PROCEEDINGS OF THE BOSTON AREA COLLOQUIUM IN ANCIENT PHILOSOPHY

PROCEEDINGS OF THE BOSTON AREA COLLOQUIUM IN ANCIENT PHILOSOPHY

VOLUME XXI, 2005

EDITED BY

JOHN J. CLEARY

AND

GARY M. GURTLER, S.J.

LEIDEN BOSTON 2006

BRILL

This book is printed on acid-free paper.

ISSN ISBN-13: ISBN-10: ISBN-13: ISBN-10:

1059-986X 978 90 04 15353 0 (Pbk) 90 04 15353 5 (Pbk) 978 90 04 15391 2 (Bound) 90 04 15391 8 (Bound)

Copyright 2006 by Koninklijke Brill NV, Leiden, The Netherlands Koninklijke Brill NV incorporates the imprints Brill Academic Publishers, Martinus Nijhoff Publishers and VSP. All rights reserved. No part of this publication may be reproduced, translated, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission from the publisher. Authorization to photocopy items for internal or personal use is granted by Brill provided that the appropriate fees are paid directly to The Copyright Clearance Center, 222 Rosewood Drive, Suite 910 Danvers MA 01923, USA. Fees are subject to change.
PRINTED IN THE NETHERLANDS

CONTENTS

Preface ...................................................................................... Introduction ...............................................................................

COLLOQUIUM 1 On Platos
STEPHEN MENN .......................................................................................

vii xi

1 54

Menn Bibliography ..................................................................................

COLLOQUIUM 2 Wishing for Fortune, Choosing Activity: Aristotle on External Goods and Happiness
ERIC BROWN ............................................................................................

57

Commentary on Brown
GARY M. GURTLER, S. J. ......................................................................... 82 87

Brown/Gurtler Bibliography ..................................................................

COLLOQUIUM 3 Aristotle on
ALFREDO FERRARIN ................................................................................

89

Commentary on Ferrarin
KLAUS BRINKMANN ................................................................................ 113 122

Ferrarin/Brinkmann Bibliography .........................................................

COLLOQUIUM 4 Enchanting the Souls: On Platos Conception of Law and Preambles
JEAN-FRANOIS PRADEAU .......................................................................

125

vi

CONTENTS

Commentary on Pradeau
GAVIN T. COLVERT ................................................................................. 138 154

Pradeau/Colvert Bibliography ................................................................

COLLOQUIUM 5 Aristotle and the Metaphysics of Metaphor
FRAN OROURKE .....................................................................................

155

Commentary on ORourke
IOANNA PATSIOTI-TSACPOUNIDI ............................................................. 178 186

ORourke/Patsioti-Tsacpounidi Bibliography .....................................

COLLOQUIUM 6 Was Aristotle a Particularist?
A. W. PRICE .............................................................................................

191

Commentary on Price
BRIDGET CLARKE .................................................................................... 213 232

Price/Clarke Bibliography ......................................................................

COLLOQUIUM 7 In Defense of Inner Sense: Aristotle on Perceiving That One Sees
THOMAS K. JOHANSEN ............................................................................

235

Commentary on Johansen
ARYEH KOSMAN....................................................................................... 277 284 287 291

Johansen/Kosman Bibliography ............................................................ About our Contributors ........................................................................... Index of Names .........................................................................................

PREFACE
JOHN J. CLEARY

Regular readers of this series will probably expect me to remark on the significance of the fact that the BACAP Proceedings has now come of age with the publication of Volume 21 given that, in some prefaces to previous volumes, I have occasionally talked about the personal significance of multiples of the number 7 for my involvement in this whole project. Within the life of a person, the age of 21 is generally seen as marking ones coming of age but, in the life of a series like this, it is more plausible to view it as old age. In any event, it is perhaps opportune for me to do some stock-taking after such a long involvement with these Proceedings of which I am the founding editor. If one adopts the age-old convention that 7 years constitutes a stage in life, then one might regard the first seven volumes as reflecting youth and enthusiasm, when the ambitions expressed in my Preface to Volume 1 were seen as achievable. For instance, I entertained the possibility of an open dialogue between radically different traditions of interpretation within the field of ancient philosophy, while suggesting that we use the classical texts themselves as a basis for ongoing discussion. Here is what I wrote then in my Introduction: While I acknowledge as an historical fact the diversity of traditions in philosophy and philosophical interpretation, I consider the exclusive dichotomy between analyst and pluralist to be a result of academic politics. I continue to hold this view, although I have become less sanguine over the years about the possibility of overcoming the institutional obstacles to dialogue. In this regard, Volumes 8 to 14 represent a coming of age for the series, marked by a growing awareness of the difficulties facing such an open dialogue, so that many compromises became necessary to keep the conversation going between quite different styles of philosophical thinking. Finally, Volumes 15 to 21 may be seen as a kind of old age, reflecting a growing disillusionment with competing academic factions. Increasingly, I have come to accept that it is almost impossible to sustain a dialogue between those who are engaged in professional competition for jobs and the control of scarce funding. It is akin to the difficulty of keeping a conversation going between the deaf and

viii

JOHN J. CLEARY

the blind, who are driven by rivalry and ill-will. 1 Instead of an open and honest exchange of views, there are citation cartels which seem designed to promote academic careers rather than to promote dialogue. Most depressing of all, there seems to be an increasing unwillingness to read the work of other scholars with sympathy and understanding, especially when they publish in languages other than English. Perhaps this reflects a contemporary trend towards new brands of scholasticism, reflecting the parochialism of graduate schools and traditions of interpretation that have become quite insular in their outlooks. But as notable exceptions to such trends, there are some scholars who sustain my belief in the possibility of a fruitful exchange between different traditions of interpretation, especially for the reading of Platonic dialogues which cry out for many different approaches. Just as keen analytical skill is necessary for examining the arguments of the dialogues, so also an awareness of their dramatic and other literary features is essential for grasping the full philosophical import of these masterpieces. Some exemplary cases of such a multi-layered approach are Mitch Millers book on Platos Parmenides and, more generally, Charles Kahns work on the Platonic dialogue. 2 In this volume Stephen Menn makes a compelling case for the necessity of such literary, historical, and contextual approaches to reading Platos dialogues. Thus we are happy to include his paper in the present volume, even though it was presented within the previous years program so that, unfortunately, it lacks the usual commentary. We are also proud to continue our tradition of including a significant number of European scholars, whose first language is not English, though they all speak and write English fluently. Of course, it is no coincidence that many of the scholars writing in this volume represent different ways of interpreting ancient philosophical texts. Most of the colloquia conform to the typical structure of our Proceedings, and this volume contains (with one exception as mentioned) the papers and commentaries that were originally presented during the 2004-5 academic year at different meetings of the Boston Area Colloquium in Ancient Philosophy. Each colloquium represents the activities of a single meeting at one of the following participating institutions: Boston College, Boston University, Brown University, Clark

_________
1 The breakdown in dialogue between different modern traditions in philosophy is nicely described by Charles Taylor, in the Introduction to his book on Hegel and Modern Society (Cambridge, 1979), p. ix. 2 M.H. Miller, Platos Parmenides. The Conversion of the Soul, (Princeton, 1986). C. Kahn, Plato and the Socratic Dialogue. The Philosophical Use of a Literary Form, (Cambridge, 1998).

PREFACE

ix

University, the College of the Holy Cross, and Dartmouth College. In most cases, we try to retain the dialogical character of such colloquia by publishing both the paper and commentary from each of the meetings. In many cases, however, these oral presentations have been extensively revised by their authors in the light of subsequent discussions, and especially in response to critical comments from our external referees. I would like to thank the following referees who helped us to maintain our academic standards by providing readers reports on the papers included in this volume: Sarah Broadie, Victor Caston. Alessandra Fussi, Francisco Lisi, Marina McCoy, Deborah Modrak, Michael Pakaluk, and James Wilberding. At the end of the volume, together with the section About our Contributors, readers will find a general index of names that was collated by our editorial assistant, Michael J. Smith. As a poor substitute for an index of contents, readers can turn to my Introduction which tries to summarize some of the main topics covered by the papers published in this volume. In conclusion, I wish to thank my colleagues on the BACAP committee for their ongoing commitment to the whole program, whose survival continually depends on their voluntary work at each of the participating institutions. But I would particularly like to thank Gary Gurtler, my colleague and co-editor at Boston College, for his cooperation and friendship which helped me through a personal bereavement during this last year. I also want to thank our editorial assistant, Michael J. Smith, for his careful work in preparing camera-ready copy for this volume in the Philosophy Department at Boston College. Finally, I wish to acknowledge the continued financial assistance provided by the administrators at Boston College, whose enlightened support for the whole project over the past 21 years has been exemplary.
BOSTON COLLEGE & NUI MAYNOOTH (IRELAND)

INTRODUCTION
JOHN J. CLEARY

By comparison with the previous volume which was dominated by papers on Plato, the balance of power in this volume has decisively shifted towards Aristotle who is the subject for five of the colloquia, while the remaining two are dedicated to Plato. Indeed, the dominance of Aristotle would have been even more complete, if we had not included Stephen Menns paper from the previous years BACAP program. The two colloquia given over to Plato focus on political topics from his Republic and Laws, while the colloquia on Aristotle cover ethical, psychological and poetical topics. As usual, my introduction is intended to function as a rather poor substitute for an index of contents.

I.
In his wide-ranging and scholarly paper, Stephen Menn suggests that Platos Republic and Laws should be read within the historical context of an ancient Greek tradition of writings On the Politeia, which was an established genre different from Socratic writings. He claims that such a broader perspective can help to resolve some perennial problems that arise both about the general interpretation of Platos views in the Republic (e.g., Is he pro-Spartan?) and about the internal relations between different parts of the text (e.g., the relationship between Book I and the rest of the work). Thus, in the first half of his paper, Menn tries to reconstruct from the extant evidence, some of the salient features of this pre-existing tradition of political writings, which provides the background for Platos Republic. Within this genre, the constitution of the Spartans seems to have played a prominent role, either as an ideal to be emulated (by aristocrats) or to be criticized (by democrats). In effect, these writings contain not so much descriptions of how cities actually govern their affairs but prescriptions for how best they ought to be governed. Whichever attitude is adopted by the writer in question, he can hardly avoid dealing with the Spartan constitution, just as Plato does in a critical fashion both in the Republic and the Laws. If one wants to consult a pro-Spartan work, one has only to look at Xenophons largely uncritical Politeia of the Spartans, or the more ambiguous anonymous work, called Politeia of the Athenians, both of which almost certainly pre-date Platos Republic.

xii

JOHN J. CLEARY

By pointing to these extant works, Menn argues convincingly against the claim that Plato was the originator of the politeia genre of writing. This also has a direct bearing on the relation between Book I and the rest of the Republic. Contrary to the standard interpretation that Plato depicts types of polis so as to provide large-scale models for types of soul, Menn points out that the link between the question of justice and types of politeia has already been raised by Thrasymachus in Book I, where he launches a specifically political challenge to justice. Thus, within the context of his Socratic dialogue on justice, Plato wants to refute the claims of Thrasymachus that rulers in existing politeiai rule in their own interests, by giving an alternative positive account of what justice is; and by establishing that a just politeia is possible because ideal justice is rooted in nature rather than merely in convention. According to Menn, the amount of attention given to the Thrasymachean conception of justice is only justified because it is the logical outcome of the standard sophistic theory of politeia, which has a great deal of empirical support in the actual conduct of different Greek cities. Despite appearances to the contrary, Menn claims that Plato began from the idealized Sparta of the politeia-literature when constructing his own ideal in the Republic. As evidence for this claim, he points to the fact that timocracy (which is a typically Spartan politeia) is held to be superior to the so-called politeiai talked about in Book I; i.e., tyranny, oligarchy, and democracy. Menn concedes, however, that Plato does not do so directly in Books II-VII, since the Spartan politeia is not thematized until Book VIII, where it is sharply criticized, even though it is closest to the good and correct politeia. Menn suggests that this is because the character Socrates is not directly concerned with pointing out what is wrong with other real or imagined politeiai, while he is constructing his own best politeia. In support of this whole interpretation, Menn cites Politics Book II where Aristotle criticizes Plato for not first examining other politeiai before proposing his own ideal politeia. But this is precisely what Plato has not done in the Republic, where he first constructs an ideal polis out the raw materials of human nature, without explicit reference to other cities; and only subsequently criticizes the Spartan, Athenian, and other politeiai. Yet, according to Menn, we should still understand Platos ideal politeiai in the Republic and Laws as emerging from a process of correction of the inadequacies of the (real or imaginary) politeiai that people ordinarily admire. It is only by seeing the contrasts between Callipolis and the other politeiai in Republic Book VIII that we can understand why he has built certain features into Callipolis in the first place; e.g., the rule against guardians owning private property.

INTRODUCTION

xiii

II.
In an aptly titled paper, Wishing for Fortune, Eric Brown re-examines the vexed question of the relationship between external goods and happiness, according to Aristotle, and proposes what he considers to be a novel answer in terms of the Aristotelian distinction between wishing and choosing. Central to Browns exegetical thesis is the claim that throughout EN I Aristotle sticks to his narrow account of happiness as virtuous rational activity. In support of this claim, Brown provides a schematic map of EN I 8-12 as a whole, while arguing that Aristotle makes external goods essential for happiness only because they are necessary for virtuous rational activity. In defence of his thesis, Brown offers an account of why Aristotle thinks that external goods are necessary because people have a psychological need for certain external goods like good friends, wealth and social status. Brown interprets Aristotles qualification (i.e., in a complete life) to his definition of happiness in a chronological sense to mean that the pursuit of happiness is a lifelong project; i.e., that happiness is wholly instantiated only by a lifetime of virtuous activity. Thus Brown claims that Aristotles insistence on a complete life concerns time and not external goods. There is a separate question about what is the relationship between happiness and external goods, which Aristotle deals with later in EN I after he has confirmed his narrow definition of happiness as virtuous activity. Thus, Brown argues, we should not take Aristotle to be defining happiness in terms of both virtuous activity and external goods, since everything which he says in EN I is consistent with his narrow definition of happiness as virtuous activity, while holding both that happiness requires a complete life and that virtuous activity requires external goods. But perhaps the most original part of Browns closely reasoned paper is the section where he develops his account in terms of Aristotles distinction between wish and choice. According to this view, Aristotle allows us to choose activity, whereas we can only wish for good fortune and the external goods that it brings. Furthermore, Brown argues that there is an intrinsic link between virtue and good fortune because virtue is partially constituted by the correct appreciation of value, and because our capacity to choose virtuously is diminished when we fail to obtain what we wish for. In this way, Brown claims that Aristotle believes that external goods are necessary for virtuous activity and, thereby, for happiness. The attractions of such a conclusion are, on the one hand, that it preserves the traditional distinction between the Aristotelian and Stoic views on happiness while, on the other hand, establishing something more than a merely contingent relationship between human happiness and the external

xiv

JOHN J. CLEARY

goods of fortune. However, Browns central argument seems to depend on attributing a psychological thesis to Aristotle for which there is little textual evidence; namely, that our failure to get such wished for goods as depend on fortune significantly diminishes our capacity to choose virtuously. While agreeing with the general thrust of Browns paper, in his commentary Gary Gurtler warns against the danger of identifying Aristotles notion of blessedness too closely with common opinions about the good life and what one might wish for. He points out that Aristotles specification of a complete life is not durational but formal; i.e., it is not a matter of how long happiness lasts but rather whether the appropriate kind of life can belong to a human being. Gurtler argues that one usage of complete is related to Aristotles technical distinction between activity as complete and motion as incomplete, which leads him to define happiness as an activity rather than as a habit or disposition. The other usage relies on the common meaning of teleios as indicating what is full-grown or mature, so that it makes sense for Aristotle to deny that children, for instance, can be called happy. In support of his own analysis, Gurtler draws on EN X where Aristotle considers the question of whether happiness, defined in terms of contemplation, is possible for human beings, given that intellect is divine rather than human. Significantly, according to Gurtler, Aristotles response to this question does not focus on duration so much as on the fact that intellect and its activity is the defining part of a human being, such that it can be regarded as ones true self. This is related to the second sense of completeness, according to which human life is characterized as an integrated whole. In such a life, contemplation is not a chance event but rather an integral part of a life that has reached a perfection that may be defined as contemplative. In EN X.1-5, Aristotle shows that happiness is like pleasure; i.e., an activity that supervenes on another specific activity, thereby motivating the agent to perform that activity more frequently. Thus happiness functions as a final cause, motivating the exercise of wisdom in its activity of contemplation and making the sage more proficient at this activity.

III.
In his thoughtful paper on phantasia in Aristotle, Alfredo Ferrarin first addresses the hermeneutical problem of recovering an historically correct and textually accurate interpretation of the role of phantasia in Aristotles thought, despite the layers of interpretation and commentary that have

INTRODUCTION

xv

accumulated within the Aristotelian tradition. For instance, he warns against the danger of bringing into play our modern presuppositions about imagination as a creative and independent faculty of the self-conscious subject. By contrast, phantasia is a very narrow notion in Aristotle that is quite puzzling precisely because it is vague and incomplete. As one might expect, Ferrarin focuses on DA III.3 which he reads as work in progress rather than as Aristotles definitive account of phantasia, as many commentators tend to do. Thus, for instance, he warns against talking about phantasia as if it were an independent faculty because he considers such talk to presuppose the modern notion of an ego-subject to which cognitions and volitions can be ascribed as a kind of inner space distinct from the body. Any such assumption runs counter to Aristotles hylomorphism which regards the soul as the form of an organic body. Furthermore, by treating mental faculties as separate and distinct, we tend to emphasize the differences between perception and phantasia rather than underlining the continuity between them. While Aristotle does treat perception and thinking as independent powers of the soul, phantasia is described as a process that is dependent on sensation but is not itself an activity. Ferrarin emphasizes that this process is fundamentally reproductive, according to Aristotle; so that the resulting phantasma is at best a copy, i.e., not a standard or model but rather a derivative proxy for the perceptual object. Thus Ferrarin rejects as un-Aristotelian all those modern interpretations of phantasia which treat it as some kind of active power of interpreting (seeing-as) passive perceptions. Such readings tend to treat the sense faculties as passive recipients of atomized pieces of information that require an active power to unify, compare and clarify them. Against all such interpretations, Ferrarin insists that for Aristotle perception is an active awareness of distinct sensory content, as well as being a capacity to identify complex perceptual states of affairs; e.g., the white of Diares son. In this way Ferrarin argues that a correct understanding of phantasia in Aristotle depends on a proper assessment of its intermediate position between perception and thinking. He claims that Aristotles notion of phantasia is not definite and systematic, as we might have expected, but instead is indefinite and open-ended, depending on the different contexts in which it is used. Consequently, Ferrarin goes beyond DA III.3 to draw on Parva Naturalia for some physiological aspects of phantasia, as well as touching on the Rhetoric and Nicomachean Ethics for Aristotles discussions of deliberative phantasia and its role in human action. As a result of his analysis of DA III.3, Ferrarin claims that Aristotle understands phantasia as a form of representation of things in their absence, which results from prior perception and which may be used in

xvi

JOHN J. CLEARY

memory and thought. Phantasia is the process by which images are left over, presented, visualized, recalled, held fast as possibly true; and all these images are traces of prior perceptions. According to Aristotle, the reason why phantasia lends itself to memory and thought is that, by contrast with perception, the preserved image is not limited to the immediacy of the given. Thus a sense-derived phantasma can become a representational image or even a disconnected phantasma, depending on how we want to use it. For the purposes of thinking, we can use images as particular examples and illustrations of intelligible forms. Ferrarin suggests that this is what Aristotle means by saying that one can see the universal in the phantasma, precisely by disregarding its particularity, which makes it an image of some perceptible thing. Similarly, Aristotle claims that memory is always of images, even when we remember intelligible things. However, Ferrarin finds unduly restrictive the traditional interpretation of Aristotle as making all thinking dependent on imagination, given that he denies (DA III.5) the corporeal basis for thinking. In his fine commentary, Klaus Brinkmann notes the curious fact that DA III.3 does not specify what indispensable contribution phantasia is supposed to make towards the whole cognitive process. Rather controversially, Brinkmann claims that Aristotles primary purpose in this chapter is to show that phantasia is a faculty sui generis, which is not reducible either to perception or to any other cognitive process. However, no detailed explanation is given of how and why phantasia plays an indispensable role in believing and thinking. Brinkmann notes that in DA III.8 Aristotle does say that thoughts could not exist without phantasmata because these in turn depend on perception. Thus Brinkmann argues that we need to answer the question of what kind of functionally necessary work phantasia is supposed to do, according to Aristotle, with reference to the whole cognitive process. He briefly reviews one answer recently given by Victor Caston who claims that Aristotle introduces phantasia in order to explain how cognitive error is possible, given that any perception of its proper object is always correct. But in the case of so-called incidental perception, there is a possibility of error because I may connect the content of my perception with the wrong phantasma. Brinkmann is convinced by Castons general point that Aristotle uses phantasia to explain the existence of error, though he complains that Caston does not go far enough in explaining the role of phantasia in cognition. Brinkmann suggests that the most fundamental and indispensable function of phantasia for cognition is that it uses sensations to generate sensible forms or universals that refer to the sensible features of things. He argues that phantasmata are sensible forms, not mere

INTRODUCTION

xvii

sensations, and claims that this is supported by DA III.8 where Aristotle says that intelligible forms are among the sensible forms. This would explain why there can be no thoughts without phantasmata; i.e. without sensible forms as their prerequisite. If phantasmata are universals that would better explain why Aristotle insists that phantasmata are needed in the act of contemplation itself.

IV.
In his aptly titled paper, Enchanting the Souls, Jean-Franois Pradeau discusses the philosophical and rhetorical function of preambles attached to particular laws within Platos Laws. As part of his philolological study of the term nomos, Pradeau alerts us to the paucity of historical information available on Greek law in general. Hence when Plato gives detailed descriptions of his proposed legislation for Magnesia in the Laws, we have difficulty in ascertaining how much he adopts or adapts from Athenian law. By means of his philological excursus, Pradeau argues that the original usage of nomos is related to the activity of distributing land and the exercise of pastoral authority. Subsequently, nomos functions as a prescription for conduct within the polis, which distributes and orders appropriate shares. In this way, nomos becomes a common prescription that can be written down, while the term also designates the habitual usage of the group according to which its life is ordered. While nomos may be used in a specific sense to refer to any law, it also has the general sense of the totality of law that makes it synonymous with politeia. Thus for Greek thinkers any discourse on law always involves both judicial and institutional factors such that it becomes a discourse on civic community and its constitutional organization. From this perspective one can understand why Platos Laws links the fate of the polis with its legislation, given that law is defined as the reasoning imposed on the city; i.e., rational discourse that teaches all souls what they must appreciate in their own interest. Pradeau argues convincingly that such an understanding of law is not uniquely Platonic but rather encapsulates the typical features of law as viewed by most classical thinkers; i.e., law as the common prescriptive discourse that is imposed on all citizens as governing their conduct in the interest of the city. Similarly, the educative conception of law is also ancient and common to classical Greek thinkers who saw the law as an instrument for education, just as much as for prohibition and surveillance. It was commonly conceived as the discourse which teaches the citizen the way to virtue through obedience to the law. Plato expands and deepens this conception of law through his analysis of rational

xviii

JOHN J. CLEARY

discourse and his inquiry into the receptive soul of the citizen, who instead of being coerced should be persuaded to obey by teaching him that obedience is in his own interest. According to Pradeau, this is one of the principal functions of the elaborate preambles that we find in the Laws. According to Plato, the law is a discourse prescribed by the legislator to the citizen who must act in conformity with it. Along with persuasive discourse, the law also prescribes penalties for not conforming to law and these must be imposed by the police or guardians. Pradeau argues, however, that Plato envisages the law as prescribing for the soul of the citizen the sort of behavior which it can actualize only when the rational faculty is properly governing the spirited and appetitive faculties. Consequently, in his paper he explores how the laws are supposed to act on souls so as to inculcate the virtues that are required in good citizens. This is the function of the legislative discourse addressed to all the citizens, which has both an edifying and pedagogical purpose in telling them what good and bad conduct is and how to distinguish them. Since this form of pedagogy must be based on persuasive rhetoric, Plato invented the preamble to the laws, or preliminary discourse that is addressed to the rational faculty of the citizen so as to persuade him by exhortation and threats to act in accordance with the law. Despite his radical critique of contemporary political rhetoric, therefore, Plato regarded a certain kind of civic rhetoric as being indispensable for persuading the citizens of Magnesia to obey the law. Thus he compares the preambles to musical preludes that charm the soul, just like an incantation that makes it receptive to the law and leads to obedience becoming internalized like a second nature. Pradeau claims that the preambles are more like sermons than purely rational discourses, and that they are akin to parental admonitions about good and bad behavior that appeal to religious authority. As illustrated by the long preamble to the law on sexuality, its purpose is to persuade the citizen through a mixture of admonition and threats to adhere to the law by suppressing through fear the irrational desires. For individuals who cannot rationally control their own desires, the law takes the place of reason; so that the preambles are intended by Plato to accomplish rational constraint and to further the moral education of the citizens. In his extensive commentary, Gavin Colvert draws our attention to the dramatic structure of Platos Laws as a dialogue between three elderly Greeks from different cities who are serving as legislators for a fictional colony called Magnesia. One ostensible reason for paying attention to its dialogical structure is that it may reveal, at different levels, Platos purpose in writing as he does. For instance, only the Athenian Stranger seems to be

INTRODUCTION

xix

aware of the deeper significance of the clear difference between the education to be given to most citizens and the philosophical training offered to members of the so-called Nocturnal Council. Drawing on such examples in the Laws, Colvert describes the complex relationship between the philosopher and the lawgivers in Magnesia. Thus the fictional audience within the Laws are the elderly non-philosophers who do not always understand the philosophical significance of the legislative program outlined by the Athenian Stranger. From this disparity Colvert infers that Platos intended audience are younger people (within the Academy) who are more suited to philosophical inquiry and with whose political education he is mainly concerned. By adopting this Straussian approach, Colverts purpose is to show that the dramatic context of the Laws sometimes lends support to Pradeaus conclusions but sometimes calls them into question. For instance, against Pradeaus thesis about the continuity between the Republic and Laws, Colvert draws attention to the different fictional audiences addressed in both dialogues. The characters in the Laws, unlike the Republic, are virtual strangers and two of them have little interest in philosophy. Thus Colvert claims that the attentive reader of the Laws (who has also read Leo Strauss) will sense that Platos target audience are people of the same nature as the philosophical Athenian Stranger, who speaks simultaneously to two different audiences. In this way, Colvert suggests, Pradeaus continuity thesis might be indirectly supported by a dramatic analysis which can account for the different presentations in the Republic and the Laws. However, Colvert does not think that such an analysis provides direct support for Pradeaus claim that Platos characters use the term nomos in the same way in both dialogues, so that there is continuity in his treatment of nomos. On the contrary, Colvert argues that considerations of dramatic context and audience suggest instead that Plato deployed his discourse about the law differently under varied circumstances.

V.
In a highly literate contribution on Aristotle and the metaphysics of metaphor, Fran ORourke transports us from the poetic to the metaphysical level with a potent cocktail of literary illustrations and philosophical argumentation. His explicit intention is to examine some of the presuppositions of Aristotles theory of metaphor, while connecting these assumptions with other aspects of his philosophy. ORourke argues convincingly that the key to this theory is Aristotles understanding of metaphor as analogy. According to this argument, analogy is the essence

xx

JOHN J. CLEARY

of metaphor because it relies on the diversity and unity both of human knowledge and of human nature, and on the diversity and interconnection of beings within the cosmos. Thus metaphor can be regarded as a kind of token for the analogous unity-in-diversity of the cosmos, and also as an indication of mans psychosomatic unity. In the Poetics, Aristotle defines metaphor as the transfer to one thing of a term belonging properly to another, according to analogy or proportion, thereby expressing a similarity of relations. Furthermore, in the Rhetoric, he explains that metaphor through proportional analogy is the most highly prized; e.g., the poetic parallel between the shield of Ares and the cup of Dionysus. The key to proportional metaphor is the perception of a novel resemblance between two pairs of coordinates which are not normally conjoined. Such perception is characteristic of the poetic ability which cannot be taught, and so must be regarded as a gift of the Muses. ORourke distinguishes between metaphors that have a limited cultural value, and those which are universal in scope and which indicate something essential to human nature. For instance, the vocabulary used in many different languages for the activity of knowing and perceiving usually involves physical metaphors like seizing and grasping. Here the basis for the transfer to mental acts of the names of physical activity must be some analogous similarity perceived between them. But any adequate explanation of metaphoric signification must also account for the unity underlying the two domains from which these expressions are drawn, and Aristotle provides such a foundation through his account of the ultimate unity and complementary distinctness of body and soul. ORourke claims that the unity of analogy (which is basic for metaphor) has broad implications for Aristotles physical and metaphysical thought. According to its root meaning, analogy is the similarity of an intrinsic proportion that is realized across many different relationships. Thus analogical unity transcends the unity of the individual, species, and genus; so that Aristotles metaphysics as a science of being qua being can discover those causes and principles that are common to all beings. ORourke argues that this is the basis for Aristotles vision of a unified cosmos which is rendered intelligible by means of analogous principles, such as act and potency, matter, form and privation. He concludes that it is this fundamental ontological relatedness among beings which provides Aristotle with the profound basis for metaphor in every sphere of being. In her useful commentary, Ioanna Patsioti-Tsacpounidi draws our attention to the apparently conflicting statements which Aristotle makes about metaphor in different texts. For instance, within the logical texts of his Organon, he seems to characterize metaphor negatively as being obscure, whereas he describes it as clear and effective for persuasion in the

INTRODUCTION

xxi

Poetics and Rhetoric. She suggests that ORourke does not provide a satisfactory account of such conflicting claims about metaphor. She also suggests that he might have paid more attention to Aristotles definition of metaphor as the transposition of an alien name to another location, which is possible because of the coordination of relationship within language itself. Her main complaint, however, seems to be that ORourke does not draw on Ricouers semantic analysis of metaphor in Aristotle, as she herself does. Patsioti-Tsacpounidi accepts ORourkes general thesis that metaphor is closely linked to Aristotles use of analogy in metaphysics and biology, but she suggests that one might also usefully examine this link within the ethical treatises. For instance, in his discussion of different types of friendship in terms of focal meaning, Aristotle uses the example of medical as an illustration of analogy. She also suggests that our understanding of these analogical relationships might be improved through an analysis in terms of likeness and unlikeness, since Aristotle uses the notion of metaphor to capture possible relations between similar and dissimilar things. Given that metaphor involves the recognition of similarity in dissimilar things through the process of perception, she accepts ORourkes thesis that it is used by Aristotle to connect the physical and psychic parts of human existence. In addition to metaphysical analysis, however, she advocates further inquiry into the epistemological significance of metaphor in Aristotle.

VI.
In his provocatively modernizing paper, Anthony Price begins with a long quotation from Newmans Grammar of Assent which he regards as being a faithful account of Aristotles conception of phronsis. Price undertakes to defend Newmans reading of Aristotle by clarifying its implications with respect to Jonathan Dancys contemporary view which he labels particularism as opposed to generalism. Thus the structure of his paper is determined by four guiding questions. (1) Does Aristotle suppose that an agents practical decisions apply some general specification of eudaimonia to a particular situation? Price gives a negative answer to this question, thereby providing support for a particularist reading of Aristotle. (2) Does Aristotle suppose that there are any principles applied without exception that guide decisions? Price gives a qualified positive answer by trying to determine how large a role is given by Aristotle to such principles, since for him all moral principles admit of exceptions. (3) Does Aristotle suppose that there are relevant factors

xxii

JOHN J. CLEARY

whose valence is invariable between different contexts? Again Prices answer is positive, though he points out that a moral agent faces situations as they arise, relying on character to determine specific judgments. Price faults both generalism and extreme particularism in ethics for failing to give an adequate role to character in moral decision-making. (4) Does Aristotle suppose that a moral agent brings nothing to a new situation except a contentless ability to discern what matters where it matters (Dancy)? Prices answer to this question is negative, thereby undermining an extreme particularist reading of Aristotle. With respect to the first question, Price reviews the evidence for and against attributing to Aristotle a Grand End view of ethical deliberation such that the rationality of an action would be shown by its derivation from an unchanging blueprint which is applied to a particular situation and its changing circumstances. After considering the textual evidence from both the Eudemian and Nicomachean Ethics, Price concludes that there is no basis for attributing this view to Aristotle. Even the most promising passage in the Eudemian Ethics has nothing to say about a blueprint for living or even a decision-procedure for action. The moral agent who opts for one kind of eudaimonia is not endorsing any Grand End that is sufficient to direct all his deliberation but rather is adopting a general moral orientation that allows for a wide spectrum of specific ends for action. Indeed, Price questions whether Aristotle regarded any practical principles as absolute in the sense of applying to all actions without exception. After considering the evidence, Price concludes that not only first-order but also second-order rules of practical action hold only for the most part, according to Aristotle. Yet, one might object, Aristotle does seem to regard as absolute some universal prohibitions on murder, adultery, and theft, which are concrete and hold without exception. The espousal of such universal moral principles would appear to rule out any attempt to read Aristotle as a particularist like Dancy. Such an objection leads Price to consider variabilism as one central aspect of Dancys particularism, which implies that features of an action may count for or against it in different contexts; e.g., the pleasure taken in an action may give it a different valence depending on the character of the action; e.g., taking vicious pleasure in bad actions. Aristotle does hold that there are good and bad pleasures linked to action but there is no unambiguous evidence for variable valence with regard to sub-types of action, as the particularist holds. An extreme particularist may deny that any general principles either do or should play any role in generating or explaining a decision to act in some determinate way in a certain situation. Thus his ideal is a person on

INTRODUCTION

xxiii

whom we can rely to make sound moral judgments; i.e. someone who gets it right case by case. Presumably such a person is the product of moral education. But does this correspond to Aristotles view? Price thinks there is no simple way of classifying Aristotle as a particularist, though he does avoid the errors of an excessive generalism that pays no attention to particular circumstances. But, according to Price, Aristotles conception of practical truth also helps him to avoid the errors of extreme particularism that pays scant attention to the humanity of moral agents. In her very long commentary, which begins to look like a counter-paper, Bridget Clarke suggests that John McDowell might have been better than Jonathan Dancy as a modern foil for Price when he asks the question of whether or not Aristotle is a particularist. By focusing on McDowells reading of Aristotle, Clarke tries to show that the particularist literature on the ethical treatises both anticipates Prices criticism of Dancy, and also provides a more convincing account of practical reason. She claims that Price, in trying to avoid attributing to Aristotle a modern top-down model of practical reason, is too hasty in giving up the idea that a practically wise person deliberates from some general conception of eudaimonia. Instead, she argues, one must keep in mind the central place that Aristotle gives to achieving an understanding of the human good within human life. Thus, by explicitly adopting McDowells approach, Clarke suggests that we can make sense of the idea that a correct concept of eudaimonia serves as the end of deliberation, without assuming that some Grand End or comprehensive blueprint is involved. In this way, practical wisdom can be seen as incorporating a synoptic understanding of what is important in human life. On this basis, she criticizes Price for adopting an overly narrow view of phronsis in Aristotle, which fails to take account of his richer notion of deliberation that goes beyond means-end calculation. Once again drawing on McDowells more expansive notion of deliberation, she attributes to Aristotle a moderate particularism that has none of the deficiencies that Price finds in Dancys extreme particularism. Central to that moderate particularism is the idea that some end may figure in deliberation, although it is impossible to formulate it definitively before it is actually applied to a particular situation. Thus, instead of a Grand End or blueprint, the phronimos has a synoptic vision of eudaimonia that helps him to make specific choices which will realize that end. Therefore, as McDowell puts it, we can speak of an agent deliberating towards a conception of eudaimonia, even when its content can only be specified through particular applications of it.

xxiv

JOHN J. CLEARY

VII.
In his comprehensive and detailed paper, Thomas Johansen provides a close analysis of a key passage in Aristotles De Anima, which he interprets as supporting a version of inner sense theory; i.e., that we are conscious of our perceptions by means of further perceptions that take those first perceptions as their objects. As an integral part of his interpretation, Johansen takes into account many of the standard criticisms of inner sense theory, while also reviewing some recent secondary literature on this Aristotelian topic. He begins with a general outline of Aristotles discussion of the faculties of perception in De Anima, so as to provide the specific context for the passage at DA III.2 which he proposes to analyse in detail. Johansen accepts that Aristotle defines the faculties of sense perception in terms of their activities, but insists that not all activities of the soul serve to define distinct faculties. For instance, at DA III.1-3, Aristotle considers a range of perceptual or quasi-perceptual faculties in order to show that these are adequately explained with reference to the activities of the five sense faculties. Johansen claims that here Aristotle is guided by an axiom of explanatory economy; namely, that faculties should not be multiplied unnecessarily. This axiom becomes important later in Johansens paper when he addresses the controversial question of whether the so-called common sense is a distinct faculty for Aristotle. But first Johansen provides a translation of the key passage at DA III.2, along with a general analysis of its argumentative structure, before offering his own comprehensive interpretation. In fact, this is the core of Johansens paper where he takes account of some of the major controversies surrounding the interpretation of the passage, while also reviewing the views of contemporary scholars like Kosman and Caston, who were also involved in this colloquium, either directly or indirectly. As a result of his thorough analysis, Johansen claims that Aristotles view is that perceiving that one sees may be treated as a function of the faculty of sight itself, and not of any different sense faculty. This is consistent with his general view that the account of the five senses, given in DA II, is adequate for explaining the phenomenon of perceiving that we see or hear or smell. But, Johansen argues, the passage at DA III.2 leaves us with a question as to how sight can be responsible for second-order perception, given that its content is more complex than simply color. Thus, he claims, the passage needs to be supplemented with others, most notably from De Somno, which discuss the way that we perceive contents that transcend special perception. Johansen argues, however, that Aristotle does not provide an answer in terms of some further senses, like the so-called

INTRODUCTION

xxv

common sense, but rather in terms of how the five senses work together as a comprehensive faculty of perception. Thus, in the latter part of his paper, Johansen develops and defends the implications of his interpretation by considering the following questions: (1) In what way is Aristotle committed to the claim that we always perceive that we see, when we see? (2) Is Johansens interpretation of DA III.2 consistent with what Aristotle says elsewhere about perceiving that we see? (3) Does Aristotle offer a general account of perceptual consciousness in DA III.2 and, if so, what sort of account does he give? Obviously, it is impossible for me to outline Johansens extensive answers to these questions, so I will focus briefly on the final question about perceptual consciousness. In light of the distinction between first-order perceptual awareness of objects in the world and second-order awareness of our own perceptions, Johansen does not find Aristotle offering any general account of perceptual consciousness by way of explaining how first-order perception is consciousness of objects in the world. However, Johansen accepts that Aristotle does give an account of perceptual selfconsciousness when he specifies by what faculty we can gain information about our perceptual activities. But, in general, Johansens conclusion is that Aristotles account of perceiving that we perceive does not amount to a general account of phenomenal consciousness as such. In his model commentary, Aryeh Kosman raises a few perceptive objections, some of which have already been taken on board in Johansens revised paper, but he also raises some questions (which remain open) about the interpretation of Aristotle. Kosman rehearses some of the differences between himself and Johansen in their understanding of what Aristotle means when he talks about perceiving that we perceive. Despite the explicit parallels drawn with modern inner sense theories, Kosman suspects that Johansens discussion remains more concerned with selfconsciousness. Kosman also raises some questions about what drives the infinite regress argument that Aristotle uses to establish the negative conclusion that no other faculty is needed beyond the sense faculty for perceiving that we perceive. Drawing on Aquinass general analysis of the regress argument for efficient causes, Kosman claims that, without the requirement that perception be perceived, the ad infinitum argument would make no sense as an objection to Aristotles supposition that the agency for such perception lies elsewhere. In this way he concludes that Aristotle is concerned with the explanation of perceptual consciousness for which iteration would be required. Thus Kosman emphasizes that Aristotles concern is with the explanation of first-order consciousness and not with the reflexive self-awareness that we call self-consciousness. Finally, he

xxvi

JOHN J. CLEARY

suggests that Johansen, by contrast, seems to be concerned with secondorder consciousness, which may not be appropriate to Aristotles concerns in the De Anima passage under discussion. According to the established practice of our Proceedings, each colloquium in this volume (with one unavoidable exception) is dialogical both in structure and content, which underlines our original goal of providing a forum for conversation between different traditions of interpreting ancient philosophical texts. Once again, the dominant analytic tradition is well represented in at least three of the colloquia, though the commentators often provide an alternative perspective on the same issues. The final colloquium on Aristotles theory of inner sense may be regarded as determining the status quaestionis on this topic, since it involves some prominent contributors to the recent debate. The first colloquium, however, shows how the analytic approach to ancient texts can be enriched by bringing literary and historical approaches to bear on the same material. The third and fourth colloquia illustrate how a keen historical sensitivity can prevent us from making anachronistic assumptions about ancient texts and philosophical theories. But, although they are very different from each other in theme and style, the fifth and sixth colloquia show how modern philosophical preoccupations can be usefully employed in reaching a new perspective on some central issues in Aristotles thought.
BOSTON COLLEGE & NUI MAYNOOTH (IRELAND)

COLLOQUIUM 1

ON PLATOS
STEPHEN MENN 1

I. I want to present here some interim results from an ongoing project of reading Platos Republic, and also Platos Laws and Aristotles Politics, in the light of the Greek tradition of writing on the . This was a well-established kind of writing in Platos day, and a kind quite different from the Socratic that Plato had mostly been writing, and we can ask why Plato in the Republic chose to take up this kind of writing (and to make the character Socrates take up this kind of talking, very different from his usual questioning style). Undoubtedly, part of Platos reason was to show that he could write this kind of text better than the people who usually do it, just as, in writing the Timaeus, part of his reason was to show that he could write an On Nature [ ] better and significantly different from the usual pre-Socratic accounts. Plato would not have written a that would be just one more instance of the usual kind, so in trying to understand his work we will want to understand not merely how it is like, but also how it is unlike, typical earlier , so far as they are preserved or can be reconstructed. I hope to present here, if in sketchy form, enough of the results of this kind of investigation to show that it can bring new illumination to Platos text. In particular, it can give new perspectives on some perennial problems, both about the interpretation of Platos views in the Republic, and about the relations between different parts and emphases of the text. By problems about the different parts of the text I mean, for instance, what is the relation between the elenctic Socrates of Book I and the more positive Socrates of Books II-X, and what is the relation between the Republics discussion of moral virtue and its discussion of politics (is one a means to the other? is the text sim-

_________
1 I am grateful for comments on various stages of this paper to Tad Brennan, Eric Brown, Myles Burnyeat, Paul Cartledge, John Cooper, Rachana Kamtekar, Nelly Lahoud, Josh Ober and Malcolm Schofield (the comments of Brennan, Cartledge and Kamtekar were very detailed and helpful), as well as to my BACAP commentator Sara Monoson and an anonymous BACAP referee, and to audiences at Brown (BACAP) and at the Montreal Political Theory Workshop.

STEPHEN MENN

ply a hodgepodge of different topics?). By problems about Platos views I mean, notably, does Plato really believe that the he constructs in Books II-VII would be the best if it ever came about, does he really believe it is possible to actualize it, and does he advocate taking political action toward that end? Alsoa somewhat less discussed but also perennial problem which will be of particular interest to mewhat should we make of Platos views on Sparta, given that the Plato constructs has striking similarities to the Spartan (not necessarily to the historical reality but to the idealized Sparta as described in Xenophons of the Spartans), but given also that Plato sharply criticizes the Spartan in Republic VIII (and also in the Laws)? Should we say that Plato was essentially a Laconizer (with Popper 1945, and, with qualifications, Cartledge 1999), or should we conclude (with Tigerstedt I 24476) that despite the similarities, the differences are deep enough to show that Platos ideal proceeds from a fundamentally different and independent inspiration? Obviously both the resemblances and the differences are real, and what is interesting for us is not to weigh them up and judge whether he is more pro- or anti-Spartan, but rather to understand why he constructs his ideal the way he does, with both Spartan and anti-Spartan elements, why he finds Sparta good to think with in constructing his ideal , and what this might imply for the meaning of his ideal. First let me say something about Greek literature and why it gives relevant comparanda for reading the Republic. The title (Latin Res publica, English Republic) is attested solidly and early for Platos text. Plato himself, at Timaeus 17c1-3, refers back to what is apparently the Republic as . This is not exactly a title, but Aristotle clearly cites the Republic under the title at Politics II 1, 1261a6, II 6, 1264b28, IV 4, 1291a12, V 12, 1316a1 and VIII 7, 1342a33, and Rhetoric III 4, 1406b32. 2 Furthermore, Aristotle makes clear that he intends this, not as a proprietary title for Platos work alone, but as a generic title like (Aristotles title for the Menexenus, Rhet. III 14, 1415b31) or . For Aristotle says that, although there is a fifth or constitution beyond monarchy, oligarchy, democracy and aristocracy, and although this fifth constitution is the one most properly called , yet because this [fifth kind] does not often

_________
2 Proclus In rem publicam v.1 p.8. Kroll, citing unnamed earlier writers who had argued that the of the text was the (best) , points out that Aristotle cites the text by that name not only in the Politics but also in his Sussitikos (presumably the Sussitikoi Nomoi mentioned in Diogenes Laertius catalogue of Aristotles works), and that Theophrastus does so in his Laws and in many other places.

ON PLATOS

come about, it escapes the notice of those who try to enumerate the kinds of , and they make use of only four [kinds] in their , like Plato (Pol. IV 7, 1293a40-b1). 3 In this passage Aristotle is using the word in three different senses: once for constitution in general, once for a particular constitution which is preeminently constitutional as opposed to despotic rule, but once for a genre or kind of writing that includes Platos Republic. The written that Aristotle is thinking of would include not only texts called simply or or , but also texts called of the so-and-sos. We might think that the first type of text would be normative and the second historical; but Aristotle often does not bother to distinguish the two types, and they would have covered heavily overlapping ranges of topics. As we have seen Aristotle say, texts called simply , like Platos, would not merely describe an ideal , but would also classify all possible types of , in order to prove by exhaustion that their ideal is the best possible, and in describing the possible types they might well describe the of the so-and-sos (so the Republic describes the second-best type, the timocracy or timarchy, as the Cretan and Spartan [ ], Rep. VIII, 544c2-3). Conversely, a text called of the so-and-sos might discuss which if any of the standardly recognized types of it fell under (cp. Plato Laws IV, 712c6-e5, on the notorious problem of classifying the Spartan ). Furthermore, many texts called of the so-and-sos are also descriptions of an ideal, since they are written in order to praise the of the so-andsos and to contrast it with how other cities are governed. This was clearly Xenophons aim in his of the Spartans, and it would also have been the aim of many other texts on the of the Spartansand many there were. Thus Aristotle speaks of Thibron [who] seems to admire the legislator of the Spartansand all the others who write about their tooon the ground that they ruled over many through exer-

_________
3 It initially seems odd that Aristotle does not mention Platos class of timocracy; but Aristotles subsequent discussion makes clear that he is counting this among governments which can be called aristocracies, although they are not aristocracies in the strictest sense. In any case, the series of constitutions described in Republic VIII does not include a mean or blending of oligarchy and democracy, of the kind that Aristotle specially calls ; the Laws, by contrast, does describe such a constitution, so it is clear that Aristotle is referring specifically to the Republic. Aristotle omits tyranny here as being not properly a constitution, but the opposite of constitutional rule. It is curious that Aristotles own collection of , going by the catalogue in Diogenes Laertius, did not include the missing mixed constitutionwhich may support his judgment that this type is rarely found.

STEPHEN MENN

cising themselves for danger (Pol. VII 14, 1333b18-21; Thibron was a general in the Spartan campaigns in Anatolia after the failure of Cyrus revolt, Xenophon Hellenica III i 4-8 and IV viii 17-19). So too Critias, who wrote a of the Spartans in prose and perhaps another in verse (DK 88 B32-7, cp. A22), said that the of the Spartans is thought to be the most noble (Hell. II iii 34). Naturally such -ofthe-Spartans texts involved much idealizing away from the often brutal Spartan realities. Aristotle treats both idealizing -of-the-so-andsos and purely ideal like Platos as proposals for how a city might best be governed, and in Politics II he sets out to examine the other , both the ones practiced in certain cities that are said to be wellgoverned [ a Spartan slogan], and any others that particular people have described and thought to be right [ ] (Pol. II 1, 1260b29-32), to find out what is right in them and to show that he himself is seeking a further not arbitrarily but because the ones that now exist are not right (b34-5), drawing no distinction between that exist only in and those that are said to exist (or to have existed) in places like Sparta. In Politics IV he complains that most of those who have spoken either seek only the very highest [ ], even if it requires impracticable material conditions, or else although they speak of a more common [ ], they take away the existing , praising the Spartan or some other (Pol. IV 1, 1288b35-1289a1). Aristotle himself, having described his own Politics as an investigation (EN X 9, 1181b14-15), argues that, just as gymnastics must investigate not only what is the best regimen for an ideal body, but also what is best for the average body or for particular types of body, so the science must investigate not only the best for an ideal city, but also what is best under more common conditions, and what best preserves each given type of (so Pol. IV 1). As Aristotle states this program in EN X 9, one crucial step is to consider, out of the that have been collected, what kinds of thing [= what laws and customs] preserve and destroy cities and what kinds [preserve , and for what causes some [cities] are and destroy] each of the governed [ ] rightly or wrongly (1181b17-20); so for Aristotle too, as for the admirers of Sparta, texts on the of the so-andsos are instruments of the normative study of how a city should best be governed. The broad outline of the history of the genre is well known at least, it is well known in continental scholarship: I am not sure how far

ON PLATOS

this discussion has penetrated the Anglophone world. 4 But the genre, and the title , seem not to have been much taken into account in the literature on Platos Republic. 5 This may be in part because of a reluctance to rely on works that are lost. But not all besides Platos are lost: setting aside the complex cases of Platos Laws and Aristotles Politics (although at least Politics VII-VIII is a classic ), we have the pseudo-Xenophon of the Athenians, the genuine Xenophon of the Spartans, Aristotles of the Athenians, and Ciceros De re publica; in addition, Book VI of Polybius Histories is a of the Romans, and most of Book II of Josephus Against Apion is a of the Jews. Of course, the majority of these texts are later than Platos, and influenced by Plato to some degree, and I have heard it suggested that if there was afterward a genre, it was Plato who invented it (in this vein one hears it said that Zeno of Citiums , just because of its title, must have been a reference specifically to Plato; against this inference see Schofield 1999). But in fact the genre is both older than Plato, and also highly ramified, with nothing to suggest that Plato is the source from which all the branches after his time are spreading out. Aristotle describes Hippodamus of Miletus as the first of those who did not themselves engage in politics [ ] to undertake to say something (Pol. II 8, 1267b29-30; Hippodamus was involved in the founding of Thurii in 444/3 BC); Hippodamus and Phaleas of Chalcedon, whom Aristotle discusses together with

_________
4 A standard German handbook account is Treu 1966, 1935-47; a distinct French approach, focussing more on uses of the word than on the genre, and with much less interest in lost works, is Bordes 1982, building on de Romilly 1959. None of these studies show much interest in philosophy. One Anglophone account, Dawson 1992, seems to have had no impact on the literature, apart from Schofield 1999 (originally a review-essay on Dawson). The only other recent Anglophone scholars of Greek political philosophy I know whose work makes use of the genre are Schofield, Paul Cartledge (see Cartledge forthcoming), G.R.F. Ferrari (see Griffith and Ferrari 2000), and Josiah Ober (see Ober 1998). There is a very quick sketch of the genre at Jacoby 1949, 211-15; even briefer by Connor 1989, 49-51 and by Gera 1993, 11-13. A volume of the continuation of the Fragmente der griechischen Historiker (Jacoby 1923-, abbreviated FGrHist) has been projected to include fragments of . The majority of the fragmentary Spartan are already in FGrHist vol. 3B, ##580-98. 5 Standard surveys such as Annas 1981, Cross and Woozley 1964, Murphy 1951, Reeve 1988, White 1979, and Hffe 1997, have no discussion of the genre, and no index entry for Sparta; the honorable exception is G.R.F. Ferraris introduction to Griffith and Ferrari 2000. Leroux 2002, 42-54 discusses the concept of and the question of Platos political or utopian program in his historical context, but does not focus on the genre or on Sparta.

STEPHEN MENN

him, seem to have written their without any pretence that they had been in practice in Sparta or elsewhere. There was also a by Protagoras, and a forged in the name of Epicharmus (two extant fragments, DK 23 B56-7), according to Aristoxenus by Chrysogonus the flute-player (datable by his involvement in Alcibiades ceremonial return to Athens in 407, Athenaeus XII 49 and XIV 59). Also the extant pseudo-Xenophon (or Old Oligarch) of the Athenians (generally dated to somewhere between the 440s and the 420s), 6 the extant genuine Xenophon of the Spartans (dated by its latest editor of the Spartans and the Thessalians by to the 390s), 7 the lost Critias, and of the Spartans by three Spartans, Thibron, Lysander, 8 and Pausanias, 9 are all pre-Platonic. After Plato we hear of works called or by Diogenes the Cynic, Xenocrates, and the Stoics Zeno and Chrysippus, a

_________
6 See the edition, translation, and brief but acute introduction in Bowersock 1968. For a sense of the apparently limitless range of views that have been defended about the date and purpose of this text, see de Ste. Croix 1972, 307-10, Mattingly 1997, and Hornblower 2000. 7 I will use Lipka 2002, an excellent scholarly instrument, as the edition of reference; see however also the less ambitious Rebenich 1998. (I abbreviate the title as Resp. Lac.). The date of the treatise has been the subject of a fair amount of controversy, and there is something to be said for Rebenichs dating to the time of the Theban revolution in the early 370s, but I accept Lipkas arguments (9-13) for a dating in the mid-390s: the evidence turns on chapter XIV, which assumes a time when the Spartans have lost their allies and their hegemony, but when they still have harmosts (military governors) in many cities. More generally, Xenophons attitude in the treatise is not simply pro-Spartan, but is taking sides in an internal debate at Sparta, on the side of his patron Agesilaus and against Lysander, whose partisans are the target of the polemic in chapter XIV (on Agesilaus and Lysander and Pausanias, see a note below). If we accept Rebenichs dating, Lysander would be a nonissue and Xenophon would have to be criticizing his patron Agesilaus, who condoned the seizure of the Theban citadel and failed to punish Sphodrias for his raid on the Piraeus. In any case the once popular dating of chapter XIV after Leuctra seems clearly impossible (harmosts? threat of Sparta regaining hegemony?). It also seems clear that chapter XIV is not, as was once commonly assumed, an afterthought reflecting Xenophons later disappointment with Sparta: the parallel with the end of the Cyropaedia, where everything has degenerated in Persia since Cyrus time, as everything has degenerated in Sparta since they stopped following the laws of Lycurgus, cannot be a coincidence. On the comparison between the end of the Cyropaedia and Resp. Lac. XIV, see Dorion 2002. 8 , found in Lysanders house after his death, according to Ephorus in Plutarch Lysander 30; written for him by Cleon of Halicarnassus according to Lysander 25 and Agesilaus 20. For full references and discussion see FGrHist #583. 9 FGrHist #582. Warning: much about Pausanias is controversial, including the text of the crucial passage from Ephorus, FGrHist #582 T3, from Strabo VIII v 5. Besides Jacoby, see references in a note below.

ON PLATOS

by Antisthenes, a by Demetrius of Phalerum, a by Theophrastus (the same title might be applied to Aristotles Politics VII-VIII, which begin with [almost] these words), and, in the imperial period, a by Tiberius teacher Theodorus of Gadara and a by the third century AD Cynic Oenomaus, also of Gadara. 10 And besides Aristotles 158 of individual cities (including of course Sparta) there were further of the Spartans by the Peripatetic Dicaearchus, by the Stoics Persaeus and Sphaerus, and by persons named Aristocles, Dioscurides, Hippasus, Molpis, Nicocles and Proxenus (and by Aristocrates and Polycrates). 11 Much of this literature may lie under Plutarchs Life of Lycurgus (which cites the works of Aristotle, Sphaerus, Dioscurides and Aristocrates); as far as we know, none of these of the Spartans, except Aristotles and possibly Pausanias, said anything negative about Sparta, except to criticize what were alleged to be post-Lycurgan deviations (Dicaearchus was so laudatory that the Spartans are said to have mandated yearly readings of it, Fr. 2 in Mirhady 2001). 12 There were

_________
10 References to all except Chrysippus and Theodorus and Oenomaus are in Diogenes Laertius in the lives of the respective authors (Diogenes VI 80, Xenocrates IV 12, Zeno VII 4, Antisthenes VI 16, Demetrius V 81, Theophrastus V 45). Theodorus and Oenomaus are in the Suda under and respectively. References to the Zeno and Chrysippus works are collected in von Arnims Stoicorum Veterum Fragmenta. 11 For all of these see FGrHist ##580-98. The title of most of the lost Spartan , as well as of the extant Xenophon, seems to have been or or variants on these, not , but I will translate as of the Spartans in all cases. Lacedaimonians or Laconians might in principle include the perioeci or other disenfranchised groups as well as the Spartiate full citizens. But these groups precisely did not participate in the , and the texts on the Spartan seem to have said little or nothing about them, concentrating on the collective mode of life of the Spartiate full citizens (especially men but also women) who are trained and lived their lives . Where necessary to disambiguate, I will say Spartiate to make it clear that I am talking only about the full citizens. 12 But see the reference in Josephus Against Apion I 221 to Polycrates Tripolitikos, which contained an attack on the Spartans (according to Jacoby FGrHist #588 and #597, this is not the same Polycrates who wrote the mentioned above). There are of course also criticisms in extant works of Plato, Aristotle and Isocrates, none of these works being primarily devoted to Sparta. Note that even Aristotle, who makes harshly critical comments on Sparta and who remains the source of our most damning information about it, also said that Lycurgus, who was honored as a god by the Spartans, was honored less than he deserved (Aristotle Fr. 534 Rose, cited by Plutarch Lycurgus 31, perhaps from Aristotles of the Spartans). The Pausanias text is unfortunately a scholarly hornets nest (on which see now van Wees 1999 and references therein, notably David 1979; Rebenich 1998, 23n87, lists which scholars line up on which side). It seems to be agreed that

STEPHEN MENN

also of the Athenians, Corinthians and Pellenians by Dicaearchus (so Cicero ad Atticum II ii) and a by Demetrius of Phalerum (DL V 80). No one will seriously maintain that Platos Republic was the generic model for all this literature. In some ways, what is striking is how little impact Plato had. Despite his (and Aristotles) criticisms of Sparta and of all other existing , it remains true after Plato as before him that the most common way to present the best was to describe, in an idealizing way, the allegedly practiced in some actual city, usually Sparta. Ciceros De re publica explicitly sets out to describe the best constitution (I xx 33), but then does so only by asserting that the best constitution is the constitution of the (preGracchan) Romans and describing that historical constitution (II i 2-3). In substituting the of the Romans for the of the Spartans as the best, Cicero is following Polybius; Josephus is emboldened to argue instead that the of the Jews (at some indefinite past time) was the best . But the Spartan presumption remains strong enough that all these authors need to describe and contrast the of the Spartans in order to prove that not the Spartan but their preferred is the best. 13 At this point it will be objected that the fact that all these texts are referred to by later sources as is not enough to show that the authors themselves, particularly in the fifth century before it had become standard for authors to give titles for prose works, thought of these texts as belonging to a determinate -kind of writing. And it is indeed true

_________
either Pausanias criticized the ephorate as a post-Lycurgan innovation, or he criticized Lycurgus for establishing the ephorate over and above the constitutional structure imposed on Sparta by the Delphic oracle. Even if the latter is true, Pausanias is still conforming to the basic pattern of criticizing later decay from an originally ideal . 13 Josephus of the Jews is Against Apion II 145-296 (Josephus coins the word theocracy to distinguish the Jewish from the standard Greek forms at II 165). Polybius examines the Spartan , to show the superiority of the Roman, at VI xlviiil, as does Cicero in the fragmentary De re publica II xxiii; Josephus discusses Sparta from a similar motive at Against Apion II 225-31, with further Spartan comparisons scattered elsewhere. Polybius also discusses Crete (VI xlv 1-xlvii 6), commonly linked with Sparta (Polybius argues that it is much worse), and also Carthage (VI li-lii), as well as explaining why Athens and Thebes are not contenders (VI xliii-xliv; the list is very close to Aristotles in Politics II). Polybius also explains why Platos , since it has never been actualized, has no right to enter the competition (VI xlvii 7-10); Josephus Against Apion II 220-24 says that while Platos and are generally regarded as unattainable by human nature, those which the Jews have actually practiced are more demanding. Cicero contrasts his procedure, describing the real Roman res publica, with Platos, making up his own perhaps impossible res publica, at II i 3 and II xi 21-2.

ON PLATOS

that the references to Hippodamus, Phaleas, Thibron, Lysander, and Pausanias are not really to titles, and that any title we did have for these texts might well be non-authorial. But we do not have to worry too much about titles of lost works. Our earliest extant , the pseudoXenophon, begins with the words ; this is not exactly a title, but it is as close to a title as we can expect for a fifthcentury text, and it sets the theme of the pamphlet, which consistently argues that while the Athenians is bad (being rule by the worst people), all of their laws and customs are good, in the sense of being well calculated to preserve that bad . While the pseudoXenophon has a reputation as crude and un-intellectual, this in fact fits very closely with everything else we know about early theorizing about the , and the text is a good witness to what -writing looked like around the time of Platos birth. The only anomaly is that, while most of which we know were works of rhetorical praise, this one is equally rhetorical blame. But a work on the worst seems to presuppose, and be a deliberate twist on, works on the best . And the pseudo-Xenophon is also, in its way, a speech of paradoxical praise, like the genuine Xenophon of the Spartans: both take up the hypothesis the customs of the so-and-sos are opposite to those of the rest of the Greeks, doing what would be considered bizarre or shameful elsewhere, but I will show that each custom, even the most bizarre, is correct, being justified by the purpose that they all serve. What is important for our purposes is that the pseudo-Xenophon assumes the theoretical distinction between the , classifiable by ruling group, and the particular customs designed to preserve that . We first meet such a classification, into monarchy and oligarchy and the rule of the demos (though without the generic term ), with characterizations of each type and arguments about which is the best and about how stable they can be, in the constitutional debate which Herodotus puts in the mouths of the Persian conspirators in Histories III 80-82. Similar classifications, and similar arguments on one side or another, can be found in other texts, notably Isocrates Nicocles 12-26, which explicitly sets out to show that monarchy (or even tyranny) is the best of (12; the other options being oligarchy and democracy, 15 and again 18), and texts in his Areopagiticus and especially Panathenaicus praising the ancestral at Athens. 14

_________
14 Xenophon also assumes the threefold classification of , in the first sentence of the Cyropaedia and in Alcibiades conversation with Pericles about laws, Mem. I ii 4046.

10

STEPHEN MENN

There is a close analogy between and , and indeed it seems to have been a commonplace that the is the of the city (Aristotle Pol. IV 11, 1295a40-b1, cp. Isocrates Areopagiticus 14, Panathenaicus 138). A of the so-and-sos will give enough details of their actions to allow us to see their , and a of so-and-so will give enough details of his actions to allow us to see his ; both contrast with history, and both, owing to a shared origin in rhetorical praise and blame, feel freer to take liberties with the truth of the details than history does. are not simply general (abstracting from the details) but classificatory and evaluative. A work might be a collection of lives of famous people, but it is as likely to be a classification of the different and an argument about which is preferable, and the same is true for a ; indeed, it is often assumed that the questions of the best and of the best are linked. 15 A discussion was not necessarily a book: it might be merely part of a book (as in Herodotus), it might be spoken rather than written, or it might be written only as an aid to a political speech (as, probably, Lysanders text); and in speaking of the discourse as a background to Plato and Aristotle, I do not mean to restrict myself to whole books. But certainly whole books, called or or or of the so-and-sos, were common enough, and it is fair to describe them, using the standard phrase, as pamphlet literature: that is, as texts describing, in more accurate or more fictionalized form, some possible mode of collective life and governance,

_________
15 The issues about ancient biography and its relations with other forms of writing (history and antiquarian writing) are complex and have been much discussed. A point of entry is Momigliano 1993; the locus classicus for the contrast between biography and history is Plutarch Alexander 1. Xenophons assumptions about the aims of -writing come through at Agesilaus I 6: I think that from his deeds his too will best be shown. It has often been observed (notably in Cartledge 1987) that Xenophon takes more liberties with the truth in the Agesilaus than in the Hellenica (itself no paradigm of historical accuracy), even when (as often) the accounts are parallel. The assimilation between and is perhaps already in Pericles funeral oration in Thucydides, which speaks conjunctively of the Athenians and and (II xxxvi 4), thus their collective way of life (posited as the inner cause of their external successes, like the Spartans at Xenophon Resp. Lac. I 1). Plato in Rep. VIII-IX, discussed in section IV below, assumes a correspondence between and (see VIII, 544d6-e2); Aristotle Pol. VII 1-3 argues that answers to the questions of the best for an individual and for a city go together. Polybius says that he will apply biographical (not historical) methods in his of the Romans (VI ii 5-6); Dicaearchus wrote a of Greece.

ON PLATOS

11

proposed for emulation (or, in the pseudo-Xenophons case, avoidance) to a citizen body that must decide how to govern itself now. 16 By far the favorite form were texts praising the of the Spartans, and even texts with some other aim would be influenced by the standard praise of Sparta. (Just how standard was the praise of the Spartan can be seen from the fact that Isocrates Nicocles 24, arguing that kingship is the best , admits that the Spartans are the best governed of the Greeks, 17 but argues that they are governed by kings when at war; while his Areopagiticus 61, arguing that [not the degenerate modern Athenian democracy but] the democracy of our glorious ancestors was the best , admits that the Spartans are the best governed but argues that this is because they are the most democratic; elsewhere and with more truth, including in the Nicocles passage, he describes Sparta as an oligarchy.) 18 Perhaps the main division of the of the Spartans literature is into texts written for non-Spartans using historical or philosophical reasoning

_________
16 Aristotles are not themselves pamphlet-literature in this sense, but are a correction of earlier texts that were pamphlet-literature as I have described it; a natural development, paralleled in Aristotles writings on other topics. Jacoby 1949, 211-5 distinguishes, within the literary form of the , three , political (pamphlet literature like the pseudo-Xenophon and the Xenophon), philosophical (like Hippodamus and Platos, not primarily about some existing but seeking to determine the best ), and scientific (invented by Aristotle and in fact represented only by him, or by him and the students who may have collaborated on his series of 158 ). I am probably not substantively disagreeing with Jacoby on the relationship between the Aristotelian and the earlier texts (and Jacoby agrees that Aristotles are ultimately intended to subserve the construction of the best state), although it seems a poor idea to posit a genre with just one author; I think he draws too great a distinction between his first two types, although of course Platos is much more reflective and sophisticated than the earlier texts. But Jacoby is not distinguishing of the soand-sos as historical from without genitive as normative: on the contrary, and rightly, he takes the of the Spartans (apart from Aristotles) as normative texts not intimately related to historical reality. 17 Or, with an emendation, that the Carthaginians and the Spartans are better governed than all others. 18 At Panathenaicus 41 he says that most people moderately praise the city of the Spartans, but some refer to it as if the demigods were there. At Panathenaicus 111, the supporters of the Spartans, being defeated by Isocrates arguments that the Athenians have benefited the Greeks and the Spartans have harmed them, try to turn instead to the issue of the , where, it is agreed, the Spartans come off well: so Isocrates investigates . He takes it as agreed that the present Athenian democracy is bad, but argues that the Athenian ancestral is better than the Spartan , and that Lycurgus in fact used it as his model, against the claim of Xenophon Resp. Lac. I that Lycurgus imitated no other city, and against the common claim (already at Herodotus I 65) that he imitated Crete.

12

STEPHEN MENN

to present an idealized Spartan as a contrast with other and so as a possible model for reform elsewhere, and texts written for Spartans, like those of Lysander, Pausanias and Thibron, relying often on oracles and legends to present an idealized past Spartan as a model for reform (presented as a return to a glorious past) at Sparta itself; but the distinction is far from absolute. 19 Aristotles numerous references to earlier discussions of the , and the references we have already noted in Plato, in Laws IV to the difficulty of fitting the Spartan under any of the standard types, and in Republic VIII to the Spartan and Cretan praised by the many (praised certainly not by the democratic masses in Athens, but by most of those who theorize about such things), show that Plato and Aristotle were

_________
19 The Lysander and Pausanias, like the rest of the -of-the-Spartans literature, were political pamphlets urging some present political aim (for Lysander, opening the kingship to those not born to it, for Pausanias, abolishing the ephorate) on the basis of arguments from alleged Spartan history. From the little we know about them, however, Lysander and Pausanias seem to have been much more interested in arguments from oracles (FGrHist #582 T3 and Plutarch Lysander 25 in FGrHist #583 T1) than is the rest of the literature, which generally treats the Delphic authorization of Lycurgus laws as a mere divine rubberstamp on Lycurgus work, if not as a cynical fabrication by Lycurgus. Undoubtedly the reason is that these works were meant to be read or heard at Sparta, and the Spartans were particularly impressed by oracles (so, on Lysanders text, Plutarch Lysander 25, and cp. Diodorus Siculus XIV xiii 3 [from Ephorus] on Lysander trying to bribe the priestess at Delphi considering that the Spartans especially paid heed to oracles; both in FGrHist #583 T1). These texts were designed to support radical change (presented as a return to a mythical past) at Sparta, while most other -of-the-Spartans literature was designed to support assimilation to a Spartan ideal at other cities, especially Athens. On the other hand, Sphaerus of the Spartans, which so far as we can tell (e.g., if, as is generally thought, it lies behind much of Plutarchs Lycurgus) was very much in the tradition of other -of-the-Spartans literature, was written in support of the revolutionary innovations (or return to Lycurgus) of Cleomenes III at Sparta, and since Cleomenes did in fact abolish the ephorate, undoubtedly Sphaerus referred back to Pausanias at least to prove that the ephorate was a later deviation. Since Sphaerus fellow-student Persaeus was the captain of the Macedonian garrison at Corinth defending the Peloponnese against Cleomenes, his of the Spartans presumably argued for a very different view of what that traditional was, and inferred (or allowed it to be inferred) that the Spartan revolutionaries were dangerous innovators. Both Sphaerus and Persaeus would presumably have been writing (at least inter alia) for a Spartan audience; so we cannot cleanly distinguish between literature for Spartans based on oracles and literature for Athenians and other Greeks based on sophistic and philosophical modes of reasoning. (However, Spartans may have been better educated in Sphaerus day than in the fourth century). Although Xenophon is writing in the first instance for an Athenian or other non-Spartan audience, he may also intend some Spartans to overhear; cp. Isocrates Panathenaicus 250-51, saying that while most Spartans will care no more what is said about them in Athens than at BACAP (beyond the pillars of Heracles), the most intelligent people there will pay attention.

ON PLATOS

13

conscious of this existing mode of discourse , with its conceptual apparatus, standard topics, and commonplaces of praise and blame; and that they could, when and if they wanted, call up for their readers the expectations of this discourse, inviting their readers to judge their own new proposals against the background of the standard classification of , the standard evaluation of laws and customs as designed to preserve the , and the standard praise of Sparta. Juxtaposing the different extant texts will help to bring out this generic background and thus to shed light on many details in Plato. When Aristoxenus said that almost the whole of Platos was written in the of Protagoras (DL III 38), he was doubtless being deliberately provocative, but he must have been thinking of something. It is curious that Aristoxenus does not say in the , a title credited to Protagoras elsewhere (DL IX 55, which also lists two books of ); this suggests that Protagoras was just a section of his , containing arguments for and against each of the standard (as in Herodotus), and perhaps for and against some more outlandish customs as well. Plato may well have taken over some arguments from Protagoras, as well as from other earlier literature, naturally with the intention of creating something new. 20 II. Why, then, did Plato choose to write a text in this modeand why did he decide to modulate what starts out looking like a standard Socratic dialogue, on justice as a virtue, into a ? The standard answer, so far as the question is raised at all, is that Plato depicts the types of city in order to give larger-scale models of the types of soul, so that, having first

_________
20 In the glory-days of Quellenforschung it used to be suggested that the debate in Herodotus was taken from Protagoras (especially since both Protagoras and Herodotus, as also Hippodamus, are supposed to have been involved in the foundation of Thurii). But it also used to be suggested that Protagoras book was the source of the utopian feminist constitution parodied in Aristophanes Ecclesiazusae, whose ridicule may be referred to at Rep. V, 452a7-e3, and this was supposed to be what Aristoxenus was thinking of. I find it hard to imagine Protagoras as proposing utopias, or as arguing, against the common sense of the society of his time, for the equality of women and men. (If Protagoras had proposed such a utopia, we would expect Aristotle to mention it in Politics II.) On the other hand, I have no trouble at all imagining Protagoras arguing for the superiority of womenhe would also have argued, in the other half of the , for the superiority of menand this, not equality, is what the Ecclesiazusae depicts.

14

STEPHEN MENN

discerned the justice and injustice writ large in the city as a whole, we will be better able to discover the justice and injustice within the individual soul and so to respond to the challenge of Glaucon and Adeimantus (cf. Rep. II, 368c4-369a7). This answer seems to me to be clearly inadequate. In the first place (as Ferrari rightly argues, in Griffith and Ferrari 2000, xxiii), Plato goes into far more detail about the different than would be needed simply to show the correspondence with each type of soul (what, for instance, does the equality of women guardians in the ideal correspond to within the individual soul?); and it is at many points clear that Platos discussion of the different is engaging with earlier literature, and not simply imagining civic equivalents for different individual psychologies. Second and more importantly, the issue of justice was already a political issue in Book I, and the standard sophistic topic of the is first raised, not by Socrates in Book II, but by Thrasymachus in Book I: dont you know that some cities are governed tyrannically, others democratically, other aristocratically (338d7-8), distinguished by their ruling part (d10)? 21 And Thrasymachus invokes the theory of , not to display the variability of laws and customs in support of a cultural-relativist challenge to morality, but as a specifically political challenge to justice. In each city, the ruling part makes laws commanding actions that are advantageous to that ruling part, and those actions are called just for the ruled. If this is the whole truth about justice, then the rulers will have no reason to limit their exploitation of the ruled by any objective norm of justice, and the ruled will have no reason, except fear or habit, to live either by what is called justice in their society, or by any other kind of justice. Plato naturally finds it very important, in the context of his Socratic dialogue on justice, to refute these claims of Thrasymachus and to give a positive alternative account of what justice is, and of why rulers and ruled have reason to follow it. And Platos task in refuting Thrasymachus is made harder by the fact that he agrees with Thrasymachus about how laws are made in any of the three standard types of . But even if all existing are Thrasymachean , where the rulers rule in their own interests, Plato wants to show that a Socratic , where the

_________
21 Thrasymachus does not here use the word , although, as we will see, someone who sounds much like Thrasymachus elsewhere in Plato does use the word. He does however speak of (339a1), and, as Bordes 1982 shows, was the precursor of as a technical term (and this is probably its sense in Protagoras title , DL IX 55), and fourth-century authors including Plato still sometimes use the term in this way.

ON PLATOS

15

rulers rule in the interest of the ruled, is possible; in such a , what is conventionally called just would be rooted in what is just by nature, and it would be in the interest of the ruled to live by what the laws declare to be just. Thrasymachus claim about the different types of and about what is called just in each is implicitly a modal claim, about all possible and not merely about all actual ones; if Plato can show the possibility of his Socratic , he will have refuted Thrasymachus, as well as displaying writ large what is just by nature. However, this just seems to push the problem back a step: why should Plato have introduced Thrasymachus and in the first place into the more typically Socratic and non-political discussion with Cephalus and Polemarchus? If Thrasymachus is just an eccentric (and, as is often thought, incoherent) immoralist, it seems strange to let his concerns, and his conspiracy theory of justice, dictate the development of the grand argument of the Republic. But this is not how Plato thinks of Thrasymachus: he thinks of him as the logical outcome of the standard sophistic theory of , and he thinks that this theory has much empirical support in the actual conduct of the different cities. The connection between Thrasymachus and the sophistic theory of the , not made especially clear in Republic I, becomes clearer in a parallel text from Laws IV, which will be important in interpreting the Republic. Laws IV, together with some references in Aristotles Politics, points us back to an early stage of -theory, which may go back to Protagoras and which Plato sees as logically leading to Thrasymachus; Laws IV also suggests a Platonic strategy for responding to this theory, which we can also see at work in the Republic. Starting at Laws IV, 712b8-c1, the Athenian Stranger and his Spartan and Cretan interlocutors are discussing what they should prescribe for their new city; a number of possibilities are discussed, and it is proposed that, since it is no longer possible for us to be ruled by daimons as in the age of Kronos, the best imitation of that is to be ruled not by one or more human beings, as in a monarchy or oligarchy or democracy, but by as much of immortality [i.e. of reason] as is in us, calling that dispensation of reason law [etymologizing as ] (713e8714a2). This of course contradicts the theory that every is a monarchy or oligarchy or democracy, and it assumes that it is possible to produce laws which simply reflect universal and impersonal dictates of reason, against views that justice is relative to the type of , and, especially, against the view that reason in legislating is instrumental to the interests of the human rulers of the . Plato decides to confront the difficulty head on:

16

STEPHEN MENN

you know that some people say that there are as many forms [ ] of laws as there are of , and we have just gone through how many forms of most people say there are [= monarchy, oligarchy, democracy]? Consider that the present contest is not about something small, but about the greatest: for it has come back to us contested where one should look for the just and the unjust. (714b3-8)

Now at first sight the commonplace sophistic thesis that there are as many forms of laws as of might not look so alarming, but Plato thinks that, when its implications are spelled out, it is a challenge to the concept of laws as dispensations of reason reflecting an objectively just order, and specifically to the Laws project of legislation aiming at promoting virtue in the citizens. The Athenian Stranger continues with the view of (apparently all of) the people who say there are as many forms of laws as of :
they say that the laws should look neither toward warfare [= toward promoting military virtue, like the Spartan and Cretan laws] nor toward virtue as a whole, but rather, whichever is established, they should look toward the advantage of this [ ], that it should rule forever and not be dissolved; and they say that the natural limit [or definition] of justice is most rightly expressed thus ... the advantage of the superior. (714b8-c6)

This theory (as stated here and further spelled out 714d1-10) is certainly meant to be the same as that set out by the character Thrasymachus in the Republic (the advantage of the superior, , is verbatim Thrasymachus formula at Rep. I, 338c2): it is possible that it was also the theory of the real person Thrasymachus, and it is possible that Plato intends to allude to Thrasymachus here in the Laws, but it is at least as likely that he is just alluding to the views of a much broader class of people who theorize about and laws, and that the character Thrasymachus in the Republic is intended to represent this broad group rather than anything peculiar to the real Thrasymachus. Certainly Plato thinks that the Thrasymachean views of law and justice logically follow from a widely held theory of , even if it was especially Thrasymachus who made the conclusion explicit. In any case, the account in Laws IV gives the proper context in -discourse for the claims which Plato puts in the mouth of Thrasymachus in Republic I, and which he is concerned to reply to not only there but in the rest of the Republic. There is something distinctive and worth noting in the way that the Thrasymachean theory of Laws IV speaks about . It seems a bit odd to speak of a as ruling [ , 714c2], especially since this theory has insisted that cities are ruled by one or more human beings rather than by laws, and likewise to speak of what is advantageous to the , rather than to one or more human beings. The explanation is that

ON PLATOS

17

this theory does not really distinguish between the and the ruling group: for something to be advantageous to the is simply for it to be advantageous to the rulers. This way of speaking about the turns up even in Aristotle and [= ruling body] mean the same thing, and the is what is sovereign [ ] in the cities, and either one or a few or the many must be sovereign (Pol. III 7, 1279a25-8, cp. III 6, 1278b8-14)although it is certainly inconsistent with Aristotles considered theory of the , and must be a residue of an earlier way of thinking. Plato does not reject this identification of the principles of legislation in the with the advantage of the ruling group, as an empirical description of the behavior of tyrannies, oligarchies, and democracies. He does reject the accompanying normative claims (the laws should [ ] look toward the advantage of this [ ], Laws IV, 714c1-2); he also rejects the universal claim that this is the behavior of all actual and possible , which helps to support the normative claims; indeed, he refuses to grant the honorific title of to any authority which does govern in this way. As the Athenian Stranger says, what his opponents describe has indeed happened all too often as a result of civil strife, where the victors make everything belonging to the city their own, excluding the losers from any share in rule and continuing to exclude their descendants for fear of vengeance (715a4b2); but we say now that those are neither nor correct [ ] laws which they had made not for the sake of the whole city; and those who acted for the sake of some, we call not citizens but civil warriors, and the legal duties [ ] which these people assert to exist, we say are said in vain (715b2-6); thus the three standard forms of , having arisen through such winner-take-all conflicts, are not , but managements of cities which are masters and slaves in different parts of themselves, and each is called the power [ ] of its master (712e10713a2). 22 The same stage of -theory that Plato criticizes here in Laws IV also seems to be alluded to in Aristotles Politics. Indeed, Aristotle seems to regard this theory as giving the default assumptions that he must modify, which is surely not how he would treat an idiosyncrasy of Thrasymachus. What is most striking (beyond the equation of and that we have already seen) is that Aristotle several times cites, and endorses, a formula very close to the one Plato had used as a summary of his opponents theory, namely that there are as many forms of laws as

_________
22

Similarly at Laws VIII, 832b10-c7.

18

STEPHEN MENN

. Thus it is clear that the laws must be laid down (Pol. III 11, 1282b10-11); it belongs to this same prudence [which studies the different ] to know both the best laws and the laws which fit with each of the : for one should [ ] lay down, and everyone does in fact lay down, the laws and not the the laws (IV 1, 1289a11-15); those who will hold the supreme offices must have the virtue and justice in each which is : for if legal duty [ ] is not the same in all , justice [ ] too must differ (V 9, 1309a36-9). For Aristotle, as for the opponents, to say that the legislator should legislate is to say that he should institute whatever practices tend to preserve the ; and Aristotle is willing to infer, with the opponents, that justice in the sense of the political virtue (the virtue of the good citizen) is different in different . But Aristotle tries, while accepting these premisses, to modify the theory so as to avoid the Thrasymachean conclusions. In part this is because he does not individuate the simply by its ruling group, so that preserving the does not mean simply preserving the given rulers in power. In particular, kingship and aristocracy and proper, being the rule of the one, the few or the many in the interests of the whole city, are distinct from tyranny and oligarchy and democracy, the rule of the one or the few or the many in the interests of the rulers themselves, and so an aristocracy (say) will not be preserved if the same group continue to rule but in pursuit of different ends; and Aristotle, like Plato, will say that tyranny and oligarchy and democracy are not right [ ] (Pol. III 7, 1279a22-b10, IV 2, 1289a26-30), or even that they are not but rather (implied Pol. VII 14, 1333a3-6). Aristotle also tries to avoid the implication that justice and other virtues are relative to the , by distinguishing moral from political virtue, and holding that the virtue of the good citizen and the virtue of the good person coincide only in the ideal . It seems very likely that this pre-Aristotelian and pre-Platonic theory of justice and laws and is in fact due to Protagoras. Protagoras is said to have given the laws of the pan-Hellenic colony of Thurii, 23 and he, like Aristotle in the Politics, would have seen training in the art of legislation as the highest part of his training of aspiring . The Prota-

of

_________
23 Although on a dubious authority, Heraclides Ponticus in DL IX 50. Still, someone must have done it, especially because there was no one mother-city whose laws the Thurians could simply copy, and who better than Protagoras? (Well, as Paul Cartledge says, Hippodamus; but then Aristotle would probably have said so in discussing him.)

ON PLATOS

19

goras of the Theaetetus explains that wise and good orators make good/useful [ ] things appear [ ] to be just in place of wretched ones; whatever things appear to each city to be just and fine are so for that city, for as long as it practices/deems/legislates [ ] them, but the wise man has made good/useful things both appear and be for each of them in place of wretched ones (167c2-7). For things to appear just and fine to a city is simply for the city to have a law or decree enjoining those things, and for something to be just is simply for it to be in accordance with the law; but while all laws are equally true, some are better and more useful, as the perceptions of healthy person are better and more useful than those of a sick person, and the Protagorean orator or legislator will replace worse laws with better ones, as the doctor replaces worse perceptions with better ones. Plato speaks here equally of the better laws as being advantageous to the city [ , 172a5-6, b1, as in Laws IV, 714b8-d3, and Thrasymachus at Rep. I, 338c2ff]. When the one, few or many sovereign in a city summon Protagoras (or one of his students) to help them make laws, this is because they think that he will be better able than they to determine what laws will be most advantageous for them: they set the end, and he determines the best means (the is in general the aim of discussion in deliberative assemblies, Aristotle Rhet. I 3, 1358b20-25). When you summon Protagoras, he will presumably say something like this: When you set out to legislate well, you must first of all know the target that legislation aims at [cp. Laws IV, 705e3-706a4, etc.]. In making laws for your city, you should not aim at commanding what is just and forbidding what is unjust: for there is nothing just or unjust by nature, but everything is just or unjust according to different circumstances and for different cities, and whatever you command will be just in your city, and whatever you forbid will be unjust. But you should have a care that what you command and enact as just, and what you therefore do in your city, will also be advantageous. For not everything that people desire turns out well for them. A patient may desire certain foods and find them sweet, and yet the doctor may know that if he takes them, they will be harmful to his health, and he will find everything bitter tomorrow. So too for some customs which the city may desire today, the legislator may know that they will be harmful and lead to civil strife, which is disease in the whole city. But what customs are advantageous? Again, you must know that there is nothing advantageous by nature, but everything is advantageous or harmful according to different circumstances and for different cities: just as one diet is advantageous for a phlegmatic person and another for a bilious person [cp., with different examples, Protagoras 334a3-c6], so one custom is

20

STEPHEN MENN

advantageous for a monarchy, another for an oligarchy, and another for a democracy. That is always harmful which leads to civil strife, and that is always advantageous which preserves the . Wise legislators in every city have taken this as their aim, and this is why different things are lawful and just in monarchical and in oligarchic and in democratic cities; and this is why, in making laws for your city, you must not simply imitate what is just and advantageous elsewhere, but must take counsel that you enact what will be advantageous here. If Protagoras said something like this, he had no sinister political agenda. Protagoras was neither a democrat nor an antidemocrat, but a travelling educator and political advisor who had to be useful to each of his clients, whether a sovereign individual or group that might ask him to make laws, or an individual who aspires to a political career (and thus wishes to acquire political virtue) in any of the different . But what begins as value-neutral social science may have sinister political implications. If what is just is what is in accordance with law, and if laws do not reflect nature but are freely enacted agreements or conventions, and if those conventions are made with a view to the advantage of the conveners, that is, of the sovereign whether one or many, and if this advantage is chiefly the preservation of the , that is, the preservation of their own rule, then, as we have seen above, law and justice give the rulers no reason, beyond the fear of provoking a revolution, to restrain them in exploiting those they rule, and the ruled have no reason except fear of punishment to follow the so-called legal duties decreed (in vain, as Plato says, Laws IV, 715b6 above) by their rulers. In the good old daysso Plato might saythe Greeks used to despise the tyrant, the man who ruled over his fellow-citizens despotically, i.e., as a master over slaves, unrestrained by law and justice; and they used to praise law and the common covenants which allow us to live together in civil peace, and not eat one another like the beasts. Now, taught by Protagoras and his kind, they contrast law to nature, not as civilization to savagery, but as the merely conventional, and they suspect the laws of an oligarchy or even a democracy of serving the partisan interests of the conveners, much like the decrees of a tyrant. 24 In the days of Solon and Tyr-

_________
24 This line of thought is beautifully illustrated in Alcibiades conversation with Pericles about laws, Xenophon Mem. I ii 40-46. (While modern readers tend to sympathize with Alcibiades, I agree with Dorion in Bandini and Dorion 2000, CLX-CLXIX, that Xenophon is horrified by him and is trying, not necessarily successfully, to show that this is not what he learned from Socrates.) The pseudo-Xenophon of the Athenians accepts some, but not all, of the amoralist conclusions of the Protagorean theory of : the author

ON PLATOS

21

taeus, the Athenians and Spartans could agree in praising the rule of law and condemning despotic rule, and when the Peisistratids were overthrown they were in concert, but then the anti-tyrannical movement split into oligarchic and democratic factions: the democrats accuse the oligarchs too of ruling despoticallyfor what does it matter whether the people are enslaved to one or to several masters?while the oligarchs accuse the of ruling as arbitrarily and lawlessly and irrationally and hubristically as a collective tyrant (this line of thought already at Herodotus III 81; Herodotus debate does not include an explicit democratic critique of oligarchy). Thrasymachus agrees with them all, and concludes that tyranny and oligarchy and democracy are despotisms alike; Plato agrees too, but rather than identifying with , rule over fellow-citizens with rule over slaves, and concluding that law and justice are merely a mask for partisan interests, he argues that tyranny and oligarchy and democracy are not , but managements of cities which are masters and slaves in different parts of themselves, and each is called the power [ ] of its master (Laws IV, 712e10-713a2 above), and that they do not enact correct laws or real . By contrast, in a real , governed by laws which reflect not the advantage of human rulers but rather the rule of Kronos, that is to say of , god would be for us most of all the measure of all things, much more than, as they say, some man (716c4-6). 25 This puts the burden on Plato to explain what a real would be like, and what law and justice in it would be; as I put the question above, if all actually existing are Thrasymachean , what would a

_________
thinks that there are objective standards of human goodness, justice and good government, but that since these would lead to the good people ruling in their own interests, overthrowing the democratic , and indeed enslaving the bad, the is right to act in its own interests and contrary to these objective standards (so I 4-9). It should be stressed that both benign and sinister Greek social contract theories begin from real social contracts, such as that imposed by Solon to head off at Athens, or the Lycurgan social contract at Sparta: these social contracts were originally sworn by the oath of all the citizens, or were afterwards remembered as having been so sworn, and at Sparta the oaths between king and city are ritually renewed every month, with the ephors standing in for the city, according to Xenophon Resp. Lac. XV 7 (cp. 2 Kings 23:1-3). Social contract theories generalize such events and project them back onto the first origin of law, of the , or even of (evaluative) language. 25 Compare the opening of the Laws, I, 624a1-6, where the Athenian asks whether a god or a man is responsible for the laws of Sparta and Crete, and Cleinias and Megillus both say a god: this is at least believed to be true in Sparta and Crete, and would be true in a real . (This passage is stressed, in an unusual alliance, both by Strauss 1975 and by Burnyeat 1997.)

22

STEPHEN MENN

Socratic , ruled in accordance with knowledge and in the interests of the ruled (or of the whole city rather than of its ruling part) be like? III. Here Sparta helps. Sparta enters the argument in rather different ways in the Republic and the Laws, but I do not think Platos real attitude toward it has changed much. 26 Sparta plays a much more explicit part in the Laws which features the only Spartan and the only Cretan in Plato, talking with Platos Athenian spokesman about the laws for a new Dorian colonyand I think the Laws can help, by comparison and contrast, in understanding the strategy of the Republic. While Sparta and Crete have been subject to examination beginning in Book I of the Laws, they are brought up again in Book IV, when the Athenian stranger asks what we should prescribe for our new colony. The interlocutors assume that the choice must be between aristocracy, oligarchy and democracy (since tyranny, if it is a , is clearly a bad one), but the Athenian stranger, in order to show them that there are other alternatives, asks them which of these types their own home in Sparta and Crete would fall under, and they are unable to answer: the Spartan and Cretan share some distinctive features with each of the given types (and the Spartans also have kings). Plato could now give a minimalist solution by saying that the Spartans and Cretans have mixed , that the standard types are idealizations and that we should expect to encounter many shades of gray between them. 27 Instead, the Athenian says that the reason the Spartan and Cretan have difficulty classifying their in the standard scheme is that

_________
26 I will say more about this below. Here let me note that the passage of time between the composition of the Republic and of the Laws is not in itself a reason to expect Platos attitude to Sparta to have changed, against the peculiar view that it took the defeat at Leuctra in 371 to convince Athenian conservatives that the Sparta of their own time was no model. If someone was going to be convinced by historical events, then the experience of Lysander and the Thirty, the imperial arrogance which led to the Spartans alienating their Theban and Corinthian allies and driving them into the arms of Athens and Persia, the Spartan betrayal of the Greeks of Asia Minor in the Kings Peace, the Spartan seizure of the Theban citadel and the attempted seizure of the Piraeus, and so on, would have convinced them. Isocrates wrote his most pro-Spartan piece, the Archidamus, after Leuctra and Mantinea. Athenian conservatives could of course still choose to believe that an earlier uncorrupted Sparta remained a good model. 27 For intermediates between the ideal types see Rep. VIII, 544c8-d4; for mixed constitutions (although not with the precise term which later becomes technical) see Thucydides VIII 97 and the Laws itself, III, 693d2-e3.

ON PLATOS

23

you really do belong to , 28 whereas the ones we have just now named are not (712e9-10)as he explains in a text we have already cited, they are forms of despotic rather than of political rule (712e10-713a2). The Spartan and Cretan (or the Spartan and Cretan in the singular, as Rep. VIII, 544c2-3 puts it) are thus important to Plato as counterexamples to the reductionist ProtagoreanThrasymachean theory of . However, to say that these are really is not to say that they are good , and while the Laws does not flatly call the Spartan or Cretan bad and erring as the Republic does (V, 449a2-3, covering all four deviant forms), here too Plato is sharply critical of their (mythical) lawgivers Lycurgus and Minos, at least as their work is commonly understood. When the Athenian, at the beginning of the dialogue, asks the Cretan about the purpose of some particular Cretan customs, the Cretan replies that the lawgiver of the Cretans arranged all our public and private lawful practices with a view to war (626a5-7), and indeed that the mark [or goal, ] of a well-governed [ ] city ... [is] that it must live ordered in such a way as to defeat the other cities in war (b7c2); the Spartan agrees and indeed says that any Spartan would agree (c45). The Athenian argues that this is misguided, and that, as peace and friendship are the best things for a city, and war merely a necessary means to them, an accurate lawgiver must legislate the things of war for the sake of peace, rather than the things of peace for the sake of war (628d7e1). On the face of it, this implies that the Spartan and Cretan lawgivers have gone badly wrong, but Plato tries to find a way of praising them. Happiness [ ], for a city as for an individual, depends primarily on the possession of virtue, rather than of health or wealth, and Lycurgus and Minos, recognizing this, have arranged all their laws with a view to instilling virtue in the citizen-body (631b3-d6, a2-4), by contrast with the laws of all the other so-called as described in Book IV, which aim merely at preserving the power and advantage of the ruling group (these laws contrasted with laws aiming at virtue, IV, 714b8-c4). Unfortunately, it appears that Lycurgus and Minos have legislated looking only to courage or military virtue, which is less important for happiness than wisdom and temperance and genuine justice (630a7-d1, 631c5-d1). 29 But Plato

_________
28 Using the language of , to be citizens or belong to the citizen body or have a share in the civic life and its rights and duties. 29 At Laws II, 660d11-661d4, the Athenian explains that, without justice, no life can be happy, and all the other so-called goods, including external goods but also courage, can

24

STEPHEN MENN

argues, or pretends, that this is not the fault of Lycurgus or Minos but of their modern interpreters: Lycurgus and Minos were in fact aiming at complete virtue and not merely at its least important part (630c1-631a8), and the Athenian stranger challenges his interlocutors to give an interpretation of the Spartan and Cretan laws that would reveal them as a rational system designed to promote virtue (632d1-e7). 30 It is this hypothesis that leads the three characters into their collective examination of laws, involving many criticisms of particular Spartan or Cretan practices, and designed to lead an admirer of a charitably reinterpreted Laconizing ideal to the revised version of that ideal which Plato presents as the laws of his imaginary new Dorian colony and as a model to be adapted (V, 739a1-b7, 745e7-746d2) by lawgivers elsewhere. The Spartan and Cretan interlocutors serve as stand-ins for the Laconizing intended reader whom Plato hopes to persuade; the Athenian stranger serves as a stand-in for Plato himself, the good Athenian by contrast to the usual products of the democracy, 31 who has the philosophical and mathematical knowledge which the Spartans and Cretans are sadly lacking. But if all other existing cities aim only at the advantage of their rulers, and the Spartans and Cretans aim at virtue, even if it is a militaristic and unphilosophical conception of virtue, these rather than the other will be the natural starting-points to

_________
only make it worse; the Cretan (and presumably also the Spartan) is politely unconvinced. Compare Isocrates Panathenaicus 182-88, arguing against those who praise the Spartans that victories won without justice are not properly speaking virtuous or noble and should not be praised, and that the Spartans have never had any concern with justice or with virtue properly so called, but only with gaining other peoples possessions by violence. (But Isocrates does not seem to take the Platonic line of denying that the Spartans genuinely benefit themselves in this way). 30 Cp. Xenophon Resp. Lac. X 4, where Lycurgus mandates the practice of all the virtues. 31 At Laws III, 698a9-701e8, the Athenian democracy, particularly as it has become since the Persian wars, is represented as a sad warning of the excesses of freedom. But at Laws I, 642b2-d2, the Spartan character turns out to be a hereditary of the Athenians at Sparta, and, unlike most other Spartans, is full of good will toward Athens: those Athenians who are good are especially so for only they are, without compulsion, spontaneously [ ] by divine allotment [ ] truly and without artifice good. (This is close to Platos descriptions elsewhere of good people arising spontaneously or even miraculously in bad environments, e.g., Rep. VII, 520a9-b4 which also uses ; in the Meno Pericles and the like have their virtue , 100b2-4, and Socrates says something similar about Glaucon and Adeimantus at Rep. II, 368a5-7, a passage I will return to below.) The Cretan character then chimes in that he too has old family connections with Athens (Laws I, 642d3-a1). The interlocutors of the Laws are thus types of the Laconizer who might be amenable to Athenian philosophical persuasion.

ON PLATOS

25

criticize and reinterpret and try to reshape if we are looking for the best (or even the second- or third-best, V, 739a1-b7) . The Republic proceeds rather differently, and its ideal is further from Sparta than is the ideal of the Laws, most strikingly in abolishing the family and private property for the guardians. Nonetheless, I think it is correct to say that in the Republic too Plato constructs his ideal by beginning with Sparta (not the real Sparta but the idealized Sparta of Laconizing literature) and modifying it as necessary. For the Republic too the Spartan and Cretan , the timocracy, is superior to the so-called that Thrasymachus talks about in Book I, tyranny and oligarchy and democracy; in the terms of Republic VIII, the timocracy is ruled by , which values the noble even if it may not have an adequate conception of the noble, while these other are governed by appetite, which values only pleasure. So the Spartan , here as in the Laws, has the value of a counterexample. It is not a merely Thrasymachean , but it is not a Socratic either, since the Spartiates do not rule in accordance with knowledge, nor do they rule in the interests of the ruled, notably the perioeci and helots. Plato thus needs to consider in what ways this would need to be modified to make it Socratic. However, Plato does not do this explicitly in constructing the ideal city in Books II-VII. The Spartan is not thematized until Book VIII, when it is sharply criticized, even satirized, but where it is also said to be, of all the bad and erring , the closest to the good and correct one. So while the character Socrates is constructing the best he is not explicitly pointing out what is wrong with other real or imagined , or how these might fail to meet their own intended aims, and so he does not avail himself of the Laws strategy for bringing admirers of other along with his argument. Indeed, Aristotle seems to be making just this criticism of the Republic in a passage that we have cited from Politics II 1, where we must examine the other , both the ones practiced in certain cities that are said to be well-governed [ ], and any others that particular people have described and thought to be right (1269b29-32), not just to draw on what they have got right, but also to show that we ourselves are seeking a further not arbitrarily but because the ones that now exist are not right (b34-5). Aristotle thus devotes Politics II to showing what is wrong with the proposals of Plato and Phaleas and Hippodamus and with the Spartan and Cretan (and for good measure Carthaginian) and with the ancestral at Athens, the models praised by different earlier writers, before going on to give his positive proposals, either for the best absolutely or for the best under certain conditions. This is precisely what Plato

26

STEPHEN MENN

has not done in the Republic, where he constructs the ideal city out of the raw materials of human nature without explicit reference to other cities, and turns to savage criticism of the Spartan and Athenian ideals only in Book VIII, probably after most readers have stopped reading. Aristotles complaint is that this makes it more difficult to motivate the reader who may be satisfied with the existing ; more to our present point, the fact that Plato criticizes Sparta and Athens only after he has presented his own ideal can make it harder to follow what he is doing in constructing his ideal. Nonetheless, in the Republic as more explicitly in the Laws and Politics, we must understand the ideal as emerging from a process of correction of the inadequacies of the (real or imagined) that people ordinarily admire. In a sense Book VIII delivers the punchline of the Republic, or at least of the specifically political project of the Republic, and the ideal must be read backwards from Book VIII: it is only by seeing the contrasts between Callipolis and the other that Plato draws in Book VIII that we understand why he has built certain features into Callipolis in the first place. And, if Plato constructs Callipolis by modifying other , it is most immediately by modifying the Spartan , which according to Book VIII is the first degeneration of the Callipolis and has the most in common with it. Furthermore, given the Laconizing background of the -genre, we might assume that Platos intended reader would initially expect the developing ideal to resemble Sparta, and that, as in the Laws, it would be the differences from Sparta (or from Sparta as usually imagined) that would be foregrounded and would be interpreted by the reader against that background expectation. As we will see, this assumption is strongly confirmed by Platos characterizations of the interlocutors, especially Glaucon, and by their interventions when Socrates exposition goes too hard against their expectations (and by their non-intervention when it does not): Glaucon and Adeimantus are stand-ins for the intended reader of the Republic, just as the Cretan Clinias and the Spartan Megillus are stand-ins for the intended reader of the Laws. What Plato presents in Republic VIII as the story of the decay of the Callipolis into the timocracy, explaining both the defects of the timocracy and the good features that it retains from the Callipolis, can be read in reverse as the story of the generation of Platos ideal, showing both what features it kept from the Spartan ideal and what features Plato found it necessary to change. The similarities between the Callipolis and the Spartan ideal are fairly obvious, and have been pointed out since antiquity by those who have viewed Plato as a Laconizer. (Dicaearchus says that Plato mixed Socrates no less with Lycurgus than with Pythagoras, Fr. 45 in Mirhady 2001;

ON PLATOS

27

Plutarch says that Plato and Diogenes and Zeno, and all who are praised for having undertaken to say something about these matters took over Lycurgus plan [ ] for the , Lycurgus 31.) 32 One way to collect similarities is to compare Platos descriptions of the ideal city with Xenophons of the Spartans, the best sample we have of the early Laconizing literature. Both Xenophons and Platos ideal cities have an elite population (for Xenophon the Spartiates, for Plato the rulers and auxiliariesI will say guardians for both) which is forbidden to engage in trade or production and devotes itself full-time to military and civic activity and to various forms of training, while the producers are excluded from political life. 33 While Xenophon may stress the military function of his elite more than Plato does, Xenophons main emphasis is not on military activity but on the all-encompassing training for virtue that occupies the whole life of the elite (Lycurgus compelled all [the Spartiates] to practice all the virtues publicly, Resp. Lac. X 4); and Plato too first introduces the guardians as military specialists (Republic II, 373e-374e) and describes the temperament and education they will need to fulfill that function well only later does he distinguish the guardians into rulers and auxiliaries, and propose to give the rulers a mathematical and dialectical training beyond the common education that they get along with the auxiliaries. Both in Xenophon and in Plato, much of the account of the is devoted to the education of the elite at different stages of their life, proceeding in roughly chronological sequence through their life-cycle (Xenophon Resp. Lac. I 3 is apparently following Critias of the Spartans DK88 B32 in saying that we must begin with the generation of offspring; Plato defers this to Book V). This does not seem to have been a standard way of describing in general: rather, it is particularly appropriate for describing Sparta (as the Laconizers imagined it) and Cal-

_________
32 Although the here seems to be the goal (the persistence of the citizens in a life of virtue and concord, allowing them to maintain their collective freedom and happiness) rather than the details of execution. Plutarchs picture of Lycurgus goals is of course itself influenced by Platonic and Stoic ideals. 33 Actually Xenophons Resp. Lac. keeps almost complete silence about the helots and perioeci (perioeci mentioned once incidentally, XV 3, slaves three times incidentally but helots never, though Xenophon uses the word in other texts); so we must fill in information about them from other sources. Resp. Lac. XI-XIII manage to give the strong impression that the Spartan army was composed entirely of Spartiate full citizens, which was far from being true. (Lipka 2002, 99 may be right that the treatise distinguishes Sparta from Lacedaemon as nouns, but, contra Lipka, Xenophon cannot intend Lacedaemonians to include the perioeci [especially at XIV 2 this would make no sense], and I have had no scruples in translating the word as Spartans.)

28

STEPHEN MENN

lipolis, in both of which the whole civic life is geared toward forming the desired type of human beings, first in producing appropriate offspring (by eugenic regulation, and for Xenophon by the regimen observed by the women) and then in educating them at each stage. Because the makes this its aim, it does not leave reproduction or education to the whims of private individuals, as ordinary cities do, but strictly regulates them and indeed carries out much of the process publicly and collectively. Both Xenophon and Plato eugenically regulate the production of offspring (Plato abolishing permanent marriages for the guardians and having the state fix temporary marriages for eugenic reasons, Xenophon accepting traditional marriage but allowing or requiring citizens to have sex with other citizens wives for eugenic reasons, I 6-9; both Plato and Xenophon are particularly concerned with regulating the ages of the parents), and both sharply limit the opportunities for sexual intercourse (Xenophon I 5). Both in Xenophon and in Plato the children of the elite are not left under the control of their parents, but live collectively and are subject to a common state education (cf. Xenophon II 1-2 on education governed not by a slave appointed by the parents, but by one of the highest magistrates, echoed Laws VI, 765d4-766b1, and cp. Rep. IV, 424bff on the guardians guarding over education much more than over contracts and lawsuits and the like). 34 Indeed, the elite citizens remain subject to a common discipline, presided over by elder persons of authority, not only in childhood but at every stage of their lives. Both for Xenophon and for Plato, the stratification of society into different age-classes, sharing much of their daily activities and subordinated to older groups, becomes more important than the division into different households (which Plato entirely abolishes); in a particularly bizarre parallel, both Xenophon and Plato encourage their young people to get into physical fights with each other (so they have to stay fit), while they must yield to their elders (Rep. V, 464e-465a, Xenophon IV 6). Both in Xenophon and in Plato, adult members of the elite remain under a strict military discipline and are required to eat in common messes. For Xenophon the common life and common discipline of each age-class of the citizenry applies primarily to the males, but he does stress that Lycurgus imposed at least a common gymnastic discipline for the women as

_________
34 Note also that Republic V, 459e3 speaks of the herd [ ] of guardians, using the word used by Ephorus (FGrHist #70 F149 = Strabo X iv 20) for the Cretan boys and by Plutarch (Lycurgus 16) also for the Spartan boys; cp. Republic V, 460c1-3 for the rearing pen [ ] established in a separate part of the city where the young are brought to be raised.

ON PLATOS

29

well (I 3-4); Plato, who explicitly criticizes the Spartans for not regulating the women citizens as much as the men (Laws VII, 805e7-806c7), brings both girls and boys, women and men, under the same discipline. For Xenophon, the common life of the Spartiates extends to common rights of use of each others property: they may use each others slaves, dogs and horses; may command and even beat each others children as if they were their own; and, as we have seen, may under some circumstances have sex with each others wives for eugenic reasons; in addition, the social effect of differences of wealth is minimized by requiring all to eat the same food in the common messes and by prohibiting the elite from owning gold or silver. Plato, of course, radicalizes this by entirely abolishing the (the household as family, and the private house with attached agricultural land and slaves) for the guardians, so that they will all say mine and thine of the same things, and of the same people. But even in Xenophon the boys in their groups and the men on military duty and in their common messes spend very little time on their private estates, and if the girls and women are subjected to the same discipline, the may have little remaining role. 35 (If we turn to Plutarchs Lycurgus we find yet other features in common with Plato: girls exercising naked like the boys; a musical and even intellectual component to the education, with much emphasis on songs of praise and blame; a strongly egalitarian picture of the life of the Spartiate citizens; rather desperate claims that the laws were originally intended for virtue in general rather than specifically for warfare, and were originally not harsh toward the helots [that came only after the revolt of 464!]. However, Plutarch is not necessarily an independent witness to the Laconizing tradition, since some of the literature he draws onperhaps especially the Stoic Sphaerus of the Spartans, written in the service of the revolutionary/restorative program of Cleomenes IIImay itself have been influenced by Platos Republic and Laws.) Plato is, obviously, well aware of these similarities between Callipolis and the Laconizers Sparta. In some cases he explicitly says that some practice is retained from the best when it degenerates into a timocracy. The timocracy, being in between, will imitate the prior [i.e., Platos ideal ] in some respects and the oligarchy in others, and will also have something peculiar to itself (Rep. VIII, 547d1-2): it will agree with the best in honoring the magistrates/rulers, and in that its fighting element abstains from farming and handicrafts and

_________
On the limited role of the Spartan , with the citizen males spending as much time as possible collectively in public space, see Cartledge 2001.
35

30

STEPHEN MENN

other moneymaking, and in establishing common messes in devotion to gymnastics and to training for war (547d4-8). That means that Plato has taken these features from what people say about Sparta, and has decided that he approves of them and wishes to preserve them in his ideal . On the other hand, the elite of the timocracy, who correspond to the guardians of Callipolis (no longer ruled by philosophers specially selected from among them) and are imagined as arising out of them, will fall short in their devotion to musical and then even to gymnastic training (546d68). They will no longer renounce private possessions, but distribute and privatize [ ] the land and the houses, and, enslaving those whom they previously guarded as free friends and nourishers [= the producers of Callipolis], keep them as perioeci and servants, and devote themselves to war and to guarding [their slaves] (547b8-c4). And the will be afraid to call wise people to the magistracies/positions of power ... and will incline to spirited [ ] and simpler people, those who are naturally disposed more to war than to peace; it will hold in honor the plots and deceptions of war and spend its whole lifetime at war; it will have most of these [features] peculiar to itself (547e1-548a3). That means that Plato notes that even the idealized Spartiates, despite how much they share and their renunciation of leisure and luxury, continue to own private houses and estates and an unfree workforce; that the Spartan is designed for war and values war as an end in itself and as a means to keeping the subject population enslaved rather than as a means to protecting the musical pursuits of peaceful leisure; and that it chooses and trains its leaders for warlike virtues rather than for musical or philosophical wisdom. And Plato has decided that these are bad features of the Spartan ideal, and must be changed to yield an acceptable Socratic . Finally, despite the official prohibition of most moneymaking activities and presumably of the possession of gold and silver, these people will be desirous [ ] of money/possessions, like those in oligarchies, and wildly honoring gold and silver under cover of darkness, since they possess storehouses and private treasuries where they can put [gold and silver] to hide them, and also possess enclosed dwellings, verily private nests, where they may expend them, lavishing much on women and on whoever else they want (548a5-b2); they are sparing of money, since they honor it and cannot acquire it openly, but on account of they are fond of spending other peoples [money], enjoying pleasures in secret, like boys running away from their father the law, since, having been educated not by persuasion but by violence, they have neglected what belongs to the true Muse who is associated with discourse and philosophy, and

ON PLATOS

31

have honored gymnastics prior to music (548b4-c2). These are features that timocracy shares with oligarchy, and show the private behavior of the Spartan elite, when not under public scrutiny, defecting from the official Spartan ideals and towards the behavior of rich ruling elites elsewhere. Here again Plato is drawing on what is commonly said about Sparta. Even as consistent a Laconizer as Xenophon, in of the Spartans XIV, says that the Spartans these days no longer observe the laws of Lycurgus, that previously the Spartans chose to live together at home with modest possessions rather than [as now] to be flattered and corrupted serving as harmosts in the [subject] cities, that previously they were afraid to be caught possessing gold, but now there are some who even pride themselves on having it (XIV 2-3), and so on. The difference is that, for Xenophon, this is an inexplicable decay from the days when the Spartans lived by Lycurgus laws, and explains why the other Greeks now resent the Spartans; this is supposed to support Xenophons general argument that Lycurgus laws were the cause of the Spartans success. 36 Plato, by contrast, traces back what is wrong with present-day corrupt Sparta to a deficiency in the original Lycurgan ideal, to the failure of its educational system to instill the right ideals (even Xenophon implicitly admits that the ideals had not been fully internalized, since he claims not that in the old days the Spartans were not desirous of gold or did not possess gold, but only that they were afraid to be caught with it; but Xenophon finds nothing wrong in the ideals themselves). Indeed, if the timocracy begins by

_________
On the comparison with the end of the Cyropaedia, and the contrast with Plato, see Dorion 2002. Xenophon is arguing not simply for the Spartan , but for the Lycurgan values of his patron Agesilaus, by contrast with the values of Lysander and perhaps also of Pausanias, whose he may well be answering. The polemics of chapter XIV against the corruption of the harmosts and the new influx of gold from the Spartan empire seem to be directed specifically against Lysander and his partisans. Xenophon stresses the obedience of the hereditary kings to the ephors, whereas Lysander wanted to make the kingship elective and Pausanias wanted to abolish the ephorate. Resp. Lac. XV stresses that the king and his contract with the city, represented especially by the ephors, have remained unchanged since the days of Lycurgus, and this seems to be implicitly polemical against anyone who maintains that any of these institutions are post-Lycurgan corruptions. (XV 1 [and its parallel Xenophon Agesilaus I 4] mean not, as is sometimes said, that only the Spartan kingship has remained unchanged and other Spartan magistracies have not, but rather that only the Spartan has remained unchanged and the of other cities have not; the contract between the king and the city, limiting the king by law, is supposed to be a key to that permanence.) Xenophon also stresses the importance of the in forming the Spartans character and their spontaneous obedience to their laws and magistrates, and this seems also to be an important part of his ideal of Spartan kingship, but, as Plutarch tells us, Spartans kings did not usually go through the , and Agesilaus was exceptional in having done so (Plutarch Agesilaus 1, see Cartledge 1987, and cp. Lipka 2002, 34).
36

32

STEPHEN MENN

enslaving its workforce, and then devotes itself to war for the sake of conquering and enslaving its neighbors and guarding those it has already enslaved (and indeed there is no firm line between the Spartans attitudes toward the Laconian helots and toward the Messenians), 37 then it is valuing collective , and it is no surprise if this leads to covert valuing of individual as well. The point is well made by Aristotle:
just as most people esteem despotic rule [i.e. rule as of a master over slaves] over many people because it brings a great supply of the goods of fortune, so Thibron seems to admire the legislator of the Spartansand all the others who write about their tooon the ground that they ruled over many through exercising themselves for danger .... [these writers] do not judge rightly about the kind of rule that the legislator should honor: for rule over free people is more noble and accompanied by more virtue than despotic rule; and one ought not to think a city happy, or praise its legislator, because he trained them to conquer so as to rule their neighbors, for these things involve great harm: for it is clear that any of the citizens who is able would also pursue this, how he might rule over his own city, which is what the Spartans accuse King Pausanias of, although he [already] had so much honor. (Pol. VII 14, 1333b16-35) 38

It follows that the Spartan is not a Socratic , and is all too close to the Thrasymachean or rather , oligarchy and democracy and tyranny. The timocracy is not a pure Thrasymachean , because it is not ruled simply for the economic advantage of its rulers, who willingly submit to a harsh discipline, and it is important for Socrates argument against Thrasymachus to show that this is possible. But the timocracy remains a , because the Spartans rule despotically over the perioeci and especially the helots, that is, ruling them as a master rules slaves and not as a citizen magistrate rules free fellowcitizens, thus not in the interests of the ruled. Plato had said earlier that every city other than his ideal is really two mutually warring cities of the

_________
not Pausanias the king of the 390s (who is the person I have called simply Pausanias elsewhere in this essay); Jacoby FGrHist #582 T2 actually brackets king as an interpolation. Compare Pol.VII 15, 1334a40-b4: the Spartans practice virtue, not because they do not agree with people elsewhere that external things are the greatest of goods, but because they think that these are best acquired through some kind of virtue. Isocrates consistently associates the Spartans with (Busiris 20, Plataicus 19-20, Philippus 147-8, Panathenaicus 241 and 243 [this last text praising Spartan ]; besides texts not using or its cognates, such as Panathenaicus 188, the Spartans look to nothing but how they can seize as much as possible of other peoples property), and this was probably just as much a commonplace of the fourth century as praise of Spartan virtue.
37 On the difficulty of distinguishing Messenians from helots see Figueira 1999. 38 Despite the word king, this must be Pausanias the regent of the 470s,

ON PLATOS

33

rich and the poor (Rep. IV 422e3-423a5), 39 and the clearest illustration would be Sparta, where the helots and perioeci are denied Spartan citizenship (the perioeci belong to their own cities, Xenophon Resp. Lac. XV 3), and (according to Aristotles of the Spartans, Plutarch Lycurgus 28 = Fr. 538 Rose) each years ephors formally declare war on the helots, so that they can lawfully be killed. The Republic is on this point more radical in its critique of Sparta than the Laws, which says that the Spartan is a true and not a . The reason for the divergence is that the Laws is considering only the Spartiates rule over fellow-Spartiates, regarding their rule over the perioeci and helots as part of their foreign relations rather than their internal , presumably on the grounds that these people are defeated Messenians, or aboriginal inhabitants of the Peloponnesus from before the return of the Heraclids (or Dorian invasion). 40 The Republic, by contrast, regards the perioeci and helots (not really distinguishing them) as enslaved members of the original Spartan community, 41 so that the oppression of the helots is a graver charge from the standpoint of the Republic than of the Laws. And this difference is because, while the Laws own ideal community would contain only citizens corresponding roughly to the Spartiates or to the guardians of the Republic (they can farm but not engage in handicrafts or trade, and they will probably not do much manual labor on their farms), the Republic wants to show the possibility of an entirely self-sufficient community, including producers as well as soldiers and rulers, which is ruled for the benefit of all its members, and so it is important for the Republic to bring out the Spartan deviation from this ideal. However, if Plato wishes to reform the Spartan so that it will not be devoted to or to warfare for its own sake, and will rule in the interests of the ruled, he will need more than narrowly political

_________
39 Since several readers have told me that Plato says this only about the oligarchic city, let me stress that this is not true: Rep. VIII, 551d5-7 says that the oligarchic city is two conflicting cities, but Rep. IV, 422e3-423a5 says this about every city other than the ideal. 40 See the (often bizarre) analysis of Peloponnesian history at Laws III, 683c8-693c5, where apparently it was all the Messenians own fault, because they corrupted their original and laws as they were established after the Dorian invasion, and broke their alliance with the Spartans. 41 However, it is possible that when Plato speaks of the guardians enslaving those whom they previously guarded as free friends and nourishers and keeping them as perioeci and servants (Rep. VIII, 547c1-3), he is thinking not of the Spartan perioeci but of a Cretan group corresponding to the Spartan helots (Aristotle Pol. II 10, 1271b40-1272a1), who might more appropriately be described as enslaved. It is nonetheless true that the Spartan perioeci had no civil rights at Sparta, could be put to death without trial, and so on.

34

STEPHEN MENN

means. Plato wants to show that a Socratic is possible, and this means showing that it is psychologically possible to get people to act in the desired way. This is already a concern of the Laconizing literature: the Spartans (allegedly) behave in ways very different from everyone else, in ways we might have thought psychologically impossible, and so we try to explain how this comes about by stressing the power of the common state education to shape peoples character into a new mold. Plato wishes to keep state control of education but change the content of that education. This is because he, like the Laconizers, thinks that the is founded on the of individual citizens, that is, on their character as formed by their education: perhaps the greatest insight of the Spartans is that the highest goal of the is to form the character of the citizens, although unfortunately it is not the right character-type that they are molding their citizens into. So Plato is concerned with different character-types as well as, and in parallel to, the different types of . This works at two levels: he is concerned with the psychology of the different members of each , especially the ruling members who determine the collective decisions; he is also interested in the psychology of the individual who admires each type of , even if he does not himself live in a of that type, and uses it to guide his actions where he does live. In the terms of the Republic, the timocracy, and its ruling members, are governed by , while the oligarchy, democracy and tyranny and their rulers are governed by ; and this is a psychological explanation of why the timocracy is not a merely Thrasymachean , since is the force in the soul that leads to action in pursuit of the ideals one has heard praised as noble, and if necessary suppression of action to satisfy bodily needs. Plato is not here simply imposing his own independently developed psychology in an attempt to explain the Spartans. On the contrary, in his account of and (though not in his account of their relation to reason), Plato is elaborating a psychology that is at least implicit in the Laconizing texts themselves. Xenophon starts his of the Spartans by speaking of Lycurgus laws as making his city preeminent in happiness [ ], and he emphasizes that happiness came to the Spartans by obedience to the law, that is, by virtue: not simply because obedience to law is a virtue (a point stressed in the Memorabilia) but because the laws of the Spartans in particular are designed to foster all the virtues. Since [Lycurgus] recognized that those who choose to devote themselves to virtue individually are not sufficient to make the nation great, he compelled all [the Spartiates] to practice all the virtues publicly: for as individuals who train [in virtue] surpass in vir-

ON PLATOS

35

tue individuals who neglect it, so naturally Sparta surpasses all cities in virtue, since it alone publicly practices excellence [ ], the laws imposing penalties on anyone who neglects to be as good as possible, and compelling all to practice every political virtue (X 4-7). Any Spartan who is virtuous in this way will also be happy: Lycurgus conspicuously held out happiness [as a reward] for the good, and unhappiness for the bad (IX 3). That sounds Platonic, or Socratic, enough, but virtue produces happiness not naturally by perfecting the soul, but artificially, because the legislator has contrived that public honor will attend virtue, and especially that public censures of all kinds will attend vice. By making honor correspond to virtue, Lycurgus brought it about in the city that a noble [ ] death is more choiceworthy than a base [ ] life .... when such dishonor [ ] is attached to the bad, I do not at all wonder that death is there preferred to such a dishonored and shamed [ ] life (IX 1 and IX 6). Xenophon does not seem to notice any conceptual gap between being noble or honorable and being in fact honored, between being base or shameful and being in fact shamed. The virtuous are those who prefer the noble, or, it seems equivalently, who prefer being praised, to pleasure or wealth or long life: having made it noble [ ] to steal as many cheeses as possible from [the altar of Artemis] Orthia, he commanded others to whip these, wishing in this too to show that it is possible by suffering pain for a short time to delight in good fame [ ] for a long time (II 9). Now the conflict between a desire for pleasure and a desire for honor or the honorable is by no means restricted to Laconizing literature (we find it for instance in Prodicus Choice of Heracles in Xenophon Mem. II i 21-34, and in Heraclitus B29). But it is particularly important there, because the Spartans as the Laconizers imagine them are the theoretical extreme of the triumph of honor-values over pleasure-values, and the Laconizing literature tries to explain this triumph by giving a psychological account of the kind of education that would produce it. In setting out his tripartite psychology, in Republic VIII and earlier in Republic IV, Plato seems much more concerned to correct this simple bipartite psychology than he is to correct what is sometimes described as the Socratic unitary psychology. Being a Socratic did not stop anyone from positing plural sources of motivation within the soul; it did not stop Xenophon from going so far as to posit two souls within each of us, a good soul pursuing noble works and a bad soul pursuing base works (Cyropaedia VI i 41), and in several dialogues before the Republic Plato had experimented

36

STEPHEN MENN

with a contrast between rational and non-rational motivation, 42 but none of this had led him to tripartition. Rather than describing Plato in the Republic as correcting Socratic psychology (which could not explain why the new psychology is tripartite, but at most why it is non-unitary), I would prefer to describe him as Socratizing the psychology of the Laconizers, as the Republic as a whole Socratizes the Laconizing -literature both in form and in content. Republic VIII describes the elite of the timocracy as enjoying pleasures in secret, like boys running away from their father the law, since, having been educated not by persuasion but by violence, they have neglected what belongs to the true Muse who is associated with discourse and philosophy, and have honored gymnastics prior to music (548b6-c2), and says of the timocratic person that such a person when young would despise money [or possessions], but as he becomes older he would embrace it through sharing in the money-loving nature and not being pure as to virtue, since he has abandoned the best guardian, namely discourse blended with music, which alone, when it has come to be in someone, will dwell within its possessor as a savior/preserver of virtue throughout life (549a9-b7). That is to say: -motivation may be sufficient to produce political courage (as Plato calls it, Rep. IV, 430c2-4), the virtue for which the Spartans are most famous, but it is not sufficient to preserve an individual or a city from degenerating into or (as Plato equally stresses) into arbitrary aggression against subordinates or neighbors, the vices for which the Spartans are most notorious. To preserve virtue reliably in an individual or a city, must be controlled by reason: not by a reason which aims merely at maximizing long-term satisfaction of or , but by a reason that has a desire of its own, for philosophical contemplation or more generally for peaceful musical

_________
At Charmides 167e1-5 the object of is pleasure and the object of is the good. At Protagoras 356c4-e4 there is a contrast between two motivating powers, the measuring art and the power of appearance; Rep. X, 602c7-603a8 picks up this passage, developing it more fully, but with no sign of rejecting the Protagoras. Gorgias 467c5-468c8 argues that is always of the good (the real good, not the apparent good); but 464b2-465e1 contrasts arts aiming at the best with pseudo-arts aiming at what is pleasant, and if the body were not governed by a soul which can distinguish the arts from their imitators, but rather the body itself judged/distinguished, measuring by the gratification [ ] for it, chaos would result; at 493a3-b3 the part of the soul where the are, and which can be persuaded in contrary directions, is like a leaky jar. At Phaedo 94b4-e6 the soul contradicts and overrules bodily affections such as hunger and thirst, and converses with and passions and fears as one thing speaking to another, as when Odysseus commands his heart to endure. I do not mean to suggest that these passages put forward a consistent theory, for instance on how far is due to the soul or to the body.
42

ON PLATOS

37

pursuits, so that aggression does not degenerate into an end in itself or into a means to . On the other hand, while Plato is saying against the Laconizers that -motivation needs to be controlled by philosophical reason, he is also saying that -motivation can be controlled by reason in a way that -motivation cannot, so that characters and governed by need only an extra layer of rational control to make them Socratic, while oligarchy and democracy and tyranny and the corresponding characters governed by would require more radical transformations. As Republic IV puts it, in the civil conflict within the soul, [ ] bears arms for the rational part, should conflict break out between reason and (440e1-6, cp. 440a8-b7); or, more cautiously, the is an auxiliary by nature to the rational part, unless [the ] is corrupted by a bad upbringing (441a2-3). The Spartan and Spartan education, as the Laconizers imagine them, are an amazingly effective machinery for developing and harnessing the power of to control ; unfortunately, the machinery is not being used for the philosophically correct purpose, but if the machinery could be captured by philosophical reason, the best would result. This plan of exploiting the power of to bring under the control of reason does not seem to occur to Plato anywhere before the Republic: the Phaedo contrasts the person who abstains from bodily from philosophical virtue with the politically virtuous people who abstain fearing bankruptcy and poverty, like the many, the lovers of money ... [or] fearing dishonor and the reputation of wickedness, like the lovers of rule and honor (82c5-8, cp. 82c2-5 and 82a11-b3), as if there was no value-difference between what the Republic will call the timocratic and oligarchic characters; reason in the Phaedo seems to dominate all the passions, when it does so, equally directly. We may say that Platos new understanding of the role of comes out of critical reflection on the Laconizing ideology, or, conversely, that it was this new understanding of that led him to reflect on the Spartan and on how it could be improved. 43 From Republic VIIIs criticisms of the Spartan and of the character of the elite it produces, we could infer, even if we did not have the

_________
43 Note that if Platos starting-point in introducing tripartition were in reflection on individual moral psychology rather than on politics, he would more plausibly have proceeded by first distinguishing rational from irrational sources of motivation, and only then subdividing the irrational soul. Instead he starts by distinguishing the producers from guardians, and from , and only then separates out the philosophers and reason from the auxiliaries and .

38

STEPHEN MENN

earlier books of the Republic, how Plato thought the Spartan ideal would need to be emended to turn it into a Socratic . First, the Spartan education does well at making their military guardian class and fierce towards enemies, but not at making them gentle toward their own people, as they must be if they are truly to guard them, and not be like sheep-dogs who attack their own sheep (Rep. II, 375b9-d1, and III, 416a27, without explicit reference to Sparta); so we must take great care over the education of the guardians so that they do not turn from benevolent allies into harsh masters [ ] of the people (III, 416b1-6). In Republic III the solution seems to be that their education must balance gymnastics, which develops the body but also the part of the soul and makes people brave but risks making them excessively harsh, with musical education, which develops the philosophical or wisdom-loving part of the soul and makes people temperate but risks making them excessively soft (410a7-412a7). 44 It is especially the ruler or overseer of the city who will have to be formed in this way, if the is going to be saved/preserved (412a9-10), and it is such a ruler who is most truly a guardian, since he will not only guard the city militarily against threats from without, but will also be the guardian of the against any changes in the practices of its citizens which could destabilize it from within (414b1-6). For this reason the city must be ruled, not simply by the military class, but by suitably formed characters carefully selected from among them. This is not in itself any criticism of Sparta, where, as in other Greek cities, there are specially selected rulers/magistrates (holding office for life, like the kings and gerontes, or for a year, like the ephors and nauarchs), and indeed the Laconizing literature stresses the prompt obedience of all Spartiates (even the kings) to those in authority over them. So it is too simple to say that Plato replaces a Spartan duality of unfree producers and armsbearing citizens with a triple division of producers, auxiliaries and rulers; Plato and the Laconizers equally accept a duality of producers and military

_________
44 Compare the criticisms of Sparta in Laws I-II: the Spartan legislator instituted many practices to develop courage, but what did he institute to develop temperance? The Spartans are, in a way, temperate, since they forgo many pleasures in adhering to their military or quasi-military discipline, and since in public none are allowed to consume more conspicuously than others; but when they are freed from this discipline or are in private, since they have not really moderated their appetites but have merely overwhelmed them with motivations, they indulge themselves without limit. This is the point of departure for the odd-sounding criticism of the Spartans for not allowing drinking or symposia; but presumably the symposium is not simply an occasion for moderation in drinking, but also a vehicle of the musical education that the Spartans are missing.

ON PLATOS

39

guardians, with rulers/magistrates selected out of the military class. 45 The question, however, is how they are to be selected, and in particular what special education is needed to make a good ruler, beyond the common education imposed on the whole military class. 46 Already in Republic III Plato stresses that the ruler, even more than other guardians, needs not only gymnastic but also musical education, or as Plato also says philosophy (so 411c5); in later books, of course, he specifies the content of this philosophy, saying that in order to preserve the the ruler must know eternal paradigms and especially the good for the sake of which everything else is done, and specifying the curriculum, mathematics followed by dialectic, that is necessary to bring the rulers to this knowledge. (We might thus reverse Dicaearchus dictum cited above, and say that Plato corrected Lycurgus by mixing him with Pythagoras [mathematics] and Socrates [dialectic].) Plato is not simply adding philosopher-rulers on the Socratic grounds that the ruler must have knowledge of the good, that stable right action requires knowledge and not mere true opinion, that the many can be guaranteed to have true opinion only if they follow someone who has knowledge. Plato is also adding music and philosophy to the Spartan to ensure that the rulers and fighters have something better to do than ruling and fighting, philosophical contemplation in the strict sense for the rulers and musical or cultural pursuits for the others; only in this way can they be trusted not to make ruling or fighting ends in themselves. Finally, to ensure that the rulers and the military class, whose power the producers are unable to check, will not use this power for but will rule in the interest of the ruled, Plato finds it necessary to abolish the and private property for the guardians, to undo the crucial step

_________
45 The Timaeus summary of the Republic, 17c1-19b2, reports a bipartition of society into producers and guardians, rather than a tripartition, suggesting that Plato sees this bipartition as the more basic division of the society. 46 Here Plato is picking up a theme of the so-called or mirrors for princes literature, describing how the ideal king will act and how he must first be trained: examples would be Xenophons Cyropaedia, Isocrates Cyprian orations, and the lost Cyrus or by Antisthenes (whether these were written before or after the Republic). There are also later lost works by Aristotle, Theophrastus, Strato, Cleanthes, Persaeus, Sphaerus and the Megarian Euphantus, and a separate On the Education of a King by Theophrastus, all cited by Diogenes Laertius in the corresponding lives, and much later extant texts, notably by Dio Chrysostom. These texts would be related to texts, but differ in being at least sometimes dedicated to a king, and in assuming unlimited monarchy rather than discussing the merits of different . For a brief but helpful discussion of Platos Republic and the literature see Ferrari in Ferrari and Griffith 2000, xviii-xx.

40

STEPHEN MENN

that turned the Callipolis into the timocracy, namely the guardians division of the land and crops and animals and human beings of the city among themselves as their own property. As Plato says at the end of Republic III, immediately after giving the standard Laconizing rules that the guardians must eat in common messes and must not possess gold or silver, in this way they would be saved/preserved and would save/preserve the city; but when they acquire their own [ ] land and houses and currency, they will be household managers [ ] and farmers instead of guardians, and will become hostile masters [ ] instead of allies of the other citizens, and they will spend their whole lives hating and being hated and plotting and being plotted against, fearing more enemies from within than from without, and fearing them more: and thus both they and the rest of the city have already come very close to destruction (417a5b6). The Spartiates do of course have their own land and houses (and at least bronze currency, with the temptation to gold and silver), and Plato is saying that this leads them to a conflict of interest with their assigned task of guarding the city and the producing class. Farming sounds innocuous, but the Spartiates are not plowing their land themselves (and if they did they would not have time for their civic-military duties); rather, they are supervising a landed estate and the workers who are bound to that land; they are trying to extract a surplus of produce beyond what they must grant for the survival and reproduction of their workforce, and their interests will conflict with those of their workers and also with the larger interests of the city. The tension that Plato evokes here between and , between managing ones estate, exercising arbitrary authority over slaves in ones own private interest, and participating in the governance of the city, exercising authority within the law over free and equal citizens in the common interest of rulers and ruled, is a Greek commonplace, as is the criticism of those who retire to preside over their estate instead of devoting themselves to the common good of the city. 47 The Laconizers especially

_________
47 For the differences between and see especially Aristotle Pol. I. Xenophon in Oikonomikos 13 defends the deliberately paradoxical thesis that the rule of a master or a slave steward over slaves requires the same skill as the rule of a statesman or a king over free citizens; the Eleatic Stranger in the Statesman starts by defending this thesis too (258e8-259c4), but then apparently repents of it after his recantation at 274e1-4; Pol. I starts by citing, and setting out to refute, this shared thesis. is often identified with the art of moneymaking [ ]; Aristotle, while mostly trying to distinguish from the more noble , also argues that the best and truest kind of is something better than . For criticism of people retiring to their estates rather than taking part in democratic politics, see Carter 1986.

ON PLATOS

41

stress the priority of the Spartiates duty to the city, and their freedom from conflicting private duties and interests which could get in the way of . But, says Plato, the continuing institution of the contradicts this. As we saw in Republic VIII, the founding act of the timocracy is when the guardians distribute and privatize [ ] the land and the houses, and, enslaving those whom they previously guarded as free friends and nourishers, keep them as perioeci and servants, and devote themselves to war and to guarding [their slaves] (547b8-c4): while the (presumably equal and inalienable) division of land in the Spartan is usually praised, 48 for Plato it represents the violent dispossession and enslavement of the producers, and the corruption of the guardians, who are now landowners with an economic self-interest instead of pure Socratic rulers ruling in the interest of the ruled. And the private is the place where the Spartiates can indulge in consumption immune to public scrutiny and to Spartan discipline: Plato speaks of enclosed dwellings, verily private nests, where they may expend [gold and silver], lavishing much on women and on whoever else they want (548a9-b2). The Spartan women are here especially singled out because they remain on the estate and are not subject to the public discipline and communal living of Spartan boys and men; indeed, they are likely to wind up managing the estate when the men are away. Plato in Laws VII criticizes the Spartans for not subjecting the women to the same discipline as the men: he admits that the Spartans do better than other Greeks in making their girls take part in athletics, but whoever wishes to praise your legislators [= Lycurgus and Minos] for these things, let him praise, but I will not speak otherwise: the legislator must be complete and not diminished by half, but if he allows the females to enjoy luxuries and to spend in a

_________
48 The argument of Hodkinson 2000, 19-64, that Spartan equality in land-ownership was a myth invented by the revolutionaries around Cleomenes III in the late third century, is stimulating and useful but exaggerated. I agree that it was a myth. Hodkinson is right that Aristotles critique of Sparta in Pol. II 9 makes clear that the Spartan landed allotment was not equal and inalienable and indivisible; it may have been close to inalienable within the owners lifetime (it is striking that Aristotle does not talk about people losing their land through debt), but it could be divided or combined with other property in inheritance, so that if the division of land among the Spartiates had ever been equal, it had ceased to be so by any time for which we have real evidence. However, the myth is earlier than Cleomenes: it is clearly there at Laws III, 684d4-685a4, despite Hodkinsons attempts at denial (32 with 61n17), and it does not look like an invention of Platos, though certainly egalitarian aspects of the Spartan legend (like other aspects) grow in the telling. Polybius also speaks of the Spartan equality of landholding, VI xlv 3 and xlviii 3, and Hodkinsons claim, 51, that Polybius was fooled into this by the Spartan revolutionaries (whom he detested) is implausible.

42

STEPHEN MENN

disordered regimen, and takes care only of the males, he has almost completely neglected half of a happy life for the city, instead of [taking care for] twice that (806c1-7; Aristotle will develop this criticism of the Spartans in Politics II 9, and sees the non-discipline of the women as a main cause leading the Spartans to honor wealth). By subjecting the girls and women to the same discipline and same communal living as the boys and men, Plato aims to bring -values into submission to values; and if there is neither desire nor leisure for either guardian men or women to consume in private on the , the elimination of the guardian altogether, and the entrusting of the oversight of the producers and their land to the state, will be an easy next step. IV. The generic background of Laconizing -literature helps us in interpreting the Republic by allowing us to see what expectations Plato assumes in his readers as he constructs his best . Plato partly confirms those expectations, sketching a with many familiar Laconizing features; but he partly defeats the expectations, pointedly rejecting other familiar features of the Spartan . His starts by sounding Spartan enoughparticularly with the separation of a specialist military class who are barred from money-making pursuits, while the others are barred from military activityand it gradually diverges; it may be some time before the Laconizing reader realizes that something has gone wrong. Where Plato does criticize the Spartan model, his criticisms are typically internal, as they are in the Laws: that is, he starts with some value which the Laconizing reader can be assumed to share (e.g., the rejection of , the importance of an all-encompassing discipline of virtue), and shows that some correction to Spartan institutions is needed to fully realize that value. (Some of these internal criticisms will only work if the reader assumes that the city which the guardians are serving comprises the producers too as citizens and not only the guardians; but this premiss is secured by the way Plato derives the guardians and the other groups in the city he is constructing, as different specializations mutually dependent and all needed for the good of the whole.) Without this background we would be liable to what Schleiermacher calls quantitative

ON PLATOS

43

misunderstanding, that is, to missing where the emphasis is supposed to fall. 49 But in discerning the expectations of Platos intended audience, we are not entirely dependent on what we can reconstruct (based especially on Xenophon) of the history of the -genre. Plato directly shows us the character of his intended readers in the persons of Glaucon and Adeimantus, who are stand-ins for the reader, as Clinias and Megillus are in the Laws (noted above). Glaucon and Adeimantus are not given strongly contrasting characterizations, and take turns functioning cooperatively as Socrates interlocutor for a single developing argument, but it is clear that Glaucon is the dominant personality (it is Glaucon who first challenges Socrates to give a more persuasive account of justice and not rest content with defeating Thrasymachus, and Socrates then comments that Glaucon is always most courageous in all encounters [II, 357a2-3]; Glaucon and not Adeimantus walks down to the Piraeus with Socrates at the beginning of the dialogue; Adeimantus is three times described as Glaucons brother [II, 327c2, 362d2, 376d4], never vice versa; the poet who praises them both at II, 368a4, is Glaucons lover; also Xenophon mentions Glaucon but not Adeimantus as a companion of Socrates, and Diogenes Laertius attributes to Glaucon a series of Socratic dialogues including a Cephalus!). Glaucon is the interlocutor for the philosophical and political high-points of the dialogue, while Adeimantus seems more concerned with culture and religion. And Plato expressly describes Glaucons character: when Socrates asks what the person corresponding to the timocratic will be like, Adeimantus immediately volunteers that hell be like Glaucon; Socrates replies that while theyll be alike in their love of victory, the timocrat will be less given to music and discourse, and therefore more inclined to be savage to slaves, and more likely to fall into a love of money (VIII, 548d6-549b10). Glaucon is thus like the timocrat but somewhat better: he is able to see the faults that Socrates points out in the timocratic city and the timocratic character, and can be persuaded to look to a better, philosophical, collective and individual way of life. The timocrat is the Laconizer. That is, when Plato describes types of soul as corresponding to types of , the individual corresponds, not

_________
49 More particularly, being democrats ourselves, we are naturally inclined to put the emphasis on Platos criticisms of democracy. But Plato has no hope of persuading the democrats, and his criticisms of democracy are mostly conventional sarcasms; his emphasis is rather on persuading the people closest to him, the Laconizers, whose ideal has very little pull on us now. (So, rightly, Tigerstedt I 274-5.)

44

STEPHEN MENN

necessarily to the he lives in (each will contain many types of individuals, although its ruling group will typically be of the corresponding type), but to the that he admires and would prefer to live in as one of its rulers, believing that these rulers are happy. 50 Thus Thrasymachus is clearly portrayed as a tyrannical person, and thinks that the tyrant is happiest (I, 344a3-c4); the philosopher will think that the philosopher-rulers of Callipolis are the happiest; and the timocrat will think that the (idealized) Spartiates are the happiest. Plato does not expect actual Spartan readers and does not care about actual Spartans (and he surely believes no more than Xenophon did that the Lycurgan ideal was realized in the Sparta of his own time); he is writing in the first instance for the Laconizers at Athens and elsewhere, and he wants to persuade them to admire something different from the Spartan ideal. After Glaucon and Adeimantus present their initial challenge to Socrates at the beginning of Book II, they for the most part allow themselves to be persuaded by Socrates construction of the ideal city; but they intervene with objections at four crucial junctures. Glaucon successfully objects to the minimalist city of pigs (II, 372c2-e1), requiring civilized luxuries and thus leading to further specialization and expansion, demands on neighboring territories, and thus the introduction of a Spartan-style military class, which motivates all the rest of the construction. Glaucon and Adeimantus accept that construction until the end of Book III, where Socrates, radicalizing the Spartan model, proposes the abolition of private property for the guardians: Glaucon accepts this, but Adeimantus, at the beginning of Book IV, objects that these men ... whose city it really is [N.B. something that Socrates has not said], enjoy none of the goods of the city, and that Socrates is making the rest of the population happy at the expense of the guardians (419a1-420a1). This prompts Socrates to defend the principle that we must construct the city (and the guardians must rule it) for the sake of the whole citys happiness rather than just the guardians, and also to try to bring his interlocutors to see the guardians of the Callipolis as truly happy. Likewise, the brothers (on behalf of the whole audience) protest at Socrates two further radicalizations of the Spartan model, the abolition of the family at the end of Book IV and beginning of Book V, and the introduction

_________
50 For the idea that persons of different character would also prefer to live in different kinds of city, compare the pseudo-Xenophon: the can be forgiven for wanting to live in a democratic city, but whoever, not being of the , chooses to live in a democratic city rather than in an oligarchic one, has prepared to do injustice and has recognised that it is easier to escape notice when one is bad in a democratic city than in an oligarchic one (II 20).

ON PLATOS

45

of philosopher-kings at V, 473-4, prompting Socrates elaboration and defense of each of these radicalizations, which take up all of Books V-VII. Glaucon promises that he will be a particularly helpful and receptive interlocutor for this argument (474a6-b2, and cp. 450b6-d7), and he is indeed persuaded. What one admires, and what citys rulers one believes to be happy, make a difference for ones own : even if one lives in some other city, one can live looking to the of some other . The democratic city in particular, where people of all varieties would most of all arise, contains all kinds of on account of its liberty, so that anyone who wishes to construct a city need only come to the democratic city and select whichever he likes, as if coming to the bazaar of , and establish/colonize [ ] the one he has selected; he would not lack for (VIII, 557c1-2, d4e1; likewise, the of the democratic man contains the most of and , 561e6-7). So we must see democratic Athens as filled, not only with democratic people admiring the democratic , but also with many dissident types each admiring, and trying to live as if they inhabited, some other type of . Thus the Athenian Laconizers will look to their idealized Sparta, the oligarchic people in the democratic city (explicitly described 565b2-c4) will look to oligarchy, and tyrannical people like Thrasymachus will look to tyranny as their . 51 And we must see a great debate at Athens, particularly among those dissatisfied with the democracy, about what counterone should look toward. Just as it was Thrasymachus and not Socrates who introduced into the discussion, so it was Thrasymachus who introduced (though not using that word), when he says that to see that injustice is more advantageous than justice, we must look not to a petty thief but to the most perfect or complete [ ] injustice, namely tyranny, to see that the most unjust person is also the happiest (I, 343e7-344c4). 52 This gives Socrates the task of

_________
51 An anonymous referee points out to me that there is a contradiction between the nobler things that the timocratic person admires, and the baser things that he has a tendency to pursue. This is true, but the same contradiction exists in the rulers of the timocratic city. The timocratic man professes to admire the Spartan devotion to civic virtue, but, as Plato and Aristotle try to bring out, this is mixed with a less open admiration for Spartan as well. 52 So too Glaucon at II, 360e1-361d3: to compare and decide whether the just or the unjust is happier, we must first posit perfect or complete and extreme versions of each, the cunning and powerful unjust person who appears just, and the just person who appears unjust and suffers the consequences. This is what sets Socrates task.

46

STEPHEN MENN

constructing a counterof a city and an individual, which will be both most just and happiest, and will allow us to see that justice is more advantageous than injustice: this will be shown to be more just and happier than the Laconizing (which is conceded to be more just and happier than the democratic ), and it will also be shown that the tyrannical , which Thrasymachus agrees to be the most unjust, is also the most miserable (more precisely, the tyrannical person is more miserable than any other character-type, and the tyrannical person who succeeds in acquiring a tyranny is even more so, IX, 578b4c3). Socrates gives this comparative evaluation of individual and collective an eschatological significance when he imagines that after death we will be given a choice among all the of (X, 618a1), and that those who have correctly studied what circumstances of make a soul most just and therefore happiest are the people who will be happy in the next life (618b6-619b1; note that someone who has lived in the previous life in an ordered , participating in virtue by habituation without philosophy is likely to choose wrong in the next life, 619b7-d3). Now while we might be tempted to connect this talk of with the Platonic theory of Forms, in fact it does not seem to be peculiarly Platonic. Xenophon says that if plumbline and straightedge [ , ] are a noble [ ] invention for human beings for producing good works, the virtue of Agesilaus seems to me to be a noble for those who wish to practice excellence [ ] (Agesilaus X 2); he goes on to speak of imitating Agesilaus and suggests that Agesilaus main task as king was not military leadership but leading his fellowcitizens to virtue, presumably by offering himself as . Presumably Xenophon also sees Lycurgan Sparta as a collective when he says that everyone praises such practices, but no city is willing to imitate them (Resp. Lac. X 8); Isocrates too speaks of , sometimes meaning merely an example in an inductive argument, but sometimes a model for admiration and imitation (Nicocles virtue or will be a for his fellow-citizens, Ad Nicoclem 31 and Nicocles 37; ancient Athens provided a of laws and to the other Greeks, Panegyricus 39). Different Athenians will admire different individual and collective , and this will make a difference to the kind of political action, or inaction, that they pursue at Athens. Socrates tells Callicles that just as to succeed politically (to avoid suffering injustice) under a tyranny, one must make oneself a friend to the tyrant by making oneself similar to him in character, so to succeed politically in the Athenian democracy one must make oneself as similar as possible to

ON PLATOS

47

the , that is, to the (Gorgias 510a6-511a3, 512e4-513c3). Since Callicles thinks himself superior to the and would not want to become like it in character, this is supposed to dissuade himand other Athenian dissidentsfrom pursuing a political career at Athens. Some may try it anyway, thinking that they can avoid the consequences; others may retreat from politics (like the father of the timocratic youth at Republic VIII, 549c, or the uncorrupted philosophical natures, and Socrates prevented by his , at VI, 496a-e) because they realize that if they intervened in politics without lowering themselves to the character of the established they would be destroyed before they could accomplish anything; yet others may attempt clandestine action to subvert the established (like the oligarchic people whose attempt to resist their dispossession by subverting the democracy leads disastrously to and the rise of tyranny at VIII, 565b2-566d3). By contrast with other dissidents admiring other , the philosophical person, who looks to the within himself, and guards lest anything overturn it by way of wealth or honors, will not be willing to practice politics [ ] (IX, 591e1-5). Or so Glaucon says, but Socrates corrects him to say that in his own city [he will practice politics], but not in his fatherland, unless some divine good fortune occurs (592a7-9): his own city is the city in which Socrates and his interlocutors have been describing, which may exist nowhere on earth (592a10-b1), but of which a is perhaps stored up in heaven for one who wishes to see it, and, having seen it, to settle himself there; it makes no difference whether it is or will be anywhere, he would practice [the politics] of this city alone and of no other (592b2-5). Presumably the way to practice the politics of the Callipolis while living in Athens is to do what Socrates is doing when he says in the Gorgias that he, and he alone of the Athenians of his time, truly practices politics (521d6-8): namely, to pursue philosophical inquiry and ask dialectical questions, and especially to cross-examine people like Callicles and Alcibiades who are planning political careers, and persuade them to pursue philosophy first or instead. The political, or anti-political, consequences of Platos critique of previous have particular significance in Platos immediate political context; and this too is reflected in the dramatis personae of the Republic. Many people in Platos immediate circle of family and friends, disaffected with the democracy, would have looked to the Spartan ; most obviously his cousin Critias (whose talk about Sparta was described above). Plato speaks of Critias, without naming him, in the Seventh Letter:

48

STEPHEN MENN

some of [the Thirty] were relatives and familiars of mine, and they straightway exhorted me [to sharing in their political undertaking] as if it were my duty. And, given my youth, what I experienced was nothing surprising: I thought they would govern the city so as to lead it from an unjust to a just , 53 and so I paid close attention to see what they would do. And [I saw] these men in a short time reveal the previous [democratic] as a thing golden [in comparison]. (Epist. VII, 325d1-8)

So Plato withdrew from politics; when the democracy was restored, and the democrats proved themselves surprisingly mild toward those who had fought for the oligarchy, Plato again felt drawn (less intensely) toward politics, but after the execution of Socrates he again withdrew, concluding that just and effective political action was impossible, that it is not possible to act without friends and faithful/reliable (325d1-2), that no such good people could be found in so corrupt a city and that it would not be easy to produce new ones; until finally I was compelled to say, in praise of right philosophy, that from it [alone] would it be possible to discern all that is politically and individually just (326a5-7), and that no good would come of the human race until philosophers became rulers or vice versa. The turn to philosophy thus appears reluctant, and as a result of the exhaustion of available political . Plato finds the democracy irredeemably corrupt, but the disastrous rule of the Thirty, beginning with noble political talk, had utterly discredited the cause of oligarchy and of the Laconizers; and Plato and Xenophon and many others, who had stayed in the city and in all likelihood fought for the oligarchy against the men from the Piraeus, find themselves suspect perhaps as soon as they engage in politics, certainly as soon as they say anything critical of democracy. Both for Plato and for Xenophon, Socrates represents a personal , an alternative to both democratic and oligarchic politicians: although Socrates was accused of having educated the most outrageous figures on both sides, Critias and Alcibiades, Plato and Xenophon say that while Socrates associated with unsavory politicians, he never encouraged their illegalities, but on the contrary challenged their credentials to rule and so stimulated them to improve themselves (and that it is not his fault if they did not always follow through on his advice). The real or mythologized events of Socrates life serve to mark the course that Plato and Xenophon want to steer, disassociating themselves from both sides of the conflict. Thus the Seventh Letter uses the Thirtys attempt to involve Soc-

_________
53 This certainly reflects the actual propaganda of the Thirty: Lysias Against Eratosthenes 5 bitterly recalls that the Thirty promised to make the city pure of unjust men, and turn the remaining citizens toward virtue and justice.

ON PLATOS

49

rates in the unlawful execution of Leon of Salamis (surely not the greatest of the Thirtys crimes) to explain Platos alienation from the oligarchy, and it uses the execution of Socrates to explain his alienation from the restored democracy; elsewhere also Socrates refusal to put the fate of the generals at Arginusae to an unlawful vote of the is used as a symbolic disavowal of the illegalities of the democracy, just as his refusal to participate in the arrest of Leon is used as a symbolic disavowal of the illegalities of the oligarchy. 54 Xenophon tried to defend his honor, linking himself with Socrates and Theramenes rather than Critias, with Agesilaus rather than Lysander; but Xenophon also left Athens, and then was exiled and could not return for decades. Plato also seems to have left Athens, but returned and set up a school in the Academy, with both Athenian and other Greek students, many of them politically ambitious people, who, like so many people that Socrates talks with in both Plato and Xenophon, have been persuaded that they need philosophy to improve themselves first before they will be able to accomplish anything in politics. Glaucon and Adeimantus represent the kind of (Athenian) student that will come to the Academy. (Glaucon is in fact one of the people that Xenophons Socrates persuades to pursue philosophy before politics, Mem. III vi.)

_________
54 Socrates in Platos Apology reminds the dikasts first of his resistance to the unlawful demands of the in the case of the generals (32a9-c4), and then of his resistance to the unlawful demands of the Thirty about Leon (32c4-e1). Xenophon Mem. IV iv 2-3 also combines what must be references to the case of the generals and to Leon in much the same way. There are explicit descriptions of Socrates conduct in the case of the generals at Mem. I i 18 (in a defense of Socrates) and Hell. I vii 15 (in a long historical account of the trial, with Socrates mentioned only briefly), and what must be a reference to the case at Gorgias 473e6-474a1. There is, however, confusion about what Socrates actually did at the assembly where the demanded a single collective vote to condemn all the generals: the two Memorabilia passages say, and the Gorgias passage implies, that Socrates was the presiding officer [ ] chosen by lot out of the tribe holding the presidency of the to preside on that day, in which case he should have been able, at least temporarily, to prevent the issue from being put to the vote of the , but the Apology and Hellenica passages imply that Socrates was merely one of the presiding group [ ] and not the presiding officer, and that while he protested he had no authority to block the vote. It seems clear that the Apology and Hellenica passages are closer to the historical facts, and that the event has been mythologized in the Gorgias and Memorabilia, as also (and even more so) in other later sources. On all this see Dodds 1959, 247-8. The Seventh Letter mentions only Leon and not Arginusae, because, writing to Dions partisans in Syracuse, Platos burden is not to explain why he does not get involved in democratic politics (Socrates execution is more than enough for that), but only why he is not going to get involved in oligarchic politics, and why he thinks those of his friends who do get involved are making a mistake (and why he did get involved both with Dion and with Dionysius II to the extent that he did).

50

STEPHEN MENN

Glaucon and Adeimantus are Platos brothers, but by the same token they are Critias cousins (and the usual scholarly guess is that the lover of Glaucon who wrote elegiac verses praising Glaucons and Adeimantus bravery at the battle of Megara, II, 368a1-4, was Critias). They could go either way: disaffected with the democracy, they could turn either to philosophical quietism or to oligarchic subversion. They are disaffected with the democracy and its talk of law and justice and social concord, not because they are partisans of some other , but because the sophistic discourse represented by Thrasymachus has quite rightly opened their eyes to the emptiness of all this talk. When Socrates hears them say what people say against justice, he marvels and concludes that they have experienced some divine favor, since Socrates knows, from their rather than from what they say, that they do not really believe that injustice is better than justice, although they are able to argue that case so powerfully (368a5-b4). Such divine favor is however notoriously unstable, and the opinion that justice is better than injustice needs to be tied down with arguments; and Glaucon and Adeimantus appeal to Socrates for help.55 When Glaucon and Adeimantus beg Socrates to convince them fully that justice is to be chosen for itself and not merely for the social consequences of appearing just, that it is better to be just and appear unjust and therefore suffer injustice than to get away with doing injustice by appearing just, they are in part asking for reasons not to become like Critias. While Glaucon and Adeimantus are often taken to be asking why I shouldnt act unjustly in such a way as to seem just to everyone else, they are also asking why we collectively shouldnt act unjustly in such a way as to seem just to everyone outside our group: as Adeimantus says, to remain concealed, we will gather and (II, 365d2-3). These or , the clubs (sometimes oath-bound, as implies) that were the basis of much political action at Athens and elsewhere, with members supporting each other in the assembly or courts according to a common plan of action, were also capable of criminal or revolutionary conspiracies, and were the nuclei of the oligarchic resemble the clubs [ ] in revolutions of 411 and 404. 56 The

_________
On Glaucon and Adeimantus compare Ferrari 2003, 11-36. On a range of issues, notably about the relation between each type of person and the corresponding type of city and its rulers, where Ferrari and I have been thinking mostly independently, and coming sometimes to similar and sometimes to divergent conclusions, discussion will have to await a fuller treatment. 56 Thucydides VIII 49 and 65 describe the role of or in the revolution of 411 (the mutilation of the Herms was also blamed on a to overthrow the democracy, VI 27). Plutarch Lysander 5 says Lysander urged people wishing revolt in the
55

ON PLATOS

51

which the Spartiates had their common meals, and the collectively could see themselves as forming the equivalent of a Spartiate citizen elite that could come to power through revolution. Thus Lysias Against Eratosthenes 43 says that the revolution of the Thirty began with five ephors [the same title and number as at Sparta; Critias was one of the five] ... established by the so-called as collectors of the citizens [ ] and leaders of the conspirators [ ]; Aristotle of the Athenians 34 says that at this time (after the Athenian surrender) the notables who belonged to the wanted an oligarchy, while the ordinary people wanted to preserve the democracy and those who did not belong to but thought themselves otherwise not socially inferior wanted a middle ground (the ). But if Platos arguments succeed with the readers represented by Glaucon and Adeimantus, they will not be tempted to follow the lead of Critias, looking to Spartan or oligarchic models and forming groups to subvert the democracy. Instead, they will do what Socrates persuades Glaucon to do at Xenophon Memorabilia III vi: instead of planning how to gain honour by becoming the leader of the city, he will work at acquiring knowledge and improving himself so that he will be worthy to lead the city, should it ever ask. Plato is not entirely rejecting the idea of : we have seen the Seventh Letter say that political action requires friends and faithful/reliable (325d1-2), that these cannot be found at Athens and that it is not easy to produce new ones; and the Academy can be seen as a way of producing such . We can thus say that Platos Republic, like the of the Athenian Laconizers, invites its readers to see themselves as members of a society of friends which might someday be the nucleus of a new (the Academy becoming the philosophers of the Callipolis). But the Academy will be a with a crucial difference; Plato sharply contrasts this kind of friendship (the kind that he had notably with Dion) with those who do not become friends from philosophy, but from the casual companionship [ ] of most friends, which they pursue from [formal] guest-friendships and from being initiated and seeing [mystery rituals together] (Epist. VII, 333e1-4, here said of Callippus, who joined Dion in his invasion of Sicily and later assassinated him). The Academy, like the Pythagorean society, is a founded on knowledge and on the justice which that knowledge is supposed to produce, and it will remain within the limits of justice. It will nurture people from many

_________
democratic cities of the Athenian alliance to form and ready themselves for political action when the time came, and Lysander put the decarchies together out of these (Lysander 13).

52

STEPHEN MENN

different Greek cities who are capable of ruling, and some of those cities may someday be in enough trouble that they will invite the Academics to give laws or settle their quarrels, as the Athenians had invited Solon (on a number of occasions Academics were in fact invited to legislate, as other philosophers had been and would be [e.g., Protagoras, Demetrius of Phalerum], in founding a colony or reestablishing a city that had been destroyed, or in making peace like Solon after ; naturally this does not mean that they were given a blank slate on which to create utopia). 57 But Platonic friends will not plot to acquire power, and if they are never invited to rule they will be just as happy contemplating: indeed, it is only because they would rather not rule that it is safe to invite them. People often prefer to read the Republic without reference to this political background, not simply because of the uncertainty of the historical evidence, but because they do not like bringing Plato so close to the Laconizers and to oligarchic revolutionaries like Dion: they are afraid that reading the Republic this way will turn it into a mere piece of political ideology, an expression of class interests and a justification for political action, failing to respect its integrity as a work of philosophy. Of course such crude readings of the Republic have been given often enough. But I do not think this is the real result of reading the Republic against the political background I have been sketching. The Republic is not a mere expression of the class ideology in which Plato grew up, but a sharp and effective critique of that ideology. It is not a justification for political action, and to the extent that it has immediate political counsels, they are counsels against political action, although it never forecloses the possibility of eventual political action, under the right circumstances, to bring about some imitation of the ideal. But political philosophy and political ideology do not have entirely independent histories; what we see here is that political ideology can provide the background from which political philosophy emerges, because a political ideology like that of the Laconizers sets up normative standards against which it can itself be judged and found wanting. Plato is responding in the first instance not to political events but to

_________
57 For a detailed survey of reports of political activity by reported Academics, see Trampedach 1994. Notably Aristotle and Eudoxus are said to have given laws to their native cities, and Plutarch Against Colotes 1126c reports the same for Menedemus of Pyrrha, Phormion of Elis and Aristonymus of Arcadia. To Trampedachs list should be added Demophanes and Ekdelus or Ekdemus, the students of Arcesilaus who are supposed to have legislated and restored peace in Cyrene, and to have been involved in the overthrow of tyrannies in Megalopolis and Sicyon (Plutarch Philopoemen 1 and Polybius X 22). Trampedach (besides doubting some of these reports) rejects the idea of an overall Academic political program, but that is another question.

ON PLATOS

53

political discourse, and his addresses the Glaucons and Adeimantuses of its audience by working within the discourse of the of the Spartans, exposing the contradictions between Spartan praise of virtue and Spartan despotism and , and bringing its audience to a rational and Socratic transformation of the Spartan ideal. This is what makes it a of political philosophy. 58
MCGILL UNIVERSITY

_________
58 Let me add here a few words in reply to Sara Monosons comments, which I appreciate not only for her kind words and accurate restatement of my main thrust, but also for her very interesting observations about Thrasymachus and the question of realizability. I agree with Monoson that the possibility of realization in an actual city is important for Platos task of answering Thrasymachus. But since such realization is improbable in the short term, it is worth stressing that the more probable second-best kinds of realization are not limited to realization within a single soul: there might be an imperfect realization in a whole city (yielding perhaps something more like Magnesia than like Callipolis), and the ideal might also be realized in a society of friends smaller than a city, within a city or cutting across the divisions between actual cities. The Academy might be one such realization, and so might the invisible community to which the Republic invites its readers to see themselves as belonging.

MENN BIBLIOGRAPHY
Annas, J. 1981. An Introduction to Platos Republic. Oxford. Bandini, M. and Dorion, L.-A. (ed. and tr.) 2000. Xnophon, Mmorables, t. 1. Paris. Bordes, J. 1982. Politeia dans la pense grecque jusqu Aristote. Paris. Bowersock, G. (ed. and tr.) 1968. Pseudo-Xenophon, Constitution of the Athenians. In Xenophon, Scripta Minora. Loeb Classical Library. Cambridge, MA: 461-507. Burnyeat, M. 1997. First words. Proceedings of the Cambridge Philological Society 43:1-20. Carter, L.B. 1986. The Quiet Athenian. Oxford. Cartledge, P. 1987. Agesilaos and the Crisis of Sparta. London. Cartledge, P. 1999. The Socratics Sparta and Rousseaus. In Hodkinson and Powell (eds.) 311-37. Cartledge, P. 2001. City and Chora in Sparta. In his Spartan Reflections: 9-20. London. Cartledge, P. forthcoming. Political Thought among the Ancient Greeks. Cambridge. Connor, W. R. 1989. Historical Writing in the Fourth Century B.C. and the Hellenistic Period. In P.E. Easterling and B.M.W. Knox (eds.), Cambridge History of Classical Literature, I/3: 46-59. Cross, R.C. and Woozley, A.D. 1964. Platos Republic. London. David, E. 1979. The Pamphlet of Pausanias. Parola del Passato 34: 94-116. Dawson, D. 1992. Cities of the Gods. Oxford. de Romilly, J. 1959. Le classement des constitutions dHerodote Aristote. Revue dtudes grecques 72: 81-99. de Ste. Croix, G.E.M. 1972. The Origins of the Peloponnesian War. London Dodds, E.R. (ed. and comm.) 1959. Plato, Gorgias. Oxford. Dorion, L.-A. 2002. La responsabilit de Cyrus dans le dclin de lempire perse selon Platon et Xnophon. Revue franaise dhistoire des ides politiques 16: 369-86. Dring, I. 1957. Aristotle in the Ancient Biographical Tradition. Gteborg. Ferrari, G.R.F. 2003. City and Soul in Platos Republic. Sankt Augustin. Figueira, T.J. 1999. The Evolution of the Messenian Identity. In Hodkinson and Powell (eds.) 211-44. Gera, D.L. 1993. Xenophons Cyropaedia. Oxford. Griffith, M. (tr.) and Ferrari, G.R.F. (ed.) 2000. Plato, The Republic. Cambridge. Hodkinson, S. 2000. Property and Wealth in Classical Sparta. London. Hodkinson, S. and Powell, A. (eds.) 1999 Sparta: New Perspectives. London. Hffe, O. (ed.) 1997. Platon, Politeia [essays]. Berlin. Hornblower, S. 2000. The Old Oligarch (Pseudo-Xenophons Athenaion Politeia) and Thucydides. In Polis and Politics: 363-84. Copenhagen. Jacoby, F. (ed.). 1923-. Die Fragmente der griechischen Historiker. Berlin and Leiden. Jacoby, F. 1949. Atthis. Oxford. Leroux, G. (tr.) 2002. Platon, La Rpublique. Paris. Lipka, M. 2002. Xenophons Spartan Constitution. Berlin. Mattingly, H.B. 1997. The Date and Purpose of the Pseudo-Xenophon Constitution of Athens. Classical Quarterly n.s. 47: 352-7. Mirhady, D. 2001. Dicaearchus of Messana: The Sources, Texts and Translations. In Fortenbaugh, W.W. and Schtrumpf, E. (eds.), Dicaearchus of Mesana: Text, Translation and Discussion, New Brunswick, NJ: 1-132. Momigliano, A. 1993. The Development of Greek Biography. Expanded ed. Cambridge, MA.

MENN BIBLIOGRAPHY

55

Murphy, N. R. 1951. The Interpretation of Platos Republic. London. Ober, J. 1998. Political Dissent in Democratic Athens. Princeton, NJ. Popper, K.R. 1945. The Open Society and its Enemies. v.1: The Spell of Plato. London. Rebenich, S. (ed. and tr.) 1998. Xenophon, Die Verfassung der Spartaner. Darmstadt. Reeve, C.D.C. 1988. Philosopher-Kings. Princeton, NJ. Schofield, M. 1999. Zeno of Citiums Anti-Utopianism. In Saving the City: 51-68. London. Strauss, L. 1975. The Argument and the Action of Platos Laws. Chicago, IL. Tigerstedt, E.N. 1965-78. The Legend of Sparta in Classical Antiquity, 3 vols. Stockholm. Trampedach, K. 1994. Platon, die Akademie, und die zeitgenssische Politik. Stuttgart. Treu, M. 1966. Ps.-Xenophon: . RE IXA/2: 1928-82. van Wees, H. 1999. Tyrtaeus Eunomia. In Hodkinson and Powell (eds.) 1-41. White, N.P. 1979. A Companion to Platos Republic. Oxford.

COLLOQUIUM 2

WISHING FOR FORTUNE, CHOOSING ACTIVITY: ARISTOTLE ON EXTERNAL GOODS AND HAPPINESS ERIC BROWN

I. Introduction In Book One of the Nicomachean Ethics (EN), Aristotle seeks to identify the human good, which he also calls eudaimonia or happiness (I 4, 1095a14-20) and which he explains as that for the sake of which one should do everything one does (I 7, 1097a22-24 and 1097a25-b21). After introducing the idea (in chapters one through three) and surveying some received accounts of it (in chapters four through six), he seems to give his definition in the seventh chapter, where he appeals to the human function and concludes that the human good is activity of the [rational] soul in accordance with virtue, and if there are multiple virtues, in accordance with the best and most complete virtue (I 7, 1098a16-18). This account is sketchy, as Aristotle admits (I 7, 1098a20-22): he needs to say what virtuous activity is, how many virtues there are, and whether some one virtue is best and most complete. But the account has enough content to suit Aristotles initial purposes (I 7, 1098a22-b8) and to court interpretive controversy. Perhaps the most obvious controversy is this: Does Aristotle really mean that the human good is just virtuous rational activity? Are health and wealth, not to mention friends and lovers, not part of the goal for the sake of which one should do everything one does? Many readers think that Aristotle does not intend such a narrow account. Some point to what he says about happiness before he comes to the human function argument, or to what he says about the good human life outside of Book One in, say, his discussion of friendship. But others point to what he says about happiness in Book One after he produces his apparently narrow definition. In EN I 8-12, Aristotle tests his account against what is commonly said about happiness, and he affirms that goods external to the soulexternal goods are necessary for happiness. Some readers insist that in these chapters he also expands his definition of happiness to include the external goods. In this essay, I tackle just this last part of the question: my exegetical thesis is that Aristotle sticks by his narrow account of happiness from its

58

ERIC BROWN

introduction in EN I 7 through the rest of Book One. What I propose to show is restricted: I leave aside concerns from Book One that precede the function argument and those from outside Book One. Moreover, what I propose to show is unoriginal: the debate over Aristotles definition of the human good is well established, and others have supported the claim that he sticks to a narrow definition of happiness as virtuous rational activity. But I do have three exotic fish to fry. First, I support my exegetical thesis by providing a map of EN I 8-12 as a whole. The thorny and muchdiscussed passages that directly pertain to the relation between external goods and Aristotles account of happiness are contestable, but I argue that the transitions throughout EN I 8-12 make Aristotles intentions plain. Second, to uphold my exegetical thesis, I maintain that in EN I 8-12 Aristotle claims that external goods are necessary for happiness only because they are necessary for virtuous rational activity, and to defend this, in turn, I offer a new account of why he thinks that external goods are necessary for virtuous rational activity. My account innovates by attributing to Aristotle the view (roughly) that virtuous people have a psychological need for certain external goods. Third, I develop my account in terms of Aristotles distinction between wish and choice. On my view, he wants us to choose activity while we merely wish for good fortune and the external goods that good fortune brings. But as I shall argue, Aristotle also believes, first, that choosing virtuously requires wishing for external goods that cannot be chosen, because virtue is partly constituted by the correct appreciation of value, and second, that our capacity to choose virtuously is diminished when we do not get what we wish for. These previously unacknowledged claims about the relation between wish and choice help to explain why Aristotle believes that external goods are necessary for virtuous activity and thereby happiness. They also seem to me both striking and quite possibly true. My primary purpose here is to bring them to light. II. In a complete life (EN I 7, 1098a18-20) I start with my exegetical thesis. I maintain that Aristotle sticks by his narrow definition of happiness as virtuous activity after he introduces it in EN I 7. There is a challenge to this thesis even before EN I 8-12. Upon completing his function argument and concluding that happiness is virtuous activity, Aristotle immediately adds, And in a complete life (EN I 7, 1098a18-20). It is possible to hear in these words a reference to the need for external goods in addition to virtuous activity, and thus to hear in them

WISHING FOR FORTUNE, CHOOSING ACTIVITY

59

an emendation of the narrow account of happiness. Thus, before I turn to EN I 8-12, I must counter this possibility and show that EN I 7 concludes with the narrow definition intact. This I will let Aristotle do, for he explains what his appendix means. He says, And in a complete life, for one swallow does not make a spring, nor does one day, and in this way neither one day nor a short time makes a man blessed and happy (EN I 7, 1098a18-20, emphasis added). Aristotles point concerns time. There is good reason for Aristotle to be making a chronological point. Once one realizes that happiness just is virtuous activity, one might wonder, Can I wholly instantiate happiness in one burst of virtuous activity? To answer this question, Aristotle clarifies that happiness is only partly instantiated by a days worth of virtuous activity. This might seem too obvious to be Aristotles point: surely it is obvious that the pursuit of happiness is a lifelong project, that happiness is wholly instantiated only by a lifetime of virtuous activity. In fact, it is not obvious, and Aristotle should have said more about it than he does. First, there is room for confusion about what constitutes a complete life of virtuous activity: is it a completed lifetime or some shorter span? But second, and more importantly, some Greeks, including Stoics, Epicureans, and quite possibly some of Aristotles contemporaries, thought that happiness is wholly realized by a short period of virtuous activity. These philosophers might have wanted to see an argument. Still, what Aristotle might have argued is a topic of another paper. For my purposes here, it is enough to see that Aristotles insistence on a complete life concerns time and not external goods. III. A Map of EN I 8-12 External goods enter in EN I 8-12. Scholars usually turn directly to the most relevant passages, but I first want to establish a map of the five chapters. The reason for this is simple. The most relevant passages are particularly thorny, which opens them to multiple interpretations, but the transitions that structure EN I 8-12 reveal Aristotles intentions much more straightforwardly. A map of EN I 8-12 shows that Aristotle means to defend his narrow definition of happiness as virtuous activity. At the start of these chapters, Aristotle announces the need to test his account of happiness (EN I 8, 1098b9-12): We must examine happiness not only from my conclusion and premises, but also from the things that

60

ERIC BROWN

are said about it, for all facts harmonize with what is true and the truth quickly conflicts with what is false. So the question is, does Aristotle take himself to confirm or reject his narrow definition of happiness? A glimpse at his next major transition, at the start of chapter thirteen, suggests that Aristotle takes himself to have confirmed his definition. For he turns to his next topic by saying, Since happiness is a kind of activity of the soul in accordance with complete virtue, we must inquire about virtue (EN I 13, 1102a5-6). Now, this appearance could be deceiving; Aristotles transition at the start of chapter thirteen might be the telescoped expression of a more complicated truth. But if it were, then we should expect to see very clear indications of that within chapters eight through twelve. Let us look. In chapter eight, Aristotle considers five things said about happiness: that (1) the goods of the soul are more properly goods than the goods of the body and the goods external to the soul and the body, that (2) the happy man lives well and fares well, and that happiness seems (3a) to some to require virtue, (3b) to others to require pleasure in addition, and (3c) to still others to require external goods, too. He makes quick work of the first two. He explains that the first (EN I 8, 1098b12-16) confirms his account, since he locates happiness in actions and activities and thereby among goods of the soul and not among external goods (EN I 8, 1098b1620). This would be especially perverse if he had a broad definition of happiness in mind. Then he asserts that the second point harmonizes with his account or definition ( , 1098b20), since his definition practically makes happiness living well and faring well (EN I 8, 1098b20-22). So he insists that both of the first two views harmonize with his narrow definition of happiness as virtuous activity. Aristotle then works through the next three views more patiently. He introduces them together and concedes some truth to them, but he explicitly backs away from saying that they are unqualifiedly true, asserting that it is reasonable that none of these people err entirely, but are right in at least one respect or even for the most part (EN I 8, 1098b28-29). Thus, when he considers the view that happiness is a kind of virtue, he draws out its harmony with his definition ( , 1098b31), but he also explains how his definition improves on the received view by insisting on virtuous activity and not the mere possession of virtue (EN I 8, 1098b30-1099a10). And when he considers the view that happiness requires pleasure, Aristotle argues that each person finds pleasure in what he loves and the virtuous agent loves virtuous activity (EN I 8, 1099a7-31). This allows him to conclude that pleasure belongs to the best activities, and these, or the single

WISHING FOR FORTUNE, CHOOSING ACTIVITY

61

best of these, is what we said happiness is (EN I 8, 1099a29-31). He argues, in other words, that both the third and fourth views harmonize with the narrow definition of happiness as virtuous activity. The fifth view that happiness needs external goodsbrings us to one of the thorny passages that I am temporarily setting aside. For now, it is enough to say that this passage concerns one of five views that are considered, and the other four are all made to harmonize with the narrow definition of happiness. Aristotles clear strategy in treating these four points is also on display in chapter nine, where he considers an old puzzle about the acquisition of happiness: is happiness acquired by practice and effort, by divine providence, or by luck (EN I 9, 1099b9-11)? The broad contours of his solution to this puzzle are clear. He rules out the possibility that happiness comes through luck; he insists that it comes through effort; and he allows that the achievement of happiness through effort might also be considered providential, although he backs away from a full discussion of this possibility (EN I 9, 1099b11-18). For my purposes here, the most important point is that Aristotle fits his solution to his narrow definition of happiness. After insisting that happiness is not acquired by luck, he notes, [The solution to the puzzle that] we seek is also perfectly clear from my definition [ ], for happiness was said to be a kind of activity of the soul in accordance with virtue (EN I 9, 1099b25-26). Then, in chapter ten, Aristotle tackles a second puzzle. Solon used to say that he would not judge any living person happy, because changes in fortune could render his judgment false (Herodotus I 30-32). So Aristotle asks whether we have to wait until a man is dead to declare him happy. His response is complicated. He does not agree that one has to wait until the very end of a persons life (EN I 10, 1101a14-16), though he acknowledges that happiness is not entirely impervious to changes in fortune (EN I 10, 1101a8-13). To make room for judging a living person happy without courting error, Aristotle insists that happiness is not easily subject to change. As he puts it, The current puzzle also bears witness to my definition [ ], for stability belongs to none of the human products in the way that it belongs to virtuous activities (EN I 10, 1100b11-13). This way of putting the point suggests that Aristotle means again to confirm his narrow account of happiness in chapter ten. Toward the end of the chapter, however, Aristotle puts his point in another way that suggests to some readers a broadened account of happiness. He asks (EN I 10, 1101a14-16), What, then, prevents one from calling happy a man who is active in accordance with complete virtue and is supplied sufficiently with external goods, not for just any length of time but for a complete life? It

62

ERIC BROWN

is important to keep this question in its context. Aristotle is not seeking a definition of happiness. He is in the middle of testing the old, narrow definition to which he has just referred ( , 1100b11), and thus far he has given no explicit indication that this definition is unsatisfactory. More specifically, he is completing his response to the test posed by the Solonic puzzle concerning whether we can call a person happy when he is alive. Having explained that happiness is especially stable because virtuous activity is especially stable, Aristotle insists that one is justified in calling a living man happy if he acts virtuously and has a sufficient supply of external goods. This does not require that one define happiness in terms of both virtuous activity and external goods. It is equally well explained if one defines happiness as virtuous activity and thinks both that happiness requires a complete life and that virtuous activity requires external goods. Allow me to explain. If one thinks that virtuous activity requires external goods, it would be redundant to tack the question Does he have enough external goods? onto the question Does he act virtuously? But Aristotles question demands that one consider whether the mans external goods are sufficient not for just any length of time, but for a complete life. It is not redundant to tack the question Does he have enough external goods? onto the question Does he act virtuously? when one wants to know whether he will continue to act virtuously for a complete life. To know whether a man will continue to act virtuously for a complete life, one needs to know not just whether he is now acting virtuously (which includes consideration of whether he has enough goods to act virtuously now) but also whether he now has sufficient resources to make virtuous activity possible into the indefinite future, all the way to the point of having lived a complete life. So, despite the suggestions of scholars to the contrary, Aristotles rhetorical question at the end of chapter ten is not in the service of defining happiness, and its answer does not require a broadened definition of happiness. The tenth chapter does indeed carry out the same strategy that we found in the previous two: it tests and confirms the narrow definition of happiness. Aristotle digresses a bit in chapter eleven to address a puzzle that does not directly challenge his definition of happiness. He does this to pick up some residue of his discussion of the Solonic puzzle. In chapter ten, he had noted that Solons position might be extended. One might think that one should wait until well after a mans death before saying that he had been happy, since the fortunes of his descendants might matter (EN I 10, 1100a18-27). But Aristotle bracketed this extended version of the puzzle

WISHING FOR FORTUNE, CHOOSING ACTIVITY

63

by declaring that it would be absurd if ones happiness were to be changed by events after ones death (EN I 10, 1100a27-29), though he allowed that it would also be absurd if ancestors were in no way affected by what happens to their ancestors after they themselves had died (EN I 10, 1100a2930). Thus, in chapter eleven, Aristotle returns to the question of what effect posthumous events have. Note, however, that he does not return to the question of whether a mans happiness can fluctuate after his death. He has put that suggestion aside as absurd, and he never returns to it. Thus he recognizes no challenge here to his narrow account of happiness. He merely concedes that there are some posthumous effects on a man without admitting that there are posthumous effects on his happiness (EN I 11, 1101b1-9). Finally, in chapter twelve, Aristotle returns to testing his narrow account of happiness as virtuous activity. He asks, Is happiness something merely to be praised, or is it something to be honored with encomia? He explains that good dispositions and potentialities are fit for praise, but better things are fit to be honored with encomia. So virtue is fit to be praised, virtuous activity fit to be honored (EN I 12, 1101b31-34). Accordingly, Aristotles conclusion that happiness should be honored with encomia supports his definition of happiness as virtuous activity against those who would identify it with virtue (for whom see EN I 8, 1098b23-25 and b30, and 1099b8). The consistency of Aristotles resolve in these chapters is remarkable. He regularly goes out of his way to link the truth in some received view or puzzle to his narrow definition of happiness as virtuous activity, and what is perhaps more impressive, he nowhere makes any explicit declaration that takes back or alters his narrow definition. This strongly supports the natural reading of the words that begin and follow EN I 8-12. Aristotle wants to concede the importance of external goods without giving up on his claim that the human good is just virtuous activity. It is time to see whether he can actually do this. IV. The Central Argument (EN I 8 1099a31-b8) A. Four Inferences Aristotles most direct argument concerning the relation between happiness and external goods comes in the thorny passage at the end of chapter eight. He has introduced the ordinary view that happiness requires exter-

64

ERIC BROWN

nal goods, and he has conceded some truth to it (EN I 8, 1098b26-29). Now he offers his analysis:
[A] Nevertheless, it is apparent that happiness also needs external goods, as we said, for [B] it is impossible or not easy to do fine actions if one is not equipped. For, on the one hand, [C] many are done by means of friends and wealth and political power as if by tools, and on the other hand, [D] men who lack some things such as good birth, good children, and beauty soil blessedness; for [E] the man who is very ugly in appearance or of low birth or solitary and childless is not entirely happy, and [F] moreover, he would perhaps be even less so if he had thoroughly bad children or friends, or if his children or friends, though good, had died. Thus, as we said, [G] happiness seems to need this sort of prosperity in addition, and that is why [H] some people identify happiness with good fortune, while others identify it with excellence. (EN I 8, 1099a31-b8)

This passage is so thorny that controversy extends even to its translation. Patience is required. I start with the explicit logical relations among the claims. In (A), Aristotle announces his thesis that happiness requires external goods. In (B), he introduces a reason for his thesis ( , 1099a32): it is impossible or difficult to do virtuous activities without external goods. (Call that inference one.) Then, in (C) and (D), he introduces two explanations ( , 1998a34 and 1099b2) of why it is impossible or difficult to do virtuous activities without external goods: on the one hand ( , 1998a34), some external goods are tools for virtuous activities, and on other hand ( , 1999b2), some external goods affect our blessedness. (These are inferences two and three.) Finally, in (E) and (F), Aristotle offers reason ( , 1099b3) for supposing that external goods affect our blessedness (inference four). In sum, (E) and (F) are reason for (D); (C) and (D) are reasons for (B); and (B) is a reason for (A). Aristotle concludes his discussion by reiterating in (G) that his thesis follows from the whole chain of inferences, and by noting in (H) that the whole chain leads some people to mistaken characterizations of happiness. This logical structure is quite clear in Aristotles Greek, but it is frequently misrepresented in English. In particular, many interpreters construe (D) not as a second reason for (B), but as a second direct reason for (A). This matters to my exegetical thesis. If (D) is a second reason for (B), then Aristotle affirms that (A) external goods are necessary for happiness precisely because (B) they are needed for virtuous activity. But if (D) is a second direct reason for (A), then it would be natural to say that happiness includes both virtuous activity (B) and the external goods that bring blessedness (D).

WISHING FOR FORTUNE, CHOOSING ACTIVITY

65

As I say, the Greeks logical conjunctions tell in favor of my reading and thereby support my exegetical thesis. But there might be reason to overlook the explicit import of the Greek. First, if there were no good defense of the four inferences that Aristotle explicitly marks, as he marks them, then charity would summon forth a looser construal of his words. Second, Aristotle might offer later textual parallels that require a looser reading of this one. To defend my exegetical thesis, I shall show that these reasons do not hold. In the rest of this section, I defend the plausibility of Aristotles four inferences, one by one, and in the next section, I consider the two important textual parallels from EN I 9-12. Of course, this work will also develop why Aristotle thinks that virtuous activity requires external goods, and thus it will introduce the relation between wishing and choosing that I am especially eager to explore. B. Inference One Aristotle first infers (A) that happiness needs external goods from the claim (B) that it is difficult or impossible to do virtuous activities without external goods. There is a valid inference in the offing if Aristotle is assuming the narrow definition of happiness as virtuous activity: (Definition) Happiness is virtuous activity. (B) Virtuous activity needs external goods. Therefore, (A) Happiness needs external goods. Unfortunately, this does not exactly capture Aristotles words. His premise (B) is weaker. He says that virtuous activity is impossible or not easy without external goods. This weaker premise can sustain only the weaker conclusion that happiness is impossible or not easy without external goods. So either Aristotle is fallaciously inferring a strong conclusion from a weaker premise or his apparently strong conclusion is actually commensurate with his weaker premise. One does not need much charity to see why the second option provides the better reading. On Aristotles view, necessity is said in many ways (Metaphysics 5), and so, then, is the sentence Happiness needs external goods. In fact, according to one, relaxed way in which one might reasonably say that happiness needs external goods, one means only that happiness is difficult or impossible without external goods. Consider: it is perfectly reasonable to say that a person needs a car in order to get to campus even though a person could bike or jog the ten miles. In a case

66

ERIC BROWN

like this, one says that a person needs x for y because we recognize that y is impossible or not easy without x. Reasonable reflection of this sort secures Aristotles inference. Now charity can kick in. If Aristotles inference is valid, as charity demands, then he must mean that happiness needs external goods with a relaxed sense of needs. He must mean, as he very reasonably can, that happiness is impossible or not easy without external goods. There is a broader lesson here. The meaning of Aristotles claim that happiness needs external goods is not fixed independently of his development and defense of the claim, and his argument requires that happiness needs external goods only in a relaxed sense of needs. C. Inference Two Aristotle next gives his first of two reasons why virtuous activity is impossible or not easy without external goods: some external goods are needed as tools for virtuous activity. The Socrates of Platos Euthydemus or some other proto-Stoic could quibble here, for it is surely possible to insist that virtuous activity is whatever activity the virtuous person would do in the circumstances, where the circumstances can be whatever you like. But Aristotle has a different position on virtuous activity, and it makes his second inference perfectly reasonable. On his view, virtuous activity must live up to aristocratic ideals, and for that reason, it needs some of the aristocrats tools. D. Inference Three The next inference is not so clear. Aristotle introduces external goods that are not instruments for virtuous activity, and he insists that they affect the blessedness of life. This might seem to say that some external goods make a direct contribution to happiness and thus are to be included in the definition of happiness. But according to the logical structure of the passage, the dependence of blessedness on external goods (in [D]) is not a direct reason for supposing that (A) happiness requires external goods. Rather, it is a reason for supposing that (B) virtuous activity needs external goods. Why should that be? Here is a possibility: (B) virtuous activity needs external goods because (D) blessedness needs external goods, (ii) blessedness is happiness, and (iii) happiness is virtuous activity. But this reading, too, fails to make sense of Aristotles inferences as he marks them. Aristotle clearly wants to get from (D) to (B) to (A). This reading, by contrast, gives him (A)

WISHING FOR FORTUNE, CHOOSING ACTIVITY

67

already with (D) and (ii); it has no use for (B) at all. In fact, any reading that treats blessedness and happiness as interchangeable will have no need of (B) to get from (D) to (A). Aristotle says that (D) is a reason for (B), which is, in turn, a reason for (A). We should at least try to understand this. John Cooper provides a better explanation of the move from (D) to (B). He points to a passage in Book Seven where Aristotle says that the lack of external goods impedes virtuous activity (EN VII 13, 1153b1719), and so he attributes to Aristotle some assumptions about the causal necessity of non-instrumental external goods for the capacity to act virtuously. This, I think, is along the right track, and what remains is to make sense of the causal mechanism that links non-instrumental external goods and the capacity to act virtuously and to explain why Aristotle puts the point in terms of blessedness. For Cooper, the causal mechanism is mainly social: the possession of certain non-instrumental external goods confers enhanced social standing and thereby the opportunity to exercise the standard Aristotelian range of excellent actions. According to Aristotles examples of non-instrumental external goods, our virtuous activity requires that we be well-born, goodlooking parents of good children. Cooper explains, Some external conditions (being good-looking, having good children, coming from a good family), while not used by the virtuous person as means to achieve his purposes (as, e.g., his money or personal influence might be), put him in the position where the options for action that are presented to him by his circumstances allow him to exercise his virtues fully and in ways that one might describe as normal for the virtues (Cooper 1999, 298-299). Some critics have flatly rejected Coopers idea (see, e.g., Botros 1986, 113). Controversy partly focuses on his concrete suggestion that ugly people do not have as many opportunities for sex and thus are capable of less grand temperance than beautiful people. But this is a needless distraction. Coopers general position is perfectly plausible, and it does not require that every possible virtuous action (or omission) be straightforwardly dependent upon non-instrumental external goods. It is enough for Cooper to illuminate the ways in which the reach of excellent activities is expanded by the opportunities afforded by good looks, good birth, and good children, for this demonstrates that without these external goods, one would find it more difficult to act in excellent ways. That is already enough, since Aristotles conclusion, as I argue above, does not insist that external goods are strictly necessary for virtuous activity. But in fact, Coopers general analysis is still more powerful in light of an especially aristocratic conception of virtuous activity. To act as a paradigmatic aris-

68

ERIC BROWN

tocrat, one must be recognized as a superior, which requires at least nobility if not beauty, and one must extend ones family honor with good children. One who lacks these goods is simply incapable of acting up to the standards of aristocratic excellence, and so, on the aristocratic conception, these external goods are straightforwardly, robustly necessary for virtuous activity. But this does not fully explain Aristotles third inference, for two reasons. First, Coopers analysis does not explain Aristotles particular locution that men who lack some things such as good birth, good children, and beauty soil blessedness (EN I 8, 1099b2-3). It explains neither why the men, and not the lack of non-instrumental external goods per se, are somehow responsible for soiling blessedness, nor why the point is expressed in terms of soiled blessedness. Second, the social mechanism that Cooper highlights does not exhaust Aristotles reasons why noninstrumental external goods are needed for virtuous activity. To fill the second gap first: there is also a psychological mechanism. The difficulties posed by lacking what our society esteems lie not merely in the diminished opportunities afforded us by others, but also in our dashed hopes. At least, this is what Aristotle says in a chapter ten passage that is clearly relevant to the text currently under consideration. He says, When many great events occur badly, they oppress and spoil blessedness, for they bring pains and impede many activities (EN I 10, 1100b28-30). Coopers social mechanism might be suggested by the mention of impeded activities, but Aristotle also clearly mentions a role for pain (or grief, as we might render ). So according to the passage in chapter ten, there are two ways in which the loss of non-instrumental external goods soils blessedness. What do these ways tell us about the third inference in the chapter eight passage? One way clearly provides support: when the loss of non-instrumental external goods soils blessedness by impeding activities, it is also diminishing virtuous activities. Coopers social mechanism specifies how this might be so. But the other way in which chapter ten reports that the loss of non-instrumental external goods soils blessedness also supports the third inference in chapter eight if the pain or grief that soils blessedness also diminishes virtuous activity. This is quite plausible, for pain cuts into the pleasure of acting well. In fact, to the extent that virtuous activity requires taking pleasure in it, pain makes it more difficult to act virtuously, and to the extent that perfectly virtuous activity requires taking wholehearted, pain-free pleasure in it, pain makes perfectly virtuous activity impossible. Thus, non-instrumental external goods are

WISHING FOR FORTUNE, CHOOSING ACTIVITY

69

psychologically necessary for the capacity to act virtuously, in addition to being socially necessary. To develop the psychological mechanism that Aristotle invokes and to explain his claim that men who lack non-instrumental external goods soil blessedness, I turn to his distinction between wish and choice. This might seem odd. Aristotle distinguishes between wish and choice only later in the Nicomachean Ethics, in Book III, chapter two, when he is trying to explain what choice is, and he does not explicitly invoke wish at all in Book One. But wish is surely relevant, because it is the attitude one has toward any good one would like to possess. As Aristotle explains in EN III 4, the good without qualification is the object without qualification of wish, and what appears good to someone is an object of wish in relation to him. So whenever we are talking about goods that one might like to possess, including external goods, we are talking about objects of wish. For example, one can wish for a friendship (EN VIII 3, 1156b29-31), and one can wish for ones friend to enjoy a good, just for the sake of ones friend (EN VIII 2, 1155b31). The distinction between wish and choice suggests, moreover, that some discussions of external goods are about what I call mere wish. Here is how Aristotle distinguishes:
But neither is choice wish, though it seems near to wish. For [1] there is no choice of impossible things, and if someone should say that he chose something impossible he would be thought silly; but there is wish <even> for impossible things, e.g., for immortality. And [2] wish can also concern things that could in no way be done by ones own efforts, e.g., that a particular actor or athlete should win in a competition; but no one chooses such things, but only the things that he thinks could come to be by his own efforts. Further, [3] wish is more for the end, choice for what promotes the end; for instance, we wish to be healthy, but we choose the things by which we will be healthy, and we wish to be happy and say that we do, but it does not sound right to say that we choose to be happy. For [4], in general, choice seems to concern the things that are in our own power. (EN III 2, 1111b19-30)

I take this last point to sum up Aristotles contrast: [4] choice concerns only the things in our power but wish ranges more broadly. The first three points spell this out: [1] one can wish for but cannot choose what is impossible; [2] one can wish for but cannot choose what is possible but not under ones control; and [3] one can wish but cannot choose to enjoy the goal of ones endeavors. The third point calls for a slight digression because it might seem obscure or, worse, a threat to my exegetical thesis. Aristotles point is that our activity often aims at a goal whose coming about the activity itself

70

ERIC BROWN

cannot ensure. This can happen in either of two ways. First, I might act so as to bring about a state of affairs even though my activity by itself will not suffice to bring that state of affairs about. For example: I eat well and exercise so that I will continue to enjoy good health, but eating well and exercising cannot ensure that I will be healthy. Alternatively, I might act so as to begin a temporally extended action whose completion my current activity cannot guarantee. For example: I might set words to paper to write a commentary on Aristotles Nicomachean Ethics, but my setting words to paper today cannot ensure that I will write a commentary on Aristotles Nicomachean Ethics. The second kind of case is especially relevant to explaining why happiness cannot be chosen though virtuous activity can be chosen, and Aristotles insistence that happiness is a complete life of virtuous activity is crucial (cf. section II above). When a man acts virtuously for the sake of happiness, he partly instantiates happiness, but he cannot guarantee the full instantiation of happiness because his acting virtuously here and now cannot guarantee a complete life of virtuous activity. The first two points about what we can wish for but cannot choose what, as I say, we merely wish forare relevant to the argument at the end of EN I 8. To judge by his examples in that argument, Aristotle has objects of mere wish in mind as the non-instrumental external goods needed for blessedness. Good birth is obviously beyond our power to choose. Physical beauty is also reasonably thought to be beyond our control there is only so much that exercise, hygiene, cosmetics, and the like can do (cf. EN III 5, 1114a23-25). Good children might seem to be a different matter, but there is reason to think that Aristotle would not think so. It is not just that some people are infertile or that death in childhood was a common feature of the ancient world, though these points go some distance toward making good children objects of wish rather than choice. Rather, even when children can be chosen, good children cannot. Socrates registers this to query the teachability of virtue: he points out that great Athenians have had neer-do-wells for children (Meno 92d-95a, Protagoras 319e-320b). Aristotle, we can be certain, knew of Socrates challenge, and it provides good reason to think of good children as something to be wished for and not chosen. Of course, Aristotle need not think that all non-instrumental external goods are objects of mere wish. It is significant enough if Aristotle does not shy away from the fact that some important examples of them are. For if this is so, then even mere wishes are relevant to Aristotles case for claiming that non-instrumental external goods are necessary for virtuous

WISHING FOR FORTUNE, CHOOSING ACTIVITY

71

activity. This suggests that what is often called idle wish because it does not give rise to action is not entirely idle, insofar as its satisfaction (or not) affects ones ability to act virtuously. It also raises questions about why one would not simply jettison ones wishes for objects one cannot choose. I return to these questions in the last section. The distinction between wish and choice can be used to articulate more fully the mechanisms by which external goods are necessary for virtuous activity. The distinction illuminates the psychological mechanism by highlighting unfulfilled wishes (and, in part, unfulfilled mere wishes) as the source of pain and difficulty. It also deepens the account of the social mechanism by making possible the idea of shared wishes (and even shared mere wishes): many of the opportunities for the good-looking well-born who have good children are available because members of society converge in thinking that good looks, good birth, and good children are features of the aristocratic ideal worth wishing for (and even merely wishing for). Indeed, these are just the points that Aristotle could make by saying that men who lack some external goods soil blessedness. On the one hand, Aristotles odd suggestion that the deprived men (and not the lack of external goods per se) are responsible for soiling blessedness is unexplained by Coopers social mechanism and, in fact, widely ignored by the translators. But the role of unfulfilled wishes helps to explain Aristotles meaning by highlighting why he would assign responsibility to mens attitudes or diminished capacities. On the other hand, Coopers social mechanism does not adequately explain why Aristotle expresses the importance of non-instrumental external goods by saying that blessedness is soiled in their absence. But if one has wishes in mind, the reference to blessedness suggests the ideal life that is the summation of what can be wished for, and thus offers a way of understanding Aristotles premise about blessedness without assuming his ultimate conclusion about happiness. These points deserve more careful consideration. Recall that Aristotle argues that (D) the need of blessedness for external goods is a reason why (B) virtuous activity needs external goods, which is a reason why (A) happiness needs external goods. So Aristotle cannot mean by blessedness exactly what he means by happiness without begging the question. What does he mean by happiness in (A)? According to the first inference, he means happiness as he has defined it, the temporally complete life of virtuous activity that is the goal for the sake of which we should do everything we do. What does he mean by blessedness in (D)? It must be something different. This much is required by the logic, and suggested by Aristotles shift in terms. It should be something closer to the

72

ERIC BROWN

Aristotles shift in terms. It should be something closer to the commonsense view of a happy life, too, because in this passage Aristotle is testing his technically derived definition of happiness against ordinary views. I suggest that the role of wishes offers a helping hand here. I suggest that Aristotle invokes blessedness as the ordinarily conceived life that optimally realizes all that one might wish for. This construal obeys the constraints Aristotle faces: he needs to infer (A) from (B) and (B) from (D) without begging the question, and he needs to test his theoretical account of happiness in (A) against ordinary views about happiness. It also highlights the wishes that help to explain the inferences. Obviously, if blessedness is a life that optimally realizes everything one might wish for, then the lack of some non-instrumental external goods that one might wish for soils blessedness. That is the basic idea of premise (D), though Aristotle recasts the point so as to bring out the role that mens attitudes or capacities play. By casting the point in this odd way, Aristotle hints at the mechanisms by which the lack of noninstrumental external goods also impairs virtuous activity: when I do not enjoy what I wish for, I experience pain, which adversely affects my capacity to act virtuously, and when I do not enjoy what we wish for, I experience diminished social opportunities, which adversely affects my capacity to act virtuously. So the role of wishes helps to explain the premise (D), the curious way in which Aristotle states that premise, and the inference from the premise to the claim that virtuous activity needs external goods. To sum up a long discussion, I want to emphasize three claims about Aristotles third inference at the end of EN I 8. First, Aristotle means to derive from an ordinary notion of happiness some support for his theoretical definition of happiness. This is plain from the logic, which requires that happiness in (A) be distinct from blessedness in (D), and it is plain from the general context of the chapter, in which Aristotle is testing his theoretical account against ordinary views of happiness. The distinction between the two conceptions of happiness can be put this way: on Aristotles theoretical account, the goal of living is not what is ordinarily recognized as happiness but something shown by analysis to be the practical aim that structures and makes possible what is ordinarily recognized as happiness. Second, Aristotle assumes that there are two mechanisms by which non-instrumental external goods are needed for virtuous activity: they are needed for the enhanced social standing that makes possible the full range of virtuous activity, and they are needed for the avoidance of psychological pain that would adversely affect the virtue of ones activity.

WISHING FOR FORTUNE, CHOOSING ACTIVITY

73

Third, my first two points can be cast intelligibly in terms of fulfilled and unfulfilled wishes, with even a role for mere wishes for things that cannot be chosen. To recast the first point: on the ordinary view, happiness is the optimal fulfillment of wishes, and on Aristotles view, the practical aim of virtuous activity makes possible the optimal fulfillment of wishes. To recast the second point: Aristotle also recognizes that unfulfilled wishes undermine our capacity for virtuous activity because our failure to enjoy what society wishes for adversely affects our opportunities for virtuous activity and because our failure to enjoy what we wish for adversely our capacity to act with painless excellence. Recasting Aristotles argument in terms of wishes is permitted by Aristotles account of wishing and choosing, though he himself does not do it. There are two reasons to recast the argument: first, to provide a richer explanation of Aristotles position in EN I 8-12, and second, to make more explicit some challenges latent in that position. I pursue the second line in section six. First, there is more work to do to secure the explanation of Aristotles position in EN I 8-12. E. Inference Four Thus far, I have tracked three of Aristotles inferences in EN I 8, 1099a31b8. He argues that happiness needs external goods because virtuous activity needs external goods, and then he argues that virtuous activity needs external goods first because it needs some external goods as instruments and second because it needs other, non-instrumental external goods (some of which are objects of mere wish), for social and psychological reasons. Now he backs up his claim about non-instrumental external goods with a final inference as follows:
and on the other hand, [D] men who lack some things such as good birth, good children, and beauty soil blessedness; for [E] the man who is very ugly in appearance or of low birth or solitary and childless is not entirely happy, and [F] he would perhaps be even less so if he had thoroughly bad children or friends, or if his good children or friends had died. (EN I 8, 1099b2-6)

Is this final inference intelligible and defensible as Aristotle presents it? Again, it is tempting to suppose that Aristotle is simply asserting that happiness simply includes the non-instrumental external goods like good looks, good birth, and good children. On the assumption that happiness and blessedness are the same thing, that would make ready sense of the relations among (D), (E), and (F). But as we have seen before, it would also ignore the earlier inferences that Aristotle explicitly marks. Accord-

74

ERIC BROWN

ing to his logical conjunctions, he is trying to show that non-instrumental external goods are necessary for happiness because they are necessary for virtuous activity. That inference would be utterly unnecessary if Aristotle were assuming that non-instrumental external goods are necessary as constituents of happiness. Again, we should construe all of Aristotles inferences as he explicitly marks them, unless we cannot otherwise make sense of them that way. As it happens, we can make sense of the last inference. The last inferences conclusion is the premise of the third inference. As a premise, it is essentially a commonsense thought. How could Aristotle support the essentially commonsense claim that men who lack good looks, good birth, and good children soil blessedness? Essentially, by asking his audience to check their commonsense intuitions. Aristotle does not give independent reasons to suppose that the ordinary view of happiness requires noninstrumental external goods. Rather, he calls to mind commonsense intuitions about how good looks, good birth, and good children are fundamental to the ordinary view of the happy life, by suggesting that being ugly, low-born, or childless would make our lives less happy and that being deprived of some external goods by death might (perhaps, ) make our lives even worse off. If we assent to these judgments, then we should agree to the more general claim that the lack of non-instrumental external goods soils blessedness. The fourth inference backs up my analysis of the third inference in two additional ways. First, it is easily recast into the vocabulary of fulfilled and unfulfilled wishes. In these terms, Aristotle is securing the claim that fulfilled wishes are necessary for blessedness as the optimal fulfillment of wishes, and he does this by isolating the significance of some particular unfulfilled wishes. Second, it again demonstrates Aristotles penchant for expressing an ordinary point in extraordinary ways. In (D), Aristotle describes the importance of non-instrumental external goods in such a way as to highlight the responsibility of mens attitudes and capacities for soiled blessedness. In (E), he expresses the ordinary thought that noninstrumental external goods are necessary for happiness with an unusual word for happy, , instead of the more common . Insofar as suggests that one tends to be or is likely to be happy rather than someone who flatly is happy, Aristotles choice might suggest further that the lack of non-instrumental external goods does not necessarily undermine happiness, but only tends to. That would serve his purposes very well, whether it accurately reflects the ordinary view or not.

WISHING FOR FORTUNE, CHOOSING ACTIVITY

75

The fourth inference also finishes off a brilliant argument. Aristotles initial premises (E) and (F) are commonsense claims that motivate the thought that blessedness or happiness, the ordinarily conceived state in which our wishes are optimally fulfilled, requires external goods such as good birth, good looks, and good children. These are the very claims that challenge Aristotles narrow definition of happiness; they make it seem implausible to say that happiness is merely virtuous activity. Aristotle shows his genius by arguing, through a series of careful inferences, that the ordinary thoughts actually support his narrow definition. He is able to do this because the external goods fundamental to the commonsense claims also make a difference to our capacity for virtuous activity, or, as I would like to put it, the fulfillment of the wishes that are fundamental to the commonsense claims also makes a difference to our capacity for virtuous activity.

V. Parallel Texts in EN I 9-12 Some readers will no doubt resist my reading of Aristotles treatment of external goods at the end of EN I 8. In spite of the logical structure of that argument and in spite of the intentions that litter the transitions of EN I 812, they will continue to insist that his distinction between two kinds of external goods differentiates between those external goods that are tools for virtuous activity and those that are constituents of happiness as Aristotle understands it. For support within EN I 8-12, they will point to two texts that parallel the argument at the end of EN I 8. So I now consider both of these texts briefly. The first occurs in chapter nine, where Aristotle is discussing the puzzle about how happiness is acquired. He notes, Of the remaining goods [viz., the external goods], some belong necessarily [viz., to the happy man], and others are naturally useful and cooperative as tools. This, taken all by itself, might suggest that some external goods are necessary as constituents of happiness (cf. Irwin 1985, 95). But the sentence does not occur all by itself. It immediately follows a sentence in which Aristotle repeats his definition of happiness as a certain sort of activity of the soul in accordance with virtue (EN I 9, 1099b25-26). So Aristotle should not be taken to deny or alter his definition of happiness by referring to those external goods that belong necessarily to happiness. He can and should be taken to

76

ERIC BROWN

claim that some non-instrumental external goods are necessary conditions of happiness because they are necessary conditions of virtuous activity. The second parallel passage occurs in chapter ten, where Aristotle is trying to oppose Solons advice without denying that fortune has an impact on happiness. I have already made use of this passage in part:
When many great events occur well, they will make a life more blessed (for they naturally add adornment, and the use of them comes to be fine and excellent), and conversely when many great events occur badly, they oppress and spoil blessedness, for they bring pains and impede many activities. (EN I 10, 1100b25-30)

The explanation of good fortunes effects might again be taken to assume that some external goods are parts of happiness independently of virtuous activity. Again, though, the context tells against this reading, for Aristotle is about to say, If activities are controlling for life, as we said, then none of the blessed could become wretched (EN I 10, 1100b33-34). It is possible to understand this as follows: goods of fortune matter, but in the right circumstances, virtuous activity makes the decisive contribution (see Irwin 1985, 102). But Aristotle does not say that virtuous activity makes the decisive contribution in the right circumstances. He says that activities are controlling on the heels of insisting that virtue shines through in terrible circumstances (EN I 10, 1100a30-33), and he reminds us (as we said) of his earlier discussion of how virtue is controlling. In the earlier discussionearlier in the same chapterAristotle argues against following a mans fortunes to assess his happiness:
Or is it not at all right to follow his fortunes? For the well or badly is not in these things; rather, human life needs them in addition, as we said, and activities in accordance with virtue are controlling for happiness, while opposite activities are controlling for the opposite. (EN I 10 1100b8-11)

Here Aristotle clearly refers back again (as we said), this time to the passage at the end of chapter eight in which he argues that external goods are necessary for happiness because they are necessary for virtuous activity. So he is saying that one should look to virtuous activity, and not to the goods of fortune, to determine whether a man is happy. As if his insistence on the narrow definition of happiness were unclear, Aristotles next point is that the current puzzle also bears witness to my definition [ ], for stability belongs to none of the human products in the way that it belongs to virtuous activities (EN I 10, 1100b11-13). In its context, then, Aristotles allowance that external goods make a difference to happiness cannot without serious inconsistency say that they make some difference to the final goal independently of their effect on

WISHING FOR FORTUNE, CHOOSING ACTIVITY

77

virtuous activity. His allowance can and should be taken in either of two ways. Aristotle might have the ordinary notion of happiness in mind, for that is the idea of a life that optimally fulfills our wishes and such a life requires external goods for reasons independent of their effect on virtuous activity. Alternatively, Aristotle might trust his audience to recall the analysis of external goods he has already given, in chapter eight. If one does this, then one will know that external goods make a difference to the final goal because some of them are tools for virtuous activity, others are crucial to social opportunities for virtuous activity, and all are objects of wish whose absence causes pain and thereby diminishes virtuous activity. Either way, there is nothing in chapter ten to require that Aristotle adds external goods to his conception of happiness, and plenty of explicit evidence that he stands by his narrow definition. Once again, then, Aristotles point is that although external goods are constituents of happiness as it is ordinarily conceived, they contribute only as necessary conditions fornot as constituents ofthe practical goal that makes life worth living. In the terms I used earlier, external goods are part of the blessedness (or happiness) that sums up all our wishes, but not of happiness as Aristotle understands it, for happiness as Aristotle understands it is the more limited practical goal of any agent who has a chance to enjoy the blessedness that sums up all our wishes.

VI. Wishing Well This ends the defense of my exegetical thesis. I have argued that Aristotle stands by his narrow definition of happiness as virtuous activity when he tries to accommodate the ordinary thought that external goods are needed for happiness. On my account, Aristotles reasoning is readily intelligible in terms of wishes. The ordinary thought rests on an account of wishes in a very straightforward way: we wish for a range of external goods, and if we do not enjoy these goods, then, quite obviously, we fail to live the life of optimally fulfilled wishes. Aristotle wants to accommodate this thought with his narrow account of happiness, and so he explains that when we fail to enjoy the objects of our wishes, our capacity for virtuous activity is diminished, by psychological and social mechanisms. In this way, Aristotles narrow definition of the human good delivers the same result that ordinary reflection on happiness does, by taking account of the causal significance, both social and psychological, of wishes.

78

ERIC BROWN

But the reasoning about wishes that I attribute to Aristotle might seem to leave him open to two serious objections, and before I conclude I want to defend the merits of Aristotles position as I interpret it. To do this, I will state the objections and develop responses that an Aristotelian might give. One problem with Aristotles account of wishesat least as I present it hereis that our wishes range beyond the goal for the sake of which we should do everything that we do. We wish for friendship, wealth, good looks, good children, and the rest, but our goal is simply virtuous activity. How can this be? If virtuous activity is the goal, we are supposed to do everything for the sake of it. Why, then, would a virtuous person wish for these other goods? Another problem concerns some of the particular goods that Aristotle identifies as objects of wish. Some of them are objects of mere wish; they are beyond our power. Why should these matter at all to a life of virtuous activity? Why should we not change our wishes to lessen the impact of fortune? The two problems are related, and one might capture the underlying worry that they share as follows: on what grounds would Aristotle identify some wishing as wishing well? The Stoics provide an excellent counterpoint. This is perhaps obvious in the case of the second problem. The Stoics think of wish as Aristotle thinks of wish without qualification: wish is reasonable desire for the good. Moreover, they are restrictive about what things are good: only virtue is, strictly speaking, good. So, for the Stoics, wishes are not risky; any wisher wishes for what cannot be easily lost. But the Stoics do not think that wishes exhaust our reasonable pro attitudes. On their account, we can and should recognize other things, like health and wealth, as valuable, and we can and should prefer to have them. But a preference is a weaker attachment than a wish, and it is not subject to the same psychological ramifications or the same social ramifications (should we all decide merely to prefer health and wealth instead of wishing for it). So although the Stoic agent can share many preferences with the Aristotelian, the Stoics preferences are mere preferences and not the risky wishes of the Aristotelian. That is one central way in which the Stoic lessens the impact of fortune. The Stoics also provide a counterpoint to Aristotle on the first problem. They explain that we have preferences for things other than the good because we naturally develop preferences before we can have any notion of the good. Most of us, unfortunately, also develop preferences in unnatural ways because we are corrupted by the teachings of our society and by the misleading appearances of things. But were we to ward off corruption and develop our preferences in a purely natural way, we would come to a cor-

WISHING FOR FORTUNE, CHOOSING ACTIVITY

79

rect apprehension of the good, and we would see that desire for the good does not replace our earlier preferences. The good is rational harmony, in the cosmos as a whole and in the mind of a human sage, and so one who has developed naturally and has apprehended the good desires the rational harmony of her own preferences and commitments as her good. She faces no choice between the goodthe rational harmony that is her knowledge and virtueand her preferences; to seek harmony is to seek to maintain and act on the preferences that are parts of a harmonious set. Nothing prevents the Aristotelian from adopting the Stoics response to the first problem. Why do we wish for things other than virtuous activity? We just do. We acquire these wishes in childhood, when we are (hopefully) learning the that of virtue before we have any grasp of the because. Why should we wish for things other than virtuous activity? We should not, except insofar as our wishing for them is a necessary constituent of our virtue. Just as a Stoic needs to have certain preferences for certain states of affairs in order to enjoy psychological harmony, so too an Aristotelian needs to have certain wishes for certain goods other than virtuous activity in order to be virtuous. The point here is not that the Aristotelian needs certain external goods for virtuous activity and so had better wish for them. Rather, the Aristotelian needs to have certain attitudes in order to have the psychological makeup required for virtue because, after all, the virtuous agent has the correct appreciation of what things are valuable. This is an important point. Much of the scholarly debate over Aristotle on external goods has been predicated on what I call the thesis of independent external value, the claim that at least some external goods have at least some value independent of virtuous activity. Scholars on all sides agree about the import of this thesis. If there is independent external value, then it seems that it must make an independent contribution to happiness, and if external value makes an independent contribution to happiness, then it cannot be right to say that happiness is simply virtuous activity. So if Aristotle accepts the thesis of independent external value, he must reject the narrow definition of happiness as virtuous activity. But this is a mistake. The thesis of independent external value is compatible with the narrow definition, because the person who acts solely for the sake of virtuous activity still needs to wish for things that he values independently of virtuous activity in order to be psychologically capable of virtuous activity. I distinguish: there are the wishessome wishes are necessary for the sake of virtuous activityand there are the objects of wishes these objects need not be valued for the sake of virtuous activity. Indeed,

80

ERIC BROWN

we must wish for some objects for their own sake, independent of our virtuous activity, in order to be virtuous, and we do this even though, as it happens, our wishing for these objects has an effect on our capacity for virtuous activity. The picture here is of an agent who wishes for all sorts of goods for their own sake (and for some goods for instrumental reasons, as well), but who always chooses activity. The agents wishes are complicated, and they include mere wishes. Otherwise, the agent would not be appreciating value correctly; he would not be virtuous. These wishes might suggest to the agent a huge range of possible activities, each for the sake of some object of wish. One of them, for example, might be to bake a cake for a friend, because the agent wishes good things for his friend, just for his friends sake (see EN VIII 2, 1155b31). But the agent chooses one action out of the many possible actions by recognizing the right thing to do in his particular circumstances. So he chooses a virtuous action for its own sake, though the action he chooses might be aimed at an object of wish for its own sake. But what about the second problem facing Aristotles treatment of wish? How would an Aristotelian justify having these wishes for these objects? Why should we not simply restrain our wishes, and adopt weaker pro attitudes towards so-called goods of fortune? The initial reply might be that the Stoics misrepresent what things are valuable. Physical beauty, say, just is a good, and anyone who does not see that is mistaken. But this reply is unlikely to be persuasive. Is there no account the Aristotelian can give to bolster her claims about what things are to be wished for? A lazy Aristotelian might try to insist that the Stoic alternative is psychologically impossible and Stoic ethics mere bluff. Such a maneuver is very weak. The presence of any Stoically minded agent threatens it, and any smart fan of Stoicism will concede that the Stoic ideal is difficult. Indeed, to refute the lazy Aristotelian, one could concede that the Stoic ideal is, practically speaking, impossible in the current environment, so long as one insisted that reformed institutions could make the ideal possible. A cleverer Aristotelian might try to meet the challenge meta-philosophically. If philosophical ethics is a search for reflective equilibrium and if our pre-theoretical intuitions strongly take health, wealth, and the rest to be goods, then the Aristotelian position is favored. Indeed, some might invoke this sort of commitment to explain Aristotles actual position: because Aristotle is so keen to save the appearances and the appearances take health, wealth, and the rest to be goods, he takes the posi-

WISHING FOR FORTUNE, CHOOSING ACTIVITY

81

tion he does. But this approach does not bode well for Aristotle, since he strove for equilibrium only in a small, culturally insulated pocket of possibilities. Worse, the meta-philosophical assumptions will get no grip on those who suspect that most people are about as well off in practical knowledge as they are in theoretical knowledge. Stoics, for example, think that most people are deeply mistaken about what things are good. I think that the Aristotelian has a better justification available if she simply meets the Stoics head on. The Stoics say that psychological harmony requires having certain preferences. This might seem easy to reject. Multiple sets of preferences could be coherent, and so a coherent psychology does not seem to require any particular preferences. But the Stoics insist that human nature constrains the possibilities: we are naturally such as to have certain preferences. So if you want to know what preferences a person with a coherent psychology must have, you should study human nature and particularly natural human development. The Aristotelian can and should say that the Stoics are wrong about human nature, can and should argue that human nature requires stronger attachments to goods other than virtuous activity. This response, too, isolates something important about Aristotles treatment of wish. Many of the external goods that lead people to say that happiness depends in large measure on good fortune are objects of mere wish, not available for choosing. An ethics aimed at action rightly focuses on choosing activities. But Aristotle does not believe that we can surrender our mere wishes, and he recognizes that what we wish for matters for what we can choose. That is, we must wish for certain things as human beings; otherwise, we will not be capable of choosing actions in accordance with human excellence. And as we wish for certain things, we must have our wishes fulfilled; otherwise, our capacity for choosing actions in accordance with human excellence will be diminished, by both a social and a psychological mechanism. We can disagree with both sides of Aristotles claim, about what wishes are required for a good human life and about the impact of unfulfilled wishes. But the important point is that the disagreement is over broadly empirical questions that cut to the heart of our chances of living well.
WASHINGTON UNIVERSITY IN ST. LOUIS

COMMENTARY ON BROWN
GARY M. GURTLER, S.J.

Professor Brown presents a careful reading of EN I 8-12 as a way of defending and clarifying that Aristotle maintains consistently the definition of happiness as virtuous, rational activity. In these chapters, Aristotle raises a variety of objections to that narrow definition, but Brown indicates that in each case Aristotle subtly uses the objection, based on common opinions, to reinforce his definition. Browns introduction of the distinction between wishing for fortune and choosing activity to help resolve some of these issues strikes me as the right kind of move to make, articulating some of the assumptions that plausibly lay behind Aristotles position. Finally, I find it confirms my findings on happiness based on a reading of EN X 1-8 that also uses activity as key for unraveling the complex interrelation of the three activities. These activities, chosen for themselves by the perfectly blessed man, 1 define happiness in various degrees for Aristotle. I would like to offer two comments as a way of clarifying the argument. One concerns what Aristotle means by a complete life and the other on the scope of Aristotles narrow definition. It seems to me that discussion of the complete life has long been off track. Brown continues this by calling the phrase chronological. He seems to mean durational, whether happiness can occur in short bursts of activity or must be prolonged over a lifetime (and even beyond it). My examination of Book X led me to realize that the phrase is being used by Aristotle not to determine how long happiness lasts, but whether a life according to virtue (EN I 7) or a contemplative life (EN X 7-8) is possible for a human being, and not merely present occasionally and extrinsically,

_________
1 Brown maintains that blessedness refers to the common opinion about happiness as including all that one might wish for (p. 70). This does not seem logically convincing nor likely. The discussion in EN I 8-9 entails an overlap with common opinion, as including things that some may wish for, but overlap does not logically demand coincidence, as Brown infers. Blessedness is not reducible to common opinion, but is related to happiness, adding those elements of fortune outside ones control. From his examples and discussion, Aristotle argues that the lack of blessedness cannot take away complete happiness. One cannot argue the inverse that blessedness is possible without the kind of happiness defined by Aristotle in terms of virtue. Some common opinions about all that one might wish for, however, can easily not include happiness or blessedness, as Aristotle understands them.

COMMENTARY ON BROWN

83

but essentially and whenever appropriate. That is, the question a complete life seeks to answer is formal rather than durational. The complete side of the phrase has roots in two complementary uses. One is related to his technical distinction between activity as complete and motion as incomplete. Seeing, for example, is a complete activity not because it lasts for a long or short time, but because it is formally the same at every instant. This usage helps us to understand why Aristotle defines happiness as an activity and not as a habit or disposition, which are accidental qualifications. The other use relies on the common meaning of as indicating what is full-grown or mature. Aristotle develops this usage in connection with the whole, where both terms are opposed to the indefinite (cf., Physics III 6 207a7ff.). This usage is illustrated in his comment in EN I 9 that we do not call children happy, properly speaking, because it is necessary that happiness is both of complete virtue and of a complete life ( , 1100a5). If we look at the text where a complete life occurs in Book I, Aristotle alludes to a popular saying, one swallow does not make a spring (I 7 1098a18), nor does one fine day. Similarly, he says, neither one day nor a short time makes one blessed and happy. This is not a matter of duration, as if more swallows or more days together would somehow make it spring. 2 Spring is defined independently and not by these accidental, temporal events; spring is more an activity than a motion, something formally the same at every instance, regardless of the presence of a swallow or a fine day. Similarly, happiness cannot be an occasional or chance event, but something that occurs in a complete life. In Book I, Aristotles comment is laconic, leading easily to misconstruing its meaning. In Book X, a variation of the phrase recurs in a context where the meaning can be determined with more precision. At the point where Aristotle has defined the nature of happiness as the activity of intellect, he adds: assuming a complete length of life, for ( ) nothing incomplete is associated with happiness (X 7 1177b25-26). While the introduction of length seems to

_________
2 In discussion, Professor Brown suggested getting rid of the poor swallow and the one fine day as befogging the clear meaning of a complete life (the very question at issue). In defense, let me quote him: But there might be reason to overlook the explicit import of the Greek. First, if there were no good defense of the four inferences that Aristotle explicitly marks, as he marks them, then charity would summon forth a looser construal of his words. Second, Aristotle might offer later textual parallels that require a looser reading of this one (Brown, p. 63-64). My defense of the poor swallow rests on Browns present analysis of EN I 8-12 and my complementary analysis of EN X 7-8, leaving no reason to ignore the clear import of the Greek and chase the swallow away like some unwelcome Canada goose.

84

GARY M. GURTLER, S.J.

nod 3 in the direction of duration, the clause emphasizes the word complete. That is, it must be a complete life because happiness is itself complete. The subsequent discussion in EN X 8-9 brings out two complementary ideas about the completeness of happiness and thus also how this length of life is complete. The first is whether happiness, now defined as contemplation, is possible for human beings at all. Aristotle reminds us that intellect, its objects, and contemplation are divine rather than human. This poses two problems. One is already implicit in Book I 7, that the activity of happiness could at most occur by chance or haphazardly for human beings, and thus a human life could not be taken as happy. In this case, we would be no better off than other animals, mentioned in his discussion in Book X 8. The second problem makes explicit why one might think happiness occurs by chance or haphazardly, stated in Book X 8 as an objection: as divine, a life of contemplation is beyond the human and thus should not be pursued (X 8 1177b33). Aristotles answer to this objection does not concern duration but the fact that intellect and its activity, however small and higher than the composite, is in fact the defining part of the human, and is even the true self. Thus, happiness is neither occasional nor extrinsic, but defines that complete activity whose pleasure is proper to a human being. The second aspect of this completeness is that a human life is an integrated whole. This aspect, in fact, is confirmed by Browns analysis of Book I 8-12, where he shows that Aristotles definition is precise but not narrow. I will add to the evidence for Browns thesis from my analysis of Book X. 4 Aristotle at X 8 introduces another variation of the phrase under review, a whole life. He uses this to refer to the contemplative activity of the gods, whose whole life is blessed (X 8 1178b26), while a human life is blessed insofar as it is similar to that divine activity and the life of animals has no happiness since they are incapable of contemplation. The implication is that a human life is happy not in terms of the quantity of time spent in contemplation, but whether or not such a life can be contemplative in some real way. Contemplation for human beings falls between

_________
Michael Davis, The Poetry of Philosophy (South Bend, IN, St. Augustines Press, 1999), pp. 52, 56, shows that refers to the wholeness of a sequence of events over time as distinct from a spatial magnitude whose wholeness can be taken in all at once. A contemplative life, like a tragedy, is not a whole that can appear all at once, but as a sequence of activities, contemplative and otherwise, that is intrinsically united under the same formality. 4 The Activity of Happiness in Aristotles Ethics, The Review of Metaphysics, 56 (2003) 801-834.
3

COMMENTARY ON BROWN

85

the case of the gods, where contemplation excludes all other activities, and the case of animals, where contemplation is itself excluded. Thus, Aristotle wants to establish that a human life can be contemplative, but is not interested in determining how long one engages in this activity nor that one do so exclusively. We are now in a position to differentiate the two possibilities that Aristotle has in mind for engaging in contemplation. One is episodic, a chance event that he admits can occur, but not as part of a complete life, like the lone swallow. The other happens in a complete life. This alternative is not a chance event but part of a life that has reached a completion or perfection that can be defined as contemplative. To explain what this means, Aristotle has shown in Book X 1-5, how happiness is like pleasure, an activity that supervenes on another, specific activity, causing the agent to perform that activity with greater facility and motivating the agent to engage in that activity more frequently. In the case of someone truly happy, happiness functions as this final cause, motivating the exercise of wisdom in its activity of contemplation and making the sage more proficient at this activity. Thus, happiness is no longer episodic or haphazard, but proceeds from the stable character of the sage as wise. In the context of Book X 7-8, this also means that contemplative activity is the center around which the other activities of the sage are integrated. In a move very similar to what Brown reveals in Book I 8-12, the other activities of concern in Book X 6-8, amusements and the activities of the moral virtues, are directed toward the end of contemplation, merely as a means or in a secondary way. If someone, for example, is centered on a life of moral virtue, such an individual can engage in contemplation only occasionally and episodically, as it were. These occasional bursts of contemplation do not proceed from wisdom as a stable character, but are like the solitary swallow mentioned in Book I 7. Even worse, if someone lives a life centered on pleasure and amusement, such an individual would be incapable of contemplation, since Aristotle sees virtue as the precondition for contemplation and happiness. These alternatives help us to see that the complete life is not about duration but the defining activity that governs ones life as a human being and integrates ones activities. 5

_________
5 There is one brief confirmation of this from Book VIII 9, where Aristotle describes the polis as aiming not at a present advantage, but one for the whole of life (VIII 9 1160a2324). A society does this by the festivals it celebrates, which allow it to look at the meaning of life as a whole. This is not a quantitative summing up, but that kind of wholeness revealed in drama, which explores the human situation against the backdrop of the rest of nature and the divine.

86

GARY M. GURTLER, S.J.

We can now conclude with a comment about the narrowness of Aristotles definition. Brown has shown that in Book I 7-12 Aristotles restriction of happiness to rational, virtuous activity is not narrow in an exclusionary sense, but that his definition actually includes those goods and activities that common opinion identifies as parts of happiness. The definition of happiness in Book X has been taken to be even narrower, since only contemplative activity is identified with happiness. In a similar way, I have argued that Aristotles strict definition of contemplation as the activity identified with happiness is combined with a constant integration of the other activities that form part of the sages life. Thus, a careful reading of both Books I and X shows that Aristotle is subtly integrating external goods as well as the variety of human activities into his account of happiness as rational virtuous activity and contemplation itself. To insist that a complete life is a chronological question about how long happiness lasts misses how carefully Aristotle use the terms, complete and whole, and the precise senses he gives them in defining a human life as happy.
BOSTON COLLEGE

BROWN/GURTLER BIBLIOGRAPHY
Ackrill, J.L. 1980. Aristotle on Eudaimonia. In Rorty (ed.): 15-33. Reprinted from the Proceedings of the British Academy 60 (1974): 339-359. Allan, D.J. 1953. Aristotles Account of the Origin of Moral Principles. Actes du XIe Congrs Internationale de Philosophie 12: 120-127. Reprinted in Barnes et al. 1977, 72-78. Anscombe, G.E.M. 1963. Intention. 2nd ed. Ithaca. Barnes, J. et al. (eds.) 1977. Articles on Aristotle. Vol. 2. New York. _______ 1984. The Complete Works of Aristotle. 2 vols. Princeton. Botros, S. 1986. Precarious Virtue. Phronesis 31: 101-131. Broadie, S. and Rowe, C. 2002. Aristotle, Nicomachean Ethics. Oxford. Broadie, S. 2002. Commentary. In Broadie and Rowe: 261-452. Brown, E. 2002. Epicurus on the Value of Friendship (Sententia Vaticana 23). Classical Philology 97: 68-80. _______ 2004. Minding the Gap in Platos Republic. Philosophical Studies 117: 275302. _______ 2005. Aristotle on the Choice of Lives. In Destre (ed.): forthcoming. _______ 2006. Stoic Cosmopolitanism. Cambridge. Burnyeat, M. 1980. Aristotle on Learning to be Good. In Rorty (ed.): 69-92. _______ 2001. A Map of Metaphysics Zeta. Pittsburgh. Bywater, I. (ed.) 1894. Aristotelis Ethica Nicomachea. Oxford. Cooper, J.M. 1999. Aristotle on the Goods of Fortune. In Cooper 1999, 292-311. Reprinted from Philosophical Review 94 (1985): 173-196. _______ 1999. Reason and Emotion. Princeton. Davis, M. 1999. The Poetry of Philosophy. South Bend. Destre, P. (ed.) 2005. (As yet untitled collection of essays on Aristotle, EN X 6-8.) Louvain. Gurtler, G. 2003. The Activity of Happiness in Aristotles Ethics. Review of Metaphysics 56: 801-834. Irwin, T.H. 1985. Permanent Happiness. Oxford Studies in Ancient Philosophy 3: 89124. _______ 1999. Aristotle, Nicomachean Ethics. 2nd ed. Indianapolis. Joachim, H.H. 1951. Aristotle, The Nicomachean Ethics. Oxford. Kleingeld, P. 1993. The Problematic Status of Gender-Neutral Language in the History of Philosophy. Philosophical Forum 25: 134-150. Korsgaard, C. 1996. From Duty and for the Sake of the Noble. In Engstrom and Whiting (eds.): 203-236. Engstrom, S. and Whiting, J. (eds.) 1996. Aristotle, Kant, and the Stoics. Cambridge. Kraut, R. 1989. Aristotle on the Human Good. Princeton. Lawrence, G. 1997. Nonaggregatibility, Inclusiveness, and the Theory of Focal Value. Phronesis 42: 32-76. Lear, G.R. 2004. Happy Lives and the Highest Good. Princeton. Menn, S. 1995. Physics as a Virtue. Proceedings of the Boston Area Colloquium in Ancient Philosophy 11: 1-33. Nussbaum, M.C. 1986. The Fragility of Goodness. Cambridge. Owen, G.E.L. 1970. Notes on Ryles Plato. In Wood and Pitcher (eds.): 341-372. Reprinted in Owen 1986, 85-103. _______ 1986. Logic, Science and Dialectic. Ed. M.C. Nussbaum. Ithaca. Reeve, C.D.C. 1995. Practices of Reason. Oxford.

88

BROWN/GURTLER BIBLIOGRAPHY

Rorty, A.O. (ed.) 1980. Essays on Aristotles Ethics. Berkeley. Rowe, C. (tr.) 2002. Aristotle, Nicomachean Ethics. In Broadie and Rowe: 95-258. Scott, D. 2000. Aristotle on Posthumous Fortune. Oxford Studies in Ancient Philosophy 18: 211-229. Thompson, M. forthcoming. Nave Action Theory. White, S.A. 1992. Sovereign Virtue. Stanford. Wiggins, D. 1980. Deliberation and Practical Reason. In Rorty (ed.): 221-240. Wood, O.P. and Pitcher, G. (eds.) 1970. Ryle. Garden City.

COLLOQUIUM 3

ARISTOTLE ON
ALFREDO FERRARIN

I. The Problem of The topic of this essay is Aristotles discussion and use of . The first task incumbent on me is to explain why I am going to leave and untranslated. The fact is that these terms are only occasionally and very roughly equivalent respectively to imagination and image. For example, when Aristotle sets out the program of the De anima explaining he is going to investigate the functions and affections of the soul (I 1, 402b23), the last thing he has on his mind is to make imagination the guiding tool for his inquiry. Here the phrase means the souls properties as presented to us, or according to the phenomena, not according to the imagination. Likewise, the association of with mental image does not seem to fit the bill. While it is mostly right in its occurrences in On Memory (De memoria et reminiscentia) and On Dreams (Insomniis)with the reservation that a mental image may be convenient as a loose umbrella term but is not anyway thoroughly identical with an after-image or a memory image or a dream-image, it simply cannot make sense in other cases, for example in De motu animalium and the Rhetoric, where the pictorial or visual connotation typically associated with mental images is virtually absent. Here means prefiguration, anticipation, or recollection of the pleasant and the painful, mostly in terms of an end to pursue, which has nothing of the pictorial character some of us associate with mental images. The problem is not only the translation, though. There is an undeniable plurivocity to the concepts imagination and image which, if unattended to, jeopardizes our understanding and makes us confuse different phenomena under a misleadingly uniform name. The neglect of the rich implications inherent in a sorely ambiguous and limited vocabulary one is bound to use indifferently is the source of misunderstandings when we conflate, for example, real and mental images, say, a photo of my brother, my memory of when I last saw him or my imaging his look of surprise when I present him with a gift for his upcoming birthday. All these images are identified differently and satisfy different criteria, such as spatio-temporal individuation, modality of existence or ontological status, our interpretation or read-

90

ALFREDO FERRARIN

ing of them (codes of decipherment that are presupposed, our varying awareness of detail), relation to the space outside them, context and material medium in which they appear and the constraints thereby put on them, separation between viewer and image, what it means for us to perceive them, our psychological and affective involvement, etc. The same is true of imagination and its functions. Integrating the discontinuities of perception into a unitary picture, anticipating the possible development of a plot, or of a shape partly hidden from view, deciphering a sketchy image and interpreting it as the two-dimensional abbreviation or snapshot of an event, giving rise to a world alternative to the perceptual one, dreaming, phantasizing and having reveries, recognizing someone in a portrait, not to mention constructing a plot, envisaging or picturing one, drawing a figure, writing a poemall seem to be very different, yet not unrelated functions of the same imagination. The several modes at work are not only disparate, often they are conflicting, too, as when we oppose an escape from reality to an effort at better understanding it. However, this cautionary tale may be applied to all philosophers, and especially to the freedom of the mind from preconceived meanings that the reader, interpreter and scholar are expected to bring to a text or work from the outset. What makes the situation markedly different, and significantly more complicated, in Aristotle is that the effort to liberate our minds from later conceptions must go deeper. If, say, we are interested in an account of the historical genesis of concepts pertaining to imagination in Descartes or Kant or Husserl, we must behave like archeologists, digging under sedimented notions in search of an original ground and documenting its later transformations across layers of meaning that are intertwined and can serve as useful directions to keep in mind. For Aristotle, by contrast, the task is the paradoxical one of freeing our mind from preconceptions and constantly watching the unwarranted inferences we read into the text, while at the same time keeping all lines of flight together in view of the subsequent exploitation of sketchy, diverse threads. In order to understand the sketchy, diverse threads that constitute the first attempt at a broad treatment of we have (Aristotles) the layers of meaning turn out to be by and large modern presuppositions best left out altogether. It is not even clear that has a unity and is not a scattered set of several notions, quite vague and hard to identify to begin with. True, has to do with memory, dreams, visualization, traces of perception. Yet, many of the traits that most often define for us what is proper to imaginationfrom the integration of what is absent in perception to the function of synthesis, from exhibition in a sensible medium to an implicit form of judgment, from the translation of thought into a symbolic system

ARISTOTLE ON

91

of representations to the various forms of spontaneity and creativityare virtually alien to Aristotles and mostly the result of Stoic, Neoplatonic, Cartesian, Humean and Kantian transformations of and departures from it. In fact for us, who come after Hume and Kant, the idea seems to go without saying that imagination helps integrate perception, fills in all or some of the gaps that other faculties leave open, puts forth tentative, more or less reliable conjectures on what the senses cannot provide, and synthesizes the manifold into a unitary synopsis. From this point of view, Aristotles , where none of that seems to happen, is a remarkably narrow notion bound to puzzle the reader. What is especially baffling is the vagueness and lack of precision of its description. The first impression is that it is not, to recall Heideggers phrase on Kants imagination, homeless (heimatlos); it seems positively messy. Here genetic conjectures do not seem to help much, because the inconsistencies and difficulties that emerge at a closer scrutiny do not span across different works (for example, between the De anima and the later De Sensu), but appear within the same text, as if Aristotle often took back what he had just established in the turn of two continuous sentences. This is obvious from a cursory analysis of De anima III 3, the putatively central text for any examination of . It reads very much like a work in progress, where the main effort is directed towards demarcating from perception and thinking, except the criteria for differentiation get modified along the way, or are strikingly and suddenly revoked. Just to name some examples: Aristotle writes that perception differs from in that it is always true, while is by and large false. This cannot be correct, since perception of incidental and common sensibles may err, while of proper sensibles is correct. Also, Aristotle claims that is up to us while opinion is not; and yet in dreams is clearly not up to us. This contrast admits of tapering off, though, for up to us may mean that we give rise to it at will or that we can disregard its reference to facts, which are the touchstone of truth. But other contrasts seem more resilient: for example, differs from opinion because if we fear something and have reason to believe the object is threatening, the fear is greater, while we are not emotionally affected in imagination because we are like spectators looking at something frightening in a picture; 1 in the second book of the Rhetoric, by contrast, imagination may generate the greatest fear, and in general is the

_________
1

An. III 3, 427b 22-4. Unless otherwise noted, all translations are my own.

92

ALFREDO FERRARIN

source of the most powerful emotions. Also, Aristotles frequent and considered statement that is the same as perception albeit not in actuality or essence (e. g., Insomn.. 459a16 ff.) is contradicted at 427b28, where he writes that belongs to thinking. But the most blatant tension is between 428a3-4, where judges true and false, and 428b4 a few lines below, where differs from opinion because the discrimination between true and false is proper to the latter only, which judges also s tentative claim to truth. This chapter has represented a sort of challenge for Aristotle scholars, so that in recent years we have seen several intelligent and helpful contributions to its understanding that are well worth reading; the common attempt is at salvaging the unity of the notion from the verdict of inconsistency that has been repeatedly leveled against it, not entirely without ground, from the time of Freudenthal up to Hamlyn and Rees. 2 I personally profited greatly from these essays; but the problem I have referred to, of imposing or taking for granted modern presuppositions in our interpretation, limits the import and the results of many of them. Let me mention the following, having to do first with the misunderstood systematic function of the De anima and then with the misconstrued import of perception. Almost invariably is understood as a faculty, in an overall philosophy of mind that the De anima is purportedly meant to articulate. This may be an exaggerated scruple on my part, but I would like to stress that the talk of faculties goes in tandem with the notion of an ego or a subject that, reflectively or spontaneously, ascribes to itself cognitions, volitions and such. The subject is in turn understood as a single inner space in which all contents are equally and uniformly mental, as opposed to an external realm in which objects are in a relation to it through the medium of the body. This picture seriously downplays Aristotles hylomorphism and forgets that the De anima is not a rudimentary philosophy of mindthere is not even a word for mind in Aristotlebut a chapter of his philosophy of nature, therefore integral to the study of (enmattered forms), with the notable exception of the chapters on the intellect, which is wholly unnatural and constitutes the object of first philosophy. More importantly, because faculties are demarcated and by their nature mutually

_________
2 See the bibliography at the end of this paper. Modraks work seems to me to deserve special mention among those essays, and I share most of her conclusions. The only qualms I have are with her definition of as a mechanism for handling internal representations (Aristotles Theory of Language and Meaning, Cambridge and New York 2001, 223). I am not sure I would emphasize the internality of representation; but I am sure is neither a mechanism nor can it handle its contents.

ARISTOTLE ON

93

separate and distinct, this language is tantamount to the presupposition of what has not yet been established: that is an independent faculty, and not a phenomenon whose limits Aristotle is trying to test and circumscribe from all sides. The unfortunate consequence is that we end up driving a wedge between perception and , instead of underlining the continuity stretching between them. If in De anima III 3 cannot be described in terms of a faculty, it is not a power either, but the result of a movement, . Aristotle is striving to find independent room for it between either end of the segment along which he locates it, perception and thinking. These are described as independent powers; Aristotle describes as a process, at least genetically derived from and subsequent to sensation, not an activity, let alone an activity referring to an ego. And the process is fundamentally reproductive: the is at best a copy, not a standard or model but a derivative proxy drawing its meaning from the thing which it is meant to reproduce, the norm with respect to which it is at best commensurate, and mostly inadequate. But the more significant superimposition of un-Aristotelian concepts, going along with the separation of faculties, is the promotion of to center stage at the cost of a demotion of the work of perception. is taken by most readers as an interpretation of the sensation, which by itself is not in a position to give an informative and grounded account of itself: lends its own voice to a dumb sensation. 3 , which is, as we shall see in the next section, intrinsically connected to appearance, , especially when appearance is unclear, is then taken as the generalization and synthesis of impressions of present situations and sequences of events (Frede); 4 the interpretation of sensory content and even resolution of Gestalt shifts (Nussbaum); a loose-knit family concept explaining the capacity for having non-paradigmatic sensory experience, with regard to which we remain non-committal, skeptical and cautious, 5 and the like. The problem is that this relevance of is achieved at a cost: the impossibility to tell when an interpreta-

_________
3 Cf. Ross, Aristotle, 5th ed., London 1949, 141 ff.; M. C. Nussbaum, Aristotles De Motu Animalium, Princeton 1978, esp. 255-69 and her use of the Wittgensteinian notion of seeing-as to describe phantasia; D. Frede, The Cognitive Role of Phantasia in Aristotle, in M. C. Nussbaum and A. O. Rorty, Essays on Aristotles De anima, Oxford 1992. For a judgment calling in question the idea of as interpretation, see A. O. Rorty, Structuring Rhetoric, in Id., Essays on Aristotles Rhetoric, BerkeleyLos Angeles 1996, p. 19. 4 D. Frede, op. cit., p. 287. 5 M. Schofield, Aristotle on the Imagination, in G. E. R. Lloyd and G. E. L. Owen, eds., Aristotle on Mind and the Senses, Cambridge 1978, quotes from pp. 101 and 117.

94

ALFREDO FERRARIN

tion is needed and is called in as a discriminating and judging power ( ), and when it is not because the senses can judge by themselves (and see-something-as what it is). More generally, the price paid by these readings is that of muting the senses, making the percipient a mere passive recipient of sensible qualia, and the sensible material an indistinct manifold waiting for the mind to compare, unify, collect and shed light on what is per se obscure and indistinct. 6

_________
6 A good example of modern, when not Kantian, assumptions at play in the reading of Aristotle is Frede, who can ascribe to the power of retention and synthesis into an overall impression because she attributes to Aristotle an atomistic theory of perception, according to which sensation is of particulars only (how admittedly arbitrary her interpolation of is can be seen in her reading of An. Post. II 19 in n. 43 at p. 292). That sensation is of particulars only is a well-known Aristotelian thesis; but Aristotle adds that sense retains or receives ( ) sensible forms without matter (An. II 12, 424a24; An. Post. I 31, 87b28-30; II 19, 100a17, 100b4-5). An additional faculty is not required to synthesize the particulars of sense and generalize the single sense-data into the type of the tokens that sensible particulars are. (Among some of the differences between Aristotle and Kant on imagination, see my Kants Productive Imagination and Its Alleged Antecedents, in Graduate Faculty Philosophy Journal 18: 1, 1995, 65-92, and my Hegel and Aristotle, CambridgeNew York 2001, 287-325.) Since Frede believes that Aristotle needs an abiding self in which are synthesized, she gives in to the recurrent temptation among Aristotle readers to find the solution in a common or inner sense (p. 283) which tradition has long tried to read back into Aristotle, but which is not a genuinely Aristotelian notion (the manipulation of Aristotles notion of a central organ allowing us to perceive common sensibles and incidental sensibles simultaneously in the unity of a thingSens. 449a5-20, Som. 455a16-8, An., III 2, 425b12-426b29into a supposed theory of a common sense begins as early as Alexander of Aphrodisias (De anima liber cum mantissa, 63, 6-28) and runs through the medieval sensus communis up to the 18th century common sense and Kants inner sense). Like all good chapters of Aristotelianism, even this doctrine is rooted in tensions and ambiguities left standing by Aristotle: in De anima he writes that all senses are self-conscious and there is nothing beyond the five senses in which we can locate awareness, for the awareness of the sensible and the awareness of my sensation of it are one and the same (III 1-2). Here something unitary ( , 426b18) is required in order to postulate that the perceiver who asserts a difference among sensibles be one (426b20-1); but something like a common sense over and above the senses is denied. Later in De sensu he reiterates that the power of perception must be one numerically, albeit divided in its functions (and calls this at 449a17-8). But in De somno et vigilia he writes that there is a common power ( , 455a16) whereby one is conscious that one sees, hears, etc. This, as well as the cursory definition of the as an affection , of the common sense in On memory ( 450a10-1), openly contrast with An. III 1-2, and are as close as Aristotle ever comes to a common sense; this however is never developed and remains in all its other occurrences the correlate of a special class of sensibles (common to more than one sense, such as size, shape, number, etc.) and not an original mode or seat of sensible consciousness.

ARISTOTLE ON

95

For Aristotle the perceiver, potentially the contraries it can sense, retains in memory different sensations because it is a disposition which becomes determinate when actualized (An. III 2, 427a6-7). Aristotle compares the perceivers awareness, numerically one but divided in its functions, to a geometric point which is both indivisible (the point unites two segments: the perceiver perceives simultaneously different sensibles) and divisible (the point separates two segments which originate from it: here the perceiver operates as a limit discriminating two sensible things, An. III 2, 427a10 ff.). If the perceiver can compare, relate and referin Aristotelian parlance, judge the truth ofsensory contents, and ascribe sensation of red to this thing, it seems that the senses must have for Aristotle a much broader range of meaning than we would be inclined to attribute to them. Indeed, they are not directed simply at their proper objects, but give us a very rich content, for they discriminate and judge common sensibles in and through movement, and they even give us a sensible awareness of relations, pluralities, and connections in incidental perception (the white we perceive and recognize as the white of Diares son; in another example, we simultaneously perceive the yellow and the bitter in the bile). Perception is highly complex in that it involves awareness of sensory content in its distinctness and the ability to identify and recognize not just qualia, but complex states of affairs. The senses are not fallible and deceptive as they are for modernity, they are not the recalcitrant, passive material and instrument of a mistrustful reason setting up experiments and testing sensible instances in light of them. They give us an active and intelligent perception, not supposedly raw sense-data or the material blind manifold for intellectual unification; and they are self-conscious, in that we are aware of our sensation through the sensible things we discriminate (An. III 2, 426b10-1, 425b12 ff.). We are the abiding continuity of a disposition, making possible the formation of habits, attitudes, first actualities by repeated actualizations; our sensory potency becomes a second nature, a formed , so that once it is formed, oncein the words of An. Post. II 19the universal has come to rest in our soul, we can recognize a thing as the token of a type, a form in matter. That is, I submit, what Aristotle means with his distinction between a purely passive alteration and the actualization of the senses as a progress into ones entelechy and a change to a positive disposition realizing the subjects nature (An. II 5, 417b6-7 and 16). This should be sufficient to show that the understanding of crucially depends on the proper assessment of its intermediate position between sensation and thinking. It should not be taken to show that we must advocate a form of nominalism and forsake all attempts at giving one

96

ALFREDO FERRARIN

definition of given that all such definitions avowedly cannot take into account all the different aspects of the notion. True, what holds for one aspect or function of does not always for others; but one-sidedness is not an inevitable result. If a systematic notion is to be ruled out, a unitary comprehensive reading of the different meanings and occurrences of may not be. It seems to me that if has an identity at all, it is not a definite one, but the more open-ended, unfinished, shifting yet not entirely equivocal identity of a construction site, in which Aristotle provides building materials, scanty instructions and bare contours, but leaves it to us to fill in the details. Seeing in such a site a finished building is our inference, for naturally our imagination tends to complete what seems incomplete and only sketched, and attributes a definite meaning to the structure; but we must watch this spontaneous and unwitting stepping over boundaries in our interpretation. Compared to his notion of , the issue of the intellect is always going to admit of several conflicting interpretations because of its mysteries and , but is not a structurally incomplete and indefinite theory. In the following sections I propose not to rely on De anima III 3 alone, but to bring in the entirety of Aristotles texts on , especially the Parva Naturalia, in order to look for the possible unity of this notion. While it would be impossible to analyze them all, they provide a much needed context for the question of the cognitive role of in An. III 3. After discussing the relation between thinking and images, I want finally to contrast the results with what at first blush appears as a new element that forces us to revise the whole context, deliberative and the explanation of teleological conduct leading to action, especially ethical conduct. I am interested here in discussing select passages from the Rhetoric and the Nicomachean Ethics in order to evaluate the freedom of . II. and Perception (De Anima III 3 and Related Texts)

When Aristotle introduces in An. III 3, he immediately mentions that is causally derived from perception. (Hobbes, translating the from the Rhetoric, famously expressed this as decaying sense in Leviathan.) Aristotle then proceeds to demarcate its function from (supposal, including different forms of thought like science, opinion, practical wisdom) and perception. First, differs from opinion. Aristotle argues that while it is not in our power to form opinions as we will, is an affection which lies in our power

ARISTOTLE ON

97

whenever we choose (427b19-20). He means by this an arbitrary, deliberate ability to visualize ( ... ), to bring before our eyes. The obvious implicit assumption here is that images, even if caused by sensation, become independent of it: we have them in our soul, and can use them whenever and however we please, unlike perception, which needs a sensible to be actualized. This does not mean we have here a spontaneous productive imagination. We can visualize at will absent things, in their residual images, and recalling them is subject to a deliberate intention. But there is no discussion of the power of combination of different images into new ones. is certainly not a creatio ex nihilo, or even a shaping power. And the closest it will explicitly get to that is in recollection, which is no more than the capacity for playing with and arranging at will given images in accordance with a design. Another reason why differs from opinion is that , unlike opinion, does not involve conviction ( ; we are not afraid of an act of imagination unless we ascribe to it an actual danger, unless, that is, we see in it more than imagination). Conviction is amenable to persuasion, and therefore belongs to rational animals who can listen to reason, while is common to most animals. As I said, this is problematic if we compare it to the Rhetoric and the idea that imagination is responsible for the strongest passions (e.g., fear at II 5); it is also unclear, because it seems as if were here non-committal, not exactly alternative to reality but unbothered by its constraints. But that would not be a necessarily universal inference, because there seems to be a radical difference between human and animal : animals need a reliable to move about their world. For them, but not for us, is coupled with a conviction of sorts. In and through , most animals have a way to negotiate what is absent, e.g., by making it present, prefiguring it, as an end to pursue or an evil to avert. But rational animals can make negotiating with absence, in the form of persuading to desire or fear goals that are beyond the immediate realm of perception, an independent task, enterprise, object (even, as in the case of rhetoricians and sophists, the object of a profession). Possessing reason in this case does not make human beings closer to truth, on the contrary: we are prone to making up stories, to telling ourselves lies, subject as we are to delusions and self-imposed beliefs depending on the most diverse motivations. Regardless of what we think of the different directions in which this point leads us with regard to truth, the theme of veridicality has silently made its appearance. And it wont leave the stage from now on. differs from sensation, too. Aristotle adduces five reasons why.

98

ALFREDO FERRARIN

(i) We have dream-images even while sensation is not going on, which shows that may be at work when the senses are not. 7 In this sense exceeds perception. (ii) Conversely, perception exceeds because it belongs to all animals (the grub is Aristotles example of a limiting case), while does not. (iii) All sensations are true, but most are false (428a12-3). Basically Aristotle means to say that because exceeds perception, and does not take its bearings by the relation to the world, it is more likely to err. Here is clearly seen in its freedom from the responsibility for veridicality. That all sensations are true does not square with what precedes in the text, but expresses the fact that sensation stands or falls with its relation to the things external to it: sensation is intentional, it refers to independent sensible things. Here it is obvious how essential it is to give up the idea, found time and again in the literature (and in Aristotle himself), that the only power capable of asserting truth and falsity is judgment as the combination of notions: if that were the case, we would miss the complex theory of book II of the De anima regarding the truth of perception, from the incorrigible one of proper sensibles to the discrimination of common and incidental sensibles, as well as the truth at play in the

_________
7 Here Aristotle treats seeing images in a dream as essentially different from seeing proper (427b7-9). In De somno et vigilia instead he talks about the functioning of the five senses regardless of their reference, and separates the sense of touch, which is the seat of wakeful consciousness and guarantees contact with reality, from the other senses because he is interested in showing how sleep suspends the sense of touch while the other senses may go on working. This implies that in sleep we do not stop having sensory, especially visual images; we are however incapacitated, for we cannot test their solidity and depth, their integrity, their being more than surface. In other words, what is ruled out in sleep and dreams is the possibility that our senses operate jointly, establishing and verifying the truth and falsity of their sensible representations; therefore the ability to recognize an image as an image, i.e., to distinguish between an image and a likeness, is out of the question. Here touch is the requisite for wakeful consciousness in that it keeps the senses in check and allows me to step out of my situation and verify whether or not I must assent to its appearance. This is as good a description as it gets to the situation of the prisoners at the bottom of the cave (and the reason why Jacob Kleins interpretation of as double seeing is wrong). The Stoics, especially Zeno and Cleanthes, will exploit this distinction between being presented with images and assenting to them: is ruled out in dreams. The image in a dream cannot refer to a thing external to it, for it is the thing; I cannot contextualize it. This is a different way to arrive at the same situation as that plaguing the lunatic Antipheron of Oreus, who hallucinates, i.e., treats his images as if they were memories, and is thus unable to tell an image ( ), free from the constraints of reference, from a likeness ( ), tied to the events that generated it in experience (Mem. 451a9-11).

ARISTOTLE ON

99

intellection of indivisibles at An. III 6 and Metaph. Q 10. 8 As I said, (iii) is a problematic point, and strangely Aristotle does not add a single word here. He passes on to the next difference. (iv) appears as a reflection on the consequences of a proper command of language and shows Aristotles determination not to depart from ordinary experience: I do not say I imagine it is a man if the perception is clear, but only when it is not distinct (428a13-5). I imagine and it appears to me become equivalent. then is here brought in to express a non-committal interpretation of an indistinct sensory experience. (v) Finally, we still see even with our eyes shut (428a16), which refers to the inertial force of images, which endure when perception is over. Elsewhere, Aristotle likens this movement to that of a projectile or a javelin, which moves on even if the physical force impressed to it is no longer in contact with the athletes arm and the javelin (Insomn. 459a28 ff.). Let me go back to (iv) for a moment. Whether this consideration arises as a reflection on linguistic practice or not, it definitely shifts our attention from our power to deal with images, residual or not, to appearance itself. If so far comprehended aspects of what we call imagination, now it is decisively a way to respond to indistinct appearance: an impression we do not trust. But it is confusing that all distinction between my response and what appears vanishes, for the word may refer to either my impression or the thing that appears to me. If does not normally add to the discriminatory power of perception, which is by and large self-sufficient for Aristotle, yet, as I anticipated, in certain limiting cases it is a sort of interpretation of appearances that are intrinsically hard to read (owing to problems either with the perceiver, e.g., illness, emotions, or with the distinctness of the object, e.g., distance, conditions of light and visibility, etc.). In this case it is a , a discriminating power ( , , Aristotle writes at An. III 3, 428a3-4), but very likely to err or go astray; and here the senses are a more passive vehicle of impressions, precisely insofar as they dont take a stand on how perceived things actually are. In this case alone does Aristotle contravene Wittgensteins dictum that you cant perceive and imagine the same thing at the same timeexcept the meaning of imagine must be properly understood as a form of conjecture. This new link between and appearance ( ) is what motivates Lycos rendition of as being ap-

_________
8

See my Hegel and Aristotle, 161-71.

100

ALFREDO FERRARIN

peared to, and justifies Schofields interpretation of as a loose family concept judging non-veridical experience in a sceptical way. Aristotles obvious reference is to Platos use of the term in the Theaetetus, Philebus and Sophist (and before that to Protagoras idea that perception is what seems to me, Metaph. 5). And the conclusion of this effort at understanding apart from opinion and perception is that it cannot be a blend of the two, as it was in Plato. Were an opinion corresponding to and indistinguishable from the sensation of the same object it is an opinion of, we could not have a true belief with regard to something which has a false appearance. Even if the sun appears a foot across, we judge it to be bigger than the inhabited earth; and the judgment does not alter the way the sun appears to us. In other words, here Aristotle wants to avoid reducing to its neighboring functions or saying that to imagine is to form an opinion exactly corresponding to a direct perception (428b2-3). Here the of the sun is the misleading appearance we verify and confront with our pondered judgment. But s role is the admittedly misleading one of presenting us with a deceiving image; and it is itself a judgment, which has nothing to do with mental images, or with imagination (save insofar as it is a sort of conjecture, a guess). It is a judgment on appearance and includes an inference and estimation of quantity and distance, therefore an interpretation and a (I do not say that the sun appears small, I say it appears to be of this magis a and nitude 9 ). In the case of indistinct perception becomes interpretation; and when judges, it is not in the form of an overlap between mental images, or pictures, and reality. does judge what appears ( ), but we preserve the freedom not to assent to this appearancewhich means that we can bracket, put out of play and override . There is thus a discrepancy between appearance and opinion that makes it possible for us to draw back from appearance and set up a distance between it and us. It is as if a realist intentionalist theory of perception suddenly became in one of its applications phenomenalist and gave rise to an otherwise virtually absent gap between givenness in our subjective experience and in itself. Along with this discrepancy, there is a conflict between the claim to truth put forth by and the resolution, the decision about it by opinion ( ). For in On Dreams Aristotle returns to the example of the sun, and argues that the controlling sense ( ) does

_________
9 I pass over the complications regarding the estimation of magnitude by the senses or by at Sens. 448b 14 ff. and Insom. 460b16-20 and the conflicts between these texts and the De anima.

ARISTOTLE ON

101

not judge these things by the same power as that by which images occur. This is proved by the fact that the sun appears to measure a foot across, but something else often contradicts this impression ( , 460b16-20, transl. Hett). That the controlling power contradicts is repeated at 462a6-7: in sleep I cannot contradict the impression in which my consists because I cannot take the image as image, and thereby judge its claim to truth. What sleep suspends is my capacity to see images as representational. If were left to itself, to its tendency to disconnectedness, its images would be all we have; because it is not, because it is kept in check and is validated as relational, it has a unique cognitive use precisely qua capacity to represent. Now this , this ability to contradict , sends us back to another trait that is behind Aristotles theory of truth, which we could express as follows: the greater the contact (howsoever understood) with presence ( ), with the thing, the greater the truth. If contact is the guarantee of faithfulness, removal and distance are the main causes of falsehood, and is all the more likely to err the more it is removed from the thing and connected with absence. At times what Aristotle says about falsehood is confusing, for it covers many shades along what is for us a heterogeneous spectrum ranging from arbitrary or spontaneous representations to inadequate appearances and misleading sensory impressions. In fact, we may find Aristotles rather sweeping claim that are for the most part false (428a11-2) baffling; after all it seems reasonable to say that at least some images, e.g., in dreams and paintings, are neither false nor true. That for Aristotle they are instead mostly false is consistent with the first of the several meanings of false in the Metaphysics ( 29, 1024b23-5), where he calls false paintings and dreams because they are not taken in themselves, but as representing a reality that is not. is therefore always understood and evaluated in terms of its basic characteristic, its claim to truth and its goal of reproducing faithfully the thing for our consideration. The partial exception to this cognitive import of is the treatise On Memory, where the contrast between image and likeness relieves from the responsibility for reference and truth and leaves it to likenesses, as we will see in the next section. In any event, this theory of truth as contact returns by the end of the chapter, where Aristotle distinguishes between the different images left over from the respective modes of perception. As we know, perception of proper sensibles is always true, while with common and incidental sensibles we may go wrong; likewise, a deriving from proper sensibles is true whenever the sensation is present, but the others may be false

102

ALFREDO FERRARIN

both when it is present and when it is absent, and especially when the sensible object is at a distance (428b28-30). It is not clear what conclusions we should draw from An. III 3. There are relatively inconsistent accounts of , but it seems clear that Aristotles preoccupation is mainly directed at a form of representation of things in their absence which is the result of a prior perception and which may be used by thinking and memory (or, as for animals, as a way to orient themselves in reality). His definition is very broad purposely, I suggest, as if to adjust to and accommodate the diverse understandings we have seen. He writes: is the movement by which we say that an image occurs in us (428a1-2). It is the process by which images are left over, presented, visualized, recalled, held fast as possibly true; and these images are the traces of prior perception. The nominal definition of expressed at the beginning of the chapter is repeated and confirmed at the end, where Aristotle writes that imagination must be a movement produced by sensation actively operating. Since sight is the chief sense, the name is derived from (light), because without light it is impossible to see ... Imaginations persist in us and resemble sensations (429a2-6, transl. Hett). This characterization has led interpreters to emphasize the visual connotation of images and the resemblance model for images. I would like to say just a few words to dispel what may be a myth. The visual paradigm is not as exclusive as in most of our philosophical tradition: the residual in us must be able to refer to all sensibles and their respective senses. It is not necessarily a visual trace, but can comprehend the memory image of everything the perception has left in us; and this ranges from proper sensibles (the of a string quartet is no oxymoron) to common and incidental sensibles, including the awareness of the relation of this white to Diares son, ascriptions, comparisons, etc. We saw that the alternative is to postulate the action of in order to generalize, attune and smooth out singular perceptions. If instead the trace of perception includes the whole of what I have perceived (and I perceive the , the ratio or form of the sensible, not isolated bits of qualia), there is no need to postulate an additional faculty, because in the experience of a singular token I already have access to its type. Another criterion that has been taken, I submit, too literally is the requirement that images resemble the percepts from which they derive. True, Aristotle does speak of , and his theory of perception does need a likeness: the virtue of likeness is that it fastens my image to the thing, and thus secures reference as a natural relation, in a way that is not conventional, arbitrary, easily changeable. However, it is a caricature to

ARISTOTLE ON

103

make Aristotle say that I must harbor in my soul a picture reproducing as if on exact scale the thing I saw. I think it is an exaggeration to attribute to Aristotle this illusion of immanence, in Sartres words, or ghost in the machine if you prefer Ryles, by which I relate to things through the medium of my fixed mental representations of them; it shows again that the visual paradigm has taken over as the exclusive model. Likeness is not visual, nor is it wholesale; it can be a matter of degree. For example, I can refer to a thing by way of an arbitrary truncated symbol, as in mnemotechnique, whereby I acquire a certain freedom towards the thing I experienced and want to recall to my mind. So we must carefully interpret the theory of reproduction of sensory experience in memory as if by substitution. The theory of images as pictures of prior experience has mainly been derived from the treatise On memory. To a fuller examination of it we can now turn. III. and Reference (Memory and Thought)

As we have seen, the definition of is general enough to accommodate visualization, memory, dreams, teleologically directed activity. s main but not exclusive effect is that of enabling us to visualize and make present to ourselves absent things. represents, in a broad sense stands for, the thing that generated the : this is implicit in the causal theory we have seen in De anima III 3. The image is the thing in its absence, the thing as a representation. Its advantage vis-avis the perception of the sensible thing is that the image is not bound or limited to the immediacy of givenness. If, as I said, presence is the necessary touchstone of truth we rely on, still presence is said in at least two ways: the presence of the sensible in perception differs from this now vicarious presence to our consideration it enjoys in . The obvious consequence is that in we are freer from givenness in that we can give ourselves, we can reproduce, presence out of deliberate intention. Yet, while an image is always a form of presence, it is not always the vicarious presence of the thing, for it may be taken in different, nonrepresentational senses: for example, as a simple image I entertain while phantasizing or dreaming. Even if the first origin of this image is in perception, here the presence of the image is not meant as a likeness. When it is so meant, we can have a memory image related to the experience from which it derives, or a likeness in more general terms, i.e., as a way to refer to a concept (the image of Coriscus may conjure up my friend for me or represent an illustration of man to my mind). The distinction the treatise

104

ALFREDO FERRARIN

On Memory introduces is a distinction, following my point (iii) above from An. III 3, internal to the sense-derived , which becomes a representational image (an ) or a disconnected according to our consideration, depending on what we mean to make of it. But let me proceed with order. In On memory Aristotle asks: if the image presents us with the thing without its physical presence, when I consider an image isnt it paradoxical to say that I am intuiting something absent? And in what more precise sense can an image come to be considered a likeness of the thing? When I remember, I have an image of the thing which I treat as a likeness ( ) of that thing: Why? Perception impresses a transcription ( , Mem. 450a32) in memory. When one remembers, is one contemplating the present affection, or that from which it is derived (450b12-3)? If it is the present affection, then remembering and sensing would be the same, and we could not remember anything in its absence. If I failed to consider my present image of Coriscus an image of Coriscus, I would be presenting myself with a new image (Insomn. 461b23 ff.): if it were not a likeness, all images would be different objects of ever renewed contemplation. A world without a stable identity is a meaningless world. When I conjure up an image of Coriscus, my imagination must then work together with my memory, which is the disposition ( , Mem. 449b25) constituted by our repeated distinction between image ( ) and the image-as-a-likeness ( ), and by the repeated consideration of an image as a copy of the thing to which it refers. When I remember, I must be seeing and hearing what is not present (450b19-20). Differently stated, memory is a real presencing of absence. But unlike imagination, which only makes me visualize images, memory presences absence qua absence. Unlike in imagining, in remembering the images are always considered as deriving from an actual perception: we are aware of having experienced the thing before, hence also of the time elapsed, and the image we envision now is regarded as a likeness precisely because of our consciousness of its temporal connotation. As copies, images can function in reference to things. Whether we regard it as likeness or as image, the image is one; what changes is not its relation to the original, as it would be for Plato (Soph. 232a1-235c7), but our different thematization, our intentionality, our mode of contemplating or considering the image ( , 450b31). If the sensory content pictured remains identical in the image and the likeness, still the theoretical import differs, because in the case of a memory the content is not perceived in

ARISTOTLE ON

105

itself, but is visualized as a representation of the past experience of the thing: it has the value of a temporal, relational index. This index value is less apparent and much less discussed by Aristotle, but becomes crucial when it is non-temporal and is used by thinking. For thinking, I suggest, uses images as particular examples and illustrations of intelligible forms. Which is to say it sees the universal in the image, but disregards precisely what makes the image an image, its particularity. Thinking ignores whatever is subjective and particular about the image and uses it as a representation of the form we think in it. Memory, Aristotle argues, is always of images, even when we remember intelligibles (450a12-3). That we think in images is a very well-known thesis (Mem. 450a1; An. I 1, 403a8; III 7, 431a16-7; 431b2; 8, 432a10). It follows from the necessity that thinking have a present object of thought. Thinking needs an intuition filling its thematic consideration just as sensation needs a sensible thing to be activated; to the thinking soul images ( ) serve as sense-images ( ) do to perception (An. III 7, 431a14-5; cf. 8, 432a4-11) 10 . I need to place the thing before my eyes ( , Mem. 450a4) and consider the absent thing in its image as if I saw it ( , An. III 7, 431b7). However, the exact nature of the dependence of thinking on images is a matter of dispute. While I believe that Simplicius, among others, was too quickly dismissive when he said that Aristotle really meant that imagination is only required by the discursive soul, since the thinking soul is in identity with its object and not related to it through otherness or images, 11 I also think that to make thinking dependent on imagination would be an undue restriction of Aristotle's position, at odds with An. III 5 and the denial of the corporeal basis of thinking. Let me explain. Obviously intelligibles are enmattered and immanent in sensibles, and therefore they are apprehended on the basis of the images left over from our sensation of them (An. III 8, 432a3 ff.). But, contrary to a widespread belief, Aristotles point that we cannot think without images does not simply translate the content of images into thought. An image is both inevitable and prior for us, for our apprehension and memory; but it is not prior by nature. Thus Aristotle writes that we cannot learn or understand anything without images (for images are like sense-images, except without matter, 432a9-11), and that concepts are not images but are not without images (432a13-5). I take the last quote to indicate that even if we cannot

_________
10

For the meaning of

see Sorabji, Aristotle on Memory, London 1972, pp. 82-

3.
11

In Libros Aristotelis De Anima, ed. by M. Hayduck, Berlin 1882, p. 267, ll. 30-32.

106

ALFREDO FERRARIN

help picturing the pure intellect, or being, or the divine to ourselves, we need not rely in our thought of what is not intuitive on what is no more than an analogical representation, which may be of help to picture, memorize or communicate, but does not adequately capture any of the essence of our object of thought. Images are inevitable because of the finitude of our thinking, which first must learn, and then knows. But thinking in itself is actual and free and does not need images, as we see in the case of the prime mover, thinking itself without the aid of images; only our thought needs images, because we need to recall, give our intelligible gaze, present ourselves with images to substantiate our thought, which for us always begins from (but does not necessarily end at) the sensible. When the intellect is embodied in us, thinking in images is what its use amounts to. But per se the active intellect is prior, and is independent of the representation of the sensible. It follows that our thinking, true to its impassive, separate and unmixed nature, must consider the sensible forms we learn in experience as no more than the illustrations of the intelligible essences enmattered in them. It follows from the principles of Aristotles noetics that the image only has an exemplary function, and that imagining and remembering are the subjective acts of the presentification of things whose content is distinct from and irreducible to their image. Even if this thesis is not discussed in a manner and to an extent comparable to the distinction between and likeness, it is nevertheless stated in no uncertain terms by Aristotle when he distinguishes between objects of memory properly so called from incidental objects of memory ( , Mem. 450a27); and when he argues, in a passage which has rarely attracted the attention of commentators in this context (An. Post. I 10, 77a1-3), that when a geometrician draws a triangle the figure only has an illustrative function. The geometrician contemplates or sees, as it were, the essence of the triangle in light of the image (we have seen the connection between sight, and at An. III 3, 429a3); but the relation between the two is not direct or necessary, let alone causal. The image is irremediably particular and determinate because through it my imagination reproduces what appeared to me in perception. The image can help me understand or remember aspects of the thing; but for Aristotle I do not simply translate into concepts what is present in an image, for there is nothing universal about an image as image. It is only when I regard it as the particular occurrence of an abstract form that it acquires for my intellect the value of an index and reference to an intelligible essence. And that is because, once again, an image is not an unchangeable content, but can vary according to the mode of consideration, so that the same content, say the visual image of

ARISTOTLE ON

107

my brother, may represent a fleeting image of him I entertain now, my memory of him, or give a concrete exhibition to my abstract consideration of man. In the last case, the image is a vehicle and an index. Here is a central passage showing what I mean:
We cannot think without an image. For the same thing occurs in thinking as in the drawing of a figure. There, although we do not make use of the triangles determinateness of quantity, yet we draw it with a determinate quantity. Similarly in thinking, although we do not think of the quantity, yet we place a quantity before our eyes, but do not think of it as of a quantity (Mem. 449b 24-450a 6).

Differently stated, thinking works against the particularity of images. It often resists misleading appearances and thus contradicts the imagination, as we saw; likewise, it goes beyond the subjective appearance of because it contradicts the subjective particularization of forms in experience. What is not stated by Aristotle is a further, crucial point, that makes the transmission model ( , from image to concept), towards which Aristotle entertains more than a simple gesture, fail: a sensible form and an intellectual essence simply cannot be identical. True, the sensible form is a ratio or ; but the logos of the essence, expressed by the definition, is intrinsically non-sensible and can hardly be arrived at by a conceptualization of the sensible form left over in the image. By treating image, sign and name as continuous and homogeneous, as he often does,12 Aristotle cannot satisfactorily take into account the gap between image and concept, between intuitive and abstract, which is deeper than he thought. IV. Deliberative and the Freedom of Practical Imagination

In the passages devoted putatively to the picture we get is of a process causally generated by perception, a process which has a derivative status. It seems that in their use for thinking images have basically an instrumental function, because of their irremediable particularity, but also, more generally, because of the fact that an image reproduces to our mind whatever state of affairs generated it, and is thus removed from the original from which we must take our bearings when understanding and truth are our goal. Unlike modern mind, the Aristotelian intellect does not legis-

_________
See for example the theory of meaning as an affection of the soul in Int. 1, 16a3-8, or the theory of recollection in the second part of On Memory.
12

108

ALFREDO FERRARIN

late over nature, which it only purports to understand and attest. If the becomes the form it thinks, it cannot have a form of its own (laws, categories, principles), but must adapt itself in a plastic way to its several objects, which invariably constitute the point of departure and the guide for it, including the criterion by which we distinguish the functions of the intellect itself: for example, it is the different modal status of the object at play (the necessity of the immutable vs. the contingency of what can be otherwise) that allows us to differentiate between scientific and calculative intellect and their respective criteria for truth. In other words, the tries to follow the nature of the thing; at best images are good bearers or reminders of them, but have no value in themselves. However, when we come to other texts (Rhetoric, De motu animalium, Nicomachean Ethics), the large picture again needs refocusing. In De motu, in the course of his explanation of the principles of animal movement, Aristotle argues that the object represented as the goal is the unmoved mover of the action (701b33-4; cf. An. III 10). This representation is formed as the thought or image of the object to pursue or avoid; and what counts here is how the object appears to me. We are sent back to the meaning of as impression, except that it is now the distinct impression of the end that is at once the starting point for action, and therefore does not derive from a prior perception, but originates a movement. For the affections suitably prepare the organic parts, desire the affections, and the desire (MA 702a18 ff., transl. Nussbaum). Very strikingly, in light of what we have seen and the derivation of from perception, Aristotle continues: and comes about either through thought or through sense-perception (ibid.). This move was actually adumbrated in chapter 10 of De anima III, where Aristotle spoke of a deliberative or rational ( ). This deliberative follows upon reasoning and only belongs to calculative animals, human beings. Here , based on the evaluation of the relative merits of different choices, combines the diverse images into one representation (this is the only occurrence of a synthesis, at 434a10), and thus helps the internal weighing of options leading to a deliberation. In this sense is the ability to see the particular in light of a goal, in view of possible choices. We have here a definitely new element: while so far was investigated in the traits that all animals possessing it shared, now we have a specifically human power to represent ends, which evaluates a particular state of affairs (of which it can have no science but at best a correct opinion) and is aimed at a deliberation. Here the images are clearly not visual, but rather forms of expectation, hope, fear, in which I prefigure to myself

ARISTOTLE ON

109

a future good. The of a revenge appeasing my rage and giving me pleasure (EN IV 11, 1126a20-3), of being reunited with my beloved, of being hurt in an accident, of having been abandoned by a friend, are phantasies, memories or anticipations I may savor, in which I may indulge, linger or even brood, feel frightened, bitter or sad; they are as such the source of an often greater pleasure (or pain) than the actual happening itself. These representations are not the pictures of remnants of sense, but psychological conditions which represent states of affairs that are not given and are mostly projected onto the future (or past). And, more importantly, they are not derived from sense but generate passions of their own accord. The practical imagination then has a very different function from the residual derivation of a trace left from perception in our memory; it orients our desires, and moves us to pursue an end. In other words, instead of being a process caused by perception, it is now the cause of movement. This can be best seen in the Rhetoric, where passions are described in their mental nature, as expectation of the pleasant and painfuli.e., as representations, which means in their complexity, involving hope and fear, evaluations and judgments on relevant states of affairs. Like perceptions, passions are discriminating; they are not blind and irrational impulses but are based on opinions and beliefs, and can change according to how beliefs change. For example, my fear depends on the danger I am convinced I perceive in something. When my conviction regarding the reality of the danger changes, so does my fear, which either vanishes or is addressed to other objects. In the new conviction, the light in which things appear to me changes; I will see things differently from now on, I will shift my perspective and the global image in a way that will alter its borders, its color. Like imagination, passions are in this sense more or less rational, to the extent to which they are permeable, or impervious, to rational arguments and persuasion. Aristotle, who does not make this trait of central at all, resists the move tempting later philosophers, from Zeno and Chrysippus to Plotinus (and Kant and Hegel), of splitting into a sensible and an intellectual mode, eventually making all human tendentially rational ( ) 13 . As Hegel would say, because thought is implicit in everything human beings feel, dream, imagine or do, we must recognize even in human beings lower functions diverse forms of rationality at work. Human imagination would then be unqualifiedly rational. Aristotle

_________
For Zeno and Chrysippus, see Aetius in Stoicorum Veterum Fragmenta II 83 and Diogenes Laertius, Lives VII 51; for Plotinus, see Ennead IV 3, 30.
13

110

ALFREDO FERRARIN

resists this, because he is more interested in the continuity between humans and animals than in the gap and demarcation of their functions and affections. But, if his entire theory of is relatively unprecendented, it is undeniable that he introduces in deliberative a groundbreaking aspect that others will exploit more exclusively. As I said, what matters in animal movement is not how things stand, but how they appear to me; it is how I see something that explains why I am afraid of it. Opinion is an integral part of the passion; because passions change according to my opinions, they can be at least in part constituted rhetorically. The rhetorician exploits the fact that what pleases in its presence does also in its absence, in memory and expectation (Rhet. I 11, 1370b10-2). This is why it is pleasant to evoke the loved one in his or her absence. Likewise, to continue with Aristotles examples, winning generates an image ( ) of superiority that gives me pleasure (1370b346); the image of revenge taken, if only in thought or in dreams, for a slight suffered generates pleasure (II 2, 1378b8-10). Fame generates the impression ( ) of possessing the qualities of an excellent person (1371a8-10); being loved is pleasant because it sends me back an image of myself as good (1371a18-9). Fear is a pain deriving from the image of a forthcoming evil ( , II 5, 1382a22-3); hope is the image of an imminent good (1383a17-8); and shame is the imagination of haunting disgrace (II 6, 1384a21-2). These images (the one word for images, imagination and impression in the previous examples is always ) are internal representations whose contours are not definite, and which therefore leave room for the rhetoricians influence. In the end it is not only the contours of the images, but my very passions, that can change, and with them the actions for which I will be responsible. And I dont need an outside persuader: a dialogue of the soul with itself will do just as fine. If this combination of three basic elements: image or representation, fear of evil and hope for good, giving rise in its permutations, additions and subtractions to different passions of the mind, reminds you of Chapter 6 of Leviathan, I think you are right. Hobbes drew his inspiration for that chapter from his study of the Rhetoric by the philosopher he much maligned and hated. This reference to Hobbes only helps me introduce one last point. In the controversy with Bishop Bramhall, Hobbes retorts to his opponent that man may well be free to do what he hath a fancy to do, though, continues Hobbes, it be not in his will or power to choose his

ARISTOTLE ON

111

fancy, or choose his election or will. 14 I am free to do whatever I please (provided I have the power), except I am not free to please whatever I want. My imagination determines my wants and desires, and is itself determined, out of my control. Aristotle does consider this problem in the Nicomachean Ethics. We have seen the link between appearance and representation. This becomes the heart of a pressing question in ethics: if all we do in action is look to the appearing good, that is, to our representation of the good, and this is the only available deciding factor, we might be inclined towards the view that appearing does not depend on us, and therefore a certain choice is not imputable to us but to an inevitably partial judgment on what could have appeared differently had I had a different perspective or vantage point. In that case the judgment would be contingent, and the ensuing action, if blamable, pardonable. I am, after all, innocent with respect to what I perceive. Interestingly, Aristotle argues to the contrary, again because perception is understood as a disposition shaping my identity and is thus more active, more educational and involves more responsibility than for many modern philosophers. He writes:
But someone might argue as follows: All men seek what appears good to them, but they have no control over how things appear to them; the end appears different to different men. If, we reply, the individual is somehow responsible for his own characteristics, he is similarly responsible for what appears to him (EN III 7, 1114a31b3, trans. Ostwald).

Even appearance is part of my ethical responsibility. The images that move me to action have moral connotation and significance; they show the kind of person I am. Images thus have more than a causal function. They can also persuade me to feel a certain passion and stabilize a certain disposition of tendencies in me. If my desire is stimulated by the representation of the end I have, my desire is not atomistic and disconnected from other desires or from me. Both my desires and my images are combined in a thoroughgoing unity defining my individuality, my mental disposition. Because what I perceive as morally salient for my choices depends on my desires and representations, I must pay the keenest attention not only to the right education of my passions, but also to the cultivation of the good

_________
14 Of Liberty and Necessity, in T. Hobbes, The English Works, ed. Molesworth, London 1839-45, vol. IV, p. 247. On imagination in Hobbes, see my Artificio, desiderio, considerazione di s. Hobbes e i fondamenti antropologici della politica (Pisa 2001, chapter 6), and Imagination and Hobbes: Distance, Possibility, and Desire, in The Graduate Faculty Philosophy Journal 24: 2, 2003, 5-27.

112

ALFREDO FERRARIN

dispositions of my imaginative life. To be more precise, there is no real distinction between these two forms of education. Think of books I and II of the Nicomachean Ethics and the circle between exercise and disposition: the activity derives from a disposition, yet may also establish one, through intentional repetition. There is an analogous circle between character and imagination. The disposition here is the abiding habitus thanks to which I relate to my images. So my images may be quite ephemeral, but they may also leave an abiding trace in me, a sedimented attitude and habit of imagining (and conversely may derive from such habits). And that is a habit which I contribute to establishing, and for which I am ultimately responsible. If I imagine something that gives me pleasure, and this pleasure in the absence of its object moves me to act, then I am not only the initiator of a desire and its subsequent behavior; I am also, and tend to remain, the person who has so imagined, desired and decided. I have in part generated a disposition and an attitude to represent ends and objects to myself. This is another way in which desire, far from being an irrational and formless drive indifferent to individuality, depends in part on my imaginations representations and on my character. V. Conclusions Is there a unity to these different activities, functions and meanings of ? They do not seem to be connected by a thorough, let alone systematic, unity. But nor are they a bundle of heterogeneous or disparate elements, for they find their leading thread and common denominator in the power of representation and presentification of absence.
UNIVERSITY OF PISA

COMMENTARY ON FERRARIN
KLAUS BRINKMANN

One of the puzzling things about De Anima III 3Aristotles most extensive treatment of is the fact that nowhere in that chapter does Aristotle seem to indicate what the indispensable contribution of to the cognitive process is supposed to be. The text remains largely silent on the rationale for . To be sure, the function of and how it differs from perception on one hand and understanding, believing and thinking on the other is described in great detail, as Professor Ferrarin has so amply shown in his paper. Thus, we are being told that presents us with (images) when no perception is occurring, as in dreams (An. III 3, 428a6-8). Also, it is up to us to call up such , whereas it is not up to us to perceive something that is not there or not to perceive something that is. And again, we are free to imagine something at will, but we are not free to believe something to be true or false at will (An. III 3, 427b16-21). For belief is based on judgment and judgments must be true or false, whereas according to Aristotle are for the most part false (An. III 3, 428a12). (In parenthesis, we may wonder why Aristotle says this since many , e.g., such as occur in day dreaming or when we simply make things up, might with equal right be called neither true nor false. Also, we may wonder why not all are simply false and why apparently sometimes they can be true. I will return to some of these points below.) Aristotles argument in De Anima III 3 is to show primarily that is a faculty sui generis, not reducible either to perception or to processes of reasoning, thinking and belief formation, even though is an ingredient in or, alternatively, a prerequisite for, all of them for the most part. But what we really would like to see is an argument explaining why is an indispensable cognitive function. What would happen, if were missing from the panoply of the souls cognitive functions and processes? What could we not do in the ? There is one place where Aristotle commits to a absence of clear answer to this question. Famously, in An. III 8, 432a12-14 he tells us that while thoughts and are not the same, thoughts would not exist without existing as well. Moreover, unless one perceived things one would not learn or understand anything, the reason

114

KLAUS BRINKMANN

being that without perception there would be no and without , An. III 8, 432a7-9). no theoretical understanding ( Unfortunately, this remark in An. III 8 remains largely unexplained. While there is an indication why are a necessary prerequisite for thinking, 1 we are still left speculating how the act of thinking itself . Or take dreaming. may be dependent on the existence of Quite explicitly, Aristotle establishes a connection between dreams and . And yet, he does not say that without there could be no dream images. In other words, he does not here exclude that memory might perhaps substitute for in making dreams possible. Even if we can make a plausible case for arguing that memory would be insufficient to explain our dream experience, it would be more reassuring to hear this from Aristotle himself. Incidentally, as far as the faculty of memory is concerned we are indeed able to specify a necessary cognitive role for it on the basis of what Aristotle says about memory in Analytica Posteriora II 19 and Metaphysics 1. The argument there is that without memory no retention of images would be possible and without retention of images no emergence of simple universals. 2 So the question remains: What kind of functionally necessary work is supposed to do in the economy of our cognitive activities? Why does play an indispensable part in our cognitive processes? What would not work, if were to be taken out of the picture? We , but we do not know that we would not have thoughts without know why Aristotle believes this. Is there anything else that we would not ? possess or be able to do without In a 1996 article in Phronesis, Victor Caston has made a persuasive proposal in response to these questions (although I believe it does not go far enough and also needs to be refined, as I shall try to show). 3 His answer is, in a nutshell, that without error could not be explained. Briefly put, Castons argument is a follows. Perception involves a causal process in which an object affects a sense organ such that the object is the cause of a change in the sense organ which change produces an affection in the soul that results in a mental content (Caston 1996, 29). Also important in this context is the theory that something in

_________
1 The intelligible forms ( ) are among the sensible forms, Aristotle says at An. III 8, 432a4-5. I will be arguing below that the are in fact the sensible forms. If so, sensible forms are a prerequisite for having thoughts, although this does not yet explain why they are also part of the act of . 2 Cf. APo. II 19, 100a3-8; Metaph. A 1, 980b28-981a1. 3 Cf. Caston, 1996.

COMMENTARY ON FERRARIN

115

the perceiver must be like the object, if cognition is to occur, otherwise we might have a disconnect between object and perceiver in the sense that the change effected by the object in the perceiver does not translate into a representation or image of the object itself. When we put these two requirements together, i.e. the causal connection and the likeness between object and something in the perceiver, we will conclude that an object is known because it interacts with something in the subject like itself, and furthermore, and more importantly, that such a mental state will be about what brings it about (Caston 1996, 29). Now on one reading of this, this means that the content of the perception will infallibly correspond exactly to the object that caused it (cf. Caston 1996, 30) and from this it may be concluded that all appearances are true (cf. Caston 1996, 31), or that perception is always veridical. That perception is always about what brings it about can thus lead to the unacceptable consequence that we can never be mistaken in judgments based on perception. At the same time it is not difficult to see how such a patently false conclusion could be avoided. We would need to find a way to decouple the object from the mental or psychic content while not denying the existence of a causal connection. If it were possible to say that the content of a perception need not be exactly like its object, we could open up the possibility of a divergence of content and cause of perception (cf. Caston 1996, 38). However, on the causal model of perception, such a separation seems impossible. This is why, Caston argues, Aristotle introduces the faculty of imagination. Phantasia generates a content that can be different from its cause:
For error to be possible, content must bear some relation to cause. But it must be possible for content and cause to diverge. The relation between a mental state and what it is about cannot be simply identified with the relation between an effect and its cause. (Caston 1996, 38)

So what happens in an erroneous judgment is that we misattribute the content or effect to an object or cause that is not really its cause. For this to be possible, however, there must be something in addition to the sensory effect which is wrongly attributed to the cause of the sensory effect. This generated by . something in addition would be the This result seems to be confirmed by Aristotle when he says that we need to distinguish between perception of the proper sensibles which is always true, and incidental perception which concerns the perceptual judgment based on the former kind of perception. Thus we cannot be mistaken that we see a color (even though that perception may still be indis-

116

KLAUS BRINKMANN

tinct, i.e., Is it really white or is what I am seeing actually gray? 4 ). What I can be and sometimes are mistaken about is that this white shape in front of me is in fact my friend who has unexpectedly donned a toga. Says Aristotle: . . . we are not mistaken on the point that there is white [in front of us], but about whether the white object is this thing or another we may be mistaken (An. III 3, 428b22-24). The fact, then, that in so-called incidental perception I may connect the content of my perception with the wrong explains the possibility of error. And without a in addition to the percept or the sensation, such misattribution could not be accounted for. I believe that although Professor Ferrarin is presumably aware of this argument by Caston, he probably had his reasons not to follow its lead. Now, I think that Castons point that has the function of explaining the otherwise inexplicable existence of error is well taken and difficult to dispute. However, I also have two reservations. First, his analysis of the relationship between cause of perception and mental or psychic content needs refinement. And second, I believe that his interpretation of the role of in cognition does not go far enough. It does not address what I believe to be the most fundamental and at the same time indispensable function of for cognition, viz., the fact that it uses sensations to generate sensible forms or universals that refer to the sensible features of things. 5 Let me briefly state what I think needs to be fine tuned in Castons account and then elaborate some more on what I take to be the most basic and indispensable function of in cognition. When Caston suggests that the possibility of error requires a decoupling of the object from the content of perception, what I believe he should have said is that it requires a decoupling of the sensation caused by the object from the content of perception. For the object or cause does not need to be distinguished from the sensation, because the two are already unlike. Clearly, we do not mistake a sensation caused by an object for the object itself. What is like here according to Aristotle, is not the object and the sensation but the form inherent in the sensible object as compared to the sensible form of the object as it appears in the soul. Their unlikeness consists in the fact that prior to perception the form in the soul exists in potentiality only whereas the form in the object exists in actuality independently

_________
4 I suggest that this is what Aristotle has in mind when he says that perception of the proper sensibles is liable to falsity to the least possible extent (An. III 3, 428b19). 5 My argument in support of this claim crucially depends on Aristotles remark at An. III 8, 432a9-10 that are like perceptions, only without the matter. If so, they must be universals.

COMMENTARY ON FERRARIN

117

of the sensation. 6 The sensation is only the vehicle for the form. The process of perception establishes a likeness in the modality of the form, not a likeness of sensation and sensible object. Consequently, the content of perception is twofold. There is a) the physical sensation, and b) the sensible form for which the sensation serves as the material substratum. What happens in the case of error, therefore, is a mistaken marriage of sensible form with a sensory substratum whose potentialities are not restricted to a particular form (e.g., the man in the toga) but which could conceivably be the substratum of other forms as well (albeit within limits). These other sensible forms are supplied by . The mistakenly associated form may be suggested to me by my memory, e.g., by what I expect to see due to an association of the present sensation with what I have already seen or perhaps merely imagined. In other words, the sensation I have may be like some other sensation I had and which is now memorized as a memory image, but which is normally correlated with some other sensible form. Note that I am tacitly assuming here that correlations between sensory images and or sensible forms must exist in my memory to explain errors in incidental perception. But what I have been arguing also implies among other things that cannot possibly be simply identical with the sensory content of perception. Hence I find myself in agreement with Professor Ferrarin when he warns explicitly that should not be identified without further ado with sensory images. Indeed, I believe this point is one of his most important contributions without which the mystery of cannot be resolved. To elaborate, I am now claiming that are sensible forms, not mere sensations, and that memory may contain both memory images, i.e., images of particulars, as well as sensible forms, i.e., universals. As I see it, to identify with sensible forms would make very good sense in the light of De anima III 8 where the intelligible forms are said to be among the sensible forms. For it would explain why then there could , i.e., without sensible forms as their be no thoughts without prerequisite. And it might help us better understand why we need in the act of contemplation itself, too, if are universals whereas it seems hopeless to try to understand why we would need images to think. are universals simply because they are sensible forms, i.e., like sense-perceptions, only without matter. 7 I would

_________
Cf. An. II 5, 417a17-20. It should be noted that Hamlyns translation of impossible to see that are indeed universals.
6 7

as images makes it near

118

KLAUS BRINKMANN

also want to argue that practically all sensible forms are , at least in the initial stages of concept formation, given that the genesis of sensible forms presupposes memory. Intelligent perception would then always involve the marriage of a sensible substratum (which, in the case of dreams, would come from memory images) with an already existing with which we are familiar. In actual perception, the coalesces with the sensible substratum, i.e., it is actualized and thus becomes the actualization of the meaning potentially inhering in the underlying sensible substratum. This is how, famously, we (sensibly) perceive the particular while (intelligent) perception is of the universal. Through the actualization of the , we experience the object in its selfgivenness as a particular of a certain kind. So while are certainly involved in the case of erroneous incidental perception as argued by Caston, they are not produced ad hoc on the occasion of making an erroneous judgment. In other words, are not invented out of the blue whenever we encounter a case of an unclear, indistinct, ambiguous percept. Rather, as I proposed, they must lie ready in the mind, to coin a phrase, i.e., in memory, in order to be called up by the judging faculty. This means, however, that error is , but only an occasion not the occasion for the production of of their actualization. This is why I suggested earlier that Castons argument, convincing as it is within its own scope, does not go far enough in . It is reasonable to assume, addressing the true origin of therefore, that are produced during the process called induction in An. Post. II 19 which involves the retention of percepts in memory. Only on the basis of such retention do universals emerge in the soul. 8 My line of argument seems to be confirmed when Aristotle insists that are parasitic upon sensory impressions. So there must be a cognitive process that is more fundamental than, or at least independent from, the formation of erroneous perceptual judgments that generates and that therefore should be referred to as the primary activity of . This is, I suggest, the process though which sensible forms, i.e., universals, emerge in the soul through induction. Let me broaden the perspective a little by first focusing on a passage in De Anima III 3 that is also discussed by Professor Ferrarin. At An. III 3, 428a24ff., Aristotle begins a somewhat convoluted argument to the effect

_________
8 I realize that is not mentioned in APo. II 19 at all, but I do not think that this necessarily invalidates my conjecture. The role of in cognition may not yet have been worked out by Aristotle at the time of the Analytica Posteriora.

COMMENTARY ON FERRARIN

119

that is neither belief together with perception, nor belief through perception, nor a blend of belief and perception. In other words, is not a form of belief. It is not a form of belief in the sense that it would interpret or judge perceptions, nor a form of belief prompted or induced by the perception, nor some synthesis of belief and perception. Whatever may be, it is not an interpretive or a judging faculty which leads to a belief. Here Professor Ferrarin might contradict me, because he holds (pp. 98-99) that can be a faculty of judgment in certain cases. It is important for my argument, however, that this not be so. (In parenthesis, Professor Ferrarin also says in the same context that becomes interpretation in certain cases, and with this formulation I believe I would want to agree.) Aristotle supports his claim that is not a form of belief by analyzing cases which would become inexplicable or paradoxical, if were indeed some form of belief. One of his examples is that of the sun appearing to be one foot across, an example which occurs also in On Dreams, as Professor Ferrarin reminds us. Aristotles reductio of the claim that is a form of belief runs as follows. If were indeed a form of belief,
somethings appearing to us will be believing what one perceives, and not incidentally. But things can also appear falsely, when we have at the same time a true supposition about them; e.g. the sun appears a foot across, although we believe it to be bigger than the inhabited world (An. III 3, 428b14).

If were a form of belief, then all our perceptually based judgments would be incorrigible, because nothing could appear falsely. The thought that the sun might be bigger than the inhabited world could not occur to us, because its apparent size would be taken to be its real size. The sensible form of the suns apparent shape would be taken to be identical with its sensory substratum. This would constitute a case of deceptive perception that could not be corrected, unless there were independent evidence to the contrary. But then this independent evidence would likely suffer from the same inherent flaw. Things could in reality not be other than what they looked like. What Aristotle is saying here is that if were a belief based on sensation, we would be systematically misled or deceived by it. Hence Aristotle should not be construed as saying that in the case of the sun we are initially misled to believe that the sun is actually only a foot across and only later come to understand that this is a deception. According to Aris-

120

KLAUS BRINKMANN

totle, such an interpretation would lead to inconsistencies. For if we were initially deceived by the appearance of the sun, it would also follow
that we shall have abandoned the true belief we had, although the circumstances remain as they were . . . [i.e., although we had no reason to abandon our true belief]; or, if we still have it, the same one [i.e., belief] must be both true and false (An. III 8, 428b 4-8).

The former alternative would be irrational, the latter absurd. So something must be wrong with believing that the appearance of the sun is deceptive and leads to or induces in us a false judgment that must later be corrected. The little word false in false appearance should therefore be treated with caution. I doubt that Aristotle here equates false with deceptive or misleading. Instead, I suggest, false here just means inadequate, but not misleading. To interpret false as deceptive means to adopt a Cartesian point of view whereas I believe the point of view to adopt here is not Cartesian, but Husserlian, not propositional but phenomenological. You may remember that Descartes uses the very same example of the sun in order to show that the senses are systematically misleading and cannot be relied upon, although our reasoning is capable of correcting this senseinduced error. 9 On a phenomenological reading, however, the appearance of the sun as a foot across is not misleading at all, just because it is understood that the sensory substratum is not an adequate representation, but only an instance of the appearance of the thing. The understanding of the sensation as an appearance of something has built into it that large objects may look comparatively small from a distance, and in general that appearances should not be taken at face value. Hence there is no deception that would need to be corrected. The sensible impression of the sun is already customarily associated with the sensible universal called sun which qua universal is no longer tied to any particular image of the sun. It could be argued that I am now identifying and the universal element in perception much too closely, against Aristotles explicit argument in De anima III 3 that and must be kept distinct. 10 And I must admit that according to my interpretation the difference between the two seems to be merely that between potentiality and actuality. However, contains a component of sensation that normally does not. So it seems we must further distinguish between a that is actualized without an accompanying sensation due to perception and one that is. Erroneous perception could then be explained

_________
9 See Descartes 1984, vol. II 10 Cf. An. III 3, 428a 5-16.

27.

COMMENTARY ON FERRARIN

121

as a mistaken perceptual judgment in which a was wrongly associated with a perceptual matter generated by sensation whereas in the case of dreaming a combination of perceptual matter and is . This may not be enough to explain all functions of generated by , although both erroneous perception and dreaming could still be explained in this way. In both cases the actualization of the is experienced as if they were perceptions. What I have not accounted for are cases such as the willful combination of to form fictitious objects or scenarios for which no counterpart exists in reality. This is the function we most readily assign to the imagination today. I have to leave this an open question for now. My response to Professor Ferrarin was more concerned with carving out a role for phantasia that might perhaps be called its most central and epistemologically irreducible function. In sensation and perception, we are in direct contact with the real thing. Hence, if all we had were sensations, we would never be able to effect the separation between the sensation and our representation of the object. enables us to do just that by putting sensations in perspective, i.e., perceiving them as appearances that must be distinguished from their sensory substratum. Professor Ferrarin says much the same thing when he comments that has a unique cognitive use precisely qua capacity to represent (p. 100). Again, I think it is worth emphasizing that this function would be difficult to identify, if are not distinguished from images. 11
BOSTON UNIVERSITY

_________
11

of

It seems to me that this distinction is neglected or even ignored in Modraks account : cf. Modrak 1987, 82, 84, 86, 87, 169.

FERRARIN BIBLIOGRAPHY
Blumenthal, H.J. 1977. Neoplatonic Interpretations of Aristotle on Phantasia, Review of Metaphysics 31: 230-41.Bods, R. 1990. "L'imagination au pouvoir," in Dialogue, 29:1: 21-40. Brann, E.T.H. 1991. The World of the Imagination. Sum and Substance, Lanham. Bundy, M.W. 1927. The Theory of Imagination in Classical and Medieval Thought. Champaign. Busche, H. 2003. "Die Aufgaben der phantasia nach Aristoteles," in Imagination FiktionKreation, T. Dewender and T. Welt (eds.), Mnchen and Leipzig: 23-43. Canto-Sperber, M. 1996. Le role de limagination dans la philosophie aritostlicienne de laction, in Corps et me. Sur le de anima dAristote. C. Viano (ed.), Paris: 44162. _______ 1997. "Mouvement des animaux et motivation humaine dans le livre III du De anima d'Aristote," in Les Etudes Philosophiques, 1: 3-40. Caston, V. 1996. Why Aristotle needs Imagination, Phronesis 18: 156-75. Castoriadis, C. 1997. The Discovery of the Imagination. In World in Fragments, Stanford: 213-45. Dugr, F. 1990. "Le role de l'imagination dans le mouvement animal et l'action humaine chez Aristote," in Dialogue, 29:1: 65-78. Engmann, J. 1976. Imagination and Truth in Aristotle. Journal of the History of Philosophy 14/3: 259-65;. Ferrarin. 1995. Kants Productive Imagination and Its Alleged Antecedents, in Graduate Faculty Philosophy Journal 18: 65-92. _______ 1997. Riproduzione di forme e esibizione di concetti. Immaginazione e pensiero dalla aristotelica alla Einbildungskraft in Kant e in Hegel, in Hegel e Aristotele. G. Movia (ed.), Cagliari 1997: 253-93. _______ 2004. Hegels Interpretation of the Aristotelian NOUS, in The Impact of Aristotelianism on Modern Philosophy. R. Pozzo (ed.), Washington, D.C.:193-209. _______ 2001. Hegel and Aristotle. New York. _______ 2004. Saggezza, immaginazione e giudizio pratico. Studio su Aristotele e Kant. Pisa. Frre, J. 1996. Fonction reprsentative et reprsentation. et selon Aristote, in Corps et me. Sur le de anima dAristote. C. Viano (ed.), Paris: 333-48. Freudenthal, J. 1863. ber den Begriff des Wortes bei Aristoteles, Gttingen. Hankinson, R. J. 1990. "Perception and Evaluation: Aristotle on the Moral Imagination," in Dialogue, 29:1: 41-63. Illuminati, A. Quasi una fantasia. Funzioni cognitive dellimmaginazione nei commentatori di Aristotele, in Imago in phantasia depicta. Studi sulla teoria dellimmaginazione, L. Formigari, G. Casertano, I. Cubeddu, eds., Rome: 149-66. Labarrire, J.L. Imagination humaine et imagination animale chez Aristote, Phronesis 29: 17-49. Lefebvre, R. 1997. "La Phantasia chez Aristote: subliminalit, indistinction et pathologie de la perception," in Les Etudes Philosophiques, 1: 41-58. Lycos, K. Aristotle and Plato on Appearing, Mind 73: 496-514. Modrak, D. Reconsidered, Archiv fr Geschichte der Philosophie 66: 4769. _______ 2001. Aristotles Theory of Language and Meaning. Cambridge. Nussbaum, M.C. 1978. Aristotles De Motu animalium. Princeton.

FERRARIN/BRINKMANN BIBLIOGRAPHY

123

Piro, F. 1999. Il retore interno. Immaginazione e passioni allalba dellet moderna, Naples. Rees, D.A. 1971. Aristotles treatment of , in Essays in Ancient Greek Philosophy, J. Anton and G. Kustas, eds., Albany. Rorty, A.O. 1996. Structuring Rhetoric, in A. O. Rorty (ed.), Essays on Aristotles Rhetoric. Berkeley, 1996: 1-33. Schofield, M. 1978. Aristotle on Imagination, in G. E. R. Lloyd and G. E. L. Owen, eds., Aristotle on Mind and the Senses. Cambridge: 99-129. Sorabji, R. 1972. Aristotle on Memory. Translation and Commentary by R. Sorabji, London. Watson, G. 1982. in Aristotle, De anima 3.3, Classical Quarterly 32: 100113. Wedin, M. 1988. Mind and Imagination in Aristotle. New Haven. Yates, F.A. 1966. The Art of Memory. London.

BRINKMANN BIBLIOGRAPHY
Caston, V. 1996. Why Aristotle needs Imagination, Phronesis 18: 156-75. Descartes, R. 1984. The Philosophical Writings of Descartes. Cottingham, J., Stoothoff, R., Murdoch, D. (trs.) 2 vols. Cambridge. Hamlyn. D.W. (tr.) 1993. De Anima Books II and III (With Passages from Book I). Oxford: Clarendon Press Modrak, D. 1987. Aristotle. Chicago and London. Ross, D. (ed.) 1967. Aristotle, De Anima. Oxford.

COLLOQUIUM 4

ENCHANTING THE SOULS ON PLATOS CONCEPTION OF LAW AND PREAMBLES

JEAN-FRANOIS PRADEAU

Before it became a major term in Platonic political arguments, was a term with a very large number of meanings; many more than those of the term law today. In ancient Athens, moreover, the properly juridical uses of the term already had such a considerable history and importance that the Platonic argument would remain incomprehensible if we did not recall them. The understanding we have today of Greek law and justice is restricted by the paucity of sources. 1 Greek law had a feeble posterity, insignificant in comparison to Roman law, and the sources we have for it remain few and for the most part exclusively Athenian. They are defective insofar as the Greeks did not leave behind any properly juridical literature, and our only primary sources are speeches that were written by orators to be pronounced at trials, for or against the defendant. 2 This literature offers nothing comparable to the treatises of the Latin jurisprudents, and it is as partial as it is biased, since the trials for which these speeches were written by professional orators represented only a fraction of court cases. Above all, our sources are deficient because the preserved texts have most often been purged of their properly juridical content: the text of the laws to which the speeches refer has not always been reproduced in the manuscripts, and is sometimes only summarized allusively. 3 Apart from the judicial speeches,

_________
In these pages, I limit myself essentially to three studies, each of which contains complementary bibliographical references and indispensable specifications. In judicial matters, I shall refer most often to the work by S. C. Todd, The Shape of the Athenian Law, (Oxford, Clarendon Press, 1993); historical and social information will be dependent on M. H. Hansen, The Athenian Democracy in the Age of Demosthenes: Structure, Principles, and Ideology, translated by J.A. Crook, 1991, (Blackwell, University of Oklahoma Press, 1999). Finally, to appreciate the specificity of the Platonic treatment of the law, we will follow J.M. Bertrand, De lcriture loralit. Lectures des Lois de Platon, (Paris, Publications de la Sorbonne, 1999). Each of these three works will be designated simply by the authors name. 2 See Todd, p. 7, and pp. 30-63 for a presentation of all the sources. 3 As Todd recalls, p. 44, the texts of the principal Athenian orators were transmitted by posterity for their literary qualities, and editors did not think it useful to reproduce the texts of the laws that might be cited in the speeches (very often, a brief lemma simply indicates that a legal text figured in the speech, which should then be read or recalled).
1

126

JEAN-FRANOIS PRADEAU

the historical material remains very meager with regard to Greek law outside of Athens. This relative poverty at least has the merit of privileging the city of which Plato was a citizen, and in the same vein we can be glad that the essential part of the judicial speeches were pronounced by contemporaries of Plato. Lysias (444-c.375) was of the same generation as Plato, like Isocrates, whom he probably opposed (436-338); and Demosthenes (384-322) wrote a good many of his speeches while Plato (428-348) was still alive. Thus, however incomplete the testimony of the orators may be, it nevertheless gives a glimpse of judicial practices that is perfectly contemporary with what Plato knew of them, and to which he objected. We must, moreover, add that the testimony of the Platonic dialogues also contributes to our knowledge of Athenian law, for the same reasons as the constitutional writings found in Aristotle or in some passages of the historians. 4 This does not fail to cause difficulties in the case of the Laws, as we shall see later, for the simple reason that not all aspects of the legislation of the Magnesians have an equivalent in what we know of Athenian legislation, so that it is often difficult to make allowances for what Plato adopts, adapts, corrects, or invents. What we have just rapidly indicated will be specified further on, as we advance in the description of the legislation of the Magnesians. In what follows, a few lexical and conceptual remarks will be devoted to Platonic usages of the term , and to the presence of the law in the dialogues. This should enable us to understand that the way the Laws define the law (and this holds for legislation and justice as well) does not depart from what is said, sometimes in a less elaborate way, in the other Platonic dialogues. There is indeed a coherent conception of the law in the Platonic corpus. 5 I. The Term

The history of their usages teaches us that the verb and the noun do not immediately designate the fact of legislating or the law. The meaning of the two terms is initially attached to the distribution of land and goods, but also to pastoral practices (the shepherd is a ). The verb can thus designate the division and distribution of land be-

_________
4 Todd, pp. 38-40 (who nevertheless emphasizes, here again, the poverty of historical testimonies we find in the works of Herodotus, Thucydides, or Xenophon). 5 This coherence was set forth most recently in the study by F. L. Lisi, Les fondements mtaphysiques du nomos dans les Lois, Revue philosophique, 1, 2000, p. 57-82.

ENCHANTING THE SOULS

127

tween human beings, that is, the delimitation of this land; but also the behavior of animals on the land thus distributed, or the pasture. The activity of distributing lands and the exercise of pastoral authority also gave its initial meaning to the term (as they did to the almost identical term , which has exclusively preserved the pastoral meaning and designates a portion of territory, a field or pastureland). If designates the law that prescribes conduct in the city, it is because this law distributes at the same time as it orders, and it remits to the group, as to a herd, its appropriate share. To be sure, this meaning turns the law into a common prescription, capable of being written down, but it also designates the habitual usage of the group, and the use to which its life is ordered. Distribution, prescription, and usage thus enter into the meaning of as terms related to it, as connotations that are found in the prose of the Classical period. 6 In the most strictly juridical sense, the term may designate the two realities that English distinguishes under the name of law and laws (or Latin under those of ius and lex), that is, such-and-such a particular law, but also the system of law: here, in other words, the ordered totality of the laws at the same time as the common principles on which they all rest because the legislator has set his world in order on their bais used in its general sense, to refer to the totality of the sis. 7 When law, it also designates the totality of prescriptions that are imposed on the city; so much so that the term may be synonymous with that of . This provides evidence, in the simplest way imaginable, for the interweaving of judicial and institutional considerations, as well as for the way in

_________
6 The examination of an argument found at the beginning of book VII will enable us to identify some of these related terms (see infra, p. 130). The term has an extraordinary semantic amplitude; its usages and meanings are excessively numerous, and this unfortunately forbids us from holding with certainty that each occurrence of designates a law or the law, rather than a simple obligation, a custom, or a habit. 7 As has been pointed out by S. C. Todd (pp. 18-19), the ancient texts more spontaneously use the plural to designate the law, that is, the whole formed by the laws (a whole that can be personalized, as Plato does in the prosopopoeia of the Crito), and they reserve the singular for the designation of a particular law. Let us add in this context that it is the merit of S. C. Todd himself to have emphasized, together with P. Millett, how prudent we must be when discussing Greek law, since the documentary lacuna make this expression fragile: Law, Society and Athens, in Nomos: Essays in Athenian Law, Politics and Society, ed. P. A. Cartledge, P. A. Millet and S. C. Todd, (Cambridge, Cambridge University Press, 1990), 10-11. In this same study, P. Millett and S. C. Todd deal with the various meanings attached to the term .

128

JEAN-FRANOIS PRADEAU

which, for the Greeks, discourse on law is always immediately a discourse on the civic community and its constitutional organization. On this point, Plato does not depart from the usages of his time. Quite the contrary, he chooses to link together, in order to fuse them, the fate of the city and that of its legislation. Right from the first definition of the given by the Laws, this link is sealed as follows:
Over and against all these [states the Athenian Stranger, speaking of the different powers of the soul] we have calculation [ ], by which we judge the relative merits of pleasure and pain, and when this is expressed as a public decision of a state, it receives the title law [ ] (I 644d1-3).

The law is thus defined, in general, as the reasoning that is imposed on the city. The great interest of this definition lies in the way it collects and announces the various statuses that the rest of the dialogue will reserve for the law: it is first and foremost a process of reasoning, and as such is molded by a thought at the same time as it seeks to be set forth in a rational, demonstrative way. This rational thought is then addressed to souls, which are liable to experience desires, pleasures, and pains, but also to perceive a process of reasoning. Finally, the process of reasoning or calculation in which the law consists seeks to dictate to all a judgment on what is and is not to be appreciated; that is, desired or experienced. Rational discourse that teaches all souls what they must appreciate in their interest: such is the initial definition of the law in the Laws. This definition has many presuppositions, which are not exclusively Platonic; far from it. For they express rather faithfully, in a concentrated form, the characteristic features of the representation of the law common to all those who pronounced an opinion on it in the Classical period, although they might do so in different perspectives or to defend opposite political arguments. Indeed, it can be argued that the definition of the law as a common prescription or prescriptive discourse imposed on all, in the interest of the city, and which prescribes conducts, is a current definition. 8 Likewise, we may concede that the educative conception of the law is not specifically Platonic: it is ancient and common, and it has justifiably been written that the law always had, in the eyes of the Greeks, a function not only of prohibi-

_________
8 It is, for instance, one of the lessons of J. de Romillys classic study of the law to show how, in the midst of the debate on the nature of the law promoted by the Sophists at the turn of the 5th and 4th centuries, this common definition of the law was preserved, even though the necessity of the law, its content, and the modalities of its writing were controversial (La loi dans la pense grecque, (Paris, Les Belles Lettres, 1971), 73-114).

ENCHANTING THE SOULS

129

tion and surveillance, but of education. 9 It has also been recalled that the law was commonly conceived as the discourse that teaches the citizen who obeys it the paths of virtue. If Plato distinguishes himself from his predecessors or contemporaries, it is in the way he seeks to collect and to found, on the same principles, these various characteristic features, and in the way he chooses to give them a form of unheard-of extent or depth. Indeed, what appears in this brief extract from Book I, is that Plato chooses to designate the soul as the recipient of the law. This stance could, once again, be attributed to other contemporary Greek authors, but Platos texts give it a genuine doctrinal justification and an unrivalled development. When, in his Gorgias, Plato defines legislation as the equivalent of a gymnastics of the soul (and justice as its medicine, 464b), he suggests that legislation must exercise the soul for virtue, and that the citizen who obeys the law finds in this obedience the opportunity for the improvement and training of his soul. Far from being incidental or full of imagery, this definition is assumed as such by the Platonic doctrine, which develops with precision a psychological definition of the law, explaining how it informs human behavior by exercising a form of persuasion and constraint upon souls, the resources of which are abundantly described here in the Laws. That the observation of the law is a kind of maintenance and care, as is the practice of gymnastics, or else that it makes known governmental decisions or principles of conduct: all this deserves some explanation. How, precisely, does the law prescribe to the soul its behavior? On what aspects or faculty of the soul is it exerted, and what sort of knowledge does it dispense? II. Morals and the Law In order to appreciate the function Plato assigns to the law, we can ask ourselves on what the laws mentioned or molded in the dialogues are exercised. The question is not yet that of knowing what these laws deal with, or what are their various objects, but to know on what they are exerted and carry out the operation proper to them. If we accept that laws have objects and addressees, and that they prescribe certain behaviors to the citizens, this is because the laws are exerted on the citizens. Certain laws concern all the citizens, while others may concern only some of them, according to

_________
9

J. de Romilly, op. cit., p. 227 (see the whole of pp. 227-250).

130

JEAN-FRANOIS PRADEAU

the activities they carry out, but the citizen is indeed the addressee of the law: that is, he is the person to whom a discourse prescribed by the legislator is addressed. The citizen must act or refrain from acting in conformity with the law, and it is the responsibility of the government and its police auxiliaries to constrain him to do so. In this very general form, this is a definition of legislation that is to be found in the Platonic texts, as it is, moreover, also in the rest of ancient and even modern judicial thought. We must try to be more precise. If we accept that the text of the law is written down, and is offered to the citizen reader who has an obligation to become aware of it, and if we also accept that the citizen must adapt his or her behavior to what this collective discourse prescribes, then we must say that the law carries out its operation only on the condition that the way the citizen conceives of his or her own behavior, and decides about it, is affected by this legal discourse. According to Plato, the laws addressee is not exactly the citizen as such, but that which, in the citizen, is the seat of knowledge and principle of behavior and actions: the soul. Better yet, that within the soul which reasons and knows is a rational faculty that is exerted beside and sometimes against an irascible faculty (the ) and a desiring faculty (the ). Thus, the law must prescribe for the soul a behavior that it can actualize only if the rational faculty within it is exerted appropriately, without being impeded by the other faculties. The law is thus exerted on the souls of the citizens, and its specific operation consists in promoting a specific economy in the exercise of the psychic faculties. This mutual arrangement of the psychic faculties is nothing other than the goal Plato assigns to legislation, explaining that it is the task of the laws to forge human morals, or . The question of the laws intention, which is thus psychological in nature, is followed by that of their objects. For if the laws act on souls and forge morals, we must ask ourselves how they achieve this, how they make psychic gymnastics possible, what are the exercises they prescribe, and with what precision or in what number they prescribe them: in addition to the legislations addressee, we must examine its extent and its precision. Because it pronounces judgment on conduct, or on ways of life (morals or ) the law is an ethical discourse, which Plato holds should not simply forbid illicit conduct, but indicate and favor good morals. Insofar as legislative discourse is the discourse that civic authorities address to the totality of citizens, it must assume an edifying and pedagogical vocation. It must tell the citizen what good and bad conduct is, and in addition tell him how to distinguish them. This, at least, is the mission Plato assigns to the legislative text, thus linking the contents and ends of his legislation to

ENCHANTING THE SOULS

131

the ethical considerations that have occupied the beginning of the dialogue and the long exposition of book V: it is the function of the law, even more than of institutions such as meals in common, civic festivals and banquets, to accomplish virtue in the city. It is the instrument thereof. According to Plato, the law serves a particular form of pedagogy, which is based on an indispensable persuasive rhetoric. This rhetoric is indispensable for the writing of laws, for if certain magistrates and guardians of the laws must govern the city according to reason, and make the entire city attain the happy end of virtue, it is indeed necessary that all the citizens subject their own conduct to rational prescriptions, or else that they be convinced that they must obey such prescriptions. With this goal in mind, Plato conceives of an unheard-of legislative instrument: the preamble or prelude. The legislator is to have some laws preceded by a preliminary discourse, addressed to the citizen and intended to persuade him, by exhortations and threats, to censure the conduct contrary to the law that follows, and to praise the conduct it authorizes. This might surprise the reader of Plato who recalls that Plato, particularly in the Gorgias and the Phaedrus, produced a critique as resolute as it was precise of the political rhetoric which, in his time, was dominant in the democratic Assemblies and the tribunals. 10 There Plato declares this rhetoric to be manipulative, deplores the fact that demagogues simply seek to persuade their public (and their electorate) by flattering it but by no means instructing it or making it better, and thus entirely condemns this political practice. Yet this vehement critique, of which we also find elements in the Laws, 11 by no means implies that Plato intends to renounce civic rhetoric. 12 Such rhetoric remains indispensable for constructing his political and legislative edifice, which requires a discursive instrument apt to make governmental decisions and legal prescriptions known to the entire population. A technique of civil discourse is thus required that can

_________
10 The texts from the Gorgias and the Phaedrus are examined, together with those from the Laws, in an important study by H. Yunis, Taming Democracy, (Ithaca, Cornell University Press, 1996) (whose chap. VI-VIII are devoted to Platos work), where the most essential relevant studies are mentioned. In addition, see the article by L. Brisson, Les prambules dans les Lois (in Lectures de Platon, (Paris, Vrin, 2000), 235-262) and above all the specifications by F. L. Lisi, Les fondements mtaphysiques du nomos dans les Lois. 11 For instance, in the condemnation of judicial rhetoric, at XI 937e-938a. 12 Plato habitually opposes a perverted rhetoric to a good and learned rhetoric which, for its part, is capable of instructing its auditor(s). The means of this rhetoric are various, according to whether the perspective is that of the pedagogic relation (as in the Phaedrus, see in particular 263b-272b), or else that of mass civic education (in the Laws, but also in the Republic, Books II and III).

132

JEAN-FRANOIS PRADEAU

explain to all the citizens what the decisions and prescriptions are, but which can also persuade the citizens to follow them: to govern is to persuade. 13 Plato chooses to confide this rhetorical mission partly to the law: more exactly, it is assumed by preambles, which consist in a text attached to some laws and intended to precede them like a musical prelude which must also play the part of an incantation, as Plato repeats several times. 14 Here the philosopher has in mind that the incantation must prepare the soul to receive the prescription and somehow to absorb it, to make it hers. Here, among several other similar passages, is how the Athenian can account for the necessity to attach to the text of the law that supplementary piece of writing that bears with it the edifying and genuinely pedagogical mission of legislation:
The real job of the legislator is not only to write his laws, but to blend into them an explanation of what he regards as respectable and what he does not [ , , , ], and the perfect citizen must be bound by these standards no less than by those backed by legal sanctions (VII 823a2-6).

These remarks from Book VII remind the reader of the way in which Book IV had justified the need to introduce the law by means of a preamble ( ):
A relevant point here is that no legislator ever seems to have noticed that in spite of its being open to them to use two methods in their legislation, compulsion and persuasion (subject to the limits imposed by the uneducated masses) [ , ], in fact they use only one. They never mix in persuasion with force when they brew their laws, but administer compulsion neat (IV 722 b4-c2).

Plato thus claims to be innovative in this subject, and this innovation is confirmed by the little we know of the codes of Greek laws, in which, in-

_________
See, among other passages, IV 711c or VI 753a. The preamble (proomion) is a considerable Platonic innovation, to which a good number of studies have been devoted. In different registers, one derives considerable profit from the terminological remarks of M. Costantini and J. Lallot, Le proomion est-il un prome?, in Le texte et ses reprsentations, (Paris, Presses de lENS, 1987), 13-27 (who show that the term designates the preliminary hymn that precedes the recitation of epic poems in rhapsody contests, with regard to which they play the part both of prelude and of exordium), from the Platonic analyses of J.-M. Bertrand, De lcriture loralit. Lectures des Lois de Platon, pp. 278-287, and finally from the contextual presentation of H. Yunis, op. cit., pp. 211-236, who for his part focuses on comparing the rhetorical vocation of the Platonic preambles with those of certain judicial discourses.
13 14

ENCHANTING THE SOULS

133

deed, all that appears is the statement of the law and the inventory of penalties incurred when it is transgressed. 15 In this way, Plato denounces a form of incompleteness of the legislation in vigor, at the same time as he presents the remedy: the text of the law, in its nakedness, cannot be sufficient unto itself to justify its reason for existence and its pertinence. The laws text includes a description and a form of coercive expectation: it defines the offence that must not be committed, then the penalty that would sanction it if the prohibition were transgressed. Penal coercion is thus announced. As such, as is indicated in its way by the comparison with the two doctors at IV 719e-720e (a doctor treats free men by trying to instruct them about their illness and persuading them to adopt a remedy, while another auxiliary doctor treats slaves and contents himself with prescribing remedies for them), the law merely prescribes. This is why Plato assimilates it to constraint and force. 16 However, it will be up to the preamble to persuade every citizen that the conduct favoured by the law is good, whereas that which it prohibits is, on the contrary, bad. Like the law, the preamble is a discourse enunciated by the legislator, 17 and intended to promote the rational and virtuous order of the city. To be sure, insofar as the law may not be transgressed, the preamble does not have the mission of persuading the citizens that it is appropriate to obey it. It has another function, otherwise more important, which is indicated concisely by the extract we have just quoted (from VII 823a): it must give the citizen an opinion or judgment on what is beautiful and what is not. In this way, the goal is to make the citizens judgment conform to what is prescribed by civic reason, and to promote the community of judgments

_________
15 As far as codes are concerned, the most complete and interesting document is that of the laws of Gortyna. M. Gargarin, who has devoted several studies to it, proposes a few elements of comparison between this code and what is contained in the Laws in Platos code and Greek law, article cited. In the same collective work, E. Lvy gives a translation and a commentary of the code of Gortyna: La cohrence du code de Gortyne, in La codification des lois dans lantiquit (Actes du Colloque de Strasbourg, 27-29 Novembre 1997, Strasbourg 2000), ed. by E. Lvy, (Strasbourg and Paris, Universit Marc Bloch de Strasbourg and de Boccard, 2000), 185-214. Finally, the essential bibliography is recalled and updated in the vast review by A. Maffi, Studi recenti sul codice di Gortina, Dike, 6, 2003, pp. 161-225. 16 See, first and foremost, IV 722b-c. 17 Its text is written down, and it may be inscribed on various supports, but above all it must be diffused orally: the herald is an important personage in the city of the Magnesians, alternately the city messenger and public reader of laws or sentences. On the public dissemination of legislation, see in particular IX 874a, XI 917c, then 921d-e, or again XII 946d).

134

JEAN-FRANOIS PRADEAU

and affections, that common, unique breathing that is evoked ( ) to explain that the city is complete and virtuous when the citizens share the same morals and the same feelings. In the Laws, this necessity was stated at the beginning of Book IV, that is at the very beginning of the foundation of the virtuous city, which is to be established as a colony in Crete:
So it wont be all that easy for the Cretan states to found their colony. The emigrants, you see, havent the unity of a swarm of bees: they are not a single people from a single territory settling down to form a colony with mutual goodwill between themselves and those they have left behind. Such migrations occur because of the pressures of land-shortage or some similar misfortune: sometimes a given section of the community may be obliged to go off and settle elsewhere because it is harassed by civil war, and on some occasion a whole state took to its heels after being overcome by an attack it could not resist. In all these cases to found a state and give it laws is, in some ways, comparatively easy, but in others its rather difficult. When a single people speak the same language and observes the same laws, you get a certain feeling of community, because everyone shares the same religious rites and so forth; but they certainly wont find it easy to accept law or political systems that differ from their own. Sometimes, when its bad laws that have stimulated the revolt, and the rebels try in their new home to keep to the same familiar habits that ruined them before, their reluctance to tow the line presents the founder and lawgiver with a difficult problem. On the other hand, a miscellaneous combination of all kinds of different people will perhaps be more ready to submit to a new code of lawsbut to get them to pull and puff as one (as they say of a team of horses) is very difficult and takes a long time. (IV 708b1-d5)

Here, it is clear enough that the community of judgement and affection is the goal of the laws. We now understand that the preamble of the law is conceived as one of the ways to obtain this unique civic way of life. It is obvious, as is attested by the long preamble addressed to the colonists and future Magnesians citizens that begins at IV 715e, that the preamble is a discourse conceived in order to be addressed to the majority of citizens, no doubt assembled for the occasion, and intended to take the place for them of judgment and conduct. 18 This presupposes a properly poetic aptitude on the part of the legislator, who will use the resources of

_________
18 This is well established in the work, already cited, by H. Yunis, pp. 217-220. The legislator thus has the opportunity to describe the principles that govern his legislation (see also the remarks by M. Gagarin, who explains in the article already cited that the Platonic preambles have an equivalent in the speeches for the defense of the orators, particularly Demosthenes, who likes to explain the legislators intentions to the jurors, and that perhaps Plato found inspiration in such explanations (p. 216).

ENCHANTING THE SOULS

135

fiction and storytelling in the preamble, to exhort his audience to observe the law. 19 The preambles are thus endowed with an educative and edifying function: they must set forth for the citizen what is praiseworthy and what is blameworthy, what he should appreciate and what should inspire revulsion in him. For the legislator, the point is not to set forth rationally the appropriateness of such-and-such a conduct, still less to teach his fellow-citizens a knowledge of a scientific type concerning the various objects with which the law may deal, but to persuade them that such conduct is praiseworthy, while another is not, 20 by delivering to them, with this goal in mind, a form of parental admonition. This is what is suggested by a remark of the Athenian, who says that instead of a tyrant and a despot that gives orders and threatens, the law must be in the attitude of a father and a mother, full of love and intelligence (IX 859a), that is, endowed with an authority over the citizen that is affective but also sacred. 21 The various characteristic features of the preambles are gathered together in a long preamble that is particularly illuminating, that of the twofold law on sexuality found in VIII 835b-842a. 22 This is the law that proscribes sexual relations with individuals of the same sex or with children. This prohibition targets homosexual practices, but also pederastic morals, both of which were in vigor in Greece. The law of the Magnesians, because it runs contrary to extant practices and morals, and because it touches upon desires that are difficult to master, must be preceded by a persuasive preamble whose principal interest is that it gathers together the

_________
19 This places the legislator in competition with the poets, who must be guided and censored so as not to impede legislative and governmental discourse, as is explained by the legislation concerning poetic creation, which finds its conclusion and moral in VII 817a-e (see also the reminder at XI 935c-936b). 20 Thus, preambles do not transmit a teaching, as some interpreters have maintained, but they are conceived in order to suggest to the citizen what is the appropriate conduct whose absence or transgression the law sanctions (see IX 857b-d). The law promises a penalty for whomever transgresses what it prohibits: it names bad conduct. The preamble, for its part, persuades its addressee that certain types of conduct are praiseworthy and others blameworthy: it describes good conduct and its benefits. For this reason, the Platonic preamble has been assimilated to preaching (so, rightly, by H. Yunis, op. cit., pp. 223-229). 21 In a way, Plato thus confers familial authority upon the law. Through the preamble, the law is assimilated to the discourse of a legislator endowed with a parental function (which the laws themselves assume in the prosopopoeia of the Crito). On this point, see the remarks by S. Gastaldi, Legge e retorica. I proemi delle Legge di Platone, Quaderni di Storia, 10, 20, 1984, pp. 84-86. 22 The law is twofold because two versions of it are given, one more constraining, the other less; see VIII 840c-842a.

136

JEAN-FRANOIS PRADEAU

totality of resources of the Platonic preamble. It makes use of recourse to religious authority as well as threats (the legislator explains to the citizen that these criminal acts are judged as such by the gods, who will punish them), appeal to tradition (which will have it that one should unite with an individual of a different sex and of the same age), recourse to edifying examples (animals do not have these morals; nor do Olympic champions, who, on the contrary, cultivate sexual abstinence in order better to triumph), and finally recourse to the threat of civic dishonour for whoever is unable to master his appetites. As we can see, the preamble is indeed a form of admonition and even of threat, which relies on historical reminders or examples to inspire a healthy repugnance towards culpable practices in the citizen. What the preamble must obtain, whatever law it may precede, is the citizens adherence. And this adherence is brought about in most citizens by means of a constraint placed upon desire by fear. This is once again clearly emphasized in the preamble of the law on sexuality, which explains that persuasion is obtained when the individual who feels the desire for culpable practices, a desire which cannot be reasoned with, renounces it because he is afraid of the consequences of his acts. As the Athenian explains in VIII 840b-e, the preamble is a kind of collective incantation that proposes to inspire a collective fear that is effective enough so that the citizen may find in it a means for constraining his own desires to the point of no longer being able to consider transgressing the law. The preambles efficacy relies on a particular psychic mechanism, which is familiar to the reader of the Republic and the Timaeus, in which Plato also explained that desire in the soul is particularly recalcitrant to the instructions of reason, so that the irascible or choleric element of the soul must be brought to bear upon it in order to constrain it by force. 23 The preamble thus dispenses, by means of a collective persuasion, those good beliefs known as right opinions, 24 and it fills souls, as is affirmed at VII 839c, with fear and obedience for the enacted laws. The law thus takes the place of reason, in individuals who lack the psychic aptitude to constrain their desires themselves. 25 More exactly, it is thus the preambles that accomplish rational constraint and take the place, in a way, of the con-

_________
23 See in particular Timaeus 69d-72e; Republic IV 440a-442c, which show clearly that the irascible element in the soul is the one that obeys reason and constrains the desiring element to obey it in turn; the legislative preamble is assigned an identical mission. 24 See also Statesman 310e-311a, where Plato also maintains that politics must promote the community of judgment among the various citizens. 25 This is what the Republic explains, in its way: see IX 571b-c and 590d-e, then X 604a.

ENCHANTING THE SOULS

137

tinuous education of the citizens, since they participate, in the same sense, for instance, as at banquets, in that constraint of desires and pleasures which the beginning of the dialogue had explained was the first step of virtue in man, that is, the first step in his education (II 653a-c). Because they enable each citizen to order his own conduct according to the same morals and the same norms, and thus to follow reason, which reigns in the city, the preambles bring about unity, and finally the construction of the city of the Magnesians: they give it its unique breath. 26 Finally, the preambles retrospectively illuminate the importance of legislation in the constitutional construction of the Laws. Because legislation incorporates that innovation known as the preliminary incantations, it is endowed with a function and content that immediately exceed what could be found in a simple code of laws. It is indeed a particular form of governmental discourse that Plato conceives under the name of , by combining the durability of prohibitions with the ethical pedagogy of a rhetoric liberated from the Greek context of debate, and henceforth suitable for the formation of morals.
INSTITUT UNIVERSITAIRE DE FRANCE UNIVERSIT DE PARIS X.

_________
See the remarks at VIII 846d on the way the citizen is the one who contributes to the good order of the city.
26

COMMENTARY ON PRADEAU
GAVIN T. COLVERT

Professor J.-F. Pradeaus scholarly and stimulating paper challenges the reader to reconsider some commonplaces of interpretation in Platonic political thought. 1 In particular, it provides the occasion to ponder Platos Laws more diligently, which is a difficult but rewarding dialogue. The fact that it is a Platonic dialogue in which the interlocutors converse about legislation makes it doubly difficult to master because of Platos understanding of the diverse purposes of the dialogue form and the work of legislation. Since Plato conceptualizes the goal of these forms of expression differently, the reader must face the complexity of a legislative discourse embedded within a philosophical dialogue that reshapes, critiques and even undermines the embedded discourse. This twin dimension of the work is surely neither incidental nor unintentional, but deliberate and central to the development of Platos argument. The attentive reader senses that the dialogue operates at different levels and that Plato intends certain readers who are very different from some of the dialogues fictional interlocutors to understand this fact. For this reason, one cannot always take what is said about law in the Laws at face value. The dramatic context of the dialogue must shape the readers understanding. The narrative structure of the dialogue involves a conversation between three elderly Greeks. The content of their conversation is legislative. At least two of the interlocutors, the Cretan Kleinias and Megillus the Spartan appear to take the interchange at face value as legislative. The principal character, the Athenian stranger, appears to be aware of deeper significance to the conversation. He describes his speeches variously as mythical and poetic, and he prescribes the reading of the Laws in the education of Magnesian children. 2 He goes on to differentiate between the education of the majority of citizens, and the evidently philosophical training

_________
I would like to express my thanks to the BACAP organizing committee and to Boston College for extending the invitation to comment on Professor Pradeaus paper. I am also grateful for the comments of an anonymous referee who has saved me from making several errors. The remaining flaws are my own. The following critique was made in response to the text of Pradeaus lecture as delivered at Boston College. Some elements of the commentary refer to remarks that have been amended or deleted in the published text. 2 See Laws VII: 811c 812a. English translations of the Laws below are taken from Thomas L. Pangle, The Laws of Plato, Chicago: University of Chicago Press, 1988.
1

COMMENTARY ON PRADEAU

139

that continues into adulthood of those who will compose the Nocturnal Council. Purposefully setting the reading of the Laws alongside other poets, he is surely aware that some will recognize how his very human founding of a city contrasts markedly with the putatively divine basis of other mythical accounts. 3 The three interlocutors are legislators in the sense that they discuss the founding of the city of Magnesia in word, but they are not presently engaged in founding that city in deed. 4 The principal architect of the virtual founding of the city in the dialogue, the Athenian, takes on the role of an advisor to those who have been formally charged with founding the city. He provides a complex and detailed model covering every aspect of Magnesian law, from birth to death, and from inculcating virtue to crime and punishment. Nevertheless, he remains mindful of the fact that their virtual foundation does not obey the order in which actual legislation must be considered. 5 Furthermore, he grants on several occasions that the legislator must adapt the virtual foundation to the practical situation and the human condition. In other words, he presumes from the start that his legislative discourse goes beyond what is possible in numerous instances. We need not look very far for potential examples of such instances throughout the dialogue. Early on, the Athenian praises extensively the idea of carefully regulated drinking parties. Shortly thereafter he waxes eloquent about tyranny. Departing from the traditional models of his interlocutors, he espouses the similar education of the citys male and female youth, including military training. In the case of punishment, he introduces a version of the Socratic paradox, that all wrongdoing is ignorance, and proposes something like philosophical discourse for the treatment of certain offenses. This leads to the establishment of the Nocturnal Council, the members of which resemble a Socratic philosophical society. With regard to his own role in the process, the Athenian appears to distance himself explicitly from the task of the practical legislator, or to consider his position as super-legislative, by describing his contribution to the interlocutors legislative conversation as a form of myth:
. . . you have made a promise to the Cretan nation that you will, with a firm spirit, carry out a founding; and I have promised to help you, with the mythic

_________
3 For a detailed discussion of this point see Pangles interpretive essay at pp. 490-1. For confirmation of the Athenians awareness of this anomaly, see his discussion of Rhadamanthus at Laws XII: 948b ff. 4 See Laws VI: 778b. 5 Ibid.

140

GAVIN T. COLVERT

discourse were now involved in. And surely when Im telling a myth I wouldnt voluntarily leave it unfinished. 6

Several points are worth observing here. The Athenians promise is to his interlocutors, while Kleinias promise is to the Cretan nation. Despite the fact that it is the Athenian who does the bulk of the legislative work in the dialogue, he describes himself as a consultant telling a myth to those who have been duly charged with the founding by the Cretans. This ambiguous position leaves Plato room for the Athenian to say and do things that transcend the work of the ordinary legislator. The conclusion of the myth of the Laws is remarkable according to the commentator Thomas Pangle, precisely because of Platos choice to silence the Athenian and give the largely silent Megillus the last word. 7 In response to Megillus suggestion that the Athenian be compelled to stay and participate in the actual work of founding the city, Plato ends the dialogue abruptly with Megillus and Kleinias agreement to remain and participate in the actual founding, but not that of the Athenian. Pangle suggests that this choice reveals Platos ambivalence about the entanglement of the philosopher in the concrete political process. To the attentive reader, the dramatic context of the dialogue, the Athenians role as expert advisor, and especially his concluding discourse concerning the constitution and authority of the Nocturnal Council should make it clear that philosophical dialogue and philosophers live in a complex relationship to lawmaking in the city. Furthermore, while the fictional audience in the dialogue is composed of elderly non-philosophers who do not always understand the deeper significance of the Athenians legislative program, it is clear that the Athenian and by extension Plato are deeply concerned with the education of a younger audience who are suited to philosophical inquiry. For this reason, the Athenian remains a stranger to his fictional interlocutors, although not to Platos audience, because of both his geographic and also his intellectual origin. One is reminded of the dramatic context of the Republic. The emphasis there was upon a dialogue with young inexperienced philosophical minds in which concrete affairs of the city were secondary. The emphasis here upon practical political life is front and center, but the dramatic context shows that concerns similar to those in the Republic are also operative. One should avoid the implausible view that the purpose of the Laws is to show that legislation is futile and contrary to the nature of serious philosophical inquiry, but we

_________
6 7

Laws VII: 751e-752a (Pangle p. 137). See Pangle, pp. 508-509.

COMMENTARY ON PRADEAU

141

must assume that the content of the work, in addition to its dialogue form, tempers and reshapes its surface legislative program. The following critique of Pradeaus paper is intended to show that the dramatic context of the dialogue lends support to some of his conclusions and calls others into question. He is certainly correct to argue that a serious reconsideration of Platos account of law in the later dialogues is worthwhile. Perhaps because of its length and the apparently minute, even tedious, attention to legislative details not found elsewhere, the Laws has received less careful attention lately than it deserves. Interpreters seem to have been relatively complacent with the stock view that the dialogue represents a pragmatic turn and a corresponding doctrinal evolution in Platos thinking about law later in life. Pradeau takes issue with this standard interpretation and makes a case for the conclusion that some interpreters of Plato have been quick to see a doctrinal change that is not supported by the texts in question. His line of argument rests upon showing that a single conception of the term can be found across several Platonic dialogues. Even if we grant that the way some central characters deploy the term in key Platonic dialogues such as the Republic and Laws is similar, this would not be enough to tell against the historical evolutionary thesis, as the foregoing considerations of audience and context must be taken into account. So, for instance, evidence in the Republic and Laws pointing to the fact that central characters conceptualize the end or purpose of similarly as the production of virtue in citizens, cannot by itself show that Platos conception of the value and role of law in shaping political life remains unchanged. A cursory examination of the Republic and the Laws reveals that there is more legislative work going on in the latter, and that the work of legislation occupies a much more central place in the dialogue. This suggests that one especially important contextual difference to be considered with regard to Platos use of is the audience to whom the dialogue is addressed. In the case of the Laws, this is true both on account of the imaginary interlocutors in the dialogue and Platos intended readers. Unlike the Republic, the characters in the Laws are virtual strangers in a conversation, not friends, and cannot therefore be expected to share the intimacy necessary for philosophical inquiry. Moreover, two of the interlocutors, the Cretan and the Spartan, seem to have little interest in or appreciation for philosophy. The attentive reader senses that Plato has in mind some of his readers will be more like the philosophical Athenian, while others will be more like Kleinias and Megillus. Platos principal interlocutor, the Athenian, must speak differently simultaneously to each of these audiences.

142

GAVIN T. COLVERT

At the same time, the historical evolutionary view is not unquestionably correct. Evidence concerning the dramatic context of the Laws suggests another plausible thesis. Platos view of the role of law in political life may not have changed much from the Republic, but the manner of presentation and points of emphasis may change greatly depending upon the intended audience and dramatic context. It is beyond the scope of this critique to demonstrate such a thesis. Suffice it to say that there is evidence in the dialogic structure of the Laws that may provide support to Pradeaus thesis, even if the thesis itself is not sufficiently demonstrated by his terminological considerations. In the following remarks, the author shall therefore not aim to disprove J.-F. Pradeaus principal thesis, which calls into question the assertion of an historical development in Platos doctrine of . The intended purpose will be to show that it remains unclear whether the similar use of the term by certain key Platonic characters implies that Plato always treats law or puts it to use in the same way. Consideration of the context and audience, both the dramatis personae of the dialogues and the intended readers, suggest that he deployed his discourse about the law differently under varied circumstances. The Republic, Statesman, and Laws may well constitute a continuous line of thought, but not an identical one about the law. A useful way to introduce this line of thinking is found in a curious passage from Pascals Penses. What he has to say there about Plato and Aristotle is evidently an exaggeration intended to amuse, but it contains a kernel of truth:
One only imagines Plato and Aristotle in grand academic robes. They were honest people, and like others, laughed with their friends. When they entertained themselves by writing their Laws and their Politics, they did it playfully. That was the least philosophical and the least serious part of their lives. The most philosophical part was to live simply and peacefully. If they wrote about politics, it was as if to regulate a lunatic asylum. And if they appeared to be speaking about an important matter, this is because they knew the fools to whom they were speaking thought they were kings and emperors. They delved into their principles, in order to moderate this folly and make it as harmless as it could be. 8

_________
8 Blaise Pascal, Penses, Texte tabli et annot par Jacques Chevalier, Paris: ditions Gallimard, 1962, pp. 139-140: On ne simagine Platon et Aristote quavec de grandes robes de pdants. Ctaient des gens honntes, et comme les autres, riants avec leurs amis; et, quand ils se sont divertis faire leurs Lois et leurs Politiques, ils lont fait en se jouant; ctait la partie la moins philosophe et la moins srieuse de leur vie: la plus philosophe tait de vivre simplement et tranquillement. Sils ont crit de politique, ctait comme pour rgler un hpital de fous; et sils ont fait semblant den parler comme dune grande chose,

COMMENTARY ON PRADEAU

143

Surely we should not say, as Pascal appears to do, that Platos political discourse in general or his Laws in particular represent the least philosophical moment in his thinking. Indeed, for Socrates and Plato especially, ethical and political considerations are at the center of the philosophical way of life. Nevertheless, a careful reading of Platos most legislative work, the Laws, reveals that legislation and practical political life exist in constant tension with the philosophical life. Pascal makes at least two points here that are worthy of our attention. First, what Plato thinks about law cannot be separated in the dialogues from the person or persons to whom his discourse about it is addressed. Second, while speaking with apparent seriousness about the law and its purposes, one detects a deep sense of ambivalence in Plato about laws ability to bring about its stated purpose. Pascal undoubtedly asserts with some hyperbole that Platos writings on politics are among his least philosophical. We might refine this point by saying that where Platos interlocutor the Athenian becomes most concerned with philosophical matters in the Laws, he is not speaking apolitically, but he does call into question the surface reading of the legislative program proposed therein. Pascals observations raise some interesting challenges for Professor Pradeaus central assertions in the paper. In his opening remarks he endorsed the idea that Plato conceives of law as a tool for the legislator to form the virtuous character of the citizens and the primary or even exclusive means by which the city is to be founded. 9 In addition, he rejected the commonplace view that Plato turns back to law in the Laws as a second-best solution after despairing about his own failure in life to found a regime modeled upon the virtuous governor in the Republic. From the structure of Pradeaus argument, it is clear that we are to reject the commonplace interpretation because it can be shown that Platos deployment of the term in the mouths of certain interlocutors is the same across important dialogues. The key to this interpretation is the idea that law is not merely a second-best expedient adopted for an imperfect world after

_________
cest quils savaient que les fous qui ils parlaient pensaient tre rois et empereurs. Ils entrent dans leurs principes, pour modrer leur folie au moins mal quil se peut. (Translation mine) I am indebted to my colleague Marc LePain for bringing this passage to my attention. 9 Some points of this critique refer to claims that were stated more sharply and at length in the version of Pradeaus paper presented originally at Boston College. In particular, a lengthy discussion of the use of in the Republic has been removed from the final text and more categorical assertions about the exclusive and sufficient role of law in founding a political community have been omitted.

144

GAVIN T. COLVERT

a loss of confidence in the program of moral education in the Republic. Indeed, according to Pradeau, for Plato the law consistently has a protreptic character throughout his mature works, ordering citizens to the attainment of full virtue. We must ask whether a consistent use of terminology throughout entails precisely these other conclusions. For instance, while granting that law has a protreptic purpose, one must ask to what extent does Plato intend to express confidence in the laws ability to accomplish the proper and complete formation of an ordinary citizens character? Alongside repeated references to the end of law as productive of virtue and the happiness of the community, we find a significant degree of skepticism about the likelihood that citizens will take up the call to virtue in the Laws. Vice is said to be sweet and easy. The gods have placed sweat and sacrifice in the way of the attainment of virtue. Few are inclined to achieve it and it seems very uncertain that we will preserve it in political communities just because they are under the discipline of laws. 10 Indeed, an entirely separate program of education (one presumes philosophical dialectic) is required for the education of members of the Nocturnal Council in the Laws. 11 Only a tiny minority of citizens will ever attain the state of virtue necessary for eligibility for the Council, and that they should attain it is to a certain extent a matter of divine intervention or luck. 12 It should be stressed that these are all features of the argument in the Laws, which if anything is more sanguine about the use of law in the political community than the Republic or the Statesman. In addition, what is needed for the success of a political community is a certain kind of ruler or rulers whose own character is sufficient to inspire and produce obedience to law. The role of the legislator and statesman is essential to the production of a virtuous community. Furthermore, while the Athenian stranger clearly articulates a perfective end or purpose to the law, he is careful to distinguish persuasive or hortatory language from the prescriptive character of legislation itself. The better educated physician converses with an intelligent patient in order to produce understanding and assent to the prescription of the cure, but this conversation is not part of the work of the prescription as such. 13 The physicians assistant provides only prescriptive guidance to patients who are not suited to the discourse of the master physician. This is analogous

_________
10 11 12 13

See Laws IV: 718-720 discussed below. See Laws XII: 965b ff. See Laws XII: 966c-d. See Laws IV: 720-723 discussed below.

COMMENTARY ON PRADEAU

145

to the Athenians conception of how the better legislator legislates for intelligent and free citizens. In addition to the legislation itself, the legislator should produce certain preambles to the law, which explain its purpose for the intelligent person. From this context and the later differentiation of those who should be the guardians of the law from ordinary citizens who are obedient to the law, it is clear that the Athenian does not think all will be able to grasp the teaching of the preambles and to understand the deeper basis for obedience to the lawgivers prescriptions. While these considerations do not lead us to doubt that the purpose of law is fundamentally the production of virtue and the happiness of the political community, it does cause us to wonder whether law is sufficient to mold the moral character of citizens. Pradeau may be successful in showing that Plato did not change how his central characters employ the term law between the Republic and the Laws, but there is evidence to indicate that he changed his presentation of the usefulness of law in light of the dramatic context and for the sake of his intended audience. Here three points must be distinguished. First, terminological considerations are not sufficient to establish Pradeaus claim about the uniformity of Platos account of law in the dialogues. Second, this is not to deny that Plato has a consistent account of the role and purpose of law that is operative from the Republic to the Laws. It has been suggested above that this thesis is in fact plausible, when the different attitudes towards legislation in the dialogues are appropriately contextualized. The Socratic dialogues and external evidence from the letters surely lend support to Pradeaus idea that Plato did not abandon a utopian dream of the prospects for moral education and personal rule, replacing it with a sober but optimistic view of the law as a second-best expedient. Third, however, Plato appears quite consistently ambivalent throughout the Republic, Statesman, and Laws about the prospects for law to establish an excellent political community without the aid of a wise ruler or rulers who have the divine spark of philosophic eros, even though Plato is always interested in good laws as a foundation for virtue. One way to make sense of this attitude, which this analysis proposes as a conjecture needing further exploration, is that Plato thought of the laws as a foundational condition for creating a stable political community within which the pursuit of excellence could be allowed to flourish. He did not think of law as an exclusively sufficient cause of the virtuous community. For various reasons, one of which is the dramatic position of the interlocutors, in the Laws he gradually and only partially reveals other necessary ingredients of the virtuous political community. Evidence for

146

GAVIN T. COLVERT

this claim can be found in Platos vagueness about the composition, purpose, authority, and mode of education appropriate to the Nocturnal Council in Book XII. Given the almost exhaustive detail in which other elements of Magnesian legislation are specified, and the Athenians efforts to convince Kleinias and Megillus of the indispensable necessity of such a body, the anomalous way in which it is described cannot be insignificant. In order to work out these general observations in some greater detail, it is worth examining two passages briefly from the Laws that exemplify some of the aforementioned assertions. The first is from Book IV. It occurs at the beginning of the detailed legislative program for the city of Magnesia. The second is from the end of Book XII. The two passages provide a set of book-ends that illustrate fairly well the project of the work. Each section begins by affirming Pradeaus basic claim about the protreptic character of legislation, but goes on to develop points that underscore the concerns that have been raised. Book IV Book IV begins with Kleinias and the Athenian engaging in a brief discussion of the topography and natural resources of the city of Magnesia. What appear initially to be certain deficiencies in the location and physical endowments of the region, such as the lack of a harbor and trees suitable for boatbuilding, as well as rugged and hilly terrain, are seen in a quite positive light by the Athenian. The lack of means to establish an easy commercial enterprise and the inability to provision a navy are pluses because these things breed vice and ill conduct in the citizens. At 705e ff. the Athenian offers a definition of the purpose of legislation as tending to produce virtue. 706d strengthens this line of thinking by asserting that citizens should never be habituated to vicious habits. At 707d the Athenian adds that their study of legislation and topography should be made with a view to excellence, and that the citys purpose should not be merely preservation but the attainment of moral goodness. On the surface, this introduction to the legislative project could not be a more perfect illustration of the thesis that the purpose of law is to produce a virtuous political community. And indeed, it appears indisputable that this is the way in which Platos characters consistently articulate their understanding of the role of law. The context of the passage, however, puts this straightforward conclusion into a more complex light. At 708d, the Athenian suggests some notes of caution. The founding may be difficult because the city will require much time and effort on the part of the legis-

COMMENTARY ON PRADEAU

147

lator to get the citizens breathing together in unison like a pack of horses. Founding a city is a great test of the legislators virtue. 709a-d suggests in addition that divine concurrence, a good deal of luck and a lawgiver who possesses truth is required. These are by no means incidental obstacles. Consider also that the Athenian dislikes what he sees as the vicious qualities that are inculcated by naval warfare. Essentially, naval invasions are fought guerilla style, with a continuous series of fight and flight maneuvers, which he apparently regards as contrary to appropriate courage in battle. He is faced with the reality, on the other hand, that the previous success of the Cretan military in battle against Athens owed a great deal to this strategy and to the Cretan navy. Apparently, founding the perfect city may have tragic consequences. One can afford to legislate this sort of city into being as an imaginary entity, like old men playing a childs game, as the Athenian describes their collective enterprise at 712b, but it is not altogether certain that one can found such a city in fact and in deed. Several similar instances throughout the succeeding books culminate in the doubts expressed in Book XII, where the attempt to found Magnesia is described in the final analysis as an act of gambling. 14 Platos ambivalence comes through loudly and clearly to the attentive reader. The difficulty of founding the city of Magnesia can only be overcome by a legislator who has certain extraordinary qualities, including knowledge and the virtues. How can we produce such an individual? Furthermore, how can the city reproduce him? Plato is sketchy about this point in the Laws, although he suggests in Book XII that the legislator can attempt to perpetuate his abilities through the provision of Nocturnal Councilors who must be given the sort of philosophic education Plato proposes for some of his Guardians in the Republic. The truly great legislator who can found or continually re-found the city as necessary must possess the poetic vision of the author of the preambles mentioned above. The discourse at the end of Book IX, where Glaucon and Socrates discuss the character of the wise legislator in the Republic is therefore pertinent to understanding this type. At Republic 590e we find a similar conception of the purpose of law as the production of perfect virtue, but we also find a skeptical portrait of the relationship of the legislator to the community. In order to maintain the appropriate balance of the various elements within the tri-partite soul necessary for perfect virtue, the legislator it seems must avoid political life. He will be unable to participate in politics because the

_________
14

See Laws XII: 968e-969a.

148

GAVIN T. COLVERT

demands of such a life inherently tend to the corruption of his soul. In the end, Socrates concludes that the ideal of perfect legislation can be contemplated as a model, but it cannot be realized in reality without some sort of divine intervention. The Laws appears to take a more optimistic view of the possibility of the legislator in Book IV. At the same time, the reader is informed that chance and divine intervention are necessary for success. Furthermore, the Athenian himself remains an advisor to Kleinias and in the final lines of the dialogue Plato chooses to silence him at the moment when his ultimate consent to be a part of the concrete founding of the city is solicited by his interlocutors. As if this situation is not improbable enough, the Athenian introduces another important element of the foundational portrait at 709e. The legislator will wish for an autocratic ruler in control of this society in order to promote rapid change. Better yet, and to the chagrin of his interlocutors to be sure, the Athenian suggests supreme power and wisdom should be combined in the legislator himself (712a). As if to underscore the improbability of this situation, he characterizes the foregoing description of the initial conditions of the state as a myth. We may be inclined to regard the demand for a virtuously perfect autocrat as merely a device to promote rapid germination of the Magnesian city in the laboratory so to speak, but not as a condition which is essential to Platos account of the role of law in founding a really existing political community. The difficulty with this assumption is that it is hard to square with the surprising turn of events that culminate in the founding of the Nocturnal Council in Book XII. This passage will be discussed momentarily. Returning to the development of Book IV, the Athenian reiterates the familiar line of argument that the purpose of law is to produce a virtuous community. At 718b he asserts that with the blessing of divine concurrence and citizens who observe the laws, a city will be happy. The law contributes to this outcome partly by persuasion and partly by coercion. The laws ability to achieve its purpose, however, is complicated by the fact that people genuinely interested in the attainment of virtue are very hard to come by because it requires a lot of toil and sacrifice, whereas vice is perceived to be easy and sweet (718d-e). The juxtaposition of these points causes one to wonder: for how many in the community is the protreptic rather than the punitive role of law really meaningful in Platos assessment?

COMMENTARY ON PRADEAU

149

In order to elaborate this point, the Athenian contrasts the poet and the legislator. 15 Whereas the poet speaks with many voices, even to the point of contradiction, the legislator must always speak with a single voice. Upon careful reflection the interlocutors realize that the legislator must actually speak with two voices conveying a single message: one exhortative and one punitive. At 720a-e a comparison previously mentioned is made between Greek physicians and their slaves who often serve as physicians assistants. The physicians assistant dispenses medical advice or prescriptions without further explanation, whereas the genuine doctor converses with the intelligent patient and produces understanding of the disease and consent to medical treatment before offering a prescription. Like the best doctor, the Athenian argues that the law should contain a persuasive or exhortative dimension, not merely a prescription or command of action. Two points about the persuasive dimension of law are worth observing. First, it should be noted that the Athenian explicitly distinguishes this persuasive dimension as a separate or distinct preamble from legislation proper. Second, it is not law as such, but something which is designed to secure docile obedience to the law. Furthermore, from what we know of Platonic psychology, such obedience to law through understanding of the preambles would already require a well ordered soul. So, the protreptic element of legislation depends for its success upon some as yet unspecified additional element of moral education. As will become clear in the case of the members of the Nocturnal Council, nothing specified in the legislative work of books 4-12 is sufficient to capture this work of moral education. A further point made by Leo Strauss about the speech of the legislator in relation to the broader context of the dialogue is worth mentioning. 16 As Socrates notes in the Phaedrus, a defect of all ordinary writing is that it must do exactly what the traditional legislator does, that is it speaks with a single voice on every subject. Strauss points out that the Platonic dialogue overcomes this limitation of ordinary writing by speaking equivocally or with polyvalent meaning. In answer to the question of whether Plato in the Laws is writing a dialogue or acting as a legislator, Strauss concludes Platos writings, including the Laws, are as remote as possible from the legislators writings. 17 Again, we hear echoes of Pascals hyperbole,

_________
See IV: 719b-e. See Leo Strauss, The Argument and the Action of Platos Laws, Chicago: University of Chicago Press, 1975, p. 62. 17 Ibid., p. 62.
15 16

150

GAVIN T. COLVERT

although notably Strauss does not say that Platos writings are apolitical. It would appear that Strausss assertion is directly contradicted by the Athenians proposal in Book VII that the children of the regime should read the Laws as a part of their education, and that the author of the Laws inaugurates a new kind of poetry with his legislative speeches. 18 Perhaps it is more accurate to say that Platos Laws is both legislative and philosophical, and that the former as well as the latter activities are thoroughly political. Plato has inaugurated a new kind of legislative discourse that transcends the traditional legislative model, precisely because it incorporates philosophical elements, and speaks simultaneously to differing audiences. He does this explicitly in the case of the preambles, but also indirectly by means of the dramatic structure of the dialogue. With this important qualification, however, Strausss assertion remains significant. Consider, for example, the experience of the youth who will be reading this new kind of poetry or myth, which is represented by the Laws. Some will surely take its prescriptive dimension at face value as the specification of the duties and responsibilities of Magnesian life. They will find in themselves affinities to Kleinias and Megillus as models. Others, however, will be struck with wonder by the very human and dialectical character of the interaction between the parties to the discussion. 19 Moreover, they will find a model in the character of the Athenian, who raises troubling questions that seem to leave his interlocutors puzzled. Depending upon their reaction to the dialogue in early life, they may or may not be suited to serve as Nocturnal Councilors when they get older. While we ought not to be inclined to read Platos Laws purely as esoteric writing, or as a comedy, there is evidently some need for circumspection with regard to the surface employment of terms offered by his interlocutors. The context and setting of the dialogue provide an intonation to those terms which shifts their role and purpose in significant ways. The peculiar turn of events that takes place in Book XII demonstrates that there is more at work beneath the surface portrait of legislations role in the production of the virtuous community.

_________
18 See Laws VII: 811c-812a. I am indebted to the helpful comments of an anonymous reader for the suggestion of this passage and some objections to Strausss assertion. 19 Pangle elaborates this point in his interpretive essay, cf. pp. 490-1.

COMMENTARY ON PRADEAU

151

Book XII We find in Book XII a similar situation to that in Book IV. In support of the definition of the purpose of law that Pradeau proposes, there is one of the clearest examples of this definition in a passage at 963a. Occurring near the close of the dialogue, an interesting exchange takes place between the Athenian (who remains a stranger) and Kleinias the Cretan. Having taken the lead in the conversation from the Athenian, Kleinias observes that law should always aim at the singular end of virtue. It is curious, of course, that this definition is placed in the mouth of Kleinias, rather than the Athenian, but the Athenian appears to concur with Kleinias assertion. What is even more noteworthy, however, is that this definition occurs within a passage that introduces the Nocturnal Council as the guarantor and salvation of the political community. In other words, that institution places the truth of Kleinias assertion in its proper context. A brief review of the key features of the Council and the character of prospective councilors is revealing. At 962b we learn that members of the Council must have perfect virtue and a genuine understanding of the true nature of statecraft. This evidently includes an understanding of the soul. In order to attain this state we learn at 965b that the program of education for the councilors must be much more rigorous and exacting than that for ordinary citizens. Indeed, at 966c we learn that the Athenian is content to allow the majority of citizens merely to observe the letter of the law, rather than follow it from a level of genuine depth and understanding. This admission appears startling when one considers the distinction between better and worse physicians from Book IV above. If the persuasive dimension of legislation is precisely analogous to the discourse of the better physician with a patient who is capable of understanding, for whom does the Athenian think the protreptic dimension of the law is truly meaningful? One must conclude that he thinks this is the case for a small minority within the political community. Plato provides only the barest sketch of the education that the Nocturnal Councilors will receive at 965-967. The fact that he deliberately condenses the treatment of the subject is itself worthy of consideration. We recognize in it the philosophic training of the wise man to see the one over many and the form of the Good in the Republic. It would appear that the appropriate sequel to the last book of the Laws may well be the latter books of the Republic. It is doubtful therefore that those who lack the sort of education that the councilors will receive will be able to attain the appropriate state of soul required for genuine human excellence. Yet, this education is clearly something other than what the discipline of the laws

152

GAVIN T. COLVERT

provides in the political community and cannot be inculcated in the ordinary citizen. Plato concludes at 966d that a prospective councilor must be divine himself or have studied divine things carefully. Perhaps the most significant single point to be observed about the legislation of the Nocturnal Council is that it is impossible to legislate its authority and purpose before it has been constituted. Furthermore, while the program of education undertaken by the councilors can be described, it cannot be prescribed in advance. The Council literally transcends the law. In short, the apparently exhaustive and detailed account of the moral and intellectual life of the Magnesian political community has just been turned upside down by the effort to create an institution of the Platonic philosophical life. What are we to make of these developments? A reasonable conjecture is that Plato is writing for a certain audience and that he adapts his presentation of the role and purpose of law accordingly. With the very same definition of the end of law as we find in the Laws, Plato displays less confidence in the ability of law to accomplish its end in the Republic. In the case of the Republic, that is because the fictional audience and dramatic setting involves a dialogue between Socrates and a group of young men in which Socrates is engaged in his typical project of trying to initiate his interlocutors into the philosophical life. The characters who participate in the latter stages of the conversation aim to cultivate the friendship appropriate to philosophical dialogue, unlike interlocutors such as Thrasymachus who have already been excluded from the conversation. The discussion of legislation in the Laws expresses more confidence in the laws ability to habituate citizens to virtue because of Platos more straightforwardly prescriptive and less philosophical expectations for a large part of its audience, or because it serves as a model for conversation with those non-philosophical types whom the attentive reader of Plato will be dealing. Thus, terminological considerations alone neither confirm nor deny the conclusion that Platos view of law changes in the dialogues over time. A Platonic dialogue must speak simultaneously to different persons with different voices. The Laws places less stress upon a program of philosophic education for the councilors and expresses less ambivalence about the possibility of participating in public life, because it is an argument made primarily by a group of strangers who mirror the characteristics of Platos intended audience. The Nocturnal Council is treated at the end without great detail because of what the strangers in the conversation are able to hear.

COMMENTARY ON PRADEAU

153

We need not draw the conclusion from these considerations that Plato has a deep mistrust of the role of law in the political community or that he is speaking in a strictly esoteric fashion in the Laws. But, it is reasonable to conclude that he thinks the laws must aim at a good or purpose that they cannot produce entirely in virtue of themselves. They are not merely a second-best solution. They create the conditions in which excellence can flourish, although that kind of human excellence requires something other than law for its achievement. Law is therefore a foundation for virtue in the political community. Pradeau presents a strong case for the claim that key characters in the later Platonic dialogues employ the term consistently throughout, but the context of his imaginary and intended audiences changes the use to which the term is put. Furthermore, the different uses of the term in the Republic, Statesman, and Laws reveal deep ambivalence about the ability of law to found and produce full virtue in a political community without the deeper and more elusive contributions of the philosophic life.
ASSUMPTION COLLEGE

PRADEAU/COLVERT BIBLIOGRAPHY
Bertrand, J.-M. 1999. De lcriture loralit: Lectures des Lois de Platon. Paris. Brisson, L. 2000. Les prambules dans les Lois. Lectures de Platon: 235-262. Costantini, M. and J. Lallot. 1987. Le proomion est-il un prome? Le texte et ses representations: 13-27. de Romilly, J. 1971. La loi dans la pense grecque. Paris. Gastaldi, S. 1984. Legge e retorica. I proemi delle Legge di Platone. Quaderni di Storia 10, 20: 84-86. Hansen, M. H. 1999. The Athenian Democracy in the Age of Demosthenes: Structure, Principles, and Ideology. J. A. Crook (tr.). Blackwell. Lvy, E. 2000. La cohrence du code de Gortyne. La codification des lois dans lantiquit: 185-214. Lisi, F. L. 2000. Les fondements mtaphysiques du nomos dans les Lois. Revue philosophique 1: 57-82. Maffi, A. 2003. Studi recenti sul codice di Gortina. Dike 6: 161-225. Millet, P. A. and S. C. Todd. 1990. Law, Society and Athens. Nomos: Essays in Athenian Law, Politics and Society. P. A. Cartledge, P. A. Millet and S. C. Todd (eds.). Cambridge. Pangle, Thomas L. 1988. The Laws of Plato. Chicago. Pascal, Blaise. 1962. Penses, Texte tabli et annot par Jacques Chevalier. Paris. Strauss, Leo. 1975. The Argument and the Action of Platos Laws. Chicago. Todd, S. C. 1993. The Shape of Athenian Law. Oxford. Yunis, H. 1996. Taming Democracy. Ithaca.

COLLOQUIUM 5

ARISTOTLE AND THE METAPHYSICS OF METAPHOR

FRAN OROURKE

Asked what his first decree would be, were he to become emperor, Confucius allegedly replied that he would fix the meaning of words. It is easy to appreciate the good intentions of the eastern sage; Aristotle may have had something similar in mind when he stated that a word which does not have a single meaning has no meaning. 1 This expresses a central truth about the nature of language, thought and reality; not, however, the full truth, since language does not lend itself to such Procrustean fixity; Aristotle recognised this better than most. Only a tyrannical philosopher king could legislate as suggested by the anecdote concerning Confucius. Perhaps the clearest challenge to such a decree is analogy; this occurs most commonly as metaphor, which is surely one of the most marvellous feats of language. Bereft of metaphor, everyday language would remain flat and univocal, each word atomically attached to a single object. Indispensable to our way of understanding and articulating the world, metaphor is richly revealing of the relationship between knowledge and reality. It deeply penetrates our way of perceiving and expressing the world. John Middleton Murry did not exaggerate when he remarked: To attempt a fundamental examination of metaphor would be nothing less than an investigation of the genesis of thought itself. 2 Metaphor means literally transfer or transport. The word is used as such by Herodotus, who relates that the Athenian tyrant Pisistratus removed all the dead that were buried within sight of the temple and carried them to another part of Delos. 3 He also uses the word to describe the use of levers for the lifting of stone in the construction of pyramids. 4 These are both strongly physical and visible uses of the term. The first, as it were, metaphorical use of the wordas a nounis found in the orator Isocrates, who describes the wealth of stylistic means enjoyed by poets, compared to the dearth of literary devices available to prose writers: The poets are granted many methods of adorning their language, for besides

_________
1 2 3 4

Metaph. , 4, 1006b8: J. M. Murry, 1931, p. 2. I, 64.2-3: II, 125, 4.

. .

156

FRAN OROURKE

the use of normal words they can also employ foreign words, neologisms, and metaphors while prose writers are allowed none of these last three, but must severely restrict themselves to such terms alone as citizens use and such arguments as are precisely relevant to the subject matter. 5 Metaphor was primarily understood by Isocrates, therefore, as a means of poetic adornment. While he was himself a master of metaphor, Plato does not name it as , meaning to transfer an object from one such. 6 He uses place to another. Interestingly he employs the expression , meaning to translate from one language into another. Aristotle was the first to offer a systematic study of the essential nature and structure of metaphor. Umberto Eco has suggested that of the thousands and thousands of pages written about the metaphor, few add anything of substance to the first two or three fundamental concepts stated by Aristotle; 7 such a claim may seem exaggerated in view of the voluminous literature that has since appeared, especially in recent decades. Another author refers to the Stagirites astonishingly modern description of metaphorical processes. 8 While Aristotle could not have anticipated the variety of theories now current, many interpretations will find support in his stated views; his perspective, however, may not be reduced to any one in particular. In the following reflections I wish to consider some of the presuppositions of Aristotles theory of metaphor, and relate them to other aspects of his philosophy, especially his metaphysics, epistemology and psychology. My focus is metaphor as a token for the analogous unity pervading the diversity of the world, and as an index of mans psychosomatic unity. The key to Aristotles approach is his understanding of metaphor as analogy; much discussion of metaphor as a linguistic or literary device has unfortunately neglected this. Analogy is of the essence of metaphor. It relies on the diversity and unity both of human knowledge and human nature, and on the diversity and interconnection of beings within the cosmos. Aristotle famously defines metaphor in the Poetics 9 as the transfer to one thing of a term belonging properly to another, i.e., an alien or strange name ( ). This may occur, he explains, in one of four ways: from genus to species, from species to genus, from species

_________
Evagoras 190 D. Translation W. B. Stanford, 1936, p. 3. See Stanford, pp. 3-4. An excellent account of Platos use of metaphor may be found in E. E. Pender, 2002. 7 U. Eco, 1984, p. 88. 8 P. Swiggers, 1984, p. 40. 9 Poet. 21, 1457b7.
5 6

ARISTOTLE AND THE METAPHYSICS OF METAPHOR

157

to species or, finally and most significantly, according to analogy or proportion ( ), 10 expressing thereby a similarity of relations. Metaphor through proportional analogy, he explains in the Rhetoric, is valued most of all. 11 While metaphor traditionally refers only to the fourth typeproportional metaphorthe first three also illustrate different levels of unity and diversity. These forms of so-called metaphor function, however, on the basis of a manifest similarity which is transferred univocally rather than by analogy. Aristotles definition of metaphor proves, if proof were needed, that there is nothing more elusive or difficult to define. The word metaphor is already metaphorical; the best Aristotle can do is coin a variant ( ), simply by changing the prefix. Metaphor is the imposition upon the object of a name belonging to another. conveys the notion of adding to, or placing something upon something elsefor Thucydides meant an additional payment. 12 Is there a tautology here? Is Aristotles definition circular? Perhaps, but not viciously so. It reveals rather a hermeneutic circle in which we find ourselves firmly centered and which allows us to extend the horizon of our world. We are on sure ground, since we spontaneously affirm the existence of diverse beings, recognise simultaneously their similarities, and deny their identity. Since we are also able to distinguish between the proper ( ) and transferred meaning of our conceptual terms, Aristotles definition merely articulates what we already experience. For a definition of proportion we may consult the Nicomachean Ethics, where it is defined as an equality of ratios, implying at least four terms. 13 In the Poetics Aristotle prescribes the following formula: Proportional metaphor is possible whenever there are four terms so related that the second is to the first, as the fourth to the third; for one may then put the fourth in place of the second, and the second in place of the fourth. 14 He illustrates this by the poetic parallel between Ares shield and Dionysus cup: as the shield is to Ares, so is the drinking bowl to Dionysus. Thus the cup is, as it were, Dionysus shield, and the shield

_________
10 Poet. 11 Rhet.

21, 1457b9. III 10, 1410b36-1411a1: . 12 See J. T. Kirby, 1997, p. 532. 13EN V 3, 1131a31-32: 14 Poet. 21, 1457b16-19:

. Translation Bywater, slightly amended.

158

FRAN OROURKE

Ares cup. 15 Again, Old age is to life as evening is to day. Thus old age is called the evening of life or sunset of life. 16 What is transferred in metaphor? A likeness of relationship between two or more unrelated pairs of individuals. Metaphor is essentially the recognition of likeness in unlike things. The merit of metaphor is to recognise deep and hidden similarities: just as in philosophy also an acute mind will perceive resemblances even in things far apart. 17 The key to proportional metaphor is the perceptionperhaps imaginativelyof a novel resemblance between two pairs of coordinates not normally conjoined. Aristotle goes so far as to declare that the gift for metaphorthe perception of unlikely likenessis a true sign of genius: it is the one thing, he states, which cannot be taught by another. 18 (It is interesting that in the Nicomachean Ethics he states that the moral vision whereby one discerns what is truly good, is also a gift of naturethe greatest and most noble which likewise cannot be acquired or learnt from another.) 19 Having enumerated all the means and literary devices which the poet has at his disposal, Aristotle declares in the Poetics: It is a great thing, indeed, to make a proper use of these poetical forms, as also of compounds and strange words. But the greatest thing by far is to be a master of metaphor ( ). It is the one thing that cannot be learnt from others; and it is also a sign of genius, since a good metaphor implies an intuitive perception of the similarity in dissimilars ( ). 20 With delightful irony George Eliot chides the philosopher: O Aristotle! if you had had the advantage of being the freshest modern instead of the greatest ancient, would you not have mingled your praise of metaphorical speech, as a sign of high intelligence, with a lamentation that intelligence

_________
Rhet. III 4, 1407a15-17: , . See Poet. 21, 1457b 20-22. 16 Poet. 21, 1457b22-25. 17 Rhet. III 11, 1412a9-12: , , , . Translation Roberts. 18 Rhet. III 2, 1405a8-10: , . 19 EN III 5, 1114b9-10: , . 20 Poet. 22, 1459a4-8. Translation Bywater. Aristotle remarks elsewhere that in the case of things that greatly diverge, much practise is needed; in other things, similarities are more easily seen. (Top. I 17, 108a12-14: .)
15

ARISTOTLE AND THE METAPHYSICS OF METAPHOR

159

so rarely shows itself in speech without metaphorthat we can so seldom declare what a thing is, except by saying it is something else? 21 It is indeed true that intelligence rarely shows itself in speech without metaphor. Aristotle remarks in the Rhetoric that alongside ordinary, regular, words ( ), everybody uses metaphor in normal conversation ( ). 22 This is evidence of a natural and universal inclination towards metaphor. The spontaneous and unreflective use of metaphor has been seen as indicating something elemental in human knowledge. According to Giambattista Vico, in order to understand how primitive man interprets the world, we need simply examine his metaphors. Benedetto Croce sums up Vicos view: Poetry is the primary activity of the human mind. Man, before he has arrived at the stage of forming universals, forms imaginary ideas. Before he reflects with a clear mind, he apprehends with faculties confused and disturbed: before he can articulate, he sings: before speaking in prose, he speaks in verse: before using technical terms, he uses metaphors, and the metaphorical use of words is as natural to him as that which we call natural. 23 While this is certainly exaggerated, it is doubtless true that everyday language is suffused with metaphor. There is no contradiction in Aristotles statement that while everybody uses metaphor, the mastery of metaphor is a sign of true genius. Cicero later distinguished between the creation or invention of metaphor, and its use; even children and fools use metaphor! Metaphor manifests itself at diverse levels of intelligence. I wish to distinguish between those metaphors with limited cultural value and those which are universal in scope and which, I suggest, indicate something essential in human nature. To illustrate I will refer to a common scene in the modern Greek capital. Standing on any busy street in Athens, you would not have to wait long to see a truck drive by with the word metaphors ( ) painted on its side. This is, of course, not a dialup delivery service for poets with writers block but more prosaically a removals company vehicle; as already noted, the word metaphor means literally transport. There is an entire stoa, or covered archway, in the center of the city, occupied by companies specializing in metaphors. It is interesting that while some advertise , i.e., national metaphors, others offer international metaphors ( );

_________
21 22

Eliot, G., 1980, p. 123. Rhet. III 2, 1404b34-35: . 23 Croce, 1964, p. 48.

160

FRAN OROURKE

yet more promote , general metaphors. (The Greeks even have a government minister for metaphors! The Athens telephone directory has ten pages advertising topical metaphors, frozen metaphors, air and sea metaphors; you can even chose between esoteric and exoteric metaphors.) The basic division which I wish to mention is that between ethnic and international or general metaphors. Speaking of ethnic or national metaphors, we can agree with Vico that we can learn much about the mentality and tradition of a people from the metaphors embedded in its language. Consider the countless maritime metaphors in English, inconceivable in the language of a landlocked nation. The frequency of nautical and maritime terms reflects the importance of the sea in English history. I have counted no less that sixty expressions originating in sailors language which are part and parcel of English. Other languages have copious terms drawn perhaps from military or agricultural life. There are thus what we may loosely call cultural metaphors, particular to a people or nation. However we may also note, above and beyond the diversity of individual languages, a host of metaphoric meanings which transcend regional boundaries. These indicate, not particular cultural, geographic or historical characteristics, but essential aspects of human nature and mans fundamental relationship with the world. They are truly international or universal metaphors. Predominant among such universal metaphors which may be observed across cultural divisions are those intended to explain mental activity by means of terms drawn from the physical world. I will give just two examples, taken from the vocabulary used to describe knowledge. Firstly perception: English and all the Romance languages adopt the Latin word perceptio, deriving from capere, to take, seize or lay hold of; the Greek word likewise derives from , also to take. Similarly the language of conceptual comprehension: the Latin concipere, its synonym comprehendere, the Greek and German begreifen, likewise understand intellectual knowledge as a seizing or grasping. The psychic activity of knowing is conveyed with terms drawn from the physical activity of taking hold of, seizing and gripping. The transfer of physical terms to intellectual activities makes perfect sense in light of Aristotles insistence that all concepts are founded upon sense experience. Because of his composite nature man needs metaphor to bind the physical and the psychic, the external and internal, the sensible and intellectual. Aristotle remarks: The beauty of the body is seen,

ARISTOTLE AND THE METAPHYSICS OF METAPHOR

161

whereas the beauty of the soul is not seen. 24 In keeping with the Aristotelian concept of man, it is entirely natural for us to elaborate abstract concepts from our knowledge of concrete objectsnatural because necessary. Man needs to figure his speech, so that it can turn from the domain where it initially belongs and for which it is properly fitted, towards the realms which surpass the physical. Aristotle continually created linguistic analogies by enlisting everyday concepts in the service of philosophy. Porphyry begins his commentary on the Categories with the question why Aristotle chose as a title for his work a term which in ordinary language refers to the speech of the prosecution against the accused in the law courts. He explains that while ordinary language communicates everyday things, philosophers are interpreters of things that are unknown to most people and need new words to communicate the things they have discovered.
Hence either they have invented new and unfamiliar expressions or they have used established ones in extended senses in order to indicate the things they have discovered . . . . So even though is applied in ordinary usage to the speech of the prosecution which presents evidence against a defendant, he adopted the word, and chose to call those utterances in which significant expressions are applied to things predications ( ). 25

Simplicius likewise recognizes the clear fittingness for Aristotle of the transition by analogy from sensible to intelligible things ( ). 26 Just as intellectual knowledge is rooted in the senses, so too are those various terms we use to describe cognition itself. We have no terms other than physical with which to denote non-physical, immaterial or psychic activities. The reason we transfer to mental acts the names of physical activity is because of the analogous similarity perceived between the two. Consider what we do with ideas and what they do to us. We can trace an idea, pursue it, get our head around it, embrace it, take it to heart and dwell on it; we may put it on the table or into someones head. Ideas dawn upon us, cross our mind and enlighten us; they trickle down and perhaps inundate us. 27 Among the natural processes which provide rich metaphoric

_________
Pol. I 5, 1254b38-39: . Translation Jowett. 25 Porphyry. 1992, pp. 29-31. 26 Simplicius. 1907, p. 74. 27 We chew on ideas, swallow and digest them, put them in our pipe and smoke them. An idea can be bright, dull or dumb, be up to scratch, have an upside or a downside. We can have sharp ideaswith a point, unless they are dull, blunt and rigid. We put a spin or a slant
24

162

FRAN OROURKE

motifs for mental or spiritual realities and activities are those of physical force, light, nutrition, growth, reproduction and birth. The world of the psyche mirrors the realm of nature. In the words of Emerson, The whole of nature is a metaphor of the human mind. 28 The language of mind is largely metaphorical and refers to phenomena of the body. An adequate explanation of metaphoric signification must account for the unity which necessarily underlies the duality of the domains from which this wide array of expression is drawn. Aristotles doctrine of the distinction of body and soul, yet their complementarity and unity, provides precisely such an adequate foundation. The unitive power of metaphor, expressing a mental function by analogy with its physical parallel, can only be explained by recognising a unitary subject of cognition, whose mode of knowledge equally involves physical and mental operations. Metaphor effects between disparate domains a unity which mirrors the relation of body and mind. Aristotle refers repeatedly to mans composite nature; man is a . 29 Metaphor is likewise a . As already stated, metaphor is the transfer of a name from one object to another on the basis of analogical similarity, i.e., of likeness through

_________
on an idea, iron it out, hammer away at it and drive it home; we focus on an idea, underline it, or have it up our sleeve. We toy with the grain of an idea, get to the kernel, but may find it hard to crack. Ideas emerge or spring to mind; they percolate and trickle down; an idea might make a splash or cause a ripple; we can be flooded by a spate of ideas; we channel, float and filter; fish and trawl for ideas. We warm to an idea, put a damper, or throw cold water on it; we put it on ice or on the back burner. An idea may be threadbare or redundant; if it has a silver lining we may cash it inunless it has become bankrupt. We convey ideas, ditch them, drop and dump them, throw them out the window. An idea can be a red herring, which we ram down someones throat. It can be pregnant and bear fruitprematurely if its before its time. If an idea adds up, we can break it down; it can measure up and outweigh the opposition. We can run it up the flagpole; it may take off, or sink like a lead balloon. It may be a milestone, perhaps even pioneering. An idea may be in the pipeline, or coming down the track. We can map it out; if it is explosive it will break new ground. An idea sometimes takes on legs and does the rounds; but occasionally it comes home to roost. Consider the language of violence attached to ideas: an idea can grab, strike and stun us; we hit on an idea; we can be floored or flattened by an idea; it can hit us like a ton of bricks. We can seize upon an idea, come to grips with it, get a handle on it, grasp it, hold it, pull it asunder, jump at it, grapple with it, toss it about, push it too far, pick it up and run with it. We struggle with ideas, pin them down, knock them on the head, bounce them off one other and take them apartit sounds just like wrestling! 28 Emerson, 1950, p. 18. See p. 15: Man is an analogist, and studies relations in all objects; p. 130: Nature is full of a sublime family likeness throughout her works, and delights in startling us with resemblances in the most unexpected quarters. 29 EN X 7, 1177 b28-29; X 8, 1178a20.

ARISTOTLE AND THE METAPHYSICS OF METAPHOR

163

equality of ratios ( , ). 30 Analogy is a similarity of relationship, a correspondence of proportionor, as both Alasdair MacIntyre 31 and Martin Heidegger 32 term it, a relation of relations. Before considering the kind of analogy which constitutes metaphor, it will be helpful to review the broader meaning of analogy for Aristotle, and the use made of it throughout his system. Aristotle uses it extensively in his overall synthesis of knowledge, particularly his metaphysics and biology. The aspect of analogy which I wish to emphasise is its power of universal reference and comprehensiveness; it is this which ultimately allows metaphors of proportion to be predicated across the most widely diverse contexts. Analogy refers not to any or every aspect of unity, but to the resemblance of relations within and among a diversity of beings. It is the agreement of correspondent relations which are diversely realized in different domains; it thus provides the widest possible framework for universal unity among diverse substances. In Topics I vii Aristotle distinguishes between three senses of sameness ( ): numerical, specific and generic. 33 Ten chapters onwithout naming it as analogyhe speaks of the likeness ( ) which belongs to different genera: As one thing is to another, so a third is to something else. For example, as knowledge is to the knowable, so is sensation to the sensible thing ( , ). And as one thing is in another, so a third is in something else. For example, as sight is in the eye so the mind is in the soul ( , ), and as a calm is in the sea, so is stillness in the air. 34 The unity of analogy is clearly not the unity of the individual, species or genus, 35 but of a likeness transcending all three. It is wider than genus:

_________
EN V 3, 1131 a31-32. A. MacIntyre, 1949, p. 45. M. Heidegger, 1976, p. 348. Top. I 7, 103a7-14: In general, sameness would seem to fall into three divisions; for we usually speak of numerical, specific and generic sameness. There is numerical sameness when there is more than one name for the same thing, e.g., mantle and cloak. There is specific sameness when there are several things but they do not differ in species, e.g., one man and another man, one horse and another horse; for such things as fall under the same species are said to be specifically the same. Similarly things are generically the same when they fall under the same genus, e.g., horse and man. Translation Forster. 34 Top. I 17, 108a8-12. 35 The relationship between these is the basis for the first three kinds of metaphor noted by Aristotle at Poet. 21, 1457b.
30 31 32 33

164

FRAN OROURKE

things that are one by analogy are not all one in genus. 36 In accordance with the root meaning of , it is the similarity of an intrinsic proportion which is repeated and realized across an endless number of disparate relationships. Geoffrey Lloyd explains: In such four-term proportional analogies, what is claimed is that the relationship within each pair is the same, a sameness distinct from sameness in number, sameness in species and sameness in genus, and labelled, precisely, sameness by analogy. 37 Analogy offers the widest possible ground for unity, overarching that of genus, which in turn embraces the more limited unity of species and individual. Analogy links different categories because it transcends them. In Metaphysics Aristotle defines both sameness and likeness as forms of unity:
Some things are one numerically, some in species, some in genus, some by analogy. Those things are numerically one of which the matter is one: those things are specifically one of which the definition is one: those things are generically one which belong to the same category; those things are analogically one that have the same relationship as two other things have to one another. 38

The unity of analogy transcends the unity of the individual, species and genus; it runs through all three because it surpasses them. In his zoological investigations Aristotle uses analogy to introduce order among disparate species on the basis of similarity of function or operation; put simply: birds have wings, fish have fins. 39 There are some animals whose parts are neither identical in form nor differing in the way of excess or defect; but they are the same only in the way of analogy ( ), as, for instance, nail to hoof, hand to claw, and scale to feather; for what the feather is in a bird, the scale is in a fish. 40 Each member of these distinct pairs performs a similar function within their respective natures. Analogy, according to Aristotle, also facilitates the work of taxonomy; for example, pounce (the internal shell of the cuttlefish), spine and bone are all analogues of animal bone, and may thus be classified together. 41

_________
Metaph. 6, 1017a2-3: , . Translation Ross. G. E. R. Lloyd, 1966, p. 138. I have benefited greatly from Chapter 7 (The Unity of Analogy), and Chapter 10 (The Metaphors of Metaphora) of this excellent work. 38 Metaph. 6, 1016b31-35. Translation Lloyd, 1966, p. 140. 39 PA I 4, 644a21-22. 40 HA I 1, 486b17-22. Translation Thompson. 41 APo. II 14, 98a20-23.
36 37

ARISTOTLE AND THE METAPHYSICS OF METAPHOR

165

Aristotle intentionally exploits analogy as a method of scientific order. To treat all common attributes separately would involve endless and needless reiteration, whereas to study the operation of a function in one animal will cast light upon a corresponding function in another. Aristotle therefore proposes to investigate all animals, insofar as possible, according to their similarities, principally that of function.42 It is, he states, a reputable opinion ( ) that among similars what is true of one is true also of the rest. 43 Most far reaching is Aristotles use of analogy in metaphysics. Going beyond species and genus he seeks those features and principles common to all beings precisely as beings. Some first principles are common to particular sciences. They are common, however, only in an analogical sense, since each is valid only insofar as it falls within the genus of the particular science. 44 He declares however: There is analogy between all the categories of being ( ). 45 In Metaphysics he states: In one sense, the causes and principles of distinct things are distinct, but in another sense, if one is to speak universally and analogically, they are the same for all. 46 The principles of corruptible bodies are form, privation and matter. They are fulfilled differently in each case, but relate similarly to one another precisely as principles in every unique instance. There are four distinct causes, but each acts in a mode proper to itself: cause is analogical. Matter causes the effect by supporting or sustaining form; form determines the matter. The efficient cause produces the effect, while the final cause attracts the efficient cause. Analogy operates most clearly in Aristotles elucidation of the distinction between act and potency. The distinction is disclosed inductively, and grasped analogically by way of example. It is the difference between that which builds and that which is capable of building, that which sees and that which has its eyes shut but has the power to see, the finished product compared to the raw material. 47 The relation of act and potency is verified

_________
PA I 5, 645a36-645b13, PA II 2, 647b14-15, PA II 2, 648a4-5, PA II 6, 652a3, PA II 7, 652b24-25. 43 Top. I 18, 108b13: , , . Translation Pickard-Cambridge. 44 APo. I 10, 76a37-40. 45 Metaph. 6, 1093b18-19. Translation Tredennick. 46 Metaph. 4, 1070a31-33. Translation Apostle. 47 Metaph. 6, 1048a35-b9.
42

166

FRAN OROURKE

analogically in the duality of prime matter and substantial form, 48 and in the distinction of substance and accident. The analogous principles of act and potency; matter, form and privation; the reciprocal and dynamic relationship of causes; all conspire to shape Aristotles vision of a unified cosmos. Nature is inherently coherent. It is not, as he expresses it, a series of episodes, like a badly constructed tragedy. 49 The perception of the world as an interrelated wickerwork of substances and causes gives foundation to the conviction that the cosmos is essentially and integrally united. All things are ordered together somehow, but not all alike,both fishes and fowls and plants; and the world is not such that one thing has nothing to do with another, but they are connected. For all are ordered together to one end. 50 Aristotle would doubtless agree with Thomas de Vio, Cardinal Cajetan, who wrote in his highly influential work De Nominum Analogia (1498): An understanding of this doctrine is so necessary that without it no one can study metaphysics, and ignorance of it gives rise to many errors in other sciences . . . . Metaphysical speculation without knowledge of [proportional] analogy must be said to be unskilled. 51 Analogy is intrinsic to our human mode of cognition, discovery and creativity; it is a mental crossing of the barriers from one science, art, or region of experience, into another. Analogy is the keya veritable passe-partoutwhich unlocks the structure of thought in its dual attitude to the unity and multiplicity of the world, as it engages in the twin approaches of analysis and synthesis. Aristotle clearly grasped the importance of his own insight that the causes of all things are the same analogically; Marie-Dominique Philippe suggests that Aristotles use of analogy best characterizes his philosophical approach. 52 Proportionality, moreover, is important for Aristotle in all areas of reality and human activity. Justice is defined as proportion. The demands of justice must take into account the circumstances of the individual situa-

_________
Analogy makes the knowledge of prime matter possible (Phys. I 7, 191a8: ) 49 Metaph. 3, 1090b19-20: , . 50 Metaph. 10, 1075a16-19. Translation Ross. 51 de Vio, 1952, p. 3: Est siquidem eius notitia necessaria adeo, ut sine illa non possit metaphysicam quispiam discere, et multi in aliis scientiis ex eius ignorantia errores procedant. See p. 29: Unde sine huius analogiae notitia, processus metaphysicales absque arte dicuntur. Translation Bushinski, pp. 9, 29. 52 M.-D. Philippe, 1969, p. 1
48

ARISTOTLE AND THE METAPHYSICS OF METAPHOR

167

tion; instead of being imposed in unbending fashion as an iron rule, it must adapt itself with equity to the situation. Aristotle aptly conveys this with the image of the leaden rule used by the builders of Mytilene to harmonize the uneven edges of the building stones. 53 Political life demands equitable harmony; in Rhetoric I he draws a parallel between the balance required between leniency and severity in a democracy, and the mean between aquiline and snub in a handsome nose. 54 There should be a certain proportion and fittingness between ones position in life and the possession of goods. 55 In friendship among unequals, love should be proportional, i.e., analogously balanced by different levels of dedication and response. 56 Now it is this fundamental ontological, analogical, relatedness among beings which provides the profound basis for metaphor. We need however to distinguish between metaphor and analogy in its proper sense; this is also to clarify the distinction implied by Aristotles definition of metaphor between the proper meaning of a word ( ) and its metaphorical or non-proper meaning. Aristotles definition of metaphor as the transfer of a name from its proper to an alien context is echoed in the medieval characterization of metaphor as improper analogy. This is found in Aquinas, and canonised in Cajetans influential work On the Analogy of Names. In order to distinguish between simple analogy and metaphor (what we might call intrinsic and incidental analogy, rather than proper and improper analogy), let us examine the various ways in which beings resemble one another analogically. Firstly, for Aristotle the metaphysical principles of being are perfectly realized in every individual. They are properly and intrinsically affirmed in the case of every particular entity in all its uniqueness. Metaphysical principles are affirmed proportionately of every entity by proper analogy, whereas metaphor is the proportional, but imperfect, transfer of a perfection or activity from its primary to a secondary subject. Secondly, similar functions, operations and actions are also predicated properly and analogically of substances belonging to different genera, because of a real similarity in the corresponding roles which they perform in accordance with their own nature: the bird flies, the fish swims both move. In metaphor, however, and here is the point, what is affirmed

_________
53 54 55 56

EN VI 10, 1137b30-32. Rhet. I 4,1360a Rhet. II 9, 1387a27-1387b2. See EN VIII 7, 1158b23-28.

168

FRAN OROURKE

is not a proper analogy but an imperfectly analogous resemblance: the quality, perfection or action belongs perfectly and intrinsically only to one substance, and is transferred to another because of some perceived but imperfect likeness. In the assertion Achilles is a lion, the poet is not attributing to the hero either the nature of a lion, nor its beastly rapacious activity as such, but rather a certain secondary likeness. Metaphor is the proportional, but imperfect, transfer of a perfection or activity from its primary to a secondary subject. In metaphor, a name which belongs intrinsically to one being is transferred to another, not by virtue of what it is properly in itself, but through a relation of proportional similarity in some secondary or accidental respect. This similarity is frequently glimpsed only through the creative imagination. At this point I wish to propose that it is action which constitutes the metaphysical foundation of metaphoric resemblance. Aristotle hints at this, but does not make it explicit. In Rhetoric III he repeatedly notes that one of the primary virtues of analogous metaphor is to place things before the eyes ( ), i.e., to bring them to life. Things are set before the eyes, he explains, by words which represent them in a state of activity ( ). A metaphor may be nominally complete, but will lack vitality unless it conveys the notion of activity ( ). Through metaphor Homer frequently speaks of lifeless things as living ( ); his poetry is thus distinguished through the effect of activity ( ). 57 Aristotle cites a number of Homeric metaphors in all of which there is appearance of actuality ( ), since the objects are represented as animate, such as the shameless stone or the eager spearpoint. He explains: Homer has attached these attributes by the employment of the proportional metaphor ( ); for as the stone is to Sisyphus, so is the shameless one to the one who is shamelessly treated. 58 The expression to place things before the eyes is itself metaphorical for the sensible character of metaphor. There are here two significant aspects worthy of note. Sensation is itself an activity, as Aristotle makes clear in De Anima III 2; hence an image or metaphor is all the more potent when it conveys an action ( ). Secondly, in line with Aristotles metaphysics of the categories, action is the most appropriate similitude to be expressed through metaphor. The substances of different genera cannot resemble one another in essence or nature. The only resemblance which

_________
57 58

See Rhet. III 10, 1411b21-11, 1411b33. Rhet. III 11, 1412a4-6. Translation Freese.

ARISTOTLE AND THE METAPHYSICS OF METAPHOR

169

may be affirmed between them is either the perfectly analogous similarity of their metaphysical principles or the imperfectly analogous resemblance of action. The similarity which metaphor conveys is not that of substance, but activity. Beings of different genera resemble one another not in what they are (essence or nature) but in what they doeach in accordance with its own nature and identity. For Aristotle, as already noted, metaphors should place an idea : before the eyes. The vast majority of everyday metaphors originate from sensible images, although most have lost their imagic character; they have become dead metaphors, no more than clichs. Happily we continue to create new metaphors, and there is delight in both inventing and recognizing these. Allow me to mention a few metaphoric images which surprised me recently, and which confirmed the validity of Aristotles remark that the vitality of metaphor is to place something before the eyes. I was struck by Hugh Kenners assertion that Language is a Trojan horse by which the universe gets into the mind, 59 and by Plutarchs suggestion that myth is the rainbow which reflects the sun of truth. 60 Hearing a woman describe her reaction to the murder of her father, overcome with the black lava of grief and hate, I recalled vivid images of carbonised bodies among the ruins of Pompeii. I was fascinated with Seamus Heaneys description of the intellectual condition of Boston in the 17th century: Nothing stirred. The future was a verb in hibernation. 61 Most recently, right on schedule and against my best hopes, I heard someone speak of a tsunami of information. Aristotles point is well illustrated: the power of metaphoric expression comes from its sensible, imagic character. The faculty of imagination, he states, is identical with that of sensation. 62 Image is defined in the Rhetoric as feeble sensation. 63 Aristotle declares that metaphorical expressions are always obscure. 64 This does not jeopardise its value but denotes its double character as clairobscur, projecting and diffusing its light, prism-like, although itself opaque. Cecil Day Lewis remarked: There are such things as unverifiable truths, and it is the unverifiable element in poetry which carries the conviction of truth. 65 (It is a strange characteristic of mystery that, while ob-

_________
59 60 61 62 63 64 65

H. Kenner, 1955, p. 117. De Iside et Osiride 358f-359a. Villanelle for an Anniversary, written for the 350th anniversary of Harvard. Insom. 459a15-16: . Translation Beare, amended. Rhet. I 11, 1370a28-29: Top. VI 2, 139b34-35: . C. Day Lewis, 1947, p. 35.

170

FRAN OROURKE

scure in itself, it has frequently a remarkable capacity to illuminate other things.) The distinction between the metaphorical and the proper ( ) use of words allows Aristotle to praise metaphor in poetry but scorn its use in philosophy. While obscurity ( ) has a place in poetry and metaphor lends an air of wondrous strangeness ( ), philosophy seeks clarity ( ). Analogy is one of its most valuable tools; from a scientific point of view metaphor is deviant, defective and wanting in definition. Aristotles sharpest criticism of Platonic participation in the Metaphysics was to dismiss it as a poetic word or empty metaphor. 66 In the Meteorology he illustrates the opposing values which metaphor has for philosophy and poetry: It is absurd to suppose that anything has been explained by calling the sea the sweat of the earth, as Empedocles does. Metaphors are poetical and so that expression of his may satisfy the requirements of a poem, but as to knowledge of nature it is unsatisfactory. 67 Empedocles metaphor provides a graphic image, but nothing of scientific value. Metaphor is for science a semantic hybrid it flourishes and blooms, but is itself infertile. Dealing in the Topics with the tactics of argument, Aristotle provides another reason for caution. By using metaphor the opponent may escape through sleight of argument. 68 Definition requires strict unity and coherence, while metaphor lives in the double entendre, a duality of denotation which may give rise to ambiguity. One may refute such an opponent, however, if one can turn his metaphoric meaning against him, on the ironic assumption that he has used words in their proper sense. 69 In rational discourse, however, one should seek clarity of definition and eschew equivocation: If we are to avoid arguing in metaphors, clearly we must also avoid defining in metaphors and defining metaphorical terms; otherwise we are bound to argue in metaphors. 70 One cannot reason syllogistically by metaphor; equivocation ensues, as does incongruity if we

_________
Metaph. 9, 991a20-22. Mete. II 3, 357a 24-28. Translation Webster amended. Top. IV 3, 123a33-37: You must also see whether your opponent has assigned as a genus a term used metaphorically, speaking, for example, of temperance as a harmony; for every genus is predicated of its species in its proper sense, but harmony is predicated of temperance not in its proper sense but metaphorically; for a harmony consists always of sounds. Translation Forster. 69 Top. VI 2, 139b35-36. 70 APo. II 13, 97b37-39. Translation Tredennick.
66 67 68

ARISTOTLE AND THE METAPHYSICS OF METAPHOR

171

confuse metaphor and literal description: Socrates has a sharp mind and a snub nose! Aristotles approach to metaphor is comprehensive and multifaceted. While various theories have emphasised one or other aspect of metaphor, Aristotles approach cannot be reduced to any in particular. Of the elements which he associates with metaphor we may note primarily, however, ornamentation, emotion, and cognition. Aristotle recognises the importance of metaphor as adorning language. It is essential to what he calls , the virtue of the word: The materials of metaphor must be beautiful to the ear, to the understanding, to the eye or some other physical sense. 71 Metaphor no doubt embellishes but cannot be reduced to ornament; these are explicitly distinguished by Aristotle.72 With regard to emotion, the states which Aristotle explicitly notes are wonder and the pleasure of knowledge. Metaphor is equally effective, it could be argued, with regard to such affective states as fear, horror or disgust. These too can simultaneously evoke the marvel of knowledge. Vital to metaphor is the contrast between the familiar and the strange, which is the hallmark of wonder. Metaphor is one of the most effective ways to give everyday speech an unfamiliar air. 73 Things which are remote are wonderful and what is wonderful is pleasant. 74 Metaphor is a continual reminder of the strangeness of things all around: the marvellous in the quotidian. Easy learning is naturally pleasant to all, and words mean something, so that all words which make us learn something are most pleasant. Now we do not know the meaning of strange words, and proper terms we know already. It is metaphor, therefore, that above all produces this effect. 75 Most discussion of metaphor considers it as an event occurring at the semantic level of the object. The effect on the speaking or listening subject, however, should not be overlooked. As well as the transfer of a name from its proper setting to a strange or inhabitual context, metaphor transports the speaker, listener or reader, beyond the confines of his present

_________
iambic verse, which most resembles spoken language, he states: Only those words are allowed which might be used in speech. These are the ordinary word, metaphor, and ornament ( ). Poet. 22, 1459a12-14. Translation Fyfe, amended. 73 Rhet. III 2, 1404b10-11. 74 Rhet. III 2, 1404b11-12: , . 75 Rhet. III 10, 1410b10-13. Translation Freese.
71 Rhet. III 2, 1405b17-19. Translation Roberts. 72 Poet. 21, 1457b2; 22, 1458a33. Referring to

172

FRAN OROURKE

experience to a new horizon. With its power of estrangement metaphor arrests our habitual relationship with the world. The miracle of metaphor is its power to evoke marvel and astonishment. According to Aristotle metaphor introduces the element of strangeness ( ); he has in mind the strangeness of expression, but beyond language it also serves to make things strange. Malebranches invitation comes to mind: I will not bring you into a strange land, but show perhaps that you are a stranger in your own country. 76 An effective metaphor can bring about a dramatic displacement in the Brechtian sense of Verfremdung. There ensues the surprise of recognition, the joy of discovery. This brings us to the cognitive function of metaphoralready implicit throughout the preceding discussion. Metaphor discerns similitude, discovers novel connections, establishes new resemblances, thus offers new insight; it deepens our understanding of what we know. It provides a cipher for the unknown. Aristotle remarks that even though there is sometimes no word for some of its terms, analogy loses none of its expressive power. 77 Metaphor too can give names to nameless things. 78 Most witty sayings, according to Aristotle, are derived from metaphor and beguile the listener in advance: expecting something else, his surprise is all the greater. His mind seems to say, according to Aristotle, How true, but I missed it. 79 Such discovery provides the pleasure of easy and rapid learning ( ... ). 80 Successful metaphors, as in the case of Homer, succeed in creating new learning and knowledge ( ). 81 Here Aristotle sees the difference between metaphor and simile. Simile does not captivate the listeners attention so powerfully as metaphor. It does not declare outright this is that

_________
76 N. Malebranche, 1965, p. 30: Non, je ne vous conduirai point dans une terre trangere: mais je vous apprendrai peut-tre que vous tes tranger vous-mme dans vtre propre pas. 77 Poet. 21, 1457b25-30. Aristotles example is the poets analogy between the casting forth of seed-corn, i.e., sowing, and the casting forth by the sun of its flame, for which there is no word. 78 Rhet. III 2, 1405a35-37: In using metaphors to give names to nameless things, we must draw them not from remote but from kindred and similar things, so that the kinship is clearly perceived as soon as the words are said (Translation Roberts). Clearly such metaphor does not give rise to obscurity: A metaphor in a way adds to our knowledge of what is indicated on account of the similarity, for those who use metaphors always do so on account of some similarity (Top. VI 2, 140a8-11, translation Forster). 79 Rhet. III 11, 1412a20-21: , . 80 Rhet. III 10, 1410b10, 21. 81 Rhet. III 10, 1410b14.

ARISTOTLE AND THE METAPHYSICS OF METAPHOR

173

( ), and thus jolt the mind to examine the strange connection between the objects. 82 The more cryptic quality of metaphor draws the listener to a closer examination of the similarity which he must discover for himself. As Aristotle notes, the difference between metaphor and simile is minimal. 83 Similes, if they are good, can also have the effect of brilliance. However, the unstated nature of the similarity in metaphor forces the listener or reader to invent it for himself; it has thus an added element of surprise and discovery. 84 The impact of metaphor is to say that this is that. The mind is aroused by a Socratic sting that shocks the mind to new recognition. Simile, moreover, is less pleasant because it is longer; metaphor is elegant and clever ( ) because it delivers rapid instruction. In its cognitive function, we can discern in Aristotle an aspect of metaphor which has rightly been emphasized in recent decades, namely its interactive character, heralded by I. A. Richards and championed by Max Black. Richards speaks of the tension between the two contexts which are juxtaposed in metaphor. These he denotes with the terms tenor and vehicle. 85 (Ernan McMullin suggests the more obvious terms target and illuminator). 86 Metaphor involves the conjugation of ideas or images from distinct domains of experience. In Dr Johnsons celebrated phrase, metaphor gives us two ideas for one. 87 (James Joyce in Ulysses offers what is itself an impressive metaphor to describe analogy: Though they didnt see eye to eye in everything, a certain analogy there somehow was, as if both their minds were travelling, so to speak, in the one train of thought.) 88 But as well as juxtaposition, there must be an element of opposition or antithesis. Aristotle notes that metaphors should be drawn between kindred objects, but emphasizes that the kinship should not be too obvious: otherwise there is no need for metaphor. The virtue of metaphor is precisely to discover likeness in unlikeness. 89 What is either too obvious

_________
Rhet. III 10, 1410b19. Rhet. III 4, 1406b20: . Rhet. III 10, 1410b15-21. I. A. Richards, 1936, p. 96. E. McMullin, 1981, p. 39. The full phrase is worth citing: And, Sir, as to metaphorical expression, that is a great excellence in style, when it is used with propriety, for it gives you two ideas in one; conveys the meaning more luminously, and generally with a perception of delight. Boswell, 1934, p. 174. 88 J. Joyce, 2001, p. 536. 89 Rhet. III 11, 1412a10-12.
82 83 84 85 86 87

174

FRAN OROURKE

or obscure conveys nothing new and is without interest, whereas a successful metaphor provides new learning and insight. 90 In metaphor the speaker assumes a certain conscious ambivalence. Metaphor asserts one thing, individual and unique, to be what it is not. The speaker is aware of this seeming contradiction, but is saved from absurdity by a concomitant awareness that it is not really asserted as such. There is a doublethink, a parallelism or duplicity of intention. When Homer refers to Achilles as a lion, he is not really asserting that he is a member of the species Panthera leo, but that in a certain aspect his actions resemble those of a lion. With poetic license metaphor implicitly exercises an existential with respect to the copula; it declares both that it is and it is not, perhaps more precisely: it is this, but not really; it affirms a substance, but intends an accident. It asserts identity, but includes otherness. Aristotles distinction between the normal and the strange use of a word is echoed by the interactive theory, which emphasises the tension between the two usages as a basic constituent of metaphor. To state that one thing is another offends the most basic principle of all discourse, the principle of non-contradiction. Of course Achilles is not a lion: should we not mean what we say? The tension of this doublethink forces the mind beyond itself. There is a fruitful tension at the heart of metaphor, which impels the mind to new discovery. Analogy is the intuitive leap by which mind connects the known with unknown experience. It is the spark that ignites the mind to light up similarities below the surface; it is a lamp borrowed from one domain to illumine the recesses of another. It brings objects from distinct arenas into a reflective relationship, that one may clarify the other. The mind shuttles between one and the other term, and back again, in a quick movement of thought which at once affirms identity and difference, thus extending our knowledge of the given. What are the metaphysical requirements of metaphor? What does the activity of metaphor reveal to us about man, in terms of Aristotles philosophy? Metaphor brings out in a unique manner the metaphysical nature of human knowledge. By metaphysical I simply mean the ability of human cognition to pass beyond the sense experience of an individual object to grasp it in its universal aspect, to view an individualhowever insignificantsub specie totalitatis. It brings an increase of metaphysical awareness, a heightened pitch of abstractive and intuitive activity: intuitive, because it grasps a concrete feature of the object, abstractive because it sets it in relation with a reality from a distinct, perhaps distant, domain. Meta-

_________
90

Rhet. III 10, 1410b12-13.

ARISTOTLE AND THE METAPHYSICS OF METAPHOR

175

phor is the embodiment in miniature of mans metaphysical knowledge, and illustrates in a unique manner his ability to surpass the physical confines of immediate experience. He may thus view any object of experience, sensible or intellectual, within a wider context according to whatever similarity he perceives. He can associate one individual with any other, even a thing unknown. His arena of reflection is ultimately the unlimited horizon of the totality of being. Summarizing his treatise on psychology, Aristotle states that the soul is in a sense all things. 91 This is the openness requisite for the spontaneous play of metaphor; the subtle tendrils of mind and imagination recognize no obstacle in their glimpse of similarity in the most unlikely places. In agreement with Aristotles view of things, metaphor indicates a duality in human nature between body and psyche, sense and intellect; but the ability also to surpass this division. It reveals a more profound unity in human nature. Just as the diversity of sense perceptions is unified by the power of the common sense, 92 so also the acts of cognition which operate in tandem to produce metaphor demand a single subject who is aware of identity in difference. The dual optic must be brought into single focus. Only a common element can bind what is diverse. Moreover, the fact that in countless metaphors the physical and psychic mirror one another indicates the underlying unity of reality itself. Mans citizenship of two worlds, material and mental, is already inscribed in the very nature of language: a material medium which carries a metaphysical meaning. Language encapsulates the human capacity and impulse for self-transcendence. Using sensible symbols man surpasses the confines of the material world. Frege has put it well: Signs have the same importance for thought as the discovery of using the wind to sail against the wind has for seafaring. 93 Words are somehow a summation of mans sensible and intellectual unity. Language is laden both with the inner tension of sense and intellect and the further struggle to express, beyond cognition, a reality which in principle it can never fully disclose. In metaphor the human impulse for transcendence achieves one of its deepest, most metaphysical, moments. More than any other mental act, analogy, including metaphor, reveals the ability to rise beyond a single individual and establish its relationship with other beings.

_________
An. III 8, 431b21: . An. III 2, 426b17-23. G. Frege, 1964, p. 107: Die Zeichen sind fr das Denken von derselben Bedeutung wie fr die Schifffahrt die Erfindung, den Wind zu gebrauchen, um gegen den Wind zu segeln.
91 92 93

176

FRAN OROURKE

The poet Cecil Day Lewis has expressed much of what I wish to conveywhich I believe to be in harmony with the fundamentals of Aristotle: Relationship being in the very nature of metaphor, if we believe that the universe is a body wherein all men and all things are members one of another, we must allow metaphor to give a partial intuition of the whole world. Every poetic image, by clearly revealing a tiny portion of this body, suggests its infinite extension poetrys truth comes from the perception of a unity underlying and relating all phenomena, poetrys task is the perpetual discovery, through its imaging, metaphor-making faculty, of new relationships within this pattern, and the rediscovery and renovation of old ones. The poetic image is the human mind claiming kinship with everything that lives or has lived, and making good its claim. 94 In keeping with its importance in the Poetics and Rhetoric, metaphor exhibits a pervasive power for creative insight; it lives in the tension between unity and diversity both in human nature and in the universe. It is moreover a token both for the simplicity of human nature which acts through a diversity of levels, and for the unity of reality throughout the multiplicity of beings. All of these elements are present though not explicit in Aristotle. They are, I suggest, the implicit background to his theory of metaphor. In the absence of genuine metaphysical analogy, which binds entities through a proper likeness and similitude, there would be no real foundation for transferred or metaphoric resemblance. One Shakespearean critic has expressed as follows the profound implications of metaphor: I believe that analogylikeness between dissimilar things, which is the fact underlying the possibility and reality of metaphorholds within itself the very secret of the universe. 95 This is close to the passages from the Metaphysics cited earlier to illustrate Aristotles vision of a unified cosmos. Metaphor is vital to daily language; it attains its fullest expression in poetic creation. Analogy, on the other hand, finds its fullest application in metaphysics. The poet suggests in metaphor what the philosopher asserts through analogy. Metaphor depends upon imagery; analogy operates by means of concepts. Each engages and activates in its own way the universal character of human intentionality: the unique relationship which human has towards the totality of being. Mans nature is sensible and intellectual. His knowledge is a unity of both, beginning with and relying upon the senses. His ability to surpass the physical is attested to primarily by the intellectual power of abstraction, which is the pulse and drive of

_________
94 95

C. Day Lewis, 1947, pp. 29, 34, 35. C. Spurgeon, 1936, p. 7.

ARISTOTLE AND THE METAPHYSICS OF METAPHOR

177

philosophy, and heightened by the associative power of imagination, reaching its highest intensity in the act of creative metaphor. Metaphor always retains an element of paradox, whether viewed as ambiguity or surplus of meaning; it uniquely blends the luminous with the obscure. It cannot enter as such into syllogistic reasoning; it is not a tool of philosophy, but a profound phenomenon which summons philosophic reflection. The process of metaphor is highly revealing of human experience and expression. It discloses a relational similarity between diverse contexts: a resemblance the significance of which is not merely rhetorical or ornamental, but essentially metaphysical. The ultimate philosophical value of metaphor, therefore, from an Aristotelian perspective, is not its argumentative role but, I suggest, its power to disclose the relational solidarity of diverse substances; this in turn calls for philosophic explanation. Aristotle does not himself offer a comprehensive explanation in these terms, but provides the concepts and principles which are required. According to Henri Bergson, if we remove from Aristotles philosophy everything derived from poetry, religion and social life, as well as from a somewhat rudimentary physics and biology, we are left with the grand framework of a metaphysics which, he believes, is the natural metaphysics of the human intellect. 96 It seems to me that metaphor, which so profoundly characterizes our intellectual cognition, as it cooperates with sense and imagination, is best explained by such a natural metaphysics. 97
UNIVERSITY COLLEGE DUBLIN

_________
H. Bergson, 1928, p. 344. I wish to thank Gary Gurtler, S.J., John Manoussakis, and Michael Smith for their hospitality and kindness during my visit to Boston. I am indebted to Patrick Sammon, Gerard Casey, and Andrew Smith for many helpful suggestions. I am grateful to the anonymous referee for the Proceedings, and especially to my commentator Ioanna PatsiotiTsacpounidi.
96 97

COMMENTARY ON OROURKE
IOANNA PATSIOTI-TSACPOUNIDI

It is a challenging task to work out the metaphysics of the Aristotelian concept of metaphor, especially if we take into consideration Umberto Ecos statement that little has been added to its understanding since its brief treatment by Aristotle. In fact, this task may prove to be even more difficult if we also consider the view of many people that the Aristotelian metaphora is not exactly equivalent to what we mean by metaphor today. 1 And yet Professor Fran ORourke seems to have accomplished this task to a fair extent, and has managed to exhibit the significance of metaphor for human knowledge. My comments will be categorized under three main areas: (a) the philosophical significance of metaphor, (b) metaphor and analogy, and (c) man and metaphor. I. The Philosophical Significance of Metaphor To start with, Prof. ORourke is right when he says that the claim that metaphor is for Aristotle merely literary ornamentation is wrong. It is evident enough to any Aristotelian scholar that the Stagirite offers an interesting semiotic theory, which involves the interaction of cognitive and emotive contexts. From the definition of metaphor in Aristotles Poetics, 1457b1ff, where we receive a detailed account of what a noun must be, with particular emphasis on the ways it can be used metaphorically, to the elaborate analysis of the types of metaphors we encounter in the third book of his Rhetoric, with emphasis being put on its connection to analogy, we notice that Aristotle does not simply use this as a linguistic weapon in the hands of a persuasion-seeking rhetorician. If, as Prof. ORourke says, Aristotle uses metaphor in order to bring about in a unique manner the metaphysical nature of human knowledge, that is, to show the ability of human cognition to pass beyond the sense experience of an individual object and to grasp it in its universal aspect, then he is trying to establish its philosophical significance and not to depreciate its value. It would not be correct either to take its definition in the Poetics,

_________
Lloyd, G.E.R., Aristotelian Explorations (Cambridge: Cambridge University Press, 1996), 205.
1

COMMENTARY ON OROURKE

179

onomatos allotriou epiphora, as Delphic and very general, as Lloyd indicates, 2 since what Aristotle here means is the transfer of a strange term and its application in a different context. Of course, Lloyd comes to acknowledge 3 that Aristotle offers later analyses of metaphor, simile and comparison. His point of departure is not the use of comparison as such but a basic contrast between terms used strictly (kyrios) or appropriately (oikeios) on the one hand, and terms not so used on the other. Lloyd also observes the negative attitude towards metaphora in most books of Aristotles Organon, as its use would undermine the validity of any chain of reasoning. 4 For example, in the Topics, it is stated that every metaphorical expression in a definition is obscure (asaphes). Nevertheless, in Rhetoric, 1405 a 8 ff., Aristotle says that the use of metaphora produces clarity (to saphes), pleasure (to hdu) and a foreign element (to xenikon), which would attribute to metaphora a more philosophical character. And if the assumption is that in Rhetoric, 10, Aristotle praises metaphora for its effectiveness in argumentation, and thats as far as he can accommodate it, then we simply ignore the quality Aristotle wants to attribute to it, and even talk about an apparent contradiction in Aristotles treatises. The possible confusion or contradiction in Aristotles thought regarding the value he attributes to metaphora is a point perhaps that Prof. ORourke could have seen more closely. Moreover, it would be very useful in his analysis if Prof. ORourke insisted more on the Aristotelian definition of metaphor as stated above. To explain metaphor, Aristotle creates a metaphor that he borrows from the domain of movement. The concept of phora indicates the notion of movement or even change of movement with respect to location. As for epiphora, it takes the meaning of bringing to or upon, carrying towards, as well as that of conclusion, what is derived from. 5 In general, the preposition of epi in Greek attaches a further emphasis on the word that takes it as its prefix. And most probably this is the case with epiphora here. The word allotrios means alien, unusual. Thus, the whole meaning of this phrase would be that an alien or unusual name is carried toward another location, that is, metaphor is the transposition of an alien name to somewhere else, to another name with which one would be more familiar.

_________
Ibid., p.206. Ibid., pp.206-207. Ibid., p.208. Liddell H.G., & Scott R., A Greek-English Lexicon, vol.1, (Oxford: At the Clarendon Press), 1925, p.671.
2 3 4 5

180

IOANNA PATSIOTI-TSACPOUNIDI

According to Paul Ricoeur, 6 the Aristotelian idea of allotrios tends to assimilate three distinct ideas: the idea of deviation from ordinary usage; the idea of borrowing from an original domain; and the idea of substitution for an absent but available ordinary word. Ricoeur considers the idea of deviation as having a rather negative sense as opposed to the idea of borrowing that implies a more positive view. 7 As for the idea of substitution, he indicates, if the metaphorical term is really a substituted term, it carries no new information, since the absent term can be brought back in; and if there is no information conveyed, then metaphor has only a decorative value. 8 It is interesting to include here the three interpretative hypotheses, Ricoeur proposes. 9 First, in all metaphor one might consider not only the name that is displaced but the pair of terms between which the transposition operates (from genus to species, from species to genus, from species to species, from the second to the fourth term of a proportional relationship). And this logical deviation, for Aristotle, produces meaning. Second, what we notice in a metaphor is also the idea of categorical transgression, that is, a deviation in relation to a preexisting logical order, a kind of disordering in a scheme of classification. And third, there is an identification between the process that disturbs a certain logical order and that from which all classification proceeds. In other words, the functioning of language operates within an already constituted order. If we accept these three interpretations by Ricoeur, then we notice that even though metaphor involves an apparent deviation from the existing logical order, the whole process takes place within the system of language and this transposition from unusual to ordinary names can be achieved because of the coordination of relationships within language itself. What follows, is that there is no real opposition between ordinary and unusual or strange names, between proper and figurative speech, and what Aristotle is trying to achieve here is to show that metaphor cannot be merely a decorative tool but one of the main semantic means a person could use. This point about the metaphysics of metaphor as a linguistic tool might be examined more closely by Professor ORourke. Of course, Prof. ORourke does indicate that metaphora is the proportional but imperfect transfer of a perfection or activity from its primary to

_________
6 Ricoeur, P., Between Rhetoric and Poetics, in Essays on Aristotles Rhetoric, edited by Amelie Oksenberg Rorty, (Berkeley: University of California Press, 1996) 324-384 (particular reference p.332). 7 Ibid., pp. 331-2. 8 Ibid., pp. 332-3. 9 Ibid., pp. 334-5.

COMMENTARY ON OROURKE

181

a secondary subject. In metaphor, a name which belongs intrinsically to one being is transferred to another, not by virtue of what is intrinsically in itself, but through a relation of proportional similarity in some secondary or accidental sense, namely a resemblance of action. It is here in fact that one could attempt to provide a solution to the question as to whether Aristotle is ambivalent about the reliability of metaphora. As Prof. ORourke rightly observes, metaphor affirms both that something is and it is not; it affirms a substance, but intends an accident. This by itself shows the twofold purpose that Aristotle wants to serve by means of metaphora in the domain of action: (i) it is asaphes, xenikon and hdu insofar as it is used to enhance the persuasive techniques of the rhetorical argument, but it is also saphes, since man uses it to understand better abstract objects and associate himself to the external reality, and (ii) through its vividness it exemplifies the connection between theory and action. Thus, even though Aristotle would be more in line with his main philosophical position that due to its ambiguity metaphora should be avoided either in demonstration or dialectic, he seems to lean on metaphora for making his points clearer in the domain of action. It would be relevant perhaps, if Prof. ORourke added here the moderate character Aristotle attributes to it. A metaphor should not be an unusual or unknown word, as it would be difficult for someone to understand it, nor too common, as it might have no impact on the hearer. 10 This would prove the philosophical significance Aristotle attributes to it more persuasively. II. Metaphor and Analogy Nevertheless, what is more important in Prof. ORourkes paper is the emphasis he puts on the connection between metaphor and analogy, and the power of the latter to achieve universal comprehension. As he rightly indicates, analogy is the agreement of correspondent relations which are diversely realized in different domains. Thus it provides the widest framework for universal unity among diverse substances. It refers to the unity of relations within and among a diversity of beings. In fact, Prof. ORourke manages to connect metaphor with analogy in terms of the following ways: (i) There is emphasis on the aspects of analogy, such as sameness and likeness, which are used as forms of unity, as well as the use of analogy as

_________
10

Rhetoric, 10, 1410 b31 34.

182

IOANNA PATSIOTI-TSACPOUNIDI

a method of scientific order. The author cites Topics I and Metaphysics V, which refer to Aristotles discussion of sameness with emphasis on his zoological treatises, in which an analogy is used in order to introduce order among disparate species on the basis of similarity of function or operation. Along with these references, we would have thought that it would be essential to the connection between metaphor and analogy, if there was also further reference to his ethical treatises, where the use of analogy makes the universal comprehension it achieves even more evident. For example, one could also refer to the famous organic theory of the state, in which Aristotle draws an analogy between an organism and its individual organs and a state and its individuals. 11 In the same analogous way, in his discussion of the proper education Aristotle states that every human being needs different treatment in accordance with his own nature, in the same way that a patients past medical record and general physique determines the treatment prescribed by a doctor once a certain disease is diagnosed. 12 It would also be relevant here if along with the examples of the katanalogian metaphora (analogical metaphor) we added some more that enhance the notions of likeness and sameness. After all, in his Metaphysics 4, 1070b10-21, Aristotle states that all things can be said to have the same elements by analogy. For example, in the Eudemian Ethics, 13 he states that all kinds of friendship relate to one another in terms of a pros hen kai mian tina philian (towards one focal meaning of the concept of friendship). According to Aristotles account, the various meanings of philia relate to one another in that understanding of the fact that they are inferior types involves reference to the primary type, in the sense that understanding of the inferior types of philia contribute to the understanding of the primary one, and the other way round. 14 It is also worth noting that the philosopher draws an analogical metaphor between the various types of philia with the word iatrikon (medical) that applies to the soul, the body, or the instrument used in a treatment. We use the word medical whether we refer to the mental state of the doctor and the knowledge he has, or the body of the patient he is set to treat, or the medical instrument

_________
11 Cf. Miller, F.D., Jr., Nature, Justice and Rights in Aristotles Politics, (Oxford: Clarendon Press, 1995) 47-50. 12 EN, K14, 1180b7-8. 13 EE, H2 1236a18, 1236b19 26. 14 Given that all the other kinds of philia are affirmed or denied in relation with the primary type (cf. EE, H2, 1236a20-23, 1237b810).

COMMENTARY ON OROURKE

183

he uses as part of the treatment, in the same way we call friendship any form of philia. 15 (ii) There is emphasis on the use of analogy in metaphysics, with particular reference to the distinction between act and potency, matter, form and privation, and the reciprocal relationship of causes. In all cases, the use of analogical metaphor as well as that of proportionality provides the main methodological approach of the Aristotelian analysis. This extensive use of analogical metaphor, as Prof. ORourke rightly indicates, confirms the philosophers attempt to present the cosmos as a coherent whole that aims at some end (as stated in Metaphysics, 10, 1075a ff.). He also observes that in metaphor what is affirmed is not a proper analogy, but an imperfectly analogical resemblance: the action belongs perfectly only to one substance, and is transferred to another because of some perceived but imperfect likeness, so he takes metaphor to be an imperfect transfer of an activity from its primary to a secondary subject, that is a resemblance of action. Here, however, we would raise the following objection: even if we take the similarity expressed through metaphor to be one of action and not of substance, it does not reduce its value as a method, since the study of the being qua being cannot be a demonstrative process but a metaphorical one. The philosophical value of metaphor is to discover the being, and it does this by showing to us every possible and endless combination of similar things in the context of dissimilarity. 16 And this cannot have an end point, as if it were a syllogistic process. Thats why Aristotle uses metaphor, in order to draw all the possible relations between similar and dissimilar things. Both definition and metaphor are based on the similarities of things, and try to reflect reality from a different perspective. As the philosopher says: metaphor places things before the eyes (Rhetoric, 11, 1411a22), it makes things closer to perception, hence it helps us structure our thoughts and achieve a clearer understanding of reality. As Moran rightly observes, what Aristotle means by this phrase is that the mind of the hearer is provoked, set into motion, and engaged imaginatively with metaphor, assuming that activity exists not only on the side of what is depicted, but what is depicted is figured as a living thing demanding some kind or response from the audience, some mental activity of its own.17

_________
EE, H2, 1236a1620. Murchland, B., O Aristotelis, i metaphora kai to ergon tis philosophias, in Andriopoulos edition, Aristoteles, Athens, 1996, pp.282-294 (p.291). 17 Moran, R., Artifice, and Persuasion: The Work of Metaphor in the Rhetoric, in A.O. Rorty, 1995, pp. 385-398 (particular reference to pp. 395-6).
15 16

184

IOANNA PATSIOTI-TSACPOUNIDI

III. Man and Metaphor The most significant emphasis Prof. ORourke places in his paper is on our understanding of the way metaphor enhances the human intellectual capacities, and in particular, the connection between the psychic activity conveyed with terms drawn from the physical activity. By means of sensible symbols metaphor helps man surpass the limitations of the material world. In fact, the perceptive element involved in metaphor should be further emphasized. Given that metaphor is the recognition of similarity in dissimilar things through the process of perception, it comes to associate the physical with the psychic part of the human existence. Aristotle draws from the world of the senses in order to illustrate the psychic activities, thus providing a unified account of the human nature. It is even more important that Prof. ORourke identifies the metaphysical element involved with analogical metaphor with an intelligent as well as an eloquent use of language, which manages to disclose what can never be fully known or comprehended. At this point, however, we would expect further explication of the epistemological significance of metaphor. If it manages to connect the dissimilar things with each other, abstractions with concrete situations, and imagination with reality, in the minds of people, we would like to know how this process takes place. What level of knowledge does this process amount to and why is it conducive to a persons selftranscendence, as Prof. ORourke indicates? Is there any spiritual element involved with metaphor? And which is the epistemological strength of it? If metaphor helps humans achieve a universal comprehension of reality, i.e., the fact that diversity is nothing else but various parts of a coherent whole, then despite its absence of clarity, it must involve an element of objectivity, or even better, of intersubjectivity. It must help the human mind understand how each situation fits into this unified whole. After all, it consists in a transposition of an unusual name to an ordinary one within the operative system of language, and this deviation from the structured logical order, being a legitimate one, must be closely understood by the audience to which it is presented. In conclusion, Prof. ORourke has presented an interesting analysis of the metaphysics of metaphor in Aristotle and has explored its dynamics as well as its connection with the human intellect to a fair extent. All in all, his analysis manages to exhibit the philosophical significance that Aristotle implicitly attaches to metaphora in his own explorative manner. What we would like to see more closely is the interpretation of the main definition that Aristotle provides in his Rhetoric, the way the Aristotelian metaphora connects to the modern concept, as well as the epistemologi-

COMMENTARY ON OROURKE

185

cal strength of it, especially if we take into consideration the fact that Aristotle is not ambivalent at all about the philosophical significance of this powerful means of conveying meaning, which promotes communication among moral agents.
DEREE COLLEGE

OROURKE/PATSIOTI BIBLIOGRAPHY Aristotle

Apostle, H. G. (tr.) 1979. Aristotles Metaphysics. Grinnell. Barnes, J. (ed.) 1991. The Complete Works of Aristotle. The Revised Oxford Translation. Princeton. Beare, J. I. (tr.) Aristotle. On Dreams. In Barnes (ed.): 729-735. Butcher. S. H. (tr.) 1951. Aristotles Theory of Poetry and Fine Art. New York. Bywater, I. (tr.) 1909. Aristotles Poetics. Oxford. Forster, E. S. (tr.) 1989. Aristotles Topics. Loeb Classical Library. London. Freese, J. H. (tr.) 1926, Aristotle. The Art of Rhetoric. Loeb Classical Library. Cambridge. Fyfe W. H. (tr.) 1960. Aristotles Poetics. Loeb Classical Library. Cambridge. Halliwell, S. (tr.) 1986. Aristotles Poetics. London. Halliwell, S. (tr.) 1995. Aristotles Poetics. Loeb Classical Library. London. Heath, M. (tr.) 1996, Aristotles Poetics. Harmondsworth. Hutton, J. (tr.) 1982. Aristotles Poetics. New York. Jowett, B. (tr.) 1921. Aristotles Politica. Oxford. Kassel, R. (ed.) 1976. Aristotelis Ars Rhetorica. Berlin. Kassel, R., (ed.) 1965. Aristotelis de Arte Poetica Liber. Oxford. Kennedy, G. A. (tr./ed.) 1991. Aristotle. On Rhetoric. New York. Pickard-Cambridge, W. A. (tr.) 1950. Aristotle. Topica. Oxford. Rapp, C. (tr./ed.) 2002. Aristoteles. Rhetorik (2 Vols.) Berlin. Roberts, W. R. (tr.) 1971. Aristotles Rhetorica. Oxford. Ross, W. D. (tr.) 1972. Aristotles Metaphysica. Oxford. Thompson, D. W. (tr.) 1949. Aristotles Historia Animalium. Oxford. Tredennick, H. (tr.) 1962. Aristotle. Metaphysics. Loeb Classical Library. Cambridge. Tredennick, H. (tr.) 1989. Aristotles Posterior Analytics. Loeb Classical Library. Cambridge. Webster, E. W. (tr.) 1951. Aristotles Meteorologica. Oxford.

Secondary Sources
Amsler, M. (ed.) 1987. Creativity and the Imagination. Newark. Bergson H., 1928. Creative Evolution. London. Black, M. 1962. Models and Metaphors. Ithaca. Black, M. 1977. More about metaphor. Dialectica 31: 431-457. Black, M. 1980. How Metaphors Work: A Reply to Donald Davidson. In Sacks, S. (ed.): 181-192. Boswell, J. 1934. Life of Johnson. Hill, G. B. and Powell, L. F. (eds.) Vol. 3, Oxford. Boyd, R. 1993. Metaphor and Theory Change: What is Metaphor a Metaphor for? In Ortony (ed.): 481-532. Boyle, R. R. 1954. The Nature of Metaphor. The Modern Schoolman 31: 257-280. Boyle, R. R. 1957. The Nature of Metaphor: Further Considerations. The Modern Schoolman 34: 283-298. Bremer, D. 1980. Aristoteles, Empedokles und die Erkenntnisleistung der Metapher. Poetica 12: 350-376. Brooke-Rose, C. 1958. A Grammar of Metaphor. London. Brown, S. J. 1927. The World of Imagery: Metaphor and Kindred Imagery. London. Burrell, David. 1973. Analogy and Philosophical Language. New Haven.

OROURKE/PATSIOTI BIBLIOGRAPHY

187

Bustos, E. 1992. La teora aristotlica sobre la metfora. In Snchez Meca, D. and Domnguez Caparrs, J. (eds.): 17-21. Cooper, D. E. 1986. Metaphor. Oxford. Coorebyter, V. 1994. Rhtoriques de la science. Paris. Cope, E. M. and Sandys, J. E. 1877. The Rhetoric of Aristotle (3 vols.). Cambridge. Croce, B. 1964. The Philosophy of Giambattista Vico. Collingwood, R. G. (tr.) New York. Danesi, M.1993. Vico, Metaphor, and the Origin of Language. Bloomington. Dasenbrock, R. W. 1993. (ed.) Literary Theory After Davidson. University Park, PA. Davidson, D. 1984. Inquiries into Truth and Interpretation. Oxford. Davis, M. 1992. Aristotles Poetics: The Poetry of Philosophy. Lanham. Day Lewis, C. 1947. The Poetic Image. London. De Rijk, L. M. 2002. Aristotle: Semantics and Ontology. Leiden. Derrida, J. 1982. Margins of Philosophy. Chicago. de Vio, T. (Cardinal Cajetan). 1953. The Analogy of Names. Bushinski, E. A. (tr.) Pittsburgh. de Vio, T. (Cardinal Cajetan). 1952. Scripta Philosophica (De Nominum Analogia, De Conceptu Entis). Rome. Eco, U. 1984. Semiotics and the Philosophy of Language. London. Eco, U., Niklas, U. and Edeline, F. 1986. Metaphor. In Sebeok, T. A. (ed.): 534-549. Eliot, G. 1980. The Mill on the Floss. Haight, G. S. (ed.) Oxford. Embler, W. 1966. Metaphor and Meaning. DeLand, FL. Emerson, R. W. 1950. Selected Writings, New York. Erickson, K. V. 1974. Aristotle: The Classical Heritage of Rhetoric. Metuchen, N. J. Fahnestock, J. 2000. Aristotle and Theories of Figuration. In Gross and Walzer (eds.): 166-184. Fraser, B. 1993. The Interpretation of Novel Metaphors. In Ortony (ed.): 329-341. Frege, G. 1964. Begriffsschrift und andere Aufstze. Hildesheim. Furley, D. J., & Nehamas, A. 1994. Aristotles Rhetoric: Philosophical Essays. Princeton. Gaipa, A. and Scholes, R. 1993. On the Very Idea of a Literal Meaning. In Dasenbrock, R. W. (ed.): 160-179. Gerhart, M. and Russell, A. M. 1984. Metaphoric Process: The Creation of Scientific and Religious Understanding. Fort Worth. Gordon, P. 1990. The Enigma of Aristotelian Metaphor: A Deconstructive Analysis. Metaphor and Symbolic Activity 5: 83-90. Gross, A. G. and Walzer, A. E. (eds.) 2000. Rereading Aristotles Rhetoric. Carbondale. Gumpel, L. 1984. Metaphor Reexamined: A Non-Aristotelian Perspective. Bloomington. Hale, B. and Wright, C. (eds.) 1997. A Companion to the Philosophy of Language. Oxford. Halliwell, S. 1987. The Poetics of Aristotle. Chapel Hill. Harries, K. 1980. Metaphor and Transcendence. In Sacks, S. (ed.): 71-88. Harries, K. 1980. The Many Uses of Metaphor. In Sacks, S. (ed.): 165-172. Hausman, C. R. 1989. Metaphor and Art. Cambridge. Haverkamp, A. (ed.) 1983. Theorie der Metapher. Darmstadt. Hawkes, T. 1972. Metaphor. London. Heidegger, M. 1976. Wegmarken. Frankfurt am Main. Herodotus, 1960 [History], Godley, A. D. (tr.) Loeb Classical Library, Cambridge.

188

OROURKE/PATSIOTI BIBLIOGRAPHY

Hesse, M. 1965. Aristotles Logic of Analogy. The Philosophical Quarterly 15: 328340. Hesse, M. B. 1970. Models and Analogies in Science. Notre Dame. Hesse, M. B. 1988. The Cognitive Claims of Metaphor. The Journal of Speculative Philosophy 2: 1-16. Johnson, M. 1985. Philosophical Perspectives on Metaphor. Minneapolis. Jordan, W. J. 1974. Aristotles Concept of Metaphor in Rhetoric. In Erickson, K. V. (ed.): 235-250. Joyce, J. 2001. Ulysses. London. Kenner, H. 1955. Dublins Joyce. London. Kirby, J. T. 1997, Aristotle on Metaphor. American Journal of Philology 118: 517554. Kittay, E. F. 1987. Metaphor: Its Cognitive Force and Linguistic Structure. Oxford. Knights, L. C. and Cottle, B. (eds.) 1960. Metaphor and Symbol. London. Kuhn, T. S. 1993. Metaphor in Science. In Ortony (ed.): 533-542. Kyrkos, B. A. 1972. Die Dichtung als Wissensproblem bei Aristoteles. Athen. Lakoff, G. 1993. The Contemporary Theory of Metaphor. In Ortony (ed.): 202-251. Lakoff, G. and Turner, M. 1980. Metaphors We Live By. Chicago. Laks, A. 1992. Zeitgewinn. Bemerkungen zum Unterschied zwischen Metapher und Vergleich in Aristoteles Rhetorik. In Rudolph, E. and Wismann, H. (eds.): 11-19. Laks, A. 1994, Substitution et connaissance: une interprtation unitaire (ou presque) de la thorie aristotlicienne de la mtaphore. In Furley and Nehamas (eds.): 283305. Lallot, J. 1988. . Le fonctionnement smiotique de la mtaphore selon Aristote. Recherches sur la philosophie et le langage 9: 47-58. Levin, S. R. 1979. Standard Approaches to Metaphor and a Proposal for Literary Metaphor. In Ortony, A. (ed.): 124-135. Levin, S. R. 1982, Aristotles Theory of Metaphor. Philosophy and Rhetoric 15: 2446. Levin, S. R. 1988. Metaphoric Worlds. New Haven. Levin, S. R. 1993. Language, Concepts, and Worlds: Three Domains of Metaphor. In Ortony, A. (ed.): 112-123. Lloyd, G. E. R. 1966, Polarity and Analogy. Cambridge. Lloyd, G. E. R. 1989, The Revolutions of Wisdom. Berkeley. Lloyd, G. E. R. 1991, Methods and Problems in Greek Science. Cambridge. Lloyd, G. 1966. Aristotelian Explorations. Cambridge. Lyttkens, H. 1953. The Analogy Between God and the World. Uppsala. MacIntyre, A. 1951. Analogy in Metaphysics. The Downside Review 69: 45-61. Maj, B. 1987. Elementi di metaforologia aristotelica. Ferrara. Malebranche, N. 1965. Oeuvres completes XII, Robinet A. (ed.) Paris. McCall, M. H. 1969. Ancient Rhetorical Theories of Simile and Comparison. Cambridge, Mass. McMullin, E. 1981. The Motive for Metaphor. Proceedings of the American Catholic Philosophical Association 55: 27-39. Marcos, A. The Tension Between Aristotles Theories and Uses of Metaphor. Studies in History and Philosophy of Science 28: 123-139. Miall, D. S. 1982. Metaphor: Problems and Perspectives. Brighton. Mio, J. S. and Katz, A. N. (eds.) 1996. Metaphor: Implications and Applications. Mahwah, NJ.

OROURKE/PATSIOTI BIBLIOGRAPHY

189

Moran, R. 1996. Artifice and Persuasion: The Work of Metaphor in the Rhetoric. In Rorty, A. O. (ed.): 385-398. Moran, R. 1989. Seeing and Believing: Metaphor, Image, and Force. Critical Inquiry 16: 87-112. Moran, R. 1999. Metaphor. In Hale, B. and Wright, C. (eds.): 248-268. Most, G. W. Seeming and Being: Sign and Metaphor in Aristotle. 1987. In Amsler, M. (ed.): 11-33. Murry, J. M. 1931. Countries of the Mind, London. Nightingale, A. W. 2004. Spectacles of Truth in Classical Greek Philosophy. Cambridge. Nth, W. 1985. Semiotic Aspects of Metaphor. In Paprott, W. and Dirven, R. 1985. (eds.): 1-16. Ortony, A. 1979. (ed.) Metaphor and Thought. Cambridge. Ortony, A. 1993. (ed.) Metaphor and Thought (Second Edition). Cambridge. Ortony, A. 1979. Metaphor: A Multidimensional Problem. In Ortony (ed.): 1-16. Ortony, A. 1993. Metaphor, Language, and Thought. In Ortony (ed.) 1-16. Ortony, A. 1993. The Role of Similarity in Similes and Metaphors. In Ortony (ed.): 342-356. Paivio, A. and Walsh, M. 1993. Psychological Processes in Metaphor Comprehension and Memory. In Ortony (ed.): 307-328. Paprott, W. and Dirven, R. 1985. (eds.) The Ubiquity of Metaphor: Metaphor in Language and Thought. Amsterdam. Pender, E. E. 2002. Images of Persons Unseen. Sankt Augustin. Perelman, C. 1969. Analogie et mtaphore en science, posie et philosophie. Revue internationale de philosophie 87: 3-15. Perelman, C. 1982. The Realm of Rhetoric. Notre Dame. Perelman, C. and Olbrechts-Tyteca, L. 1971. The New Rhetoric. A Treatise on Argumentation. Notre Dame. Petit, A. 1988. Mtaphore et mathsis dans la Rhtorique dAristote. Recherches sur la philosophie et le langage 9: 59-71. Philippe, M-D, 1969, Analogon and Analogia in the Philosophy of Aristotle. The Thomist 33: 1-74. Porphyry. 1957. In Aristotelis Categorias Commentarium. Busse, A. (ed.) Berlin. Porphyry. 1992. On Aristotle Categories. Strange, S. K. (tr.) London. Richards, I. A. 1936. The Philosophy of Rhetoric. London. Ricoeur, P. 1974. Metaphor and the Main Problem of Hermeneutics. New Literary History 6: 95-110. Ricoeur, P. 1986, The Rule of Metaphor. London. Rodrigo, P. 1988. LEuphorie de la langue. Sur le statut aristotlicien de la mtaphore. Recherches sur la philosophie et le langage 9: 73-90. Rorty, A. O. 1992. Essays on Aristotles Poetics. Princeton. Rorty, A. O. 1996. Essays on Aristotles Rhetoric. Berkeley. Ross, J. F. 1981. Portraying Analogy. Cambridge. Rudolph, E. and Wismann, H. (eds.) 1992. Sagen, was die Zeit ist: Analysen zur Zeitlichkeit der Sprache. Stuttgart. Ryan, E. E. 1984. Aristotles Theory of Rhetorical Argumentation. Montral. Sacks, S. 1980. (ed.) On Metaphor. Chicago. Snchez Meca, D. and Domnguez Caparrs, J. (eds.) 1992. Histora de la relacin filosofa-literatura en sus textos. Barcelona. Searle, J. R. 1993. Metaphor. In Ortony, A. (ed.): 83-111.

190

OROURKE/PATSIOTI BIBLIOGRAPHY

Sebeok, T. A. (ed.) 1986. Encyclopedic Dictionary of Semiotics. Berlin. Sewell, E. 1964. The Human Metaphor. Notre Dame. Shibles, W. 1971. An Analysis of Metaphor in the Light of W. M. Urbans Theories. The Hague. Shibles, W. 1972. Essays on Metaphor. Whitewater. Simplicius. 1907. In Aristotelis Categorias Commentarium. Kalbfleisch, C. (ed.) Berlin. Sinnreich, J. 1969. Die Aristotelische Theorie der Metapher. (Dissertation) Mnchen. Slattery, M. 1955. Metaphor and Metaphysics. Philosophical Studies 5: 89-99. Soskice, J. M. 1985. Metaphor and Religious Language. Oxford. Spurgeon, C. 1936. Shakespeares Imagery and What It Tells us. New York. Stanford, W. B. 1936. Greek Metaphor. Studies in Theory and Practice. Oxford. Swiggers, P. 1984. Cognitive aspects of Aristotles theory of metaphor. Glotta 62: 40-45. Tamba-Mecz, I. and Veyne, P. 1979. Metaphora et Comparaison selon Aristote. Revue des tudes grecques 92: 77-98. Thomas, O. 1969. Metaphor and Related Subjects. New York. Vegetti, M. 1994. Quand la science parle vide: procds dialectiques et mtaphoriques chez Aristote. In Coorebyter, V. (ed.): 7-32. Vico, G. 1988. New Science. Bergin, T. G. and Fisch, M. H. (trs.) Ithaca. Waggoner, J. E. 1990. Interaction Theories of Metaphor: Psychological Perspectives. Metaphor and Symbolic Activity 5: 91-108. Ward, A. 1987. The Unfurling of Entity: Metaphor in Poetic Theory. New York. Wheelwright, P. 1960. Semantics and Ontology. In Knights and Cottle (eds.): 1-9. Wheelwright, P. 1962. Metaphor and Reality. Bloomington. Wheelwright, P. 1954. The Burning Fountain. Bloomington. Wilson, M. 2000. Aristotles Theory of the Unity of Science. Toronto. Wu, K.-M. 2001. On Metaphoring. Leiden.

COLLOQUIUM 6

WAS ARISTOTLE A PARTICULARIST?

A. W. PRICE

I. Preface I open with a slightly abbreviated quotation which is faithful to Aristotles conception of phronsis, or practical wisdom, though it offers no explicit answers to some recent questions:
What it is to be virtuous, how we are to gain the just idea and standard of virtue, how we are to approximate in practice to our own standard, what is right and wrong in a particular case, for the answers in fullness and accuracy the philosopher refers us to no code of laws, to no moral treatise, because no science of life, applicable to the case of an individual, has been or can be written. An ethical system may supply laws, general rules, guiding principles, a number of examples, suggestions, landmarks, limitations, cautions, distinctions, solutions of critical or anxious difficulties; but who is to apply them to a particular case? whither can we go, except to the living intellect, our own, or anothers? What is written is too vague, too negative for our need. It bids us avoid extremes; but it cannot ascertain for us, according to our personal need, the golden mean. The authoritative oracle, which is to decide our path, is something more searching and manifold than such jejune generalizations as treatises can give. It comes of an acquired habit, though it has its first origin in nature itself, and it is formed and matured by practice and experience; and it manifests itself, not in any breadth of view, any philosophical comprehension of the mutual relations of duty towards duty, or any consistency in its teachings, but it is a capacity sufficient for the occasion, deciding what ought to be done here and now, by this given person, under these given circumstances.

The writer is John Henry Newman, and the passage comes from An Essay in Aid of a Grammar of Assent (268-269). Where he sees them to diverge, he quietly follows the truth, not Aristotle: he proceeds to question whether phronsis is a single faculty or a family of faculties, and whether the virtues form such a unity that a man cannot be just and cruel, brave and sensual, imprudent and patient. I cite him as a moral, not a philological, authority. One may be persuaded by what I have quoted without accepting that one is thereby being persuaded by Aristotle. So far, however, I believe that they coincide. My task today is to defend this as a reading of Aristotle, and to clarify it within the always shifting context of current philosophy.

192

A. W. PRICE

A recent line of thinking which may further our understanding of Aristotle is particularism. This label, being applicable to somewhat different claims, rather identifies a tendency than a dogma. We may oppose particularism to generalism, trying to identify where he stands, through a variety of issues. The questions that I shall try to answer, each within a section of this paper, are the following: (1) Does Aristotle suppose that an agents practical decisions apply to a particular situation an articulate general specification of eudaimonia? (2) Does he suppose that there are any principles to guide decisions that apply without exception? If he does, how large a role does he allow them? (3) Does he suppose that there are factors to be taken into account whose valence is invariable between different contexts? (4) Does he suppose that, as Jonathan Dancy has put it (1993: 50), there is nothing that one brings to the new situation other than a contentless ability to discern what matters where it matters? An extreme particularist reading of Aristotle would answer no to the first three questions, and yes to the last. I shall argue for positive answers to (2) and (3), and negative ones to (1) and (4). However, I fully accept as Aristotelian what I have quoted from Newman. This is compatible with my answers to (2) and (3), so long as they allow that most principles admit exceptions, and that relevant factors may need to be quite specific if they are not to vary in valence between contexts. I hope to make out that an agent brings to situations as they arise a character that infuses exercises of judgement. A failure to do justice to the ineliminable role of character may be common, ironically, to forms of generalism and of particularism that may otherwise seem polar opposites. II. The Grand End View We read in the Eudemian Ethics that choice is of something and for the sake of something (tinos kai heneka tinos, II 2, 1227b37; cf. II 10, 1226a11-13). 1 It thus equally relates an agent to an end, and a way or means (some act that subserves the end in the circumstances). Choice comes of deliberation, which takes an end as given: No one deliberates about the end (telos)that is there for everyone; men deliberate about the

_________
I take, or adapt, my translations of the Eudemian Ethics from Woods (1992), of the Nicomachean Ethics either from Ross (1925) or from Rowe (2002).
1

WAS ARISTOTLE A PARTICULARIST?

193

things that lead towards it, whether this or that contributes to its attainment, or else, when that has been decided, how it will come about (II 10, 1226b9-12). It is inferred that those who have no goal (skopos) before them are not in a position to deliberate (b29-30, cf. 1227a5-9). The virtue that directs deliberate action is practical wisdom, and its endas we read in one of the books common to the Eudemian and Nicomachean Ethics is a broad one: It is thought to be a mark of a man of practical wisdom to be able to deliberate well about what is good and beneficial for himself, not in some particular respect, e.g. about what sorts of thing conduce to health or to strength, but about what sorts of thing conduce to the good life in general (EN VI 5, 1140a25-8). The practically wise man is the man who is without qualification good at deliberating, being capable of aiming in accordance with calculation (logismos) at the best for man of things attainable by action, presumably living well (VI 7, 1141b13-14). The term logismos connotes more than the derivation of an action or decision from two premises. We read in the De Motu Animalium of appetite saying I must drink, and perception or imagination or reason adding This is a drink (7, 701a32-3); which exemplifies a pattern of explanation that may be equally applicable to a thirsty man, and a thirsty dog. Yet Aristotle denies to animals other than men choice, deliberation, calculation, and a supposition about the why (hypolpsis tou dia ti, EE II 10, 1226b23, within b21-5). I take the last to have two aspects: it introduces something specific, but it subsumes it within the good life in general (EN VI 5, 1140a28). Other animals lack that global determinable end, and hence do not act in ways explicable by ascribing to an agent not only a goal, but the subsumption of a goal within that end. They pursue, on occasion, drinking, but never drinking for the sake of living well. Here the specific drinking mediates between the general acting well and the particular doing this. What counts as acting well needs to be provisionally specified (to some greater degree of determinacy) before the availability of an acceptable way or means can be worked out. This may lead one into interpreting the supposition about the why as an attempt by the agent to provide a general determination of the determinable living well, that is, as J.L. Austin put it, to spell out what fills the bill of achieving eudaimonia (1967: 277). This would be a concrete specification of how to live well that only awaited implementation from occasion to occasion of action. The rationality of an action would be displayed by its derivation from the blueprint against a background of factual circumstance. The blueprint itself would be unchanging, though one wouldnt need to cite more than the relevant parts of it in relation to any one action. Ethical virtue, or excellence of character, would then ensure that desire fell in line behind reasons dictates, so that nothing internal

194

A. W. PRICE

could cloud their reception or impede their implementation; their content would owe nothing to desire. 2 Such an understanding of the starting-point of deliberation has been termed by Sarah Broadie the Grand End view (1991: 198). The label intimates that she places it within an ethics of fantasy; it need not follow that it is un-Aristotelian. David Bostock comments, In so far as Broadie is claiming that no one does, in practice, have a fully worked-out picture of what eudaimonia consists in, and that no one does, in practice, refer to such a picture in all their deliberations, I agree with her entirely. But the question is not whether I agree but whether Aristotle agrees, and surely the signs are that he did not (2000: 84). He then cites a passage from EE I 2 that was also used by Richard Kraut (1993) in an early response to Broadies book. Aristotle has just mooted that eudaimonia may consist in one or more of three things that are taken to be the most choiceworthy: wisdom, or virtue, or pleasure. People disagree about whether one, or two, or three of these really are part of eudaimonia, and, if more than one of them are, how to rank their contributions (I 1, 1214a30-b6). He continues (I 2, 1214b6-14),
Taking note of these things, everyone who can live according to his own choice should adopt some goal for the fine life, whether it be honour or reputation or wealth or cultivationan aim that he will have in view in all his actions; for not to have ordered ones life in relation to some end is a mark of extreme folly. But above all, and before everything else, he should distinguish in his own mindneither in a hurried nor in a dilatory mannerin which human thing living well consists, and what those things are without which it cannot belong to human beings.

It is tempting, reading this passage within the context that I have set, to take it to be inviting each agent to elaborate for himself a conception of how to live that will then be the starting-point of all his deliberations. That the invitation may reasonably be declined as impracticable becomes an objection to Aristotle. Nothing quite like that is said in the Nicomachean Ethics (from which Kraut infers that the two Ethics may diverge). However, there are parallel discussions, initially in I 4. There Aristotle writes, Verbally there is very general agreement; for both the general run of men and people of superior refinement say that it [the highest of all goods achievable by action] is happiness, and identify living well and doing well with being happy; but with regard to what happiness is they differ, and the many do not give the same account as the wise (1095a17-22). He then cites such things as pleasure, wealth, honour, and health (a23-5). In the next chapter, he at-

_________
2

Such a view is sketched, and resisted, in McDowell (1998: 11).

WAS ARISTOTLE A PARTICULARIST?

195

taches himself to an old tradition in distinguishing the three lives of pleasure, politics, and theory (I 5, 1095b14-19). He there advances what one may call structural points: honour depends on the bestower (b24-5); virtue is a disposition, and implies neither action, nor success in action (1095b32-1096a2). Rather later he argues that the life of virtue is also pleasant, and that it needs external goods as instruments or enhancers (I 8, 1099a7-b8). When he returns to a comparison between the lives of theory and of practice much later (X 7-8), he is more reflective than decisive, and only imprecisely prescriptive even at his most decisive. It is a noble sentiment that we must, so far as we can, make ourselves immortal, and do all we can (panta poiein) to live in accordance with the best thing in us (X 7, 1177b33-4). I intend the phrase do all we can (literally, do all things) to be neutral between perform every action, and strain every nerve (Ross) or go to all lengths (Irwin). Yet the second is more idiomatic. 3 And it alone makes any sense of the transition between X 7 and X 8: whatever Aristotle is prescribing, it is not that we should attend to theory whenever we (literally) can. He is assigning to theorizing a unique value, but not absolute priority. He is being demanding, not demented. Is the Eudemian passage so different? It consisted of two sentences. Let me take the second first: But above all, and before everything else, he should distinguish in his own mindneither in a hurried nor in a dilatory mannerin which human thing living well consists, and what those things are without which it cannot belong to human beings (I 2, 1214b11-14). These questions are answered neither by a blueprint nor by an inventory, but by some commonplaces: For being healthy is not the same as the things without which it is not possible to be healthy; and this holds likewise in many other cases too. So, living well also is not the same as the things without which living well is impossible These are the reasons for the dispute over being happywhat it is and the means by which it comes about: things without which it is not possible to be happy are thought by some to be parts of happiness (b14-27). So the instruction is not to elaborate a full specification of eudaimonia (which isnt what dihorizesthai could signify anyway), but to distinguish components from necessary conditions. Thus, in the other Ethics, if we view external goods rightly, as

_________
3 This ambiguity is absent from I 13, 1102a2-3 (since there panta are qualified as ta loipa), and VI 5, 1140b18-19 (where we read haireisthai panta kai prattein). It is present in a debated sentence of the Republic (VI, 505e1-2). Terence Irwin has recently urged that Socrates meaning there is that every soul is tireless in pursuit of the good, but not (which his detachment of appetite from reason may make problematic) that all its acts are for the sake of the good.

196

A. W. PRICE

facilitating fine action (EN I 8, 1099a31-b8), we shall lack the ambitions of a Croesus, and may be content without becoming acquisitive. Getting that distinction right is a precondition of acting wisely on the first sentence: Everyone who can live according to his own choice should adopt some goal for the fine life, whether it be honour or reputation or wealth or cultivationan aim that he will have in view in all his actions; for, not to have ordered ones life in relation to some end is a mark of extreme folly (EE I 2, 1214b6-11). This too must be interpreted in context, and not over-interpreted. As in the other Ethics, we may think of the three lives, the political, the philosophical, and the pleasure-loving, that are chosen by all who have the opportunity to choose (I 4, 1215a35-b1). Anaxagoras can be cited with respect for thinking that it was the man who led a life without pain and free from stigma in matters of justice, or participated in some kind of godlike speculation, who was, humanly speaking, divinely-happy (b11-14), even though his first goal is complex, and his second unlikely to be exclusive (participating in doesnt sound full-time). The one goal for the fine life that one always has in view may neither be ones only goal, nor exhaust ones conception of eudaimonia. Even if my life takes on a cohesion from some limited set of goals or values, it need not follow that I am fixated upon them exclusively. The agent who is always committed to justice must remain ever awake to its exigencies; yet he need not view any choice between equally just options as indifferent. 4 What Aristotle is advising is some way of living (such as one of the three lives) focused upon a distinctive end that is rather a priority than an object of monomania. His prescription is not vapid, for not all our lives display any such unifying and distinguishing focus; yet it is neither eccentric nor extraordinary. If this is right, the passage has nothing to say about a blueprint for living, or even a decision-procedure for acting. The man who decides to be an expert scientist, or an honest entrepreneur, is not endorsing any grand end sufficient to direct all his deliberations. And this may warn us of the likely ambiguity of much of the evidence. Typically ambivalent in itself, though not (I believe) in context, is a sentence from the common books: Practical syllogisms have a starting-point, viz. since the end, or what is best, is such-and-such, whatever that may be (let it for the sake of argument be whatever you like) (EN VI 12, 1144a31-3). 5 If this end is nothing less than eudaimonia, then the agents starting-point is his own charac-

_________
4 Which would constantly reduce him, in Aristotles view, to the predicament of Buridans ass; see Cael. 2.13.295b30-34. 5 Or could for the sake of a logos (logou charin) mean to make a piece of reasoning possible?

WAS ARISTOTLE A PARTICULARIST?

197

teristic determination of it in specific terms; whether this be right or wrong, it will guide him in any situation of action. Now the end, or what is best can indeed signify no less: compare the phrase the best for man of things attainable by action (VI 7, 1141b13-14), where for man indicates what is a goal for everybody. Yet the passage continues, And this is not evident except to the good man (1144a34). This variable and contingent end cannot be read as a general goal in life; for the end that is only evident to the good man must be a good end, and not a standing end that may be either good or bad. I take it to be whatever concrete goal a good agent would adopt in the circumstances; if so, the variation is between not different men, but different situations. 6 A spectrum of ends, all subordinate to eudaimonia, may be roughly mapped as follows. At one extreme, there are the life-goals of those who accept, or anticipate, the advice given in EE I 2 by ordering their lives by relation to some end. At the other, there are the goals thrown up from occasion to occasion. When we read that The agents themselves must in each case attend to what is appropriate to the occasion (ta pros ton kairon skopein, EN II 2, 1104a8-9), we should be put in mind of an equal flexibility in the selection of ends and of means. Within his discussion of mixed actions, Aristotle tells us, The end of the action is according to the occasion (III 1, 1110a13-14). Thus a sea-captain may discover, in a storm, that he cannot retain his cargo without risking the lives of his men. If he then jettisons the cargo in order to secure his mens safety, he adopts an unwelcome means of escape from an unexpected crisis. Yet Aristotle may have in mind something between those two extremes when he writes, Virtue and vice respectively preserve and destroy the starting-point, and in actions the final cause is the starting-point (VII 8, 1151a15-16). This must be an end that is more than sudden, but need not be lifelong. We may think of periodic ends, such as writing a book, or weaving a tapestry, which influence an agent through a stretch of his life. What then of Aristotles distinction between deliberating in some particular respect, say with regard to health or strength, and in respect of what sorts of thing conduce to the good life in general (VI 5, 1140a258)? The former sounds like deliberation about production rather than action, and we have been told that technical thinking is always subordinate

_________
6 More equivocal is Aristotles chapter on phronsis (VI .5). When we read The starting-points of things to be done are that for the sake of which they are to be done (1140b1617), and that pleasure obscures the starting-point, so that it is no longer evident that for the sake of this or because of this he ought to choose and do whatever he chooses and does (b17-19), the reference could be to a Grand End; but it could equally be to whatever end for the occasion is prompted by a proper appreciation of the plurality of human goods (b9).

198

A. W. PRICE

to practical: Everyone who produces produces for an end, and what is an end without qualification is not the end of production (being relative to something else, and the end of something), but the end of action; for acting well is an end (VI 2, 1139b1-4). That medical reasoning rests upon some specification of health is stated in the Metaphysics: A healthy patient is produced as a result of the following reasoning: since health is soand-so, if the patient is to be healthy, he must have such-and-such a quality (= 7, 1032b6-7, cf. EE I 8, 1218b19-20). 7 Analogy may then suggest that a practical syllogism starts from a specification not of some goal of production, but of the goal of action, which is eudaimonia. This is what the Grand End view supposes. I have already discussed what seems to me indecisive evidence of that. And I have elsewhere traced Broadies footsteps in a contrary directionsaying things that I have managed not to repeat here. 8 Her explanation contrasts a technical goal not with a blueprint for life, but with a wider practical receptivity. She characterizes practical deliberation, within a given situation, by two features: a focus upon a limited and accessible target, and an openness to whatever considerations come into play: The practical agent differs not by being focused on another special sort of good that is special because unrestricted and categorically demanding, but by being focused on a restricted good (not always the same one, either) with a focus that sets no limit on the considerations that could affect which way he goes with regard to that good or to the points of view that might make a difference (1991: 211). His particular action is oriented not towards a segment of a lifeplan, but towards some good that he can acceptably achieve in the context. His immediate target may be narrow (and, as Broadie emphasizes, often quite commonplace); but he selects it, and aims at it, wisely and without blinkers. 9

_________
I adopt a translation from Kenny (1979: 131). See Price (2005: 266-70). Kraut rightly presses me on the practical role of ethical theory. After all, we read We are not inquiring into what excellence is for the sake of knowing it, but for the sake of becoming good, since otherwise there would be no benefit in it (EN II 2, 1103b27-9). Aristotles ethical writings offer us no blueprint, but have practical implications even at their most abstract. Just two illustrations: the doctrine of the mean, itself purely formal, suggests a policy of steering towards either the less distant, or the less tempting, extreme (II 9, 1109a30-b7); the marks of eudaimonia (I 7-8) turn out also to privilege contemplation above other components of eudaimonia (X 7). And there is far more to concede than can be mentioned here. How this is consistent with the provision of ends by excellence of character (which I stress in 5 below) is not to be settled in a footnote; yet it must be that, while the profitable study of ethics presupposes a good upbringing (I 4, 1095b4-8), there are actual risks and possible refinements to a good character that demand the attention of a not wholly conservative theorist.
7 8 9

WAS ARISTOTLE A PARTICULARIST?

199

III. Principles Acts fall under principles. Are any practical principles absolute, that is, exceptionless? Certainly characteristic of Aristotle is an awareness of exceptions both to practical principles, and to ethical generalizations. That fine and just actions, which political science investigates, admit of much variation and fluctuation is a thought immediately applied to the second rather than the first: an illustration is that men have perished by reason of their wealth, and others by reason of their courage (EN I 3, 1094b14-19). This actually implies a degree of invariance: evidently an act of a kind generally brave may still count as brave itself even if it turns out fatal (though that may make it rash if nothing sufficiently grave is at stake). Hence it is not implied that principles of courage are always to be accompanied by the qualificationwhich would still be a general oneif it isnt fatal. However, we soon find that it is typical of practical principles to take on the indefinite qualification for the most part (hs epi to polu). This is exampled in EN IX 2 by the ranking principles that fail always to resolve conflicting obligations. These give rise to many variations of all sorts in respect both of the magnitude of the service and of its fineness and necessity (1164b28-30). We must for the most part return benefits rather than oblige friends, but there are difficult cases: Should a man who has been ransomed out of the hands of brigands ransom his ransomer in return or should he ransom his father? (1164b31-1165a1). Generally (katholou) the debt should be paid, Aristotle repliesbut katholou here evidently means generally, not universally. That similar considerations must apply not just to second-order rules ranking first-order ones, but to first-order rules themselves, is explicit in the better-known discussion of equity in EN V 10. Equity is a correction of legal justice, but what is said about the imperfection of positive laws must apply equally to personal principles. Aristotle concludes with a famous analogy: When the thing is indefinite the rule also is indefinite, like the leaden rule used in making the Lesbian moulding; the rule adapts itself to the shape of the stone and is not rigid, and so too the decree is adapted to the facts (1137b29-32). Of course, there are exceptions to this rejection of the exceptionless. The least interesting instances are truisms of the kind One should act finely. EN IX 2 has a sentence that moves in a narrow circle: We ought

200

A. W. PRICE

to render to each class what is appropriate and becoming (1165a17-18). 10 Equally evident tautologies are generated by the virtues and vices, given Aristotles view, which is a corollary of his doctrine of the unity of the virtues, that all acts displaying a virtue are right, and all acts displaying a vice wrong. 11 Slightly more interesting are rules prescribing a regard for values even in contexts that tempt their neglect; thus Terence Irwin justifiably infers from EN IV 6-8 that we ought never to make fun of people simply to raise a laugh, without any regard for what is fine or expedient (2000: 111). There are also principles that prescribe making distinctions: We should not give preference in all things to the same person (IX 2, 1164b30). More substantive are a handful of universal and concrete prohibitions to which Aristotle expresses allegiance. One may forbid matricide: Some acts, perhaps, we cannot be forced to do, but ought rather to face death after the most fearful suffering; for the things that forced Euripides Alcmon to slay his mother seem absurd (III 1, 1110a26-9). However, this doesnt entail that Orestes was wrong to kill his; it is left open what range of acts is being excluded, and whether it is neutrally identifiable. More openly generalizing is another passage: Not every action admits of a mean; for some have been named in a way that combines them with badness from the start, e.g. adultery, theft, murder; for all of these and suchlike things owe their names to the fact they are themselves bad, and not the excesses or deficiencies of them. It is not possible, then, ever to be right with regard to them; one must always be wrong. Nor does goodness or badness with regard to such things depend on committing adultery with the right woman, at the right time, and in the right way, but simply to do any of them is to go wrong (II 6, 1107a8-17). 12 It is unclear what range of cases Aristotle has in mind by suchlike things (ta toiauta, a12). Some might argue, precisely from the clause These and suchlike things imply by their names that they are themselves bad (a12-13), that all that he intends here is a series of tautologies. Take his three cases in turn. That one should not commit theft risks tautology: theft is feloniously depriving a man of his property, that is, of what is legally his and cannot

_________
This is introduced by a substantive principle, We ought to render different things to parents, brothers, comrades, and benefactors (1165a16-17). However, this isnt falsified by a case where I should do for my brother precisely what I do for my father; so it cannot be interpreted as a universal truth about everything that one ever does for a parent or a brother. 11 Thus Aristotle interprets the assessment of an option as, in some way, virtuous or vicious as entailing a practical judgement prescribing or proscribing it; see Price (2005: 260). 12 I slightly abbreviate the passage to omit examples of prohibited passions, which raise special issues.
10

WAS ARISTOTLE A PARTICULARIST?

201

justly be taken from him against his will (except, perhaps, in some public emergency). So long as we count what is unjust as thereby wrong, as Aristotle does, it appears to go without saying that theft is wrong. (If we hold that theft can be the lesser of two evils, we had better either reject the unity of the virtues, or deny that, in its use within a definition of property, the term just has any necessary implications for what is virtuous.) Murder signifies something like unlawful, and hence unjust, homicide. Again, given the unity of the virtues, and the use of just to signify a virtue, it looks to be a tautology that one should not commit murder. Of the three, adultery may best appear definable in non-evaluative terms: Aristotle elsewhere counts as an adulterer (moichos) a man who has intercourse with married women (i.e., no doubt, any married woman other than his own wife, EE II 4, 1121b20-21). However, it would seem that a speaker who doesnt think that the marriage ceremony creates any obligations cannot use the term adultery any more seriously than a person who rejects the institution of property can use the term theft. In each case, it may appear, non-ironic and truth-directed use of the term theft, murder, or adultery is impossible unless the speaker disapproves of the acts to which they apply. However, I believe that the argument is mistaken, even if Aristotle is right in each case to make the prohibition absolute. 13 Use of the term murder commonly conveys assent to a moral consensus; yet it also thereby connotes a set of kinds of act. To claim that all acts of those kinds are wrong is not to state a tautology. An act of homicide counts as murder so long as it is of one of those kinds, and hence lacks what, in defining those kinds, we agree in recognizing as exculpatory features. On this view, it is as meriting our condemnation that those kinds of act are appropriately placed within that set; but that a particular action is of one of those kinds, and hence an act of murder, is a matter of fact. So if someone calls an action murder, it doesnt follow logically, even if he is telling the truth, that it was wrong, or even unjust. It is indeed shared attitudes that give

_________
13 The argument attempts to assimilate the three terms theft, murder, and adultery, and I shall challenge it where it is strongest, which is with murder. However, it may well be that applications of the three terms (and their Greek equivalents) reflect and attract attitudes rather differently. We commonly think of any murder as bad overall. (However, think of a consequentialist who advises A to murder C if the alternative is that B murders C and D.) Yet certain acts of theft may be bad not overall, but either in a way (i.e., not overridingly), or only presumptively (i.e., in their concrete reality, not at all). And adultery, in its literal sense (for the OED recognizes an extended sense in Scripture), may be a pejorative term whose sense, determining reference, is descriptive in origin as in upshot, since its extension has a unity even within a neutral perspective; cf. Price (2000: 142) on the Greek word barbaros, applicable to all non-Greeks.

202

A. W. PRICE

such a term its sense; but to apply it, say in asserting That was murder, is not to say anything that entails that the attitudes are justified both in general, and in excluding exceptions. Fully to grasp the sense of the term probably requires sharing the attitudes that give its application a shape; it is at least to be able to enter into them sympathetically. Yet it is not part of its sense that all (or, indeed, any) of the acts to which it applies merit the attitudes that mold that sense. Hence it is paradoxical to say That was a murder that had nothing against it, since this at once exploits the attitudes that have determined the sense and extension of the term, and distances the speaker from them in this caseand yet without treating it as a marginal instance to which the concept only contestably applies; yet it is not, I think, self-contradictory. 14 So Aristotles confidence in these exceptionless principles remains striking, and is not to be explained away. It is no accident they are all negative as well as concrete: Do nothing that counts as murder is a practicable rule, whereas Do everything that respects life is not. Yet is the universality surprising, perhaps even inconsistent? Broadie finds it so: It contradicts some of his best-known positions to hold that any neutrally described kind of action (let alone all the members of some set of kinds) is always right to perform (1991: 210). She concedes that what she takes to be a ban on matricide (EN III 1, 1110a26-9) may carelessly contradict this, but comments, We should not on that account ascribe to him an official doctrine inconsistent with another of his official doctrines (1991: 261 n. 19)whatever the force of official is meant to be. She takes her view to be supported by the very doctrine of the mean, supposing it to imply that the decision whether to do a certain thing is never taken care of

_________
14 As I have noted elsewhere (2005: 276, n. 33), it is no accident that the terms that enter into such principles have extensions that are contestable at the margin. This makes the principles flexible without rendering them anodyne. Compare One should never tell a lie, a rule that remains controversial even outside debate about whether, for example, equivocation is a subtle form of truth-telling or a blatant kind of lying. See Foot (2001: 77-8) for a defence of lying through ones teeth in certain exceptional circumstancescircumstances that may totally exculpate the agent even though they do nothing to make the term inapplicable; for another example, cf. Hugos Les Misrables (1862), I 8.5, where the excited author is led to exclaim to one of his characters, a nun who has just told a lie for the first time in her life, Que ce mensonge vous soit compt dans le paradis! White lie is a clich; good murder is not (as yet). However, the student in Dostoyevskys Crime and Punishment (1866), whose advocacy of killing and robbing the old moneylender Raskolnikov overhears to his ruin (I 6), surely views that one little, insignificant transgression as meriting the phrase. We mayor may notwish that Aristotle had recognized such possibilities in principle.

WAS ARISTOTLE A PARTICULARIST?

203

by the description of what might be done (1991: 210). 15 The implication is hardly self-evidentthough it would take an exposition of the doctrine to assess it properly. 16 Does what I quoted from Newman really commit him, if he thinks things through, to conceding that it can be right, on occasion, to commit adultery, or even to kill ones mother? Surely not. We may recall a remark in Chapter 7 of his novel Loss and Gain (1848): The moral world is not an open country; it is already marked and mapped out; it has its roads. You cant go across country; if you attempt a steeplechase, you will break your neck for your pains. How much is implied in The Grammar of Assent by his emphasis upon practical judgement? No more than Aristotle does he claim that judgement is always required for correct practice. Even when it is required, it can hardly be the case that every conceivable option has to be assessed by an exercise of judgement. Most of the options that an agent might otherwise consider are evidently unacceptable, whether universally or in context: they just dont enter his deliberationsunless to be rejected at once. This in no way implies that the option that should be privileged is yielded by the mechanical application of a rulethough this is one way in which he may be spared a state of uncertainty that requires resolution by deliberation, 17 a condition (happily not ubiquitous) which demands that he make up his mind. One is right to reject the ideal of precision in ethics by recognizing how often and to what extent it fails to apply, and not by pretending that all policies and decisions, negative and positive, are difficult. 18 The matter is less transparent, of course, just because the very notion of judgement excludes any simple procedure for distinguishing good from bad judgement. If we understood better what fixed the mean, we would also be better placed to explain when rules can be absolute, and when

_________
A later note ad 1107a11-12 (Broadie and Rowe, 2002: 306) is ambivalent: Actions falling under these descriptions are wrong prima facie, which is enough to distinguish these from neutral descriptions, but here Aristotle takes the stronger position that they are always wrong. However, his discussion of hard choices in III 1 holds out the possibility that he might consider them excusable under certain circumstances. This claims to know Aristotles mind better than Aristotle himself; and the reference to III 1 is curious in the light of the occurrence precisely there of the other passage that causes difficulty (1110a26-9). 16 There is no such implication in the most sophisticated and satisfactory treatment I know, which is Mller (2004). 17 I quote from Watson (2004:125); cf. EN III 3, 1112a33-b9. 18 This is usefully brought out, in application to Aristotles craft analogies, in Pakaluk (unpublished). Wiggins (2004: 482, n. 3) well distinguishes deciding upon a particular omission as itself the solution to some practical problem, and excluding certain general ways of acting from the space of deliberation.
15

204

A. W. PRICE

(more often) they need to be applied with discretion. Aristotle does lack one ground for making exceptions, which is the notion of negative responsibility that is part of consequentialism. Take any unmentionable act: consequentialism implies that we have only to imagine a situation in which either I perform it once, or someone else performs it twice, to make my performing it not only permissible but obligatory. Aristotle is not entirely without such a notion of vicarious responsibility: the good man may even give up actions to his friend; it may be finer to become the cause of his friends acting than to act himself (IX 8, 1169a32-4). Yet his view that, in such a case, the good man is seen to assign to himself the greater share in what is fine (a34-b1) shows that it is still his action or restraint that is important from his point of view. 19 And that it may be fine to hold back from acting finely, if this prompts another to act finely, does not entail that it must be base to hold back from acting basely, if this prompts another to act basely. IV. Variable Valence One aspect of Dancys particularism is more precisely labelled variabilism. The particularist claims that features of an act may count for or against it in different contexts. This gives them what is called variable valence. An extreme view would be that all features vary in valence in this way; a weak view would be that some features do. We find that, while certain of Dancys examples are present in Aristotle, they are otherwise grounded; and he offers counter-indications to others. The most plausible instance is pleasure. As Dancy justly observes (1993: 60), Pleasure at a wrong action compounds the wrong. Why should this be? Why should the object of the pleasure alter its valence? Aristotles view of the matter is explicit: Since activities differ in respect of goodness and badness, and some are worthy to be chosen, others to be avoided, and others neutral, so, too, are the pleasures; for to each activity there is a proper pleasure. The pleasure proper to a worthy activity is good and that proper to an unworthy activity bad (EN X 5, 1175b24-8). This is not a piece of free-floating moralism, but a plausible inference from a central principle within his account of pleasure, which is, as he then restates it, that as activities are different, so are the corresponding pleasures (b36). Once we have described some good and enjoyable mode of activity sufficiently to identify what pleasure is taken in it by the agent,

_________
19

This may seem unattractive, wrongly; see Mller (1977).

WAS ARISTOTLE A PARTICULARIST?

205

that pleasure will have invariant valence: it will be a good, and count in favor, most obviously, of the activity itself. If the activity is by nature bad, then any pleasure proper to it will be characteristic of a bad agent, and make it worse. Aristotle goes so far as to claim that pleasures of a reprehensible kind may make it preferable not to be alive (EE I 5, 1215b25-6). However, such remarks relate rather to the type of the activity than to the context of the token. Nothing just cited implies, say, that if I take a gourmets pleasure in eating a marron glac on an occasion when I should abstain, the pleasure makes the eating worse. That Dancys grounding is different is plain from another, and to my mind less plausible, example. He instances (1993: 55-6) extracting a seaurchin from ones daughters foot, proposing that, if there is no alternative, the pain caused does not tell at all against the action taken. Here there is nothing inappropriate about the pain: anyone would find the operation painful to undergo. Dancy must hold that the necessity removes the disvalue from the pain, even though it is a perfectly natural corollary of the operation; and yet surely, we may object, the father acts reluctantly. 20 That Aristotle would disagree with Dancy is indicated by his discussion of unusual acts required by unusual situations in EN III 1. He writes of throwing goods overboard in a storm. In the abstract no one throws goods away voluntarily, but on condition of its securing the safety of himself and his crew any sensible man does so. Such actions, then, are mixed, but are more like voluntary actions; for they are worthy of choice at the time when they are done, and the end of an action is relative to the occasion (1110a9-14). Crucial here is the force of in the abstract (hapls). Rowe translates, or rather glosses, if it is just a matter of throwing it away, which permits a variabilist view that, in the storm, the act loses its disvalue. However, it appears that it is the act-token that Aristotle counts as mixed. When he concludes, Such actions are voluntary, but in the abstract perhaps involuntary (a18), he is shifting between not different tokens of the same type, but different perspectives upon the same act-

_________
20 Someone might say, That an operation is painful is no reason against it if it is really necessary. But if no reason here means no good reason (as it idiomatically may), this does not indicate that the pain of a life-saving amputation was not a bad thing before the invention of anaesthetics. (If so, why were they invented? Only for inessential surgery?) A better example for Dancy might be this: we are told that Nero fiddled as Rome burned, doubtless for fun; did not his aesthetic pleasure become bad in context? This raises two questions. First, was he taking an innocent pleasure in playing the fiddle, but at the wrong time, or a perverse pleasure in playing the fiddle inopportunely? (I presume that it was not from absence of mind that he felt like fiddling then.) Secondly, if the first was the case, should we say that, in context, the pleasure itself became simply bad, or that, though still mildly good in itself, it became very bad in context as unfitting to the occasion?

206

A. W. PRICE

token. His likely meaning is not that the type has varying valence, but that some tokens of it are unwelcome in themselves, but accepted in context. If so, he assumes that the act-type throwing goods overboard has an invariant negative valence, though one that is overridden in certain cases of necessity. Why else should it be plausible, in English, to describe the captain as acting unwillingly? One may, admittedly, find this rather confused. Nothing that I have reported, from Aristotle or Dancy, is really about tokens rather than types. Better stated, the issue is whether the general act-type (enjoying oneself, or throwing goods overboard) has a valence that is invariable through specific sub-types (e.g., enjoying being bad, or throwing goods overboard in a storm). The variabilist might hold, strongly and implausibly, that any acttype, however specific, has further sub-types that differ in valence. Or else, weakly and indeterminately, he might hold that some act-types vary in valence between sub-types. In either case, we must ask for his ground. Aristotle offers an intelligible one in the case of pleasure: pleasures that vary in nature may also vary in valence. However, it is then tempting to suppose that the whole value inheres in the determinate pleasure, so that no value attaches to pleasure as a determinable. If so, it isnt the thing that varies in valence; rather, pleasure is a general term that applies mostly to desirable experiences, but also, though secondarily, to undesirable experiences of enjoying the badness of bad things (X 5, 1176a26-9). And then there is nothing that has variable valence. Thus the grounding that Aristotle may appear to offer for one instance of variable valence actually tells against it. 21

_________
A modified variabilism finds room for the concept of a default valence. (On this, see Dancy 2004: 112-17, which is carried further in Dancy, forthcoming.) We might say that pleasure carries a default valence which is positive. If so, pleasure is good, ceteris paribusno positive extra is needed in a situation to make it good; and yet, when its object is evil, its valence gets reversed. Aristotle makes a corresponding contrast: Pleasure is good is true of what really count as pleasures, viz. those of good people (X 5, 1176a17-19); it is only in a derivative sense that there are bad pleasures, for these are only pleasures for bad people (a21-2). However, there is then no single type of mental state that is variably realized in good and bad pleasures. These stand to each other as real and relative, primary and secondary, and not as varieties of a single type. Even within the class of good pleasures, the ground of value is not the fact, but the object, of enjoyment. The fact that an activity gives pleasure may provisionally indicate that it has value; yet it never constitutes its value. On this conception, hedonic value is highly specifica view of it that may be thought less generalist, and hence in a correlative sense more particularist, than a variabilist one.
21

WAS ARISTOTLE A PARTICULARIST?

207

V. Virtue and Action At the heart of particularism is the relation between a decision to act in some determinate way in a certain situation, and any general concerns or principles that may enter into that decision or be thrown up by it. An extreme particularism might deny that such concerns or principles do, or should, play any role in generating or explaining the decision. In the light of some choice by the agent, we might offer the generalization that he cares about things of a certain kind. Yet his having that concern will be explanatory neither of that decision, nor of any others; it will just be a corollary of the particular decisions that he takes. I quoted from Dancy a claim that makes no reference to general concerns at all: There is nothing that one brings to the new situation other than a contentless ability to discern what matters where it matters, an ability whose presence in us is explained by our having undergone a successful moral education (1993: 50). He also writes as follows (1993: 64):
Our account of the person on whom we can rely to make sound moral judgements is not very long. Such a person is someone who gets it right case by case. To be so consistently successful, we need to have a broad range of sensitivities, so that no relevant feature escapes us, and we do not mistake its relevance either. But that is all there is to say on the matter. To have the relevant sensitivities just is to be able to get things right case by case. The only remaining question is how we might get into this enviable state. And the answer is that for us it is probably too late. As Aristotle held, moral education is the key; for those who are past educating, there is no real remedy.

It is unclear how to interpret such phrases as nothing other than, and just is. They may just exclude any attempt to rest the ability to get things right upon some mechanism generating true answers to practical questions. If so, Dancy is already supported by Section II above. Or they may exclude any attempt to relate exercises of that ability to anything else that the agent brings to the situation. That some agents consistently get things right becomes a product of education that has no other describable effect. That others dont may be explained either simply by lack of education, or by impediments and interferences; the second case invites a richer present characterization than bare denial of the ability. But do exercises of the ability really proceed in the kind of accompanying mental void that one associates with the deliverances of telepathy, or the calculations of an idiot savant? Rationalist interpreters of Aristotle have difficulty with his recurrent, and clearly not careless, claim that the end or goal of action is initially set by ethical virtue (telos, EN VI 13, 1145a5-6, EE II 11, 1228a1-2; skopos, EN VI 12, 1144a8, EE II 11, 1227b22-4). This must give desire much

208

A. W. PRICE

more than the subordinate role allowed it by the Grand End view. Success in finding whether there is an acceptable way or means to achieving the end, which can alone settle whether the end is acceptably pursuable in the circumstances, is then owed to practical wisdom (EN VI 13, 1145a46). 22 Its exercise is commonly described as involving a process of deliberation (of the kind described in III 3), though Aristotle does once allow that a right choice may be characteristic without deriving from actual deliberation or reasoning (III 8, 1117a20-22). That variation may not be crucial here: whether the agent just chooses and acts rightly, or deliberates rightly first, his getting things right is an exercise of his ability to do so. What is there to be said about the exercise either of virtue or of practical wisdom, other than by identifying the kind of thing that it achieves? About the structure of deliberation Aristotle says much, though not alwaysbecause of his tendency to treat one thing at a timein a manner that clarifies its relation to perception, and the selection of an end. And some crucial passages, notably in EN VI, are famously opaque. W.D. Ross translates one of them (VI 11, 1143b2-5), rather freely, as follows:
The intuitive reason (nous) involved in practical reasonings grasps the last and variable fact, i.e. the minor premiss. For these variable facts are the starting-points for the apprehension of the end, since the universals are reached from the particulars; of these therefore we must have perception, and this perception is intuitive reason. 23

I prefer (with D.J. Allan, in the margin of his copy of Bywaters EN) the alternative manuscript reading to katholou (singular) to ta katholou (plural) in b5. Rosss the universals is innocuous so long as we connect it with a later observation, There are two kinds of universal term: one is predicable of the agent, the other of the subject (VII 3, 1147a4-5). Yet it is easier, in the earlier passage, if we read the universal, taking this to be whatever major premise is prompted by the perceived particulars. Aristotle had better not have in mind a practical induction that would derive general principles from particular cases, for two reasons: first, his focus is upon the minor premise, not the conclusion, and yet any generalization to practical principles must derive from conjunctions linking a minor premise to a conclusion (doing this, in these circumstances, may begin to suggest

_________
See Broadie (1991: 244-5). Since I accept the interpretation that it conveys, I should warn the reader that Rosss these variable facts are the starting-points for the apprehension of the end is really a gloss upon Rowes more literal these are the starting points of that for the sake of which (1143b4). If these signified actions, the thought would simply be that actions realize ends; since these signifies minor premises, the thought must rather be that ends are suggested by circumstances.
22 23

WAS ARISTOTLE A PARTICULARIST?

209

that, in such and such circumstances, I should always or usually do so and so); secondly, as Michael Woods well observed (1986: 159), talk of induction would separate perception (as providing the data) from nous (as generalizing from them), while Aristotle wishes to equate the two. Rather, the universal is reached from the particulars in that it is the concrete features of the situation which activate the general concern that is appropriate to them, and then gets expressed in the major premise. David Wiggins offers a helpful gloss: For here, in the capacity to find the right feature and form a practical syllogism, resides the understanding of the reason for performing an action, or its end. For the major premise and the generalizable concern that comes with it arise from this perception of something particular. So one must have an appreciation or perception of the particular, and my name for this is intuitive reason (2002a: 236). 24 The particularism expressed in the two sentences I quoted from Dancy at the beginning of this section is too exclusively cognitivist if it asserts baldly that the good agent is the man who, upon inspection of a situation, can identify what is to be done. This suggests a purely rational Kantian will, though one that responds to situations as they arise and not to universal imperatives. Wiggins is less unreal: It is hard to conceive of there being an evaluation of x and y in the absence of a structure of pre-existing concerns that will direct the imagining of what x amounts to and of what y amounts to and that will focus the evaluators attendant perceptions of the circumstances (2002b: 244; cf. 2004: 481). The roles of the virtues of character and of practical thinking are multiple and inter-connected. They supply the agent with distinctive standing concerns to supplement the common-garden concerns that we all carry around with us. They lend salience to certain aspects of a situation (those which invite, or endanger, the realization of his distinctive concerns). They determine that the goal selected is both generally acceptable, and acceptably pursuable in the situation. A tenable particularism must concede a need for standing concerns whose objects can only be general; what it can properly deny is that these concerns have a precision that yields universal principles whose citing in context entails particular decisions. Rather, practical principles are typically (if not universally) qualified by a for the most part which is not shorthand for exceptions that could be fully spelled out. One agent is distinguished from another by a mindset that he derives from his nature and experience and brings to a situation. Its origins will surely include some element of general exhortation as well as particular precept and actual

_________
24

Cf. McDowell (1998: 5).

210

A. W. PRICE

practice; 25 and it will be constituted by general dispositions of desire (and the pro tanto evaluations that inform and express them) as well as capacities of discrimination. Indeed, the two are not separable: discrimination, as an element in judgement, is not neutral, but involves detecting not just fine differences, but relevant ones. The agents pre-existing concerns provide criteria of relevance. A man who is keen on making mischief may be equally perceptive of opportunities, though the wrong ones. The eye of experience (VI 11, 1143b13-14) that is as trustworthy here as demonstrations are elsewhere is not ethically neutral: it is perceptive not of anything, but of the right things. While Aristotle could not consistently spell out its working as if they were demonstrative, we can say rather more about it than that it enables one to get things right. Judgement is exercised not only within a situation, but against the background that is constituted by ones character, which, in a mature agent, is something fairly constant that he carries around with him. 26 Distinguishing the concerns that are brought to bear upon the situation from the judgement that emerges, we may say that the agents distinctive sensibility is general and pre-existent, though its application is often irreducibly present and particular. Of course a mere difference in time and place can make no moral difference. To this extent, the particular judgement applies beyond its own case. Yet the perception that prompts desires selection of a goal need not be articulate beyond identifying the feature that connects the goal to the action proposed. Thus I may need to perceive that this is chicken if I am to connect my present situation to a goal of healthy eating (VI 7, 1141b20-21). Yet it is perceiving with pleasure that motivates pursuit. 27 If I am to perceive this with pleasure in a way that is sufficiently sensitive to possible complications (here and now may not be the time and place for eating, or healthy eating), I must be open to a wider range of contextual features than I can single out. Hence I may not know, when I decide to eat this, what future decisions are implicit in my present one. The premises of a practical syllogism specify what the agent is doing, and for what end, but themselves contain no precise answer

_________
Cf. EN X 9, 1179b4-9. This leaves open options that are surely to be selected not in accordance with a blanket prejudice, but with a reasoned discrimination between cases. A general evaluation expressive of a standing attitude may, whether in general or according to the context, apply either pro tanto (identifying a value or disvalue to be weighed in the balance), or prima facie (identifying a value or disvalue that on occasion disappears). On principles that are normative and explanatory, though defeasible and not exceptionless, see Irwin (2000: 10613) for a reading of Aristotle, and Lance and Little (unpublished) for a contemporary elaboration. 27 Compare what I say elsewhere (2005: 273-4) about An. III 7, 431a8-11.
25 26

WAS ARISTOTLE A PARTICULARIST?

211

to the question what situations are sufficiently similar to merit the same response. We may say that, to the extent that it is characteristic of the agent to respond in this way to this context, his response expresses an implicit conception of eudaimonia; but then having a conception of eudaimonia must irreducibly include being disposed to characteristic perceptions involving pleasure or pain. 28 To this extent, we can confirm what follows immediately after my initial quotation from Newmans Essay (269): practical wisdom, he continues, decides nothing hypothetical, it does not determine what a man should do ten years hence, or what another should do at this time. It may indeed happen to decide ten years hence as it does now, and to decide a second case now as it now decides a first; still its present act is for the present, not for the distant or the future. At the heart of Aristotles conception is his notion of practical truth. A full explanation of that would involve an analysis of one of the most condensed chapters of the Ethics (EN VI 2). This confirms the importance not only of choices in context, but of background desires. The deliberate desire that is choice (1139a23) must be distinguished from the desire that is the origin (arch) of choice, and linked to it by reasoning for the sake of something (logos hou heneka tinos, a32-3). This desire precedes any syllogizing; ideally, it is an expression of the virtue of character that makes the goal right (e.g., VI 12, 1144a7-8). What kind of desire does Aristotle have in mind when he identifies the good state (to eu) of practical thinking with practical truth, which is truth in agreement with right desire (1139a29-31)? It is surely desire in general, and thus both the choice which pursues what reasoning proposes (a24-6), and the wish which provides the reasoning with a target (a32-3). What constitutes the rightness of the wish? It is partly contextual: this goal is acceptably achievable in these circumstances. Yet the eudaimonism sets a wider background: acceptably realizing such goals contributes to the happy life, which is a way of life such that one who obtains it will have his desire fulfilled (EE I 5,

_________
28 Note that Aristotle contrasts the operation of perceptive nous with the deployment of a logos (EN VI 11, 1143a36-b1), even though a logos may well contain implicit premises (MA 7, 701a25-8). So, if we are to speak here of the agents expressing an implicit conception of eudaimonia, we must not construe this conception as the internalization of a set of propositions that may then be put to explicit or implicit use. The uncodifiability of such conceptions (a term that one may look up in the index to McDowells 1998 collection) goes much further than that: it often leaves the agent with having to rely upon how he feels about a situation as it presents itself to him on a particular occasion, since he can only in part articulate why he feels as he does. Cf. a paragraph of Dancys (2004: 149) that concludes as follows: It should be possible first to discern that an action is wrong, and only later to come to recognize what is wrong with it.

212

A. W. PRICE

1215b17-18). ). Not that eudaimonia comes of achieving whatever one desires, good or bad; rather, it consists of actions that together make up a life that is humanly both desired and desirable, a life free of the conflicts and regrets begotten by moral deficiency (EN IX 4, 1166b6-29). Good standing desires dont determine right actions by any decision procedure, and yet it is partly by reference to them that an action counts as right: acting well is the expression in context of a good ethical character (II 4, 1105a32-3; VI 2, 1139a33-4). This character is owed to a shared culture that is communicated within lasting friendships (IX 12, 1172a10-14). It is not as a product extended in time, but as conserving a constancy of character through contextual discontinuities, that a life takes on a coherent shape. Its being of a piece is not imposed upon change and chance by a procrustean design, but elicited by them from a practiced judgement whose decisions time will not translate into regrets (IX 4, 1166a27-9). The persisting agent whose life it is, with his characteristic aspirations and potentialities, enters into the rightness of his current action as much as its concrete setting. Aristotles ethics is richly agent-centered. The good man is not just the man who acts well: rather, he is the standard and measure of good action (III 4, 1113a33). It is as much for him as of him, as an ethical subject, that it is good. 29 So is Aristotle a particularist? There is no simple answer. He certainly avoids the errors of a generalism that pays too little regard to the passing contingencies of circumstance; perhaps he also avoids the errors of a particularism that regards too little the persisting humanity of agents.30
BIRBECK COLLEGE

_________
29

For related reflections, presented in a modern terminology, cf. Bakhurst (2000: 173-

6).
30 This paper has benefited, no doubt insufficiently, from the writings of Martha Nussbaum, conversations with David Wiggins, comments from Richard Kraut, and discussions at Dartmouth College and Northwestern University. The latter were hosted, respectively, by Margaret Graver and Richard Kraut, and enlivened by Elizabeth Asmis, Tad Brennan, Justin Broackes, Walter Sinnott-Armstrong, and several research students whose names I shall surely hear in future. I also enjoyed talking with my BACAP respondent, Bridget Clarke. Subsequently, I learnt much (more than can be adequately incorporated here) from a weeklong conference on particularism held, in ideally beautiful and relaxing surroundings, at Bled, Slovenia. Finally, I benefited from the restrained but pertinent comments of an anonymous BACAP referee.

COMMENTARY ON PRICE
BRIDGET CLARKE

Some of the most influential work on Aristotles Ethics in recent years has attached special importance to Aristotles scepticism about the purchase of action-guiding principles (Nicomachean Ethics I 3,1094b11-23; II 2, 1104a1-10) and his corresponding emphasis upon the proper grasp of particulars (EN II 9, 1109b14-23; IV 5, 1126a30-1126b4; VI 8, 1142a23-30; VI 11, 1143a25-b5).1 A natural way, then, to ask Was Aristotle a particularist? is to inquire how well this body of work succeeds as a reading of Aristotle. 2 Professor Price has asked the question in a somewhat different way, with a focus on whether Aristotle subscribed to the theory of moral particularism set out in the work of Jonathan Dancy. Although there are indications that Dancy would claim Aristotle as a kindred spirit, his particularism does not originate in a reading of Aristotles Ethics, so this is to ask a rather different question. 3 By not clearly distinguishing these questions, Price risks leaving the impression that the issue of Aristotles particularism turns on far narrower grounds than it actually does. For whatever distance there is between Dancys particularism and Aristotles Ethics need not extend to the relation between particularist readings of the Ethics and the Ethics itself. In what follows, then, I will try to keep the question of Aristotles particularism appropriately open by showing that the par-

_________
1 Such a reading has been most extensively developed in the work of John McDowell with many of the relevant papers appearing in McDowell (1998a). In the version of the paper Price presented at the meeting in May, he made no mention of McDowell. He has since drawn brief comparisons (perhaps as a result of our exchange); while I welcome these, I hope to make clear that they do not succeed in taking the measure of McDowells account for the issues at hand. Particularist readings of Aristotle may also be found to varying degrees in: Nussbaum (1986, 1990), Wiggins (2002a, 2002b, 2004) andat least potentially Broadie (1991). I do not mean that everyone in this group identifies his or her view as particularist, or that their views are identical in their essentials, but that all of them develop readings of Aristotle that give special prominence to the passages indicated. By Aristotles Ethics I mean both the Nicomachean and Eudemian works. 2 Irwin (2000) considers this question but, misleadingly in my view, includes Dancy in the mix with McDowell, Wiggins and Nussbaum. See the next note. 3 To my knowledge, the closest Dancy comes to claiming an Aristotelian pedigree for his own view is to endorse McDowells view as Aristotelian while aligning his own view with McDowells in select respects (Dancy 1993, esp. 50-51). He also, in a footnote, takes a view about the meaning of EN II 9, 1109b23, but only in passing; it is Ross, not Aristotle, he is seeking to explicate there (1993,107, n. 6).

214

BRIDGET CLARKE

ticularist literature on the Ethics anticipates Prices criticisms of Dancy while also furnishing what I take to be a more convincing account of practical wisdom than Price himself. I. Dancys latest work is entitled Ethics Without Principles and in it he maintains:
[M]orality has no need for principles at all. Moral thought, moral judgment, and the possibility of moral distinctionsnone of these depends in any way on the provision of a suitable supply of moral principles. This claim is what I call particularism. (2004, 5)

Price gives us good reasons to distinguish Dancys wholesale rejection of principles from what we find in Aristotle. At the same time, he distances Aristotle from the kind of generalism that is a familiar feature of the modern moral landscape. I think that Price is quite right to suggest that Aristotle occupies a middle ground between these extremes and, moreover, to identify this ground with an especially tenable form of particularism, one that gives the character of the agent an indispensable role in the exercise of practical reason. I will argue, however, that he concedes too much to the form of generalism he opposes, and that this keeps him from locating this middle ground as successfully as he might. II. The particularists concern is: to what extent and in what ways do general notions or principles, broadly speaking, inform morality? 4 Price considers the place of generalities in Aristotles Ethics in three main connections: as ends ( ) in deliberation, as exceptionless rules of conduct, and as generalizations about morally relevant features of actions. These different contexts join up to suggest the outlines of the overall place of principles in practical wisdom on Aristotles account. Price argues in favor of the idea that Aristotle allows for substantive exceptionless principles, such as Dont murder your mother. He also suggests that Aristotle admits of action-features with fixed moral valence. He naturally denies, then, that

_________
4 I am using morality interchangeably with ethics to refer to the sphere concerning how it is best for humans to live.

COMMENTARY ON PRICE

215

the phronimos brings nothing more to a situation than an acquired ability to get things right case by case. At the very least, he can bring some precepts and ethical generalizations to bear. Beyond that, and more importantly in Prices view, the phronimos ability to light on the mean is structured by standing ends of a general nature, such as caring for ones family members. These not only provide appropriate goals for action, they also focus the agents attention and ensure that what strikes him about the situation is what genuinely matters about the situation. In Prices view, it is this kind of end, in contrast to the Grand End of eudaimonia, that we should treat as the focal point of excellent practical deliberation.5 So even though, on Prices view, the phronimos does not begin from a general conception of how to live when deliberating about what to do, he is not rudderless; general standing concerns, practical precepts and ethical generalizations all play a role in ordering his thinking. At the same time, they reveal this thinking to be intimately shaped by the agents character; for ones standing ends flow from ones character. This is in many ways an attractive picture, particularly in the way that it avoids attributing to Aristotle a suspiciously modern top-down model of practical reason. 6 I think that Price is too quick, however, to give up the idea that the practically wise person deliberates from a general conception of eudaimonia above and beyond his narrower standing ends. This is not to endorse the Grand End view, but to propose a different form of resistance to it, one that would solidify the attractions of Prices picture while squaring it more directly with the central place that Aristotle gives to the achievement of an understanding of the human good within the good human life. By these lights, John Henry Newmans words would still stand as a description of phronsis, except that we would no longer exclude breadth of view, consistency in teaching and philosophical comprehension from its manifestations for fear of inordinate intellectualism. 7

_________
5 Throughout, I shall mean by end ( ) the aim or starting-point of deliberation as distinct from its issue in the form of a choice ( ). 6 As Price notes, it is preceded in some respects by Broadie (1991) and Wiggins (2002a, 2002b). 7 Newman, An Essay in Aid of a Grammar of Assent (1870), quoted by Price, p. 189. The whole passage is relevant, but particularly these lines: The authoritative oracle, which is to decide our path, is something more searching and manifold than such jejune generalizations as treatises can give. It is seated in the mind of the individual, who is thus his own law, his own teacher, and his own judge in those special cases of duty which are personal to him. It comes of an acquired habit, though it has its first origin in nature itself, and it is formed and matured by practice and experience; and it manifests itself, not in any breadth of view, any philosophical comprehension of the mutual relations of duty towards duty, or any

216

BRIDGET CLARKE

III. Aristotle cites excellence in deliberation ( ) about what conduces to the good life in general as the principal mark of practical wisdom (EN VI 5, 1140a25-28). The Grand End view represents an interpretation of this claim with two main parts: (a) the idea that a correct conception of eudaimonia figures as the end ( ) of the phronimos deliberations; (b) the idea that this conception consists in an explicit, comprehensive, substantial vision of [the human] good. . . .This blueprint of the good guides its possessor in all his deliberations, and in terms of it his rational choices can be explained and justified (Broadie 1991, 198). Now I want to suggest, taking my lead from McDowell, that we can make sense of the idea that a correct conception of eudaimonialiving well in generalfigures as the end of deliberation without supposing that it takes the form of a comprehensive blueprint; in other words, we can affirm (a) without affirming (b). 8 The point of this would be two-fold. First, it would enable us to take at face value the passages in which Aristotle certainly seems to posit living well in general as the end of deliberation [( ) EN VI 5, 1140a25-28 and Eudemian Ethics I 2, 1214b614; ( ) EN VI 7, 1141b13-15 and VI 12, 1144a31-34; ( ) Rhet. I 9, 1366b20-23 and indirectly EN I 12, 1102a2-3; ( ) VI 2, 1139b3]. 9 Second, it would enable us to see practical wisdom as incorporating a synoptic understanding of what is important in human life without imputing to Aristotle what Price terms an ethics of fantasy (3). The suggestion, in short, is that we can distinguish the following three ways to conceive of the place of a correct conception of eudaimonia in deliberation, and that the third is preferable: (C1) As the proper end of deliberation, where it is a Grand End as defined above. (C2) As a limiting factor in deliberation, but not an end proper. (C3) As the proper end of deliberation, where it is not a Grand End. It seems to me that Prices readings of the pertinent passages in 2 of his paper are perfectly consistent with the third line of interpretation but that

_________
consistency in its teachings, but it is a capacity sufficient for the occasion, deciding what ought to be done here and now, by this given person, under these given circumstances. 8 McDowell (1980) and (1996). 9 As it is Broadie, and perhaps Price as well, has to assume that a number of these passages apply only to the ideal statesman and not to the ground-level practically wise person going about his daily rounds (1991, 202-204; 1998, 305; 1999, 240-243).

COMMENTARY ON PRICE

217

he follows the second line because he falsely assumes that the only way for living well to figure as the end of deliberation is for it to figure as a Grand End. 10 Consider his reading of the opening passage of I 2 of EE. Aristotle says that everyone who can live according to his own choice should adopt some goal for the fine life an aim that he will have in view in all his actions; for, not to have ordered ones life in relation to some end is a mark of extreme folly and then urges the listener to consider carefully in which human thing living well consists (1214b7-13).11 As Price notes, this passage certainly seems to support the idea that the practically wise persons deliberations will be undertaken with a view to eudaimonia where this (or a principal component thereof) is specified in terms such as honor, wealth or virtue. 12 And Price does not actually query this; rather, he argues, plausibly enough, that the passage gives no grounds for concluding that ones conception of eudaimonia, so specified (at least in large measure), must be ones only end. He proposes instead that the end in question be thought of as a pivotal focus or priority consistent with the pursuit of many narrower ends but always capable of conditioning ones pursuit of those ends. I want to urge that nothing in this reading need fly in the face of the idea that the agent is deliberating from a general conception of eudaimonia when he pursues these narrower standing ends in his conscientious way. I would suggest that it rather points us in the direction of an alternative understanding of what it is for a conception of eudaimonia to figure as the end of deliberation. Moreover, the alternative I have in mind can readily accommodate Prices reading of EN VI 12,

_________
10 This claim may need to be softened in light of revisions Price has made to the version he read at the meeting, particularly the addition of footnote 28. But even if we take these to indicate a receptivity to C3, it would seem to be an inchoate receptivity at best, for Price has kept passages that place his view much closer to C2. Thus, for instance, he seeks to help us to resist the temptation to read EE I 2, 1214b6-14 as inviting each agent to elaborate for himself a conception of how to live that will then be the starting-point of all his deliberations (4, original emph.). This certainly reads like a rejection of idea (a) as well as idea (b), especially in view of the emphasis upon allunless we take elaborate to indicate something grand in the relevant sense. There is no ambiguity three pages later when he rejects the suggestion that a practical syllogism starts from a specification not of some goal of production, but of the goal of action, which is eudaimonia. This, he continues, is what the Grand End view supposes. That is true, but it is also, in essence, what C3 supposes. 11 Translation by Woods (1992). 12 At EE I 1, 1214a14 Aristotle seems to take it for granted that eudaimonia just is living well and at 1214a33-1214b5 he implies that it is the greatest good. He lays out these same connections more systematically in EN I 1-4, first introducing the idea that actions aim at a highest good in I 1-3 and then identifying this good with eudaimonia understood as living well and doing well at I 4, 1095a17-20 (also EE II 1, 1219b1-2).

218

BRIDGET CLARKE

1144a31-34. In this passage Aristotle speaks of practical syllogisms as having what is best ( ) as their end; on Prices reading, we should understand what is best to be situation-specific, as the Grand End could never be. The question is how a correct conception of eudaimonia could figure as the end of all the deliberations that display practical wisdom without figuring as a blueprint. To start, we might conceive of a conception of eudaimonia as a considered sense of what matters, one that can be indicated with phrases such as Live Free or Die or Activity in Accordance with the Virtues but not definitively spelled out, or codified, as the Grand End picture requires. 13 For such a sense to figure as an end of practical deliberation would be for practical deliberation to be stereoscopic in the following way: in considering what to do here and now, one would ipso facto be considering how to answer to ones sense of what is humanly important quite generally. For example, suppose one would like to reduce the amount of noise in ones apartment building. The suggestion is that in deliberating about how to secure more quiet, one would also be seeking to realize another end, that of having ones priorities straight when one decides such things, or, living well. The idea is not of course that the practically wise person leads his deliberations with the question So what would living well in general require of me here and now? but that a correct view about what living well requires is manifest in the lower-order ends he pursues and in the careful way that he pursues them. 14 In order to see why this is more than a merely verbal dispute, imagine that you pressed the practically wise person about why it is not acceptable, say, to poison the neighbors dog in order to secure peace and quiet. He might very well tell you that dogs have feelings, too, and that there are far better ways to go about this. And if you pressed him further he could say something about how, quite apart from the poor dog, this would be a terrible thing to do to the dogs owner. If we follow Price, as I understand him, we will treat these explanations as indices of further standing ends of

_________
13 Codifiable is a term of art in McDowells work meaning (of a principle or end that it is) capable of being definitively written down, in a shape suitable for deduction of particular practical conclusions (McDowell 1988, 93). 14 McDowell makes this point in terms of the practical syllogism by distinguishing between (a) a core syllogism, where the end in question is narrow, and (b) its supplement, where the end is eudaimonia, corresponding to two stages in the explanation of an action (1979, 5). I assume that an agent need not have undergone an actual process of deliberation in order for his action to be seen as issuing from deliberation (as in EN III 8, 1117a1722). See Cooper 1986, 6-10; Bostock 2000, 39-40; cf. Broadie 2002, 44-45.

COMMENTARY ON PRICE

219

the agent, such as being decent to ones fellows (and fellow-creatures), keeping ones calm in a nerveracking situation, maintaining neighborhood harmony, etc. In hearing his explanations, then, we will begin to see how his deliberation was undertaken with a view to all the other things that he cares about, and so how it is revelatory of his character. We will begin, in other words, to trace how his given end branches out to link up, in a coherent way, with all of his other ends. 15 But at this juncture it is important to ask: how does it link up coherently with all of his other endsin such a way as to avoid routine conflictif there is no overarching conception of their interrelations? And if there is such a conception, then why not allow that his deliberation was undertaken with a view to eudaimonia as he conceives it? Why banish it to the environs of deliberation? What we are getting when we ask the agent for explanation or justification are windows onto his sense of what matters in a human life quite generally. The fact that he cannot press this conception into the mold of a blueprint need not mean that he did not deliberate from it, or so I would urge. He doesnt have a Grand End, but he gets, as we say, the big picture. And it would seem that this, the big picture, is precisely what Aristotle wants to keep in play as a kind of focal point when he claims in EN VI 7 that the person who is without qualification the good deliberator is the one whose calculations make him good at hitting upon what is best for a human being among practicable goods ( 1141b13-15). 16 IV. Price is likely to object to this account on the grounds that, for Aristotle, a of deliberation has to be articulable enough to permit a survey of its causal relations (else there would be no way to arrive at the available means). 17 As Broadie puts it: what can there be to deliberate about when there is no articulable end? (1991, 198). 18 So Price could very well ac-

_________
I borrow this apt phrasing from Richardson 1992, 358. All translations from the EN are from Rowe (2002). For Rowes among practicable goods ( ), Ross (ROT) has of things attainable by action and Irwin (1999) that is achievable in action. 17 In EN III 3 Aristotle characterizes deliberation in part as the search for means to ends that are within ones power to bring about. Thus we do not deliberate about things brought about purely by chance or by natural processes. 18 Broadies principal ground for concluding this is the fact that Aristotle likens deliberation to geometrical analysis, where one starts with a definite problem (EN III 3, 1112b2015 16

220

BRIDGET CLARKE

cept the identification of one's conception of eudaimonia with an uncodifiable sense of what matters, but claim that this equation only offers all the more reason why such a conception could not figure as an end of deliberation, although it might inform deliberation in some other way. 19 At the same time, Price has made clear that the practically wise persons ends need not be short term or merely circumstantial on his account. They can be something in between the ad hoc end of the sea-captain in a storm who suddenly finds himself needing to jettison his cargo and the everpresent end of living well, however one conceives it. We may think of periodic ends, such as writing a book which influence an agent through a stretch of his life as well as life-long goals, such as the pursuit of science (6). What distinguishes an agent as practically wise is the adroitness with which he pursues these less than grand ends. What does this restriction amount to? Lets grant that one can deliberate toward comparatively narrow ends without anything in particular following about what it means to deliberate. In that case, the narrowing of the scope of phronsis to limited ends need not entail radical restrictions on the kind of thought represents. However, this narrowing of scope does lead to such restrictions when it is coupled with a very familiar view of the division of labor between virtue of character (virtue for short) and phronsis. (I take it that the conjunction of these positions is not fortuitous though I cannot explore this important point here.) And once this more circumscribed view of what it is to deliberate is in place, we end up with a conception of phronsis that looks suspiciously akin to a modern instrumentalist conception of reason. Alternatively, if we take it that deliberation proceeds from a conception of eudaimonia and we reject the blueprint picture of a correct conception of eudaimonia, we are swiftly led to a richer notion of deliberation and therewith a conception of phronsis that is arguably more compelling both textually and philosophically. At stake, then, is the proper understanding of Aristotles conception of practi-

_________
25 and VI 8, 1142a28-29; Broadie 1991, 236-237). She cites further supporting passages in Broadie 1998, 306. 19 The assumption I am steering clear of here is made by Richard Kraut when he says, in a thoughtful review of Broadie, A practically wise person, on her reading, does not need a conception of happiness in order to make good decisions (Kraut 1993, 362). (This would be a fourth possible viewto add to those enumerated in III aboveaccording to which a conception of eudaimonia plays no role in good deliberation.)

COMMENTARY ON PRICE

221

cal wisdom and the potential of this conception to provide an alternative to an instrumentalist conception of practical reason. 20 Let me sketch brieflyand tentativelyhow I think Price arrives at what I take to be an overly narrow view of phronsis. The sketch is only tentative because it concerns issues in the background of Prices piece that are, however, essential to the question of Aristotles particularism. Even if it should turn out to be wide of the mark, then, it will serve to get decisive matters on the table. On Prices view, as I have understood it, deliberation must be toward an end that is narrower than eudaimonia because only such an end can have sufficiently determinate causal relations to make a search for means a viable project. So far, this view is consistent with the search for means taking a variety of forms. In particular, we might follow Wiggins and McDowell in distinguishing between means-end calculation, rule-case reasoning and specification of an end. 21 In one sense, all deliberation has a means-end form (EN III 3). The question is how to understand this form, and means-end calculation is one way. Means-end calculation is the determination of means to ends that are settled in the sense that it is clear what it would be for them to be realized; there is no question, to take an example from Aristotle, what it would be to have a coat. 22 Similarly, in rule-case reasoning, one subsumes a particular instance under a general rule (Keep to the sidewalk) where it is clear in advance what it would be to have an instance of the rule (or its violation). In contrast, specification of an end refers to the determination of what a given end amounts to; in this case, the uncertainty about achieving the end is primarily an uncertainty about what it would be for the end to have been achieved. 23

_________
20 By instrumentalist I mean a conception according to which practical reason is both goal-directed and divisible into separate cognitive and desiderative elements. Cf. Vogler (2002), ch. 1. 21 Wiggins (2002a) and McDowell (1998b). 22 There is of course component (or constitutive) means-end reasoning as well the instrumental species (Greenwood 1908, 46-48). Wiggins (2002a) more or less equates the former with specification while McDowell cautions against this (1998b, 33, n. 19). The point I wish to make can, mutatis mutandis, accommodate either view of the matter. 23 If, for example, my end is resting and I know that in order to feel rested I must be in a darkened room with my feet off the ground and an aloe compress over my eyes for at least ten minutes, then all of my deliberations to this end will concern how to obtain these conditions. But if I dont initially know what I need to do to feel rested (compare: napping, reading a novel, listening to Hank Williams, dinner with friends) then my deliberations will center on the question what would resting here and now be?

222

BRIDGET CLARKE

It is important to note that conceptualizing deliberation as specification implies that deliberation is as much a process of setting ends as it is of finding means to them. 24 To specify what a general end (resting) amounts to is, in effect, to set more detailed ends (e.g., getting a hammock, recovering the Hank Williams CD one loaned out) at which to aim, until one arrives at what Aristotle calls the last thingthe action itself that will satisfy the description resting in these particular circumstances (III 3, 1112b16-25). 25 Since the conceptualization of deliberation as specification presupposes that deliberation entails the setting of ends, one cannot allow for this way of thinking about deliberation if one considers the setting of ends to be the prerogative of virtue as distinct from phronsis. And this is just what Price appears to do. Although he does he not explicitly endorse this division of labor, it

_________
24 This is bound to appear problematic since Aristotle makes clear both that deliberation presupposes an end and that it seeks to ascertain that which is toward the endas distinct from the end itself (EN III 2, 1111b27; III 3, 1112b11ff.; III 5, 1113b3). Thus, he says, a politician does not deliberate about whether he will produce good public order but about how he will do this (1112b14). While this sounds decisive, Aristotle then goes on to note that one will typically need to find means to the means one has settled on (1112b18-19), and this implies that deliberation typically involves the setting of some endsfor where there is a chain of interlinked ends, the means to one end will itself usually constitute a further end. It might seem that this does not really tell against the restriction of deliberation to a search for means, however, on the grounds that it assigns to deliberation the setting of merely secondary ends, i.e., ends which are purely instrumental to the achievement of primary ends. But I would argue that this is to assume in advance what is in question, viz. the possible shapes of deliberative thought. If we conceive of the setting of secondary ends as an exercise in specification the activity in question need not consist of merely instrumental calculation; it could involve seeing that breaking the law, say, is what the given end of living well requires in these circumstances, or again, given the end of civil disobedience, seeing that this way of breaking the law is far more suitable than the alternatives. But note that the distinction between primary and secondary ends begins to dissolve from this point of view, for lower order ends represent specifications of higher order ends and so on up the chain until one reaches the end of eudaimonia (which Aristotle seems to see as a kind of condition of rational choice as such). I do not mean to suggest, however, that there is nothing to be made of Aristotles repeated association of deliberation with means as distinct from ends, and in n. 31 I give some indication of how one might make sense of this association consistent with the account I am sketching. See also the next note. 25 As this example shows, to conceive of deliberation as specification is not to rule out means-end calculation or rule-case reasoning; it is to suppose that, as Wiggins puts it, the whole interest and difficulty of the matter is in the search for adequate specifications, not in the technical means-end sequel or sequels (2002a, 225). The plurals remind us that any actual deliberation will likely involve numerous sequences of one or the other type of thoughtwith, for instance, one specification leading to further specifications and chains of means-end reasoning interwoven throughout.

COMMENTARY ON PRICE

223

is central to his critique of Dancy and generalism alike in 5 that virtue sets the practically wise persons ends (see also n. 9 in Price). This is how he proposes to insure that character is kept in play with respect to deliberation and choice. To the extent that one accepts this division of labor, deliberation will be confined to means-end calculation or rule-case reasoning, i.e., a search for means where it is clear in advance what it would be for the end to be realized. This does not mean that the kinds of sensitivity to ones circumstances that figure in specification and enable one, crucially, to know which ends would be appropriate to pursue here and now will not have any place in Prices picture, but it does mean that they will not be part of phronsis proper. These kinds of sensitivity will, according to Price, need to be supplied by virtue: 26
Virtues roles are multiple and inter-connected. It supplies the agent with distinctive standing concerns to supplement the common-garden concerns that we all carry around with us. It lends salience to certain aspects of a situation (those which invite, or endanger, the realization of his distinctive concerns). It determines that the goal selected is both generally acceptable, and acceptably pursuable in the situation. 27

My concern is that this would appear to leave nothing for practical wisdom to do but to light on the best means to ends that are set independently by virtue. To this extent, the kinds of discernment that mark out the virtuous person as morally acute versus merely clever will fall outside the purview of practical wisdom. 28

_________
26 Fortenbaugh (1969) sets out this basic view in opposition to a rationalistic reading of the Ethics. He provides a summary of the 19th and early 20th century debate about the respective roles of virtue and phronsis on 177, n.2. 27 I quote from the version of Prices essay given to me as final. It has just come to my attention that in the version to be published Price has changed the first sentence of this passage to The roles of the virtues of character and of practical thinking are multiple and interconnected (p. 16, emph. added). I am not sure what to make of the addition of practical thinking here. It could indicate, contrary to what I have been suggesting, a lack of sympathy for the means-ends division of labor and a corresponding receptivity to the notion of deliberation as specification. In that case, it is difficult to see why Price wont allow that a conception of eudaimonia can serve as a telos of deliberation. I think it is more likely that the addition is meant simply to stress the interdependence of the tasks assigned to virtue, on the one hand, and phronsis, on the other, while leaving the division of labor between them relatively intact. I discuss a subtle version of this position in the next note. 28 A more nuanced account of the division of labor Price probably has in mind is provided in Broadie 2002, 49-50. Here virtue decides on the merits of pursuing a given end in the light of the consequences of so doing, as these are revealed by phronsis. Accordingly, an agents search for means can result in the modification or even rejection of the end that

224

BRIDGET CLARKE

It must be admitted that this division of labor is consistent with wellknown passages in which Aristotle credits virtue with making the ends correct and phronsis with making the means to the end correct (EN VI 12, 1144a7-9; VI 13, 1145a5-6). In turn, these passages fit with Aristotles claim that deliberation concerns , not ends themselves. However, there are passages that make real trouble for this understanding of the relation between virtue and phronsis, starting with Aristotles definition of virtue in terms of practical wisdom (EN II 6, 1106b361107a2); it is not clear why virtue would be a state that chooses with reference to (a standard set by) the practically wise person if the latter were merely good at calculating the means to ends set independently by virtue. In addition, Aristotle seems n at least one occasion to equate practical wisdom with a true conception of the end (VI 9, 1142b31-33). 29 Taken together, these suggest a far wider scope to practical wisdom and the activity of deliberation at which it excels than Prices reading allows. Within this broader scope there may well be room for deliberation to take as its end nothing less than living well itself while maintaining a distinctively particularist alternative to the Grand End picture. Indeed, this more expansive view of deliberation is at the heart of a version of particularism that suffers none of the defects Price finds in Dancys version. V. The key to the alternative version of particularism lies in the idea that an end may figure in deliberation even though it is not possible to formulate the end definitively in advance of its application. This allows for the midlevel standing ends Price highlights, as well as for a more general end

_________
initially set deliberation in motion. There is a sense in which ends are no longer set independently by virtue on this account, since in evaluating hypothetical ends virtue relies upon factual information provided by phronsis. But, if I have understood Broadie, it remains the case that the work of making crucial moral assessments is for virtue alone; given the question how can I achieve end y? the capacity designated phronsis delivers the same (factual) report that the capacity designated cleverness ( ) would deliver. The basis for the different designations is that in the former case, the report will go on to be assessed in the light of exacting moral standards while in the latter case it will not. It seems to me there is a significant respect in which phronsis understood in this way does not rise above cleverness. 29 It is possible to read this passage as reaffirming the assignment of phronsis to means only, but that is not the most plausible reading. See Irwin 2000, 249; Kenny 1979, 106-107; McDowell 1998b, 31, n. 17; cf. Broadie (2002), 376.

COMMENTARY ON PRICE

225

above and beyond them that explains their interconnections. But it conceives both orders of ends in such a way that the ability to see what would count as realizing them in any given case may well require a special sensitivity to particulars, the kind of sensitivity Aristotle seems to identify with practical wisdom when he likens it to a perceptual capacity (EN VI 8, 1142a23-30; VI 11 1143a25-b5 and 1143b13-14). On the particularist account I have been sketching, it is this capacity that enables the phronimos to read from the specifics of his situation which standing end is appropriate to pursue now, and what would count as best realizing that end; this just is the capacity to deliberate well. There is no blueprint here, but there is a synoptic vision directing the specific choices, manifest in the way one reads the details of situations. Of course the phronimos will almost always have a relatively narrow end in view, but these ends will be activated (when they are, in the way that they are) in the light of his conception of eudaimonia. They will, more precisely, be means to-cumspecifications of that final end. This view will make sense only if we can speak of an agent deliberating toward a conception of eudaimonia even when it is not possible to identify the content of that conception without recourse to the particular applications the agent makes of it (McDowell 1996, 5). This is the natural counterpart to the thesis of specification, for where deliberation toward an end is a matter of inquiring into what it would be for the end to be realized, there is no way to specify the end except through its application.30 It should be emphasized that deliberation construed along these lines is keyed to the character of the agent; the uncodifiability of the end leaves it to the agent to make the relevant specifications in the very way that he conceives his circumstances, based on the kind of person he is. So virtue of character and practical wisdom do not operate independently on this alternative, with the one setting the ends for the other to pursue. They

_________
30 In other words, there are different kinds of distance between end and action that deliberation can bridge, and an independently articulable end speaks to only one of those kinds of distance. Anscombe put the idea this way: The mark of practical reasoning is that the thing wanted is at a distance from the immediate action, and the immediate action is calculated as the way of getting or doing or securing the thing wanted. Now it may be at a distance in various ways. For example, resting is merely a wider description of what I am perhaps doing lying on my bed; and acts done to fulfill moral laws will generally be related to positive precepts in this way; whereas getting [into power] the good government is remote in time from the act of pumping, and the replenishment of the house water-supply, while very little distant in time, is at some spatial distance from the act of pumping (Intention, 41, second emphasis added). I have obviously exploited her example of resting in n. 23 above.

226

BRIDGET CLARKE

rather represent different aspects of the same ability to respond appropriatelyin the right way, for the right reasonto whatever happens (EN VI 8, 1142a23-30 in conjunction with VI 9, 1142b31-33 and X 8, 1178a1620). 31 Accordingly, it would be a serious mistake to read the particularist interpretation of Aristotle as advancing the kind of exclusively cognitivist picture that worries Price in 5 (p. 16). Since Price directs the worry at two remarks by Dancy (quoted on p. 14 of Price), this may seem unremarkable. But the more extended of Dancys remarks acknowledges the influence of McDowell in a footnote, and its wording closely resembles McDowells claim that, for Aristotle,
there is nothing for grasp of the content of the universal to be except a capacity to read the details of situations in the light of a way of valuing actions in which proper upbringing has habituated one. (McDowell 1996, 23)

In fact, I think it is highly unlikely that Dancy himself was advancing the kind of rationalistic picture that worries Price in the passage that he examines. 32 For in that passage, the broad range of sensitivities that enable

_________
31 McDowell 1998b, esp. 6 and 11. One might worry that this picture fails to find an appreciable role for virtue and so to make good sense of Aristotles distinction between virtue and phronsis. There are at least two points that Aristotle could be making (consistent with this account) when he advances the means-end division of labor. In the first place, he could be stressing that one cannot have phronsis (understood as the capacity to discern what living well requires here and now) unless one has had the requisite upbringing; this would be to gloss virtue sets the ends as we acquire correct ends only with the right kinds of practice in our formative years. This blocks the thought that mere argument or an exercise of spontaneous rational intuition could account for the development of phronsis (ibid., 31-32). These passages could equally well be read (in conjunction with VI 5, 1140b11-20 and VI 12, 1144a36) as underlining that nothing less than the ongoing practice of virtuous acts is necessary to preserve ones grasp of the correct ends (Sorabji 1980, 213). It is noteworthy that Aristotle emphasizes both of these points in the final chapter of the EN (X 9) stressing that nothing short of habituation through proper upbringing can bring about goodness and that nothing short of continued habituation to good deeds through proper legislation can insure its preservation. Both points suggest that practical excellence is the outcome of a lifelong process of development that is at once desiderative and intellectual. Burnyeat (1980) elaborates this developmental view. Cf. Cooper 1996, esp. IV and Vasiliou 1996. 32 Namely: Our account of the person on whom we can rely to make sound moral judgements is not very long. Such a person is someone who gets it right case by case. To be so consistently successful, we need to have a broad range of sensitivities, so that no relevant feature escapes us, and we do not mistake its relevance either. But that is all there is to say on the matter. To have the relevant sensitivities just is to be able to get things right case by case. The only remaining question is how we might get into this enviable state. And the answer is that for us it is probably too late. As Aristotle held, moral education is the key;

COMMENTARY ON PRICE

227

one to get things right case by case are the precipitate of a moral education as conceived by Aristotle. And we know that the outcome of a successful moral education, in Aristotles view, just is good character. This suggests that Dancys point is that we should equate good character with such sensitivities, not that such sensitivities might exist in the absence of good character. If my reading of Dancy is correct, then (contra Price) nobody on the particularist side of things is suggesting that good judgment proceeds in the absence of character; the question at issue is whether there is anything but character, with its general standing concerns, to direct judgment. 33 VI. In contrast to Dancys version of particularism, the version I have sketched makes room for character and for principles that offer guidance from without in its account of judgment. It does this by supposing that the consistently correct application of a principle (that of eudaimonia, in the final instance) requires the correctly formed dispositions that comprise good character. In the space that remains, I want to indicate briefly how this form of particularism can readily build on Prices arguments to elucidate the place of exceptionless principles and features of unvarying valence in the Ethics. Along the way, I shall make some further observations about Prices analysis. Although Aristotle holds that substantive practical principles hold only for the most part ( ) (EN I 3, 1094b21), he does seem to proscribe adultery, theft and murder absolutely (EN II 6, 1107a8-15). Price argues for the idea that these prohibitions are not tautological on the grounds that the extension of a concept such as murder is informed by social attitudes (about, e.g., mitigating circumstances) that may be unwarranted. Accordingly, on Prices view, when Aristotle endorses these principles, he is in fact endorsing the social attitudes that have fixed the exten-

_________
for those who are past educating, there is no real remedy (Dancy 1993, 64, quoted by Price, p.14). 33 If this is right, it means that when Dancy rejects the idea that principles have any role to play in moral thought, he is working with a narrower conception of principle than simply a general notion (which I introduced in II above). I take it he has in mind something akin to general propositions that specify morally relevant considerations across contexts. (See Dancy 1993, 66-72; 2004, 76.) It is with this more familiar conception of moral principle that I will be working in the next section.

228

BRIDGET CLARKE

sion of the concept just where it has been fixed. He is saying not merely murder is wrong but murderunderstood as the taking of life in circumstances x, y, and zis wrong. Only the former is a tautology. Since Aristotle does not make use of the notion of a description under which an action can be evaluated, this is a possible interpretation of his meaning. But I see virtually no evidence that Aristotle actually takes this view of these principles. A review of his words in context may be helpful. He has just defined virtue as a mean state and is warning (those looking for a pretty big loophole, it would seem) that we should not conclude from his definition that all actions admit of a mean.
Not every action admits of intermediacy, nor does every affection; for in some cases they have been named in such a way that they are combined with badness from the start, as e.g. with malice, shamelessness, grudging ill-will, and in the case of actions, fornication, theft, murder; for all these, and others like them, owe their names to the fact that they themselvesnot excessive versions of them, or deficient onesare bad. It is not possible, then, ever to get it right with affections and actions like these, but only to go astray. (EN II 6, 1107a8-15)

He seems, clearly, to be affirming that certain kinds of action and affect really are ill-advised in all circumstances and that this is built into the very concepts of them; I see no implication one way or the other about whether the dominant views about what counts as murder, etc., should be accepted. Presumably, Aristotle thinks these viewslike the endoxa generally should be accepted only insofar as they hold up under the distinctive kind of critical scrutiny he applies throughout the Ethics. What are the implications of interpreting these prohibitions to be as tautologous as they appear? Prices comments suggest that such an interpretation may encourage one to attribute to Aristotle some implausible views, including the view that moral judgments must always be difficult. But this view need not follow, and it is not a commitment of the particularist reading of Aristotle that I have defended. What that reading is committed to, I would suggest, is two-fold. First is the idea that even absolute principles may require discrimination for their correct application on occasion. 34 Thus even if, by Aristotles lights, murder is always wrong, it does not follow that it is always easy to know when something counts as murder, theft, or adultery (however we see the role of convention in fixing the ex-

_________
34 McDowell makes the further point that, in Aristotles view, absolute principles on their own could never suffice to account for the virtuous persons knowledge of how to act in particular cases (1998b, 27, n. 9).

COMMENTARY ON PRICE

229

tensions of the terms). 35 (Its hard to say, for instance, whether Euthyphros father murdered the servant.) Second is the idea that the requisite discrimination will often depend on excellence of characterit wont be something to which the nasty person, however clever, can lay claim. It is significant that this way of thinking about principles gives us no reason to restrict cases which may require special discernment to the sphere of the merely personal, as Newman does when he states that the individual must be his own law and judge in those special cases of duty which are personal to him (quoted in Price, 1). 36 But this line of thought does not thereby support the idea that, as Price puts it, all policies and decisions, negative and positive, are difficult (11). It suggests rather that hard cases can crop up anywherein contexts of justice as well as friendshipand that when they do, ones ability to judge will be conditioned for better or worse by the character one has. 37 Thus the definitive measure of correctness in moral matters is the good man ( ) (EN 1 8, 1099a22-24; III 4, 1113a31-33; X 5, 1176a15-19). Now consider moral variability. Dancy claims that the same feature can count for or against an action depending upon the circumstances; this is the thesis of reason holism (1993, ch. 4). 38 Accordingly, the fact that an action causes pleasure can be a reason not to undertake it, and the fact that an action causes pain does not necessarily tell against it, even when its victim is innocent and unwilling (1993, 65-66; 2004, 208-209). Price makes the point that for Aristotle pleasures do vary in their valence, but only insofar as they are indexed to activities of distinct but fixed valence. 39 This means, as he puts it, that once we have described some good and enjoyable activity sufficiently to identify what pleasure is taken in it by the agent, that pleasure will have invariant valence (12). This sounds right, but the particularist will hasten to point out that it does nothing to fix the value of a given pleasure for deliberative purposes. For one, the consistently pro valence of an activity and its supervening pleasure cannot imply that one should always perform actions that embody that activity. It must depend on whether it is good to be undertaking this

_________
Cf. EN V 9, 1137a5ff. See n. 7 above for more of the passage. EN V 10 suggests that Aristotle does not see issues of justice as conforming to universals anymore than anything else in the sphere of action so conforms. 38 Elsewhere Dancy concedes the possibility of features of invariant valence with the proviso that the invariance be understood as an incidental, not logical, fact about such features and the reasons they constitute (2000, 136-137). 39 Aristotle attributes to activities different levels of purity at EN X 5, 1176a1-3.
35 36 37

230

BRIDGET CLARKE

activity here and now. 40 The pleasures of contemplation are best, but they should probably not be enjoyed at a meal. 41 And according to Aristotle, it is no easy matter to know when ones activities are genuinely good in the first place. 42 We think they are good because they give us pleasure; pleasure is an apparent good (EN III 4, 1113a35-b2; EE II 10, 1227b3-4; MA 700b29, cf. EE VII 2, 1235b26-29). By nature, then, we are not impartial judges of the activities we enjoy. This vulnerability in our make-up leads Aristotle to formulate some stringent practical principles in EN II 9: we should consider the things we enjoy and then drag ourselves away in the contrary direction, the way people try to straighten warped pieces of wood (1109b5-6). Even more emphatically, he urges his audience to regard pleasure as the elders of Troy regarded Helen and repeat on every occasion what they uttered (1109b9-11). 43 He then follows this advice with the famous caveat that the right relationship is not easy to achieve, for such things depend on the particular circumstances, and the judgment of them lies in our perception (1109b24). Thus qualified, the notion that Aristotle assigns invariant valences to pleasures poses no threat to the particularist reading of Aristotle. To do that, one would need to provide evidence that these fixtures simplify the deliberative search in such a way as to obviate the need for a special sensitivity to the specifics of ones situation, the kind of sensitivity that will elude the nasty person, however clever she might be.

_________
40 In EN X 5 Aristotle notes that pleasures which are proper to one activity may impede the performance of another activity to which a different set of pleasures are proper. Thus lovers of pipe-music are incapable of paying attention to a discussion if they happen to hear someone playing the pipes. . . So the pleasure in pipe-playing destroys the activity of discussion (1175b3-5). It seems clear from the example that the pleasures that impede a good activity may be good in themselves. 41 How one develops this line of thought will obviously affect how one makes sense of EN X 6-8. 42 Aristotle distinguishes between pleasures that are good without qualification ( ) and those that are good only under certain conditions ( ) (EN VII 12, 1152b27). See Annas (1980) and Gottlieb (1991). 43 Terrible is the likeness of her face to immortal goddesses/Still, though she be such, let her go away in the ships, lest/she be left behind, a grief to us and our children (Iliad III, 158-160).

COMMENTARY ON PRICE

231

VII. Price has defended a conception of practical wisdom in Aristotles Ethics that seeks to incorporate the insights of both particularism and generalism while avoiding their respective pitfalls. My critical suggestion has been that a modified answer to the first of the four questions he raises can help to bring into view a sense of particularism that is well developed in the literature on the Ethics and that complements the essentials of his answers to the other three questions. At the same time, it allows us (contra Newman) to credit the phronimos with a breadth of view, a consistency in outlook and a philosophical comprehension of moral requirements that is surely a desideratum of any practical agent. 44 To locate that comprehensive understanding, as Price does, in a separately conceived virtue (viz., of character) is, I have suggested, to accept a markedly narrow view of deliberation and its signature excellence. Put another way, it is to miss an opportunity to see how an excellence of reason in the practical sphere might encompass far more than instrumental reasoning. The phronimos does not need a fantastic master plan that anticipates all of lifes vicissitudes, but he does need a general understanding of the human good if his ability to act well whatever the circumstances is not to seem mysterious. The challenge is to conceive of that understanding properly, with a sensitivity textual and philosophical. That challenge is both the origin and promise of a truly Aristotelian particularism. 45
WILLIAMS COLLEGE

_________
44 In keeping with EN I 3 I take it that the philosophical comprehension at issue would not involve anything more theoretical than Aristotle provides in the Ethics and might well involve something less. 45 I am grateful to Sarah Broadie, Paul Muench, Amlie Rorty, Jennifer Whiting and James Wilberding for very helpful comments on earlier drafts of this paper. I am also indebted to Sarah Broadie and John McDowell for valuable conversation about these topics.

PRICE/CLARKE BIBLIOGRAPHY Primary Texts

Barnes, J. (ed.) 1984. The Complete Works of Aristotle: The Revised Oxford Translation. Princeton. Broadie, S. (com.) and Rowe, C. (tr.) 2002. Aristotle, Nicomachean Ethics. Oxford. Bywater, I. (ed.) 1894. Aristotelis, Ethica Nicomachea. Oxford. Irwin, T. (tr. & com.) 1999. Aristotle, Nicomachean Ethics, 2nd edn. Indianapolis. Jaeger, W. (ed.) 1957. Aristotelis, Metaphysica. Oxford. Nussbaum, M.C. (ed. & com.) 1978. Aristotle, De Motu Animalium. Princeton. Ross, W.D. (tr.) 1925. Aristotle, Nicomachean Ethics. Oxford. _______. (ed. & com.) 1961. Aristotle, De Anima. Oxford. Slings, S.R. (ed.) 2003. Platonis, Respublica. Oxford. Walzer, R.R. and Mingay, J.M. (eds.) 1991. Aristotelis, Ethica Eudemia. Oxford. Woods, M. (tr. & com.) 1992. Aristotle, Eudemian Ethics, Books 1, 2, and 8, 2nd edn. Oxford.

Secondary Literature
Annas, J. 1980. Aristotle on Pleasure and Goodness. In Rorty (ed.): 285-299. Anscombe, G.E.M. 1957. Intention. Oxford. _______. 1965. Thought and Action in Aristotle. In Bambrough (ed.): 143-158. Austin, J.L. 1967. Agathon and Eudaimonia in the Ethics of Aristotle. In Moravscik (ed.): 261-96. Bakhurst, D. 2000. Ethical Particularism in Context. In Hooker and Little (eds.): 15777. Bambrough, R. (ed.) 1965. New Essays on Plato and Aristotle. London. Bostock, D. 2002. Aristotles Ethics. Oxford. Broadie, S. 1991. Ethics in Aristotle. New York. _______. 1998. Interpreting Aristotles Directions. In Gentzler (ed.): 291-306. _______. 1999. Aristotles Elusive Summum Bonum. Social Philosophy and Policy 16(1): 233-251. Burnyeat, M. 1980. Aristotle on Learning to be Good. In Rorty (ed.): 69-92. Cooper, J. 1996. Reason, Moral Virtue, and Moral Value. In Frede and Striker (eds.): 81-114. _______. 1986. Reason and Human Good in Aristotle. Hackett. First published in 1975, Harvard. Dancy, J. 1993. Moral Reasons. Oxford. _______. 2004. Ethics Without Principles. Oxford. _______. 2000. The Particularists Progress. In Hooker and Little (eds.): 130-156. Oxford. _______. (forthcoming). Defending the Right. Journal of Moral Philosophy. Engstrom, S. and Whiting, J. (eds.) 1996. Aristotle, Kant, and the Stoics. Cambridge. Foot, P. 2001. Natural Goodness. Oxford. Fortenbaugh, W.W. 1969. Aristotle: Emotion and Moral Virtue. Arethusa 2(1): 163185. Frede, M. and Striker, G. (eds.) 1996. Rationality in Greek Thought. Oxford. Gentzler, J. (ed.) 1998. Method in Ancient Philosophy. Oxford. Gill, C. (ed.) 2005. Virtue, Norms, and Objectivity. Oxford. Gottlieb, P. 1991. Aristotle and Protagoras: The Good Human Being as the Measure of Goods. Apeiron 24(1): 25-45.

PRICE/CLARKE BIBLIOGRAPHY

233

Greenwood, L.H.G. 1909. Aristotle: Nicomachean Ethics Book Six. Cambridge. Hooker, B. and Little, M.O. (eds.) 2000. Moral Particularism. Oxford. Irwin, T.H. 2000. Ethics as an Inexact Science: Aristotles Ambitions for Moral Theory. In Hooker and Little (eds.): 100-129. Kenny, A. 1979. Aristotles Theory of the Will. London. Kraut, R. 1993. In Defense of the Grand End. Ethics 103: 361-74. Lance, M. and Little, M. (unpublished). Where the Laws Are (accessible through Mark Lances website). McDowell, J. 1979. Virtue and Reason. The Monist 62(3): 331-350. _______. 1980. The Role of Eudaimonia in Aristotles Ethics. In Rorty (ed.): 359-376. _______. 1988. Comments on Some Rational Aspects of Incontinence. The Southern Journal of Philosophy Supplement 27: 89-102. _______. 1996. Deliberation and Moral Development in Aristotles Ethics. In Engstrom and Whiting (eds.): 19-35. _______. 1998. Mind, Value, and Reality. Harvard. _______. 1998. Some Issues in Aristotles Moral Psychology. In his Mind, Value, and Reality, Cambridge, Mass.: 23-49. Moravcsik, J.M.E. (ed.) 1967. Aristotle. New York. Mller, A.W. 1977. Radical Subjectivity: Morality versus Utilitarianism. Ratio 19: 115-32. _______. 2004. Aristotles Conception of Ethical and Natural Virtue: How the Unity Thesis sheds light on the Doctrine of the Mean. In Szaif and Lutz-Bachmann (eds.): 18-53. Newman, J.H. 1870. An Essay in Aid of a Grammar of Assent. London. Nussbam, M. 1986. The Fragility of Goodness. Cambridge. _______. 1990. Loves Knowledge. Oxford. Pakaluk, M. (unpublished). Are there Moral Absolutes in Aristotles Ethics? Price, A.W. 2000. On Criticising Values. In Philosophy, supplement 47: 141-58. _______. 2005. Aristotelian Virtue and Practical Judgement. In Gill (ed.): 257-78. Richardson, H. 1992. Review of S. Broadie, Ethics with Aristotle. Mind 101(402): 358361. Rorty, A.O. 1980. Essays on Aristotles Ethics. Berkeley. Sorabji, R. 1980. Aristotle on the Role of Intellect in Virtue. In Rorty (ed.): 201-219. Szaif, J. and Lutz-Bachmann, M. (eds.) 2004. Was ist das fr den Menschen Gute? Berlin. Vasiliou, I. 1996. The Role of Good Upbringing in Aristotles Ethics. Philosophy and Phenomenological Research 56: 771-797. Vogler, C. 2002. Reasonably Vicious. Harvard. Watson, G. 2004. The Work of the Will. In his Agency and Answerability, Oxford: 123-57. Wiggins, D. 2002a. Deliberation and Practical Reason. In his Needs, Values, Truth, amended impression of 3rd edn, Oxford: 215-37. _______. 2002b. Weakness of Will, Commensurability and the Objects of Deliberation and Desire. In Needs, Values, Truth: 239-67. _______. 2004. Neo-Aristotelian Reflections on Justice. Mind 113: 477-512. Woods, M. 1986. Intuition and Perception in Aristotles Ethics. Oxford Studies in Ancient Philosophy 4: 145-66.

COLLOQUIUM 7

IN DEFENSE OF INNER SENSE: ARISTOTLE ON PERCEIVING THAT ONE SEES

THOMAS K. JOHANSEN

Inner sense theory holds that we are conscious of our own perceptions by virtue of further perceptions which have those first perceptions as their object. Such second-order perception is understood in parallel with firstorder perception, with the difference that whereas first-order perception is (normally) directed at external objects, second-order perception has as its object ones own perceptual states. By calling the theory one of inner sense it is implied that the faculty responsible for second-order perception is itself perceptual, rather than, say, intellectual. 1 Aristotle, Locke and Kant are traditionally taken to be inner sense theorists, while more recently David Armstrong, William Lycan, and Hugh Mellor have advocated similar views. 2 As an interpretation of Aristotle, inner sense theory has been criticized in recent years, most notably by Aryeh Kosman and Victor Caston. 3 In this paper, I present a new reading of the central passage at DA III.2 425b12-25 which supports a version of inner sense theory, while countering some of these criticisms. Setting the Context for DA III.2 In Book II, Chapter 2 of the DA, Aristotle sets out the agenda for his psychology. He points out that the soul is the principle of life. There are, however, different ways for different living beings to be alive. Some living beings have some capacities or faculties (dunameis) which others do not. We need therefore to understand the different faculties first in order to

_________
1 It may also be thought that inner sense theory claims the inner sense is distinct or separate from the outer senses (sight, hearing, and so on). In my interpretation of Aristotle, I shall not take this claim to be an essential element of inner sense theory: as we shall see the sense responsible for second-order perception and the sense responsible for first-order perception are not separate. 2 D. Armstrong, What is Consciousness, in The Nature of Mind, Ithaca 1981: 55-67; W. Lycan, Consciousness and Experience, 1996; H. Mellor, Conscious Belief, Proceedings of the Aristotelian Society 78 (1978) 87-101. 3 L. A. Kosman, Perceiving That We Perceive: On the Soul III, 2, The Philosophical Review 84 (1975) 499-519; V. Caston, Aristotle on Consciousness, Mind 111 (2002) 751815.

236

THOMAS K. JOHANSEN

understand the different kinds of soul. The rest of the DA consequently accounts for the faculties that distinguish the major life forms: nutrition, sense-perception, locomotion and the intellect (nous). First, nutrition is tackled in Book II.4, then each of the five senses receives a chapter in the rest of Book II. In Book III.4-11 Aristotle accounts further for the capacities of intellect (III.4-8) and locomotion (III.9-11). The work ends with two chapters (III.12-13) on why the various faculties belong to different kinds of living being. Books II and III thus seem to be structured according to the plan of explaining each of the capacities that are characteristic of the different kinds of living being, with a conclusion showing why these capacities are required or beneficial for those kinds of being. I have not yet mentioned the first three chapters of book III of the DA. How do they fit into the plan of the work? In III.1 Aristotle undertakes first to show the adequacy of the five senses to perceiving the various kinds of special sensible. He then argues the related point that there is no special perception of the common sensibles, and hence no requirement for a further special sense for these sensibles. Finally, having accounted for the adequacy of the five senses to perceiving the proper sensibles and the distinctness of proper perception from common perception, Aristotle explains why we need several senses capable of perceiving the common sensibles. III.1 has thus given us a defence both of the adequacy and the necessity of the five senses, in relation both to the proper sensibles and the common sensibles. I read III.2 as a continuation, and conclusion, of III.1, which is meant to show how the five senses are sufficient to account for other perceptual phenomena. The two main phenomena to be accounted for are 1) perceiving that we see and hear (and smell, etc.) and 2) perceiving the difference between the special sensibles, such as white and sweet. Plato in the Theaetetus had used the difference between sensibles as an example of the sort of object that could not be grasped by perception, but only by the soul itself. The aim of Aristotles argument is to show that these phenomena are perceptual; they can therefore be explained by reference to the perceptual faculties. Moreover, we do not need to posit further sense faculties than the five already acknowledged for us to explain how living beings can engage in these activities. We do, however, need to consider how the five senses are related such that they can perceive objects beyond those available to them as special senses. The answer to the question, as we shall see, lies in the common sense, by which the five senses are united. The end of III.2 and the beginning of III.3 mark a break from the account of the senses to that of the other faculties. Thus Aristotle ends Chapter 2 by saying concerning the principle (arkh) by which we say that the animal is capable of perceiving, let it have been defined in this manner.

IN DEFENSE OF INNER SENSE

237

Chapters 1-2 have thus dealt with the last issues required here in order to define the principle of perception, by which Aristotle clearly means the faculty of perception. The main business of Chapter 3 is to prepare the account of the intellect in III.4-5. Aristotle does this by showing the distinction of the intellect first from perception: unlike the activities considered in III.1-2 thinking cannot be understood as an exercise of the perceptual faculty. On this point many of his predecessors, Aristotle claims, were wrong. However, nor can intellectual thought be assimilated to belief or phantasia. So the intellect is a distinct faculty of cognition, requiring separate discussion in III.4-5. Let me end this section by stressing the importance of seeing Aristotles discussion of perceiving that we see within his thinking about the faculties. When accounting for the faculties Aristotle takes his starting point in their activities: the activity of a faculty is definitionally prior to the faculty (dunamis) (DA II.4 415a18-20). However, not all activities of the soul serve to define distinct faculties. In the case of the perceptual activities considered in Book II, seeing, hearing, smelling and so forth, each of these activities clearly defines a distinct faculty. Here Aristotle seems to have already assumed that there are five distinct sense faculties. 4 He has already decided, it seems, when he considers the activity of seeing that this activity will serve to define sight as a faculty distinct from the other senses. However, in Book III.1-3, we find Aristotle considering a range of perceptual, or quasi-perceptual, activities with a view not to defining further distinct sense faculties but rather to showing how the sense faculties already accounted for in Book II are sufficient for explaining these activities. We may see this stage of the DA as motivated by a concern with explanatory economy that is characteristic of what is commonly known as faculty psychology. Aristotle wants to show how a variety of psychological phenomena can be explained by reference to a few fundamental faculties. Since the explanatory power of this kind of theory comes in large measure from positing simple causes for complex behavior, the guiding principle is: no faculties beyond necessity. We see the principle at work when Aristotle shows that the cognition of common and accidental sensibles should be seen as an exercise of the perceptual faculty, when the activities of phantasia are explained as movements arising from percep-

_________
4 Aristotle is highly critical of the atomist view, which, he thinks, makes the other senses dependent on touch (cf. De Sensu 3 440a). On the other hand, there are cases where Aristotle himself struggles to differentiate the senses, as in the case of taste and smell, cf. De Sensu 5 with T. K. Johansen, Aristotle on the Sense of Smell, Phronesis 41, no.1 (1996) 1-19, as well as cases where he struggles to maintain the unity of a sense-faculty (especially that of touch, cf. DA II.11 422b18-34).

238

THOMAS K. JOHANSEN

tion, and hence at least indirectly as activities of the perceptual faculty, and when belief is shown to be an exercise of the intellect. Perhaps the culmination of the drive to economy is Aristotles attempt to group together even the perceptual and the intellectual faculties in a single discerning faculty (427a17-21). The discussion in DA III.2 should be read in this context. Aristotle here takes his starting point in the activity of perceiving that we see and hear and asks what faculty explains how we engage in this activity. As we shall see, Aristotle worries about this activity partly because it threatens to generate an infinite regress of faculties, clearly a faculty psychologists nightmare. Aristotle attempts to settle the worry by showing that we do not need another sense facultylet alone an infinite regressin order to explain our ability to perceive that we see; rather the senses that we have already been introduced to in Book II will be shown to be adequate to the task. Translation and Analysis of III.2 I translate the key passage at DA III.2 425b12-25 as follows:
Since we perceive that we see and hear, it is necessary [b12] that one perceives that one sees either by sight (opsis) or by some other [sense]. But the same [sense] will be [b13] of sight and the underlying color, so that either there will be two [senses] [b14] of the same thing or it [the sense] will be of itself. And furthermore, if indeed the sense that is of sight were different [b15], then either it will go on to infinity or some sense will be of itself, [b16] so that we should do this in the case of the first [sense]. But there is a puzzle (aporia): for if [b17] perceiving by sight is seeing, and color or what [b18] has color is seen, [then] if something is going to see what sees (to horn), then what first sees [or the first thing that sees] [b19] will also have color. However, it is clear that perceiving by sight is not one thing; for even [b20] when we are not seeing, we discriminate both darkness and [b21] light by sight, but not in the same way. Moreover, what sees is actually (kai) colored in a way [b22]: for the sense organ is in each case receptive of the sense object without [b23] the matter. That is why perceptions (aisthseis) and appearances (phantasiai) are present in the sense organs [b24] even when the sense objects have departed. 5

_________
5

In the Greek of W.D.Rosss 1961 edition: , , , , . , ,

. . ,

IN DEFENSE OF INNER SENSE

239

I start by offering a brief analysis of the argument. I shall then provide a more detailed interpretation. 1. Premise: We perceive that we see. 2. Inference: Therefore it is either A) by sight that we perceive that we see or B) by another sense. 3. Inference: 6 The same sense will be of both i) sight and ii) color. Interpretative note: 3) applies to both options 2) A) and B) above. 4. Inferences: if 2 A), sight will be of itself (by 3 i)); if 2 B), two senses will have the same object (=color by 3 ii)). 5. Further inferences: if 2 B), either a) for each perception that one is perceiving there will be another sense by which one perceives that perception, in which case there will be an infinite regress of senses, or b) some sense will be of itself. But not a), so b). If b), best that the sense that is of itself is sight, so 2 A). But if 2 A), there is an aporia: sight is of color or what has color [cf. DA II.7 418a29]; but if so, if what sees (to horn) is seen, it too must have color. 6. Aporia defused: a) There is another way of seeing by sight than seeing colors, for when it is dark we discern by sight that we are not seeing. b) What sees is in a way colored since the sense-organ receives the sense-object without its matter [i.e. the sense-organ receives the form]. That is why the perceptions and phantasiai can remain in the senseorgans to be perceived later.

_________
,
6

. The distinction between steps marked here as premise and inference is pretty loose. Clearly a number of further premises would have to be supplied with every new inference step to make the argument deductively valid. Nonetheless, it is worth marking in this outline how Aristotle presents the structure of the argument by means of words such as , , , .

240

THOMAS K. JOHANSEN

Detailed Interpretation The interpretation of this passage is disputed at almost every turn. We therefore need to consider the passage almost line-by-line to achieve any clarity and conviction. B12. Aristotle first presents the phenomenon to be explained: we perceive that we see and hear. The necessity mentioned in b12 is the necessity of the consequence rather than of the consequent. Aristotle is not saying that we necessarily perceive that we see when we see, or even that we always perceive that we see when we see. So even if passages elsewhere are taken to imply either of these claims, neither is mentioned as a premise of the argument here. Secondly, I take it that it is significant that Aristotle says that we perceive that we see and hear, rather than that we perceive that we perceive. 7 Aristotles phrasing does not imply (nor of course does it exclude) an interest in general perceptual awareness. Rather, the phrasing suggests an interest in the question of how we are aware of perceiving with specific sense-modalities, such as seeing and hearing. I shall have more to say about this subject later. B13. The question of how we perceive that we see is presented as a question of by what? we perceive that we see, where the alternatives are by sight (opsis) or by another. I translate opsis here as sight. Seeing in the sense of the act of seeing would be a possible alternative, as Bonitzs Index Aristotelicus shows. However, none of the examples cited by Bonitz are from the DA and other passages in the DA make it more likely that Aristotle has the potentiality of sight in mind rather than its actuality. On at least three occasions in DA Aristotle draws an explicit contrast between opsis meaning the faculty or potentiality of sight and horasis meaning the actuality of sight. The distinction is first made at DA II.1 412b27413a2, where Aristotle, wanting to define the soul as a potentiality, says as waking is actuality in the manner of cutting and seeing (horasis), so the soul is [actuality] as sight (opsis) and the potentiality. Again at DA III.2 426a12-15, some thirty Bekker lines after the passage we are considering, Aristotle says that in some cases there are distinct words for the actuality of the sense-object and the sense-faculty, in other cases there are not. So sounding (psophsis) and hearing (akousis) indicate the activity of the sense-object and the sense-faculty respectively, but in the case of vision there is only a distinct word for the actuality of sight, and not one for

_________
As emphasized by C. Osborne, Aristotle, De Anima 3.2: How Do We Perceive That We See And Hear, Classical Quarterly 33 (1983) 401-411, at 406.
7

IN DEFENSE OF INNER SENSE

241

that of color: for seeing (horasis) is said to be the actuality of sight (opsis). 8 And again at DA III.3 428a6-7 Aristotle draws the explicit distinction between opsis as the dunamis and horasis as the actuality of seeing. These passages strongly suggest that we should take the capacity of sight to be the default meaning of opsis in the DA. 9 If Aristotle meant to refer to the actuality of sight at DA III.2 425b13, it is quite unlikely that he would use the word opsis unmarked. As far as the alternative by another ( ) is concerned, it is not immediately clear what the complement of another would be: the feminine suggests aisthsis or, possibly, dunamis. Aisthsis is more likely both because we were told at b12 that we perceive that we see and because the alternative at b13 is that one perceives by sight that one sees, that is, the contrast is between perceiving this by sight and perceiving it by some other faculty of the sort by which we perceive. The elliptical reference to aisthsis by another is understandable in extension of the discussion in DA III.1 of why we have the aisthseis we do and neither more nor fewer. Moreover, once we provide the reference we can see why Aristotle would raise this question after this discussion. DA III.1 showed the adequacy of the five senses; the question that now arises is: do we need more senses to account for our ability to perceive that we see and hear? This question is at its most pointed when we envisage the possibility of an infinite regress of senses at b16. B14-15. The conclusion that the same sense will be of sight and of the underlying color follows whether or not it is by sight that one perceives that one sees or by another sense. 10 However, the further implications of

_________
I am grateful to Michael Pakaluk for stressing the significance of this passage to me. Metaphysics IX.9 1050a24-25 makes it clear that Aristotle would generally expect his readership to understand opsis to mean the faculty of sight, and horasis to mean its actuality. Indeed, he assumes that the reader is so familiar with the difference between opsis and horasis that the terms themselves can on their own be used to convey the distinction between potentiality and actuality. 10 So also Osborne op.cit. 401. Caston op.cit. 765 reconstructs the argument at 425b1315 on the capacity reading as follows: 1. [If a sense other than sight is involved,] there will be a sense for both sight and the azure color of the sky. 2. But if there is a sense for both sight and the azure color of the sky, then there will be two senses that have azure as their object. 3. But there cannot be two senses that have azure as their object. Therefore, 4. No sense other than sight is involved. Caston argues, then, that the tension between 1 and 3 is too great to make it plausible that anybody could subscribe to both. However, it is unclear why he saddles the capacity reading with the mistaken construal of (1), since he himself notes (op.cit. 765, n.33) that, The restriction to cases where there is a distinct sense is not actually in Aristotles text, but
8 9

242

THOMAS K. JOHANSEN

this conclusion differ in the two cases. In the first case it will be by sight that we perceive sight. Therefore it follows that there will be a sense that is of itself (aut hauts). In the second case, however, the sense will not only have sight as its object but also the underlying color that is sights object. In the second case there will therefore be two senses with color as its object: sight and the sense that perceives sight. 11 There are two important difficulties raised by the passage. The first concerns the expression it of itself. On the reading I have given, the expression should imply that the sense of sight is of itself in that it perceives itself. But, as Victor Caston and others have pointed out, what is seen is strictly speaking not the sense of sight but the activity of seeing a color. So, Caston argues, to avoid a change of reference between the pronoun it and the reflexive itself we had better take Aristotle to mean that the activity of sight sees itself. Having a change of reference within the phrase would be too harsh, as he puts it. 12 This point in turn supports Castons overall argument that the passage as whole does not, as traditionally thought, concern the question by which capacity do we perceive that we see and hear? but rather the question by which activity do we perceive that we see and hear? However, Castons point relies on an artificially narrow idea of how to read reflexive expressions such as seeing oneself. When we say, for example, that I see myself, we do not necessarily mean that the respect with which I see is the same as the respect with which I am seen. I may see my arms with my eyes and we would still be right to say that I see myself because both the eyes and the arms belong to the same thing, me. Similarly, we may say that the sense of sight sees itself, when it sees something that belongs to it, in this case the activity of seeing. There are, of course, special cases of reflection of the sort highlighted by Socrates in Alcibiades

_________
is supplied by many translations and commentaries . . . . The translations of Hicks and Smith are more faithful to the text: they represent the inference as following instead from the initial premisenamely, that we perceive that we seewhich is neutral between the options Aristotle is considering. 11 It is Aristotles considered view that it is more correct to say that we perceive color by sight rather than that our sight perceives color (cf. DA 408b11-15, 414a5-12). Nevertheless, he talks repeatedly in this passage, and elsewhere, of the sense faculty perceiving. Presumably he thinks that such talk can always be appropriately rephrased into talk of us perceiving. I shall follow Aristotle in repeatedly referring to the sense faculty perceiving below, but one should be aware of Aristotles own reservations with this idiom. 12 it [will be] of itself ( , b15; cf. b16). To read an alternation within this phrase would be too harsh; and the reflexive pronoun precludes it entirely. In this argument, then, Aristotle must be speaking either solely of capacities or solely of activities (Caston op.cit. 768).

IN DEFENSE OF INNER SENSE

243

I (133a) where the seeing thing (the pupil) is exactly the same as what is seen. But there is no need to think that Aristotle in DA III.2 has this special sort of case in mind when he sees that the sense of sight will be of itself. Secondly, commentators have puzzled over why Aristotle assumes that the second sense will also have color as its object. So Hamlyn says: It is not clear why Aristotle supposes the consequence to follow. He seems to assume that if I perceive by sense Y that I see X, I must therefore perceive X by Y. 13 To understand the point we need to go back to DA II.12. Perception, Aristotle explained, occurs when the sense faculty takes on the sensible form of the sense object. Therefore one cannot perceive the perception of that sensible form without also perceiving the sensible form as it is taken on by the sense faculty. 14 So since seeing red, for example, occurs when the sense of sight becomes red in a certain way, one cannot perceive seeing red without also perceiving the sense of sight become red in this way. Notice here that Aristotle says that the sense by which we perceive that we see red has to perceive both sight and the underlying color. I take it he has in mind here exactly that, since seeing color is a matter of the sense of sight becoming colored, both sight and color have to be perceived by the sense by which we perceive that we see color. The clarification so put anticipates what Aristotle will make explicit a few lines further down (b22-23), namely, that the organ of sight in a way becomes colored in vision. Second-order perception is more complex than first-order special perception: both color and sight are perceived. The point will turn out to be significant later. The way Aristotle presents the two consequences, either there will be two [senses] of the same thing or it will be of itself has the ring of dilemma to it. So we should ask why either disjunct might seem problematic. In the first case, we may take it that the apparent problem comes from Aristotles own idea that each special sense has one proper object, color, for example, in the case of sight. 15 A special sense is defined by its exclusive ability to perceive color as its proper object. It may therefore seem impossible that there could be two senses by which we could perceive

_________
D. W. Hamlyn, Aristotles De Anima. Books II,III, Oxford 1968, 121-22. Cf. R. D. Hicks note, Aristotle De Anima, Cambridge 1907, ad loc.: one can clearly be aware that one is seeing without being aware of what one is seeing. But properly speaking seeing is never just seeing for Aristotle; it is, at least, seeing red or yellow or some color, since seeing is a matter of some such sensible form being received by the sense faculty. 15 The claim is problematized by Aristotle himself in DA II.11 423a23-33 because touch, it seems, has several objects.
13 14

244

THOMAS K. JOHANSEN

color, if we think of those two senses as both perceiving color as their proper object. As for the other option, that there will be a sense that is of itself, the air of paradox comes from at least two directions. Partly it comes from the worry aired in Platos Charmides 165cff, namely, that if knowledge is of something, it ought to be of something other than itself, and partly it comes from the apparent conflict between the notion of a sense that is of itself with Aristotles own comments in DA II.5 (417a2-9) that the senses do not perceive themselves. Aristotle at this stage is setting up aporiai concerning perceiving that we see and hear, aporiai of the sort that might occur to a reader of Book 2. B15-16 goes on to present a further ( b15) issue that arises from saying that there is another sense of sight: either a) there will for each sensefaculty be another sense by which one perceives that sense, in which case there will be an infinite regress of senses, or b) some sense will be of itself. In the first case, a), the infinite regress seems generated by applying the premise that wherever an act of seeing is perceived the faculty by which that act is perceived must be other than that responsible for the first seeing. So for any sense, s, and any act of perception, p, if s is responsible for p, and p is itself perceived in p1, then there will be another sense s1, different from s, such that s1 is responsible for p1. Apply the principle to any act of perception, pn, there will be a sense sn+1 responsible for perceiving pn. Note that Aristotle is not committed here to saying that any act of perception, pn, must itself be perceived; only that any act of perception may be perceived, and if it is to be so perceived, then according to 2 B) the perceptibility of that perception must be explained by the stipulation of a further sense. For the regress of faculties to be generated it is sufficient that pn for any valuable of p be perceptible not that it is actually perceived. Aryeh Kosman, in his commentary, raised the crucial objection that unless there is a requirement that every perception is itself actually perceived there is no reason why the regress should continue. For unless each perception pn is perceived by a perception pn+1 there seems no reason why the regress should not simply stop at any given level. However, this requirement does not arise on the current reading. On this reading what generates the regress is not the requirement that any act perception actually be perceived, but rather the explanatory principle that any perception exercised by a faculty can only itself be perceived by a different faculty. The thought is that at each level of perception, where that perception is itself perceptible, a further faculty has to be posited in order to explain how that perception can be perceived. The regress argument thus serves to refute the conjunction of the following claims: a) any act of perception is perceptible, b) any act of perception at level n can only be perceived by a

IN DEFENSE OF INNER SENSE

245

sense faculty of level n1. Taken as an argument about faculties, a), rather than the stronger any act of perception is actually perceived, is sufficient with b) to generate the regress. 16 B17 draws the consequence that we should make the first sense responsible for perceiving that we see. The previous line sought to establish that, on pain of an infinite regress, there had to be some sense that was of itself. Now one might imagine somebody saying that we should go on postulating higher order senses until a certain level. Say, once we get to the level of the sense that is capable of perceiving that we perceive that we see, we stop and say that this sense is capable of perceiving its own perceptions. One reply to such a person is of course to say that stopping at this level seems arbitrary unless some further story can be told about why the sense that is capable of higher perception of this order should also be capable of perceiving itself. In the absence of such a story, however, it seems not only arbitrary but also uneconomical to postulate a higher order sense with the ability to perceive itself. For if we ascribe the ability to perceive itself simply to sight, then we need postulate no higher senses at all. Aristotles preference here for this account underscores the principle no faculties beyond necessity. B18-19 presents an aporia arising from the previous lines conclusion that sight perceives that it sees: as we know from DA II.7 (418a29), sight is of color or what has color. Therefore, if what sees (to horn) is seen, it too must have color. In sight is of color or what has color it is what has color that is picked up in the argument (cf. it too must have color). The implication, then, of saying that sight sees that it sees is that sight must somehow have color when it sees. Caston would here take the expression to horn (what sees) to refer not to the faculty of sight qua seeing, but to the activity of seeing as such. This is difficult for reasons of both meaning and language.

_________
16 I am grateful to Mary Louise Gill for pointing me in the direction of a parallel between the regress argument and the Third Man Argument, at least on one simplified reading of that notoriously tricky argument. If you postulate a different entity from Fred in order to explain why Fred is a man, namely the Form of Man, then you also need to postulate a second Form of Man in order to explain why the first Form of Man is a man, and so on. Similarly, the regress argument says: if you postulate a different faculty from sight in order to explain why sight is perceptible, then you will need a further faculty in order to explain why that faculty is perceptible, and so on. The regress argument thus understood employs analogues to both of Vlastoss Self-Prediction (SP) and Non-Identity (NI) assumptions: every faculty of perception is itself perceptible (SP) and the faculty by which a faculty of perception is perceptible must be different from itself (NI). Aristotles response to the regress argument is to deny NI: some faculties are capable of perceiving themselves.

246

THOMAS K. JOHANSEN

As far as meaning goes, it is unclear what sense could be made of saying that the activity of seeing has color. On my reading, the claim is that the sense-faculty has color insofar as it is seeing color: the sensefaculty assumes the color in the act of seeing it, but it is not the act of seeing the color that has the color. Compare: the fence becomes colored in the act or activity of being painted, but it is not the act of being painted that has the color. We can talk, then, of what sees as having color insofar as it sees color (where seeing color for Aristotle is a way of becoming colored), but that does not mean that it is the act of seeing color that has the color. Thus when you are seeing yourself seeing, you are seeing what sees as colored, because it becomes colored in seeing; but again that does not mean that you are seeing the act of seeing rather than what sees. Another way of putting the point is to say that you see the color in your sense faculty where that color is present by the sense facultys seeing the color. One sees the color or coloration of that which sees. Secondly, as far as language goes, taking to mean the (act of) seeing rather than what sees or the seeing (thing) is, to put it mildly, difficult. would be the normal Greek for seeing. Presumably it was in order to get that sense that Ross in his first edition changed the text to only to return in his second edition to the far better attested on the two occasions at b18. 17 However, Caston (788) suggests that the use of the neuter leaves it open what is being referred to: given Aristotles use of neuter substantives, it could refer either to the person, the sense organ, the visual capacity, the visual act. Rosss emendation of to is therefore unnecessary. The grammatical construction using a neuter participle with the definite article instead of the infinite with definite article is commented upon by Smyth ( 2051): Thucydides often uses in an abstract sense a substantival neuter participle where the infinitive would be more common, e.g., to dedios fear, to tharsoun courage (for to dedienai, to tharsein) 1. 36. 18

_________
See Kosman op.cit. 502. Smyth refers back 1153 b, N. 2: N. 2.: Thucydides often substantivizes the neuter participle to form abstract expressions: ts poles to ti_mmenon the dignity of the State 2. 63. Such participial nouns denote an action regulated by time and circumstance. Contrast to dedios fear (in actual operation) 1. 36 with to deos (simply fear in the abstract). Cf. also Horn 30, n.36 (cited by Caston): Bei dem Streit, ob 425b19 beide Male oder zu lesen ist, sollte bercksichtigt werden, dass und fast synonym verwendet werden knnen. Man vgl. die bei SchwyzerDebrunner, Griechiesche Grammatik Bd. II (=Hdb. d. Altertumswissenschaft II.I.2, Mnchen 1966, 409 zitierten Beispiele). However, Schwyzers examples of the substantivization of the neuter participle do not imply, and are not used by him to imply, the synonymy (near or otherwise) of the two grammatical constructions.
17 18

IN DEFENSE OF INNER SENSE

247

Initially, this may look like grist to Castons mill. However, outside of Thucydides idiosyncratic prose, the use of the neuter participle with the definite article to express the verbal act is unusual in classical Greek literature. 19 Caston quotes no examples from Aristotle and I am aware of none. Moreover, even if we take the two occurrences of at 425b19 as examples of the idiom, it is doubtful if the idiom, on closer inspection, will serve Castons purposes. Denniston has shown that the idiom while designating the verbal act still retains its partial reference to that part of the being which does so-and-so:
The use of the neuter article with present participle . . . to denote an abstract idea, is well known, particularly in Thucydides . . . . The participle almost always denotes a mental or emotional state: and it is usually in the accusative. Exact analysis of the idiom is difficult. The original force is clearly that part of the being which does so-and-so. (1) In some passages the participial expression has an almost purely abstract force. Thuc. VI.24.2 . Here virtually stands for : but not quite: they did not lose (not their desire for the voyage, but, more vividly) that part of their being which desired the voyage. (2) In others, that part of the person virtually includes the whole person, and the expression practically means the person, in so far as he does so-andso. 20

Denniston here gives two different uses of the idiom, (1) and (2). Let us imagine that at DA III.2 415b19 was an example of the idiom. If so, would in case (1) still also refer to the part of us which sees, whereas in case (2) it would refer to the entire person in so far as he sees. The key point for our purposes is that in either case the locution, while having abstract force, still retains its partial reference to that part of the being which does so-and-so. 21 I conclude as follows: Firstly, given the rarity of the locution in classical Greek, it is highly unlikely that Aristotle uses it here; secondly, even if we did take the idiom to be used here, it would still retain, according to Denniston, a reference to the part or the whole of the person in its activity of

_________
19 So Popp comments that, while the construction is fairly frequent in Thucydides, it is apud antiquores praetor Th. scriptores satis rarum, J.M.Popp, Thucydidis de Bello Peloponnesiaco, p.147 ad I.36.1. His only philosophical citation is the cyclical argument in Platos Phaedo 72b: . But here it seems important to Platos argument that should not just refer to the state of being asleep but also to the thing that is asleep since he wants to argue that it is the very things that are dead that come back to life (cf. 72c6; c7; d2, etc.). 20 J. Denniston, Two Notes Classical Review 45 (1931) 7-8. 21 It might be worth noting also that in DA III.2 is only accusative once (425b19) and otherwise nominative, twice (425b19, 22).

248

THOMAS K. JOHANSEN

seeing, that is to say the faculty or the person seeing. 22 So even in the unlikely eventuality that Aristotle was using the idiom, it would still be unsuited to supporting an activity reading rather than a capacity reading. It is right to note with Hicks and Caston (772-3) that Aristotle at b18-19 is introducing an aporia that has its origin in Platos Charmides 167c169c. Socrates is here finding difficulty with the proposal that temperance might be understood as a kind of self-knowledge, understood as a knowledge that is of knowledge without also being of some other object known. To highlight the difficulty Socrates introduces two kinds of analogous case. The first is that of a power such as greater than being greater than itself (and thereby simultaneously being smaller than itself) and double than being twice itself (and thereby simultaneously being half itself). The second is the case of sight which if it saw itself would itself have color. Socrates distinguishes clearly between the degrees of difficulty attached to understanding these two kinds of cases:
you observe then, Critias, that of the cases we have gone through, some appear to us to be absolutely impossible, whereas in others it is very doubtful if they could ever apply their own faculties to themselves? And that magnitude and number and similar things belong to the absolutely impossible group, is that so? Certainly. Again, that hearing or vision, or, in fact, any sort of motion should move itself, or heat burn itselfall cases like this produce disbelief in some, though perhaps there are some in whom it does not (168e). 23

_________
Cf. also A. C. Moorhouse, The Syntax of Sophocles, (Leiden 1982) on the substantive (articular) use of the participle in the neuter in Sophocles (my underlining): A.229 the future. E. 40 all that is going on. Here is a notable abstract use which was especially favoured in the latter half of C. Vth., and is prominent in Thuc . . . . Comparison is prompted with the articular use of the infin. (as ); but it is possible to see a difference of sense between the two. The difference is touched on by Smyth, Greek Gram. Para. 1153, who speaks of an action regulated by time and circumstance with reference to the participle: in other words, there is an element of the concrete present. P. 675 is more concrete than would be, so that my infirmity includes the notion of my infirm part. T. 144 : is not purely abstract youth, as is seen by its coupling with and the further physical references to the effects of heat, rain, etc.; it is closer to a collective what is young. 23 I do not see why Caston thinks that the parallels with the Charmides require an activity reading (773) to the exclusion of the sort of capacity reading I have developed here for Aristotle. What the passage requires, in Socrates words, is the application of the dunameis to themselves. So hearing will have to have a sound when it hears itself. But this has no stronger claim than that the sense perceives itself qua hearing. The plurals that Caston notes (772) at Charmides, hearing of hearings ( . . . , ), perception of perceptions are quite compatible with this claim: they are simply
22

IN DEFENSE OF INNER SENSE

249

It is therefore not accurate to use the Charmidess discussion of the idea of a sight that sees itself to claim, as does Caston, that the conclusions stated at 425b19-20 are understood to be patently absurd, just as it was in the Charmides. 24 The aporia is just that, a difficulty or source of perplexity which, as so often with Aristotelian aporiai, will be dissolved once we have a proper understanding of the subject matter. Lines b20-25 accordingly proceed to offer a two-fold solution to the aporia. In b20-22 Aristotle points out that there is another way of seeing by sight than seeing colors, therefore we do not have to conclude that what sees must have color if it is seen. In b22-24, on the other hand, he argues that there is a way in which what sees becomes colored in sight. There need be no difficulty, therefore, in accepting that what sees has color. Aristotle here turns out to be one of those in whom the idea of vision seeing itself does not produce disbelief, as Socrates put it. 25 B20-22. The first sentence establishes that while seeing may entail perceiving by sight, perceiving by sight does not entail seeing. We may thus assert the first disjunct in line b13 (i.e. one perceives by sight that one sees) without implying that one sees that one sees, with the consequent aporia. What shows that we may perceive by sight without seeing is the case of discerning darkness and light, even when we are not seeing. In understanding this point it helps to remember that in DA II.7 light is not considered the object of sight but its medium. Color is the proper object of sight, light being that through which the color appears. Perceiving the light is in this sense not strictly speaking a case of seeing. A fortiori perceiving darkness does not count as seeing. 26 Yet Aristotle also acknowledges that we do not discern light and darkness in the same way ( ,

_________
multiple instances of the activity whereby the sense faculty would perceive itself. Note here the singulars as well. 24 V. Caston, More on Aristotle on Consciousness: Reply to Sisko, Mind 113 (2004) 525. 25 Notice also that Socrates after this passage continues to explore the idea of a knowledge that is of itself on the assumption that it is possible to make progress with the difficulties. Caston therefore overstates his case (More on 542) when he says that Socrates treats this [sc. the idea that the act of seeing would be colored] as a reductio ad absurdum of the notion that sight might see itself, and more generally that any attitude might be reflexively directed. 26 Note, incidentally, that discerning ( ) does not itself imply a higher cognitive achievement than mere seeing, which is also referred to by Aristotle as (e.g. 426b10, 427a20).

250

THOMAS K. JOHANSEN

b22). 27 The contrast he has in mind here is not clear. One way of explaining it would be to say that when one perceives light (but is not seeing anything colored) one perceives in the light the absence of any color, whereas when one perceives darkness, one perceives the absence of the light. If so, the difference between the two kinds of perceiving also relates to the difference between two ways of perceiving that one is not seeing, one by way of perceiving that there is no sense-object, the other by way of perceiving that there is no light. We have then a connection between the claim that we perceive both darkness and light by sight and the claim that it is by sight that we perceive that we are seeing. For if it is by sight that we perceive that we are not seeing (either by the absence of a color or by the absence of light), then that seems to lend support also to the idea that it might be by sight that we perceive that we are seeing: for the same faculty would then be able to monitor both kinds of case. The proper distinction between ways of seeing is then a way not only of dodging the implication that perceiving by sight has to be perceiving color; it is also a way of showing, via the negative example, how the sense of sight may be involved in registering both its own activity and its own inactivity with respect to first-order perception. B22-24 considers the alternative case where we are seeing. One may take this as a rival solution to the aporia insofar as it accepts what b20-22 seeks to avoid, namely, the implication that what sees is colored. However, we may also read b20-22 and 22-4 as offering complementary answers to the aporia: b20-22 concerning the case where we are not seeing, but still exercising our vision, and b22-24 dealing with the case where we are seeing. The two answers, from this point of view, are offered not as contrasting or competing answers to the aporia. Rather they are meant to exhaust the two possible scenarios where we either perceive that we see or that we do not see, where the second scenario turns out to strengthen Aristotles case that such second-order perception is due to sight and not to another sense. B22-24 says first that what sees becomes in a way ( ) colored. For colored Aristotle uses the perfect ( ), which indicates that what sees is colored as a result of having been colored, more specifically, I suggest, as the result of having been colored through the activity of seeing. Thus Aristotle goes on to explain the claim ( ) by referring to the idea

_________
With Simplicius, Sophonias and Hicks, and against Themistius and Rodier, I take to indicate a contrast between discerning light and darkness by sight, and not between discerning color and darkness by sight.
27

IN DEFENSE OF INNER SENSE

251

familiar from DA II.12 that each sense-organ is capable of receiving the form of the sense-object without its matter. What it means for the sense-organ to take on the form of the senseobject without its matter is, of course, the subject of controversy. 28 Are we to understand the reception in terms of the sense-organs becoming literally colored or in some non-literal sense, as argued by Myles Burnyeat, according to whom the sense organ becomes colored insofar as the color appears to the eye of the perceiver. 29 Or are we to interpret the reception in some third way? 30 We may take Aristotles careful insertion of hs, in a way, as meant to steer us away from a literal reading. However, a literal coloration seems at first to give us a straightforward understanding of the passage. We are able to perceive the sense-organ because it becomes literally colored in vision just as we are able to perceive the external sense object because it is literally colored. Moreover, the eyes becoming literally colored in perception explains how the sensible form can remain in the sense organ after the external object has ceased to affect the eye, as one might imagine a stain remaining on a piece of cloth after the source of the stain has been removed. However, a literal reading is not required to make sense of Aristotles argument in our passage. What is required is a notion of being colored that would allow a) for the sense organ itself to be perceptible by being so colored and b) for the color to remain in the sense organ after the departure of the external sense object. Burnyeat addresses a) by saying that the sense-organ becomes colored in so far as color appears in the eye, where the appearance of color would be understood analogously to the way in which the external medium of sight appears to be colored in ones line of sight towards a colored sense object. 31 Requirement b) might seem more problematic on the non-literal reading

_________
28 For a start, see the essays collected in A. O. Rorty and M. C. Nussbaum (eds.), Essays on Aristotles De Anima, Oxford 1995. 29 Cf. M. F. Burnyeat, How Much Happens When Aristotle Sees Red And Hears Middle C? Remarks on De Anima 2.7-8 in Rorty and Nussbaum (eds.) op.cit. 427: Given our earlier discussion about the transparent medium being colored in a derivative way, it will come as no surprise to learn that, when we see, the eye is colored in a way (3. 2, 425b22-3). All this implies that the effect on the eye is the same as the effect on the medium, as one could already infer from lines 419a13-15 in 2. 7. The color appears to the eye of the subject who perceives it. The effect on the eye is a quasi-alteration just like the quasi-alteration in the transparent, because the eye too is actually transparent. 30 For the range of options, see V. Caston, The Spirit and the Letter: Aristotle on Perception, in R. Salles (ed.), Metaphysics, Soul, and Ethics in Ancient Thought. Themes From the Work of Richard Sorabji, Oxford 2005, 245-320. 31 Compare Burnyeats example of the water in a transparent glass which appears red when you hold a red card close behind it, Burnyeat op.cit. 425.

252

THOMAS K. JOHANSEN

since the external sense object is no longer present so as to bring about the appearance of color in the eye. On the other hand, notice that Aristotle says that perceptions and phantasiai are present in the sense organs also when the sense object are absent. Here we may take him to mean that once the external object has caused the appearance of red this appearance of red continues to be present in the sense organs by way of perceptions and phantasiai. The sensible form is capable of continuing to be present in the sense-organs in this way exactly because it is first received as a sensible form without the matter of the sense object. The continued perceptions and phantasiai of red do not require the continued presence of an external red object exactly because the sense organs reception of the sense objects form was a perceptual reception. Put differently, the perceptions continue in the sense organ because the color was first received as an appearance of color. 32 I do not sketch this answer here in order to endorse it, but rather to show how the current passage needs not be read as offering exclusive support to a literalist reading. For current purposes we may, thankfully, leave the issue undecided: all I shall need in my interpretation is a notion of the sense organs being colored in perception which, literalist or not, meets the two requirements I mentioned. The reference to the aisthtrion at 425b23, if it is to make any sense in the argument, has to pick up on to horn. 33 The claim that the aisthtrion takes on the sensible form is supposed to show the way in which to horn is colored. But if this is supposed to help solve the aporia at b17-20, then to horn must in turn refer to the same thing in b22 as in b19. This point tells once more against the activity reading of to hron: if the sense organ picks up on to hron, then to hron is surely what sees rather than the activity of seeing. However, it might be held also to be a problem for the faculty reading. For do we not have to say that to horn in b19 must refer to the sense organ and not to the sense faculty, as I argued earlier? I think not. Recall that Aristotles primary conception of the aisthtrion at DA

_________
32 Cf. R. Bolton, Perception Naturalized in Aristotles De Anima, in Salles (ed.) op.cit. 228-29. Without endorsing Burnyeats interpretation in general, Bolton emphasizes that the as if coloration in perception must involve an item of phantasia. He cites the passage at De Insomniis 2 (459b13-18) where Aristotle refers to the afterimage one has of the sun when, having looked at it, one shuts ones eyes. Bolton says it makes sense here to say that my eye is as if colored red because of the way my eye appears or presents itself to me in phantasia when I am aware of the afterimage (229). 33 Cf. Osborne op.cit. 404, n. 15: There is a textual discrepancy here which enables some editors to read in place of in one or both occurrences at 425b19, but this does not provide a satisfactory solution, since if the comment at 425b22-3 (which must refer to ) is to be any kind of solution to the difficulty in 425b19 it had better refer to at least a possible referent of the terms in b19, and preferably to the same term.

IN DEFENSE OF INNER SENSE

253

II.12 is that in which the dunamis to receive the sensible form is present. 34 Referring to the sense organ may therefore be taken to include reference to the sense faculty. Taking the sense faculty to be included in the reference to the sense organ seems particularly plausible at b23 since Aristotle mentions the sense organ as what is capable of taking on the sensible form of an object without its matter, the very capacity that in DA II.12 defined the sense faculties (aisthseis, 424a18). However, even if there is no contradiction between to aisthtrion at b23 and taking to horn at b19 to refer to the faculty of sight, we still need to explain why Aristotle at b23 shifts to referring to the matter of sight. I take it that the primary reason for the reference to the sense organs here is the attempt to locate the perceptions and imaginings. If you ask where the sensory impressions remain after the sense object has departed it seems appropriate not to point to the sense faculty, which qua faculty has no location, but to its organ, which as corporeal does have a location.35 So we know from the Physics (212b7ff) that it is the body that has place per se while the soul has place only accidentally. 36 The suggestion at DA III.2 425b24-5 is that the perceptions and phantasiai are capable of remaining in the sense organs because the sensible form is received by a sense organ which has a location and is therefore such as to be able to provide a location for the perceptions and phantasiai. 37 There is in b24-25 an implicit contrast between the absence of the sense objects ( ) and the presence of the perceptions

_________
where picks up in 424a18. 35 I note the parallel references in the De Insomniis (459a26, 459b6) to the sense organs as that in which the perceptual affection happens and remains after perception. 36 Cf. DA I.3 406a4-12, where it is suggested that the soul only moves locally insofar as the body it is in so moves, and DA II.12 424a25-28, where Aristotle first identifies the aisthtrion as that in which ( ) the sensory faculty is located and then says that it is the same as the faculty but different in being: for what perceives must be a magnitude (megethos), but neither the being of what can perceive nor the sense (aisthsis) is a magnitude but a certain logos and faculty (dunamis) of that thing [sc. the magnitude]. Insofar as having place requires having magnitude (points have no place), and only the sense organ has extension, it is proper to ask for the location of a sense faculty by asking where its organ is. 37 I take aisthseis here to be equivalent to the affections (path) or changes (kinseis) of the sense-organ as such. Compare the use of alloisis and pathos in a similar context at De Insomniis 459b4-9 (cf. also DA II.12 424a30). This understanding of aisthseis goes naturally with its coupling here with phantasia, which Aristotle understands as (DA III.3 429a1-2). The use of the plural aisthseis at 425b25 should therefore occasion no confusion with the use of aisthsis in the singular at 425b16 to refer to the sense faculty. on
34

254

THOMAS K. JOHANSEN

and phantasiai in the sense organs. I take this contrast to be relevant to the overall argument of the passage in the following way: even when the external sense object has gone, its sensible form is available for perception or cognition because it has been received by the sense organ. The sense organ thus becomes the vehicle for the sensible form vice the external sense object. Where the sensible form in normal perception is realized in the external sense object, it is, upon actual perception, also realized in the sense organ. That ultimately is why we can ascribe the ability to perceive our own seeing to sight understood as the faculty by which we perceive a specific sensible form, color. For when sight perceives itself seeing, it is also perceiving color, the sensible form that it also perceives in first-order vision. 38 However, perceiving that one sees cannot simply be a matter of perceiving color; for it also requires perceiving ones own sight as colored. In this respect, the content of second-order perception is more complex than that of first-order perception, where vision as a special sense can simply be of color. Hence if perceiving that one sees red is an exercise of the sense of sight, then it cannot simply be an exercise of sight as a special sense, that is, as the sense by which we perceive color. DA III.1 thus leaves us with the impression that it is legitimate to hold sight responsible for perceiving that one sees. The conclusion of the passage underlines that perceiving that one sees may be thought of as a function of the faculty of sight and not of a different sense faculty. We have thereby strengthened the argument in DA III.1 that no senses have been left out of the account in Book II. Aristotle has made a case for thinking that the five senses as defined in Book II are adequate also to explaining the phenomenon of perceiving that we see or hear or smell, and so on. However, DA III.2 has also left us with a question as to how sight can be responsible for second-order perception insofar as its content is more complex than simply color. In this respect the passage points forward to the discussion both later in DA III.2 (426b8-427a16) and in the De Somno of the way in which we perceive contents that go beyond those of special perception. As we shall see, the answer to the question lies not in positing

_________
To the objection that there is now another problematic shift of reference in the notion of the sense perceiving itself, this time between the sense faculty and the sense organ, I reply by referring to the above answer: just as there need be no problem in accepting that the sense faculty can see itself insofar as it is actualized, so there should be no problem in admitting that the sense faculty sees itself insofar as it materially realized. Aristotles idea that the sense organ is the sense faculty at a lower level of potentiality, while the actuality of perception is the sense faculty at a higher level of actuality serves after all to emphasize that these are three different states or aspects of the same thing.
38

IN DEFENSE OF INNER SENSE

255

further senses but in understanding the way in which the five senses work together. Implications and Defense of Interpretation At DA III.2 425b12-25, I have argued, Aristotle discusses the question of which faculty or capacity (dunamis) is responsible for perceiving that we see (or hear or smell, etc.). The outcome of the discussion was that we can say that it is by the sense of sight that we perceive that we see, and, by implication (though this is left unstated), that it is by the sense of hearing that we perceive that we hear, and so forth. 39 I want now to defend and develop this reading by considering three questions: 1) In what sense, if any, is Aristotle committed to the claim that we always or necessarily perceive that we see, when we see? 2) Is the interpretation of DA III.2 consistent with what Aristotle says elsewhere about perceiving that we see? 3) Does Aristotle, in the final analysis, offer a general account of perceptual consciousness in DA III.2 and if so, of what sort? 1) Inner sense theories of consciousness generally take consciousness to be a feature that first-order mental states acquire in relation to an act of perception that has the first-order mental state as its object. Inner sense theory thus distinguishes clearly between first-order and second-order mental states. Moreover, given the distinctness of the two states inner sense theory allows, in principle at least, for the first-order mental state to occur without the second-order perception. Inner sense theory thus in principle allows for first-order mental states which we are not conscious of having. This possibility is vividly illustrated by David Armstrongs example of the absent-minded truck driver who, having driven for many miles, comes to. Armstrong uses the example to distinguish the consciousness we have of the world from the consciousness we have of our own mental lives. Thus, in the example, the truck driver was all along seeing the traffic lights and avoiding dangers but he was unaware that he was seeing the traffic lights. Does Aristotle conform with inner sense theory on a) the distinctness of first and second-order perception and b) the possibility of having first-order perceptions without accompanying second-order perceptions?

_________
39 Castons comment (Aristotle on Consciousness 779, n.59) that Aristotle could not plausibly say that we taste that we taste or smell that we smell (i.e. smell that we are smelling) seems to me to beg exactly the question we are trying to answer. Cf. also Kosman, op. cit. 503.

256

THOMAS K. JOHANSEN

First of all, does he think that first and second-order acts of perception are distinct kinds of perception? There is a tendency in the literature to read Aristotle in such a way that perceiving that one sees becomes an essential part of first-order seeing. One prominent reading takes Aristotles account in DA III.2 to be an account not only of perceptual selfconsciousness but also an account of perceptual consciousness full stop. So Kosman argues that awareness that one is perceiving is what distinguishes being perceptually affected by sense-objects from being nonperceptually affected by them. 40 This conclusion rests on Kosmans reading of DA III.2 425b26-426a26, where Aristotle asserts the sameness in number of the activities of the sense-object and the sense-faculty. So he says, We cannot locate awareness outside the perception as an awareness of the perception different from an awareness of what is immediately perceived. For given that the activity of the object of perception as such is the same as the activity of perception itself, there must be a single awareness of both. 41 However, as Caston has argued, 42 it no more follows that we perceive our own activity of perceiving just because that activity is the same as that of the sense-object than it follows that we think of the road from Thebes to Athens when we think of the road from Athens to Thebes, where these two roads are also one in number but two in being. 43 Kosman, moreover, analyzes DA II.12 as implying that the difference between the cheeses being affected by the odor of the onion so as to smell of onion and my being affected by the odor so as to smell the onion is that I am aware of being affected by the odor whereas the cheese is not. However, the introduction of self-awareness at this point in the interpretation is premature. 44 It is true that perceiving is contrasted with merely being affected insofar as the former implies awareness. However, the awareness implied

_________
40 On Kosmans reading DA II.12 424b16 sets the question which DA III.2 answers. The question is, how does perceiving differ from merely being affected? The answer is, perceiving is being consciously affected in the sense that one is aware of being affected. Kosman, op. cit., 508: To perceive, we might say at the risk of circularity, is not simply to be affected but to perceive that one is affected, or to be affected, or to be affected and perceive that one is affected . . . . The question then at 424b16 is just the question: what is the nature and seat of the awareness which we recognize to be a necessary condition of perception in contrast to some mere affection? (his emphases). 41 Kosman, op. cit. 514. 42 Caston, op. cit. 784-5. 43 The key problem with this argument, as Caston rightly points out (784-5), is that since the actuality of the sense and of the sense-object are logically distinct (or distinct in being) it does not follow that if one perceives the activity of the sense-object (a dubious claim in itself, as Caston also points out) one also perceives the activity of sight. 44 My criticism here is not meant to cast any shadow on the brilliance of Kosmans article, which has illuminated Aristotles theory of sense perception in so many ways.

IN DEFENSE OF INNER SENSE

257

is just the awareness of the sense-object by which one is affected; it is not also the awareness of oneself being affected by the sense-object. It is implied by Aristotle throughout Book II that first-order perception is a form of awareness (aisthanesthai) of its object. DA III.2, however, explicitly addresses the different question by what faculty we are aware of our firstorder perceptions. This is a question not about perceptual awareness as such; it is specifically a question about our awareness of our own perceptual processes, about our second-order awareness. 45 For Aristotle the fundamental reason why first-order and second-order perception differ in being is that their contents differ. For it is by their kinds of content that we determine the different kinds of perception. In first-order vision you see color or what has color, where in second-order vision you see not just color or what has color; you see that you are seeing color, where that perception is of both color and sight, as b13-14 said. Notice, initially, that this difference is not best expressed by saying that second-order perception has a propositional content, that you are seeing, whereas first-order perception does not. For Aristotle is prepared to express the content of first-order perception of colors by a proposition, e.g. that [there is] color (418a15). Moreover, there is no reason to think that Aristotle is only interested in those cases where we perceive that we are perceiving the so-called special sensibles, such as color or sound; presumably he is as interested in cases such as perceiving that we are seeing that this white thing is the son of Diares, or that this white object is moving towards me, that is, in cases involving awareness of accidental or common sense-objects. First-order perception may, then, particularly in the case of accidental perception, have a complex content. So the difference between the contents of first-order and second-order perception need not lie in the one being simple and the other complex. Rather the difference lies in the fact that the content of second-order perception will always have greater complexity than the first-order perception that it takes as its

_________
45 This second-order awareness that one is seeing should in turn be distinguished from the reflective consciousness associated with deliberate introspection of ones own states (cf. Armstrong, op. cit.). Kosman in his commentary suggested that my discussion in this section assimilated Aristotles notion of perceiving that one sees to this third level of reflective self-consciousness. This criticism may reflect Kosmans own preference for a first-order theory of consciousness, which by taking awareness that one perceives to be an element of any first-order perception leaves reflective self-consciousness as the only alternative to firstorder perception. It is important, however, to insist that perceiving that we see for Aristotle is a function of our perceptual faculty; there is no indication that it requires any of the higher-order concepts typically associated with reflection and required to ask questions such as what am I thinking?, what is this feeling? and so on. Nor is there any indication that animals and children do not perceive that they see.

258

THOMAS K. JOHANSEN

object, since the second-order perception conveys not just the content of the first-order perception but also the information that one is perceiving that content. One way of bringing out the greater complexity of second-order perception is by pointing to the way in which second-order perception allows for error, in a way which mere vision of color, or that there is color, does not. Unlike the first-order perceiving of white, the perceiving that [it is] is white, perceiving that one sees white seems to be something one could conceivably be wrong about, in two ways. Firstly, one might conceivably be wrong in perceiving that one sees white. For example, one might, conceivably, have an accidental perception of white that one confuses with a special perception of white. In that case, it would be true that one was perceiving white, but not that one was seeing white. Secondly, we could imagine being wrong about whether one is perceiving oneself seeing red, or whether ones sense of sight was perceiving itself seeing white. One such situation in which reflexivity might fail would be if ones sense of sight was in fact perceiving another or anothers sense of sight seeing white. We need not, in order to make this point, import any notion that one in second-order perception perceives oneself as a certain sort of thing, let alone that one perceives ones self. However, it does seem required for X to perceive that he sees that X is somehow aware that it is he who is seeing. Perceiving that one sees does not seem simply to be a matter of perceiving that seeing is happening; it seems to be a matter of perceiving that one is oneself seeing. Indeed, in the formulation we perceive that we see the subject of the verb in the object clause (that we see) can only be determined by reference to the subject of the main verb. It seems implied by the formulation, therefore, that the perceiver in the second-order perception is also identifying himself as the subject of the (first-order) seeing. 46 It seems in principle possible that this identification should be mistaken. 47

_________
46 In Nicomachean Ethics IX.9 1170a29-b3 Aristotle argues from saying that one perceives that one perceives, and more generally that one perceives that one is alive, to saying that perceiving that one is alive is pleasant since life is good and to perceive that the good belongs in oneself is pleasant. It seems clear that this argument would not work unless Aristotle thought that in perceiving that one perceives one is aware that it is oneself that is perceiving. For if second-order perception did not provide the awareness that perceiving and more generally living belonged to oneself, it would provide no support for the claim that one perceives that the good belongs in oneself ( , 1170b3). 47 There may appear to be a contrast between my reading of Aristotle on this point and Sydney Shoemakers claim that there are certain first-person thoughts, I-thoughts, that are immune to error through misidentification. He claims, for example, that I cannot think that I am experiencing pain and think something false because I am mistaken about who it is that feels pain. See S. Shoemaker, Self-reference and self-awareness, Journal of Philosophy

IN DEFENSE OF INNER SENSE

259

Neither of these two sorts of second-order misperception is, I take it, a genuine physical or causal possibility for Aristotle. Given that perceiving that one sees white arises as a consequence of white being present in the sense of sight it is perhaps hard to see how one could be wrong about seeing being the sense-modality with which one is perceiving white. Also, given that one has no perceptual access to the sensory processes of others, it is hard to see how one might see anothers sense of sight seeing. 48 However, the claim here is not that such mistakes are physically possible. Rather the claim is that they are conceptually possible, where this conceptual possibility arises because the content of second-order perception is richer and more complex than that of first-order perception. First-order and second-order perceptions have different truth conditions because they have different contents. The fact that they have different contents shows that they are distinct kinds or types of perception. However, the fact that first-order and second-order perceptions have different truth conditions also suggests that the token act of perceiving white and the token act of perceiving that one is perceiving white cannot in general be the same. 49 Firstly, it seems clear that we ascribe truth or falsehood to tokens

_________
65 (1968) 555-567. Notice, however, that Shoemakers point is specifically one about firstperson singular statements such as I am in pain. Aristotles regular use of the third person, one perceives that one sees (DA 425b13, De Somno 455a17, Nic.Eth. 1170a29), suggests, however, that he is not interested in the peculiar semantics of first-person statements. Saying this is of course consistent with that possibility that Aristotle holds that all cases where somebody perceives that he sees are such that he would represent the perception to himself in the first person as I perceive that I see. I am grateful to Julian Kiverstein for comments on this point. 48 Even if the sensible quality is literally instantiated in the eye when one sees, e.g., even if the eye-jelly literally goes red when one sees red, then looking into somebodys eye when he sees red would still not allow us to see the seeing, in that we would still just see another instance of the sensible quality, the quality qua perceptible (to us) not qua seen. (For a similar modern thought experiment, see G. Gzeldere, Is Consciousness the Perception of What Passes in Ones Own Mind? in N. Block, O. Flanagan, and G. Gzeldere (eds.), The Nature of Consciousness. Philosophical Debates, Cambridge Mass., 1997: 789806 at 796). If, on the other hand, the color was not literally instantiated in the eye, but only occurred in the perceivers eye insofar as the color appeared to him, it would seem that anyone with access to that appearance of red would ipso facto have be that perceiver. 49 Contrast Caston, op. cit. 777-8, who suggests that [Aristotle] agrees that there is a higher-order contentperceiving that we perceiveas well as the first-order content of the original perception. But this is independent of how many token mental states are involved. And he believes that no other token state is required to make the original state conscious. The original state instantiates both lowerand higher order contents (his italics). Castons interpretation is subtle and detailed. I make the point here in the spirit of defending the claim important to inner sense theory that first-order perceptions are rendered conscious by

260

THOMAS K. JOHANSEN

of perception and not just to types. So we say that it is particular tokens of seeing an oak tree that are true or false according to the circumstances; we do not say that the type seeing an oak tree is true or false. Similarly, we say that it is tokensand not just the typeof seeing that [there is] white that are true even if in general such acts of special perception are true, or are least prone to error (DA 428b19). However, if seeing white and perceiving that one sees white are realized by a single token act, a problem arises when we want to say that these two perceptions have different truth values. 50 For it is possible, I have suggested, for it to be true that one is seeing white but false that one is perceiving that one is seeing white. However, it becomes impossible to ascribe different truth values to these two perceptions if we treat them as realized by a single token act: for surely we cannot say that one and the same token perception is at once both true and false. The implications of our identification of different contents for first- and second-order perception do not end here. For it seems that second-order perception will generally qualify as a form of accidental perception, insofar as it involves not just perceiving color but attributing this color to ones own sense of sight. Alternatively, we may take second-order perception to locate the color in ones own perceptual faculty. Even so, the second-order perception still looks like a case of accidental perception. For accidental perception is involved not just in saying what the colored thing is but also in saying where it is, cf. DA II.6 418a16. Now as far as Aristotles argument in DA III.2 is concerned, he should be able to admit that second-order perception is a form of accidental perception without that undermining his argument that we perceive that we see by sight. Aristotle suggested that we can say that we perceive by sight that we see because in vision our sense of sight becomes in a way colored. This suggestion does not commit Aristotle to saying that we perceive that we see by sight understood exclusively as a special sense. He may say that we perceive accidentally by sight that we see and what makes it appropriate to say that it is by sight that we see this is that part of this content is something we see by

_________
distinct tokens of second-order perception rather than in order to refute Castons own interpretation which requires and deserves a much fuller discussion than I can give it here. 50 Pace Caston, op. cit, 778, n.55, who notes that Rosenthal Thinking that One Thinks, in Martin Davies and Glyn W. Humphreys (eds.), Consciousness: Psychological and Philosophical Essays, Oxford 1993, pp.197-223 at 212 and A Theory of Consciousness, in N. Block, O. Flanagan and G. Gzeldere (eds.), op. cit. 746] argues that the two [sc. first-order and second-order mental states] must be distinct, because of the difference in their truth conditions. But this again only requires a difference in type, not in token. Caston, however, provides no justification of his rebuttal.

IN DEFENSE OF INNER SENSE

261

means of sight as the special sense it is. For part of what we see is a color, the color of our sense-faculty. The fact that in second-order perception this color is further attributed to my sense of sight makes the overall content of the perception an accidental perceptible. However, we can still insist that it is by sight that we perceive this accidental perceptible. Compare the discussion of accidental perception in DA II.6 418a20-25:
A thing is said to be an accidental perceptible, for example, that the white thing is the son of Diares. For it is perceived accidentally, because the son of Diares belongs to the white, which is perceived. That is also why no sense is affected as such by the [accidental] sensible. Rather amongst the per se sensibles the proper sensibles are primarily sensibles, and they are the sensibles in relation to which the substance of each sense is by nature.

Now when I see that the white thing is the son of Diares, it is clear that what affects my sense of sight as such is the white. What makes the perception as a whole an act of accidental perception is that that perception also attributes son of Diares to the white. It is appropriate to say that I see that the white thing is the son of Diares because part of the content, the white, is a proper sensible of sight. However, it is only an act of accidental perception because the content is not exhausted by the per se sensible, white; rather an attribute which is not a per se sensible, son of Diares, is attributed to the white. It is clear, then, that because of the presence of color in the content of my perception we can say here that I see that the white is the son of Diares, but also that we can only say that I see the whole content, that the white is the son of Diares, accidentally. 51 I think the case of perceiving that we see is similar. Because there is a way in which our sight becomes colored in sight we can rightly say that it is by sight that we perceive that we see. However, the content of such secondorder perception is not exhausted by the color since one perceives what has color as ones own sense of sight, 52 we should say that we perceive that we see accidentally by sight. This reading of the relationship between the contents of first-order and second-order perception allows us to reply, on Aristotles behalf, to an objection to inner sense theory formulated by Gven Gzeldere. The ob-

_________
Compare the manuscript reading at DA III.1 425a28-30: . 52 Aristotle expresses the content of accidental perceptions such as the white (thing) is the son of Diares. Here the term indicating the proper sensible appears first as the subject of the proposition, reflecting, it seems the causal primacy in such perception of the color as a proper sensible. Accordingly, we may prefer to express the content of second order perception as, for example, the red is my sense of sight. I am grateful to Pavel Gregoric for comments on this point.
51

262

THOMAS K. JOHANSEN

jection takes the form of dilemma: either inner perception simply provides the perceiver with the content of the first-order perception, in which case it is hard to see how it differs from a first-order theory of consciousness; or the inner perception provides information that one is seeing in which case the higher order perception theory looks no different from a higher thought theory, that is to say, a theory which takes it to be a thought or belief, rather than sense-experience, which endows first-order mental states with consciousness. 53 The Aristotelian way with this dilemma is to resist being caught on either horn. The Aristotelian will insist that the second-order perception combines both contents, that of first-order perception, the color, and the information that this color is being seen. Moreover, it provides this information as an exercise of perception, not as an exercise of thought or belief, exactly because part of what is perceived in secondorder perception is color. 54 There is a strong parallel, on the account I have offered, between perceiving that one sees and perceptual memory: remembering is perceiving that one has seen or heard in the past, etc., where perceiving that one sees or hears is perceiving that one is currently seeing or hearing. In De Memoria 1, 449b Aristotle says,
memory is of what happened, whereas nobody would claim to remember the present when it is present, such as this here white when one sees it, nor what is being contemplated, when one happens to be contemplating it and having it in mind, but one says only that one perceives the one and knows the other. But when one has knowledge and perception without the actual objects (aneu tn ergn) one thus remembers of a triangle that its angles are equal to two right angles, in the one case because one learnt or contemplated it, in the other because one heard or saw it or some such thing. For it is necessary

_________
53 Gzeldere, op. cit. 795, if the output of the second-order scanner simply provides the system with the content of the first-order state, this collapses the double-tiered structure of the HOP accounts into a unilevel account . . . . If, on the other hand, the output of the scanning process is taken merely to deliver information about whether the system is currently in possession of a certain first-order state (that a particular first-order state is tokened), then the HOP account transforms into a species of a HOT account. 54 To a modern reader it may seem that the content of Aristotles second-order perception that I am seeing white or that I am seeing the son of Diares involves concepts of the sort that only higher-order cognition such as thinking and believing, and not perception, could be expected to process. If so, Aristotles theory may look like it really ought to be characterized as a higher order thought theory rather than a higher-order perception theory. However, as Richard Sorabji has argued, part of what allows Aristotle not to ascribe beliefs and thoughts to animals is the extent to which he endows perception with conceptual content, cf. his Intentionality and Physiological Processes: Aristotles Theory of SensePerception, in Rorty and Nussbaum (eds.) op.cit.196-203. I am suggesting that it is the conceptual richness of perception that allows him to insist that also the perception of perceiving that one sees is, indeed, a case of perceiving.

IN DEFENSE OF INNER SENSE

263

when one actualises ones memory to say thus in ones soul, that one heard or perceived or grasped this earlier.

Aristotle goes on to explain that this ability to recognise that one has perceived something earlier requires a sense of time, and is therefore not available to all animals: for always whenever ones memory is active, one perceives in addition to perceiving that one saw this or heard or learnt this, that one did so earlier, and earlier and later are in time (450a). What is striking for our purposes is the parallel between second-order perception and memory. Both are an exercise of higher-order perception, involving perceiving ones own acts of vision or hearing, and so forth and both involve the use of a common sense. 55 The key difference is that in memory one is perceiving that one saw or heard ( , : aorist), in the case of normal second-order perception we perceive that we see or hear (DA 425b 11, : present tense). It is tempting therefore to see memory as a parallel case to the second-order perception identified in DA III.2, namely the case in which one perceives ones own perceptual states with the added information that these occurred in the past. It is possible that the final lines of the passage at 425b11-25 is inviting us to make exactly this sort of connection with memory: the occurrence of the sensible form in the sensorium is such that not only can it be perceived currently that we are perceiving, the continued presence of the sensible form also ensures we can also perceive at a later stage both what we perceived in the past and that we perceived it. 56 I have argued for the distinctness of first-order and second-order perception as kinds of perception. However, can we infer from this difference that first-order perception can occur in the absence of a second-order perception? Can we see without perceiving that we see? The distinctness between first-order and second-order perception as kinds of perception suggests that there is no conceptual necessity that second-order perception should always accompany first-order perception. However, a number of passages suggests that Aristotle thinks that we do always perceive that we see whenever we see. Here, in quick succession, are the three passages which seem to show the point, with some comments:

_________
Cf. De Memoria 450a. Stating the parallel does not require me to say that it is the common sense in the same capacity that is involved in both memory and second-order perception of current perceptual states. See P. Gregoric, Aristotle on the Common Sense, Oxford (forthcoming), who distinguishes more clearly than most scholars between the common sense as what he calls the unified perceptual capacity of the soul and the higher-order perceptual power that emerges from it, the latter including the capacity for phantasia. 56 Compare here at DA 425b24 with at De Memoria 449b.
55

264

THOMAS K. JOHANSEN

a) Nicomachean Ethics IX.9 1170a29-b21:

but the one who sees perceives that he sees, one who hears that he hears, one who walks that he walks, and similarly for all the other human activities there is something that perceives that we are active, so that we perceive that we perceive, and we think that we think, and [to perceive or think] that we are perceiving or thinking is [to perceive or think] that we exist (for being, as we saw, is perception or thought). 57

There are some unclarities here regarding both text and meaning. However, the passage seems clear that seeing or hearing is accompanied by perceiving that one sees and hears. Nonetheless, the passage is not explicit that we always perceive that we see or hear and the conclusion of the passage, that we take pleasure in our own existence, does not seem to require the premise that we are continuously aware of our own perceptual states. 58 b) Physics VII.2 244b12-245a2:
For actual perception is a change through the body, the sense-faculty being somehow affected. The animate, then, is altered in as many ways as the inanimate, but the inanimate is not altered in all the same ways as the animate (for it is not altered with respect to the senses), and (kai) whereas the inanimate is not aware of being affected, the animate is. For nothing prevents the animate from also being unaware of being altered, whenever the alteration does not happen with respect to the senses. 59

Caston interprets the passage as follows:

Aristotle begins with the obvious point that inanimate things do not perceive objects in their environment. But he adds a further and more interesting difference. Animate things are not unaware of undergoing change ( ) when alteration occurs in the manner of the senses, whereas nothing inanimate is aware of any such change (244b15-245a1). The participial construction with assures us that this is not the same point as the first. Animate things are not only aware of objects in their envi-

_________
Reading with the manuscripts , , , , ( ) 58 Bywaters emendation , , [understand ], makes the point more difficult to resist. 59 , . , , , , . .
57

, ,

IN DEFENSE OF INNER SENSE

265

ronment through perception; they are also aware of undergoing this alteration itself. This isnt proprioception either: it is not a question of being aware of eye movement or the like, but of being aware of an alteration that in some sense constitutes perception (244b10-12). To be aware of the changes one undergoes in the manner of the senses is to be aware in some sense of ones perceiving. 60

The key question is whether, as Caston claims, Aristotle is here pointing to two differences between the affections of the animate and inanimate or just one. On Castons reading Aristotle is articulating a double contrast between the animate and the inanimate: firstly, the animate notices objects in the world when it is affected with respect to the senses, whereas the inanimate does not (because it does not have any senses). So, for example, I feel the heat, whereas the steak in the oven does not. Secondly, the animate notices itself undergoing the perceptual change, whereas the animate does not. So I perceive myself feeling the heat, whereas the steak in the oven does not. But the second contrast now seems oddly irrelevant: of course inanimate objects do not perceive themselves perceiving since they do not perceive in the first place. The alternative to Castons construal would be to take and whereas the inanimate is not aware of being affected, the animate is as epexegetic (introduced by ), articulating one way in which the animate is affected and the inanimate is not. We could then read the Greek to mean it does not escape the notice of the animate when it is affected where the focus is not on noticing being affected but on noticing the thing that affects one. c) Metaphysics XII.9 1074b33-6:
Mind, then, thinks itself, if it is the best, and its thinking is a thinking of thinking. But it seems that knowledge and perception and opinion and reasoning are always of something else, but only of itself on the side. 61

Again this passage may be taken to suggest that perception alongside its primary object always also takes itself as a secondary object. 62 However, there is no need to read always across the contrast between primary and second function. One natural association of the expression on the side ( ) is with something you do not always or necessarily do. So I

_________
60 Caston, 61

op. cit. 757, his italics. ,

. So Caston, op. cit. 786: while our intentional states are always directed at something else and directed at themselves, he claims, they are primarily other-directed (his italics).
62

266

THOMAS K. JOHANSEN

cook for a living and do some fishing on the side suggests that fishing for me is an occasional activity. Our passage would then give us the double contrast, expressed by a pleasing chiasmos, of a) being of another ( ) vs. being of oneself and b) being always of something vs. being of something on the side ( ). 63 Read in this way the passage would rather suggest that we do not always perceive our own perception. The three passages can plausibly be read as saying that we always perceive that we see when we see, 64 although I have suggested some possible alternative readings which avoid this conclusion. However, let us grant for the sake of argument that Aristotle does think that we always perceive that we see, whenever we see. Is such a concession fatal to the inner sense interpretation of Aristotle? Given the distinctness of first- and second-order perception as types and tokens, on the inner sense interpretation the most natural explanation of the concurrence of first-order and second-order perception is that there is causal connection between the two, more particularly, that first-order perception causes second-order perception. If the occurrence of a first-order perception were causally sufficient for the occurrence of second-order perception, then the one would generally (in the absence of any unusual obstacles) be followed by the other. Is there any reason to think that Aristotle saw the relationship between the two in this way? DA III.2 says little about the processes of second-order perception.

_________
63 God, of course, emerges as a special case since for him thinking of himself is not something . 64 Caston, op. cit. 758, also adduces De Sensu 2 437a26-9: But this view faces another difficulty, since if it is not possible to be unaware of perceiving and seeing something seen, then necessarily the eye will see itself. Why, then, doesnt this happen when it is left alone? (his translation and emphasis). Caston suggests that this argument depends crucially on the assumption that it is impossible to be unaware that one is perceiving something while one is perceiving it. It is not clear, however, that this assumption is crucial to the argument: the conclusion follows equally if we take Aristotle to be sayingon another plausible reading of the Greekthat it is not possible for somebody who is seeing to be unaware of what is being seen. When Aristotle asks why does not this happen when one does not rub the eye?, he is implying that the fire in the eye should be causing vision even when no-one is bothering the eye, that is, on the theory criticized, which is ignorant that fire is not a standing condition of vision but a special phenomenon caused by a reflection set up by rubbing the eye quickly. And if the fire is there causing vision we cannot be unaware of it since it is not possible to be unaware of what is seen when one is seeing. It is then not the seeing that we cannot be unaware of but, as Aristotle himself says, the eye that we cannot avoid seeing (cf. ). The objection is then relying on essentially the same point as the one mentioned at DA II.5 (417a2-9): if fire was actually present in the eye it would continuous be offering itself as an object of vision without the need for an external object to actualize the eyes potential to see.

IN DEFENSE OF INNER SENSE

267

Two clues that we are offered are, firstly, that we can say that we see our vision of color because the sense-organ itself in a way becomes colored in vision and, secondly, that such perceptions and imaginings remain in the sense-organ even after the sense-object has gone. The first clue suggests that there is color available for second-order vision whenever we are engaged in first-order vision. So whenever I see a colored object before me, that color will itself be impressed on my sense-organ and therefore my sense organ can itself be seen, as colored. However, given that vision is the reception of the color by my sense faculty we may wonder how the reception of the color could also result in a further second-order act of vision. Here the second clue may help: that is why perceptions and appearances (phantasiai) are present in the sense-organs even when the sense-objects have departed (425b24-5). For an example of such phantasiai we may refer to Aristotles account of dreaming. 65 This account is based exactly on the idea that the sense organs in perception receive and retain impressions of the sense objects. However, what it means for the impressions to remain in the sense organs is not simply for them to lie dormant as potential objects of cognition. Rather the sense-impressions are understood dynamically as motions or changes. Aristotle uses the analogies of a javelin that keeps moving after it has left the hand of the thrower (compare the external sense-object), or a hot object that heats up a succession of adjacent objects. Aristotle thus takes dreaming to be the result of the continued motions of the sensory impressions in the sense organs, where those continued motions are understood as the efficient cause of dreaming. This dynamic conception of the movements in the sense organs may help us understand better how the sense impressions act on us so as to bring about second-order perception. The idea would be this: just as the movement of the sense impression brings about dreams after the external

_________
Cf. for the following Aristotle De Insomniis 2 459a23-b7: What a dream is, and how it occurs, we may best study from the circumstances attending sleep. For sense-objects corresponding to each sense-organ provide us with perception. And the affection produced by them persists in the sense-organs, not only while the perceptions are being actualized, but also after they have gone. For the affection in their case would seem akin to that of objects moving in space. In their case too there is motion, even when the moving agent is no longer in contact with them. For the moving agent moves a certain portion of air; and that, on being moved, in turn moves another. And in that way motion continues to be produced, in air and liquids alike, until a standstill is reached. This, one must assume, applies in the case of alteration likewise. For what has been heated by something hot heats its neighbour, and this passes it on successively, until the starting-point is reached. So this must happen in perceiving as well, seeing that actualized perception is a kind of alteration. Hence the affection persists in the sense-organs, both in depth and on the surface, not only while they are actually perceiving, but even after they have ceased to do so. (Translation D. Gallop.)
65

268

THOMAS K. JOHANSEN

sense-object has ceased to act on us, so this movement brings about second-order perception while the external sense-object is acting on us. The notion of the impression on the sense organ as a movement may go some way towards explaining how the impression is such as to cause us to perceive that we see whenever we see. 66 However, this account no doubt still falls short of a full explanation of the mechanics of second-order perception. For example, we may worry how the sense-impression in the sense faculty can cause the same faculty to perceive it, given that there is generally only perception at a distance? 67 The upshot of these considerations is this. Aristotle may have a causal story according to which perceiving that one sees will happen whenever one sees. It may be that perceiving that one sees happens whenever one sees because seeing is causally sufficient for perceiving that one sees. Even so, however, the difference in properties between first-order seeing and second-order seeing, particularly with respect to truth and falsehood, suggests that second-order perception is a distinct kind of perception, characterised by a different kind of content, from that of first-order perception. In this way, second-order perception does not seem to part of the nature or essence of first-order perception. This conclusion goes well with Aristotles observation in Metaphysics XII.9 that perception is always of something else but only of itself en parergi, as a secondary task (1074b35-6). I have explored one way in which we may understand firstorder perception to be causally sufficient for second-order experience. And so it may be that, for Aristotle, first-order perception is always in fact

_________
66 We may still worry whether the movement of the sense-impression would be a sufficient condition for second-order perception. Compare the case of dreaming: while the movement of the sense-impression may later bring about dreaming the actual occurrence of dreams clearly depends on a range of other factors relating to sleep. Similarly, the movement of the sense-impression may be just one of a range of factors required to bring about second-order perception. I dont pretend to have an answer to this worry other than to speculate that Aristotle may think that, given that first-order perception is taking place, the sense faculty is already primed for perception in such a way that all it takes for second-order perception to take place is the action of the sense-impression on the sense faculty. 67 A stop-gap answer to this question may be to refer to the further mediation of the sense-impression to the common sense faculty (cf. below on the role of the common sense in second-order perception). However, this only appears a stop-gap answer insofar as firstorder perception too requires this mediation. So we may ask: how does the common sense faculty perceive itself perceiving, if perception in general requires distance between sensefaculty and sense-object? Another reply may be to say that perception requires distance between the material object perceived and the sense-faculty but not between its sensible form and the sense-faculty. So it is perhaps significant that Aristotle at DA II.11 423b21-22 says that there would be no perception if somebody placed a white body directly on the eye.

IN DEFENSE OF INNER SENSE

269

followed by second-order perception, even though there is no conceptual necessity that this should be so. 68 Aristotle would then differ from those modern inner sense theorists who insist that first-order perception often does occur in the absence of secondorder perception, a claim supported by examples such as Armstrongs truck driver. Moreover, such inner sense theorists would take it to be an important advantage of their theory that it explained just how it was possible to have mental states of which we were unconscious. Aristotle seems to differ from those inner sense theorists on this point. However, he does not differ from all inner sense theorists: Locke, for example, agrees with him. 69 There is scope for disagreement, then, amongst inner sense theorists as to whether second-order always follows first-order perception. I take it, therefore, that for the purposes of identifying Aristotle as an inner sense theorist it is more important that for him first-order perception and second-order perception remain distinct kinds of activity, and that it is the second order perception that makes us aware of the first order perception.

_________
68 If we accept that seeing is always accompanied by perceiving that we see, does the infinite regress not appear again, this time not as a regress of faculties (of the sort that I claimed DA III.2 was concerned with) but as a regress of token acts of perception (of the sort that Castons interpretation of DA III.2 is concerned with)? At this point we may relevantly recall the point Kosman made in his commentary (cited above): the senses can decide to stop and order drinks all round; nothing more needs to happen. . . . For there to be a genuine infinite regress, there will have to be some reason requiring the series to continue. On my interpretation there is for Aristotle no general requirement that all acts of perception have to be perceived, and so the infinite regress of acts of perception does not follow. There is, on my interpretation, a particular causal story that suggests why first-order perception should be followed by second-order perception, but the story is not of the sort of generality that gives reason to think that an act of perception of any order has to be itself perceived. If we think of Aristotelian causes as including the final cause, it is easy to think of potential final causes for perceiving that one sees, such as the coordination of behavior, storage of information for memory, the enjoyment of life mentioned in Nic. Eth. IX.9 (cited above). It is much harder to see how such final causes would apply to third or even higher order perception. What would be the benefit to an animal of perceiving that it perceives that it sees? 69 Essay on Human Understanding, II.i.19, That a man should be busy in thinking, and yet not retain it the next moment," very improbable. To suppose the soul to think, and the man not to perceive it, is, as has been said, to make two persons in one man. And if one considers well these mens way of speaking, one should be led into a suspicion that they do so. For they who tell us that the soul always thinks, do never, that I remember, say that a man always thinks. Can the soul think, and not the man? Or a man think, and not be conscious of it? This, perhaps, would be suspected of jargon in others. If they say the man thinks always, but is not always conscious of it, they may as well say his body is extended without having parts. For it is altogether as intelligible to say that a body is extended without parts, as that anything thinks without being conscious of it, or perceiving that it does so.

270

THOMAS K. JOHANSEN

2) DA III.2 argues that it is by sight that we perceive that we see. But at De Somno 2 455a12-22 Aristotle seems to deny this explicitly:
But since in the case of each sense there is something peculiar and something common, for example, on the one hand, seeing is peculiar to sight, hearing to the sense of hearing, and in the same way for each of the other senses, while on the other hand there is also a certain common capacity that attends upon all of the senses, by which one perceives both that one sees and that one hearsfor surely it is not by sight that one sees that one sees and it is not by taste or sight or both together that one discerns and is capable of discerning that the sweet things are different from the white but by some part that is common to all the sense organs. For there is one sense and the controlling sense organ is one, but the being is different for the sense of each kind, such as sound and color. 70

How are we to explain the apparent contradiction between the De Somno and DA III.2? Hicks and Osborne underline the inconclusiveness of the discussion in DA III.2: Aristotle never asserts in DA III.2, in so many words, that sight is responsible for perceiving that we see. So they take his considered view instead to be that of the De Somno: it is the common sense by which we perceive that we see. Caston, meanwhile, takes the contradiction to be a result of reading DA III.2 as concerned with capacities: once we realize that this chapter deals with the activity of perceiving that we see, there is no contradiction between taking this activity to be a case of seeing, while also in De Somno presenting the common sense as the capacity responsible for this activity. Caston thus presents a challenge to any capacity reading of DA III.2 to avoid the inconsistency with De Somno. My response shall be a compatibilist one according to which the conflict between DA III.2 and the De Somno can be resolved while maintaining a capacity reading of DA III.2. Let us start by considering the context of De Somno 2. Aristotle wants to argue in this chapter that sleep is an affection of the sensory faculty. His view is that since sleep affects all the senses together, sleep must be an affection of something the senses have in common, that is, the common sense. If not, it would be possible to perceive by one of the senses even while asleep, but we observe that there is no perception by any of the senses when we sleep. But what does Aristotle mean by a common sense?

_________
70

, , , ,

, , , ) ,

, (

IN DEFENSE OF INNER SENSE

271

He stresses the fact that the common sense is common in that it belongs to all of the five senses ( , 455a16); similarly, the common sense organ is a part that is common to all the sense organs ( , 455a1920). More precisely, however, Aristotle also says that there is one sense whose being differs for each kind of sensible. The common sense is thus not a sense in addition to, or over and above, the individual senses; it is rather the sense in virtue of which the five senses are one. Alternatively, we could say that there is one sense that is also distinguished as five senses insofar as there are five different kinds of proper sensible. Aristotles comment that it is not by sight that one sees that one sees should be read in the context of this emphasis on the unity of the senses. Aristotle is contrasting what can be said of the individual senses as special (idiai) or different and what can be said of them as having something in common (koinon). What Aristotle wants to deny is that it is by sight as a special sense that we see that we see that we see. I note two points here. First, consider the other role Aristotle assigns in the passage to the common sense, discerning that sweet is different from white. This too Aristotle says is a kind of perception that can be put down to neither taste nor vision. Now clearly what Aristotle means here cannot be that vision and taste are not involved in the perception that white is different from sweet. For white and sweet are after all the proper objects of vision and taste. Rather what makes the discrimination of white from sweet a matter for the common sense is that neither of these senses as special senses is capable of discerning the difference between their proper objects, whether you take them individually or together ( , 455a19). Now lets return to the question concerning, as Aristotle first put it, that by which one perceives both that one sees and that one hears ( , 455a16-17). It is natural to take Aristotles phrasing here to indicate the capacity by which one perceives both that one sees and that one hears. What we have here, is the analogue, at the level of second-order perception to the case, at the level of first-order perception, of perceiving something as both sweet and white. It is the analogue because it brings together in one perception, two different sense-objects, that one sees and that one hears. In this case the perception that one both sees and hears does not bring together the proper objects of taste and sight, sweet and white, but the activities of the senses themselves. It is, thus, as reasonable to say that one and same special sense cannot coordinate the perceptions that one sees and that one hears as it is to say that one special sense cannot coordinate the perceptions of white and sweet. Now lets consider again, in extension of this reading, the claim that surely it is not by sight that one sees that one sees. Here there is appar-

272

THOMAS K. JOHANSEN

ently no coordination of the activities of sight and other senses, like hearing. Nonetheless, in extension of the previous lines, it is natural to read Aristotle as denying that it is by sight as a special sense that we perceive that we see. If so, he is not excluding that sight is involved in perceiving that we see; he is just saying that insofar as sight is involved it is not involved as a special sense, but rather by virtue of its integration with the other senses. Now in the previous section I argued that perceiving that we see red appeared to be a form of accidental perception in that the content involved not just red but also an attribution of that red to ones own sense of sight. I think we can see now that the implication of this point was already that perceiving that one sees could not simply be a function of sight as a special sense. For sight as a special sense was concerned merely with color, and not with information such as whether this or that thing is colored. This notion of perceiving by sight as a form of accidental perception came out particularly clearly in the case mentioned by Aristotle of perceiving by sight that we are not perceiving, whether in the absence of light or a senseobject. There is therefore nothing new in De Somnos suggestion that it is not by sight as a special sense that we see that we see. Rather the question that arises is whether DA III.2 is compatible with the De Somnos suggestion that it is by the common sense that we see that we see, given the way DA III.2 highlighted the role of sight in perceiving that we see. I think two more comments will help show that the two works are compatible. Firstly, there is the key point that the common sense does not constitute a further sensory faculty over and above the special senses. The De Somno is clear that the common sense is not different in number from the special senses, it is only different in being. The common sense does not constitute another sense (hetera aisthsis) of the sort referred to in DA III.2. Given the way DA III.2 sets up the question, Is it by sight that we perceive that we see or by another sense?, there is therefore no contradiction between the answers given by the two works. The admission of the common sense does not add to the number of senses; nor, therefore, does it constitute the first step in a regress argument. Secondly, it makes good sense to say that the way in which the content of second-order perception goes beyond that of first-order perception points exactly to the involvement of the common sense. Once we accept the notion that we can perceive by sight that we see as a matter of accidental perception, it is easier to understand exactly why this kind of accidental perception would draw on the input of the common sense. It is the common sense, after all, that allows us to make contrastive judgments between perceptibles. It is, as De Somno points out, by the common sense that we discriminate white things from sweet things. We can now see the rele-

IN DEFENSE OF INNER SENSE

273

vance of Aristotle using exactly this parallel at the point when he wants to say that it is by the common sense that we perceive that we see. For if perceiving that we see involves identifying vision amongst a range of sense modalities, hearing, taste, touch, etc., then we would expect the common sense to be involved just as we expect it to be involved in discriminating (krinein) among a range of kinds of first-order sensible, flavor, color, sound, etc. 71 When it comes to making discriminations between different generaas opposed to speciesof sensible we expect the involvement of the common sense, not just in first-order discriminations such as white is different from sweet, but also in second-order discriminations of the sort that seeing is happening rather the tasting. Moreover, it is exactly this element in the second-order perception that makes it other than a case of special perception. That is, it is the element of our perceiving that we see red rather than that we see red which gives second-order perception a richer content that special perception. Therefore the manner in which second-order perception involves accidental perception goes well, on my reading of DA III.2, with the point in De Somno that it is by the common sense that we are capable of grasping that we see. The two works do seem to be compatible, then, even on a capacity reading. 3) Finally, does Aristotle have a general account of perceptual consciousness? Given the proliferation of senses of consciousness,72 it is far from clear what we should expect from such an account. I have in my interpretation tried to separate the question of first-order perceptual awareness of objects in the world and the question of second-order awareness of ones own perceptions. Given my defense of this distinction, I do not think that Aristotle is offering either in DA III.2 or in De Somno 2 a general account of perceptual consciousness, if that is meant to explain also what makes first-order perception consciousness of objects in the world. However, Aristotle does give us an account of perceptual self-consciousness insofar as he tells us by what faculty it is we are capable of gaining information about our own perceptual activities. This account may only be an account of what, thanks to Ned Block, has become known as access con-

_________
It is natural, as do Ross and Siwek, to place a comma at 455a17 and read , However, we know from elsewhere that is also considered (DA 426b10). It is tempting to take Aristotle as implying this point here, since it would ease the transition in thought in the sentence from seeing to discerning. Moreover, it would support the point I am making here that Aristotle thinks that perceiving that one sees involves a discrimination amongst the different sense-modalities. I am grateful to Philip van der Eijk for comments on this sentence. 72 For a useful survey, see W. Lycan, Consciousness, Cambridge Mass., 1987, ch.1.
71

274

THOMAS K. JOHANSEN

sciousness, that is, the mechanism by which the contents of our mental state are available, or poised, for rational control, verbal report, and reasoning. 73 Aristotles account of perceiving that we perceive may not also constitute an account of phenomenal consciousness in perception, that is, the fact, as it seems, that our perceptual experiences have peculiar qualitative feel to them, what is sometimes referred to as the qualia of experience. 74 If so, there are at least three explanations of Aristotles failure to address the issue of phenomenal consciousness in the context of perceiving that we see. One is that he did not recognize phenomenal consciousness as a phenomenon and therefore not as an explanandum. Another is that he recognized it as an explanandum but not as an explanandum of a theory of inner perception. His theory was meant as a theory of how we gain perceptual access to our own perceptions, and no more. If his scope were so limited, he would not be alone among inner sense theorists. 75 A third possible explanation is that Aristotle did not think there were further questions to be answered about the qualitative aspects of experience other than the question of what it is represented in our awareness of such experiences. In articulating this option, it helps (again) to keep in mind the distinction between first-order and second-order perception. The qualia of the first-order vision of red (what red is like when it is seen) would, on this account, be none other than the attributes of red. 76 The qualia of the second-order perception that one is seeing red would be the peculiar way in which red appears in ones act of seeing red. 77 Put differently, the qualia of seeing red are those of red as instantiated in ones own visual system. Thus for Aristotle the qualia of first-order and second-order

_________
73 N. Block, On a confusion about a function of consciousness, Behavioral and Brain Sciences 18 (1995) 227-47. 74 See, however, Andy Clarks A case where access implies qualia?, Analysis 60 (2000), 30-38, for an attempt to show how qualia might be explained as an aspect of ones access to the specific perceptual modes of experience. 75 Cf. W. Lycan, Consciousness as Internal Monitoring, in N. Block, O. Flanagan, and G. Gzeldere (eds.), op. cit. 755-772 at 756. 76 Such a reading would be facilitated if it is correct to say with S.Broadie that Aristotle thinks of the so called secondary qualities as literally qualifying the physical objects perceived in terms of them andhe thinks of the objects qua thus qualified as causing the corresponding perceptions (Aristotles Perceptual Realism, in J.Ellis (ed.), Southern Journal of Philosophy, Supp.Vol. Spindel Conference 1992: Ancient Minds 31: 137-60 at 144). 77 On the importance of keeping these two questions apart in the case of phenomenal consciousness, see Peter Carruthers warning against the tendency to elide the conceptual distinction between what the world is like for an organism, and what the organisms experience of the world is like for the organism. Phenomenal Consciousness, Cambridge 2000.

IN DEFENSE OF INNER SENSE

275

vision would be similar insofar as they are both qualia of red but differ insofar as in second-order vision those qualia of red are invested with a sort of internal character: it is red as appearing in my sense of sight rather than as a feature of the external world.78 This difference may express the peculiar introspective feel that focusing on ones own perceptions has. Aristotles account of perceiving that we see could thus provide some elements, if no more, for a theory of the qualia of perception. Moreover, such a theory would be able to address at least one objection that is commonly leveled against inner sense theory. The objection is that if the qualia of seeing blue are generated by the experience of a distinct inner sense, we would expect the qualia of seeing blue to be significantly different from the qualia of blue. 79 But, as Moore famously pointed out, the experience of seeing blue is strikingly like the experience of blue: when we try to introspect the sensation of blue, all we can see is the blue: the other element is as it were diaphanous. 80 The Aristotelian answer to this objection is, firstly, to accept the observation that the experience of seeing blue is very much like that of blue, while resisting complete transparency. The Aristotelian explanation of the similarity, as we saw, is that blue forms part of the content of seeing that one sees blue, although it does

_________
78 G. Gzeldere, Is Consciousness the Perception of What Passes in Ones Own Mind?, in Block, Flanagan and Gzeldere (eds.) op. cit. 798, argues that the idea of inner perception might work if the properties of ones representational states resembled (or were identical to) the properties that those states represented. . . . Here is how it would go: By first-order visual perception, the subject S forms, in her mind, a representation of the cup in front of her. This representation is her visual state, V. By second-order introspective consciousness, she goes on to internally perceive this representation. The representation of the cup has cuplike properties that resemble properties of the cup itself. Thus, by secondorder perception, S gets to see the cup in her minds eye. He goes on to associate this view with that of Scholastic Aristotelians, according to which S would have in her mind a little replica of what she saw in front of her, and with the aid of an internal third eye . . . were able to view this replica, which to Gzelderes mind is more like fairy tales. However, Lycan (as Gzeldere points out, 804, n. 21) disagrees, supporting the Aristotelian notion of a resemblance between the sense-object and the perceptual system in terms of their isomorphism. 79 Thus Peter Carruthers objects: If the sort of subjectivity produced by inner sense were really like the sort of subjectivity of the world produced by the operations of the firstorder senses, then it is mysterious how our phenomenally conscious experiences could have the property of transparency. If inner sense picks up on, and represents in a particular manner, properties of our experiencesthen surely we would expect there to be a distinct (non worldly) set of properties of phenomenally conscious experiences on which introspection could concentrate. But there is not. Forconcentrating on your experience of red just is concentrating more closely on the redness represented. Carruthers op. cit. 238. 80 G. E. Moore, The Refutation of Idealism, Mind XII (1903), 449,450; quoted by Kosman, op. cit. 516-7.

276

THOMAS K. JOHANSEN

not exhaust it. Secondly, however, since perceiving that one sees is not the function of a distinct inner sense, but a function of sight in cooperation with the other senses by way of the common sense, we should not expect inner perception to generate alien perceptual qualia. The visual feel also of inner perception is then as we would expect it on this Aristotelian-style theory of qualia. The theory may therefore have something to commend it; nonetheless I have spoken of it as Aristotelian rather than Aristotles in the awareness that there is no strong reason to think that Aristotle ever meant to account for phenomenal consciousness as such. 81
UNIVERSITY OF EDINBURGH

_________
81 I am grateful to Michael Pakaluk and the other organisers of the Boston Area Colloquium in Ancient Philosophy for inviting me to speak, to my hosts at Brown University, Mary Louise Gill, in particular, for warm hospitality and instructive discussion, and to Aryeh Kosman, who as the ideal commentator has helped me both correct and construct my argument. In revising the paper for publication, I have benefited from Michael Pakaluks analysis on his weblog, dissoiblogoi.blogspot.com, and the written comments of Pavel Gregoric and Julian Kiverstein. I also owe a debt of gratitude to Victor Caston not only for acute criticism of my paper but also for the stimulation of his Mind article, which has raised the bar in this area of Aristotelian studies.

COMMENTARY ON JOHANSEN
ARYEH KOSMAN

In this carefully thought out paper, Thomas Johansen bravely sets out to elucidate one of Aristotles super-cryptics; those are the texts in which its unclear not simply what Aristotles arguments are, nor what it is that they establish, but in addition what the phenomenon is that Aristotle sets out to explain in the first place. So were led to ask about the opening paragraphs of De Anima III.2: what is it to perceive that we see? as well as the more obvious questions: Does Aristotle think that whatever it is is accomplished by sight or by something other than sight? and How do the textually different arguments he brings to bear establish this or that conclusion? By overtly linking Aristotles discussion with such inner-sense theorists (or higher-order perception theorists as Peter Carruthers suggests we should call them) as David Armstrong and William Lycan, Johansen reinforces our sense that Aristotle is talking about perceptual consciousness. But what features of perceptual consciousness? One central question about how to understand Aristotles discussion, the question concerning the phenomenon he means to explain, is, I think, this: does perceiving that we seerefer to a phenomenon of consciousness, which is to be explained by a theory of inner sense, or does it refer to a phenomenon of reflective self-consciousness, the exact nature of which and whose relation to ordinary instances of consciousness Aristotle here sets out to explain? Inner sense theory argues that we have not only a set of first-order senses that generate representations of our surroundings and states of our own bodies, but a faculty of inner sense as well, which, in Peter Carruthers words scans the outputs of those first-order senses and generates higher-order . . . representations of (some of) them in turn. And, he goes on to say, these higher-order representations are responsible for the feel of our phenomenally conscious states. That is to say, our first-order perceptual states get to be phenomenally conscious by virtue of being targeted by higher-order perceptions, produced by the operations of our faculty of inner sense. 1

_________
HOP over FOR, HOT theory, in R.Gennaro (ed.), Higher Order Theories of Consciousness, Philadelphia: John Benjamins, 2004.
1

278

ARYEH KOSMAN

Such theories can clearly claim as their ancestor the theory that Socrates gets Theaetetus to agree to at Theaetetus 184c, according to which our eyes are not but rather : not that by which, but through which we see, not seers or agents of seeing, but instruments of our sight, sense organs, as we like to say. It is some inner power, whether, as Socrates delicately puts it, we should call it soul or something else, by which we should properly be said to see, using these organs, exactly as he . puts it, as instruments: Notice that here, and this is true as well for Armstrong and his cohort, a theory of second-order perception is not a theory of second-order consciousness; that is to say, its not a theory of reflective self-consciousness, but a theory of consciousness, in which the orders differ categorically, and it is only by virtue of the higher order perception that the term conscious applies to anything. First-order perceptual states play an important role in this perceptual system, so that Lucretius cheeky jibe at similar theories:
If eyes are doorways might it not be better to remove them, sash, jamb, lintel and let the spirit have a wider field? 2

doesnt apply. But first-order perceptual states by themselves, such as those of Armstrongs abstracted truck driver, are still in sense some unconscious. But my saying in some sense unconscious should alert us to the respect in which linking our reading to discussions like those of Armstrong and Lycan makes the issue more complex and subtle than Ive just made it out to be, and more complex and subtle than my discussion of thirty years ago. 3 That earlier effort in understanding this chapter was founded on a very rough distinction among three sorts of phenomena. 1) The image of an apple is projected, say upon a screen. No one would say that the screen saw the apple. 2) The image of an apple affects the functioning eye of a sentient animal, like a dog or a philosopher, such that we would want to report that the dog or philosopher is aware of the apple. 3) A philosopher, reflecting on his experience of seeing an apple, is aware of his awareness. Two connected questions then recommended themselves. A) What, according to Aristotle, is the nature of the distinction between the first two phenomena; thats the question that I think activates II.12 of the De

_________
2 De Rerum Natura, III.367-369; the translation is by Rolfe Humphries, Lucretius: The Way Things Are, Bloomington: Indiana University Press, 1969. p. 97 3 Perceiving That We Perceive: On the Soul III, 2 Philosophical Review, 84 (1975).

COMMENTARY ON JOHANSEN

279

Anima. B) Is our text in III.2 about that distinction, or is it about the nature of the third phenomenon? Johansen is unhappy with my citation of De Anima II.12 as an element in the analysis of De Anima III.2; he is of the opinion that
the introduction of self-awareness at this point in the interpretation is premature. It is true that perceiving is contrasted with merely being affected insofar as the former implies awareness. However, the awareness implied is just the awareness of the sense-object by which one is affected; it is not also the awareness of oneself being affected by the sense-object.

But I saw II.12 as of interest only as setting the issue, and took the issue to be in fact simply the issue of object awareness. That is, II.12 shows, as Johansen says, that perception is not adequately explained by understanding it to be an affection by the form of the perceptible object, even if we specify, of the object qua perceptible. This is simply to repeat the argument of the Theatetus: theres more to seeing than meets the eyeball. The question at the end of II.12 is: what more? And the question Aristotle then goes on to pose at the beginning of Book III is whether something like inner sense theory can provide the explanatory addition. The answer I suggested was: yes or no, depending on how we understand the nature of inner sense. But however we parse this answer, my suggestion was that what is at stake is the explanation of consciousness, which it is true, is assumed throughout Book II, but which now, at the end of the book, is questioned. A finer set of distinctions than the crude one on which my argument was based is generated by Lycan, who reminds us, as Johansen notes, of the variety of senses and levels of the notion of being conscious. And of even greater interest, it seems to me, is the distinction David Armstrong attends to in the case of the abstracted truck driver, a distinction again cited by Johansen, between a minimal perceptual registering and the kind of perceptual awareness that we are willing to call consciousness. For Armstrongs distinction allows us to speak of a theory of first- and secondorder perception in the service of explaining first-order consciousness, that is, in the service of allowing us to see the distinction between the kind of awareness we have on automatic pilot, and the kind of awareness we have when attending to our environment or to our tasks, neither of which is the second-order self-consciousness that occurs when we attend reflexively to the acts of our perceiving. What Im most uneasy about in Johansens discussion is that despite its appeals to an inner sense theory that is interested not in selfconsciousness, but in consciousness, of which it may give an interestingly nuanced account, it is precisely the latter, self-consciousness, with which Johansens discussion seems concerned. Thus he writes:

280

ARYEH KOSMAN

The awareness implied [in II.12] is just the awareness of the sense-object by which one is affected; it is not also the awareness of oneself being affected by the sense-object. It is implied by Aristotle throughout Book II that first-order perception is a form of awareness (aisthanesthai) of its object. De Anima III.2, however, explicitly addresses the different question by what faculty we are aware of our first-order perceptions. This is a question not about perceptual awareness as such; it is specifically a question about our awareness of our own perceptual processes, about our second-order awareness.

Thats why he takes the phenomenon that Aristotle is out to explain to be an occasional or at best incidental feature of our perception:
The necessity mentioned in II.12 is the necessity of the consequence rather than of the consequent. Aristotle is not saying that we necessarily perceive that we see when we see, or even that we always perceive that we see when we see.

But if this were true, why would the infinite regress argument unseat one of the suggested explanations of the phenomenon? Recall Aristotles argument; if it were by some other sense that we accomplish the phenomenon in question, then, Aristotle says, either there will be an infinite regress, or some sense will be of itself: (425b17). Here it seems clear that only if meta-perception, itself an instance of perception, is being considered as an explanation and condition of perception does the argument make sense. Johansen suggests:
the infinite regress seems generated by applying the premise that wherever an act of seeing is perceived the sense by which the act is seen must be other than that responsible for the first seeing. So for any sense, s, and any act of perceiving, p, if s is responsible for p and p itself is perceived, then there will be a sense s1, different from s, such that s1 is responsible for perceiving p. Apply the principle to any act of perceiving, pn, there will be a sense sn+1 responsible for perceiving pn. Note that Aristotle is not committed here to saying that any act of perception, pn, must itself be perceived; only that any act of perception may be perceived, and if it is to be so perceived, then according to 2 B) it must be explained by the stipulation of a further sense. For the regress of faculties to be generated it is sufficient that there be just one act of pn being perceived at any level n.

But why on this account will there be a regress? Johansen says the premise is this: if an act of seeing is perceived, then the sense by which it is perceived is different from the original sense that sees. So if s1s act of perceiving, p1, is perceived, its perceived by s2, and s2 will perceive p1 in p2. But then what? At that point the senses can decide to stop and order drinks all around; nothing more needs to happen. On the other hand, it may go on; s3 could perceive p2 with p3, and if the senses were stubborn or patient, this could go on quite a while. But at any moment they might decide to stop; thats what it means to say, as Johansen does, Aristotle is not committed here to saying that any act of perception, pn, must itself be per-

COMMENTARY ON JOHANSEN

281

ceived. So having grown weary of their little game, theyll call it a day: enough reflection for a while. For there to be a genuine infinite regress, there will have to be some reason requiring the series to continue. In order to see this, we need only to recall the nature of an ad infinitum argument. In Aristotle, such an argument is characteristically not about the existence or non-existence of an actual infinite; it concerns a series, in which any term of the series is essentially dependent upon a prior term. So the ad infinitum argument of Physics VIII.5 (256a4-256b2), of Metaphysics II.2 (994a1-19), of Nicomachean Ethics I.2 (1094a18-21), and of Posterior Analytics I.3 (72b 8-14). Without that dependency, there is no ad infinitum argument; the series could stop at any moment, or for that matter could go infinitely. Recall the argument Thomas Aquinas gives against those who thought that the eternity of the world was disprovable, and who made the following argument:
If the world were eternal, then generation would also have occurred from eternity. Therefore, one man would have been generated by another ad infinitum. But as Physics II says, a father is an efficient cause of his child. Therefore, it would be possible to proceed to infinity among efficient causes. But this is disproved in Metaphysics II. 4

Thomas replies to this argument as follows:

It is impossible to proceed to infinity per se among efficient causes, i.e., it is impossible for the causes that are required per se for a given effect to be multiplied to infinityas, for instance, if a rock were being moved with a stick, and the stick were being moved by a hand, and so on ad infinitum. But it is not impossible to proceed to infinity per accidens among agent causes, i.e., it is not impossible if all the causes that are multiplied to infinity belong to a single order of causes and if their multiplication is per accidensas, for instance, if a craftsman were to use many hammers incidentally, because one after another kept breaking. Then it is incidental to this hammer that it acts after the action of some other hammer. In the same way, it is incidental to this man, insofar as he generates, that he be generated by another. For he generates insofar as he is a man and not insofar as he is the son of some other man, since all the men who generate belong to the same order of efficient causality, viz., the order of a particular generating cause. 5

_________
4 Summa Theologiae, Part 1, Question 46, article 2: si mundus fuit aeternus, et generatio fuit ab aeterno. Ergo unus homo genitus est ab alio in infinitum. Sed pater est causa efficiens filii, ut dicitur in II Physic. Ergo in causis efficientibus est procedere in infinitum: quod improbatur in II Metaphys. 5 dicendum quod in causis efficientibus impossibile est procedere in infinitum per se; ut puta si causae quae per se requiruntur ad aliquem effectum, multiplicarentur in infinitum; sicut si lapis moveretur a baculo, et baculus a manu, et hoc in infinitum. Sed per accidens in infinitum procedere in causis agentibus non reputatur impossibile; ut puta si omnes causae quae in infinitum multiplicantur, non teneant ordinem nisi unius causae, sed earum multipli-

282

ARYEH KOSMAN

So without the requirement that perception be perceived, the ad infinitum argument would make no sense as an objection to Aristotles supposition that the agency of such perception be elsewhere. And if thats true, Aristotles concern cannot be with an occasional phenomenon such as reflective consciousness explained by higher perception, but must be with an explanation of perceptual consciousness for which iteration would be required. So my worry finally is this: Aristotles concern still seems to me to be with an explanation of first-order consciousness. If we are inclined to think of this as self-consciousness because of the syntax of Aristotles expression, we must understand that here self-consciousness is not the reflexive self-awareness that we experience in the special circumstances of shame, pride, embarrassment, or surprised delight at our present experience; it is simply the self-presence of all conscious experience. Some philosophers, Sartre for example, distinguish such self-presence from reflexive self-awareness by describing it as an awareness in which the self is not an object; but thats merely an attempt to mark its special nature (marked in Being and Nothingness by placing the of of objectivity in parentheses, so: conscience (de) conscience, as though to create a visible sign of objectivity sous rature.) But just as in Metaphysics Lambda Aristotle continues to speak of thought as an object of thoughtas thought of ( )so here he speaks of perception as an object of perception; but in neither case does he intend by this reflective self-awarenessmy vivid thought that I am now gazing at the gates of Providence I had so long imaginedbut the self-awareness that characterizes simple first-order consciousnessmy knowing what I am thinking, or seeing, or saying, or doing. Johansens concern on the other hand sounds as though it is with second-order consciousness, and that appearance is strengthened by remarks such as this:
Aristotles phrasing does not imply an interest in general perceptual awareness. Rather, the phrasing suggests an interest in the question of how we are

_________
catio sit per accidens; sicut artifex agit multis martellis per accidens, quia unus post unum frangitur. Accidit ergo huic martello, quod agat post actionem alterius martelli. Et similiter accidit huic homini, inquantum generat, quod sit generatus ab alio: generat enim inquantum homo, et non inquantum est filius alterius hominis; omnes enim homines generantes habent gradum in causis efficientibus, scilicet gradum particularis generantis. The distinction between causal series ordered per se and those ordered per accidens is frequently noted in medieval proofs for the existence of God; I find very clear Scotus explanation in his proof: John Duns Scotus, De primo principio. Edited by Allan Wolter, Chicago: Franciscan Herald Press, 1966. I take this to be the suggestion of Victor Caston, in his influential essay Aristotle on Consciousness, Mind, 111 (444) (2002), 785ff.

COMMENTARY ON JOHANSEN

283

aware of perceiving with specific sense-modalities, such as seeing and hearing.

But mine may be an uncharitable interpretation; for it may be the case, as I noted earlier, that the range of modalities of consciousness is more nuanced than can be captured by the simple distinction Ive here offered. There may be a mode of self awareness that involves neither the mere self presence of first-order consciousness nor the explicit self-objectification of reflective self-consciousness. Or it may be that the reading suggested by Johansens paper is designed only to capture the distinction between preoccupied truck drivers and those of us attentive to the world in a more explicit way, and to build an inner sense theory on that fact. In that case, the view that Johansen is attributing to Aristotle may be more like the view that (at least temporarily) occupies center stage in the middle of the Theaetetus, the view that we read in Epicharmus (DK B12): ; : it is mind that hears, and mind that sees; everything else is deaf and blind. That view, with its radical distinction between sensation and conscious perception, is comfortably at home with higher-order perception theories. I find such views finally less convincing than first-order representational theories. But those are philosophical considerations, not hermeneutical ones; no one promised us that Aristotle would always be right, merely that he was a god. In any case, Johansens pointing us in the direction of contemporary theories that link first-order consciousness to second-order perception seems to me to have this virtue. It suggests the possibility of a more nuanced understanding of Aristotles argument than either Johansens own traditional reading of it as concerned with second-order consciousness, as I construe it in my less charitable moments, or my fanciful reading here and in later discussions, that dismisses Aristotles talk of second-order perception as what I called a faux reflexive. It seems to me that Aristotle is about explaining consciousness, but it may be that he explains it, as I once thought was not the case, by some distinction in orders of what may be loosely called perception. What continues, however, to concern me is this question: if we assimilate Aristotles argument to this kind of explanation, can we still account for the ad infinitum argument and its apparent conclusion that what is responsible for the consciousness of seeing cannot be another sense?
HAVERFORD COLLEGE

JOHANSEN/KOSMAN BIBLIOGRAPHY
Armstrong, D. 1981. What is Consciousness. The Nature of Mind: 55-67. Ithaca. Block, N. 1995. On a confusion about a function of consciousness. Behavioral and Brain Sciences 18: 227-47. Bolton, R. 2005. Perception Naturalized in Aristotles De Anima. Metaphysics, Soul, and Ethics in Ancient Thought: Themes From the Work of Richard Sorabji, R. Salles (ed.): 228-29. Oxford. Broadie, S. 1992. Aristotles Perceptual Realism. Southern Journal of Philosophy, Supp.Vol. Spindel Conference 1992: Ancient Minds 31, J. Ellis (ed.): 137-60. Burnyeat, M. F. 1995. How Much Happens When Aristotle Sees Red And Hears Middle C? Remarks on De Anima 2.7-8 in Rorty and Nussbaum (eds.). Carruthers, P. 2000. Phenomenal Consciousness. Cambridge. _______. 2004. HOP over FOR, HOT theory. In R.Gennaro (ed.), Higher Order Theories of Consciousness. Philadelphia. Caston, V. 2002. Aristotle on Consciousness. Mind 111: 751-815. _______. 2004. More on Aristotle on Consciousness: Reply to Sisko. Mind 113: Denniston, J. 1931. Two Notes. Classical Review 45: 7-8. _______. 2005. The Spirit and the Letter: Aristotle on Perception. Metaphysics, Soul, and Ethics in Ancient Thought: Themes From the Work of Richard Sorabji, R. Salles (ed.): 245-320. Oxford. Clark, A. 2000. A case where access implies qualia? Analysis 60: 30-38. Gzeldere, G. 1997. Is Consciousness the Perception of What Passes in Ones Own Mind? The Nature of Consciousness. Philosophical Debates, N. Block, O. Flanagan, and G. Gzeldere (eds.): 789-806. Cambridge. Hamlyn, D. W. 1968. Aristotles De Anima: Books II,III. Oxford. Hicks, R. D. 1907. Aristotle De Anima. Cambridge. Humphries, R. 1969. Lucretius: The Way Things Are. Bloomington. Johansen, T. K. 1996. Aristotle on the Sense of Smell. Phronesis 41, 1: 1-19. Kosman, A. 1975. Perceiving That We Perceive: On the Soul III, 2 The Philosophical Review 84: 499-519. Lycan, W. 1987. Consciousness. Cambridge. _______. 1996. Consciousness and Experience. Cambridge. _______. 1997. Consciousness as Internal Monitoring. The Nature of Consciousness. Philosophical Debates, N. Block, O. Flanagan and G. Gzeldere (eds.): 755-772. Mellor, H. 1978. Conscious Belief. Proceedings of the Aristotelian Society 78: 87101. Moorhouse, A. C. 1982. The Syntax of Sophocles. Leiden. Osborne, C. 1983. Aristotle, De Anima 3.2: How Do We Perceive That We See And Hear. Classical Quarterly 33: 401-411. 523-533. Rorty, A. O. and M. C. Nussbaum (eds.). 1995. Essays on Aristotles De Anima. Oxford. Rosenthal, D. M. 1993. Thinking that One Thinks. Consciousness: Psychological and Philosophical Essays, Martin Davies and Glyn W. Humphreys (eds.): 197-223. Oxford. Rosenthal, D. M. 1997. A Theory of Consciousness. The Nature of Consciousness. Philosophical Debates, N. Block, O. Flanagan and G. Gzeldere (eds.).

JOHANSEN/KOSMAN BIBLIOGRAPHY

285

Shoemaker, S. 1968. Self-reference and self-awareness. Journal of Philosophy 65: 555-567. Sorabji, R. 1995. Intentionality and Physiological Processes: Aristotles Theory of Sense-Perception. In Rorty and Nussbaum (eds.): 196-203.

ABOUT OUR CONTRIBUTORS

Klaus Brinkmann is Associate Professor of Philosophy at Boston University, where he has been teaching since 1990. He was educated at the Universities of Bonn and Tbingen and at Wolfson College, Oxford. He has published on Aristotle, Hegel, Fichte, Schelling, and Jaspers, among others, and is currently finishing a book entitled Idealism Without Limits: Hegel and the Problem of Objectivity. He is the editor of Critical Concepts: German Idealism (to appear with Routledge) and is working on a translation of Hegels Encyclopaedia Logic for Cambridge University Press. Eric Brown is Associate Professor of Philosophy at Washington University in St. Louis, where he has taught since 1997, after studying philosophy and classics at the Universities of Cambridge, Pittsburgh, and, principally, Chicago. He has published articles on several different issues and figures in ancient philosophy, and he is the author of Stoic Cosmopolitanism (Cambridge, 2006). Bridget Clarke has been Assistant Professor of Philosophy at Williams College since 2003. She will join the Department of Philosophy at the University of Montana beginning in fall 2006. She was educated at Oxford and the University of Pittsburgh and has articles forthcoming on Descartes Meditations and the moral philosophy of Iris Murdoch. John J. Cleary is Professor of Philosophy at Boston College and Associate Professor of Philosophy at NUI Maynooth (Ireland). He received his B.A. and M.A. from University College Dublin, and his Ph.D. from Boston University. He was director of the Boston Area Colloquium in Ancient Philosophy from 1984 to 1988, and is the founding general editor of this series of Proceedings. He has published extensively on ancient philosophy, including a monograph on Aristotle and Mathematics (Leiden, 1995). Currently, he is writing a book on the role of paideia in ancient political thought.

288

ABOUT OUR CONTRIBUTORS

Gavin T. Colvert is Associate Professor of Philosophy at Assumption College in Worcester, Massachusetts. He received an M.A. and Ph.D. from the University of Toronto, and a B.A. from Santa Clara University. His areas of research include ethics, political philosophy, philosophy of religion and medieval philosophy. He has published essays on the thought of Thomas Aquinas, William Ockham, Jacques Maritain, Hilary Putnam and John Paul II. He is currently working on a study of the concept of practical wisdom in classical and contemporary ethics and political philosophy. Alfredo Ferrarin received his doctorate from the Scuola Normale Superiore in Pisa in 1990. He is currently teaching at the University of Pisa. He has published over twenty-five articles on the history of philosophy (esp. Plato, Aristotle, Descartes, Hobbes, Kant, Hegel, Husserl and Heidegger) in Italian, German and American journals and edited volumes. His latest books include Artificio, desiderio, considerazione di s. Hobbes e i fondamenti antropologici della politica (Pisa, ETS 2001), Hegel and Aristotle (Cambridge and New York, Cambridge University Press 2001), and Saggezza, immaginazione e giudizio pratico. Studio su Aristotele e Kant (Pisa, ETS 2004). He is preparing a book on Kant and imagination in English. Gary M. Gurtler, S.J., is Associate Professor of Philosophy at Boston College. He was educated at St. John Fisher College, at Fordham University, and at the Weston School of Theology. He has published on ancient philosophy, with special attention to Neoplatonism, including a book Plotinus: The Experience of Unity (1988). Most recently, his article The Activity of Happiness in Aristotles Ethics appeared in The Review of Metaphysics (June, 2003). Currently, his continued research on alienation and otherness in Plotinus is published in two articles, Plotinus: Matter and Otherness, On Matter (II 4[12]), Epoche 9 (2005) 197-214, and Plotinus: Self and Consciousness, History of Platonism: Plato Redivivus (New Orleans, University Press of the South, 2005). Thomas K. Johansen is Reader in Philosophy at the University of Edinburgh. He received his BA and Ph.D. from Cambridge. His publications include Aristotle on the Sense-Organs (Cambridge 1998) and Platos Natural Philosophy: A Study of the Timaeus-Critias (Cambridge

ABOUT OUR CONTRIBUTORS

289

2004). He is currently writing a book on the faculties of the soul in Aristotle. Aryeh Kosman was educated at The University of California at Berkeley, at Hebrew University, and at Harvard University. He has taught at various institutions in the United States and has been at Haverford College since 1962, where he is John Whitehead Professor of Philosophy. He is the author of a number of essays in the history of philosophy, primarily on Plato and Aristotle. Stephen Menn is Associate Professor of Philosophy at McGill University. He was educated at Johns Hopkins University and the University of Chicago. He is the author of Plato on God as Nous (Carbondale, 1995) and Descartes and Augustine (Cambridge, 1998), and articles on ancient and medieval philosophy and science, and is completing a book, The Aim and the Argument of Aristotles Metaphysics, to be published by Oxford University Press. Fran ORourke is Senior Lecturer in the School of Philosophy, University College Dublin. He has held Fulbright and Onassis fellowships, and in 2003 was Visiting Research Professor at Marquette University. His book Pseudo-Dionysius and the Metaphysics of Aquinas was reprinted in 2005 by University of Notre Dame Press. He has published widely on Plato, Aristotle, Neoplatonism, Aquinas, and Heidegger; he is currently researching the influence of Aristotle and Aquinas on James Joyce. A monograph Allwisest Stagyrite. Joyces Quotations from Aristotle was recently published by the National Library of Ireland. Ioanna Patsioti-Tsacpounidis is Professor I of Philosophy at Deree College, Athens. She studied at the Philosophy School of the University of Athens, and at the University College London. She has published on ancient Greek philosophy, especially on Aristotles ethics, psychology and epistemology, as well as on applied ethics, and on pragmatism. Her recent book in Greek examines the classical American pragmatists. She has also translated various articles and two books from English into Greek and vice versa, as well as some ancient Greek medical authors. She is currently working on a book entitled The relevance of Aristotles ethical theory in the domain of business ethics.

290

ABOUT OUR CONTRIBUTORS

Jean-Franois Pradeau teaches history of ancient philosophy in the University of Paris X Nanterre. He studied Philosophy in Paris. Phd on Platos Critias in 1995 (University of Lille). He is a member of the Institut Universitaire de France. He translated in French several of Platos dialogs (Hippias maior, Ion, Philebus, and, with Luc Brisson, the Statesman and the Laws) and published studies on Platos philosophy, among which Plato and the City: a new Introduction to Platos Political Thought (Exeter University Press, 2002) and most recently Platon, les dmocrates et la dmocratie. Essai sur la rception contemporaine de la pense politique platonicienne (Napoli, Bibliopolis, 2005). Jean-Franois Pradeau published a French translation and commentary of Heraclituss Fragments 5paris, Flammarion, 2002), and is the co-editor, with Luc Brisson, of the French translation of Plotinuss treatises in the Flammarion series (five volumes published in 2006, from treatise 1 to 37). He also published an essay on Plotinus: Limitation du principe. Plotin et la procession, Paris, Vrin, 2003. A. W. Price is Reader in Philosophy at Birkbeck College, University of London. He was educated at Winchester and Oxford. He has published two books on Greek ethics, Love and Friendship in Plato and Aristotle (1989, 1997), and Mental Conflict (1995); also papers on Greek and contemporary ethics. He is currently preparing a book on practical reasoning and reasons.

INDEX
Annas, J., 5, 229 Anscombe, G.E.M., 225 Aristotle, 1-8, 10-11, 13, 15, 17-19, 26, 32-34, 39, 41-42, 46, 51-52, 57-81, 82-86, 89-111, 113-116, 118-120, 126, 142, 155-177, 178-184, 191-197, 199212, 213-217, 219-230, 235-238, 240257, 259-276, 277-283 Armstrong, D., 235, 255, 257, 269, 277-279 Austin, J.L., 193 Bakhurst, D., 212 Bandini, M., 21 Bergson, H., 176-177 Bertrand, J-.M., 125, 132 Black, M., 173 Block, N., 259, 260, 273-274 Bolton, R., 252 Bordes, J., 5, 14 Bostock, D., 194, 218 Boswell, J., 173 Botros, S., 67 Bowersock, G., 6 Brisson, L., 131 Broadie, S., 194, 198, 202-203, 208, 213, 215-216, 218-220, 223-224, 231, 274 Burnyeat, M., 1, 22, 226, 251-252 Carruthers, P., 274-275, 277 Carter, L.B., 41 Cartledge, P., 1-2, 5, 10, 19, 30, 32, 127 Caston, V., 114-116, 118, 235, 241242, 245-249, 251, 255-256, 259-260, 264-266, 268, 270, 276, 282 Chrysippus, 109 Clark, A., 274 Cleanthes, 98 Cleomenes, 12, 30, 42 Confucius, 155 Connor, W.R., 5 Constantini, M., 132 Cooper, J.M., 66-68, 71, 218, 226 Critias, 4, 6, 28, 48-52 Croce, B., 159 Cross, R.C., 5 Dancy, J., 192, 204-207, 209, 211, 213-214, 222, 224, 226-227, 229 David, E., 8 Davis, M., 84 Dawson, D., 5 Day Lewis, C., 169, 175 Demosthenes, 125-126, 134 Denniston, J., 247 Descartes, R., 90, 120 Diogenes Laertius, 2-3, 7, 40, 44 Dodds, E.R., 50 Dorion, L.-A., 6, 21, 31 Dostoyevsky, F., 202 Eco, E., 156, 178 Eliot, G., 158-159 Emerson, R.W., 162 Eudoxus, 52 Ferrari, G.R.G., 14 Figueira, T.J., 32 Foot, P., 202 Fortenbaugh, W.W., 223 Frede, D., 93-94 Frege, G., 175 Freudenthal, J., 92 Gargarin, M., 132 Gastaldi, S., 135 Gennaro, R., 277 Gera, D.L., 5 Gill, M.L., 245 Gottlieb, P., 229 Greenwood, L.H.G., 221 Griffith, M., 14

292

INDEX OF NAMES

Gurtler, G., 84 Gzeldere, G., 259-262, 274-275 Hamlyn, D.W., 92, 117, 243 Hansen, M.H., 125 Hegel, G.W.F., 94, 99, 109 Heidegger, M., 162-163 Herodotus, 126, 155 Hicks, R.D., 242-243, 250, 248, 270 Hobbes, T., 96, 110 Hodkinson, S., 42 Hffe, O., 5 Hornblower, S., 6 Hugo, V., 202 Hume, D., 91 Husserl, E., 90, 120 Irwin, T.H., 75, 195, 200, 210, 213, 219, 224 Isocrates, 7, 9-12, 22, 24, 33, 39, 47, 126 Jacoby, F., 5, 7, 11, 33 Johansen, T.K., 237 John Duns Scotus, 282 Johnson, M., 173 Josephus, 5, 8-9 Joyce, J., 173 Kant, I., 90-91, 94, 109 Kenner, H., 169 Kenny, A., 198, 224 Kirby, J.T., 157 Kosman, A., 235, 244, 246, 255-257, 269, 275-276 Kraut, R., 194, 198, 212, 220 Lallot, J., 132 Lance, M., 210 Leroux, G., 5 Lvy, E., 132-133 Little, M., 210 Lipka, M., 6, 28, 32 Lisi, F.L., 126, 131 Lloyd, G.E.R., 163-164, 178-179

Lucretius, 278 Lycan, W., 235, 273-275, 277279 Lycos, K., 99 Lysias, 126 MacIntyre, A., 162 Maffi, A., 133 Malebranche, N., 171 Mattingly, H.B., 6 McDowell, J., 194, 209, 211, 213, 216, 218, 221, 224-226, 228, 231 Mellor, H., 235 Miller, F.D., 182 Millet, P.A., 127 Mirhady, D., 7, 27 Momigliano, A., 10 Modrak, D., 92, 121 Moorehouse, A.C., 248 Moran, R., 183 Mller, A.W., 203-204 Murchland, B., 183 Murphy, N.R., 5 Murry, J.M., 155 Newman, J.H., 191-192, 203, 211, 215, 228, 230 Nussbaum, M.C., 93, 108, 212, 213, 251, 262 Ober, J., 1, 5 Osborne, C., 240-241, 252, 270 Pakaluk, M., 203 Pangle, T., 138-140, 150 Pascal, B., 142-143, 149 Pender, E.E., 156 Plato, 1-10, 13-45, 47-50, 52-54, 66, 100, 104, 125-137, 138, 140152, 156, 170, 236, 244, 247-248 Plotinus, 109 Plutarch, 6-8, 10, 12, 27, 29, 30, 32-33, 51, 53 Polybius, 5, 8-9, 11, 42, 53 Popp, J.M., 247

INDEX OF NAMES

293

Popper, K.R., 2 Porphyry, 161 Price, A.W., 198, 200-201 Proclus, 2 Protagoras, 6, 13, 15, 19-21, 52, 100 Rebenich, S., 6, 8 Rees, D.A., 92 Reeve, C.D.C., 5 Richards, I.A., 173 Richardson, H., 219 Ricoeur, P., 180 de Romilly, J., 5, 128 Rorty, A.O., 93, 231, 251, 262 Rosenthal, D.M., 260 Ross, D., 93 Ryle, G., 103 Sartre, J.-P., 282 Schoemaker, S., 258-259 Schofield, M., 1, 5, 93, 99 Simplicius, 161, 250 Socrates, 1, 14, 26-27, 33, 40, 44-52, 66, 70, 143, 147, 149, 152, 170, 195, 243, 248-249, 278 Sorabji, R., 105, 226, 251, 262 Spurgeon, C., 176 Stanford, W.B., 156 de Ste. Croix, G.E.M., 6 Strauss, L., 22, 149-150 Swiggers, P., 156 Tigerstedt, E.N., 2, 43 Theophrastus, 2, 7 Thomas Aquinas, 281 Thucydides, 10, 23, 51, 126, 246-247 Todd, S.C., 125-127 Trampedach, K., 52-53 Treu, M., 5 Vasiliou, I., 226 Vico, G., 159-160 de Vio, T., 166 Vogler, C., 220

Watson, G., 203 van Wees, H., 8 White, N.P., 5 Wiggins, D., 203, 209, 212, 213, 215, 221-222 Wittgenstein, L., 93, 99 Woods, M., 192, 209, 217 Woozley, A.D., 5 Xenophon, 2-6, 9, 11, 27-29, 3133, 35-36, 44-45, 47, 49-52, 126 Yunis, H., 131-132, 134-135 Zeno, 98, 109