31/32–40 trans. Lai).

What do the Zhuangzi’s Skill Stories Tell Us about Life?

301

(zhiren 㠣Ӫ) expunges her tracks (20/33–34). How is the theme of not leaving tracks relevant for us? This question is important especially as the accounts of mastery are personalized. The language used in many of the stories involve personal referents, marked by the characters for ‘me’ or ‘I’ (wo ᡁ; wu ੮), and a self-referential diminutive based on rank (chen 㠓).32 They indicate that the masters are offering their personal accounts, not intended as general statements of guidance. We could say that the experts in the Zhuangzi are not leaving tracks, even though in some of the stories, Confucius ironically takes it upon himself to appropriate and distil the masters’ ‘wisdoms’ for his followers. The idea of not leaving tracks is instructive as it seeks to ‘return’ initiative to individuals: they should develop, with authorial integrity, skills for their lives. My view resonates with the eudaimonistic account of dao offered by Chris Fraser, who notes that the well-lived life in the Zhuangzi is substantively thin: its focus is rather on ‘a distinctive mode or style of activity, accompanied by certain characteristic attitudes’ (2014: 544). Further, I agree with Fraser that the substantively thin account should be welcomed, rather than be a source of worry.33 In the vocabulary here, one can be liberated from having to follow somebody else’s tracks. To carve out one’s own path is of course not a licence to freedom, for the key idea is to develop skills to be appropriately responsive. Learning is no easy street either, as the stories highlight the ongoing nature of cultivation and its exacting demands on us to attend to detail. The demands may be cognitive, emotional, and/or physiological. The Zhuangzi’s stories supplement our imagination with examples of different and unique ways to work and hone an individual’s distinctive capabilities and traits. Each actively developed dao is unique. In conclusion, in my examination of dao in the context of the ‘skill’ stories in the Zhuangzi, I have explored different ways in which dao – creating and walking a path – might be related to accomplished performances. In presenting nuanced differences across the stories examined here, I suggested that the mastery stories are inspirational rather than prescriptive. Dao exceeds skill, as Butcher Ding says. The stories incite curiosity and arouse imagination about the possibilities for human action. Dao is ‘beyond skill’ in that we should not expect it to be fully explicable with reference to any one of the skill stories.34 Earlier, I posed the

32 33 34

Even in the case of the buckle-forger, one can do better than simply succeed in forging. Fraser writes, ‘The substantive thinness of this ideal is an advantage, not a shortcoming’ (ibid: 562) This phrase is from Dan Robins (2011). Robins argues that dao is not any one specific way of acting and therefore, to pin it down to one specific way runs against the conceptions of dao in the Laozi and Zhuangzi.

302

Skill in Ancient Ethics

question of what exactly the masters are models of. The response to this question is not that each master is a model of any one activity recommended to us (even though they are activities in their own right). Rather, they present accounts of initiatives taken by each master, often with details about his orientation to the task, and how he developed capacities for better handling the task. The thinking behind this approach to individual initiative is straightforward: a text that denounces compliance with tracks cannot also require that we follow the skill stories as tracks. Many of the stories share a personalized account of mastery, a vision that is possible only if individual initiative is encouraged. Insofar as the masters display agential authority in developing their own programmes, readers of the Zhuangzi might be inspired do likewise. In doing so, we would not be walking someone else’s dao but developing our own in the activities we undertake. Sometimes, as with the cicada-catcher, we might say, ‘I have dao!’ or, at other times, we might have reason like the swimmer to say ‘I have no dao!’.

References Ames, R.T., ed (1998), Wandering at Ease in the Zhuangzi, Albany, NY: State University of New York Press. Billeter, J.F. (2011), Zhuang Zi Shi Jiang ljᒴᆀഋ䇢NJ. Translated from the original Leçons sur Tchouang-tseu by Song Gang ᆻࡊ, Taipei: Linking Publishing. Chiu, W.W. (2019), ‘The Forger: The Use of Things’, in K. Lai and WW. Chiu (eds), Skill and Mastery: Philosophical Stories from the Zhuangzi, CEACOP East Asian Comparative Ethics, Politics and Philosophy of Law Series, 259–277, London: Rowman & Littlefield International. Chong, K. (2011), ‘The Concept of ‘zhen’ ⵏ in the Zhuangzi’, Philosophy East and West 61 (2): 324–346. Coutinho, S. (2004), Zhuangzi and Early Chinese Philosophy: Vagueness, Transformation, and Paradox, Aldershot: Ashgate Publishing. Fraser, C. (2014), ‘Wandering the Way: A Eudaimonistic Approach to the Zhuāngzˇı ’, Dao 13: 541–565. Galvany, A. (2019), ‘Radical Alterity in the Zhuangzi: On the Political and Philosophical Function of Monsters’, Philosophy Compass 14 (9) (online journal: https:// onlinelibrary.wiley.com/doi/abs/10.1111/phc3.12617; accessed on 15 April 2020). Graham, A.C. (2001), Chuang-tzu: The Inner Chapters, Indianapolis: Hackett Publishing. Huang, Y. (2010), ‘Respecting Different Ways of Life: A Daoist Ethics of Virtue in the Zhuangzi’, Journal of Asian Studies 69: 1049–1069. Kohn, L., ed. (2015), New Visions of the Zhuangzi, St Petersburg, FL : Three Pines Press.

What do the Zhuangzi’s Skill Stories Tell Us about Life?

303

Lai, K., and W.W. Chiu, eds (2019), Skill and Mastery: Philosophical Stories from the Zhuangzi, CEACOP East Asian Comparative Ethics, Politics and Philosophy of Law Series, London: Rowman & Littlefield International. Li, P. ᵾള et al. (1999–2004), Gu wenzi gulin ljਔ᮷ᆇ䁱᷇NJ [Thesaurus of semasiological glosses on paleographic characters], 7 vols., Shanghai: Shanghai Education Press к⎧᭾㛢ࠪ⡸⽮. Liu, A. ࢹᆹ, Huainanzi, Huainan Hong Lie Jie lj␞ই卫⛸䀓NJ, Sibucongkan series (ഋ䜘਒࠺ࡍ㐘), Vols. 425–428, Shanghai: Han Fen Press (Commercial Press) (Ჟк⎧⏥㣜⁃㯿). Liu, X. ࢹੁ (comp. and comm.), Shuo Yuan 䃚㤁 [Garden of stories], Shanghai: Commercial Press ୶࣑ঠҖ侶, 1937. Moeller, H.-G., and Paul D’Ambrosio (2017), Genuine Pretending: On the Philosophy of the Zhuangzi, New York: Columbia University Press. Møllgaard, E. (2007), An Introduction to Daoist Thought: Action, Language and Ethics in Zhuangzi, London and New York: Routledge. Perkins, F. (2014) Heaven and Earth are Not Humane: The Problem of Evil in Classical Chinese Philosophy, Bloomington and Indianapolis: Indiana University Press. Raphals, L. (2019), ‘Wheelright Bian: A Difficult Dao’, in Karyn Lai and Wai Wai Chiu (eds), Skill and Mastery: Philosophical Stories from the Zhuangzi, CEACOP East Asian Comparative Ethics, Politics and Philosophy of Law Series, 129–142, London: Rowman & Littlefield International. Robins, D. (2011), ‘It Goes beyond Skill,’ in Chris Fraser, Dan Robins and Timothy O’Leary (eds) Ethics in Early China, 143–157, Hong Kong: Hong Kong University Press. Schwitzgebel, E. (2019), ‘The unskilled Zhuangzi: Big and Useless and Not So Good At Catching Rats’, in Karyn Lai and Wai Wai Chiu (eds), Skill and Mastery: Philosophical Stories from the Zhuangzi, CEACOP East Asian Comparative Ethics, Politics and Philosophy of Law Series, 101–108, London: Rowman & Littlefield International. Wang S. ⦻㚵 (comp.), Kongzi Jiayu ljᆄᆀᇦ䃎NJ [Sayings of Confucius for the Confucian School], Taipei: Zhonghua Publishing 㠪⚓ѝ㨟ᴨተ, 1979. Watson, B., trans. (1968), The Complete Works of Chuang Tzu, New York and London: Columbia University Press. Xu Shen 䁡᝾ et. al., Shuowen Jiezi 䃚᮷䀓ᆇ [Etymological gloss of characters and words]. ˄ᆻ˅ᗀ䡹ㅹཹᮅṑᇊ (Song dynasty edn. by Xu Xuan et. al.), Beijing: Zhonghua, 1963. Zhang Zhan ᕥ⒋ (comm.), Liezi ࡇᆀ, Taipei: Yiwen Publishing 㰍᮷ঠᴨ佘, 1971. Zhuangzi 㦺ᆀ, Zhuangzi Yinde 㦺ᆀᕅᗇ [A Concordance to Zhuangzi], Hong, Ye ⍚ᾝ (ed), Harvard-Yenching Institute Sinological Index Series, Suppl.20, Cambridge MA : Harvard University Press, 1956. Ziporyn, B., trans. (2009), Zhuangzi: The Essential Writings with Selections from Traditional Commentaries, Indianapolis: Hackett Publishing.

304

Part Six

Comparative Perspectives on Skill in Ethics

305

306

15

Can One Become Wise by Learning to Catch Cicadas? Analogies between Crafts and Wisdom in Daoism and Stoicism David Machek

Introduction and texts It has become a well-established view in the scholarship that some of the most influential ethical theories in two ancient philosophical traditions, Chinese and Greek, share one fundamental feature: they envisage human flourishing as the ultimate reference point of ethical self-cultivation or moral development.1 Typically, flourishing largely depends on the acquisition of certain individual excellences, virtues or powers that enable one to recognize the true value of things, persons or situations and to respond to them adequately. In both traditions, these qualities have often been exemplified by excellence in crafts, arts or skills, such as archery, carving, weaving or navigation.2 The analogies between excellence in life and excellence in skills were perhaps most extensively and systematically explored by Stoic thinkers and the authors of the Zhuangzi, one of the most philosophically significant texts of Daoist thought.3 Recent comparative studies of Stoicism and Daoism have established that there is a fundamental similarity between the Daoist and Stoic ideals of flourishing that can perhaps be best characterized as a joint commitment to a broad ideal of ‘following nature’. As

1

2

3

This view informs, for instance, two recent volumes on Sino-Greek comparison: King 2015; Lai, Benitez and Kim 2019. For a useful overview and discussion of different types of crafts and arts in both these traditions, see Raphals 2005. The volume by Lai and Chiu 2019 testifies to the centrality of craft analogies in the Daoist text Zhuangzi, and to the thriving scholarly interest in it. Whereas I shall be using the broad and somewhat anachronistic category of ‘Daoism’ in my interpretation, the evidence I shall consider really boils down to a single text from the Warring States period, the Zhuangzi, which was probably written by several different groups of authors (which I shall collectively refer to as ‘Zhuangzi’).

307

308

Skill in Ancient Ethics

a result, one might expect that the shared interest in craft analogies is nonincidentally linked with this shared philosophical outlook – as I argue it is.4 Building on these similarities, I focus on one significant – and striking – structural difference between Stoic and Daoist analogies. It concerns the relationship between specialized skills as the explanans and excellence in life, or ‘wisdom’, as the explanandum. The Stoics make clear that the expertise in skills like archery or acting is merely analogical to, but certainly not identical with, wisdom. This is why the process of becoming a good archer or a good actor supposedly does not contribute in any significant way to becoming a wise person. Rather, one can become wise only by submitting to a special, wisdom-conducive educational curriculum including the study of scientific disciplines such as logic or physics. In contrast, the Zhuangzi is much less explicit about this distinction. In fact, some passages even suggest that learning mundane skills such as bellstand carving or cicada catching is the route to wisdom. In this chapter I try to explain why, for all their similarities, Stoics and Zhuangzian Daoists took such different views of the practical relevance of learning skills for wisdom. I begin (Section 2) by noting significant differences in how craft analogies were employed in each context, including a Stoic preference for so-called stochastic crafts, and a Daoist preference for carving-like crafts. I propose (Section  3) that a single common denominator can explain these different preferences. I suggest that two dichotomies informed the Stoic and Daoist theories and structured their respective philosophical terrains: a Greek dichotomy of rational and non-rational and a Chinese dichotomy of natural and artificial. These dichotomies help clarify (Section 4) why expertise in a specialized skill could be seen as a possible road to wisdom in Zhuangzian Daoism but not in Stoicism. I begin (Section  1) with an account of general underlying similarities between Stoic and Zhuangzian Daoist ethical theories, with particular interest in the explanatory role of craft analogies within them. These broad similarities make all the more salient the differences I shall point out. Limited space does not permit extensive discussion of many controversial issues of interpretation in the texts, or of possibly controversial claims about differences between these two philosophical traditions overall. I consider the potential benefits of comparative ventures well worth the risk, and I hope that even those who are not persuaded by my suggestions will find them worthy of further discussion.

4

Yu 2008; Machek 2018.

Can One Become Wise by Learning to Catch Cicadas?

309

The follow eight passages from the Zhuangzi and from Stoic texts establish an analogy between excellence in life and excellence in a skill. This selection is not exhaustive, but it does bring out the most fundamental aspects of the analogy in both schools.

The Butcher [D1] Cook Ding was cutting up an ox for Lord Wenhui. As every touch of his hand, every heave of his shoulder, every move of his feet, every thrust of his knee – zip! zoop! He slithered the knife along with a zing, and all was in perfect rhythm, as though he were performing the dance of the Mulberry Grove or keeping time to the Jingshou music. ‘Ah, this is marvellous!’ said Lord Wenhui. ‘Imagine skill reaching such heights!’ Cook Ding laid down his knife and replied, ‘What I care about is the Way, which goes beyond skill. When I first began cutting up oxen, all I could see was the ox itself. After three years I no longer saw the whole ox. And now – now I go at it by spirit and don’t look with my eyes. Perception and understanding have come to a stop and spirit moves where it wants. I go along with the natural makeup, strike in the big hollows, guide the knife through the big openings, and follow things as they are. So I never touch the smallest ligament or tendon, much less a main joint. . . . There are spaces between the joints, and the blade of the knife has really no thickness. If you insert what has no thickness into such spaces, then there’s plenty of room – more than enough for the blade to play about it. . . . However, whenever I come to a complicated place, I size up the difficulties, tell myself to watch out and be careful, keep my eyes on what I’m doing, work very slowly, and move the knife with the greatest subtlety, until – flop! the whole thing comes apart like a clod of earth crumbling to the ground . . . .’ ‘Excellent!’ said Lord Wenhui. ‘I have heard the words of Cook Ding and learned how to care for life!’ Zhuangzi jijie 1954: 3:18–19. Trans. modified from Watson 2013: 19–20.

The Ferryman [D2] Yan Yuan asked Confucius, ‘I once crossed the depths at Goblet Deeps. The ferryman handled the boat like a spirit. I asked him, “Handling a boat, can it be learned?” He said, “It can. Good swimmers are quickly able, and as to divers, without having seen a boat, they can handle one.” I asked him about this but he didn’t tell me. ‘I ask, what was he referring to?’ Confucius said, “Good swimmers

310

Skill in Ancient Ethics

are quickly able – it’s that they forget the water. As to divers’ handling a boat without ever seeing one – they regard the depths as like land and the boat’s capsizing as like their cart rolling backward. Boats could capsize and carts roll backward all around them without it penetrating their chest. Where could they go and not be at ease? Shooting for tiles, you’re skilled. Shooting for silver buckles, you’re uneasy. Shooting for gold, you’re flustered. Your skill’s the same, but there’s something you’re worried about – this is putting weight on the external. Anyone who puts weight on the external gets clumsy with the internal.” Zhuangzi jijie 1954: 19:8. Trans. modified from Watson 2013: 148.

The Wheelwright [D3] Duke Huan was reading a book in the upper part of his hall. Wheelwright Bian was cutting a wheel in the lower part of the hall. Putting down his hammer and chisel, he went up and asked Duke Huan: ‘I venture to ask what words Your Highness is reading?’ The duke replied: ‘The words of the sages.’ The wheelwright said: ‘Are the sages still alive?’ The duke replied: ‘They’re already dead.’ The wheelwright said: ‘If that is so, then what My Lord is reading is the ancients’ dregs and leavings.’ Duke Huan replied: ‘How can a wheelwright offer comment on what I am reading? If you have an explanation, very well, but if you have none, you shall die.’ Wheelwright Bian said: ‘Your servant will observe it from the perspective of his own occupation. Now in cutting a wheel, if the spokes are loose, they’ll fit sweet as a whistle, but the wheel won’t be solid. If they’re too tight, you won’t be able to insert them, no matter how hard you try. To make them neither too lose nor too tight is something you “get” in your hands and respond to in your mind. The mouth cannot put it into words, and there is a knack to it. I cannot explain it to my son, and my son cannot receive it from me, which is why I am seventy years old but am still cutting wheels. The ancients and that which they could not transmit have both died, so as a result, what My Lord is reading is only ancients’ dregs and leavings.’ Zhuangzi jijie 1954: 13:79. Trans. Watson 2013: 106–107.

The Cicada-catcher [D4] When Confucius was on his way to Chu, he passed through a forest where he saw a hunchback catching cicadas with a sticky pole as easily as though he were grabbing them with his hand. Confucius said, ‘What skill you have! Is there (a) Way?’ ‘I have (a) Way,’ said the hunchback. ‘For the first five or six months I practise balancing two balls on top of each other on the end of the pole and, if they don’t fall off, I will lose very few cicadas. Then I balance three balls and, if

Can One Become Wise by Learning to Catch Cicadas?

311

they don’t fall off, I will lose only one cicada in ten. Then I balance five balls and, if they don’t fall off, I know it will be as easy as grabbing them with my hand. I position my body like a stiff tree trunk and hold my arm like an old dry limb. No matter how expansive heaven and earth are, or how numerous the ten thousand things, I’m aware of nothing but cicada wings. Not wavering, not tipping, not letting any of the other ten thousand things take the place of those cicada wings – how can I help but succeed in taking them?’ Confucius turned to his followers and said, ‘He keeps his intention undivided and concentrates his spirit – is this not what we say about the venerable hunchback?’ Zhuangzi jijie 1954: 19:3. Trans. modified from Watson 2013: 147.

The Dice and Ball Players [S1] Materials are indifferent, but the use that one makes of them is by no means indifferent. How, then, can one preserve firmness and calmness of mind, and at the same time the attentiveness that saves us from careless and thoughtless action? By following the example of those who play at dice. The counters are indifferent, the dice are indifferent. How can I know in what way the throw will fall? But to be attentive and skilful in making use of whatever does fall, that is now my task. And so likewise, my principal task in life is this: to distinguish between things, and establish a division between them and say, ‘External things are not within my power; choice is within my power. Where am I to seek the good and the bad? Within myself, in that which is my own.’ But with regard to what is not my own, never apply the words good or bad, and benefit or harm, and any other word of that kind. Experienced ball players can also be seen to act in such a way. None of them is concerned about whether the ball is good or bad, but solely about how to throw and catch it. It is there accordingly that the player’s agility, and skill, and speed, and good judgement are demonstrated . . . . But if we’re anxious or nervous when we make the catch or throw, what will become of the game, and how can one maintain one’s composure. Epictetus, Discourses 2.5. Trans. Hard and Gill 2014: 258–260.

The Actors [S2] Everyone, therefore, should acquire knowledge of his own talents, and show himself a sharp judge of his own good qualities and faults; else it will seem that actors have more good sense than us. For they do not choose the best plays, but those that are most suited to themselves. Those who rely on their voice choose the Epigoni and the Medus, those who rely on gesture, Melanippa and Clytemnestra; Rupilius, whom I remember, was always doing Antiope, while

312

Skill in Ancient Ethics

Aesopus did not often take part in the Ajax. If an actor, then, will observe this on the stage, will not a wise man observe it in his life? We shall, therefore, exert ourselves above all in those things to which we are most suited, if necessity has on occasion pushed us towards things that are beyond our natural talents, we shall have to apply all possible care, preparation and diligence so that we can perform them, if not in a seemly fashion, still with as little unseemliness as possible. Cicero, On Duties 1.114–16. Trans. Griffin and Atkins 1991: 44–45.

The Archer [S3] For just as, if it is someone’s purpose to direct a spear or arrow at something, we say that his highest goal is to do everything he can in order to direct it at [the target], in the same sense that we say that our highest goal is the good. The archer in this comparison is to do all that he can to direct [his arrow at the target]; and yet doing all that he can to attain his purpose would be like the highest goal of the sort that we say is the highest good in life; actually striking [the target], though, is as it were to be selected and not to be chosen. Cicero, De fin. 3.22. Trans. Inwood and Gerson 2008: 153.

1. Why is excellence in skills paradigmatic for the art of living? All these passages presuppose that there is a robust and non-accidental analogy between life and exercise of an art or skill, and that the excellence in the former can therefore be usefully exemplified by the excellence in the latter. The most fundamental similarity is that both kinds of excellences primarily aim at achieving a goal, or producing a change in the world, by means of an efficient use of the available materials, tools or resources in the context of given constraints and opportunities. So, just like skilful ball players aim at playing well by means of skilful catching and throwing of balls as they come (S1), and just like the best butcher carves up the ox by playing along rather than struggling against its bone structure (D1), so can a person skilled in living – the ‘sage’ (shengren 㚆Ӫ or sophos) – live well by means of skilful use of opportunities and constraints as they emerge in various circumstances of her life.5

5

Or as they are inherent in one’s personal psychophysical make-up, as (S2) suggests.

Can One Become Wise by Learning to Catch Cicadas?

313

Within this general framework, there are some more specific attributes of skill virtuosity that appear to have explanatory significance for excellence in living. One is that skill virtuosity typically goes with a certain effortlessness or ease, which is manifested in a smooth flow of the skilled person’s activities. The butcher does not get stuck on a complex joint, and the ball player does not get overwhelmed by a difficult ball. The more effortless the activity, the more precise and efficient it is. It is significant in this context that one of the Stoic definitions of happiness is ‘smooth flow of life’.6 One attractive aspect of the skill paradigm on the Stoic side might be precisely that the ease characteristic of the performance of analogies exemplifies the unruffled course of the sage’s life. The connection between the effortlessness of virtuoso exercise and effortlessness as the attribute of ideal action is even more explicit in the Daoist case: ‘non-interference’ or ‘effortless action’ (wuwei ❑⛪) as exemplified by virtuoso performances is to be achieved in the realm of human agency in general.7 This characteristic ease depends on another important attribute of skill virtuosity, namely on an evaluative detachment of the practitioners from external factors that are related to the exercise of the skill but beyond their power or competence. These include quality of materials, circumstantial constraints on one’s efficiency in handling them, and the positive or negative consequences of success or failure. Skilful ball players do not worry about whether their balls are good or bad; good actors don’t care about whether the characters they play are good or bad; a good ferryman does not worry about the possibility of capsizing the boat and its consequences. In both Stoic and Daoist texts, this evaluative detachment is articulated by reference to a dichotomy between what is internal, i.e. one’s skill and its exercise, and what is external, i.e. various circumstances that bear on the exercise of the skill but are ultimately not under one’s control. As Epictetus puts it, they are not ‘up to us’. There is, of course, a close causal connection between the evaluative detachment and the effortless efficiency: being relaxed about what one cannot control is the prerequisite for being able to focus fully on those things that one can control. These convergences are grounded in a more fundamental philosophical outlook that can be dubbed as ‘living with nature’ or ‘following nature’.8 Roughly, this ideal is based on two claims. The first is that the entire world is pervaded by a sublime, creative force which ensures that everything happens in accordance 6 7 8

Diogenes Laertius, Lives of Eminent Philosophers (=DL) 7.87. E.g. Zhuangzi jijie 1:6; 7:48. The Stoic definition of eudaimonia is to ‘live in accordance with nature’ (e.g. DL 7.87–89); Daoist texts talk about ‘following what-is-so-of-itself ’ (yin ziran ഐ㠚❦) (Zhuangzi jijie 5:35).

314

Skill in Ancient Ethics

with the cosmic order. Stoics and Daoists conceived this force very differently. Stoics talk about the God or Reason, whereas Daoists speak about Dao 䚃 or the Way – but we can see why some commentators were inclined to refer to this as ‘nature’ in both cases. For the Stoics, ‘nature’ (phusis) is one of the synonyms for the God; in the Daoist texts, the characteristic attribute of the Way is ‘being-soof-itself ’ (ziran 㠚❦) or ‘free from effort’ (wuwei ❑⛪), i.e. free from artifice.9 The second claim is that human beings have epistemological access to this force, and that they can use this access to consciously shape their agency to follow along with this force. So, according to the Stoics, the idea of living in agreement with nature means ‘always doing everything on the basis of the concordance of each man’s guardian daemon with the will of the administrator of the whole’ (DL 7.88=LS 63C). The inner daemon refers to a fragment of God in the human mind, which has a natural ability and tendency to guide humans to action that follows nature. Similarly, in Zhuangzi’s story of the butcher, the transition from ordinary to virtuoso performance is described as a shift from action based on superficial cognition of sense perception to action animated by a deeper cognitive grasp by a daemonic faculty (shen ⾎). One widely noted consequence of the ideal of following nature is the injunction to calmly accept even adverse circumstances as an inevitable and ultimately meaningful part of cosmic operations. This injunction is perhaps most suggestively brought out by the Stoic simile of a dog tied behind a cart.10 The dog will have to follow the cart, whether by being dragged, when it does not want to follow of its own will, or by willingly running along; so it is practically better, but also wiser, to ‘make its autonomous acts coincide with necessity’ (poion kai to autexousion meta tes anankes). The injunction to willingly accept or follow whatever is mandated by ‘fate’ (ming ભ) also appears at various places in the Daoist texts.11 But if this is the main consequence of the ideal of following nature for human action, then why were Stoics and Daoists so keen on the craft analogy? For the agency characteristic of skills or crafts does not typically simply accepts things as they are; quite to the contrary, it seeks to intervene into the given state of affairs and to produce a change in the world: to carve up an ox as yet uncarved, to hit a target as yet unhit. To answer this question, we should first point out that ‘nature’ that should be followed is not merely a static constellation of things; it is, as noted, a creative

9 10 11

Zhuangzi jijie 6:38. Hippolytus, Refutation of all heresies 1.21 (=LS 62A) E.g. Zhuangzi jijie 5:31; 6:42.

Can One Become Wise by Learning to Catch Cicadas?

315

force with a characteristic modus operandi. Whatever this creative force does or causes, it does it in a peculiar style or manner. To act or live in accordance with nature means, then, acting or living in precisely this manner. One’s actions must embody certain attributes or qualities that are characteristic for the way nature operates. Perhaps the most general account of this cosmic modus operandi is what I have called ‘creativity’, and in both Stoicism and Daoism this is understood in terms of a craftsmanlike force. In the Stoic texts, this is brought out clearly by the definition of craft (technē), i.e. a ‘tenor which achieves everything methodically’, which is a definition that applies also to nature.12 The Zhuangzi repeatedly refers to nature by using the term ‘that what creates things’ (zaowuzhe 䙐⢙㘵) or ‘that what creates transformations’ (zaohuazhe 䙐ॆ㘵), and likens this force to a smith.13 So if creativity or craftsmanship is the characteristic modus operandi of nature, and this is what humans should follow or embody, then it is not at all surprising that excellence in crafts and skills was regarded as being structurally akin to the excellence in living. There are further attributes on both Stoic and Daoist side that specify the creativity as the characteristic modus operandi of nature, such as proceeding ‘methodically’ on the Stoic side and ‘effortlessly’ on the Daoist side.14 These attributes characterize also the activity of virtuoso practitioners and, at the same time, human agency at its best. So the ideal of following nature does not necessarily promote complacency or resignation, but rather valorizes the kind of agency that embodies the modus operandi of nature. Still, the injunction to accept things as they are is fully compatible with this more fundamental perspective on following nature. Again, skill virtuosity is well suited to bring out why this is the case. Even though the objective of exercising craft or skill is to produce a change in the outside world, this can be achieved most effectively when the craftsmen fully focus on their internal exercise of the task at hand, rather than on external variables. The ability to accept circumstances as they are is what distinguishes ordinary from virtuoso performance of craft. It is also the prerequisite for fully enacting in person the modus operandi of cosmic nature. Insofar as the virtuosity is skills exemplifies how detachment and non-intervention can maximize achievement and efficacy, skills indeed appear to be an attractive explanatory paradigm. When individual human agency itself embodies the modus operandi of cosmic nature, then it

12 13 14

Olympiodorus, On Plato’s Gorgias 12.1 (=LS 42A). Zhuangzi jijie 6.41(2x); 7.46. Ibid.

316

Skill in Ancient Ethics

becomes virtually at one with this creative force, and hence it does not, strictly considered, intervene in the natural development of affairs, but rather helps to realize it. The change produced by human artifice then becomes one of the universal transformations caused by the quasi-artificial creative force of nature.15 It is when one’s action is most focused, most precise and most efficient that it ceases, properly speaking, to be an activity of an individual human being and becomes at one with the universal creative force. Having established the common motivation for embracing skill analogies on the background of the broad ideal of following nature, I shall now turn to several differences pertaining to how the analogy is employed on each side.

2. The differences in the Stoic and Daoist craft analogies The first point of difference concerns the favoured types of crafts or skills. On the Daoist side, we can note a preference for carving-like crafts, whereas on the Stoic side there is a pronounced interest in the so-called stochastic crafts, i.e. those concerned with aiming at a target or goal, such as archery, but also navigation or medicine. It is characteristic for the stochastic crafts that whether they count as being exercised well does not depend on whether they actually achieve their goal.16 A virtuoso archer may fail to hit the target without compromising his virtuosity, for the wind may intervene and divert his arrow. A doctor may fail to save his patient, and yet this does not detract from the fact that he has practised his skill impeccably. The stochastic arts, notably archery, also appear in the Daoist texts, but this peculiar feature of stochastic arts is never philosophically exploited. We find no interest in the Daoist sources in the fact that even a virtuoso practitioner can fail. Rather, the chosen type of crafts, particularly carving-like crafts, imply that one’s virtuosity can be measured by the actual success or failure in achieving of one’s goal. The butcher has a direct contact with the bone structure, and there is no room for an external intervening element that would do the job of the wind in the archery analogy. Translated to the context of the human life, this indicates a special concern on the Stoic side with the possibility that even sage’s ‘unruffled’ life may become difficult when it comes to the balls that the fate throws at them. Aristotle famously 15

16

This is suggestively captured by the engraver’s account (not included in the above selection of texts; Zhuangzi jijie 19:11), who describes his agency as an activity through which nature (or ‘Heaven’) acts upon itself. Inwood 1986; Striker 1996, ch. 14.

Can One Become Wise by Learning to Catch Cicadas?

317

argued that virtue is necessary, but not sufficient condition for happiness.17 In order to be able to fully exercise virtuous activities, the virtuous person also needs a reasonable amount of external goods (health, wealth etc.). Against this view, the Stoics vigorously defended the view that virtue is not only necessary but also sufficient for happiness. In other words, life may be extremely unfortunate but still happy.18 The structure of the stochastic crafts is well suited to bringing out this point. The experts in stochastic crafts may embody the cosmic modus operandi (e.g. shoot methodically), and yet fail to achieve their goal (e.g. to hit the target). Now, if the ultimate goal of the sage is to act in accordance with nature, which is in his power, and if happiness lies in that, then the fact that the sage fails to achieve certain goals, which one would for the most part achieve by means of acting in accordance with nature, but whose achievement is not fully in his power, does not detract from his happiness. This claim rests on a distinction between two different goals, internal and external: the internal goal of archer’s exercise is to exercise his skill impeccably, i.e. to shoot well; the external goal is to hit the target (S3).19 On the level of the Stoic craft of living,20 the external goals belong to the realm of what the Stoics called ‘preferred’ (proegmena) but ultimately ‘indifferent’ (adiaphora) things, such as health or wealth.21 For Aristotle, these count (at least to some degree) as necessary for happiness; for the Stoics, the possession of these things is preferred, given the natural constitution of humans, but ultimately indifferent for happiness. Other things being equal, it makes good sense to aim at them, but what ultimately (and exclusively) matters for happiness is how we shoot – how we deliberate and act in order to obtain them – not whether we actually obtain them. Our happiness is defined not so much by what we do or what we achieve, but how we proceed to achieving it. This idea may seem quite counterintuitive in the context of skills (what ultimately matters in medicine is to save patient’s life, not to do everything in our power to save it), but it has a more prima facie plausibility when we think of games. It is quite plausible to say that the ultimate goal of playing tennis, for instance, is not to win a match, but to play well, or to enjoy a good game. It makes good sense to say that we try to win the match in order to play well. The prospect of winning provides a referential point that motivates and structures our play, but it is not the actual goal of the play.

17 18 19 20 21

E.g. Nicomachean Ethics 10.8, 1178b34 ff. E.g. DL 7.101–105 (=LS 58A) See n.7. The Stoic idea that philosophy is a ‘craft of living’ is attested, e.g., in Epictetus, Diss. 1.15.2. See n. 9.

318

Skill in Ancient Ethics

The second point of difference is closely related to the first and concerns the materials of craft. Daoist accounts of crafts that involve carving consistently tend to describe the material cut or carved as being structured or patterned in ways that guide the craftsman’s hand. Master practitioners do not impose new structures on their material, but rather bring out the patterning that is already inherent in them (D1, in particular). In the Stoic account, the attitude toward materials is, at best, indifferent. Fate often throws curve balls at us; but a cow’s bone structure lends itself to the knife, provided the knife respects it. There are difficult spots, but if they are handled carefully, success is guaranteed. Is the implication that the circumstances encountered by the Daoist sages in the course of their lives always permit them to get through unscathed? And is this condition of staying unscathed somehow equivalent to the Stoic external goal, i.e. that of actually hitting the target? There is, ultimately, no conclusive textual evidence to answer these questions, but I am inclined to answer both questions in the affirmative on Zhuangzi’s behalf. There is a recurrent emphasis in the text on physical self-preservation as something that the sages can achieve with greater efficacy than non-sages, even though, ultimately, death is not to be feared.22 The idea that sages will never meet with misfortune, as long as they follow the natural patterns, might seem unrealistic, but we should bear in mind that the Daoist sage is often portrayed as a character that transcends the notion of the human being, e.g. does not get burned.23 My third contrast concerns the transmissibility of skill, especially by means of language. Zhuangzi’s point in the story of Wheelwright Bian (D3) is that skill comes from first-hand experience with material, and that experience cannot be mediated easily by means of language. If this passage refers more broadly to expertise in living, it would follow that the art of living cannot be transmitted from person to person, and that words are the least suitable vehicle of transmission. Rather, this expertise must be developed gradually by means of long-term lived experience of close contact with the material. This implication is supported throughout the Zhuangzi by a fairly reserved view of language as a suitable epistemological means to understand and navigate the world.24 One of the attractions of the skill analogy from the Zhuangzian point of view is precisely that there are obvious limitations in what linguistically mediated knowledge can

22

23 24

See Zhuangzi jijie 3:18; 19:7 for the greater efficacy of sages; and Zhuangzi jijie 2:16; 6:41-3 for the idea that death is not to be feared. Zhuangzi jijie 19:6. See, especially, the second chapter of the Zhuangzi. A more concessive view on the role of language is implied at Zhuangzi jijie 27:66, where words are likened to stakes used to catch fish.

Can One Become Wise by Learning to Catch Cicadas?

319

contribute to mastering a skill. Virtuoso craftsmen need trained eyes, sensitive ears and responsive hands – none of which can really be achieved by means of the language-based erudition. This outlook is in a stark contrast with the Stoic view of crafts in general, and the craft of living in particular. No matter how much extralinguistic practice is needed to fully master a skill, for the Stoics technē essentially amounts to a system of true propositions relevant to its subject.25 Being fully expressible in language, these propositions can be verbally transmitted from person to person, and hence one can (indeed must) go a long way towards mastering a skill by means of language-based instruction. This applies also to the craft of living, which, the Stoics insist, consists of a body of knowledge comprising three basic disciplines of philosophy: logic, physics and ethics. Some scholars have rightly emphasized that the excellence of the Stoic sage has an important dimension of ‘knowing how’, rather than merely ‘knowing that’, insofar as it amounts to a skill of assenting correctly to the impressions that impinge on one’s mind.26 But besides this ‘dispositional’ aspect, the Stoic craft of living also has a ‘cognitional’ aspect.27 And it is this cognitional aspect that ultimately grounds the dispositional aspect.28 The sage can infallibly decide whether a given impression should be assented to or not on the basis of judging whether the propositional content of the impression is logically compatible with the system of propositions that constitute his appropriated body of knowledge, which amounts to ‘wisdom’ or ‘philosophy’.29 I shall return to the specific body of knowledge that constitutes wisdom in the last section of this chapter.

3. The common denominator behind the differences I suggest that all the above differences in employment of craft analogies can be explained by a different understanding of the cosmic modus operandi in Stoic and Daoist thought, and that this difference, again, points to a difference in fundamental dichotomies that informed philosophical debates in both traditions. Perhaps the most direct pointer to the Stoic understanding of the cosmic modus 25

26 27 28 29

Technē is defined as a ‘systematic collection of cognitions (katalepseis)’; ‘cognitions’ are assents to cognitive impressions that have always propositional form in case of human adults (DL 7.51). Insofar as wisdom (or virtue) is a technē, it is teachable by means of linguistic instruction (DL 7.91). Liu 2009. Brouwer 2014: 32–33. Vogt 2008: 127–129. Aetius 1, Preface 2 (LS 26A); DL 7.39–41 (LS 26B).

320

Skill in Ancient Ethics

operandi is the idea that expertise can be transmitted by means of language or speech (logos). This testifies to the view that the creative force of nature, as well as the creative force operative in crafts, is conceived in terms of rationality, i.e. the force that finds its privileged expression in language and speech. Language is the suitable medium to transmit the knowledge necessary for acting in accordance with nature because nature itself is essentially linguistic. It is a peculiar and uniquely Stoic claim that the universal, physical creative force is logical in its nature, which is why the physics and logic as parts of philosophy are merely two different aspects of the same body knowledge, rather than two different disciplines.30 The Stoic understanding of rationality draws on the association of reason with teleology that is prominent, among others, in the philosophy of Plato and Aristotle.31 A rational ordering of elements within a whole is such that these elements stand to each other in hierarchical relations that are conducive to the achievement of the end of this whole. But Stoics accentuate those aspects of rationality that are specifically linked to speech or discourse, particularly consistency, firmness or irrefutability. The Stoic universe holds together in the same sense that a well argued, consistent rational argument does, and the necessity that arranges the worldly events is the same necessity with which certain conclusions inevitably follow from certain premises. The notion of rationality played a central role in Ancient Greek and Roman philosophy, and it was typically contrasted with non-rationality. In the domain of the human psychology, this is reflected in the division of the soul in rational and non-rational parts, as we find it, most notably, in Plato and Aristotle.32 This dichotomy is of course value-laden. The activities of the rational part of the soul, such as thinking or reasoning, have more worth, and are more directly and substantially conducive to the human flourishing, than the activities of the nonrational parts, such as perception or digestion. Similarly, on the metaphysical or cosmological level, it is those aspects of reality that have more rational or reasonrelated attributes such as regularity, form, and determination that exceed in worth those less rational aspects of reality that are more irregular, more material and less determined by form. When the imagery of skills or crafts was invoked in the Greek philosophical tradition, it was typically understood and exploited in terms of the dichotomy 30

31 32

‘The cosmos is logical’ (Sextus Empiricus, Against Math. 9.104); ‘hence it follows that the cosmos possesses wisdom, and that the nature that embraces all things is pre-eminently and perfectly rational (perfectione rationis excellere)’ (Cicero, On the nature of gods, 2.30; see also ibid. 1.39). For the relationship between different disciplines within philosophy, see n. 18 above. E.g. Cicero, On the nature of gods 2.37–9. E.g. Plato, Resp. 4 or 9; Aristotle, Nicomachean Ethics 1.13.

Can One Become Wise by Learning to Catch Cicadas?

321

between rational and non-rational. The paradigmatic case is Plato’s demiurge from the Timaeus, who imposes rational forms and proportions on disordered matter. Both Plato and Aristotle conceived of matter as essentially recalcitrant, with a significant potential to impede or thwart the craftsmanlike workings of reason. This holds also for the ‘matter’ in the sense of life circumstances that encounter us. In the Nicomachean Ethics 1, for instance, Aristotle draws an analogy between a shoemaker’s or general’s ability to make the best possible use of the limited material resources at their disposal with the virtuous person’s ability to live as well as possible in the adverse circumstance of life. Aristotle makes two points here: first, like master practitioners, the virtuous person can always make the best possible use of available resources; second, a virtuous person’s performance will be significantly limited if these resources are seriously limited. From this perspective, we can appreciate more clearly why the stochastic skills were attractive for the Stoics. Unfavourable external circumstances can prevent the practitioners of stochastic skills from achieving external goals, but they cannot thwart the achievement of internal goals such as exercising one’s skill impeccably. A doctor’s expertise isn’t necessarily compromised if her patient dies; what matters for her excellence qua doctor is that her interventions were lege artis. Similarly, the expert in the craft of living always deliberates impeccably when it comes to the selection of her naturally preferred matters of indifference and everything that is conducive to it under the given circumstances. Whether this deliberation actually turns out to be successful in terms of achieving the intended result does not add to or detract anything from her happiness, which lies entirely in the rational deliberation with view of achieving that end. The threat of non-rationality inherent in the contingency and unpredictability of materials is eliminated by insulating one’s happiness from factors outside of one’s control. The Daoists in general, and the Zhuangzi in particular, were sometimes labelled in the scholarship as anti-rationalists, mainly on the grounds of their sceptical stance towards language.33 If this is right, then we might have to do with two antithetical perspectives on the ‘art of living’: rationalist and anti-rationalist. In the anti-rationalist or Daoist version, skills exemplify wisdom insofar as they are empirical knacks, whereas the rationalist or Stoic version emphasizes the underlying concept-based knowledge. But the interpretation of Zhuangzi as anti-rationalist is seriously flawed because the importation of notions of ‘rationality’ and ‘anti-rationality’ into early 33

Carr and Ivanhoe 2000; Graham 1989: 7, 176.

322

Skill in Ancient Ethics

Chinese context is deeply problematic. These notions, insofar as they are closely associated with the notion of teleology, and insofar as they are associated with the view that the essence of reality, or at least some domain of reality, is essentially linguistic, is distinctly Greek, and cannot be implanted into a Chinese context without being deprived of much of its richness and distinctness. Likewise, the early Chinese tradition had its own distinctive dichotomies and concerns, which could not be implanted into the Greek tradition without a loss. It is, at the most fundamental level, the dichotomy between ‘what is so-of-itself ’ (ziran 㠚❦) and what is ‘made’ (wei ⛪). This dichotomy corresponds to two different modalities of agency, a spontaneous and/or effortless action (wuwei ❑⛪) that ‘gives birth’ (sheng ⭏) to things and artificial action that involves effort (youwei ᴹ⛪) and ‘creates’ (zuo ֌) things.34 The artificial human effort was often (though not always) understood as the characteristic modus operandi of humans (ren Ӫ), in contrast to spontaneous generative processes associated with nature or Heaven (tian ཙ).35 Many major philosophical debates in early China are more or less directly embedded in these dichotomies. For instance, the well-known debate in the Confucian camp about whether the ‘human nature’ (xing ᙗ) is good or evil was essentially one about whether there is some moral value in our innate motivations, and hence whether there is value in acting spontaneously, or whether moral value is something that must be construed by human artificial effort in order to discipline or supersede the wayward natural inclinations. Unlike the dichotomy between rational and non-rational, where the rational was clearly valorized over the nonrational, much of the discussion in the Chinese context revolved around the question whether it is the natural or the artificial that should be more highly valued. As in the context of Greek tradition, the adoption of the imagery of crafts and skills in ethical contexts can be traced back to the underlying dichotomy. In the Greek tradition, craftsmen’s methodical, goal-oriented way of handling materials was the paradigmatic expression of the value of rationality and its powers, and limits, in containing the non-rational disorder and contingency. In the Chinese context, craft was paradigmatic for the artificial human effort that intervenes, for better or for worse, into how things naturally are of themselves. As such, it was invoked, for instance, in the Confucian texts to bring out the view that the process of self-cultivation consists in refining the unhewn human nature.36

34

35

36

Two monographs that offer, from different perspectives, narratives of early Chinese thought as informed by this dichotomy are Slingerland 2003 and Puett 2001. In Mohist thought, Heaven was envisaged as a creative force that operates in the modality of artificial effort (see. esp. the chapter Tianzhi zhong). Xunzi jijie 1–2.

Can One Become Wise by Learning to Catch Cicadas?

323

The significance and uniqueness of Zhuangzi’s adoption of craft analogies lies in challenging the widely held view that crafts, paradigms of human artificial effort, must be interventionist; only mediocre practitioners accomplish their goals by struggling with their materials. The master butcher’s way of carving the ox is so efficient precisely because it does not interfere with the animal’s bone structure, but rather follows it. The favourable outlook on material as inherently user-friendly is not so much intended to support the claim that the sage is immune to misfortunes caused by unfavourable external circumstances, as in Stoicism, but rather that any superior human intervention cannot be understood in terms of refining the way things are. For the things the way they are of itself are already supremely refined. Likewise, the critical attitude to language has to be understood in the context of the dichotomy between nature and artifice. Whereas in the Greek context, language is associated with rationality as metaphysical or cosmological category, in the Chinese tradition language has an affinity with the domain of what is human, made and artificial. Correspondingly, the prescriptivist, constructivist role of language is much emphasized. In the early Chinese context, language is predominantly understood not so much as a tool for disengaged description of reality, but as a tool for construction and implementation of social, political and moral norms. Perhaps the most exemplary and familiar case of this outlook is Confucian idea of ‘rectification of names’ (zheng ming ↓਽).37 To rectify names means that we shall describe as ‘ruler’ or ‘son’ only those rulers or sons that actually live up to the demands of these roles. The correct names should be those that do not describe but prescribe for the sake of political and educational purposes.38 Now, since the goal of the Daoist craft of living is not constructivist, it is not surprising that language as the tool of moral construction cannot be of much use in acquisition of this excellence. Being mere ‘guests of reality’ (shi zhi bin ሖѻ䌃) (Zhuangzi jijie 1:3), names cannot claim for themselves any authority to prescribe what the reality should look like.

4. Why the specialist crafts can (or cannot) be conducive to wisdom One of Zhuangzi’s reasons for the wide employment of craft imagery is to provide an account of how one becomes wise. The butcher, the engraver, the 37 38

Xunzi jijie 411–433. For an interpretation along these lines, see Hansen 1992: 65–71.

324

Skill in Ancient Ethics

cicada-catcher all offer a narrative of their gradual advancement towards virtuosity. The training in a manual craft or skill encompasses a training of mind, or, more precisely, a training in how to ‘use one’s mind’ (yong xin ⭘ᗳ)39 and how to ‘respond to things’ (ying wu ៹⢙).40 The goal is to develop such a psychophysical hygiene that one becomes a ‘dwelling place for the spirit (shen ⾎)’.41 This realizes the potential of the mind to become at one with all existing things.42 Doing so eliminates one’s personal biases and puts one in a position to use one’s mind ‘as a mirror’ (ruo jing 㤕䨑; Zhuangzi jijie 7:49), ‘without effort’ (bulao нऎ; Zhuangzi jijie 22:28), and to respond to things ‘without constraints’ (wuqiong ❑マ; Zhuangzi jijie 2:10). Elsewhere, Zhuangzi describes this process as ‘fasting of the mind’ (xinzhai ᗳ啻), where one ‘listens’ to other things by ‘vital energy’ (qi ≓) that runs through all of them (Zhuangzi jijie 4:23). Learning a craft or skill is, presumably, not the only way to build such receptivity of mind, but Zhuangzi no doubt presents it as an exemplary way. It serves well their intention to affirm a way to wisdom that would serve as an alternative to the Confucian model of self-cultivation in terms of an appropriation of noble and normative ideas from the ancient classics. Rather than expanding our ‘knowledge’ (zhi ⸕), we should purify our mind and ‘forget’ (wang ᘈ).43 Learning a skill or craft, in particular a manual one, can serve as a regimen of spiritual transformation in which the sustained repetition of certain movements, and the concentration on a clearly defined set of tasks, helps one to liberate one’s mind from of all emotional baggage that gets into the way of free interaction with things. By learning a skill, we build those muscles in our mind that are engaged in the craft of living. The idea that a psychophysical regimen associated with learning a craft is well suited to develop the capacity to ‘use one’s mind’ well is clearly embedded in and informed by the dichotomy between natural and artificial. The major obstacle for using one’s mind well are the recalcitrant sediments of conventional values and man-made discriminations in it, as well as excessive emotions that they generate, illustrated by the ‘thickness’ of knife of the mediocre butcher that fails to find its way through the narrow spaces inbetween the bones. Unlike the Zhuangzi, Stoics were quite explicit (as we have seen above) that there is a ‘craft of living’ which is independent from individual crafts; this is

39 40 41 42 43

Zhuangzi jijie 5:30; 7:49. Zhuangzi jijie 14:83; 22:28. Zhuangzi jijie 4:22; 22:27. Zhuangzi jijie 6:41–45. Zhuangzi jijie 2:17; 3:20; 6:45.

Can One Become Wise by Learning to Catch Cicadas?

325

wisdom (sophia), philosophy (philosophia), or knowledge (episteme). From the perspective of its content, wisdom is defined as an ‘unshakable and consistent system of cognitions’ (Stob. 2.73.19–4.3=SVF 3.112). This system of cognitions is specified in terms of its scope and subject matter. It consists of a triad of excellences in physics, ethics and logic: ‘Physics is practised whenever we investigate the world and its contents, ethics is our engagement with human life, and logic our engagement with discourse, which they also call dialectic’ (Aëtius 1, Preface 2=LS 26A). These excellences are closely interrelated, and this interrelatedness is reflected in the Stoic idea of unity of philosophy, where logic is likened to ‘bones and sinews, ethics to fleshier parts, and physics to soul’ (DL 7.39–40=LS 26B). The capacity of the wise person to assent as she should (i.e. the dispositional aspect), as well as the related disposition to act well, are owed to a specific body of knowledge. So, for instance, the sage will not assent to an impression like ‘Death is bad for me, and hence it is appropriate for me to be afraid of dying’ as true, because she has the relevant pieces of knowledge from the domain of ethics, such as ‘only virtue is good and only vice is evil’, as well as physics, such as that everything is providentially determined by the god, that enable her to infallibly – also due to her expertise in logic – recognize this impression as false. So the important contrast in conceiving the epistemic condition of wisdom is that for the Stoics, it is content-dependent, whereas for the Zhuangzi, it is not necessarily so. Zhuangzi’s wisdom as the excellence in knowing how to respond to natural patterns is not premised on a specific knowing that in any obvious sense, whereas for the Stoics it is. This explains why the Stoics believed that the road to wisdom cannot consist in learning any skill whatsoever. For Zhuangzi, in contrast, the content of one’s knowledge does not ultimately matter. But what explains these different views about the content of wisdom? Why do the Stoics insist on the idea that wisdom is constituted by a special, unique body of knowledge? There may be several different lines of explanation available, and I would like to suggest only one of them. The content-dependence of the Stoic wisdom follows from the Stoic (and more generally Greek) valorization of rationality. Rationality as a cosmological and ethical ideal is, in its very definition, inexorably tied with the phenomenon of language and speech. To the extent that language is the privileged vehicle of knowledge, knowledge will be predominantly understood in propositional terms, i.e. in terms of knowing that something is the case. In contrast, Zhuangzi’s ideal of wisdom wasn’t informed by the distinctly Greek and Roman notion of rationality and its valorization, but rather by a valorization of what-is-so-of-itself, in contrast to what is the product of artificial

326

Skill in Ancient Ethics

human effort. This naturally led to the ideal of wisdom in terms of a gnosis that is not language based, and hence not tied to a particular content.

References Brouwer, R. (2014), The Stoic Sage: The Early Stoics on Wisdom, Sagehood and Socrates, Cambridge: Cambridge University Press. Carr, K.L., and P.J Ivanhoe (2000), The Sense of Antirationalism: The Religious Thought of Zhuangzi and Kierkegaard. Seven Bridges Press. Graham, A.C. (1989), Disputers of the Tao, La Salle: Open Court. Griffin M.T., and E.M. Atkins, eds (1991), Cicero: On Duties, Cambridge: Cambridge University Press. Hansen, C. (1992), A Daoist Theory of Chinese Thought: A Philosophical Interpretation, Oxford: Oxford University Press. Hard, R., and C. Gill, eds (2014), Epictetus: Discourses, Fragments, Handbook, Oxford: Oxford University Press. Inwood, B. 1986. ‘The Goal and Target in Stoicism’. The Journal of Philosophy 83: 547–556. Inwood, B., and L. Gerson (20080, The Stoics Reader, Indianapolis: Hackett. King, R.A.H., ed. (2015), The Good Life and Conceptions of Life in Early China and Graeco-Roman Antiquity, Berlin: De Gruyter. Lai, K. and W.W. Chiu, eds (2019), Skill and Mastery: Philosophical Stories from the Zhuangzi, Lanham, MD : Rowman & Littlefield. Lai, K., R. Benitez and H.J. Kim, eds (2018), Cultivating a Good Life in Early Chinese and Ancient Greek Philosophy, London: Bloomsbury. Liu, I. (2009), ‘Nature and Knowledge in Stoicism. On the Ordinariness of the Stoic Sage’, Apeiron 41: 247–275. Machek, D. (2018), ‘Stoics and Daoists on Freedom As Doing Necessary Things’, Philosophy East and West  68 (1): 174–200. Puett, M. (2001), The Ambivalence of Creation: Debates Concerning Innovation and Artifice in Early China, Stanford: Stanford University Press. Raphals, L. (2005), ‘Craft Analogies in Chinese and Greek Argumentation’, in Eric Ziolkowski (ed.), Literature, Religion, and East/West Comparison: Essays in Honor of Anthony C. Yu, Cranbury, NJ : Rosemont Publishing, 181–201. Slingerland, E. (2003), Effortless Action: Wu-wei as Conceptual Metaphor and Spiritual Ideal in Early China, New York: Oxford. Striker, G. (1996), Essays on Hellenistic Epistemology and Ethics, Cambridge: Cambridge University Press. Vogt, K. (2008), Law, Reason and the Cosmic City: Political Philosophy in the Early Stoa, New York: Oxford University Press.

Can One Become Wise by Learning to Catch Cicadas?

327

Watson, B. (2013), The Complete Works of Chuang Tzu, New York: Columbia University Press. Yu, J. (2008), ‘Living with Nature: Stoicism and Daoism’, History of Philosophy Quarterly 25 (1): 1–19. Xunzi jijie 㥰ᆀ䳶䀓, (1988), ed. Wang Xianqian, Beijing: Zhonghua shuju. Zhuangzi jijie 㦺ᆀ䳶䀓, (1954) ed. Wang Xianqian, Beijing: Zhongua shuju.

328

16

Gendered Skill: Chinese and Greek Skill-Knowledge Analogies from Archery and Weaving Lisa Raphals

Weaving and archery are strongly gendered skills, and both occur repeatedly in both Chinese and Greek accounts of skill and ethics.1 I examine both metaphors and narratives that liken these skills to various aspects of ethics, wisdom and government, with particular interest in how or whether the account of the skill reflects the experience of the gender of its typical expert. Before turning to archery and weaving metaphors, it is useful to make some important distinctions between accounts of skill in similes, analogies and metaphors. A simile is an explicit comparison in which one thing is likened to another on the basis of some shared quality, for example: ‘my love is like a red red rose.’ Analogy is a mode of argument by which one thing is likened to another with the implication that the similarity is structural or systematic. An argument by analogy infers from one similarity between two things that they will correspond in other respects. Analogical reasoning draws conclusions about an unknown thing from information about a known thing it resembles. A metaphor is a figure of speech in which one thing is described in terms of another, to which it is understood to be analogous, for example: ‘my love is a rose.’ Putting these together, we can understand metaphors as compressed similes and similes as expanded metaphors.2 This is not quite correct, because a metaphor may be – and often is – literally false, but metaphorically true. For example, the simile ‘my spouse is like an angel’ may be literally true, while the corresponding metaphor, 1

2

This paper is dedicated to Anthony C. Yu (Yu Guofan ։഻㰙, 1938–2015). It is an expansion of two sections of Raphals 2005. In the Poetics, Aristotle described analogy as one of four kinds of metaphor and describes similes as metaphors needing an explanatory word (Poet. 1407a). In Institutio Oratoria, Quintillian described metaphors as shortened similes (Inst. 8.6.8–9).

329

330

Skill in Ancient Ethics

‘my spouse is an angel’, may be metaphorically true but literally false. An analogy is thus an argument or mode of reasoning based on simile, and also must be literally true. Against a history of being treated as a literary or rhetorical device, recent research has underscored the importance of metaphors in scientific reasoning, philosophy, and as a basic principle of human cognition and understanding.3 Under the influence of George Lakoff and Mark Johnson (1980 and 1999), the study of conceptual metaphors has emerged as a major field. Lakoff and Johnson claim that most of our normal conceptual schemes are ‘metaphorically structured’ (1980: 115). They argue that ordinary language is so immersed in conceptual metaphor that it would be very hard to do without it: ‘If we consciously make the enormous effort to separate out metaphorical from non-metaphorical thought, we probably can do some very minimal and unsophisticated nonmetaphorical reasoning. But almost no one ever does this’ (1999: 59). They also argue that conceptual metaphors are universal because they are shaped by the shared physical experience upon which they all draw, for example of ‘up’ and ‘down’, ‘hot’ and ‘cold’, spatial metaphors for time, etc. They describe this shared basis in the physical experiences as ‘embodiment’ in the claim that ‘the mind is inherently embodied’, and to that extent, derive conceptual commonality from shared experience (Lakoff and Johnson 1999: 5). In addition to the core claim that root metaphors draw on bodily experience, many of these structures either describe the body, mind, spirit or person in other metaphors, or use elements of these in other metaphorical structures. A range of studies have shown the importance of metaphors in China.4 Cognitive metaphor theories and accounts of supposedly universal cognitive metaphors have been criticized for inattention to the cultural context of analogies and metaphors. Literary metaphors, by contrast, draw on nuances and resonances that depend on the social, historical and literary contexts of the literary and cultural traditions that use them. (They may also draw on the embodied experiences that underlie conceptual metaphors.) For all these important differences, both conceptual and literary metaphors share several common features that bear importantly on the understanding of skill and ethics. Both analogies and metaphors are based on perception of similarity; such perceptions can be surprising or striking if the similarity is not immediately obvious.

3

4

Black 1954–55 and 1977; Burrell 1973; Fauconnier and Turner 2002; Johnson 1981; Searle 1979; Turner 1988. Allan 1997; Link 2013; Raphals 2015; Reding 1997 and 2004; Slingerland 2019.

Gendered Skill: Chinese and Greek Skill-Knowledge Analogies

331

Culturally shared – or incompletely shared – similes, metaphors and analogies can provide an important basis for comparison. Analogies between wisdom, virtue and good government and craft(s) or skills are pervasive in both Greek and Chinese accounts of virtue and political excellence. These analogies invoke a variety of established arts, crafts, technē or branches of specialized knowledge. Greek craft metaphors such as the ‘body politic’ and ‘ship of state’ have become so commonplace as to conceal their sources in Platonic analogies (Bambrough 1956: 101). A wide range of craft analogies are equally well established in Chinese Warring States (475–221 bce ) and Han dynasty (202 bce– 221 ce ) philosophical works and in metaphors hidden in common expressions. The role of analogy has also become an element of broad arguments about the nature of reasoning in China and Greece.5 Many analogies and metaphors based on skill and craft are what I will call skill-knowledge analogies that compare mastery of a skill or craft to the practice or process of acquisition of knowledge, either of a specific skill, or to broad ‘skills’ of wisdom. Most skill-knowledge analogies describe skills that historically are the purview of men, and very few draw on skills primarily gained by women. Weaving is a significant exception. Chinese philosophical works frequently use the mastery, acquisition or skilled performance of crafts or skills as analogues for sageliness or good government. These analogies identify knowledge, wisdom or realization of dao either with the mastery or acquisition of the skill or with the exemplary character of the skilled individual. These analogies praise (or blame) a variety of experts (or maladroits) in the arts of: archery, cooking and butchering, engineering and technological innovation, farming, hunting and trapping, medicine, prediction or prognostication, military strategy, music, swordsmanship, weaving and wheelmaking (Raphals 1992: 204). Some Chinese accounts highlight characters who have been marginalized from, or who have rejected, the political world and the public domain. They also include accounts of the lack of some crucial skill. Many Chinese narratives describe wisdom and virtue as a skill, identified positively with ‘brightening virtue’ (ming de ᰾ᗧ, Maspéro 1933), but other narratives describe virtue as the absence of positive skill, and linked to sage rulers who ‘act by not acting’ (wuwei ❑⛪). Analogies to specific crafts first become prominent in the Mengzi and Zhuangzi. Mencius uses craft analogies from farming, carving, archery and engineering in order to advocate socially constructive technical skills that work in accordance with the nature of their 5

Cikoski 1975; Maspéro 1928; Reding 1986; Volkov 1992.

332

Skill in Ancient Ethics

objects. The Zhuangzi skill stories liken the mastery of the Way (dao 䚃) to the process of mastery of a craft (rather than on skilled performance). These analogies are striking because of their emphasis on the skills of commoners; they bypass the elite skills of archery, hydraulics or government in favour of the skills of butchers, wheelwrights and cicada-catchers (Raphals 2005; Lai and Chiu 2019). Greek analogies between virtue, craft and skill begin with concepts of the world as the product of intelligent design. Plato’s cosmology repeatedly invokes images of the craftsman of the cosmos (Lloyd 1966: 275–286). Most Greek craft imagery focuses on practitioners of received crafts. Given Plato’s hostile depictions of craft and art, it is of some interest that he refers to practically every technical art extant at his time (Brumbaugh 1989: 196–198; Lloyd 1966: 343– 349). Turning to specifically philosophical craft analogies, Plato (Rep. 495de, 522b, 590c) and Aristotle (Pol. 1277a37, 1278a6, 1328b39) show a striking contrast between their low regard for crafts and craftsmen and their repeated use of crafts imagery. I examine a range of Chinese and Greek analogies and metaphors based on comparison of the mastery, practice, or acquisition of two technical crafts, archery and weaving, and the mastery, practice or acquisition of wisdom or ethical or political virtue. Archery and weaving are both embodied practices that have strong cultural and even mythological resonances. Each is culturally important in both Chinese and Greek contexts. Both are gendered, and the performance of each is an important aspect of gendered virtue. Skill in archery is primarily attributed to men, while weaving is primarily performed by women. Both are also the basis of powerful metaphors. Finally, analogies to both are used extensively in philosophical argument. The next four sections turn to discussions of Chinese and Greek metaphors based on archery and weaving.

1. Chinese archery metaphors Accounts of archery in Chinese textual and material sources are both rich and ancient. Archery appears in early Chinese sources in a wide range of metaphors and analogies: for government, for virtue, for spontaneity or naturalness, and also for small or trivial knowledge. It is usually, but not always, portrayed as practised by men. References to bows and arrows first appear in the oracle inscriptions, for example:

Gendered Skill: Chinese and Greek Skill-Knowledge Analogies

333

Ⲩᐣঌᆀདྷ ሴ ੺䶎 㢡 Divined on [the day] Guisi: Our lord dreamed of (shooting with) bow and arrow; reporting it, it will not (spell affliction) Schwartz 2019: 87

By the Zhou dynasty (1046–256 bce ), archery had both military and ritual aspects. It was among the ‘Six Arts’ (Liuyi ‫ޝ‬㰍) of the gentleman or junzi ੋᆀ. The Shijing 䂙㏃ or Book of Odes describes court archery contests at banquets (for example, Mao 220) and ritual seasonal hunts. The Zuozhuan ᐖۣ (Zuo Transmissions) and other historical records attest to the use of bow and arrow in battle. The Xici 㒛䗝 commentary to the Yijing ᱃㏃ (Book of Changes) ascribes the invention of the bow and arrow to the legendary sages Huangdi 哳ᑍ, Yao ๟ and Shun 㡌: ᕖᵘ⛪ᕗˈ࢑ᵘ⛪⸒ˈᕗ⸒ѻ࡙ˈԕေཙл They strung wood to make bows and sharpened wood to make arrows. The benefit of bows and arrows was that by which they imposed their power over the world. Zhouyi yinde, Xici xia, II, 456

Accounts of archery first take a philosophical turn in the Lunyu 䄆䃎 (Analects), where Confucius describes archery as an important form of ritual expertise in which the practice of archery manifests the virtue of the archer. This virtue is expressed not by victory in the contest, but by the manner in which the archer shoots. Several passages illustrate this point: ᆀᴠ˖Njੋᆀ❑ᡰ⡝ˈᗵҏሴѾʽᨆ䇃㘼ॷˈл㘼伢ˈަ⡝ҏੋᆀDŽnj The Master said: ‘Surely archery can serve as an illustration of the fact that the gentleman does not compete! Before mounting the stairs to the archery hall, gentlemen bow and defer to one another, and after descending from the hall they mutually offer up toasts. This is how a gentleman “competes”.’ Lunyu 3.7/5/9. Trans. Slingerland 2003: 19 ᆀᴠ˖NjሴнѫⳞˈ⛪࣋н਼、ˈਔѻ䚃ҏDŽnj The Master said: ‘In archery, one does not emphasize piercing the hide of the target because people’s strengths differ. Such is the way of the ancients.’ Lunyu 3.16/6/4. Trans. Slingerland 2003: 23

A similar view appears in the Liji ⿞䁈 (Book of Rites). The chapter ‘The Meaning of Archery’ (Sheyi ሴ㗙) describes a hierarchy of methods to regulate one’s intentions while practising archery, from the ruler down through a range of officials: 6

Translations are my own unless otherwise indicated.

334

Skill in Ancient Ethics

᭵᰾Ѿަㇰѻᘇˈԕнཡަһˈࡷ࣏ᡀ㘼ᗧ㹼・ˈᗧ㹼・ࡷ❑᳤Ҳѻ⾽ ⸓DŽ࣏ᡀࡷ഻ᆹDŽ᭵ᴠ˖ሴ㘵ˈᡰԕ㿰ⴋᗧҏDŽ In this way, they were clear about the intention behind these regulating measures and thus were not wanting in their service, and so they were successful in their undertakings, and their virtue and conduct was established. When their virtue and conduct was established, there were no evils of oppression and disorder. When their undertakings were successful, their states were at peace. Thus it is said: archery is that by which we see the completeness of [the archer’s] virtue. Liji 47.3.31–32

In skill-knowledge analogies, archery is compared to virtue (Mencius), to the ‘idiocy’ of true Daoist masters (the Daodejing and Zhuangzi), and to the regularity of nature (Xunzi). Chinese archery analogies are ambivalent, especially in stories about the legendary Archer Yi 㗯. He is portrayed both as a saviour of humankind and as a neglectful ruler. He shoots down seven suns that threaten humankind, but he also usurps the throne of Yin, follows bad advice, neglects his kingdom for pleasure and hunting, and teaches his skill to an unworthy man who eventually kills him.

Archery as government The Guanzi compares Yi’s archery to the craftsman’s intuitive grasp of axe and adze to cut along his marking. Archer Yi’s dao did not merely lie in shooting arrows (Yi zhi dao, fei she ye 㗯ѻ䚃ˈ䶎ሴҏ): he hit the mark by careful adjustment of bow and arrows, and accurate assessment of the height of the target. According to the Guanzi, enlightened rulers harmonize laws, judge what to keep or discard, and guard their laws securely. Archer Yi hits the mark and rulers govern well, each by mastering a respective way (Guanzi 20.64 (‘Xingshi’ ᖒऒ): 5b; Rickett 1985: 70). The Guanzi passage focuses on skilled performance and analogizes the skills of the ruler to those of the archer. But each skill consists of full mastery of a way (dao), not merely the successful performance of a limited task. Other Guanzi accounts portray archery as at best a military skill, and at worst a distraction from proper rule, when archery and hunting cause rulers to neglect the empire. For example, in ‘Admonitions’ (Jie ᡂ) Guan Zhong admonishes Duke Huan for shooting arrows, neglecting the empire and oppressing the people (10.26: 3a–b). The Guanzi also compares archery to poor or misdirected government. In a series of analogies demonstrating the importance of ‘Seven Standards’ of government. For example, not being clear about the nature of transformation (bu ming yu hua н᰾ᯬॆ), but wanting to change customs and

Gendered Skill: Chinese and Greek Skill-Knowledge Analogies

335

teachings, is like bending wood for wheels in the morning and hoping to ride in a chariot by night. Not being clear about punishments and rewards (juesai ⊪ຎ), but wanting to exhort and move the people, is like trying to make water flow against the current. The sixth analogy compares understanding of government to competence in archery: н᰾ᯬᗳ㺃ˈ㘼Ⅲ㹼ԔᯬӪˈ⥦‫ؽ‬ᤋ㘼ᗵᤈѻDŽ Not being clear about the arts of the mind, but wanting to have people carry out your orders, is like turning your back on the target and being confident of hitting the mark. Guanzi 2.7 (‘Qifa’ г⌅): 2b, cf. Rickett 1985: 129

Another passage compares skill in archery to military preparation. However, unlike the previous passage, which stressed the importance of the arts of the mind (xinshu ᗳ㺃), this one stresses military training: ሴ㘼н㜭ѝˈ㠷❑⸒㘵਼ሖDŽѝ㘼н㜭‫ˈޕ‬㠷❑䧳㘵਼ሖDŽሷᗂӪˈ㠷‫ؤ‬ 㘵਼ሖDŽ⸝‫ޥ‬ᖵ䚐⸒ˈ㠷඀㘼ᖵ↫㘵਼ሖDŽ Having archers who cannot hit the target is the same thing as having no arrows. Their hitting the target but not piercing it is the same thing as having no arrowheads. Being a general over untrained men is the same thing as having no armour. Using short weapons against long-distance arrows is the same thing as sitting down to wait for death. Guanzi 10.28 (‘Canhuan’ ৳ᛓ): 9a; Rickett 1985: 394

In all these examples, for the ruler to act without essential political knowledge is like performing archery without the requisite skills or materiel, and unrealistically expecting a satisfactory outcome.

Archery as virtue Mencius uses archery in a skill-knowledge analogy to elucidate the component aspects of skill wisdom. He compares wisdom (zhi Ც) to skill (qiao ᐗ) and sagacity to strength. Ცˈ䆜ࡷᐗҏ˗㚆ˈ䆜ࡷ࣋ҏDŽ⭡ሴᯬⲮ↕ѻཆҏˈަ㠣ˈ⡮࣋ҏ˗ ަѝˈ䶎⡮࣋ҏDŽ Wisdom can be likened to skill, and sageness can be likened to strength. As with shooting from beyond a hundred paces, its reaching [the target] is [because of] your strength, but its hitting it is not [because of] your strength. Mencius 5B1, 10.1/51/12–13

336

Skill in Ancient Ethics

He also compares benevolence to archery. The archer should check his stance before shooting. If he misses the mark, he should seek the cause of the error within himself, rather than begrudge the victor (2A7). The purpose of this analogy is to elucidate the components, or correct manner of practice, of virtue (wisdom or benevolence). The quality common to both is the attitude or emotional disposition of the benevolent individual and the skilled archer. Other archery analogies stress the importance of ‘naturalness’ (ziran 㠚❦) – not interfering with the natural tendencies of things or forcing them against their natures – in the teaching of the expert archer and the intuitive learning of the student. Xunzi compares Archer Yi’s expertise at training archers to an enlightened official training men (11/53/11). He also uses archery as a metaphor for the consistency of human behaviour: ⢙਴ᗎަ于ҏDŽᱟ᭵䌚Ⲵᕥˈ㘼ᕃ⸒㠣✹DŽ Things follow their own kind. When targets are measured out, bows and arrows follow. Xunzi 1/2/6–77

Another passage demonstrates the necessity of appropriate means to achieve acceptable ends: 㗯㘵ˈཙлѻழሴ㘵ҏˈ❑ᕃ⸒ࡷ❑ᡰ㾻ަᐗDŽ Yi was the greatest archer in the world, but without bow and arrows, he would have had no way to make his skill visible. Xunzi 8/32/4–5

By contrast, Mencius (6A20) compares the teaching of Archer Yi to that of a master carpenter. Yi’s students imitate his ‘natural’ drawing of his bow; the carpenter’s students imitate his ‘natural’ use of compasses and square. Mencius stresses the central role of naturalness (ziran) in the teaching of the expert archer and the intuitive learning of the student. Similarly, a narrative from the Liezi uses archery as a metaphor for the broadest kind of efficacy: ‘Equalizing the give and the pull is the ultimate principle of dealing with the world.’8 It goes on to describe the skills of the fisherman Zhan He 䂩օ, who explains to the king of Chu that he had heard of the shooting (yi ᔻ) of Puqizi 㫢фᆀѻ: 7

8

Similarly: ‘If the lord is an archer his ministers will wear thumb rings’ ੋሴࡷ㠓⊪ (Xunzi 12/58/11– 12). ൷ཙлѻ㠣⨶ҏ. Zhuangzi 5: 171; Graham 1960: 105.

Gendered Skill: Chinese and Greek Skill-Knowledge Analogies

337

ᕡᕃ㓆㒣ˈ҈付ᥟѻˈ䙓䴉叜ᯬ䶂䴢ѻ䳋DŽ⭘ᗳሸˈअ᡻൷ҏDŽ Using a weak bow and thin line, and shaking the line so that it rode with the winds, he transfixed both of a pair of black cranes on the edge of a dark cloud. Using the concentration of his mind, the movement of his hand equalized the give and pull. Liezi 5: 172–173; cf. Graham 105

He explains that he took this technique as his model for fishing: ⮦㠓ѻ㠘⋣ᤱㄯˈᗳᰐ䴌ឞˈୟ冊ѻᘥ˗ᣅ㏨⊸䢔ˈ᡻ᰐ䕅䟽ˈ⢙㧛㜭 ҲDŽ冊㾻㠓ѻ䠓似ˈ⥦⊸ෳ㚊⋛ˈ੎ѻн⯁DŽᡰԕ㜭ԕᕡࡦᕺˈԕ䕅㠤 䟽ҏDŽབྷ⦻⋫഻䃐㜭㤕↔ˈࡷཙлਟ䙻ᯬаᨑˈሷӖྊһૹ˛ When I overlook the river holding my rod, there are no distracting thoughts in my mind. I contemplate nothing but the fish. When I cast the line and sink the hook, my hand does not pull too hard nor give too easily, so that nothing can disturb it. When the fish see the bait on my hook, it is like sinking dust or gathered foam, and they swallow it without suspecting. This is how I am able to use weak things to control strong ones, light things to bring in heavy ones. If Your Majesty really is able to rule his state in the same way, he can turn the Empire within the span of his hand; what can give you trouble? Liezi 5: 173; Graham 1960: 106

The Zhuangzi also describes the skill of Archer Yi, but in very different terms. It uses his skill as a metaphor for the limitations of trivial skill and small knowledge. Yi could hit the smallest target, but could not avoid praise. He is contrasted to the sage, who is skilful in affairs of Heaven, rather than in human affairs (23/67/17–18). In the Zhuangzi, Yi’s archery also demonstrates the dangers of a purely perspectivist epistemology: if we call an archer skilled who accidentally hits the target, everyone can be an Archer Yi. If everyone is his own moral authority, everyone could be a Yao (24/69/16–17). Here the common quality is accidental success without true understanding. Finally, the Zhuangzi uses the example of Archer Yi to illustrate how to navigate between fate and choice. Only a person of virtue can understand what cannot be avoided and be at peace with it, as with fate. If you wander into the centre of Archer Yi’s target range and don’t get hit, that’s fate.9

Instruction stories about archery Two longer narratives concern the ethics of instruction. The Mengzi and the Liezi present contrasting examples of master archers’ choices of the students to 9

䙺ᯬ㗯ѻᕰѝˈѝཞ㘵ˈѝൠҏˈ❦㘼нѝ㘵ˈભҏ. Zhuangzi 5/14/2.

338

Skill in Ancient Ethics

whom they teach their skills. Mencius (4B24) recounts how Archer Yi’s student Peng Meng 䙒㫉, having learnt everything Yi could teach, killed him. Mencius blames Yi for having chosen his student poorly, and offers the counterexample of the upright student of Zizhuo Ruzi ᆀ☟ᆪᆀ of Zheng 䝝. Pursued by an archer from Wei and too ill to hold a bow, he anticipates death. When he learns that his antagonist is his student’s student, he is confident that his student would only instruct an upright man. He is correct, and his pursuer resolves the conflict between his obligation to a teacher’s teacher and his obligations to the state of Wei by shooting arrows harmlessly in the air (4B24). Another counterexample of an upright student is the Liezi story of the master Gan Ying ⭈㸵, his student Fei Wei 伋㺋, and Fei’s student Ji Chang ㌰᰼. Fei Wei tries to kill his teacher Gan Ying, but he catches the arrow in his teeth. Ji Chang tries to kill his teacher Fei Wei, and their arrows meet mid-air; Fei then blocks Ji’s last arrow with a thorn. Fei Wei and his upright student Ji Chang become like father and son, and both vow to instruct no one else (5: 184; Graham 1960: 113). The Liezi also describes how Ji Chang followed Fei Wei’s instruction to train his eyes. He returned home and lay down under his wife’s loom ‘in order to train his eyes with the raising and lowering’ of the heddle shafts (yi mucheng qianting ԕⴞ᢯⢭ᥪ). After two years, he did not blink even when the sharp point of the mechanism dropped to the corner of his eye (5: 182; Graham 1960: 112). The idea of balancing conflicting elements also occurs in the Lienüzhuan ࡇ ྣۣ (Lives of Exemplary Women), in one of a few stories about women knowledgeable in archery. The wife of the bow-maker of Jin intercedes with Duke Ping of Jin ᱹᒣ‫ ޜ‬when her husband is condemned for making a bow the duke cannot shoot. She argues that, since the bow’s components are all of top quality, the fault must be with the man, not the bow. Yet she can hardly afford to criticize a powerful ruler directly. She resolves the situation by addressing the cause of Duke Ping’s real dissatisfaction: his own inability to hit the target. For this, the remedy is not a better bow, but rather instruction in archery, and she provides it: ‘Make your left hand as if pushing away a stone and your right hand as if leaning on a branch. The right hand lets fly and the left hand does not know [remains firm]. This is the way of archery’.10 This tacit metaphor correlates Duke Ping’s failure in archery to his failure in moral judgment (condemning the innocent bow-maker). The bow-maker’s wife remedies both.

10

Lienüzhuan 6.3/53.7–10. For discussion, see Raphals 2002a.

Gendered Skill: Chinese and Greek Skill-Knowledge Analogies

339

2. Greek archery metaphors Chinese accounts of Archer Yi freely combine analogies and metaphors. Archery does not figure in Greek craft analogies, possibly because it is not materially productive in the sense of Plato’s other chosen analogies. That said, it does figure in Greek texts as a metaphor for accurate foresight and focused goals, for the resourcefulness and implacability of desire and love, and as a powerful metaphor for the ‘mean’.

Archery and foresight Archery metaphors appear in the plays of Aeschylus and Sophocles. In the Agamemnon (1195), Cassandra uses the archer hitting the mark as a metaphor for accurate prophecy. Sophocles uses arrows metaphorically for both the false divination of greedy prophets and the true grief foretold. In Antigone, Creon accuses Teiresias of shooting prophetic ‘arrows’ at him, like archers at their mark (toxotai skopou toxeuet’, 1033). Teiresias responds ‘archer-like in anger’ (hōste toxotēs aphēka thumōi) to Creon’s provocation by launching ‘arrows for [his] heart’ (kardias toxeumata), that fly true, from which he cannot run (1085–1086). Plato’s account of hamartia in the Theaetetus is an archery metaphor; he compares the origin of false opinion to the bad archer who misses the mark for lack of accurate perception (Tht. 193e). Other Platonic metaphors address the goals of archery, rather than its attendant skills. In the Laws, Plato compares archery to a good law that aims solely at its proper target and hits nothing else (Leg. 705e). Elsewhere in the Laws, he compares the accuracy of a good archer to judges and lawgivers who accurately mete out punishment in correct amount (Leg. 934b). Finally, Plato associates archery with love and desire, in both their resourceful and predatory aspects. The Symposium (197a) describes Apollo’s invention of archery, medicine and divination under the guidance of Desire and Love. Plato (Symp. 203cd) describes Eros in turn as the son of Plenty (Poros), the son of Mētis and Poverty (Pēnia), and like his father and grandfather, a famous hunter, wise, always weaving stratagems, and a master of jugglery and witchcraft.

Archery and the mean Plato’s archery metaphors are apt but haphazard, in the sense that archery is not a repeated or especially powerful image in Plato’s hands, nor is archery especially

340

Skill in Ancient Ethics

associated with virtue. Aristotle, by contrast, uses the metaphor of archery for what is arguably one of his most important doctrines: the doctrine of the mean. He also uses it to represent virtue as an ‘archery-like quality’.11 Analogies to archery recur throughout the Nicomachean Ethics. One arises from Aristotle’s claim that every activity has a goal or end, and takes archery as an analogue for knowledge of the good and the best good (t’agathon kai to ariston): Will not the knowledge of it, then, have a great influence on life? Shall we not, like archers who have a mark to aim at [kathaper toxotai skopion ekhontes], be more likely to hit upon what we should? If so, we must try, in outline at least, to determine what it is, and of which of the sciences or capacities it is the object. EN 1.2 1094a22–2612

The point is that both activities, understanding of the good and archery, have a clear goal.13 Archery also appears as an analogue to the mean, specifically the claim that virtue is a mean between extremes. Aristotle argues that, as in running and wrestling, masters of any art avoid excess and deficit, and seek the mean (to meson), and: Virtue [aretē] has the quality of hitting the mean. I refer to moral excellence [ēthikēn], for it is this that is concerned with emotions and actions, and in these there is excess, defect, and the mean [to meson]. EN 2.6, 1106b14–18

For emotions, the mean is to feel the right feelings at the right times, with the right objects, towards the right people, with the right aim, and in the right way (1106b20–22): Virtue is thus a kind of mean [mesotēs] insofar as it is able to hit the mean [stokhastikē ge ousa tou mesou]. 1106b27–2814

The implicit analogy between virtue and ‘hitting the middle point’ (to stokhastikē tou mesou) in feelings and actions moves from the ‘numeric’ mean to the spatial centre of a circle in geometry (1109a22). The analogy between the virtuous agent and the skilled archer continues in an implicit archery metaphor: 11

12 13 14

For a comparison of virtue, the mean and archery in the thought of Aristole and Confucius, see Yu 2010. Translations are based on Barnes 1991, but with slight modifications. For a useful discussion of craft analogy in Stoic ethics, see Inwood 1986. μεσότης τις ἄρα ἐστὶν ἡ ἀρετή στοχαστική γε οὖσα τοῦ μέσου.

Gendered Skill: Chinese and Greek Skill-Knowledge Analogies

341

There is a mark [skopos] to which the man who possesses reason [ton logon ekhōn] looks, and heightens or relaxes his activity accordingly, and there is a standard which determines the mean states [mesotētōn]. EN 6.1, 1138b22–23.

Here the analogy is between the accurate archer and the reasoning mind. These analogies differ in their details and points of emphasis, but they all point to archery as ‘virtue like’ in having a goal (skopos) that is met best by hitting the mean, and linked to the use of clear perception and reason.

3. Chinese analogies from weaving As a craft, weaving differs from archery, not only because women were its primary practitioners, but also in that it was one of the major activities of women that had direct economic benefits to themselves and to their households. And as a primary activity of women, it is to be expected that weaving appears in analogies that compare the roles of men and women.

Weaving as virtue Most famous is the story of Mencius’ mother, who took a knife to her weaving to deter her young son from neglecting his studies.15 She compares women’s weaving to men’s study and self-cultivation: ᆀѻᔒᆨˈ㤕੮ᯧᯟ㒄ҏDŽཛੋᆀᆨԕ・਽ˈ୿ࡷᔓ⸕ . . . Ӻ㘼ᔒѻˈ ᱟн‫ݽ‬ᯬᔍᖩˈ㘼❑ԕ䴒ᯬ⾽ᛓҏDŽօԕ⮠ᯬ㒄㑮㘼伏 . . . ྣࡷᔒަᡰ伏ˈ ⭧ࡷໞᯬ㝙ᗧˈн⛪ヺⴌˈࡷ⛪㲌ᖩ⸓DŽ Your abandoning study is like my cutting this weaving. A gentleman studies to establish his name and asks questions to become broad in knowledge . . . Now if you abandon it, you will not avoid becoming a servant or labourer, and will have no means to avoid calamity and suffering. In what way is this different from weaving and spinning for one’s food? . . . As to a woman who abandons her means of sustenance or a man who slips in cultivating his virtue, if they don’ t become robbers or thieves, they will end up as convicts or labourers. Lienűzhuan 1.9/6/28–1.9/7/3

15

Lienüzhuan 1.9/6.21–7.22, Men. 1A12.

342

Skill in Ancient Ethics

Both are sources of livelihood and recognition and protections against misfortune in an unpredictable world.

Weaving as government Other texts compare weaving to government. The Guanzi lists ornament in weaving (and its male equivalent, carving) as one of five signs of political disorder a prince should heed (Guanzi 1.4 (‘Lizheng’ ・᭯): 14a; Rickett 1985: 103). An extensive analogy in the Lienüzhuan compares knowledge of the component processes of weaving and government. Here, Jing Jiang of Lu 冟ᮜဌ teaches her son, the minister Wenbo ᮷՟, how to govern his state: governing a country is ‘entirely in the warp’.16 She makes a detailed and coherent analogy between eight parts of a loom and eight offices. (1) The selvage (fu ᑵ), the finished edge of woven cloth, corresponds to the General (Jiang ሷ). (2) The pattern (hua ⮛) of the weaving is like the Director (Zheng ↓). (3) The object (wu ⢙) that measures fibre and woven cloth corresponds to the Prefect of the Capital (Dudafu 䜭བྷཛ). (4) The batten (kun ᥶), the movable piece that is used to strike down and tighten the weft of the weaving, is like the Director of Messengers (Daxing བྷ㹼). (5) The heddles (zong ㏌), the apparatus of cords that separate the warp into two sets to open space for the passage of the shuttle, corresponds to the Regional Mentor of Guannei 䰌‫ޗ‬ѻᑛ, a liaison officer. (6) The reed comb (jun ൷), which keeps the threads of the warp at equal distances from each other, is like the Royal Annalist (Neishi ‫ޗ‬ਢ). (7) The axle (zhou 䔨), the cloth beam that maintains the density of the warp threads, corresponds to the Prime Minister (Xiang ⴨). (8) Finally, the warp beam (di ᪈), a revolving beam in a frame that played out the rolled warp threads and maintained an even pressure on them, is like the virtuous and capable Three Dukes (Sangong й‫)ޜ‬.17 This passage describes good government by comparing the activities of men and women. Each activity has specialized component functions, all of which must be performed adequately and correctly for the overall activity to succeed. Each component must fulfil its unique function; for example, the reed comb must be notched finely enough to separate hundreds of threads; the warp beam must be strong enough to bear the tension of all of the threads wound around it. Similarly, if a state is to be governed effectively, the component offices must be 16 17

⋫഻ѻ㾱ⴑ൘㏃⸓. Lienüzhuan 1.10/8/9. Lienüzhuan 1.10/8/9–14. For the logic of this analogy, see Raphals 2002a and Rothschild 2014. For a slightly different account of these components, see Kuhn 1990: 46–52 and Kinney 2014: 203–204, nn. 96–101.

Gendered Skill: Chinese and Greek Skill-Knowledge Analogies

343

staffed by men whose qualities are those required by the specialized tasks of the offices.

Warp and weft Another group of Chinese metaphors focus not on the process of weaving, but on the structure of warp and weft (jingwei ㏃㐟). Han bibliographers used the structure of woven cloth as a metaphor for the distinction between textual canon (warp) and apocrypha (weft). In a very different context, medical writers used warp as a metaphor for the conduits and vessels of the body, for example, in the Six Warps (Liujing ‫ޝ‬㏃) chapter of the Huangdi neijing (Yellow Emperor’s Classic of Internal Medicine), the conduits (jing ㏃) or conduit vessels (jingmo ㏃㜸).18 The joints of the ox carved by the skilful butcher in the Zhuangzi are also described as jing (Zhuangzi 7/3/2, 4, and 7).

4. Greek analogies from weaving Whether practised in the home or at the service of a patron, the skill of weaving produced objects of clear economic value, as did the ‘equivalent’ (in Plato’s sense) received crafts of shoemaking and shipbuilding (Bambrough 1956: 100). Given the specialization of weaving as a female craft and the low status of women in both Greek and Chinese society, the prominence of analogies and metaphors based on weaving is striking. Weaving is compared to government, language and the human body. Most weaving analogies portray the weaver in a positive light, and the same is probably true for the other received crafts of Plato’s time. By contrast, weaving metaphors may be more negative, for example, the metaphors of ‘weaving’ deceits, traps and snares, discussed below.

Weaving as government Plato makes repeated analogies between weaving and government. The Statesman (277d) uses the paradigm of weaving to discover the ‘kingly art’ of government, and it presents (279b–283b) a detailed comparison of the arts of kingship and weaving: ‘What example (paradeigma) could we apply which is very small, but has the same kind of activity (pragmateia) as statesmanship?’ (Pol. 279b). 18

See Huangdi neijing suwen 21 (Jingmo bielun ㏃㜸ࡕ䄆) and Shiji 105. For jing in medicine, see Unschuld 1985: 75 and 81–83.

344

Skill in Ancient Ethics

The analogy begins by classifying weaving according to two kinds of motives for human action: making something and preventing suffering. Preventatives include charms and protections (military and otherwise), screens against wind and heat, housing or shields for the person, blankets spread below or garments wrapped around, one-piece or stitched, and made of vegetable fibre or hair. Weaving (and clothes-making) are preventatives, presumably against the violence of the climate (Pol. 279cd). Weaving, like statecraft, depends on contributive and productive arts (Pol. 281e–282); these include carding and weaving (281a), combining and separating (282b), and twisting and plaiting (282d), with the result that wool is created through the intertwining of warp and weft in a set of processes distinct from other productive arts such as carpentry and magic (283b). The analogy next moves to the web of the state, with the claim that all ‘arts’ (epistēmē) that combine materials try to reject bad materials (Pol. 308c). Just as weavers provide raw materials to the carders they oversee, a true statesman supervises the educators of children and permits only education that leads to the web of state (Pol. 308e). Finally, the courageous are the warp of the state; others are the weft; the statesman is the weaver who weaves both together (Pol. 309b). A similar argument appears at Laws (734e), which argues that different materials should be used for warp and weft. The warp should be tough and of superior quality, while the weft may be more pliant. The point of the analogy is to make a similar distinction between magistrates (the warp) and ordinary citizens (the weft). The Timaeus deploys a different weaving metaphor: of plaiting (plekein) and weaving together (uphainein) the soul and body (Ti. 36e), the veins (77e) and the mortal and immortal parts of human beings: For the rest, do ye weave together the mortal with the immortal, and thereby fashion and generate living creatures, and give them food that they may grow, and when they waste away receive them to yourselves again. 41d

Weaving as naming Weaving is used metaphorically in early Greek literature, on the one hand for counsel, plans and traps, and on the other, for language itself. Metaphors that identify weaving with both language and planned action first appear in the Homeric poems. Helen describes the brevity and clarity of Menelaus in the

Gendered Skill: Chinese and Greek Skill-Knowledge Analogies

345

counsels of the Achaeans ‘weaving the web of speech’ (muthous kai mēdea uphainon, Il. 3.210). In the Cratylus (387d–390b), Socrates extends the weaving metaphor to argue that a name is a kind of instrument: things are woven or named by means of something: a shuttle or a name. Just as we use the shuttle to separate the threads of warp and weft when we weave, we separate things according to their natures when we use names. A name is thus an instrument for separating reality, just as a shuttle is an instrument of separating the threads of the web. Just as weavers use shuttles well, teachers use names well (388c). By contrast, the Laws (806a) mentions weaving as the trivial occupation of girls and women, who would be better educated in gymnastics, music and the military arts, so as to ‘weave themselves instead a life that is not trivial or useless’ (806a).

Weaving as duplicity In the Homeric metaphors of weaving a web of speech and weaving counsel, mentioned above, weaving plots and traps is a pejorative form of weaving counsel (huphainein mētin). Nestor ‘weaves a web of counsel’ (Il. 7.320, 9.90). Laertes might ‘weave some plan in his heart’ (Od. 4.735) to plead with the enemies of Odysseus. Medon overhears the suitors ‘weaving their plot’ (Od. 4.675). Athena and Odysseus discuss ‘weaving a plan’ to hide his Phaeacian treasure (Od. 13.303) and take revenge on the suitors of Penelope (13.385). Similarly, Aeschylus refers to ‘weaving riddles’ (emplekōn ainigmata, PV 609). Euripides’ Andromache describes Spartans as ‘masters of falsehoods and crafty weavers of ill’ (pseudōn anaktes, mēkhanorraphoi kakōn, 447). Here the metaphor is stitching, rather than weaving, since a raptēs is someone who sews or repairs clothing, rather than someone who weaves cloth. In other words, the mēkhanorraphos is someone who cobbles together falsehoods, rather than weaving them from raw materials. Other metaphors are specific to weaving. Aristotle calls Aphrodite a ‘weaver of wiles’ (doloplokos, EN 7.5, 1149b16). Pindar refers to crafty citizens weaving utter destruction (atan pagkhu diaplekei, Pyth. 2.82) and prays to Poseidon for help in weaving future prosperity (huphainein loipon olbon, Pyth. 4.141). He enjoins the lyre to weave song (exuphaine, Nem. 4.44), and prays to Herakles to harness together youth and old age and ‘weave them together in eudaimonia’ (diaplekois eudaimon’ eonta, Nem. 7.96–101).

346

Skill in Ancient Ethics

5. Conclusions Archery and weaving appear in a range of analogies between wisdom, knowledge, virtue and craft(s) in both Chinese and Greek philosophical writing. These skillknowledge analogies compare wisdom or virtue in a variety of ways. Some analogies compare the virtue to the overall craft; others focus on analogous components or common elements. Yet others focus on the skilled performance of virtue or craft. How do weaving and archery analogies compare? Greek archery analogies seem to focus more on accuracy than on skill. Plato compares accurate archery to good lawgivers, who hit their proper target and nothing else. Aristotle uses archery to describe knowledge of the Supreme Good as like the archer’s target, and also to describe archery as a metaphor for the Mean in moral inquiry. Warring States archery metaphors take account of both skill and accuracy; they compare archery to skilled government, wisdom and virtue. Archery analogies compare skill in archery to the virtues of wisdom and benevolence (Mencius), expert government (Guanzi) and the management of subordinates (Xunzi), but also to small knowledge and failure to see a larger picture (Zhuangzi). Yu Jiyuan has argued that the metaphor for archery as a Mean is common to Aristotle and Confucius. Both analogize a virtuous agent to a skilled archer: the virtuous agent forms and exercises virtue just as the skilled archer develops and exercises archery (Yu 2007: 79–80, cf. Camus 2017). This comparison fits both the Mencius and the Xunzi. Other Chinese archery metaphors are very different. The Liezi compares the embodied skill of the archer to efficacy in government through understanding of the ‘give’ and ‘pull’ of the line, as does the instruction story in the Lienüzhuan. The Zhuangzi uses the archer as an example of skill that is entirely misdirected. Chinese and Greek weaving-statecraft analogies differ in focus and context. One Chinese analogy demonstrates the need for well-selected components that act in harmony to form a well-regulated whole by comparing the components of a loom to the offices of the state. Another addresses the regulation of individuals by showing how the analogous activities of men and women lead to livelihood, recognition, and protection against misfortune. The Chinese weavinggovernment analogies appear as maternal instructions to sons, who are their inferiors in generational hierarchy but their superiors in gender hierarchy. Their motivations are immediately practical, rather than rhetorical. Although the Platonic Socrates occasionally claims to have been instructed by a woman, notably Diotima in the Symposium, Plato’s weaving analogies are attributed

Gendered Skill: Chinese and Greek Skill-Knowledge Analogies

347

purely to Socrates. Greek weaving analogies vary in their details, but tend to focus on the hierarchical discrimination of different ‘natures’. The Statesman compares kingship and weaving overall, as ‘the same kind of activity’ with analogous motivations and processes, but the contributive arts common to both create a hierarchy of supervision. The Laws defends social hierarchy by asserting different qualities to warp and weft, analogous to the different natures of rulers and citizens. The Cratylus analogy between weaving and teaching the correct use of language (names) focuses on the ‘discriminating’ role of the shuttle, and compares the skilled performance of the expert weaver and the expert teacher. Weaving metaphors are strikingly different from weaving analogies. Chinese warp and weft metaphors classify complements; Greek metaphors weave strategy (counsel, plans, traps), language and the body (the veins, the body and soul, and the mortal and immortal components of human beings). Recent research on Chinese root metaphors and comparative metaphorology have shown the possibility of and need for comparative study of the use of metaphor, and particularly its use in philosophy. To that end, the present study is a partial prolegomenon to comparative discussion of Chinese and Greek craft metaphors.

References Allan, S. (1997), The Way of Water and Sprouts of Virtue, Albany : State University of New York Press. Barnes, J. (1991), Nicomachean Ethics, in Complete Works (Aristotle). Princeton: Princeton University Press. Bambrough, R. (1956), ‘Plato’s Political Analogies’ in Peter Laslett (ed), Philosophy, Politics and Society, 98–115, Oxford: Basil Blackwell. Black, M. (1954–1955), ‘Metaphor’, Proceedings of the Aristotelian Society n.s. 55: 273–294. Black, M. (1977), ‘More About Metaphor’, Dialectica 31 (3–4): 431–457. Brumbaugh, R.S. (1989), ‘Plato’s Relation to the Arts and Crafts’, in W. H. Werkmeister (ed.), Platonic Studies in Greek Philosophy: Form, Arts, Gadgets and Hemlock, 195–204. Albany : State University of New York Press. Burrell, D. (1973), Analogy and Philosophical Language, New Haven: Yale University Press. Camus, R.M. (2017), ‘Comparison by Metaphor: Archery in Confucius and Aristotle’, Dao 16: 165–85.

348

Skill in Ancient Ethics

Chunqiu Zuozhuan zhu ᱕⿻ᐖՐ⌘, ed Yang Bojun ὺ՟ጫ, Gaoxiong: Fuwen tushu chubanshe, 1991. Cikoski, J.S. (1975), ‘On Standards of Analogical Reasoning in the Late Zhou’, Journal of Chinese Philosophy 2 (3): 325–357. Fauconnier, G., and M. Turner (2002), The Way We Think: Conceptual Blending and the Mind’s Hidden Complexities, New York: Basic Books. Graham, A.C. (1960), The Book of Lieh-tzu, New York: Columbia University Press. Guanzi ㇑ᆀ, Sibu Beiyao edn. Inwood, B. (1986), ‘Goal and Target in Stoicism’, The Journal of Philosophy 83 (10): 547–556. Johnson, M., ed. (1981), Philosophical Perspectives on Metaphor, Minneapolis: University of Minnesota Press. Kinney, A.B. (2014), Exemplary Women of Early China: the Lienü Zhuan of Liu Xiang, New York: Columbia University Press. Kuhn, D. (1990), Zur Entwicklung der Webstuhltechnologie im alten China: kommentierte Übersetzung der entsprechenden Kapitel der Geschichte der chinesischen Textilwissenschaft und Textiltechnik. Das Altertum, Zhongguo fangzhi kexue jishu shi (gudai bufen). Heidelberg: Edition Forum. Lai, K., and W.W. Chiu, eds (2019) Skill and Mastery: Philosophical Stories from the Zhuangzi, London: Rowman & Littlefield. Lakoff, G., and M. Johnson (1980) Metaphors We Live By, Chicago: University of Chicago Press. Lakoff, G. and M. Johnson (1999) Philosophy in the Flesh: The Embodied Mind and its Challenge to Western Thought. New York: Basic Books. Lienüzhuan ࡇྣۣ, in A Concordance to Gulienüzhuan (ਔࡇྣۣ䙀ᆇ㍒ᕅ), eds D.C. Lau et. al., ICS series, Hong Kong: Commercial Press, 1993. Liezi jishi ࡇᆀ䳶䟺, ed Yang Bojun ᶘ՟ጫ, Beijing: Zhonghua, 1979. Liji ⿞䁈, in A Concordance to the Liji (⿞䁈䙀ᆇ㍒ᕅ), eds D.C. Lau et. al., ICS series, Hong Kong: Commercial Press, 1992. Link, Perry (2013) An Anatomy of Chinese: Rhythm, Metaphor, Politics. Cambridge MA : Harvard University Press. Lunyu 䄆䃎, in A Concordance to the Lunyu ( 䄆䃎䙀ᆇ㍒ᕅ), eds D.C. Lau et. al., ICS series. Hong Kong: Commercial Press, 1995. Lloyd, G.E R. (1966), Polarity and Analogy: Two Types of Argumentation in Early Greek Thought, Cambridge: Cambridge University Press. Maspéro, H. (1928), ‘Notes sur la logique de Mo-Tseu et de son école’, T’ oung-Pao 25: 1–64. Maspéro, H. (1933), ‘Le Mot Ming’ Journal Asiatique 223: 249–296. Mengzi ᆏᆀ, in A Concordance to the Mengzi (ᆏᆀ䙀ᆇ㍒ᕅ), eds D.C. Lau et. al., ICS series, Hong Kong: Commercial Press, 1995. Raphals, L. (1992), Knowing Words: Wisdom and Cunning in the Classical Traditions of China and Greece, Ithaca, NY: Cornell University Press.

Gendered Skill: Chinese and Greek Skill-Knowledge Analogies

349

Raphals, L. (2002a), ‘Arguments by Women in Classical Chinese Texts’, Nan Nü 3 (2): 157–195. Raphals, L. (2002b), ‘A Woman Who Understood the Rites’, in B.W. Van Norden (ed.), Essays on the Analects of Confucius, 275–302, Oxford: Oxford University Press. Raphals, L. (2005), ‘Craft Analogies in Chinese and Greek Argumentation’, in E. Ziolkowski (ed), Literature, Religion, and East/West Comparison: Essays in Honor of Anthony C. Yu, 181–201, Newark: University of Delaware Press. Raphals, L. (2015), ‘Body and Mind in Early China and Greece’, Journal of Cognitive Historiography 2: 132–182. Reding, J.-P. (1986), ‘Analogical Reasoning in Early Chinese Philosophy’, Asiatische Studien 40.1: 40–56. Reding, J.-P. (1997), ‘L’ Utilisation Philosophique de la Métaphore en Grèce et en Chine’, Revue de Théologie et de Philosophie 129: 1–30. Reding, J.-P. (2004) ‘Light and the Mirror in Greece and China: Elements of Comparative Metaphorology’ in Comparative Essays in Early Greek and Chinese Rational Thinking, 127–166. Aldershot and Bloomington, IN : Ashgate. Rickett, W.A. (1985), Guanzi: Political, Economic and Philosophical Essays from Early China, Vol. 1, Princeton: Princeton University Press. Rothschild, N. H. (2014), ‘Rhetoric of the Loom: Discursive Weaving Women in Chinese and Greek Traditions’, Sino-Platonic Papers 24: 1–22. Schwartz, A.C. (2019), The Huayuanzhuang East Oracle Bone Inscriptions: A Study and Completely Annotated English Translation, Berlin and Boston: Walter de Gruyter. Searle, J.R. (1979), ‘Metaphor’ in A.P. Martinich (ed), The Philosophy of Language, 76–116, Oxford: Oxford University Press. Slingerland, E.T. (2003), Confucius Analects: with Selections from Traditional Commentaries, Indianapolis: Hackett Publishing. Slingerland, E. T. (2019) Body and Mind in Early China: Beyond Orientalism and the Myth of Holism. Oxford Turner, M. (1988), ‘Categories and Analogies’ in D.H. Helman (ed.), Analogical Reasoning: Perspectives of Artificial Intelligence, Cognitive Science, and Philosophy, 3–24, Dordrecht: Kluwer. Unschuld, P.U. (1985), Medicine in China: A History of Ideas, Berkeley : University of California Press. Volkov, A. (1992), ‘Analogical Reasoning in Ancient China: Some Examples’, ExtrêmeOrient, Extrême-Occident 14: 16–45. Xunzi 㥰ᆀ, in A Concordance to the Xunzi ( 㥰ᆀ䙀ᆇ㍒ᕅ), eds D.C. Lau et. al., ICS Series, Hong Kong: Commercial Press, 1996. Yili ܰ⿞, in A Concordance to the Yili (ܰ⿞䙀ᆇ㍒ᕅ), eds D.C. Lau et. al., Taiwan: Commercial Press, 1996. Yu, J.Y. (2007), The Ethics of Confucius and Aristotle: Mirrors of Virtue, New York: Routledge.

350

Skill in Ancient Ethics

Yu, J.Y. (2010), ‘The Mean, the Right and Archery’, Selected Papers of the Beiijing Forum 2007, Procedia Social and Behavioural Sciences 2: 6798–6804. Zhouyi yinde ઘ᱃ᕅᗇ, in A Concordance to Yi Ching, ed. William Hung, HarvardYenching Institute sinological index series Suppl. 10, Cambridge MA : Harvard Yenching Institute. rpt. Taipei, 1965. Zhuangzi 㦺ᆀ, in A Concordance to the Zhuangzi (㦺ᆀ䙀ᆇ㍒ᕅ), ed. D.C. Lau et. al., ICS Series, Hong Kong: Commercial Press, 2000.

17

The Skilful Wanderer: On the Risks and Rewards of Travel in Plato and Zhuangzi Rohan Sikri

Introduction The problem of travel marks a fertile ground of inquiry in the canons of both Ancient Greek and classical Chinese philosophy. It relates to questions of desire, subject-formation, the construction of place, and, in its most rarefied formulation, to thinking as such. It is only to be expected, then, that we encounter a diversity of positions on the merits and the perils that accompany an act that even today, in a world more or less secure, often emerges as a fraught experience. Why does the thought of travel so vex the minds of some of these ancient philosophers and delight the hearts of others? What is the nature of the risk that lurks in every decision to travel, and how is it to be managed? What are the skills that define the traveller in these ancient worlds, and what dictates the preferential treatment of a certain kind of traveller over another? Implicit in each of these questions are both ancient prejudices and ancient anxieties related to the manner in which we constitute ourselves as coherent, clear-thinking, ethical subjects. Travel can be a bane or a boon – or a more nebulous mixture of both – in the context of this endeavour. I want to entertain here the general presumption that to travel, far from being an uncomplicated privilege to feed one’s curious mind and one’s desires, is to invite risk that is inexorably given in the very act. Such a framework, in turn, dictates a spectrum of possibilities, bookended by a conservative management of this risk that forsakes travel in favour of rest, at one end, and by a form of ‘reckless’ pursuit at the other. Two philosophers, in particular, aid us in conceptualizing these bookends. One is Plato; the other, Zhuangzi. Where the former casts aspersions on the sophist, his bête noire, by identifying him with ‘a

351

352

Skill in Ancient Ethics

class that wanders (to genos . . . planēton)’ (Timaeus 19e),1 the latter delights at every stage of his argument in the art of wandering (you 䙺), the ostensible goal of most of his philosophical exercises. To both, wandering is a mode that distils the act of travel down to its elemental quality of risk. The risk, for Plato, is ultimately that of self-annihilation, that is, of un-fixing stable constituents of selfhood, of exposure to only endless varieties of life without recourse to the underlying form that gathers and binds these phenomena. To wander, in Zhuangzi’s frame of reference, also involves an act of dispersion, of loss of self, of ambiguation, confusion, and the need for constant revision of one’s horizons of intelligibility. But these are the very nutrients that make for a life lived well in the therapeutic vision the authors of the text delineate, and an emphatic mode of itinerancy thus emerges as an integral panacea. To be sure, Plato builds a central pillar of his philosophical lexicon upon an extant culture of travel in classical Greece, modelling his use of the word theōria, the clear sighting of the truth that is the philosopher’s art, upon the civic institution of the theōros, the official traveller, who, fulfilling the demands of his public office, journeys to witness religious festivals, oracular events or spectacles of colonial settlement, and returns to the home-city with a report of ‘ritualized visuality’ (Nightingale 2009: 4).2 This is to tolerate a travel under constraints, where there is always a return to a source in a further act of self-constitution. The art of travel celebrated in Zhuangzi appears diametrically opposed to the Platonic conception as it calls for an exercise that continually works against constraints. As an unfettered, ‘free and easy wandering’ (xiao yao you 䘽䚉䙺), this modality of travel translates into an act of ‘self-forgetting’ (wang qi shen ᘈަ䓛) (ZY 10/4/43–44).3 My aim in this chapter is to articulate these competing models of the wanderer and make explicit the skills that each of the philosophers attribute to the act of wandering. I begin by analysing a selection of wanderers in Zhuangzi and construct a set of typologies amongst the various cases of itinerant humans that

1

2

3

All citations to the original text in the Platonic corpus are from the Loeb Classical Library editions and are marked in the body of the essay. All translations are mine unless noted otherwise. Plato’s explicit recognition of the civic institution of the theōros occurs in Laws 950d–951a, where he adapts the model to suit the strictures of the hypothesized Magnesian constitution. His prohibition on travel as such by any citizen under the age of forty, as well as the strict constraints under which theōroi undertake their travels, emphasizes both Plato’s reluctance to the idea of traveling as well as the philosophical ends to which he ultimately subjects an extant civic and cultural institution. See also Montiglio 2005: 160. All citations are to the Zhuangzi Yinde Harvard-Yenching Concordance in Zhuangzi 1956 and are included in the body of the essay as ZY . All translations are mine unless noted otherwise.

The Skilful Wanderer on the Risks and Rewards of Travel

353

populate the text. First, I conceptualize an explicitly ateleological mode of wandering, which takes shape in the context of exemplars involved in an apparently unfocused movement in spaces of wilderness ‘beyond the four seas’ (you hu si hai 䙺Ѿഋ⎧) (the trope that establishes the boundaries of civilization in the classical Chinese world). I then describe a pragmatic mode of wandering, in which adepts display a skill in negotiating normative spaces of political action. Finally, I draw the reader’s attention to a third, and arguably most prized, version of the wanderer who is able to transform the normative bases for epistemic criteria and rethink their application. In the second part of the analysis, I turn to Plato’s treatment of the wanderer and present his strong aversion in the context of a broader cultural estimation of such figures and the connotations they carry. With a focus on Phaedrus, I distill Plato’s criticisms of wandering down to the sophistic method of to eikos, or ‘plausibility,’ and demonstrate the distinctly epistemological dimension to the problem of wandering. Where Plato identifies the act of wandering between cities exclusively with the class of sophists, it is seen to involve a merely plausible, and ultimately dissimulating, dialectic that is mired in false beliefs and opinions. This is diametrically opposed to the Socratic philosopher, who remains within the city’s walls and eschews travel so he may hone his dialectical skill on a solid foundation of hē anankē, or ‘necessity’. Between Plato’s aversion to the wanderer’s skills in dissimulation and Zhuangzi’s unabashed celebration of wandering as a dialectical skill par excellence, there are both similarities and differences in the treatment of a common philosophical problem. In the final part of the essay, I explore the intersections of a Platonic lexicon, organized around the criticism of the act and the agent of planē, and Zhuangzi’s model of you, with the intention of building a more nuanced understanding of the problem in each of these figures.

1. The a(nti)teliology of wandering Variously rendered in English as ‘to travel,’ ‘to roam,’ ‘to wander’ or ‘to ramble’, and equally inclusive of the sense of enjoyment that accompanies or emerges from such movement, the semantic range of the character you is harnessed in Zhuangzi to a degree unmatched by any other philosophical text from the Warring States period (5th–3rd century bce ). In contrast to Confucian classics like Lunyu or Mengzi, which use the character only four and eight times respectively, Zhuangzi’s multiple, and multivalent, uses of you immediately alert the reader to a conceptual

354

Skill in Ancient Ethics

depth in the text that distinguishes it from most other Chinese philosophical classics.4 One point of purchase in the text’s complex world of wanderers is offered up by sage-like figures who do not sit still. Be it the case of Nan Bo who ‘wanders about the heights of Shang’ (you hu shang zhi qiu 䙺Ѿ୶ѻш), or Tian Gen ‘wandering on the sunny side of Yin’ (you yu yin yang 䙺ᯬ⇧䲭), the ubiquitous scene in the text is that of an individual or a group exploring spaces far removed from the official quarters of the world and doing so with no apparent purpose. These rambles, to be sure, are productive of insight, but the wisdom thereby attained is assuredly not a consequence of a premeditated course of action. There is a particular figure in the text who is, arguably above all others, a skilful practitioner of the aimless jaunt. Hong Meng is a wanderer par excellence in that his ramblings have an additional quality of the absurd that accentuates the semantic range covered by the character you. He does not merely roam about idly, but indulges in a kind of slapstick rambling: Yun Jiang was traveling east and, when he crossed the branches of the Fuyao, he encountered Hong Meng. Hong Meng at the time was wandering about [you], slapping his thighs and hopping around like a sparrow. Yun Jiang, on seeing this, stopped in shock, stood motionless, and said, ‘Old Sir, who are you? What is this you’re doing?’ Hong Meng, without ceasing his thigh slapping and sparrow hopping, replied to Yun Jiang, ‘I am wandering about, enjoying myself [you].’ ZY 27/11/44–45

We might speak of a certain choreography of wandering in the strange case of Hong Meng. It is, evidently, not enough to ramble in the merely ordinary fashion of placing one foot in front of the other. Here, the act of wandering involves a bewildering thigh-slapping, sparrow-hopping dance, which lends an added sense of enjoyment to the experience. Equally important is Hong Meng’s resistance to all attempts on the part of his interlocutor to distill this art of choreographed movement down to a set of principles or directives that might benefit one’s position. Asked for instruction, he once again slaps his 4

As I proceed to show above, the quality of the text’s inordinate reliance on the term further establishes the central role it plays in the overall conceptual scheme. Scholars are in general agreement on the pivotal function of the concept of ‘wandering’ in the text and the formative role Zhuangzi plays in the subsequent development of literary genres such as youxian 䙺ԉ (‘wandering into immortality’) and youlan 䙺㿭 (‘sightseeing’) verse. Diverse positions have been taken in the debate over the meaning of you, such as the following: Mair (1983) views the character through the playful prism of the homo ludens; Fraser (2014) translates ‘wandering’ into a eudaimonistic model; Levinovitz (2012) presents the heuristic of the ‘unreliable narrator’ that enables one to describe the model of you without contradiction. On the historical development of you as a literary tradition and in neo-Daoist xuanxue ⦴ᆨ thought, see Hawkes 1967 and Kroll 1996.

The Skilful Wanderer on the Risks and Rewards of Travel

355

thighs, hops about, and shakes his head, saying, ‘I do not know! I do not know!’ (wu fu zhi ੮ᕇ⸕ˈ ੮ᕇ⸕). Refusing to offer a judgment on the nature of his wandering, Hong Meng replicates the text’s overarching philosophical scepticism that continually undermines criteria for zhi ⸕ or knowledge. I will return shortly to this epistemological dimension to the art of wandering. Here, I simply wish to emphasize the corresponding quality of aimlessness that the text attributes to the un-knowing wanderer. Upon encountering his interlocutor again after three years, and in a slightly more voluble mood, Hong Meng remarks, Drifting (fu ⎞), wandering (you), I do not know (bu zhi н⸕) what I seek; utterly wild, I do not know where I go[.] ZY 27/11/49

Yet, is there something disingenuous about Hong Meng’s response? He might be aimlessly wandering without knowledge of any objective he seeks, but he certainly has a clear idea of what he does not wish to be involved with. In a word, this is the business of zhi ⋫, or governing, the singular source to which he attributes all incidence of misfortune or disaster, even that which ‘reaches down to worms’ (huo ji zhi chong ⾽৺→㸢). Such a sensibility pervades the text, its various archetypes often portrayed in their explicit rejection of the world of official business, involved political action, and concerted effort.5 As a general statement of the ethic of wuwei or ‘non-action’, Hong Meng’s privative claims are hardly surprising. The passage unmistakably restates the positions of Laozi 1 (on the inconstancy of a way that can be walked), of Laozi 2 (where sages ‘practice the teaching that is without words’), and of Laozi 48 (which affirms the method of ‘gaining the world by having no affairs’), amongst numerous other chapters.6 The general thrust of these positions in Laozi amounts to the text’s rejection of the episteme of ‘worldly affairs’ (shi һ), constructed upon specifically Confucian criteria like benevolence (ren ӱ), righteousness (yi 㗙) and ritual propriety (li ⿞) that dictate how the world can be known authentically. In being without words (bu yan н䀰), however, the sages of both Laozi and Zhuangzi do not transcend the strictures of discursive thought (and linguistic expression), understood in terms of the dyad of zhi ⸕ (knowledge) and yan 䀰 (speech), per se. The challenge that the authors present us with is one of re-imagining the discursive operation so that language might be freed from

5

6

See, for instance, ZY 6/2/73–74, where we are told that ‘sages do not involve themselves with worldly affairs . . . and [instead] wander beyond the dust and dirt’ (er you hu chen gou zhi wai 㘼䙺Ѿລිѻ ཆ). Cf. LZ 5, LZ 14, LZ 25 and LZ 32.

356

Skill in Ancient Ethics

the totalizing, and erroneous, categories of a Confucian doxa. Such a challenge appears to be front and centre in the case of Hong Meng, whose wandering comes to be defined primarily in terms of what it is not. If scenes of wilderness, and all that they convey (aimlessness, absurdity, the blurring of species difference, etc.), are predicated upon the negation of a normative episteme (located in the space of culture or wen ᮷), then the wanderer’s skill involves mastering an apophatic mode of expression. Hong Meng exemplifies this ability in his emphatic claim, ‘I do not know what I seek . . . I do not know where I go’.

2. Wandering inside the world If wanderers in wilderness, outside the ‘four seas’, are engaged in an epistemic subversion that dismantles the categories of Confucian zhi, then what are we to make of cases in Zhuangzi in which the genre of court-travel, inside the ‘four seas’, is not wholly abandoned? Any examination of models of wandering in this text bears the burden of an unavoidable tension between the likes of Hong Meng and figures like Zigao and Yanhui, who introduce the well-established trope of ‘travel’ (you) as official business – that is, of itineraries to courts that are animated by the anxiety of a potentially fatal outcome in exercising the demands of one’s position. The text appears to grant a certain inevitability to the circumstances of these obligations, and instead of recommending an abdication of duty, it offers a revised model of action. In the two paradigmatic cases, Confucius is the ironic sage who relieves his two traveling students from their worries by suggesting that they ‘wander’ (you) in the courts of fickle rulers. Zigao is made aware of the unavoidability of his obligations and is told to let his ‘heart wander in the circumstances of [his] position’ (фཛ҈⢙ ԕ䙺ᗳ) (ZY 10/4/52–53). In a similar fashion, Yanhui is counselled ‘to enter and wander in the cage [of the ruler’s court]’ (‫ޕ‬䙺ަ›) (ZY 9/4/29) and thereby avoid any potentially calamitous outcome to his person. If one reduces Confucius’ remarks to a mere metaphor, one risks missing the full force of the re-conceptualization of a more literal order of travel that is being undertaken here. In transposing the referent of you to the cognitive and behavioural register (i.e. in relation to a xin ᗳ that wanders), the text remains wedded to a ubiquitous empirical frame of travel. The objective, in fact, is to investigate the manner in which travel to a court, with its accompanying political praxis tainted with risk, might still remain a permissible commitment within the horizons of the text. Li (2018) has recently derived an order of landscapes in Zhuangzi, which aids us in negotiating this apparent impasse between the wanderer outside and the

The Skilful Wanderer on the Risks and Rewards of Travel

357

wanderer inside. She points to a clear distinction in the sceneries described in the text that are either fangwai ᯩཆ, outside ‘bounded space’ (fang), or fangnei ᯩ‫ޗ‬, inside such space. Boundaries in either context are both physical as well as normative. Thus, around one pole of this dichotomy (fangwai) we can locate wildernesses and cosmic expanses (mountains, rivers, oceans, clouds, ether, etc.), which are interpreted to be ‘beyond conceptualization’ (Li 2018: 11). Around the other pole (fangnei) are classed the courts, the royal parks and various other built-up spaces that are institutionalized through the apparatus of culture and that constitute the world that can be known. Using this heuristic, Li identifies the cases of Zigao and Yanhui as expressly fitting the ‘game space’ (ibid., 5) of the court, wherein the actors are confined by ‘strict rules’ that conceptualize outcomes as involving the very likely risk of grievous loss. To wander within this ‘game space’, then, amounts to a use of skill and stratagems to secure a steely exit. This does appear to be the import of Confucius’ message, considering the additional advice he offers Yanhui: [Wander in the cage] without impinging on his concern for a good name. When he’s receptive, do your crowing, but when he’s not, let it rest. Do not let him get to you, but do not harm him either. Seeing all possible dwelling places as one, let yourself be temporarily lodged in whichever cannot be avoided. ZY 9/4/29–30

The risk inherent to spaces that are fangnei, to be sure, is bound up with the very credible threat that the authors of Zhuangzi identify with bian 䗟, ‘dialectical argument’ or ‘disputation’. The text associates this genre of eristic debate almost exclusively with the philosophical exchanges of the Confucians and the Mohists, who remain locked in the mire of ‘affirming (shi ᱟ) what the other denies (suo fei ᡰ䶎), and denying what the other affirms’ (ZY 4/2/26). Shi and fei are the dialectical poles that order the discourse of bian, where each party to a debate carves up the matter at hand in terms of fixed categories of ‘what is the case’ (shi) and ‘what is not’ (fei). This elemental act of categorization is the basis for the evaluative judgment of approval and disapproval, the additional senses conveyed by the shi-fei binomial. Zhuangzi is insistent on the epistemologically deficient quality of such divisions, a deficiency that ultimately mars the body of knowledge that each disputant lays claim to.7 7

See, for instance, ZY 4/2/24–25 and 5/2/57–58. The motivations underlying the text’s dismissal of the dialectic of bian, however, are not adequately captured by reducing them to the epistemological scepticism usually attributed to its authors. I am, in this respect, in agreement with the reading offered in Ivanhoe 1993, which identifies categories of scepticism in Zhuangzi beyond the epistemological kind and attributes, above all, a ‘therapeutic scepticism’ to its authors.

358

Skill in Ancient Ethics

Confucius’ sage words on wandering within the court, in a space that is fangnei, are predicated on this critique of bian and the understanding that spaces inside the boundaries of the socio-political world are often fabricated in and through a dangerous game of dialectical showmanship. To wander in this context amounts to a skilful activity that avoids the danger that is always given in the confined ‘game space’. Adopting a passive role inhibits the wanderer’s participation in the eristic encounter of debate. Moreover, the two orders of fangwai and fangnei, as we learn elsewhere in the Neipian, are disjunctive in their relation. Confucius, on being given a report of certain persons who are tout autre, proclaims, ‘these are men who wander outside the lines (fang zhi wai ᯩѻཆ) while I wander inside the lines (fang zhi nei ᯩѻ‫)ޗ‬. Outside and inside can never meet’. (ZY 18/6/66–67)

3. Wandering as transformative: Between ‘inside’ and ‘outside’ Yet, Confucius is, after all, Confucius. For all the appropriations of his voice and the sacrilege of the Confucian faith carried out at his behest, his adaptation into the reckless world of Zhuangzi remains, ultimately, a maladaptation. While his insight paves the way to grasping two orders of ‘wandering’ in the text, it does so only in a disjunctive configuration. Such a view, to be sure, does not exhaust the semantic range of you in Zhuangzi. We can indeed identify an alternate class of wanderers in the text that undermines the disjunction between the two foregoing modalities and instead positions itself in an ambivalent space between the inside and the outside. To wander, in these cases, implies an inherently transformational skill by means of which normative spaces are continually disaggregated and the criteria by which we can know the world continually re-thought. Representative cases of this third type of wandering, one that occurs neither entirely outside in the wilderness nor entirely inside the spaces of culture like the court, are to be found among ‘perfected persons’ or zhiren 㠣Ӫ. While the text often describes these models as wanderers who roam ‘beyond the four seas’,8 there are instances in which they remain thorougly bound to a ‘world’ (shi ц) in which one’s conduct (xing 㹼) is scrutinized, evaluated and ultimately contested on the basis of competing normative standards for ethical action. In other words, to wander in such cases entails a critical engagement in the space of debate, and, accordingly, we must set this skill apart from the kind of apophatic or quietistic 8

See, for example, ZY 2/1/21 and 6/2/72

The Skilful Wanderer on the Risks and Rewards of Travel

359

version of the wanderer identified above. Consider, for instance, the following description from the text’s twenty-sixth chapter, ‘External Things’ (Waiwu ཆ⢙): Zhuangzi said: ‘If persons are able to wander (neng you 㜭䙺), can anything prevent them from wandering? And if they are unable to wander, can anything lead to them wandering? The will to hide oneself away and actions meant to cut oneself off – alas, these are not the duties of perfected knowledge and deep virtue! . . . Hence it is said: Perfected persons do not fixate on conduct. Venerating the past and despising the present is the drift of scholars (xue zhe ᆨ㘵) . . . Only perfected persons are able to wander in the world (you yu shi 䙺ᯬц) and not be one-sided (pi ‫)ܫ‬, go along with others and not lose themselves, [hear] their teaching and not become doctrinaire, receive their ideas and not convert to their disposition.’ ZY 74/26/33–37

What is striking about the description of you here is the conspicuous manner in with it is decoupled from the option of retreat or withdrawal. ‘Perfected persons’, when they ‘wander in the world’ (you yu shi 䙺ᯬц), are thus unlike Hong Meng. They demonstrate no desire for the wilderness and thus no desire for physical withdrawal from the world of practical affairs. Crucially, neither do they demonstrate the kind of bizarre skills that the ‘wild’ wanderer exemplifies – skills, as we have seen, that involve a spectacularly unconventional comportment as well as a concerted apophatic articulation that denies any normative standard by which the former can be known. But to ‘wander in the world’ also differs in a fundamental sense from the risky ‘travels’ undertaken by courtiers like Yanhui and Zigao. Whereas in the latter cases we have equated you with the quietism that animates the disposition of the ‘court-traveller’ (a quietism that sometimes suggests a method of appeasement in order to avoid harm), the wandering of zhiren is tantamount to an open and critical engagement with contesting knowledge-schemes. Skills enumerated as part of this mode of ‘worldly wandering’ thus underscore, primarily, the actor’s dialectical sophistication. The ‘perfected person,’ as wanderer, does not remain ‘one-sided’ or biased, engages with various points of view without collapsing into dogma, and maintains a dynamic awareness to the changing circumstances of the ‘present’ (jin Ӻ) rather than ‘venerating the past’ (zun gu ሺਔ). Ultimately, these skills identify the wanderer, in this avatar, with the possession of ‘perfected knowledge’ (zhi zhi 㠣⸕). Such descriptions of wandering expressly recall the qualities of a sage’s dialectic articulated in the text’s second chapter, ‘Debate that Equalizes Things’ (Qiwulun 啺⢙䄆). There, the authors train their sights on critically appraising,

360

Skill in Ancient Ethics

and eventually dismantling, the apparatus of bian as a means to securing knowledge. Sages identified in the context of this critical project remain bound by the limits of linguistic expression (yan 䀰) and, in this respect, their skill involves a certain facility with the very dialectical structures that constitute the hollow epistemic judgments characteristic of bian. In contrast to the unyielding, fixed dichotomies of eristic debate, the sage is able to categorize and assert a knowledge claim with due recognition of its context dependency. This perspectivism is a defining feature of a dialectics that Zhuangzi is still able to tolerate within the wider parameters of its critique. Sages continue to carve up their world through ‘divisions’ (fen ࠶) and they still make knowledge claims – but these claims, as we read, are adaptive in nature and constantly open to revision.9 Several figures in the text appear to confirm this transformational model of the wanderer, wherein you is exemplified through an art of dialectics.10 In all cases, we are offered a critical skill-set that disaggretates fixed criteria for knowledge in order to develop a more adaptive capacity for context recognition and response. The wanderer achieves an effect here that is not dissimilar to the Socratic gadlfy, constantly sending interlocutors into aporetic throes and logical dead-ends, and forever antagonizing the normative grounds upon which truth, or ‘genuineness’ (zhen ⵏ), comes to be distinguished from falsity, or ‘artifice’ (wei ‫)ڭ‬.11 This does not prevent the wanderer from establishing positive criteria for knowledge—indeed, as we have just read, the zhiren are guided by the ostensible goal of attaining zhi zhi, ‘perfected knowledge’! Yet, to know the world as such a wanderer does is to master the two-foldness of a positive articulation as well as the refusal to grant it normative validity. The wanderer certainly establishes criteria for truth, but these criteria are constantly revisable and, thus, adaptable.

9

10

11

As we read in ZY 4/2/27–29, the sage’s adaptive articulation—a claim of ‘what is so on the basis of [circumstance]’ (yin shi ഐᱟ)—is defined by its context-dependency and the corresponding insight that allows for things to be both ‘this’ (shi ᱟ) and ‘that’ (bi ᖬ). See also ZY 5/2/37. See, for instance, Jie Yu, madman of Chu, whose wandering is associated with critical arguments with Confucius over the nature of de ᗧ (ZY 12/4/86–89); Yierzi, who, wandering at the edge of the world, is instructed in a revision of the conceptual scheme of de that disaggregates ideas of virtue from a calcified Confucian episteme (ZY 17/6/85–89); and the concluding descriptions of Zhuangzhou in the final chapter, ‘The World’ (Tianxia ཙл), which associate his ‘wandering’ with a capacity to inhabit both the spaces outside and inside the world – he “wanders with the Creator of Things” and yet always returns to the confines of a world to which access is necessarily granted through the “rights and wrongs” of bian (ZY 93/33/63–69). ZY 4/2/25

The Skilful Wanderer on the Risks and Rewards of Travel

361

4. The privations of the wanderer in Ancient Greece Wandering in archaic Greece, as intimated primarily by the cast of epic and tragic actors, is an experience yoked to privation.12 At its core, it connotes the lack of home and thus the lack of a self-constituting centre, the myriad strands of family relations, physical place, community and the city all absent. The privations of the wanderer are therefore both material and transcendent in kind, and a pervasive sense of loss continually threatens the victim with the dire prospects of unsettled movement without end. The logic of wandering, in other words, always includes, in this tragic formulation, the potential for a perpetual affliction. Medea renders this logic in exemplary fashion when, sharpened by her rage, she distils her fate down to the singular penury of a wanderer: I have you, a marvellous husband and trustworthy in my wretchedness – if, at any rate, I will flee this country, without friends, in solitude, with only my children alone. Such a beautiful disgrace for a new groom, with your children and your saviour like beggars wandering (alasthai).13

For Medea, this beggarly condition of wandering translates into the explicit potential for constant movement without respite. ‘What city will receive?’ she asks, ‘Is there someone somewhere to shelter me, give me asylum, a safe place to live?’ (ibid., 386) The tragic fate of actors like Medea becomes particularly pronounced in their valiant, but failed, attempts to avoid the misfortune that is typical of a vagabond or one in exile (alētēs, planētēs), figures who are woven together in the very fabric of language with words that describe the literal act and condition of ceaseless wandering (alaomai, planē).14 Consider, in this light, Oedipus, who must weigh the consequences of his actions in the shadow of having been and once again becoming a wanderer. Both in his ignorance as well as in the knowledge of his hamartia, he ascribes his condition to a past in which he has ‘wept and walked many roads in the wanderings of his thought’ (phrontidos planois).15 Such a condition, to be sure, has as much to do with the physical act of wandering that makes him that ‘wayfarer’ (ho hodoiporos) (ibid., 66–67) he unwittingly associates with the killing of Laius as it does with the ubiquitous workings of his wandering

12

13 14 15

Montiglio (2005: 30) gives us an insightful and synoptic presentation of this theme of privation in her study of ‘exile’ as a focal experience of wandering in Ancient Greece. Medea 510–15 in Euripides 1984. Montiglio (2000: 86) Oedipus tyrannus 66–7 in Sophocles 1990. Henceforth, all citations made in the body of the essay.

362

Skill in Ancient Ethics

mind. Furthermore, this noetic dimension of wandering displays a vexed quality that grows more acute as Oedipus gains knowledge of his flaw. He speaks of the ‘wanderings of his soul (psukhēs planēma), the tumult of his mind’ (ibid., 727) as he submits to a future that is framed by the terminal punishment of exile. And, on his departure from Thebes, he pleads with Creon to care for his daughters, lest they be abandoned ‘beggarly, husbandless, of his kin, wandering’ (alōmenas) (ibid., 1506). The privations of the wanderer thus multiply across a diversity of fronts, both tangible and intangible. But there is more to the wanderer than just his or her abject suffering and loss, whether in body, mind or soul. As we find in the case of that great, exemplary wanderer of antiquity, Odysseus, there is less a quality of mental disturbance than there is one of resourcefulness that distinguishes the person associated with the condition or action of planē or alaomai. In Odysseus, we find a particular model of this resourcefulness that is distinctly epistemic in nature.16 On his return to Ithaca, he presents himself to Eumaeus, the swineherd and caretaker of his property, as one who, because of his ‘wanderings far and wide’ (polla d’ alēthēn), potentially brings important news or tidings (angellein).17 Odysseus, though, is not just any angelos, or messenger, but one who is set apart by the exclusivity of his message. His tidings are predicated on a summation of empirical situations that are available only to him, and, in this regard, he gains stature with his putative body of knowledge as a wanderer. The lasting impact of a Homeric model of planē may be surmised in various quarters of archaic and Ancient Greek culture that continue to uphold the rudimentary association between travel and the acquisition of knowledge. We are justified, for example, in connecting the basic epistemological skill of Odysseus qua wanderer and the institution of theōria, the ‘spectating’ or ‘witnessing’ and subsequent reportage carried out by the civic or religious office of the theōros who travels in order to gain knowledge.18 An early account of theōria, which inheres in a concrete experience of travel rather than in the more rarefied conceptions held by philosophers (which tend to assume a purely noetic

16

17 18

Marcel Detienne and Jean-Pierre Vernant identify this resourcefulness in terms of a metic intelligence, a quality that, in their words, sets apart ‘the tricks of a crafty character such as Odysseus’ (Detienne and Vernant 1991: 2). This skilful use of one’s knowledge, in other words, amounts to what Lisa Raphals, in her illuminating study of mētis in Ancient Greece and early China, has called a ‘cunning intelligence’ – a mode of intelligence ‘that relies on skill, strategy, and a general knack for handling whatever comes along’ (Raphals 1992: xii). The point that warrants our attention here is that mētis, as a skill in deploying one’s knowledge, finds particular traction in the figure of the ‘wanderer’. Odyssey, Bk 14, 120 in Homer 1995. See Nightingale 2009 for an illuminating analysis of the broader cultural institutions of theōria.

The Skilful Wanderer on the Risks and Rewards of Travel

363

spectacle), is available in the case of Solon. His departure from Athens, according to Herodotus, is kata theōriēs prophasin, ‘for the purpose of sightseeing’, and this motivation is subsequently identified by Croesus of Lydia as the explicit grounds for Solon’s reknowned wisdom: ‘[W]ord of your wisdom (sophiēs) and wandering (planēs) has reached us even here’, he tells his Athenian guest. ‘As a lover of wisdom, you have traveled much of the world for the purpose of seeing it.’19 Yet, the knowledge of the wanderer remains a double-edged sword. If its roots lie in a singular experience of bearing witness to the world, this singularity is equally grounds for suspicion. Eumaeus will thus respond in a common refrain that, inspite of the positive value that is endowed in the skill of wandering, nevertheless conjoins wandering and deceit. ‘[W]andering men (andres alētai), in need of care, speak lies and refuse to speak the truth’, he tells Odysseus.20 Between the prized wisdom of the sight-seeing Solon and the potential dissimulations of Odysseus, the wanderer in disguise, the epistemology of wandering is left mired in the ambivalence of motivations both noble and suspect. In the debate over the potential merits and demerits of wandering, Plato assumes arguably the strongest of available positions. His unequivocal loathing for the wanderer is manifest not only in his personal distaste for travel, as when he speaks of ‘having hated [his] Sicilian wandering (Sikelian planēn) and bad luck’21 (in reference to his sojourn in the court of the profligate Dionysus II), but also, and more importantly, in his theoretical objections to the sophists. Their misology, simply put, is attributable to their wandering, and their wandering, in Plato’s account, has been entirely emptied of any Solonic echo. Thus, we are told in Timeaus 19e that the sophistic class, because they ‘wander from city to city’, ‘stray from the mark’ in producing coherent and truthful definitions. Sophist 224a–b further spells out the motivations of this wandering in terms of the sophist’s professional identity as ‘merchant’ or ‘trader’ (ho emporos), where the products sold and passed around ‘into one city and out of another’ (polin te ek poleōs) are not physical objects but rather mathēmata, ‘learnables’.22 The goal of this sophistical wandering is, accordingly, a ‘merchandising of the soul’ (hē psukhemporikos) (ibid.). If I have earlier spoken of the ambivalence of wandering, understood in terms of its potential for creating a truly revalatory experience, on the one hand, as well

19 20

21 22

Histories, Bk 1.30 in Herodotus 1975. Odyssey, Bk 14, 124. See also Montiglio 2000 for a discussion of these Homeric and Herodotean modalities of wandering. Seventh Letter in Epistles, 350d Following the translation by Brann et. al. in Brann 1996: 25.

364

Skill in Ancient Ethics

as entailing the threat of dissimulation, on the other, we must now confront Plato’s unmitigated critique of wandering as a sophistical skill. Where the Solonic model of itinerancy endows the wanderer with wisdom, Plato’s assessment appears to empty the act of wandering of any positive epistemological content. Indeed, the rupture between wandering and knowledge is evidently severe in Plato’s estimation in that the wanderer, as sophist, is mired in a dialectic that remains committed only to the vagaries of doxa or mere opinion. This makes wandering a thoroughly unphilosophical skill.

5. Wandering in mere eikos: Sophistry in Phaedrus It is a rare occasion, then, in Phaedrus, when we encounter Socrates wandering beyond the city’s dromoi under the spell of the sophist’s beautiful, but untruthful, speech. This is a dialogue that is worthy of our attention simply by virtue of its mise en scène – the countryside, the many, indefinite paths to explore, and the ambulating discussion of Phaedrus and Socrates, one that repeatedly collapses into a palinode seeking forgiveness for its unphilosophical character. The dialogue is framed by an opening question and a final resolve that bluntly force the matter of the wanderer front and centre. Socrates begins with a request for information, or rather, for an accounting that must be given: ‘Beloved Phaedrus, where are you going, and where have you been’ (poi dē kai pothen)? (Phaedrus 227a). While this is a seemingly pedestrian question, one could argue that the formulation poi kai pothen has a peculiar rhetorical force across philosophical, religious and literary texts in emphasizing both the perils and the rewards of an experience of itinerancy.23 Phaedrus, young and impressionable, responds to the question by drawing on the advice of his older lover and sympathizer of sophists, Akoumenos, who believes that it is more refreshing to walk in the country than in the city’s dromoi. He tells Socrates, accordingly, that he is ‘going for a walk outside the city-

23

Plato appears particularly interested in this question, asking of several of his characters the question of wither (poi) and whence (pothen). Ion, like Phaedrus, begins with a version of this question addressed to the dialogue’s eponymous protagonist: ‘Whence have you come now to the city to visit us’ (pothen ta nun hemin epidedēmēkas)? (Ion 530a). See also Lysis 203a. The question resurfaces verbatim (‘poi kai pothen’) in the biography of Diogenes the Cynic (Diogenes Laertius, 6.59), a figure whose itinerancy is the bedrock upon which his identity as a kosmopolitēs stands. Later, the Neoplatonist Aeneas of Gaza will begin his Theophrastus, in an unmistakable resemblance, if not outright homage, to Plato’s Phaedrus, with the following words: ‘Poi dē kai pothen, Euxithee?’ (Aeneas Gazaeus & Mitylenaeus, 1836: 1) Consider, also, the accounting that Hagar must give to God’s messenger in Genesis 16.8, when, wandering in the wilderness, she is asked: ‘Where have you come from and where are you going (pothen erkhēi kai pou poreuēi)?’ (Alter 2019: 52).

The Skilful Wanderer on the Risks and Rewards of Travel

365

walls’ (peripaton exō teikhous) (ibid.). Rather than a meaningless exchange of pleasantries, Phaedrus’ response will instigate a singularly inauspicious episode in the Platonic oeuvre in which Socrates will wander beyond the environs of the agora. Their peripaton, to be sure, will be anything but refreshing. Socrates will repeatedly feel atopōtatos (228c),24 ‘most out of place’, and he will famously compare the entire episode to a hallucinatory experience brought on by the pharmakon of the sophist’s written speech and the promise of learning that it contains. But he will also overcome the drug’s spell and wrest control from Phaedrus, knowing where to ‘turn off ’ (ektrepein) (229a) so they can reorient themselves away from the countryside and back to the city’s dromoi. If Socrates’ inaugural question exposes him to the threat of wandering in the company of an aspiring sophist, his final, hortatory resolve puts a decisive end to this threat. ‘Let’s be going’ (iōmen) (279c), he tells Phaedrus, in an ironic echo of the question that opens their encounter. Let’s be going, that is, back to the city and back to philosophy. That the emphasis of Socrates’ critique in Phaedrus settles on the question of dialectical method is only a logical choice on Plato’s part. There is but an additional move in language that gets us from the dispute on which path, or hodos, to tread to which methodos to follow. The dialogue’s principal contribution to the debate on wandering arguably lies in exploiting this play in language. For if the sophist is, as we read in Timaeus 19e, a ‘class that wanders’ (genos . . . planēton), his wandering in Phaedrus emerges in light of a damning portrayal of the sophistic method of plausibility (to eikos). In the parenthetical space of the countryside, in between one city and the next,25 Socrates submits the tainted epistemological wares of the sophist, those deceptive mathēmata that he sells, to a rigorous test and finds them to be lacking in necessity, hē anankē, the ‘true essence (tēn ousian)’ of argument. Instead, the sophist is faulted for proceeding ‘as if ’ his arguments follow, ‘resorting to what seems plausible’ (237c). The specific mathēmata Phaedrus is hoping to learn during his countryside peripaton concern the nature of love and the associated problem of granting sexual favour. These are the themes of the written speech by the sophist Lysias he is closely guarding, until, of course, Socrates shows up insurmountably ‘sick with desire to hear speeches’ (238b). Their ensuing examination identifies eikos to be Lysias’ overriding method,26 and Socrates is quick to train his sights on the

24 25

26

See also Phaedrus 229c and 229e. Socrates identifies the pair’s location as such, telling Phaedrus: ‘I have in fact set my heart on hearing his words to such an extent so that if you were to go about a walk to Megara . . . ’ (227d). The term eikos appears numerous times in a short span (231c, 231e, 232c, 233a), underscoring the merely plausible status of argumentation in Lysias’ speech.

366

Skill in Ancient Ethics

problematic nature of the sophist’s (wilfully) unexamined premises. Responding to the argument that it is better to grant favours to non-lovers than lovers, Socrates draws Phaedrus’ attention to Lysias’ originating move – namely, to begin with the ‘plausib[ility] that lovers . . . will be excited to talk and show off about their affairs’ (232a). Gregory Vlastos has called such premises ‘iffy theses’ (Vlastos 1994: 8), ones that simultaneously entail unasserted premises that are equally unasserted counterfactuals. Socrates’ response to Phaedrus’ recitation of the speech begins with an oblique jab meant to convey the equal viability of such a counterfactual, which continually makes available an equally ‘truthful’ (read plausible) counterargument. He tells Phaedrus, ‘[t]here is a swelling in my chest and I feel as if I am able to compose a different speech, against his, and none the worse either’ (235c). The success of this different speech, as he continues in his acerbic tone, need not require him to possess knowledge ‘about these things’. Eikos, as a result, speaks to a condition of both ethical and epistemological compromise. In the context of his eristic encounters, the sophist is compelled by the desire for competitive victory in debate, and his mathēmata function within the constraints of a dialectical exercise decoupled from the goal of a true epistēmē. In its sophistical arrangement, a merely plausible argument is thus an act of intellectual deceit that simultaneously holds open the possibility of arguing otherwise, as determined by one’s baser desires (for victory, fame, wealth, etc.). The ignorance, or amathia, underlying this deceit, moreover, is decidedly unsocratic in nature, differing from the kind of productive aporias that mark an advance in the philosopher’s pursuit of truth. Where the latter’s intellectual exercises journey up the epistemological ladder, from categories of doxa (like ‘imagistic representation’/eikasia and ‘belief ’/pistis) to the ultimately paradoxical sphere of epistēmē (denoted by the objects of ‘discursive thinking’/dianoia and ‘intelligence’ or ‘intuition’/noesis), the sophist’s dialectic lacks any such telos and remains mired in the morass of ‘false beliefs’ (pistis pseudes).27 Phaedrus is emphatic in its portrayal of the sophist’s amathia as a product of a well-worked out methodological scheme to evade inquiry into the true nature of things. A philosophical dialectic, on the other hand, stands diametrically opposed to the sophist’s method, and requires that one begin with an inquiry into ‘a definition of love, showing what it is (hoion t’estin) and what power (dunamin) it has’ (237c–d). As a corrective to Lysias’ speech, Socrates thus substitutes the ‘iffy’ ground of to eikos with hē anankē, ‘necessity’, and thereby begins to extricate himself from the web of mere plausibility spun by the sophist’s 27

Gorgias 454d

The Skilful Wanderer on the Risks and Rewards of Travel

367

argument. Inquiring into the definition of what love is, and what dunamis it has, sends Socrates and Phaedrus back to the originating sets of premises in Lysias’ argument. For the philosopher, ‘agreement,’ ho homologia, on these fundamental concepts is necessary if the subsequent body of the argument is to have any epistemic viability. Socrates impresses upon Phaedrus the importance of this point. ‘Because they have not reached an agreement at the beginning of an investigation’, he says of Lysias and his company, ‘they proceed as if they know, resorting to what seems plausible, when they have agreed neither with themselves or with others’ (237c). The demands of anankē, in other words, require this crucial standard of ‘agreement’ to be met, and it is only through a Socratic logos that a discourse can approach true homologia and thus ground its arguments in necessity. We can return then to the literal scene of the countryside with a keener appreciation of just what is at stake for Plato. To wander in the country, outside the city walls, in between one city and the next, is to wander in the intellectual wilderness of false beliefs. The full force of the sophistical threat thus comes to the surface, for Plato, in this act of wandering outside the city. For it is here that Socrates catches the shape-shifter, the deceiver, the ‘merchandiser of the soul’, doing what he does best – sharpening his arguments so they may twist this way and that, from fact to counter-fact, as the opportunity warrants. If Phaedrus is simply following Akoumenas’ advice that it is ‘more refreshing’ to walk in the country than in the city, Plato serves up these refreshments for what they truly are. From one plausible thesis to the next, the sophist must refresh his arguments as he wanders from one city to the next.

6. Toward an ethics of wandering: Zhuangzi and Plato compared I have attempted to demonstrate above that the condition of the wanderer presents a salient philosophical problem in the works of both Plato and Zhuangzi. We may now consider whether, and how, these two models can be brought into dialogue. My comparative efforts, I should note at the outset, do not subscribe to the assumption that there is a coherent ‘Greek’ position on wandering, a singular ‘ethics’ we might capture, and likewise, a ‘Chinese’ one. I consider the individual analyses of Plato and Zhuangzi to have shown clearly enough that this is not the case. The goal of the comparison, then, is not to argue for identity or difference between some contrived monoliths of culture. Rather, successfully identifying

368

Skill in Ancient Ethics

the points of intersection between the two articulations of wandering involves establishing both the similarities and differences that obtain in the comparative space of analysis. This spectrum of points, both converging and diverging, serves two ends. First, it affords us greater conceptual depth of the problem in each of the cases. And second, we potentially stand to gain a more nuanced, multi-valent understanding of the problem of wandering in general. Recall that, in Zhuangzi, there are at least three discernible modalities of wandering. There is the aimless wanderer who eschews the built-up spaces of culture and political action, repairing instead to the wilderness; the political actor who must wander, as a passive strategy, within the ‘cage’ (fan ›) of the court; and the skilful wanderer, who is able to continually disaggregrate categories and re-think them as part of a larger, transformational ethic. In each of these models, wandering relates as much to a physical negotiation of space as it does to the epistemological criteria by which such spaces are made knowable. I have, in this respect, identified an apophatic model, where the act of wandering entails an explicitly negative orientation to questions of knowledge; a pragmatic mode in which the wanderer’s epistemic claims are situationally determined and often resort to a passive orientation to normative epistemes; and, finally, what we may designate as a transformational model, which continually re-thinks criteria for knowledge through an artful wandering within normative spaces. In the Qiwulun, the skill in continually transforming the criteria for epistemic judgment is articulated in relation to what the text calls the inherent so-ness (ran ❦) of things: Dao is walked and thus formed, things are named (wei 䄲) and are thus so (ran ❦). How are they so? So-ing makes them so (ran yu ran ❦ᯬ❦). How are they not so? Not so-ing makes them not so. Things (wu ⢙) surely have what is so, things surely have that which makes them plausible (ke ਟ). There is no thing that is not so (❦), there is no thing that is not plausible. Thus being so, a twig or a pillar, a leper or Xi Shi–broad, fluctuating, deviant, strange, dao connects them all as one. Their separation (fen ࠶) is their completion (cheng ᡀ), their completion is their ruin . . . Judge what is so on a given basis (yin shi ഐᱟ) and stop. Stopping without knowing (zhi) it is so, this is called dao. ZY 4/2/33–5/2/37

The outstanding feature of this passage is that it rejects the ‘separations’ (fen) of eristic debate (bian) while simultaneously granting a reality to ‘things’ in its positive ascription of so-ness (ran). ‘Things surely have what is so’, we are told, suggesting that the text upholds a certain mode of delimitation in which ‘things’ (wu ) can be named (wei). And whereas the shi–fei binary, in the context of

The Skilful Wanderer on the Risks and Rewards of Travel

369

eristic debate, organizes our judgments along the fixed categories of truth/zhen ⵏ–falsity/wei ‫( ڭ‬an overall structure that has a pejorative sense in Zhuangzi), a thing’s existence in terms of ran is deemed plausible (ke ਟ). We are reminded here of the Platonic critique of the sophistic method of to eikos, and one must entertain the possibility that Plato’s reasons for dismissing the wandering sophist equally apply to the wanderer’s judgments in the Zhuangzi. Are we, in other words, afforded a better grasp of the discussion surrounding the issue of ke when read through the prism of Plato’s dismissal of to eikos? The character ke, to be sure, reveals a texture of meanings, from implying stronger senses of assent or approval at one end of the spectrum to weaker notions of probability, at the other. The Confucian text Xunzi, for instance, works with a robust sense of ke, using the character to establish the deliberate act of assent or approval that emanates from the heart bent on overcoming the vagaries of one’s spontaneous dispositions and desires.28 In Zhuangzi, a decidedly weaker sense is at play. The plausibility of things, as discussed above, emerges from a ground of spontaneous generation, the so-ness of things, which is contextually given and inherently dynamic. The valence that ke carries in Zhuangzi should thus be heard in the context of the authors’ commitment to a perspectival epistemology that mandates a provisional basis for judgment rather than an exhaustive or absolute set of veridical categories. But, then, is Zhuangzi’s wandering sage a replica of the sophist, who likewise peddles a provisional logos? The similarities are superficial, and yet instructive. Plato’s stark appraisal of the sophist’s method affords us a clearer view of the manner in which Zhuangzi fashions its own deployment of a language of plausibility against competing frameworks, like those available in Xunzi. To a certain extent, then, the sophistic method of to eikos and Zhuangzi’s naming of things provisionally or with plausibility (ke) intersect in their common rejection of a singular Truth or Form with respect to this or that thing. Elsewhere in the Qiwulun, we encounter a shift in the terminology of debate, where the bankrupt dichotomies of shi/fei give way to the sage’s capacity to articulate a thing as both ‘this’ (shi ᱟ) and ‘that’ (bi ᖬ) owing to the fluid and contextual nature of experience.29 The ensuing plausibility that animates each of the Zhuangzi’s dialectical accounts, often assuming contradictory positions on a given set of issues, thus bears a thoroughgoing resemblance to the shifting accounts of the sophists.

28 29

On Xunzi’s use of the character ke see Stalnaker 2016. ZY 4/2/27–29

370

Skill in Ancient Ethics

Yet this remains a superficial resemblance. Nowhere in Zhuangzi does the plausible dialectic of the wandering sage exhibit any of the commodification of knowledge, its mathemata, that we witness in the case of the sophist. Instead, underlying the text’s provisional accounts is a deeper ontological commitment to the nature of dao as a dynamic, spontaneous, generative continuum in which ‘things’ exist in provisional ways. Consider the following description of an experience of wandering that Zhuangzi invites one of his interlocutors to undertake together: Try and wander together in the palace of not-even-anything (wu he you zhi gong ❑օᴹѻᇞ), debating (lun 䄆) together in unity, without ever approaching exhaustion (qiong マ). Try and [wander] together in non-action . . . Nowhere do I go and yet I do not know where I end up. I leave and I return and I never know where I will come to a stop; I’ve already gone and returned, and I do not know if [the wandering] is over . . . Great knowledge enters there and I do not know where it reaches exhaustion. ZY 59/22/47–50

The refrain of inexhaustibility (wu qiong ❑マ) across the text30 resurfaces here in the idea of a free and unlimited wandering in a world that elicits endless responses from an adept like Zhuangzi. Moreover, the distinctive ontology of this world, rendered here as the ‘palace of not-even-anything’, immediately catches our eye. It admits a class of beings unmoored from fixed, unchanging criteria and thus free from the bias of a totalizing episteme. Zhuangzi, wandering in the ‘palace of not-even-anything’, is much like the zhiren wandering in the ‘world’ (shi), both actors articulating a ‘perfected’ (zhi 㠣) or ‘great’ (da བྷ) knowledge that consists entirely in a skilful responsivenes to the dynamic and emergent quality of things. Zhuangzi’s inexhaustible wandering in ‘debate’ once again suggests the capacity to rethink, revise, and reapply dialectical categories. In doing so, however, his wandering marks a complete and total exit from an eristic space. This can never be the case with that ‘merchandiser of the soul,’ the sophist. One last point warrants our attention. In spite of Plato’s antipathy to the figure of the wanderer, he nevertheless occasionally speaks of the philosopher himself, and his art, in terms of a planē. Thus, in Apology 22a, Socrates calls his investigation of the Oracle’s assertions tēn emēn planēn, ‘my wandering’, essentially attributing to each Socratic encounter the quality of ‘wandering.’ Yet, even if we were to put such an expression down to a mere turn of phrase, Parmenides 135e alerts us to

30

See, for example, ZY 2/1/21, 4/2/31, 21/7/32, etc.

The Skilful Wanderer on the Risks and Rewards of Travel

371

another incidence of the language of wandering in connection with the Socratic encounter, and with something decidedly more crucial at stake. Here, Parmenides praises the young Socrates for urging Zeno to not remain invested ‘in a world of visible things (en tois horōmenois) or inspect them in relation to their wandering’ (peri tauta tēn planēn episkopein). The ontological ramifications of the problem of wandering, as discussed in the case of Zhuangzi, afford us an additional comparative vantage from which we might gain a better understanding of these seemingly atypical notions of the wandering Socrates. I want to suggest that, while these two citations appear to point in different directions, they are in fact linked in their common reference to the ontological dimension of wandering in Plato’s dialogues. In this frame of analysis, Plato describes the movement of the phenomenal world, the ‘things seen’, as a ‘wandering’. This is the world of embodied existence and constant change – of birth, growth and decay – in which the philosopher is unavoidably thrown. But if this ‘wandering’ realm of the visible is unavoidable, it is certainly not insurmountable. The philosopher’s skill resides precisely in the ability to navigate this phenomenal world and reorient oneself, like Socrates does with Zeno, in the direction of an ontological realm unavailable to the senses. In this place of true Being, the philosopher’s planē transforms into a more orderly, teleological movement akin to the uniform revolutions of the gods who tour the place that is huperouranion, ‘beyond heaven’ (Phaedrus 247c). Silvia Montiglio, alerting us to the shift in Plato’s language from planē to periodos, a ‘perfect cycle’, has described this celestial journey as one in which ‘there are no detours, shifts, irregularities’.31 In Zhuangzi, where no ontological difference obtains between the seen and the unseen, between the Form and its phenomenal copy, the sage must negotiate movement in the absence of a foundation of eternal, unchanging Being. Where Plato seeks the uniform movement of the heavens, Zhuangzi embraces a life of wandering in the only world given, one that arranges and rearranges itself across multiple axes.

Works Cited Aeneas Gazaeus, and Z. Mitylenaeus (1836), Aeneas Gazaeus et Zacharias Mitylenaeus de immortalitate animae et mundi consummatione (J. F. Boissonade Ed.): J. A. Mercklein.

31

Montiglio 2005: 167.

372

Skill in Ancient Ethics

Alter, R. (2019), The Hebrew Bible: A Translation with Commentary, New York: W. W. Norton. Brann, E. et al. (1996), Plato: Sophist or The Professor of Wisdom, Newburyport: Focus Publishing. Detienne, M. and Vernant, J. (1991), Cunning Intelligence in Greek Culture and Society, Chicago: University of Chicago Press. Euripides. (1984), Euripidis fabulae, ed. J. Diggle, Oxford: Clarendon Press. Fraser, C. (2014), ‘Wandering the way: A eudaimonistic approach to the Zhuangzi’, Dao 13 (4), 541–565. Hawkes, D. (1967), ‘The Quest of the Goddess’, Asia Major, 13 (1–2): 71. Herodotus (1975), Herodotus. Books I and II , trans. A.D. Godley, Cambridge, MA : Harvard University Press. Homer (1995), Odyssey, Books 13–24, ed. J. Henderson, Cambridge: Harvard University Press. Ivanhoe, P.J. (1993), ‘Zhuangzi on Skepticism, Skill, and the Ineffable Dao’, Journal of the American Academy of Religion, 61 (4): 639–654. Kroll, P.W. (1996), ‘On “Far Roaming” ’, Journal of the American Oriental Society, 653–669. Levinovitz, A. (2012), ‘The Zhuangzi and You 䙺: Defining an Ideal Without Contradiction’, Dao 11 (4): 479–496. Li, X.A. (2018), ‘Playful You in the Zhuangzi and Six Dynasties Literati Writing’, Journal of the British Association for Chinese Studies, 8 (2): 1–28. Mair, V.H. (1983), ‘Wandering in and through the Chuang-tzu’, Journal of Chinese Religions 11 (1): 106–117. Montiglio, S. (2000), ‘Wandering Philosophers in Classical Greece’, The Journal of Hellenic Studies 120: 86–105. Montiglio, S. (2005), Wandering in Ancient Greek Culture, Chicago: University of Chicago Press. Nightingale, A.W. (2004), Spectacles of Truth in Classical Greek Philosophy: Theoria in its Cultural Context, Cambridge: Cambridge University Press. Plato (2005), Plato: Timaeus. Critias. Cleitophon. Menexenus. Epistles, trans. R.G. Bury, Cambridge, MA : Harvard University Press. Plato (1925), Plato: Lysis. Symposium. Gorgias, trans. W.R.M. Lamb, Cambridge, MA : Harvard University Press. Plato (2001), Plato: Euthyphro. Apology. Crito. Phaedo. Phaedrus, trans. H.N. Fowler, Cambridge, MA : Harvard University Press. Plato (1921), Plato: Theaetetus. Sophist, trans. H.N. Fowler, Cambridge, MA : Harvard University Press. Raphals, L. (1992), Knowing Words: Wisdom and Cunning in the Classical Traditions of China and Greece, Ithaca, NY: Cornell University Press. Sophocles (1990), Sophoclis fabulae, ed. H.W. Lloyd-Jones, N.G., Oxford: Clarendon Press.

The Skilful Wanderer on the Risks and Rewards of Travel

373

Stalnaker, A. (2016), ‘Xunzi on Self-Cultivation’, in E.L. Hutton (ed.), Dao Companion to the Philosophy of Xunzi (Vol. 7), Dordrecht: Springer. Vlastos, G. (1994), Socratic Studies, New York: Cambridge University Press. Zhuangzi (1956), Zhuangzi Yinde (A Concordance to Chuang Tzu) (Vol. Supplement No. 20), Cambridge, MA : Harvard University Press.

374

Index according (yin ഐ) 246–247, 251, 360 accumulation 222, 226, 233, 235 actors 311–313, 357, 361, 370 Aëtius 171–174, 176–177, 180 agency 206, 219, 223–225, 227, 234, 237, 246, 249, 256, 258, 281, 291, 295, 299, 313, 322, 340 aisthēsis (perception) 167, 169 aitia (cause, explanation) 144, 148, 159 alaomai (to wander) 361–362 Analects 220, 267, 274–276, 279, 333, 349 anankē (necessity) 353, 365–367 Annas, J. E. 3, 32, 81, 83–84, 98, 115, 119, 181, 187, 193, 223–224, 236–237 Apology 13–15, 18–21, 24–26, 28, 47, 68 Aquinas, St Thomas 148 archē (origin, principle) 126, 143, 149, 155, 159 Archer Yi 334, 336–339 archery 9, 307–308, 316, 329, 331–341, 346–347, 350 architecture 119, 186, 191, 194 aretē (virtue) 53–54, 115–118, 120, 131–132, 143, 155, 157 Aristeides 14–15 Aristotle 2, 4–5, 10, 16, 66, 85, 88, 90, 95, 97–99, 102, 106, 111, 114–133, 136–139, 141–162, 187, 199, 223, 316, 317, 320–321, 329, 332, 340, 345–347, 349 Arius Didymus 176, 179, 181 attention 206, 210–214, 261, 265, 270–271, 282, 284–285, 330, 353, 362, 364, 366, 370 ball-players 313 bian 䗟 (eristic debate) 357–358, 360 bu dong xin нअᗳ (unmoved heart/ mind) 203–204, 209, 216, 219 bu neng н㜭 (inability; impossibility) 208 butchery 8–9, 245, 255, 276, 281, 288, 291–293, 295–296, 298, 301, 309, 312–314, 316, 323–324, 331–332, 343

Callicles 13, 15–16, 20, 24, 27, 55, 67–73 carpentry 7, 22, 37, 78, 87, 91, 160, 222, 226, 232–234, 241, 243–244, 248, 261, 263, 284, 336, 344 CBT (cognitive behavioural therapy) 2, 5, 89, 117, 129, 135–137 character 2, 5, 38, 98–9, 115–117, 119, 1, 21–24, 126, 129, 131–132, 137, 147, 149, 156–157, 161, 185, 193, 224–225, 228, 256, 261, 263, 318, 331, 360, 362, 364 Charmides 13, 77 Chrysippus 5, 165–182 Cicada-catcher 261, 266, 281–282, 286–287, 289–291, 293, 295–296, 298, 302, 310, 324, 332 Cicero 169–170, 173–174, 176, 180 Cleanthes 174 cobbling 53–57, 68, 91, 142, 154, 186 Confucius 210, 220, 265, 267, 270, 274–276, 279, 286–287, 291–294, 296, 299–301, 303, 309–311, 333, 340, 346–347, 349, 356–358, 360 Cook Ding 245, 248–251, 254, 256, 261, 266–273, 277–278, 309 cookery 43, 88, 111–113, 249, 266, 269 courage 92, 100, 117–118, 122–123, 129–130, 180, 344 Cratylus 31, 34–36, 44–46 cultivation (of virtue) 5, 7, 8–9, 115–116, 132–134, 171, 177, 204–209, 219–223, 227–230, 233–236, 238, 261–262, 265, 290, 298, 301, 307, 322, 324, 341, 373 daimonion (Socrates’s) 21–22, 24, 28 dancing 18, 85–87, 241, 245, 291–292, 309, 354 dao 䚃 219–220, 226, 237–238, 241–259, 261–264, 267–268, 272–273, 275, 278–279, 281–296, 298–303, 307–308, 313–316, 318–319, 321, 323, 326–327, 331–332, 334, 347, 354, 368, 370, 372–373

375

376

Index

Daoist 249–250, 279, 285, 287, 290, 298, 302–303, 307–308, 313–316, 318–319, 321, 323, 326, 334, 354 de ᗧ 206, 211, 242, 255–257 deliberation 4, 32, 60, 79, 91, 97–98, 100–107, 109–110, 112–114, 120–121, 133, 149, 154, 157–158, 198, 206, 215–216, 222, 224–225, 228–9, 231, 235, 300, 317, 321, 369 democracy 14 detachment 6–7, 262, 267–268, 277, 285, 313, 315 Detienne, M. 152–153 Dewey, J. 4, 76, 86–92 dexterity 291–292 Diogenes Laërtius 166–169, 174, 176, 178, 180 disposition 2, 7, 87, 116–120, 130, 132–133, 136, 148, 150, 156, 166, 168, 175–176, 178–180, 198, 204–205, 209–211, 215, 219, 224–225, 232, 236, 282–283, 287, 319, 325, 336, 359, 369; see also hexis Doris, J. M. 224 Dreyfus, H. L. 223–224, 228 duan ㄟ (sprout; tip) 203, 205, 209, 218 dunamis (power, capacity, potentiality) 83, 93, 143, 145, 158–159

gentleman, see junzi golden age 186, 190–191, 193 goods, internal/external 233–234, 317 Gorgias 3, 13–15, 17, 19–20, 27, 29, 31–32, 34, 43, 45, 49, 51–73, 85 government 6, 9, 186, 218, 221–222, 261, 267, 273–274, 276, 329, 331–335, 342–343, 346, 355 gradability (of craft) 1, 16–19, 24 Guo Xiang 246 gymnastics 17, 24, 38, 43, 52, 87, 151, 154, 345

ease 227, 231–233, 256, 268–269, 282, 310, 313 eikos (probability) 353, 364–366, 369 elenchos (Socratic) 26–27, 56, 59, 70 empeiria (knack) 43–44, 53, 62, 72, 119 engagement 258, 262, 275, 283–284, 300, 325, 358–359 epistēmē (knowledge) 144, 168, 171, 174–176 ergon (function) 18, 77–78, 80, 83 eudaimonia (happiness) 23, 50, 83 Euthydemus 18, 22–24

initiative 283, 289, 293–294, 300–302 Inner Chapters 261–262, 266, 277–279, 281, 302 intellectualism 16–18, 94, 115–116, 119–121, 126, 137, 160

flow experience 229 flute-playing 34, 36–39, 41, 45, 48–49, 53, 57, 118–119, 151, 160, 187 friendship 77–78, 105–106, 109, 111–113, 134, 158, 187, 196–197, 230, 257, 263, 361

kairos (right time) 150–153 katalēpsis (apprehension) 169, 174–175 kathēkon (appropriate) 5, 166, 179 ke ਟ (plausible) 368–369 kinēsis (movement) 158–159 King Xuan 208, 217

Galen 168, 171, 177 generosity 4, 100, 103, 105–106, 110–112, 117

habit/habituation 2, 4–6, 86–92, 116–118, 121–123, 126–127, 131–133, 136–137, 159, 185, 212, 222, 233, 235, 271 Hacker–Wright, J. 115, 120–121, 123 Hadot, P. 172, 180–181 happiness (in ancient Chinese ethics) 254, 313, 317, 321 hexis (disposition) 122–123, 129, 143, 155–156, 159, 161, 174 Homer 153, 344 hormē (impulse) 168, 175, 177–179 horse/horse-riding 26, 43, 57, 65–66, 79, 100–101, 155, 173–174, 230 Hui Shi (Huizi) 257, 264, 266–267, 274

junzi ੋᆀ (gentleman) 225, 281, 333, 341 justice 2, 4, 14, 16–17, 24–25, 27, 35, 37–38, 43, 51–52, 54–56, 58–60, 62–73, 76–84, 86–87, 91–93, 100, 106–107, 155–156, 170, 179–180, 217, 231

Laches 13, 15, 17, 19, 49, 77, 82 legislation 14, 17, 24–25, 27, 43, 45, 87

Index li (ritual) 203–218, 220–238, 241–259, 261–279, 281–303, 307–327, 329–372 Li ⨶ (pattern) 247, 249–256, 291–292, 297, 318, 325 Lienü Zhuan 348 Liezi 286, 290, 303, 336–338, 346, 348 logos (reason) 121, 144–145, 150, 153–156, 160, 165, 167 looms 326 MacIntyre, A. C. 233–234 manifestation 206, 209–219, 293 Mean 2, 98, 113, 122, 130, 339–341, 346 medicine 4, 17, 24, 43, 52, 54–55, 58, 62, 64, 72, 76–77, 85, 87–90, 92, 142, 144, 147–148, 151, 153–154, 156, 182, 194, 316–317, 331, 339, 343 Mencius 203–220, 238, 270, 274, 279, 331, 334–336, 338, 341, 346 Meno 3, 14–15, 17, 31–34, 36–38, 41 Mo Di, see Mozi Models (fa ⌅) 243–245, 247, 296 Moss, J. 27, 102, 115–117, 122, 124–129, 131–132, 187 Mozi 241–243, 245, 247, 249, 251, 253, 255, 257, 259 Nature, see Xing navigation 17, 43, 86, 88, 151, 153–154, 294, 307, 316 Nussbaum, M. C. 85, 125, 127, 151, 190 oikeiōsis (appropriation) 166–167, 173, 180 orektikon (desiring part of the soul) 116, 121–123, 125–126 pathos (emotion) 115–116, 118, 121–123, 127, 129, 131 Pattern, see li Pericles 14, 33, 61–62 periodos (perfect cycle) 371 Phaedo 41 Phaedrus 73, 88, 353, 364–367, 371–372 phantasia (imagination) 5, 116–117, 121, 123–129, 131, 168–169, 171, 175 phronēsis (practical wisdom) 5, 104, 115–116, 124, 126, 131–133, 141, 143–144, 146–148, 151, 153, 161–162, 187

377

planē (wandering) 352–353, 361–365, 370 Plutarch 167, 171, 177, 180 poiēsis (making) 141–142, 146–147, 149–151, 154, 158 politics 3, 13–28, 37, 88, 116, 138, 151, 158, 262, 273, 275–276, 278, 302–303, 347–348 Posidonius 186–187, 193 potentiality 215, 272, 276–277; see also dunamis practical intelligence 224–225, 236, 238; see also phronēsis praxis (action) 77, 85, 118, 141, 147, 151, 153 proficiens (striver after virtue) 6, 185–186, 194–198 prohairēsis (choice, decision) 118, 131–132 propriety 203–206, 210, 212, 214–218, 225, 296–297, 299, 355 Protagoras 2, 15, 17, 31–50, 70 punishment 37–38, 55, 132, 136, 335, 339, 362 Pythagoras 41 qi ≓ (vital energies) 204, 211–214, 216–217, 219, 274, 276, 298, 324, 352 qiao ᐗ (skill) 283, 286, 335 Republic 2, 13, 70, 72–73, 76, 90, 92 rhetoric 3–4, 43, 51, 56, 59–73, 127–129, 144, 151, 154, 176, 330, 364 ritual 203, 207–308, 221–223, 225–227, 229, 231–238, 245, 250, 253, 289, 296–297, 324, 326, 333, 352, 355 ritual learning 223, 229, 234 Roochnik, D. 52, 83–85 Ryle, G. 8, 160 sage 2, 5–6, 221–222, 225–30, 233–236, 241, 243, 248, 263–267, 270–272, 274, 277, 289, 298, 307, 310, 312–313, 316–319, 325, 331, 333, 354–355, 360 science 1, 6, 31–32, 36, 39–43, 48–49, 174–176, 192, 198, 308, 330, 340 sculpting 154 Seneca 6, 166, 185–199 senses/sense-perception 125, 128, 157, 160, 171, 205–206, 210–212, 215, 231, 285, 357, 369

378

Index

Sextus Empiricus 168–169, 174–178 Sherman, N. 119–120, 124, 131, 133 shi ᱲ (timeliness) 262, 265, 270, 277 shi/bi ᱟᖬ (‘this’/‘that’ ) 360, 369 shi/fei ᱟ䶎 (‘so/‘not so’ ) 203, 297, 357, 368–369 shoe-making, see cobbling si ᙍ (think, deliberate, reflect) 205–206, 218, 235 situationism 238 sophist/sophistry 9, 33, 43, 87, 100, 269, 351, 353, 359, 363–367, 369–370, 372 soul 5, 24, 26–27, 53–56, 58–59, 62, 67–73, 79, 93, 109, 116, 121–123, 126–127, 131–132, 145, 148–149, 159–161, 171, 178, 188, 320, 325, 344, 347, 362–363, 367, 370 Sprague, R. K. 79, 84–5, 91 telos (end, purpose) 142–143, 145, 148, 152, 157, 166–167, 174, 176, 181–182, 320, 322, 353, 366, 371 Theaetetus 31, 35–39, 41, 45, 48–49, 57, 118–119, 151, 160, 187 theōria (contemplation, spectacle, theory) 352, 362 Timaeus 157 tracks, leaving 298–302 tuchē (chance, luck) 2, 22, 150–152 unity (of virtues) 17–18, 98 uselessness 255, 261–9, 271–279 Vernant, J.–P. 152–153 Vlastos, G. 14, 35–37, 83, 85 wandering 255, 257–258, 261, 268–269, 272–273, 277, 279, 302, 352–356, 358–365, 367–372

weaving 142, 186–187, 189, 307, 329, 331–332, 339, 341–347, 349 wheelwright 241, 243–244, 247–248, 259, 261, 266–267, 274, 276, 281–282, 296, 298, 310, 318, 332 women 331–332, 338, 341–343, 345–346, 348–349 xin ᗳ (heart/mind) 203–222, 227, 243, 263, 278, 282, 284, 290, 297–298, 322, 324, 334–335, 342, 352, 356, 358 xing ᙗ (nature) 221–223, 322–323 Xunzi 220–223, 225–238, 266–267, 279, 297, 322–323, 327, 334, 336, 346, 349, 369, 373 yi 㗙 (propriety) 203–204, 207, 212, 218, 221, 267, 296, 355 you 䙺 (wandering) 213, 232, 249, 255–256, 268–269, 272, 274, 276, 283, 285–288, 290–291, 293, 297, 300, 309–310, 322, 335, 337–338, 341, 344–345, 352–361, 363–365, 370–372 Zeno 167, 174 zhi ⸕ (knowledge) 225, 289, 324–325, 355–356, 359–360 zhi ᘇ (will; goal of the heart/mind) 207, 211–219 zhi ren 㠣Ӫ (‘perfected persons’) 358–359 zhi zhi 㠣⸕ (‘perfected knowledge’) 289, 359–360 Zhuang Zhou 245, 261, 269, 274 Zhuangzi 241–303, 307–311, 313–327, 331–332, 334–337, 343, 346, 351–360, 367–373

379

380

381

382