Ancient Philosophy of the Self
The New Synthese Historical Library Texts and Studies in the History of Philosophy VOL...
228 downloads
1330 Views
1MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
Ancient Philosophy of the Self
The New Synthese Historical Library Texts and Studies in the History of Philosophy VOLUME 64
Managing Editor: SIMO KNUUTTILA, University of Helsinki Associate Editors: DANIEL ELLIOT GARBER, Princeton University RICHARD SORABJI, University of London Editorial Consultants: JAN A. AERTSEN, Thomas-Institut, Universität zu Köln ROGER ARIEW, University of South Florida E. JENNIFER ASHWORTH, University of Waterloo MICHAEL AYERS, Wadham College, Oxford GAIL FINE, Cornell University R. J. HANKINSON, University of Texas JAAKKO HINTIKKA, Boston University PAUL HOFFMAN, University of California, Riverside DAVID KONSTAN, Brown University RICHARD H. KRAUT, Northwestern University, Evanston ALAIN DE LIBERA, Université de Genève JOHN E. MURDOCH, Harvard University DAVID FATE NORTON, McGill University LUCA OBERTELLO, Università degli Studi di genova ELEONORE STUMP, St. Louis University ALLEN WOOD, Stanford University
For other titles published in this series, go to www.springer.com/series/6608
Pauliina Remes • Juha Sihvola Editors
Ancient Philosophy of the Self
Editors Pauliina Remes Uppsala University Sweden and University of Helsinki Finland
ISBN 978-1-4020-8595-6
Juha Sihvola Helsinki Collegium for Advanced Studies University of Helsinki Finland
e-ISBN 978-1-4020-8596-3
Library of Congress Control Number: 2008928521 © 2008 Springer Science + Business Media B.V. No part of this work may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, microfilming, recording or otherwise, without written permission from the Publisher, with the exception of any material supplied specifically for the purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the work. Printed on acid-free paper 9 8 7 6 5 4 3 2 1 springer.com
Acknowledgements
This collection of essays emerged from a colloquium on self in ancient philosophy we organized at the University of Helsinki in summer 2003. We are grateful for the speakers who accepted our invitation to contribute to this volume, but we also wish to express our gratitude to all the other participants of the colloquium, particularly to Simo Knuuttila, Martha Nussbaum and Holger Thesleff for their presence and share in the discussions that greatly promoted this project. We thank, further, the anonymous reader of Springer for his insightful and thorough comments on the manuscript, as well as Anssi Korhonen and particularly Timo Miettinen for their most diligent work as our editorial secretaries. The editors
v
Contents
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
v
Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ix
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Pauliina Remes and Juha Sihvola
1
Part I
Approaches to Self and Person in Antiquity
1
Graeco-Roman Varieties of Self. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Richard Sorabji
13
2
The Ancient Self: Issues and Approaches . . . . . . . . . . . . . . . . . . . . . . . . Christopher Gill
35
3
Assumptions of Normativity: Two Ancient Approaches to Agency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Miira Tuominen
Part II
57
From Plato to Plotinus
4
Socratic Authority . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Raphael Woolf
5
Protean Socrates: Mythical Figures in the Euthydemus . . . . . . . . . . . . 109 Mary Margaret McCabe
6
Aristotle on the Individuality of Self . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125 Juha Sihvola
7
What Kind of Self Can a Greek Sceptic Have? . . . . . . . . . . . . . . . . . . . 139 Richard Bett
77
vii
viii
Contents
8
Inwardness and Infinity of Selfhood: From Plotinus to Augustine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155 Pauliina Remes
Part III
Christian and Islamic Themes
9
Philosophy of the Self in the Apostle Paul . . . . . . . . . . . . . . . . . . . . . . 179 Troels Engberg-Pedersen
10
Two Kinds of Subjectivity in Augustine’s Confessions: Memory and Identity, and the Integrated Self. . . . . . . . 195 Gerard J.P. O’Daly
11
The Self as Enemy, the Self as Divine: A Crossroads in the Development of Islamic Anthropology . . . . . . . . . . . . . . . . . . . . 205 Taneli Kukkonen
12
Locating the Self Within the Soul – Thirteenth-Century Discussions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225 Mikko Yrjönsuuri
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243 Name Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255 Subject Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259 Index Locorum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263
Contributors
Richard Bett Johns Hopkins University, USA Troels Engberg-Pedersen University of Copenhagen, Denmark Christopher Gill University of Exeter, UK Taneli Kukkonen University of Jyväskylä, Finland Mary Margaret McCabe King’s College London, UK Gerard J.P. O’Daly University College London, UK Pauliina Remes Uppsala University, Sweden and University of Helsinki, Finland Juha Sihvola Helsinki Collegium for Advanced Studies, University of Helsinki, Finland Richard Sorabji King’s College London, UK Wolfson College, University of Oxford, UK and New York University, USA Miira Tuominen Helsinki Collegium for Advanced Studies, University of Helsinki, Finland Raphael Woolf King’s College London, UK Mikko Yrjönsuuri University of Jyväskylä, Finland ix
Introduction Pauliina Remes and Juha Sihvola
In the course of history, philosophers have given an impressive variety of answers to the question, “What is self?” Some of them have even argued that there is no such thing at all. This volume explores the various ways in which selfhood was approached and conceptualised in antiquity. How did the ancients understand what it is that I am, fundamentally, as an acting and affected subject, interpreting the world around me, being distinct from others like and unlike me? The authors highlight the attempts in ancient philosophical sources to grasp the evasive character of the specifically human presence in the world. They also describe how the ancient philosophers understood human agents as capable of causing changes and being affected in and by the world. Attention will be paid to the various ways in which the ancients conceived of human beings as subjects of reasoning and action, as well as responsible individuals in the moral sphere and in their relations to other people. The themes of persistence, identity, self-examination and self-improvement recur in many of these essays. The articles of the collection combine systematic and historical approaches to ancient sources that range from Socrates to Plotinus and Augustine. Some contributions offer us broad overviews of the philosophical landscape around the problem of selfhood and outline innovative generalizations about the ancient approach to the topic, while others focus on particular philosophical problems, thinkers and schools. The volume also explores the influence of ancient philosophy on Western and Islamic philosophy in the medieval era. There are two basic lines of interpretation in terms of which we can try to understand the ancient philosophy of self. On the one hand, we can argue that the ancient reflections around the topic express the insights of different philosophical approaches as to what is fundamentally one and the same philosophical problem about one and the same self. When the ancient philosophers were offering different answers to questions such as, what in the human being merits the name of self, what is the fundamental truth about human nature, what defines the fundamental identity of an individual, what is the relation between what we value as our ideal goals of life and what we actually are, they, so this view supposes, were nonetheless talking about the very same thing. On the other hand, we could think that there is no unambiguous problem of the self to which there are different answers, depending upon the philosophical vocabularies that the ancient philosophers happen to use. On the contrary, P. Remes, J. Sihvola (eds.) Ancient Philosophy of the Self, © Springer Science+Business Media B.V. 2008
1
2
P. Remes, J. Sihvola
according to this line of interpretation, the ancient philosophers reflect on a wide variety of different problems. If the latter view were accepted, one could wonder whether there is, after all, such a thing as the ancient philosophy of self, if related inquiries break down into philosophical problems and dilemmas that are concerned with incomparable and incommensurable issues. For example, one may ask whether there is any common ground shared by discussions about human bodily identity, in which the determinacy and permanence of an embodied being is problematicized, and by questions about rational agency, whereby the theoretical foundations behind the idea of human beings as subjects of rational judgements and purposive plans comes to the forefront. Philosophy of self could also be understood in a new way, if we approach it from the idea that selfhood is not a single thing but a many-dimensional phenomenon. If this is accepted, we can neither explain selfhood within a single branch of philosophy nor give an unambiguous and clear-cut definition of what a self is. An example of the many-dimensional approach is Jerrold Seigel’s recent book The Idea of the Self in which three dimensions of self are distinguished: the bodily or material, the relational; and the reflective.1 Our material needs, drives and temperaments are distinguished from our cultural and interpersonal conditioning, such as shared values, and finally attention is paid to our capacity for examining and restructuring ourselves and our lives. Selfhood is located in the continuous dialectic between these aspects. The many-dimensional approach, although it has been developed outside the field of studies in ancient philosophy, may be helpful for understanding the ancient discussions, which were themselves widely dispersed in the field of philosophy without assuming that the self could be subsumed under a single, clearly structured definition. The scheme does not, however, directly lend itself to application in ancient philosophy. The interpretation of self that it yields, in any case, does not correspond well with some of the central features in the teleological and eudaimonistic framework of ancient philosophy. For example, the idea that the human striving towards a flourishing life regulates the interplay between the material, cultural and reflective aspects of the self is not so easily captured by a merely classificatory framework. As such, the significance of the many-dimensional approach lies not so much in its success in explaining particular doctrines of ancient philosophy, but in the methodological possibilities it opens. It shows that the contributions of this volume do not need to be taken as merely alternative ways of approaching the philosophical problem of selfhood. At least some suggestions that are made here could work together towards a comprehensive understanding of a phenomenon that is essentially multifaceted. It seems that some of the authors in this volume have in fact approached their topic in a many-dimensional way, as they explore a cluster of separate philosophical problems, the solutions of which are together intended to provide a more comprehensive conception of the notion of self.
1
Jerrold Seigel, The Idea of the Self. Thought and Experience in Western Europe Since the Seventeenth Century, Cambridge: Cambridge University Press 2005.
Introduction
3
Whichever interpretative line we follow, it is clear that we shall easily find numerous interesting arguments around the theme of selfhood in ancient philosophy. Among the answers offered, we find some that look familiar to us and seem to correspond with our own intuitions, whereas some ancient reflections might look strange and implausible. Even the arguments in this latter group, can, however, be philosophically interesting from our contemporary perspective, since their strangeness may lead us to challenge our own tacit assumptions and intuitions. Some of the essays refer to differences between ancient and modern thinking, but in the final essays of this volume it is also shown that the ancient discussion of selfhood had an interesting afterlife in medieval and Arabic philosophy. It is instructive to see which elements in the ancient tradition its immediate followers regarded as worthy of being preserved and further developed in their own reflections. It should also be asked whether there is such a thing as the ancient approach to the topic of selfhood, in distinction to the philosophies of self during the other periods in the history of philosophy. Behind the wide variety of viewpoints to the topic, there indeed seem to be some basic assumptions that at least most ancient philosophers share. First, in ancient philosophy the problem of self was usually discussed in terms of metaphysics and ontology with an aim to locate the self among the basic entities of reality. So the Platonists put a lot of effort into arguing that each of us is fundamentally the soul alone, not the body or the composite of body and soul (see especially Alcibiades I, 129–130), whereas the Aristotelians denied this and advocated the composite as the most likely candidate for what the human subject is (see especially DA I 4, 408b11–15). Second, although the acquisition of truth is of paramount concern for ancient philosophical schools, the notion of self is not construed as a domain of epistemological certainty, as the Cartesians later claimed.2 Rather, in antiquity selfhood has to be traced in the junctures of metaphysics, philosophical psychology and ethics. Third, at least until Plotinus and Augustine, selves are primarily understood as being constituent parts of an objective world and having capacities that, at least ideally, enable them to conceive of this world just as it objectively exists. The ancient philosophers had very little if any interest in the private and subjective aspects of human experience.3 Fourth, conceptions about the good, happiness, and flourishing create a space between every-day existence and normatively regulative ideal existence. For human beings, the latter coincides with the use and well-being of our rational or intellectual nature. The problems of self were thus approached
2
Myles Burnyeat, “Idealism and Greek Philosophy. What Descartes Saw and Berkeley Missed”, Philosophical Review 90(1982), 3–40. 3 See Stephen Everson, “The Objective Appearance of Pyrrhonism”, in Stephen Everson (ed.), Psychology. Companions to Ancient Thought 2, Cambridge: Cambridge University Press 1991, 121–147; Gail Fine, “Subjectivity, Ancient and Modern”, in Jon Miller and Brad Inwood (eds.), Hellenistic and Early Modern Philosophy, Cambridge, Cambridge University Press 2003, 192–231; as well as Amber Carpenter’s and Pauliina Remes’s contributions in Sara Heinämaa, Vili Lähteenmäki and Pauliina Remes (eds.), Consciousness: From Perception to Reflection in the History of Philosophy, Dordrecht: Springer, 2007.
4
P. Remes, J. Sihvola
within the overall teleological framework typical of ancient philosophy.4 Fifth, the very strong assumptions concerning the natural sociability of human beings, so prominent in Plato’s and Aristotle’s ethical and political arguments, also influenced the way in which the themes around selfhood were conceived. Before outlining each individual contribution, let us briefly reflect on those critical voices who have doubted the applicability of the notion of self to ancient philosophy. These criticisms can be schematically divided into three groups. First, some philosophers have claimed that there is no need to distinguish such entities as selves as ontologically or even conceptually distinct from human beings, persons or souls.5 Second, it may also be argued that there are no terms corresponding to the notion of self in ancient philosophy, and therefore there is nothing to study. A third group of critics consists of those who hold that the doctrinal differences between ancient philosophy and the more recent discussions of selfhood are so great that they are no longer talking about one and the same concept or entity.6 To the members of the first group we can say that even if there were no selves in reality, it would be important to study why earlier philosophers assumed that such things existed, just as it might be useful to study theological beliefs even if we did not believe that God exists. Moreover, the 20th century attempts to discard the notion of self have not been widely accepted. On the contrary, the concept of self
4 Several scholars have pointed out that this feature of ancient selfhood renders a conception of selfhood in antiquity which is constructed or achieved rather than simply given. It is also called an “honorary” conception of selfhood – something that is striven for, and that can be had in different degrees. Cf. Mary Margaret McCabe, Plato’s Individuals, Princeton, NJ: Princeton University Press 1994, ch. 6; Alexander Nehamas, The Art of Living. Socratic Reflections from Plato to Foucault, Berkeley and Los Angeles, CA: University of California Press 1998, esp. p. 4; A.A. Long, “Ancient Philosophy’s Hardest Question: What to Make of Oneself?”, Representations 74(2001), 19–36; Lloyd P. Gerson, Knowing Persons. A Study in Plato, Oxford: Oxford University Press 2003, esp. at 11. 5 Anthony Kenny treats “self” as a grammatical mistake, A.J.P. Kenny, The Self, The Aquinas Lecture, Marquette University, Milwaukee, WI: Wisconsin 1988; Elisabeth Anscombe claims the “I” does not refer into anything; see her “The First Person”, in Samuel Guttenplan (ed.), Mind and Language, Wolfson College lectures, Oxford: Clarendon 1975, 45–65. Kathleen V. Wilkes argues that “self”, “consciousness” and “mind” are all terms without which it is possible to cope, and indeed to cope better, and that especially the concept of ‘person’ with its connotations of different human capacities and especially the capacity to change in time is much more useful. K.V. Wilkes, “Know Thyself”, in Shaun Gallagher and Jonathan Shear (eds.), Models of the Self, Thorverton: Imprint Academic 1999, 25–38; cf. Eric T. Olson, “There is No Problem of the Self”, in Gallagher and Shear 1999, 49–61. 6 Although Charles Taylor (Sources of the Self. The Making of the Modern Identity, Cambridge, MA: Harvard University Press 1989) does not conceive his task as a refutation that there is such a thing as the notion of self in antiquity, he does emphasise the differences between ancient and, say, Cartesian approaches. The lack of the centrality of inwardness as well the commitment to a pre-existing rational order, in which the agent knows and loves, sharply distinguish ancient thinking from that of the modern.
Introduction
5
has recently become a fashionable topic within both psychology and philosophy.7 The value of the anti-self challenge lies, rather, in its insistence that there might be deep background assumptions behind certain conceptions of selfhood, inherited from our philosophical tradition, that should not be accepted at face value. The criticism of the second group is relatively easy to refute. Although there is no direct equivalent of ‘self’ in Greek or Latin, terms such as autos (‘same’, emphatic ‘himself’) and the reflexive heautos (‘himself’) often come quite close.8 Aristotle’s description of a friend as ‘another self’, allos autos, is a paradigmatic example (EN IX, 4.1166a32; 1169b7). Plotinus’ question of who ‘we’ (h¯emeis) are is a reflective turn towards an exploration on the nature of the inquirer, and involves distinguishing this study from the study of the soul (Plotinus, Enn. I.1 [13]; VI.4 [14] 16).9 It is also evident that anthr¯opos (‘human being’) is frequently used in contexts in which it does not denote species membership but something closer to ‘self’ or ‘person’ (e.g. Plato, Republic IX, 589a–b). So it seems that, at least at the terminological level, the distance between the ancients and ourselves is not unbridgeable. The members in the last group of critics, often specialists in early modern philosophy, provide, by far, the most serious challenge. They claim that the conceptual transformation has altered the philosophical landscape beyond recognition, even if there might be some superficial resemblances, e.g., at the level of terminology. It is true that Descartes used the notion of self in such a different role from that of Plato that it can be reasonably questioned whether they were talking about the same thing. Yet, the most exciting studies in the history of philosophy are often those concerned with just these kinds of transformations. They first attempt to explicate the kinds of conceptual, terminological and doctrinal similarities and differences there are between the explored historical periods or between the historical context and our own time, and then ask how the continuities and discontinuities could be explained.10 The contributions of this volume take part in this fruitful and continuing dialogue on the development of the notion of self and the theories of selfhood in the history of philosophy. *** The volume begins with a central debate concerning the outlook of ancient thinking about self and person as a whole, begun in Christopher Gill’s works Personality in 7 E.g. Ulric Neisser, The Perceived Self. Ecological and Interpersonal Sources of Self-Knowledge. Emory Symposia in Cognition 5, Cambridge: Cambridge University Press 1993; Shaun Gallagher and Jonathan Shear (eds.), Models of the Self, Thorverton: Imprint Academic 1999. Antonio R. Damasio has brought the discussion of self also to neuroscience: The Feeling of What Happens: Body and Emotion in the Making of Consciousness, New York: Harcourt Brace 1999. 8 Richard Sorabji, Self. Ancient and Modern Insights About Individuality, Life and Death, Oxford: Oxford University Press 2006, 32. 9 E.g. Gwenaëlle Aubry, “Introduction. Structure et thèmes de la traité”, in Plotin. Traité 53. Paris: Les éditions du Cerf 2004, 15–61, esp. at 23. 10 E.g. Raymond Martin and John Barresi, The Rise and Fall of Soul and Self. An Intellectual History of Personal Identity, New York: Columbia University Press 2006.
6
P. Remes, J. Sihvola
Greek Epic, Tragedy and Philosophy (Oxford: Oxford University Press 1996) and The Structured Self in Hellenistic and Roman Thought (Oxford: Oxford University Press 2006) as well as Richard Sorabji’s study Self: Ancient and Modern Insights about Individuality, Life and Death (Oxford: Oxford University Press 2006). In the articles collected here, these scholars elaborate and expand this ongoing dialogue. In “Greco-Roman Varieties of Self” (Chapter 1), Richard Sorabji expounds his view that human beings could not cope in the world unless they saw it in terms of the me and me again, and that ancient philosophers did show an interest in the individual person, and in the individual seeing itself in terms of me. Sorabji further shows that ancient philosophy exhibits a large variety of discussions on selfhood, and gives 16 examples. What is often involved is privileging one aspect of humanity as the self, that is, choosing a locus of importance within the whole anthr¯opos. Against the common assumption, Sorabji argues that the aspect chosen as the self in antiquity is seldom the soul – it can be the body or an aspect of the soul, its reason or will. In response to Christopher Gill’s classification of subjective-individual and objective-participant ideas of personality, Sorabji suggests that none of the categories of subjective, individual, objective and participant discounts any other, and that they would, rather, seem to appear in different combinations. Further, he claims that subjective and individual presuppose the objective and participatory, and that in later antiquity, there is an increasing interest in me-ness, self-awareness and individuality. Sorabji underlines the importance of individuality in, among other things, the Stoic theory of four personae. For Stoics moral decision making, he claims, presupposes an understanding of one’s individual character and position in the world. In his “The Ancient Self: Issues and Approaches” (Chapter 2), Christopher Gill takes this approach to overemphasise individuality at the expense of the ancient philosophers’ core focus, which he sees in terms of searching for objective ethical norms. He interprets the same passages with the intent of contextualising them within an ancient ethical agenda, and arrives at a notion of selfhood which is not primarily individual or subjective, but rather objective-participant. For example, the Stoic individual persona is, indeed, something that has to be taken into account in moral decision making and in the quest for self-improvement, but the aim is to make it consistent with the first persona, universal reason. The distinctive qualities of persons need to be taken into account in the moral development, but our shared rationality and ethical humanity are – or should be – the overriding aspect of who we are. Gill argues also that Epictetus’ discussions on prohairesis convey an essentially ethical point, that of developments towards complete virtue and happiness, rather than a claim about personal identity. Rather than conveying a new interest in individuality or subjectivity, these discussions should be understood as expressing an objective-participant conception of personhood. Gill puts forward a conception he has called “the structured self”. This conception combines psychological and psychophysical holism and naturalism with radical ethical claims that have their roots in Socratic thinking, and that underline the importance of rationality and coherence.
Introduction
7
A predominantly ethical approach is exemplified by Miira Tuominen’s article “Assumptions of Normativity: Two Ancient Approaches to Agency” (Chapter 3) in which the author discusses descriptive and normative criteria for moral agency in Plato and the Stoic Epictetus. Tuominen shows that for Plato, descriptive and normative criteria coincide to the extent that in failing to be a good and effective moral agent, one may fail as an agent tout court, although some minimal notion of agency applies to all sane human beings since they are nonetheless held responsible for their actions. In Epictetus, everyone is an effective agent in the sense of being capable of using impressions, of exercising prohairetic power, but sages attain extreme efficacy in reaching their goals of action. Even though normative standards of rationality apply specifically to an ideal agent, even elementary rationality involves at least some conceptions about normative notions. In general, ancient discussions on agency differ from contemporary ones in the attitudes taken towards objective moral standards. For both Plato and the Stoics, there are truths about goodness and these truths are embedded in human reason. In his article “Socratic Authority” (Chapter 4), Raphael Woolf argues that there is a fundamental asymmetry between self-questioning and the questioning of others, between first- and third-person authority. Diametrically opposite to Descartes, Plato conceives of third-person authority, of examining and interpreting the contents of other minds as something fairly unproblematic, something everyone can in principle accomplish, whereas first-person authority, the idea of the privileged view a subject has to his or her own mental states, does not carry any special weight for him. Similarly, ascribing beliefs to oneself has no special privilege: because beliefs are considered to be not just those surface opinions that the person claims to have but also the veiled ones that entail his openly uttered beliefs, the success of both self-ascription and other-ascription rely on a scrutiny of the beliefs involved, of their logical relations and entailments. This gives rise to what Woolf calls the problem of distance: since self-examination calls for distancing oneself from the fixity of our beliefs, even, and perhaps particularly, from those that constitute our identity, examined life may amount to living without a self. One critical question about selfhood concerns the relationship between ontological and epistemological claims. What, if any, is the connection between the dilemmas about the stability and persistence of our nature and questions about beliefs and knowledge? In “Protean Socrates: Mythical Figures in the Euthydemus” (Chapter 5) Mary Margaret McCabe argues that the two major themes of Euthydemus constitute the metaphysical question of what it is for someone to persist through change and the epistemological question of what it is to be able to tell the truth. In a methodologically interesting approach, McCabe shows that the significant philosophical purport of the dialogue, i.e., the postulates about persistence, change and personal identity, are explicit only in the mythical figures of Proteus, Marsyas and Cronos presented in the dialogue. The figures embody a complex discussion about persistence and truth. Persistence and systematic wisdom are interconnected, and both are embedded in the discussion of the desire to be a good and happy man, with a firm state of mind and character. Through the figures Plato argues for a particular
8
P. Remes, J. Sihvola
state of the soul, systematic wisdom, and against a merely episodic condition propounded by the Sophists. In his article “Aristotle on the Individuality of Self” (Chapter 6) Juha Sihvola explores the place of individuality in the Aristotelian hylomorphic framework. He argues that although Aristotle was in many ways much more interested than Plato in the particular contingencies human beings undergo in their lives, the expressions of an individual aspect of self are not presented as a central source of value. The restricted attention to the individual self is explained against the background of hylomorphic categories Aristotle applied in his metaphysics and philosophical psychology, both based on a strong explanatory priority of the formal and structural. Aristotle recognizes that each human soul has a distinctively individual career and life-history as far as it animates a particular body in a particular environment, but what remains philosophically most interesting in being this kind of human being is the aspect of human life that expresses universal human nature. Richard Bett’s article on “What Kind of Self Can a Greek Sceptic Have?” (Chapter 7) returns to the issue of the relation between an ontological aspect of selfhood and beliefs. He highlights the idea that for personal identity and selfhood, important are characteristics that have some degree of stability, that is, personal dispositions and tendencies. There is thus no need to postulate something other than an idea in which human beings would persist as unchangeable over time, nor anything rigidly identical over time. The problem with Scepticism is whether we are allowed to attribute even this weaker notion of personal continuity to the proponents of this school. Bett shows that the question depends upon a much debated issue concerning the extent of a sceptic’s withdrawal from belief. If the sceptic really has no beliefs at all – beyond those that consist in a mere registering of current experience – then his mental life would seem to be drastically restricted. And if the sceptic has no core commitments, if he cannot identify even with the sceptical procedure, has he any core at all? For Plato, as Woolf boldly claims earlier in the volume, minds present no special problem over bodies. In “Inwardness and Infinity of Selfhood: From Plotinus to Augustine” (Chapter 8), Pauliina Remes studies later developments within Platonism. Plotinus combines the Stoic methodology of an inward turn and the examination of one’s appearances to the Neoplatonic understanding of the existence of metaphysical layers and entities within the human mind, thus making the inward turn crucial for knowledge acquisition. Remes argues that in Plotinus this involves the emergence of the idea of privacy of every-day discursive reasoning. What is objective is universal, ontological truth which remains normatively regulative and desirable. Remes relocates, further, the significance of Augustine’s Confessions to the Western conception of selfhood. Whereas within the topic of inwardness Augustine contributes relatively modestly to the Stoic and Neoplatonic heritage, his true legacy is in making the self temporal, changing and of infinite possibilities. Studies on ancient conceptions of selfhood have had relatively little to say about Christian thinkers before Augustine. “Philosophy of the Self in the Apostle Paul” (Chapter 9) by Troels Engberg-Pedersen inquires into selfhood in Paul, discriminating
Introduction
9
two concepts of self: an abstract concept that underlies change and is the source of second-order reflection (resembling the Kantian conception), as well as a “filled I”, an identity that varies according to the act of interpretation with which it is filled (closer to a hermeneutic conception). The changing “I” in Paul can be either distinctly self-oriented and egotistical or, in the post-conversion situation, concerned primarily with others. The latter, spiritual “I” is stripped of any ethical significance of the “I” perspective: ideally, the self has no role in setting up the proper goals of action. Engberg-Pedersen’s study shows how, in Christian thought, other-concern challenges the primacy in ancient philosophy of ethical discussions that are mainly self-directed, like the preoccupation with self-control or the becoming a unified and effective agent. Gerald O’Daly’s “Two Kinds of Subjectivity in Augustine’s Confessions: Memory and Identity, and the Integrated Self” (Chapter 10) concentrates on two aspects of thinking regarding the Church Fathers: the role of memory in personal identity and the account of the self in relation to others. O’Daly shows that for Augustine, memory contributes to the formation of individual identity. This identity is something we preserve even in the afterlife. At the same time, Augustine’s theory of memory is situated within the framework of moral and religious self-scrutiny, which is dependent upon goals exterior to the “I” who engages in it. The self is discussed by Augustine in its relation to other human beings or to God. In this regard, human agency is hardly treated as unqualifiedly autonomous. Memory has God as its condition, and moral progress coincides with understanding our relation to the divine. The author explicates these linkages by a comparison of two accounts of friendships provided in the Confessions. In contrast to false friendship tainted with false sensibilities in the face of the friend’s death, Christian friendship involves objectification of grief, and thus encompasses help with overcoming self-absorption and integrating the self. The volume concludes with the medieval developments of ancient themes. In “The Self as Enemy, the Self as Divine: A Crossroads in the Development of Islamic Anthropology” (Chapter 11), Taneli Kukkonen shows how the Arab philosopher al-Ghazâlî explicitly treats the essence of humanity as an underlying principle and co-ordinator of four different expressions or distinct features of existence. Neither heart, animating spirit, intellectual apprehension nor the deliberate pursuit of worldly happiness is strictly identical with the principle with which it is associated, but nonetheless all functions refer to features of our existence. Kukkonen follows the many-dimensional approach himself by discussing such questions that are relevant for selfhood as self-knowledge and self-identification, loss of self in Islamic mysticism, as well as the role of intellect and emotions in selfhood. He detects, among other things, Neoplatonic influences in the understanding of selfhood as bipolarised between reason and body, God and evil, and the connected idea that the object of care of the self and its ideal end is the rational and divine aspect of our being. Kukkonen also revisits the theme of the significance of outside evidence in determining one’s spiritual state. In “Locating the Self Within the Soul: The Thirteenth-Century Discussions” (Chapter 12) Mikko Yrjönsuuri singles out two special areas in which the medieval
10
P. Remes, J. Sihvola
philosophers of the West interpreted and built upon ancient views. One is the soul-body relationship and the other is the topic of self-consciousness. Yrjönsuuri introduces the early medieval philosophy of the self, based largely on an Augustinian brand of the Neoplatonic-Stoic thinking, as well as the thirteenth century views of Thomas Aquinas, a relatively faithful Aristotelian thinker, and of Peter John Olivi, a radical innovator from the Franciscan camp with some Stoic leanings. Gradually, the understanding of human beings as both bodily and intellectual gives rise to questions about the unity of self. How does human multiplicity fall together into one self, and where is the centre of this self, if there is one? Peter John Olivi claims that there is a single unified centre of the self, a self-consciousness which appropriates every action of the person as its own. By emphasising individuality of human selfpresence and the way in which we attend to our thoughts and volitions as our own and dependent on us, Olivi may go well beyond ancient philosopher’s interest in reflexivity of mental states.11
11
The authors wish to thank Sara Heinämaa and Joona Taipale for their comments on the earlier versions of the introduction.
Graeco-Roman Varieties of Self Richard Sorabji
Some Recent Interpretations In a recent book,1 I argued for the idea that there is such a thing as the self. Part of what I meant by the self was the individual embodied owner of a body and of psychological states. I contrasted this conception with the idea that there is only an embodied stream of consciousness, without any owner of the consciousness. I rejected the claim that the only alternative to an embodied stream of consciousness would be some disembodied owner of consciousness, and I found the concept of disembodied ownership of consciousness problematic, even though I did not finally rule out the belief of some religions in an embodied human owner becoming disembodied. So far, even the simplest animals might meet my description of the self as an embodied individual owner. But I added something else into my account of the self, that for the preservation of a human (or higher animal) way of life, it was necessary to view the world in terms of its relation to me and me again, not just in terms of its relation to a member or members of a stream. A self, I suggested, is an embodied individual owner who sees himself or herself as me and me again, and human or higher animal life would be impossible without this viewpoint. So far, I have spoken only of the most basic metaphysics of what the self consists in. But every individual human develops a growing picture of itself, as male or female, son or daughter, American or Indian, baker or teacher, resourceful or victim of circumstances, public or private figure, subject to fate or past incarnations, or free. These pictures are not dictated by the metaphysical conception of self, which is too narrow to determine which pictures will be adopted. Nor are the pictures essential: they could be changed under pressure. But they are very important to a complete picture of selfhood, although they are typically studied separately nowadays by different philosophers. Among the most able recent studies, the metaphysics
1 Richard Sorabji, Self: Ancient and Modern Insights about Individuality, Life and Death, Chicago, IL: Chicago University Press 2006. My book benefited from the most helpful and perceptive comments of Christopher Gill, whom I take pleasure in thanking again.
P. Remes, J. Sihvola (eds.) Ancient Philosophy of the Self, © Springer Science+Business Media B.V. 2008
13
14
R. Sorabji
has been treated by Derek Parfit, the pictures of self by Charles Taylor.2 I celebrated both treatments, though entirely disagreeing with the stream-of-consciousness view of the first. So far, these claims were not historical. I do not know how many ancients thought of the self as an embodied owner of a body and of psychological states. I think Aristotle did, and such a view has been ably defended in modern times by Peter Strawson in his Individuals,3 but there were other ancient philosophers who did not think that this was the right account. As for the pictures which each individual builds into the self, many of them involve social relations. Not all do so. One may also see oneself in relation to God, to one’s past and future incarnations, or as conqueror or victim of the purely physical world. But I do not think there could be humans who did not build social relations into their idea of self, such as son, teacher or American. Christopher Gill has distinguished two approaches to selfhood, which he labels “objective-participant” and “subjective-individualist”. The four terms in these two pairs could mean different things, but I am saying that any completed view of human selfhood will be participant, since it will include social relations among other things. And I further take it for granted that Greek thinkers accepted this. The surprising thing would be if the modern West had gone so far in the direction of thinking that any individual was free to do whatever they felt like at any time that they never saw the individual self in terms of any social relation, such as daughter, teacher or American. As regards the objective versus the subjective, my basic account of the self as the embodied owner of a body and of psychological states so far would be classified, if anything, as objective, since it has nothing very subjective about it. But I have added to that the idea of individuality, and the further idea of the individual seeing itself in terms of me. The latter could be said to import a subjective element, as also could the interest in self-awareness. This does not dictate whether philosophical accounts of selfhood will contain a subjective element. As to how far Greek thinkers recognised one, my account would be that increasingly from the end of the second century, BC Greek thinkers did get interested in the me aspect of selfhood and in self-awareness. I do not see them as approaching in this a modern or Cartesian interest in subjectivity. What Charles Taylor has called Descartes’ “inward turn” may be foreshadowed in Plotinus and Augustine, but other ancient thinkers had quite other reasons for taking an increased interest in the me aspect and in self-awareness. Nor do I see the me aspect as providing any kind of core to the idea of self. Do I agree or disagree with Christopher Gill’s deployment of his pairs, “objectiveparticipant” and “subjective-individualist”? He has used these pairs in two masterly books on the self in earlier and later ancient thought, from which I have learnt a
2
Derek Parfit, Reasons and Persons, Oxford: Clarendon 1984; Charles Taylor, Sources of the Self, Cambridge, MA: Harvard University Press 1989. 3 Peter Strawson, Individuals, London: Methuen 1959.
Graeco-Roman Varieties of Self
15
very great deal. His comprehensive classification is based on the most extensive knowledge of texts and intellectual daring, and the result is of the greatest heuristic value. Each of the four headings is illuminating, and their combination into pairs forces everyone to learn new things as they discover whether they agree. Gill’s interpretation in his books and the present chapter is that the subjective-individualist view of the self is either absent from Greek and Roman thought or present in such a weak form as to be negligible. Insofar as he thinks that it is absent, I should disagree, and some of the examples below, drawn from Greek and Roman thought, are meant to show why I disagree. But the idea that the view is so weak as to be negligible is more nuanced. In pursuing this line of thought, Gill allows that the ancient view of the self is not exclusively objective-participant and he allows what I also believe, that the individual aspect can be combined with a number of the others, not only with the subjective, but also with the objective. But his classification has led me to think that any of his four headings can be combined with any. In my own account, I believe I have combined the subjective with the objective in one story, and I believe some ancient texts combine these approaches. In fact, interest in me-ness and in self-awareness would not arise, unless there were such things as persons and higher animals which could be described in more objective terms. This suggests to me that there may be certain priorities among the four terms. Giving an objective account does not commit one to noticing the subjective aspects, although without the subjective aspects, the word ‘self’ has less point and might be replaced by talk of the individual person. Any subjective account must at least presuppose that there is an objective one. Any attempt to include pictures of self must recognise the element of participation in society. Individuality can be discussed at a subjective or objective level, as Gill agrees. If one does not discuss the individual, the word ‘self’ may again have less point, except in connexion with the idea of the true self, which can be discussed in terms of humans as a type, as in Plato, as well as in terms of the individual, as in Homer and Epictetus. If these reflections are right, I would still find Gill’s headings illuminating, but would not myself put them in pairs. Gill does not agree with me that there was any shift in the thought of later antiquity towards a more subjective or individualist interest, and he supports his view by suggesting a certain exclusivity between the pairs. He allows that the presence of the objective-participative pair does not altogether exclude the subjective-individualist pair. But he speaks as if it can be used to discount it. At any rate, after urging, quite rightly, that Epictetus’ case for an inviolable self presupposes a background of objective-participatory factors, he speaks as if this is an alternative and better than, or to be understood rather than, any subjective-individual interpretation. Similarly, in the discussion of personae, the presence of an objective-participant treatment of Regulus in another context is seen as somehow discounting the interest in the individual in the theory of personae. It discounts it, if not in the sense of showing it to be an illusion, at least by showing it to be in some other way negligible. But I am not persuaded by the view that the presence of one pair can exclude or in any way discount the presence of either of the other categories. The difference between us, then, is that I do not see any of these four categories as excluding or discounting
16
R. Sorabji
any other, and I actually expect the subjective and the individual to presuppose the objective and the participatory. Although I believe that any of Gill’s four categories can occur in combination with any other, it is not so easy for each to occur in separation from any other. At least subjectivity is less likely to be discussed in separation from individuality. Many of the texts discussed by Gill are about what it is to fall under the human type, rather than about the individual. To that extent, we have in the past focused on different texts. Texts that examine the human type and exclude the individual are less likely to discuss me-ness. They might still discuss other forms of subjectivity, such as self-awareness, or forms of subjectivity that I have not brought into the discussion, such as what it feels like to have a human form of consciousness. When I say that interest increases in the individual and in me-ness and selfawareness in the later period, I do not deny that each period has accounts of each kind, so there is no more than an increase of interest. Moreover, objective accounts are still presupposed and participation in society remains as important as ever at least up to the time of Neoplatonism, when the ‘civic’ or social virtues are somewhat downgraded. But along with these presuppositions, I believe that discussions of the individual and of the subjective become more central than before. I want now to survey some 16 accounts of the self from different periods of antiquity, so that we can better assess what kinds of account there are.4 Only some of the accounts are at all subjective. The variety is astonishing.
The Varieties of Self The ancient philosophers often express their ideas of self just as we do, by the use of pronouns. They talk of ‘I’, ‘we’, ‘each’, the reflexive ‘oneself’ (heautos), or the emphatic ‘himself’ (autos), or (Plato, Laws 959b3) ‘that which is each of us truly’. Where autos is used without an accompanying noun, it sometimes demands to be translated by the English ‘self’, and autos is sometimes combined with hekastos, meaning each self. What is happening is that the self is being identified with an aspect of the human being. In some cases, the self selected is something which has been deliberately fostered, rather than being there all ready for inspection in Hume’s manner. Moreover the self selected varies not only according to different purposes and contexts of discussion. But even within a single sentence, there may be radically different aspects selected as self, because one aspect is often seen as working on another, where each is regarded as a self. This means that I cannot agree with any account which says that, for the ancient Greeks, self meant so and so, because the
4
These 16 accounts are derived from chapter 2 of my book Self: Ancient and Modern Insights about Individuality, Life and Death.
Graeco-Roman Varieties of Self
17
notion can refer to different things in the same sentence. Often, but not always, what is picked out is viewed as a locus of importance, the importance differing with different purpose and context. The self in the ancient philosophers is seldom identical with the soul. Often it is only one aspect of soul, its reason or will, for example, or a part of soul to be distinguished from the shade or ghost. In theories of reincarnation, the same soul may be successively borrowed by entirely different people, and so outlasts any one self. Sometimes the self is the body, or includes the body along with the whole person. Although the pronouns pick out only a thin self, the specifications of what the self consists in are thick, and this contrasts with some of the very thin conceptions of selfhood passed on to us by certain 17th and 18th writers on selfhood. Only some of the examples to be given below involve discussion of the individual and of me and I. Socrates awaiting execution is represented in Plato’s Phaedo (115c) as offering his friends the reassurance: I am not my body, but my rational soul. But whereas this discussion at 115c demands expression in terms of ‘me’, it is closely related to one in Plato Republic IX which is couched in terms not of me, but of what the inner human is (anthr¯opos). The two discussions are so closely related that it would be artificial to confine oneself to the first of them. But I want to pass briefly in review a whole range of discussions. 1. We can start with Homer’s Odyssey, or the possible interpolation at 11.601–603, where it is said of the dead Heracles that Odysseus is talking to his shade, but he himself (autos) is with the gods. Then I spied the mighty Heracles [his shade, but he himself enjoys festivities with the immortal gods]. This led to many Platonist discussions of what his true self was.5 The next two writers in terms of date will be mentioned below, Epicharmus and Heraclitus, who come from the fifth century BC. Heraclitus’ interest in the self is suggested by his saying that he went in search of himself and looked for the logos of the soul.6 I shall also cite both below for the denial of continuous selves. But I shall take next in the order of exposition Plato in the fourth century BC and subsequent Platonism. 2. Plato held that the true self is the reason or intellect (e.g. Phaedo 63b–c; 115c; 1st Alcibiades 133c4–6; Republic IX, 589a6–b6). This raises the worry whether the true self is sufficiently individual. Do we differ from each other in our reason in distinctive ways? On one interpretation, Plato is even conscious of a contrast between the true self and individuality at 1st Alcibiades (130d). In the passage already mentioned from Phaedo, Plato has Socrates remind the interlocutors that
5
See Jean Pépin, “Héraclès et son reflet dans le néoplatonisme”, in E.O. Bel (ed.), Le Néoplatonisme, Paris: Colloque du CNRS, Royaumont 1971, 167–192; Harold Cherniss, note c to Plutarch, Mor. De facie quae in orbe lunae apparet 944f, Loeb (vol. 12). Besides the Plutarch reference, see Plotinus IV.3 [27] 27.1–25; Proclus, in R. 1. 119,23–120,15. 6 Heraclitus, Fragments 45; 101; 116 in Diels–Kranz.
18
R. Sorabji
he is his rational soul, and this time the rational soul in question is meant to be individual and is characterised in terms of ‘me’: CR: How are we to bury you? SOC: However you like, provided you can catch me and I do not escape you. (Phaedo 115c) 3. I doubt if Aristotle agreed with Plato that our true self was the intellect, in quite the way that Plato meant. Each of the four times he reports the doctrine, he reports it as something that is thought, rather than as something to which he is committed. He does not believe the doctrine of Plato’s Phaedo, that Socrates can survive after death. So he does not believe that there will be a time when a human can become a pure philosopher, instead of an embodied social being who has to eat. At the very least the human needs the social and practical intellect, along with the theoretical intellect. I think that he puts the case dialectically for Plato’s view of the true self as theoretical intellect in Nicomachean Ethics X 7, but then in the next chapter, X 8, puts his case for the human as embodied social being who always needs to eat, although the more he uses theoretical reason, the happier he is. This fits with his view in On the Soul, that it is as wrong to say that our soul pities or is angry as it is to say that it weaves or builds. It is the human (anthr¯opos) who is angry, in virtue of having a soul.7 It fits too with Aristotle’s description of a friend as another self (allos or heteros autos). The pseudoAristotelian Magna Moralia also says ‘another I’ (allos or heteros eg¯o).8 Finally, it fits with Aristotle connecting the human with practical reason, when in Nicomachean Ethics VI 2, speaking of practical policy decisions (prohaireseis), he says, “such a source [of action] is the human”. 4. It was Plotinus, the founder of Neoplatonism six hundred years after Plato in the third century AD, who really wrestled with the problem of whether the Platonist self is sufficiently individual. He was torn between two concerns. One was that of thinking that we should not separate ourselves out from the timeless universal Intellect from which we derive, or we will lose our identity as much as those who do not know their father (V.1 [10] 1.1–17). The other concern was seeking to retain some separate individuality after all when we return to Intellect. It is only souls which do not attain that identification with intellect, but remain within time that can exercise memory, and Plotinus is anxious to show that they at least could still recognise each other through personality, even if they all received spherical bodies (IV.4 [28] 5.11–31). But what about the individuality of souls that do escape from time and achieve identification with the timeless Intellect? They can be accorded individuality only on the analogy with a theorem in mathematics, which has a certain uniqueness, but is intelligible only as part of a whole system.9
7
Arist. DA I 4, 408b5–11. Arist. EN IX 4, 1166a32, 1169b7, 1170b6; “of one’s children” see EN VIII 12, 1161b28–9; EE VII 12, 1245a30; “pseudo-Aristotle” see MM II 15, 1213a13. 9 Plotinus IV.3 [27] 2.49–58; IV.9 [8] 5.7–26; VI.2 [43] 20.4–23. 8
Graeco-Roman Varieties of Self
19
Plotinus believed we have a number of selves, and can identify ourselves with a higher or lower one. If we succeed in identifying ourselves with the timeless Intellect, which is one of the selves within, we shall have escaped to a life outside of time in which our obsession with prolonging life makes no sense (I.5 [36] 7.1–30). Augustine, who was inspired by Plotinus, was also torn in his Confessions in two directions, between on the one hand love of his mother as an individual, and hopes that his unnamed dead friend will remember him,10 and on the other hand aspiration towards a heaven in which there is no genetic relationship and no memory.11 5. In the next century, the philosopher-rhetorician Themistius repeats Plotinus’ view that it is harder to differentiate intellects than souls (On Aristotle’s On the Soul 104,14–23). Yet to the relief of Thomas Aquinas in the 13th century, he interprets Aristotle On the Soul III 5 as saying that there is a plurality of distinct human intellects illuminated by the one intellect (103,32–104,6). Thomas Aquinas quotes Themistius’ view in his attack on the Islamic philosopher Averroes. Since it is the intellectual soul that Thomas Aquinas takes to be immortal, it is important that human intellect should be not only immortal, but also individually distinct. One type of human intellect, the “active” intellect, is treated differently by Themistius as a single intellect shared by all humans, and Philoponus and pseudo-Philoponus follow him by saying of the human active intellect that it is immortal and non-intermittent only through the unending succession of mortal thinkers and their thoughts.12 Averroes went on to apply this denial of individual distinctness to all human intellect, which is why the Christian, Thomas Aquinas, had to oppose him. This famous “Averroist” controversy did not arise out of the blue in the 13th century. The seeds were sown in Plato, and the issue was brought to the fore by Plotinus, and reflected in the rival interpretations of Themistius and Averroes. 6. A number of the foregoing views about the self offer solace in the face of death, either, as in Plotinus, by making the happiest life exempt from time, or by offering immortality. But an opposite solace was offered by Epicurus who set up his school in Athens at the end of the same century in 307 BC. The soul is a bunch of material atoms, which will be dispersed at death, so there is no need to fear that we will be punished or otherwise suffer after death.13 Epicurus’ expositor in Latin, Lucretius, in the first century BC, considered the possibility that despite the dispersal of our atoms, they might come together again in the infinity of time, so that we could be punished after all (3, 843–864). Lucretius thinks that the interruption of memory solves this, but whether by preventing it being us who suffer, or by making it a matter of no concern despite its being us, is hard to determine.
10
Augustine, Conf. 9.3; 9.10. Augustine, Conf. 9.13; De vera religione 46.88–89. 12 Philoponus, in de Intellectu 52,23–29; 91,40–49 CLCAG, Verbeke; pseudo-Philoponus, On Aristotle On the Soul 3, 538,32–539,12. 13 Epicurus, Sent. 2: Lucretius 3.31–93; 830–1094. 11
20
R. Sorabji
7. The Stoics believed that I would return each time the universe repeats its history, though only Seneca regarded this as a solace. At the first mention in his letters, he suggests that our return might be seen as comforting, but later as the letters get more demanding, he offers only the solace that the preceding annihilation is shared by everything in the universe.14 On the standard view, the universe will repeat its history exactly, and its being I who return15 is guaranteed, despite the difference of time, by the recurrence of my supposedly unique and unshareable characteristic, whatever that may be.16 Variants on the view were that it would not be a same me, but a different me each time, or that it would not be me, but only someone indistinguishable from me.17 8. A different solace is the idea that there is no continuous self anyhow. Your infant self, and even your yesterday’s self, are already dead, so why lament the last in a series of deaths, when there have been so many deaths already? The idea that there is no continuous self was introduced early in the fifth century BC by the playwright Epicharmus, if the text is his, but only as a joke. The idea was that just as the number 7 is replaced when it grows to 8, so a person is replaced when he or she grows. This is used by Epicharmus’ characters, to avoid responsibility for debts incurred and for retaliation against defaulting debtors. But philosophers are good at taking jokes seriously. The idea was used later by Platonists against the Stoic School under the name of the Growing Argument and answered with what I would call a Shrinking Argument, of which I have offered an interpretation elsewhere, by the Stoic Chrysippus in the third century BC.18 Philo, the Jewish philosopher of the first century AD, objected to a certain aspect of Chrysippus’ Shrinking Argument that the survival of one person cannot depend solely on whether another person survives.19 This principle has been used in modern philosophy by Bernard Williams20 and much discussed by others. It is not impossible that Chrysippus used it himself for an opposite purpose to Philo’s. In the meantime, Aristotle had raised the problem about persistence through growth in a particular form (GC I 5, 321a18–22; b26–8; 322a28–33). If there is not to be a collision between incoming food and what receives it, the persisting receiver had better be form, not matter.
14
Seneca, Ep. 36,10–11. Alexander, in An. Pr. 180,33–36; 181,25–32; Tatian, Ad Gr. ch. 5 (SVF 1.109). 16 Alexander, loc.cit.; Origen, C. Cels. 5.20. 17 Origen, C. Cels. 5.20; Simplicius, in Phys. 886,12–16. 18 Epicharmus in Diogenes Laertius (Lives of Eminent Philosophers) 3.12; Chrysippus against Platonist Academics in Plutarch, Mor. Comm. not. 1083a–c. Interpretations in David Sedley, “The Stoic Criterion of Identity”, Phronesis 27(1982), 255–275; Richard Sorabji, The Philosophy of the Commentators 200–600 AD: A Sourcebook (vol. 3), Logic and Metaphysics, London: Cornell University Press 2004b, 6(h)(ix); see also Sorabji 2006, ch. 4. 19 Philo, On the Eternity of the World 48 (SVF 2.397). 20 Bernard Williams, Problems of the Self, Cambridge: Cambridge University Press 1973, ch. 4, discussed by Derek Parfit, Reasons and Persons, Oxford: Clarendon 1984, Part 3, note 47. 15
Graeco-Roman Varieties of Self
21
The idea that everything is in flux was supported by the Presocratic philosopher Heraclitus in the fifth century BC, when he said that you cannot step into the same river water twice (fragment 12). His follower, Cratylus, is supposed to have said that you cannot do so even once.21 Heraclitus’ view was developed by Plato who refers to Heraclitus and, for the purpose of his argument, applies the point also to persons. But Plato finishes by reducing the argument to absurdity.22 Plato further puts into the mouth of Diotima in the Symposium23 the idea that our attributes are for ever changing, so that we do not even retain the same knowledge, but have to replace it by repeating it to ourselves, as it slips away. I have not been convinced by an interesting suggestion that the Cyrenaic philosopher Aristippus, who developed Socrates in a hedonistic direction in the fifth to fourth centuries BC, prefers pleasure over happiness because of uncertainty whether there is a continuing self to enjoy the happiness.24 His argument that pleasurable motion in the soul gets exhausted over time suggests to me rather that he believes in a continuing self. The consequences for fear of death are drawn out in a Buddhist text written in Pali, the Questions of Milinda.25 This purports to be a dialogue between a Buddhist monk and a Greek king of Bactria of the second century, Menander, and it explains the Buddhist theory that there is no continuous self and the conclusion, among others, that one should not fear death. There may be traces of a similar argument having entered the consciousness of Greek and Roman philosophers in the first century AD, because this reason for not fearing death is found both in the Stoic Seneca and in the Platonist Plutarch. But if they really mean the same, it will be incompatible with the rest of what they say, which suggests that it would be an alien growth.26 They both treat it as an extension which goes beyond what Heraclitus had said. But they throw it in as an extra argument, despite its incompatibility with Seneca’s view that we have a life-long self (ego, me),27 and Plutarch’s belief that we have genuine memories which we can use to weave our lives into a unity.28 Seneca may be saying only that one’s past state and time of life has perished, and Plutarch might be expected to say no more than that one acquires a new identity in an everyday sense.
21
Arist. Met. IV 5, 1010a10–15. Plato, Theaetetus 152e–186e. 23 Plato, Symposium 207c–208b. 24 Terence Irwin, “Aristippus Against Happiness”, The Monist 74(1991), 55–82; fuller reply by Voula Tsouna-McKirahan, Cyrenaic Epistemology, Cambridge: Cambridge University Press 1997. 25 Milinda’s Questions, trans. I.B. Horner, Sacred Books of the Buddhists (vols. 22 and 23), London: Luzac 1964. 26 Seneca, Ep. 24,19–21; 58,22–23; Plutarch, Mor. De E apud Delphos 392c–e, all translated in Richard Sorabji, Emotion and Peace of Mind, Oxford: Oxford University Press 2000a, 247–248. 27 Seneca, Ep. 121,16. 28 Plutarch, Mor. De tranq. anim. 473b–474b, translated in Sorabji 2000a, 232–233. 22
22
R. Sorabji
9. Different from this is the denigration of the bodily self by the Stoic Emperor of the second century AD, Marcus Aurelius.29 He uses the phrase that Aristotle sometimes uses for a thing’s matter, ‘whatever it is that is so-and-so’, and speaks as if he were just this matter, namely flesh, breath and the governing mind. It is actually completely wrong to suppose that something is just its matter. The form or organisation makes all the difference, as Aristotle always insisted, or else we would all be Shakespeares, since we can all produce the letters of the alphabet. Marcus does not denigrate the governing mind, but he is not offering it immortal release from the body either. He is merely recommending that one imagines and welcomes the eventual ending of its enslavement to the denigrated body. 10. In the first centuries BC and AD, I believe there was an explosion of new ideas about the self. We have already reached Seneca and Plutarch, who belonged to this time, and it was a particularly fruitful era for the subject. Plutarch in the first century AD, writing about tranquillity, is the first philosopher I know to have made the connexion, which has recently been popular,30 between self and narrative, although Seneca had had some of Plutarch’s idea,31 and we shall see shortly that the Stoic Panaetius had already connected selves with personae, or roles. Plutarch’s view is that, to secure tranquillity, we need to use our memories to weave our life into a unified whole.32 Otherwise we will be like the man in the painting who is plaiting a rope in Hades. As he plaits it, he throws it over his shoulder, but does not notice that a donkey is eating it up behind him. We shall also be like the people described in the Growing Argument, who have no continuous self. We must weave in the bad parts, as well as the good, for a picture needs dark patches as well as bright, and music needs low notes as well as high. But we must not wallow in the bad parts, like beetles in the place called “Death to Beetles”. How are we to understand this argument? The appeal to the Growing Argument and its denial of a continuous self suggests that it is about how to secure the continuation of one and the same self. But on the other hand, Plutarch retreats later in the passage and recommends that we weave not a self, but a life. This suggests that his subject may be identity in a different sense, not sameness of person, but the taking on of an identity, that is, acquiring a conscious or unconscious conception of oneself based on appropriating certain past actions and experiences rather than others as one’s own. Marya Schechtman,33 without
29
Marcus Aurelius, Med. 2.2. Alasdair Macintyre, After Virtue, London: Duckworth 1981, 200–201; Taylor 1989, 47, 51; Daniel Dennett, Consciousness Explained, London: Penguin 1992, ch. 13; Marya Schechtman, The Constitution of Selves, Ithaca, NY: Cornell University Press 1996. 31 Seneca, De brevitate vitae 10, 3–6. 32 Plutarch, Mor. De tranq. anim. 473b–474b. 33 Marya Schechtman, The Constitution of Selves, Ithaca, NY: Cornell University Press 1996, 94. 30
Graeco-Roman Varieties of Self
23
reference to Plutarch, uses the very same idea of weaving the story or narrative of one’s life. Further, she sees the resulting kind of identity as not so different from the other kind because it confers a unity over time that is the very opposite of the disintegration that afflicts victims of memory loss in Alzheimer’s disease (p. 150), and, in her view, to some extent infants who as yet lack a narrative. That Plutarch has in mind the same topic as Schechtman, the adoption of an identity by a person who already has a continuous history, is suggested also by the fact that the memories to be woven into the life are meant to be genuine memories of what the selfsame person experienced earlier. As to whether Plutarch’s advice on achieving tranquillity through narrative is sound, I think it often may be, but, as Plutarch recognises, those who have been through traumas will not find memories congenial and will need far more help. Again, it is perhaps even more important to weave in future projects. The late Russian neuropsychologist, A.R. Luria, wrote a book about a man who had lost his memory of who he was through being shot in the brain in the Second World War. His life’s project was trying to discover who he was and writing a diary at the rate of about a word a day describing his efforts. Luria comments that those of his patients who lost the ability to plan future projects disintegrated far more than those who had lost their memories.34 11. Also in the first century AD, we find explications of the Stoic theory of Justice as based on a naturally felt attachment (oikei¯osis). Justice is argued to be natural, because based on the natural attachment that the newborn feel first for their own persons and then for their nearest. The Stoics therefore examine the newborn closely and Seneca argues that, although at any time the attachment is felt towards the current constitution, which will change, it is the life-long self (me, ego) that is entrusted to my concern.35 Around the end of the first century, Seneca’s Stoic near-contemporary, Hierocles,36 discussing the same subject, imagines the mind (dianoia) as being the centre point of a set of concentric circles. He thus equates the mind with the self (each of us, hekastos h¯em¯on) which is entirely surrounded by circles. What draws the circles, to express the degrees of attachment it feels, is also a self (a given self, autos tis), and this is described not so much as identical with the mind as possessing it (heautou). Perhaps it is the composite of mind and body. The first circle outside the mind includes one’s body to which a sense of attachment is directed, and circles further out represent other people. If Hierocles thinks, like Chrysippus37 and Seneca, that the first target of attachment is a self, then one self will feel attachment to the body as self, and altogether three aspects of the person will have been treated as selves.
34
A.R. Luria, Man with a Shattered World, London: Jonathan Cape 1973. Seneca, Ep. 121,16. 36 Hierocles, Elements of Ethics in Wachsmuth and Hense (eds.), Stobaeus: Florilegium (vol. 4), 671, lines 7–16. 37 Hautous, Chrysippus ap. Plutarchum, Mor. Sto. Rep. 1038b. 35
24
R. Sorabji
It has been said that the idea of the body as external to the mind does not come up in epistemological contexts in ancient philosophy,38 but it is common enough in other contexts, another example being the highly influential “inner man” of Plato, Republic Book IX. But Hierocles gives us a particularly vivid example. The circles further out represent one’s family, friends, fellow-citizens, and foreigners. The context is ethical: one should learn to pull the circles inwards, so that one feels as much attachment to family as to one’s own body, as much to friends as to family, and so on. The Stoic idea reached the Christians, and we find Tertullian, who died c. 220 AD, saying that there is a natural progression from giving to one’s brothers to giving to everyone.39 12. In the same century as Plutarch, Seneca and Hierocles, the Stoic, Epictetus, gives us the idea of the inviolable self. Epictetus had had his leg broken when he was a slave. He imagines the following dialogue. “I will put you in chains”. “What did you say, man? Put me in chains? My leg you will put in chains, but my will (an imperfect paraphrase of prohairesis) not even God can conquer”.40 The implication is that he is his will, although this needs qualification. Epictetus distinguishes good and bad will, and his idea is that ideally, but not without effort and training, he could identify himself with the one inviolable thing, his unperverted will. Your will can be frustrated only by an opposing will in you, but a good will cannot be frustrated at all. The will is something that the tyrant cannot put in chains. Since the will can be moulded, the self is also, on Epictetus’ view, something that can by effort be fashioned. The Stoic Hierocles went in the opposite direction from his older contemporary Epictetus when he expanded the concept of ‘mine’ so that it could extend to the whole of mankind. But whereas Hierocles recommends this for social reasons, Epictetus in effect rejects it as making one violable, even though one owes justice to all humans. 13. Towards the end of the second century BC, an earlier Stoic, Panaetius, is probably the source for Cicero’s account of personae in the first century BC.41 This is a view about what you must take into account in deciding what it is right to do. You must consider not only the fact that you are a rational being. That is only the first persona, although it is what Kant tells us to consider, when deciding how it is right to act. Complaints have been made about how thin a conception that of rational being is for the purpose of moral decision making.42 It is not that of being a rational Indian or American, nor that of being a rational male or female. Panaetius wants a thick concept. You need to make moral decisions in the light also of your individual persona, that is, of the position you have been born into, 38 Myles Burnyeat, “Idealism and Greek Philosophy: What Descartes Saw and Berkeley Missed”, Philosophical Review 91(1982), 3–40, see 29–30. 39 Tertullian, Adv. Marc. 4.16. 40 Epictetus, Diss. 1.1.23; cf. 1.18.17; 1.19.8; 3.18.3; 4.5.23; 4.5.12; 3.1.40; Handbook 9. 41 Cicero, Off. 1. 107–115. 42 Bernard Williams, Ethics and the Limits of Philosophy, London: Fontana 1985; Macintyre 198.
Graeco-Roman Varieties of Self
25
the choices you have made and what fortune has brought you. These personae are again, some of them, chosen, rather than being already there awaiting introspection. Acting according to personae has something in common with acting according to a life-narrative. It is further said that when Julius Caesar defeated his opponents at Thapsus, it was right for Cato to commit suicide in those circumstances, even though it would not have been right for anyone else in the same circumstances. This again sounds like the opposite of Kant’s view, for Kant tells us that something cannot be the right thing to do, unless it would be the right thing for anyone to do in the same circumstances. Such a view is already suggested by Aristotle’s account of natural, as opposed to legal, justice, as conceived by Empedocles. What is just for one is just for all (Rhet. I 13, 1373b14–17). In one way, I presume that Panaetius would agree that if there were anyone exactly like Cato (and a Kantian could make that one of the relevant circumstances), then it would be right for that person in those circumstances to commit suicide. But that is not the interesting point. The interesting point is that there was no one else like Cato among those defeated at Thapsus. He had always stood for a kind of austerity that no one else began to match. Cicero calls it gravity. One could say that he is presented as an example of authenticity. And that is why it would be right for him, but for none of the others defeated, to commit suicide in the circumstances that prevailed. That this is the interesting point about people has been brought out in modern times by Peter Winch.43 Cicero would admittedly have had an extra reason for insisting that the need for suicide applied to no one else, given that he too was among the losers in the civil war, was a leading defender of the Republic and was also endowed with gravity of a kind, although he was not involved in the precise circumstances of the battle of Thapsus. Epictetus in the next century agrees in Discourses 1.2. The general point here and in Cicero (Off. 1.109, 112–113) is that different characters may call for different action in the same circumstances without one way of action being treated as right or wrong universally, independently of character. Epictetus brings out, incidentally, what a commitment he took it to be to adopt the persona of philosopher. It was then a badge of being a philosopher to wear a beard, and Epictetus says it is better for a philosopher to have his head cut off than his beard.44 Henry Chadwick has described the philosopher’s beard as exercising the same attraction and repulsion as the modern priest’s dog collar.45 Its origin is said to have lain in a misconception that arose when the clean-shaven Romans were smitten by a delegation from Athens in 155 BC of three bearded Greeks who were the leading philosophers of the day. The Romans assumed that the beard was the mark of a philosopher.46 43
Peter Winch, “The Universalisability of Moral Judgement”, The Monist 49(1965), 196–214. 1.2.20; cf. 3.23.4–5. In modern times, Robert Nozick recognises both the possibility of contributing to self-formation and of doing so in the light of persona or profession. See Nozick, Philosophical Explanations, Oxford: Oxford University Press 1981, 106. 45 Henry Chadwick, The Early Church, Harmondsworth: Penguin 1967. 46 Nicely told by John Sellars, The Art of Living: The Stoics on the Nature and Function of Philosophy, Aldershot: Ashgate 2003, see introduction. 44
26
R. Sorabji
14. Porphyry’s Isag¯og¯e, or Introduction to Philosophy, provides yet another view of the individual person. The individual is here defined from the more impersonal point of view of the logician as a bundle of qualities that cannot be shared by any other individual.47 15. I will take a further example from the Platonist tradition. Plato in the Symposium (206e–209e) suggests that we want to have children, or to leave other works behind, because this is the nearest we can come to immortality. Here the self seems to be invested in offspring. But Plotinus makes the opposite point, already noted, which also contains some truth, when he describes how souls emanated from Intellect, because they wanted to belong to themselves (heaut¯on einai). In fact, however, this search for selfhood got them the opposite result. For in forgetting their origin and their Father, the Intellect, they failed to see themselves (heautas) or to value themselves (heaut¯on). Plotinus is making the point that an important part of one’s conception of oneself is a conception of one’s parents or forebears.48 16. A final context in which there is relevant talk of self, as I have mentioned, is that of self-awareness. A star example is provided by Augustine’s Cogito argument which appears in many of his works starting a little before 400 AD. In the version found in On the Trinity (10.10.14 and 16) the argument, as in Descartes twelve hundred years later, has two parts, and the second part is about the nature of the self, and not only about awareness of it. Augustine’s first Cogito argument is that one knows about one’s own thinking with certainty. The second is that such knowledge implies one knows one’s essence, but since one does not know whether it is bodily, bodily nature cannot belong to one’s essence, which must be incorporeal. In Descartes, these arguments are couched in the first person singular, “I think”, but in Augustine they are not. The first person formulation is more dramatic and rhetorically effective. On one view, Descartes’ version of the first argument actually requires the first person, because, on this view, “I think” is to be compared with “I promise”.49 By saying “I promise” in the right circumstances, you actually do promise, whereas saying “he promises” creates no commitment. In my view, however, the argument that I cannot be mistaken in supposing “I think” turns on the point that “I think” claims no more than is presupposed in its being thought. Lichtenberg was to complain that all I can meaningfully say is “there is thinking”.50
47
Porphyry, Isag¯og¯e 7, 6–24. Plotinus V.1 [10] 1.1–17. 49 Jaakko Hintikka, “Cogito, ergo sum: Inference or Performance?”, Philosophical Review 71(1962), 3–32. 50 G.C. Lichtenberg, Schriften und Briefe. Sudelbücher II, München: Carl Hanser Verlag 1971, 412. 48
Graeco-Roman Varieties of Self
27
Earlier and Later Developments in Ancient Concepts of Self True Self, Whether Type or Individual I have identified treatments of the self as starting in antiquity, unlike the many authors who believe the idea to belong to a much later age. But what is true, I think, is that there were earlier and later developments in the concept of self. The idea of a true individual self goes back as far as Homer, with the distinction between Heracles’ shade (eid¯olon) and his true self. The First Alcibiades ascribed to Plato discusses the true self of humans in general, and Epictetus the true self once again of the individual. As already remarked, individuality is a subject of interest at various different periods.
Individual Personal Identity Over Time The interest in individual personal identity over time became urgent only later in philosophers of the third century BC onwards, when objections were raised to the Pythagorean and Stoic theory that the same people will reappear when the universe repeats its history exactly, and when the Epicureans conceded that there might in the course of infinite time be chance reassembly of a person’s atoms. The Epicurean Lucretius, who discussed this in the first century BC, almost certainly influenced John Locke,51 who has been taken to represent a brand new modern treatment of personal identity. Dispensing with Christian reliance on the idea of a continuing soul or substance, Locke appealed to memory as a source for deciding questions of personal identity over time, and this influenced many modern treatments which also rely on psychological linkages instead of on a continuing substance. But Lucretius, whom Locke read, had made the same move of appealing to memory instead of a continuing soul or substance. Before that, a threat relevant to personal identity had been raised in the fifth century BC, with the idea that all things are discontinuous. But persons, we saw, provided only one example of this threatened discontinuity.
The Idea of Me The Socrates of Plato’s Phaedo already puts a stress on the idea of me, when Socrates implies, “my dead body is not me”. But the interest in the idea of me is much more widespread later. Someone might think that all the talk of me could be
51
James Warren, “Lucretian pangenesis Recycled”, Classical Quarterly 51(2001), 499–508; see also Sorabji 2006, ch. 5.
28
R. Sorabji
replaced by more “objective” talk of persons, or humans, and aspects of them. Certainly not all. That would completely ruin many of the points being made. The existence of persons, humans and their aspects is rather a presupposition in these cases. Epictetus is not saying, “you cannot chain the person, or human”. Of course you can. Nor is he saying, “there is an aspect of me you cannot chain, my will”. The tyrant knew that. He is rather saying, “the aspect you cannot chain is me”. Similarly, Plutarch is not saying that a person or human can be woven. Those are already there. What he is tempted to say is that the thing which can be woven is me. Nor is Socrates in the Phaedo reassuring his friends that some person will continue to exist. He is saying, “it will be me”. Cato is precisely not to think only about what it would be right for a person to do. He must also ask himself, “What would be right uniquely for me?” Seneca and Plutarch seek to relieve their own anxiety by denying that there is a continuous me. Marcus does so by denigrating the bodily me. Hierocles wants us to extend our ‘me’ concerns to other people as well. To the Epicurean Lucretius, the alarming thought is that it might be we who are by chance reassembled after death. And to Seneca it is encouraging that it precisely will be we who return when the universe repeats its history. Plotinus was clearly anxious to avoid a total loss of individuality, in his discussion of souls retaining recognisable personality, or at least having the uniqueness of theorems. And this anxiety derives, of course, from ‘I’-thoughts.
Self-Awareness The interest in another subjective element, self-awareness, had a chequered career. Plato’s Charmides made the case for self-knowledge being impossible (167a–169c) and his First Alcibiades, if it is his, argues that self-knowledge needs to be achieved by seeing oneself reflected in another person (132c–133c), a theme taken up by Aristotle and most strongly in the pseudo-Aristotelian Magna Moralia (II 15, 1213a13–16). For a head-on denial, and assertion that instead the soul has selfknowledge by being present to itself, we have to wait more than seven centuries to Augustine’s On the Trinity.52 Plato’s insistence on self-knowledge through another person strengthens the relevance of participation in society. It is also supported, against the tradition of Augustine and Descartes, by modern developmental psychology, which argues that the infant only gets the idea of its own gaze by recognising the possibility of divergence from its carer’s gaze, in a game of “are we looking at the same thing?”, played uniquely by infants of the human species at the age of about nine months, and essential to the subsequent full acquisition of language.53 This theory reverses 52
Augustine, De trin. 10.3.5; 10.7.10; 10.8.11; 10.9.12; 10.10.16 (72–85). Michael Tomasello, “On the Interpersonal Origins of Self-Concept”, in Ulric Neisser (ed.), The Perceived Self, Cambridge: Cambridge University Press 1993; see also his The Cultural Origins of Human Cognition, Cambridge, MA: Harvard University Press 1999. 53
Graeco-Roman Varieties of Self
29
Descartes’ idea that I know my own psychological states, but with others I see only their hats and coats. The theory also makes self-awareness possible only through participation in social activity. In these discussions of self-awareness, Plato is talking in the First Alcibiades of knowing what type of thing a human is not of the knowledge of individuality. But there were others, the Pythgoreans, who practised self-interrogation of the individual self, perhaps as early as the fifth century BC, and passed the practice on to Seneca and other Stoics and eventually through Origen to Christianity.54 Socrates in the fifth century BC is also represented as encouraging the individual to self-interrogation. But although it is intended to improve character, it remains at a very much more intellectual level than in the more thoroughgoing self-interrogation of Pythagoreans, Stoics and Christians. Self-inspection developed a different role again in Neoplatonism from the third century AD and hence in the Christian Augustine in the late fourth century. They find God and higher reality by looking within themselves. Augustine tells us that he learnt his technique of looking inwards from the “books of the Platonists” (Conf. 7.10.16). He had read Latin translations of Plotinus, who repeatedly tells us to look for the chief divinities, the Intellect and the One, within ourselves,55 and who provides a famous autobiographical account of experiencing union with the divine Intellect, an account in which he says that he often withdraws into himself.56 It was a controversial view that the intelligible realities reside within us. Plotinus’ pupil, Porphyry, tells us in his biography of Plotinus, Life of Plotinus (chs. 18 and 20), that he was made to re-write his essay three times, with counter-essays by another pupil, before he came to acknowledge that the intelligibles lie within us. This is not a selfevident interpretation of Plato. Despite the interest of Plato’s Socrates in the saying of the Oracle at Delphi, “Know thyself”, the intelligibles on which Plato concentrates, the Forms or Ideas, are not spoken of as lying within. It took the Neoplatonist commentators on Plato to interpret in this spirit the First Alcibiades. They conclude that the starting point for studying Plato and the whole of Philosophy must be the Delphic injunction, “Know thyself” (Proclus in Alc. 5,13–14, following Iamblichus, ap. Proclum, in Alc. 11,11–17, and similarly, Anonymous Prolegomena to Platonic Philosophy, ch. 26). But Plotinus was not the first to look for truth within. Cicero in the first century BC borrows from the Stoics, when he says in connexion with natural law and justice that one finds them within (in se) (see Leg. 1.22.58; cf. Rep. 3.22.33). And in the next century, the Stoic Epictetus (Diss. 3.22.38–9) says that we find preconceptions about the good within ourselves. But the Stoics are not necessarily the first either. Among the earliest Presocratic philosophers of the fifth century BC,
54
Seneca, De ira 3.36; Epictetus, Diss. 4.6.35; Origen, Fragment on Psalms 4,5, ed. Cadiou (p. 74), 1936. 55 See I.6 [1] 9.8; V.8 [31] 10.31–43; VI.9 [9] 7.16–23. 56 See IV.8 [6] 1.1–11.
30
R. Sorabji
Heraclitus says, “I went in search of myself” (fragment 101 in Diels-Kranz) and “You would not by travelling every path find the furthest limits of the soul, so deep is its logos” (fragment 45). If this logos is the same as the universal truth that he also calls logos in fragments 1 and 2, he will have anticipated the view that the most important truth is to be found within.
Participation in Society Participation in society was never denied, although it became slightly modified in Neoplatonism, where Plotinus and Porphyry may seem to recommend that the philosopher should transcend the “civic” or social virtues and turn away from the world to the purificatory or contemplative virtues.57 Some ascetics and hermits might become more interested in their relation to God than to other humans. But the modifications in Neoplatonism were small. Plotinus and Porphyry still saw themselves as teachers and some later Neoplatonists saw themselves as priests.
The Classification of Ancient Treatments of Self I have been saying that Gill’s four aspects of selfhood may crop up in different combinations at different times, but that there are shifts towards new directions of interest. It seems to me that interest in the individual, in the idea of me and in selfawareness increases over time. Only some of the 16 cases I surveyed briefly earlier would illustrate this tendency. Gill takes two of those that might and interprets them differently. The two chosen are Panaetius’ idea of persona and Epictetus idea of the inviolable self.
Personae The first persona is not individual but species-wide. It is rational human nature and, as Gill rightly says, the ethical requirement that we take that persona into account in all our decisions is very important. I think the requirement is important because it is meant to rule out the kind of modern individualism that Charles Taylor has
57 Plotinus I.2 [19] 7.13–28; Porphyry, Sent. 32, 30,1–31,11, Lambert, discussed in Richard Sorabji, Philosophy of the Commentators 200–600 AD: A Sourcebook (vol. 1), Psychology, London: Duckworth 2004a, ch. 17a. See also John Dillon, “An Ethic for the Late Antique Sage”, in Lloyd P. Gerson (ed.), The Cambridge Companion to Plotinus, Cambridge: Cambridge University Press 1996, 315–335. For qualifications, see Dirk Baltzly, “The Virtues and ‘Becoming Like God’: Alcinous to Proclus”, Oxford Studies in Ancient Philosophy 26(2004), 297–321.
Graeco-Roman Varieties of Self
31
called “the malaise of modernity”58 – the idea that each person should be free to do their own thing compatibly with others also doing their own thing. On the contrary, in the Stoic view, human nature calls on a person to behave in accordance with right reason and with virtue, or at least to progress in the direction of virtue by performing right actions. But individual differences come in, because that injunction leaves us with too little guidance about which we should pursue among the many ways of performing right actions. We have noticed this complaint being made in modern times about Kant’s injunction that one should act in the way that a rational being would act. That leaves too much open to guide us. The late Stoic theory of personae avoids this criticism, because it supplies the guidance of more specific personae. There is not one right way of acting. Each of us must also take into account his or her individual talents and temperaments, for example. This already displays an interest in the individual. These talents and temperaments will not be shared by all humans, and each individual must consider his or her own. Admittedly, they will not normally be unique and I do not need to claim that they are. But there are two cases in which the texts do suggest uniqueness, I think. One is that in the Greek war against Troy only Achilles could fight a duel against The Trojan Hector (Epictetus, Diss. 3.22. 7–8). The other example is in Cicero. Gill holds that Cicero does not call it a case of uniqueness. But Cicero implies uniqueness by using the phrase ‘in the same situation’ (in eadem causa), and ‘in a different situation’ (alia in causa). No one but Cato should commit suicide in the same situation, the situation in which there was a surrender to Caesar, who would become tyrant and end the Roman Republic. And this difference of natures has such force that sometimes one man ought to commit suicide, while another in the same situation (in eadem causa, only in some mss) ought not. For was Marcus Cato in a different situation (alia in causa) from the others who surrendered to Caesar in Africa? But perhaps with the others it would have been attributed to moral failure if they had killed themselves, because their lives had been less austere and their habits more easy-going. Since nature had conferred on Cato an incredible gravity, and he had strengthened it by unceasing consistency, and had always persisted in his resolved purpose, it was right for him to die rather than to look on the face of a tyrant. (Off. 1.112)
I should admit that the comparison between Cato and Cicero in my earlier book, as both having kinds of gravity and both having championed the Republic and lost in the civil war against Caesar, does not show that they were both in quite the situation envisaged, because Cicero was not defeated in the battle of Thapsus. But given the situation of the battle, it was unique to Cato that suicide was the right course, because his character was unique among those defeated there. The interest here is not only in the individual, but in an individual whose character in the situation was unique. The quite different example of Regulus’ exemplary conduct, cited by Gill, is not relevant to what I want to say, both because I am not claiming that interest in the
58
Charles Taylor, The Malaise of Modernity, Concord, ON: Anansi 1991.
32
R. Sorabji
individual is always in uniqeness, and because the case of Regulus is not introduced by Cicero in order to explain personae, and so he does not trouble to tell us what personal considerations may have been added in to his heroic decision to return to his death in order to keep his word to the enemy. The presence of an objectiveparticipant example in the different context of Regulus cannot remove the interest in the individual in the theory of personae. Gill rightly points out that participation in society is involved in many decisions, including Cato’s decision on suicide. I would go further, because I think all four of his headings are involved. The personae of human rationality and of individual character might both be classified as objective. But the subjectivity of self-awareness hovers in the background, because the personae are introduced in order that we may attend to our own personae in making decisions. But this illustrates my suggestion that the four useful categories are best not separated into pairs, because they can all combine. I agree further with Gill that the first persona of rationality is not a rival of the others, but is meant to fit seamlessly with the other personae. For the other personae tell you in what way to exercise your human rationality. But I nonetheless think that human rationality is not best classified in his way as individual, since it is shared by all. The interest in the individual is revealed by our having to reflect on our further personae which are not shared by all. And that interest in individuality is once again combined with something else, an interest in the rationality that is not individual, but common to all. We may be able to see where this new interest in individuality comes from. For Panaetius is quoted by Cicero as switching attention from the earlier Stoic portraits of the ideal type, the Stoic sage. The Stoics had come under attack on the grounds that no one had yet existed who lived up to that model.59 So Panaetius urged that we should look at ordinary people who are merely progressing towards virtue. I think that gave later Stoicism its interest in all the varying anxieties and foibles of ordinary people, which makes it more immediately relevant, in my view, to modern daily life than any other ancient ethical system and possibly than any Western system. Here is what Panaetius is reported is reported by Cicero as saying: But since life is lived not in company with perfect people who are truly sages, but with those who are doing very well if there are likenesses of virtue in them, I think this too must be understood, that no one should be entirely neglected in whom any hint of virtue appears. (Off. 1.46)
I have one qualification. It was pointed out to me by Alex Cohen that there is an earlier example of Socrates in Plato’s Crito basing a decision on what is in effect individual persona. Socrates is presented there as refusing to escape from prison in Athens partly on the ground that he had never left Athens except on military service. This example could actually have influenced Panaetius. But it is a single example whose relevance needs to be detected. Panaetius has made this kind of case the basis of a whole ethical theory. 59
Philodemus, Sto. col.XII, 4–6; Sextus, Math. 7. 432–434.
Graeco-Roman Varieties of Self
33
Epictetus’ Inviolable Self The other example of mine that Gill takes up is Epictetus’ idea that I can make myself inviolable, if I identify myself with my will rightly directed only to my character and rationality. The tyrant can then override my body or any other possession, but me he cannot override. I have argued that this represents an interest in the individual and in the subjective idea of me. Gill classifies Epictetus’ view instead as objective-participant, but once again I think that all four of these useful categories are present. I would find the interest in social participation not only in Epictetus’ broader programme, although Gill is right to find it there, but even closer at hand. For the inviolable will must be directed to my character only and the good character towards which one progresses will most typically manifest itself in interaction with other people as brother, mother, or citizen. There will be exceptions where it manifests itself in one’s attitude to God or to solitary death, but these are exceptions. There is a qualification.60 An earlier Stoic, Antipater had said that the goal was the good character, whereas the right interaction with other people was only the subject matter (hyl¯e) in which that virtue was exercised. So although the participation in society is essential, its role here is subordinate. What I am not persuaded of is that the relevance of social participation is to be seen as an alternative which gives us a better interpretation, one that we should take rather than attending to the interest in individuality and me-ness. The appeal to the objective-participant background here is like the appeal to the non-individualised case of Regulus, in the earlier consideration of personae. In some way the presence of an objective-participant interest at one point is presented by Gill as discounting the presence of a subjective-individualist one elsewhere. But I once again see all four factors as equally present, not only the individual, the subjective me-ness and the participatory, but also the objective. The training required to identify yourself with a rightly directed will is described by Epictetus,61 and it involves sending a physical trainee out into the physical streets to observe his own reactions to the people he meets there. Is the bereaved father encountered really in a bad state, or the consul in a good one? This trainee is what I called an individual owner of a body and of psychological states, most relevantly of impressions about what is important in his or her daily encounters, impressions which call for reflection and correction. The physical body is, if anything, something to be classified as objective. At the same time, the need for reflection brings in another kind of subjectivity. My conclusion is that when I say that there is a greatly increased interest in later periods of antiquity in me-ness, self-awareness, and individuality, these new emphases do not give us a subjective-individualist pair that replaces interest in
60 61
Cicero, Fin. 3.22; Plutarch, Mor. Comm. not. 1071 A–B. Epictetus, Diss. 3.3. 14–19.
34
R. Sorabji
social participation and things objective. In the two examples under discussion, I see all factors as equally present. King’s College London, UK Wolfson College, University of Oxford, UK and New York University, USA
The Ancient Self: Issues and Approaches Christopher Gill
In this essay, I consider the implications of two themes in ancient philosophy as a way of exploring current issues and approaches in the study of the self in antiquity. The first theme is Cicero’s presentation of Cato in On Duties (1.112), considered in the context of his account of the theory of the four roles (personae). The second is Epictetus’ recurrent theme in the Discourses that we should focus our lives on exercising prohairesis (rational agency or will), taken together with his three-topic programme of practical ethics. Both examples fall within the area of HellenisticRoman practical ethics and are based on Stoic thinking. In examining these themes, I draw on the frameworks developed in two books on ancient conceptions of personality or selfhood, one centred on Homer, Greek tragedy, Plato, and Aristotle, and the other on Hellenistic and Roman philosophy.1 These two themes are also treated by Richard Sorabji in his contribution to this volume and in a recent book on the self,2 and I consider certain salient differences between our treatments. Sorabji also discusses my interpretation in his essay and so the two chapters constitute a kind of dialogue between our approaches to this topic.3 I also refer to other recent treatments of Hellenistic and Roman thinking on the self, including those of Michel Foucault and A.A. Long. In a topic as complex and many-layered as study of the ancient self, it would be naïve and misguided to suggest that there is a single “right” approach. In presenting my standpoint alongside that of others, my aim is to illustrate certain representative approaches to this fascinating – but rather elusive – topic and to reflect on the conceptual and interpretative issues raised by these approaches.
1
Christopher Gill, Personality in Greek Epic, Tragedy, and Philosophy: The Self in Dialogue, Oxford: Oxford University Press 1996; The Structured Self in Hellenistic and Roman Thought, Oxford: Oxford University Press 2006. 2 Richard Sorabji, Self: Ancient and Modern Insights about Individuality, Life, and Death, Oxford: Oxford University Press 2006. 3 I am grateful to Richard Sorabji for showing me a copy of his chapter (which I refer to here), for his thoughtful responses to my paper, and for the (characteristic) generosity of spirit with which he has engaged in this debate despite our differences of view.
P. Remes, J. Sihvola (eds.) Ancient Philosophy of the Self, © Springer Science+Business Media B.V. 2008
35
36
C. Gill
Study of this topic raises at least three general questions on which we need to formulate a distinct view if we are to make progress with this inquiry. Questions of this type are raised by any attempt to study notions of self across cultures and periods. One is what concept of self we take as our starting-point and how we consider this to be related to other such ideas, including those of person or individual. A second is that of the relationship we see as existing between the modern concept of self and ancient ideas. For instance, we might suppose that this concept is available, in a broadly similar form, in all cultures and periods or that it is restricted to certain periods and cultures (for instance, modern Western thought). A further option is that ‘self’ is regarded as one of a number of concepts that we (modern thinkers) must inevitably use as “bridgeheads” to engage in dialogue with the thought-world of other cultures, even though their concepts are quite differently formulated.4 A related question is what kind of history of ideas of self we see as appropriate within the period or culture in question. For instance, we may see this as involving development or evolution of thinking, and, if so, we may think that this evolution leads towards a modern concept of self, however this is understood. A further prerequisite, I think, of effective study of ideas of self in ancient and other cultures is that we can, in some sense, take these ideas seriously. That is, we can imagine adopting these ideas as our own, either in their original form or after some process of interpretative translation. If the ideas remain, ineluctably, primitive or alien for us, it seems unlikely that our interpretative engagement with them will be either revealing or profound.5 In considering the implications of the examples of ancient thought selected for special attention here, I refer especially to the issues raised by these larger methodological questions.
Cicero’s Cato and Unique Individuality Indeed, such differences of natures have so great a force that sometimes one man ought to choose death for himself, while another ought not. For surely the case of Marcus Cato was no different from that of the others who gave themselves up to Caesar in Africa? And yet
4 A useful volume on cross-cultural study of ideas of person and self is Michael Carrithers, Steven Collins, and Steven Lukes (eds.), The Category of the Person: Anthropology, Philosophy, History, Cambridge: Cambridge University Press 1985; see Collins in that same volume; especially on “bridgeheads” see Lukes on the same volume, 297–298. 5 Issues about development are often intertwined with the question of taking ideas seriously. For instance, part of Williams’s critique of the developmentalist approaches to Greek ideas of mind adopted by Bruno Snell and Arthur Adkins was that they encouraged us (inappropriately) to treat certain ancient ideas as primitive or alien and so discouraged us from taking them seriously. See Bruno Snell, The Discovery of the Mind: The Greek Origins of European Thought, trans. T.G. Rosenmayer, New York: Harper & Row 1953; Arthur Adkins, Merit and Responsibility, Oxford: Clarendon 1960; Bernard Williams, Shame and Necessity, Berkeley, CA: California University Press 1993, chs. 2–3; see also Gill 1996, 29–41, 61–68, 89–92.
The Ancient Self: Issues and Approaches
37
it would perhaps have been counted as a fault if they had killed themselves, for the very reason that they had been more gentle in their lives, and more easy-going in their behaviour. But since nature had assigned to Cato an extraordinary seriousness, which he himself had consolidated by his unfailing consistency, abiding always by his adopted purpose and policy, it was necessary for him to die rather than look upon the face of a tyrant. (Cicero, Off. 1.112)6
This comment falls within Cicero’s presentation in Book 1 of On Duties (de Officiis) of the theory of the four roles (personae), based on the thinking of Panaetius, head of the Stoic school in the second century BC. What implications does it have for our understanding of ancient conceptions of selfhood? Richard Sorabji, in his contribution to this volume, finds in this passage a striking expression of the importance of unique individuality. He highlights the contrast with Kant’s conception of moral obligation. Whereas Kant sees moral rules as, by their very nature, applying universally, Cicero maintains that Cato’s suicide was morally right only for him. Sorabji allows that, in principle, it would also be right for anyone else exactly like Cato in this situation. But “[t]he interesting point is that there was no-one like Cato among those defeated at Thapsus. He had always stood for a kind of austerity that no-one else began to match […] And that is why it would be right for him, but for none of the others to commit suicide in the circumstances that prevailed”. Sorabji adds subsequently: “it was unique to Cato that suicide was the right course, because his character was unique among those defeated here. The interest here is not only in the individual, but in an individual whose character in the situation was unique”.7 Why does Sorabji regard this passage as specially significant? It is not because the idea represents a standard feature of ancient, or indeed Stoic ethics. In fact, he sees the passage as exceptional, though reflecting a more individualizing version of Stoic ethics that seems to go back to Panaetius.8 Rather, the passage is significant because, alongside other Greek and Roman examples he cites, it displays an idea that, for Sorabji, lies at the heart of the concept of self. Crucial for this concept is what is often characterized as a “first-personal” standpoint or perspective, an understanding of being “I” or “me”. At the start of his discussion, Sorabji presents this standpoint as being fundamental to the experience of humans and higher animals, along with the still more basic sense of being the owner of one’s body that is shared with even the simplest animals.9 Sorabji believes that this first-personal standpoint, while in one sense basic and intuitive, underlies more advanced ideas, including those of personal uniqueness and individuality.
6 Translated in M.T. Griffin and E.M. Atkins, Cicero: On Duties, Cambridge: Cambridge University Press 1991, modified. 7 See 25 and 31 above; his discussions here amplify his comments in Sorabji 2006, 158–159. 8 See 24, 32 above; also Sorabji 2006, 165. 9 “A self […] is an embodied individual owner who sees himself or herself as me and me again”, 13 above.
38
C. Gill
It is this whole set of ideas, centred on the first-personal viewpoint and unique identity, that Sorabji sees as making up a concept of self, by contrast with that of person or human being.10 The idea suggested in this passage of Cicero that certain moral claims apply not universally but by virtue of one’s unique individual character is, clearly, a complex and sophisticated one. But its significance for Sorabji’s argument inheres in the fact that it depends on this core notion of self – that is, a consciousness of being (uniquely) me, and to this extent a unique individual. Passages of this type are taken to show that the idea of unique individuality formed a key strand in the ancient, as well as modern, thought-world. What general picture does Sorabji offer of the relationship between modern versions of the concept of self and ancient thought? Although he does not argue for a strong form of universalism (that all cultures hold the same conception of self), he does suggest that the core idea of self is widely available in ancient, and other, cultures and periods. He dissents from the view that the idea of unique individual self emerges only as a product of specific cultural or intellectual conditions, for instance, those obtaining in later medieval or modern Europe.11 On the other hand, Sorabji thinks that certain strands in thinking about the self (in the relevant sense) emerge more strongly in specific phases of ancient thought. For instance, he thinks that there was an intensified interest in unique individuality and self-awareness in the late Hellenistic and early Roman Imperial periods, from the late second century BC to the second century AD. Cicero’s stress on the idea that Cato’s unique individuality constitutes a source of moral authority thus expresses a distinctive feature of the period while also indicating the presence of this idea in ancient thinking more generally. I now juxtapose Sorabji’s interpretation of this passage, and the intellectual framework that he presupposes, to my own. Whereas his interest is focused specifically on the self, as defined earlier, I have taken as my subject ancient thinking on personality, personhood, and selfhood, more broadly understood; I treat these terms as evoking an overlapping nexus of ideas, rather than as having sharply determinate meanings. The core focus, in my case, is on normative ideas about human nature and psychology. Put differently, my interest centres on the thought that certain natural or metaphysical features of our nature or psychology carry ethical implications, “ethical” being here understood in a broad sense. The question whether these features should be understood as consisting in unique individuality and a first-personal viewpoint or psychological (or psychophysical) cohesion or the capacity for social engagement is an important but secondary one as regards the specification of my topic. Within this broad subject area, I have distinguished two patterns of thinking about the person, both bearing on human psychology and ethics. One pattern is focused on (unique) individuality, “I”-centred self-consciousness and subjectivity. The other is focused on the cohesion of psychological functions, the capacity for interpersonal and social engagement, and the development of objective understanding,
10 11
Sorabji 2006, 20–22, 47–48; see also 13–14 above. Sorabji 2006, 32–33, 48–50.
The Ancient Self: Issues and Approaches
39
as this bears on the question of what counts as a good human life. I have characterized one pattern as a “subjective-individualist” and the other as an “objective-participant” conception of personality.12 In drawing this contrast, I have not claimed that one pattern, the subjectiveindividualist, corresponds uniformly and exclusively to modern Western thinking or that the other, the objective-participant one, does so to ancient Greek and Roman thinking about the person. Rather, my thesis has been that, if we aim to engage with ancient thought in this area, we should seek to counteract the more individualist and subjective dimensions of our own thinking and accentuate the more participant and objective dimensions. This reflects my assumption that there are specific features of modern thinking, derived from our religious, cultural, and conceptual frameworks, which have promoted the adoption of a subjective-individualist conception of person and which do not have exact equivalents in Greek and Roman culture. I have also argued, building on certain strands in contemporary philosophy, that we have, in purely modern terms, good conceptual reasons to prefer a more objective approach to human psychology and a more participant (and more objectivist) approach to ethics. Hence, engagement with a more objective-participant conception of personality in ancient thought has conceptual advantages for us, apart from enabling us to understand Greek and Roman thinking better.13 On the question whether we should see development within ancient thinking on personality, I have followed Bernard Williams in his scepticism about the claims of Bruno Snell and A.W.H. Adkins that Greek (or ancient) thought embodies a stepby-step evolution towards the modern concept of self, the latter being understood in Cartesian or Kantian terms.14 In a more recent book (2006), I have also expressed doubt about whether there is a significant shift within Hellenistic and Roman thought towards a more subjectivist or individualist conception of self, as has sometimes been maintained. This is not because I question the existence of any shifts in ancient thought in this area. On the contrary, I have argued for the emergence in Stoic and Epicurean thought of a (partly) new conception that I have characterized as that of the “structured self”. The core feature here is the combination of psychological (and psychophysical) holism and naturalism with certain radical ethical claims that have their roots in Socratic thinking. There are significant differences between this conception and much Platonic and Aristotelian thought, as well as
12 See Gill 1996, 1–18, and ch. 6, especially 455–469. On modern thinking on ‘person’ as a normative category and the relationship of this category to ancient thought, see also Gill, The Person and the Human Mind: Issues in Ancient and Modern Philosophy, Oxford: Oxford University Press 1990, especially Introduction; Gill, “Is There a Concept of Person in Greek Philosophy?”, in Stephen Everson (ed.), Psychology: Companions to Ancient Thought 2, Cambridge: Cambridge University Press 1991, 166–193. 13 See Gill 1996, 6–10; also 41–46, 60–69 and, more broadly, ch. 6; also Gill 2006, ch. 6. On the relationship between ancient and modern thinking about ethical objectivity, see Christopher Gill (ed.), Virtue, Norms, and Objectivity: Issues in Ancient and Modern Ethics, Oxford: Oxford University Press 2005. 14 Cf. n. 5. See also Williams 1993, chs. 2–3; Gill 1996, 29–78.
40
C. Gill
later Hellenistic and Roman theories such as those found in Middle Platonism and Neoplatonism. But I see these points of difference as reflecting variation within the (broadly) objective-participant conception of personality that is characteristic of ancient thought, rather than as marking a move towards the more subjectiveindividualist view that has been more typical of modern Western thought.15 How does my approach compare with Sorabji’s? As indicated, there are substantial differences in the nature of our conceptual starting-points and assumptions, which do not make it altogether easy to say whether we agree or disagree on specific points. Sorabji’s thesis that an intuitive sense of selfhood (centring on a firstpersonal view and ownership of one’s body) is fundamental to human and higher animal life, seems plausible, taken on its own.16 At least, it seems plausible to us, modern Western thinkers,17 though it is much more open to question whether it has seemed fundamental in all human cultures, including those of Greek and Roman antiquity. It is also questionable, I think, whether in fact ancient philosophy ever evolved the idea that selfhood is uniquely individual and “I”-centred, as Sorabji claims.18 In Hellenistic and Roman philosophy, for instance, we find examples of (what is often called) the cradle argument. This argument depends on identifying certain key features as fundamental to the motivation of all animals at birth or in infancy.19 Although it has been maintained, by Troels Engberg-Pedersen, that one version of the Stoic form of this argument, in Cicero On Ends (Fin.) 3.16, does express an “I”-centred sense of self, this claim remains controversial. On the face of it, the Stoic examples of this argument, including the most elaborate account, by Hierocles (second century AD), focus on the idea that animals function, from birth, as unified psychophysical organisms, with an in-built motivation to seek their own benefit and to maintain their constitution. Although Stoic sources sometimes present this process as involving self-consciousness or self-perception, this term seems to be used to convey the idea of organic or animal cohesion, without invoking the idea of an “I” or the first-personal viewpoint.20 In making this point, I am not
15
On “the structured self” see Gill 2006, chs. 1–3. On the question whether there is a development in the Hellenistic-Roman period towards a subjective-individualistic conception, see ch. 6. 16 Sorabji 2006, 20–23; also 13 above. 17 However, as Sorabji notes, the validity and coherence of the notion of self has been strongly challenged by a number of modern Western philosophers (2006, 17–20). 18 See Sorabji 2006, 32–33, 48–50; also 14, 27–28, 30–31 above. 19 See e.g. Cicero, Fin. 1.30, 3.16; the classic treatment of this argument is Jacques Brunschwig, “The Cradle Argument in Epicureanism and Stoicism”, in Malcom Schofield and Gisela Striker (eds.), Norms of Nature: Studies in Hellenistic Ethics, Cambridge: Cambridge University Press 1986, 113–144. The main explicit aim of this argument, by Epicureans and Stoics, is to support a certain account of the overall human good or goal of life. 20 For Hierocles’ account, see A.A. Long and D.N. Sedley, The Hellenistic Philosophers, 2 vols., Cambridge: Cambridge University Press 1987 (=LS), 53 B. See Troels Engberg-Pedersen, The Stoic Theory of Oikeiosis: Moral Development and Social Interaction in Early Stoic Philosophy, Aarhus: Aarhus University Press 1990, 64–80, especially 66–71; on Hierocles, see A.A. Long, Stoic Studies, Cambridge: Cambridge University Press 1996, 350–363. For responses to EngbergPedersen’s view, see Gill 1996, 332–333, 365–370; 2006, 359–370.
The Ancient Self: Issues and Approaches
41
claiming that ancient culture wholly lacks a sense of the kind of selfhood that Sorabji regards as basic. But I am suggesting that this sense, if present, feeds into ancient theory in a less obvious or explicit way than we might expect. What about Cicero’s presentation of Cato (in Off. 1.112), which Sorabji sees as expressing the – obviously, more and more complex and theorized – idea that our uniquely individual qualities can be seen as constituting a moral claim? In reflecting on this striking passage, I consider first some relatively localized points and then seek to relate the topic to my larger framework of thinking about HellenisticRoman (and more generally ancient) conceptions of self. First, I want to qualify some of Sorabji’s emphases. Cicero never explicitly says that the moral obligation of suicide applies uniquely to Cato, only that one (type of) person (alius) should, while another (alius) should not, commit suicide in this situation. Although, as Sorabji underlines,21 it is stressed that Cato was exceptional in the seriousness (gravitas) of his nature and his life-long consistency of purpose, the inference that this makes his decision to die uniquely appropriate is not drawn by Cicero himself.22 It is worth noting here the (much more extensive) treatment in Book 3 (99–115) of On Duties of the third-century BC Roman general Regulus and of his decision to follow a principled course of action that, as he well knew, entailed his death at the hands of his Carthaginian captors. Two relevant points arise from this discussion of Regulus. One is that the character-type ascribed to Cato is not actually unique but has parallels in other Roman exemplary figures. (A similar inference can be drawn from the Greek parallel to Cato cited in 1.113, Ajax, who is presented as someone who would prefer death to enduring the indignities endured by Ulysses.)23 The other implication is that Cicero has the resources to construct an argument for a principled choice of death in On Duties without referring to the idea of unique individuality (or the second persona) at all. On the contrary, the implications of Cicero’s arguments – against imaginary counterclaims – for the choice taken by Regulus are that this is a course of action that anyone should make in this circumstance.24 So we should
21
Sorabji 2006, 159. For Sorabji’s treatment, see his 2006, 158–159, and 24, 31 above. On Cicero’s thinking about the grounds for legitimate self-killing and his idealization of Cato, see also Timothy Hill, Ambitiosa Mors: Suicide and the Self in Roman Thought and Literature, London: Routledge 2004, 31–71, esp. 64–67. 23 Interestingly, the Ajax-Ulysses contrast (which reinforces the contrast between Cato and the other Roman leaders defeated by Julius Caesar) recurs just before the discussion of Regulus (Off. 3.97–8), which might imply a linkage in Cicero’s mind between Ajax, Cato, and Regulus. 24 See Cicero, Off. 3.100–110, a series of arguments for and against Regulus’ decision, in which Cicero explicitly favours Regulus’ stance. Regulus is offered as an illustration of Cicero’s overall project in Off. 3, that of helping ordinary well-motivated people (i.e., in principle, anyone) to discriminate between what is just and what is expedient (3.7–16, 99, 110, 115). For a response to the connection drawn here between Cicero’s use of the examples of Cato and Regulus, see Sorabji, 31–32 above. 22
42
C. Gill
not be too ready to conclude that Cicero’s approach in On Duties necessarily focuses on unique individuality (or what is right for me).25 However, this response might be seen as too narrowly focused on Cicero’s mode of presentation and as failing to address the implications of the four-personae theory, especially the second (individual) persona. It could be argued that Panaetius, in formulating this theory, which is generally taken as his innovation, was deliberately giving Stoic ethics a more individual focus and that Cicero’s depiction of Cato at least (though perhaps not that of Regulus) reflects this feature. To some degree, this suggestion must be right. Although it is difficult to gauge just how far Panaetius goes beyond Chrysippus in his treatment of the theme of appropriate actions and in his development of (what we call) practical ethics, the four-personae theory surely marks some increase in differentiating and particularizing what is appropriate (kathˆekon or officium).26 However, it is not clear that the overall aim or effect of this innovation is to signal an interest or valuation of unique individuality. The main explicit emphasis in Cicero’s account of the theory (which in this respect seems faithful to Panaetius) is on the idea of consistency (constantia or aequabilitas).27 This is presented as important, in turn, because it enables us to achieve what is fitting (decorum or prepon), that is, to express the outward face of virtue.28 Consistency is achieved by co-ordinating the first two personae, our universal role as rational agents capable of virtue and our distinctive qualities and inclinations (107–114). It also depends on taking into account the third and fourth personae, that is, the status given us by birth and social situation and the role we adopt by choice (if we have the opportunity to make such choice).29 The predominant thought is not that any one persona constitutes an overriding moral claim or constraint – although, if any one persona is overriding, it is the universal one.30 25
In Off. 3.51–57, for instance, Cicero favours the more “universalist” ethical approach attributed to Antipater (especially 52–53), rather than the “conventionalist” ethic attributed to Diogenes. 26 On the four-personae theory, see Cicero, Off. 1.107–21; see also Gill, “Personhood and Personality: The Four-Personae Theory in Cicero, De Officiis 1”, Oxford Studies in Ancient Philosophy 6(1988), 169–199; and, on the relevant background in Stoic theory, see Brad Inwood, “Rules and Reasoning in Stoic Ethics”, in Katerina Ierodiakonou (ed.), Topics in Stoic Philosophy, Oxford: Oxford University Press 1999, 112–127; see also Brad Inwood and Pierluigi Donini, “Stoic Ethics”, in Keimpe Algra, Jonathan Barnes, Jaap Mansfeld and Malcom Schofield (eds.), The Cambridge History of Hellenistic Philosophy, Cambridge: Cambridge University Press 1999, 675–738, see especially 690–699. However, Tieleman argues that Panaetius’ theory of personae does not contain any significant innovations or express any greater interest in individuality, as compared with earlier ideas, Teun Tieleman, “Panaetius’ Place in the History of Stoicism with Special Reference to his Moral Psychology”, in A.M. Ioppolo and D.N. Sedley (eds.) Pyrrhonists, Platonizers: Hellenistic Philosophy in the Period 155-86 BC, Naples: Bibliopolis 2007, 104-142, esp. 130-140. 27 See e.g. Cicero, Off. 1.98, 110–111, 114, 119–120, 125. 28 Cicero, Off. 1.93–95, 98, 107, 110–111. 29 See Cicero, Off. 1.115–121. See further Gill 1988, 173–176. 30 The presentation of this persona as ‘first’ implies this priority: see also Off. 1.98, 100–102, 107. Cicero explicitly stipulates that the retention of individual features must not be “contrary to universal nature”; hence the individual features must be “not defective” (non vitiosa), and “least of all to be criticized” (minime vituperandorum), 1.109–110.
The Ancient Self: Issues and Approaches
43
Rather, the point is that these personae must be correlated with each other, so that our life acquires the overall consistency that is typical of the virtuous or fitting life (107, 111, 115, 125). In other words, the point is not so much that unique individuality constitutes a moral claim. Rather, the way in which we seek to realize our shared rationality and ethical humanity (the first personae) should take account of our distinctive qualities and inclinations, as well as the normative implications of our social situation and role. This point should be borne in mind in interpreting Cicero’s presentation of Cato in Off. 1.112. The stress falls on the importance of consistency with one’s particular qualities, and on this factor as helping to determine what is appropriate in specific situations. Cato, in fact, is depicted as exceptional in his consistency (constantia) as well as in the seriousness (grauitas) of his nature; and this made suicide right for him when it would not have been right for those with a more flexible character. But we should not overlook the significance of the point already made by Cicero that we also need to aim at consistency with the first (universal) human persona. What is being stressed here (112–113), and subsequently in the case of the third and fourth personae, is that the way in which we seek to realize this universal persona should take the other personae into account, with a view to achieving the kind of consistency that is presented as our overall goal. If our actions are in conflict with the first persona, they count as faults, regardless of whether they are consistent with our individual nature.31 So far, I have simply qualified Sorabji’s emphasis on the importance of individuality in this passage, or at least tried to place this in a larger context. I now suggest ways in which our understanding of this passage and theory can be increased by reference to my alternative approach to the study of ancient personality and selfhood, outlined earlier. In Gill (2006), I have stressed the importance in Stoicism (and Epicureanism) of certain “Socratic” ethical claims. One of these is that all human beings are constitutively capable of developing towards complete virtue and happiness, regardless of their inborn nature or social context.32 In partial contrast, in certain Platonic and Aristotelian passages (and sometimes in Middle Platonism), virtue is presented as depending on the combination of the right kind of inborn nature, social habituation, and rational education.33 A further Stoic theme with Socratic roots is that only the wise person embodies complete consistency and coherence of character, whereas non-wise people display incoherence and inconsistency.34 In Aristotle, by contrast, we sometimes find the idea that defective as well as virtuous people have stable dispositions, a theme again sometimes adopted by
31
See Cicero, Off. 1.109–110, taken with the preceding note. See e.g. LS 61 K–L; relevant here is that the process of ethical development described in Cicero, Off. 3.20–22 is presented as, fundamentally, natural; see also LS 60 B–F. See further text to nn. 82–84 below. 33 See e.g. Plato, Republic 490d–498b; Arist. EN I 3–4; II 1; X 9; Plutarch, Mor. De virt. mor. 443c–d. 34 See e.g. the contrast between (unified) virtue and (conflicted) vice in LS 61 A, O, also Seneca, Ep. 120.22. 32
44
C. Gill
Middle Platonists such as Plutarch.35 These are two of the Socratic ideas which, taken together with a naturalistic and holistic view of personality, make up the Stoic-Epicurean conception that I have termed “the structured self”.36 This background can help us to define more precisely the distinctive character of Panaetius’ four-personae theory. While retaining the core features of Stoic thinking, Panaetius seems to take account of the (partly contrasting) Platonic-Aristotelian approach in a way that is consistent with his ancient reputation as a “lover of Plato and Aristotle”.37 Without repudiating the Stoic thesis that “all human beings have the starting-points of virtue”,38 he also stresses that, in their pursuit of virtue, people should take account of their distinctive natural qualities (the second persona) as well as of their inherited social situation and chosen role (the third and fourth personae). He retains the standard Stoic view that the ideal character-state is that of cohesive structure and consistency, centred on the possession of the virtues as a unified set but including other qualities based on the possession of the virtues, such as psychic health and harmony.39 But he adds the idea that, in working towards this ideal condition, we need to make reference to our distinctive inclinations and talents. In other words, to achieve the kind of consistency involved in the realization of the first persona, we need also to maintain consistency with the second persona, as exemplified in the case of Cato (and Ajax) in Off. 1.112–113.40 The stress on the significance of our individual nature, to which Sorabji draws attention, can thus be explained in part by reference to Panaetius’ adoption of a version of Stoic thinking on ethical development which is informed by Platonic-Aristotelian thinking, without losing its substantively Stoic character and the combination of features that I associate with the idea of the “structured self”. I think that Panaetius’ four-personae theory can also be helpfully analysed as expressing what I have called an “objective-participant” conception of personality. In my earlier book (1996), I took as ancient philosophical examples of this conception the kind of framework found in Plato’s Republic and Aristotle’s ethical treatises. Key relevant features include the idea that the search for (what we should call) objective ethical norms depends on participation both in social and communal life and in the shared life of an intellectual community. Understanding the good, in Plato, or what
35
See e.g. Arist. EN VII 8 (the contrast between stable self-indulgence and unstable akrasia), and I 13, 1102b13–1103a10; also Plutarch, Mor. De virt. mor. 445e, 446c–d. 36 See Gill 2006, 73–96, 129–138, 231–233. 37 Panaetius, fragments 56–57 in M. van Straaten, Panaetii Rhodii Fragmenta, Leiden: E. J. Brill 1962. On the intellectual context of Panaetius’ interest in Plato, see Michael Frede, “Epilogue”, in Keimpe A. Algra, Jonathan Barnes, Jaap Mansfeld and Malcom Schofield (eds.), The Cambridge History of Hellenistic Philosophy, Cambridge: Cambridge University Press 1999, 771–797, especially 783–784. 38 LS 61 L (trans. modified); also n. 32 above. 39 See Cicero, Off. 1.95–6, 98, 100, 102. On the cohesive character of virtue, see Cicero, Fin. 3.20–1, LS 61 A–H, and, on related, virtue-based qualities such as psychic health and harmony, Stob. 2.62.15–63.5, section 5b4. See also Gill 2006, 150–157. 40 See references in n. 27 above.
The Ancient Self: Issues and Approaches
45
it means to be, normatively, human or divine, in Aristotle, depends on this combination of social and philosophical participation.41 Panaetius’ four-personae theory can be seen as “objective-participant” in the same sense. The normative status given to the third and fourth personae is just one expression of the fundamental role given to social participation in the development of virtue (115–121). More broadly, the kind of exercise in practical ethics presented in On Duties presupposes that the reader, like Cicero himself, will make a serious attempt to relate his social life to the kind of theoretical ideas presented, if the work is to have a useful outcome. In other words, the goal of establishing and living by (what we should call) objective ethical norms depends on co-ordinating all four normative dimensions or ‘roles’ (personae), including our distinctive individual nature, both at the theoretical level and in our social and communal practice (107–125).42 In commenting on my approach (in 1996 and 2006), Sorabji suggests that the contrast between objective-participant and subjective-individualist conceptions of personality is drawn too strongly and that individuality in certain senses can figure in both types of conception.43 The present discussion, to some extent, supports Sorabji’s suggestion. The four-personae theory shows clearly how a significant role can be given to individuality (in the form of the second persona), without undermining the objective-participant framework of thinking involved. However, it is important to recognize that the ideal of realizing one’s individual nature or selfhood can also be conceived in (radically) subjective-individualist terms, as has happened in certain strands of modern thought – but not in this ancient example.44 So, if we want to avoid giving the impression that realizing individuality necessarily carries this kind of connotation, we need to be quite explicit on this point. If individuality can indeed be seen as compatible with an objective-participant framework of thought, I see no objection to finding it in works such as Cicero’s On Duties.
41
See Gill 1996, 266–287, 346–383, 430–443. See further, on the correlation of universal and particular motifs in ancient theory with a view to establishing what we call objective ethical norms, Gill, “In What Sense are Ancient Ethical Norms Universal?”, in Christopher Gill (ed.), Virtue, Norms, and Objectivity: Issues in Ancient and Modern Ethics, Oxford: Oxford University Press 2005, 15–40, especially 35–40 on Cicero, Off. On the idea that coming to know the good is a matter of combining theory and communal practice in Stoicism, see Gill, “The Stoic Theory of Ethical Development: In What Sense is Nature a Norm?”, in Jan Szaif and Mathias Lutz-Bachmann (eds.), Was ist das für den Menschen Gute? Menschliche Natur und Güterlehre/What is Good for a Human Being? Human Nature and Values. Berlin: De Gruyter, 2004, 101–125, see 113–114. 43 See his 2006, 48–49, and 14–16 above. Sorabji’s view is that all four ideas can combine with each in shaping conceptions of selfhood. 44 “Authenticity” and the search to “be yourself” or “be true to yourself” have sometimes been conceived in radically subjective-individualist terms, e.g. by Sartre or Nietzsche. See Gill 1996, 109–113, 125–129, 446. 42
46
C. Gill
Epictetus, Prohairesis, and the Self The second ancient example considered here is Epictetus’ presentation of prohairesis, including his claim, sometimes, that we are our prohairesis. Although I start again from some comments by Sorabji, I discuss several interpretations of Epictetus’ conception of self, especially those that imply that Epictetus holds (some version of) a subjective-individualist conception of personality. In responding to these scholarly views, I locate Epictetus’ comments on prohairesis in the context of his three-topic programme of practical ethics. I take Epictetus’ thinking, like Cicero’s version of the four-personae theory, as exemplifying the features I associate with the idea of the “structured self” in Hellenistic and Roman thought, and also as expressing an “objective-participant” conception of personality.45 The scholarly views outlined shortly highlight certain distinctive features of Epictetus’ terminology, when considered against the background of Stoic conceptual language more generally. One is his use of prohairesis, the standard Aristotelian word for “choice” or “decision”, where we might have expected to find “reason” (logos) or “control-centre” (h¯egemonikon). A related theme is the emphatic contrast between what is or is not “up to us” (eph’ h¯emin), sometimes expressed as what is or is not “within the scope of prohairesis” (prohairetika or aprohaireta). Another theme is that we should exercise our prohairesis in “using” or “examining” impressions (phantasiai). This extract from Handbook 1 illustrates some of these features: Practise, then, from the start to say to every harsh impression, “You are an impression, and not at all the thing you appear to be.” Then examine it and test it by the rules which you have, and firstly and chiefly, by this: whether the impression has to do with the things which are up to us, or those which are not; and, if it has to do with the things that are not up to us, be ready to reply, “It is nothing to me.”46
Sorabji draws attention to the claim that we should not only use but in some sense are our prohairesis. This comes out in passages such as this imagined exchange: “I will put you in chains”. “What did you say, man? Put me in chains? My leg you will put in chains, but my will (prohairesis) not even God can conquer.” (Epictetus, Diss. 1.1.23)47
Sorabji sees such passages as expressing the idea that Epictetus can identify himself with an “inviolable” aspect of himself, roughly, his “will.”48 Another feature of
45 This part of my discussion draws on the more detailed treatment in Gill 2006, especially in 328–344, 371–391. 46 Translated by R. Hard with introduction and notes in Christopher Gill (ed.), Epictetus: The Discourses, Handbook, Fragments, London: Dent 1995. On innovations of terminology in Epictetus, see Brad Inwood, Ethics and Human Action in Early Stoicism, Oxford: Oxford University Press 1985, 115–126, also Appendix 2 (224–242); for comments on Inwood’s treatment, see A.A. Long, Epictetus: A Stoic and Socratic Guide to Life, Oxford: Oxford University Press Long 2002, 126–127. 47 Translated by Sorabji, 24 above, also his 2006, 44, 181. 48 See 24 above; also his 2006, 181–185.
The Ancient Self: Issues and Approaches
47
Epictetus’ teachings highlighted by Sorabji is the emphasis on the importance of acting on the basis of commitment to our role (pros¯opon). For instance, in Discourse 1.2, as in Cicero’s presentation of Cato (Off. 1.112), Sorabji finds the idea that “different characters may call for different actions in the same circumstances without one way of action being treated as right or wrong universally, independently of character.” Sorabji sees both these features as reflecting the more individualizing approach to the self that seems to derive from Panaetius, at least in Stoic thought.49 The idea that Epictetus’ thinking about prohairesis embodies a new, and in some sense, more individualist, conception of the self, has also been advanced by some other scholars. Charles Kahn, for instance, offers this formulation of what is innovative in Epictetus’ usage: [Prohairesis] is presented not only as the decisive factor in practical existence but as the true self, the inner man, the ‘I’ of personal identity. By contrast, for Plato and Aristotle, the ‘I’ or true self was nous, the principle of reason most fully expressed in theoretical knowledge. This shift is a momentous one for the evolution of the idea of person and selfhood. For theoretical reason is essentially impersonal, and the Platonic-Aristotelian identification of the person with his intellect offers no basis for a metaphysics of the self in any individual sense. Epictetus, on the other hand, identifies himself with something essentially personal and individualized: not with reason as such but with the practical application of reason in selecting his commitments, in keeping his emotional balance, his serenity, by not extending himself to goals and values that lie beyond his control.50
Another formulation of this idea is offered by A.A. Long in a discussion of “representation and the self” in Stoicism.51 Long begins by locating certain key features of Stoic psychology, notably the role given to “impressions” and “assent”, within the larger perspective of the history of ideas about the self. This stress on phantasia (“representation” or “impression”) “is best interpreted […] as a new focus on consciousness, on the individuality of the perceiving subject, as the fundamental feature of the mental”.52 Long elaborates this view of the significance of the Stoic concept of phantasia in these terms: “[in so far as phantasiai are] appearances to this individual, they have an irreducible particularity – they are mental affections of this and only this person […] what it is for [a given] person to assent to [a given phantasia] will remain something unique […] how to deal with the representations one has is a matter for each person’s individual decision”.53 Long sees Epictetus’ emphasis on the idea of “examining impressions” as prefiguring a post-Cartesian emphasis on self-consciousness or reflexivity and as forming part of the pre-history
49
See 25 above; also his 2006, 161–163. Charles H. Kahn, “Discovering the Will: From Aristotle to Augustine”, in John Dillon and A.A. Long (eds.), The Question of “Eclecticism”: Studies in Later Greek Philosophy, Berkeley, CA: California University Press 1988, 234–259. 51 Long 1996, ch. 12. 52 Long 1996, 266. 53 Long 1996, 275, referring also to Kahn 1988, 253. 50
48
C. Gill
of the idea of the person or self as one who possesses a uniquely “first-personal” viewpoint.54 Another scholar who has underlined the significance of Epictetus’ stress on the idea of “examining impressions” is Michel Foucault.55 Foucault’s work, especially his book on “the care of the self”, has been influential in promoting the idea that a new type of (in my terms) “subjective-individualist” conception of self emerges in Hellenistic and Roman thought. But his views on this subject differ from what has often been maintained by scholars and reflect a very particular intellectual standpoint. Foucault expresses scepticism about the idea that the Hellenistic-Roman period sees the emergence of “individualism” – at least in two senses of this notion. These are “the individualistic attitude, characterized by the absolute value attributed to the individual in his singularity” and “the positive valuation of private life, as distinct from public life”.56 However, there is a third sense of “individualism” which Foucault does see as characteristic of the early Roman Empire, namely “the insistence on the attention that should be brought to bear on oneself”, and, more precisely, “an intensification of the relationship to oneself by which one constituted oneself as a subject of one’s acts”.57 Foucault also refers to “the intensity of the relations to self” as expressed in “taking oneself as an object of knowledge and field of action, so as to transform, correct, and purify oneself”.58 Related features of the period are the emergence of more elaborate techniques of self-scrutiny and selfexamination to ensure that desire matches “the aesthetic and ethical criteria of existence”.59 To this extent, Foucault lends support to the view of Kahn and Long that Epictetus and other writers of this period introduce a greater focus on individual reflexivity and self-consciousness.60 What conception of self underlies Foucault’s conception of the care of the self in Hellenistic-Roman thought, and should we see it as, in my terms, “subjectiveindividualist” or “objective-participant” in approach? The answer to this question is not wholly straightforward, in part because Foucault himself takes a strongly critical view about the notions of subjectivity and objectivity (as he also does about individualism). Much of his work (1973, 1977) has focused on the way in which political, legal, and religious constraints have shaped how people think about their
54 Long 1996, 282, referring to Rorty’s vivid outline of the post-Cartesian conception in A.O. Rorty, The Identities of Persons, Berkeley, CA: California University Press 1976, 11. On the postCartesian self and its relationship to ancient thought, see Gill 1996, 34–37, 405–407; 2006, 330–344, 391–407. 55 Michel Foucault, The Care of the Self: The History of Sexuality vol. 3, trans. R. Hurley. London: Penguin 1988a, 63–64, see also 57–58. 56 Foucault 1988a, 42. 57 Foucault 1988a, 41. 58 Foucault 1988a, 42. 59 Foucault 1988a, 67. 60 See also Charles Edwards, “Self-Scrutiny and Self-Transformation in Seneca’s Letters”, Greece and Rome 44(1997), 23–38, referring to a range of view of this type, in connection with Seneca.
The Ancient Self: Issues and Approaches
49
mental states and selfhood. He has used the term ‘assujettissement’ to suggest a linkage between the ideas of subjection and subjectivity, implying that people are subjected to certain norms of what it means to be a (post-Cartesian) subject. The idea of objectivity is interpreted by him as implying the (false) objectification of the self which is the result of the imposition of external norms. These views reflect scepticism both about the validity of the theories of mind and subjectivity developed since Descartes and about the idea of objective moral norms. Hence, in analysing the kind of care of the self found in Hellenistic-Roman thought, Foucault makes virtually no reference to the larger political or religious context of social and family life. He also shows scant interest in the kind of theoretical frameworks (for instance, those of Stoic and Epicurean ethics and epistemology) that might be interpreted as analysing the basis of objective norms or knowledge.61 As a result, as Pierre Hadot has pointed out, Foucault’s account of the care of the self offers a rather “aesthetic” picture of what self-fashioning involves in Hellenistic-Roman thought.62 The focus is on the individual subject (despite Foucault’s scepticism about this notion), as one who defines his own identity, especially his sexual identity, with very few normative reference-points other than personal preference and experience. As a result, Foucault’s conception of self seems to be appropriately characterized as “subjective-individualist” (rather than “objective-participant”), even though he himself might not have embraced this description.63 So far, I have outlined a number of ways of interpreting the significance of Epictetus’ thinking about prohairesis and about the examination of impressions for our understanding of selfhood in this period, which stress in different ways individualistic or subjective senses of self. I now offer an alternative line of interpretation, which links Epictetus’ teachings with Stoic thinking about (what I am calling) the idea of the structured self and which also analyses this idea in objectiveparticipant terms. A preliminary question is how far Epictetus should be seen as expressing standard Stoic theory (as was maintained strongly by Bonhöffer, the author of two important early studies of Epictetus),64 or as reflecting a modified
61
On ancient thought about what we call ethical “objectivity”, see Gill 2005, especially part 2. Pierre Hadot, Philosophy as a Way of Life: Spiritual Exercises from Socrates to Foucault. Trans. M. Chase. Oxford: Blackwell 1995, 206–213, especially 211–213. 63 This discussion of Foucault is indebted to Jean-François Pradeau, “Le sujet ancien d’une politique moderne: à propos des exercises spirituels anciens dans L’Histoire de la sexualité de Michel Foucault”, in Frédéric Gros et al. (eds.), Foucault: le courage et la vérité, Paris: Presses Universitaires de France 2002, 131–154; “Le sujet ancien d’une politique moderne: sur la subjectivation et l’éthiques anciennes dans les Dits et écrits de Michel Foucault”, in Pierre-Francois Moreau (ed.), Lectures de Michel Foucault, vol. 3. Sur les Dits et Écrits, Lyons: École normale supérieure Lettres et sciences humaines 2003, 35–51. See also Wolfgang Detel’s critique of Foucault’s approach in Foucault and Classical Antiquity: Power, Ethics and Knowledge, trans. D. Wigg-Wolf, Cambridge: Cambridge University Press 2005, focused on the second volume of Foucault’s history of sexuality (1987). 64 Adolf Bonhöffer, Epictet und die Stoa: Untersuchungen zur Stoischen Philosophie, Stuttgart: Frommann 1890, reprinted 1968; Die Ethik der Stoikers Epictet, Stuttgart: Frommann 1894, repr. 1968, trans. W. O. Stephens, The Ethics of the Stoic Epictetus, New York: Peter Lang 1996. 62
50
C. Gill
version of Stoicism, as is assumed in most of the recent scholarship just outlined. Long (2002), in the first sustained study of Epictetus since Bonhöffer’s, offers a reasonable view on this question. It is important to bear in mind that Epictetus’ discourses are reports of largely non-technical discussions on practical ethics and do not represent formal instruction in Stoic doctrine (which Epictetus seems also to have provided). Epictetus presupposes standard Stoic theory even if his mode of expression, in these practical discussions, is distinctive or singular in certain respects, as, for instance, in his usage of the term prohairesis and related ideas.65 In exploring the conception of selfhood expressed in Epictetus’ discourses, I focus on the implications of the three-topic programme of practical ethics that is widely recognized as a distinctive feature of his thought. Here is Epictetus’ fullest statement of this programme, which underlies his ethical teachings in the Discourses and Handbook: (1) There are three topics, in which the would-be honourable and good person needs to have been trained. (2) That of desires and aversions, to ensure that he succeeds in getting what he desires and does not encounter what he seeks to avoid. (3) That of impulses and repulsions, or the appropriate (kath¯ekon) as a whole, to ensure that he acts in ways that are orderly, well-reasoned and not thoughtless. (4) The third has to do with infallibility and uncarelessness, or assents in general. (5) Of these [three] the most important and urgent is the one concerned with the passions [i.e. the first]. A passion only occurs if a desire is unsuccessful or an aversion encounters [what it seeks to avoid]. This is the topic which brings up disturbances, confusions, misfortunes, disasters, sorrows, lamentations, envies […] through which we are unable even to listen to reason. (6) The second has to do with the appropriate; I should not be impassive like a statue, but maintain my natural and acquired relationships, as a religious person, as a son, a brother, a father, and a citizen. (7) The third topic applies to those who are already making progress, and concerns security in just those matters mentioned, to ensure that even in dreams or intoxication or depression an impression (phantasia) should not slip by which has not been tested. (Epictetus, Diss. 3.2.1–5)66
I take it that this programme, like a similar one outlined by Seneca (Ep. 89.14, LS 56 B), represents practical advice for promoting ethical development, conceived in Stoic theory as “appropriation” (oikei¯osis). More precisely, it offers advice for advancing and correlating the two aspects of development, namely personal and social, presented in famous passages of Cicero’s On Ends (Fin.).67 The key feature of personal appropriation is the motivational shift from pursuing things normally
65 Long 2002, 31–4, 43–45; on practical ethics in Stoic philosophy in this period, see also Gill, “The School in the Roman Imperial Period”, in Brad Inwood (ed.), The Cambridge Companion to the Stoics, Cambridge: Cambridge University Press 2003, 33–58, especially 40–44. On distinctive features of Epictetus’ psychological vocabulary, see text to nn. 45–46 above. 66 See LS 56 C, trans. modified, numbers as in LS. On one or more of these three topics, see Diss. 1.4.11, 2.17.14–18, 3.12.13–15, 4.10.13. For discussion, see LS, vol. 2, p. 342; Bonhöffer 1890, 19–28; Inwood 1985, 116–119; Hadot 1995, 12–13, 193–195; Robert F. Dobbin, Epictetus: Discourses Book I. Translation with Introduction and Commentary, Oxford: Oxford University Press 1998, 91–94; Long 2002, 112–118; Sellars 2003, 134–142. 67 Cicero, Fin. 3.17, 20–22 (LS 59 D), 3.62–68 (LS 57 F).
The Ancient Self: Issues and Approaches
51
thought of as goods, such as health and wealth, to recognizing that these are matters of indifference in comparison with virtue, which is the only thing that is good in itself. The salient feature of social appropriation is progressing from the desire to benefit those who are “our own” (oikeios) in a straightforward way, above all, our own children, to benefiting, in principle, anyone. Both types of appropriation are presented as, fundamentally, natural to human beings as rational animals. Although there is no surviving theoretical discussion of the exact relationship between these two kinds of appropriation, I think that Stoic practical ethics are typically shaped so as to promote the co-ordination of these two types of development.68 Epictetus, here and elsewhere in the Discourses, mostly avoids the technical language of Stoic ethical theory (such as indifferents).69 But his programme makes best sense, I believe, if understood against the background of Stoic thinking about development. Thus, the first topic can be understood as identifying the key theme of personal appropriation, that we should direct our desire at progress towards virtue (which is, fundamentally, “up to us”) rather than at obtaining conventional “good things”, which are not “up to us” in the same way.70 It is the failure to secure such goods that, as Epictetus highlights, leads to passions such as anger or grief. By contrast, the completion of personal appropriation leads to the kind of cohesive and structured character-state that is marked by “absence of passion” (apatheia).71 The second topic expresses themes that bridge personal and social appropriation. It partly reflects the idea, crucial for personal appropriation, that development towards complete virtue consists in selection of “appropriate actions” (kath¯ekonta).72 But it also expresses the idea that we are naturally disposed to benefit others, or, as Epictetus puts it, “to maintain my natural and acquired relationships” and the “roles (pros¯opa) to which these give rise”.73 By inference, the second topic also involves the combination and correlation of the two types of development. This consists, for instance, in ensuring that our relationships are informed by our progressive understanding of the absolute value of virtue in relation to indifferents, a theme we can see expressed elsewhere in the Discourses.74 The third topic presents the outcome of the programme as the kind
68 For this suggestion, see also Gill 2004, 120–124; on the relationship between personal and social appropriation, see Gretchen Reydams-Schils, “Human Bonding and Oikei¯osis”, Oxford Studies in Ancient Philosophy 22(2002), 221–251; The Roman Stoics: Self, Responsibility, and Affection, Chicago, IL: Chicago University Press 2005, 53–82. 69 But see, exceptionally, on indifferents, Epictetus, Diss. 2.6; on Epictetus’ conceptual idiom, see Inwood 1985, 115–119; Long 2002, especially chs. 6–9. 70 See Epictetus, Diss. 3.2.1 (LS 56 C(2) ); also Cicero, Fin. 3.20–1 (LS 59 D(4–6) ). 71 Epictetus, Diss. 3.2.3 (LS 56 C(5); on the structured character-state and apatheia that is the mark of wisdom, see LS 63 L–M, 65 W. 72 Cf. Epictetus, 3.2.2, 4 (LS 56 C(3, 6) ) and Cicero, Fin. 3.20 (LS 59 D(3) ). On ‘appropriate actions’ and social behaviour. see LS 59 E(2), Q; see also Inwood 1999, 100–126. 73 Cf. Epictetus, 3.2.4 (LS 56 C(6) ) and Cicero, Fin. 3.62–3 (LS 57 F(1–2) ). 74 See e.g. Epictetus, Diss. 3.3.5–10. 3.24.84–8, also 1.2.1–11. See further Inwood 1985, 123–124; Gill 1988, 187–192; 2000, 608–611; Long 2002, 232–244, 247–249.
52
C. Gill
of cognitive security that ensures that we do not rashly accept or assent to appearances that have not been tested. The qualities indicated here, such as infallibility and uncarelessness, are those which are associated elsewhere with the state of wisdom that is the target and ultimate outcome of ethical development.75 The features of Epictetus’ thought and language noted earlier (and highlighted by scholars such as Kahn and Long) can be linked with this three-topic programme. It is plausible to connect the first stage of Epictetus’ programme with his recurrent advice to direct desire towards what is “up to us”, “ours” and “within prohairesis”, rather than what is not “up to us”, what is not “external”, and what “falls outside prohairesis”.76 More broadly, Epictetus’ emphasis on the importance of prohairesis and on what is “up to us” conveys the point that the whole programme – that is, the whole development towards complete virtue and happiness – depends on the exercise of rational agency that is part of our fundamental human character.77 When Epictetus says that we are our prohairesis,78 this is best taken as a way of conveying the same – essentially ethical – point, rather than being a characterization of our personal identity. (In fact, it does not hold good of our personal identity, since in Stoicism we are not just “minds” or loci of rational agency, but psychophysical wholes and rational animals, as Epictetus himself acknowledges in some of his phraseology.)79 The theme of “examining impressions” also makes better sense when conceived as part of this programme. What is involved is not just selfexamination for its own sake or as an exercise in introspective reflexivity. We should examine our impressions before giving assent as part of a programme of development in ethical understanding. The ultimate outcome is the kind of “infallibility and uncarelessness” that marks complete wisdom (and the correlated cohesion and structure of character), and it is this kind of goal that shapes the project of “examining impressions”.80 The points made so far about Epictetus’ three-topic programme may help to explain why it is plausible to present the programme, and the other features of his thought, as reflecting an “objective-participant” conception of personality. The most obviously “participant” dimension of the programme is the second topic, which centres on motivation towards the “appropriate actions” (kath¯ekonta) that
75 Cf. Epictetus, Diss. 3.2.5 (LS 56 C(7) ) and Diogenes Laertius, 7.46–8 (LS 31 B, especially (3) and (6) ). See also Long 1996, 92–5. 76 See Epictetus, 3.2.1 (LS 56 C(2) ); also Ench. 1, 2, 5, Diss. 1.1.7–12, 1.4.1–3, and text to n. 46 above. 77 For the view that Epictetus’ emphasis on what is and is not “up to us” reflects an ethical message rather than an independent claim about voluntariness or agency, see S. Bobzien, Determinism and Freedom in Stoic Philosophy, Oxford: Oxford University Press 1998, 331–338. 78 See e.g. Epictetus, Diss. 1.1.23, 1.20.17–18, 2.22.19. On this type of passage, see Sorabji, 24 and 33 above. 79 See e.g. Epictetus, Diss. 2.23.3, 3.3.22. For this line of interpretation, see Long 2002, 158–62, 199, 207–10, 222–9. See further Gill 2006, 66–73, 96–100. 80 See n. 75 above.
The Ancient Self: Issues and Approaches
53
derive from one’s social relationships (for instance, “as a son, a brother, a father, and a citizen”).81 The idea of “objectivity” emerges most strongly in two features of Stoic thinking that are implied in the programme. One is the claim that human beings are constitutively or “naturally” capable of carrying through the kind of personal and social appropriation that makes up ethical development.82 The other is the correlated claim that human beings are naturally capable of the cognitive development that leads them to recognize the true nature of the good, a claim associated with Stoic thinking on “common notions” and “preconceptions” (prol¯epseis).83 These claims are implied in Epictetus’ repeated theme that the pursuit of virtue and happiness is “up to us” (through completing this kind of programme), that is, it is a fundamental part of our nature. As highlighted earlier, in Stoic theory, personal and social development are two correlated aspects of this (objectively conceived) process,84 and in this sense appropriation also involves social and interpersonal participation. To put the point differently, social and personal ethical development go hand in hand and need to be co-ordinated, an idea implied in the second topic of Epictetus’ programme and made more explicit elsewhere.85 In this way also, Epictetus’ programme, and the Stoic theory lying behind it, can be seen as presupposing an objective-participant conception of personality. Epictetus’ programme also serves to illustrate key features of the conception of personality that I associate with the structured self in Stoic (and, in a different way, Epicurean) philosophy. The salient mark of this conception is a striking, even paradoxical, combination of a strongly naturalistic picture of human beings as psychological and psychophysical wholes with certain radical ethical claims stemming from Socratic thought. The main ethical claims are these. (1) The achievement of happiness is “up to us” through virtue and rational reflection in a way that is not constrained by one’s inborn nature or social situation. (2) Happiness involves a time-independent, invulnerable perfection of character, marked by freedom from distress or passion. (3) Only the fully rational and virtuous person is fully integrated and coherent while non-wise people are psychologically and ethically incoherent and live incoherent lives. (This is combined with the belief that we should all, none the less, take the normative rational person as our model.86) This combination of features is expressed both in Epictetus’ programme and the Stoic theories presupposed by this programme. As indicated earlier, this programme
81 Epictetus, Diss. 3.2.1, 4 (= LS 56 C(3, 6) ), also 1.4.11, 3.12.13, 4.10.13. On “appropriate actions” and social behaviour, see further LS 59, especially E(2), Q; also Inwood 1999, 100–126. 82 The naturalness of the process involved in each aspect of appropriation is emphasized in, for instance, Cicero, Fin. 3.21–2 (59 D(4–6), 3.62–63 (LS 57 F(1–2) ). 83 See LS 40 N, S, 60 B–D, F. On Epictetus’ reliance on this aspect of Stoic theory, see Long 2002, 79–85. 84 See n. 74 above. 85 See text to nn. 72–74 above; also Reydams-Schils 2002; 2005, 53–82. 86 See further Gill 2006, especially 380–384, also 75–96, 130–132.
54
C. Gill
rests on central Stoic ideas, especially the conception of ethical development as appropriation and the belief that all human beings have the cognitive capacity to know the good. These theories are “naturalistic” in the sense that these capacities are attributed to human beings as embodied rational animals, with the potential to develop though social interaction and reflective living.87 They also presuppose psychological holism in that the development of ethical understanding is taken to bring about changes in affective reactions without assuming the need for independent training of a supposedly non-rational “part”. The development of complete wisdom brings about both a unified and structured type of knowledge and a stable and coherent character, marked by “absence of passion” (apatheia) and by rational “good emotions” (eupatheiai).88 Those theories also provide a naturalistic and holistic framework for the radical Socratic claims outlined earlier. The idea of development as appropriation presupposes as a fact that human beings are constitutively capable of developing towards perfect virtue and happiness, regardless of their inborn nature and social situation.89 The completion of this process is taken to bring “invulnerable” perfection of character and absence of passion. The wise person’s perfection is presented as the only possible form of complete psychological and ethical coherence, by contrast with which the characters and lives of nonwise people are all (and equally) incoherent. None the less, this state of perfection is appropriately taken as the goal for ordinary non-wise people (in effect, all of us) because it represents a target that is in principle attainable and is the only fully coherent norm.90 Epictetus’ programme of practical ethics, taken with other recurrent motifs in his Discourses, in reflecting these Stoic theories, thus illustrates the idea of a “structured self”, as outlined here, as well as an “objective-participant” conception of personality. The interpretation of these features I have suggested can be taken as an alternative to the (in my terms) “individualist” or “subjective-individualist” readings of these themes offered by Sorabji, Kahn, Long, and Foucault. Epictetus’ demand to focus on prohairesis (or his claim that we are prohairesis), taken with the refocusing of desire in the first topic of the programme, implies the idea that it is “up to us” – and open to us – to develop towards complete virtue and happiness, regardless of our inborn nature and social situation. The first topic sets out an overall target or orientation for our interpersonal and social engagement and for the type of affective reactions especially associated with this engagement, as indicated in
87
See nn. 82–83 above. Contrast, for instance, the (non-naturalistic) Platonic theory of learning as “recollection” of prenatally implanted truths (on the relationship between this theory and the Stoic conceptions of gaining knowledge, see Dominic Scott, Recollection and Experience: Plato’s Theory of Learning and its Successors, Cambridge: Cambridge University Press 1995. 88 This is implied in Cicero, Fin. 3.20–1 (LS 59 D(3–5) ); see also LS 61A–B, 65 F, W. See further Tad Brennan, “The Old Stoic Theory of Emotions” in Juha Sihvola and Troels Engberg-Pedersen (eds.), The Emotions in Hellenistic Philosophy, Dordrecht: Kluwer 1998, 21–70, see 21–39. 89 See LS K–L (also references in n. 82 above). 90 See LS 61 A, B(8–11), I, N–O, T, 65 R, U–W. See further Gill 2006, 75–96 and 129–145.
The Ancient Self: Issues and Approaches
55
Epictetus’ second topic.91 Implied here and in the third topic is the idea that our affective state (our emotions and desires) are shaped by the development of our understanding and objectives, without any need for training of a supposedly separate, non-rational “part”.92 Epictetus’ recurrent emphasis on the importance of “examining impressions” before giving assent, taken with the advice in the third topic on establishing “security” in assenting to impressions, reflects the idea that all human beings have the cognitive resources to achieve knowledge of the good and the kind of coherent, structured understanding that constitutes wisdom. As Epictetus suggests in characterizing this state as one of “infallibility and uncarelessness” (and also absence of passion), this also represents the state of “invulnerable” perfection of character that we should all take as our goal.93 Failure to achieve this condition leaves us all in a state of relative psychological and ethical incoherence, though also permanently capable of aspiring to perfect wisdom.
Conclusion In this essay, I have used the discussion of two features of Hellenistic and Roman thought as a way of exploring the implications of some current approaches to the ancient idea of self and the issues, both conceptual and interpretative, raised by these approaches. In the first part, I considered competing readings of Cicero’s presentation of Cato, and of the four-personae theory, by Richard Sorabji and myself. The main focus was on the question whether Cicero’s discussion points towards an intensified interest in unique individuality or whether it is better understood as expressing an “objective-participant” conception of personality – and also whether the latter conception could accommodate a form of interest in individuality. In the second part, I considered competing interpretations of the significance of recurrent features of Epictetus’ Discourses for understanding his version of Stoic thinking about selfhood or personality. Epictetus’ stress on the importance of prohairesis and on examining impressions is taken by a number of scholars as expressing a new interest in individuality or subjectivity in one or other sense of those notions. I have suggested that these features, especially when taken alongside his distinctive three-topic programme of practical ethics, express, rather, an “objective-participant” conception of personality, and, more specifically, the idea of the “structured self”, as this is formulated in Stoic thought. However, my overall aim
91 See especially, “I ought not be impassive like a statue, but maintain my natural and acquired relationships”, Epictetus, Diss. 3.2.4 (LS 56 C(6) ). 92 See especially, ‘The third has to do with infallibility and uncarelessness, or acts of assent more generally’ (Epictetus, Diss. 3.2.2 (LS 56 C4) ), taken with the references to passions and their avoidance in 3.2.3 (LS 56 C(5) ). 93 See Epictetus, Diss. 3.2.2, 5 (LS 56 C(4, 7) ); also nn. 75 and 89 above.
56
C. Gill
has not been to argue for a specific line of interpretation, but rather, by reference to these examples, to display the character and implications of different ways of analysing ancient material in terms of – what we call – “the self”. University of Exeter, UK
Assumptions of Normativity: Two Ancient Approaches to Agency Miira Tuominen
Introduction In the recent scholarly literature concerning the notion of self in antiquity, increasing attention has been paid to the fact that those ancient discussions that we can best conceptualise by using the notion of a person (or ‘self’) involve a strong normative dimension. Tony Long has put it vividly: the crucial ancient question related to personhood is what we are to make of ourselves.1 In the ancient framework, if we succeed in transforming ourselves into ideal agents, we become virtuous and virtuous activities taken as a whole form a good human life. In order for us to be able to mould ourselves in a relevant way, we need to be able to make overall decisions concerning our life. Such decisions also require freedom of choice. These capacities imply responsibility for one’s actions and the possibility of being good or bad. In other words, human beings are moral agents. In some cases the demarcation line between agents and non-agents is easy to draw. To borrow Mary Margaret McCabe’s example: we are not to blame lampposts – even though we often do, I would add – if we bump into them, but it is perfectly reasonable to ethically evaluate the conduct of (adult) human beings.2 However, there are cases like madmen, children, and animals in which the notion of agency is more problematic. This essay focuses on the question of how normative considerations or suppositions enter ancient discussions about moral agency in two authors,
1
A.A. Long, “Ancient Philosophy’s Hardest Question: What to Make of Oneself?”, Representations 74(2001), 19–36. Cf. also Mary Margaret McCabe, Plato’s Individuals, Princeton, NJ: Princeton University Press 1994, ch. 9; and John M. Cooper, “Stoic Autonomy”, Social Philosophy and Policy 20(2003), 1–29. This article was reprinted in his book Knowledge, Nature, and the Good. Essays on Ancient Philosophy Princeton, NJ: Princeton University Press 2004, 204–244 (page numbers are to this reprint). Christopher Gill’s recent monograph The Structured Self in Hellenistic and Roman Thought (Oxford: Oxford University Press 2006) offers a detailed examination of the Hellenistic theory of the self. 2 McCabe 1994, 264.
P. Remes, J. Sihvola (eds.) Ancient Philosophy of the Self, © Springer Science+Business Media B.V. 2008
57
58
M. Tuominen
Plato and Epictetus. To approach this topic, two more specific questions must be formulated: 1. What criteria must a creature satisfy in order to be a moral agent? Or, what distinguishes moral agents from other creatures? 2. What criteria must a creature satisfy in order to be an ideal moral agent? That is: What distinguishes best moral agents from other moral agents? Of these two questions, only the second one is explicitly normative; the first one may be understood as being descriptive. Even though not formulated in terms of moral agency, the second question has a prominent role in ancient ethics: all descriptions of a virtuous person, an ideal human being, can be seen as providing an answer to it. The first question is not equally central, but we do find rather clear indications that at least some implicit criteria were used for moral responsibility. For example, madmen were not considered fully responsible for their actions, neither were drunkards. Our discussion in the following centres on the question of how these two questions were interrelated and in what sense they may be seen to involve normative suppositions. Are some normative standards necessary already in the answers to the first question? In more concrete terms, is it the case that to become a virtuous person, an ideal agent, one just needs to maximise the general criteria of a moral agent? Or is it so that the first question needs to be answered in valueneutral terms and only the second one allows for an answer referring to ideals? In the following, it will be shown that the two questions can, to some extent, be separated in both Plato and Epictetus. Even though the answers that these two authors give to the questions differ, both share the supposition that the general criteria of agency, i.e. in the sense of the first question, are not entirely value-neutral or non-normative. The most important normative elements are contained in the notion of reason. Reason is not understood instrumentally and it has a specific connection with truth. There has been, and still is, some scepticism towards the idea that the notion of self was used in antiquity at all. Rarely, and with good reason, such scepticism is formulated as pertaining to the notion of persons as agents. There does not seem to be much point in doubting whether such writers as Plato and Epictetus operated with at least some implicit notion similar to that of a moral agent.3 Yet it is undeniable that this notion was not formulated as a philosophical topic as such; no Greek term would correspond to that of an agent. The following discussion can be read as a clarification of how assumptions of agency function in Plato and Epictetus.
3 Cf. also McCabe who simply says: “[…] of course, Plato concedes that persons are moral agents” (1994, 264).
Assumptions of Normativity: Two Ancient Approaches to Agency
59
Criteria of Agency in Plato Parts of the Soul In the fourth book of the Republic, Plato presents the idea that the human soul has distinct sources of motivational energy that can, at a given moment, be in conflict with each other.4 However, it is not only at moments of conflict that a split opens up in us. Rather, it is assumed to be a permanent fact that these “parts of the soul” exist. Plato’s discussion is very well known and it will only be presented in outline here. The aspects of the soul form a hierarchy of parts of which the lowest, namely the appetitive one, basically confines itself to pulling the person towards something; its appetites are probably not even propositional.5 The middle one, the spirited part, involves more advanced cognitive functions and acts as reason’s ally against the lowest part (e.g. IV, 441a).6 Reason is the highest part and it has a special role in
4
Motivational conflict is in today’s discussion on the criteria of agency considered a kind of hard case for the theories. Such conflicts are typically construed as conflicts between a simple desire and a higher order desire concerning that particular desire. For example, I might want to tell a lying promise but not want that desire to guide my action. However, as Michael Bratman puts it: such cases are difficult because the mere idea that a volition is about another volition does not seem to provide us with a sufficient answer to the question of why it is that that particular desire in the conflict situation is the agent’s desire. See Bratman, “Reflection, Planning, and Temporally Extended Agency”, The Philosophical Review 109/1(2002), 35–61. For the contemporary analysis of motivational conflicts cf. also Charles Taylor, Philosophical Papers 1: Human Agency and Language, Cambridge: Cambridge University Press 1985, 15–44. The seminal work on higher order volitions and agency has been published by Harry Frankfurt, “Freedom of the Will and the Concept of a Person”, Journal of Philosophy 67/1(1971), 5–20 (cf. the essays in his The Importance of What We Care About: Philosophical Essays, Cambridge: Cambridge University Press 1988). For a discussion of Frankfurt’s analysis, see, e.g. Gary Watson, “Free Agency” in Watson (ed.), Free Will, Oxford: Oxford University Press 1982, 96–110. Note that when Plato discusses motivational conflict by means of the famous example of Leontius, he is explicit in saying that Leontius is overpowered by his desire to look at the corpses. This means that Leontius’ desire is the one which goes along the judgment of reason (the desire not to look at the corpses). However, the conflict is not formulated in terms of first order and second order volitions. 5 Cf. Julia Annas, An Introduction to Plato’s Republic, Oxford: Oxford University Press 1981, ch. 5. Annas, however, expresses some reservations concerning this point. For example, the appetitive part must agree to being ruled; this agreement seems to be propositional or at least it involves more advanced cognitive functioning than mere pulling towards (a drink, for example, in IV, 439d). In addition, later on in the Republic (VIII, 580d–581a) the appetitive part is said to pull one towards money and gain. Annas infers that this implies that at least some elementary form of means-ends reasoning is supposed. However, Plato is rather clear that means-ends reasoning of this sort is a function of reason and is a case where the lower parts have enslaved it (553d1–3). In any case, Annas also concludes that the appetitive part’s desire itself looks like “a blind craving” (p. 139). 6 Cf. also the conclusion of the Leontius example in IV, 440a: “It certainly proves that the spirited part sometimes makes war against the appetites, as one thing against another.” (Trans. Grube rev. by Reeve from John M. Cooper (eds.), Plato Complete Works, Indianapolis/Cambridge: HackettCooper 1997.)
60
M. Tuominen
the hierarchy. It is not merely restricted to desiring some definite set of objects – even though it does have desires and pleasures of its own (e.g. IX, 581b–e) – but it has a long term perspective towards life and reality, and the motivational structure itself. Reason’s superiority in the hierarchy holds even if the person’s action is constantly guided by motivations of the lower parts. The assumption of reason’s superiority is not only understood in the sense that reason is able to perform more complicated and more demanding cognitive functions than the lower parts. Reason also is so constituted as to recognise the truth. This is not solely the claim that reason alone is capable of entering the realm of forms. Rather, the claim involves the assumption that the lower parts have a tendency to guide the person in a wrong direction. Either they are likely to desire things that are not really worth desiring, or they desire the objects excessively. Christine Korsgaard has contrasted Plato’s model for agency in the Republic to what she calls the “Combat-model”.7 According to this model, there are conflicting energies or drives in the soul, and at moments of conflict the agent comes to identify herself with one of them. In order to live rationally, in the Combat-model, we should always reject any desires that might go against our reason’s dictates. Interestingly, Korsgaard argues that in the Republic we do not encounter this Combat-model but, rather, an alternative “Constitutional-model”. In the Constitutionalmodel, the agent is not identified with any of her parts or drives, but she is something over and above them. Korsgaard likens this to the way in which a city-state’s constitution is something over and above the citizens and officials of the polis. When, in the Constitutional-model, the agent is acting according to reason, it is not that the agent would identify herself with reason. Rather, it is because it is according to the ‘constitution’ of the just soul that reason should rule. As many have pointed out, however, the Republic also contains references to the Combat-model.8 Yet, in book VIII corrupted souls are distinguished from just ones on the basis of the condition that only in an unjust soul parts of the soul are enslaved (for the oligarchic soul, VIII, 554d).9 In a just soul such forceful enslavement should not take place and, hence, the Combat-model would only apply to inferior souls. According to Korsgaard, justice in the Republic offers normative standards for action. As she notes, however, this approach seems to entail the following problem: an agent that fails to satisfy these standards fails to be a good agent to the extent that it can be asked whether that person acts at all. If only good action counts as action in the proper sense, it seems that there is no such thing as bad action.
7
Christine Korsgaard, “Self-Constitution in the Ethics of Plato and Kant”, The Journal of Ethics 3(1999), 1–29. She traces the Combat-model back to Hume (A Treatise of Human Nature book II, part 3 section 3). According to Korsgaard, the model is described and not explicitly rejected in the Treatise. 8 For the idea that reason should use force to rule the lower parts, see, e.g., IV, 430e. 9 Cf. Jonathan Lear, “Inside and Outside the Republic”, Phronesis 37/2(1992), 184–215; Pauliina Remes, Plotinus on Self: The Philosophy of the “We”, Cambridge: Cambridge University Press 2007a, section IV.2.1.
Assumptions of Normativity: Two Ancient Approaches to Agency
61
In a complex section, the details of which cannot be considered here, she argues that this problem can be circumvented for the following reason. Choosing good and bad actions are not two distinct activities, but in essence choosing bad actions is the same activity as choosing good actions; it is just done badly.
Are Normative and Descriptive Criteria of Agency Distinct in Plato? Let us now go back to the two questions distinguished in the introductory section, namely: (1) What are the criteria that distinguish agents from other kinds of things or creatures? (2) What are the criteria that distinguish best agents from other agents? In the light of these questions, Korsgaard’s claim can be formulated as follows: the description of a just person in the Republic offers an answer to the second, normative question. This seems undeniable. It can now be asked whether or in what sense failing to satisfy the standards of agency in the strong normative sense (i.e. failing to be virtuous) entails failing the standards of agency in the first sense (i.e. failing to be an agent)? There is some evidence that Plato would take the two standards as being closely connected. For example, an oligarchic person who is in a state of internal civil war is on the threshold of falling apart and ceasing to be one agent (VIII, 551d). Similarly, a democratic person, who simply pursues whatever desires or impulses happen to occur in her, is not properly acting and not making proper choices: she is merely a slave to chance. The desires come and go “as if [they] were chosen by lot” (VIII, 561b).10 Conversely, Plato also seems to indicate that a just person, ideal agent, is not only morally superior to the vicious ones but an agent in a fuller sense. He first describes the just person’s inner harmony and peace: the parts of the soul do not meddle in each other’s business (m¯e […] allotria prattein). Then he concludes that only a person who is in such inner harmony is capable of acting in a proper sense (hout¯o d¯e prattein ¯ed¯e, IV, 443e2). In order to illustrate the difference between the two standards, we can borrow Korsgaard’s example. In the example a person (called Jeremy) is too restless to read for an exam. He goes for a walk but ends up going to a book-store in front of which he meets a friend with whom he goes for a beer. In the bar, however, he notices that the noise gives him a headache and decides to go back home.11 In the end, Jeremy does not read for the exam, he barely goes for a walk, does not buy the book – and does not even drink the beer. He fails to satisfy standards for agency because
10 This is compared to how in democratic cities people are assigned to ruling positions by lot (VIII, 557a). 11 Korsgaard 1999, 19; she borrows the example from another article of hers “The Normativity of Instrumental Reason”.
62
M. Tuominen
he is not properly acting but is rather driven by external circumstances. In this case we might ask whether Jeremy’s – or Plato’s democratic person’s, for that matter – doings count as action at all. Even though it is clear that in some sense Jeremy is a very unsuccessful and inefficient agent, it may be asked whether his failure is ethical in any very strong sense. In fact, what Jeremy is doing seems rather harmless. Further, if we tie the general and normative standards of agency together too closely, the normative force of effectiveness might become too strong.12 Mere persistence, when understood as sticking to one’s principles in every situation, is not necessarily good without qualifications. Persistence is valuable only if the principles themselves are reasonable. If we go back to the Republic,13 where the emphasis is on the moral quality (or the character) of the agent, not the action, similar concerns can be raised in the following way: Is it conceivable, on the one hand, that someone could be very effective and agent-like and yet fail to be virtuous? Prima facie, at least, it could seem that being a very effective agent can make one rather insensitive to other people’s needs. To some extent it would seem unjustified or anachronistic to read this distinction into Plato at all. In the middle books of the Republic, justice is defined as inner harmony and the just person is claimed to be acting in a full sense (IV, 443e2). Does this not show that the two aspects coincide in Plato? To some extent we could say so, but a specification needs to be made. The passage (IV, 443e) does show us that the questions of who is a most agent-like agent and who is the best agent in the strong sense are answered in the same way: the just person. However, the very same passage implies that the two questions must be separated from each other. The point that Plato makes requires this distinction: in order to make the claim that more effective or agent-like action follows from being a superior agent in the ethical sense, one must differentiate between the two aspects. Even though the discussion of a just person in the fourth book seems positive on the point that it is not possible to be an efficient agent without also being virtuous, the eighth book is not equally unambiguous. On the one hand, Plato seems to confirm that if reason rules, it cannot rule unjustly. In inferior souls the criterion that reason should rule is not satisfied, since calculating about how to reach bad ends implies enslavement of reason (cf. 553d1–3). On the other hand, however, Plato hesitates about whether reason’s rule guarantees justice. In the course of the argument, Socrates adds that within a just soul reason should rule like a king, not as a tyrant (580c). If it were utterly impossible for reason to rule in any other manner than well, this remark would seem rather pointless. Therefore, the requirement that
12 An analogous question is today posed in connection with autonomy; for the discussion, see, e.g., Onora O’Neill, Bounds of Justice, Cambridge: Cambridge University Press 2000; Christine Korsgaard, The Sources of Normativity, Cambridge: Cambridge University Press 1996; Henry S. Richardson, “Autonomy’s Many Normative Presuppositions”, American Philosophical Quarterly 38/3(2001), 287–303. In the background of the discussion we find Kant’s famous claim according to which autonomous action will be moral action. 13 What justice really is, is not the external action of that person but the inner action (h¯e entos praxis, IV, 443c10–d1). Therefore, persons, not their action, are the main objects of evaluation.
Assumptions of Normativity: Two Ancient Approaches to Agency
63
reason should rule within the soul provides almost sufficient normative standards for action, but not entirely. Something more seems to be required. At this point it is useful to note that the emphasis in the discussion about justice is clearly on the inner state – or, as Plato puts it “inner action” (h¯e entos praxis) – of the person and not social aspects of justice. Nonetheless, Socrates also pays attention to some traditionally guided expectations concerning justice and claims that the person who is just from the internal perspective satisfies those expectations (see, e.g., IV, 442e–443b).14 The argument, however, goes very quickly and appears rather feeble. Socrates simply seems to assume that it follows quite trivially that if someone is just from the point of view of inner action, then all of his or her external actions can also be described as just in a traditional sense: the just person is not violating laws and so on.15 This argument is extremely interesting but can not be treated in more detail in this context. We have now considered the two questions concerning agency on the upper scale of human agents, i.e. in the case of virtuous and just souls. What about the lower types of agency, then? Does moral weakness entail reduced responsibility? When Plato points out that a tyrannical person is in the state of internal civil war, he suggests that the person’s agency has been diminished. However, he does not seem to imply that the person would completely fail to satisfy the criteria of agency. Even though in a sense a tyrannical person is mad, he or she is not mad in the sense of not being responsible for his or her action (for the ‘madness’ of the tyrannical person, see, e.g., 573a–c).16 By contrast, when someone has literally gone out of his mind, then that person is not responsible for whatever he does. Socrates argues (331c) that if someone has borrowed weapons from a friend who, in the meantime, has gone mad, the weapons should not be returned to their owner. All in all, Plato leaves room for bad action with responsibility.
14 For a discussion of the traditional assumptions, see Rachel Barney, “Callicles and Thrasymachus” in Edward N. Zalta (ed.), The Stanford Encyclopedia of Philosophy Fall 2004 Internet Edition, see http://plato.stanford.edu/archives/fall2004/entries/callicles-thrasymachus; cf. Gregory Vlastos, “Justice and Happiness in the Republic” (p. 132), in his Platonic Studies, Princeton, NJ: Princeton University Press 1981, 111–139. 15 As Barney (2004) shows, however, the traditional background is more complex than initially might seem. An additional twist comes from the amoralist challenge to morality presented by Thrasymachus in the Republic and Callicles in the Gorgias. Further, it is important that some of the amoralist revisions to traditional morality are left uncontested by Socrates. 16 Cf., however, the Timaeus, where Plato claims (86d) that moral weakness is essentially caused by inherited physical weakness and bad upbringing, and that even though it is common to blame people for, e.g., excessive sexuality, such properties are not in our power and we are not responsible for them. Bad political circumstances may also contribute to our corruption (87a7–b4), but it is not in our power either in what kind of environment we were born. In that connection Plato also lumps together clinical madness and ignorance as forms of irrationality (anoia). Some scholars, e.g. Alfred E. Taylor in his old Commentary on Plato’s Timaeus (Oxford: Clarendon 1928) have argued that the position that is outlined in the Timaeus is incompatible with the other dialogues. For the position that the contrast is not as strong as Taylor claims, see Christopher Gill, “The Body’s Fault? Plato’s Timaeus on Psychic Illness” in M.R. Wright (ed.), Reason and Necessity in Plato’s Timaeus, London: Duckworth 2000, 59–84.
64
M. Tuominen
Is the difference between a mad person and the tyrannical person a qualitative one, or is it one in degree? Plato would probably concede that madmen are bad agents in the sense of being deficient, and so are those whose soul is in a state of tyranny. Further, at times he seems to indicate that for such deficient agents as people with tyrannical or democratic souls, the rights or decision-making powers within the polis should be smaller than for those who are in a state of harmony. This idea seems to indicate that the tyrannical and democratic persons are not fully responsible for their own action either. Problems related to this idea have, of course, been much disputed and we shall not go deeper into that discussion here. Suffice it to say that both of the considerations just presented point in the direction that, in Plato, the criteria of agency would come in degrees: someone can be a better or worse agent – both with respect to virtue and inner harmony and in the sense of being more or less agent-like. Further, diminished success in satisfying the strong normative criteria seems to entail diminished responsibility, up to the point of clinically insane people, who are not responsible for their actions at all. If we want to abstract some descriptive criteria of agency from Plato’s discussion in the Republic, we will find at least the following: agents consist of nonhomogenous hierarchically ordered parts with motivational energy and cognitive powers. The assumption according to which agents are always complex structural wholes entails that in a sense human action for Plato is always interaction between these parts. This assumption, as is often noted in the literature, makes unity a pressing issue for Plato. Good agents are assumed to be more unified than bad ones.17 Greater unity is achieved when the parts function for the well-being of the whole, not only selfishly pursuing their own benefit. Analogously, even a group of thieves falls apart unless they share common goals (Rep. I, 351c). Therefore, rather than emphasising that the unity of the whole is explained by reference to the unity of the parts,18 it seems more important that the whole is the more unified the less selfish the constituent parts and the more they function for the good of the whole. We may now draw the following conclusion from the discussion about the Republic above. Plato often supposes that the answers to the two questions concerning agency distinguished in the introductory section coincide and that the second kind of criteria are prior to the first ones (no efficient agency without virtue). Further, this seems to imply that agency is a matter of degree: bad agency is deficient both from the point of view of the notion of action and from an ethical point of view. Yet, some doubt has been cast over the idea that the criteria for agency – most importantly the criterion that in a just soul reason should rule – would constitute a sufficient source for normativity tout court. This is because of the additional qualification in a just soul reason should rule like a king, not like a tyrant, made in book VIII. The limits of agency do not coincide at the lower end of the scale either: there are bad actions for which the agent is responsible.
17 18
E.g. McCabe calls unity “an honorific title” for Plato (1994, 268). For the assumption, see McCabe 1994, 269; cf. Vlastos 1981 ch. 5.
Assumptions of Normativity: Two Ancient Approaches to Agency
65
Criteria of Agency in Epictetus Responsibility and the Power of Making Decisions Epictetus makes it quite clear that what is distinctive of agents is the rational capacity of giving assent to or withholding it from impressions (phantasiai). He calls this capacity prohairesis.19 From Epictetus’ perspective, we are responsible for those things or actions that are in our power (ta eph’ h¯emin). The only thing, according to him, that is completely in our power is the capacity to use the impressions (1.1.7), namely our prohairesis. The idea that we accept or reject impressions and it is in our power to do so is, of course, not Epictetus’ invention. It goes back at least to Chrysippus. However, as Susanne Bobzien has argued, Epictetus proposes a new interpretation of this common Stoic idea. Epictetus wants to exclude from the class of things that are in our power all such things that might be externally hindered.20 Whereas, according to Chrysippus, it is in my power to go for a walk, according to Epictetus it is not.21 Because it is possible that a tyrant ties me up, in which case I will be unable to go for a walk, going for a walk is not completely in my own power but I need external reality to contribute. Similarly, all other bodily movements are beyond my power as well. Epictetus also uses the following expression to underline that our prohairesis is in our power. According to him, not even Zeus can conquer my prohairesis (t¯en prohairesin de oud’ ho Zeus nik¯esai dynatai, Diss. 1.1.23). Bobzien seems to be correct in saying that Epictetus is not by this phrase referring to the distinction between some things Zeus could make me do and others in which I could not be conquered like this. However, the quote is very important in stressing that there are things that are my doing, not Zeus’. There are things that I can and indeed must leave to Zeus to decide but there also are decisions that I must make myself. For those decisions I am responsible. In the present context we cannot go deeper into the question of whether this understanding of Epictetus’ formulation of the prohairesis has implications on the relationship between determinism and freedom of choice in Stoicism. In any case, it is a common Stoic tenet that determinism and freedom of choice are compatible and that we have full responsibility for our decisions in a deterministic world.
19 I shall leave prohairesis untranslated here. A.A. Long uses the term ‘volition’ in his Epictetus. A Stoic and Socratic Guide to Life, Oxford: Clarendon 2002. 20 Epictetus also modifies the way we speak of that which is in our power. Chrysippus speaks of ‘being in our power’ (to eph’ h¯emin) whereas Epictetus refers to those things that are in our power (ta eph’ h¯emin); cf. Susanna Bobzien, Determinism and Freedom in Stoic Philosophy, Oxford: Oxford University Press 1998, 332. 21 Cf. Bobzien 1998, 331–338; see also her article “Stoic Conceptions of Freedom and their Relation to Ethics”, in Richard Sorabji (ed.), Aristotle and After, Bulletin of the Institute of Classical Studies, Supplementary volume 68(1997), 71–89.
66
M. Tuominen
From today’s perspective Epictetus’ border-lines for agency are quite radical. We already paid attention to the idea that bodily movements are outside of our power. However, Epictetus also widens the scope of agency: he denies that a threat on my life could ever liberate me from my responsibility. Even in those cases, it is completely in my power to assent or not to assent to an impression concerning the value of my life, and I am fully responsible for the decision I make in the situation.22 Once we recognise that agents have the power to give or withhold assent and that some things are in our own power, there is no escape from the responsibility for the things I do, that is, the decisions I make. However, quite surprisingly, Epictetus can be seen to question this point concerning responsibility. When he discusses Medea’s case, he suggests that we should pity rather than be angry with her (Diss. 1.28.10). If we assume that Epictetus at least to some extent follows the conditions for pity Aristotle lays out in the Rhetorics (II, 8)23 this entails that he assumes that Medea’s bad situation is not principally her own fault and, hence, she is not fully responsible for it. Epictetus’ remarks concerning Medea are, of course, parts of his complex rhetoric, and it is not entirely clear whether he actually claims that Medea should not be held responsible for her action. It is possible that Epictetus expects that the student should make the thought complete and add something like the following. We should not pity her because she is responsible for her condition and pity excludes this possibility. Further, Epictetus has taught us not to get emotional at all and, therefore, pity would be misguided in all cases.24 Elsewhere Epictetus in fact seems to expect this response. He points out, somewhat moralistically, that what Medea got wrong was that she did not understand where our power to do as we wish lies (pou keitai to poiein ha thelomen, Diss. 2.17.21). He probably implies that even though Medea was wrong in how she understood this power, she still had it in her. A normal Stoic attitude towards a situation like Medea’s would be that even though it is a common folly among human beings that we esteem wrong things and are hence driven to misguided action, we are responsible for our evaluations. Ordinary folly like this, however, has to be distinguished from clinical madness.
22
There are plenty of stories illustrating this point in the Discourses (see, e.g., 1.1.24–32, 1.2.29, 1.2). For a discussion of Aristotle’s conditions for pity, see Martha Nussbaum, Upheavals of Thought. The Intelligence of Emotions, Cambridge: Cambridge University Press 2001, ch. 6; see also her article “The ‘Morality of Pity’: Sophocles’ Philoctetes and the European Stoics” in Vesa Hirvonen, Toivo Holopainen and Miira Tuominen (eds.), Mind and Modality: Studies in the History of Philosophy in Honour of Simo Knuuttila, Brill’s Studies in Intellectual History 144, Leiden: E. J. Brill 2006, 3–18. 24 Tony Long claims (114) that Epictetus’ reasons for suggesting us to pity Medea is that, given her character, she could not help her value judgments. See A.A. Long, “Representation and the Self in Stoicism”, in Stephen Everson (ed.), Companions to Ancient Thought 2. Psychology, Cambridge: Cambridge University Press 1991, 102–120. But isn’t it one of Epictetus’ basic points that our moral character is in our own power? 23
Assumptions of Normativity: Two Ancient Approaches to Agency
67
The latter does affect our responsibility, at least our legal responsibility, according to the Roman law.25 Even though Epictetus advises the students to pity rather than blame Medea, he does not suppose her to be clinically insane or otherwise free from responsibility. In any case, his attitude towards Medea is most interesting because he is actually rather soft towards her. This attitude might be necessary if any success in the area of philosophical therapy is hoped for, and Epictetus makes an attempt towards this direction. “Stop wanting your husband and you will never fail to get what you want” (II.17.22), he says and thus indicates that it is still in her own power to make a difference and change everything. At the beginning of the very same passage (Diss. 2.17.19–20) Epictetus also praises Medea for her great spirit: she at least knew what it is when one’s wishes are not fulfilled. In that situation she made the decision of killing her children to take vengeance upon her husband, but that did not benefit her in any way. Epictetus is often quite harsh with his students who have minor problems: “wipe your nose, what do you think you have hands for”, he exclaims to a student who complains about a running nose (1.6.30).26 Now, when he has a very difficult case in his hands he transposes to this therapeutic key with paternalistic undertones. Does he assume that Medea has traversed the border-line between agency and non-agency? Or, does he assume that Medea, at the point of making her fatal decision would be convinced by his arguments according to which she will get all she wants if she simply stops wanting her husband? Perhaps neither of these alternatives is quite right, and Epictetus simply provokes his students to think for themselves. As already indicated, Epictetus’ focus probably is on his students. He may be praising Medea in order to contrast her to the students who complain about minor losses and ask for pity. Medea at least was brave27 and did not ask for pity for her serious losses: she lost her husband, her home country, her social role, her children, and her whole fortune. In that situation she made her terribly erroneous decision but at least she did not weep for herself. In a frightening way she understood that, despite having lost all external power, she still had the inner power of making decisions left in her – “Fortune can rob one’s money but not the spirit” (Seneca’s Medea, line 177) – but she got its nature wrong. She understood that some external things are not worth sticking to but she did not understand this about others because she wanted vengeance on Jason at any cost. Independently of how we read Epictetus’ reaction towards Medea, he clearly suggests an answer to the descriptive question concerning agency. What all agents
25 See Cicero, Tusc. 3,11; for the non-responsibility of clinically mad people, see also Seneca, Ben. 2.35,2 and 7.16,5–20,5. Marke Ahonen has discussed the evidence in detail in her licentiate thesis – written in Finnish (Fall 2005, University of Helsinki). She emphasises the ancient Greek and Roman philosophers’ general tendency to make a sharp distinction between ordinary stupidity or moral weakness and clinical madness. Whereas the latter implies non-responsibility, the former does not. 26 Or, with the students who do epistemology or logic for the sake of epistemology or logic alone; he likens them to corpses (1.6.6–8) or advises them to go hang themselves (2.17.34). 27 Cf. Nussbaum 1994, ch. 12.
68
M. Tuominen
share and what distinguishes them from non-agents is that aspect of our rational soul which gives assent to impressions or withholds it from them. At times Epictetus makes it clear that it is rationality that distinguishes us from animals and, hence, indicates that rationality is the decisive feature of agency. It is important to note, however, that even though both Plato and Epictetus consider rationality being central for agency, the connotations are not quite the same. At least in the Republic, the Platonic reaction would be: yes, an agent needs to have rational powers but the spirited and the appetitive part must be there, too. (It might be objectively better if they were not, but it is in any case the human lot to have them.) The motivational structure is intrinsically non-homogenous, and all the desires are not flowing from the same source, so to speak. In the Stoic framework, by contrast, reason itself is the source of all the desires; all our mistakes in action are mistakes of reason. In addition, Epictetus lays much more emphasis on the reason’s decision-making power than Plato does. In Epictetus, decision-making is not confined to action. From the Stoic point of view, all mental activity involves an aspect of decision. Both in the case of impulsive and non-impulsive impressions, we decide whether the impression should be accepted or not. In Plato, by contrast, reason is characterised by its desire and ability to know complex and abstract objects, its capacity to reflect upon the whole structure of the soul and, hence, to rule it. Decision-making, by contrast, is an issue where the whole soul participates.
Normative Standards for Agency As to the normative question about ideal agency, we do not need very much argument to show that Epictetus does consider the question of ideal agency. Expressions concerning an ideal agent, the wise person, are ubiquitous in the sources. One phrase that is especially characteristic of Epictetus is to say that a wise person makes correct use of his or her impressions.28 This way the wise person best uses what is peculiar to moral action in general. Now we need to ask, which of the two different senses of normativity distinguished above is at stake when the wise person is characterised as an ideal agent. Is the wise person a most efficient agent or a most ethical agent? If he or she is both efficient and ethical, which of the two aspects is primary? Or, is it that the two aspects can not be separated in the Stoic framework? Again, it seems rather clear that the wise person is ideal at least in the ethical sense. Only full wisdom counts as virtue, according to the Stoics, and it cannot be lost. The virtuous person does not get involved in any ethically inferior action because she will not assent to a false impression suggesting her to do so. The wise and virtuous person provides the ethical ideal for human action.
28
The expression is frequent in the Discourses. For an instance, see, e.g., Epictetus, Diss. 1.1.7.
Assumptions of Normativity: Two Ancient Approaches to Agency
69
As regards the second point, namely the effective aspect of agency, we can note immediately that some form of elementary freedom that could be called “negative freedom” belongs to all agents. Our decision making capacity is completely in our power and cannot be externally hindered. A Stoic equivalent for Korsgaard’s Jeremy or Plato’s democratic person would be one who is constantly assenting to impressions that one or another external thing is good. From the Stoic point of view, however, this person is constantly acting in the sense that the decisions he or she makes are his or her decisions. Such a person is using that decision making power very badly, but anyone cannot literally be driven by the external circumstances because no external force can conquer our prohairesis. However, in a sense exercising the prohairetic power is more fully actualised in dissenting: there is something so powerfully appealing in the way external things present themselves to us – or how we present them to ourselves – that they seduce us to say “yes, external things are good and I need them in order to flourish”. Therefore, according to Epictetus, all adult human beings are efficient agents in the sense of being capable of using impressions. This capacity is completely in our own power and, hence, guarantees a basic freedom of choice and responsibility for our actions. However, this power can be used more or less effectively – or in a better or worse manner. This entails that even though the capacity of prohairesis itself does not come in degrees, there are degrees in how well we can use it. Those who easily concede any chance impressions that occur to them, are using the decision making power in a lazy and slavish manner (sometimes Epictetus exclaims that everyone who is not wise is a slave, cf. 2.1.24). Nonetheless, this does not entail lack of negative freedom and responsibility for ordinary fools. Rather, it is only compared to the positive freedom (eleutheria)29 of the wise person that all ordinary people are slaves. The idea of positive freedom brings along with it a form of extreme efficacy for the wise person: as Epictetus says very clearly, the wise person never fails to get what he or she wants and never fails to avoid objects of aversion (1.4.1–4). This means that the wise person is effective, not only in the sense of acting in a fuller sense than someone who – like Korsgaard’s Jeremy – is simply floating around, but in the sense that wisdom and ethical agency make us reach our goals more effectively. This claim, of course, involves a certain very specific understanding of what the right goals are, and many might be disappointed when hearing the answer: the only true goal is to attain virtue. In any case, the claim might have been a great advantage for the Stoics in the mutual competition among the Hellenistic philosophical schools all of which tried to establish that their suggested
29
Eleutheria, as Bobzien argues (1998), is a concept altogether different from the elementary ‘freedom’ peculiar to all agents; the latter is not freedom properly speaking but only involves some things’ being in the person’s power.
70
M. Tuominen
form of philosophical therapy is the most effective one.30 As to the extreme efficacy of the virtuous person, many philosophers would disagree with the Stoics. Many would perhaps concede that more acute observation concerning the circumstances and ethical value of our objectives makes us more responsible agents or agents in a fuller sense. However, much fewer would unquestionably agree that such responsibility and care would make us less vulnerable to losses. Efficacy in this sense can be related to truth at least in two relevant ways. Firstly, according to the Stoics, the only way in which it is possible for us to fail to get what we want is that we want wrong kinds of things, namely, something that is external to ourselves. If we manage to grasp the truth concerning what is really worth striving for, the possibility of failure disappears. Secondly, it might seem that if we are wiser or closer to the truth, we will be more likely to find the correct means to reach the ends that we set for ourselves. However, this instrumental sense of rationality and truth is not at all typical of the Stoic analysis. Rather, Epictetus’ advice about how to understand an apparent loss is of the following kind: when analysing what has happened, focus on your inner condition. For example, when people are splashing water at a public bath and I start to get nervous and think that my bathing experience is destroyed, I should remind myself that taking a bath is not the only thing I wanted; what I really wanted is to take a bath in such a manner that my inner peace is maintained (Ench. 4; cf. 24, 2–4). Even though there is a clear difference between the normative standards for rationality provided by the wise person and the elementary rationality that belongs to all adult human beings, not even the elementary rationality is entirely valueneutral. As Epictetus makes clear, all human beings have some true preconceptions (prol¯epseis) about what goodness and justice are.31 This means that all rational creatures, in this case adult human beings, have some true conceptions about normative notions and about human ideals. As Epictetus makes clear, all agree that what is good is profitable and something that should be chosen in all circumstances. Further, there is no disagreement that justice is good or beautiful (kalon). (Diss. 1.22.1) However, as it happens, these preconceptions of the good are often falsely applied; one needs education in order to understand which things are good and which are bad. Going back to the example of Medea, we might ask whether there is a sense in which she succeeds in being an efficient agent even though her conceptions of what is good and what should be done are misguided. As becomes clear in the play, I am referring now to the version by Seneca, Medea is quite efficient in carrying out her plans. Even at the end when Jason tries to catch her and persuade her not to kill their sons, she escapes and manages to do what she wants. However, from the Stoic
30 The mutual competition between Hellenistic philosophical schools is often brought forward by Simo Knuuttila in his oral presentations; cf. also Simo Knuuttila, Emotions in Ancient and Medieval Philosophy, Oxford: Oxford University Press 2004. 31 For a discussion of shared truths, see Michael Frede, “The Stoic Concept of Reason”, in I.K. Boudouris (ed.), Hellenistic Philosophy, Athens: Ionia 1994, 50–63.
Assumptions of Normativity: Two Ancient Approaches to Agency
71
point of view, it would be rather strange to suppose that pursuing a misguided end would be successful – however efficient the person might be in attaining that end. Efficient use of reason requires that one is well informed about the ends, not just about the means of attaining these ends. Therefore, even though there is a sense in which normative and functional rationality may be distinguished in the Stoics32 – all adult human beings are functionally rational but only the wise person is (fully) rational in the normative sense – functional rationality is not completely devoid of normativity.
Reflections on the Relation Between Normative and Descriptive Criteria We are now in a position to pull the main threads together and discuss the material from a more general point of view. In recent literature on agency, philosophers have debated the source of normativity in human action. There is much disagreement concerning even the most basic outlines, but many agree that autonomy is a distinctive mark of human agents and that it is good to be autonomous (at least in some sense of ‘good’). One crucial question is whether normativity can be derived entirely from the notion of autonomy, particularly if autonomy must be understood in value-neutral terms. The suggestion that autonomy is a sufficient source for normativity has been defended by Christine Korsgaard.33 If she is right, our task as moral agents is simply to maximise our autonomy. Her suggestion has met some criticism. One response that is relevant for the present purposes is Henry Richardson’s article where he argues that autonomy cannot function as the sole source of normativity. The reason is that, in order to derive normativity from autonomy, we already need to have some normative standards for autonomy. If we, for instance, understand autonomy as following one’s principles, this kind of autonomy is sufficient as a normative standard only if the principles are reasonable.34 Hence, any attempt to secure normativity on the basis of autonomy is in danger of being circular. The following example from Onora O’Neill illustrates the problem of defining autonomy in value-neutral terms.35 In the example, a young female student comes
32 For the distinction, see, e.g., Christopher Gill, “Did Galen Understand Platonic and Stoic Thinking on Emotions?”, in Troels Engberg-Pedersen and Juha Sihvola (eds.), The Emotions in Hellenistic Philosophy. The New Synthese Historical Library 46, Dordrecht: Kluwer 1998, 113–148. 33 Korsgaard 1996. 34 Richardson 2001. 35 O’Neill (2000) does not articulate the point exactly like this but seems to conceive it similarly.
72
M. Tuominen
to O’Neill’s office. She has taken part in a student provocation and joined young male students in running naked around the campus. When O’Neill asks why the student did this, she argues as follows. Through maximising her negative freedom she also maximised her autonomy, and – she apparently assumes – the professor should be satisfied with her conduct. It would seem questionable to claim that the student’s ethical quality or character was improved by the show.36 However, her assumption that the more autonomous we are, the better agents in an ethical sense we become, is interesting. When commenting on this incident O’Neill does not question this assumption but, rather, argues that the example shows that it is mistaken to assimilate autonomy with negative freedom. A similar point could be made about rationality, if it is conceived in instrumental terms. To reason more effectively offers a normative standard for action only if the ends about which we reason are valuable. Mere instrumental rationality will not be sufficient in satisfying this requirement. Therefore, it seems that we should abandon the supposition according to which autonomy or rationally understood in value-neutral terms should provide us with a sufficient source for normativity. Autonomy only serves as such a source, if some further, normative standards are employed to define it. It is important to note that a similar problem can be formulated independently of the notion of autonomy. In the preceding discussion we made the distinction between normative and descriptive questions concerning agency and found reactions to both questions from the ancient discussion. In the spirit of the example just discussed, we can go back to our questions about agency and ask whether the normative standards of agency can be achieved by fulfilling the descriptive standards simply to a greater degree. Further, does this mean that the general criteria of agency as such provide a sufficient source of normativity? Within the Platonic framework, we noted, an agent consists of a three-fold motivational structure that involves rational capacities. What would it mean to satisfy these criteria in a greater degree? As to the first criterion, it would seem a little difficult to understand how it could come in degrees. How could something be a three-fold structure in a greater degree than something else? Being a three-fold structure seems, rather, to be an all or nothing affair. In its core this seems a sensible reaction to the question. However, this is not all there is to the matter. Plato in fact seems at times to use the first criterion in the way that it does allow of degrees. The possibility for this lies in the structural aspect of the criterion. According to Plato, in ethically better agents (and most of all in a just person), the inner three-fold complex is more coherently structured and hence more unified because the parts have different functions and one characterisation of
36
In her literary memoir Reading Lolita in Tehran, Azar Nafisi, an Iranian professor of English literature who studied in the U.S. in the 1970s and emigrated there in the 1990s, remarks with gentle sarcasm that running naked around the campus was a rather common idea of freedom among American students at that time.
Assumptions of Normativity: Two Ancient Approaches to Agency
73
justice is that all the parts concentrate on their own business and do not intrude into those of the others. Nonetheless, because in Plato’s framework unity is an honorific title, the normative dimension is not excluded from this criterion either. The second criterion, rationality, seems at first sight a much more promising candidate for comparison in degrees. It surely makes sense to say that someone – or even something – is more rational than someone (or something) else. It also seems natural to say that, according to Plato, we should, as agents, aim at being maximally rational. Or, if the claim is specified according to reason’s role in the structure, it makes sense to say that someone is more or less ruled by reason. However, as we saw above, this is not completely correct as a final analysis of the second criterion. In the eighth book, Plato leaves the door open for the possibility that reason’s rule is not a sufficient account for ideal agency. Even if we did not see the remark in book VIII as a threat to the idea that reason’s rule guarantees justice, we need to grant that reason is not the ultimate source of normativity. What makes reason a special ingredient in the soul’s hierarchy is its link with the truth and with the good. On the one hand, this means that reason is motivated by the truth and by the good, but it also entails that reason has some truths within it. This supposition is visible, for example, behind the technique of refutation: if someone is holding false conceptions on the discussed topic, a contradiction can be inferred, since that person also must have at least some true conceptions about it as well. Therefore, reason’s normative role in the structure is derived from its connection with truth and goodness. This also entails that reason is never, unless violated in an unjust soul, a merely instrumental capacity: its intrinsic function is to consider the ends one sets for oneself, only secondarily is it concerned with the means of attaining these ends. It follows from this general ethos of the discussion that the threat of circularity is not a serious one for Plato. The threat, as we saw, comes about if value-neutral criteria of agency are assumed to be the primary source of normativity. In the framework of moral objectivism – if this is the right way to describe the assumption that there is a form of good that can be apprehended intellectually and that is the source of order and structure in the whole nature – there is no need to reduce normativity to the criteria of agency. Rather, it makes perfect sense that the source of normativity, also for the criteria of agency, lies in what goodness and justice are. Being good involves being truthful, ordered and structured. Being better, then, means being more truthful, more ordered and structured in a more unified way, and this is basically what we find behind the criteria of agency. The Stoics might at first sight seem to take the normative standards of agency as increased or intensified descriptive criteria. As we saw, according to the Stoics, all agents must be rational and have the capacity to assent to impressions (and dissent); the best agents are the wise ones, who are most rational of all. One way of formulating the difference between ordinary rationality and ideal rationality is to say that the rationality that belongs to all can be characterised as functional, whereas the wise person’s rationality is normative. The most important consequence of this formulation is to stress the fact that all mental functions of adult human beings are reason’s functions. However, reason is never purely
74
M. Tuominen
functional for the Stoics. As mentioned above, from a Stoic point of view, acquiring rationality necessarily means acquiring some basic truths in the soul in the form of so-called preconceptions (prol¯epseis). Therefore, the functional rationality is not completely different in kind from the normative one. The crucial difference between the ancient and the contemporary discussions about agency is related to philosophers’ attitudes towards objective moral standards. In the ancient context where goodness and truth (together with beauty and unity perhaps) go hand in hand, there is no need to locate the source of normativity in value-neutral criteria of agency. Even for the Stoics, who do not accept the existence of the form of the good, it makes perfect sense to say that there are truths about what goodness is, and that these truths are embedded in human reason – even though we often apply them mistakenly. Helsinki Collegium for Advanced Studies, University of Helsinki, Finland
Socratic Authority Raphael Woolf
We are more able to observe our neighbours than ourselves Aristotle, Nichomachean Ethics 1169b33-4
Four Theses about Socrates If anything is clear in the interpretation of Plato’s Socrates, it is that he lays claim to a certain epistemic superiority over others, despite his famous disavowal of knowledge (or wisdom). His superiority turns out to consist precisely in the fact that he is aware that he lacks knowledge, whereas others mistakenly think they are wise. What is more, Socrates can show that others are not wise, notwithstanding their own view of themselves. So it is natural to suppose that Socrates’ epistemic superiority can be explicated (at least in part) in terms of third-person authority. Socrates, as the inventor of the elenchus, and in virtue of his capacity to wield that formidable weapon for the critical testing of beliefs, is better able to judge the mind of another than the subject is himself, revealing intellectual confusion in place of an imagined wisdom.1 If this captures something of Socrates’ distinctiveness (albeit in a rough and ready way), then the succession of elenctic defeats that Plato portrays him inflicting on a variety of interlocutors seems merely to confirm the point. But I want to argue that the appearance is deceptive, and to demonstrate that Socrates’ epistemic edge is confined at most to a certain first-person authority. The position may be set out as follows: 1. Socrates is able correctly to identify his own state of mind: he is ignorant, at least to the extent of lacking knowledge, and is (veridically) aware that he is.2
1 The nature and scope of elenchus have been much discussed, and for the moment I lay out just the bare bones of a standard case: the interlocutor makes an assertion (typically, but not invariably, an attempted definition of a virtue or some substantive moral thesis); Socrates then elicits from the interlocutor other assertions which he shows to be inconsistent with the original one. 2 I take Socrates’ disavowals of knowledge as sincere, though he is sometimes challenged on the point by interlocutors (most trenchantly perhaps by Thrasymachus: see Republic I, 337a–e). It is not relevant for my purposes whether his interlocutors acknowledge any special authority on Socrates’ part (it is clear in many cases that they do not). My interest, rather, is in the basis on which such authority as his claims imply might rightfully accrue to him: a normative matter.
P. Remes, J. Sihvola (eds.) Ancient Philosophy of the Self, © Springer Science+Business Media B.V. 2008
77
78
R. Woolf
2. No one else is able correctly to identify their own state of mind. So Socrates is unique in this regard. To the extent that others have a view about their own state it is false, since they take themselves to have knowledge of matters on which they are ignorant (particularly in the moral sphere). 3. Socrates is able correctly to identify the state of mind of others: that is, in particular, to show (using elenchus) that they are ignorant in matters of which they take themselves to have knowledge. 4. Socrates is not the only one able correctly to identify the state of mind of others. So he is not unique in this regard. He has plenty of imitators who can do this too. If this account is right (I shall proceed to elaborate and justify it below) there is a certain asymmetry at work. With the exception of Socrates, there is a dearth of firstperson authority, insofar as the claims of others to have knowledge are defeasible when subject to examination. By the same token, third-person authority is relatively commonplace. Anyone who can competently wield the elenchus has the authority in principle to over-ride the claims of others that they have knowledge. But one wielder of elenchus – Socrates himself – makes first-person claims that, as we shall see, are resistant to challenge even by the highest of outside authorities, namely god. If Descartes found that “I am thinking” was a first-person claim that not even the machinations of an evil demon could call into doubt, Plato’s Socrates can state “I lack wisdom” with an authority that faces down the all-seeing Oracle, who declares that no one is wiser than Socrates. Indeed, the two encounters have a certain affinity, since both involve the establishment of the subject’s authority for selfascription of a certain kind by showing how the truth of what is so ascribed is resistant to challenge from a maximally powerful opponent. The respective sources of such authority are, however, quite different, and this is in line with the differing content of what is ascribed in either case. Thus it is far from obvious that one can confirm that one lacks wisdom simply by the act of thinking that one does – or, for that matter, by doubting it.3 Rather, it is natural to suppose that Socrates’ authority in this regard arises to the extent that he can (and does) perform elenchus on himself, whereas others do not (and perhaps cannot) perform it on themselves. Socrates alone is in the business of critical self-examination, and both the content and the authority of his judgement that he lacks wisdom is its result. Any account that would grapple with Socratic self-examination must deal with at least one peculiarity, which seems to me to have been rather underplayed: we are never allowed to see Socrates explicitly engaging in this process, though his
3 True, Socrates holds that only god possesses genuine wisdom (see Apology 23a5–6), so that it might be a proof of one’s lack of wisdom to so much as doubt that, as a human, one lacks wisdom. But the proof (such as it is) does not, as in the Cartesian case, rest on the indubitability of the cognitive act itself, but on the background assumption about humans (in turn confirmed by the outcome of elenchus). Of course the one thing Socrates does not do is doubt that he lacks divine wisdom. It is others who make this kind of mistake about themselves.
Socratic Authority
79
examinations of others are described in lavish detail in the Platonic corpus.4 I shall argue that this contrast (if it can be established) underlies and sharpens the initial asymmetry that I sketched above. Plato’s instincts here are sound, if disturbing. Socratic self-examination is a rare bird not (or not just) because Socrates is special, but because it is itself a highly problematic notion, so much so that the title even of Socrates to a distinctively first-person authority, while apparently genuine, must be called into question. Plato does, as we shall see, implicitly but unmistakably call it into question, and (so I shall argue) with good reason. At one level, then, this enquiry into the nature of Socrates’ epistemic authority will be aporetic, illustrating a way in which one might capture the sense, gleaned by many readers of Plato, of Socrates as a figure compelling yet elusive.5 More positively, it will attempt to describe a striking complex of views about knowledge, authority and the mind within which Socrates conducts his operations. The position that will emerge casts a sceptical eye on the notion of first-person authority itself, and gives corresponding weight to its third-person counterpart. But in contrast to contemporary scepticism about the first person, Plato’s doubts arise in a context in which first-person authority is already regarded as a token of intellectual (and perhaps moral) achievement; and not something which, if it exists at all, will be the property of just any regular human subject.6 Since the bedrock of my argument is the claim of asymmetry encapsulated in the four theses listed above, I shall seek to demonstrate, in section II below, that they are to be found in Plato’s text. In section III the focus will be on self-examination, its character and the puzzling absence of examples from Plato’s work. Sections IV and V develop my positive account of how Plato thinks about epistemic authority, in part by contrast with a broadly Cartesian conception. Section VI returns to the puzzle, and seeks to clarify if not resolve it by arguing that Plato’s view is one that, in effect, disallows a first-person perspective. 4 To anticipate later discussion of two possible objections: it is not, I think, correct to assume that Socrates examines himself just in virtue of his examining others. And the case of the Hippias Major, which does appear to show Socratic self-examination, is not a straightforward one. 5 This echoes what Gregory Vlastos rightly saw as the importance of “keep[ing] faith with Socrates’ strangeness” in Socrates, Ironist and Moral Philosopher, Ithaca, NY: Cornell University Press 1991, 3. Yet Vlastos’s work on the Platonic Socrates, as much as on Plato in general, evinces the entirely creditable aim of raising puzzles in order to, in his word, “crack” them. Any strangeness that lingers is in spite of the vigour with which he presses his solutions. The torch of faith is borne in a rather different way by Alexander Nehamas, The Art of Living, Berkeley, CA: University of California Press 1998. His Socrates has features which remain mysterious right the way down, though I think that Nehamas fails in the end to characterise the mystery correctly (see n. 63 below). 6 I do not mean to suggest that contemporary analysts of the concept of the first person are insensitive to its normative character – and much depends on what one considers a “regular human subject”. Burge, for example, has argued that first-person authority is partly constitutive of our ability to reason critically about our beliefs and other attitudes. See Tyler Burge, “Our Entitlement to Self-Knowledge”, Proceedings of the Aristotelian Society 96(1996), 91–116. Plato takes a certain kind of failure of rationality (prevalent amongst humans in his view), that of not tracking the logical relations that hold between one’s own beliefs, to undercut a subject’s authority with regard to those beliefs. See further Section V below.
80
R. Woolf
Epistemic Authority in the Apology First, then, a defence of my four theses, evidence for which comes chiefly from the Apology, where Socrates offers the Athenian jury a sustained account of his method and mission. Thesis (1) is fairly straightforward to establish. The following two statements encapsulate it (Socrates is the speaker): “I am aware that I am wise in no matter great or small.” (Apology 21b4–5) “I was aware that I knew virtually nothing” (22c9–d1)
The confidence with which Socrates makes these claims is striking. That they are intended as veridical is indicated by his use of “I am aware”, synoida (cognate with oida, “I know”). And they are non-negotiable even when apparently challenged by god. For the Oracle at Delphi has just been reported as saying that no one is wiser than Socrates (21a6–7), which Socrates takes to mean that he (Socrates) is the wisest man (21b5–6). Not even this is enough to cause him to doubt the truth of his claims about his ignorance.7 In fact, if he doubts anything, it is the very word of god: he is going to “put the prophecy to the test and show the Oracle” that there are others wiser than he (21c1–2). Socrates actually says rather delicately at 21b9–c1 that “here [i.e. with the well-reputed politician he first questions] if anywhere I will put the prophecy to the test […]”. No doubt he is being as respectful as he can be of the Oracle’s divine provenance; and his aim indeed is that it should turn out unrefuted (22a7–8). But he goes on to imagine himself telling the god: “This man is wiser than me, but you said I was [wiser]” (c2). And when one notes his use of the loaded elenx¯en for “put to the test”, it seems hard to doubt that Socrates envisages the possibility, at least in principle, of the Oracle’s refutation.8
7
For this reason Stokes seems to me mistaken in arguing that 21b4–5 might be read, in effect, negatively (as the Greek would perhaps allow), as the claim that Socrates has nothing he is aware of knowing (or being wise in). See Michael Stokes, Plato; Apology, Warminster: Aris & Philips 1997, 18 and 117. Mere lack of awareness of wisdom would not explain the tenacity of his disavowal in the face of the Oracle’s statement, as awareness of a lack of wisdom does, see. For a parallel criticism of Vlastos on the point, see James Doyle, “Socrates and the Oracle”, Ancient Philosophy 24(2004), 19–36, at 31 n.13. The passage 22c9–d1 is in any event unambiguous. 8 As is pointed out by Nehamas 1998, 249. Commentators who deny this must set aside the literal sense of the text, as Mark McPherran, one of the deniers, is prepared to admit in his The Religion of Socrates, University Park, PA: Pennsylvania State University Press 1996, 224. The point is sometimes obscured when Socrates is made to claim at 21b6 that the god cannot be wrong (see among others McPherran himself, loc. cit.; Stokes 1997, 49; and C.D.C. Reeve, Socrates in the Apology, Indianapolis, IN: Hackett 1989, 23). Even Nehamas translates modally as “can’t be lying” (loc. cit.), though of course this is compatible with the god’s being wrong. Doyle asserts flatly that “Socrates […] accepts the conventional belief that the oracle […] cannot lie or be mistaken” (2004, 20); and he goes on to endorse what he calls the “exertion” required to square this with Socrates’ apparent preparedness to refute it (23 n. 7). But the text at b6 simply says “Presumably he [the god] isn’t wrong” (ou […] d¯epou pseudetai ge), which seems perfectly phrased to leave a sliver of doubt. See also n. 9 below.
Socratic Authority
81
Fortunately, of course, nothing eventually has to give. Socrates is wiser than others, but just insofar as he is aware of his own (first order) ignorance or unwisdom (23b2–4), whereas others take themselves to be knowledgeable in areas where they are not (e.g. 21d2–8, 22d5–e1). It is important to recognise, then, that without being mistaken itself (that would not be “lawful” (themis) for a god, 21b6–7),9 the Oracle turns out to confirm and vindicate Socrates’ disavowals. In its affirmation of his superior wisdom Delphi has at the same time certified their truth. One may comfortably infer that if these can withstand challenge even when under pressure from god, then they are hardly defeasible by humans.10 At the dramatic level Socrates’ encounter with the Oracle shows an uncanny, and not uncharacteristic, mix of humility and arrogance. Philosophically (and perhaps underlying the drama) it suggests that the claims he makes about himself have a special authority (whose nature remains to be investigated). The defence of thesis (2) is more complex. Socrates reports (21b–22e) that he goes to politicians, poets and craftsmen, and finds time and again when he examines them that they falsely believe themselves to have a knowledge that they do not in fact possess. He reports no case of discovering someone who has a correct view of his own state of mind. There is an intriguing, though rather obscure, section which may, however, suggest a different emphasis: As I made my enquiries in accordance with god, those who had the highest reputation seemed to me to be pretty much the most deficient, whereas others who were considered less grand seemed to be more respectable, as far as possessing moral wisdom is concerned. (Apology 22a3–6)11
On the face of it, this seems to imply that not all those whom Socrates encountered had an exaggerated view of themselves. Some of those he tested not only seemed
9
I take Socrates’ claim here to be normative, whether one renders pseudetai at b6 as “is mistaken” or “is lying” (either translation is possible). Socrates does not say the god cannot be wrong, and as we have just seen, is prepared to call his veracity (or exactitude) into question. A thing’s not being themis means not that it is impossible, but that it is impermissible, though Socrates is happy to draw the modal inference in a different context at 30c10–d2. Again, the text is nuanced so as almost (but not quite) to rule out divine error. 10 It may seem odd that a claim of ignorance should need vindicating in this way. But one of Socrates’ main motivations in context (remember he is on trial for, among other things, “corrupting the young”) is to rebut the common opinion that he has first-order wisdom, be it scientific or moral (18b7–8, 23a3–5), which he might therefore teach to others. No dangerous intellectual he – at least on those grounds. 11 I perhaps over-translate to phronim¯os echein as “possessing moral wisdom”, but do so to capture the fact that possessing phron¯esis, or being phronimos, is to have both an intellectual quality and, in the Apology (as of course elsewhere), one with a specifically moral connotation (cf. its use in the context of Socratic exhortations to virtue at 29e1 and 36c7). The phrase cannot in context refer to a state of awareness of one’s ignorance such as that possessed by Socrates, as claimed by Shigeru Yonezawa, “Socratic Knowledge and Socratic Virtue”, Ancient Philosophy 15(1995), 349–358 at 356–357. The reputation that the politicians fail to live up to is not (of course) for knowing their own mental state but for having some substantive moral knowledge in virtue of which they might be entitled to their position.
82
R. Woolf
to know something (21e5–22a1) but actually did know something, and specifically something in the moral realm, in contrast, say, with the craftsmen, who have genuine knowledge of their craft which misleads them into thinking they have moral knowledge too (22d–e). One cannot infer from this, however, that these good people correctly took themselves to have the amount of knowledge that they did in fact have. The passage says nothing about their view of their own mental state, nor suggests that they had any view at all. Rather, it notes, in their case, a smaller (or perhaps inverted) gap between what others think of them (their reputation) and their actual state. So although the passage remains suggestive,12 it cannot be used as evidence against the thesis that anyone other than Socrates has a correct view of his own cognitive state. The following passage is also relevant to the question of whether anyone, other than Socrates, has the requisite self-awareness: [The god is] using me as an example, just as if he were to say, “This man is wisest amongst you, gentlemen, who, like Socrates, has recognised that in truth he is worth nothing with regard to wisdom.” (Apology 23b1–4)
These lines set out god’s criterion (in Socrates’ view) for the “human wisdom” (20d8) that Socrates takes himself to possess, namely the recognition of one’s lack of “divine” wisdom (20d9–e1, 23a5–6) that would be constituted by moral knowledge.13 The text does not rule out that others could possess this human wisdom, and leaves open the question of whether anyone else does possess it. None the less, the indications are that we are to regard Socrates as the only possessor. For the Apology implies that no one other than Socrates possesses self-awareness even after elenchus. Socrates says at 21c8 that he tried to show the first politician he encountered that he (the politician) was not wise. At 23b6–7 he says more generally that when someone seems to him not to be wise, he serves the god by showing that he (the interlocutor) is not wise. Notice: he does not in this latter case say that he shows the interlocutor that he is not wise, just that he shows that the interlocutor is not wise.14 The difference is crucial. Whereas to have shown the interlocutor his lack of 12 It is intriguing not least because these unknown persons would, to the extent that they have some phron¯esis, even if it is only relative to their lowly reputation, seem to be Socrates’ most serious rivals in the contest for wisdom, superior to the craftsmen whose (mere) craft-knowledge is insufficient to counter-balance their resulting lack of self-knowledge (22e1–6). The unknowns cannot (pace Emile de Strycker and Simon Slings, Plato’s Apology of Socrates, Leiden: E. J. Brill 1994, 278) be the craftsmen (nor phron¯esis refer to their skills; cf. n. 11 above). Their error (pl¯emmeleia, 22d8, a strong word) Socrates criticises in a tone widely at variance with the one he adopts towards the unknown persons. 13 And not, as Vlastos had it, the possession of some less than infallible first-order moral knowledge (Vlastos’s “knowledgeE”) acquired through elenchus. See “Socrates’ Disavowal of Knowledge”, in Myles Burnyeat (ed.), Socratic Studies, Cambridge: Cambridge University Press, 1994, 39–66. The lines just quoted confirm what Socrates had already spelled out at 21d2–8, that his superiority in wisdom consists in his not thinking he has knowledge of what is “fine and good” (i.e. moral knowledge) when he does not. 14 A similar point is made by Doyle 2004, 28.
Socratic Authority
83
wisdom would suggest that the interlocutor has become aware of it, to have merely shown that the interlocutor lacks wisdom carries no such connotation.15 And when Socrates in the earlier passage does talk of showing the interlocutor, it is qualified by “I tried”. Socrates never says that he succeeds in rendering his interlocutors self-aware. Some other remarks of his indicate positively his failure in this regard.16 Just after he reports that he tried to show the politician his ignorance, Socrates says the following: I went away and reckoned to myself that I am wiser than this man. For neither of us happens to know anything fine or good, but he thinks he knows something when he does not, whereas I, just as I do not know, so I do not think that I do. (Apology 21d2–6)
Note the present tense: Socrates reckoned after his encounter with the politician that he (Socrates) is wiser because he (the politician) mistakenly thinks he knows. The encounter has made the politician (and onlookers) hostile to Socrates (d1–2), but has not generated self-awareness. The politician remains convinced that he knows, despite having been waylaid into aporia by Socratic elenchus.17 But does not Socrates’ celebrated comparison of himself to a gadfly (30e–31a) show otherwise? The gadfly wakes up the sluggish horse: and it might be thought that this corresponds in the Socratic case to his making the interlocutors (who of course are already literally awake) self-aware. The gadfly awakens its victim by stinging, so what the metaphor suggests is indeed a certain jolting awareness – that 15
Compare “I showed that Fermat’s last theorem is true”, which I might have done without anyone else becoming aware of its truth, with “I showed my colleagues that Fermat’s last theorem is true”, which suggests that they came to see that it was true too. 16 Let me add, though, a note of caution. I am resting my case with the Apology, since its format provides the closest thing we have to a general account of Socratic procedure. But one should not commit Plato in advance to a uniform portrayal of particular Socratic encounters. The courses and outcomes of elenchus are as subtly different as the range of interlocutors. Still, the Apology does purport to summarise Socrates’ experience; and one would in fact be hard pressed to find a single unequivocal example in the dialogues of an interlocutor brought to self-awareness by an encounter with Socrates. The closest case may be the famous experiment with Meno’s slave, but its heavily idealised conditions – an interlocutor with no axe to grind and an uncontentious subject-matter – only emphasise a more regular pattern of failure. A similar moral, though on different grounds, is to be drawn in the case of Theaetetus, who seems even to begin his encounter with Socrates without the false conceit of wisdom (Theaetetus 148e), but who in turn is intellectually and morally gifted to an exceptional degree. The slave episode is of course pitched directly against Meno’s complaint of being “numbed” by elenchus, a description that runs somewhat counter to the idea that, outside the realm of a controlled experiment, brushes with Socrates represented much of an intellectual awakening for the interlocutor. 17 A different story may appear to be suggested at 23d, where Socrates (in the context of elenchus practised by his followers, on which see further below) says that his accusers use the charges against him as a smokescreen: “For I think they would not be willing to speak the truth, that they manifestly turn out to claim to know, but to know nothing.” We do not, however, have to take it that these people realise that their claims have proven hollow. The words after “that” represent Socrates’ (no doubt correct) assessment of his accusers, but Socrates does not say that the truth is manifest to them. Being unwilling to speak the truth is consistent with doing anything not to recognise it. The accusers may be self-deceiving; but that hardly counts as elenctic success.
84
R. Woolf
of sharp and unexpected pain – which perfectly captures the experience of a typical encounter with Socrates. The interlocutor receives a sharp shock from the human gadfly: confident in his wisdom, he finds himself at a loss as to how to rebut his interrogator. None of this entails (unfortunately) that the interlocutor becomes aware that he is ignorant. Being made to look a fool is a self-consciously painful experience, sufficient in itself to explain the interlocutor’s hostility; but it need not and (to judge from Socrates’ reckoning) typically does not inspire the reflection that one is as one looks. What, though, of the following bold claim by Socrates? He [the Olympic victor] makes you seem happy, but I make you be happy […]. (Apology 36d9–10)
This sentiment seems to pull in a somewhat different direction from the gadfly simile, though of course there is no reason to deny that a measure of discomfort might be a necessary precondition for the Athenians’ happiness (or that Socrates thought it so). On the other hand, Socrates also famously maintains that the unexamined life is not worth living (38a5–6). If (as I maintain) Socrates does not succeed in turning the Athenians to self-examination, how can he be making them happy? Socrates’ dire warnings that they are neglecting their souls (e.g. 29d–e) at any rate sit rather oddly with the notion that he has made them happy – though this is not, of course, exactly what he says. We must, I think, interpret Socrates here as sloganising (I do not use the term in a pejorative sense). “Milk makes you strong” need not imply that one’s target audience has taken so much as a drop of the stuff. The Athenians’ problem is that they are unable or unwilling to take the Socratic medicine: to adopt a receptive attitude towards elenchus, realise that their souls are in bad shape, and start from there on to look after them.18 Socrates makes you happy – but only if you let him.19
18 Socrates does call himself a “benefactor” at 36d5, but one might say that milk is of great benefit too, without this implying that anyone in particular has taken a draught. At 36c3–7 he is more nuanced, claiming only to go in the direction of (epi, c3) effecting the greatest benefit, in endeavouring (epicheir¯on, c5) to persuade the Athenians to take care of themselves. He deserves free meals in the Prytaneum more for the sort of person he is (ton toiouton andra, d7) than for his results. 19 Socrates’ claim at 30a6–7, that no greater good ever happened to the Athenians than his mission, may be read similarly. Compare “She was the greatest thing that ever happened to him, but he rejected her”. The gifts of a benefactor are not always taken. Socrates knows this better than most. Nehamas (1998, 182) is therefore right to point out that, as the Apology sees it, Socrates’ interlocutors do not become convinced of their ignorance, but wrong (or at least misleading) to argue (181) that there “Socrates’ primary object of care is his own self, his own soul, not the souls of others.” Nehamas continues, “That is not to say, of course, that Socrates disregards others, that he does not care for them. But to care for others is not the same as to devote oneself to them” (181– 182). This seems to me a rather strained distinction, and the apparently motivating thought, that having oneself as primary object of care would mean one is not devoted to others, is a non sequitur. In fact Socrates declares in the strongest possible terms that he is devoted to the Athenians (see e.g. 30a–b, 31a–b). Devotion to even a good cause is perfectly compatible with failure, though we may be reluctant to countenance this regrettable fact.
Socratic Authority
85
There is a passage which confirms that in one particular regard Socrates is indeed unique in his self-awareness: Fearing death, gentlemen, is nothing other than seeming to be wise while not being so. For it is to think one knows what one does not know. For no one knows whether death is not perhaps the greatest good for man, but they fear it as if well knowing it is the greatest of evils. But surely this is the most reproachable ignorance, to think one knows what one does not know. Perhaps it is here, gentlemen, that I am different from the majority of men, and if I am wiser than anyone it is in this regard, that when I do not know sufficiently about matters in Hades, I think that I do not know. (Apology 29a5–b6)
Here Socrates implies that he does not fear death (he is explicit on the point at 29b8–9). He does not know what death is like, and takes himself not to know, in contrast to those who fear death, which he diagnoses as a case of thinking one has a knowledge one does not possess. Who are the people who fear death? Certainly “the majority of men”, and an even more non-specific “they” which probably implies “everyone” (excepting himself), since it is co-ordinate with the “no one” who knows the true nature of death. Socrates later (35a–b) points out that even those of high repute tend (unlike himself) to throw dignity to the winds when facing a capital charge. It seems reasonable to infer that it is only a distaste for hyperbole which prevents Socrates from saying that his awareness of his own ignorance, at least as far as the matter of death is concerned, is unique. And it would, I think, be understating the case to conclude that the failure of others to attain self-awareness is an accident. In its relentless repetitiveness, the failure is presented as indicative of a general human incapacity, with Socrates a lone exception, and prompts the question (which I shall attempt to reinforce in section III below) whether there may not be something peculiar about what Socrates is apparently able to achieve. Let us move on to thesis (3). Socrates evidently takes himself to get the state of mind of others right when he examines them, and in so doing to countermand their false view of themselves. Some examples: Then I tried to show him [the politician] that he thought he was wise, but was not. (21c8–d1) At the same time I noticed that [the poets], on account of their poetry, thought they were the wisest of men in other matters too in which they were not. (Apology 22c4–6)
Again, Socrates’ confidence that he has it right is striking. His talk of “trying” to show the politician indicates not that he thinks that the politician might after all be wise (Socrates affirms that he is not at 21d2–6, quoted earlier), but that he may have failed to see that he is ignorant.20 Similarly the poets are not wise, though they thought that they were. Misled by their poetic ability, they claim wisdom in other respects too. But not even as poets do they rank as wise, since they cannot speak fluently about their own poems (22b3–c3). There is no suggestion that our author, Plato, fails to share Socrates’ confidence in his analysis. Socrates gets the poets 20
Compare “I tried to show my colleagues that Fermat’s last theorem is true”, which implies not that I doubt that it is, but that they may not have accepted it.
86
R. Woolf
and politicians right because that is what proper use of elenchus enables one to do. Likewise, we may infer from the way he writes the encounter with the Oracle that Plato has no doubt that Socrates gets his own mental state right too. But let me reiterate what I take to be the crucial difference: Socrates is shown getting others right on many occasions in the dialogues. He is not, however, shown explicitly practising elenchus on himself. Nor is Socrates the only one who practises elenchus on others. This brings us to thesis (4), and it finds clear support in the Apology: Often they [Socrates’ youthful followers] imitate me, and endeavour to examine others. And then, I suppose, they discover a great deal of resentment from people who think they know something, but know little or nothing. And then those who are examined by them get angry with me rather than with themselves, and say that Socrates is an absolute pest and corrupts the young […] for I think they would not be willing to speak the truth, that they manifestly turn out to claim to know, but to know nothing. (23c5–d9)
The followers of Socrates who practice the elenchus are not said to do the job any less stringently than Socrates himself.21 Socrates gives no reason to doubt that they expose others in their false pretensions to wisdom, arousing an anger against Socrates as supposed mentor that would (in Socrates’ view) be more profitably directed at themselves. What is not said even about the followers, however, is that they examine themselves. Nor are they said to be aware of their own ignorance, or to have any self-awareness at all. By contrast, on two occasions in the Apology Socrates tells us that he examines himself, as well as others (28e6, 38a5). So, with our four theses intact, I want now to look more closely at the question of what Socratic self-examination consists in.
Examination and Self-Examination It is tempting to suppose that in examining others Socrates thereby examines himself. On the two occasions in the Apology when he speaks of examining himself, he uses the phrase “examining myself and others” (and refers on the second occasion to the Athenians hearing him do this) which might suggest he is talking of a single process. But why should examining others entail that one is examining oneself? Testing the views of others for consistency (as Socrates does, “in my accustomed
21 Socrates later issues a warning (39c–d) that after his death people younger than he (presumably those same youthful followers), freed from Socrates’ restraining influence, will put the Athenians even more severely to the test (elenchontes, d1). In respect of elenchus on others, the main difference between Socrates and his young followers seems to be one of attitude. Socrates insists that he practises it to benefit the interlocutor (e.g. 30a–b, 36c), whereas his followers do so for their own enjoyment: at any rate their imitation is said to follow on from their enjoyment of hearing people being examined (23c4–6, cf. 33c2–4).
Socratic Authority
87
manner” (27b2), with his accuser Meletus) need not show anything about one’s own views. There are, certainly, occasions when Socrates’ own views do seem to be on the table, and contraposed to those of the interlocutor.22 Thus Meletus claims that Socrates does not believe in gods (26c8), which Socrates vehemently denies (23c–d, 35d). Socrates proceeds to argue (26e–28a) that Meletus’s own claim that Socrates believes in “divine things” (daimonia) entails that Socrates does believe in gods. Now Socrates surely concurs that he believes in daimonia (given his talk of his own divine aide, his daimonion, at 31c–d, 40a).23 So if the entailment goes through, one might conclude that Socrates’ denial of atheism has been confirmed and that his belief structure has been tested and found, in that regard, to be sound. This cannot be the whole story, however. For there seems to be nothing in the structure of (inter-personal) elenchus that requires the questioner’s views to be on the table in any given case; nor, I think, do our texts suggest otherwise. Even when they are, this will be a matter of contingency, since, in order for the interlocutor to be properly tested, the questioner must rely on premises that the interlocutor himself accepts, which constrains the existence of any particular relation, be it agreement or contradiction, between the views of the questioner and those of the interlocutor. The questioner cannot simply “plant” a set of views with the interlocutor which (for example) directly contradict his own, such that the refutation of one set would entail the confirmation of the other.24 Moreover (and as we saw in the above example), when Socrates’ views are on the table, the elenchus that he performs on others serves if anything to confirm them. It is not generally presented as a vehicle by which Socrates discovers that he
22 The Gorgias is perhaps the most extensive example of this, in particular Socrates’ debates with Polus and Callicles (the latter explicitly billed as a touchstone for Socrates’ views at 486d–487e). And the Protagoras confirms at its close (361a–c) that a similar dynamic has been operating with regard to the discussion between Socrates and Protagoras. 23 Detailed consideration of the daimonion, Socrates’ divine voice or sign, is beyond the scope of this paper, though it clearly played a major role in his life, preventing him often, as he tells us himself (Apology 40a4–7), from doing the wrong thing. Given that they are both voices of the divine, Socrates’ attitude towards the daimonion and the Oracle ought to be similar. Thus the daimonion (albeit in the form of negative injunctions) is the source of perhaps surprising information for Socrates (most notably in its view that he had better avoid a political career, 31c–d), for which Socrates himself seeks to provide rational confirmation (e.g. 31d–33a; see here McPherran 1996, 187–189). Socrates does not tell us that he ever challenged the daimonion as he does the Oracle, but he is evidently just as unsatisfied simply to take its word and leave things at that. 24 Crito (48b–49e) provides a case of Socrates explicitly putting up for review a set of moral principles that have previously been agreed by Crito and himself. In this way Socrates offers the possibility of his own views being tested in dialectical debate (cf. 46b–c on his adhering to whichever argument seems best to him on reflection). In practice debate never gets going. The passage has Crito unquestioningly assent to Socrates’ recapitulation.
88
R. Woolf
has mistaken views.25 Doubtless, such views as are confirmed do not amount to knowledge (the confirmations are presumably too piecemeal); but still, one cannot without considerable distortion look to elenchus on others as the grounds for Socrates’ disavowal – as if one were to say: I can tell that I lack knowledge because my views keep being confirmed!26 The closest Plato comes to showing Socrates’ own views overturned are some episodes in the Hippias Major, but these (albeit in problematic fashion, as I discuss below) are precisely episodes of Socrates in dialogue with himself.27
25 The examination performed on a youthful Socrates by Parmenides in the eponymous dialogue is another matter (perhaps a referent of Socrates’ rather mysterious allusion to his multiple defeats “by a Heracles or Theseus” at Theaetetus 169b, cf. 183e–184a). More like an exception in a standard case is Laches 194c–d, where Nicias introduces what Socrates admits is a view of his own to the effect that one is good in those things one is wise in, which then forms the basis for a discussion of courage as a kind of wisdom or knowledge. (I owe this observation to Sara Rappe, “Socrates and Self-Knowledge”, Apeiron 28(1995), 1–24 at 14; cf. also Vlastos 1991, 113 n. 28.) However, the more specific proposal actually discussed and refuted (courage is knowledge of the fearful and hopeful) is presented as Nicias’s own (194e11–195a1), and represents at least a development of the Socratic precept, in particular by its specification of what the appropriate object of knowledge is (194d–195a). It therefore seems to me questionable that one should regard this proposal too as Socrates’ own here, his argument for a similar formula in the Protagoras (359b–360e) notwithstanding. Contrast Paul Woodruff, who also claims, without specific citation, a Socratic provenance for the proposals of Critias in the Charmides. See “Expert Knowledge in the Apology and Laches: What a General Needs to Know”, Proceedings of the Boston Area Colloquium in Ancient Philosophy 3(1987), 79–115 at 104–105. Rappe (1995, 9 and 14) cites in this regard 165b and 166c, which seem to me to show no such thing. Some commentators maintain, oddly, both that Socrates’ examination of Critias is a species of self-examination and that Critias represents a distorted view of Socratic thinking (e.g. Harold Tarrant, “Naming Socratic Interrogation in the Charmides”, in T.M. Robinson and Luc Brisson (eds.), Plato: Euthydemus, Lysis, Charmides, Sankt Augustin: Academia Verlag 2000, 251–258 at 258; Walter Schmid, Plato’s Charmides and the Socratic Ideal of Rationality, Albany, NY: State University of New York Press 1998, 50). 26 In the Hippias Minor Socrates confesses to being confused about the topic at hand (376b), as befits one who disavows knowledge (372d7–e1). But it is clear that his being in this condition predates the encounter with Hippias (376c2–3); what is discovered, rather, is that Hippias is in no better state (c4–6). A similar picture is suggested by those passages which show the interlocutor “catching” aporia from an already bewildered Socrates (Charmides 169c, Meno 80c and perhaps Euthyphro 11d). 27 In turn, the closest analogues to the Hippias Major are (as Vlastos recognised) probably the Euthydemus and the Lysis. The latter does not formally present a case of self-examination. But, as Vlastos puts it, “Socrates proposes all the theses which are discussed and refutes all the theses which are refuted” (see in Vlastos, “The Socratic Elenchus: Method Is All”, in Myles Burnyeat (ed.), Socratic Studies, Cambridge: Cambridge University Press, 1994, 1–37 at 31). No link here between genuine elenchus on others and the grounds of Socrates’ disavowal. With the Euthydemus, Vlastos does not quite get to the heart of the matter. Cleinias is indeed a largely docile interlocutor whose rather substantive, and critical, contributions at 290b–d cause surprise. But more specifically, Crito hints to a bashful Socrates that these must have been Socrates’ contributions after all (290e–291a) – particularly, we might add, as Cleinias’s remarks on mathematics and dialectic seem, albeit aporetically, either to anticipate or reflect on the Republic. This suggests an even more remarkable version of the Hippias Major scenario, Socrates in effect disguising his own selfcriticism as an exchange between a named (not just anonymous) other and himself.
Socratic Authority
89
Let us, then, look beyond Socrates’ examinations of others for the key to Socratic self-examination. If it did not originate with the Oracle, elenchus on others acquires its shape and direction in response to it: so much is clear from the text of the Apology.28 But even at the time of the Oracle Socrates is precociously selfaware. How should one explain this? As we saw earlier, he has no inclination to question the veridicality of his self-awareness even when it is under pressure from the Oracle’s pronouncement. Indeed his first inclination is to question the pronouncement itself. A prior life of rigorous self-examination would be the obvious way of justifying the confident humility with which Socrates disavows knowledge despite the Oracle’s apparent demurral.29 This will also explain how he is able to practise elenchus on others with such dexterity once the call of the Oracle is heard. (His youthful followers, by contrast, will have someone to imitate – namely him.) Prior examination of others will have been patchy, if we are to give the Oracle its allotted role; but nothing rules out a life to this point as his own harshest critic, and this would better account for his remarkable facility as a critic of others. Support for this idea may be found, appropriately enough, in the intellectual autobiography that Socrates recounts in the Phaedo, telling of his early brushes with natural science. Here we are offered a succinct summary of how he came to undergo a process of critical self-examination. He tells us how he at first “both seemed to himself and to others” to have knowledge (96c4–5) – phraseology that closely echoes the language used in the Apology to describe the initial state of Socrates’ first elenctic interviewee after the Oracle (21c6–7) – but that as his investigation proceeded he ‘unlearned’ even what he had thought he knew (96c6–7). No one interrogated him to reveal that his pretensions to knowledge were misplaced. The challenge to his supposed wisdom was provided by himself.30 Further indications of Socratic self-examination taking place independently of inter-personal elenchus are to be found outside the Apology. The Symposium reports two occasions on which Socrates stood alone for lengthy periods contemplating some problem (175a–d, 220c–d) – the first report tells us that this was a 28 Commentators have often pointed out that Socrates does not say that he did not practise elenchus on others before the Oracle. On the other hand, the text conveys the unmistakable impression that the Oracle marks a sea change in Socrates’ life. As several passages make clear (e.g. 21e–22a, 23b, 29e–30a), systematic adversarial elenchus – the examination of anyone with some pretension to goodness or wisdom – begins here. The Laches (187e–188a) suggests that Socrates was examining others even as a young man, but that this was limited to “whoever is very close to Socrates and approaches for dialogue” (187e6–7). The contrast with Socrates, post-Oracle, actively seeking out subjects for testing, is clear. 29 Along these lines see McPherran 1996, 216; though it is unclear whether McPherran treats this, as I think one must, as an activity independent of elenchus on others. Reeve is also unclear on the point: Socrates already knew by the time of the Oracle that he lacked wisdom “by means of the elenchus […] for elenctic examination is always self-examination” (1989, 32). None of the texts Reeve cites seems to me to establish the claim that elenchus on others always involves self-examination, and one (Charmides 166c7–d2, discussed below) suggests the contrary. 30 One may note in this regard the pointed phrase “I do not accept from myself” (ouk apodechomai emautou) at 96e7–8. It remains the case that Socrates is shown reporting, not undergoing, his self-examination.
90
R. Woolf
habit of his (b1–2).31 And the Charmides has Socrates make the following remark to Critias: How can you think, however much I test (elench¯o) you, that I do so for any other reason than that for which I would thoroughly question myself as to what I say, fearing that unawares I might be thinking that I know something, when I do not know it. (166c7–d2)
Here Socrates contrasts the examination of Critias with the examination of himself. His reason for testing Critias is the same as the reason he would question himself. So in testing Critias he is not questioning himself, although, as we saw above, and as is confirmed here, results that bear upon Socrates’ own mental state may emerge. Indeed, Socrates (at least here) gives the prevention of mistaken self-attribution as his main reason even for testing others.32 He will certainly not take the pronouncements of others either as pieces of knowledge to be uncritically absorbed or as points of view unworthy of investigation. With this attitude, the examination of others, as much as that of oneself, may well have as its upshot revisions to one’s own set of beliefs: to the discovery, if you will, that one mistakenly thought one knew what one did not. But one can accept this and still maintain that to examine the beliefs of others is not ipso facto to examine one’s own. Our passage is quite clear in identifying Socrates’ questioning of himself as an activity independent of his questioning of others.33 Socratic self-examination neither originates nor consists in his examinations of others. And it is striking that, for all its importance in explaining Socrates’ encounter with the Oracle and its aftermath, we are never given an example of Socratic self-examination as such. The Symposium (perhaps) and the Charmides (more definitively) allude to Socratic self-questioning as carried on separately from his examination of others. And certainly, it would not be appropriate to expect an example within Socrates’ defence speech in the Apology. Admittedly, too, in his insistence in the Apology that he is a great benefactor to the Athenians, Socrates presents his mission as primarily other-directed. If a major part of Plato’s project is to portray this mission, then perhaps it is not so surprising that it is Socrates’ interaction with others that is chiefly described in those dialogues in which he is the leading character.34 Yet the power of the portrayal should not mislead us into 31
Cf. also the lengthy spell of self-reflection reported at Phaedo 95e8–9. Compare Cratylus 428d, where Socrates prefaces his interrogation of Cratylus’s position on the correctness of names with the admission that he may himself have been guilty thus far of selfdeception and the false conceit of wisdom. On the dangers of self-deception in a similar context see also Phaedo 91c4–5. 33 Charmides 163e6–7 is also suggestive in this regard. “Let us not yet examine my view but what you are now saying,” Socrates tells Critias with regard to the question whether one who does bad things could be temperate. Even if the “not yet” envisages a potential future examination of Socrates’ views, his remark evidently implies that one can (and perhaps should) examine the interlocutor’s view independently. 34 The short stretch of the Gorgias (506c–507c) in which Socrates, at least formally, carries on dialogue with himself when Callicles withdraws in frustration from the role of interlocutor, is clearly not intended as an example of self-examination, but as a setting-out of Socrates’ position. As he says himself in summary (507c8–9): “These are the claims I make, and I say that they are true.” 32
Socratic Authority
91
thinking there was anything inevitable about this. To judge by his exhortations at Apology 30a–b and 36c, Socrates’ use of elenchus in other-directed terms seems to have the objective of getting interlocutors to examine, and take care of, their own souls – and so is a mere prelude, as it were, to the real business. If elenchus on others is meant as a stimulus towards the goal of self-examination, then why just keep revisiting the stimulus? Should not one occasionally be shown, by way of obvious and edifying example, Socrates directly questioning and examining himself?35 At this point, a dialogue that I have referred to only briefly calls for mention, namely the Hippias Major. For here, surely, Socratic self-examination is a dominant theme. Socrates refers constantly to an anonymous “someone” (tis, first mentioned at 286c5) who turns out to be none other than “the son of Sophroniscus” (298b11) – i.e. Socrates himself. This someone asks Socrates awkward questions, and subjects his answers to withering critique. We are given illustrations of their interaction within the dialogue itself. Sometimes Socrates suggests an answer, and is refuted by the “someone” (as with the definition of “the fine” (to kalon) as pleasure through sight and hearing, 297e–300b). Sometimes the “someone” suggests an answer, which is refuted by Socrates in his own name (as with the definition of the fine as the fitting, 293d–294e). Here, then, we have Plato devoting a substantial portion of text to showing Socratic self-examination in action. I do not take issue with the dialogue’s authenticity (siding with the majority of scholars who deem it genuine). The problem, rather, is that it shows – goes out of its way to show – what is purportedly Socratic self-examination as a third person exercise. The exercise is presented throughout as a double-act between Socrates and another person, albeit an anonymous one, and even the eventual identification (nearly three-quarters of the way through) is rather coy: “the son of Sophroniscus” could, in theory, refer to someone other than Socrates. It is presumably not intended to, except that the coyness conforms with the dialogue’s general reluctance to identify Socrates’ partner in dialogue
35 Despite appearances, there is no dramatic reason for self-examination not to have been portrayed (see n. 58 below). Vlastos (1991) notes the lack and remarks that “the procedural form of elenctic argument” is what “prevents him [Socrates] from making any of his own doctrines the target of elenctic refutation by himself” (113 n. 28). But talk of “procedural form” looks like a fancy way of begging the question. Why that form? Why, given the aim of self-examination, should Socrates (or Plato on his behalf) feel restricted to serving up enquiry in the inter-personal mode? One might, with Christoper Gill, appeal to a notion of Socratic enquiry as “shared search” as part of a larger argument that Greek thinking (particularly ethical thinking) was of a fundamentally “participant” character. See Gill, Personality in Greek Epic, Tragedy and Philosophy, Oxford: Oxford University Press 1996, 282–283. And Socrates is certainly capable of presenting himself as engaging in dialogue in order to discover the truth (Gorgias 453b–c), a common benefit to all (Charmides 166d). But I remain unconvinced that the real work of self-examination is either displayed in, or owes any great debt to, Socrates’ encounters with others.
92
R. Woolf
here (the hapless Hippias aside) as himself. The double-act continues, even after the identification, to the end (see 304d–e).36 Two things now need explaining: the general reluctance of Plato to show us cases of Socratic self-examination. And the specific reluctance of the Hippias Major to show it as self-examination. I propose to link these two levels of reluctance and suggest that elenchus as a first-person exercise is found genuinely problematic by Plato. It is for good reason that he gets no further than showing it as (just another) third-person routine. This may in turn have some bearing on that venerable question: why did Plato write dialogues? Though it has been much discussed, the question can have a somewhat shopworn air – it is not as if there were a fixed template for writing or doing philosophy from which the dialogue form would represent a surprising or problematic departure. There is, however, a way of posing the question which does carry a flavour of genuine paradox: why did Plato write, of all things, dialogues with a leading character who advocates the virtue of examining oneself? The medium and the message seem out of synch. From this perspective, there is indeed reason to investigate why Plato sticks so resolutely to his chosen format.
Authority and Other Minds To pave the way for consideration of these difficulties, let me sum up the discussion thus far and throw it into sharper relief by considering in a little more detail the view of epistemic authority that underlies it. I have argued thus far for an asymmetry in Plato’s conception of authority. Thus third-person authority is in principle the possession of anyone: it just takes the requisite skills (which, to judge by the prevalence of Socratic imitation, are not so hard to pick up) to countermand the view of another about his own mental state. First-person authority, on the other hand, is depicted in the dialogues as very rare. Only Socrates, as we have seen, has the authority to make epistemic self-ascriptions that are immune from challenge.
36 A variety of explanations can be given for the presence of Socrates’ alter ego, a common one being that it enables Socrates to conduct a much more rigorous investigation of the question than would be possible via engagement in his own name with the preening, prickly Hippias – though as Tarrant has pointed out this rather begs the question: why Hippias? See Harold Tarrant, “The Hippias Major and Socratic Theories of Pleasure”, in Paul Vander Waerdt (ed.), The Socratic Movement, Ithaca, NY: Cornell University Press 1994, 107–126 at 110. Of course the blows would not be softened were the alter ego simply to have been identified as Socrates at the outset. What does the work in these accounts is the anonymity of the alter ego as fierce critic – in other words, the portrayal (without downright deceit) of Socrates’ interaction with him as a third-person exercise. Yet one might still wonder whether this procedure was a particularly effective way of dealing with Hippias, whose parting remarks are pretty dismissive (304a–b). Might not a straight admission from Socrates (with appropriate declarations of inexpertise) that he was engaging primarily in an act of self-criticism have done more to keep Hippias on board?
Socratic Authority
93
Such an asymmetry is roughly the inverse of that made familiar by the Cartesian tradition,37 and a brief investigation of this phenomenon may help elucidate the Platonic view.38 Within the Cartesian tradition the default position, so to speak, is first-person authority. Each subject has privileged access to his own mental states just in virtue of being their subject. By contrast, access (let alone authoritative access) to another’s mental states is problematic. If I can report authoritatively on my own just because, and only insofar as, they are mine, then by the same token I may have cause to be sceptical about those of others. I hear sounds and see movements apparently emanating from another human being, for example. But can these offer secure grounds for concluding that the being is in possession of certain, or indeed any, mental states? Perhaps the sounds and movements are just that, with no thinking thing behind them. (Perhaps there is no thinking thing in the world except me!) At any rate, my own mental states I can directly experience; those of others I must infer on the basis of evidence which perhaps can never be conclusive. Plato tells a different tale, not by being free from scepticism but by inverting the sceptical order. While carrying no brief for the idea that the view a subject has of his own mental states carries any special weight (self-ascriptions can by and large be overturned), he displays a robust lack of scepticism about the existence and interpretation of other minds. A good example of this outlook is found in the following passage from the Charmides: “It is clear [said Socrates] that if temperance is present to you, you are able to form some belief about it. For it is necessary, I suppose, that if it is in you, it produces some sensation, from which you would form a belief about what it is and what sort of thing it is. Don’t you think so?” “I do,” said Charmides. “And since you know how to speak Greek, you would be able to report what you think just in the way it appears to you?” “Probably,” he said.39 (Charmides 158e7–159a8)
37 I use the phrase “Cartesian tradition” here with due caution, since it is not my aim to provide exegesis of Descartes or any follower (though see n. 46 below). None the less, I think that what is described may be categorised in this way without too much strain; and that it captures, albeit very roughly, a familiar and influential way of thinking about epistemic authority. I do not wish to suggest that scepticism about the Cartesian viewpoint must lead to scepticism about a distinctively first person mode of access altogether. Though the former has sometimes resulted in the latter, it can equally motivate new ways of conceptualising such access – the task would be “disentangling the basic idea of first person access from the Cartesian picture” (see Richard Moran, Authority and Estrangement, Princeton, NJ: Princeton University Press 2001, 4). My purpose is to use the position outlined as a convenient foil for bringing out some of the more philosophically noteworthy aspects of the Platonic position. 38 The claim that the Cartesian viewpoint differs in fundamental respects from Platonic (and, in general, Greek) ways of thought is one made with increasing frequency. See e.g. Gill 1996 (esp. 6–10); Myles Burnyeat, “Idealism and Greek Philosophy: What Descartes Saw and Berkeley Missed”. Philosophical Review 91(1982), 3–40 . The contrast between dualism in the Phaedo and in Descartes has recently been stressed by Sarah Broadie, “Soul and Body in Plato and Descartes”. Proceedings of the Aristotelian Society 101(2001), 295–308. My analysis offers a complementary perspective to these approaches both in philosophical outlook and choice of text. 39 The word translated “probably” (is¯os) may also be rendered more weakly as “perhaps.” But any hesitancy that Charmides is expressing I take to represent not some general sceptical doubt about communication, but a specific concern (in line with his modesty, cf. 158c–d) about his own ability to report accurately what he thinks he has: a practical rather than theoretical difficulty.
94
R. Woolf
Here then is Charmides, who may or may not have a certain virtue (namely temperance: a mental state, something “in the soul”, 157a5–6). Whatever that state is, if present it will cause a certain sensation which will enable him to form a belief about its nature; and since he shares a common language with his audience, he can report this to them intelligibly. It is as simple as that. But, although not doubting that Charmides can report accurately to others what he thinks is in his soul, Socrates regards him as having no special access to what is actually there. For he clearly takes himself to have the authority to rebut Charmides’ (albeit tentative) claim that what he has discovered is the presence of temperance, reported in the first instance as “a kind of quietness” (159b5). He shows that temperance cannot be quietness (159b–160d), and Charmides is then asked to inspect himself again and see if this time he can correctly identify what he finds (160d–e).40 The second attempt fares no better than the first (160e–161b). What entitles Socrates to countermand the results of Charmides’ self-inspections? The short answer, of course, is dialectic41: the qualities that Charmides reports can be shown argumentatively not to be equivalent to temperance. But dialectic in turn owes its free run of the territory to some fairly explicit presuppositions about authority and other minds. In this world inspection of another’s mind is no more theoretically problematic than inspection of another’s body. Plato lacks the Cartesian notion of introspection,42 with its accompanying doctrine of privileged first person access. Thus Charmides is asked at 160d6 simply to “look at” himself (eis seauton emblepsas), Plato employing a phrase that has no special first person connotation, as the similar expressions at 155c8 (Charmides looks at Socrates: eneblepsen te moi) and 162b11 (Charmides looks at Critias: eis ton Kritian apeblepen) show. Nor, it seems to me, would Plato have any use for such a notion; the whole thrust of his position is that the individual has no special epistemic competence with regard to his or her own mental states. The prospective subject of inspection has only to make his mind available to the inspector (he has to be willing to engage in dialogue), as the subject of a physical inspection has to make his body available (he has to be willing to strip down, for example). The early pages of the Charmides include a discussion of the relation between soul and body (156a–157c) in which the legendary Thracian doctors are praised for being able to treat souls (with words) as a pre-condition for treating bodies (with medicaments). And when Charmides first arrives on the scene and Socrates, admiring his face, is told that if Charmides
40 Socrates is not of course necessarily claiming that Charmides has misidentified what is within himself as quietness, but that, whatever is producing the sensation he is picking out, he has misidentified as temperance. 41 I use the term in effect inter-changeably with “elenchus”, to pick out the process of critical questioning of beliefs exemplified by Socratic practice. 42 See Voula Tsouna, “Socrate et la connaissance de soi: quelques interpretations”, Philosophie Antique 1(2001), 37–64. Tsouna argues, from a somewhat different perspective than mine, for an “objective” conception of the self and self-knowledge in Plato.
Socratic Authority
95
were to undress he wouldn’t even notice his face so fine is his body (154d), he remarks that perhaps they should undress his soul and examine that first (154e). Similarly, at Protagoras 352a–b Socrates compares the exposure of Protagoras’ mind to the uncovering of a patient’s chest for examination by a doctor.43 These forms of description indicate that for Plato other minds are to be regarded as no more opaque to examination than other bodies. Both, of course, may be opaque. An interlocutor may decline to engage frankly in dialogue; a patient may keep himself demurely wrapped up. But minds present no special problem over bodies. Thus, with the requisite expertise, another may have more authority to judge what my mental states are than I do myself, as much as a doctor may have more authority to judge what my bodily states are than I do myself. This is the framework that allows the Socratic questioner his leading role. So the parallelism between soul and body as objects of examination should not be viewed simply as the innocent by-product of a pre-Cartesian way of thinking. It suggests, rather, a theoretical stance that provides vital underpinning for the Socratic mission.44 If the first half of the Charmides sketches a framework within which the dialectician has the authority to operate, the second half hints, within its aporetic mesh, at a theoretical outline of the dialectician’s role. Critias offers as a definition of temperance “knowing oneself” (164d3), which, with some assistance from Socrates, soon metamorphoses into “knowledge of knowledge and lack of knowledge” (166e5–9). No doubt Critias’ references to Delphi in his initial formulation (164d– 165a) are a Platonic cue for us to connect the definition with Socrates’ own activity. And of course the reformulation calls to mind Socrates’ prowess at discerning whether both he and others possess knowledge or not. Socrates’ subsequent refusal to allow the inference from “knowing that someone does (or does not) know” to “knowing what someone does (or does not) know” (169e–171c), although it raises problems of its own, none the less succeeds in marking off the special function of the Socratic dialectician. He is one who, without necessarily possessing first order expertise, understands the conditions for knowledge itself, and so (given a willing interlocutor) can test whether others meet them (170b–d). Those with ordinary first
43
The stripping analogy recurs at Theaetetus 162b, cf. 169a–b. For a striking though more remote case of the parallelism operating across dialogues, compare Socrates’ peek at Charmides’ pectorals (Charmides 155d3) with Alcibiades’ glimpse of the treasures in Socrates’ soul (Symposium 216e6–7). In both examples an ardent observer catches sight of “the things within” (ta entos). To judge from Alcibiades’ preceding remarks at 216d–e, Socrates’ cloak of irony may in general be less liable to slip than Charmides’ tunic; but that marks a difference between Charmides and Socrates, not a theoretical distinction between the accessibility of bodies and souls. I would add that Socrates’ disinclination to reveal himself, while quite consistent with the hypothesis that he cultivates his soul largely independently of his inter-action with others, does not suffice to explain why we never see an explicit case of Socratic self-examination (see further n. 58 below). 44 In using the term “theoretical” in this context I am not treating the Cartesian position as if it were available for Plato to react against. I do argue, however, that Plato is exploring the conditions of possibility for the practice of Socratic dialectic; and this is already to set out his stall on a range of issues which may not sit comfortably with Cartesian modes of thought.
96
R. Woolf
order expertise (such as doctors) will have knowledge of (say) health and sickness but not (as such) of knowledge (170e–171a), and so will be able to test for the former but not for the latter even within their own domain. Does Socrates backtrack at 171c when he asserts that any craftsman can identify fellow-craftsmen in his field? He has just said at 171b that the doctor can tell, in the domain of health and sickness, whether another’s statements are true and his actions correct. Socrates then denies at 171c4–9 that the putatively temperate man with knowledge of knowledge can tell who has medical knowledge (as opposed to just who has some knowledge or other), but without quite affirming that the doctor (who has knowledge of health but not of knowledge) can tell who has medical knowledge. He says less committally that the craftsman can identify his fellowcraftsman. What this amounts to has been stated, in the case of the doctor, at 171b. The doctor, just in virtue of his knowledge of sickness and health, can identify (say) whether another is describing and treating a disease correctly. But to do this is not to demonstrate a prerequisite for identifying knowledge.45 Socrates implies, with his talk at 172b4–5 of one who has knowledge of knowledge being able to “discern the knowledge over and above each individual thing he learns”, that one identifies knowledge by discerning a certain formal property of the subject’s belief-set, namely its structure. This constitutes the dialectician’s proprietary task. His role is underlined by a specific problem Socrates raises at the end of the dialogue (175b–c) for the notion of knowledge of one’s lack of knowledge. He asks how this could be possible without the absurd consequence that one know the very things one does not know. The putative absurdity perhaps rests on the thought that only one who has knowledge in a given area will have sufficient warrant to determine who lacks it. But the point can be met if one’s criterion for knowledge is a formal property (such as structure) whose absence can be determined without an assessment of how any of the subject’s particular beliefs stand in relation to external facts. At the same time, it should be noted that the problem applies, to the extent that it arises at all, only to the first person case, since there is nothing even prima facie absurd about knowing that someone else lacks knowledge. With respect to the identification of knowledge or its absence, then, far from privileging first person authority, the Charmides recognises it only to the extent that the subject possesses dialectical expertise which may be applied as well (or better) to others as to oneself. The non-expert subject must yield to the authority of the dialectician.
45
For a different view see Robert Ketchum, “Plato on the Uselessness of Epistemology”, Apeiron 24(1991), 81–98. Ketchum claims (94) that Socrates argues “that only one who can (a) determine whether or not one who speaks truly about health and acts correctly vis à vis sick and healthy people can (b) determine whether or not someone knows health.” (My lettering.) But the move from (a) to (b) – from, if you like, truth to knowledge – is one which Socrates stops short of making; and he is about to suggest his own model of knowledge at 172b, as we shall see. Ketchum’s provocative paper is vitiated to an extent by its failure to mention 172b, which at least attributes some use (cf. agathon, b1) to knowledge of knowledge.
Socratic Authority
97
Authority and Belief One should not, to be sure, press the contrast between Platonic and Cartesian viewpoints too hard. As we have seen, Socratic enquiry paradigmatically investigates an interlocutor’s claims to (first-order) knowledge; and perhaps even from a Cartesian perspective one need not deny that another may be a more authoritative judge of whether a subject possesses such knowledge than the subject is himself.46 (Whether a Cartesian should have any truck with Socrates’ claim that only the dialectician has this authority is not, I suppose, something his Cartesianism could help him decide.) None the less it would be an equal mistake to under-estimate the difference in outlook. The approach to first-person claims and to other minds exemplified by the Apology and Charmides suggests a conception of authority that is far from Cartesian in spirit.47 And this is strikingly borne out when we consider that Socrates’ investigations of a subject’s claims to knowledge characteristically turn on whether the subject’s beliefs are consistent or not.48 On pain of losing that title, a Cartesian should certainly bridle at the idea that another might be better able to determine what beliefs the subject possesses than the subject is himself. One might want to insist that the Cartesian will only allot a subject authority with respect to what the latter is occurrently thinking. But this will in fact leave Socrates and the Cartesian still facing squarely up to one another, given that Socrates tests an interlocutor’s beliefs precisely on the basis of what the interlocutor is moved to express in live dialogue.49 For the Cartesian, I take it, no one can be in an epistemically better position to declare that a subject believes, or does not believe, p than the subject himself. There will, on this view, be no gainsaying the sincere report of a competent subject that he does, or does not, have that belief.
46 It is, though, interesting to note that Descartes himself (Meditation III, 1) treats as components of his being a thinking thing (res cogitans), on a par with wishing, imagining and so on, his knowing a little (pauca intellegens) and being ignorant of much (multa ignorans). Descartes remarks that, as modes of thinking, he can be certain even of his sensings and imaginings, though their objects may be nothing real. Will his knowing or being ignorant, by implication, likewise (as modes of thinking) be things of which he can be introspectively certain and which are not subject to the countermand of others? If so, the opposition between Descartes and Socrates is both direct and radical. 47 The case of pleasure and pain, though not my main focus, deserves brief mention, since from a Cartesian point of view these are paradigmatically states to which the subject is thought to have authoritative access. When Socrates raises the issue of the subject’s relation to these states, it is to secure the interlocutor’s agreement that there are many cases, for example dreaming or madness, where a subject may mistakenly take himself to be experiencing pleasure or pain (Philebus 36e). Although Socrates then states that this assumption should be tested (e11), it is not overturned in the subsequent discussion of false pleasures. 48 As we have seen, Socrates warns the jury that his interrogation of Meletus will be “in my accustomed manner” (Apology 27b2), and he proceeds to expose Meletus in contradiction. 49 So with Charmides, the testing of his identification of the putative virtue within him proceeded on the basis of a belief reported by him, derived in turn from a certain sensation.
98
R. Woolf
Socrates begs to differ. In furtherance of his investigations, he is willing to claim authority with regard to a subject’s beliefs in just this sort of context. I want to demonstrate that this is so, and the grounds for his claim, by looking at one text in particular: the section of the Gorgias in which Socrates debates with Polus whether it is better to do injustice than to suffer it. This section has received its fair share of discussion, but its methodological implications have not, I think, been fully teased out. Doing so will enable us to see how the authority of the dialectician in the case of belief rests on much the same principles that we have already seen to be operative in the case of knowledge. The following passage is especially pertinent: SOC: I think that both I and you and everyone else holds that doing wrong is worse than being wronged, and that not paying the penalty [for wrongdoing] is worse than paying it. POL: And I think that neither I nor anyone else holds this. Would you prefer to be wronged than to do wrong? SOC: Yes, and so would you and everyone else. POL: Far from it: neither I nor you nor anyone else would. (Gorgias 474b2–10)
It is clear from this remarkable exchange that neither Socrates nor Polus have any problem about stating what another believes (or, for that matter, prefers). Nor indeed do they have a problem countermanding that other’s view about what he himself believes. Socrates feels entitled to claim that Polus and everyone else thinks that doing wrong is worse than being wronged, though it is hard to imagine everyone else agreeing that that is what they think, and Polus explicitly denies that that is what he thinks.50 By the same token, Polus can pronounce with confidence that no one holds that doing wrong is worse than suffering it, despite Socrates’ assertion that this is what he holds. What Polus finds outrageous is not the principle that Socrates should contradict the view of others about what they believe (for he does so himself) but the content of the belief that Socrates, in so doing, ascribes to others – that is all. What has no purchase at all in this exchange is the idea that the subject has any special authority to identify what he himself believes. Since both Socrates and Polus are in agreement on the matter, it is perhaps reasonable to infer that Plato shares the view. Asymmetry being an underlying theme, it is worth noting a certain asymmetry in the statements of Socrates and Polus here. What Socrates claims is the authority to attribute a belief to a subject which the subject does not take himself to hold (and would deny, if asked, that he does hold). Polus claims authority to deny a belief to 50 Vlastos found the claim so striking that, in typical problem-solving mode, he appealed to a distinction between “overt” and “covert” belief to account for it: “[Socrates] is telling them [Polus and the rest] that, along with their (overt) belief in p, they have certain other (overt) beliefs which entail not-p. In this sense they do (covertly) believe not-p” (1994, 23). In fact in this exchange Socrates takes Polus to hold a belief that the latter denies holding without Polus asserting his own (“overt”) belief. But should one none the less call the belief in question “covert” in Socrates’ assertion and “overt” in Polus’s denial? The text gives no sign of recognising such a distinction, and to read one in would rob the exchange of much of its adversarial force. Socrates affirms, and Polus denies, that the latter has a certain belief. We should not seek to reduce their clash to a kind of talking at cross-purposes. (SOC: (Covertly) you believe it. POL: (Overtly) I do not.)
Socratic Authority
99
a subject that a subject does take himself to hold. Polus’s authority claim seems the more ambitious.51 For it seems more plausible that one might have a certain belief which (say) is repressed or just escapes one’s notice, than that one should not have a certain belief which one takes oneself to have. It might even seem plausible that taking oneself to have a certain belief is a sufficient condition for having it, whereas one would be less inclined to say that taking oneself not to have a certain belief is sufficient for not having it.52 But it is Socrates who leads the discussion. So let us follow him. Socrates is able to show that Polus all along believes it is worse to do wrong than to be wronged by showing (474c–475e) that Polus believes propositions which entail that conclusion. Thus Socrates induces Polus to agree that it is more base to do wrong than to be wronged, and that what is more base is either more painful or worse. Doing wrong is evidently not more painful than being wronged. So, since it is more base, it must be worse. The final conclusion (475e3–5) is that Polus would not therefore prefer doing wrong over suffering it. The dialectic is in fact slightly complicated here, though I think harmlessly so for present purposes. Taking “A” to stand for “doing wrong”, “B” to stand for “being wronged” and “PO” to stand for “Polus”, Socrates makes two assertions at 474b: 1. PO believes that A is worse than B. 2. PO would prefer B to A. His final conclusion is formulated as: 3. PO would not prefer A to B. Given Polus’s assent to the interim conclusion that A is worse than B, Socrates is entitled, other things being equal, to the stronger (though admittedly clumsy):
51 One should leave open the possibility (cf. 471e1) that Polus (wrongly) takes Socrates to be dissembling – asserting something which he does not in fact take himself to hold. Polus’s denial would then be merely that Socrates was sincere. But I take it that Socrates’ long speech at 471e–472d is designed to convince Polus that he is in deadly earnest, and does so, for we find thereafter plenty from Polus about the outlandishness of Socrates’ position, but nothing more about him not meaning it. 52 We should in fact be cautious about taking Socrates’ project to involve not just attributing a certain belief to an interlocutor which the latter denies he holds but also denying that the interlocutor holds the contrary belief. So rightly, on the present passage, see Thomas Brickhouse and Nicholas Smith, Plato’s Socrates, Oxford: Oxford University Press 1994, 76. This would, for one thing, fail to recognise that for Socrates contradiction on the part of the interlocutor is a genuine (and undesirable) phenomenon (e.g. Gorgias 482b–c). (Brickhouse and Smith go on to muddy the waters somewhat by identifying the attributed belief as what the interlocutor “really believes” (1994, 79) ). Earlier in the Gorgias (466e) Polus had claimed that having great power is doing whatever seems best to one, to which Socrates replies (e4) “No; or so Polus says”; Polus ripostes (e5) “I say no? I say so!”; and Socrates responds (e6–7) “You certainly say no. For you say that great power is a good for the one who has it.” Here Socrates is doing no more than asserting that Polus believes that those who do whatever they think best do not (necessarily) have great power (cf. the phrasing of his conclusion at 468e3–5 in these terms). He is not also denying that Polus does believe that they do have great power. Polus says (is made to say): yes and no. Hence the interlocutor’s characteristic bewilderment. In sum, Socrates holds that: B(p&q)→B(r), where (p&q)→r; but not that, additionally, B(p&q)→∼B(~r).
100
R. Woolf
(3*) PO would disprefer A to B (3*) is equivalent to (2), and on certain plausible assumptions about the relation between preference and belief, translates fairly comfortably into (1), giving Socrates a proof of his initial assertion The validity of his argument need not concern us here. What is of present interest is the principle underlying Socrates’ procedure, which may be stated as follows: S: A subject believes a proposition p if he believes propositions that entail p. In the present case p would be (in our current terminology), “A is worse than B”. (1) is established by having Polus assent to propositions that (purportedly) entail p. Let us consider the force of S using a simple example. Say a subject believes that (a) All men are mortal, and (b) Socrates is a man. Does this entail he believes (c) Socrates is mortal? One would suppose not, as a matter of logic or even psychology. Though in practice it might be hard to find examples of an actual subject who would affirm (a) and (b) but deny (c), S thus far seems to have little bite as a general principle. But this is too quick. Since (a) and (b) jointly entail (c) one might say that the concepts of all men being mortal and of Socrates being a man jointly include the concept of Socrates being mortal: the latter conveys no information that is not already conveyed by the former pair. Thus, if “All men are mortal” and “Socrates is a man” are genuinely the objects of one’s belief, then one cannot (whatever one thinks!) fail to have “Socrates is mortal” as an object of one’s belief. It is already there, in virtue of one’s having (a) and (b) as objects of one’s belief. Believing (c) is, necessarily, part of what is involved in believing (a) and (b).53 The position I am outlining here represents a way of thinking about the relation between the mind and its objects that runs quite deep in Plato’s thought, and surfaces occasionally to take centre stage. It lies, for example, behind the strain in the Theaetetus which puzzles that having something as an object of one’s belief is incompatible with getting anything about it wrong, since to have a given thing as an object of one’s belief in the first place is to have in mind precisely it, that very thing, exactly as it is; else the hypothesis that that thing is the object of one’s belief is false. Thus the passage on “other-judging” at 189b–190e offers the view that being so much as in mental “contact” (190c6–7, d10–11) with a given object excludes mistake; and at 209c Socrates asserts that Theaetetus cannot be an object of one’s judgement unless one has his particular snub-nosedness and all his other qualities in mind. I do not claim that what one might call the “transparency” view of belief is correct (though I do not think its assessment is a straightforward matter). My aim
53 There is already a hint of this approach in Socrates’ testing of Meletus in the Apology, examined briefly in Section III above. Socrates claims at 27d that one who holds that there are divinities holds that there are gods given agreement that divinities are gods or the children of gods. Kamtekar notes a similar example at Symposium 201e–202d, in which Socrates is said by Diotima to believe that love is not a god given that this follows from other things he assents to. See Rachana Kamtekar, “Plato on the Attribution of Conative Attitudes”, Archiv für Geschichte der Philosophie 88(2006), 127–162 at 148 n. 42.
Socratic Authority
101
is to show how it might, in appropriate contexts, ground a certain view of epistemic authority. The case of Polus is one such context. Failure to have (c) as an object of one’s belief would, on this view, simply demonstrate that one did not in fact have (a) and (b) as objects of one’s belief. 54 There is, indeed, some plausibility about this position. For if it were independently established that Polus did not believe (c), then his claim to believe (a) and (b) would be a kind of cognitive mistake, a misidentifying of (a) and (b) as propositions with a different content, one that did not include (c): that is, as different propositions. What he actually believed, therefore, despite what he thought, would not be (a) and (b). The interlocutor, then, is faced with a dilemma. He must either concede that he does not believe (a) and (b) – though he takes it that he does; or that he does believe (c) – though he takes it that he does not. When he presses the point, as with Polus, Socrates generally takes it that the interlocutor does hold (a) and (b), and so (c). This, as we saw above, has the advantage of a kind of philosophical modesty, in that ascribing a belief to one who denies he holds it seems less contentious than the converse. Either way, mere self-ascription turns out to have no special privilege in the domain of belief. That short cut is unavailable. Instead, the question for Polus (or any subject) is whether he can wield the elenchus with precision on himself. The pretender to first-person authority about belief must be willing and able to scrutinise rigorously the propositions he asserts, together with their logical relations and entailments.55 If no one except Socrates can do this, then none except Socrates can safely claim such authority.56 In this way we can trace a clear link between Socrates’
54 The threat which the transparency view is seen in the Theaetetus to pose for the possibility of false belief, and therefore contradiction, has obvious implications of its own for the practice of Socratic method (in which the dialogue shows a renewed interest). Though the question is controversial, it seems to me (as 209c indicates) that the Theaetetus does not abandon the transparency view in the course of its attempts to tackle the problem of falsity. I offer a more detailed defence of the transparency reading (this point included) in my article “A Shaggy Soul Story: How Not to Read the Wax Tablet Model in Plato’s Theaetetus”, Philosophy and Phenomenological Research 69(2004), 573–604. Here I emphasise the positive contribution that it can make to the explanation of an important aspect of Socratic method. Note that Socrates’ strategy against Polus does not in itself undermine the reality of contradiction (cf. n. 52 above). Polus’s believing propositions p and q entails his believing r just if p and q entail r. His (also) believing ∼r is a matter for independent assessment. 55 So too, it may rightly be argued, must the pretender to third-person authority with respect to the propositions of the interlocutor. Why, then, the asymmetry? Why is the scrutiny of one’s own thought more difficult or problematic than the scrutiny of others’? The topic is discussed in section VI below. 56 An issue that calls for discussion at this point, but which I can do no more than flag here, is that of relative versus absolute authority. If A scrutinises B’s thoughts with a greater degree of rigour than B does, then, relative to B, A has authority with respect to B’s thoughts. But A’s scrutiny may well be less than perfect, and less good than that of some other person, C. Socrates, I have suggested, is the only figure in Plato who gives serious attention to the scrutiny of his own beliefs. But this need not be enough on its own to endow him with even relative authority over them. He must be able to examine himself with greater precision than outsiders can. I do think the tale of the Oracle in the Apology comes close to presenting Socrates as possessing absolute first-person authority. On the other hand, self-scrutiny is evidently an ongoing process (see also n. 68 below).
102
R. Woolf
seemingly unique capacity for elenctic self-examination and the authority of his self-ascriptions. In examining himself, Socrates will doubtless have discovered that he has contradictory beliefs, betokening a lack of knowledge.57 But in so doing he will have got clearer about what he actually does believe. The examination of Polus is analogous: the uncovering of Polus’ contradictions is revealing of what, despite Polus’ protestations, the latter actually believes. The moral is plain, and reinforces the divide between Platonic and Cartesian outlooks. To get an accurate fix on what it is one believes is a reward for intellectual effort; the privilege of those who have analysed, and not merely experienced, their own thoughts.
Plato and the First Person Grant that Socrates has a unique capacity for rigorous self-examination; let firstperson authority be hard to come by and as hard won as you please. The earlier puzzle remains. I have argued that for Plato first-person authority (to the extent that it exists) is grounded on the subject’s capacity for self-examination, and conferred only on Socrates. Yet we have seen that Plato never gives an unambiguous example of Socratic self-examination. Even the most promising candidate, found in the Hippias Major, is presented in third person terms. Why so?58 Let me highlight one issue in particular which, I think, begins to address Plato’s reluctance to show cases of Socratic self-examination as an explicitly first person exercise. This I shall call the problem of distance. The elenchus is designed to test the beliefs that an interlocutor holds. But interlocutors are notoriously reluctant to let go of their beliefs. We saw earlier how Socrates, for all his efforts, typically does not induce self-awareness in his interlocutors. They go away still convinced that they know, still, therefore, holding to whatever moral or other position they began with, even if this has been shown to entangle them in contradiction. If we have
57 As Chris Hughes points out to me, if principle S applies, then an interlocutor who holds contradictory beliefs will, within a framework of classical logic, believe every proposition. Relevance logics, designed in part to model real-life reasoning, would not carry this implication. 58 It is not as if an appropriate dramatic setting is hard to devise. The Hippias Major itself (with a little fine-tuning; cf. n. 36 above) already offers one such. Or imagine (to adapt the Symposium) Socrates late for dinner chez Agathon. A slave, sent to search for him, discovers him rapt in contemplation on a neighbour’s porch. On his eventual arrival Agathon insists he divulge the contents of his reverie, for the edification of all. Of course Socrates might refuse; his aversion to selfrevelation will be highlighted by Alcibiades later in the dialogue (see n. 43 above). But one would be asking him not so much to reveal the treasures in his soul as the way he sifts out the dross: applied to his own case rather than that of an interlocutor. That, it seems to me, would not breach any self-imposed taboo. After all, he advocates self-examination publicly and vigorously, and even, so it seems, does it in public. The secret is already out! In any event, other dramatic possibilities beckon: a straight first-person narrative (as in, for example, the Lysis and Charmides) would obviate the need for Socrates to be revealing himself to anyone in particular.
Socratic Authority
103
Socratic sympathies, we can hurl a variety of pejorative labels in the direction of this reluctance to yield: pride, complacency, stupidity! But let us consider how demanding the alternative is. What I am calling the “problem” of distance can certainly seem to flag something admirable about Socrates in this context. In principle, he seems able so readily to distance himself from his first-order beliefs. As he says to Critias, “never mind whether Critias or Socrates is refuted” – what matters is the truth (Charmides 166d–e).59 A similar sentiment may be found at Gorgias 458a–b, where Socrates says he would even rather be refuted than refute, since it is better to be freed from the evil of false belief oneself than to free another.60 Any first-order beliefs one might have are up for scrutiny. Indeed this willingness and ability to subject his first-order beliefs to scrutiny is, it seems, just what gives Socrates’ second-order belief about his (lack of) wisdom its authority. Now Socrates certainly holds moral beliefs, sometimes expressed with great conviction. And the fact that any given belief of his will, to the best of his efforts, have been subject to critical scrutiny grants him an epistemic entitlement which others do not possess. But at the same time his attitude requires that any given belief be subject to discard just in case the outcome of self-scrutiny mandates it. This willingness in principle to hold his beliefs at arm’s length is admirable, and understandably rather exceptional, just because, psychologically, it is hard to achieve the requisite distance. Issues of pride aside, most of us have something like a core set of beliefs, particularly in the moral arena where Socrates’ critical forces are primarily engaged, whose very fixity gives us a sense of who we are and a means of navigating our way through the practical business of life.61 Given this, it is hard in advance to treat such beliefs as occupying a no more than provisional place in one’s
59
It may seem remiss of me not to discuss the Charmides further in this context, since on the face of it the dialogue is Plato’s most explicit treatment of problems of self-knowledge. But despite its complex (and ultimately aporetic) discussion, it does not directly confront first person issues. Once self-knowledge is reformulated as knowledge of knowledge, the discussion treats knowledge of one’s own knowledge (or ignorance) on a par with knowledge of another person’s (see here Julia Annas, “Self-Knowledge in Early Plato”, in Dominic O’Meara (ed.), Platonic Investigations, Washington, DC: Catholic University of America Press 1995, 111–138, at 134), raising problems that apply equally to both. This parity is signalled repeatedly (e.g. 170b, 170d, 171d, 172b), enough to invite the suspicion that we are being encouraged to ask questions about it, though the dialogue does not do so itself. A brief exception is the puzzle about knowledge of one’s ignorance at 175b–c, discussed above. 60 In a typical case, it is the interlocutor’s initial assertion that Socrates regards as refuted, or shown false, by its contradiction with the entailment of the other assertions that he subsequently elicits. I shall not here enter the controversy, initiated by Vlastos, over whether Socrates is entitled to take this determinate line. Suffice it to say that some or other portion of the interlocutor’s belief-set has to go; and each option might involve the abandonment of some cherished dogma. The same, in principle, will apply to the examination of oneself. 61 See here Michael Frede, “Plato’s Arguments and the Dialogue Form”, in James C. Klagge and Nicholas D. Smith (eds.), Methods of Interpreting Plato and his Dialogues, Oxford Studies in Ancient Philosophy, Supplementary volume, Oxford: Oxford University Press 1992, 201–219, 215.
104
R. Woolf
belief-set, but this is what preparedness for self-examination requires. There is no comparable difficulty, of course, in being ready to kick others’ beliefs around. Herein lies the practical basis of the asymmetry between examination of self and other. If we may none the less speak of Socrates’ stance as an ideal, then a transcendence of the asymmetry beckons.62 However, Socrates alone is able to maintain the proper detachment. There is, I think, no more than superficial paradox in the idea of a kind of provisional conviction in Socrates’ case. Socrates’ beliefs, while lacking the systematicity required for knowledge, will have a firmer rational basis than those of others, just because they are constantly subject to his critical gaze.63 If so, it would be churlish not to grant, and admire, the uniqueness of Socratic first-person authority. But what is admirable (because difficult) can very easily shade into what is mysterious (because apparently impossible). It is arguable that it is not just difficult to attain the proper detachment required for self-examination, but impossible, or incoherent, to do so. There is a practical dimension here, though it does not get to the heart of the matter. As I just indicated, the fixity of some of our beliefs – or perhaps our holding them as fixed – plays an important role in the constitution of our identity. Actually to live the examined life is thus in a sense to live without a self, or at least be prepared to do so. If this stance continues to strike us as admirable, it may not be one we can think of embracing, whether through weakness of character, simple ignorance of the right way to live, or our own kind of human wisdom.64 But this practical awkwardness rests in turn on an underlying theoretical difficulty. To be genuinely questioning a belief of one’s own is (at least temporarily) to have suspended its membership of one’s belief-set. This is, it seems to me, a logical point about what it means to “question”, not an issue about the psychology of the agent. The agent has no practical choice about whether to retain the proposition as a belief while it is under critical consideration. That whose truth I question, I necessarily refrain from endorsing. And that which I refrain from endorsing cannot be something I believe. What began as a belief takes on, for the duration of questioning
62 Indeed if Socrates’ specialness lies in his treating his own beliefs as he does those of anyone else, then what seems puzzling, on the practical level, is why this translates into a preference for his own beliefs to be hygienic over those of others: to be refuted rather than refute. One might plausibly say that his distinctiveness is to be even harsher on himself than on others, without this explaining why his special concern should be with himself. 63 Nehamas (1998, 154) claims that “since that [viz. expert knowledge of virtue] was a knowledge he [Socrates] never acquired, his actions, which Plato depicts as invariably moral and right, were not produced by such reasons; they constituted a mystery: they had no source.” (His emphasis.) The inference from Socrates’ lacking systematic moral expertise to his having no (rational) source of action is something of a logical leap, and not one which I think Plato is guilty of. Socrates himself tells us (Apology 32d2–4) that all his concern (including, presumably, his powers of intellect) was directed at the avoidance of wrongdoing. A challenging position, but not inherently mysterious. 64 I include this latter possibility in spite of Nehamas’s trenchant discussion (1998, 41–44) of how Plato’s irony reaches out to his readers, condemning us for “avoiding Socrates” (42) precisely as we congratulate ourselves for being so much better than most of his victims.
Socratic Authority
105
(and possibly thereafter), the status of a mere proposition. If so, then the object of one’s examination is no longer oneself. When turned on itself in this way, the self that is supposed to be under examination seems to vanish, as surely (though for different reasons) as Hume’s did. From this point of view, Socratic self-examination turns out to be a suspect notion.65 The problem does not arise in the case of examination of others’ beliefs. Here the third person structure allows the questioner to be distanced from the beliefs he is questioning, as he must (logically) be, without the interlocutor resiling from his own beliefs. Thus there is nothing incoherent in questioning the beliefs of another. The beliefs can consistently remain the beliefs of that other, and be examined as such. One still has a self for examination, albeit not one’s own – nothing vanishes here. Interlocutors can (and generally do) persist in holding to their beliefs before, during and even after Socratic examination. Even a willing examinee can, so to speak, wait on the results of the examination without having already suspended judgement. But one cannot thus (sincerely) be questioning oneself. To embark on such a feat would already be to render one’s self elusive. I suggest that this is the fundamental asymmetry between self-questioning and the questioning of others which undergirds the asymmetry between first and third person authority that Plato presents. To return to Plato’s own beloved medical analogy, in the case of bodily states there is in the end a merely contingent asymmetry of authority. A doctor is better able to diagnose my bodily states than I who am not a doctor. But if I too happened to have medical training the asymmetry (logistical difficulties aside) would disappear. Here the analogy with the soul breaks down. Even to begin with, it would be a mistake of emphasis to say that only Socrates happened to have the character as well as the skill to examine himself, as if such a combination of gifts might easily have been prevalent. But this does not amount to saying that another of equal gifts is a flat impossibility. Where we end up, though, is with an asymmetry of full strength from the modal point of view: anyone in principle can examine others; no one, however gifted, can coherently undertake self-examination. This, at any rate, is the thesis I offer to explain the otherwise peculiar dominance of the third person perspective in Plato’s works. So in the Hippias Major the third person structure remains even when it is purportedly self-examination that we are dealing with. So too the Euthydemus, in the section reviewed above (n. 27), represents Socratic self-criticism as critique by a third party. And even a dialogue such as the Theaetetus, which famously describes thinking as the soul’s dialogue with itself (189e–190a), offers this picture as an account of belief formation, not
65 It is worth distinguishing this problem from a critique that has been made most notably by Bernard Williams (of Kant but also of Plato), to the effect that the very idea of stepping back from a given set of ethical commitments and assessing them from a standpoint of disinterested reason is incoherent or illusory. This is a point about the status of the examining rather than the examined self, and not one, I think, which Plato would be inclined to grant. Contrast in turn the practical difficulty, discussed above, of seeking to occupy such a standpoint.
106
R. Woolf
examination of beliefs already formed. Indeed the picture emphasises that so long as a proposition is being debated with oneself it is not a belief; and this seems to me conceptually correct.66 A similar picture is to be found in the dialogue’s immediate successor, the Sophist (263e–264b). Even the Philebus, with its more elaborate account of imagination and thought, continues to reserve the notion of internal question-and-answer for belief formation (38b–d). We can search high and low in Plato for a theoretical account of Socratic self-examination, but we search in vain when even the most promising passages refrain from extending their model of internal dialogue to an appropriate case.67 If the very notion of self-examination turns out to be incoherent, so too does the concept of a distinctively first-person authority in Socratic terms. We can certainly explain some troubling phenomena: Plato’s reluctance to portray self-examination as such despite its centrality in Socrates’ life, his inclination to convert it to the third person in some notable cases, and his unwillingness to offer a theoretical account of it when one seems to beckon. Against this, I have argued that Socratic first-person authority is portrayed as unimpeachable in the Apology, and that this authority is defended by reference to Socratic self-examination. Plato surely offers us Socratic self-examination as being on some level real and important. But he cannot account philosophically for its first person character. So the particular authority on which Socrates’ uniqueness is founded is mysterious.68 There is an integrity – that of a
66 The soul is said simply to be examining “things” in this situation (189e7). Compare also Theaetetus 154e–155a, where Socrates speaks of examining whether or not one’s thoughts harmonise with one another, and calls these thoughts not beliefs but “appearances”, using the relatively unusual word phasmata (a2). 67 It can hardly be objected that such an account would not be relevant to these dialogues’ concerns. The Theaetetus is keenly interested in questions of Socratic method; the Philebus restores Socrates to the role of questioner-in-chief after the temporary reign of the Eleatic Stranger; even the Sophist, with its category of the “sophist of noble lineage” (230a–231b) who purges souls of their false conceit of wisdom (the medical analogy turns up again, 230c), displays an interest in something resembling traditional Socratic practice. Indeed it has been suggested by Mary Margaret McCabe in Plato and His Predecessors (Cambridge: Cambridge University Press 2000) that part of the project of these dialogues is to reflect on (and to some extent defend) the Socratic practice of philosophy. If so, it is even more striking that its first person exercise does not come within the purview of their reflections. McCabe herself emphasises their interest in dialectic as an inter-personal activity. 68 At Phaedrus 229e5–6 Socrates says intriguingly: “I am not yet able to know myself, as the Delphic inscription bids”. In this, perhaps the most enigmatic of Plato’s dialogues, does Socrates himself confess failure? If so, it is strictly qualified. He has not yet attained self-knowledge: enough, though, to reaffirm awareness of his ignorance (235c7–8). Self-examination is a continuous process, and, as usual for Socrates, it is real enough: “I do not examine [extraneous matters], but myself” (230a3). Socrates investigates whether or not he is conflicted (so the normatively contrasted pairs of complex/ fierce and simple/gentle at a3–6 suggest). But the Phaedrus is no more forthcoming with illustrations than other dialogues, despite its marked taste for Socratic monologue (especially the magnificent discourse on love, madness – and psychic conflict, 243e–257b). On the Phaedrus passage see George Kateb, “Socratic Integrity”, in Ian Shapiro and Robert Adams (eds.), Integrity and Conscience. Nomos 40, New York: NewYork University Press 1998, 77–112 at 98. Kateb’s remarks on Socratic self-examination are pertinent but his claim that “perfect self-knowledge would be incompatible with having a self” seems to me to mislocate the incompatibility.
Socratic Authority
107
good Socratic disciple - in Plato’s refusal to play down either its importance or its mystery. But in the end it is doubtful that his Socrates can be judged an exception to the authority of the third person perspective. Is this altogether a negative conclusion? One might consider that the conception of authority that even Socrates fails to elude should command attention for at least two reasons. It challenges us to make sense of the first person perspective; at the same time, and in contrast to what would become the dominant philosophical tradition, it offers an unabashed primacy to the third person character of epistemic authority.69 King’s College London, UK
69 This paper has its origins in a sabbatical year spent at the Institute for Advanced Study, Princeton, in 2002/3. Earlier versions were read, in variously abridged formats, at the University of Helsinki, Temple University, King’s College London and the University of Michigan. I would like to thank the audiences on those occasions for stimulating and helpful discussion, and Peter Adamson, M.M. McCabe and an anonymous referee for the Archiv für Geschichte der Philosophie for further stimulus in writing.
Protean Socrates: Mythical Figures in the Euthydemus Mary Margaret McCabe
Myths and Curlicues At the midpoint of the Euthydemus Socrates tries to calm Ctesippus who is enraged by the twin sophists’ arguments about truth and contradiction: Ctesippus, I say the very same things to you as I said just now to Cleinias,1 that you don’t recognise the amazing nature of our visitors’ wisdom. But they are not prepared to give us a serious demonstration; instead they imitate Proteus the Egyptian sophist, and cast spells on us.2 So let us imitate Menelaus, and not let these chaps go until they manifest to us their serious side.3 For I think that something quite beautiful in them will appear once they start to be serious. So let us beg and advise and pray them to make themselves manifest. Indeed, I think I should begin by giving a pointer to what sort of persons they will, I pray, clearly turn out to be. (Plato, Euthydemus 288b–c)
You might be immediately sceptical of my title. This image of sophists – as slippery shape-shifters – is exactly what we should expect of Plato4; surely it is not Socrates, but the sophists, who are Protean in this tale? And surely there is little more to be got from the mythical image than a point which is already being made in the arguments: that these characters really can’t be tied down to a single definitive thesis, can’t be persuaded to stay the same even for a moment, are quite disqualified from entering upon serious5 dialectical exchange? 1 The possibility of such repetitions has been questioned by the sophists in the preceding episode. 2 Notice here the echo of Meno’s prediction of Socrates’ fate – that if he goes to another country he will be arrested as a wizard, Meno 80b. 3 Adapting Sprague’s version in John M. Cooper (ed.), Plato: Complete Works, Indianapolis, IN: Hackett 1997, 726; the construction is clearly one which anticipates an explanation for the sophists’ seriousness. 4 See Politicus 291a–b. 5 Seriousness is a running theme; see the opening of the section, 283b, or 278b–c. There is also a running connection between two pairs of ideas: seriousness and play, and age and youth, perhaps via the complex pun pais/paideia. This is, I think, centred on the central sophistic episode and on the significance of persistence. See here David Roochnik, “The Serious Play of Plato’s Euthydemus”, Interpretation 18(1990–1), 211–232.
P. Remes, J. Sihvola (eds.) Ancient Philosophy of the Self, © Springer Science+Business Media B.V. 2008
109
110
M.M. McCabe
This response might be thought to generalise. The Euthydemus – a dialogue which, after all, is not one of those famous for a big myth – might be thought to use mythological material as part of its literary apparatus, as a shorthand set of descriptors of the positions adopted by each of the characters in the piece. So Socrates (and a rather uncertain Ctesippus) would be cast as Menelaus because, true to the Homeric paradigm, they are going to be doughty in their pursuit of the truth. This, of course, is a rather different Menelaus from the dismal character of the Iliad, or from the weedy figure unfavourably compared to his big brother at Symposium 174c. This, instead, is the Menelaus of Odyssey 4, who tells Telemachus the tale of his own journey home from Troy. He and his men were stranded off Egypt; but Menelaus got Proteus, the old man of the sea, to tell him the future and the past – and so how to escape home. (The sea-nymph Eidothee tells Menelaus how to find out from Proteus how to get home): […] The moment you see that he is asleep seize him; put forth all your strength and hold him fast, for he will do his very utmost to get away from you. He will turn himself into every kind of creature that goes upon the earth, and will become also both fire and water; but you must hold him fast and grip him tighter and tighter, till he begins to talk to you and comes back to what he was when you saw him go to sleep; then you may slacken your hold and let him go; and you can ask him which of the gods it is that is angry with you, and what you must do to reach your home over the fishy sea. (Odyssey 4.415–424, trans. Butler)
The trick, of course, is to hold on to Proteus while he shifts his shapes6 until he capitulates and reverts to his infallible self (4.384): then he will tell the truth. So if the view that these mythologies are convenient descriptors is right, Socrates is to be seen as the brave pursuer of the truth (like Menelaus), and the twin sophists as able to give truth to him, just so long as they may be persuaded to be serious (like Proteus). This view of Plato’s use of such mythical figures as literary devices, as shorthand for direct descriptions, tends to subordinate the literary – or, to put the matter in a more inflammatory mode, the merely literary – to the philosophical. Someone who held this view might also suppose that the dialogues are written in elaborate and varied ways just as pedagogical, or even rhetorical aids: to enrich, illuminate, make compelling, the philosophical arguments, themselves conceived as bare argumentative structures. The literary aspects of the dialogues, on such a view, make no difference to the dialogue’s philosophical content; from the philosophical point of view they are just curlicues. If, then, these mythical figurings belong to the literary aspects, then they are not themselves crucial to our philosophical understanding of the piece.7
6 Proteus “becomes everything”. Notice the repeated use of the verb gignesthai, ‘become’; 4.417–8 says, in a more literal than elegant version: “he will make a trial, becoming everything that becomes a creeping thing [or every creeping thing that comes to be] on the earth.” 7 We might, carelessly perhaps, find ourselves saying the same sort of thing about Platonic imagery. This, of course, has no implications one way or the other about what we should say about Platonic stories – for example, the big eschatological myths.
Protean Socrates: Mythical Figures in the Euthydemus
111
Three Figures Let us put on one side whether such an impoverished conception of philosophy can ever satisfy, and put on one side, also, the huge question of the relation between Plato’s ways of writing and the philosophical content of the dialogues.8 Consider, instead, a more limited and a more particular question: whether in fact these mythical figurings here do, or do not, make any difference to our understanding of the arguments that are put forward in the dialogue. I shall claim that they do; and I shall argue for this claim on the basis of three figures: Proteus, Marsyas and Cronos.9 All three appear in the middle of the dialogue, in the course of the central sophistic section (283c–288a) which is marked not only by the density of its sophistic arguments, but also by the rich mythological allusions which surround them. The sophists force their interlocutors to concede that: 1. If they wish Cleinias to become wise, they wish him dead (283b–e). 2. Falsehood is impossible (everything is true) (283e–284e). 3. Contradiction is impossible (285d–288b). and they resist Socrates’ counter-argument that the latter claims are self-refuting (288a). The mythical figures are interleaved. First, after the demonstration that everything is true, Socrates suggests that they might chop him up, boil him, do anything to him, just so long as he becomes excellent.10 Ctesippus, simultaneously infuriated and enchanted by the sophists’ tricks, promises to hand himself over to them, to be destroyed and skinned, like Marsyas, just so long as they fill him up with virtue:
8 I have attacked the view that the literary frames of the dialogues are insignificant to the philosophical content on several occasions, see e.g. Mary Margaret McCabe, Plato and His Predecessors: The Dramatization of Reason, Cambridge: Cambridge University Press 2000. 9 I leave aside for another occasion two other figures which recur in the dialogue: Heracles and Medea. But each of these, too, is somehow integrated into the argumentative structure of the dialogue. Heracles appears as a hero in need of support from his companion (297b–e), a motif played on towards the end of the dialogue where Ctesippus vainly tries to come to the aid of Socrates by shouting “bravo, Heracles”; and is convicted by the double act of the brother sophists of having said nothing at all (303a). Medea’s murderous activities in boiling her enemies (alluded to here, 285c4) are recalled in a later argument about cookery (301c–d) which turns on questions about crafts and the difference between subject and object. In the overall account of the dialogue, each of these characters plays a role; for present purposes my aim is to show how much we have to gain from taking these figures seriously, by means of the example, especially of Proteus. I take the view that a large proportion of the Euthydemus is concerned with matters of metaphysics, see McCabe, “Persistent Fallacies”, Proceedings of the Aristotelian Society 1993, 73–93; “Silencing the Sophists: The Drama of Plato’s Euthydemus”, Proceedings of the Boston Area Colloquium in Ancient Philosophy 1998, 139–168; the Medea figure fits with the dialogue’s discussion of properties, parts and wholes, the Heracles figure with the dialogue’s account of counting individuals. 10 This is his “Carian crisis”, where he might just as well be one of Medea’s victims, 285c; see previous note.
112
M.M. McCabe
I too, Socrates, am ready to give myself up to our visitors, and if they wish they may flay me even more than they have already done, just so long as my skin ends up, not full of wine, as Marsyas’ was, but full of virtue. (Euthydemus 285c7–d1)
Second, when Socrates insists that the thesis that contradiction is impossible is inconsistent with what the sophists have said before, Dionysodorus replies that Socrates is a Cronos, passé, in his old-fashioned way recalling what the sophists said before, last week or last year: “This is my bothersome question,” I said. “If indeed we do not make any mistakes, either when we act or when we speak or when we think, then as for you, by Zeus, if this is so, what do you come here professing to teach? Didn’t you say just now that if anyone wished to learn virtue, you would hand it over best?” “So, Socrates,” said Dionysodorus taking me up, “are you such a Cronos that you now recall what we said first, and if we said something last year you would now recall it, but you will have nothing of what is needed to deal with what is said in the present?” (Euthydemus 287a–b)
Thirdly, at the end of this central section, Socrates offers himself as Menelaus, to catch hold – we seem to be invited to say – of the Protean sophists. He will – as the passage I quoted above suggests – get hold of their serious side, find something beautiful in them. It is not hard to read these representations straight. Ctesippus is willing to be filled up with whatever the sophists will provide; so Ctesippus, like Marsyas, becomes whatever is poured into him (it is no coincidence, perhaps, that Plato chooses this figure: Marsyas gave his name to a river11). Socrates, on the other hand, is an old curmudgeon, unwilling to shift from his old plodding ways to adopt the glittering modes of sophistry. But that is double-edged: Socrates also, perhaps, represents the golden age of the right methods of philosophy.12 And in his likeness to Cronos, as well as to Menelaus, Socrates is the antithesis of the shape-shifting Proteus: the point, then, of these pieces of mythology is to give us a character sketch, a quick view of just how each of the dramatis personae is developing. Socrates is still no Proteus. There is, of course, just a bit more to these mythical figurings than this, even at first sight. Let me cast the first doubts by wondering about the provenance of Proteus. You might suppose that the shape-shifter image of the old man of the sea is a standard piece of Greek myth. The earliest version of the story is the one we find in Odyssey 4, although even there the shape-shifting features of the Old Man of the Sea are secondary to his ability to tell the truth about the past and the future. But Proteus also figures elsewhere, in places that do not force the shape-shifting view upon us. On the contrary, they focus attention on Proteus’ exemplary intellectual and moral qualities. So in both Herodotus and Euripides, Proteus is honourable 11
Cf. Herodotus, 5.119.1, 7.26.3 and Xenophon, An. 1.2.8. There is another tale to tell about how, and whether this is meant to tie in with matters Heraclitean; I shall not pursue that here. 12 Compare the role of Cronos at Politicus 272bff. Of course that myth too may be ambivalent and certainly requires some interpretative work.
Protean Socrates: Mythical Figures in the Euthydemus
113
and wise, a mortal Egyptian king, a man of honesty and excellent judgement.13 In Plato, however, he turns up several times; each passage is consonant with the shapeshifting Proteus here.14 If Plato, in painting a dismal picture of Proteus as an inconstant and changing character, hopes thus to attack sophistic reasoning, what is his readership to do with their inherited view of the same character: the rational, wise figure who tells the truth to the brave? I shall suggest that the recalling of the inherited view is fundamental to our understanding of his significance.
Who and of What Sort? The plot may thicken in the next section of the dialogue. Now the discussion is being led by Socrates, and is apparently of a quite different character and depth. It is no more positive in its conclusions, however, than its sophistic predecessor. There has been an exchange between Socrates and Cleinias about some skill or craft, the possession of which will make us happy; they find themselves unable to identify just what it will be. Then Cleinias makes an unexpectedly telling point15 as a consequence of which the argument ends up in some difficulties. Once Cleinias’ point has been made, the frame dialogue breaks in,16 and Crito (to whom Socrates is giving an account of his encounter with the sophists) asks: CR: What are you saying, Socrates: did that young man say this? SOC: Do you think not, Crito? CR: I certainly think not, by Zeus. And for my part I think that if he did say it, he needs no education from Euthydemus nor from any other man.
13 Herodotus (2.112 ff.) has him delivering some judgments of Solomon, while in Euripides’ Helen he is the original kind host (in the play, now dead) who protects Helen from the slings and arrows (see especially Helen 46–47). 14 Most notable, perhaps, is Republic 381d, where Socrates castigates those who represent divine figures as shifty characters: an extraordinary piece of self-criticism, perhaps? At Ion 541e Proteus is again a shape-shifter, and is associated with the kind of overturning of reason we might associate with a sophist – or Heraclitus. At Euthyphro 15d the emphasis is on not letting Proteus go until he answers (and that is a general theme which recurs throughout the dialogues: compare e.g. Philebus 20a). 15 One, indeed, which should provoke close reflection in the reader upon what is said in the central books of the Republic, see McCabe, “Developing the Good: Prolepsis or Critique in the Euthydemus?”, Plato: Electronic Journal of the International Plato Society 2(2001). available online at http://www.nd.edu/~plato/plato2issue/mccabe.htm 16 Again this feature of the “literary” structure of the dialogue has an impact on how we read it; for here, as often elsewhere, the intrusion of the frame reminds us of the fictionality of the piece: and that, of course, should make a considerable difference to its philosophical reading (for example, the anhistorical features of the figure of Socrates might discourage a view that Plato here gives us a body of Socratic doctrine).
114
M.M. McCabe
SOC: Well, perhaps it was Ctesippus who said it, and I just didn’t remember? CR: What sort of Ctesippus?17 SOC: Well, I am sure it wasn’t Euthydemus or Dionysodorus who said it. But, my dear Crito, perhaps some more powerful being was there, and uttered those words? I am certain that I heard them said. CR: Yes indeed, Socrates, it seems to me that it was some more powerful being, very much so […]. (Euthydemus 290e–291a) This exchange is usually read as a piece of piety: if something clever was said, surely Socrates said it. So it is Socrates who is the more powerful being in question (hence Crito’s rather knowing “Yes indeed, Socrates, it seems to me that it was some more powerful being, very much so”). But in the context of the argument itself it is not at all clear that Socrates should take responsibility for what Cleinias is reported to have said, not least because it buys into a model of ethical reasoning which fits ill with Socrates’ own conclusions at 281.18 Instead, the whole sequence invites us to wonder who is who, and of what sort (and this is where Proteus will come back to haunt us: he, after all, “becomes everything”). Begin with Ctesippus: “what sort of Ctesippus?” It is Ctesippus who is willing to be Marsyas – who was flayed alive for daring to compete with Apollo on the flute. Marsyas’ skin was then hung up at the spring of the river named after him – and, on Ctesippus’ account, filled with wine. Ctesippus is willing to undergo the same fate, just so long as he is filled with virtue (not wine): and his behaviour as the dialogue progresses indicates that he seeks the sort of virtue provided by the pancratiast sophists – the aristocratic virtue that brings success. But nonetheless the image Ctesippus offers is a ghoulish one. When Marsyas is flayed alive, we imagine that the alive Marsyas is the man without his skin. Ctesippus, however, takes the skin to be the man, and its character to be determined by whatever substance fills it up.19 His view of virtue is a peculiarly materialist one, therefore: and facile – what
17 Michael Trapp points out to me that this remark would ordinarily be construed as just the expression of incredulity: “Ctesippus said that?” On my account, Plato’s interest in the problem of personal identity has revived the metaphor lurking in the odd grammar of the expression and makes us think about it rather more expressly: hence my translation here. Further, the Euthydemus as a whole focuses attention on character and disposition as an account of the nature of moral value; one of the aspects of its drama is the way in which Ctesippus changes – far too fast under the influence of the sophists – into a slick sophist, by the close of the dialogue: so “what sort of Ctesippus?” Compare Protagoras 311bff.: “who will you become?” under the influence of a sophist? 18 I have discussed this passage in more detail elsewhere in my “Indifference Readings: Plato and the Stoa on Socratic Ethics”, in T.P. Wiseman (ed.), Classics in Progress: Essays on Ancient Greece and Rome, Oxford: Oxford University Press 2002, 363–398. 19 At Symposium 215aff. Alcibiades characterises Socrates as Marsyas; but he makes rather a different use of the contrast between the inner and the outer man in so doing: that picture of Socrates as Marsyas is far more positive than this picture of Ctesippus as Marsyas. However both exploitations of the myth turn on just how, and where, we are to see true character.
Protean Socrates: Mythical Figures in the Euthydemus
115
you need for virtue is to get filled up with it.20 And on this view of who he is, Ctesippus is of whatever sort are his contents – “what sort of Ctesippus” could have made the point Crito so admires?21 The same sort of indeterminacy surrounds the figure of Socrates. Of course Crito’s words do invite pietistic interpretation precisely because Crito himself is liable to that attitude to Socrates himself: the intention Crito has is to allude to Socrates. But, reading the dialogue from outside Crito’s perspective – as the interruption itself invites us to do – we may be more wary. From our perspective, Crito’s words invite the question whether Socrates is, or is not, one of the “more powerful ones”. The sophists, you might recall, have suggested that he is; although their identification of him as the god Cronos looks rather more like an insult than a compliment, from the perspective of the sophists. But Socrates himself prefers to imitate a man – Menelaus; and later a demigod, Heracles, at 297 (although here he portrays himself as a hopeless kind of Heracles). In taking on these guises Socrates is Protean – not a lion or water, but a trio of different people (Cronos, Menelaus, Heracles) each with a different ontological status. Is there any point to this multiplicity?22 Well, here is one answer to that question: if the sophists are to be Proteus and Socrates is to be Proteus, this is an old Platonic story. After all, one of the common features of Plato’s account of sophistry is his insistence that there is in fact rather a narrow and complicated line to be drawn between the tricks of the naughty sophists and the serious business of Socrates, the noble sophist (Sophist 230–231). The arguments of the eristics, which force the interlocutor into inconsistency and despair, are not far from the Socratic elenchus, which so often does the same. So if the figure of Proteus does indeed represent both the sophists and Socrates that is hardly news. It is – on that account – merely a challenge thrown down for Plato’s reader to take up by disentangling what it is (if anything) that makes Socrates the good guy, and what it is that makes the sophists so naughty. Does that explain the multiplicity of Proteus? It still fails to account for the gap between Plato’s presentation of this figure and the inherited view: is there more to be said about that?
20 On this see J.S. Green, Reflections on Selfhood in Gorgias’s Encomium of Helen and Plato’s Apology of Socrates, unpublished Ph.D. dissertation, University of London 2005. 21 Notice that a similar question lurks in the Odyssey: 4.421 – you must hold on to Proteus until he is such as he was when he first went to sleep. 22 Socrates places heavy emphasis on the fact that Proteus and Menelaus are being imitated. Cf. Louis Aryeh Kosman, “Silence and Imitation in the Platonic dialogues”, in James C. Klagge and Nicholas D. Smith (eds.), Methods of Interpreting Plato and his Dialogues, Oxford Studies in Ancient Philosophy, supplementary volume, Oxford: Oxford University Press 1992, 73–92. He argues that Plato’s account of imitation stresses this kind of subterfuge, where some real characters stand behind the mythical masks. There may be something to made of the fact that the sophists, by contrast, identify Socrates with Cronos. Once again, there is no space to pursue this matter here; but let this be a promissory note for a future discussion.
116
M.M. McCabe
Serious Proteus: Truth and Persistence Return to the figure of Proteus and Plato’s close reference to the story in the Odyssey. It is a commonplace to observe that figures of myth are ambivalent; and Proteus is no exception. For he is indeed a shape-shifter; but he is also the immortal who, restored to himself, tells the truth. He does so after the struggle to make him resume his own shape; and when carefully questioned he is able to tell his interlocutor the truth about the past and the future. It is this Proteus whom Socrates alludes to when he says he is going to hang onto the sophists until they reveal their own seriousness. It is this Proteus who, from our perspective as we read the dialogue, we can only expect to find in Socrates: in the Socrates whose continuous rationality is made evident in the parts of the dialogue where he leads the discussion. This is not mere pietism, as I hope to persuade you. Instead, this Proteus resonates clearly with two major themes of this dialogue, which appear repeatedly both in the frame dialogue23 and in the sophistic stretches of argument. The first is metaphysical: What is it for someone to persist through change?
The second is epistemological: What it is to be able to tell the truth?
Both themes, however, are firmly located in the ethical context of the dialogue: the context in which the interlocutors are (whether seriously or not) seeking the skill which will make them happy, make them good men. In particular, the dialogue tries to find a skill which Cleinias may learn, in order to be happy and a good man. The first question, then, focuses on the history of Cleinias, and asks: What is it for someone to become better, and so persist through change?
The second focuses on how Cleinias is to become better, and attends to the various proposals for how it is that our lives can be organised by a rational plan. So it asks: What is it to tell the truth about what is best?
I suggest that Plato uses these mythical figures to force reflection on the connections between the metaphysical and epistemological questions, on the one hand, and their ethical counterparts on the other. For it is in these figures that he asks how truth and persistence are to be construed in the terms of someone’s life; and in this way defends a view radically different from that advanced by the sophists.
23 Cf. the running theme of Cleinias’ development and change (e.g. 275a9); and the young/old material emphasised by Socrates at 272b–d. Compare also the discussion at the outset of the transformations the sophists have worked on themselves (272b); and the way in which Ctesippus too is suddenly changed into a sophistic arguer by the last sophistic episode (298b).
Protean Socrates: Mythical Figures in the Euthydemus
117
If something changes, does it persist through the change? This ancient question of the Greek cosmologists is given a dangerous slant by the sophists of the Euthydemus.24 In a famous argument (283c–d) they claim that if Socrates and Ctesippus want Cleinias to become wise, they want him to no longer to be ignorant: “Well then,” he said. “You say that you want him [Cleinias] to become wise?” “Certainly.” “And is Cleinias now wise, or not?” he asked. “He is not”, I said, “yet. For he is not boastful.” “But you”, said he, “want him to become wise, and not to be ignorant?” We agreed. “Surely who he is not, you want him to become, and who he now is, no longer to be?” And I, when I heard this, was troubled; and he, seeing my trouble, took the point up again. “So since you wish him no longer to be who he now is, you wish him, so it seems, to perish. What worthwhile friends you are, and lovers too, who value their beloveds to death.” (283c–d)
The damage is done by the conclusion: “Surely who he is not, you want him to become, and who he now is, no longer to be?” For in that case, if Socrates and his friends want Cleinias no longer to be who he is now, they wish him dead. The Odyssey might ask the same question. For when Proteus tries to escape the person who captures him, “he will turn himself into every kind of creature that goes upon the earth, and will become also both fire and water” (4. 417–418). If Proteus turns into everything else, we might ask, what ensures that each incarnation is an incarnation of Proteus (not a mere replacement)? Or that Menelaus will ever get hold of the “real” Proteus at the end? The sophists’ argument about Cleinias would go through if they subscribe in addition to the premise that any change is death. But we need construct no sophistic metaphysics. For the sophists are interested, not in putting forward theses of their own, but rather in exploring what follows from what their interlocutor says.25 So it is Socrates, their interlocutor, who had better have the right set of premises. And there is something at stake in having the right premises here. It is not enough to walk away, or to suppose, as Socrates and Ctesippus are at one point inclined to do, that they can just accept a sophistic death, so long as they end up wise and virtuous: for who will end up wise and virtuous? Once again, Proteus may show us the way. For if we can hang on, and get the real Proteus in our grasp, he is bound to tell us the truth, and show us the way to go home. And that matters, if we, like Proteus, can persist through the journey. Who then is the real Proteus, or the real Cleinias? The sophists’ question “Surely who he is not, you want him to become, and who he now is, no longer to be?”
24 The similarity with the puzzles of early cosmology motivates, I take it, the frequent observation of the commentators that these sophists are somehow Eleatic; see Rosamond Kent Sprague, Plato’s Use of Fallacy, London: Routledge & Kegan Paul 1962. 25 At the outset they boast that whatever Cleinias says he will end up refuted (275e).
118
M.M. McCabe
focuses attention on the issue of personal identity. It invites Socrates to agree that who Cleinias is, is delimited by the character and properties that he has at some particular time; when once one of those may change, he is no longer who he was before. To deny that view, Socrates must suppose that there is one case at least where Cleinias persists through some kind of change. And to do that defensibly he needs not only to insist that Cleinias does persist through change but also to explain how, and under what constraints.
Easy and Hard Persistence What do I mean by “constraints”? There are, of course, notorious puzzles about persistence and change for cases of ordinary material objects – think of Theseus’ ship. But that puzzle is not apparently at issue in the Euthydemus. Instead, the sophists seem to offer Socrates an all or nothing choice:26 between what I shall call the episodist approach offered in the “Cleinias dies” argument, and what I shall call an absolute persistence. Suppose, first, that we concede that any change is a death, as the “Cleinias dies” argument supposes. This might allow us to think of what seems to be ordinary change as a succession of states of affairs. Each state of affairs will persist only so long as it remains exactly the same; once any feature of it alters, the whole state of affairs simply ceases to exist. What appears to be change, therefore, is nothing but a series of episodes, discrete, self-contained and liable to perish. At the other extreme from this episodism would be a conception of each state of affairs as invariant (perhaps on the grounds that its going out of existence would be a change for it, and thus incoherent). This is peddled by the argument that if Socrates is knowledgeable at all, then he is omniscient, always (293b–296e).27 That argument28 supposes that if someone is (now) thus and so, or of such and such a character, then that state of affairs is invariant, both in various respects and over time. You might see the choice that is offered thus. If you are now ignorant, then either any alteration will amount to your going out of existence (who you are is thus reduced to a single episode, or gerrymandered to cover a succession of discrete states of affairs); or there is no such thing as alteration at all (existence is absolute). In order to avoid the conclusions (that Cleinias dies, or that Socrates is omniscient)
26 As they do in other contexts: see e.g. the discussion of truth and contradiction at 286a; or the discussion of family relations at 298dff. There is a continuous discussion here about the nature of relations which reflects further on “all or nothing” choices of this kind. 27 An argument which ends, surprisingly, in the discomfiture of the sophists. 28 Itself fraught with various contentious claims, not least a running dispute about the laws of noncontradiction, see McCabe, “Does Your Plato Bite?”, in John Dillon and Monique Dixsaut (eds.), Agonistes: Festschrift for Denis O’Brien, London: Ashgate 2006, 107–120.
Protean Socrates: Mythical Figures in the Euthydemus
119
Socrates needs to offer an account of persistence through change which allows Cleinias to survive, and precludes his own turning into an immortal. He – or someone – needs to explain (to give a serious account of) what it is to answer Crito’s question “who is who”. And yet within the so-called argumentative sections of the dialogue, Socrates only offers theories of his own where he leads the discussion; and in those passages he seems to restrict himself to questions about value and virtue. In the sophistic sections, by contrast, he is confined to the role of interlocutor, in which he strikes various ironical attitudes, but offers little of theoretical substance. So where is the Socratic account of persistence to be found? Nowhere, you might say, even though most of the dialogue is narrated by Socrates. But Plato certainly has something to say on the matter. For it is Plato who juxtaposes the figures of myth with the arguments in the order they appear; and who provides us with Crito’s startling interruption. In the puzzling figure of Proteus, he offers a complex view of persistence. As far as the sophists are concerned, Proteus embodies the shape-shifting features of an episodist account of change; but hold on to him in all seriousness – as Socrates urges us to do – and the inherited view comes into its own. For the story in which Proteus is embedded relies on the assumption that individuals can persist and change too. Proteus, that is to say, gives the lie to the sophists’ insistence that we must choose between episodes and absolute persistence. One of the difficulties attached to the absolute conception of persistence, as Socrates himself points out (294a) is that it makes everyone wise; or even, as Dionysodorus points out, everyone wise and not wise at the same time. Socrates admires: […] For if indeed I have escaped my own notice being wise29 but you show me that I know everything and for ever, what greater godsend could I find in my whole life? (Euthydemus 295a)
The ethical tone of the question is set by the last phrase; it is the search for what could make a whole life happy or fine (cf. 293a) which dominates Socrates’ sections of the dialogue.30 Socrates and Crito have, from the beginning, rejected the idea that this could occur piecemeal or unsystematically; instead they are searching for some kind of wisdom. Central to this is the notion of telling the truth about the past and the future – for which Proteus, restored to himself, is famous. But then why should not the sophists’ suggestion, that they are (all) already provided with what Socrates has insisted to be the source of happiness, end the quest? Because it is just too easy: child’s play.
29
I leave the slight oddity of the Greek to stand in an awkward English – it shows up the metaphysical overtones of Socrates’ remark, where Sprague’s “if I am unaware of my own wisdom” presses only the epistemological. 30 Note that in fact Socrates only talks explicitly about lives in these two contexts; but the question about lives is continuous with the question about persons, who is who. See further below.
120
M.M. McCabe
You might say the same of the account of truth that the sophists offer: an account which has affinities to the episodist account of persistence (284a).31 When someone makes a statement, they make it about the thing it is about; and the thing it is about is (and so is true). The thing it is about, moreover, is distinct from other things. So if anyone else makes a statement, they make it either about that thing (in which case the two agree) or about some other, distinct thing (in which case they are talking at cross-purposes). So anything that someone might say, so long as it is about some thing, is true; and not subject to revision or change by virtue of some other thing that they, or anyone else might say. The intentional object of any statement, thus, determines both the content and the truth-value of the statement; and the intentional objects just are whatever determine this or that statement. They are, thence, distinct from each other, and one has no bearing on the truth-value of another. There is, therefore, no relation between statements other than separation; no statement is related in terms of consistency (or any other logical relations, including consistency over time) to another. Truths are discrete and self-contained; and any statement that succeeds in stating, states something true.32 And in that case, telling the truth is as easy as speech: anything you say is true, and there is no risk of being wrong. The Odyssey story about how to get Proteus to tell us the truth suggests, by contrast, that it is extremely hard, and something you have to take very seriously. You need to hang on to him, even when he is a lion, or fire, and only when you have done that will he answer all your questions: only when you have done that will he reveal himself in seriousness. And even then his omniscience is something quite extraordinary, not, like the omniscience offered by the sophists, derived simply and easily from his knowledge of one thing. The Proteus who persists, then, embodies a serious (normative) notion of knowledge and of the systematic telling of truth, and one which extends over time, in time with his own (immortal) persistence. So telling the truth can be easy, or it can be hard. Learning the truth, too, can be easy or hard. Recall Marsyas and Cronos. Ctesippus is Marsyas. He starts out as the young admirer of Cleinias; but then finds himself captivated by the ways of sophistry and fast becomes an adept. The speed with which this happens coheres with his own image of just how wisdom and virtue are to be acquired – by having them poured into yourself as into a wineskin (where “yourself” is just the wineskin). Ctesippus is right to see himself as Marsyas; he is taken over by the sophists because this is who he is. For Ctesippus persistence – being who he is – and learning are easy, just if there are sophists there to pour their sort of virtue into him.
31
This is, of course, familiar from elsewhere, notably the Theaetetus and the Sophist. I have discussed this sophistic account of truth at greater length elsewhere, see McCabe 1998. 32 It is worth noticing that the basis for the sophists’ argument, despite Socrates’ citation of Protagoras at 286c, is not a relativist one, but instead a kind of distorted realism – whatever some statement is about, it is about exactly that and nothing else, and so that thing about which it is uniquely makes it true: their episodism, therefore, is more of a metaphysical claim, where Protagoras’ is epistemological.
Protean Socrates: Mythical Figures in the Euthydemus
121
Socrates is Cronos. He presents himself as a slow and deliberate old man33 – too old to be any good at learning the cithara. The sophists think of him as a slow-top; for them, to say that Socrates is Cronos is an insult. This insult is slung at the end of the arguments about truth and contradiction: If everything is true, then no one can contradict anyone else (they can merely talk at cross-purposes); contradiction is impossible (286aff.). This too might seem to offer an easy way round the issue of teaching and learning: no one has to do anything much. And so Socrates observes. If truth is as the sophists suggest, then no one can make any mistakes; and this is inconsistent with the sophists’ claim to teach. The sophists famously respond that what they said earlier has nothing to do with it; they insist, that is to say, that inconsistency does not matter (or else that there is no such relation). Are they right? If the sophists give us an easy way to lead our lives – infallible just because everything is true – why should we reject their services? What price the hard road of Socratic learning over the primrose path of sophistry?
Mythical Figures and the Leaders of Lives The choice boils down to who is who; and this question is not merely academic. For it matters to show that someone persists through change, learns the truth over time and seeks consistency. We mind about hard truth, because we want to know how best to live. And it is a life that is represented by the Proteus who persists through the shape-shifting, by the serious Proteus whom Socrates seeks. The sophists, if they fall for their own arguments, cannot be this Proteus: for they survive only when there is no change at all (either they are mere episodes, or they are invariant). On the principles they invite us to believe, there is no serious, real Proteus at all: and nothing that we could reasonably call a life. So then is Socrates this Proteus either? You might be inclined to say not. Socrates, surely, is Menelaus: the person who asks the questions. The person who answers them is exactly the person whom Socrates claims not to be; for Socrates disavows being able to tell the truth, and disavows being the teacher upon whom we can base our opinions for the best life. The person who answers the questions (and, in doing so, gives us the truth) is someone whom we never meet – for all the interlocutors we encounter end up being unable to say anything worthwhile much at all.34 And yet this person is someone we may aspire to be: someone whose constancy for the truth persists throughout his changes, his circumstances and his interlocutor. However he shifts his shapes, there is someone who he is, after all.
33 At 285c Socrates concedes his own antiquity, as the ground for offering himself to the sophists as to Medea; they may cut him up and boil him, just so long as they make him better. 34 John Beversluis argues the case for the interlocutors, see Cross-Examining Socrates, Cambridge: Cambridge University Press 2000.
122
M.M. McCabe
In the protreptic to philosophy which is this dialogue (see e.g. 275a), I have suggested, we are offered a choice between the easy ways of sophistry, and the hard business of Socratic inquiry.35 The basis for the choice, and its consequences in both metaphysics and philosophical logic, is not explicit within the arguments as they are commonly demarcated. But they are represented by the mythical figures. Indeed, the postulates about persistence, change, personal identity and so forth are explicit in this dialogue only in the mythical figures. In this respect, those figures are ineliminable: because they represent the unspoken assumptions upon which a satisfactory resolution of the sophists’ view may be based.36 For recall that the arguments forced on Socrates can only be rebutted by adding some extra metaphysical assumptions. Within the immediate context of the “Cleinias dies” argument, Socrates provides us with no such assumptions. Widen the scope of what counts as relevant to the argument, however, and the metaphysical postulates appear in the mythical figures. Marsyas is a figure for an easy conception of change; and Cronos provides us with the postulate of identity over time which is vital to underpin the possibility, and the importance, of consistency. Proteus is the most significant: an ambiguous figure. Take him playfully and he is the sophists: easy truth and persistence, nothing to be made of the ethical progress of a life. Take him seriously and he persists through change and tells the truth over time: he is the paradigm for a life conducted according to reason. But notice, finally, how this trick is turned. The metaphysical postulates which allow Socrates a defence against the sophistic conclusions are mythical figures; and what they thus represent is the lives those figures live. Their appearance in the argument startles; and it provokes. Think about it this way. Suppose that, on a technicality, we can dispose of the philosophical problem of persistence, and make it trivial (either episodism is true, or persistence is absolute). Why not? The answer comes from outside the narrow purview of metaphysics; instead it comes from the ethical conception of what it is to live a life. Why does it matter that we should not call becoming wise a death? Because of the life which such a change affects. If episodism is true there is no such life, just a series of deaths. In that case, there is for me no ongoing project which transcends those deaths (because I do not survive them): the search for happiness, or for the rational skill which will provide it (the search on
35 The contrast is exemplified in the difference between Socrates’ alter ego, Cronos, and Ctesippus’ alter ego, Marsyas. For Ctesippus is searching for a quick remedy, a pour-in virtue which will make him and his life come out right. Socrates, on the other hand, is insistent on how hard it is to learn anything (the cithara, perhaps); and it is a slow process that takes place through change. 36 I have benefited greatly from extended discussion of this point with Hugh Benson. When I say that the mythical figures are ineliminable, I do not mean that the point they make could not be made in any other way; but that the point is only made in this dialogue in this way. The relation, further, between myth and protreptic resides in the fact that the use of myth to make these points is indirect, and demands the constant attention of the reader. In that sense, the myths are also ineliminable – for they are part of the strategy of indirection which Plato uses to make his reader do philosophy.
Protean Socrates: Mythical Figures in the Euthydemus
123
which this dialogue is engaged) will be vacuous. Likewise, if persistence is absolute, I am eternally changeless; and then I have no project to live my life in the best possible way – effectively I have no life at all. Likewise again, if change is easy, as Ctesippus supposes, my life cannot really be understood as my own project: Marsyas just waits for something to come along that will fill his skin. It is only if my life is construed as aspiring to the life of Proteus, or the life of Cronos, that the search for happiness can get under way. So is Socrates Proteus? Never explicitly. But the dialogue presents a series of puzzles about identification; and a series of arguments from the sophists which may cause the reader to recall the inherited view of Proteus – the wise old man of the sea. Socrates’ age, of course, is a running theme, especially in this dialogue.37 The other running theme of the dialogue is the protreptic to philosophy: and here it offers the contrast between the tricky ways of sophistry and the hard graft of working with Socrates. The sophists’ easy conception of persistence fits Proteus the shapeshifter; the Socratic method is represented by Proteus the wise teller of truths. For the Socratic sections of the dialogue make much of the search for truth by the difficult means of checking for consistency, and rejecting what fails that check38; and of setting aside the means for immediate gratification in favour of the search for wisdom itself.39 The dialogue invites us to see the ambivalence of Proteus by recalling the inherited myth; and it shows us what is at stake in the choice between the shapeshifter and the wise old man of the sea. If we are to be serious, it is the latter we must prefer: and if we do that, we can only turn to Socrates.40 King’s College London, UK
37
See, notably, 272c. Cf. 287eff. 39 Cf. 281eff. 40 I am grateful to many people for comments on versions of this paper. It was originally written for a conference at the University of Arizona; I profited tremendously for the care and the courtesy of Hugh Benson’s commentary – on that occasion and afterwards. I should also like to thank Peter Adamson, George Boys-Stones, Ursula Coope, Jonas Green, Verity Harte, Mark McPherran, Christopher Rowe, Michael Stokes, Michael Trapp and an anonymous reader for this volume for comments and suggestions, as well as to the audiences of the paper on the three occasions it has been given, in Durham, Arizona and London. 38
Aristotle on the Individuality of Self Juha Sihvola
Aristotle undeniably diverged from Plato in his view of what a human being most truly and fundamentally is. Plato, at least in many of his dialogues, held that the true self of human beings is the reason or the intellect that constitutes their soul and that is separable from their body. Aristotle, for his part, insisted that the human being is a composite of body and soul and that the soul cannot be separated from the body. Aristotle’s philosophy of self was constructed in terms of hylomorphism in which the soul of a human being is the form or the structure of the human body or the human matter, i.e., the functional organization in virtue of which human beings are able to perform their characteristic activities of life, including growth, nutrition, reproduction, perception, imagination, desire, and thinking. When interpreting human beings (as well as other living things) as hylomorphic composites, Aristotle stressed the fundamental role of matter in their constitution. With a reference to the various functions of the soul he also insisted that there is a lot more to being a human than merely intellectual activities. One might be inclined to think that this attention to the material and non-rational aspect of life would also mean Aristotle was more interested than Plato in the uniquely individual and contingent features in human life in his metaphysical psychology. In the following, I shall argue that this is only partially true. I shall suggest that Aristotle’s philosophical aims and the conceptual framework he applied directed the ways he approached the issues of individuality and accidental properties in human lives. First, Aristotle held that philosophy as far as it is a science (epist¯em¯e) is always primarily concerned with the universal and the essential, not the accidental and the contingent. By definition there is no specific branch of philosophical study concerned with the accidental characteristics of human beings or any other species. Human beings obviously have accidental properties that distinguish them from each other, but they are not in the focus of Aristotelian science. Moreover, Aristotle applied a hylomorphist framework based on the notions of form and matter in his studies of human psychology. Hylomorphism gives a strong priority to structures and functions in explaining the characteristics and activities of living things.1 1
On hylomorphism in general, see, e.g. Bernard Williams, “Hylomorphism”, Oxford Studies in Ancient Philosophy 4(1986), 189–199; Edwin Hartman, Substance, Body, and Soul, Princeton, NJ: Princeton University Press 1978. See also Martha C. Nussbaum and Amélie Rorty (eds.), Essays on Aristotle’s De Anima, Oxford: Clarendon 1992.
P. Remes, J. Sihvola (eds.) Ancient Philosophy of the Self, © Springer Science+Business Media B.V. 2008
125
126
J. Sihvola
Aristotle was thereby led to interpret the true self of each human being as essentially a substantial form or a soul that may be in a sense an individual, but that, as far as it is a form or a soul, is similar to the form or the soul of any other human being. However, Aristotle also recognized that each human soul (or the soul of any other living thing) has a distinctively individual career and life-history as far as it animates a particular human body acting in a particular environment. It might seem that in principle it would have been open for Aristotle to give quite a lot of attention to this aspect of individual careers and life-histories within his hylomorphist framework. I shall, however, show that this is not quite the case. Aristotle’s approach to individuality is also directed by his conviction, expressed in his ethical works, that the goal of human life should be understood as a realization of universal humanity. Aristotle, unlike many other ancient philosophers, pays a lot of attention to the various particular and contingent conditions under which we have to realize our humanity, but this does not imply any positive value being ascribed to the expressions of individual uniqueness. All this does not mean that Aristotle did not have resources to approach problems related to the philosophy of self or that he did not have any interest in them. On the contrary, in a few remarks in his ethical works he raises, for example, interesting questions about finding one’s true self. In my first section, I shall discuss these remarks and their role in Aristotle’s ideal good life, which focuses on the realization of the universal human nature. In my second section, I shall approach Aristotle’s philosophy of self at a more general level. I shall ask in what sense the hylomophist framework allows us to speak of individual selves of human beings.
Aristotle on Finding One’s True Self Aristotle outlines ideas and problems related to the individuality of human life in various contexts in his works. In his discussion on moral responsibility in the Nicomachean Ethics III, he distinguishes various ways in which one may be ignorant of facts related to one’s action. A human being can be ignorant of who she is, what she is doing, what or whom she is acting on, as well as with what instrument, to what end and how she is doing it. However, Aristotle thought that only mad people can be ignorant of all of these things, and he does also not seem to allow the possibility of being ignorant of the agent (agno¯esei ton prattonta), for how could it happen that a human being does not know oneself? (EN III 1, 1111a3–7). The passage makes a distinction between cases in which ordinary human beings can make mistakes, on the one hand, and cases in which such errors are normally impossible, on the other hand. As to self-knowledge, Aristotle holds it as obvious that one can recognize oneself as the subject of one’s acts, beliefs, judgments, perceptions, memories, emotions, and desires without much possibility of error. He in fact subscribes to a very strong form of immediate self-consciousness. Whenever a human being acts, or has a belief or a perception, he is more or less necessarily aware of both the fact that he is acting or having a belief or a perception and the fact that he is the
Aristotle on the Individuality of Self
127
subject of the act, belief, or perception in question.2 Aristotle seems to suggest in the EN passage that the Delphic exhortation “Know thyself” is a mere truism as regards this type of self-knowledge. However, Aristotle also pays attention in the same passage to obvious ways in which one may fail to know who one is. An extreme example is Oedipus, who was ignorant of whose son he was, which to the tragic consequences of killing his own father and having sex with his own mother. Even an ordinary human being may, in principle, quite easily ascribe mistaken predicates to oneself, and thereby have a false conception as to who one fundamentally is. There are also a couple of passages in which Aristotle distinguishes interesting cases of people who seem to be uncertain of who they really are and who could be understood as searching for or being unaware of their true selves. In the EN IX 4, Aristotle remarks that bad people who have done terrible deeds and are hated for their wickedness are “at variance with themselves” and “retreating from life and from themselves” (1166b5–26). In the EN VII 7, Aristotle mentions keen and melancholic persons as special types of incontinence. Whereas the standard form of incontinence expresses itself as an inability to stand by the conclusions of deliberation, the members of these two groups, due either to their impatience or the strength of their passions, do not deliberate at all but follow their immediate impressions (1150b19–28). It is possible to understand this lack of deliberation, which is, according to Aristotle, typical of poets and artists (see Problemata, XXX 10), as a vague and unspecified view of one’s true self. Whereas most incontinent human beings have a relatively stable conception of who they are and who only suffer from a conflict between what they aspire to be and what they are able to accomplish, the members of this last group do not recognize any ideal goal of their lives at all. They do not really know who they are and who they wish to be. One might think that the acquisition of a correct conception of oneself, in the sense of getting straight the relevant facts of one’s life and understanding one’s individual properties in the correct way is an important aspect of human moral development. Many later philosophers have thought so too. But although Aristotle recognized the problem of acquiring adequate self-knowledge, he did not give it a central role in his ethical works when discussing the development of virtues. One might expect, for example, that the idea of acquiring self-knowledge would appear in Aristotle’s discussions concerning the value of friendship in which he refers on several occasions to the idea of a friend as another self. Aristotle argues that we need friends since we can more easily look at other people’s actions than at our own and that we thereby get something important for ourselves. However, it does not seem that this something would be more information about ourselves. Only in
2 I discuss the problem of self-consciousness in Juha Sihvola, “The Problem of Consciousness in Aristotle’s Psychology”, in Sara Heinämaa, Vili Lähteenmäki and Pauliina Remes (eds.), Consciousness. From Perception to Reflection in the History of Philosophy, Dordrecht: Springer 2007, 49–66. Seriously deranged people are exceptions to this self-consciousness. There is an interesting remark in the De memoria (1, 451a9–12) about Antiphon of Oreus and other mad people who refer to other people’s imaginations and memories as their own, and so make mistakes as to who is the real subject of certain mental states.
128
J. Sihvola
the Magna Moralia, which is most probably not authentic Aristotle but written by his followers, it is explicitly said that what we get from looking at our friends’ action is important for learning something about ourselves that we would not otherwise be able to acquire (MM II 15, 1213a13–26). In other passages in which Aristotle refers to the friend as another self, he seems to assume that the virtuous person has to be aware of his own goodness before he can recognize the same characteristics in his friend.3 Why does the Aristotelian virtuous person not need so much self-inspection in the sense of finding out and clarifying the particular facts about one’s life? Although Aristotle’s conception of practical rationality stresses the importance of particularity and contingency in ethical deliberation, his ideal of a good human life is based on the idea of virtuous perfection of human capabilities that belong to the human species, not to an individual human being.4 A good human being has an objectively correct conception of himself and the goals of his life: he actualizes human virtues in such a way that he can be regarded as a good member of both his species and his polis. The subjective choices of the individual are not that important, even though one may argue that the Aristotelian ideal of the good life allows wide variation in individual specifications depending on both external conditions and subjective preferences.5 In his ethical and political writings, Aristotle subscribes to a certain type of individualism. He emphasizes the importance of promoting through political and social institutions the flourishing of each citizen as a separate individual, so that each individual citizen would be able to rationally control his life.6 The rational control of life is only possible through moral and intellectual virtues, which are of course realized in a particular context and in an appropriate way related to one’s individual characteristics and situation. Space is even allowed for purely personal choices in the Aristotelian ideal of the good life, but still, there is no special value given for the way in which these choices are made. Important value is, by contrast, seen in the individual’s ability to exercise contemplation, practical reason and moral virtues 3
When discussing friendship and self-knowledge in the EN, Aristotle states pleasure, not information, as the result from looking at a friend’s good and just action (IX 9, 1169b3–1170a4; 1170a29–b14). In a difficult passage in the EE (VII 12, 1244b24–1245b1), Aristotle’s seems to refer to selfawareness in the sense of recognizing oneself, in distinction to other people, as the subject of one’s acts and mental states, rather than in the sense of recognizing particular facts about oneself. See Richard Sorabji, Self. Ancient and Modern Insights About Individuality, Life and Death, Oxford: Clarendon 2006, 233–239; cf. Aryeh Kosman, “Aristotle on the Desirability of Friends”, Ancient Philosophy 24(2004), 135–154. 4 On Aristotle’s view of practical deliberation, see, e.g., Martha C. Nussbaum, “Discernment of Perception: An Aristotelian Conception of Private and Public Rationality”, Proceedings of the Boston Area Colloquium in Ancient Philosophy 3(1985), 151–201. 5 On pluralism and choice in Aristotelian ethics and political thought, see especially Martha C. Nussbaum, “Aristotelian Social Democracy”, in R. Bruce Douglass, Gerald M. Mara and Henry S. Richardson (eds.), Liberalism and the Good, New York and London: Routledge 1990, 203–252, especially 234–240. 6 On individualism in Aristotle’s political philosophy, see Fred D. Miller, Nature, Justice and Rights in Aristotle’s Politics, Oxford: Clarendon 1995; Martha C. Nussbaum, Fragility of Goodness. Luck and Ethics in Greek Tragedy and Philosophy, Cambridge: Cambridge University Press 1986, 343–372.
Aristotle on the Individuality of Self
129
through his own choices and thereby control and direct the contingent aspect of his existence, but, from the viewpoint of this aim, the individual’s power to express subjective preferences in a purely arbitrary way is not regarded as important or valuable. In this respect, Aristotle’s thought diverges in important ways from many more modern varieties of individualism. In modern individualism influenced by romanticism and existentialism, the particular expressions of subjective freedom are often understood as important sources of value in themselves, as the point of human life is to create a uniquely individual composition through the free choices of the will.7 Aristotle was inclined to think that human beings are all endowed with the same specifically human capabilities, if they are to qualify as human beings. These capabilities in principle enable all human beings to establish a correct conception of what it is to be a human being and what it is to live a flourishing human life. It is true that he also held that there are wide differences in the capacities of human beings to realize the objective ideal of human perfection: his notorious theories of natural slavery and gender differences between males and females are attempts to classify these differences, and so are the less notorious categories of virtue, continence, incontinence and vice. However, a single ideal of human perfection is the perspective from which all these classifications are made. Moreover, Aristotle did not only think that there is a single ideal of human perfection that is rather easy to recognize for human beings. He also subscribed to the idea that there is a strong natural inclination towards this form of perfection. The natural sociability means, among other things, that human beings have so-called natural virtues, innate dispositions that direct them towards a virtuous development. In this sense, finding one’s true self as an individual should not be too problematic for a person who has potential for a virtuous and flourishing life.8 The ultimate contents of human perfection and the aims of human life are then not central problems for Aristotle.9 He is optimistic that most human beings have more or less correct view of the human good; they do not have to put a lot of effort to find out what it is to be a human being or to what kinds of human beings they wish to be. There are in fact very few people who are strictly speaking vicious, i.e., who have a mistaken conception of the good and systematically pursue wrong goals. Most of those who are regarded as vicious are in Aristotle’s view in fact often 7 This is, according to G.W.F. Hegel, the crucial difference between ancient and modern political thought. Hegel argues (Philosophy of Right § 185), mainly with a reference to Aristotle, that the ancients do not recognize any value or validity in the expressions of arbitrary, subjective and individual will, whereas in modern societies it is just this principle of subjective freedom that is the most important element in political liberty. See John M. Cooper, “Justice and Rights in Aristotle’s Politics”, Review of Metaphysics 49(1996), 859–872. 8 On Aristotle’s doctrine of natural virtue, see, especially EN VI 13, 1144b1–9. 9 There is of course the traditional problem of the relation between contemplative and practical ideals in Aristotle’s ethics, which indeed concerns the contents of human flourishing, but it is not central for the topics discussed in this essay. Of the extensive literature on Aristotle and the role of contemplation see, e.g., John M. Cooper, “Contemplation and Happiness: A Reconsideration”, Synthese 72(1987), 187–216; Richard Kraut, Aristotle on the Human Good, Princeton, NJ: Princeton University Press 1989; Anthony Kenny, Aristotle on the Perfect Life, Oxford: Clarendon 1992.
130
J. Sihvola
only incontinent, i.e., they have a correct view of the good but who fail to pursue it because of their strong appetites or emotions. So even the incontinent, except the extreme types mentioned above, do not normally have to ask themselves who or what kind of people they fundamentally are, although they recognize a gap in themselves between what they are and what they should ideally be. In Aristotle’s theory, it is a sign of a rather serious ethical deficiency if one really has to search for his true self and ask who he fundamentally is.
Hylomorphism and the Soul Aristotle paid quite a lot of attention to the problems in the recognition of one’s individual self, but he also seemed to think that these kinds of problems are not very common among humans, since humans are by nature directed to a unified ideal of perfected humanity. Let us now, however, approach Aristotle’s philosophy of self from the viewpoint of his hylomorphist framework. The central question will be how Aristotle applied the notions of matter and form in his psychology and what kind of influence this had for his resources for approaching the problems of the self. I shall show that quite serious philosophical problems emerge, if we apply the hylomorphist framework in explaining how two individuals in the same species differ from each other and so have different selves. Aristotle introduces the notions of matter and form in the notoriously difficult middle books of the Metaphysics to clarify the notion of substance, through which he refers to the most persisting entities that remain stable when changes happen to them and to what distinguishes a thing from its environment and defines it as what a fundamentally independent particular. Aristotle argues that substances both in the sense of persisting subjects and in the sense of essences are forms or structures realized in suitable sorts of matter. Substantial forms are the sets of capacities things need in order to perform their characteristic activities. Unfortunately, the theory of substantial forms Aristotle presents in the Metaphysics VII–IX is one of the most actively discussed controversies about Aristotle’s philosophy, and there is no consensus even on the main lines of argument. The wide variety of interpretations can be divided into those according to which the substantial forms of the Metaphysics are universal species forms and those claiming that substantial forms are particular forms each of which belongs to the concrete particular whose form it exclusively is.10 10
Of the extensive literature, see, e.g., G.E.L. Owen, “Particular and General”, Proceedings of the Aristotelian Society 79(1978), 1–21; Alan Code, “Aristotle: Essence and Accident”, in Richard E. Grandy and Richard Warner (eds.), Philosophical Grounds of Rationality: Intentions, Categories, Ends, Oxford: Clarendon 1986, 411–439; Michael Loux, Primary Ousia. An Essay on Aristotle’s Metaphysics Z and H, Ithaca, NY: Cornell University Press 1991; Michael Frede and Günther Patzig, Aristoteles, “Metaphysik Z” Text, Übersetzung und Kommentar I, München: Verlag C.H. Beck 1988; Charlotte Witt, Substance and Essence in Aristotle. An Interpretation of Metaphysics VII–IX, Ithaca, NY: Cornell University Press 1989.
Aristotle on the Individuality of Self
131
There is no need to go deeper into the controversy over universals and particular forms here. From the viewpoint of the Aristotelian self, it only needs to be pointed out that neither line of interpretation gives much space for individuality at the formal level. If Aristotle’s forms are universal, it is clear that one and the same form is predicated of all the individuals falling under a certain species. But even if the forms are particulars, they are only numerically distinct from the other forms in the same kind; there would be no qualitative differences between the substantial forms of Socrates and Callias. It is true that Aristotle also uses the notion of form in a looser sense. His theory of perception, for example, is centered on the idea that the sense-organ receives the perceptual form of the perceived object without matter. This form is obviously not a substantial form, if and when it is assumed to transmit the particular perceptible properties of a particular perceptible object to the perceiver.11 It has also been argued that a different notion of form is applied in the Generation of Animals, in which Aristotle is concerned with explaining the resemblances of children with their parents.12 It is, however, questionable whether he here applies a notion of individual form that would also contain accidental characteristics of individuals (such as colors of eyes and hair) in addition to the essential ones.13 In any case, the substantial form is the most important application of the notion of form in Aristotle and its explanatory importance drew attention to species-related essential properties instead of accidental and individual ones. Aristotle uses the notion of form to separate a thing as an independent entity from its environment, but if he has to say what it is that distinguishes an individual thing from other things in the same species, he has to refer to the notion of matter. Callias and Socrates differ from each other in matter, but are the same or similar in form (Met. VII 8, 1034a5–8). If we move from pure metaphysics into Aristotle’s psychology, the doctrine of substantial forms has to be interpreted in terms of the hylomorphist account of soul and body applied in the De Anima and the Parva Naturalia. It is stated in unambiguous words at the beginning of the DA II that the soul is a substance and a form: “The soul must be a substance in the sense of the form of a natural body having life potentially within it”. (I, 412a19–21)14 Most commentators of Aristotle seem to have thought that at least souls, if not substantial forms, are particulars, not universals. It has seemed to many of them natural to understand Socrates’ soul to be not
11 On Aristotle’s theory of perception there is also a lot of recent discussion, see, e.g., Nussbaum and Rorty 1992; Stephen Everson, Aristotle on Perception, Oxford: Clarendon 1997; Victor Caston, “The Spirit and the Letter: Aristotle on Perception”, in Ricardo Salles (ed.), Metaphysics, Soul, and Ethics. Themes from the Work of Richard Sorabji, Oxford: Clarendon 2005, 245–320. 12 See David M. Balme, “Aristotle’s Biology Was Not Essentialist”, Archiv für Geschichte der Philosophie 62(1980), 1–12. 13 See Charlotte Witt, “Form, Reproduction, and Inherited Characteristics in Aristotle’s Generation of Animals”, Phronesis 30(1985), 46–57; cf. Sorabji 2006, 137–138. 14 See also Met. VII 11, 1037a5–6.
132
J. Sihvola
only numerically but also qualitatively different from Callias’ soul. However, if souls are forms, and if forms are either universal or particular, but only numerically different from other forms in the same species, it would be difficult to understand how souls within a certain species could be qualitatively different from each other. The defenders of particular souls have referred to some passages of Aristotle to support their position. There is a passage in the Met. VII 11 in which Aristotle wonders whether the soul of Socrates is Socrates and remarks that opinions on this diverge so that some people hold that Socrates is his soul, whereas others think that he is the composite of body and soul (1037a7–9). It is, however, clear that Aristotle is here only reviewing established positions on this issue but definitely not stating his own position in unconditional terms. Even if he had accepted the view that Socrates is his soul, this does not necessarily imply that he identified Socrates with his soul so that all the accidental properties of Socrates would be parts of his soul.15 Even in this case the verb “is” could easily be interpreted as “is constituted by” rather than “is identified with”. The passage would then claim that Socrates’ true self, what makes Socrates Socrates is his soul. However, this does not imply that the soul would include any of the accidental properties that could distinguish it from Callias’ soul in qualitative terms. Another interesting passage in this context is in the Met. VIII 3, in which Aristotle writes: [It is not clear] whether the term ‘animal’ refers to a soul in a body or merely a soul, for the soul is both a substance and an activity of some body. It might also be (ei¯e d’an) that the term ‘animal’ can be applied to both, but it will not be spoken of in one account but with a reference to one thing. (1043a34–37)
Aristotle seems to suggest here that the word ‘animal’ can be used of both the soul and the particular compound of soul and body, although we can see from the use of the optative in the sentence that Aristotle does not want to state his view quite unconditionally. But the ambiguity of the term ‘animal’ does not imply anything about the nature of the soul. If ‘animal’ is ambiguous, it refers to a soul in a body, which is uniquely particular, and to a soul which can be shared by or at least similar in all the members of an animal species. The soul itself is said to be both a substance and an activity of some body, corresponding to the definitions of the soul in the DA II 1. This does not yet imply that the soul as an activity should have any qualitatively distinctive features even if it is an activity of a particular body. The defenders of particular souls have also referred to the larger issues of the soul as an efficient cause of motion and the soul as a source of perceptual and intellectual activities. Aristotle undeniably states in the DA II 4 that the soul is the cause of its body in three senses of the word, as the formal and the final, but also the efficient cause (415b9–28). It is apparent that the role of the soul as the original source of local movement makes the soul the cause of motion of an individual body in a
15
Cf. Shields 2003.
Aristotle on the Individuality of Self
133
specific place and time. It might be that this role of the soul is easier to conceive if the soul of Socrates is numerically distinct from the soul of Callias, but it does not require any qualitative difference between them. Perceptual and intellectual activities are obviously activities of an individual living thing situated in a specific time and place. It does not follow that the subject of these activities should be a qualitatively individual soul. Aristotle in fact emphatically points out that it is not the soul that should be regarded as the subject of perceptual and intellectual activities but rather the human being as a composite whole: To say that it is the soul which is angry is as if we were to say that it is the soul that weaves or builds houses. It is doubtless better to avoid saying that the soul pities or learns or thinks, and rather that it is the human being who does this with his soul. (DA I 4, 408b11–15)
This important passage is often referred to as implying that Aristotle did not regard the human self as being constituted by the soul but by the concrete individual human being that is a composite of form and matter, or soul and body. It also indicates that for Aristotle the questions of whether one should ascribe particular perceptual and intellectual activities to the soul and whether one should regard the soul as their subject are intelligible possibilities to be seriously reflected upon. Although his conclusion is that it is not the soul but the composite that is really the subject of these activities, Aristotle also seems to think that the one can at least to conceive the idea of an individual soul as being the subject of particular mental states and activities. The passage, however, leaves it unclear how much and in what sense Aristotle really did ascribe individuality to the soul embodied in a human being. Let us now consider the relation between form and matter in Aristotelian hylomorphism. The soul as the substantial form or the essence of a living being is a set of those capabilities that enable the living being in question to perform its characteristic life-activities. Having these capabilities and being able to perform the corresponding activities are necessary conditions for regarding an individual living being as a member of a certain species. It might look as if they were not sufficient conditions since one also might think that it is, in principle at least, possible for some form to be realized in different kinds of matter. So having a human form or soul would not be sufficient for a thing to be a human being, since one would also need human matter. However, there is a crucial difference here between living things and non-living composites of form and matter. The forms of non-living things can be realized in various kinds of materials, but with living things, having a certain kind of soul as a substantial form necessarily implies having a corresponding body.16 Aristotle strongly emphasizes that, for example, a human soul and its capabilities can only be realized in a human body. He does not allow much flexibility over the material in which a form and a soul can be actualized. The human soul can only be situated in a human body, which consists of suitable human matter such as human flesh and bones (Met. VII 11, 1036a31–b28).
16 On flexibility allowed for non-living things see, e.g., Phys. II, 200a11–15; Met. VIII 3, 1043a31–33.
134
J. Sihvola
The dependence between the human body and the human soul also goes to the other direction. The human soul cannot occur in any other material than the human body, but the same is true conversely: the human body is not a human body if it is not structured and animated by a human soul, i.e., if it cannot perform its characteristic activities. The same is true of body parts. A body which has lost its soul and cannot perform its functions is not a body except homonymously. Likewise, an eye that cannot see is not an eye any more than an eye in a sculpture or a painting, and a hand that cannot perform its function is not really a hand (DA II 1, 412b10–24. Cf. Met. VII 10, 1035b25; 11, 1036b28–33).17 Some commentators have held that Aristotelian hylomorphism, involving the strong reciprocal dependence between the soul and the body of a living thing from each other, faces a dilemma that has problematic consequences for its philosophical applicability.18 On the one hand, the soul as a form may well be able to inhere in qualitatively different bodies, since this is not prevented by Aristotle’s denial of flexibility of material in which the soul may be realized. Although the soul of a living thing needs a body of a certain species, there can a wide variety of differences in accidental properties of bodies that possess the necessary constituents in being a body of the species in question. But if this is true, and if the differences between individuals in a certain species are reduced to differences in the accidental properties of their bodies, is there any longer space for any distinction between the souls of Socrates and Callias, for example? Are the souls universals? For many people, this looks a rather unattractive position. To be a soul of an individual would be more or less the same as to be a property of an individual. On the other hand, if individual or particular forms are introduced into order to establish the distinction between the souls of Socrates and Callias, does the soul then have sufficient independence from the particular body whose soul it is, given the strong reciprocal dependence between the body and the soul in Aristotelian hylomorphism? There is not too much evidence in Aristotle on particular substantial forms in which particularity could be understood in a strong sense (forms that are qualitatively different from other forms in the same species). It is true that the passages discussed above suggest that he regarded the idea of an individual soul as intelligible, but he did not explicitly develop his view in this direction. Let me now, however, suggest in what sense we could speak of individual souls in Aristotelian terms. The soul as the substantial form of a human being is the same or at least similar in all human beings, but it does have a uniquely particular career in each human body it animates. It is possible to distinguish Socrates’ soul from
17 See John L. Ackrill, “Aristotle’s Definitions of psuch¯e”, Proceedings of the Aristotelian Society 73(1972/73), 119–133; Christopher Shields, Order in Multiplicity. Homonymy in the Philosophy of Aristotle, Oxford: Clarendon 1999; cf. Michael Woods, “The Essence of a Human Being and the Individual Soul in Metaphysics Z and H”, in Theodor Scaltsas, David Charles and Mary Louise Gill (eds.), Unity, Identity and Explanation in Aristotle’s Metaphysics, Oxford: Clarendon 1994, 279–290. 18 See especially Williams 1986.
Aristotle on the Individuality of Self
135
Callias’ soul, when we are thinking of it as the human form in its occurrence in a particular body.19 The same or similar capabilities persist throughout the lives of them both, but the history of the soul in the individual body is a unique one. The distinction between the soul as a substantial form and the soul as occurring in the life-history of an individual body could be characterized with a reference to the notions of first and second actuality or activity, although Aristotle himself does not do this explicitly. In the famous definitions of the De Anima II (412a27–28, 412b4–5), the soul is said to the first actuality of a potentially living or organic body. The soul in this sense is a substantial form, which indicates that, within one species, it is always one and the same set of capacities that makes each member of that species potentially alive. There are no differences between the souls of Socrates and Callias in this sense. However, the soul can also be understood with a reference to the activities (so-called second actualities) through which the form has actualized in different ways and to different degrees in different individuals. It is in this secondary sense, to which Aristotle seems to refer in the Met. VIII passage quoted above (1043a34–37), that the souls can be said to be unique and individual in each of their occurrences.20 Human beings and other living things are both bodies animated by souls and souls actualized in bodies, and as such they have uniquely individual life-histories. However, Aristotle does not focus his attention on their accidental individual characteristics. This is, as we have seen, partly due to the nature of Aristotelian science in which the emphasis is always on the essential and the universal, and not the accidental and the particular. This also follows, in a more specific sense, from his hylomorphist framework, which makes it at least difficult if not impossible to interpret selfhood in terms of individual characteristics. Let me illustrate this claim with a reference to the way in which the persistence of an individual over time is explained in hylomorphist terms. Socrates remains Socrates as long as he continuously retains the human form, the essential characteristics that make him a human being capable of living. If Socrates loses any of these capabilities that constitute his substantial form, Socrates’ soul goes out of existence and he dies. What remains is no longer the same individual. Unlike Plato, Aristotle does not allow Socrates to reemerge once he has ceased of existing. If a new creature formally and qualitatively similar emerged, it would not be the same
19
Woods 1994, 289–290. I would like to thank the anonymous referee used by Springer for the suggestion of applying the distinction between first and second actualities in this context. It should be mentioned, however, that Aristotle, even though he clearly makes this distinction, nowhere uses the term “second actuality”. It was only introduced by the commentators in the late antiquity. The only occurrence for the term “first actuality” in Aristotle is the DA II 1 (412a27, b5). Cf. Myles F. Burnyeat, “De Anima II 5”, Phronesis 47(2002), 28–90. 20
136
J. Sihvola
Socrates. Whoever he would be, he would only be specifically but not numerically the same as Socrates (See GC II 11, 338b11–18).21 So, on the one hand, a continuous history of having the human form is necessary for an individual to remain one and the same, but, on the other hand, it is also sufficient for this. Aristotle is positive that Socrates’ capabilities have to be realized in a human body that consists in the relevant type of matter without any interruption, but he also allows plenty of space for variation for changes in his material constitution. Aristotle in fact assumes that the matter that composes Socrates is in a continuous movement. Despite all his metabolic changes Socrates as an individual remains one and the same. Strictly speaking, Aristotle’s theory is even more radical: there is in principle no limit to changes Socrates may undergo while remaining Socrates, except that the matter in which the Socrates’ form is actualized has to be suitable to constitute a human being. Whatever accidental properties Socrates may acquire and lose, he will remain one and the same individual. The only thing needed is a continuous history of a form realized in an appropriate but continuously changing human body. Even if it happened that the matter that today composes Callias somehow migrated to the body of Socrates and replaced all the matter there is now, it would seem that Aristotle has to call that human being Socrates on the basis of the particular historically continuous development that the composite of form and body has undergone.22 Socrates’ identity as Socrates is exclusively constituted by his form; accidental features that in fact distinguish him from Callias and others do not count much. The true self of Socrates is his species-related humanity. We can only speak of the individuality of Socrates with a reference to the particular ways his humanity, i.e., his capabilities for contemplation, practical reason and moral virtues, are actualized in his unique and non-repeatable life-history. To summarize: it might first seem that Aristotle’s hylomorphist theory, as applied in his metaphysics and psychology, provides neat and well-defined answers to some problems in the philosophy of self. Each human being is a composite of
21 Aristotle diverges from many ancient philosophers, most conspicuously from Plato, in thinking that the individual human being perishes at death without any possibility of returning to existence as the same individual. For Plato certain parts of the soul or even the soul as a whole are capable of separate existence and the idea of reincarnation is present in many of his dialogues. Aristotle also discusses the problem of an individual returning to existence after a temporal break in many occasions, but clearly denies this possibility. In the De Anima Aristotle says that no perishable thing can remain as numerically the same for ever (II 4, 415b4–7), and in the De Generatione et Corruptione he argues that perishable things can only reemerge specifically after a temporal break but not remaining numerically one and the same (II 338b11–19). The background doctrine in both contexts is the view that everything in the universe strives for eternal existence but entities in the sublunary world can only reach it as species but not as individuals. 22 There are difficult philosophical problems involved with the idea of radical material migration. They are perceptively discussed by Kit Fine, “A Puzzle Concerning Matter and Form”, in Scaltsas, Charles and Gill 1994, 13–40; cf. Mary Louise Gill, “Individuals and Individuation in Aristotle”, in Scaltsas, Charles and Gill 1994, 55–71.
Aristotle on the Individuality of Self
137
form and matter. It is just this composite, not either the form or the matter alone, that is the subject of the various mental states and activities each of us has. It is, however, the form, or the soul as it is called in psychology, defined with a reference to the capacities that a human body needs in order to live a specifically human life that distinguishes the composite from its environment and gives it its identity. In this sense, the soul as a form is the true self of each individual. The soul as a form, however, is either the same or at least similar in all individuals within each species. Although Aristotle occasionally says that individuals in a species differ in virtue of their matter (e.g., Met. VII 8, 1034a5–8), accidental material properties do not after all define the individuality of an individual. Socrates remains Socrates as long as he retains the human form, whatever changes occur in his matter. Strictly speaking, it is the unique and non-repeatable history through which the human soul goes in the individual’s life, conditioned by the accidental matter in which the soul happens to be realized, that makes Socrates different from Callias or any other human being. It is only from the viewpoint of individual life-histories that we can speak of individual souls in Aristotle’s theory. One may complain that the hylomorphist framework is not particularly well adapted for explaining the individual aspect of human life. At least it is true that explaining what makes an individual the individual it is raises serious philosophical problems that are difficult to solve within a hylomorphist theory. Aristotle may have been aware of these problems but he did pay too much attention to them, since, he was more concerned with the formal and structural explanations of reality. Helsinki Collegium for Advanced Studies, University of Helsinki, Finland
What Kind of Self Can a Greek Sceptic Have? Richard Bett
I have always had difficulty understanding talk about the self; precisely what is being referred to by the term has always seemed to me elusive. And so, in beginning work on this paper, I did what I do not generally encourage my students to do; I looked up the word in the dictionary. The Oxford English Dictionary gives the following overlapping series of definitions: “That which in a person is really and intrinsically he (in contradistinction to what is adventitious); the ego (often identified with the soul or mind as opposed to the body); a permanent subject of successive and varying states of consciousness”. Even aside from the sexism of the first definition, this set of definitions has a very archaic ring to it; it is perhaps no surprise that, although it was the 1971 edition that I consulted, the most recent of the examples of the word’s usage that followed the definitions was from 1909. For the claim that there is some group of features that constitute the real or intrinsic character, or the essence, of a person; the notion of a strong form of personal identity inhering solely in psychological, and not at all in physical, features of a person; and the suggestion that there is any such thing as “a permanent subject of successive and varying states of consciousness” – all these ideas have been the objects of severe and widespread philosophical suspicion for a long time (indeed, since long before 1909, one of their most vocal detractors being Hume). If that is what is meant by the self, it might well seem that the question whether an ancient Greek sceptic can lay claim to such a thing is of no great interest. If that is what the self is supposed to be, one might say, then of course no ancient Greek sceptic would want such a thing; but this creates no problem for sceptics in particular, since we all manage quite satisfactorily without it. Yet many scholars have felt that there is something problematic about what we could call the personhood of the Greek sceptic. Myles Burnyeat, for example, has spoken of the sceptic as committed to “a detachment from oneself”.1 Despite all that the sceptics themselves, especially Sextus Empiricus, say about how normal the sceptic’s existence is, it has often seemed that there is something not quite
1
Myles Burnyeat, “Can the Skeptic Live His Skepticism?” in Myles Burnyeat (ed.), The Skeptical Tradition, Berkeley, CA: University of California Press 1983, 117–48, at 129.
P. Remes, J. Sihvola (eds.) Ancient Philosophy of the Self, © Springer Science+Business Media B.V. 2008
139
140
R. Bett
human about the sceptic’s approach to life. In what follows, I want to try to get a better grasp of this widespread feeling, and to see how far it may be justified. And I think it is helpful to begin by focusing on a rather different definition of the self from the one we just looked at. Webster’s Collegiate Dictionary defines the self as “The union of elements (as body, emotions, thoughts, sensations) that constitute the individuality and identity of a person”. Like the previous set, this definition seems to assume a relatively robust conception of personal identity; and it assumes that the mental life of a person is a crucial element in that identity. But it does not presuppose that this identity is constituted solely by psychological as opposed to physical features; it does not posit anything like a permanent subject of states of consciousness; and it does not depend on any hard-and-fast distinction between essential and accidental features of a person. This conception of the self is certainly less clear-cut than the other one, but it is not open to the kinds of objections to which the other one was liable; to that extent, at least, it seems worth exploring. Let us call this kind of self the Webster-self, and the other kind the Oxford-self. I shall return to the Oxford-self at a couple of points later on; but from now on my main interest will be in the Webster-self. My question, then, will be what problems, if any, there may be in supposing the Greek sceptic to possess a self of this kind. Let us try to spell out some implications of the Webster definition. A person’s identity consists in a cluster of features. In order for there to be anything worth calling “the individuality and identity of a person”, these features must presumably have some stability and duration. Indeed, where Webster says “emotions, thoughts, sensations”, I would prefer to speak of dispositions or tendencies to have emotions, thoughts and sensations of particular kinds. It is hard to see how individual, occurrent emotions, thoughts, or sensations could constitute someone’s identity – unless we were to speak of someone’s identity as consisting simply in his or her mental history (perhaps along with physical history). Obviously it is in some sense correct to say that what makes person A the person that she is, is that she is the unique person who has had a certain specific (and very long) list of experiences. But the more interesting conception of a person’s identity – and the one, I think, that is bound up with this notion of the self – is that of a certain set of characteristics that make A the person that she is; and here, as I say, we are surely talking of characteristics that have some degree of stability. But nothing in the definition, as I am now understanding it, requires that these characteristics be unchangeable over time. Websterselves cannot change overnight; but we need not have the same Webster-selves for our entire lifetimes, either. In fact, on this understanding of the term “self ”, it would be very surprising if we did. What, more precisely, are the kinds of characteristics that make up this stable, yet not immutable construct that we are now calling the self? Without denying that bodily characteristics may belong on the list – for example, the fact that someone is tall and skinny or, more significantly, female or male – let us focus on what we would usually think of as the mental ones (though the distinction between physical and mental features is not absolutely clear-cut). And, before I leave the dictionary behind, let me draw attention to one other suggestive component of the Webster definition. The self, according to Webster, is a union of elements; it is a cluster of
What Kind of Self Can a Greek Sceptic Have?
141
features that are not only relatively stable, but that are somehow related to one another in a coherent way. Indeed, it is plausible to think of the stability and the interrelatedness of the features as themselves connected.2 Understood in this way, the self seems close to what we also call, in more everyday language, the personality; for this too is surely understood as a cluster of related and relatively long-term characteristics that make a person the unique and distinctive person that he or she is. Now, it seems clear that two central elements in any full-blown conception of the Webster-self, or the personality, are rationality and values. We have already spoken of thoughts and emotions, or the dispositions to have thoughts and emotions of particular kinds, as important parts of the complex cluster of features in question. That thoughts presuppose rationality is obvious; that emotions presuppose rationality is perhaps not obvious on its face, but has been argued so frequently and so cogently in recent decades that it too may be taken as a given. But emotions also presuppose values – and this has also been very widely recognized. To have emotions involves caring about things – wishing that things had happened differently, hoping that they will happen in a certain way, feeling that one has done something important and done it well, and so on. Sometimes the values involved are of the nature of personal preferences; one might care passionately about music, for example, and be profoundly disappointed at having to miss a certain concert, without necessarily feeling that music is valuable in some impersonal sense – though this is possible as well. But sometimes the values that are bound up with emotions clearly do aspire to some kind of impersonal or universal status, as in the case of guilt, remorse or pride. To have, or to be, a Webster-self is, then, among other things, to be a rational being, whose rationality is ordered or disposed in a particular way, and to have a particular collection of values, cares and concerns; these various features form a recognizable unity and have a stability over time, though not so much so as to preclude gradual (or even, in exceptional cases, sudden) changes. Now, let us turn to Greek philosophy and particularly to the sceptics. Before tackling the sceptics, though, it is worth emphasizing that, although my discussion has taken off from dictionary definitions of a modern English word, the kinds of issues that I have been wrestling with so far are by no means alien to Greek philosophy in general. The importance of a continuing personal identity – and the oddity of views that deny or eliminate it – is clearly hinted at in, for example, Plato’s Theaetetus. In the first major section of this dialogue Theaetetus proposes that knowledge is perception (151e), and Socrates immediately connects this with Protagoras’ famous claim that “a human being is measure of all things” and with an idea, attributed to a variety of people, that everything is in constant change (151e–152e). Developing this set of
2
It is not clear that there is any metaphysical connection between interrelatedness and stability. But at least in the case of human dispositions and character traits, the interrelatedness of the features in question would seem to result in a mutual reinforcement among them, making the whole cluster resistant, though certainly not impervious, to change.
142
R. Bett
ideas, Socrates lays out a theory of perception according to which every act of perception involves a different perceiver from every other; hence Socrates ill is a different person from Socrates well, and indeed Socrates perceiving something at time t is a different person from Socrates perceiving something at any time distinct from t (156a–160a). This aspect of the theory is not directly attacked, but we are clearly supposed to find something very peculiar about it. In a similar vein, when later introducing the final reductio ad absurdum of this thesis of constant change, Socrates has fun with the idea that the proponents of the theory themselves never remain the same (179e–180c). As for the notion of a self constituted by a cluster of stable and interrelated features, something close to this seems to be taken for granted in Aristotle’s ethics. First of all, eudaimonia is actually defined so as to be applicable only over the long term – not necessarily over an entire lifetime, although this is what it may sound like, and although such conceptions do exist elsewhere in Greek thought, but at least over a substantial period of a person’s life.3 Second, the long list of virtues of character, all of them involving characteristic patterns of emotion, and all of them ideally connected with each other by means of their intimate connection with practical reason, seems to add up to a paradigm case of a Webster-self. One might say that Aristotle’s goal is to depict, or to fashion, the best possible Webster-self; and, to the extent that people fail to achieve this exceptional unity through deficient virtue or deficient rationality, we can think of them as less successful Websterselves.4 At the same time, though, there is at least a hint of something more like an Oxford-self when Aristotle says, in Book X of the EN, that each person seems to be his nous (1178a2). And such conceptions are of course much more common in Plato; the immortality of the soul and the dispensability of the body, prominent in several dialogues, together clearly require that there be something that is non-bodily and that can be described as “That which in a person is really and intrinsically he” (or, for that matter, she – but of course, once the body is out of the picture, gender would seem to be out of the picture as well). These are just prominent examples; others could be cited from other authors and periods in Greek philosophy. There is nothing anachronistic, then, in raising questions
3 EN I 7, 1098a16–18. It is not entirely clear what is meant by the phrase “in a complete life” (en bi¯oi telei¯oi) that occurs in this sentence. But unless Aristotle contradicts himself, it must mean something less than an entire lifetime; for he later allows that someone may lose, and perhaps later regain, eudaimonia over the course of a lifetime (EN I 10, 1101a8–13). 4 Should we think of the attainment of a Webster-self as a matter of degree – so that some people actually possess a Webster-self to a lesser degree than others (and some people perhaps not at all)? It depends on how far we press the normative implications of the notion of a “union of elements”. If we think of a person’s “union of elements” as simply the totality of the relevant elements, whatever they are, that constitute that person, then it is trivially true that everyone has a Webster-self (to the same degree) – though there may of course be differences in, say, how integrated, or how well-rounded, one person’s self is compared with another’s. But if, as I have been suggesting, we think of a union as a collection of elements that are in some way inherently interrelated, then it is no doubt true that some people may attain such a union to a greater degree than others.
What Kind of Self Can a Greek Sceptic Have?
143
about the character of the sceptic’s self (not that, even if this were anachronistic, the issue would be immediately deprived of interest). As I noted earlier, many people have felt that the life, or the person, of a sceptic is somehow defective – that the sceptic is somehow not in a position to live a full-blown human life. And, given what I have said so far, it looks as if one way to put these worries is to say that there is a question as to whether the sceptic, if he is to be consistent, has the resources with which to construct a complete or genuine Webster-self. How much is there to this charge? In the space available, I cannot hope to survey the entire history of Greek scepticism, within which there are arguably a number of significant differences. I shall focus mainly on the best-known Greek sceptic, Sextus Empiricus, closing with a few remarks about Pyrrho. Sextus Empiricus says that the sceptic lives adoxast¯os, “without beliefs”. He is an expert at assembling opposing arguments and impressions on all kinds of topics and placing them together so that they have the feature of isostheneia, “equal strength”; as a result of this, he suspends judgement on all the topics in question, and this suspension of judgement in turn results in ataraxia, “freedom from disturbance”. Now, there is a much debated question as to the precise extent of this withdrawal of belief; does it apply to all beliefs whatever, including everyday beliefs such as my belief that my check book is in my brief case, or does it only apply to more general, theoretical beliefs about the nature of the world? In a well-known passage of Outlines of Pyrrhonism (1.13) Sextus addresses the issue of the senses in which the sceptic does and does not hold beliefs; but this is frustratingly unhelpful on the present issue. He says that the sceptic does not hold beliefs in the sense in which belief is “assent to some unclear object of investigation in the sciences” (trans. Annas and Barnes), and this may seem to suggest that it is only theoretical beliefs from which the sceptic withdraws. But he also says that the sceptic does hold beliefs in the loose sense of “acquiescing to some matter”; the example given is that he will not deny that he is hot or cold, when in fact he is hot or cold. The trouble is that this tells us nothing about beliefs such as my belief that my check book is in my brief case, which are not theoretical beliefs, but not a mere registering of one’s current experience either. Now, the extent of the sceptic’s beliefs is clearly not unrelated to the question of what kind of self the sceptic can be understood to have. If the sceptic really has no beliefs at all beyond those that consist in a mere registering of current experience, then his mental life – and therefore, one might think, his self – would seem to be drastically restricted; if, on the other hand, he has all the kinds of beliefs most people have except for theoretical beliefs, then the prospects for a self of more or less normal proportions seem much better. A proper treatment of the issue of the sceptic’s beliefs would certainly require a paper in itself, and I cannot directly pursue it any further. But I think that we can make some progress on our current topic by looking at the chapters of Outlines of Pyrrhonism called “On the Sceptic’s Criterion” and “What is the Sceptic’s Goal?” (1.21–30), which address some related issues in talking about the sceptic’s way of life and the reasons for his tranquillity. Acknowledging that there is a question as to how the sceptic can lead a recognizably human life, Sextus says that the sceptic does so by acting in light of the way things appear; whether things actually are as
144
R. Bett
they appear is another question, but the fact that they appear in certain ways is not up for discussion, and this is a good enough basis on which to shape a sceptical practice. He lists four major ways in which the sceptic habitually responds to appearances, which he calls “guidance of nature”, “necessity of feelings”, “handing down of laws and customs” and “teaching of skills”. About “guidance of nature” he says simply that in virtue of it, “we are naturally capable of sensation and thought”; we have certain perceptual and cognitive capacities that lead us to react in certain ways to how things appear. Examples given of “necessity of feelings” are hunger and thirst, which lead us to eat and to drink. The “handing down of laws and customs” is what leads us to, for example, “accept piety as good and impiety as bad, in terms of ordinary life”. And the “teaching of skills”, obviously enough, is what enables us to master various kinds of expertise – such as medicine, which Sextus himself practised, as did several other lesser-known sceptics. Now, this seems to offer a much more promising set of resources for a sceptical self than appeared to be available on the more extreme interpretation of the sceptic’s withdrawal from belief. “Necessity of feelings” is the exception here. For hunger and thirst to lead us to eat and drink by itself requires nothing beyond a sequence of occurrent states; no additional cluster of stable and interrelated psychological characteristics seems to be needed – in this respect there need be no difference between a sceptic and many non-human animals. But the other three all seem to assume or refer to some set of stable psychological dispositions. And, in particular, it seems clear that included in this set of stable characteristics are the two elements that I suggested are particularly important in a self, rationality and values. As we saw, the thinking capacity is one of those natural endowments that, according to Sextus, shape our ways of reacting and behaving. And this should be no surprise. For it can hardly be denied that the sceptical practice of assembling opposing arguments – a practice in which Sextus relentlessly engages in much of his surviving work – is a highly developed exercise of rationality. The same can be said of the sceptic’s response to the “equal strength” of these arguments, suspension of judgement. Admittedly, on most readings, the sceptic does not decide that suspension of judgement is rationally required; rather, suspension of judgement is simply an effect of being presented with this collection of arguments.5 But it is an effect that depends on one’s capacity to be impressed with the arguments’ “equal strength”, and this already requires rationality. Equally, it is an effect that depends on a sense that these arguments are in fact opposed to one another, and that one cannot simply accept all of them; in other words, as has often been noticed,6 the sceptic must in
5 This has recently been challenged by Casey Perin, “Pyrrhonian Scepticism and the Search for Truth”, Oxford Studies in Ancient Philosophy 30, Oxford: Oxford University Press 2006, 337– 360, especially section III. If Perin is right, then there is even less reason than I suggest to worry about the sceptic’s rationality. But see n. 10 below. 6 See, e.g., Martha C. Nussbaum, The Therapy of Desire, Princeton, NJ: Princeton University Press 1994, 307–308.
What Kind of Self Can a Greek Sceptic Have?
145
some sense adhere to one of the corner-stones of rationality, the Law of NonContradiction. As for values, Sextus makes clear in the passage we just looked at – and this is repeated elsewhere – that the sceptic in some sense accepts the values of the society in which he has been raised. This may be a depressingly conformist approach to values; but it is clear that the sceptic, as Sextus portrays him, does not lack values. At this point, then, it is hard to see why there should be any barrier to a sceptic of Sextus’ stamp possessing a Webster-self, just as everyone else does. The sceptic is a rational being, with a particular collection of values, cares and concerns. Some of these, as just mentioned, will be a product of his society; some may be products of his engagement with scepticism itself; and some may be features of his individual personality. And so far, at least, it is not clear why these features, in the person of the sceptic, should not have the same kind of unity and stability over time that they have for other people.7 The sceptic’s personal identity does not seem to be at issue. So what more needs to be said about the sceptic’s Webster-self? And why should there be felt to be a worry about the sceptic’s personhood or humanity? The answer is that there is more to be said; and part of the trouble is contained in the phrase “in some sense” that I twice made use of just now. For it is of course an important question how the sceptic can take on board all of the various characteristics and dispositions just referred to without at any point taking a stand on whether things actually are as they appear. In general terms the answer is not too hard to supply.8 How, for example, does one train to be a doctor without acquiring a set of theories about the real workings of the human body? It can only be by treating medicine as simply a series of routines to be applied in appropriate circumstances, without any theories about the underlying causes of their success; if someone has a fever, one performs routines A, B and C – they usually seem to work, and never mind why. The medical theory of the time did indeed include such approaches; on this model learning medicine, or any other skill, is like learning to ride a bicycle – it is know-how without any accompanying propositional knowledge. And how does one “accept piety as good and impiety as bad” without accepting that certain things really are the case? Again, the answer must be that one conforms to the practices of someone who believes that piety really is good and impiety bad, without actually believing this oneself. This is the point of the qualification “in terms of ordinary life” (bi¯otik¯os); one behaves in accordance with the values of one’s
7 If we think of the achievement of a Webster-self as a matter of degree (see n. 4), then of course any given sceptic may very well have less of a Webster-self (i.e., a less integrated and unified personality) than some non-sceptics. But there is no obvious reason to think that sceptics as a group will have Webster-selves to a lesser degree than non-sceptics as a group. 8 For a more detailed discussion of this question, with which I am in general agreement, see Jonathan Barnes, “The Beliefs of a Pyrrhonist”, in Myles Burnyeat and Michael Frede (eds.), The Original Sceptics: A Controversy, Indianapolis, IN: Hackett 1997, 58–91, esp. 82–88. (Originally published 1982.)
146
R. Bett
society, but without fully adopting them, in the sense of making an intellectual or emotional commitment to their correctness.9 Finally, the same must be true of the canons of rationality itself. The sceptic does indeed proceed in accordance with the Law of Non-Contradiction and other constitutive features of rationality. But it by no means follows that the sceptic would be prepared to defend these constitutive features as warranted, justified or correct; indeed, in his logical works Sextus vigorously attacks the whole idea of deduction, induction, proof, as well as numerous other central aspects of both philosophical and ordinary reasoning, with a view to inducing suspension of judgement about their validity.10 In adhering, then, to the canons of rationality, Sextus would say that he is simply following the inclinations with which his natural thinking capacity has endowed him, without any conviction (one way or the other) as to whether these canons are the right ones. Now, it is this stand-offish attitude – which applies, as we can now see, to both rationality and values – that is, I think, close to the heart of what has to many people seemed problematic about Sextus’ brand of scepticism. And although, as I mentioned, Sextus several times insists that he is the supporter of ordinary ways of thinking as against the theoretical abstractions of non-sceptical philosophers (e.g., PH 2.102), on this central issue he goes conspicuously in the opposite direction. Immediately following the chapter on the sceptic’s criterion is the chapter on the
9
It is a serious question whether this is indeed in tune with the attitudes of ordinary, nonphilosophical people (as Sextus usually seems to intend in his references to the sceptic’s affinity with “ordinary life” (e.g., PH 2.102) ). For ordinary people presumably do believe that piety really is good and impiety bad; this is not just some theoretical obsession of philosophers. Certainly Sextus thinks that ordinary people hold that things are good and bad in the nature of things (PH 1.30, to which I return shortly), and it is hard to see why piety and impiety would be an exception. So to “accept piety as good and impiety as bad, in terms of ordinary life”, cannot include adopting the same attitude towards piety and impiety as ordinary people do. And the only remaining alternative seems to be that the sceptic does what ordinary pious people do – including saying the kinds of things that one says at religious ceremonies – without having the beliefs that ordinarily go along with doing and saying those things. Julia Annas has argued for a different interpretation, drawing a distinction between the religious beliefs of ordinary people, and the theological beliefs, concerning the real existence and true nature of divinity, of philosophers; see Annas, “Ancient Scepticism and Ancient Religion” (unpublished: available online at http://www.u.arizona.edu/∼jannas/). But whatever the merits of this distinction as regards ancient pagan religious belief and practice in general, Sextus himself, in his treatment of the existence and nature of the gods, as well as of people’s beliefs about these things, seems not to admit any such distinction. I have discussed this in a chapter on Sextus Empiricus to appear in Graham Oppy and Nick Trakakis (eds.), History of Western Philosophy of Religion, Ancient Philosophy and Religion Vol. 1, Acumen, forthcoming. 10 This seems to create trouble for any reading such as Perin’s (cf. n. 5) which attributes to Sextus a robust commitment to rationality. Perin’s article very persuasively follows out the implications of Sextus’ claim to be a searcher for truth (a skeptikos in its etymological sense). However, there are other strands in this thinking that do not easily cohere with this, and that speak in favor of a commitment to rationality considerably more tenuous than this would seem to require. But even if we allow that Sextus’ commitment to rationality goes deeper than I am suggesting, this does not, as far as I can see, affect what is really the centerpiece of my argument, namely, the tenuousness of his commitment to the values to which he adheres.
What Kind of Self Can a Greek Sceptic Have?
147
sceptic’s telos or goal. This goal is said to be ataraxia, freedom from disturbance. Sextus is careful not to suggest that this is a telos in the sense intended by nonsceptical philosophers – that is, something that all humans should ultimately seek, or are by nature inclined to seek; instead, ataraxia is simply said to be what the sceptic as a matter of fact ultimately seeks.11 Now, this ataraxia is said to be a byproduct of the sceptic’s suspension of judgement; though the sceptic originally hoped to achieve ataraxia through the discovery of the truth, he in fact achieves it only when his search for the truth leads him, instead, to abandon all claims concerning how things really are. The reason for this is as follows. Someone who is not a sceptic does have beliefs about how things really are; and, in particular, he or she has multiple beliefs about what kinds of things are really, or by nature, good or bad. But someone who has beliefs of that sort is bound to be in a state of constant turmoil or obsession, desperate to get, or to hold on to, the good things and to get rid of, or to keep away, the bad things. The sceptic, by contrast, is free from all such “intense” pursuits and avoidances, as Sextus puts it; he simply does not care to anything like the same extent about what does or does not happen to him. This is not to say that he is immune to external events. Sextus acknowledges, perhaps in response to incredulity about the truly inhuman level of tranquillity attributed to Pyrrho, that the sceptic does suffer pain and other inevitable bodily woes. But even in these cases, he says, the sceptic is better off than the ordinary person. For the sceptic simply feels pain, hunger, or whatever it may be; the ordinary person feels pain, hunger, or whatever, and also has the opinion that this is a really bad condition to be in – with all the turmoil or obsession that this brings with it.12 And in this case he is quite explicit that it is ordinary people (idi¯otai, 1.30), and not just philosophers or theorists,
11 Dan Moller has recently argued that the specification of a telos cannot be easily detached from the kind of philosophical theorizing the sceptics claimed to be getting away from, see “The Pyrrhonian Skeptic’s Telos”, Ancient Philosophy 24(2004), 425–441. A similar point was pressed by the anonymous reviewer. But I remain unconvinced that the sceptic cannot simply say that ataraxia is what he wants and likes, with no further explanation or justification. It is what the sceptic chooses to aim for, but that is not the same as saying that the sceptic must regard it as in some objective sense choiceworthy; nor does the word telos itself carry any such implication, even though most other philosophers who use the term do intend this. There may indeed be a causal story about why ataraxia is the sceptic’s goal, and his earlier history as a non-sceptic seeking ataraxia on the grounds that it is really, or by nature, worth seeking, may indeed be part of this story. But this does not mean that the sceptic now has, or needs to have, grounds for this preference for ataraxia – or, for that matter, to feel embarrassment about the fact that he used to think he had grounds for it; it may, at this point, simply be a brute fact about him and his orientations. Of course, this means that anyone who does not share his brute preference has no reason to be interested in scepticism. But that fact itself is nothing with which the sceptic need be concerned – although Sextus does seem to think that, as it happens, a great many other people, especially philosophers, do share this preference, and the milieu of Hellenistic philosophy would no doubt encourage him to think this. 12 Indeed, elsewhere he even suggests that this additional opinion is more of an affliction than the pain, hunger, etc. itself (PH 3.236, Math. 11.158–60).
148
R. Bett
who are subject to this higher level of affliction, owing to their beliefs about things being really good or bad, and the intensity of feeling that this carries with it. The sceptic has no commitments to anything’s being really good or bad, and so nothing really matters to him to more than a minimal degree. So where does this leave the self? Once again, it is not that the sceptic does not have a Webster-self. He is a rational being with a cluster of values, and there is no clear reason why this should not add up to the sort of “union of elements” that constitutes a Webster-self.13 But his self is not, as we might put it, something in which he is invested in the same way as other people are invested in their selves. And this, in turn, is because the character of the sceptic’s self is itself somewhat different from that of most other people. In most cases a person’s Webster-self – an item that, as I suggested, is closely related to the personality – would seem to include, and to be shaped to a considerable extent by, what really matters to that person; “who you really are”, as we often put it, is especially bound up with your core commitments. But if the sceptic has no core commitments – and this seems to be precisely the advantage that Sextus is claiming for the sceptic – then it is natural to think of the sceptic’s self as much more lightweight, much less substantial than most other people’s selves. It may not in fact be any more subject to change than other people’s selves; it is not as if the sceptical outlook automatically renders one more susceptible to outside influences than are non-sceptics. So in one sense, at least, it need not lack stability. But the sceptic’s self, because it lacks, by design, depth of commitment to anything in particular, is itself something to which the sceptic lacks depth of commitment. The sceptic may have a set of values, namely those of the society in which he was raised, in the sense that he has a set of inclinations or preferences that shape his actions in a way that is in conformity with those values. However, as Myles Burnyeat has put it, “he does not identify with the values involved”.14 It is worth trying to specify more precisely what this means. The notion of lack of identification might be taken to suggest a kind of duality or split in the sceptic’s self: there is a body of attitudes making up what would normally be the Websterself, and then there is another self – and a peculiarly contentless self, one might think – that holds this set of attitudes at arm’s length. And if this is the case, then one might indeed question whether the sceptic had a genuine Webster-self at all; for the unity implied by the notion of a “union of elements” would seem to be spectacularly lacking.15 But if this is what the charge of “lack of identification” amounts to, then the sceptic has good reason to resist it, and so should we. The self that, in this story, holds at arm’s length the cluster of attitudes that would normally add up to the Webster-self could only be something like an Oxford-self, a “a permanent
13
See again n. 7. Burnyeat 1983, 132 (emphasis added). 15 See again n. 4; one might conceive this either as the lack of a Webster-self altogether, or as the possession of an unusually ill-integrated Webster-self. My choice of the term “genuine Websterself ” in the main text implies a preference for the former; this kind of hypothetical scenario does seem to suggest that a self worthy of the name has to be at least somewhat integrated. 14
What Kind of Self Can a Greek Sceptic Have?
149
subject of successive and varying states of consciousness”; and, as I suggested at the outset, both the sceptic and we ourselves have good reason to doubt the existence of any such thing. However, the non-identification does not have to be understood in this way. To say that the sceptic “does not identify with” the values that make up his Webster-self may simply mean that these values do not constitute core commitments. That is, the fact that he has these values, as opposed to the values of some other, quite different society, is a matter of indifference to him; as far as he is concerned, he might just as well have had the other set. This surely requires second-order attitudes as well as first-order ones, but it does not require a second self, isolated from the values in question and passing judgement on them. We may say, on this picture, that the sceptic does not treat his values as crucial to “who he really is”. But this is not to say that there is some additional entity that does constitute “who he really is”, of which these values are not part; rather, it is just to say that he does not care deeply and fundamentally about them (or about anything else) – they are not aspects of himself that he just could not contemplate doing without. And in the same way, he does not identify with, although he does adhere to, the canons of rationality. So on this second, less objectionable scenario, there is not, to repeat, a worry about the very existence of the sceptic’s Webster-self. The concern is rather about its character or quality. If you do not treat certain values as core commitments, not to be abandoned without major struggle and turmoil, then there is something peculiarly thin, and also peculiarly unanchored, about your Webster-self. If your values are such that you do not care deeply about anything, then you will also not care deeply about having the specific set of values, etc. – or in other words, the Webster-self – that you in fact have. Or better: you will not care deeply about being the specific set of values, etc. – or in other words, the Webster-self – that you in fact are; for again, unless we are to reintroduce problematic notions like the Oxford-self, there is no further entity that is “the real you” beyond this cluster of (first- and second-order) attitudes itself. And this lack of investment (as I put it earlier), both in the values themselves and in the self of which those values are an important part, is bound to strike many of us as unappealing. Are we being too hard on the sceptic? Is there perhaps something of which the sceptic would be willing to say “Yes: this is a core commitment of mine”? One might think that this would have to be true of the sceptical attitude, or the sceptical procedure, itself. But here again we have to be careful. For the sceptic, if he is being careful, is not going to make any strong claims about being committed even to the sceptical attitude or procedure. Sextus’ most succinct and pithy description of scepticism comes in a well-known sentence near the beginning of Outlines of Pyrrhonism: Scepticism is an ability to place in opposition appearances and things thought in any way whatever, from which, because of the equal strength in the objects and words that stand in opposition, we come first to suspension of judgement, and after this to freedom from disturbance. (PH 1.8)
This is a purely factual description of what the sceptic does – just as, to repeat, Sextus’ mention of ataraxia as the sceptic’s goal is couched purely in terms of what
150
R. Bett
the sceptic in fact aims for. But to describe how the sceptic actually proceeds is not necessarily to express any intellectual or ethical commitment to that procedure. And in fact, Sextus occasionally takes the trouble to make clear that he has no such commitment. In speaking of the goal, he says that the sceptic’s goal is ataraxia “up to now” (achri nyn, PH 1.25), and “up to now” and related phrases are a recurring feature of Sextus’ writing. The point is that there is no reason to assume that the situation will remain the same. As he makes clear from the outset, the sceptic simply reports how things seem to him at the time. “Up to now” the sceptics have found themselves inclined to seek ataraxia; and “up to now” they have found that a successful route to this is suspension of judgement. But they have no intention of clinging to this goal and this method at all costs. If at some point they found themselves more inclined to seek something other than ataraxia, then that is what they would seek; or if (supposing ataraxia was still their goal) they found that some other method – Prozac, for example – was more effective in delivering ataraxia than the sceptical method, they would switch to that. These changes of allegiance would come about without any soul-searching or angst; they would be purely a result of how things struck the sceptic at different times. And so, if one were to ask Sextus “Is the sceptical attitude or method itself a core commitment of yours?” his response would have to be “Certainly not; there is no set of attitudes or activities that constitutes a ‘core commitment’ for me. Scepticism and its result, ataraxia, are just what I find myself attracted to right now.” But to say this seems to be tantamount to saying that the sceptic regards his Webster-self as dispensable in its current form; and that, in turn, bespeaks a Webster-self, and an attitude towards the Webster-self (and these two are, as I hope has become clear, closely connected), that most people would regard as profoundly disorienting. Again, the changes just mentioned might very well not actually occur. What is important is the sceptic’s attitude towards their possibility. It is, of course, another question whether Sextus actually lives up to (some might say, “down to”) what his outlook seems officially to commit him to. For, as has often been noticed, this terminally open-minded attitude seems to be belied by the elaborate edifices of argumentation that Sextus and his predecessors have erected in the service of scepticism itself. Sextus is as much an active reasoner as any of his non-sceptical opponents.16 He argues tirelessly against the Stoics, the Epicureans and others, with a view to bringing about the desired situation of “equal strength” among the various positions on offer on any given topic; anyone who has read through the entirety of, for example, the two long books that comprise Against the Logicians is hardly left with the impression of someone not committed to what he is doing. Again, the various sets of sceptical Modes, which Sextus sketches in Book I of Outlines of Pyrrhonism, are pre-arranged means by which the sceptic is to achieve, but then also to maintain, the attitude of suspension of judgement; to devise and employ these Modes would seem to be a mark of someone who cares about keeping the sceptical outlook intact in himself, even when it shows signs of
16
I thank Oliver Thorndike for impressing this point on me.
What Kind of Self Can a Greek Sceptic Have?
151
waning. It may be, then, that Sextus does have more in the way of commitments than he would, or in consistency should, admit; perhaps he does, in Burnyeat’s phrase, “identify with the values involved” in scepticism itself to a greater degree than his official position would allow. And if this is the case, then perhaps his self is not as thin, and his attitude towards it not as insouciant, as appeared to be the case just now. If we stick with the official story, though, the verdict seems to be as follows. It is certainly the case, as we saw earlier, that the sceptic does have values and does exhibit rationality; and this may give us reason to believe that there is indeed a cluster of psychological elements, which may very well be both unified and of considerable duration, that could be said to “constitute the individuality and identity” of any given sceptic. However, this seems to give the sceptic a Webster-self of only a rather etiolated kind. For these elements – again, if the sceptic is being fully consistent – must all ultimately be regarded as contingent and dispensable. What is lacking is any anchoring of these elements around a set of core commitments – “core” not in the sense of being absolutely unchangeable, but in the sense of what might over some extended period of a person’s life be claimed, either by that very person or by others, to matter fundamentally to that person. There is, of course, no reason why the sceptic should regard this as an objection. “I am not interested in core commitments”, he might say; “those are just what lead to the ‘intense’ attitudes I am trying to get away from. Tranquillity is what I am after, and if a robust self with depth of commitment is an obstacle to that, so much the worse for the robust self.” However, to repeat, there is a widespread sense among non-sceptics that the sceptic’s existence is missing something important; and what it is missing, as we can now see, is a Webster-self of the robust character I have tried to describe. It may be that such a pared-down self is not even possible; as I just observed, Sextus seems in practice to display more significant commitments than he officially admits. But regardless of that, the sceptic’s life is frequently felt, at least by us today, to be a thoroughly unattractive one; and the thinness of the sceptic’s self is, I suggest, a major part of the reason for this. And, although it may be tempting to connect our leanings towards a more robust self with Romantic notions of authenticity, I suspect that the sense of unattractiveness was no less common, and not so different in kind, in the ancient world. The objection regularly raised against sceptics was that a sceptical life was humanly impossible. But whether or not people found these charges convincing – for the sceptics had plenty to say in response to them – it looks as if they generally saw the sceptical life as undesirable, at least for any normal human being; certainly scepticism never achieved anything like the popularity enjoyed by either Stoicism or Epicureanism. It is interesting, in this connection, that being a normal human being is precisely what Pyrrho, the supposed original sceptic, appears to have adopted a conscious policy of avoiding.17 We are told by two different sources that Pyrrho proposed as
17 In these remarks on Pyrrho I am heavily indebted to James Warren, Epicurus and Democritean Ethics: An Archaeology of Ataraxia, Cambridge: Cambridge University Press 2002, ch. 4.
152
R. Bett
an ideal “stripping off the human being” (ekdynai ton anthr¯opon).18 In both sources this remark follows an anecdote about Pyrrho being scared by a dog, and departing from his usually tranquil demeanor; in explaining himself, Pyrrho says that “stripping off the human being” is difficult. One of these sources then has Pyrrho elaborate by saying that nevertheless “one should struggle against things as much as possible in one’s deeds, and if not that, in one’s words” (D.L. 9.66); presumably what this means is that one should struggle against the normal, engaged human reactions to things. One might say that being scared by a dog is more an instinctive reaction than any indication of the kinds of cares and concerns that, as I have been arguing, give robustness to a Webster-self. However, the general picture of Pyrrho’s lifestyle and attitude that emerges from the entire series of anecdotes about him (in Diogenes Laertius and elsewhere) is of someone who really has shed all such cares and concerns to an extraordinary degree; it is not implausible, therefore, to see the ideal of “stripping off the human” as including deliberately ridding oneself of the kind of robust self that, as we have seen, the sceptic of Sextus’ model also seems to lack. Sextus’ sceptic seems to differ from Pyrrho in his adherence to convention; he follows prevailing laws and customs, whereas Pyrrho seems to have been radically unconventional. But both of these stances are an expression of a similar kind of withdrawal from any of what I have been calling “core commitments”. This point is reinforced by another anecdote. The terrifying dog is not the only animal to find its way into the sources about Pyrrho; pigs also appear several times. Pyrrho washes a pig and takes a pig to market (D.L. 9.66), thereby illustrating his indifference to conventional ideas about the activities appropriate to different social classes. But the pig I am particularly interested in is the one that was on board with Pyrrho on a ship in a storm. According to this anecdote, all the other passengers were frightened. But on this occasion Pyrrho did keep his calm, and pointed to a pig who continued eating its food as if nothing was wrong. This is the state of ataraxia, Pyrrho remarked, that the wise person should maintain; in an exact reversal of Mill’s remark about a human being dissatisfied versus a pig satisfied,19 the untroubled pig is a model for us all to emulate. “Stripping off the human”, then, is in certain respects a striving for the sub-human, as we would normally call it.20 But of course, Pyrrho would object to the prefix “sub-”. His whole point is that there is nothing inferior about the condition he is recommending, including its dismantling of the robust self – on the contrary, this is the most desirable state to be in.
18
Diogenes Laertius 9.66; Aristocles in Eusebius, PE 14.18.26. John Stuart Mill, Utilitarianism, in Alan Ryan (ed.), John Stuart Mill and Jeremy Bentham: Utilitarianism and Other Essays, New York: Penguin 1987, 281. 20 Pyrrho is also reported to have admired Homer because of his comparisons of humans with wasps, flies and birds, as well as leaves (Diogenes Laertius 9.67; cf. 9.71, where Homer is listed as a proto-sceptic). But here the emphasis seems to be on the frailty and vanity of human beings; there is no clear sign of a recommendation on Pyrrho’s part to emulate these other species. 19
What Kind of Self Can a Greek Sceptic Have?
153
Consistent with Pyrrho’s recommendation of “stripping off the human”, but pointing in the opposite direction, is the description of Pyrrho, by his disciple Timon of Phlius, as god-like in his tranquility – specifically, as like the sun-god whose circular journey is absolutely regular and imperturbable.21 But of course gods – or at least, gods as Greek philosophers tend to portray them (Homer is another matter) – do not have robust Webster-selves either. That is, gods do not have a cluster of cares, concerns and core commitments that make them unique and distinctive. Rather, the Greek philosophers’ gods are entirely trouble-free; they are also, relatedly, entirely unaffected by anything that happens in the world – even when they exert themselves on the world’s behalf, as not all of them do. There may be occasional exceptions to this picture; but it seems to me to cover all the conceptions of god offered by the major Greek philosophers, starting with Xenophanes and stretching through Plato and Aristotle to the Stoics and Epicureans and then to the neo-Platonists. If such gods have selves of any kind, it is surely Oxford-selves; whatever may be the case with humans, the notion of a permanent subject of consciousness that constitutes the essence of someone seems quite appropriate to these divine beings. It is no accident that in the same context in which Aristotle speaks of nous as what each person really is – which is also reminiscent, as we observed, of the Oxford-self – he also speaks of the importance of striving to be as divine as possible; and similar points might be made about the homoi¯osis the¯oi, “likeness to god”, that appears as an ideal in many other Greek philosophers. Timon’s praise of Pyrrho as god-like is clearly appealing to this way of conceiving of the divine. In doing so, then, he is perhaps hinting at a form of Oxford-self as a goal to be strived for (though not, of course, as what we all are anyway). But at any rate, this picture of Pyrrho is just as much a rejection of the robust Webster-self as is the image of the pig or the advice to “strip off the human”. Neither Sextus nor any other Greek sceptic follows Timon in his suggestion of the sceptic’s existence as divine; nor, therefore, is there any hint in their ideas of an Oxford-self as what the sceptic should aspire to. But leaving that aside, Sextus and Pyrrho seem, on this issue, to be very much on the same wavelength. A number of scholars, including myself, have argued for interpretations of Pyrrho’s thought that create a significant philosophical distance, as well as a historical distance, between Pyrrho and Sextus.22 But as regards their striving for a condition in which a key
21 A seven-line fragment of Timon’s Indalmoi can be assembled from overlapping quotations: Diogenes Laertius 9.65 (which includes the information about the source), Sextus, Math. 11.1 and 1.305 (which includes the specific lines referred to in the main text). The complete fragment appears as text 2D in A.A. Long and D.N. Sedley, The Hellenistic Philosophers: Translations of the Principal Sources with Philosophical Commentary, Cambridge: Cambridge University Press 1987, and as text 61 in F. Decleva Caizzi, Pirrone: Testimonianze, Naples: Bibliopolis 1981. 22 Richard Bett, Pyrrho, his Antecedents and his Legacy, Oxford: Clarendon Press 2000. See also, e.g., Decleva Caizzi, Pirrone; Long and Sedley, The Hellenistic Philosophers, section 1.
154
R. Bett
aspect of what makes up the Webster-self of any full-blooded human being is discarded, there seems to be little or no difference between them.23 Johns Hopkins University, USA
23 In addition to the 2003 Helsinki conference on the self, a version of the paper was presented at Philosophy in Assos (Turkey) in July 2004. I thank the audience at both conferences for their helpful comments; the paper has benefited especially from remarks by Richard Sorabji, Troels Engberg-Pedersen, Gisela Striker, Julia Annas, Charlotte Witt, Sara Conley, Lucas Thorpe, and Roberto Polito. Finally, I thank the anonymous reviewer for comments that prompted a number of final improvements.
Inwardness and Infinity of Selfhood: From Plotinus to Augustine Pauliina Remes
Several scholars have drawn attention to the fact that to think of the self in terms of inner space or an inner realm is not universal. It is not an inevitable part of our selfdescription like perception or mortality, but can be traced back in history, especially to early modern philosophy. While ancient philosophers sometimes talked of an “inner man” or, rather, “human being within” (Plato, Republic IX, 589a),1 this was not the primary means to describe the nature of subjects or agents, nor did they underline the inner nature and privacy of one’s own mental functioning. It has been further argued that Augustine and especially his discussions on memory present the first real steps towards the notion of an inner self.2 A relatively recent study by Phillip Cary, entitled Augustine’s Invention of the Inner Self. The Legacy of the Christian Platonist,3 concentrates on this development in Augustine. As the title already indicates, Cary claims that the inner self is an Augustinian invention, but he takes also into account Augustine’s deep Platonism or, to be more precise, his Neoplatonism. The purpose of this article is to offer an alternative reading of the relation between Augustine and Plotinus on this issue. I will argue that the Neoplatonic influences are even deeper than Cary allows, and that most novelties attributed to Augustine are either explicit or implicit in Plotinus.4 However, it is not
1 Cf. Charmides 160d; Alcibiades 132d–133c; Theatetus 189e; Philebus 8e–40c; Sophist 263e. And as Holger Thesleff pointed out to me, the ancients considered the seat of emotions to be the k¯er, the heart within. 2 Charles Taylor, Sources of the Self. The Making of Modern Identity, Cambridge, MA: Harvard University Press 1989, at 127–142. 3 Oxford: Oxford University Press 2000. 4 In Plotinus-scholarship, it is widely held that Plotinus in some sense originated the notion of self as distinct from that of the soul. Inwardness and privacy of this self is often rejected. The kind of self-knowledge that opens up in the inward turn does not reveal a primarily particular person. Cf. e.g. Gerard O’Daly, Plotinus’ Philosophy of the Self, Shannon: Irish University Press 1973, 73–74: man is himself the intelligible universe; Lloyd P. Gerson, “Introspection, Self-Reflexivity, and the Essence of Thinking”, in John J. Cleary (ed.), The Perennial Tradition of Neoplatonism, Leuven: Leuven University Press 1997a, 153–173: Intellect is primarily oriented to the good or the One and as a result it cognises all Forms, which in turn produces the result that it cognises itself; Gerard
P. Remes, J. Sihvola (eds.) Ancient Philosophy of the Self, © Springer Science+Business Media B.V. 2008
155
156
P. Remes
my intention to devalue Augustine’s originality.5 Towards the end of the paper I will propose some suggestions as to where, exactly, a new shift of emphasis happens when we move from Plotinus to the Augustine of the Confessions.
The Privatisation of the Mind Cary starts his exposition with an interesting remark, namely that there are varieties of inwardness, of which the idea of the self as a private inner space or world is one type.6 Philosophically, both the ideas of mind as space and the idea of it as private were under heavy criticism for the whole of the 20th century. The shared nature of language and in general social conditioning and cultural influences have gained much deserved attention. Today, the self is often seen as developing in a culture rather than as a fixed subject outside the world. However, it seems unlikely that the idea of privacy can altogether be abandoned. The mere fact that we can conceal things from the look of others seems to indicate that there is, somewhere within the self, a dimension that is our very own: “what I am inwardly (quid ipse intus sim), where they [other people] cannot pierce with eye or ear or mind (Aug. Conf. 10.3.4). Historically, this dimension has sometimes been seen as a whole dimension of being of its own, a private space that, as it were, contains the cognitions, experiences and feelings of a particular person. According to Cary, this understanding of mind and self as private, inner space is Augustinian in origin. Let us first review the ambitious outline of historical development Cary draws. It is focused on the following question: What happens, or ideally should happen, when the soul looks at the world? Cary distinguishes four main kinds of answers that present the crucial steps taken in the development:
Verbeke, “Individual Consciousness in Neoplatonism”, in John J. Cleary (ed.) The Perennial Tradition of Neoplatonism, Leuven: Leuven University Press 1997, 135–152: self-consciousness is at once knowledge of the divine principle and of cosmic reality as a coherent totality animated by the divine Spirit. For an argument that self-knowledge does nonetheless reveal a (singular) subject, cf. Pauliina Remes, Plotinus on Self. The Philosophy of the ‘We’, Cambridge: Cambridge University Press 2007a, ch. III.2. 5 There is also a slightly different but connected discussion on Augustine’s significance, namely the origin of the reasoning from what each mind can know about itself to the idea that minds are non-corporeal entities, which Gareth B. Matthews (“Internalist Reasoning in Augustine for MindBody Dualism”, in John P. Wright and Paul Potter (eds.), Psyche and Soma. Physicians and Metaphysicians on the Mind-Body Problem from Antiquity to Enlightement, Oxford: Clarendon 2000, 133–145) attributes to Augustine. Although Plotinus holds that (a) the phenomenon of unified consciousness calls for an immaterial soul; (b) knowing that one is a thinker happens by thinking, his reasoning does not amount to an internalist argument of the sort Matthews finds in Augustine. 6 Phillip Cary, Augustine’s Invention of the Inner Self. The Legacy of the Christian Platonist, Oxford: Oxford University Press 2000, 3.
Inwardness and Infinity of Selfhood: From Plotinus to Augustine
157
1. Plato: the mind’s eye sees the eternal and pure ideas. This is an intellectual vision that is shared.7 2. Plotinus: the mind’s eye sees the eternal and pure ideas. This does not happen primarily by gazing out towards the world, but ultimately by directing the gaze towards the inner, to the ideas within the mind or soul. The resulting intellectual vision is shared (because the objects are universal or shared). 3. Augustine: the soul looks first inside, into its own mind’s space, and only then up towards God. As such, the soul cannot attain God. It can only approach God through its memory and inner space where God and Christ are present. This inner space is private. 4. Locke: there is solely one’s own inner space, which is private. The subject sees only private images of the things that are outside. In the course of history, the inner realm shrinks step by step. Divinity and reality are progressively externalized from it. This is a process of privatization, and it leads to the Western understanding of the self as an inner and private space. It is also the very process that enhances the problem of mind’s relation to the world and creates the problem of other minds. The mind that in antiquity has a shared and universal side and is by its very nature linked to the essential features of the world, becomes cut off from the world, and hence private and potentially inaccessible to others. Cary’s emphasis is on the leap from Plotinus to Augustine, in which he detects the most significant change. The first third of the book focuses on Ciceronian8 and especially Neoplatonic influences, thus acknowledging Augustine’s background. Current consensus among scholars is that Augustine had access to several books of the Enneads. After abandoning (Manichean) materialism, Augustine became a Platonist, among other things, in believing human nature to be essentially double: corporeal body and incorporeal soul. Furthermore, Augustine seems to follow the “books of Platonists” (Conf. 7.9.13) in arguing that the reason human beings are not at all times in touch with their better nature is in the embodiment and in our interest in the sensible realm. For both Plotinus and Augustine, the way to lead the soul beyond its fascination with the senses is to turn its attention inward, and thereby to get a reflective grasp of its own incorporeal nature. The description of this incorporeal nature, as Cary argues, differs in the two thinkers. Where Plotinus emphasizes the intellect (nous) and its intelligible objects, Augustine talks not only of divine illumination but of memory as a place where God and Christ are present but which is the mind and the self: “And this is mind, this is myself” (et hoc animus est, et hoc ego ipse sum; 10.17.26).
7 The point about universalism and an ancient, especially Stoic, emphasis on the cosmic dimension is also argued by, among others, Pierre Hadot, Philosophy as a Way of Life, Oxford & Cambridge, MA: Blackwell 1995, ch. 7. 8 The Ciceronian influence pointed out by Cary (2000, 83–85): soul turning away from the bodily (Tusc. 1.75), the nature and power of the soul unique, celestial and divine (Tusc. 1.66), eyes do not see but the soul (Tusc. 1.46).
158
P. Remes
Plotinus’ inward turn has its own background in, among others, Plato, Aristotle and Alexander of Aphrodisias. The doctrine of noetic objects within every human soul is a version of Platonic recollection (anamn¯esis). Aristotle’s claim that in thought and knowledge, the mind becomes identical with the object known, with its form, is also of utmost importance to Plotinus’ theory.9 In thinking about an object in the world, the intelligibility of the form becomes actualised in one’s mind. The same form is thus present both in the world and in the soul, but in a different mode. Another background is the no¯esis no¯ese¯os of Aristotle’s God. In Metaphysics, God is said to think of itself, of intelligibility, and thereby becoming that intelligibility. (Met. XII 7, 1072b20–25; XII 9, 1074b34–35.) Plotinus’ synthesis is to treat only the perfect knower, nous and its objects as identical, but this ideal thinker resides in every human soul.10 The step Plotinus takes in positing the intelligibles within the soul and thus placing the perfect divine knower, a kind of god, within each and every human being has twofold significance to the idea of inner self. First, and conducive to the development of the inner self: to gaze at the intelligibles becomes a matter of an inward turn (eis to eis¯o; epistrephein pros to eis¯o; e.g. I.6 [1] 8.4; V.8 [9] 11.12–13; VI.9 [31] 7.18).11 In order to attain knowledge and to live a life in accordance with it, with one’s better nature, the person must direct itself towards the inner. Second, in Plotinus’ picture human nature is still understood as essentially divine and the soul as immutable. This, Cary argues, makes it radically different from a modern private self that has its origin in Augustine. Augustine retains the inward turn as the means to find one’s true self and God (for the inward turn, self entering into itself, Conf. 10.8.15), but as a good Christian he separates God from the soul, the Creator from the created, thus generating an inner space, an understanding of the soul as its own, private dimension. (E.g. Conf. 7.10.16.) To the question “Who am I?” Augustine replies that I am a human being that consists of body and soul. Of these parts, the body belongs to the external things which are further away from God, but even the internal part closer to God, the soul, is not a god itself: The inner I knows this – I, I the mind through the sense perception of my body (ego interior cognovi haec, ego, ego animus per sensum corporis mei). I asked the mass of sun about my God, and it replied to me: “It is not I, but he made me.” (Conf. 10.6.9)
9
In teaching and learning, for instance, there is only one activity taking place, although teaching and learning are not of identical essence. Numerically, there is, however, only one activity. Cf. Phys. III 3; DA III 4; Met. XII 7, 9; cf. Richard Sorabji, Time, Creation and the Continuum. Theories in Antiquity and the Early Middle Ages, London: Duckworth 1983, on Arist. 144–146; on Plotinus, e.g. 153. 10 For the full significance of this theory, cf. e.g. Lloyd P. Gerson, “Epistroph¯e pros heauton: History and Meaning”, in Documenti e studi sulla tradizione filosofica medievale, 8(1997b), 1–32; “Introspection, Self-Reflexivity, and the Essence of Thinking”, in John J. Cleary (ed.), The Perennial Tradition of Neoplatonism, Leuven: Leuven University Press 1997a, 153–173. 11 Cary (2000, 47) also points out instances of the language of inwardness in the New Testament, especially the “inner man” in the Pauline Epistles: Romans 7:22; II Corinthians 4:16; Ephesians 3:16.
Inwardness and Infinity of Selfhood: From Plotinus to Augustine
159
In short, so the story goes, the inner self is generated by Augustine’s separation of knowledge and the divine from the soul. This move results in a privatisation of human mind, visible in his discussion of memory in the 10th book of the Confessions. For Augustine, memory not only denotes the human capacity to retain experiences and cognitive contents, but the mind in general. This mind is depicted as an inner space where Augustine walks, and where he meets his memories, his plans for the future, his self as well as God (Conf. 10.8.12–15). Only the subject himself can see directly inside this space and reveal it to others, if he so chooses. When Augustine turns towards truth, he turns first towards this private inner space, and only then up towards God who, although present in the soul, is ultimately separate from it. Cary’s main argument for the Augustinian origin of inner self is this double move, first in and then up. It is at the heart of what became the early modern and modern inner self.12 A more subtle Augustinian novelty Cary that pins down is in the metaphorical language. The new understanding of mind as something like space is detectable in spatial rather than bodily metaphors. Although already Plato in the Theaetetus uses the analogy of aviary for the human mind (197b–200d), Cary is right that the metaphor of a wax block (also from the Theaetetus, at 191c–196c) is more common and influential in antiquity, and that Plotinus still sees in his inward vision a statue rather than an inner dimension or space. For Augustine, the soul has a magnitude, quantitas, memory is “vast palaces” or “a vast hall”, “a huge cavern”, and even “a vast and infinite profundity” (penetrale amplum et infinitum, Conf. 10.8.13–15).13 The alternative reading offered here reverses the order of looking at the two philosophers. Where Cary interprets Plotinus from the point of view of Augustine and early Christian thought, this article will look at Augustine from the point of view of Plotinus and ancient philosophy. The interpretation does not so much challenge Cary’s sketch of the development as it does, building on it, target the details of what is and is not Neoplatonic in the new account (I). As will hopefully become clear in the course of the paper, this is not merely to quibble, nor is this an exercise to redeem Plotinus, but it is for the more important purpose of seeing correctly what steps are needed in order to come up with a notion of inner self that starts to look like the early modern self, and which still, in spite of all the criticisms, includes features that many people do think crucial for self-descriptions. Moreover, once the Neoplatonic influence is correctly analysed, the Augustinian inventions may also surface in a new and brighter light.
12 In this he seems to follow Étienne Gilson, The Christian Philosophy of Saint Augustine, New York: Knopf 1960, at 20. 13 The spatial metaphors are discussed already by Gerard O’Daly, “Remembering and Forgetting in Augustine, Confessions X”, in Anselm Haverkamp and Renate Lachmann (eds.), Memoria. Vergessern und Erinnern, Poetik und Hermeneutik XV, München: Wilhelm Fink Verlag 1993, 31–46.
160
P. Remes
What Does the Neoplatonic Inward Turn Reveal? Let us first focus on and assess the following four points, important for the theory propounded: 1. 2. 3. 4.
The shift from looking merely inwards into looking first in and then up The Stoic influences The idea of soul’s universal and shared origin in Plotinus The notion of true self in Plotinus (which also sums up aspects from the previous points)
1. We have seen that both Plotinus and Augustine, inspired by such Platonism as we find, for example, in the Phaedo, demand turning our attention from the body to the soul, and from soul’s lowest faculties to its highest and better nature. Cary claims that for Plotinus, the inner space of the soul is nothing other than the intelligible world which is the divine mind, and not a thing of its own. Plotinus’ turn would thus be just a turn inwards, away from the body and to the intelligible. The second movement upwards would be redundant since the soul one turns to has a highest and essential part which is identical with immutable truth.14 Augustine, Cary argues, functions with an altogether different conception of mind and one that is not shared but private. Indeed, Augustine does not accept the Aristotelian notion of mind being what it knows, thus creating space between consciousness and truth.15 Undeniably, there are certain peculiarities, noted by several scholars, that mark ancient philosophy. It has been pointed out, for instance, that in antiquity reasoning and knowing meant slotting oneself into reasons and knowledge already present in the universe, “having one’s life shaped by a pre-existent rational order”.16 Personhood, many argue, was understood as communal and tied to the cosmic and divine, rather than private or individual.17 But while the move to place truth outside human mind is of crucial importance, the reading overlooks the complexity of Plotinus’ – and in general the ancient’s – understanding of the soul and its functions. It treats a part of the soul which is a paradigm and a perhaps unattainable ideal as a whole picture of the human mind. If we follow its implications, it runs the risk of rendering each and every soul in the world as entirely divine. The reading does not pause to explain why it is that human beings are not all infallible thinkers. Cary overlooks the fact that Plotinus, too, rejects the view that mind simply becomes the objects of knowledge. Only the paradigmatic knower inside our soul
14
Cary 2000, e.g. 101. Cary 2000, e.g. 30, 42. 16 Taylor 1989, 124; Timothy J. Reiss, Mirages of the Selfe. Patterns of Personhood in Ancient and Early Modern Europe, Palo Alto,CA: Stanford University Press, 2003, e.g. the introduction 2–3; Hadot 1995. 17 Christopher Gill, Personality in Greek Epic, Tragedy and Philosophy, Oxford: Oxford University Press 1996, esp. the introduction; Reiss 2003, the introduction. 15
Inwardness and Infinity of Selfhood: From Plotinus to Augustine
161
is capable of such identity with being, while ordinary consciousness functions differently. This calls for the following distinction18: Nous
Dianoia
– Knowledge – Eternal/non-temporal – Objects: internal, the forms – Unmediated, direct
– Opinion (fallible) – Temporal, moving in time (e.g. VI.1 [42] 4.7–19) – Objects: external, both the sensible and the forms – Mediated by images (eidola), imprints of phantasia, or by concepts (e.g. V.3 [49] 2; VI.5 [23] 7) – Inferential – Particular to one soul in one body
– Non-inferential (V.5 [32] 1.38–41) – Connected to the hypostasis Intellect and same in form and content with it, yet perhaps retaining somehow some individuality (IV.3 [27] 5.6–9) – Its own “higher” consciousness
– Everyday consciousness19
It is not the case that when the self turns her gaze inside, she invariably sees the intelligible. What the self sees are the imprints of the sensible world, the often emotional images that phantasia has formed of its encounters with the world, as well as the fallible propositions through which she interprets the world and strives towards knowledge. Only for the rare moments that the philosopher comes to realise itself as an intellect does his “lower” consciousness become close to the unity with the intelligible objects. The inward turn is thus not enough: reaching truth and the divine requires a lengthy process in which one’s gaze is further directed towards the highest parts of human nature. This is nicely visible in Plotinus’ famous sculpting metaphor of looking within. What we immediately see inside our soul may not be beautiful, but it is our task to act as sculptors of our selves: to cut away all that is excessive, to straighten what is crooked and to illuminate all that is overcast (I.6 [1] 9). Nous may be a part of our nature, but it is largely unknown to and hidden from our everyday consciousness and thought.20 And even during the rare moments of wisdom, embodied consciousness only mirrors the objects of knowledge with which the higher self, the Intellect, is identical. Interestingly, Plotinus also claims that disembodied souls are such that for them, there is nothing “hidden or fabricated, but before one speaks to another, that other has seen and recognised” (IV.3 [27] 18.22–24). This feature is said to resemble the 18
For the differences in general, Eyjólfur K. Emilsson, “Discursive and Non-Discursive Thought”, in H. Fossheim et al. (eds.), Non-Conceptual Aspects of Experience, Oslo: Unipub 2003, 47–66; John Bussanich, “Non-Discursive Though in Plotinus and Proclus”, Documenti e studi sulla tradizione filosofica medievale 8(1997), 191–210. 19 For consciousness in Plotinus, cf. Remes 2007a, ch. 2; Edward W. Warren, “Consciousness in Plotinus”, Phronesis 9(1964), 83–97. 20 Let it also be noted that Augustine’s memory is not exclusively personal but it contains thoughts shared with other human beings, like ideas of numbers, scientific knowledge of the disciplines etc. See e.g. Gerard O’Daly, Augustine’s Philosophy of Mind, Berkeley & Los Angeles, CA: University of California Press 1987, esp. pages 136–137.
162
P. Remes
phenomenon that ordinary people may know many things by merely looking in another’s eyes, but it is described as a peculiarity of disembodied souls which display the nous’s perfect and self-sufficient intellection. Although the contrast with discursive reason and nous is not explicitly made, it is evident from the context that this feature is one of the aspects of such functioning of the (disembodied) soul that happens in close unity with the Intellect, and as distinct from and more perfect than the discursive thought activity in bodies. Discursive thought must usually be communicated to others by speech, but the disembodied souls understand and recognise one another by some sort of intuition (oud’ an epitattoien oud’ an symbouleuoien, gin¯oskoien d’an kai ta par’ all¯el¯on synesei, ll. 18–19). Accordingly, we can add one more difference to the above table, not emphatic in Plotinus but nonetheless based on textual evidence: – Open access to the shared objects of knowledge, and thereby to the contents of all other intellects (ouden krypton oude peplasmenon; IV.3 [27] 18.22)
– Obscure and potentially hidden or veiled from others, disclosed (and possibly also concocted) in speech (and in action)
This is related to an understanding of truth as something which, once acquired, is entirely lucid to everyone, whereas our personal thoughts can be both obscure as well as obscured from others. There is an occurrence of a similar idea in John Cassian (Conf. 10.11.4–5), who describes an ideal state of the heart in which one grasps the significance of a psalm in beforehand rather than afterwards.21 Once the soul is sufficiently prepared, truth in a sense pervades the soul and is transparent to it, thereby rendering inquiry, discussion, dialectic and learning futile. In contrast, discursive reason happens in the seclusion of one’s own mind to the extent that it is possible to lie and to hide one’s thoughts from other people. Doxa, unlike epist¯em¯e, may also be fabricated. The state of knowledge is such that all minds share its secure objects and therefore have open access to one another. Although this point about crucial difference in the two modes of cognition is made in passing and left unemployed elsewhere by Plotinus, it forces us to reconsider some of the widespread conceptions about the lack of any notion of privacy in ancient psychological theories. Universality is essentially tied only to ideal modes of cognition. Cary also suggests at least implicitly that the relation of Augustine’s externalized God to human mind is of the following sort: God is present as an eternal and divine light in the mind, bestowing the very capacity for judgement by which it is a mind. Unlike theories of mind in antiquity, this mind is never a passive recipient but its nature is to understand and to judge.22 Other scholars, too, have emphasised Augustine’s memory as more than a storage, a deliberate and willed activity of imposing order upon a chaotic collection of memories.23 As to the extent to which 21 Simo Knuuttila (Emotions in Ancient and Medieval Philosophy, Oxford: Clarendon 2005, 105) links this to the Origenist idea of the super-intellectual capacities of ascetics. 22 Cary 2000, 65–67; 114; 126. 23 Cf. e.g. Janet Coleman, Ancient and Medieval Memories. Studies in the Reconstruction of the Past, Cambridge: Cambridge University Press 1992, 90–91.
Inwardness and Infinity of Selfhood: From Plotinus to Augustine
163
this makes Augustine different and more modern than Plotinus, I would like to make two qualifications. First, against Aristotelian theories of perception and thought as predominantly receptive functions, Platonism – already pre-Augustinian Platonism – has the tendency to emphasise the mind’s functions as powers or activities of the soul. Accordingly, Plotinus argues for an understanding of memory as a power to remember, and a power that can be exercised (IV.6 [41] 3). (I shall return later to the question as to whether there nonetheless is a new dynamic in Augustine’s notion of memory.) Moreover, what Plotinus means by intellect is not so far removed from the Augustinian “light” supplying the mind its dynamic power. When Plotinus is not examining the Intellect’s proper functioning from the point of view of the whole metaphysical hierarchy, but, rather, its significance to the human mind in the composite, he gives intellect a role as constituting the laws according to which discursive reason functions (V.3 [49] 3.31–32; 44–45; 4.1–4). Discursive reason is receptive of impressions from the world through phantasia, but it is not a passive recipient, but a power to recognise individuals, groups and differences, to formulate propositions and to engage in dialectic. These are powers that the intellect bestows upon it.24 If Augustine’s Christ is something other than the soul and yet present in it, Plotinus’ intellect, too, is something other than our everyday consciousness, but has a beneficial impact on it (V.3 [49] 9.1–10). In Plotinus, the Intellect is the true self in the sense of being both that which gives us our rational capacities and a normative ideal, the perfection towards which we should strive. Being a part of our nature, it is nonetheless not something open to our every-day consciousness without moral and especially intellectual efforts. To the similarities between Plotinus and Augustine must further be added that, if Augustine’s God functions as a light in the mind, the metaphor for the Intellect’s presence in the soul in Plotinus is also light (V.3 [49] 8.20–30; V.6 [24] 4.18).25 Second, to reverse the direction of argumentation, it is not evident to which extent rational capacity is, even for Augustine, a feature of the personal self, especially as compared with modern or contemporary ideas. In the beginning of the Soliloquies, for instance, Augustine professes not to know where the voice of reason, with which the solitary conversation takes place, comes from, that is, whether it is internal or external to him (Sol. 1.1: sive ego ipse sive alius quis, extrinsecus sive intrinsecus, nescio). Although it is later stated that understanding comes into being through the person who understands (e.g. Sol. 1.13), a submission to a greater force is required: “Give up your desire to be, as it were, just your own person and under your own power (quasi proprius et in tua potestate), and profess yourself to be the slave of this most merciful and mild master” (Sol. I.30, translated by Gerard Watson). As in Neoplatonism (and unlike in some pronouncements of Stoicism),26 24
For a full argument to this effect, cf. Remes 2007a, ch. 3. To which extent both are reliant on the sun in Plato’s Republic is an interesting question. 26 E.g. Epictetus (Diss. 1.1.23–24) claims that our prohairesis (moral purpose and some kind of choice or will) is such that even Zeus cannot command it, and his teaching suggests that its firm exercise is the highest form and purpose of human life. 25
164
P. Remes
intellectual activity and being a master of one’s own thoughts are only the second best option, whereas the ultimate end is a vision of God which fulfils one’s mind and being. Knowledge comes about through divine illumination. Plotinus paves the way for Augustine, who is simply more radical in never, even in discussing normatively ideal states of the soul, identifying truth with human selfhood. Truth is, rather, Christ or God, and the knowledge of it happens through the teacher within, Christ present in the soul (e.g. De trin. 8.2; De mag. 10.32; 11.38).27 Furthermore, Cary draws attention to a methodologically significant factor, that is, to the way in which Augustine suggests that we must move step by step from body and perception to the interior power of perception, and only then to reasoning power (Conf. 7.17.23). This shows the gradual way in which the mind becomes acquainted with its true and better nature, ultimately approaching God. In my view, this methodology seems entirely Neoplatonic. For Plotinus, too, turning inwards is a beginning of a long journey towards truth and ultimately the One, and it starts by stripping the soul from the body and then step by step from those of its functions connected to the body and to this world (V.3 [49] 9.1–10). In a very similar fashion, Porphyry ordered Plotinus’ Enneads so that the lengthy books on soul’s functions such as perception come before the treatises on intellect and the intelligible, and the Enneads as we have them end with treatises on the One. The order points to an ever deepening self-knowledge that starts from faculties most familiar to us, proceeding to those that are more difficult and more divine.28 Both in method and content, then, the inner vision of Plotinus and Augustine share more features than are obvious at first sight.29
27 See also Gareth B. Matthews, “Knowledge and Illumination”, in Eleanore Stump and Norman Kretzmann (eds.), The Cambridge Companion to Augustine, Cambridge: Cambridge University Press 2001, 171–185. 28 As a scientific method this has a precursor in Aristotelian methodology, see Stephen K. Strange, “Plotinus on the Nature of Eternity and Time”, in Lawrence P. Schrenk (ed.), Aristotle in Late Antiquity. Studies in Philosophy and the History of Philosophy 27, Washington, DC: Catholic University of America Press 1996, 22–53. What is important in the context of self-knowledge is the idea that both Plotinus and Augustine hold, namely that this method is a kind of gradual subtracting of that knowledge which does not concern the real or true self, rather than a process of gaining more information about it, since the true self within also and already knows itself. Cf. Stephen Menn, Descartes and Augustine, Cambridge: Cambridge University Press 1998, 252. 29 Although there is no proper space here to tackle the question of introspection in Augustine, let it be noted that Suzanne Stern-Gillett (“Consciousness and Introspection in Plotinus and Augustine”, Proceedings of the Boston Area Colloquium in Ancient Philosophy 22(2006), 145–174) makes a useful distinction between contemplative and confessional introspection. Contemplative introspection has God as its intentional object, and resembles Plotinus’ mystical experiences with unity, while confessional introspection is an investigation of one’s inner self. In his commentary of Stern-Gillett, Peter John Kenney (“Commentary on Stern-Gillett”, Proceedings of the Boston Area Colloquium in Ancient Philosophy 22(2006), 175–181) acknowledges the distinction but points out that the relationship between the two is special: confessional introspection is not simply preparatory for contemplative introspection, since contemplative introspection leads Augustine to confessional practice. One might therefore perhaps conclude that the vision of God has a different place in Plotinus’ and Augustine’s systems: for the former, it is the end of a long preparatory and intellectual road, for the latter something that also guides and motivates one to travel that road.
Inwardness and Infinity of Selfhood: From Plotinus to Augustine
165
2. From the point of view of ancient philosophy, one of Cary’s merits is to emphasise the influences of Cicero in Augustine. But from the same point of view, it seems that he has a slight tendency to oversimplify the Stoic understanding of mind and self. He claims that the self we find in, for example, Epictetus, is our own power of choice rather than any inner realm (Diss. 3.22.38–39), and that the idea that God and truth are to be found within is weighed down by Stoic materialism.30 Although the former claim seems in broad outline correct (I am not at all sure how directly metaphysical materialism is opposed to the idea of inwardness of mental life), I wish to underline two important steps made by the Stoics, necessary for both Plotinus and Augustine to come up with their theories of a turn inwards. First is the idea that the role of phantasia is not confined to just imagination or dreaming. For the Stoics, phantasia is responsible for more than dreams or other misinterpretations. In some Stoic thinkers, this involves a distinction between appearance, phantasia, and figment, phantasma, of which only the latter is of the pejorative, untrustworthy kind.31 The former seems to be involved in all or most mental functioning and has therefore some kind of unifying role in the commanding faculty, the h¯egemonikon.32 Phantasia makes the mind which receives perceptions aware of both the external object of perception and the mind itself. Some scholars further hold that phantasia is a kind of unifying consciousness.33 Furthermore, in the case of adult human souls, it renders an awareness of the objects of thought and perception which is rational, that is, in a form that can be further formulated in propositional language.34 Plotinus follows the Stoics in making phantasia indispensable for completion of all sensations. He even postulates another phantasia, closely related to the first, which intercedes between discursive reason and objects of knowledge. (IV.3 [27] 29.25; IV.3 [27] 30–31; IV [28] 4.8.) Neither Stoics nor Plotinus make explicit what seems to be one obvious implication, namely that if each person forms his or her own phantasiai about external objects, these are at least potentially different, proper to the particular mind that formed them.35 Undoubtedly, taking 30
Cary 2000, 10. LS, 39 A&B. 32 LS 53, and the comments at 321. 33 E.g. A.A. Long, “Representation and the Self in Stoicism”, in Stephen Everson (ed.) Psychology. Companions to Ancient Thought 2, Cambridge: Cambridge University Press 1991, 102–120, at 106–107. This has its background already in the Aristotelian idea that the soul never thinks without appearance, DA III 7, 431a14–17. 34 Cf. also Troels Engberg-Pedersen, The Stoic Theory of Oikeiosis. Moral Development and Social Interaction in Early Stoic Philosophy, Aarhus: Aarhus University Press 1990, esp. 145–152. 35 At least it can be noted that the Stoics held that an expert perception and resulting appearance differs from the layman’s perception of the same object. (Diogenes Laertius, 7.51 = LS 39 A(7) ). I am grateful to Håvard Lokke for the reference. Cf. Pauliina Remes, “Ownness of Experience in Antiquity”, in Sara Heinämaa, Vili Lähteenmäki, and Pauliina Remes (eds.), Consciousness: From Perception to Reflection in the History of Philosophy, Dordrecht: Springer 2007b, 67–94. In his discussion of Marcus Aurelius, Pierre Hadot distinguishes several levels of representation, of which the first one is a passive and truthful impression of the external object (phantasia katal¯eptik¯e), after which the soul starts to add and mould the impression. See Pierre Hadot, The Inner Citadel. The Meditations of Marcus Aurelius, Cambridge, MA & London: Blackwell 1998[1992], esp. 102–104. 31
166
P. Remes
appearances as some kind of inner mental objects would be anachronistic and overstated. Yet the development of phantasiai as something in the middle between the subject and the objects in the world can be seen as one step among many in the gradual movement towards the idea of an individual soul which interprets the world in a manner particular to just him or her – an inner realm proper to one individual. Appearances are not mental representations about objects, but their existence makes individual differences and errors about the shared objects of perceptions possible. Importantly, in addition to this psychological theory, the Stoics make an ethical and therapeutic suggestion that helps in conceiving the mind as a space within, namely the idea of examination and proper use of appearances (e.g. Epictetus, Diss. 1.1.7; 2.1.4; 2.22.29; 4.6.34).36 Although these appearances are the way in which the world appears to me, somehow I can, with other aspects of my h¯egemonikon, take distance to the appearances and assess and judge them, as it were, from above – that is, either assent to them or withdraw my assent from them. The above usage of ‘distance’ and ‘above’ is of course intentional: there seems to already be a “mental space” where appearances are examined by other parts or aspects of the same mind – a subject of some sort. Also the inward looking turn recurs in the Stoic authors. It is connected to this practice of concentration upon oneself in the sense of examination and evaluation of one’s possibly misleading impressions, as well as to the demand for a withdrawal to one’s inner peace and innate familiarity with truth and goodness. To quote Marcus Aurelius: Men seek retreats for themselves in country places, on beaches and mountains, and you yourself are in the habit of longing for such retreats, but that is altogether unenlightened when it is possible at any hour you please to find a retreat within yourself. For nowhere can a man withdraw to a more untroubled quietude than in his own soul, especially a man who has within him things the study of which will at once put him perfectly at ease, and by ease I mean nothing other than orderly conduct. (Med. 4.3)
Once this idea of internal peace is combined with the Platonic and Neoplatonic understanding of truth and being as internal to the soul’s higher aspects,37 the inward turn reveals the things of true significance – but only correctly executed and after much philosophising. These Stoic and Platonic doctrines form a basis on which both Plotinus and Augustine – the latter partly through Neoplatonism, partly independently – build their notions of mind and self. To some extent, Augustine’s project of Confessions book 10 is a dialectic of these two inheritances. First he enumerates the kinds of things that memory contains (experiences, liberal sciences, innate conceptions, principles and laws of numbers and dimensions, feelings etc.; Conf. 10.8–16), in 36
E.g. Long 1991, 111 ff. The idea that truth is to be found within also appears in some Stoic authors (e.g. M. Aur., Med. 7.59), but in Neoplatonism it grows into that of a full-fledged theory. Here, too, Neoplatonism amalgamates the Stoic idea, the independent power of reason internal to us, to that of the Platonic idea of innate and internal truths. 37
Inwardness and Infinity of Selfhood: From Plotinus to Augustine
167
order to later confess and analyse the different kinds of possible temptations involved in different experiences, thought and feeling (10.30 onwards). In doing this, he provides, in a manner of speaking, a map for the examination of our impressions about the world and its temptations. This project is tied to the search for God within: “But where in my memory do You abide?” (10.25.36.) Only with God does one conquer the temptations and see truth and light. God, like Platonic intelligible being, resides everywhere, both in the universe and the soul, only to be revealed by the human mind. (10.26.37.) 3. The former two points have concentrated on philosophical psychology and ethics. The third point is metaphysical, and concerns individuality. The question is whether souls and especially intellects are essentially individual. Cary notes that the souls’ origin in Plotinus is in a kind of eternal state of “togetherness”. He rightly points out that Plotinus advocates a look back to this origin of souls, to a moment before time when the souls were pure and good, and in some kind of unity. What we have here is the famous doctrine that there is an “unfallen” part of the soul which one should direct oneself to, the view of the soul’s fundamental ontological origin in the higher metaphysical levels, in eternity. Cary then argues that the only thing that separates Plotinus’ pure intellects from one another is the attachment they come to have to a particular body. Hence individuality is a result of the fall from that state of unity. Whatever the souls look at when looking back to their origin, and thereby into their own depths, is a shared unity, and not anything individual. Therefore our true nature cannot be private.38 However, in the light of Plotinian scholarship this question is highly vexed. Much of the evidence points into another direction. Plotinus has a habit of talking of several intellects, as if there was something that separates them from one another as such, not merely in and by the embodiment. He also explicitly claims that the intellects are many even without the body (IV.3 [27] 5.1–15).39 Famously, this creates a problem when combined with Plotinus’s doctrine of the intellect’s identity with its objects of thought. If every disembodied intellect is identical with the very same (Platonic) forms, it seems impossible for them to have any differentiae.40 This is not the right place to tackle this problem, and for our purposes it is enough that the dilemma is there, and that the differentiation and individuation of human beings is not entirely – or even primarily – due to matter and (material) body.
38
Cary 2000, 120–121. “Since the intellects there too do not perish into unity, because they are not corporeally divided, but remains each/distinct in difference, having the same essential being”. Most emphasis is put to individuality of souls and intellects in research literature by Robert Bolton, Person, Soul and Identity. A Neoplatonic Account of the Principle of Personality, Washington & London & Montreaux: Minerva Press 1994. 40 Cf. Richard Sorabji, “Is the True Self an Individual in the Platonist Tradition?”, in Marie-Odile Goulet-Cazé (ed.), Le Commentaire entre tradition et innovation, Paris: Librairie Philosophique J. Vrin 2000b, 293–300; The Philosophy of the Commentators 200–600 AD. A Sourcebook (vol. 3) Logic and Metaphysics, London: Cornell University Press 2004b, ch. 18. 39
168
P. Remes
In a similar fashion, every soul has some “otherness”, apparently even without or regardless of the influences of the body (e.g. V.1 [10] 1). Disembodied souls recognise each other “by their characters and particular nature of their behaviour” (IV.4 [28] 5.11–22).41 It is also Plotinus’ contention that the individual souls contribute to the creation of the sensible world. They are not just essentially similar items located in and differentiated by different bodies but have some seeds for their own characteristics, which further likely determine which bit of matter, i.e. which body, they choose to be realised in within the sensible realm. Every soul has something peculiar to it with which it helps in completing the universe, in contributing to its wholeness (e.g. VI.7 [38] 7.9–17). All this means that the origin and inner realms Plotinus advises people to turn to are not exclusively universal and shared but contain aspects of, if not privacy, individuality. 4. As to what the human self is, Plotinus is admittedly drawn in two different directions. Sometimes he does depict the intellect as our innermost nature, and hence a kind of true or core self. This idea supports Cary’s claim that Plotinus’ self is not primarily personal or an inner space but connected to the shared and universal intelligibles. However, I hope to have shown that this is not an exhaustive picture of Plotinus’ philosophical psychology. I have indicated that he devotes long treatises to human psychology, and he does once claim explicitly that our truest nature is not the intellect which is “above” us but its temporal production and counterpart, discursive reason, the rational capacity that functions in between our bodily and intelligible natures. (e.g. VI.5 [23] 7; I.1 [53] 9; V.3 [49] 3.32 ff.) Since this reason is erring, its contents partly determined by external circumstances and one’s particular personal history and character,42 and even to some extent private (as was argued above), there is a dimension of selfhood which is deeply personal. Cary is, however, right in maintaining that this dimension is not the best and ideal part of us, and that both the causally and the normatively important dimension of selfhood resides in higher metaphysical levels. With this enriched view of Plotinian psychology and anthropology, it may be concluded that the relation that the rational human being bears to the true being and the intelligible structures of the universe is not entirely simple, and that the turn inwards does not automatically guarantee a contact with what is shared and universal. Merely looking within does not reveal truth for Plotinus any more than it does for Augustine.
41 He also refers to spherical bodies, a doctrine originating in Plato’s Timaeus where each soul is attached to its respective vehicle, a star (41d–42a). One possibility is that it is this luminous vehicle which individuates disembodied souls – an idea the later Neoplatonists were fond of (cf. Sorabji 2005, 8b). Yet in this text Plotinus seems to think that a spherical body is less promising than character in this respect. 42 I have argued for this elsewhere: Remes 2007a, ch. IV.2.
Inwardness and Infinity of Selfhood: From Plotinus to Augustine
169
Augustine’s Originality: Time, Change and Infinity The purpose of these critical comments has not been to undermine Cary’s central tenets. The move to remove God from the human mind significantly contributes to individualisation and privatization of the inner realm. Conceiving of the self as not in itself divine is undoubtedly a major change in the history of philosophy. Moreover, Augustine does make explicit claims about inward presence Plotinus never ventures to make, like claiming that one perceives oneself without any phantasia, by some inward presence (De trin. 10.16). Yet something more than seeds for the development of the inner realm lie in antiquity, especially in Stoicism and Plotinus: the methodology of the inward turn as well as the understanding of mind as including both personal and universal layers. I would next like to suggest that Augustine’s originality does not lie solely in the new emphasis he places on the personal inner realm. In addition to the privateshared and inner-outer –distinctions, the following three distinctions map the gradual generation of the notion of self: Temporality–eternality Mutability–immutability Finity–infinity First, let us tackle the issue of temporality. There is a clear contrast with the soul’s ontological origin in eternity in the different outlooks of Plotinus and Augustine. For Plotinus, soul is still eternal, an entity like god, although its activities are temporal. The soul is temporal to the extent that its activities and life happen in time (time is the life of the soul), but since the soul itself must be immortal, unchanging and indestructible, it cannot have a temporal nature (IV.4 [28] 15.15–18; 16.1–7). This middle position of the soul is explicated much more clearly by Augustine, with the difference that for him the soul has a temporal generation (e.g. Ep. 166.3.6) and is entirely temporal. He has a threefold hierarchy: the nature mutable in space and time, namely body; the nature not mutable in space but only in time, namely the soul; and the nature that cannot be changed in space or in time, God. (Ep. 18.2.) Plotinus paves the way for this division in maintaining, as different parts of his philosophical system, that bodies and bodily actions are both in time and place, whereas discursive reasoning happens in time but is not essentially extended, and finally that noetic contemplation is neither temporal nor extended. But where Plotinus lifts the soul itself above time, Augustine makes the mind entirely temporal, thus preceding much later conceptions of selfhood, which cannot be understood without the horizon of time. Human beings are such that they exist in time, and their mental states are essentially temporal.43
43 Cf. Genevieve Lloyd, “Augustine and the ‘Problem’ of Time”, in Gareth B. Matthews (ed.), The Augustinian Tradition, Berkeley & Los Angeles & London: University of California Press 1999, 39–60, esp. at 39.
170
P. Remes
The same development may be seen in the way one’s past and personal history are treated and evaluated, that is, in the emphasis placed on the temporality of one’s inner mental life. For many ancient philosophers, our being in time and especially the fact that we have a past seems a dimension that does not, as one might expect, function as a layer that would constitute a true dimension of selfhood, but serves a different purpose. It reminds us of perishability, it teaches about mortality. In Marcus Aurelius, for instance, the past is seen as having perished rather than contributing to the present, and when it does have significance in the present, this is idealised. Let us look at two examples. Marcus Aurelius praises present actions, thoughts and dispositions (Med. 9.6), and points to the past as a means to find consolation. This consolation is to be found in the fact that even though the past phases of one’s life have ceased to be, there was nothing terrible in this passing, and so there will be nothing terrible in death (9.21). Both past and future are things that one does not possess, that one cannot alter, and thereby should be considered of limited importance. What matters for well-being is the present disposition and present life.44 Another difference is idealisation. In Roman thought, one’s past and one’s family traditions were idealised. Ancestors were seen as moral authorities and their actions were treated as exemplary.45 When Marcus Aurelius begins his Meditations by relating persons from whom he has learned important things, he is not primarily engaged in constituting a personal past for himself. In the Roman stoic picture, the past is not a personal dimension of one’s self but a story of one’s revered and ideal ancestry, an ideal up to which one should try to live. A whole new emphasis on the temporality of human nature is visible in both Augustine’s lengthy exposition of his past as well his curious way of lifting memory as a theory, in a sense, of the whole mind in the Confessions (esp. 10.13.20: we call memory itself the mind).46 Augustine’s memory contains, as we have seen, not just impressions of the sensible objects but also ideas, principles of sciences etc. It is thus much wider in scope than the memory in Plotinus which, as the Neoplatonic philosopher claims, belongs to phantasia.47 The choice of memory as the whole 44
Michel Foucault distinguishes between two kinds of self-writing: the intimate or confessional one which seeks to reveal the hidden or unsaid, and the other which aims at capturing the already said, collecting what one has managed to hear or read, and at shaping the self (“Self Writing”, in his Ethics: Subjectivity and Truth, New York: The New Press 1997, 207–221). Marcus Aurelius engages in the latter kind of self-shaping. His interest is to appropriate Stoic wisdom to the purposes of his own life. 45 R.B. Rutherford, The Meditations of Marcus Aurelius. A Study, Oxford: Clarendon 1989, 56–59. 46 Cf., e.g. Jean-Marie Le Blond, Les conversions de saint Augustine, Théologie 17, Paris: Aubier 1950, 6, 17; Roland Teske, “Augustine’s Philosophy of Memory”, in Eleanor Stump and Norman Kretzman (eds.), The Cambridge Companion to Augustine, Cambridge: Cambridge University Press 2001, 148–158, esp. at 151; O’Daly 1987, 136. 47 Enn. IV.3 [27] 30–31; see also IV.6 [41] 3. This does not mean that Augustine would not have been influenced by Plotinus’ discussion of memory. In addition to departures, several crucial interests of Plotinus seem to guide Augustine’s expositions (e.g. superiority of soul to the physical, innateness of ideas). See James McEvoy, “Does Augustinian Memory Depend on Plotinus?”, in John J. Cleary (ed.), The Perennial Tradition of Neoplatonism, Leuven: Leuven University Press 1997, 383–396.
Inwardness and Infinity of Selfhood: From Plotinus to Augustine
171
mind or self may result from Augustine’s contention that the temporal horizon is peculiar to the functionings of the human mind. Scholars have drawn attention to the novelty in Augustine’s attempts to illustrate the time sense through the concepts of memory and anticipation.48 His treatment of memory as not solely the retention of the past but also as the place where the self forms and encounters its plans and protentions for the future shows, further, how significant he thinks that temporality is for human selfhood. Obviously writing confessions that start from the earliest childhood enhances this understanding of selves as things with a past, evolving and being constituted in time, and as beings that have to relate themselves to their past. This relation to one’s temporal nature is personal. Augustine’s autobiographical remarks in the Confessions go far beyond idealisation. One’s past is more than a potential source of paradigms of good behaviour – a story of who one is and on what kind of necessary foundation one must build all future efforts at self-improvement. For Plotinus, personal memories have a role to play in the constitution of one’s embodied identity, but the true, stable and pure identity does not lie in memory but in the unchanging intellect.49 Second, the contrast between mutability and immutability. For both Plotinus and Augustine, truth is changeless. Plotinus still considers soul as a kind of form, immutable and eternal, but Augustine removes from the soul this special status (e.g. Conf. 10.25.36). The soul has a temporal generation and existence. This is certainly a new turn. Plotinus operates with the idea that the soul itself is immutable, although its life and activities are not, and even Augustine himself still treated the soul as changeless in the Soliloquia. Ultimately this change will make the self something that can be moulded, that can change and develop. Where Plotinus’ self is fixed to the extent that its core is immutable and eternal and that this core is also the normative ideal towards which the self is directed, Augustine’s memory offers different possibilities for self-realization. He turns to the cavern of memory, producing from its secret storages things that he wants. There [in the halls of memory] also I meet myself (ibi mihi et ipse occurro) and recall what I am, what I have done, and when and where and how I was affected when I did it. […] Out of the same abundance in store, I combine with past events images of various things, whether experienced directly or believed on the basis of what I have experienced; and on this basis I reason about future actions and events and hopes, and again think of all these things in the present. (Conf. 10.8.14)50
48 This connects with Edmund Husserl’s notions of impression, retention and protention, associated with now, past and future. Husserl starts his own lectures on time-consciousness by examining Augustine’s theory of memory. Zur Phänomenologie des inneren Zeitbewusstseins (1893–1917), Husserliana 10 (ed. Rudolph Boehm), The Hague: Martinus Nijhoff 1966. In English translation: On the Phenomenology of Consciousness of Internal Time (1893–1917), trans. John Barnett Brough, Dordrecht: Kluwer 1991. See also Simo Knuuttila, “Augustine on Time and Creation”, in Eleanor Stump and Norman Kretzman (eds.), The Cambridge Companion to Augustine, Cambridge: Cambridge University Press 2001, 103–115, at 113. 49 I have discussed this elsewhere; Remes 2007a, 112–119. 50 The translations of Augustine’s Confessions are those of Henry Chadwick.
172
P. Remes
The presence of these past events is not of them as unchanged: Augustine claims explicitly that thinking is a process in which the deliverance of the senses may be altered, interpreted by the thinker (Conf. 10.8.12). The storage of impressions functions as the space and material within which the person weaves both the story of his past and his plans for his future. While Augustine is not, of course, free from a normative Christian ideal, and this ideal to some extent regulates his narrated selfhood,51 the vast possibilities of memory offer a flexible tool for pursuing the ideal that does not reside already at hand in him. The metaphorical language of the two authors underlies this difference: Plotinus sculpts an inner statue according to a paradigmatic model, and his intellect is fixed on a set of intelligible forms (Enn. I.6 [1] 9.8–16).52 When Plotinus cuts away that which does not belong to the ideal core self, Augustine shows much more liberty in walking in the endless halls of his memory, making choices, interpreting and reinterpreting his memories: I run through all these things, I fly here and there, and penetrate their working as far as I can. But I never reach the end (per haec omnia discurro et volito hac illac, penetro etiam, quantum possum, et finis nusquam). So great is the power of memory, so great is the force of life in a human being whose life is mortal. (Conf. 10.17.26)
In this, Augustine has post-Plotinian predecessors. Both temporality and mutability of the soul seem to belong to the developments of later Neoplatonism. Later Neoplatonists rejected Plotinus’ dogma of the unfallen part of the soul, hence contributing to the humanity of the human soul by stripping it of divinity (e.g. Proclus, Elements of Theology, Prop. 211). What is left is not something that has an unquestioned access to truth nor a place among eternal entities. This has profound repercussions on the nature of the soul. Interestingly, Damascius (In Parm. 254) seems to hold that it is not merely the life and activities of the soul that change, but souls may change as a substance, thus pushing the limits of the concept of substance as something that can only come into being and perish.53 This is not something Plotinus could ever have accepted, and, again, a small but significant step in the making of the modern soul – or self. Augustine’s truly original contribution, one 51 For Augustine the moral agent is not indifferent and entirely autonomous, but the good will is determined by the good, and in the ideal case, that is, in paradise, self-knowledge is actually God’s seeing in us. See O’Daly’ contribution in this volume. 52 The sculpting passage abounds with literary references to Plato, see Remes 2007a, 211. Both Holger Thesleff and the anonymous reader at Springer suggested an allusion to Symposium (agalma at 215b and 216e–217a; Alcibiades looking inside the ugly Socrates and finding within something divine and beautiful). 53 Damascius, In Parm. 254 ff. Cf. Carlos Steel, The Changing Self. A Study on the Soul in Later Neoplatonism: Iamblichus, Damascius and Priscianus. Verhandelingen van den Koninklijke Academie voor Wetenschappen, Letteren en Schone Kunsten van België, Klasse der Letteren, Jaargang 40, n. 85, Brussels 1978. Proclus opts for substantial diversity rather than one changing substance. Matthias Perkams, “An Innovation by Proclus. The Theory of the Substantial Diversity of the Human Soul”, in Matthias Perkams and Rosa Maria Piccione (eds.), Proklos. Methode, Seelenlehre, Metaphysik, Leiden & Boston: E. J. Brill 2006, 165–185.
Inwardness and Infinity of Selfhood: From Plotinus to Augustine
173
might claim, is to do two things at the same time: to adopt and develop further this mutability of the mind and to show, through the autobiographical confessions, how this mutability appears in action. This yields selfhood which may change both gradually, as a result of confessional self-inquiry, and radically – in conversion. Third, Plotinus’ paradigm is a fixed set of intelligible forms, the beautiful structure of the intelligible universe and hence a stable and solid object of intelligible vision with which to become identified. The statue he sculpts within the confines of his interiority reveals gradually an already existing figure of amazing beauty – not one that would be the result of an individual artistic creation. Augustine’s cavern is very different, it is of infinite depth, a “manifold vastness full of innumerable riches in a wonderful way” (multiplices amplitudines plenas miris modis copiarum innumerabilium), “a field and a huge court of my memory” (in campos et lata praetoria memoriae), “a profound and infinite multiplicity” (profunda et infinita multiplicitas), “full of innumerable things” (plenis innumerabilium rerum) and the “depth or abyss of human consciousness” (abyssus humanae conscientiae) (Conf. 10.40.65; 8.12; 15; 17.26; 2.2). To an extent, Augustine is humble about this vastness, as the opacity of our nature to ourselves is the consequence of original sin.54 Human beings are finite in the sense of not being able to penetrate to the true depths of their own being, created and accessible only to God. The self-knowledge possible for them only imitates God’s perfect and atemporal self-knowledge in time.55 But as the above quotations also prove, the infinity of that quest also testifies about the wonderful richness of the human nature and consciousness. The nature of human memory, that is, mind, calls for wonder and exploration just as (or more than) the vast world invites us to wonder it. Mind’s ultimate impenetrability to itself is a challenge. (Conf. 10.9.15; 16.25.) What kind of infinity does Augustine mean, then? The collection of memories in the human soul cannot be infinite in a literal or actual sense of the word, nor is our capacity to remember unlimited. Memory is, rather, experientially endless or unlimited in that, first, we cannot access all or even major parts of it in any one moment, and many things in it escape our present attention. Second, it is vast enough to offer ever new possibilities of remembering and interpreting its own contents: depending upon the order of recollecting things in it, upon the connections we make in between distinct items in the memory, we come always to new ideas about our past. These are the “endless new likenesses” Augustine weaves at 10.8.14. The possibility of different future plans and protentions make this power even greater. Human nature, he tells us, is characterised by diversity, by life of immeasurable forms (10.17.26). Third, memory involves things more complex than imprints of past sense-impressions. Augustine’s interesting example is forgetfulness. He quickly dismisses that forgetting could be a case of either a) not remembering anything (since one does know that one does not remember); b) actively 54
Cf. O’Daly in this volume. Cf., e.g. Anne Harle, “Augustine and Rousseau: Narrative and Self-Knowledge in the Two Confessions”, in Gareth B. Matthews (ed.), The Augustinian Tradition, Berkeley & Los Angeles, CA: University of California Press 1999, 263–285, at 264, 269–270; Lloyd 1999, 40; 59. 55
174
P. Remes
remembering forgetfulness (since it is a case of forgetting rather than remembering); or c) having an image of forgetfulness rather than forgetfulness itself (since the presence of an image requires that one has once had the presence of the thing itself, and hence the case ultimately would, as I understand it, fall under b). (Conf. 10.15.23) The problem Augustine poses for himself seems to gain additional difficulty by how he defines, challengingly, both memory as more or less the whole mind with its vast abilities, and forgetfulness as a force that deletes everything, rather than, for example, a lack. For our purposes, it is significant that human mind includes not just memory imprints of experienced things but the powers that deal with them. It even encompasses this strange power of deletion, of not just transforming and revising things but of making them disappear, erasing them. One might say that memory or mind extends also into another kind of infinity, into nothingness. Now, of Augustine’s predecessors, Plotinus does attach some kind of idea of infinity to the highest summit of his metaphysical hierarchy and the final ideal end of human existence, the One (V.5 [32] 10.18–22),56 and thereby also to the final steps of an inward turn: If you have become this, and see it, and are united with yourself in purity, having nothing obstructing you from becoming in this way one, nor having anything else inside mixed with it, but [being] wholly yourself, only true light, not measured by magnitude, or defined by shape into [being something] less, or increased into magnitude by unlimitedness, but everywhere unmeasured because greater than all measure and superior to all quantity; if you see that you have become this, then from this time onwards you have become sight; feel confident about yourself, for having already ascended you no longer need anyone to show you; look intently and see. For this eye alone looks at the great beauty. (Plotinus, Enn. I.6 [1] 9.16–25)
The ideal infinity that the human soul should try to attain transcends the human mind’s own nature as a subject of thought, a dianoetic or noetic thinker. This unlimitedness is not personal,57 nor is it meant to create the kind of limitless and infinite room where Augustinian self-constitution happens. It does not free the self from an agreed and shared end but, rather, from the limits of the perceptible spatiotemporal realm. Thus, I would argue, Augustinian emphasis on the mutable and infinite nature of the soul makes self-constitution possible on a deeper level. Augustine cannot change his past, but by explaining his behaviour and by learning from it as he does in the Confessions, he can give a new meaning to it, and thus constitute his identity in a new manner. As we can read, he weaves a story of his past where significant childhood occurrences have their own meaning to the adult man he is. The late Neoplatonists make the human being a finite creature, no longer directly connected
56
I am grateful for this idea to Eyjólfur K. Emilsson. It requires “giving oneself up” (e.g. V.8 [31] 11.17). Brian Stock argues, to my mind correctly, that the Confessions highlight a process which is always given a personal content. Brian Stock, Augustine. The Reader. Meditation, Self-Knowledge, and the Ethics of Interpretation, Cambridge, MA & London: The Belknap Press 1996, 215. 57
Inwardness and Infinity of Selfhood: From Plotinus to Augustine
175
to the divine, whereas Augustine, one might claim, gives humans back their infinity, but in a new form, as mental and personal rather than divine or universal. In the course of the history of the subject, this “internal” infinity frees the person from any fixed character and limits, thus enabling a self-shaping that is not merely directed to a pre-existing end or ideal, but one which has innumerable possibilities. For Augustine, this nature of human self as innumerable memories and powers of “observation and enumeration” of them is not just richness but a scattered state in need of God to collect into one (Conf. 10.29.40). The self-inquiry sought by both Neoplatonists and Augustine is mingled with the discovery of truth, one’s source and goal – God. When we arrive at 20th and 21st century thinkers, e.g. Sartre or theorists of the so-called narrative self, the self’s capacity to change itself has become central and almost without boundaries. There is no longer any agreed goal that would make only some of the choices worthy and good. This development would not have been possible had not the normative and fixed character of the self been gradually given up. Augustine’s synthesis of the Stoic examination of images, of Plotinus’ inward and upward turn, and his own replacement of divinity in the self with the temporal and infinite depths of memory, open up for him a possibility of narrating the self through his Confessions in a manner which is characterised by an extraordinarily modern flavour. To sum up, Plotinus’ influence in the history of the inner self is even deeper than it has been acknowledged in the research literature. Privacy does gain a new momentum in Augustine, not only in the ways already envisaged but also in Augustine’s recognition of a kind of phenomenal world, the world which every subject perceives and which he or she cannot make mistakes about, as potentially distinct from a real world (C. acad. 3.11.24–5).58 Even though the Stoic-Plotinian doctrine of phantasia makes this phenomenal world possible, ancient philosophers do not seem too interested in the possibility of the subjectivity of experience.59 But alongside the new privacy and inner character of the mind, Augustine initiates a process in which the self is freed from its fixed character and opens up the possibilities of non-normative selfhood realised by much later thinkers near our era. From
58
Cf. e.g. Gareth B. Matthews, Augustine, Malden & Oxford: Blackwell 2005, 20. In a well-known article, Myles Burnyeat claims that the ancients did not consider the subjective as a realm about which there is or could be knowledge, and only after the Cartesian discovery “of the truth of statements describing the subjective states involved in the process of doubt” does subjective knowledge become central for epistemology. Knowledge and truth are thus always connected to the objective realm. It has even been claimed that for the ancients, even appearances are always wholly objective rather than subjective. For these claims and some discussion, cf. Myles Burnyeat, “Idealism and Greek Philosophy. What Descartes Saw and Berkeley Missed”, Philosophical Review 90(1982), 3–40; Stephen Everson, “The Objective Appearance of Pyrrhonism”, in Stephen Everson (ed.), Psychology, Companions to Ancient Thought 2, Cambridge: Cambridge University Press 1991, 121–147; Gail Fine, “Subjectivity, Ancient and Modern. The Cyrenaics, Sextus, and Descartes”, in Jon Miller and Brad Inwood (eds.), Hellenistic and Early Modern Philosophy, Cambridge: Cambridge University Press 2003, 192–231; Remes 2007b. 59
176
P. Remes
today’s perspective, the private inner space generated by Plotinus, Augustine and the thinkers of the early modern era may be an outdated and even dangerous conception, and thus temporality, mutability, and infinity of self may, finally, be Augustine’s more interesting and lasting marks on the notion of self.60 Uppsala University, Sweden and University of Helsinki, Finland
60 I am grateful to Simo Knuuttila, Gerard O’Daly, Holger Thesleff, Sara Heinämaa, and the anonymous reader of Springer for their insightful comments on different versions of this paper. I have further benefited from the comments of the two following audiences: the colloquium held on Ideals of Knowledge, autumn 2003, University of Oslo, Norway (organised by Eyjólfur K. Emilsson), and the research seminar of the department of theoretical philosophy, spring 2007, University of Helsinki, Finland.
Philosophy of the Self in the Apostle Paul Troels Engberg-Pedersen
Is there such a thing? Is there at all an awareness of “the self” in Paul, the apostle of Christ? Is there even “philosophy” of the self in him? These questions might seem uninteresting to students of ancient philosophy, who have traditionally defined their subject in opposition to various forms of specifically “religious” thought in antiquity. I begin from a set of considerations that will hopefully make the possibility of giving a positive answer to those questions appear reasonably interesting. First, following the Danish philosopher Dan Zahavi1 we may distinguish between three modern ways of conceiving the self: a Kantian one, which sees the self as a totally abstract, transcendental and substantive (self-identical) pole of identity behind all experience; next a 20th century, hermeneutical one connected with names like Paul Ricoeur and Charles Taylor, which sees the self as the constantly changing result of a process of narrative construction of identity;2 and finally an equally 20th century, phenomenological one, which sees the self as a dimension that is always already implied in any kind of experience. Do we find anything like these ideas in Paul? Initially, the likelihood seems slight. After all, there are manifest changes from antiquity to the modern period in the overall understanding of self and world. Indeed, there does not even seem to be any obvious Greek or Latin way of rendering the notion of “the self”. I shall argue, however, that we do find, in practice, a fairly strong, abstract concept of the self in Paul, which may even remind one of Kant – but also an extensive use of spatial talk about the “filled” self and identity which may remind one of the line adopted, for instance, by Charles Taylor in Sources of the Self.
1 “Selvet i filosofisk belysning” [The Self in Philosophical Light], in P. Bertelsen, M. Hermansen and J. Tønnesvang (eds.), Vinkler på selvet – en antologi om selvbegrebets anvendelse i psykologien [Perspectives on the Self. An Anthology on the Use of the Concept of the Self Within Psychology], Aarhus: Klim 2002, 17–36, esp. 20–22. 2 For Ricoeur, Zahavi refers to Temps et récit III: Le temps raconté, Paris: Éditions du Seuil 1985. Very relevant is also Charles Taylor, Sources of the Self. The Making of the Modern Identity, Cambridge: Cambridge University Press 1985, ch. 2 “The Self in Moral Space”, 25–52.
P. Remes, J. Sihvola (eds.) Ancient Philosophy of the Self, © Springer Science+Business Media B.V. 2008
179
180
T. Engberg-Pedersen
This claim should be seen against the background of a more or less traditional consensus that one can at most speak of an emerging consciousness of “the” self in antiquity and only in a fairly rudimentary form of a genuine “philosophy” of the self. Thus Taylor, in the book just referred to, basically took a giant leap from Plato to Augustine, and even that only as a sort of background to the later invention or “making” of the self. Against this I shall argue that although there is certainly no explicit and self-conscious philosophy of the self in Paul, we do find in him a use of (pre-existing and well-prepared-for) reflections on something which may very well be called “the self”.3 Interestingly, we shall see that this use emerges in situations of (intellectual) conflict. Second, it has been a constant preoccupation in the late Bernard Williams’ engagement with ancient thought to get behind our modern all too taken-forgranted conceptions, e.g. moral psychological ones, which are precisely a result of a historical development. Instead, Williams aimed to go back to the ancients and see how little of the later baggage we actually need to make sense of our lives. How much do we need to get along decently? Perhaps the later accretions might then be put aside as transgressing the “limits of philosophy”?4 This approach is eminently valuable. Ancient thought was here treated as an “other”, with which we nevertheless stand in some fairly intimate relationship. As is well known, Williams also preferred to go back behind Socrates who was one of the fathers of the “morality system” (basically, our own moral system), parts of which Williams detested. With Paul not being a philosopher and with the self possibly only gradually emerging in ancient thought, could he too be a window to a less grandiose understanding of the self, even though central parts of what we find in Paul, the Christian, would be rejected out of hand by the not very pious Williams? I shall argue, however, that Paul decidedly belongs after Williams’ dividing line, and moreover as an integrated part of Hellenistic reflections on the self; still, his understanding of the self is a less than fully developed (though quite sophisticated) one that only emerges in the context of specific intellectual conflicts. Third, the relationship between self and body is a classic issue in the philosophy of the self, not least in 20th century thought. Again one may think of Williams’ discussions in Problems of the Self,5 in which he argued, basically, that persons 3
In this context mention should at least be made of the work of late Foucault that began with the two last volumes of his Histoire de la sexualité vol. 2: L’usage des plaisirs and vol. 3: Le souci de soi (both Paris: Gallimard 1984) and was continued with “Technologies of the Self” in the volume of the same name edited by L.H. Martin, H. Gutman and P.H. Hutton (Amherst, MA: University of Massachusetts 1988b, 16–49). Also highly relevant is the work of Pierre Hadot, e.g. in, Philosophy as a Way of Life. Pneumatic Exercises from Socrates to Foucault, Oxford: Blackwell 1995, an extended version of Exercises spirituels et philosophie antique (2nd ed., Paris: Études Augustiniennes 1987). Together, and in their very different ways, both authors have done much to fill in the Hellenistic gap left open by Taylor. 4 Relevant titles by Williams are Ethics and the Limits of Philosophy, London: Fontana 1985; Shame and Necessity, Sather Classical Lectures 57, Berkeley, CA: University of California Press 1994. 5 Cambridge: Cambridge University Press 1973.
Philosophy of the Self in the Apostle Paul
181
(“selves”) are bodies. I shall argue that Paul both ties the self to some form of body all through (there is no idea of a bodiless, purely psychological or even immaterial self in Paul) and also operates with the self as an entity that relates itself, in some way from the outside, to the body. Fourth, there is a set of prejudices in the modern mind that contrast “philosophy” with “religion”, “Greek” with “Jewish” thought, “Hellenism” with “Judaism” and the like. I shall argue that, for Paul at least, these contrasts are obsolete.6 There is enough philosophy of the self in Paul for it to be interesting, even to modern philosophers of the self.
The Basic Picture: Galatians 2:19–20 We may begin from a few lines in Paul’s letter to the Galatians (a group of Christbelieving people who lived somewhere in the interior of modern Turkey). In a passage in which he is describing a conflict between himself and the apostle Peter over a crucial ethnic issue between Jewish and non-Jewish Christ-believers, he states the following: (19) I, by contrast (eg¯o gar), died to the (Mosaic) law through the (same) law in order that I might live for God. I have been crucified together with Christ. (20) What lives is no longer I (eg¯o), but in me lives Christ. To the extent that I now live in the flesh, I live in faith, that of the son of God who felt love for me and gave himself over for my sake. (Gal. 2:19–20, my translation)7
The crucial phrases are these: “I have been crucified together with Christ. What lives is no longer I (eg¯o), but in me lives Christ. To the extent that I now live in the flesh, I live etc.” We may note a number of points about the passage. As already noted it arises out of a description of an intellectual conflict. Thus the “I, by contrast” is in opposition to Peter, whom Paul has been referring to (though only implicitly) in what immediately precedes (2:18). Out of this rhetorical contrast springs some further thought about how to describe Paul’s own “I”.8
6 Compare my own books Paul and the Stoics, Edinburgh: Clark 2000 and Engberg-Pedersen (ed.), Paul Beyond the Judaism/Hellenism Divide, Louisville, KY: Westminster John Knox 2001. 7 Gal. 2:15–21 is a storm centre in Pauline scholarship. Apart from the translation of the end of v. 20 (“that of the son of God […]”), however, there should be general agreement about the translation given here; and those concluding words are immaterial to the present purposes. 8 Note, however, that although Paul is clearly describing a personal experience, he sees and describes it as the kind of experience all Christ-believers should have. It is not personal in the sense of being merely his own. Earlier in the letter (at 1:15–16), Paul has described the same event as a calling by God when He decided to “reveal His son in me” (1:16). That was something that happened to Paul alone and made him stand out. Now, by contrast, he is attempting to spell out the meaning of that event. And that pertains to all Christ-believers.
182
T. Engberg-Pedersen
That thought has several sides to it. There is first the idea that “I” am dead, namely, “crucified together with Christ”. This “I” stands for the “old” “I” which, as v. 20 shows, was in some way connected to the flesh. Call this “I” I1. That “I” is left behind and gone. But then there is another “I”: the one that was meant to live for God. This “I” is in fact a dual one. It is partly an “I” which now does live for God, is “crucified together with Christ” and henceforth lives in Christ faith. Call this “I”, which is in direct contrast with I1, I2. However, I think we need to say that another “I” is involved, too. This “I” has two sides to it. First, it is that same “I” which before conversion was there as being alive to the law (now, by contrast, it is dead to the law) and which now is alive to God. Or again: which has been crucified together with Christ, but now also is crucifiedtogether-with-Christ (whatever that means).9 Or again: which no longer lives – but still is there for Christ to live in it (in “me”). We may see this self-identical “I” as an entity that underlies the change of which Paul is speaking in a sense of “underlying” that harks back to Aristotle’s notion of the hypokeimenon that “underlies” substantive change. Second, this underlying “I” also seems to appear in the role of a kind of superior “I” of second-order reflection: an “I” that from the very beginning enables Paul to give his description of that change in himself. Both the general description of the change with its intense focus on a very marked “I” (eg¯o, emoi) and in particular the markedly second-order character of the phrase “to the extent that I now live in the flesh” seem to presuppose such a kind of self-reflective awareness. Now these two features – of an “I” that underlies the change and of the second-order reflection on the change – would appear to be two sides of the same idea in the precise way that the postulate of an underlying “I” may be derived from the second-order reflection. We may call this dual “I” I−/+. It expresses on the one hand the notion of an empty, abstract, underlying “I” (hence “I−”), which may in one instantiation be “filled in” to become the fleshly I1 and in another instantiation become the Christ-believing I2. On the other hand, it is also an “I” of a more powerful kind of second-order reflection, hence “I+”. What we see here is that in order to describe his own conversion – in a setting of concrete intellectual conflict – Paul in fact introduced and presupposed in his description a notion of two different kinds of “I”: a concrete, “filled” one, which may take two opposed forms as either I1 or I2, and an abstract, empty one, I−, which underlies the change from I1 to I2 and is also an “I” of second-order reflection, I+. This is the essence of my proposal. I suggest that the underlying “I” with its secondorder superstructure (I−/+) has a certain structural similarity to Kant’s substantive, transcendental self, even though it is of course not at all transcendental; rather, it is a formal self of second-order reflection on how to define oneself, on where to place one’s identity by filling in an underlying self in one or the other way. That last theme (of the actual, concrete, first-order self-definition and self-identification, the filling in of the underlying self) belongs with the two other “I”s, and that theme, I suggest, may well be understood in terms of the idea of a narrative construction of a (changing) self along the lines introduced, for instance, by Charles Taylor. 9
Note the perfect tense in synestaur¯omai: I have been crucified, but also now am – crucified.
Philosophy of the Self in the Apostle Paul
183
Here is a simple drawing to fix the idea:
Let me face an objection here. Could we not do with only two “I”s to make sense of the Pauline text: I1 and I2? Some second-order reflection is evidently involved in the text, but do we need to proliferate “I”s by postulating an extra awareness of a supposedly underlying “I”, “I−”? My answer will be that the answer should not be an either-or. The facts are there. Paul is obviously engaged in some kind of selfreflection. He is describing a change from one kind of “I”, I1, to another, I2. And he is clearly implying that it is one and the same “I” that was present in one form and now is present in another form. The logic of these facts taken together seems to require the kind of picture I have given. On the other hand, since Paul is not himself explicitly doing the kind of philosophical reflection engaged in here, we might well settle for the simpler picture if we want to stay closer to the immediate level of the text itself. What speaks for the fuller picture, however, is the fact that it draws out some rather evident implications of what is actually being said. In addition, we shall see later that in his reflection on the body that will be raised at the resurrection (this is the theme of 1 Corinthians 15:35–55), Paul insists on two separate ideas: first, that the body of “flesh and blood” will be completely changed into another body, which is a “pneumatic” one (badly translated as a “spiritual” one, see below), and second, that it is, as he repeatedly says, this particular body of “flesh and blood” that is completely changed into this other one. Here the emphasis on “thisness” (see 1 Cor. 15:52–53 quoted in note 24 below) seems to call for – and to translate very easily into – the idea of an underlying something that is, in a genuinely substantive change, first this and later that. So far, so good. A single passage is hardly enough. Fortunately, there are several other passages one may draw on. In the rest of this essay, I shall introduce one such passage and try to develop the content of the two “filled” conceptions of the “I” with which we have been presented. In particular, (1) how is each related to the body and (2) to what extent may each in fact qualify as a genuine “I”?
184
T. Engberg-Pedersen
The Fleshly “I”: Romans 7:14–25 The passage is the famous one in the letter to the Romans chapter 7, in which Paul describes – in the light of his new Christ experience – what it was like to live in the best way available before Christ, that is, under the Jewish law: the good that I will I do not do. I have discussed the passage in detail elsewhere.10 Here we must focus on the essentials. But first we need to have the text in front of us: (14) For we know that the (Mosaic) law is pneumatic (pneumatikos), but I (eg¯o de) am fleshly, sold under sin. (15) For I do not realize (gin¯osk¯o) what I bring about. For I do not do that which I will, but what I abhor, that I do. (16) But if I do what I do not will, I agree with the law that it is good. / (17) As it is, however, it is no longer I who bring it about, but sin, which lives in me. (18) For I know that no good thing lives in me, that is, in my flesh. For willing is present for me, but bringing the good about is not. (19) For I do not do the good that I will, but the bad that I will not, that I do. (20) But if I do what I do not myself will, then it is no longer I who bring it about, but sin which lives in me. / (21) Thus I find with regard to the law11 when I will do the good that the bad is present for me. (22) For I rejoice with God’s law in the inner man, (23) but I see a different law in my limbs (mele-), one that fights against the law of my mind (nous) and takes me prisoner in the law of sin, the one in my limbs. / (24) Wretched man that I am. Who will save me from the body of this death? (25) Thanks to God through Jesus Christ, our Lord. So then, I myself (autos eg¯o) serve God’s law with my mind (nous), but with my flesh (sarx) (I serve) the law of sin. (Rom. 7:14–25, my own translation)12
Paul’s basic rhetorical gambit in this text is to describe life under the Mosaic law as one in which there is a constant risk of akrasia or weakness of will (7:15–16) and to develop the description of the state of akrasia (7:17–20) in a special direction that makes it come out as nothing less than a genuine case of schizophrenia (7:21–23 and 25b, culminating in 7:24–25a).13 Again Paul is speaking of an “I” and again of life and death. The “I” that is described here is basically, as Paul states at the beginning of his argument (7:14), “fleshly” (sarkinos): “I (eg¯o) am fleshly, sold under sin.” In other words, he is again talking of I1. But here Paul goes into far more detail about the way this “I” is put together internally. Indeed, the whole point of the development of 7:15–16, 17–20, 21–23/25b and 24–25a is to spell out the exact way in which the “I” is fleshly.
10 See my “The Reception of Graeco-Roman Culture in the New Testament: The Case of Romans 7.7–25”, in Mogens Müller and Henrik Tronier (eds.), The New Testament as Reception, Sheffield: Sheffield Academic Press 2002, 32–57. One point needs special emphasis: In Romans 7 Paul is definitely not describing a case of post-conversion akrasia. Instead, as I said, he is describing a person who is in the best possible state that was available before conversion: that of a Jew, who strongly wills God’s own law. He, Paul claims, necessarily risks succumbing to cases of akrasia. 11 A more traditional rendering runs: “Thus I find the law […] that the bad etc.” For the alternative rendering see Engberg-Pedersen 2002, 49. 12 The last phrase here is probably misplaced. It may be a clarifying additional note that has crept into the text at the end of v. 25 instead of where it was meant to be: between vv. 23 and 24. 13 For this way of dividing up the argument – which I have even brought into the translation itself – see Engberg-Pedersen 2002, 46–48.
Philosophy of the Self in the Apostle Paul
185
Basically, this is an “I” which at the moment of acratic acting “does not realize what it brings about” (7:15). This lack of control is then further described as implying that “it is no longer I who bring it about, but sin, which lives in me” (7:17). “For I know that no good thing lives in me, that is, in my flesh. For willing (the good) is present for me (parakeitai moi), but bringing the good about is not” (7:18). There are two moves here. On the one hand, the fleshly “I” is split up in such a way that the good “I” that is part of the fleshly “I” and with which the person himself identifies is dissociated from some other element in the fleshly “I”: sin, which also lives in it. On the other hand, even though the good “I” dissociates itself from that other element in the fleshly “I”, it also realizes the existence of that element in itself. As Paul goes on to say (7:21), “I find (heurisk¯o) […] that the bad is present for me (again emoi […] parakeitai)”. Thus there is a complete split in the fleshly “I” at the same time as this split is clearly seen to occur within that “I” itself. The (fleshly) “I” identifies “in the inner man” with the good, but sees evil fighting “in its limbs” (mele-) against its own reason (nous), “trying to take me prisoner” in sin, which is present “in my limbs” (7:22–23).14 Paul summarizes this whole description as follows (7:25): “Thus I, being one and the same (autos eg¯o), slave with my reason (nous) for [the good], but with my flesh for [sin].”14 This passage shows a number of things that are of relevance to our query. First, there is a very heightened sense of a self or an “I”, the fleshly one (I1) of which Paul may at the very end (7:25, just quoted) say that “I myself” (autos eg¯o), in the sense of “I, being one and the same”, am put together in such a way that I am made up of several elements that interact in various disharmonious ways. This heightened sense of self more or less directly reflects the second-order view on “myself” that we called I+, the one that also gave us the idea of a self-identical self underlying the change from the state Paul is describing, I−. Remember here, moreover, that although Paul is clearly describing the akratic experience from within, the clarity with which he describes it derives from the fact that he provides this description in retrospect, based on the perspective he has achieved after his conversion. Second, the passage makes clear that the fleshly “I”, I1, is intimately tied to the human body, even where – as in the present case of akrasia – it also contains a strong element of willing the good. That is why Paul speaks repeatedly of the “limbs”. And that is why he ends with the famous exclamation: “Miserable man that I am. Who will save me from this body of death?” (7:24). We may conclude that this passage is both in complete conformity with and also adds to the earlier one. It clearly implies the existence of the second-order “I”, I+, which is the “I” that may (now) give that intense and philosophically quite careful
14
Note that in this paraphrase I have left out Paul’s references to the Jewish law, speaking instead of “the good”. The references to the law are vital to the point of Paul’s argument in its wider context. It is noteworthy, however, that it is quite possible to leave out the reference to the law in an analysis of the structure of the argument itself. Moreover, of course, Paul does take the law as standing for “the good” in the ordinary case of akrasia that he is describing, compare vv. 16, 18, 21 and 22.
186
T. Engberg-Pedersen
description of its own state as living under the Jewish law.15 It is also this “I” that may say of itself in Paul’s self-description that “I myself (autos eg¯o) serve” two opposed parts of the self. Admittedly, the fact that the “I” is split probably does not in itself quite yield the idea of an underlying “I” in the way we have been speaking of that. But representing oneself as split does. And while the autos eg¯o certainly has its primary meaning in pointing to two opposed tendencies in the single self, it also seems to point directly forward to the exasperated cry of “Wretched man that I am”, which very much expresses the self-conscious perspective of somebody looking at him- or herself from the outside. That outside perspective is the perspective of I+. In fact, what Paul does in 7:14–25 as a whole is to write out in a more and more intense way how the experience of the acratic person gradually forces upon him a clear awareness of his own cognitive situation and so practically generates the second-order perspective on oneself – which, however, is only fully adopted from the outside position that is achieved after conversion. The passage also shows in far more detail than the Galatians passage how we should understand the fleshly “I”, I1, that is now a thing of the past. That “I” is one that is insolubly tied to the human body of “flesh and blood” (sarx kai haima – as Paul himself calls it elsewhere, 1 Cor. 15:50), even where – as in the present case of akrasia – it also contains an element of reason (nous) which vainly attempts to make the “I” stick to the good. We are considerably wiser, therefore, on how to understand the ordinary human pole of Paul’s basic picture. But what about the other pole, the one we called I2? (1) Does this “I” have a body? (2) And when, as Gal. 2:19–20 had it, “I” am crucified together with Christ and Christ lives in “me”, is there in fact any “I” left at all? Is it at all proper to speak here of an “I” – other than the underlying “I”, I−/+, that also provides the description?
The Pneumatic “I” in Its Material Manifestation: Romans 8:1–13 This is one place where the modern interpreter, whose world view will be quite different from Paul’s, is forced to make an intellectual somersault if he wishes to stay in touch with what Paul himself intended to say. To begin with the first question: Yes, this “I” has a body. The situation now, after conversion, is structurally parallel with what it was before conversion. There we had an underlying “I”, I−, which was also able to look at itself (I+) as an entity that was tied to the human body of flesh
15 The point of adding (“now”) is to indicate that the whole description of the “plight” given in Rom. 7:14–25 presupposes – as I keep repeating – the realization of the “solution” as described in Rom. 8:1–13 and consequently the complete conversion from the earlier to the later state. Incidentally, this is an insight with wide ramifications that plays an important role in the so-called “new perspective” on Paul that has been developing since around 1980.
Philosophy of the Self in the Apostle Paul
187
and blood – for instance, in the way Paul spelled out in Romans 7. Now we have the same “I”, I−/+, which is similarly tied to a different kind of body, the result being a different kind of “I”, I2, which is the Christian “I” or, as Paul had it in Gal. 2:20, the “I” which, to the extent that it still lives in the flesh, lives in Christ faith. The passage that immediately follows Paul’s exasperated cry at the end of Romans 7 tells us more about this new “I”. Again, we need to have the text in front of us: (1) Thus there is no condemnation of those in Christ Jesus. (2) For the law of the pneuma (“spirit”) of life in Christ Jesus has liberated you from the law of sin and death. (3) For what was impossible in the law, the point where it was weak because of the flesh –: God sent his own son in the likeness of sinful flesh and for the sake of sin and condemned sin in the flesh, (4) in order that the just requirement of the law be fulfilled in us who do not walk in accordance with flesh, but with pneuma.16 / (5) For those who are in accordance with flesh think the things of the flesh, but those (who are) in accordance with pneuma (think) the things of the pneuma. (6) For the thought (phron¯ema) of the flesh is death, but the thought of the pneuma is life and peace, (7) since the thought of the flesh is enmity against God. For (the flesh) does not subject itself to God’s law – it just cannot do so – (8) and those who are in flesh cannot please God. / (9) You, however, are not in flesh but in pneuma, (x) if indeed God’s pneuma lives in you. (y) But if somebody does not have Christ’s pneuma, that man does not belong to him. (10) (y) If, however, Christ is in you, the body (s¯oma) is dead (nekron) because of sin, but the pneuma is life (z¯o¯e) because of righteousness. (11) (x) But if the the pneuma of Him who raised Jesus from the dead lives in you, then He who raised Christ from the dead will also make your mortal bodies alive through His pneuma which lives in you.17 / (12) So, then, my brothers, are we duty-bound – not to the flesh, to live in accordance with flesh; (13) for if you live in accordance with flesh, you will die; but if you put the acts of the body to death by means of pneuma, you will live. (Rom. 8:1–13, my own translation)
In this passage Paul describes a person who is “in Christ Jesus”. The crucial thing about this person is that he possesses something that Paul calls pneuma, the “spirit”. Such a person “is in accordance with” pneuma (8:5) in contrast with those described in chapter 7, who are here said to be “in accordance with” the flesh (8:5) or even “in the flesh” (8:8). The latter description is also transferred to those on the other side when Paul speaks of “being in spirit”, of having God’s spirit “live in” (oikein en) one (8:9a), of “having” Christ’s spirit (8:9b), and indeed of having Christ “in” one (8:10). This is all not only in principle alien to our ways of thinking, but also concretely very difficult to understand. Here is an attempt at drawing out an intelligible meaning. Pneuma in Paul is pretty certainly to be understood as a kind of body, like the matter of the stars in Stoicism.18 The idea then seems to be that human beings who
16 In this rendering nothing is left out from the original text. The dots indicate the presence in Paul’s text of a syntactical anakolouthesis – at a crucial turning-point in his argumentation. 17 The point of introducing (x) and (y) is to draw attention to the care with which Paul differentiates between God and Christ for the purposes of his argument. God cannot be present in a person. Christ can. 18 Compare for this 1 Cor. 15:35–49 with the discussion in Dale B. Martin, The Corinthian Body, New Haven, CT: Yale University Press 1995, 123–129.
188
T. Engberg-Pedersen
have Christ faith will also receive – literally from above – the divine stuff, the pneuma, which in the passage just quoted is both called God’s pneuma, Christ’s pneuma – and Christ himself.19 Thus the situation initially appears to be that where there previously was a body of flesh and blood, there now is a body of pneuma. That is in fact something Paul does want to say, but the matter is a little more complicated, as we shall see. First, let us note that he does want to say that this new body is also (some kind of) an “I”. That becomes clear, for instance, at the beginning of Romans 8 when Paul states that “the law of the pneuma of life in Christ Jesus has liberated you from the law of sin and death” (8:2), namely as described in chapter 7. Thus it is the same “I” which described itself in chapter 7 as “I myself” (autos eg¯o) being divided between reason (nous) and flesh – which is now said to have been liberated from that death by having come to possess pneuma. Similarly, of course, Paul continues throughout the Romans 8 passage to address his interlocutors, who are taken to be Christ- and pneuma-people, as individuals: “you are” and “if somebody does not”. The picture seems reasonably clear, then: where previously the underlying “I”, I−, was tied to a body of flesh and blood, resulting in an I1, it is now tied to a body that is made up of pneuma, the result being an I2.
The Pneumatic “I” in Its Cognitive Manifestation: Romans 8:1–13 But as I said: the matter is a bit more complicated both we regard to our first question concerning the embodiment of I2 and also to the second one concerning the extent to which I2 is a genuine “I”. The first part of this comes out in Rom 8:10: “If Christ is in you, the body (s¯oma) is dead (nekron) because of sin, but the pneuma is life (z¯o¯e) because of righteousness”. Well, the ‘body’ (s¯oma) – meaning the body of flesh and blood – is in fact not dead now, even in the person who has been “crucified together with Christ”. It is still there and Paul even reckons with this when he continues a few lines later (8:13): “For if you live in accordance with flesh, you will die; but if you put the acts of the body (again s¯oma) to death by means of pneuma, you will live”. How can the body of flesh and blood both be dead and still there to be reckoned with? Here we need to add to what we said about the pneuma. We said that it was a material entity, divine stuff, and it is that. But it is also connected with mind – just as flesh is. According to Paul, both pneuma and flesh have intentions (a phron¯ema, 8:6). The phron¯ema of the flesh is “for death”, that of the pneuma is
19 For the latter point compare 2 Cor. 3:17 where Christ is said to be pneuma. The point about the relationship between faith and reception of pneuma is implied in various passages in Galatians, see Gal. 3:14, 26–29; 4:6.
Philosophy of the Self in the Apostle Paul
189
“for life” (8:6).20 In a way this should come as no surprise. For as the fleshly “I” was described in chapter 7, it very much involved an intentional, cognitive element of willing one thing and “fighting in one’s limbs against” the good that one wills in one’s reason. In exact parallel with this, the divine stuff of pneuma also has a cognitive side to it. In relation to our question about the persistence of the body of flesh and blood even in the Christ believer, 8:10 will then mean, not that the body of flesh and blood is literally dead, but that once one has “Christ”, that is, pneuma, in one, the body of flesh and blood will be (completely) dead as delivering any input for valuation. It is seen as “dead” in the sense of being totally irrelevant to any value judgement. (This idea, incidentally, differs from Stoicism, where the body is rather seen as being “indifferent” (adiaphoron); as we know, this does allow for seeing certain of its expressions as at least “preferable” (pro¯egmena). In Paul, by contrast, the body of flesh and blood is seen as being totally irrelevant. It literally does not matter at all whether the body of flesh and blood is in this or the other kind of state.) The solution to the question of the persistence of the body of flesh and blood in the new, post-conversion state is, then, that having the pneuma in one is not only a material event (it is that), but also a cognitive one. Considered in the latter light, it does make sense to say that the body of flesh and blood – though still actually present – is dead and that its acts must be put to death.21 There is a nice way of putting this point, namely, by saying that the post-conversion situation envisaged by Paul mirrors the pre-conversion situation as he had described it in chapter 7. There we had, within the fleshly “I” (I1), an element of reason (nous) which willed the good but was overcome by sin in the limbs. In the new situation, we have basically the same being (with a vital addition, however), namely, a human being with a body of flesh and blood (but then also with the addition of pneuma); but now willing the good is not overcome by “sin in the limbs”, since now sin is dead (in the way I have explained), vanquished by the new element of pneuma. Previously, reason (nous) was present in the fleshly “I” – but ineffectually so: its willing was curtailed by the “other” thing in the mind. Now the role of reason has been taken over by the pneuma, which is effective: pneuma curtails the “other” thing by driving it out completely.
20 We need not consider exactly what this means. My guess is that Paul wishes to say that what flesh and pneuma put up as aims of action will eventually lead to death and life, respectively. 21 Here, if nowhere else, the perspicacious reader might ask about the relationship between saying that the body is dead and that the addressees should put its acts to death. This is the issue of the relationship in Paul between “the indicative and the imperative” that has exercised scholars for more than a hundred years. In Paul and the Stoics (Engberg-Pedersen 2000) I attempt to show why the issue should not have been seen as a problem.
190
T. Engberg-Pedersen
The Pneumatic “I” in Its Cognitive Manifestation: Galatians 5:13–26 and Philippians 2:2–4 Now to the second question I raised concerning the extent to which I2 is a genuine “I”. Here we need to ask about the content of what the pneuma both wills and is also able to put into practice. In particular, does the element of an “I” that in some form remains present in I2 have any specific role to play here? This question reflects the fact that in Paul’s construction of the fleshly “I”, I1, the element of “I” plays a very important role in the description of the content of what the flesh wills, thereby helping to give this “I” a very distinct “I”-character. Conversely, one may ask whether this is also the case in connection with the pneumatic “I”, I2. Let me explain. In the fleshly “I”, the “I”-perspective is central in the way that it is the only one that is active in the valuations performed by this “I”. This does not come out very clearly in the two passages we have considered to articulate the fleshly “I”, that is, Gal. 2:19–20 and Romans 7, since here Paul is more concerned with certain formal features of this “I”, with its logic. Elsewhere, he also explains the content of that “I”, the intention or aim of the flesh which, as we know, in the end means death. I cannot go into detail, but passages like Gal. 5:13–26 or Rom. 1:18–32 show that what is basically wrong with the fleshly “I” is that it is distinctly self-oriented or -centred, that is, egotistical. Not only does it contain a formal “I”-perspective (as all “I”s necessarily do), but it contains nothing but that. The fleshly “I” is, rather straightforwardly (in terms of the ancient moral system, that is), one that is concerned with getting as many bodily and other “material goods” as one possibly can for oneself (or one’s own group) in contrast with others. It is no wonder, therefore, that the “acts of the flesh” (erga te-s sarkos, Gal. 5:19) may be expressed in a list of vice-acts like the following: sexual immorality (porneia), depravity (akatharsia), licentiousness (aselgeia), idolatry (eid¯ololatria), magic (pharmakeia), enmities (echthrai), strife (eris), rivalry (z¯elos), fits of anger (thymoi), personal ambitions (eritheiai), dissensions (dichostasiai), factionalisms (haireseis), cases of envy (phthonoi), fits of drunkeness (methai), revellings (k¯omoi) “and things similar to those” (5:19–20). From Paul’s point of view, these may all be referred back to a desire of the bodily “I” for as many goods as possible for that “I” (or its group).22 In contrast with this, in the new, post-conversion situation described by Paul people will be concerned with nothing but the others. As he says elsewhere (Phil. 2:2–4),
22 I have explored this aspect of Gal. 5:13–26 and Rom. 1:18–32 in Engberg-Pedersen 2000, 152– 155 and 209–212. The basic point is that the list of vice-acts may be broken down into two types: acts of transgression vis-à-vis others and in relation to the body. The latter type falls under vice, not because there is anything intrinsically wrong with the body and its uses, but because the body is intimately connected with the notion of ‘I’ in its vicious form that also underlies the other type. (For an analysis of Gal. 5:13–26, see my essay “Stoicism in the Apostle Paul: A Philosophical Reading”, in Steven K. Strange and Jack Zupko (eds.), Stoicism: Traditions and Transformations, Cambridge: Cambridge University Press 2004, 52–75).
Philosophy of the Self in the Apostle Paul
191
they will be people who “have the same love, are of one heart (sympsychoi), think the same, people who do nothing from personal ambition (kat’ eritheian) nor from vain self-conceit (kata kenodoxian), but instead in (genuine) self-deprecation (tapeinophrosyn¯e) consider one another above oneself (all¯elous h¯egoumenoi hyperechontas heaut¯on), not looking each one to his own interests, but all together to those of the others”.23 Correspondingly, in the passage just quoted from Galatians, Paul lets his list of vice-acts be followed by one of virtues (attitudes proper, not just acts), which he even states to constitute the “fruit of the pneuma” (5:22): “love (agap¯e), joy (chara), peace(fulness) (eir¯en¯e), magnanimity (makrothymia), goodness (chr¯estot¯es), kindness (agath¯osyn¯e), faithfulness (pistis), mildness (praot¯es), self-control (enkrateia)” (5:22–23). Then the question should be clear: does the “I” have any specific role to play in this post-conversion “I”, I2? And then the answer should also be clear: no, not in any way that is comparable to the role of the “I” in the fleshly “I”. When an “I” is filled with pneuma, thereby becoming our I2, the “I”-perspective, which will of course continue to be there and which is also somehow tied to the idea of an underlying “I” (I−/+), plays no role whatever for filling in what the “I” takes him- or herself to be. Here there just is a “we” (or Christ). This observation corresponds very exactly with what we said about the relationship of the pneumatic “I” to the body of flesh and blood that this “I” will continue to inhabit. There we saw that the body of flesh and blood will be given no role whatever, as distinct from what one finds, for instance, in Stoicism. Similarly, we have just seen that the “I”-perspective, which continues to be part of the pneumatic “I”, is given no role whatever in setting up the proper goals of action. This is the point where Paulinism radicalizes the already quite radical other-regarding position of Stoicism. So: I2 is an I – but in a “reduced” form (as compared with I1). Conversely, we may say that it is the body of flesh and blood that provides a full self.
A Pneumatic Body in a Body of Flesh and Blood: 2 Corinthians 4:7–13 Before summarizing the implications of all this for Paul’s conception of the self, I will stay briefly with the question of how we should imagine his understanding of the post-conversion body, forgetting now about the cognitive side of this (the pneuma as a cognizer), on which we have spent some time, and focusing instead on the pneumatic body as a material entity in its relationship with that body of flesh and blood which remains in place even in the person who has received the pneuma.
23 For argument concerning this passage, see my analysis in “Radical Altruism in Philippians 2:4”, in J.T. Fitzgerald, T.H. Olbricht and L.M. White (eds.), Early Christianity and Classical Culture, FS Abraham Malherbe. SupNovT 110, Leiden: E. J. Brill 2003, 197–214.
192
T. Engberg-Pedersen
How are we to imagine this? To us the question will seem baffling, and it is far from clear that there is any very good answer to it. What matters here is mainly to show and insist that Paul did have this peculiar idea, of a body of flesh and blood which was also in some way a pneumatic body, with the pneuma being construed as a material entity. In 2 Cor. (4:7 ff) Paul speaks of being in possession of a certain “treasure”, which – to conclude from 4:6 – consists in having been illuminated in one’s heart with knowledge of the glory (doxa) of God (as seen) on the face of Christ. There can be little doubt that Paul is here describing a vision. He then states that he has this treasure “in earthen vessels” (ostrakina skeue-), meaning his body of flesh and blood (4:7). However, the idea is definitely not that he “has” some psychic experience – the vision – in something else, namely, the body. Instead, he speaks (4:10) of “always carrying around in the body the death (or dying, nekr¯osis) of Jesus, in order that the life too of Jesus may become visible (phaner¯othe-i) in our body (s¯oma)”. Indeed (4:11), though still alive, he is constantly being handed over to death, “in order that the life too of Jesus may become visible (again phaner¯othe-i) in our mortal flesh (sarx)”. A line later (4:13) this is then explicitly connected with Paul’s “having the pneuma of faith”, and indeed, “the same” pneuma, presumably the same as Jesus either has, had, was or is. So we are back again with our earlier picture of the pneuma, which is also Christ, being present in this mortal body of flesh and blood. But how are we to imagine it? Here, anybody’s guess is as good as mine. The idea may be something like this. In the human body of flesh and blood that Paul of course still has, there is also present that other stuff, the pneuma. Now elsewhere (1 Cor. 15:35–49) Paul connects the body of pneuma with a special physical appearance, which he calls “glory” (doxa) and which is something that may be directly seen (e.g. in the heavenly bodies, the sun, the moon and the stars). Thus the idea may be that pneuma as present in the human body of flesh and blood may also be concretely seen – e.g., as a certain shine or whatever (think of the modern movie E.T.) – at the same time as this human body is also very clearly betraying its mortality. Such a picture at least has the advantage of rendering intelligible Paul’s ideas about what will happen to pneuma-possessed people after the death of the mortal body. Then the pneuma, which was present in that body already here on earth, will take over completely, thereby transforming the same body into an altogether new entity where there is no longer any traces left of the mortal flesh and blood.24 For as Paul also states, “flesh and blood cannot inherit the kingdom of God” (1 Cor. 15:50). Let this conclude our speculation regarding the presence of the pneumatic stuff in a human body of flesh and blood.
24 Paul insists on this idea of “the same” body, compare 1 Cor. 15:52–53: “For this perishable body (to phtharton touto) must put on imperishability and this mortal body (to thn¯eton touto) (must) put on immortality. / But when this perishable body (to phtharton touto) has put on imperishability and this mortal body (to thn¯eton touto) has put on immortality […]”.
Philosophy of the Self in the Apostle Paul
193
Philosophy of the Self in Paul We have now reconstructed a reasonably coherent picture of our three entities, I1–2 and I−/+, of the role of bodies in connection with I1 and I2 and of the extent to which either qualifies as an “I”. Let us summarize and draw some conclusions concerning philosophy of the self in Paul. I have suggested that we should speak of two different types of self in Paul. One is a self of the type of I−. This is an abstract, empty and formal self (or “I”), which comes to light as being presupposed all through when Paul speaks of the change from one filled “I” (in this case I1) to another filled one (I2). It is also an “I” of heightened self-reflection on that same change, I+. I have suggested that there is a certain similarity, at least a structural one, with Kant’s substantive (self-identical) pole of identity behind all experience: one and the same “I”, which while remaining self-identical all through, in one act of interpretation becomes filled in one way, say as I1, and in another act of interpretation becomes filled in another way, say as I2 – and which is also able to describe that change. With this account I have already implied that in addition to the formal, self-identical “I”, I−/+, there also is another type of “I”, the filled “I” that is instantiated here by I1 and I2. That is a changing “I”, one that varies according to the act of interpretation with which it is filled. I have suggested that there is a certain similarity here, and in fact a rather strong one, with Charles Taylor’s hermeneutical self, which is the constantly changing result of a process of narrative construction of identity. In this connection it is worth noting that Paul himself constantly engages in narration when he aims to bring before the minds of his addressees his own Christ-believing self. Galatians chapters 1–2 are the best example, with Paul engaging in a bit of autobiography concerning his behaviour before being called by God, the conversion itself and its aftermath. But also such different letters as 1 Thessalonians (chapters 1–3) and 2 Corinthians (chapters 1–7) are filled up with narration about Paul’s own Christ-filled behaviour from the moment he left the two congregations until the time of writing. We may also note the extent to which Taylor’s manner of speaking in quasi-spatial terms of that narrative construction of identity is strongly matched by Paul, for whom the essence of the new identity consists in being “in” Christ.25 We have also gone into some detail concerning the relationship between the self and the body and the “I”-character of either of the two filled “I”s. On the one hand,
25 For instance, Taylor writes as follows: “I define who I am by defining where I speak from, in the family tree, in social space, in the geography of social statuses and functions, in my intimate relations to the ones I love, and also crucially in the space of moral and pneumatic orientation within which my most important defining relations are lived out” (Taylor 1985, 35, my italics). This could almost count as a description of what Paul does all through.
194
T. Engberg-Pedersen
the Pauline self of type I−/+ is not just identical with this or the other body. It is tied to a body, but it has its essence in enabling the person to reflect on and handle his or her body. To that extent it differs from the body. It stands at one remove from it in just the way any kind of reflection on “oneself” does that. On the other hand, the Pauline self of either type I1 or type I2 is through and through a bodily one. We have seen that this holds for I1, which is not at all surprising. It also holds for I2, which may seem more surprising. But that is one area where we must leave behind any modern understanding in order to grasp what Paul intended to say. Finally, we have seen that the “I”-character of either of the two filled “I”s differs rather drastically. We may conclude that there is in fact something like a real philosophy of the self in Paul. It is evidently not just a matter of a theoretical enterprise – philosophy in that sense. Instead, it is an almost explicit reflection on various elements in the self for the purposes of concrete use in situations of intellectual conflict. Apparently, in his battles not least against his ethnic, Jewish kinsmen Paul felt that he had little to fall back upon other than his own experience and the way this could be constructed. His own self – with its direct line through Christ to God – was his most powerful card. Is this something that Paul has just cooked up for his own purposes? Hardly. It is difficult to believe that a passage like the extremely careful one on akrasia in Romans 7 could have been written without its author having some fairly precise grasp of what the philosophers had been discussing in their various accounts of that phenomenon. Personally, I would say that the basic picture of two types of self that was summarized in my drawing is – in one way or another – derived from the Stoic analysis of valuation as encapsulated in their doctrine of oikei¯osis. But we know that the general philosophical situation in Paul’s time was far from clear-cut, and a broadly Middle Platonic context (which would in any case include a good bit of Stoicism) cannot be ruled out. As for Judaism and Hellenism, I trust that it has become clear that contrasting the two in relation to Paul is extraordinarily unhelpful. For instance, in Romans 7 he attempted to spell out the failures of the very best state anybody could live in before conversion to Christ – that of living under the Jewish law – by analysing that state in terms of the distinctly Greek discussion of the phenomenon of akrasia. Similarly, by drawing in Romans 8 on the specifically cognitive character of the pneuma, he developed the fundamentally apocalyptic (and hence Jewish) picture that he gave of the new state after conversion to Christ in a way that made it sit comfortably with contemporary Graeco-Roman philosophy. The result of all this is a quite sophisticated philosophy of the self in the first Christian thinker known to us – a very long time before Augustine. University of Copenhagen, Denmark
Two Kinds of Subjectivity in Augustine’s Confessions: Memory and Identity, and the Integrated Self Gerard J.P. O’Daly
There are several explorations of subjectivity in Augustine’s thought.1 One is the so-called Augustinian Cogito, where the phenomenon of my awareness of my mental activities (even the awareness of being mistaken) is the basis of epistemological claims about the attainability of certainty. But in one of his Cogito discussions, in Book 10 of his work On the Trinity, Augustine seems to take the argument further, developing the concept of the human mind as a self-thinking subject that is totally present to itself, an argument that is heavily influenced by Plotinus and Porphyry.2 To these one should add Augustine’s accounts of inwardness, the concepts of the mind as an “inner space” and of the “inward and upward” turn of its discovery of itself and God, in texts where the themes of interiority and mental ascent are, once again, influenced by the Neoplatonists.3 This paper considers two further exercises in Augustinian subjectivity, one familiar, on memory and identity, one less so, on the congruity (or lack of it) between “inner” and “outer” in certain emotionallycharged states. In both cases, the most important texts are in the Confessions. In my conclusion, I consider briefly some general implications of these texts.
1 Versions of this paper were delivered at a conference in honour of Peter Brown at University College Dublin in 2001, and at a CNRS colloquium on aspects of the self in ancient philosophy at Paris-Villejuif in 2004. 2 On Augustine and Descartes’ Cogito see Gareth B. Matthews, Thought’s Ego in Augustine and Descartes, Ithaca, NY: Cornell University Press 1992; Stephen Menn, Descartes and Augustine, Cambridge: Cambridge University Press 1998; Emmanuel Bermon, Le Cogito dans la pensée de Saint Augustin, Paris: Vrin 2001 (the fullest discussion of Augustine, also containing detailed comparisons with Husserl); Richard Sorabji, Self: Ancient and Modern Insights About Individuality, Life, and Death, Oxford: Oxford University Press 2006, 212–229 (an important discussion of On the Trinity 10); De trin. 9 and 10: see Gerard O’Daly, “Subject and Substance: Self-Knowledge in Augustine’s On the Trinity”, in Matthias Vorwerk (ed.), Augustine and Augustinianism. Contributions on Augustinian Philosophy and Its Reception. Catholic University of America School of Philosophy Fall Lecture Series 2006 (Washington, DC, forthcoming). 3 The terms “inner space” and “inward and upward” turn are taken from Philip Cary, Augustine’s Invention of the Inner Self. The Legacy of a Christian Platonist (New York: Oxford University Press 2000), whose account of Augustine’s views is persuasive, but who underestimates the influence of Plotinus.
P. Remes, J. Sihvola (eds.) Ancient Philosophy of the Self, © Springer Science+Business Media B.V. 2008
195
196
G.J.P. O'Daly
Augustine on Memory and Identity The discussion of the phenomenal memory of a certain Simplicius in On the Nature and Origin of the Soul 4.9–10 raises a question about identity. The issue is: what has changed, if Simplicius knows something about himself that, though being apparently the same person, he did not know before – if, for instance, he realizes the power and extent of his memory, which enables him to recite passages of Cicero or Virgil from any given place in the works chosen by his questioners? Augustine makes heavy weather of this phenomenon. He realizes that it raises the same question as does the case of forgetting what I once knew. Something has changed in me, but if this is a change of our identity, of “what we are”, then we are changing all the time. A concept of the unconscious or of latent awareness – such as that of Plotinus – might have solved Augustine’s dilemma here. Augustine employs such a concept in On the Trinity 14, arguing that when the mind is not actively thinking of itself, it none the less knows itself, just as someone who knows geometry is not ignorant of it if he is not actually thinking of it. His knowledge is stored in his memory.4 Alternatively, Augustine might have tried to distinguish between the kind of memory or knowledge that makes me what I am, and other kinds that do not. As we shall see, he appears to make this distinction in other texts. I turn now to some aspects of Augustine’s account of memory, which raise the question of memory’s relation to the individual’s identity. Of the “power of memory” (vis […] memoriae) Augustine says “my mind is this thing; I am this thing” (et hoc animus est, et hoc ego ipse sum, Conf. 10.17.26).5 On a minimalist reading of this kind of assertion, we could argue that Augustine is talking here about the status of memory in humans as a mental activity, as the mind engaged in certain activities. Previously I adopted such a minimalist reading, suggesting that ‘mind’ and ‘memory’ have here, in Frege’s terms, the same reference (Bedeutung) but a different sense (Sinn).6 I am not sure now that this fully exploits what Augustine means when he says that my mind – that I – am my memory. To begin with, Augustine considers parts of his life of which he has no conscious memory, such as infancy, to be no more “his” than the period prior to his birth: “what does it matter to me now, seeing as I cannot recall any traces of it?” (quid mihi iam cum eo est, cuius nulla vestigia recolo? Conf. 1.7.12). So “I would be […] reluctant to consider that life to be of a piece with the one I now live in this world (piget me adnumerare huic vitae meae, quam vivo in hoc saeculo, ibid.)”. This puzzle remains. My sense of what I can remember defines the parameters of what I can
4 Eric R. Dodds, The Ancient Concept of Progress and Other Essays on Greek Literature and Belief, Oxford: Oxford University Press 1973, 135–136; Gerard O’Daly, Augustine’s Philosophy of Mind, Berkeley & Los Angeles, CA: University of California Press/London: Duckworth 1987, 149–150. See Plotinus, Enn. IV.4 [28] 8, IV.8 [6] 8; Augustine on latent memory, De trin. 14.8–9. 5 Here and in the rest of this paper I use (modifying it occasionally) the excellent recent translation of Conf. by Philip Burton: Augustine: The Confessions, London: Everyman 2001. 6 O’Daly 1987, 135–136.
Two Kinds of Subjectivity in Augustine’s Confessions
197
intelligibly call “mine”. Augustine appears to be saying that what I cannot remember does not contribute to my identity, even if it has been part of my experience.7 Consider another passage in Augustine’s account of memory in Book 10 of the Confessions. It is I who remember, I, the mind. It is not so strange if what I am not is far from me; but what is closer to myself than I am? And yet the power of my memory is not comprehended by me, since I cannot name myself without it. (Conf. 10.16.25)8
The context here is the phenomenon of forgetting. Martha Nussbaum observes on this passage: “A really successful dissociation of the self from memory would be a total loss of the self – thus of all activities to which a sense of one’s identity is important”. And she adds that this sense of oneself as a temporal being with a history is crucial to the individual’s progress in the quest for goodness and knowledge (of self, of God), even if “this temporal history may be a nonnecessary property of his soul”.9 This is a nonPlatonic view of human moral progress, in which temporal memory is crucial to moral self-knowledge, because memory recalls what we were, what we have become, and in what ways the force of habit still constrains us. Thus, in Confessions (in a part of the Confessions to which I return later in this paper), reflecting on Monnica’s death and his inability to weep or be helped by prayer, Augustine writes: […] All day long my deep grief was hidden and with a disturbed mind I besought you, as best I could, to heal my grief. But you did not heal me, perhaps wishing to impress on my memory by this one example how the chain of every habit is opposed to a mind which feeds on the Word in whom there is no deceit (Conf. 9.12.32)10
That is to say, we are vulnerable, not merely to our own imperfections, but also to divine grace, and particularly to divine use of our memory as a means of understanding both ourselves and God. At the same time, Augustine stresses our lack of self-knowledge, the opacity to ourselves of what we really are. For him, this is a consequence of original sin. So is our lack of transparency to each other.11 The early sections of Confessions 10 explore aspects of this. Confessions 10.2.2 speaks ambiguously of the “depths of human conscience” (or consciousness, abyssus humanae conscientiae).12 7 On memory and personal identity over time see Sorabji 2006, 94–111, here 99–100 on Augustine. For memory in the afterlife see the passages in De civ. D. 22 and De trin. 14 discussed below. 8 “ego sum qui memini, ego animus. non ita mirum, si a me longe est quidquid ego non sum: quid enim propinquius me ipso mihi? Et ecce memoriae meae vis non comprehenditur a me, cum ipsum me non dicam praeter illam”. 9 Martha Nussbaum, “Augustine and Dante on the Ascent of Love”, in Gareth B. Matthews (ed.), The Augustinian Tradition, Berkeley, CA: University of California Press 1999, 61–90, see 68. 10 “[…] toto die graviter in occulto maestus eram et mente turbata rogabam te, ut poteram, quo sanares dolorem meum, nec faciebas, credo, conmendans memoriae meae vel hoc uno documento omnis consuetudinis vinculum etiam adversus mentem, quae iam non fallaci verbo pascitur”. 11 Conf. 10.8.15, 10.16.25, 10.32.48, 10.37.61, 13.31.46. 12 For the ambiguity see Cornelius Mayer, “Conscientia”, in C. Mayer (ed.), Augustinus-Lexikon 1, Basel: Schwabe 1994, cols. 1218–1228.
198
G.J.P. O'Daly
Others cannot know, and can only believe, on the basis of what I confess, “what I myself am within” (quid ipse intus sim, Conf. 10.3.4). We seem to be dealing with a concept of the private ‘me’, of a ‘me’ known only to God. This is to be distinguished from the ‘inner man’ (homo interior) of Conf. 10.6.9, who knows that he is a created being, just as it is to be distinguished from memory. Now, although these opening chapters of Conf. 10 immediately precede the account of memory, they are a prelude to the book in general, and in particular they anticipate the moral self-examination of Conf. 10.28.39–43.70. For the self which Augustine finds puzzling at the beginning of Conf. 10 is the morally complex and contradictory self analysed in these later chapters, where dreams (10.30.41), curiosity (10.35.54–7), and attitudes to the approval of others (10.37.61) surprise and dismay Augustine. The self-analysis is an examination of conscience. O’Donovan stresses Augustine’s moralizing interpretation of Platonic spirituality through the identification of the inner self with conscience-consciousness (conscientia). Self-scrutiny becomes a constant spiritual discipline, of which Conf. 10 is perhaps the prime example.13 In paradise, by contrast, our lack of self-knowledge and of knowledge of each other will give way to reciprocal knowing: “the thoughts of each of us will lie open to the other” (patebunt […] cogitationes nostrae invicem nobis, De civ. D. 22.29). We will see things through the Spirit. But Augustine suggests elsewhere, adapting 1 Corinthians 2.11–12, that when we do so, it is not we who see, but God who sees in us (Conf. 13.31.46). What Augustine appears to be arguing is the Pauline belief that I should aspire to a state where “my” vision is supplanted by God’s vision in me. However, it is clear that Augustine does not wish to suggest that this involves a loss of personal identity. For he also maintains that we continue to have memory of temporal things in the afterlife. We will recall the contents of our temporal religious faith then, just as we now recall past events of our lives (De trin. 14.4–5). Faith, and presumably other recollected experiences of our lives, will leave “a trace in the imagination” (imaginarium vestigium, De trin. 14.5). Why does Augustine maintain this? Passages like De civ. D. 22.20–1 and 22.30 suggest that these traces of our temporal lives have something to do with the fact that our resurrected state is bodily, that our final state, though eternal, will be durational, with memories. The damned will have the same sense of duration and memory as the saved.14 At the same time, Augustine suggests that awareness of the difference between, for instance, former faith and present knowledge (seeing God face to face) marks the continuing distinction between creator and creature. Awareness of what I once was, compared with awareness of what I now am, is not merely a feature of this life (as Conf. 9.12.32, referred to above, shows), but also of the afterlife. In Confessions 10.40.65 Augustine argues that memory entails a ‘force’ (vis) of the mind, distinguished from ego: “nor was it I who discovered […] nor was it I who did all this – that is, the force that is in me by which I did it” (nec ego ipse inventor
13
Oliver O’Donovan, The Problem of Self-Love in St. Augustine, New Haven, CT: Yale University Press 1980, 70–72. 14 For the damned see De civ. D. 21.9–13, 17–24.
Two Kinds of Subjectivity in Augustine’s Confessions
199
[…] nec ego ipse, cum haec agerem, id est vis mea, qua id agebam). This “force” scans sense-perception and ideas, and is aware that it is not God – “nor were you that force” (nec ipsa eras tu) – but that God is (a) the light or teacher of the mind, working through the “force” whereby the mind remembers, and (b) the “place” in which the mind subsists, and finds coherence and stability. God is the sine qua non of the memory’s operation. Augustine’s characteristic illumination-doctrine is implied here.15 The “I” that memory reveals is not an autonomous agent. Confessions 10.2 summarizes, and clarifies, some of these ideas in a striking way, by considering the purpose of writing confessions. It is not to reveal something about me to God of which God would otherwise have been unaware – for God is omniscient – but to prevent my hiding God from myself, and hence to prompt my moral self-awareness. In a context of self-scrutiny, confession uncovers some of the secrets of myself to me. My sense of my moral failings and inadequacy is highlighted by an understanding of divine perfection, and of the divine as an object of love and longing. Confession is exposure to divine goodness. Self-examination and self-revelation are means to this end. The language has Platonic overtones, but it also reminds us of the kind of self-scrutiny that we find, for instance, in Seneca. Augustine has a concept of identity, and memory plays a crucial role in it. But some kinds of memory are more important than others. The memories that matter are those that reveal to us our moral condition, and our progress in understanding our relationship to the divine.
Friendship and the Self I turn now to what I read as two related accounts of friendship (amicitia) in the Confessions, accounts that express something important about Augustine’s sense of self in relation to another person. One of these is the well-known narrative of the death of a friend of Augustine’s youth in Book 4 of the Confessions, and Augustine’s reaction to it. The other is not usually read as an account of amicitia, but I shall argue that it is: I refer to Augustine’s description of the death of his mother Monnica in Book 9, and his reaction to it. What we have to do with in these two episodes, as is frequently the case in the Confessions, is the use of narrative, with reflective inserts, as a way of philosophizing. Philosophically delineated friendship, properly understood (a non-selfish, non-possessive relationship), is Augustine’s model for inter-personal relations.16
15
O’Daly 1987, 204–207. Cicero’s On Friendship is Augustine’s principal source for philosophical ideas on friendship. On Cicero and his Greek sources see Jonathan G.F. Powell, Cicero: On Friendship and the Dream of Scipio, Warminster: Aris & Philips 1990. Christian friendship in Augustine’s day, see Carolynne White, Christian Friendship in the Fourth Century, Cambridge: Cambridge University Press 1992. Friendship in Augustine, above all, Letter 258; Marie A. MacNamara, Friends and Friendship for St. Augustine, Staten Island, NY: Alba House 1964. 16
200
G.J.P. O'Daly
In Confessions 9, Augustine and Monnica exemplify this approved friendship (the contrast with Confessions 4 is deliberate, as we will see), and their dialogue (the so-called Ostia vision) conjures up an image of what eternal life must be like. The insight that the conversation of friends achieves has a similar function to the uses of memory in moral self-awareness discussed above.17 Now, in Confessions 4 the easing role of grief at his friend’s death reveals what I shall call an inner/outer congruity, that is, there is a consistency between Augustine’s feelings and his external behaviour. But this congruity is then analysed as flawed and unstable. In Confessions 9, the difficulty of grieving at Monnica’s death reveals an inner/outer incongruity that is only partly overcome by the release of grief through weeping that is (we must assume) divinely granted, and that is finally resolved in an objectifying transformation of grief, as Augustine realizes what true friendship is. But before we turn to these two passages in more detail, let us recall some relevant aspects of philosophical friendship, particularly in Cicero’s account in his On Friendship (De Amicitia), which influences Augustine strongly, both verbally and conceptually. The characteristic features of friendship that Cicero enumerates are considered by Augustine to be those of the virtuous life in general. Friendship provides stability (stabilitas, constantia), friends see into each other’s heart (apertum pecus), the life of true friends exemplifies the workings of rationality in their lives. Augustine sees in friendship the true realization of the scope of our moral and intellectual development.18 By contrast, false friendship – like evil – is, for Augustine, “incomprehensible” (investigabilis, Conf. 2.9.17), ultimately not accessible to rational investigation. Moreover, discussions of friendship in the tradition of which Cicero’s On Friendship and Augustine’s accounts are a part express the friendship relation in terms of one’s relation to oneself: one merges one’s identity with the friend’s; a friend is a second “me”; friends have all things in common; two friends share one soul; one can talk to a friend as if with oneself – and so on.19 In Confessions 4.4.7, Augustine characterizises his friendship as false friendship (by contrast with the true variety, understood here in explicit Christian terms, with
17 Dialogue permeates the Confessions. Dialogue between God and Augustine is the means of unfolding the workings of divine grace in a human life. Even the conversion-scene (Conf. 8), intensely personal though it is, has dramatic dialogue features, involving Augustine and others: his closest friend Alypius (8.8.19), his old “girl-friends” (amicae) the vices (8.11.26), personified Continence (8.11.27), Scripture (8.12.29). We have here one real person, two personifications, and one interaction with a text: but the element of dialogue is common to them all. The act of will that is conversion – a subjective, personal moment – is thus contextualized socially. Its consequences are implemented socially, in Augustine’s change of career and lifestyle. Augustine’s conversion may be personal, but it is in no sense private. It carries Alypius in its wake, and Augustine’s completes its narration by the involvement of Monnica in it (8.12.30). 18 Stability, see Cicero, De am. 62–65. “Open heart”, ibid. 97. Friendship and the good/wise life, ibid. 18–21 and passim. Augustine on friendship and the life of reason/wisdom, see Soliloquies 1.7, 20, 22; Ep. 10 and 258. 19 For references to these topoi see White 1992, 273.
Two Kinds of Subjectivity in Augustine’s Confessions
201
a link made between love and the Holy Spirit), but none the less “sweet” because of the “interests” shared, and “ripened” (cocta), by those interests, namely, the shared Manichaeism of the two friends. When his friend dies, Augustine writes that he could not account for his sense of grief and loss (4.4.9). Yet he grieves spontaneously. Tears flow. There is an inner/outer congruity, but it is not understood by Augustine. In retrospect, he sees his wretchedness as an indulgence, based on too great attachment (4.5.10). His “other self” – the traditional description of the friend since Aristotle20 – was dead. Did he stay alive to keep his friend alive? “I did not wish to live, a half” (4.6.11). When he reviews this passage late in his life, in Retractations 2.6.2 he finds it to be “a frivolous oratorical display” (declamatio levis), not a “serious avowal” (gravis confessio). There is a correlation, he implies, between false sensibility and the falsity of his rhetoric.21 To return to the Confessions narrative: after his friend’s death, “[…] everything was an object of horror […] I had become to myself a place of unhappiness […] my error was my god” (horrebant omnia […] et ego mihi remanseram infelix locus […] error meus erat deus meus, Conf. 4.7.12). The “place of unhappiness” (or, possibly, “barren place”) here should be linked to Augustine’s descriptions elsewhere of the human condition as a “place of want” or “place of unlikeness”.22 In retrospect, he finds this state of mind compounded of “tainted emotions” (Conf. 4.6.11).23 Grief, nevertheless, runs its course, subsides (Conf. 4.8.13). Augustine has not understood the nature of mortality. Other living friends provide solace, but his love for them – again seen in retrospect – is a substitute for love of God. His condition is a “great myth, a long lie” (ingens fabula et longum mendacium, Conf. 4.8.13). He loved a mortal person as if he would never die, “pouring out his soul on to the sand” (ibid.): there was no stability (constantia) in his condition (see above). Yet into this highly self-critical narrative Augustine inserts (4.8.13) a description of friendship that is uncomplicated and spontaneous, and that is not, for once, evaluative. It is the apparent acceptability of friendship of this kind – the sense in which it is, in part, understandable – that makes Augustine’s critique of his friendship with his dead friend complex. To a certain extent, it imitated true friendship.24 It was reciprocal. Augustine and his friend loved each other (amare-redamare, 4.9.14); they lived Cicero’s friendship, wishing the other well (benivolentia, 4.8.13); their plurality had become a unity (ex pluribus unum, 4.8.13),25 even if, in retrospect, it is seen to have been inadequate. 20
Arist. EN IX 4, 1166a 31–32 (for further references to Aristotle and others see Powell 1990, 91). Cf. Augustine’s critique of oratory in Conf. 1.9.14, 4.2.2, etc. 22 Burton translates infelix locus as “a barren land”. For the “place of want” (regio egestatis) see Conf. 2.10.18; for “place of unlikeness” (in regione dissimilitudinis), Conf. 7.10.16, influenced by Plotinus 1.8.[57]13.16–17, which derives in turn from Plotinus’ reading of Plato, Statesman 273d6–e1. 23 “me mundas a talium affectionum inmunditia”, lit. “the taint of such feelings”, Conf. 4.6.11. 24 The subtext here is Augustine’s belief that evil is a “perverse imitation” of the good, a travesty of self-love. For this theme see O’Donovan 1980, 93–111. 25 Cf. Cicero, De am. 92 (unus […] animus […] ex pluribus); benivolentia in De am. 26; redamare, ibid. 49. 21
202
G.J.P. O'Daly
The narrative of the death of his mother Monnica in Confessions 9 is a meditation (not often recognized as such) on Christian friendship. Christian attitudes forbidding expression of grief and mourning26 create tension in Augustine, who initially will not, and subsequently cannot, grieve for his mother: he “weeps” internally, but sheds no tears (9.12.29). The boy Adeodatus, on the other hand, can weep unselfconsciously (ibid.). In 9.12.30 Augustine defines his and Monnica’s relationship in terms of the ‘habit’ (consuetudo) – not necessarily a positive term – of friendship. At the end of the passage the theme of friendship (the “one life” of Augustine and Monnica) dominates.27 His sense of loss is linked to his feelings for his dead friend in Confessions 4. In the later narrative the metaphors for his grief are vivid: he is wounded, torn to pieces (9.12.31, 9.13.34). What, if anything, has changed in the converted Augustine of Book 9? As in Confessions 4, Augustine’s reflective narrative is self-critical. “I rebuked my affection […] for its lack of toughness” (increpabam mollitiam affectus mei, 9.12.31). But there is a radical change, due to his suppression of his grief. Now there is a contrast and tension between his inner and outer selves, between the unemotional exterior and the turbulent interior. His sense of his human weakness causes in him a “twofold sadness” because he grieves for the very existence of his grief.28 At Monnica’s funeral Augustine still experiences this inner/outer incongruity. He is still dominated by his passions, by habit (9.12.31). Later, however, he will weep. The “carnal passion” (carnalis affectus, 9.13.34) of his incongruous state gives way to “another kind of tears” (ibid.) for Monnica and for himself (de illa et pro illa, de me et pro me, 9.12.33). Augustine alludes here to the view that ritual expressions of grief are a psychological comfort for the survivors.29 But more is happening at this stage of the narrative. The linkage of grief for himself with grief for his mother reworks the theme of the friend as one’s other self. Moreover, in weeping for Monnica and for himself, he weeps for imperfect humans, who are mortal, and need divine mercy. Monnica’s death and Augustine’s grief are objectified. In the increasingly liturgical context of the prayer for the dead (which Augustine extends to his dead father) in 9.13.35–7, she becomes God’s servant (famula tua, 9.13.37) – at one level, depersonalized: but she is also given an identity separate from the mother-son relationship by being named (as appropriately in the
26 Attitudes shared with, and often influenced by, Greek, especially Stoic, philosophical positions. see J.H. David Scourfield, Consoling Heliodorus: A Commentary on Jerome Letter 60, Oxford: Clarendon 1993: 22, 130–132. 27 Cf. Conf. 4.8.13. 28 His grief in Conf. 9.12.31 is the passion of tristitia. Cf. the earlier instances in this passage of the language of the passions: “the softness of my affection” (mollitiam affectus mei), “flood of sorrow” (fluxum maeroris). 29 Cf. De civ. D. 1.12–13; Gerard O’Daly, Augustine’s City of God: A Reader’s Guide, 2nd edn, Oxford: Oxford University Press 2004, 76–77. For Jerome’s attitude to the appropriateness of Christian grief for the dead, in many respects like Augustine’s, but more inclined to accept expression of grief, see his Ep. 60.2, 7, and Scourfield 1993, ad loc.
Two Kinds of Subjectivity in Augustine’s Confessions
203
liturgical context here as it would be in a tomb-inscription) “Monnica” here (ibid.), the only time in Augustine’s writings that she is so named. The “chain of faith” (vinculum fidei, 9.13.36) supersedes the “chain of habit” (vinculum consuetudinis) of 9.12.32. The emotion of grief is not eradicated, but transformed: the inner/outer tension is resolved. Augustine has achieved a metamorphosis of grief that he could not have achieved at the stage of his friendship of Confessions 4. By situating his grief in the appropriate religious context, he overcomes its selfabsorption, overcomes also the “unlikeness” (dissimilitudo) between his inner and outer selves (see above): we could speak of an integrated self. As well as this integration of the inner and the outer, there is a realization that we must act in the world, in relation to others, and not in isolation. At the same time, the experience is humanizing, for, rather than suppressing his grief, he can now appropriately give way to it. The experience is one of self-transformation, through realization of what true friendship entails.
Concluding Thoughts The exercise of memory is quintessentially subjective and personal (if one sets aside such phenomena as folk memory and collectively remembered historical experience, which Augustine does not consider). In that sense, Augustine’s account of memory in the Confessions and some other texts is like those kinds of selfscrutiny – the Cogito and the inward turn and ascent – mentioned at the beginning of this paper. On the other hand, the theme of inner/outer congruity and its opposite in Confessions 4 and 9 depends on an account of the self in relation to others. But common to all the kinds of subjectivity referred to or discussed in this paper, with the exception of the Cogito (to which, in turn, the discussion of self-knowledge in On the Trinity 9 and 10 is an exception), is the linking of this strand of reflection to the individual’s moral and religious progress. Augustine is often less concerned to formulate a concept of the self than to exploit the moral implications of various forms of subjectively oriented discourse. University College London, UK
The Self as Enemy, the Self as Divine: A Crossroads in the Development of Islamic Anthropology Taneli Kukkonen
Towards the end of his life the Muslim theologian, jurist, and mystic Abu- H.a-mid Muhammad al-Ghaza-lī (1058–1111 CE) composed a streamlined version of his . principal work, The Revivification of the Religious Sciences (Ih. ya-’ ‘ulu-m al-dīn). The original title of the work, written in Ghaza-lī’s native Persian, is The Chemistry of Happiness: yet the common European custom of translating the first word as “alchemy” is not entirely out of place. Ghaza-lī’s stated intention, after all, is to isolate the human being’s most valuable component from amidst the many things out of which the individual is composed. He likens this procedure to the process whereby base metals are transmuted into gold.1 Why this interest in the human constitution? Ghaza-lī draws on a Qur’a-nic citation and a saying of the Prophet. The Qur’a-nic verse goes as follows: “We will show them Our signs in the horizons and in themselves (fī anfusi-him), so that the truth will become known to them.” (Q. 41:53.) This comes across as so much creation theology and many Muslim thinkers, Ghaza-lī included, would interpret it as an argument from design of sorts.2 From the way in which the body’s members and organs are put together to the way in which they finely co-operate to, finally, the way in which they aid us in coming to terms with outward reality (the “horizons” mentioned in the verse), there is no end to signs of intelligent design in creation.3 Further implications may be teased out once it is recognised that the grammatical
Kīmiya--yi sa‘a-dat, 2 vols., ed. H.usayn Kh.adīv-jam, Teheran 1983 (hereafter Kīmiya--yi), 1:6. An Arabic version of the introductory chapter on self-knowledge has been edited by Muh.ammad ‘Abd al-‘Alīm (Cairo, 1986): I have checked my findings against this text but will refrain from citing it, since it is uncertain if the translation is in al-Ghaza-lī’s own hand (or even authorised by him). 2 See, e.g., Al-maqs.ad al-asna- fī sharh. ma‘a-nī asma-’ Alla-h al-h.usna- (“The Highest Purpose in Explaining the Meanings of the Beautiful Names of God”), ed. F.A. Shehadi, Beirut: Da-r al-Mashriq 1982, 107 (–hereafter Maqs.ad). 3 See Maqs.ad, 106–107; Ih.ya-’ ‘ulu-m al-dīn, 5 vols. (henceforth Ih.ya-’), Beirut: Da-r al-kutub al-‘ilmiyyah, 2002, bk. 21, ch. 2, 3:6.13–21; and the Book of Gratitude (=Ih.ya-’, bk. 32), passim. 1
P. Remes, J. Sihvola (eds.) Ancient Philosophy of the Self, © Springer Science+Business Media B.V. 2008
205
206
T. Kukkonen
placeholder for ‘self’ in this context, i.e. nafs, is the same expression that is customarily used to designate ‘soul’.4 On this interpretation of the verse it is the human soul that is filled with signs of God’s own truth. It is the Prophetic saying, which has the unmistakable ring of the Delphic maxim to it, that really attracts our attention. In Ghaza-lī’s phrasing this goes as follows: “One who knows himself (or again “his soul”, nafs) will come to know his Lord”.5 And what will such self-examination encompass? According to Ghaza-lī the following questions are in order: What art thou in thyself, and from whence hast thou come? Whither art thou going, and for what purpose hast thou come to tarry here awhile, and in what does thy real happiness and misery consist? Some of thy attributes are those of cattle, some of predators, some of devils, and some of angels: thou hast to find out which of these represent the reality of thy substance and which are extraneous and handed to thee on lease. Till thou knowest this, thou canst not find out where thy real happiness lies. (Kīmiya--yi, 1:13–14)6
My purpose in this essay is to unpack this concise mission statement and to outline the path Ghaza-lī takes in answering the questions he poses. An examination of Ghaza-lī’s Revivification and related works reveals the extent to which he came under the spell of the ancient ideal of self-knowledge. At the same time, idiosyncratic features in Ghaza-lī’s account indicate a departure from that self-same tradition. If it is largely Ghaza-lī whom we have to thank for the popularity of the self-examination theme in later Islamic literature as well as for the way it became couched in Platonic and Peripatetic terminology, as I believe we do, then it also remains true that a peculiarly Islamic perspective informs Ghaza-lī’s reading of the philosophers. Ghaza-lī’s thought, I will argue, is representative of a particular moment in the development of Islamic anthropology, one at which the Greek intellectualist ideal as developed by Avicenna (980–1037) begins to meld with the relentless self-questioning that characterises the moralist strand of Muslim piety. Later Islamic thinkers will more confidently speak of the self, of the “I”, and of our unhindered immediate access to both: but Ghaza-lī occupies a unique position in holding at once that self-knowledge is absolutely crucial for our now and future well-being, and that it is an exceedingly difficult task.
4
This is the way the verse is rendered, e.g., in N.J. Dawood’s popular Penguin interpretation. For variations on this h.adīth see Mīza-n al-‘amal (“The Criterion of Action” – hereafter Mīza-n), ed. S. Dunya-, Cairo Da-r al-ma‘a-rif bi al-mis.r 1964b, 200; Ma‘a-rij al-quds (“The Jerusalem Ascent” – hereafter Ma‘a-rij), Cairo: al-Maktaba al-tija-riyya al-kubrī 1963, 2–3; on the authorship of the latter work, n. 13 below. 6 The antiquated but charming English translation derives from The Alchemy of Happiness, translated by C. Field (London: MES 1991), having been emended for precision. 5
The Self as Enemy, the Self as Divine
207
Cognitive Powers Ghaza-lī’s unquestionable theocentrism allows him to draw some preliminary conclusions based on the h. adīth just cited. Human life properly conceived is directed towards the divine, with no other path leading to true happiness.7 At the same time, a certain ethical naturalism is assumed. Even if aligning one’s life with God’s revealed will constitutes the believer’s mission, the fact that this course of action is conducive to felicity is grounded in the correlation of God’s prescriptions to what one requires in one’s true being (h. aqīqa) or essence (dha-t).8 And, while dha-t itself is a notoriously slippery term, with a semantic field covering everything from ‘essence’ to ‘innermost being’ to ‘self’ in the literature of the period,9 this naturalistic tendency does help to explain why both chemical/analytic and introspective overtones should frame Ghaza-lī’s quest for self-knowledge. If it is the human essence that is properly designated by the term nafs, then it is this that should command our attention more than any accidental features that may or may not accrue to the soul or to embodied life more generally.10 But this terminological elision also suggests an immediate problem, for thought no less than for translation. Is Ghaza-lī merely making the point that we are not our bodies? Such a claim would scarcely raise an eyebrow among the Neoplatonically inclined Arabic Aristotelians (even if from the point of view of mainstream Ash’arite theology it would be radical: although the details were disputed, the theologians generally treated soul as a corporeal principle, if one of a particularly subtle nature). Ghaza-lī’s preliminary remarks on the substance of soul indicate that he does indeed take this view: whatever the reality of nafs may otherwise turn out to be, minimally we are dealing with an incorporeal principle.11 By way of negation it can be established that this essence, like the Truth from whence it stems, is indivisible and immaterial in its essence, without a bodily or perceptible quality that would attach to it. It is at once everywhere that its influence extends, and yet without a specific location that one could pinpoint (Kīmiya--yi, 1:16–18, 1:50–52).
See, further Ih.ya-’, bk. 21, 3:6.8–10 and bk. 22, 3:57.24–28, both of which cite verse 51:5 from the Qur’a-n: “I created Jinn and man only to serve Me.” 8 See, e.g., the introduction to the Criterion of Action: Mīza-n, 180–181. 9 See Fazlur Rahman, “Dha-t”, in Th. Bianquis, E.J. Donzel and W. Heinrichs (eds.), The Encyclopaedia of Islam. New Edition, vol. 2, Leiden: E. J. Brill 1965, 220. 10 Despite the many gifts God has given us in the form of our bodily constitution, we are not to identify primarily with the body, as even the beasts are aware of their limbs and organs and know that they be belong to them: Kīmiya--yi, 1:13. For this overall picture see also Ghaza-lī’s polemical treatise against the Isma-‘īlīs, Fad. a-’ih. al-ba-.tiniyya wa fad. a-’il al-mustaz. hiriyya (hereafter Fad. a-’ih.), ed. ‘A. Badawī, Cairo: Al-maktaba al-‘arabiyya 1964a, 198–200. 11 Kīmiya--yi, 1:16; see further Timothy J. Gianotti, Al-Ghaza-lī’s Unspeakable Doctrine of the Soul, Leiden: E. J. Brill 2001, 68–87; on Ghaza-lī’s relation to the Ash‘arites, see Richard M. Frank, Al-Ghazali and the Ash‘arite School, Durham, NC: Duke University Press 1994. 7
208
T. Kukkonen
Further than this it is difficult to venture. According to Ghaza-lī no fewer than four expressions are commonly used in association with the sought spiritual subtlety (lat. īfa), and this variegation easily leads to exasperation: The variation in these expressions, along with what pertains to them, has left most scholars bewildered. Consequently, you will see them discussing [these] notions and say: “This is the notion of the intellect (‘aql); this is the notion of the spirit (ru-h. ); this is the notion of the heart (qalb); this is the notion of the soul (nafs).” Yet [these] thinkers do not perceive the actual differences between these names. (Ih. ya-’, bk. 21, 3:5.21–23; cf. 3:4.2–3.)
In Ghaza-lī’s analysis, the problem lies in confusing the reality (h. aqīqa) of the essence (dha-t) under consideration with the bodily functions and organs with which it is regularly associated. The heart is correctly thought to enjoy a connection with the entity Ghaza-lī means to pinpoint; so is the animating spirit which according to the physicians courses through our veins; so are the twin phenomena of intellectual apprehension and the deliberate pursuit of worldly happiness. Yet none of these functions is strictly identical with the principle with which they are associated. Instead, the meanings of the terms add up to five in all: inasmuch as these different expressions point to diverse functions, they refer to distinct features of our existence, but inasmuch as their underlying principle and coordinator is one, all four names refer to a single entity.12 For this concealed reality (al-h. aqīqa al-ba-.tina) Ghaza-lī’s preferred expression is ‘heart’, and in both the Chemistry and in the Revivification we see a corresponding move from a consideration of the soul/self (nafs) to one concerning the “wonders of the heart” (‘aja-’ib al-qalb). The move represents a strategy common to Ghaza-lī’s later career, in that he attempts to distance himself from Greek-derived philosophical terminology by putting forward an Arabic-Islamic alternative whenever he can.13 This should not be allowed to distract from the main issue; Ghaza-lī’s “pectoral psychology”, to use Ebrahim Moosa’s term, is no less a psychology for being pectoral.14
Ih.ya-’, bk. 21, 3:5.23–30; cf. Ma‘a-rij, 10–13. This calculus makes understandable al-Ghaza-lī’s otherwise strange arithmetic, according to which four expressions, each of which has two meanings, add up to five meanings in all. This can only mean that there are in reality five intentional objects, i.e. five objects of thought in the minds of the learned, even if the people involved do not recognise this but instead keep on using fewer or more numerous expressions. On outward, mental, and verbal existence see Maqs.ad, 18–19. 13 In the Jerusalem Ascent the term nafs is retained, which has led some scholars to doubt its authenticity. A survey of the different opinions on the matter, along with a defence of the attribution of Ma‘a-rij to al-Ghaza-lī, is now offered by Frank Griffel, review article of Gianotti (2001), Journal of the American Oriental Society 124/1(2004), 108–111. Drawing on Afifi al-Akiti, “The Three Properties of Prophethood in Certain Works of Avicenna and al-G.aza-lī” (in Jon McGinnis and David C. Reisman (eds.), Interpreting Avicenna: Science and Philosophy in Medieval Islam, Leiden: E. J. Brill 2004), Griffel points out that al-Ghaza-lī’s attempts to dress up Greek philosophical psychology in Islamic garb are less than systematic. 14 See Ebrahim Moosa, Ghaza-lī and the Poetics of Imagination, Chapel Hill, NC: The University of North Carolina Press 2005, 224 ff. 12
The Self as Enemy, the Self as Divine
209
Just how cosmetic the terminological shift is that Ghaza-lī initiates can be gleaned from the way his treatise On the Marvels of the Heart compares with later works of Islamic psychology. We have already mentioned how nafs serves a dual function as both self (in the reflexive sense) and soul: a later kala-m thinker such as Fakhr al-Dīn al-Ra-zī (d. 1210 CE) can exploit this very linguistic ambivalence in his own work On the Soul and the Spirit, with an Explication of their Powers, where he argues for the functional identity of self-examination and the counting off of the various psychic faculties (Gr. dynamis/Ar. quwwa) that we possess. For what could self-examination be, except for knowledge of those capacities and movements of the soul that remain hidden to the outward eye but are manifest to inward reflection?15 This in fact is the line adopted by Ghaza-lī as well, in the Ih. ya-’ and elsewhere. The capacities in question are branded by him “the hosts of the heart” (junu-d al-qalb), and an understanding of their role is presented as a necessary first step in the quest for self-knowledge.16 Such an investigation will incorporate the sensory faculties that make possible our apprehension of the world around us, as well as an understanding of the bodily instruments that facilitate this process; it will also encompass the inner senses that serve to fashion a unified experience out of the disjointed jumble of sense-impressions, and the motive faculties that enable us to orientate ourselves appropriately in the world thus disclosed. We may term this the epistemological reading of the Delphic maxim and of the microcosm-macrocosm motif that often accompanies it.17 Within the framework of a naturalised epistemology, an account of our perceptual apparatus can provide us with a working map of how the world at large is laid out: Ghaza-lī’s habit of talking about the outer and inner senses and the intellect as revealing distinct “worlds” (‘a-lam) is especially telling in this regard.18 It is in accordance with this understanding of the Delphic tradition, which ultimately can be traced all the way back to Plato’s Phaedrus (229e–230a), that Avicenna can declare that knowledge of one’s own nafs is a prerequisite to acquiring all the sciences.19 Ghaza-lī’s Jerusalem Ascent makes the same
15 See his Kita-b al-nafs wa al-ru-h. wa sharh. qawa--huma-, ed. M. Saqhīr H.asan al-Ma‘su-mī, Islamabad: Islamic Research Institute 1968, 27. 16 See Ih.ya-’, bk. 21, chs. 2–3; Kīmiya--yi, 1:18–19; Ma‘a-rij, 80–81; Mīza-n, 202–203 ff. 17 On the different interpretations of the Delphic maxim see Alexander Altmann, Studies in Religious Philosophy and Mysticism, London: Kegan Paul, 1969, 1–40. 18 See, e.g., the famous chapter on prophecy in Ghaza-lī’s autobiography, the Al-Munqidh min ald. ala-l (“Deliverance from Error”), eds. K. Ayya-d and J. Saliba, Beirut: Librairie Orientale 1969. 19 See Maqa-la fī al-nafs ‘ala- sunna l-ikhtis.a-r (“Treatise on the Soul”), ed. Samuel Landauer under the title, “Die Psychologie des Ibn Sīna-”, Zeitschrift der deutschen morgenländischen Gesellschaft 29(1875), 335–418, pp. 340 and 374. Remi Brague, who locates an equivalent movement in alKindī’s Book of Definitions, regards this as a shift in the way the Delphic maxim was understood: from the ancient practically oriented reading we have moved on to a theoretical concern (see his article “Cosmological Mysticism: The Imitation of the Heavenly Bodies in Ibn T.ufayl’s H.ayy Ibn Yaqz. a-n”, Graduate Faculty Philosophical Journal 19.2–20.1(1997), 91–102). If this is right, then Ghaza-lī’s project in fact represents a return to the more ancient ideal, as we shall see.
210
T. Kukkonen
point, as does the Criterion of Action and indeed the Revivification. Knowledge of the psychic faculties not only provides us with a start on the path to God, but also to all the sciences.20
Motive Parts The line of enquiry I have pursued thus far must appear disappointing to anyone who has come to the Islamic tradition in hopes of finding materials related to the themes of introspection and self-awareness, to say nothing of more socially constructed aspects of selfhood. Does self-knowledge for the Muslims thinkers consist merely in a catalogue of the various perceptual and cognitive powers? Assuredly not: for on the terms laid out in the Revivification we have thus far been at most dealing with the hosts of the heart, not at all with its true nature. Ghaza-lī makes clear that all of this barely qualifies as a start in our quest of self-examination: it is not enough to know what is ours, we must also know what in us actually is us. Ghaza-lī’s most explicit attempt at defining nafs, situated right at the onset of the book On the Marvels of the Heart, sheds fresh light on the matter. According to Ghaza-lī nafs has several meanings, two of which are of consequence: One of them indicates the irascible and appetitive human powers together […] it is according to this usage that the Sufis mean by nafs the root of all of a human being’s reprehensible qualities. Accordingly, they say that it is necessary to wage war (muja-hada) against one’s nafs and break it. This is what is referred to by the blessed saying, “Your worst enemy is your own nafs – what can be found between your two flanks.” […] The second meaning [indicates] that subtle thing which we have mentioned, the true human, i.e. one’s nafs and essence. (Ih. ya-’, 3:4.31–5.1)
The passage brings back into focus the normative aspect of Ghaza-lī’s search for self-knowledge. On the one side we have the lower self or soul, which immediately gets labelled as something working contrary to the divine purpose; on the other, we have the true reality (h. aqīqa) of the human being, which is an essence defined in terms of soul (cf. similarly Ma‘a-rij, 10). One aspect forms an object of identification, while the other is fit primarily for reprimand; the one is to be promoted, while the other must on every occasion be kept in check. This time, however, we have more to go on. The division drawn here is ultimately of Neoplatonic provenance: appetite and anger together constitute the brute soul, the nafs ba-himiyya, a term that derives from the Arabic adaptation of Plotinus (the so-called Theology of Aristotle).21 It is See Ma‘a-rij, 2–5; Mīza-n, 221; Ih.ya-’, bk. 21, 3:17.18 (“the one who knows his soul/self knows his Lord; [conversely], when a human is ignorant of this [i.e. the heart] one is ignorant of oneself, and one who is ignorant of one’s self is ignorant of one’s Lord; and one who is ignorant of one’s heart is all the more ignorant of other things”). 21 See Peter Adamson, The Arabic Plotinus, London: Duckworth 2002, 61–62; the appellation has Plotinian (I.1 [53] 10.6–7) as well as Platonic (Republic IX, 588c) roots. 20
The Self as Enemy, the Self as Divine
211
contrasted with the specifically human soul (al-nafs al-insa-niyya), which is the seat of reason. The latter is what we should identify with, as it is all that sets us apart from brute animals. Just enough Peripatetic materials exist to justify attributing such a dichotomy to Aristotle: but the moralising tone adopted both by the Theology and by Ghaza-lī is almost entirely Platonic.22 According to Ghaza-lī, the human being lies situated midway between the bestial and the angelic: we have, so to speak, a leg in both camps, living at once in both the sensible and the intelligible worlds.23 Insofar as we focus on our animal functions, we form part of the animal kingdom, while insofar as we partake of the angelic life we may be said to be or to become “angels in human form” (Mīza-n, 210). This schizophrenic condition is painted in the starkest of terms in Ghaza-lī’s writings; it motivates the better part of his reflections on human psychology. What might constitute an angelic mode of existence? According to Ghaza-lī, this is a life of contemplation. What separates us from mere bestial impulses and allows for a share in the angelic nature is knowledge of the immutable realities (h. aqa-’iq) of things.24 The heart’s specific task is to teach the human being about eternal truths, most prominently about the reality of God25: ultimate happiness lies in what is specifically human, namely, knowledge and intellection (Mīza-n, 305–310). This line of thinking allows Ghaza-lī to argue again for the incorporeality of the soul in a manner reminiscent of Avicenna. Authentic knowledge of God relies on the affirmation of a special power of apprehension in the human heart. But because God is not a body nor associated with anything bodily, we cannot come to an authentic knowledge concerning divine reality through any of the earthly powers of apprehension, not even on the view that all perception hangs on a certain degree of abstraction (tajrīd: see Ma‘a-rij, 48–49). Instead, acquaintance (ma‘rifa) with the divine has to occur through a faculty whose object lies altogether beyond the material; and, due to Ghaza-lī’s adherence to the Empedoclean principle of “like knowing like”, this means that the apprehending subject must be immaterial too. It is due to the heart having something of the lordly (rubba-nī) in it that it loves lordliness (rubu-biyya) by its very nature (Ih. ya-’, 3:249.18–19). The reality of the self is of the genus of the angelic substance (Kīmiya--yi, 1:15), and this allows for calling the wise among us angelic and lordly (Ih. ya-’, 3:9.19–21).
22
According to the self-professedly crude precepts of the ethical psychology sketched in the Nicomachean Ethics (I 13, 1102a26–32) a basic division can be drawn between rationally and irrationally motivated actions, with the latter covering actions resulting both from the appetitive and from the irascible impulses. Elsewhere, it is recommended that we identify with reason (EN IX 4, 1166a16–17; 1168b34–1169a3) and with the intellect (X 7, 1178a2–3). 23 See, e.g., Mīza-n, 209; Maqs.ad, 44–46. 24 Ih.ya-’, 3:9.12–18; also Ma‘a-rij, 40–45; for the Muslim philosophers, cp, e.g., al-‘Âmirī (d. 992 CE), Kita-b al-amad ‘ala- al-abad, 92.3–4; for the equation of the angels with the separate intelligences, Avicenna, al-Shifa-’: al-Ila-hiyya-t, bk. 10, ch. 1. 25 See, e.g., Ih.ya-’, bk. 21, 3:8.11–15; bk. 22, 3:57.28–31; for God as the ultimate reality (al-H.aqq), Maqs.ad, 137–138.
212
T. Kukkonen
Thus intellectual knowledge is what reveals God and His attributes; it is what allows for approximation of God and perfects faith (Mīza-n, 331). The lower parts of the soul, meanwhile, have a place in the natural order of things in preserving our earthly existence. Ghaza-lī explains in his book On Disciplining the Soul: Desire has been created for a purpose, and is an indispensable part of human nature. Should the desire for food cease, man would die; should the desire for sexual intercourse cease, man would die out; and should man feel no anger, he would not be able to defend himself from those things which threaten his life. While the basis (as.l) of desire remains, the love of property must necessarily remain also, which encourages one to guard it. What is required is not the total extirpation of these things, but rather the restoration of their balance and moderation, which is the mean between excess and defect. (Ih. ya-’, bk. 22, 3:52.23–26)26
So, metriopatheia rather than apatheia. The troubles begin only when these bodily desires get out of hand, as according to Ghaza-lī they inevitably will: for in so doing, they pervert the natural order of things. Letting one’s passions rule oneself is to submit to becoming a slave – a state of inauthentic existence, because in this case one’s accidents assume the place of one’s essence.27 The natural order of things is that the intellect (‘aql) rule, while appetite (shahwa) and passion (ghad. ab) follow. Ghaza-lī likens the situation to a kingdom where both the tax-collector and the police officer (sharīf) are needed, but must submit to the authority of the king and his vizier, i.e. reason.28 So what is it that drives a wedge between reason and the passions? Because the nutrition of the heart is wisdom, understanding, and the love of God, this is what its nature dictates that it seek out, and any deviation from this inclination can only bespeak a terrible affliction (bk. 22, 3:54.31–55.2 and 57.24–28). The details are obscure, perhaps deliberately so, but Ghaza-lī seems to want to postulate a genuinely demonic power at the opposite extreme from the rule of reason, a profoundly malicious agency that at every turn actively seeks to thwart and contravene the divine order of things. This satanic power manifests itself in those sinister whisperings (wasa-wis) of which the Qur’a-n speaks (Q. 114:4–6). The agencies vying for influence within a single human being thus number four, all in all. The divine and satanic powers stand opposed at either end, with the appetites and passions alternately falling under the sway of one force or the other.29
26 Translation by Tim Winter, Al-Ghaza-lī: Disciplining the Soul, Refining the Character, and Curing the Sicknesses of the Heart & Breaking the Two Desires, Cambridge: Islamic Texts Society 1995, 27–28, modified. 27 Mīza-n, 240 (correcting the mistyped pagination); cf. Maqs.ad, 74.7–10; Ih.ya-’, 1:58.3–5 raises the same point in reference to a lust for instrumental goods such as riches. 28 Ih.ya-’, 3:7.11–25; also Kīmiya--yi, 1:19–20; Ma‘a-rij, 80–81; Mīza-n, 235–238. 29 See Ih.ya-’, 3:10.17–11.8, and cf. 3:249.16–18; a particularly colourful later description of the warring factions in the soul as the angelic and satanic is found in Mulla- S.adra-, Elixir of the Gnostics, pt. 3, ch. 8.
The Self as Enemy, the Self as Divine
213
And where are we in all of this? According to Ghaza-lī, it is in our activities that our true affiliation stands revealed. To cite Claud Field’s translation of the Chemistry of Happiness one more time, The occupation of animals is eating, sleeping and fighting; therefore, if thou art an animal, busy thyself in these things. Devils are busy in stirring up mischief, and in guile and deceit; if thou belongest to them, do their work. Angels contemplate the beauty of God, and are entirely free from animal qualities; if thou art of angelic nature, then strive towards thine origin, that thou mayest know and contemplate the Most High, and be delivered from the thralldom of lust and anger. (Kīmiya--yi, 1:15)
It is clear where our loyalties are supposed to lie; we are asked to identify with our contemplative self and to resist the temptation to identify with our animal urges (to say nothing of those devilish murmurings that tell us that our self-realisation lies in setting ourselves against the divine plan). This, and only this, can lead to any meaningful and lasting bliss.
Indentification and Annihilation Attractive as the picture painted by Ghaza-lī may be, a few persistent problems remain. Firstly, there are puzzles about the intellect that are not easily resolved: learned people have disagreed as to what the definition and real nature of this intellectual principle is (Ih. ya-’, 1:83.9), for instance quarrelling over whether it is an accident or a self-standing substance (Ih. ya-’, 1:82.3–4). One would think that without some basic level of agreement on such a fundamental point, the whole theoretical edifice would threaten to topple under its own weight. Nonetheless, in the Revivification Ghaza-lī refuses to take an explicit stand on the issue, all the while insisting that the answer can only be decided on the basis of divine disclosure and therefore does not have a place in a work on practical religion. Different proposals have been advanced in order to explain Ghaza-lī’s reticence to take a stand on this point, ranging from sincere indecision to wilful obscurity and blatant misdirection: the emerging scholarly consensus is that Ghaza-lī’s philosophical psychology is more Avicennian and dualist than he is willing to let on.30 The most serious accusation that this raises has to do with the question of likening the divine with some aspect of creation (shirk, tashbīh), a mortal sin according to Islamic law and yet something that Ghaza-lī’s theory of cognisance of the divine seems to require. Does Ghaza-lī’s theory of a higher, intellectual self that is able to associate with the lordly require a robust theory of divinisation? Ghaza-lī certainly does not shy away from using the term (see Ih. ya-’, 3:9.19–21 and Maqs.ad, 65).
30
For a careful setting of the problem that comes down on the “esotericist” side of the debate see Gianotti 2001; for a differing viewpoint, e.g., Griffel 2004.
214
T. Kukkonen
Secondly, and on a related note, if the ultimate goal of aligning oneself with the divine purpose should ever be reached, then this would seem to entail the annihilation of any individual perspective whatsoever. The problem has received a great deal of commentary from scholars working on late ancient Platonism: if true enjoyment of the intelligible world hinges upon the identity of the knower and the known – if the mirror of the soul is polished to the point of becoming entirely transparent – then who in the realisation of such a beatific vision is the “I” that does the enjoying? In the medieval period the problem famously crops up in the “Averroist” debates in Paris in the 1270s. Admittedly, not all premodern philosophers would have greeted either one of these conclusions as altogether undesirable. Take the following concise statement by the Baghdadi Christian philosopher Yah.ya- Ibn ‘Adī (d. 974 CE) in his Reformation of Morals: Men are a single tribe, related to one another; humanity (al-insa-niyya) unites them. The adornment of the divine power is in all of them and in each of them, and it is the rational soul. By means of this soul, man becomes man. It is the nobler of the two parts of man, which are the soul and the body. So man in his true being is the rational soul, and it is a single substance in all men (jawhar wa-h. id fī jamī‘al-na-s). All men in their true being are a single thing (kullu-hum bi-l-h. aqīqa shay’ wa-h. id), but they are many in persons. Since their souls are one (ka-nat nufusu-hum wa-h. ida), and love is only in the soul, all of them must show affection for one another and love one another.31
Clearly in this example the notion of a hive-mind of sorts (perhaps on the lines of Averroës’ later unity of the intellect) is held out as a promise, not as a threat. When we recognise that reason unites us, then a healthy disdain for personal predilections will develop and a recognition that the interests of others are ours as well will take its place.32 The appeal to a common humanity on the basis of a shared capacity of reasoning somewhat resembles that of the Stoics. The Sufi tradition would similarly view the loss of the self as a desirable outcome in our worldly struggles against our own bad impulses, albeit from very different motives (mainly, the desire to emphasise God’s absolute power). In this tradition the overtones of ‘surrender’ (to the will of God, that is) that are discernible in the very term isla-m were taken to indicate that faith entails giving up every notion of autonomous selfhood. Many contemporary translators and students of Islamic mysticism accordingly end up translating nafs as ego, and the annihilation of the self as an eradication of the same; the choice is understandable, but obscures from sight the true radicalism of the Sufi programme. Typical is the formulation of
31 Tahdhīb al-akhla-q, in The Reformation of Morals. A Parallel Arabic-English Edition, ed. S.K. Kussaim, trans. S.H. Griffith, Provo, UT: Brigham Young University Press 2002, 106.2–9. Ethicists in the modern era have, of course, argued for the value of an impersonal standpoint for long: for one recent example see Derek Parfit, Reasons and Persons, Oxford: Oxford University Press 1984. 32 See the comments on the importance of overcoming one’s irascible impulses, etc. in the continuation to the Reformation.
The Self as Enemy, the Self as Divine
215
the 14th-century Indian mystic Shara-f al-Dīn Manīrī in the 37th of his Hundred Letters (Maktubat-i Sadi). According to Manīrī, the only thing standing between us and God is our nafs, so that when the seeker “engages in austerities and struggle with self and turns away from following his selfish inclinations, he emerges from the veil of his ego. Then there occurs revelation upon revelation, vision upon vision[…]”33 In Ghaza-lī’s analysis, too, dominion consists in exercising dominance over one’s enemies: one’s foremost enemy is one’s self, which is between one’s two flanks (Maqs.ad, 86.11–12). Becoming angelic means divesting one’s human attributes (al-s.ifa-t al-bashariyya) altogether.34 One must be freed of oneself (yuslam min nafsi-hi: Maqs.ad, 74.10–11) if one is to achieve authentic existence as part of the intelligible domain. Indicative of the way that the two traditions – the intellectualist and the Sufi – meld in the later Islamic tradition is Mulla- S.adra-’s (d. 1640 CE) appropriation of the thought of the Andalusian mystic Ibn al-‘Arabī (d. 1240 CE). Mulla- S.adra- in his Elixir of the Gnostics urges his reader to embark on a journey of self-discovery, and for the greater part prescribes an Aristotelian/Platonist programme in plotting its course; yet when it comes to describing the final stage on the wayfarer’s journey, Mulla- S.adra- paradoxically reveals that this has to do with “removing from your road to Him the harm of your existence.”35 The sting of this revelation is somewhat lessened once one acknowledges that one’s individual existence was illusory in the first place: all that the rational soul and human self is, is a mirror for reflecting aspects of the divine self-disclosure, not an independent reality of its own (a play on Aristotle’s contention that the material intellect is potentially all things, while actually none of them).36 The point is essentially the same that Roderick Chisholm makes in a more secular setting, while commenting on the way the notion of the subject is present in the Aristotelian tradition. Our expectations of the results of introspection notwithstanding, the discovery that the perceiving subject is in essence transparent is a positive result, and genuinely informative.37 This appears to be what Ghaza-lī is getting at when he discusses (in very allusive fashion, mind) the pregnant Islamic saying according to which God created Adam, i.e. the first archetypal human, “according to His form”.38
33
Maneri, The Hundred Letters, trans. P. Jackson, New York: Paulist Press 1980, 142–143. Mīza-n, 210; this reproduces a theme in Miskawayh, on whom see Roxanne Marcotte, “The Role of Imagination (mutakhayyilah) in Ibn Miskawayh’s Theory of Prophecies”, American Catholic Philosophical Quarterly 73(1999), 37–72, see p. 54. 35 Elixir of the Gnostics, pt. 2, ch. 10 (§64 Chittick). 36 For details see, e.g., Ibn al-‘Arabī, “Wisdom of the Heart”, in A. ‘Afifī (ed.), The Bezels of Wisdom: Fus.u-s. al-h.ikam, Beirut: Da-r al-kita-b al-’arabī 2002, 119–126. 37 See Roderick M. Chisholm, “On the Observability of the Self”, Philosophy and Phenomenological Research 30(1969), 7–21. 38 See, e.g., Mishka-t, 21.12–22.4; the passage is a particular favourite of Ibn al-‘Arabī’s, with a large part of the Fus.u-s. dedicated to its exegesis. 34
216
T. Kukkonen
Struggling with the Self We appear to have arrived at an impassé in our quest for self-knowledge. The divine self is elusive, inscrutable, and may well turn out to be entirely impersonal; the lower self, meanwhile, is indisputably mine, but at the same time it is mean and base and hardly worthy of attention. Where to go from here? Ghaza-lī’s tactic, I will argue, is to make a virtue out of necessity, and this in quite the literal sense. The shift is instituted through a newfound focus on the practical intellect, a part of Aristotelian psychology to which Avicenna among Ghaza-lī’s predecessors had paid relatively little attention, though the moralist literature coming out of Baghdad had shone some light on it.39 A suitable starting point is Ghaza-lī’s belief, central to his entire epistemological enterprise, that any disclosure of the divine reality can in the final analysis occur only through an extension of divine grace from the top down; it is not something that the human thinker can achieve merely by way of inference and willpower.40 This effectively puts the attainment of ultimate bliss beyond human control and human achievement – a welcome result for Ghaza-lī, who strenuously defends the Islamic tenet that God “guides whom He wills and leads astray whom He wills.” The details of this departure from Avicennian epistemology are beyond the scope of this study, but its significance cannot be overlooked. For all of this, there is still something we can do to help our case. The soul’s ultimate happiness lies in the contemplation of the realities of things divine and unification with them, true; however, such a state cannot be attained except through subjugating the appetitive and the irascible powers, and this again requires spiritual warfare and good works (Mīza-n, 221). While such preparation in itself is not enough, and while any preparatory work done in anticipation of the divine selfdisclosure does not on its own guarantee its arrival, one may still ready oneself for God’s gracious descent through a conscious effort to curb the animal impulses and to eradicate the demonic propensity for destruction and perversity (Mīza-n, 400). Spiritual warfare – Ghaza-lī’s celebrated larger jiha-d – is said to be the sole key to unlocking the hidden sciences, even if in the final analysis this occurs through God’s grace, not through man’s own efforts (Ih. ya-’, bk. 1, 1:43.20). Knowledge of God, after all, will not enter an impure heart (Ih. ya-’, 1:51.20). Accordingly, the principal aim of Ghaza-lī’s Criterion of Action, to take but one example, is to outline the practical knowledge necessary for spiritual warfare and the struggle against the passions.41 In fact, Ghaza-lī positions the main part of his
39
On Avicenna’s scant remarks on ethics see Majid Fakhry, Ethical Theories in Islam, 2nd expanded ed., Leiden: E. J. Brill 1994, 107–110. 40 This feature is noted already by Farid Jabre, La notion de la ma’rifa chez Ghazali, Beirut: Lettres Orientales 1958. 41 See Mīza-n, 231; such a struggle necessarily precedes any correct understanding of the spiritual realities, Mīza-n, 399.
The Self as Enemy, the Self as Divine
217
later authorial output squarely in the “spiritual medicine” genre of which Abu- Bakr al-Ra-zī’s (d. 925) al-T.ibb al-ru-h. a-nī is a prime early example.42 In many ways, Ghaza-lī’s exposition seems to be modelled directly on Ah.mad Ibn Muh.ammad Miskawayh’s (d. 1030) earlier work On the Refinement of Character; another major source is Abu- T.a-lib al-Makkī’s Nourishment of the Hearts.43 What both these works have in common is a dynamic approach to the construction of the self. Despite occasional warnings against the unhealthy urges of the self/soul, this principle remains fundamentally malleable and subject to the control of the rational soul. It is here that the Aristotelian notion of the practical intellect gains in prominence. According to Ghaza-lī the term ‘intellect’ is used equivocally of the faculties of knowledge (‘a-lim) and action (‘a-mil), both of which attach to this power equally (Mīza-n, 203). The rational soul is thus rather like the god Janus, inasmuch as “it has as it were two faces: one is [turned] towards the body […] the other towards the exalted principle.”44 And, strangely enough, whereas all that the theoretical intellect can do is patiently wait for the chance to witness the divine, the practical intellect is constantly busy in managing worldly affairs. This in fact is something Ghaza-lī will have learned from Avicenna: with respect to the intelligible universe, the theoretical faculty is essentially passive, whereas the practical intellect at least gets to be the active partner in relation to the body and its various powers.45 There is something paradoxical about this, seeing as the theoretical intellect with its connection to the intelligible world is supposed to enjoy a “separate” (mufa-riq) subsistence,46 while the practical intellect must needs remain inextricably intertwined with the Aristotelian definition of the soul as the form of the living body, with all of its worldly attachments.47 Another way of contrasting the theoretical and the practical intellect in Ghaza-lī is to point out that there does not seem to be anything particularly personal about an investigation that has at its crosshairs a definition of “essential humanity”. To the contrary, a single scholar should in principle be able to prescribe a common ideal to an audience encompassing all humankind. In a marked contrast, the obligation
See, e.g., Ih.ya-’, 1:9.32–39; the Chemistry of Happiness bears the same stamp. 43 The two works in the original are called Fī tadhhīb al-akhla-q and Qu-t al-qulu-b; see here Timothy Winter’s informative comparison (1995, lvi–lvii). 44 Ma‘a-rij, 41.14–16; see Mīza-n, 205; cf. Ibn Sīna-, Avicenna’s De Anima (Arabic Text). Being the Psychological Part of Kita-b al-Shifa-’, ed. F. Rahman, London: Oxford University Press 1959, bk. 1, ch. 5; similarly, e.g., Mulla- S.adra-, The Elixir of the Gnostics, pt. 3, ch. 10 (§56 Chittick). 45 See al-Shifa-’: al-Nafs, bk. 5, ch. 1 (202–209 in Rahman); the same interplay between active and passive is described in Ghaza-lī’s Taha-fut al-fala-sifa (in Incoherence of the Philosophers 2nd ed., trans. Michael E. Marmura, Provo, UT: Brigham Young University Press 2000, 181.7–15). 46 Ghaza-lī in the Revivification makes the important observation that the heart qua intellect differs from the knowledge that the human being possesses: as the locus (mah.all) of knowledge and the receptacle for divine revelation, the heart is perceiver rather than perception itself, see Ih.ya-’, bk. 21, 3:5.13–18. 47 On soul as substance and soul as form see, e.g., Avicenna, al-Shifa-’: al-Nafs, bk. 1, ch. 1. 42
218
T. Kukkonen
to engage in spiritual warfare emerges as an intensely personal one. One’s faults and failures are one’s own, and even if outside counsel is welcome in this struggle, what can safely be said is that no two diagnoses and prescriptions will ever be the same.48 The search for the human essence is common and scientifically guided, while the questioning of one’s lower instincts and motivations takes on a much more individual character. Here, the particular overtakes the universal.49 Take for instance the way in which Ghaza-lī treats the divine attribute of awareness in his explication of the Beautiful Names of God. Ghaza-lī first defines awareness as a subset of knowledge – the type of knowledge that attaches to the inner dimension of things, to be precise, as opposed to their external features. God is thereby said to be supremely aware, since He possesses such knowledge to the utmost degree. After such an introduction, it would seem natural for Ghaza-lī to play up once more his love of knowing “the realities of things” whilst explaining what the human being’s share in such an attribute can be. If God’s awareness is said to consist in his knowledge concerning the world He has created, then is not our awareness, too, tied to a knowledge of that same world? No! The servant’s share in this [attribute] lies in being aware of what occurs in his world. A servant’s world in turn [consists of] his heart, his body, and those hidden things that characterise his heart: treachery, deception, preoccupation with the present life, evil intent [wrapped up in] good appearances, and the pretence of sincerity where it is lacking. These cannot be known except through extensive experience. [The experienced person will be] aware and mindful of his own self and know its [ways of] deceiving and deluding and holding out illusions: he will hold guard against himself and toil in opposition to it, assuming watch over it. Among the servants [of God], such a man deserves to be called ‘aware’. (Maqs.ad, 112, emphasis added.)
One’s dominion over one’s world, by which is meant primarily one’s body, resembles the Creator’s dominion over the universe at large (Ma‘a-rij, 148). Recall once more also the simile involving the vizier, the tax-man, and the police officer: the boldly drawn analogy between God’s world and the world of humankind reinforces the impression that it is precisely our inner life – our self-governance – that has been allotted as our primary dominion.
48 Thus Ghaza-lī in his book On Disciplining the Soul: “Were a physician to treat all of his patients with a single medicine he would kill most of them; and so it is with the Sufi master, who, were he to charge all his aspirants with one kind of exercise, would destroy them and kill their hearts. Rather, attention should be paid to the illness of each aspirant, his circumstances, his age, his constitution, and the capacity of his body to perform such exercises, which should be prescribed on this basis” (Ih.ya-’, bk. 22, 3:56.23–26); the translation is Tim Winter’s (1995). 49 Cp. Aristotle, Met. I 1, 981a13–20, where it is laid down as a rule that the job of the physician is not to heal “man, but […] Callias or Socrates or some other called by some such individual name.”
The Self as Enemy, the Self as Divine
219
Mirrors and Masks This brings us to a few observations regarding the relation between the solitary self and the societal one. The Revivification was written primarily with the individual in mind who had come to recognise a void in his or her own spiritual life, much like Ghaza-lī himself had done some ten years previous.50 Would he then greet the eager wayfarer with the exhortation: “Healer, heal thyself?” Evidently not: in fact, Ghaza-lī’s recipes for finding out what state one’s soul is in invariably involve other people, the testimony of friends and countrymen. All too often we are blind to our own faults.51 The preferable thing to do, Ghaza-lī says as a good Sufi should, is to find a reliable master or Shaykh: however, these are few and far between. (Ih. ya-’, 3:59.1–3) The second option is to find a good friend, one who will not hesitate to point out one’s weaknesses; the potential pitfall is that friends are often soft-hearted and willing to think only the best of us (3:59.4–22). It is because of this that one’s enemies are a particularly valuable sort of friend in this sort of situation. People who routinely assume the worst of us more often hit the mark than not; therefore “keeping one’s friends close, but enemies closer” is a dictum that could find use in areas besides politics (3:59.23–26). Finally, Ghaza-lī reminds us that it is good for us to mingle among people in general and to learn from each other’s foibles, weaknesses, and subtle personality quirks. As the Prophet had reminded his people, “The believers are mirrors for one another” (3:59.27–30). The mirroring relation is an unmistakably Platonic theme, although here again the source is likely to be pseudo-Aristotelian – this time, the Treatise of the Apple.52 Why, though, should outside evidence be given preponderance in determining our spiritual state? We may here glance back at where we first began. It is in our actions that our character stands revealed, Ghaza-lī has told us: there simply is no better witness to our state – certainly not any internal monologue, which according to Ghaza-lī is always prone to delusional and defensive editorialising anyway. In fact, Ghaza-lī holds that God and one’s self are alike in that both are best observed indirectly:53 just as God can be seen everywhere and nowhere, 50
On these developments see the remarks made in the Munqidh under the “Sufism” chapter. Ih.ya-’, bk. 22, 3:58.33–36, following the Gospel of Matthew 7:3. 52 The Treatise of the Apple purports to describe a deathbed dialogue between the philosopher and Simmias. Wisdom is achieved through the spirit, which is to say the immaterial, so a first requisite of wisdom is knowledge of self or soul (nafs). This is attained first through fortification, i.e. improvement of character, which again is achieved through letting others describe one’s self for oneself. The same way that the ill must consult physicians and the blind must have their complexion described to them by their companions, we need to have our condition outlined for us by our fellow human beings – preferably, other seekers after wisdom. For an English translation of the Arabic De pomo see Chittick 2001, 106. 53 There is ample ancient precedent to the notion that the soul, like God, cannot be perceived directly: both manifest themselves in works and in the orderliness of the kosmoi that form their respective domains of influence. For rich documentation spanning the philosophical literature, Philo, and the Greek Fathers see Philip Lyndon Reynolds, “The Essence, Power and Presence of God: Fragments of the History of an Idea, from Neopythagoreanism to Peter Abelard”, in Haijo J. Westra (ed.), From Athens to Chartres, Leiden: E. J. Brill 1992, 351–380, pp. 351–362. 51
220
T. Kukkonen
The intention (ma‘na-) by which a human being is a human being is both evident and obscure. It is evident when inferred from his orderly and wise actions; it is obscure when sought by what the senses perceive. […] This quiddity is obscured from the senses, but evident to the intellect by way of inferring it from what [a human being] achieves and how he acts. (Maqs.ad, 149–150)
The plain reading here, of course, is that the true self (the untainted rational soul) cannot be perceived by the outer senses but only by something structurally similar, viz. another intellect. And this is quite right; this is indeed what we are meant to infer. But I would suggest that more is going on in the text. Take note of Ghaza-lī’s chosen examples: it is by observing a person’s actions over a sustained period of time that we affirm her or his humanity. If someone consistently acts in an orderly and wise manner, we may conclude that we are dealing with a human being – clearly here a normative notion: conversely, bestial or destructive behaviour also reveals to us something about the perpetrator’s true character, something that in the resurrection shall become evident to all.54 And again, an outside observer is more likely to be able to cut through the thicket of thin justifications and excuses to get at the real motivating factors than is the perpetrator him- or herself. We are what we do: among other things, this conviction grounds Ghaza-lī’s further faith that in the afterlife, we shall each receive the fate best befitting to us. Ghaza-lī thus explains those visions that people sometimes have of their acquaintances being magically transformed, e.g., into dogs and pigs. What such visions disclose are intentions or meanings (ma‘a-nī) that already exist in their subjects. It is merely that in death, the material conditions that cloak such intentions fall off, revealing for all to see what such people already in fact are.55 The same interpretive framework explains al-Ghaza-lī’s insistence, both in the Chemistry and elsewhere, on judging the tree by its fruit.56 When it comes to explaining how good acts and mortification of the soul create and enforce positive character traits – for instance, it is in constant acts of giving that the attribute of munificence is acquired (sifa alsakha-’: see Ih. ya-’, 3:210.25–27) – and how, conversely, wicked actions conspire to produce wicked people, Ghaza-lī freely avails himself of the Aristotelian doctrine of habituation (see Ih. ya-’, 3:55 ff.). The treatise On Disciplining the Soul, long considered a centrepiece in the Revivification, in fact reads like a monotheist’s guide to the Nicomachean Ethics in many places. Special attention is paid to the doctrine of the mean and its attendant complications; another favourite theme is how character is nothing but a condition or state (hay’a) of the soul-self that is particularly wellestablished (3:49.19).
54 With a few exceptions, al-Ghaza-lī advocates suspending judgement in this world concerning another person’s piety or impiety, see Sherman Jackson, The Boundaries of Theological Tolerance in Islam: Abu- Ha-mid al-Ghaza-lī’s Faysal al-Tafriqa Bayna al-Islam wa al-Zandaqa, Oxford: Oxford University Press 2002. 55 See Ih.ya-’, 1:52.8–12; Kīmiya--yi, 1:23; Fad. a-’ih., 201; the apprehension of the intentional aspects of reality is an Avicennian trope, on which see Deborah L. Black, “Estimation (wahm) in Avicenna: The Logical and Psychological Dimensions”, Dialogue 32(1993), 219–258. 56 Ghaza-lī puts it that “the fruits signify the one that grants them”, Maqs.ad, 57.15–16.
The Self as Enemy, the Self as Divine
221
One aspect that emerges from Ghaza-lī’s treatment of moral persons is how tentative and fragile is the harmony that any of us can achieve within ourselves. Because our different impulses really do constitute distinct autonomous forces within us, brought under one banner only in the event that one faculty manages to bring the others under its sway, uniting our theoretical beliefs with our inner states and furthermore with our actions is a constant challenge, never to be taken lightly let alone for granted (Maqs.ad, 155.9–14; cf. Ih. ya-’, bk. 1, 1:74.26–27). Self-identity and unity for Ghaza-lī are the ideal, not a presumed starting-point: if it is true that “Aristotle views psychic unity as the result of a slow process of integration which is broadly co-extensive with the acquisition of moral virtue,” and if likewise for Plato personal unity is not a given but an honorific title, then he continues in a venerable line of discussion.57
Conclusion Much has been made over the years of al-Ghaza-lī’s emphasis on ethics, both as regards the practical orientation of his own chief work (the Revivification) and his insistence on the unity of theory and practice.58 Yet the theoretical underpinnings to this tendency have generally not been well understood. It is not enough simply to state that for al-Ghaza-lī, correct belief always comes accompanied by right action: it is also important to note why this is. I propose to read these remarks in light of Ghaza-lī’s ruminations on the self and the soul. The true self for Ghaza-lī can only be the human quiddity, and a naturalist analysis is enough to establish that it is (1) our capacity for reason and (2) our power of will that set us apart within the animal kingdom. The respective proofs for the existence of these two realities can be found in (1) the presence in us of an access point to a world beyond the material, and (2) a more detached viewpoint existing in relation to our worldly dealings than is afforded by the impulses of the animal passions. At the same time, Ghaza-lī takes quite seriously the Sufi tradition and its penchant for insisting that one’s worst enemy is one’s self, “what lies between one’s two flanks.” What is us in this second sense is what ours, what in this world is attributable to us: and these are more often than not the reproachable character traits and actions.59
57 See Suzanne Stern-Gillet, Aristotle’s Philosophy of Friendship, Albany, NY: State University of New York Press 1995, 26; Mary Margaret McCabe, Plato’s Individuals, Princeton, NJ: Princeton University Press 1994, 300. 58 On theoretical and practical knowledge as the two pillars of faith see Ih.ya-’, 1:26.27–28.23. For Ghaza-lī, it is a mark of the truly learned individual that one’s deeds do not contradict one’s professed beliefs (Ih.ya-, 1:64.1 ff.). 59 There are important questions here to be raised about the way in which actions are ascribed to us (iktisa-b), given that according to Ghaza-lī God is the sole true Agent in everything: but these questions will have to await a later date.
222
T. Kukkonen
The real novelty in Ghaza-lī’s account, at least as far as the philosophical mainstream is concerned, lies in the emphasis he places on patient, life-long observation and refinement of one’s character.60 In his book on Vigilance and self-examination, for example (mura-qaba wa muh. a-saba: Ih. ya-’, bk. 38), Ghaza-lī lays great emphasis on the need for self-rebuke (tawbīkh al-nafs wa mu‘a-tabati-ha-). Quoting from the Qur’a-n, Ghaza-lī warns his readers that not a single soul will escape admonishment: because the straight path down the middle is thinner than a hair’s breadth and sharper than a sword’s edge, everyone will stumble at some point.61 No-one is exempt from the character-building exercise of repentance (tawba: e.g., Ih. ya-’, bk. 31, 4:9.19–20). Furthermore, according to Ghaza-lī, the wayfarer’s soul is ever a work in progress, with work always remaining to be done, hence the need for constant and never-ending vigilance.62 Such a life of constant vigilance may seem to some less than rewarding, as Ian William Miller writes of this spectacle of self-criticism: I am never turned away at the ticket booth for a sold-out performance; I am condemned, unless alcohol or true fun intervenes to put self-consciousness to sleep, to play the smalltown newspaper critic to my own performances in a high school play, while bitterly wishing I could make a living as an author rather than as a critic.63
Certainly many people with a propensity for self-doubt will sympathise. But there is something revealing about Miller’s remark as well, considering that Ghaza-lī took real joy (or “true fun”) to mean precisely such an intoxicated state as would result from another power forcefully overtaking the soul.64 The Sufis’ “taste for the divine” is the one thing that can offer respite from the yoke of one’s own unhappy consciousness. Where Miller has gone wrong, I imagine Ghaza-lī would say, is in thinking that release consists in becoming an author rather than a critic; rather, our true contentment lies in recognising how God is the sole author of all our actions.65
60 What Jonathan Jacobs says of Maimonides applies equally to Ghaza-lī: “Virtue and even repentance do not free us of the potentiality to sin, and the obligation to examine oneself is never once and for all discharged. Likewise, sin does not exhaust our capacity for repentance and virtue, because the potential for ethically significant change is at no time exhausted. […] This idiom of virtue and perfection is very similar to Aristotle’s, but it is used to elaborate a quite different philosophical anthropology, one which reconfigures the ethical perfections and their possibility” (Jacobs, “Aristotle and Maimonides: The Ethics of Perfection and the Perfection of Ethics”, American Catholic Philosophical Quarterly 76/1(2002), 145–163, p. 163). 61 See Ih.ya-’, bk. 22, 3:58.20–26, citing Q. 19:71. 62 See Ih.ya-’, bk. 21, 3:40–42; Mīza-n, 400–401. The one exception to this may be the prophets and the saints (see Mīza-n, 241): but this hardly registers as an objection, as Ghaza-lī is always quick to remind his reader of the exclusivity of this class. 63 Miller, Faking It, Cambridge: Cambridge University Press 2003, 129. 64 For some remarks on Ghaza-lī’s take on divine intoxication see Kukkonen, “Ibn T.ufayl and the Wisdom of the East: On Apprehending the Divine”, in Stephen R.L. Clark and Panayiota Vassilopoulou (eds.), Late Antique Epistemology, London: Macmillan, 2008 (forthcoming). 65 The first half of Ghaza-lī’s Book of Unity and Trust (bk. 35 of the Ih.ya-’) is devoted to the subject.
The Self as Enemy, the Self as Divine
223
Where this is not possible, we must resign ourselves to the fact that we are neither pure angels nor simple beasts, but something forever suspended in between.66
Appendix In this essay I have tried to show how Ghaza-lī employs a two-pronged approach in interpreting the Delphic maxim, filtered through the traditions related to the Prophet. On the one side we have the mystery of the heart when viewed as a link to the intelligible universe; on the other, the self as a source of all that can go wrong in one’s relation to the universe (indeed, all that can go wrong in the universe as a whole: animals, lacking will, never consciously contravene the divine plan, after all). But Ghaza-lī is often equally as interesting for the paths he chooses not to take. Most crucially, I have not found in Ghaza-lī’s major works any extensive use of the argument from self-awareness so characteristic of Suhrawardī and the later Illuminationist tradition. The treatise called the Jerusalem Ascent does put forward an early version of the argument, essentially similar to the one that is found in Avicenna; and though the treatise’s authorship has traditionally been disputed, I have in this essay treated it as authentic and therefore would be remiss not to mention the argument as well. Very briefly, this begins from the self-evident nature of our role and existence as cognising subjects: it is something whose truth cannot escape any rational observer.67 “Even while asleep you will not be ignorant of your being or your reality”,68 al-Ghaza-lī contends: the reason is that even dreams, which assuredly are not in one’s conscious control, will nonetheless be identified as one’s own dreams after the fact. What this points to is a notion of the self as something akin to a logical subject, a principle to which all these various properties, beliefs, and functions are attributed but of which little else is otherwise known. The unity of the soul that undergoes all these different states is something that simply has to be assumed, even if it eternally escapes the spectator’s gaze: otherwise it would be impossible to make sense of such sentence constructions as for instance “I saw this bread and became hungry, then when I saw it being snatched away I got angry instead.” The brand of self-awareness that this entails is immediate, since it does not depend on sense-perception; in the case of intellection, “its quiddity is equal to its cognition,
66 Compare this with Pascal, according to whom “man is neither angel nor beast; and the misfortune is that he who would act the angel acts the beast” (Pensées, §358). 67 Cp. Kīmiya--yi, 1:16: the existence of the reality of the heart is so self-evident (z. a-hir) that not a single human being is given over to doubts about her or his own existence. 68 ’anniyyatu-ka wa-h.aqīqatu-ka: Ma‘a-rij, 18 (correcting an obvious typographical error). Anniyya could denote existence as well as being or essence (See Marie-Therese d’Alverny, “AnniyyaAnnitas”, in Mélanges offerts à E. Gilson, Paris: J. Vrin 1959, 59–91); here I have chosen the middle term as the most neutral interpretation.
224
T. Kukkonen
and its cognition to its quiddity” (Ma‘a-rij, 19). Clearly these comments find inspiration in Avicenna’s so-called “flying man” argument, especially its more developed versions;69 equally as clearly, the purpose of al-Ghaza-lī’s ruminations is to provide the main fruits of Avicenna’s labours without going to the metaphysical intricacies of the philosopher’s mature work. Much further study would be needed in order to put this minimal sketch in context, either in relation to its sources in Avicenna or its possible afterlife among the Illuminationist philosophers.70 University of Jyväskylä, Finland
For Avicenna’s flying man see, e.g., al-Shifa-’: al-Nafs, ed. Rahman 15–16; al-Isha-ra-t wa altanbīha-t, ed. S. Dunya-, Cairo, 4 vols., 1960–1968, 2:343–344; for comments and analysis, most recently by Jari Kaukua, Avicenna on Subjectivity, Jyväskylä: University of Jyväskylä 2007. 70 The research for this article was completed during my tenure as the Canada Research Chair in the Aristotelian Tradition. The CRC Programme’s role in enabling this work is hereby gratefully acknowledged. 69
Locating the Self Within the Soul – Thirteenth-Century Discussions Mikko Yrjönsuuri
According to the traditional picture of the history of Western philosophy the High Middle Ages was intellectually Aristotelian, dominated by the Thomist approach. To some extent, this picture was formed already in the Early Modern Era, when many important thinkers distinguished their own philosophy from that of the scholastics. The university philosophy rejected by Descartes, for example, was indeed characteristically Aristotelian, and to a considerable extent even based on a thirteenth-century interpretation of Aristotle by Thomas Aquinas. It may be true that the scholastic philosophy, superseded in the seventeenth century by new approaches, was a direct extension of certain Classical trends. However, as a description of what really happened in thirteenth-century Western philosophy, simply categorising it as “unoriginal Aristotelianism” is clearly inappropriate. In fact, many of the crucial philosophical innovations typically associated with early modern thinkers were already established in the thirteenth century or at the latest in the beginning of the fourteenth century. We should not, thus, locate the borderline between Classical and Modern thinking at the Renaissance, as is often done. But it seems equally inappropriate to locate it at the fall of the Roman Empire, like Jonathan Swift did in his tale Battle of the Books. On the contrary, medieval philosophers were in deep debt to classical civilization. The dark centuries in the latter half of the first millennium did imply a significant break in Western European philosophical thinking, but it is nevertheless clear that the medieval schools were established on the basis of the literary heritage of Classical tradition, in the Latin community almost as directly as on the Arabic side. It was only little by little that the discussions in medieval universities were able to formulate from this material new kinds of philosophical thinking that could be called distinctively modern. The historical picture looks remarkably similar to this also in the case of the philosophy of the self. Medieval philosophers brewed modern thought from classical materials. The first full century of university life, the thirteenth century, initially saw a radical expansion in the knowledge and command of the ancient literary material, and then a radical re-evaluation of the deeper philosophical issues involved. The century began with an approach that can broadly be regarded as Platonist. The philosophy of self, in particular, was at first largely based on an Augustinian brand of the neoplatonic–stoic thinking, which was dominant in the P. Remes, J. Sihvola (eds.) Ancient Philosophy of the Self, © Springer Science+Business Media B.V. 2008
225
226
M. Yrjönsuuri
last centuries of the Roman Empire. After the introduction of the Arabic philosophical materials, however, more Aristotelian ways of thinking took hold. However, towards the end of the century, a genuinely new philosophy of self emerged among Franciscan intellectuals. This originally rather radical ideology contained the crucial philosophical innovations that made possible the development that can nowadays be called modernity. My aim here is to look at these two steps in the development of thirteenthcentury thought. After a sketch of the older Platonist picture, I base my discussion on two main examples: Thomas Aquinas as a relatively faithful Aristotelian thinker, and Peter John Olivi as a radical innovator from the Franciscan camp. The discussion will, I hope, shed some light on the origins of the modern individualist conception of the human being, namely the conception that human personhood consists of being a rational subject capable of making free decisions. A better grasp of how this still influential conception was originally devised will enable us to pinpoint and perhaps even review some of the philosophical misconceptions involved in it. There is, however, also a historical lesson to be learned. A careful look at the thirteenthcentury discussion shows how history of philosophy proceeds. New innovative ideas are not summoned from scratch, but rather stumbled upon as apparently minor changes in understanding the old ways of thinking. Hopefully my discussion can in some manner contribute to the process of revising the problematic elements of our contemporary ways of thinking about the self.
Soul–Body Dualism Early medieval Christian philosophy contained a deep tension within its philosophy of the self. On the one hand, the metaphysical account of human existence was characteristically Platonist and involved a belief in an immortal soul that could exist as an individual thing in separation of the body. On the other hand, the Early Fathers of the Church had already taken a strong position against the Platonists in favour of believing in the resurrection of the body, thereby connecting personal identity of a human being with the body as well as the soul. The metaphysical side of the story has been studied, e.g., by Carlos Bazán.1 If we simplify the picture a bit, we may say that the crucial distinction was between understanding the human soul as a particular thing (hoc aliquid) and as a form (forma corporis). The former became the catchword for the view that the soul is a separable immortal entity capable of existing on its own. The latter, for its part, became the catchword for emphasising the strong connection that the soul has to the body and the understanding of the soul as the principle of life.
1 B. Carlos Baz´ an, “The Human Soul: Form and Substance. Thomas Aquinas’ Critique of Eclectic Aristotelianism”, Archives d’histoire doctrinale et littéraire du moyen âge 64(1997), 95–126.
Locating the Self Within the Soul - Thirteenth-Century Discussions
227
As Bazán shows, early medieval thinkers up to Thomas Aquinas tended to emphasise the former identification of the soul at the expense of the latter. The soul was understood as an individual capable of existing alone. When extended into a full account of the metaphysics of human existence, the result was a dualism of two distinguishable entities, the soul and the body. In the Platonist framework, the latter would be weaker in its metaphysical status, or not even considered a thing at all. Thus, if we look at early medieval systematic presentations of this picture, such as in William of Auvergne’s On the soul (De anima), we find discussions of the role of the body in human existence. In what sense can the body be called the instrument of the soul? Why exactly can the soul not be called a full human being without the body? These discussions reflect rather clearly an underlying approach in which the human self is, from the metaphysical point of view, to be identified with the soul. The soul is a particular individual thing and, from the metaphysical point of view, it is the thing that one would consider as oneself. From the moral point of view the situation looks different. As was recognized already in the Platonic Alcibiades major, to have the status of a moral agent, a person must have a body. We act in the world only as embodied beings. In this sense, the body is crucial for moral identification. Nevertheless, the Platonic picture allows one even in this context to look at the body as an instrument of the soul. The role of the body is that of a necessary instrument, like a horse is necessary for the rider. Without a horse, there is no actual rider. In this sense, the Platonists could allow that real human existence does require a body – although the survival of the soul is not threatened by the loss of the body any more than the loss of the horse kills the rider. Already the Early Fathers of the Church had tried to distance the Christian outlook from the Platonist view. The most important discussions concerned resurrection. While the Platonists believed in the natural immortality of the soul and the corruptibility of the body, the Church officially adopted the position that Christian belief involves a promise of the resurrection of the body. Eternal reward and punishment concern human beings as bodily beings, not as mere souls. If we look at somewhat later Western Fathers like Augustine, the position has become clear. The body is a crucial part of one’s personal identity. The promise of the resurrection concerns Christians as individuals who have their personal characteristics. This means that they have their personal bodily characteristics, like the face, bodily constitution, disposition, sex, and – in the case of the martyrs – even the scars achieved in saintly actions. The soul as such is not personal enough, and indeed the main creeds do not mention the eternal life of the soul at all. The doctrine became such that in the resurrection God gives life back to the very bodies that have been buried, just as it happened in the case of Jesus. In this sense, the self is assimilated to the body rather than to the soul.2
2
For further discussion and references, see e.g. Caroline W. Bynum, The Resurrection of the Body in Western Christianity, 200–1336, New York: Columbia University Press, 1995.
228
M. Yrjönsuuri
The idea that personal identity depends on the body rather than on the soul is still clearly present in early medieval texts. The dialogue On the soul (De anima) by Aelred of Rievaux (1109–1167 CE) provides a suitable example. He emphasises that in human morality, one must recognize one’s character as a bodily being and not only as a separable soul. The point is not that Aelred would think that human beings are mere bodies. On the contrary, he provides arguments to show that there is an incorporeal part in the soul. One such argument deserves mention here. In the dialogue, the teacher asks the student to turn to himself and endeavour to locate the different functions of the soul at different parts of the body. The crux of the argument is that while one can feel the location of many of the functions, in the case of intellectual operations this is impossible. It is clear that one sees with the eyes. But if one understands what justice is, this understanding does not appear to take place anywhere in the body. We have no spatially located awareness of the understanding at all. Thus, Aelred argues, understanding is not dependent on the body.3 But a person is not merely understanding. According to Aelred, the body is an important part of a person as an individual. He attacks certain heretics who reject the resurrection of the body. They believe that one should, in the face of God, care only for one’s soul. As Aelred points out, among these heretics this belief has resulted in promiscuous sexual life. Marriage is not honoured. When the heretics think of themselves as mere souls, they have no reason to take care of themselves as bodily beings. But when they neglect their bodies, they neglect something crucial for eternal life, claims Aelred. It is important to notice that Aelred’s argument clearly differs from the traditional arguments. Traditionally, the Platonists had argued that the main motivation for being moral is to take care of one’s soul. By injustice one would harm one’s soul. Aelred seems to think that it is not at all obvious that promiscuous sexual life damages the immaterial soul, but it is for him relatively obvious that it damages one as a bodily being. While the intellect may not be harmed by improper sexual behaviour, one’s bodily life is. In this respect, therefore, the motivation for being moral is taking care of one’s body.4 Aelred’s soul–body dualism is neither of the Platonist nor of the Cartesian kind. While it is fairly self-evident that the chastity he calls for is not of an intellectual kind, it is almost as evident that he is not thinking of an inappropriate sex life as ruining the body in the Cartesian sense. For example, Aelred does not have in mind sexually transmitted diseases, but rather the role of sexuality in human flourishing as intended by God. As he sees it, the heretics at issue neglect God’s wishes on how
3
Aelred of Rievaulx, Dialogue on the Soul, trans. C. Hugh Talbot, Kalamazoo, MI: Cistercian, 1981, 48. 4 Aelred of Rievaulx 1981, 66.
Locating the Self Within the Soul - Thirteenth-Century Discussions
229
to live our bodily lives. This results, according to Aelred, in damage to them as bodily beings. The body that Aelred is speaking of is not just a corporeal system following the rules of mechanics. Rather, he means the ensouled body, the living body. Having renounced their identities as things of that kind, the heretics fail to see that living a flourishing human life implies requirements upon the kind of sexual life one leads. When they think that in God’s eyes they are just incorporeal, separable souls, they lose the possibility of living a fully satisfactory human life. People who think that the body is just an instrument for achieving experiences may end up in spontaneous short-term pleasures, which ultimately damage their persons. Aelred’s dualism is, thus, a dualism dividing an incorporeal intellectual soul that is not dependent on bodily life from an ensouled living body. From the Cartesian viewpoint it may look like a dualism within the soul rather than a soul–body dualism. As a dualism of this kind, it is quite understandably apt to look like a dualism within the self, a division of the self into the bodily self and the intellectual self. Aelred does not really address the issue of a division within the self. This was, however, to become a topic of discussion in the next century. Given the dualism, one may seem entitled to ask questions concerning the unity of the self: Since there are (at least) two parts in a human being, but only one person or self, how does the single self relate to the parts? Am I the soul or the body? And if I am both, how is it that the self appears to be a unity? How does human multiplicity fall together into one self, and where is the centre of this self, if there is one? Aelred’s picture shows no direct familiarity with Aristotle. The rise in Aristotle’s importance in the thirteenth century nevertheless made the problem even more clearly visible. Aristotelian psychology divides psychological faculties into two or three parts. There is the vegetative soul responsible for functions like nutrition and growth. Then there is the sensitive soul, which is responsible for the sensomotoric systems. Third, there is the intellectual soul, which is responsible for the higher rational functions like understanding. Since the vegetative and sensitive are typically lumped together in medieval readings of Aristotle, a dualism results. What is more, this dualism is very much like the dualism we see in Aelred’s older picture. But now the choice is between the sensitive functions and the intellectual functions. Which ones are more proper to me as an individual self? Putting it crudely and following Aelred’s footsteps: am I, as a self, primarily sexual or intellectual? How is it that I am able to be both? An Arabic author Avicenna (980–1037 CE) who became known among Latin speaking philosophers in the twelfth century had provided a scheme of argumentation that suited the purpose of discussing questions of this kind. In his so-called De anima, he takes up the general problem of unity in the framework of a faculty psychology. Assuming that there are, for example in a lamb, distinct faculties for the cognitive recognition of the dangerousness of a wolf and for the fearful emotional response, why is it the same thing that both sees the danger and fears the wolf? Avicenna even transfers the question to the human sphere and uses an Arabic word translatable as the pronoun “I”. As he puts it:
230
M. Yrjönsuuri
Again, we say “when I perceived such and such a thing, I became angry” and it is a true statement, too. So it is one and the same thing which perceives and becomes angry.5
As Avicenna notes, this is “due to its being in possession of a faculty by which it is capable of combining both these things.” The explanation is that the thing referred to by the pronoun “I” has a faculty of apprehending the acts of both of the faculties as its own acts. This faculty of self-consciousness then yields the apprehension of a unified self. Avicenna’s problem of unity is of a rather general kind. Amidst all the faculties distinguished in the Aristotelian faculty psychology, how is it that there is only one self? However, thirteenth-century Latin authors began to use the same argument scheme to ask the rather specific question of unity between the faculties of the three Aristotelian levels: vegetative, sensitive and intellectual. The example became: I eat, I see, and I understand. How can I apprehend that the same self is doing these three things? It seems that the Avicennian strategy of answering the question was generally accepted and most philosophers thought that the unity of the self is somehow given in self-consciousness. Disagreements loomed, however, about the nature of the faculty playing the role of self-consciousness. Let us turn to looking at these disagreements through two examples, Thomas Aquinas and Peter John Olivi.
Aquinas: I Have a Sensory Feel of Everything I Do as My Own Deed Let us start with Thomas Aquinas and, more particularly, his theory of conscience. Until quite recently, the medieval use of the Latin term conscientia was thought to have been exclusively moral in its connotation and context. Now it is becoming recognized that the term also had a use that must be described as cognitive. Aquinas’ discussion of conscience can be found in the De veritate, question 17. He begins with a twofold division of the uses of the term conscientia. The second use is more or less the common moral understanding of the word, while the first use derives from the simple fact that we cannot lay moral judgement on anything that we are not aware of. That is, moral conscience presupposes that one is cognitively conscious of the action in question. This use of the word-form ‘conscience’ seems to have disappeared from contemporary English, and I suggest that we translate conscientia in this use as ‘awareness’, or habere conscientiam as ‘being aware of.’ In such a translation, Aquinas’ description of the first use of the word conscientia runs as follows: 5
Avicenna, Avicenna’s Psychology (an English translation of Kit¯ab al-naj¯at, Book II Chapter IV with historico–philosophical notes and textual improvements on the Cairo Edition), trans. F. Rahman, London: Oxford University Press, 1952, 65–66. For discussion on the issue see e.g. Simo Knuuttila, Emotions in Ancient and Medieval Philosophy, Oxford: Oxford University Press, 2004, 224–225; Jari Kaukua, Avicienna on Subjectivity. A Philosophical Study, Jyväskylä: Jyväskylä University Printing House, 2007, 84–85.
Locating the Self Within the Soul - Thirteenth-Century Discussions
231
[…] According to the first way of using the word, we are said to be aware of some action in so far as we know whether the action is being performed or not.6
As if this explanation had not been clear enough, Aquinas picks out from the Latin Bible three phrases where the word conscientia occurs: Gen. 43:22, Eccl. 7:23 and Rom. 9:1. The first of these three examples shows that conscientia can refer to being aware of something external – just like “being aware of” still can. In the other two, we see that it may also refer to awareness either of one’s own action or of one’s own state of mind.7 Aquinas’ point seems rather simple. In order to lay moral judgement on our own action (this is what we call conscience in accordance with Aquinas’ second use of conscientia) we need to have a grasp of what we are doing (this in turn is the first use of the word in his distinction). Now, we are evidently dealing with an important dimension of any comprehensive self-image. We can, therefore, put the question that we are after in this essay thus: where among our cognitive capacities, in accordance with Aquinas, should we locate this awareness of our own actions? In the De veritate passage at issue, he says that in this regard the term is used to refer to “particular acts of sensory knowledge”. As examples, he mentions the memories of earlier deeds and the sensation of present action.8 Thus, Aquinas locates selfconsciousness among the functions of the sensory faculties. As he understood it, one’s own image of oneself as an agent acting in the temporal world is formed in the common sense, whose function is to draw together the information gathered by the senses,
6 On a slightly broader scope the text is as follows: “et secundum quidem primum modum applicationis dicimur habere conscientiam alicuius actus, in quantum scimus illum actum esse factum vel non factum; sicut est in communi loquendi usu, quando dicitur, hoc non est factum de conscientia mea, idest nescio vel nescivi an hoc factum sit vel fuerit. et in secumdum hunc modum loquendi intelligitur quod habetur; Gen. xliii, 22: non est in conscientiis nostris quis pecuniam posuerit in saccis nostris; et Eccle. vii 23: Scit conscientia tua quia tu crebro maledixisti aliis; et secundum hoc dicitur conscientia testificari aliquid; Rom. ix, 1: testimonium mihi perhibente conscientia mea […]” de veritate qu 17 ar 1 resp. The translation is mine. 7 In the King James Version they are as follows: “[…] we cannot tell who put our money in our sacks.” (Gen. 43:22); “[…] for oftentimes also thine own heart knoweth that thou thyself hast cursed others” (Eccl. 7:23); and “I say the truth in Christ, I lie not, my conscience also bearing me witness in the Holy Ghost, that I have great heaviness and continual sorrow in my heart” (Rom. 9:1). It may be of interest to note that Olivi refers to the same passage from Gen. 43, but says that this is an uncommon use of the word: “sumatur […] aliquando, licet rarius, pro actu reminiscentiae vel scientiae facti proprii vel alieni, ut, cum Genesis capitulo 43 dicitur […]”, Peter John Olivi, Quaestiones in secundum librum Sententiarum, Bibliotheca Franciscana Scholastica Medii Aevi 4–6, vol. 3, ed. B. Jansen, Quaracci: Collegium S. Bonaventurae, 1926, 175. 8 “[…] in prima applicatione qua applicatur scientia ad actum ut sciatur an factum sit, est applicatio ad actum particularem notitiae sensitivae, ut memoriae, per quam eius quod factum est, recordamur; vel sensus, per quem hunc particularem actum quem nunc agimus, percipimus.” de veritate qu 17 ar 1 resp.
232
M. Yrjönsuuri
and common sense belongs to the sensitive part of the soul.9 The self-conscious self is, according to Aquinas, the bodily self, not the separable intellect. We have to reject an apparent counter-argument immediately. It may seem here that Aquinas was speaking only of bodily actions and not of the intellectual self. However, we may first note that this would imply that his conception of moral conscience concerns only bodily activity. This may not have been the case, although it is, on the other hand, possible that it never occurred to Aquinas that one could sin by way of the intellect. We need not resolve the issue here, since we have evidence elsewhere that Aquinas assigned a crucial role to common sense also in the awareness of our own cognitive and even intellectual functions. Common sense plays an important part in Aquinas’ theory of how we can attend to our own intellectual activity and make it an object of thought. In the central formulations of this theory, he refers to the need of phantasms, which he locates in the common sense, and thus the sensitive part of the soul. The theory is very complicated, and consequently I cannot here go very deep into this issue.10 Let me now just pick out a clear and straightforward remark concerning remembering our thoughts. In his commentary on Aristotle’s On memory and remembering, Aquinas classifies memory in the traditional way as a faculty of the sensory soul. This concerns also remembering intellectual things, as the text specifically mentions. The example is that of remembering having understood.11 Aquinas’ explanation goes approximately as follows. The intellect deals with universals. Now, there are only three ways in which a universal can be said to be present in the mind: 1. As a mere possibility when we have not yet actually understood the issue 2. As a habitual disposition once we have command of the concepts but are not actually thinking about the issue 3. With full presence when we actually understand something
9 Aquinas in clear that only the sensitive powers can grasp human existence in a temporal way. “[…] Cognoscere autem praeteritum ut est praeteritum, est illius cuius est cognoscere nunc ut nunc: hoc autem est sensus. Sicut enim intellectus non cognoscit singulare ut est hoc, sed secundum aliquam communem rationem, ut inquantum est homo vel albus vel etiam particularis, non autem inquantum est hic homo, vel particulare hoc: ita etiam intellectus cognoscit praesens et praeteritum non inquantum est hoc nunc et hoc praeteritum” de veritate qu 10 ar 2 resp. 10 For a book-length discussion see François-Xavier Putallaz, Le sens de la réflexion chez Thomas D’Aquin, Études de philosophie Médiévale 66, Paris: Vrin, 1991. 11 Sentencia De sensu, tr. 2 l. 2 n. 10: “[…] Memoria autem non solum est sensibilium, utputa cum aliquis memoratur se sensisse, sed etiam intelligibilium, ut cum aliquis memoratur se intellexisse. Non autem est sine phantasmate. Sensibilia enim postquam praetereunt, a sensu non percipiuntur, nisi sicut in phantasmate: intelligere etiam non est sine phantasmate, ut supra habitum est. Unde concludit quod memoria sit intellectivae partis animae, sed per accidens; per se autem primi sensitivi, scilicet sensus communis. Dictum est enim supra, quod intelligens proponit in phantasmate quantum determinatum, licet intellectus secundum se consideret rem absentem; ad memoriam autem pertinet apprehensio temporis secundum determinationem quamdam, secundum scilicet distantiam in praesenti nunc. Unde per se memoria pertinet ad apparitionem phantasmatum, per accidens autem ad iudicium intellectus”.
Locating the Self Within the Soul - Thirteenth-Century Discussions
233
Aquinas makes it clear that this classification is exclusive. There is no other way in which a universal concept could be in the understanding.12 It follows that the intellect would not be capable of the kind of repetition included in remembering something as a past event. A habitual disposition of the type (2) allows the intellect to repeat an earlier act of understanding. This, however, produces the full presence of the universal (i.e., presence of type (3) ), while memory essentially produces merely simulated perceptions. When I, for instance, remember seeing a dog, the event is different from actually seeing a dog. Similarly, when I remember understanding a mathematical theorem, the recollection of the actual event need not be of an intellectual kind. Just think of the student in a math exam, who well remembers having once understood what the issue was but cannot grasp it again, thus failing the exam. Remembering how the theorem goes would of course be intellectual. That is, presence of type (2) in the mind would enable presence of type (3) too, and consequently the student would pass the exam. But the student who remembers having understood may not be able to repeat the event. For Aquinas, a simulation of the original phantasms connected with the intellectual operations seems the natural way to explain how we remember having understood. From this perspective, our personal self-image of what kinds of things we have understood seems to be built in the common sense. The intellectual repetition of the acts of understanding is not of such a personal kind.13 There is, however, an apparent counterargument to this interpretation of Aquinas’ view. I already pointed out that Aquinas had a theory of how the intellect cognises itself and its own acts. Did he not imply by such a theory that the intellect is keeping track of its own activity and having second-order acts directed (perhaps through phantasms) at first-order acts of the intellect? We now should make a three-fold distinction. It is (1) one thing to understand what it is to think and what the essence of a thought is, but it is (2) another thing to actually understand that it is you yourself who is presently understanding, and (3) another thing to have a self-image picturing what kinds of thoughts you have been
12 Sentencia De sensu, tr. 2 l. 2 n. 6: “[…] intellectus possibilis habeat species intelligibiles etiam cum actu non intelligit, non est sicut in potentiis sensitivis, in quibus propter compositionem organi corporalis aliud est recipere impressionem, quod facit sentire in actu, et aliud retinere, quando etiam res actu non sentiuntur, ut obiicit Avicenna; sed contingit propter diversum gradum essendi formarum intelligibilium, vel secundum potentiam puram sicut invenire vel addiscere, vel secundum actum purum sicut quando actu intelligit, vel medio modo inter potentiam et actum, quod est esse in habitu. Non ergo propter hoc solum indiget intellectus possibilis humanus phantasmate ut acquirat intelligibiles species, sed etiam ut eas quodam modo in phantasmatibus inspiciat.” 13 Cf. also Summa Theologiae, I, q. 79, a. 6, resp.: “Sic igitur, si memoria accipiatur solum pro vi conservativa specierum, oportet dicere memoriam esse in intellectiva parte. Si vero de ratione memoriae sit quod eius obiectum sit praeteritum, ut praeteritum; memoria in parte intellectiva non erit, sed sensitiva tantum, quae est apprehensiva particularium. Praeteritum enim, ut praeteritum, cum significet esse sub determinato tempore, ad conditionem particularis pertinet.”
234
M. Yrjönsuuri
entertaining at different instants of time and at a variety of places. With this distinction in mind we can easily see that the intellect cognising itself is capable of the first and perhaps the second.14 The third requires sensory powers. The function of the intellect is to understand universal essences, not singular events – but as persons we are temporal individuals and our consciousness of ourselves necessarily concerns singular events occurring in time. The intellect does understand what it is to think, and what it is to be an intellect. However, Aquinas’ conception of the functions of the intellect excludes it having the third kind of second-order activity. Being a particular person with particular thoughts is something that must be grasped by the lower parts of the soul. In Summa theologiae (I, q. 76, a. 1, resp.), Aquinas looks closer at the way in which we have consciousness of understanding as our own activity. The aim of the passage is to show that the intellect has a strong connection to the human body. In fact, Aquinas reveals in the text that he identifies the person with the body rather than with the intellect, but does not want to admit that the intellect is anything outside the person. He opens the problem as follows: But if anyone says that the intellectual soul is not the form of the body, he must explain how it is that this action of understanding is the action of this particular man; for each one is conscious that it is he himself who understands.15
The crucial fact upon which Aquinas builds his discussion is at the end of the sentence. An act of understanding includes the experiential feel that it is one oneself who understands. That is, Aquinas thinks that understanding takes place selfconsciously (at least sometimes). Aquinas’ formulation of the problem can be found in the first part of the sentence. If the intellectual soul is not the form of the body, or in other words, if the intellectual soul does not have a strong unity with the body, an explanation is needed why an act of understanding should in the first place be attributed to the person himself. This way of putting the question seems to presuppose that the person is identified with the body rather than the intellectual soul. Otherwise the problem would be why the operations of the body are to be attributed to the person himself. Aquinas starts his solution of the problem by making a threefold distinction of the ways in which an action may be attributed to a person. Either the person acts (1) “by virtue of its whole self ”, (2) “by virtue of a part”, or (3) “through an accidental quality”. Aquinas excludes the last alternative (3) first. Understanding is an
14 Cf. Summa Theologiae, 1a, q. 87, a. 1, resp.: “Uno quidem modo, particulariter, secundum quod Socrates vel Plato percipit se habere animam intellectivam, ex hoc quod percipit se intelligere. Alio modo, in universali, secundum quod naturam humanam mentis ex actu intellectus consideramus.” 15 “[…] Si quis autem velit dicere animam intellectivam non esse corporis formam, oportet quod inveniat modum quo ista actio quae est intelligere, sit huius hominis actio, experitur enim unusquisque seipsum esse qui intelligit.” ST Ia q. 76 a. 1 co.
Locating the Self Within the Soul - Thirteenth-Century Discussions
235
operation that belongs to the essence of being a human, understanding must be predicated of Socrates, for example, essentially rather than accidentally, as Aquinas points out. So it must be either that Socrates understands by virtue of his whole self, or by virtue of some part which, of course, must be by virtue of his intellectual soul. Like many other thirteenth-century thinkers, Aquinas ascribes to Plato the view that “man is an intellectual soul”. This view would allow one to pick alternative (1), making understanding an operation of the whole self, but Aquinas does not accept the identification of a human person with a mere intellectual soul. As he remarks: […] it is one and the same man who is conscious both that he understands and that he senses. But one cannot sense without a body, and therefore the body must be some part of man.16
At this point, only the second alternative (2) remains. Understanding is attributed to an individual man, because the intellectual soul is a part of him. From our viewpoint here, the central parts of the argument can be formulated as follows: 1. 2. 3. 4. 5. 6.
One sees with the bodily eyes One understands with the intellectual soul The intellectual soul cannot see Understanding is not a bodily operation We see and understand with self-consciousness of the same self in these acts Therefore, the self has at least the body and the intellectual soul as its distinct parts
For Aquinas, the self is a thing consisting of parts. As his discussion continues, he turns to a metaphysical description of the unity of these different parts, and specifically the nature of the soul-body union. From our viewpoint here, it is especially interesting to note that this unity is given in experience. By having self-consciousness of both bodily and intellectual activities one finds oneself as a unified person consisting of different parts. Furthermore, it seems that it is no accident that Aquinas appears to start from the assumption that at least the body can be identified with the person. For he locates self-consciousness among the functions of the sensitive soul, as we saw above. The ensouled, living body houses the part of the person enabling self-consciousness. The intellectual soul as such could not in Aquinas’ mind have self-consciousness of the particular different parts of the person.
16
“[…] propter hoc quod ipse idem homo est qui percipit se et intelligere et sentire, sentire autem non est sine corpore, unde oportet corpus aliquam esse hominis partem.” ST Ia q. 76 a. 1 co.
236
M. Yrjönsuuri
Peter John Olivi Olivi’s way of looking at the same argument scheme is different. He calls for full self-reflexive unity of the person, and finds that the understanding is the only faculty capable of guaranteeing such unity: Thus he says, “I who understand, see or eat”; and only the intellectual potency can say both of these things, because no other potency can apprehend both of these actions.17
Olivi follows Avicenna’s formulation relatively closely, although the example has changed from perceiving and becoming angry to understanding, seeing and eating. Avicenna’s “one and the same thing which perceives and becomes angry” must, in Olivi’s view, be the highest part of the soul, the understanding. In comparison to Avicenna and Aquinas, Olivi is original in requiring that the faculty that has self-consciousness of the activities at issue, must be able to apprehend those activities. Olivi seems to wonder how any sensory faculty could be selfconscious of an act of understanding, since there is no way in which the sensory faculty could even apprehend the incorporeal act of understanding itself. Intellectual acts cannot be sensed. The piece of reasoning appears to be formulated against a feature of Aquinas’ theory. As we saw, Aquinas thought that self-consciousness, or at least memory of one’s own intellectual activity, is located in the sensitive soul. It seems that Olivi wants to take this issue up in order to show that the faculty of understanding is the only human faculty that is capable of apprehending all the activities of all the faculties. No other faculty is capable of this. Sensations cannot be digested, nor intellections seen. But both seeing and eating can be understood. For this reason, only the faculty of understanding is able to apprehend self-consciously all of one’s own activities. From the viewpoint of modern conceptions of subjectivity, it is very noteworthy that unlike Aquinas, Olivi thinks that human self-consciousness is connected to the highest functions of the individual soul. At “the root of our own subsistence”, as he puts it, we are intellectual beings having an “intimate sense” of our own beings.18 Such formulations show a very different approach to problems of self-apprehension than Aquinas’ approach. The Classical background of Olivi’s way of looking at the structure of the human soul could perhaps be found in Stoic rather than in an Aristotelian
17
“[…] Unde dicit: ego qui intelligo video vel comedo; et utique non potest dicere nisi per potentiam intellectivam, quia nulla alia potentia potest apprehendere utrosque actus nisi ipsa.” Peter John Olivi, Quaestiones in secundum librum Sententiarum, Bibliotheca Franciscana Scholastica Medii Aevi 4–6, vol. 2, ed. B. Jansen, Quaracci: Collegium S. Bonaventurae, 1924, 122. 18 Olivi 1924, 122: “[…] experimentum intimum et certissimum quo intra nos sentimus sensitivam teneri et regi et dirigi a parte superiori tanquam aliquid in sua natura intime plantatum; in tantumque sentitur esse plantata in radice superioris partis nostrae quod radix nostrae subsistentiae, ipsa scilicet pars superior, sentit intime et decit actus sensitivae esse suos.” See also Olivi 1924, 121: “[…] experimentum naturae nostrae intimum quo sentimus nos in radice liberi arbitrii per se et altissime consistere et omnia alia in ipsa stabiliri altiori modo quam partes corporis in corde vel cerebro.”
Locating the Self Within the Soul - Thirteenth-Century Discussions
237
context.19 This comes as no surprise, since Olivi is well known for his fierce opposition to the Aristotelian tendencies of his age. It is clear that many of Olivi’s Stoic tendencies come through Augustine, but to show the more direct influences of e.g. Seneca, more research needs to be done. In any case, Olivi locates the cognitive function that is forming our individual self-image at the highest faculty of the soul. There are two basic, though related obstacles for such a view in the Aristotelian approach: the doctrine of conversion to phantasms, and the definition of the intellect as an exclusively universal faculty. Neither of these obstacles seemed relevant to Olivi, whose viewpoint was much closer to what Augustine had written in De trinitate. Olivi took it almost for granted that the intellect is capable of cognising singular objects. His conception of changeable singulars was an Aristotelian one, and he did think that all created singulars are material objects. However, he took the intellectual part of the human soul to consist of a non-extended kind of matter, which made it unproblematic to allow the intellect the capability to cognise singulars. For Olivi, the intellectual soul was a material thing, but not a corporeal thing, because it consisted of spiritual matter.20 Matthew of Aquasparta had argued a little before the appearance of Olivi’s work that it would be humiliating to require in the Aristotelian sense phantasms in the cognition of incorporeal reality, like God, angels, and the human intellect. Olivi followed this line of thought. He thought that phantasms are needed in human understanding, if only in order to reach sensory objects. In the Augustinian manner, he thought that the need of phantasms is based on the need that the object of thought has to be present to the mind. Since sensory objects themselves cannot be in the soul, they are represented by phantasms.21 In his systematic discussion of how the mind apprehends and knows itself, Olivi makes a distinction between two kinds of self-knowledge. On the one hand, our soul experiences in an indubitable way all its actions: living, thinking, willing, seeing, hearing and bodily movements. On the other hand, we achieve general knowledge of the essence of the soul through reasoning. Olivi compared our way of achieving the first kind of self-knowledge to sensory perception: “the mind by itself senses and touches itself in an immediate way.”22 According to him, we are intellectually aware of our own thoughts in a way comparable to the sense of feeling. Coming to a universal kind of knowledge of the essence of the soul requires complicated processes of reasoning – but that is not what forming a self-image is
19
Here I have in mind especially the Stoic defence of the unitary character of the soul, and the assimilation of ourselves to the “ruling part” (h¯egemonikon). See also Seneca, Ep. 121. 20 Olivi 1924. 21 Matthaei ab Aquasparta, Quaestiones disputatae de fidei et de cognitione. Biblioteca Franciscana Scholastica medii aevi 1, Quaracchi: Florence, 1957. 22 Olivi 1926, 146.
238
M. Yrjönsuuri
primarily about. For Olivi, human beings have a direct and immediate grasp on their own thoughts in a quasi-sensory way in the intellectual part of the soul.23 The crucial idea in this explanation of the formation of self-image is the conversion of the mind’s gaze (intellectualis aspectus) upon itself. Such self-reflexivity is one of Olivi’s arguments to prove the incorporeality of the mind. He argued that corporeal things, like the particular senses and the common sense, could only be directed to something other than themselves. Olivi added emphatically that he means more than otherness in the essence of the thing (secundum essentiam): spatial otherness (secundum positionem et situm). He summarises: “Thus no corporeal thing can reflect immediately on itself, but only something next to it”.24 In a later question, Olivi considers an objection to this view, proceeding in a strikingly Stoic manner. He points out that all animals can conceive of themselves by their common sense. The idea is that in order to move in the controlled way in which animals move, they need to have some ruling part in their souls. Furthermore, this ruling part has to have a grasp of the whole. Olivi’s example is that of a dog or a serpent, who chooses to lose some less important member in order to save the whole. This implied in Olivi’s eyes that these animals must have in their common sense a self-image by which they evaluate their parts and decide their actions. How could this be if only the intellectual part of the soul could form a self-image? Does this not show that also common sense is capable of self-reflexivity?25 23
“Sciendum quod anima scit se vel potest scire duplici modo. Primus est per modum sensus experimentalis et quasi tactualis. Et hoc modo indubitaliter sentit se esse et vivere et cogitare et velle et videre et audire et se movere corpus et sic de aliis actibus suis quorum scit et sentit se esse principium et subiectum. […] Hanc autem scientiam sui habet anima per immediatam conversionem sui intellectualis aspectus super se et super suos actus. Qui quidem, quamdiu est in pervigili usu liberi arbitrii, semper et continue stat super eam conversus. Quia tamen non est sufficienter clarus ad omnes essentiales rationes et proprietates animae contuendas et discernendas: ideo, licet mens per ipsum immediate sentiat se et palpet, non tamen sciat suam naturam per genera et differentias discernere ab omnibus aliis generibus et differentiis aliarum rerum. […] Secundum modus se sciendi est per ratiocinationem per quam investigat genera et differentias quae per primum modum non non novit […]” (Olivi 1926, 146). 24 “Potentia autem non potest apprehendere suum obiectum, nisi prius sit conversa ad ipsum, conversio autem potentiae non potest fieri sine applicatione et directione virtuali suae materiae ad illud ad quod est conversa. […] Unde aspectus corporum, quantumquam sint spirituales, semper sequuntur leges corporeales, sicut patet in aspectu visus et etaim sensus communis. […] Impossibile est autem quod materia corporealis possit immediate converti nisi ad aliquid quod est extra se non solum secundum essentiam, sed etiam secundum positionem et situm; unde pars corporis non potest converti immediate ad se ipsam, sed solum ad partem sibi propinquam.” (Olivi 1924, 112). 25 “Oportet enim dare aliquam potentiam appetitivam, imperantem motum animalium et imperando moventem nunc ad hoc nunc ad oppositum; quod facere non potest, nisi habeat secum aliquam potentiam sibi dictantem omnia quae imperat et imperandi modum. Ergo sicut illam appetitivam oportet dominari omnibus membris et sensibus quos ad suos actus applicat vel ab eis retrahit: sic oportet unam iudicativam sibi assistere quae de omnibus actibus eorum iudicet et eorum delectationes vel dolores advertat et alteram alteri praeferat vel praeferandam ostendat. Praeterea, quando canis vel serpens pro conservatione totius exponit aliquam partem, tunc prefert totum parti et caput alteri membro. Ergo oportet in eis esse aliquam communem potentiam quae in simul ambo extrema et mutuam eorum comparationem et unius ad alteram praeferentiam ostendat, quamvis non cum illa plenitude et altitude reflexivi iudicii cum qua fit hoc ab intellectu.” (Olivi 1924, 587–588; see also pages 615–616 and 620).
Locating the Self Within the Soul - Thirteenth-Century Discussions
239
The reference in Olivi’s discussion is to Augustine De libero arbitrio, book II, where chapters 3–12 are especially relevant. More than three centuries later, Antoine Arnauld referred this very text to Descartes as an anticipation of his cogito ergo sum.26 In the text Augustine discusses the certainty that we have of our own existence and tries to build a picture of how we know our own cognitive activity. Augustine’s picture does not make much distinction between human beings and other animals, since for him the common sense was capable of reflecting on the activities of the other faculties of the sensitive soul and also on itself. However, Augustine seems to admit that in the case of human beings it is the reason that is responsible for the self-reflective activity. He specifically rejects the idea that there is one shared reason responsible for human rationality, and purports to prove that we each have our own reason just like we have our own eyes, although it is true that there is only one wisdom. Altogether, Augustine’s discussion seems to draw from Stoic sources, where no difference between the levels of human soul was admitted. Augustine’s text is no easy reading and allows different interpretations. In the above-mentioned text, Olivi presents his own reading of Augustine and gives a clear and straightforward answer to the problem of animal self-consciousness, but unfortunately it seems that his solution is not very well in line with his other views. Despite having elsewhere denied self-reflexivity to all corporeal things, he here admits that non-rational animals are capable of knowing themselves, and that common sense is a self-reflexive faculty. He points out, though, that intellectual selfreflexivity is of a subtler and more refined kind. Furthermore, he also draws a connection between self-reflexivity and self-control, and locates the self-image of non-rational animals in the ruling part of their soul. Locating self-reflexivity in the highest part of human soul was an important issue for Olivi in his theory of the will and its freedom. For example, he imposed as a strong condition on freedom that the agent is self-controlled. Freedom is not merely lack of external determination, but positive self-determination. By this he meant that in order to be free, the human being has to reflect upon him- or herself and make the choice as one’s own choice. Random choices are not free, but only those which one has made oneself are free. Only self-reflexive minds can cognise their own choices and produce them in a self-controlled manner.27 There is an important sense in which Olivi thought us to be self-reflexively present to ourselves as individuals. In his defence of the freedom of the will he made it clear that the most crucial aspect of our existence as human beings is exactly in the capacity to reflect on ourselves both in the cognitive and the active
26
Antoine Arnauld quotes Augustine in his “Fourth Objections”, AT 7, 197–198. For discussion and references, see my “Free Will and Self-Control in Peter Olivi”, in Henrik Lagerlund and Mikko Yrjönsuuri (eds.), Emotions and Choice from Boethius to Descartes, Dordrecht: Kluwer Academic, 2002, 99–128. 27
240
M. Yrjönsuuri
sense. As human beings, our thoughts and choices have the specific character of being present to us as our own thoughts and choices. Here I want to emphasise two dimensions of this self-presence. First, for Olivi the specifically human selfpresence was clearly of an individual kind although he located it in the incorporeal rational part of the soul. Second, he emphasised the way in which we attend to our thoughts and volitions as distinctively our own and dependent on us. This is a very different thing than mere objective perception and thus we have to say that he strongly emphasised the subjective character of the self-image. My self-image is an image of me, and not simply of some particular person.
Conclusion According to Olivi, there is a single unified centre of the self. This centre appropriates every action of the person as its own action by having self-consciousness of those actions as its own. As we saw, such experiential unity of the self had become a generally accepted philosophical view already earlier, although it remained a topic of discussion well into the fourteenth century. Among our three examples, Aelred of Rievaux might not have accepted that we can directly experience the unity of the self. We saw him paying more attention to the fact that different actions are performed and experienced at different bodily locations. Also, he emphasised the role of the body as a carrier of moral responsibility. But already Aquinas accepted the peculiar phenomenon of self-consciousness by which we perceive certain activities as our own activities and by which we experience the unity of our whole selves. Unity is not, however, among the most important features of our selfhood in Aquinas’ theory. He does not locate self-consciousness in the highest part of our souls, nor would he assimilate one’s personal individuality with self-consciousness. But Olivi does take these two distinctively modern points of view. For Olivi, self-consciousness is an incorporeal phenomenon that is distinctively human. No other animal has self-consciousness in the way humans have. And furthermore, self-reflexive consicousness is that which makes us what we are. In Olivi’s theory, the self has more structure than it has in the genuinely modern theories like Descartes’ or Locke’s theory. At the core of the self, however, Olivi posits full simplicity in a distinctively modern way. By being self-conscious in all my activities, my mind collects them together as actions of one, essentially simple subject – the self. Furthermore, Olivi claims that it is an essential feature of human personhood that one enjoys full freedom in everything one does. In Olivi’s eyes, this freedom is lost only in mental disorders and other special states of that type. We humans are free intellectual subjects making our choices with self-consciousness of the choices as our own choices. This is a distinctively modern understanding of what it is to be a human person. According to Aquinas’ Aristotelian theory of choice, voluntary action results from characteristically universal reasoning as applied to the particularities of the
Locating the Self Within the Soul - Thirteenth-Century Discussions
241
situation where one is. In this model, one’s choices do not carry any special relation to one’s individual personality but rather spring from impersonal decision-making processes. As we saw, Olivi theorizes in a modern way also in this matter. According to him, we always make our choices freely as our own choices, and no universal reasoning can have binding force over the individual choice. We reveal who we are through such free choices. In comparison with the classical tradition, our three exemplary philosophers can be seen to draw from different types of sources. While Aelred as a twelfth century author drew from the Platonist sources, Aquinas was an Aristotelian. Olivi, for his part, can be said to have leaned on Stoic kinds of thinking in his discussion of human selfhood. On the other hand, all of these authors were original, and working towards what deserves to be called modern thought. But in the case of Olivi there seems no avoiding the label. He was unequivocally a modern thinker in his theory of the self. University of Jyväskylä, Finland
Bibliography
Ackrill, John L. (1972) “Aristotle’s Definitions of psuch¯e”, Proceedings of the Aristotelian Society 73, 119–133. Adamson, Peter (2002) The Arabic Plotinus, London: Duckworth. Adkins, Arthur (1960) Merit and Responsibility, Oxford: Clarendon. Aelred of Rievaulx (1981) Dialogue on the Soul, trans. C. Hugh Talbot, Kalamazoo, MI: Cistercian. Afifi al-Akiti, M. (2004) “The Three Properties of Prophethood in Certain Works of Avicenna and al-G. aza¯lı¯”, in John McGinnis and David C. Reisman (eds.), Interpreting Avicenna: Science and Philosophy in Medieval Islam, Leiden: E. J. Brill. Altmann, Alexander (1969) Studies in Religious Philosophy and Mysticism, London: Kegan Paul. Annas, Julia (1981) An Introduction to Plato’s Republic, Oxford: Oxford University Press. Annas, Julia (1995) “Self-Knowledge in Early Plato”, in Dominic O’Meara (ed.), Platonic Investigations, Washington, DC: Catholic University of America Press, 111–138. Annas, Julia, “Ancient Scepticism and Ancient Religion”, forthcoming, available online at http:// www.u.arizona.edu/∼jannas/. Anscombe, Elisabeth (1975) “The First Person”, in Samuel Guttenplan (ed.), Mind and Language, Wolfson College lectures, Oxford: Clarendon, 45–65. Aquinas, Thomas (1882–) S. Thomae Aquinatis Doctoris Angelici Opera Omnia (Leonine). Summa Theologiae (vols. 4–12); Quastiones disputatae de veritate (vol. 22); Sentencia libri de sensu et sensatio, II, de memoria et reminisgencia (vol. 45,2), Rome. Aubry, Gwenaëlle (2004) “Introduction. Structure et thèmes de la traité”, in Plotin. Traité 53, Paris: Les éditions du Cerf. Augustine (2006) The Confessions, trans. Philip Burton, London: Everyman. Avicenna, see Ibn Sı¯na¯, Abu¯ ‘Alı¯. Balme, David M. (1980) “Aristotle’s Biology Was Not Essentialist”, Archiv für Geschichte der Philosophie 62, 1–12. Baltzly, Dirk (2004) “The Virtues and ‘Becoming like God’: Alcinous to Proclus”, Oxford Studies in Ancient Philosophy 26, 297–321. Barnes, Jonathan (1997) “The Beliefs of a Pyrrhonist”, in Myles Burnyeat and Michael Frede (eds.), The Original Sceptics: A Controversy, Indianapolis, IN: Hackett, 58–91. Barney, Rachel (2004) “Callicles and Thrasymachus”, in Edward N. Zalta (ed.), The Stanford Encyclopedia of Philosophy, Internet Edition: http://plato.stanford.edu/archives/fall2004/ entries/callicles-thrasymachus. Baz´an, B. Carlos (1997) “The Human Soul: Form and Substance. Thomas Aquinas’ Critique of Eclectic Aristotelianism”, Archives d’histoire doctrinale et littéraire du moyen a¯ge 64, 95–126. Bermon, Emmanuel (2001) Le Cogito dans la pensée de Saint Augustin, Paris: Vrin. Bett, Richard (2000) Pyrrho, His Antecedents and His Legacy, Oxford: Clarendon.
243
244
Bibliography
Beversluis, John (2000) Cross-Examining Socrates, Cambridge: Cambridge University Press. Black, Deborah L. (1993) “Estimation (wahm) in Avicenna: The Logical and Psychological Dimensions”, Dialogue 32, 219–258. Bobzien, Susanne (1997) “Stoic Conceptions of Freedom and their Relation to Ethics”, in Richard Sorabji (ed.), Aristotle and After. Bulletin of the Institute of Classical Studies, Supplementary volume, 68, 71–89. Bobzien, Susanne (1998) Determinism and Freedom in Stoic Philosophy, Oxford: Oxford University Press. Bolton, Robert (1994) Person, Soul and Identity. A Neoplatonic Account of the Principle of Personality, London: Minerva Press. Bonhöffer, Adolf (1890) Epictet und die Stoa: Untersuchungen zur Stoischen Philosophie, Stuttgart: Frommann, reprinted 1968. Bonhöffer, Adolf (1894) Die Ethik der Stoikers Epictet, Stuttgart: Frommann, reprinted 1968, trans. W. O. Stephens, The Ethics of the Stoic Epictetus, New York: Peter Lang 1996. Brague, Remi (1997) “Cosmological Mysticism: The Imitation of the Heavenly Bodies in Ibn T.ufayl’s H. ayy Ibn Yaqz.a¯n”, Graduate Faculty Philosophical Journal 19.2–20.1, 91–102. Bratman, Michael (2002) “Reflection, Planning, and Temporally Extended Agency”, The Philosophical Review 109, 35–61. Brennan, Tad (1988) “The Old Stoic Theory of Emotions”, in Juha Sihvola and Troels EngbergPedersen (eds.), The Emotions in Hellenistic Philosophy, Dordrecht: Kluwer, 21–70. Brickhouse, Thomas and Smith, Nicholas (1994) Plato’s Socrates, Oxford: Oxford University Press. Broadie, Sarah (2001) “Soul and Body in Plato and Descartes”, Proceedings of the Aristotelian Society 101, 295–308. Brunschwig, Jacques (1986) “The Cradle Argument in Epicureanism and Stoicism”, in Malcom Schofield and Gisela Striker (eds.), Norms of Nature: Studies in Hellenistic Ethics, Cambridge: Cambridge University Press, 113–144. Burge, Tyler (1996) “Our Entitlement to Self-knowledge”, Proceedings of the Aristotelian Society 96, 91–116. Burnyeat, Myles (1982) “Idealism and Greek Philosophy: What Descartes Saw and Berkeley Missed”, Philosophical Review 91, 3–40. Burnyeat, Myles (1983) “Can the Skeptic Live His Skepticism?” in Myles Burnyeat (ed.), The Skeptical Tradition, Berkeley, CA: University of California Press, 117–148. Burnyeat, Myles (ed.) (1994) Socratic Studies, Cambridge: Cambridge University Press. Burnyeat, Myles (2002) “De Anima II 5”, Phronesis 47, 28–90. Bussanich, John (1997) “Non-Discursive Thought in Plotinus and Proclus”, Documenti e studi sulla tradizione filosofica medievale 8, 191–210. Bynum, Caroline W. (1995) The Resurrection of the Body in Western Christianity, 200–1336, New York: Columbia University Press. Caizzi, F. Decleva (1981) Pirrone: Testimonianze, Naples: Bibliopolis. Carpenter, Amber (2007) “On Plato’s Lack of Consciousness”, in Sara Heinämaa, Vili Lähteenmäki and Pauliina Remes (eds.), Consciousness: From Perception to Reflection in the History of Philosophy, Dordrecht: Springer, 29–48. Carrithers, Michael, Collins, Steven and Lukes, Steven (eds.) (1985) The Category of the Person: Anthropology, Philosophy, History, Cambridge: Cambridge University Press. Cary, Phillip (2000) Augustine’s Invention of the Inner Self. The Legacy of a Christian Platonist, New York: Oxford University Press. Caston, Victor (2005) “The Spirit and the Letter: Aristotle on Perception”, in Ricardo Salles (ed.), Metaphysics, Soul, and Ethics. Themes from the Work of Richard Sorabji, Oxford: Clarendon, 245–320. Chadwick, Henry (1967) The Early Church, Harmondsworth: Penguin. Cherniss, Harold (1957) notes, in Plutarch, Moralia, On the Face in the Moon, Loeb Classical Library vol. 12, ed. Harold Cherniss, Cambridge, MA: Harvard University Press.
Bibliography
245
Chisholm, Roderick M. (1969) “On the Observability of the Self”, Philosophy and Phenomenological Research 30, 7–21. Chittick, William K. (2001) The Heart of Islamic Philosophy. The Quest for Self-Knowledge in the Teachings of Afd.a¯l al-Dı¯n Ka¯sha¯nı¯, Oxford: Oxford University Press. Code, Alan (1986) “Aristotle: Essence and Accident”, in Richard E. Grandy and Richard Warner (eds.), Philosophical Grounds of Rationality: Intentions, Categories, Ends, Oxford: Clarendon, 411–439. Coleman, Janet (1992) Ancient and Medieval Memories. Studies in the Reconstruction of the Past, Cambridge: Cambridge University Press. Cooper, John M. (1987) “Contemplation and Happiness: A Reconsideration”, Synthese 72, 187–216. Cooper, John M. (1996) “Justice and Rights in Aristotle’s Politics”, Review of Metaphysics 49, 859–872. Cooper, John M. (2003) “Stoic Autonomy”, Social Philosophy and Policy 20, 1–29. Cooper, John M. (2004) Knowledge, Nature, and the Good. Essays on Ancient Philosophy, Princeton, NJ: Princeton University Press. d’Alverny, Marie-Therese (1959) “Anniyya-Annitas”, in Mélanges offerts à E. Gilson, Paris: J. Vrin, 59–91. Damasio, Antonio R. (1999) The Feeling of What Happens: Body and Emotion in the Making of Consciousness, New York: Harcourt Brace. Descartes, René (1897–1913) Oeuvres de Descartes (cited as AT), 12 vols, eds. C. Adam and P. Tannery, Paris: Cerf. de Strycker, Emile and Slings, Simon (1994) Plato’s Apology of Socrates, Leiden: E. J. Brill. Dennett, Daniel (1992) Consciousness Explained, London: Penguin. Detel, Wolfgang (2005) Foucault and Classical Antiquity: Power, Ethics and Knowledge, trans. D. Wigg-Wolf, Cambridge: Cambridge University Press. Dillon, John (1996) “An Ethic for the Late Antique Sage”, in Lloyd P. Gerson (ed.), The Cambridge Companion to Plotinus, Cambridge: Cambridge University Press, 315–335. Dobbin, Robert F. (1998) Epictetus: Discourses Book I. Translation with Introduction and Commentary, Oxford: Oxford University Press. Dodds, Eric R. (1973) The Ancient Concept of Progress and Other Essays on Greek Literature and Belief, Oxford: Oxford University Press. Doyle, James (2004) “Socrates and the Oracle”, Ancient Philosophy 24, 19–36. Edwards, Charles (1997) “Self-Scrutiny and Self-Transformation in Seneca’s Letters”, Greece and Rome 44, 23–38. Emilsson, Eyjólfur K. (2003) “Discursive and Non-Discursive Thought”, in H. Fossheim et al. (eds.), Non-Conceptual Aspects of Experience, Oslo: Unipub, 47–66. Engberg-Pedersen, Troels (1990) The Stoic Theory of Oikeiosis: Moral Development and Social Interaction in Early Stoic Philosophy, Aarhus: Aarhus University Press. Engberg-Pedersen, Troels (2000) Paul and the Stoics, Edinburgh: Clark. Engberg-Pedersen, Troels (ed.) (2001) Paul Beyond the Judaism/Hellenism Divide, Louisville, KY: Westminster John Knox. Engberg-Pedersen, Troels (2002) “The Reception of Graeco-Roman Culture in the New Testament: The Case of Romans 7.7–25”, in Mogens Müller and Henrik Tronier (eds.), The New Testament as Reception, Sheffield: Sheffield Academic Press, 32–57. Engberg-Pedersen, Troels (2003) “Radical Altruism in Philippians 2:4”, in J.T. Fitzgerald, T.H. Olbricht and L.M. White (eds.), Early Christianity and Classical Culture, FS Abraham Malherbe. SupNovT 110, Leiden: E. J. Brill, 197–214. Engberg-Pedersen, Troels (2004) “Stoicism in the Apostle Paul: A Philosophical Reading”, in Steven K. Strange and Jack Zupko (eds.), Stoicism: Traditions and Transformations, Cambridge: Cambridge University Press, 52–75. Everson, Stephen (1991) “The Objective Appearance of Pyrrhonism”, in Stephen Everson (ed.), Psychology, Companions to Ancient Thought 2, Cambridge: Cambridge University Press, 121–147.
246
Bibliography
Everson, Stephen (1997) Aristotle on Perception, Oxford: Clarendon. Fakhry, Majid (1994) Ethical Theories in Islam, Leiden: E. J. Brill. Fine, Gail (2003) “Subjectivity, Ancient and Modern. The Cyrenaics, Sextus, and Descartes”, in Jon Miller and Brad Inwood (eds.), Hellenistic and Early Modern Philosophy, Cambridge: Cambridge University Press, 192–231. Fine, Kit (1994) “A Puzzle Concerning Matter and Form”, in Theodor Scaltsas, David Charles and Mary Louise Gill (eds.), Unity, Identity and Explanation in Aristotle’s Metaphysics, Oxford: Clarendon, 13–40. Foucault, Michel (1984) Histoire de la sexualité, 2 vols, Paris: Gallimard 1984. Foucault, Michel (1988a) The Care of the Self: The History of Sexuality vol. 3, trans. R. Hurley, London: Penguin. Foucault, Michel (1988b) “Technologies of the Self ”, in L.H. Martin, H. Gutman and P.H. Hutton (eds.), Technologies of the Self, Amherst, MA: University of Massachusetts, 16–49. Foucault, Michel (1997) “Self Writing”, in Ethics: Subjectivity and Truth, New York: The New Press, 207–221. Frank, Richard M. (1994) Al-Ghazali and the Ash‘arite School, Durham, NC: Duke University Press. Frankfurt, Harry (1971) “Freedom of the Will and the Concept of a Person”, Journal of Philosophy 67, 5–20. Frankfurt, Harry (1988) The Importance of What We Care About: Philosophical Essays, Cambridge: Cambridge University Press. Frede, Michael (1992) “Plato’s Arguments and the Dialogue Form”, in James C. Klagge and Nicholas D. Smith (eds.), Methods of Interpreting Plato and His Dialogues, Oxford Studies in Ancient Philosophy, Supplementary volume, Oxford: Oxford University Press, 201–219. Frede, Michael (1994) “The Stoic Concept of Reason”, in I.K. Boudouris (ed.), Hellenistic Philosophy, Athens: Ionia, 50–63. Frede, Michael (1999) “Epilogue”, in Keimpe A. Algra, Jonathan Barnes, Jaap Mansfeld and Malcom Schofield (eds.), The Cambridge History of Hellenistic Philosophy, Cambridge: Cambridge University Press. Frede, Michael and Patzig, Günther (1988) Aristoteles, “Metaphysik Z” Text, Übersetzung und Kommentar I, München: Verlag C.H. Beck. Gallagher, Shaun and Shear, Jonathan (eds.) (1999) Models of the Self, Thorverton: Imprint Academic. Gerson, Lloyd P. (1997a) “Introspection, Self-Reflexivity, and the Essence of Thinking”, in John J. Cleary (ed.), The Perennial Tradition of Neoplatonism, Leuven: Leuven University Press, 153–173. Gerson, Lloyd P. (1997b) “Epistroph¯e pros heauton: History and Meaning”, Documenti e studi sulla tradizione filosofica medievale, 8, 1–32. Ghaza¯lı¯, Abu¯ H. a¯mid Muh. ammad al- (1963) Ma‘a¯rij al-quds f ¯ı mada¯rij al-ma‘rifa al-nafs, Cairo: al-Maktaba al-tija¯riyya al-kubrı¯. Ghaza¯lı¯, Abu¯ H. a¯mid Muh. ammad al- (1964a) Fad. a¯’ih. al-ba¯t. iniyya wa fad. a¯’il al-mustaz.hiriyya, ed. ‘A. Badawı¯, Cairo: Al-maktaba al-’arabiyya. Ghaza¯lı¯, Abu¯ H. a¯mid Muh. ammad al- (1964b) Mı¯za¯n al-‘amal, ed. S. Dunya¯, Cairo: Da¯r alma‘a¯rif bi al-mis. r. Ghaza¯lı¯, Abu¯ H. a¯mid Muh. ammad al- (1969) Al-Munqidh min al-d. ala¯l, eds. K. Ayya¯d and J. Saliba, Beirut: Librairie Orientale. Ghaza¯lı¯, Abu¯ H. a¯mid Muh. ammad al- (1982) Al-maqs.ad al-asna¯ fı¯ sharh. ma‘a¯ni asma¯’ Alla¯h al-h. usna¯, ed. F.A. Shehadi, Beirut: Da¯r al-Mashriq. Ghaza¯lı¯, Abu¯ H. a¯mid Muh. ammad al- (1983) Kı¯miya¯-yi sa‘a¯dat, 2 vols, ed. H. usayn Khadı¯v-jam, Teheran: T. ulu¯ra-zarrı¯n. Ghaza¯lı¯, Abu¯ H. a¯mid Muh. ammad al- (1998) Mishka¯t al-anwa¯r, in The Niche of Lights, ed. and trans. D. Buchman, Provo, UT: Brigham Young University Press.
Bibliography
247
Ghaza¯lı¯, Abu¯ H. a¯mid Muh. ammad al- (2000) Taha¯fut al-fala¯sifa/The Incoherence of the Philosophers. A Parallel English-Arabic Text, ed. and trans. Michael E. Marmura, Provo, UT: Brigham Young University Press. Ghaza¯lı¯, Abu¯ H. a¯mid Muh. ammad al- (2002) Ih. ya¯’ ‘ulu¯m al-dı¯n, 5 vols, Beirut: Da¯r al-kutub al-‘ilmiyyah. Gianotti, Timothy J. (2001) Al-Ghaza¯lı¯’s Unspeakable Doctrine of the Soul, Leiden: E. J. Brill. Gill, Christopher (1988) “Personhood and Personality: The Four-Personae Theory in Cicero, De Officiis 1”, Oxford Studies in Ancient Philosophy 6, 169–199. Gill, Christopher (1990) The Person and the Human Mind: Issues in Ancient and Modern Philosophy, Oxford: Oxford University Press. Gill, Christopher (1991) “Is There a Concept of Person in Greek Philosophy?” in Stephen Everson (ed.), Psychology, Companions to Ancient Thought 2, Cambridge: Cambridge University Press, 166–193. Gill, Christopher (ed.) (1995), Epictetus: The Discourses, Handbook, Fragments, London: Dent. Gill, Christopher (1996) Personality in Greek Epic, Tragedy, and Philosophy: The Self in Dialogue, Oxford: Oxford University Press. Gill, Christopher (1998) “Did Galen Understand Platonic and Stoic Thinking on Emotions?”, in Troels Engberg-Pedersen and Juha Sihvola (eds.), The Emotions in Hellenistic Philosophy, The New Synthese Historical Library 46, Dordrecht: Kluwer, 113–148. Gill, Christopher (2000) “The Body’s Fault? Plato’s Timaeus on Psychic Illness”, in M.R. Wright (ed.), Reason and Necessity in Plato’s Timaeus, London: Duckworth, 59–84. Gill, Christopher (2003) “The School in the Roman Imperial Period”, in Brad Inwood (ed.), The Cambridge Companion to the Stoics, Cambridge: Cambridge University Press. Gill, Christopher (2004) “The Stoic Theory of Ethical Development: In What Sense Is Nature a Norm?”, in Jan Szaif and Mathias Lutz-Bachmann (eds.), Was ist das für den Menschen Gute? Menschliche Natur und Güterlehre/What Is Good for a Human Being? Human Nature and Values, Berlin: De Gruyter. Gill, Christopher (2005) “In What Sense Are Ancient Ethical Norms Universal?”, in Christopher Gill (ed.), Virtue, Norms, and Objectivity: Issues in Ancient and Modern Ethics, Oxford: Oxford University Press. Gill, Christopher (2006) The Structured Self in Hellenistic and Roman Thought, Oxford: Oxford University Press. Gill, Mary Louise (1994) “Individuals and Individuation in Aristotle”, in Theodor Scaltsas, David Charles and Mary Louise Gill (eds.), Unity, Identity and Explanation in Aristotle’s Metaphysics, Oxford: Clarendon, 55–71. Green, J.S. (2005) Reflections on Selfhood in Gorgias’s Encomium of Helen and Plato’s Apology of Socrates, unpublished Ph.D. dissertation, University of London. Griffin, M.T. and Atkins, E.M. (1991) Cicero: On Duties, Cambridge: Cambridge University Press. Hadot, Pierre (1992) The Inner Citadel. The Meditations of Marcus Aurelius, London: Blackwell. Hadot, Pierre (1995) Philosophy as a Way of Life. Pneumatic Exercises from Socrates to Foucault, Oxford: Blackwell. Harle, Anne (1999) “Augustine and Rousseau: Narrative and Self-Knowledge in the Two Confessions”, in Gareth B. Matthews (ed.), The Augustinian Tradition, Berkeley and Los Angeles, CA: University of California Press, 263–285. Hartman, Edwin (1978) Substance, Body, and Soul, Princeton, NJ: Princeton University Press. Heinämaa, Sara, Lähteenmäki, Vili and Remes, Pauliina (eds.) (2007) Consciousness: From Perception of Reflection in the History of Philosophy, Dordrecht: Springer. Hill, Timothy (2004) Ambitiosa Mors: Suicide and the Self in Roman Thought and Literature, London: Routledge. Hintikka, Jaakko (1962) “Cogito, ergo sum: Inference or Performance?” Philosophical Review 71, 3–32.
248
Bibliography
Husserl, Edmund (1966) Zur Phänomenologie des inneren Zeitbewusstseins (1893–1917), Husserliana 10, ed. Rudolph Boehm, The Hague: Martinus Nijhoff. English translation: On the Phenomenology of Consciousness of Internal Time (1893–1917), trans. John Barnett Brough, Dordrecht: Kluwer 1991. Ibn ‘Adı¯, Yah. ya¯ (2002) The Reformation of Morals. A Parallel Arabic-English Edition, ed. S. K. Kussaim, tr. S. H. Griffith, Provo, UT: Brigham Young University Press. Ibn al-‘Arabı¯, Muh. yı¯ al-Dı¯n (2002) Fus.u¯s. al-h. ikam, ed. A. ‘Afifı¯, Beirut: Da¯r al-kita¯b al-‘arabı¯. Ibn Sı¯na¯, Abu¯ ‘Alı¯ (1875) Maqa¯la fı¯ al-nafs ‘ala¯ sunna l-ikhtis.a¯r, ed. S. Landauer under the title “Die Psychologie des Ibn Sı¯na¯”, Zeitschrift der deutschen morgenländischen Gesellschaft 29, 335–418. Ibn Sı¯na¯, Abu¯ ‘Alı¯ (1952) Avicenna’s Psychology, English translation of Kit¯ab al-naj¯at, ed. and trans. F. Rahman, London: Oxford University Press. Ibn Sı¯na¯, Abu¯ ‘Alı¯ (1959) Avicenna’s De Anima (Arabic Text). Being the Psychological Part of Kita¯b al-Shifa¯’, ed. and trans. F. Rahman, London: Oxford University Press. Ibn Sı¯na¯, Abu¯ ‘Alı¯ (1960) al-Isha¯ra¯t wa al-tanbı¯ha¯t, 4 vols, ed. S. Dunya¯, Cairo: Da¯r al-ma¢a¯rif bi al-mis. r. Inwood, Brad (1985) Ethics and Human Action in Early Stoicism, Oxford: Oxford University Press. Inwood, Brad (1999) “Rules and Reasoning in Stoic Ethics”, in Katerina Ierodiakonou (ed.), Topics in Stoic Philosophy, Oxford: Oxford University Press, 112–127. Inwood, Brad and Donini, Pierluigi (1999) “Stoic Ethics”, in Keimpe A. Algra, Jonathan Barnes, Jaap Mansfeld and Malcom Schofield (eds.), The Cambridge History of Hellenistic Philosophy, Cambridge: Cambridge University Press, 675–738. Irwin, Terrence (1991) “Aristippus against Happiness”, The Monist 74, 55–82. Jabre, Farid (1958) La notion de la ma’rifa chez Ghazali, Beirut: Lettres Orientales. Jackson, Sherman (2002) The Boundaries of Theological Tolerance in Islam: Abu¯ Ha¯mid alGhaza¯lı¯’s Faysal al-Tafriqa Bayna al-Islam wa al-Zandaqa, Oxford: Oxford University Press. Jacobs, Jonathan (2002) “Aristotle and Maimonides: The Ethics of Perfection and the Perfection of Ethics”, American Catholic Philosophical Quarterly 76, 145–163. Kahn, Charles H. (1988) “Discovering the Will: From Aristotle to Augustine”, in John Dillon and A.A. Long (eds.), The Question of ‘Eclecticism’: Studies in Later Greek Philosophy, Berkeley, CA: California University Press, 234–259. Kamtekar, Rachana (2006) “Plato on the Attribution of Conative Attitudes”, Archiv für Geschichte der Philosophie 88, 127–162. Kateb, George (1998) “Socratic Integrity”, in Ian Shapiro and Robert Adams (eds.), Integrity and Conscience, Nomos 40, New York: NewYork University Press, 77–112. Kaukua, Jari (2007) Avicienna on Subjectivity. A Philosophical Study, Jyväskylä: Jyväskylä University Printing House. Kenney, Peter John (2006) “Commentary on Stern-Gillett”, Proceedings of the Boston Area Colloquium in Ancient Philosophy 22, 175–181. Kenny, Anthony (1988) The Self, The Aquinas Lecture, Milwaukee, WI: Marquette University Press. Kenny, Anthony (1992) Aristotle on the Perfect Life, Oxford: Clarendon. Ketchum, Robert (1991) “Plato on the Uselessness of Epistemology”, Apeiron 24, 81–98. Knuuttila, Simo (2001) “Augustine on Time and Creation”, in Eleanor Stump and Norman Kretzman (eds.), The Cambridge Companion to Augustine, Cambridge: Cambridge University Press, 103–115. Knuuttila, Simo (2004) Emotions in Ancient and Medieval Philosophy, Oxford: Oxford University Press. Korsgaard, Christine (1996) The Sources of Normativity, Cambridge: Cambridge University Press.
Bibliography
249
Korsgaard, Christine (1999) “Self-Constitution in the Ethics of Plato and Kant”, The Journal of Ethics 3, 1–29. Kosman, Louis Aryeh (1992) “Silence and Imitation in the Platonic Dialogues”, in James C. Klagge and Nicholas D. Smith (eds.), Methods of Interpreting Plato and His Dialogues, Oxford Studies in Ancient Philosophy, supplementary volume, Oxford: Oxford University Press, 73–92. Kosman, Louis Aryeh (2004) “Aristotle on the Desirability of Friends”, Ancient Philosophy 24, 135–154. Kraut, Richard (1989) Aristotle on the Human Good, Princeton, NJ: Princeton University Press. Kukkonen, Taneli, “Ibn T. ufayl and the Wisdom of the East: On Apprehending the Divine”, in Stephen R.L. Clark and Panayiota Vassilopoulou (eds.) (2008), Late Antique Epistemology, London: Macmillan, forthcoming. Le Blond, Jean-Marie (1950) Les conversions de saint Augustine, Théologie 17, Paris: Aubier. Lear, Jonathan (1992) “Inside and Outside the Republic”, Phronesis 37, 184–215. Lichtenberg, G.C. (1971) Schriften und Briefe. Sudelbücher II, München: Carl Hanser Verlag. Lloyd, Genevieve (1999) “Augustine and the ‘Problem’ of Time”, in Gareth B. Matthews (ed.), The Augustinian Tradition, Berkeley: University of California Press 1999, 39–60. Long, A.A. (1991) “Representation and the Self in Stoicism”, in Everson (ed.), Psychology, Companions to Ancient Thought 2, Cambridge: Cambridge University Press, 102–120. Long, A.A. (1996) Stoic Studies, Cambridge: Cambridge University Press. Long, A.A. (2001) “Ancient Philosophy’s Hardest Question: What to Make of Oneself?”, Representations 74, 19–36. Long, A.A. (2002) Epictetus. A Stoic and Socratic Guide to Life, Oxford: Clarendon. Long, A.A. and Sedley, D.N. (1987) The Hellenistic Philosophers, 2 vols, Cambridge: Cambridge University Press. Loux, Michael (1991) Primary Ousia. An Essay on Aristotle’s Metaphysics Z and H, Ithaca, NY: Cornell University Press. Luria, A.R (1973) Man with a Shattered World, London: Jonathan Cape. Macintyre, Alasdair (1981) After Virtue, London: Duckworth. MacNamara, Marie A. (1964) Friends and Friendship for St. Augustine, Staten Island, NY: Alba House. Maneri, Sharafuddin Ibn Yah. ya¯ (1980) The Hundred Letters, trans. P. Jackson, New York: Paulist Press. Marcotte, Roxanne (1999) “The Role of Imagination (mutakhayyilah) in Ibn Miskawayh’s Theory of Prophecies”, American Catholic Philosophical Quarterly 73, 37–72. Martin, Dale B. (1995) The Corinthian Body, London: Yale University Press, 123–129. Martin, Raymond and Barresi, John (2006) The Rise and Fall of Soul and Self. An Intellectual History of Personal Identity, New York: Columbia University Press. Matthaei ab Aquasparta (1957) Quaestiones disputatae de fidei et de cognitione, Biblioteca Franciscana Scholastica medii aevi 1, Quaracchi: Florence. Matthews, Gareth B. (1992) Thought’s Ego in Augustine and Descartes, Ithaca, NY: Cornell University Press. Matthews, Gareth B. (2000) “Internalist Reasoning in Augustine for Mind-Body Dualism”, in John P. Wright and Paul Potter (eds.), Psyche and Soma. Physicians and Metaphysicians on the Mind-Body Problem from Antiquity to Enlightement, Oxford: Clarendon, 133–145. Matthews, Gareth B. (2001) “Knowledge and Illumination”, in Eleanore Stump and Norman Kretzmann (eds.), The Cambridge Companion to Augustine, Cambridge: Cambridge University Press, 171–185. Matthews, Gareth B. (2005) Augustine, Malden & Oxford: Blackwell. Mayer, Cornelius (1994) “Conscientia”, in C. Mayer (ed.), Augustinus-Lexikon 1, Basel: Schwabe. McCabe, Mary Margaret (1993) “Persistent Fallacies”, Proceedings of the Aristotelian Society 1993, 73–93.
250
Bibliography
McCabe, Mary Margaret (1994) Plato’s Individuals, Princeton, NJ: Princeton University Press. McCabe, Mary Margaret (1998) “Silencing the Sophists: The Drama of Plato’s Euthydemus”, Proceedings of the Boston Area Colloquium in Ancient Philosophy, 14, 139–168. McCabe, Mary Margaret (2000) Plato and his Predecessors: The Dramatization of Reason, Cambridge: Cambridge University Press. McCabe, Mary Margaret (2001) “Developing the Good: Prolepsis or Critique in the Euthydemus?”, Plato: Electronic Journal of the International Plato Society 2, available online at http://www. nd.edu/~plato/plato2issue/mccabe.htm. McCabe, Mary Margaret (2002) “Indifference Readings: Plato and the Stoa on Socratic Ethics”, in T.P. Wiseman (ed.), Classics in Progress: Essays on Ancient Greece and Rome, Oxford: Oxford University Press, 363–398. McCabe, Mary Margaret (2006) “Does Your Plato Bite?”, in John Dillon and Monique Dixsaut (eds.), Agonistes: Festschrift for Denis O’Brien, London: Ashgate, 107–120. McEvoy, James (1997) “Does Augustinian Memory Depend on Plotinus?”, in John J. Cleary (ed.), The Perennial Tradition of Neoplatonism, Leuven: Leuven University Press, 383–396. McPherran, Mark (1996) The Religion of Socrates, University Park, PA: Pennsylvania State University Press. Menn, Stephen (1998) Descartes and Augustine, Cambridge: Cambridge University Press. Milinda’s Questions, Sacred Books of the Buddhists (1964) vols. 22 and 23, trans. I.B. Horner, London: Luzac. Mill, John Stuart (1987) Utilitarianism, in Alan Ryan (ed.), John Stuart Mill and Jeremy Bentham: Utilitarianism and Other Essays, New York: Penguin. Miller, Fred D. (1995) Nature, Justice and Rights in Aristotle’s Politics, Oxford: Clarendon. Miller, William Ian (2003) Faking It, Cambridge: Cambridge University Press. Moller, Dan (2004) “The Pyrrhonian Skeptic’s Telos”, Ancient Philosophy 24, 425–441. Moosa, Ebrahim (2005) Ghaza¯lı¯ and the Poetics of Imagination, Chapel Hill, NC: The University of North Carolina Press. Moran, Richard (2001) Authority and Estrangement, Princeton, NJ: Princeton University Press. Nehamas, Alexander (1998) The Art of Living, Berkeley, CA: University of California Press. Neisser, Ulric (1993) The Perceived Self. Ecological and Interpersonal Sources of Self-Knowledge. Emory Symposia in Cognition 5, Cambridge: Cambridge University Press. Nozick, Robert (1981) Philosophical Explanations, Oxford: Oxford University Press. Nussbaum, Martha C. (1985) “Discernment of Perception: An Aristotelian Conception of Private and Public Rationality”, Proceedings of the Boston Area Colloquium in Ancient Philosophy 3, 151–201. Nussbaum, Martha C. (1986) Fragility of Goodness. Luck and Ethics in Greek Tragedy and Philosophy, Cambridge: Cambridge University Press. Nussbaum, Martha C. (1990) “Aristotelian Social Democracy”, in R. Bruce Douglass, Gerald M. Mara and Henry S. Richardson (eds.), Liberalism and the Good, New York & London: Routledge, 203–252. Nussbaum, Martha C. (1999) “Augustine and Dante on the Ascent of Love”, in Gareth B. Matthews (ed.), The Augustinian Tradition, Berkeley & Los Angeles, CA: University of California Press, 61–90. Nussbaum, Martha C. (2001) Upheavals of Thought: The Intelligence of Emotions, Cambridge: Cambridge University Press. Nussbaum, Martha C. (2006) “The ‘Morality of Pity’: Sophocles’ Philoctetes and the European Stoics”, in Vesa Hirvonen, Toivo Holopainen and Miira Tuominen (eds.), Mind and Modality: Studies in the History of Philosophy in Honour of Simo Knuuttila, Brill’s Studies in Intellectual History 144, Leiden: E. J. Brill, 3–18. Nussbaum, Martha C. and Rorty, Amélie (eds.) (1992), Essays on Aristotle’s De Anima, Oxford: Clarendon. O’Daly, Gerard (1973) Plotinus’ Philosophy of the Self, Shannon: Irish University Press. O’Daly, Gerard (1987) Augustine’s Philosophy of Mind, Berkeley & Los Angeles, CA: University of California Press/London: Duckworth.
Bibliography
251
O’Daly, Gerard (1993) “Remembering and Forgetting in Augustine, Confessions X”, in Anselm Haverkamp and Renate Lachmann (eds.), Memoria. Vergessern und Erinnern, Poetik und Hermeneutik 15, München: Wilhelm Fink Verlag, 31–46. O’Daly, Gerard (2004) Augustine’s City of God: A Reader’s Guide, 2nd edn, Oxford: Oxford University Press. O’Daly, Gerard, “Subject and Substance: Self-Knowledge in Augustine’s On the Trinity”, in Matthias Vorwerk (ed.), Augustine and Augustinianism. Contributions on Augustinian Philosophy and Its Reception. Catholic University of America School of Philosophy Fall Lecture Series 2006 Washington, DC, forthcoming. O’Donovan, Oliver (1980) The Problem of Self-Love in St. Augustine, New Haven, CT: Yale University Press. O’Neill, Onora (2000) Bounds of Justice, Cambridge: Cambridge University Press. Olivi, Peter John (1924) Quaestiones in secundum librum Sententiarum, Bibliotheca Franciscana Scholastica Medii Aevi 4–6, vol. 2, ed. B. Jansen, Quaracci: Collegium S. Bonaventurae. Olivi, Peter John (1926) Quaestiones in secundum librum Sententiarum, Bibliotheca Franciscana Scholastica Medii Aevi 4–6, vol. 3, ed. B. Jansen, Quaracci: Collegium S. Bonaventurae. Olson, Eric T. (1999) “There Is No Problem of the Self”, in Shaun Gallagher and Jonathan Shear (eds.), Models of the Self, Thorverton: Imprint Academic, 49–61. Oppy, Graham and Trakakis, Nick (eds.), History of Western Philosophy of Religion, Ancient Philosophy and Religion, vol. 1, Acumen, forthcoming. Owen, G.E.L. (1978) “Particular and General”, Proceedings of the Aristotelian Society 79, 1–21. Oxford: Oxford University Press. Parfit, Derek (1984) Reasons and Persons, Oxford: Clarendon. Pépin, Jean (1971) “Héraclès et son reflet dans le néoplatonisme”, in E.O. Bel (ed.), Le Néoplatonisme, Paris: Colloque du CNRS, Royaumont, 167–192. Perin, Casey (2006) “Pyrrhonian Scepticism and the Search for Truth”, Oxford Studies in Ancient Philosophy 30, 337–360. Perkams, Matthias (2006) “An Innovation by Proclus. The Theory of the Substantial Diversity of the Human Soul”, in Matthias Perkams and Rosa Maria Piccione (eds.), Proklos. Methode, Seelenlehre, Metaphysik, Leiden & Boston: E. J. Brill, 165–185. Powell, Jonathan G.F. (1990) Cicero: On Friendship and the Dream of Scipio, Warminster: Aris & Philips. Pradeau, Jean-Francois (2002) “Le sujet ancien d’une politique moderne: à propos des exercises spirituels anciens dans L’Histoire de la sexualité de Michel Foucault”, in Frédéric Gros et al. (eds.), Foucault: le courage et la vérité, Paris: Presses Universitaires de France, 131–154. Pradeau, Jean-Francois (2003) “Le sujet ancien d’une politique moderne: sur la subjectivation et l’éthiques anciennes dans les Dits et écrits de Michel Foucault”, in Pierre-Francois Moreau (ed.), Lectures de Michel Foucault, vol. 3. Sur les Dits et Écrits, Lyons: École normale supérieure Lettres et sciences humaines, 35–51. Putallaz, François-Xavier (1991) Le sens de la réflexion chez Thomas D’Aquin, Études de philosophie Médiévale 66, Paris: Vrin. Rahman, Fazlur (1965) “Dha¯t”, in The Encyclopaedia of Islam. New Edition, vol. 2, Leiden: E. J. Brill. Rappe, Sara (1995) “Socrates and Self-Knowledge”, Apeiron 28, 1–24. Ra¯zı¯, Fakhr ad-Dı¯n ar- (1968) Kita¯b al-nafs wa al-ru¯h. wa sharh. qawa¯-huma¯, ed. M. Saqhı¯r H. asan al-Ma¢su¯mı¯, Islamabad: Islamic Research Institute. Reeve, C.D.C. (1989) Socrates in the Apology, Indianapolis, IN: Hackett. Reiss, Timothy J. (2003) Mirages of the Selfe. Patterns of Personhood in Ancient and Early Modern Europe, Palo Alto, CA: Stanford University Press. Remes, Pauliina (2007a) Plotinus on Self: The Philosophy of the “We”, Cambridge: Cambridge University Press. Remes, Pauliina (2007b) “Ownness of Experience in Antiquity”, in Sara Heinämaa, Vili Lähteenmäki and Pauliina Remes (eds.), Consciousness: From Perception to Reflection in the History of Philosophy, Dordrecht: Springer, 67–94.
252
Bibliography
Reydams-Schils, Gretchen (2002) “Human Bonding and Oikei¯osis”, Oxford Studies in Ancient Philosophy 22, 221–251. Reydams-Schils, Gretchen (2005) The Roman Stoics: Self, Responsibility, and Affection, Chicago, IL: Chicago University Press. Reynolds, Philip Lyndon (1992) “The Essence, Power and Presence of God: Fragments of the History of an Idea, from Neopythagoreanism to Peter Abelard”, in H.J. Westra (ed.), From Athens to Chartres, Leiden: E. J. Brill, 351–380. Richardson, Henry S. (2001) “Autonomy’s Many Normative Presuppositions”, American Philosophical Quarterly 38, 287–303. Ricoeur, Paul (1985) Temps et récit III: Le temps raconté, Paris: Éditions du Seuil. Roochnik, David (1990) “The Serious Play of Plato’s Euthydemus”, Interpretation 18, 211–321. Rorty, Amélie (1976) The Identities of Persons, Berkeley, CA: California University Press. Rutherford, R.B. (1989) The Meditations of Marcus Aurelius. A Study, Oxford: Clarendon. Schechtman, Marya (1996) The Constitution of Selves, Ithaca, NY: Cornell University Press. Schmid, Walter (1998) Plato’s Charmides and the Socratic Ideal of Rationality, Albany, NY: State University of New York Press. Scott, Dominic (1995) Recollection and Experience: Plato’s Theory of Learning and Its Successors, Cambridge: Cambridge University Press. Scourfield, J.H. David (1993) Consoling Heliodorus: A Commentary on Jerome Letter 60, Oxford: Clarendon. Sedley, David (1982) “The Stoic Criterion of Identity”, Phronesis 27, 255–275. Sellars, John (2003) The Art of Living: The Stoics on the Nature and Function of Philosophy, Aldershot: Ashgate. Shields, Christopher (1999) Order in Multiplicity. Homonymy in the Philosophy of Aristotle, Oxford: Clarendon. Sihvola, Juha (2007) “The Problem of Consciousness in Aristotle’s Psychology”, in Sara Heinämaa, Vili Lähteenmäki and Pauliina Remes (eds.), Consciousness. From Perception to Reflection in the History of Philosophy, Dordrecht: Springer, 49–66. Snell, Bruno (1953) The Discovery of the Mind: The Greek Origins of European Thought, trans. T.G. Rosenmayer, New York: Harper & Row. Sorabji, Richard (1983) Time, Creation and the Continuum. Theories in Antiquity and the Early Middle Ages, London: Duckworth. Sorabji, Richard (2000a) Emotion and Peace of Mind, Oxford: Oxford University Press. Sorabji, Richard (2000b) “Is the True Self an Individual in the Platonist Tradition?” in MarieOdile Goulet-Cazé (ed.), Le Commentaire entre tradition et innovation, Paris: Librairie Philosophique J. Vrin, 293–300. Sorabji, Richard (2004a) Philosophy of the Commentators 200–600 AD: A Sourcebook, vol. 1 Psychology, London: Duckworth. Sorabji, Richard (2004b) The Philosophy of the Commentators 200–600 AD: A Sourcebook, vol. 3 Logic and Metaphysics, London: Cornell University Press. Sorabji, Richard (2006) Self: Ancient and Modern Insights About Individuality, Life, and Death, Oxford: Oxford University Press. Sprague, Rosamond Kent (1962) Plato’s Use of Fallacy, London: Routledge & Kegan Paul. Steel, Carlos (1978) The Changing Self. A Study on the Soul in Later Neoplatonism: Iamblichus, Damascius and Priscianus. Verhandelingen van den Koninklijke Academie voor Wetenschappen, Letteren en Schone Kunsten van België, Klasse der Letteren, Jaargang 40, Brussels. Stern-Gillet, Suzanne (1995) Aristotle’s Philosophy of Friendship, Albany, NY: State University of New York Press. Stern-Gillett, Suzanne (2006) “Consciousness and Introspection in Plotinus and Augustine”, Proceedings of the Boston Area Colloquium in Ancient Philosophy 22, 145–174. Stock, Brian (1996) Augustine. The Reader. Meditation, Self-Knowledge, and the Ethics of Interpretation, Cambridge, MA: The Belknap Press. Stokes, Michael (1997) Plato: Apology, Warminster: Aris & Philips.
Bibliography
253
Strange, Stephen K. (1996) “Plotinus on the Nature of Eternity and Time”, in Lawrence P. Schrenk (ed.), Aristotle in Late Antiquity, Studies in Philosophy and the History of Philosophy 27, Washington, DC: Catholic University of America Press, 22–53. Strawson, Peter (1959) Individuals, London: Methuen. Tarrant, Harold (ed.) (1994) “The Hippias Major and Socratic Theories of Pleasure”, in Paul Vander Waerdt (ed.), The Socratic Movement, Ithaca, NY: Cornell University Press, 107–126. Tarrant, Harold (2000) “Naming Socratic Interrogation in the Charmides”, in T.M. Robinson and Luc Brisson (eds.), Plato: Euthydemus, Lysis, Charmides, Sankt Augustin: Academia Verlag, 251–258. Taylor, Alfred E. (1928) Commentary on Plato’s Timaeus, Oxford: Clarendon. Taylor, Charles (1985) Philosophical Papers 1: Human Agency and Language, Cambridge: Cambridge University Press. Taylor, Charles (1989) Sources of the Self, Cambridge, MA: Harvard University Press. Taylor, Charles (1991) The Malaise of Modernity, Concord, ON: Anansi. Teske, Roland (2001) “Augustine’s Philosophy of Memory”, in Eleanor Stump and Norman Kretzman (eds.), The Cambridge Companion to Augustine, Cambridge: Cambridge University Press, 148–158. Tieleman, Teun, “Panaetius’ Place in the History of Stoicism” with Special Reference to his Moral Psychology, in A.M. Ioppolo and D.N. Sedley (eds.) Pyrrhonists, Patricians, Platonizers: Hellenistic Philosophy in the period 155–86 BC, Naples: Bibliopolis 2007, 104– 142, esp. 130–140. Tomasello, Michael (1993) “On the Interpersonal Origins of Self-Concept”, in Ulric Neisser (ed.), The Perceived Self, Cambridge: Cambridge University Press. Tomasello, Michael (1999) The Cultural Origins of Human Cognition, Cambridge: Harvard University Press. Tsouna, Voula (2001) “Socrate et la connaissance de soi: quelques interpretations”, Philosophie Antique 1, 37–64. Tsouna-McKirahan, Voula (1997) Cyrenaic Epistemology, Cambridge: Cambridge University Press. van Straaten, Mark (1962) Panaetii Rhodii Fragmenta, Leiden: E. J. Brill. Verbeke, Gerard (1997) “Individual Consciousness in Neoplatonism”, in John J. Cleary (ed.), The Perennial Tradition of Neoplatonism, Leuven: Leuven University Press, 135–152. Vlastos, Gregory (1981) Platonic Studies, Princeton, NJ: Princeton University Press. Vlastos, Gregory (1991) Socrates, Ironist and Moral Philosopher, Ithaca, NY: Cornell University Press. Vlastos, Gregory (1994) “The Socratic Elenchus: Method Is All”, in Myles Burnyeat (ed.), Socratic Studies, Cambridge: Cambridge University Press, 1–37. Warren, Edward W. (1964) “Consciousness in Plotinus”, Phronesis 9, 83–97. Warren, James (2001) “Lucretian pangenesis Recycled”, Classical Quarterly 51, 499–508. Warren, James (2002) Epicurus and Democritean Ethics: An Archaeology of Ataraxia, Cambridge: Cambridge University Press. Watson, Gary (1982) “Free Agency” in Gary Watson (ed.), Free Will, Oxford: Oxford University Press, 96–110. White, Carolynne (1992) Christian Friendship in the Fourth Century, Cambridge: Cambridge University Press. Wilkes, K.V. (1999) “Know Thyself”, in Shaun Gallagher and Jonathan Shear (eds.), Models of the Self, Thorverton: Imprint Academic, 25–38. Williams, Bernard (1973) Problems of the Self, Cambridge: Cambridge University Press. Williams, Bernard (1985) Ethics and the Limits of Philosophy, London: Fontana. Williams, Bernard (1986) “Hylomorphism”, Oxford Studies in Ancient Philosophy 4, 189–199. Williams, Bernard (1993) Shame and Necessity, Berkeley, CA: California University Press. Winch, Peter (1965) “The Universalisability of Moral Judgement”, The Monist 49, 196–214.
254
Bibliography
Winter, Tim (1995) Al-Ghaza¯lı¯: Disciplining the Soul, Refining the Character, and Curing the Sicknesses of the Heart & Breaking the Two Desires, Cambridge: Islamic Texts Society. Witt, Charlotte (1985) “Form, Reproduction, and Inherited Characteristics in Aristotle’s Generation of Animals”, Phronesis 30, 46–57. Witt, Charlotte (1989) Substance and Essence in Aristotle. An Interpretation of Metaphysics VII–IX, Ithaca, NY: Cornell University Press. Woodruff, Paul (1987) “Expert Knowledge in the Apology and Laches: What a General Needs to Know”, Proceedings of the Boston Area Colloquium in Ancient Philosophy 3, 79–115. Woods, Michael (1994) “The Essence of a Human Being and the Individual Soul in Metaphysics Z and H”, in Theodor Scaltsas, David Charles and Mary Louise Gill (eds.), Unity, Identity and Explanation in Aristotle’s Metaphysics, Oxford: Clarendon, 279–290. Woolf, Raphael (2004) “A Shaggy Soul Story: How Not To Read The Wax Tablet Model in Plato’s Theaetetus”, Philosophy and Phenomenological Research 69, 573–604. Yonezawa, Shigeru (1995) “Socratic Knowledge and Socratic Virtue”, Ancient Philosophy 15, 349–358. Yrjönsuuri, Mikko (2002) “Free Will and Self-Control in Peter Olivi”, in Henrik Lagerlund and Mikko Yrjönsuuri (eds.), Emotions and Choice from Boethius to Descartes, Dordrecht: Kluwer, 99–128. Zahavi, Dan (2002) “Selvet i filosofisk belysning”, in P. Bertelsen, M. Hermansen and J. Tønnesvang (eds.), Vinkler på selvet – en antologi om selvbegrebets anvendelse i psykologien, Aarhus: Klim, 17–36.
Name Index
A Adkins, A.W.H., 39 Aelred of Rievaux, 228–229, 240 al–Dı¯n Manı¯rı¯, Shara¯f, 215 Alexander of Aphrodisias, 158 Al-Ghaza¯lı¯, Abu¯ H·amid M · uhammad, 205–224 Al-Makkı¯, Abu¯ T·a¯lib, 217 Al-Ra¯zı¯, Abu¯ Bakr, 217 Annas, Julia, 59n5, 146n9 Antipater, 33 Aquinas, Thomas, 19, 225–236, 240–241 Aristippus, 21 Aristotle, 4–5, 14, 18–22, 25, 28, 43–47, 66, 125–137, 142, 153, 158, 182, 201, 211, 215, 225, 229, 232 Arnauld, Antoine, 239 Augustine, 1, 3, 14, 19, 26–29, 155–176, 180, 194–203, 227, 237–239 Averroës, 19, 214 Avicenna, 206, 209, 211, 216–218, 220, 223–224, 229–230, 236
B Bazán, B. Carlos, 226–227 Beversluis, John, 121n34 Bobzien, Suzanna, 65 Bolton, Robert, 167n39 Bonhöffer, Adolf, 49 Bratman, Michael, 59n4 Brickhouse, Thomas, 99n52 Burnyeat, Myles, 139, 148, 151, 175n59
C Cary, Phillip, 155–169, 195n3 Chadwick, Henry, 25 Chisholm, Roderick, 215
Chrysippus, 20, 23, 42, 65 Cicero, 24–25, 29–33, 35–47, 50–55, 67n25, 165, 196, 199–201 Cratylus, 21
D Damascius, 172 Descartes, 5, 14, 26–29, 49, 78, 93n37, 97n46, 225, 239–240
E Engberg-Pedersen, Troels, 40 Epicharmus, 20 Epictetus, 15, 24–33, 35, 46–55, 58, 65–70, 163n26, 165–166 Euripides, 112
F Field, Claud, 213 Foucault, Michel, 48–49, 180n3
G Gill, Christopher, 14–15, 16, 30–33, 91n35 Griffel, Frank, 208n13
H Hadot, Pierre, 49, 180n3 Hegel, G.W.F., 129n7 Heraclitus, 21, 30 Herodotus, 112 Hierocles, 23–24, 28, 40 Homer, 17, 27, 153 Hume, David, 16, 60n7, 105, 139 Husserl, Edmund, 171n48
255
256 I Ibn ‘Adı¯, Yah·ya¯, 214 Ibn al-‘Arabı¯, 215 Ibn Muh·ammad Miskawayh, Ah·mad, 217 J Jacobs, Jonathan, 222n60
K Kahn, Charles, 47–48, Kant, Immanuel, 25, 37 Kenney, Peter John, 164n29 Ketchum, Robert, 96n45 Knuuttila, Simo, 70n30 Korsgaard, Christine, 60–61, 69, 71
L Lichtenberg, G.C., 26 Locke, John, 27, 157 Long, A.A., 47–48, 50 Lucretius, 19, 28 Luria, A.R., 23
M Marcus Aurelius, 21, 166, 170 Matthew of Aquasparta, 237 Matthews, Gareth B., 156n5 McCabe, Mary Margaret, 57, 106n67 McPherran, Mark, 80n8, 89n29 Mill, John Stuart, 152 Miller, Ian William, 222 Moosa, Ebrahim, 208 Mulla¯ S·adra¯, 212n29, 215 N Nafisi, Azar, 72n36 Nehamas, Alexander, 80n8, 84n19, 104n63 Nussbaum, Martha, 66n23, 128, 197
O O’Donovan, Oliver, 198 O’Neill, Onora, 71–72
P Panaetius, 22, 24–25, 30, 32, 37, 42–47 Parfit, Derek, 14 Parmenides, 88n25
Name Index Pascal, Blaise, 223n66 Paul, 179–194 Perin, Casey, 144n5, 146n10 Peter John Olivi, 226, 230, 236–241 Peter, 181 Philo, 20 Philoponus, 19 Plato, 4–5, 15–21, 24–32, 35, 44, 47, 58–64, 68–73, 77–83, 86–107, 109–116, 119, 125, 135–136, 141–142, 153, 155, 157–159, 180, 209, 221, 235 Plotinus, 3–5, 14, 18–19, 26–30, 155–176, 195–196, 201, 210 Plutarch, 21–22–23, 28 Porphyry, 26, 29–30, 164, 195 Proclus, 29, 172 pseudo–Philoponus, 19 Pyrrho, 143, 147, 151–153
R Rappe, Sara, 88n25 Richardson, Henry, 71 Ricoeur, Paul, 179
S Sartre, Jean Paul, 175 Schechtman, Marya, 22 Seigel, Jerrold, 2 Seneca, 20–24, 28–29, 50, 67, 70, 199, 237 Sextus Empiricus, 139, 143–153 Simplicius, 20n17, 196 Smith, Nicholas, 99n52 Snell, Bruno, 39 Socrates, 1, 17–18, 21, 27–29, 32, 62–63, 77–107, 109–123,131–137, 141–142, 172n52, 180, 235 Sorabji, Richard, 35–46 Stern-Gillett, Suzanne, 164n29 Stock, Brian, 174n57 Stokes, Michael, 80n7 Strawson, Peter, 14 Swift, Jonathan, 225
T Taylor, Charles, 4n6, 14, 30, 179–180, 182, 193 Tertullian, 24 Themistius, 19 Thesleff, Holger, 155n1
Name Index Timon of Philius, 153 Trasymachus, 77n2
257 Williams, Bernard, 20, 39, 105n65, 180 Winch, Peter, 25 Woodruff, Paul, 88n25
V Virgil, 196 Vlastos, Gregory, 82n13, 88n27, 91n35
X Xenophanes, 153
W William of Auvergne, 227
Z Zahavi, Dan, 179
Subject Index
A Action, 7, 25, 31, 41–43, 47–48, 57–58, 60–69, 126–127, 170–171, 220–222, 230–234, 240 Agency, 7, 52 criteria of, 58–74 ideal, 57–63 Anthro¯pos, see Human being Appropriation, 50–54 Aristotelianism, 128, 134–135, 215–217, 226–229 Ataraxia, 147 Augustine, 1, 3, 8, 9, Authority, epistemic, 77–81, 92–107 moral, 38, 170
B Belief, 7–8, 53–54, 79n6, 93–106, 126–127, 143–148
C Care of the self, 48–49 Change, 116–123, 136–137, 141–142, 169–176, 182–183, 196, 202 Character, 22, 31–44, 52–54, 66n24, 140–145, 168, 175, 190–194, 218–222 ideal, 44–45 Choice, 41–42, 104, 118–123, 128–129, 239–241, see also Prohairesis Conscientia, 173, 197–198, 230–231, see also Consciousness Consciousness, 13–16, 47–48, 160–165, 173, 197–198, 234 states of, 139–140, 149 Core commitments, 148–153
D Death, 18–22, 41, 85, 117–118, 122, 136n21, 184–192, 199–202
E Elenchus, 77–92 Emotions, 140–142, 161, 201–203
F First-person perspective, 37, 79, 92–107, see also I-perspective Friendship, 18–19, 117, 127–128, 199–203, 219
H Human being, 4–5, 17–18, 43–44, 53–55, 57–58, 69–71, 125–130, 133–137, 151–154, 158–159, 168, 210–215, 220–221, 226–229, 235–241 Hylomorphism, 125, 130–137
I Ideal self, 3, 126–130, 152–153, 160–164, 170–175, see also Character, ideal; Agency, ideal Identity, 22–23, 27, 47–49, 104, 118–124, 136–137, 139–142, 172, 174–175, 196–199, 226 loss of, 18 narrative, 179 Ignorance, 80–86, 103n59 Individual, 1, 94, 196–197, 214–219, 226–229, 236–240 Individuality, 13–19, 26–33, 36–49, 125–137, 140, 160–161, 167–168
259
260 Infinity, 159, 173–176 Inner self, 155, 158–159, 198 Interaction, 33, 90–92 between the parts of the soul, 64, 185 I-perspective, 186, 190–191, see also First-person perspective
J Judgement, moral, 66n24, 189, 230–231 suspension of, 105, 143–150, 220n54
K Kantian self, 39, 179, 182, 193 Knowledge, 48–49, 54–55, 95–98, 102–104, 120, 141, 158–165, 175n59, 196–197, 209–212, 217,218 disavowal of, 77–90 moral, 82
L Loss of self, 197–198, 214
M Many-dimensional approach (to self), 2 ‘Me’ and ‘I’, 17–28, 37–38 Memory, 9, 18–19, 23, 155–163, 166–175, 195–203, 232–233 Mutability, see Change
N Neoplatonism, 29–30, 155–175 Normativity, 45, 57–64, 68–74
O Oracle, 29, 80–81, 87n23, 89–90 Other-regard, 190–191 Oxford-self vs. Webster-self, 139–145, 148–154
P Perfection, 54–55, 128–130, 162–163, 199, 222n60 Persistence, 20, 62, 116–123, 189 Personae, see also Personality
Subject Index Personality, 141 objective-participant and subjectiveindividualist conceptions of, 39–46, 52–56 Phantasia, 46–47, 65, 161–166, 237 Platonism, 3, 17, 155–157, 160–163, 214–215, 225–228 Practical ethics, 35, 45–55, 128–130 Prohairesis, 18, 24, 46–55, 65–69, 163n26 see also Choice
R Rationality, 2–3, 17–18, 24, 30–33, 43, 53–55, 68–74, 116, 128, 141–146, 163–165, 168, 211n22, 214–217, 239–240, see also Second-order reflection, Wisdom Reflexivity, 33, 47–48, 236–241 second-order, 182–185
S Scepticism, 139–153 Self, see Care of the self, Ideal self, Loss of self, Oxford self vs. Webster self, Social self, Subjective-objective self, Terminology of self, True self, Unity of selfhood Self-awareness, 14–16, 26–32, 82–86, 199–200 Self-consciousness, 38, 40, 47–48, 127n2, 155n4, 230–241 Self-examination, 78–79, 89–92, 102–106, 198–199, 206–210 Self-knowledge, 28–30, 103n59, 126–128, 164, 172–173, 197–198, 206–210, 237 Social relations, 14, 38, 43–45, 52–54 Stability, 44n35, 140–148, 199–201, see also Persistence Stoicism, 20–24, 29–33, 35–56, 65–74, 165–166, 238–239 Subjectivity, 3, 14–16, 33, 38–40, 48–49, 175n59, 195–203, see also Personality, objectiveparticipant and subjective-individualist conceptions
T Teleology, 3–4, see also Perfection Temporality, 161, 169–171, 197–198 Terminology of self, 1–5, 17–18,
Subject Index Third-person perspective, 78–79, 91–92, 105–107 True self, 15–18, 27, 47, 125–130, 168
U Unity of self, 141–142, 167, 221–223, 229–236
261 V Virtue, 30–32, 42–45, 51–54, 68–69, 81n11, 92–94, 111–120, 127–130, 222n60, 234–235
W Wisdom, 52–55, 119–123, 219n52, see also Rationality lack of, 78–86
Index Locorum
AETIUS 4.12.1–5: 165n31
ALEXANDER OF APHRODISIAS De fato 199.14–22: 54n90
in An. Pr. 180,33–36: 20n15 181,25–32: 20n15
ARISTOCLES IN EUSEBIUS PE 14.18.26: 152n18
ARISTOTLE De an. I 4, 408b5–11: 18n7 I 4, 408b11–15: 3 and 133 II 1: 132 II 1, 412a19–21: 131 II 1, 412a27: 135n20 II 1, 412a27–28: 135 II 1, 412b4–5: 135 II 1, 412b5: 135n20 II 1, 412b10–24: 134 II 4, 415b4–7: 136n21 II 4, 415b9–28: 132 III 4: 158n9 III 7, 431a14–17: 165n33
De mem. 1, 451a9–12: 126n2
EE VII 12, 1244b24–1245b1: 128n3 VII 12, 1245a30: 18n8
EN I 3–4: 43n33 I 7, 1098a16–18: 142n3 I 10, 1101a8–13: 142n3 I 13, 1102a26–32: 211n22 I 13, 1102b13–1103a10: 44n35 II 1: 43n33 III 1, 1111a3–7: 126 VI 13, 1144b1–9: 129n8 VII 7, 1150b19–28: 126 VII 8: 44n35 VIII 14, 1161b28–9: 18n8 IX 4, 1166a16–17: 211n22 IX 4, 1166a31–32: 201n20 IX 4, 1166a32: 5 and 18n8 IX 4, 1166b5–26: 126 IX 4, 1168b34–1169a3: 211n22 IX 4, 1169b7: 5 and 18n8 IX 4, 1169b33–4: 77 IX 4, 1170b6: 18n8 IX 9, 1169b3–1170a4: 128n3 IX 9, 1170a29–b14: 128n3 X 7, 1178a2: 142 X 7, 1178a2–3: 211n22 X 9: 43n33
GC I 5, 321a18–22: 20 I 5, 321b26–8: 20 I 5, 322a28–33: 20 II 11, 338b11–18: 136 II 11, 338b11–19: 136n21
263
264 Met. I 1, 981a13–20: 218n49 IV 5, 1010a10–15: 21n21 VII 8, 1034a5–8: 131 and 137 VII 10, 1035b25: 134; VII 11, 1036a31–b28: 133 VII 11, 1036b28–33: 134 VII 11, 1037a5–6: 131n14 VII 11, 1037a7–9: 132 VIII 3, 1043a31–33: 133n16 VIII 3, 1043a34–37: 132 and 135 XII 7: 158n9 XII 7, 1072b20–25: 158 XII 9: 158n9 XII 9, 1074b34–35: 158
MM II 15, 1213a13: 18n8 II 15, 1213a13–26: 28 and 128
Phys. II 9, 200a11–15: 133n16 III 3: 158n9
Problemata XXX 10: 126
Rhet. I 13, 1373b14–17: 25 II 8: 66
AUGUSTINE C. acad. 3.11.24–5: 175
Conf. 1.7.12: 196 1.9.14: 201n21 2.9.17: 200 2.10.18: 201n22 4.2.2: 201n21 4.4.7: 200 4.4.9: 201 4.5.10: 201 4.6.11: 201 and n23 4.7.12: 201 4.8.13: 201 and 202n27 4.9.14: 201 7.9.13: 157
Index Locorum 7.10.16: 29, 158 and 201n22 7.17.23: 164 8.8.19: 200n17 8.11.26: 200n17 8.11.27: 200n17 8.12.29: 200n17 8.12.30: 200n17 9.3: 19n10 9.10: 19n10 9.12.29: 202 9.12.30: 202 9.12.31: 202 and n28 9.12.32: 197 and 203 9.12.33: 202 9.13: 19n11 9.13.34: 202 9.13.35–37: 202 9.13.36: 203 9.13.37: 202 9.32: 196 10.2: 199 10.2.2: 173 and 197 10.3.4: 198 10.6.9: 158 and 198 10.8–16: 166 10.8.12: 172–173 10.8.12–15: 159 10.8.13–15: 159 10.8.14: 171 10.8.15: 158 and 197n11 10.9.15: 173 10.11.4–5: 162 10.13.20: 170 10.15: 173 10.15.23: 173–174 10.16.25: 173, 197 and n11 10.17.26: 157, 172–173 and 196 10.25.36: 167 and 171 10.26.37: 167 10.28.39–43.70: 198 10.29.40: 175 10.30: 167 10.30.41: 198 10.32.48: 197n11 10.35.54–57: 198 10.37.61: 197n11 and 198 10.40.65: 173 and 198 13.31.46: 197n11 and 198
De civ. D. 1.12–13: 202n29 21.9–13: 198n14 21.17–24: 198n14 22.20–21: 198
Index Locorum 22.29: 198 22.30: 198
De mag. 10.32: 164 11.38: 164
De trin. 8.2: 164 10.3.5: 28n52 10.7.10: 28n52 10.8.11: 28n52 10.9.12: 28n52 10.10.14: 26 10.10.16: 26 and 28n52 10.16: 169 14.4–5: 198
Ep. 10: 200n18 18.2: 169 60.2: 202n29 60.7: 202n29 166.3.6: 169 258: 200n18
Med. 9.6: 170 9.21: 170
Nat. et or. an. 4.9–10: 196
Retr. 2.6.2: 201
Sol. 1.1: 163 1.7: 200n18 1.13: 163 1.20: 200n18 1.22: 200n18 1.30: 163
Vera rel. 46.88–89: 19n11
265 CICERO Acad. 2.30–31: 53n83
De am. 18–21: 200n18 26: 201n25 49: 201n25 62–65: 200n18 92: 201n25 97: 200n18
Fin. 1.30: 40n19 3.16: 40 and n19 3.17: 50n67 3.20: 51n72 3.20–21: 44n39, 51n70 and 54n88 3.20–22: 50n67 and 53n82 3.22: 33n6 3.33–34: 53n83 3.62–63: 51n73 and 53n82 3.62–68: 50n67
Leg. 1.22.58: 29
Off. 1.46: 32 1.93–95: 42n28 1.95–96: 44n39 1.98: 42n27, n28, n30 and n39 1.100: 44n39 1.100–102: 42n30 1.102: 44n39 1.107: 42n28 and n30 1.107–115: 24n41 1.107–121: 42n26 1.109: 25 1.109–110: 42n30 and n31 1.110–111: 42n27 and n28 1.112: 31, 35–37, 41, 43 and 47 1.112–113: 25 and 44 1.113: 41 1.114: 42n27 1.115–121: 42n29. 1.119–120: 42n27 1.125: 42n27 3.100–110: 41n24 3.110: 41n24 3.115: 41n24
266 3.20–22: 43n32 3.51–57: 42n25 3.7–16: 41n24 3.99: 41n24
Rep. 3.22.33: 29
Tusc. 1.46: 157 1.66: 157 1.75: 157 3.11: 67n25 4.29, 34–35: 54n90 5.40–41: 51n71 5.81–82: 51n71
DAMASCIUS In Parm. 254: 172 and n53
DIOGENES LAERTIUS 3.12: 20n18 7.46–48: 52n75 7.49–51: 165n31 7.51: 165n35 7.53: 53n83 7.89: 54n88 and n90 7.91: 43n32 and 54n89 7.116: 54n88 7.127: 54n90 9.65: 153n21 9.66: 152 and n18 9.67: 152n20 9.71: 152n20
EPICTETUS Diss. 1.1.7: 65, 68n28 and 166 1.1.7–12: 52n76 1.1.23: 24n40, 46, 52n78 and 65 1.1.23–24: 163n26 1.1.24–32: 66n22 1.2: 66n22 1.2.1–11: 51n74 1.2.20: 25n44 1.2.29: 66n22 1.4.1–3: 52n76
Index Locorum 1.4.1–4: 69 1.4.11: 50n66 and 53n81 1.6.30: 67 1.6.6–8: 67n25 1.12.20–21: 54n90 1.18.17: 24n40 1.19.8: 24n40 1.20.17–18: 52n78 1.22.1: 70 1.22.1–3: 53n83 1.22.9–10: 53n83 1.28.10: 66 2.1.4: 166 2.1.24: 69 2.17.14–18: 50n66 2.17.19–20: 67 2.17.21: 66 2.17.22: 67 2.17.34: 67n25. 2.22.19: 52n78 2.22.29: 166 2.23.3: 52n79 3.1.40: 24n40 3.2.1: 51n70, n72, 52n76 and 53n81 3.2.1–5: 50 3.2.3: 51n71 3.2.4: 51n72, n73 and 53n81 3.2.5: 52n75 3.3.2–4: 53n83 3.3.5–10: 51n74 3.3.14–19: 33n61 3.3.22: 52n79 3.12.13: 53n81 3.12.13–15: 50n66 3.18.3: 24n40 3.22.7–8: 31 3.22.38–39: 29 and 165 3.23.4–5: 25n44 3.24.84–88: 51n74 4.5.12: 24n40 4.5.23: 24n40 4.6.34: 166 4.6.35: 29n54 4.10.13: 50n66 and 53n81
Ench. 1: 52n76 2: 52n76 4: 70 5: 52n76 and 54n90 24.2–4: 70
Index Locorum EPICURUS Sent. 2: 19n13
EURIPIDES Helen 46–47: 113n13.
GALEN De plac. Hipp. et Plat. 5.2.3–7: 54n90
HERACLITUS Fragments (in Diels–Kranz) 45: 17n6 54: 30 101: 17n6 and 30 116: 17n6
HERODOTUS 2.112 ff: 113n13 5.119.1: 112n11 7.26.3: 112n11
HIEROCLES 1.5–33: 40n20 4.38–53: 40n20
267 3.830–1094: 19n13 3.843–864: 19
MARCUS AURELIUS Med. 2.2: 22n29 4.3: 166 7.59: 166n37
ORIGEN C. Cels. 5.20: 20n16 and n17
Fragment on Psalms 4,5: 29n54
PANAETIUS 56–57: 44n37 107–125: 45 115–121: 45
PHILO De aeternitate mundi 48: 20n19
PHILODEMUS Sto. XII, 4–6: 32n59
ap. Stob. Florilegium vol. 4, 671.7–16 (Wachsmuth and Hense): 23n36
HOMER Odyssey 4.384: 110 4.415–424: 110 4.417–18: 110n6 and 117 4.421: 115n21 11.601–603: 17
IAMBLICHUS ap. Proclum, in Alc. 11,11–17: 29
LUCRETIUS 3.31–93: 19n13
PHILOPONUS in de Intellectu 52,23–29: 19n12 91,40–49: 19n12
PLATO Alcibiades I 129–130: 3 130d: 17 132c–133c: 28 132d–133c: 155n1 133c4–6: 17
Apology 18b7–8: 81n10 20d8: 82
268 20d9–e1: 82 21a6–7: 80 21b–22e: 81 21b4–5: 80 and n7 21b5–6: 80 21b6–7: 81 21b9–c1: 80 21c1–2: 80 21c6–7: 89 21c8: 82 21c8–d1: 85 21d1–2: 83 21d2–6: 83 and 85 21d2–8: 81 and 82n13 21e–22a: 89n28 21e5–22a1: 82 22a3–6: 81 22a7–8: 80 22b3–c3: 85 22c4–6: 85 22c9–d1: 80 22d–e: 82 22d5–e1: 81 22d8: 82n12 22e1–6: 82n12 23a3–5: 81n10 23a5–6: 78 and 82 23b: 89n28 23b1–4: 82 23b2–4: 81 23b6–7: 82 23c–d: 87 23c4–6: 86n21 23c5–d9: 86 23d: 83n17 26c8: 87 26e–28a: 87 27b2: 87 and 97n48 27d: 100n53 28e6: 86 29a5–b6: 85 29b8–9: 85 29d–e: 84 29e–30a: 89n28 29e1: 81n11 30a–b: 84n19, 86n21 and 91 30a6–7: 84n19 30c10–d2: 81n9 30e–31a: 83 31a–b: 84n19 31c–d: 87 31d–33a: 87n23 32d2–4: 104n62
Index Locorum 33c2–4: 86n21 35a–b: 85 35d: 87 36c: 86n21 and 91 36c3–7: 84n18 36c7: 81n11 36d5: 84n18 36d9–10: 84 38a5: 86 38a5–6: 84 39c–d: 86n21 40a: 87 40a4–7: 87n23
Charmides 154d: 95 154e: 95 155c8: 94 155d3: 95n43 156a–157c: 94 157a5–6: 94 158e7–159a8: 93 and n39 159b–160d: 94 159b5: 94 160d: 155n1 160d–e: 94 160d6: 94 160e–161b: 94 162b11: 94 163e6–7: 90n33 164d–165a: 95 164d3: 95 165b: 88n25 166c: 88n25 166c7–d2: 89n29 and 90 166d: 91n35 166d–e: 103 166e5–9: 95 167a–169c: 28 169c: 88n26 169e–171c: 95 170b: 103n59 170b–d: 95 170d: 103n59 170e–171: 96 171b: 96 171c: 96 171c4–9: 96 171d: 103n59 172b: 96n45 and 103n59 172b4–5: 96 175b–c: 96 and 103n59
Index Locorum Cratylus 428d: 90n32
Crito 46b–c: 87n24 48b–49e: 87n24
Euthydemus 272b: 116n23 272b–d: 116n23 272c: 123n37 275a: 122 275a9: 116n23 275e: 117n25 281eff: 123n39 283b–e: 111 283c–d: 117 283e–284e: 111 284a: 120 285c: 111n10 and 121n33 285c4: 111n9 285c7–d1: 112 285d–288b: 111 286a: 118n26 286aff: 121. 286c: 120n32 287a–b: 112 287eff: 123n38 288a: 111 288b–c: 109 290e–291a: 88n27 and 114 293a: 119 293b–296e: 118 294a: 119 295a: 119 297b–e: 111n9 298b: 116n23 298dff: 118n26 303a: 111n9
Euthyphro 11d: 88n26 15d: 113n14
Gorgias 453b–c: 91n35 458a–b: 103 474b2–10: 98 474c–475e: 99
269 475e3–5: 99) 471e–472d: 99n51 482b–c: 99n52 486d–487e: 87n22 506c–507c: 90n34 507c8–9: 90n34
Hippias Major 286c5: 91 293d–294e: 91 297e–300b: 91 298b11: 91 304a–b: 91n36 304d–e: 92
Hippias Minor 372d7–e1: 88n26 376b: 88n26 376c2–3: 88n26 376c4–6: 88n26
Ion 541e: 113n14
Laches 187e–188a: 89n28 187e6–7: 89n28 194c–d: 88n25 194e11–195a1: 88n25 194d–195a: 88n25
Laws 959b3: 16
Meno 80b: 109n2 80c: 88n26
Phaedo 63b–c: 17 91c4–5: 90n32 95e8–9: 90n31 96c4–5: 89 96c6–7: 89 96e7–8: 89n30 115c: 17–18
270 Phaedrus 229e–230a: 209 229e5–6: 106n68 230a3–6: 106n68 235c7–8: 106n68 243e–257b: 106n68
Philebus 8e–40c: 155n1 20a: 113n14 36e11: 97n47 38b–d: 106
Politicus 272bff: 112n12 278b–c: 105n5 283b: 109n5 291a–b: 109n4.
Protagoras 311bff: 114n17 352a–b: 95 359b–360e: 88n25 361a–c: 87n22
Republic I, 331c: 63 I, 337a–e: 77n1 I, 351c: 64 II, 381d: 113n14 IV, 430e: 60n8 IV, 439d: 59n5 IV, 440a: 59n6 IV, 441a: 59 IV, 442e–443b: 63 IV, 443c10–d1: 62n13 IV, 443e: 62 IV, 443e2: 61 and 62 VI, 490d–498b: 43n33 VIII, 551d: 61 VIII, 553d1–3: 59n5 and 62 VIII, 554d: 60 VIII, 557a: 61n10 VIII, 561b: 61 VIII, 580d–581a: 59n5 IX, 573a–c: 63 IX, 580c: 62 IX, 581b–e: 60 IX, 588c: 210n21 IX, 589a: 155
Index Locorum IX, 589a–b: 5 IX, 589a6–b6: 17
Sophist 230–231: 115 230a–231b: 106n67 230c: 106n67 263e: 155n1. 263e–264b: 106
Statesman 273d6–e1: 201n22.
Symposium 174c: 110 175a–d: 89 201e–202d: 100n53 206e–209e: 26 207c–208b: 21n23 215aff: 114n19 215b: 172n52 216d–e: 95n43 216e–217a: 172n52 216e6–7: 95n43 220c–d: 89
Theaetetus 148e: 83n16 151e–152e: 141 152e–186e: 21n22 154e–155a: 106n66 156a–160a: 142 162b: 95n43 169a–b: 95n43 169b: 88n25 179e–180c: 142 183e–184a: 88n25 189b–190e: 100 189e: 155n1 189e–190a: 105 189e7: 106n66 190c6–7: 100 190d10–11: 100 191c–196c: 159 197b–200d: 159 209c: 100 and 101n54
Timaeus 41d–42a: 168n41
Index Locorum
271
86d: 63n16 87a7–b4: 63n16
VI.9 [9] 7.16–23: 29n55 VI.9 [31] 7.18: 158
PLOTINUS Enn. I.1 [13]: 5 I.1 [53] 9: 168 I.1 [53] 10.6–7: 210n21 I.2 [19] 7.13–28: 30n57 I.5 [36] 7.1–30: 19 I.6 [1] 8.4: 158 I.6 [1] 9: 161 I.6 [1] 9.8: 29n55 I.6 [1] 9.8–16: 172 I.6 [1] 9.16–25: 174 I.8 [51] 13.16–17: 201n22 IV [28] 4.8: 165 IV.3 [27] 2.49–58: 18n9 IV.3 [27] 5.1–15: 167 IV.3 [27] 5.6–9: 161 IV.3 [27] 18.22: 162 IV.3 [27] 18.22–24: 161 IV.3 [27] 27.1–25: 17n5 IV.3 [27] 29.25: 165 IV.3 [27] 30–31: 165 and 170n47 IV.4 [28] 5.11–22: 168 IV.4 [28] 5.11–31: 18 IV.4 [28] 15.15–18: 169 IV.4 [28] 15.16.1–7: 169 IV.6 [41] 3: 163 and 170n47 IV.8 [6] 1.1–11: 29n55 IV.9 [8] 5.7–26: 18n9 V.1 [10] 1: 168 V.1 [10] 1.1–17: 18 and 26n48 V.3 [49] 2: 161 V.3 [49] 3.31–32: 163 V.3 [49] 3.4.1–4: 163 V.3 [49] 3.32 ff.: 168 V.3 [49] 3.44–45: 163 V.3 [49] 8.20–30: 163 V.3 [49] 9.1–10: 163 and 164 V.5 [32] 1.38–41: 161 V.5 [32] 10.18–22: 174 V.6 [24] 4.18: 163 V.8 [9] 11.12–13: 158 V.8 [31] 10.31–43: 29n55 V.8 [31] 11.17: 174n57 VI.1 [42] 4.7–19: 161 VI.2 [43] 20.4–23: 18n9 VI.4 [14] 16: 5 VI.5 [23] 7: 161 and 168 VI.7 [38] 7.9–17: 168
PLUTARCH Mor. De E apud Delphos 392c–e: 21n26 Mor. De virt. mor. 440e–441d: 54n88 and n90 443c–d: 43n33 445e: 44n35 446c–d: 44n35 Mor. De tranq. anim. 473b–474b: 21n28 and 22n32 Mor. De facie quae in orbe lunae apparet 944f: 17n5
Mor. Sto. rep. 1038b: 23n37 1041e: 53n83
Mor. Comm. not. 1063a-b: 54n90 1071a–b: 33n60 1083a–c: 20n18
PORPHYRY Isago¯ge¯ 7, 6–24: 26n47
Sent. 32, 30,1–31,11: 30n57
PROCLUS El. theol. 211: 172
in Alc. 5,13–14: 29
in R. 1.119,23–120,15: 17n5
272 PSEUDO–PHILOPONUS Comment. in Arist. De Anima 3, 538,32–539,12: 19n12
SENECA Ben. 2.35,2: 67n25 7.16,5–20,5: 67n25
De brev. vit. 10, 3–6: 22n31
De ira 3.36: 29n54
Ep. 24,19–21: 21n26 36,10–11: 20n14 58,22–23: 21n26 89,14: 50 121,16: 21n27 121,16: 23n35 120,22: 43n34
Index Locorum 2.102: 146 and n9 3.236: 147n12
Math. 1.305: 153n21 7.432–434: 32n59 11.1: 153n21 11.158–60: 147n12
SIMPLICIUS in Phys. 886,12–16: 20n17
STOBAEUS 2.62.15–63.5: 44n39 2.65.8: 43n32, 44n38 and 54n89 2.155.5–17: 51n71, 54n88 and n90
TATIAN Ad Gr. 5: 20n15
TERTULLIAN Adv. Marc. 4.16: 24n39
Medea 177: 67
SEXTUS EMPIRICUS PH 1.13: 143 1.8: 149 1.21–30: 143 1.25: 150 1.30: 146n9 and 147
THEMISTIUS in Arist. de An. 103,32–104,6: 19 104,14–23: 19.
XENOPHON An. 1.2.8: 112n11