Foucault and Classical Antiquity: Power, Ethics and Knowledge (Modern European Philosophy) 9780521833813, 0521833817

Citation preview

This page intentionally left blank

FOUCAULT AND CLASSICAL ANTIQUITY This book is a critical examination of Michel Foucault’s relation to ancient Greek thought, in particular his famous analysis of Greek history of sexuality. Wolfgang Detel offers a new understanding of Foucault’s theories of power and knowledge based on modern analytical theories of science and concepts of power. He offers a fresh and complex reading of the texts which Foucault discusses, covering topics such as Aristotle’s ethics and theory of sex, Hippocratic dietetics, the earliest treatises on economics and Plato’s theory of love. The result is a philosophically rich and probing critique of Foucault’s later writings, and a persuasive account of the relation between ethics, power and knowledge in classical antiquity. His book will have a wide appeal to readers interested in Foucault and in Greek thought and culture. Wolfgang Detel is Professor of Philosophy at Johann Wolfgang GoetheUniversit¨at Frankfurt.

MODERN EUROPEAN PHILOSOPHY

General Editor R O B E R T B . P I P P I N , University of Chicago Advisory Board G A R Y G U T T I N G , University of Notre Dame R O L F - P E T E R H O R S T M A N N , Humboldt University, Berlin M A R K S A C K S , University of Essex

Some Recent Titles: Daniel W. Conway: Nietzsche’s Dangerous Game John P. McCormick: Carl Schmitt’s Critique of Liberalism Frederick A. Olafson: Heidegger and the Ground of Ethics Gu¨ nter Z¨oller: Fichte’s Transcendental Philosophy Warren Breckman: Marx, the Young Hegelians, and the Origins of Radical Social Theory William Blattner: Heidegger’s Temporal Idealism Charles Griswold: Adam Smith and the Virtues of the Enlightenment Gary Gutting: Pragmatic Liberalism and the Critique of Modernity Allen Wood: Kant’s Ethical Thought Karl Ameriks: Kant and the Fate of Autonomy Alfredo Ferrarin: Hegel and Aristotle Cristina Lafont: Heidegger, Language and World-Disclosure Nicholas Wolsterstorff: Thomas Reid and the Story of Epistemology Daniel Dahlstrom: Heidegger’s Concept of Truth Michelle Grier: Kant’s Doctrine of Transcendental Illusion Henry Allison: Kant’s Theory of Taste Allen Speight: Hegel, Literature, and the Problem of Agency J. M. Bernstein: Adorno: Disenchantment and Ethics Will Dudley: Hegel, Nietzsche, and Philosophy: Thinking Freedom Taylor Carman: Heidegger’s Analytic: Interpretation, Discourse, and Authenticity in Being and Time Rudiger ¨ Bubner: The Innovations of Idealism

FOUCAULT AND CLASSICAL ANTIQUITY Power, Ethics and Knowledge

WOLFGANG DETEL Johann Wolfgang Goethe-Universit¨at Frankfurt

Translated by

DAVID WIGG-WOLF

   Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo Cambridge University Press The Edinburgh Building, Cambridge  , UK Published in the United States of America by Cambridge University Press, New York www.cambridge.org Information on this title: www.cambridge.org/9780521833813 © Suhrkamp Verlag Frankfurt am Main 1998 This book is in copyright. Subject to statutory exception and to the provision of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published in print format - -

---- eBook (NetLibrary) --- eBook (NetLibrary)

- -

---- hardback --- hardback

Cambridge University Press has no responsibility for the persistence or accuracy of s for external or third-party internet websites referred to in this book, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate. Originally published in German as Macht, Moral, Wissen: Foucault und die Klassische Antike. First published in English by Cambridge University Press 2005 as Foucault and Classical Antiquity: Power, Ethics and Knowledge English translation © Cambridge University Press 2005

CONTENTS

page viii

List of abbreviations Introduction 1 2 3 4 5 6

1

Morals, knowledge and power The ethical teleology The scientific regimen The asymmetrical relationship The epistemic eros Gender, nature and reference

6 58 93 118 163 223

References Index

264 276

vii

A B B R E V I AT I O N S

Of Foucault’s works

M. Foucault, The History of Sexuality HS UP CS

Vol. 1: The Will to Know Vol. 2: The Use of Pleasure Vol. 3: The Care of the Self For ancient authors and texts

DK Frg. Epigramm. Graec.

Diels and Kranz (1964) Fragments Epigrammata Graeca

Aesch. Ar. An. An. Post. An. Prior. Ath. Polit. EE GA

Aeschines Aristotle De Anima (On the Soul) Analytica Posteriora (Posterior Analytics) Analytica Prioria (Prior Analytics) De Athenorum Republica (Athenian Constitution) Ethica Eudemia (Eudemian Ethics) De Generatione Animalium (On the Generation of Animals) viii

l i s t o f a b b r e v i at i o n s GC HA Iuv. Met. MM NE PA Phys. Pol. Resp. Rhet. Sens. SE Top. Aristoph. Ekkl. Lysistr. Nub. Thesm. Cael. Aur. Acut. Cic. Iuvent. Corp. Hipp. Epid. Deinarch. Dem. Demosth. Diog. Laert. Eur. Electr. Troj. Gal. Progn. Hes. Hyper. Hipp. Prisc. Med. Epid. Aphor.

ix

De Generatione et Corruptione (On Generation and Corruption) Historia Animalium (History of Animals) De Iuventute et Senectute (On Youth and Old Age) Metaphysica (Metaphysics) Magna Moralia (Great Ethics) Ethica Nicomachea (Nicomachean Ethics) De Partibus Animalium (Parts of Animals) Physica (Physics) Politica (Politics) De Respiratione (On Respiration) Rhetorica (Rhetorics) De Sensu (On Sense) Sophistici Elenchi (On Sophistical Refutations) Topica (Topics) Aristophanes Ecclesiazusae Lysistrata Clouds Thesmaphoriazusae Caelius Aurelianus Acutae passiones Cicero De Iuventione Corpus Hippocraticum Epidemiae (Epidemics) Deinarchus Democritus Demosthenes Diogenes Laertius Euripides Electra Trojan Women Galenus Prognosticum de Decubitu ex Mathematica Scientia Hesiod Hypereides Hippocrates De Priscina Medicina (On Ancient Medicine) Of the Epidemics Aphorisms

x Hor. Sat. Ep. Iambl. Isocr. Plato Charm. Gorg. Lys. Nom. Phaed. Phaedr. Polit. Prot. Symp. Tim. Porph. Vit. Pyth. Pseud.-Ar. Mech. Oec. PP Soph. Trach. Xen. Mem. Symp. Oec.

l i s t o f a b b r e v i at i o n s Horace Satyrae (Satires) Epistulae (Letters) Iamblichus Isocrates Charmides Gorgias Lysis Nomoi (Laws) Phaedo Phaedrus Politeia (Republic) Protagoras Symposium Timaeus Porphyrius Vita Pythagorae Pseudo-Aristotle Mechanica [of dubious authenticity] Oeconomica Problemata Physica [of dubious authenticity] Sophocles Trachiniae Xenophon Memorabilia (Socratic Memoirs) Symposium Oeconomicus

INTRODUCTION

The main aim of this book is to clarify the moral concern about sexuality and sexual activities expressed by leading authors in classical antiquity. Michel Foucault was, of course, the first thinker to undertake this clarification, and one aim of his stimulating study was to demonstrate the way in which his new ethics programme worked. Therefore my starting point is the reconstruction of claims about the control of the desires and the creation of the moral subject in classical philosophical, medical, economic and rhetorical texts which Michel Foucault presents in the second volume of The History of Sexuality (The Use of Pleasure (UP)). There is now a mass of literature on Foucault’s programme of ethics, some of which also deals with his interpretation of ancient sources. But most authors treat his analyses of these texts as documents illustrating his intellectual development, which is their primary interest.1 None of them subjects his thoughts on ancient sources to a detailed critical assessment in order to suggest alternative readings that are based not only on the general theoretical premises of Foucault’s project, but also on modern standards of textual interpretation

1 See e.g. Bernauer (1988); Bernauer and Mahon (1994); Dreyfus and Rabinow (1983); Mahon (1992); Rajchman (1991); Sch¨afer (1995); Schmid (1990); Smart (1991) or the articles in the volumes of essays Erdmann et al. (1990); Ewald and Waldenfels (1991); Rabinow (1986); Seitler (1993), A. Davidson (2001).

1

2

introduction

and the present state of philosophical and historical knowledge of the ancient world.2 What follows is intended to reduce this deficit. Much of what I have to say is a critical reconstruction of Foucault’s own reconstruction – while at the same time I try to maintain the systematic aspects of his perspective. It is a bold attempt to say what Foucault’s historical perspective and better historical knowledge should have led him to say about the texts he studied, but which he did not. In my view, the main reason why Foucault did not say what, in his own best theoretical interest, he should have said is that, when he first introduced his programme of ethics at the beginning of UP, he held that the ethical creation of the subject could more or less be separated from the archaeological and genealogical dimension developed in his earlier works. Thus he almost completely neglects what the ancient texts have to say about epistemology and the analysis of power, or in the case of Plato attributes far too little significance to it. In chapter 1 I point out that Foucault’s restrictive approach to the ancient sources was an initial misunderstanding of his own general project that he later corrected. However, if the incorporation of the analysis of power and archaeology into Foucault’s ethics is to be successful, a suitable reappraisal of his ideas on the intrinsic connection between power and knowledge, and the concept of ‘productive’ power, is required. The rest of the first chapter is devoted to a lengthy and detailed explication of these ideas. One of the most important results is the introduction of a concept of regulative power that is not necessarily repressive, but is productive in a clearly definable way. It is primarily this kind of power that is intimately related to the games of truth that determine the historical forms of knowledge. This interpretation explains why Foucault’s relativistic neutrality is nothing more than systematic coquetry and why his critics, as well as leading post-modernists, are all mistaken in their vociferous complaints and cynical observations that power permeates even reason. After this introduction the next four chapters turn to the ancient texts that Foucault studied in UP. In chapter 2 I prove that in focusing on the 2 The two best exceptions are Geoffrey Lloyd’s review of UP in The New York Review of Books 33, ¨ (4 March 1986), 24–8 and Pierre Hadot’s study ‘Uberlegungen zum Begriff der Selbstkultur’, in Ewald and Waldenfels (1991), 219–28 (see also the afterword in Hadot (1991)). Lloyd concentrates mainly on a critical discussion of Foucault’s analysis of ancient medical texts, while Hadot criticises the fact that Foucault prunes ancient ethics to an ethics of desire, self-observation and stylisation of the self, and neglects the extent to which it is concerned with eudaimonia. Both critiques are on the right track, but do not pay enough attention to Foucault’s general theoretical perspectives; all told they are too brief, too one-sided and lack detail.

introduction

3

ancient analysis of sexual desire, Foucault neglects the comprehensive role played by eudaimonia. This was always an important concept in antiquity, incorporating sexual desire into a broader register of desires in order to optimise them. At the same time it becomes clear that Foucault’s one-sided view leads to an inappropriate exaggeration of the part played by aesthetic stylisation. Classical dietetics is then the central topic of chapter 3. Foucault regards it as a comprehensive art of existence and a thoroughly regulated practice of the self, but it is soon apparent that, at the most, this view can only be applied to an excessive, prophylactic form of dietetics that was in no way representative of the classical world. Above all, during the course of the fifth century philosophical ethics separated itself from medicine, with the result that medicine was reduced to being just one of several important elements of the good life while at the same time attaining a new theoretical status as an empirical natural science. Foucault would doubtless have had no difficulty in describing this historical process in archaeological terms, but he pays no attention to this archaeological aspect, just as he pays no attention to how dietetics, as it became increasingly empirical, lost all reference to eudaimonic practices of the self, which now became the domain of philosophy. This restricted view of ancient teleology and dietetics had the effect that Foucault arrived at an incorrect model of the regulation and practice of sexual desire: ancient authors did not recommend a model of domination and restriction intended to rein in the dangerous dynamic of sexuality, but entertained instead the model of self-controlled integration of sexual desire into the overall complex of higher and lower desires which had been optimised and kept in equilibrium as a constitutive part of the good life (if not as its goal). All told, specifically sexual desires played only a relatively unimportant role in the works of Aristotle and the Hippocratics that deal with human desires. Foucault’s interpretation of pederasty and marital relations in the ancient world has quite rightly attracted a great deal of attention, but it is flawed by a number of serious misjudgements. This I demonstrate in chapter 4. Historical studies have shown that in classical Greece pederasty as a primary love relationship was restricted to the small circle of political elites, and otherwise was regarded with a great deal of moral and legal reservation. The moral concern that manly activity might be endangered in pederastic relationships, which Foucault identified in the ancient texts, in fact only refers to a small group of speakers who played a leading part in the political business of the polis. Other parameters generally applied to relationships between full male citizens than to those between freeborn

4

introduction

and slaves. Ancient moralists could only think of pederastic relationships in terms of a relation of dominance, endangering the status of the beloved as a freeborn citizen, as well as threatening the most important boundary that the classical world knew – that between the freeborn and the slave. Reciprocal homoerotic relationships were a taboo for ancient moralists, and they could only conceptualise homosexuality in a particularly low form – as an exchange of desire and benefit that for the lover was generally materially expensive and permitted him no more that limited bodily contact and occasional sexual gratification between the thighs of the beloved, while the young beloved was allowed to accept expensive presents, but never to experience sexual arousal. Foucault remains silent on these aspects of homoerotics in the ancient world, just as he is not interested in the fact that reciprocal marital relationships would have posed a threat to the patriarchal dominance of the husband, and that for this reason ancient moralists were only able to think of marriage as an asymmetrical relationship between an older husband and a younger wife – structurally analogous to homoerotic relationships. However, the circumstances were more complicated in the case of marriage than pederasty, above all since marriage was the central institution of the most important social and economic unit in ancient city-states – the farm or estate (oikos), an institution that ancient politics was particularly concerned to ensure continued to flourish. Thus the moral concern for the husband’s sexual behaviour on the part of ancient moralists was not focused, as Foucault claims, on the conditions that enabled the husband to qualify for a political career, but on the threat to the oikos and the husband’s honour, which might lead him to a total loss of self-respect. In order to deal with the complex problems involved in this context, the Greeks invented a new science – the science of management of the oikos, economics. As a science economics was concerned essentially with the art of leadership in the dealings the head of the household had with his wife, manager, employees and slaves. But again, Foucault paid no attention to the dimensions of archaeology and the analysis of power that are involved here. The way in which Foucault looks at Plato’s theory of eros is particularly unsatisfactory, as I show in chapter 5. Although he makes some correct and interesting observations, for example about the exchange of roles between the lover and the beloved, Foucault misses the decisive points of Plato’s thinking about eros, and his neglect of the exact epistemological background and the corresponding games of truth has particularly negative consequences. Plato’s view of eros can only be understood against

introduction

5

the background of advanced interpretations of Platonic epistemology, and these reveal the deficiencies of Foucault’s ascetic reading of Plato. The final chapter is more of an appendix or excursus. It is an attempt to reconstruct the naive reference to different sexes that Foucault relies on in his historical analysis of the relationship between sexuality and truth, and seeks to defend this reference in the face of criticism, in particular from post-modern feminists who try to incorporate the concept of (biological) sex completely into the notion of (social) gender. In the face of this criticism, I defend a thin concept of sex that takes into account all ideological and anti-essentialist feminist reservations, and plead for a complete dissolution of the category of gender as an answer to the fraying of this category that is particularly apparent in post-modern feminism. Should the ideas and suggestions proposed in this book indeed be correct, and improve Foucault’s analysis of classical Greek texts, then of course this does not mean that critical interpretation has priority over creative theoretical work. Rather I would like to see the comments presented here as an example of a rational approach to post-modern thought, that takes its intuitions seriously, but without renouncing exact explication and critical examination according to the usual standards of philosophy and science. One of the most important results of the book is the realisation that Foucault’s theoretical approach in no way prevents this, but in fact makes it possible in the first place. For although regulative power may indeed permeate language and reason, this does not render rationality obsolete; on the contrary, it creates space for and stabilises it. Thus this book develops its own methodology for a critical assessment of important aspects of post-modern thought.

1

MORALS, KNOWLEDGE AND POWER

According to one influential interpretation, in his later years Foucault turned his attention to the ‘practice of an intellectual freedom’ which ‘goes beyond the modern relationships between will, power and subjectivity’.1 This reading claims that Foucault’s ethics aims to be – at least partly – free of the process of regulation and discipline which the analysis of power had revealed to be the all-pervasive element of the modern age, and maintains that it is concerned with ‘effective resistance to a wide-spread type of power’ and to modern ‘forms of dominance and exploitation’.2 It is in precisely this sense that Foucault’s ethics is supposed to be political, while at the same time it is the last, perhaps even the most important, step in his attempt to subvert psychoanalytically based interpretations of the self, and to demonstrate that they are a form of coercion and political power to which psychoanalysis subjects our desire and our unconscious. Finally, the ethics as a practice of liberation is intended to provide an aesthetics of existence ‘which resists the science of life’ and ‘frees us from the empire of scientific knowledge’.3

1 See Bernauer and Mahon (1994), 603. Their article is undoubtedly one of the best recent studies of the ‘ethics’ of Foucault’s later years. 2 Bernauer and Mahon (1994), 599. 3 Bernauer and Mahon (1994), 606 f.

6

morals, knowledge and power

7

The general direction of this interpretation is quite clear. Foucault’s opposition to universal anthropological, moral and rationalist categories has reaped criticism from influential authors such as Charles Taylor and Jurgen ¨ Habermas, who accuse him of abandoning all critical potential, with the result that he is unable to appreciate his historical analyses as an integral part of an emancipatory movement.4 Faced with his typical relativistic coquetry, some of the best Foucault scholars, and many of those who interviewed him, have posed disturbing questions about the central political, moral and socio-political message of his works.5 His ethics is regarded as an important element of the answer to these questions – an answer which is intended to combine in a consistent manner an emancipatory message with the denial of universal anthropological, moral and rationalist categories.6 At the same time, nearly all Foucault’s interpreters agree that he regarded ‘ethics’ as a research programme which went beyond the limits of the archaeological analysis of discourse and the historical analysis of power. At best it had a negative systematic relationship to the analytical tools of his earlier studies. For example, it is noted that the ‘ethics’ represents a realignment of his project to write a history of sexuality, which initially had prevented him from working on a comprehensive analysis of bio-power and drawn his attention to a subversion of the Christian– Freudian hermeneutic subject.7 It is in connection with the programme of ethics that we hear of a ‘subject-theoretical realignment’, which marks a ‘radical break’ with Foucault’s archaeological and genealogical studies.8 The general opinion is that the ‘three axes’ of Foucault’s historical studies – archaeology, genealogy and ethics – each specifies quite different 4 See the articles by Taylor (1986) and Habermas (1986a), esp. 93 and 108. Sch¨afer (1995) has recently presented an intelligent and detailed assessment of this and other criticisms. 5 See e.g. Dreyfus and Rabinow (1983), 205 ff. on the questions of ‘resistance’ and ‘truth’. See also the interview ‘On the genealogy of ethics’, in Rabinow (1986), 340–72 (esp. 344, 347, 350) on, 265–81 (esp. 268, 271, 273) the ‘alternatives’ which ethics as Foucault understood them can offer ‘today’, or the question of what ‘we today’ can learn from them. The enquiries by Martin (in Martin, Gutman and Hutton (1988), 9–15) and various US philosophers (e.g. Jay and L¨owenthal) (in the interview ‘Politics and ethics’, in Rabinow (1986), 373–80) have the same tenor. 6 See e.g. also Sch¨afer (1995), 141: ‘The concept of an “aesthetics of existence” is in fact more than just a consequence of Foucault’s general strategy to promote an anarchic form of subjectivity. For that is associated, above all, with a denial of any form of universal ethics.’ 7 Dreyfus and Rabinow (1983), 3 f. 8 Fink-Eitel (1989), 98–100; a similar view is presented by Visker (1991), 104–16. Schmid (1990) talks of Foucault’s ‘Apollonian turning-point’ (23).

8

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

systematic orientations, and proceed from a discussion of epistemic discourses and their truth-goals, via the analysis of sub-institutional power, to an enquiry into the constitution of the subject.9 Support for this view can indeed be found in a number of Foucault’s own statements. His justification for the analysis of ancient techniques of the self at the beginning of The Use of Pleasure (UP) is a particularly impressive and prominent example. Under the title ‘Modifications’, it assumes there are three axes of sexuality as a historical experience – the formation of the knowledge of sexuality; of the power systems which regulate its effect; and of the forms in which individuals recognise themselves as subjects of sexuality. At the same time Foucault emphasises that he introduced these three analytical levels into his work successively as ‘theoretical shifts’, and that above all the last of these shifts, towards work on a programme of ethics, was an attempt at introducing a ‘new approach’ to his analyses.10 Shortly before his death Foucault was still distinguishing between four major types of technology, each a ‘matrix of practical reason’, and between ‘three traditional problems: the relations to truth, the relationship to others, the relationships between truth, power and self’.11 In the process, the formation of subjectivity in particular would seem to be identified with the field of study of ethical investigations. However, there are good systematic reasons, as well as evidence in Foucault’s own works, that suggest that the prevailing picture of the systematic role of ethics within the framework of the Foucauldian programme is very much one-sided. The ‘modifications’ that introduce and justify the programme of ethics in UP indicate that the three axes, or aspects, of the analysis of sexuality as a historical experience ‘should also be investigated 9 Even authors like Arnold Davidson, who certainly accept that there is a tension between archaeology, genealogy and ethics, do not attempt a more exact reconstruction of the relationship of the three levels to each other. Above all they do not investigate how the archaeological analysis of discourse and the genealogical analysis of power keyed into the ethics (see on this Davidson (1986), who essentially deals with the three levels in isolation, and in particular discusses ethics separately from archaeology and the analysis of power (227–32). However, at least in the case of modern sexuality he notes: ‘An attempt to study modern sexuality would have to combine all three axes of analysis, the self, power, and knowledge, and would take many more volumes than the six Foucault originally planned’ (230)). In his report on Foucault’s last course at the Coll`ege de France in the summer of 1984, Flynn (1988) reports that in the context of the discussion of parrhesia (‘truth-telling’) as an ancient cultural practice, Foucault attempted to bring all three methodological layers into play – which Flynn rightly described as a theoretical desideratum (see above all Flynn (1988), 114 f.). 10 UP 4–5. 11 Martin, Gutman and Hutton (1988), 15 and 18.

morals, knowledge and power

9

in their context’. They can all be integrated into an overall project, a ‘history of truth’ in a specific sense – a history of ‘games of truth by which man proposes to think his own nature’.12 Accordingly, Foucault feels that a specific relationship to ‘games of truth’ is at the centre of each of the axes of his studies. The coy terminus ‘games of truth’ apparently refers to rules and methodological norms for the acceptance of claims. Therefore, in trying to write a ‘history of truth’ Foucault seems to opt for a definition of truth that today is often classed as ‘verificationist’. Thus archaeological, genealogical and ethical studies all analyse, each in its own way, historical ‘games of truth’ which are linked with the constitution of scientific selfexperience, with an intrinsic reference to power relationships, and with the constitution of moral self-experience.13 In Foucault’s later works the tendency towards a systematic integration of archaeology, the analysis of power and ethics into one all-embracing programme becomes stronger, and there is an increased emphasis on how they are interwoven. For example, in his interview ‘On the genealogy of ethics’, the title of which combines two of these three axes, Foucault tries to describe the three layers as separate forms of a ‘historical ontology of the self’, within which we constitute ourselves as subjects in three respects.14 Not only are the games of truth one of the leading analytical aspects of all three areas, so too is the constitution of the subject. Among the most succinct texts on this topic is the terse explanation in the so-called Autobiography, not all of which was actually written by Foucault himself.15 It identifies and explains ‘the relationship between subject and truth as the main thread’ of all of Foucault’s analyses. More precisely, this relationship is supposed to be ‘an analysis of the ways in which the subject can be introduced as the object into the games of truth’. In other words, Foucault’s central question is ‘what are the processes of subjectivisation and objectification which can lead to the subject, in its role as subject, becoming the object of knowledge?’ In this context ‘subjectivisation and objectification are not independent of each other’, but rather ‘their reciprocal 12 UP 6, 7. 13 See the comment in Foucault’s late article ‘Technologies of the self’, in which he sums up his entire programme: ‘My objective for more than twenty-five years has been to sketch out a history of the different ways in our culture that humans develop knowledge about themselves . . . the main point is . . . to analyze these so-called sciences as very specific “truth games” related to specific technologies that human beings use to understand themselves’ (Martin, Gutman and Hutton (1988), 17–18). 14 Rabinow (1986), 351. 15 In the following I refer to the German translation, which first appeared as: ‘Autobiographie’, Deutsche Zeitschrift f¨ur Philosophie (42 (1994), 699–702).

10

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

relationship leads to what can be called games of truth’. At the same time it is assumed that power relationships as ‘different and particular forms of the “rule” of individuals’ play a leading role ‘in the various manners of objectification of the subject’. It goes without saying that the process of ‘subjectivisation’, or of the ‘constitution of the subject’, is to be understood neither in terms of traditional subject philosophy nor in terms of the constitution of personal identity. Here too Foucault is, among other things, a nominalist – the existence of human individuals is taken for granted. It is only through the double process of subjugation to power relationships and the reflexivity of epistemic self-reference that individuals become subjects. Subjectivisation in this sense also implies an objectification of the subject and its adaptation to specific historical games of truth. However this systematic connection is to be understood exactly, one thing should now be clear: neither is the constitution of the subject specific to ethics, nor in the context of the programme of ethics are the analysis of power and the archaeological study of the formation of (scientific) fields of knowledge irrelevant. Inasmuch as it claims exactly the opposite, the traditional characterisation of the ethics programme is thus indeed one-sided and misleading. For this reason, in what follows I assume that Foucault’s work is to be regarded essentially as a historical theory of knowledge. His main concern is a historical description of how the fields of study and subjects of human science have taken shape,16 and how this process was interconnected with forms of power relationships. It would seem that Foucault had some remarkable intuitions in mind, but his thoughts are often terminologically somewhat unclear and unsystematic, so that it often takes a great deal of explicatory work to reconstruct them in a satisfactory form. Particular attention must be paid to a precise clarification of the intrinsic links between subjectivisation, objectification and power, which is the intention of this chapter. I shall proceed as follows: (1) first of all I shall reconstruct the concept of the subject in the Foucauldian definition of power in terms of a nominalistic approach, as well as clarifying what the intelligibility of the forms of power comes down to; (2) then I shall turn to a number of indications that Foucault attempted to link power and 16 This aspect is introduced, for example, by Gutting (1989) in his excellent study. See also the works by Hacking (1979, 1986a and 1986b), Rorty (1986) and Rouse (1987 and 1994). As for the controversy between Hacking and Rorty as to whether Foucault developed and presented a new (form of the) theory of knowledge, I agree with Rorty in assuming that Foucault effectively makes no contribution to questions and problems of the classic theory of knowledge, but follow Hacking in believing that fruitful insights into aspects of it can be obtained from Foucault’s analyses of the relationship between knowledge and power.

morals, knowledge and power

11

knowledge more closely than is normal in standard models of the relationship between the two; (3) a precise statement of a productive definition of power in terms of regulative power is then necessary in order to be able to illuminate the true and deeper form of the relationship between power and knowledge; (4) finally, some general results can be presented concerning the character of the programme of ethics and the orientation of Foucault’s project. 1 One of Foucault’s central ambitions is to produce an analysis of power which promises to reveal interesting relationships between power and knowledge. The examination of these relationships is what distinguishes his analysis from most other influential modern approaches. The recent debate on the concept of power developed in the 1950s within the systematic framework of behaviourism, and this in turn led to a causal paradigm of power based on the idea that power is an ability to produce causal effects in the world. It was a natural step for behaviourists to link this causal approach of a power to bring something about with the idea of power over persons: person A has power over person B when A’s behaviour induces B’s behaviour and when B’s behaviour would not have occurred without the behaviour of A. Of course, this explanation was qualified with behaviouristic criteria: A and B are in a strict sense individuals, the behaviour of A and B can be observed empirically, causality can be reduced to empirical regularity and the behaviour of A regularly precedes the behaviour of B. This model essentially sees the exercise of power as a communication of preferences. Person A exercises power over person B when A communicates her own preference to B and B adapts her own behaviour to that preference.17 A number of later theories remained committed to the causal paradigm of power, but were careful to distance themselves from the positivist premises of behaviouristic concepts. For example, the realistic model of power is based on a strong philosophical realism, which, in contrast to positivism, has the confidence to talk about natural necessity, intrinsic nature 17 See for example Simon (1957); Dahl (1969), 79–93; Cartwright (1959); Dahl (1961), 763–72. In particular, Robert Dahl’s behaviouralistic definition of power was developed from criticism of Mills’s (1956) influential book, in which the latter spoke of a ‘power elite’ in modern societies who occupy the strategic command posts in their social structure. For Mills they have power in so far as ‘they are in a position to make decisions having major consequences’ (3 f.). Dahl criticised this definition of power for not being scientific, since it was vague and could not be empirically operationalised. His own behaviouralistic definition was intended to satisfy scientific, theoretical standards of precision and empirical verification.

12

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

and irreducible real causality. At its most fundamental, this model regards power as a disposition to causal influence, just as copper has the power to conduct electricity. However, people have this causal-dispositional power within the context of a social world and as such, power is exercised by individuals as performers of social roles. The realistic model has a number of interesting consequences. For example, it implies that under suitable conditions power (as a disposition) can be realised, but it can never be completely exhausted – the exercise and possession of power must be kept analytically separate. Power also remains in principle an ability to bring about something (power to . . .) and any form of power over a person is based on this ability. Finally, power is not of necessity repressive. Under suitable conditions it can be in the interest of those over whom it is exercised and, above all, without power relationships societies would fall apart.18 A more recent variation of the causal paradigm defines power in terms of rational decision theory. Rational decisions in a social context require of the actors the ability to bring about specific circumstances (outcome power), and in particular to deliberately change the motivational structure of other actors (social power). Both cases require cooperation with other actors, and this is part of the decision-taking. However, since in this complex context the actual benefit of rational decisions is also dependent on chance and the unintended success of actions, the decision-theory model of power requires us to differentiate between power and benefit: someone who has power does not always benefit from it.19 The causal model of power was criticised from a number of angles. For example, it was claimed that the notion of power has an irreducible 18 See on the subject Harr´e (1970), 81–101; Harr´e and Madden (1975). The best and most comprehensive account of the realistic model of power is presented by Isaac (1987). The first part of the book (in particular ch. 3) develops the realistic theory of power, while the second part demonstrates that Marxist and neo-Marxist theories of power are insufficient for an analysis of power in late capitalist societies. Influential articles by Backrach and Baratz expressed other reservations about the behaviourist model. They pointed out that power has ‘a second (hidden) face’, in that an effect of power can be that nothing happens or that everything continues unchanged. The behaviourist model, which can only understand the effects of power as observable changes in circumstances or behaviour, is thus too restrictive (see Backrach and Baratz (1962), 451–60; (1963), 641–51; (1970), esp. part i). 19 See Dowding (1991). The fundamental distinction between causal power as an ability to bring about something, and social power as power over other persons, is prominent in more recent work, e.g. Wartenburg (1990), 17–27; Fink-Eitel (1992), 35–56 esp. 37–9. Fink-Eitel’s criticism of the causal model of power is not in the least convincing. It is based on the observation that power as a causal disposition is not empirically observable, and that power as an ability is destroyed by its own realisation, so that ‘in the theory of power the criterium of the empirical verification of causally determined behaviour has to be abandoned’ (1992, 41).

morals, knowledge and power

13

normative content, which can be defined more or less as follows: person A only exercises power over person B when A brings about that B acts in a way which runs contrary to B’s objective interests.20 But power was also seen in a positive light – as political power, manifested in the political institutions which are constitutive of public and social order. This political form of power must be distinguished from violence or authority; it can only arise in groups in which equals communicate with each other and then act accordingly and gain power.21 According to a watered-down version of this communicative or consensual model of power, power is influence over processes of social decision-taking. It relies on the threat of sanctions, but is nevertheless based on consensuality: only those who do not submit themselves to consensual or democratic decision-taking procedures are actually threatened with negative consequences.22 Among the points emphasised by communicative models of power is the function of particular forms of power for society as a whole. A radical form of this view is found in holistic concepts of power as proposed, for example, in system theories of society. Such theories understand power as a symbolically generalised medium of communication that plays an important integrating role, for example in achieving the commensurability of different needs or motives, or the coordination of courses of action. According to the holistic view, power plays a significant part in maintaining social systems in an all too complex environment by reducing their complexity. In particular, power subverts the possibility of refusal and resistance through selective pre-formation of the premises on which decisions are based.23 20 See above all Lukes (1974). Like Connolly (1974) and others, Lukes regards the concept of power as being ‘essentially contested’, in that there are no purely theoretical criteria free of normative values for its application. Gallie (1955) introduced the term ‘essentially contested concepts’. 21 This is the influential concept of power proposed by Arendt (1958) and (1980). Within the framework of a new variation of critical theory, Fay developed Arendt’s ideas further to produce a ‘dyadic’ concept of power. This sees power as a relationship between the powerful and the powerless which can change when the powerless use ‘communicative’ power to attain ‘dominant’ power; see Fay (1986), 130 ff. Wartenburg (1990), ch. 2 is a respectful, but critical analysis of Arendt’s theory. Habermas wanted both the instrumental and the communicative concept of power to be included in the theory of power: instrumental power (in the sense proposed by Max Weber as the ability to impose one’s own will on others) is linked to strategic action, communicative power (as developed by Hannah Arendt) with communicative action; see Habermas (1981). 22 See in particular Lasswell and Kaplan (1950). 23 See Parsons (1967), 297 ff.; Luhmann (1973). Fink-Eitel (in Angehrn and Fink-Eitel; (1992)) regards Luhmann’s theory as the most complex and productive approach to date. However, he overlooks the fact that Luhmann’s theory of society employed an excessively intensive system-theoretical vocabulary. The concept of autopoietic systems cannot be applied to societies. Barry Barnes proposed a more cautious, but credible, holistic

14

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

One of the more effective recent theories of power is based on an action environmental model. Its aim is a fundamental definition that is on the one hand sufficiently exact, but on the other hand sufficiently open, to be able to generate the best of the existing definitions of power by means of terminological restriction of the basic model. The idea is to use three theoretical components to develop an action environment for an individual: the alternative courses of action, comprehension of the situation and of the alternative courses of action, and evaluation of the assessment of the alternatives (if necessary with a utility function). Then it can be said that person A exercises power over person B when A intentionally and with significant consequences changes the action environment open to B, i.e. when A intentionally and with significant consequences changes at least one of the three components of the action environment of B.24 This form of power is necessarily asymmetrical and need not be repressive, although apparently the assumption that two people exercise mutual power over each other, and that there are forms of repressive power, is compatible with the model. At the basic level of definition the model says nothing about the forms of power, the means of the exercise of power or the accompanying interests. However, such specifications can be derived from the basic model by means of additional conditions. For example, physical violence can be understood as a form of the exercise of power which involves A physically preventing B from pursuing a particular course of action that B wishes to pursue, or else A physically inducing B to pursue a particular course of action that B wishes to avoid. On the other hand, A exercises force over B when A, by means of threats and the generation of fear, changes the action environment open to B. Finally, A exercises influence over B when A, by means of communication and information, induces B to agree to pursue a course of action which is in A’s interest (manipulation is then a special case of influence – a form of influence that uses morally dubious means). However, in the action environment model normative concepts of repression and domination are not additional specific forms of power, but rather specific forms of the use of power. One of the most important advantages of the action environment model is that it pays attention to the broader environment of elementary dyadic conception of power. Essentially the possession of power is dependent on the distribution of the use of the ability to act; above all, power and the free disposal of power are a characteristic of the distribution of knowledge in a society, and thus are a characteristic of the society as a whole (see Barnes (1978)). 24 This theory was proposed by Wartenburg (1990).

morals, knowledge and power

15

power relations. The reaction of social actors outside a dyadic power relation (and thus ‘peripheral’ to it) is a vital element in its formation and preservation. In this model the social structure which induces the particular social actors peripheral to a dyadic power relation on the one hand to align themselves to this power relation, and on the other hand to stabilise it and to ensure its functionality through the manner of their reaction, is called social alignment.25 All dyadic power relations are dependent on the social alignment of peripheral actors and are thus embedded in a broader social structure. The most important theoretical advantage of the concept of social alignment is that it allows us to formulate a dynamic definition of power. If power relationships only exist inasmuch as they are constantly reproduced anew by the actions of the persons involved, and if the actions of the peripheral actors are a particularly important factor in this, then the constant change of social alignments will lead to a constant new formation of dyadic power relationships. And vice versa, a change in the structure of dyadic power relationships can also change their peripheral social alignment. As has already been indicated, the action environment model assumes a basic definition of a dyadic power relationship that is neutral as far as the question of whether or not power is repressive is concerned. Forms of repressive power can be derived from the basic definition by means of additional restrictions. One of the advantages of this approach is that it permits a conceptualisation of positive power that benefits those who are subjected to it. A common form of positive power is paternalism: the dominant actors use their power to help other actors who are not fully capable of rationally determining their actions. One particularly interesting variant of positive power has attracted the attention of feminist theory: the dominant actors use their power to ensure that the subjugated actors can develop in such a way that the differences upon which the power differential between the dominant and the subjugated was originally based disappear.26

25 The dyadic power relation between teachers and students depends, for example, on the actors outside this relationship interpreting the manifestations of the relationship suitably (the school report that the teachers write), and reacting in a particular way, e.g. by refusing to give a student with bad marks a job. It is this reaction that turns the relationship between teacher and student into a power relationship (see Wartenburg (1990) 142–6). 26 The classic example is, of course, the good parent–child relationship; see Wartenburg (1990) 183–222 (in this context Wartenburg talks of ‘transformative power’). For the relevant feminist literature on the subject see Trebilcot (1983) 213–30; Ferguson (1983); and Kuykendall (1983); also Hartsock (1983).

16

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

Foucault did not explicitly base his analysis of power on those elements of this tradition of theories of power that had already been developed when he wrote UP. But it is remarkable that he shares some of its most important intuitions, as well as several from subsequent theories of power. In particular the action environment model reveals a close systematic affinity with Foucault’s analysis of power. He emphasises the ‘strictly relational character of power relationships’, and makes it clear that within the context of the analysis of power ‘one needs to be nominalistic, no doubt’. It is ‘not an institution, and not a structure, neither is it a certain strength we are endowed with’, but rather a ‘complex strategical situation’ in a society which of course consists of ‘a multiplicity of force relations’ within various social groups, and thus ‘it is produced, in every relation from one point to another’.27 Quite clearly for Foucault the concept of power was essentially individualistic and nominalistic, and he regarded it primarily as a relation between concrete individuals. Formally it is asymmetrical, and can be reflexive,28 while of semantic importance is the fact that Foucault did not see repression as a conceptual consequence of power relations. We can capture the essence of this as follows: (M) A person P1 has the power to induce a person P2 to do X when (a) P1 has the chance and the means to induce P2 to do X, (b) P1 intends to induce P2 to do X, (c) as a result of P1 ’s influence P2 occasionally does X without anyone else forcing P2 to do X, (d) P2 would not have done X without the influence of P1 , (e) P1 is not normally identical with P2 , although it is not impossible that P1 = P2 , (f) the fact that P1 has power over P2 in this sense does not imply that P2 also has power over P1 . The asymmetry of the power relation as qualified in (f) is not contradictory to its reciprocity, that is with the situation whereby two people exercise power over each other. Similarly, love is apparently an asymmetrical relation, but fortunately that does not exclude the possibility that two people love each other. The conditions (a)–(f) in (M) would seem to be sufficient for the existence of a power relationship, but not all of the conditions are necessary for it (e.g. see condition (b)). But the existence of a power relationship as defined by (M) does not imply that important preferences, 27 See HS 93, 95. 28 See the commentary on explication (M) below.

morals, knowledge and power

17

needs or interests on the part of P2 are frustrated, or that resistance by P2 would be sanctioned. The individualistic explication of power (M) defines a power free of repression, but could, of course, be modified into an explication of repressive power by the addition of further conditions. The set of all power relations as defined by (M) (in a particular society at a particular time) Foucault calls ‘local force conditions’, and regards them as the basis of all forms of power – a basis that is apparently like a ‘moving substrate of force conditions’ that are ‘always local and unstable’, and constantly shifting.29 However, it is not yet really clear what status the general form of the exercise of power as defined by (M) has. Above all, what does it mean that one person can, by means of power, induce another person to do something? According to (a) and (b) in (M) the matter in question is an action, i.e. an activity that can be described with reference to an intention on the part of the actor. Furthermore, the action of the person exercising power, P1 , can generally be rationalised: P1 herself has an intention Z, and believes it can be realised when she induces P2 to do X. However, the perspective of P2 , over whom power is exercised, must be differentiated as well. For when P1 exercises power over P2 this generally means that P1 is able to bring about causally circumstances under which it is rational for P2 to do X, and that P1 intends to do just that, to bring about circumstances under which P2 reassesses the rationality of her own actions so that P2 does what P1 wants her to do. This is what (c) and (d) in (M) really come down to. These intuitions can be summed up as follows: (M*) A person P1 has the power to induce a person P2 to do X when (a) P1 has the chance and means to bring about causally circumstances U such that under U it will be rational for P2 , given her own intentions and beliefs, to do X, (b) P1 intends to bring about causally circumstances U such that under U it will be rational for P2 , given her own intentions and beliefs, to do X, (c) P1 has aims Z such that she is convinced that P2 doing X is necessary for the realisation of Z, (d) on the basis of the circumstances U which P1 has brought about, P2 occasionally does X, without there being anyone other than P1 for whom this applies under conditions (a)–(c), 29 See HS 93.

18

f o u c au l t a n d c l a s s i c a l a n t i q u i t y (e) without circumstances U P2 could not describe action X so that, given her intentions and beliefs, X could be rationalised, and therefore without U P2 would not have done action X, (f) = (e) in (M), (g) = (f) in (M).

(M*) provides the action-theoretical reformulation of the regulation of power according to (M). Now and then Foucault would like to say that somebody exercises power over herself. If one rejects the traditional or psychoanalytical division of the soul, as Foucault undoubtedly does, then power over oneself would seem to be a reflexive relation. This is problematic, as a symmetrical (reflexive) relation can hardly be regarded as power, and it is hard to see how Foucault could solve this. Perhaps he feels that one and the same person can be looked at under different descriptions, and that the reflexive local power relationship can be regarded as a relation between a person under one of her descriptions and the same person under another description. More important is the fact that Foucault sticks to four general conditions which must be consistent with explication (M*): – First, power may not be hypostasised ontologically; nevertheless in a certain sense it is not based on subjects.30 – Secondly, power exists fundamentally on a level beneath that of states, even at a sub-institutional level, but it can nevertheless consolidate itself to greater aggregates, systems of rules, or ‘strategies’.31 – Thirdly, as a negative force power often entails regulation, restriction and regulation, yet at the same time as a positive force it releases energies, connects itself with ‘games of truth’, and continually conquers ever wider areas.32 – Fourthly, in its most important forms power is historically and systematically contingent, yet it is also intelligible and can be revealed through the analysis of power.33 Each of these four conditions evidently contains systematic tensions; hence any acceptable interpretation of the Foucauldian concept of power 30 ‘Power relations are both intentional and nonsubjective’; HS 94. 31 Foucault describes this process, among other things, as ‘mobilising groups or individuals’; HS 96. 32 See the exemplary discussion of the repressive hypothesis in Foucault HS 15–49, as well as the division between positive and judicial power in HS 133–59. 33 See HS 94–5 on the ‘rationality of power’.

morals, knowledge and power

19

would have to show how these conditions can be satisfied consistently.34 As to conditions one and four, I hope to do this in this section; the remaining two conditions will be discussed in sections 2 and 3 below. Foucault analyses power on three different levels. The fundamental level, which consists of a multiplicity of force relations or local power relations, is analysed in terms of their specific historical forms, or structures. Foucault is not interested in exploring which individual exercises power over other individuals, or which social groups or classes exercise power over other groups or classes; rather he wants to describe, as precisely as possible, the forms or structures that certain power relations take on in certain historical situations, quite independently of the way in which these forms depend on the individuals between whom the power relations hold. In this sense the analysis of power concentrates on local forms of power, while the individual humans who constitute the reference points in local power relations remain ‘mere points’: power is ‘exercised from innumerable points’, and it is ‘produced from one moment to the next, at every point, or rather in every relation from one point to another’.35 To be sure, for the analysis of power the individuals involved in a power relationship remain uninteresting ‘points’, but at the same time from an ontological perspective these points must constantly instantiate and reproduce power relations. On a second level the analysis of power investigates the dynamic of local forms of power, that is their ‘play’, their historical shifts (according to the ‘rule of continual variations’).36 On this level the analysis of power claims to do no more than to describe or narrate which shifts of a marginal or dramatic form have taken place historically. In particular, in this setting the analysis of power has no explanatory pretensions, but rather its narrative modesty corresponds to the conviction that the play of power forms is contingent. Perhaps it may not be entirely clear whether this contingency is ontological or merely epistemic; however, even if the dynamic of power forms turned out not to be ontologically contingent, but partly controlled by causal mechanisms (as is suggested, for example, when we speak of non-intended side-effects), the analysis of power still proceeds on the assumption that on the epistemic level the most that is possible 34 A typical example of the uncertainty of even the best scholars of Foucault about the terminological explication and the methodological status of the Foucauldian concept of power, are the helpless questions that Dreyfus and Rabinow ask at the end of their long and detailed interpretation of the concept (see Dreyfus and Rabinow (1983) 206 f.). 35 HS 94, 93. 36 HS 99.

20

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

are existence assumptions about causal mechanisms, but that it is hopeless to attempt to assign definite causal mechanisms to the complex play of the shifts of power forms. And even if this were possible in individual cases, from the viewpoint of the analysis of power it would not be very interesting. It goes without saying that Foucault regards neither intentional nor functional (i.e. system-theoretical) patterns of explanation as appropriate. It is true that at times he lapses into these patterns in his historical accounts, but it seems certain that he does not and cannot want to regard the processes that interest him as events which were intended by individual agents, and that he explicitly rejects structuralist and systemtheoretical approaches in history and social theory. Rather, the central and most illuminating feature of Foucault’s analysis of power is a detailed differentiation of the forms of power and a narrative of their complex dynamic. On the third level the analysis of power forms of confinement and systems play roles which are characterised as ‘props’ and ‘strategies’ that result from the dynamic of local power forms and – according to the rule of ‘double conditioning’37 – are embedded in the power base. Undoubtedly this level involves the most difficult theoretical problems in the analysis of power. This is all the more important as it is on this level that Foucault establishes his systematically central concept of power: ‘it is the name that one attributes to a complex strategic situation in a particular society’.38 But the development of this strategic situation – including powerful institutions and state apparatuses – is described more metaphorically: as mutual support, crystallisation, coordination or confinement of local power bases to global strategies, global societal disjunctions or state apparatuses.39 Talk of a logic, and even of intentions and goals that can be disclosed by the analysis of power at this global level, is thoroughly confusing. Apparently one has to assume that here too neither individuals with their goals nor systems with their imperatives of self-preservation and development are meant to play an explanatory role. And finally even ‘double conditioning’ between the local power base and the global strategies on the one hand, and the fundamental character of the power base on the other, leads to blatant systematic tension.

37 HS 99–100. 38 See HS 93; significantly this explication is directly connected to the ‘nominalism’ which forms the basis of the analysis of power. 39 HS 93–6.

morals, knowledge and power

21

Power on the global level could be conceived of as the sum of all instantiated forms of power, that is as the total network of local power bases; or as the total network of local power bases together with their (possibly structural) secondary conditions; or finally as a global structure or a general apparatus in contrast to which the particular forms of power represent ‘tactical blocs’ within the apparatus. Presumably all three aspects come into play, and the main concern seems to be to conceive of the grid of local power forms so that in spite of their variety and effects they fit into a general global apparatus, and thus can also be newly interpreted in their conditions, relations and particularity. To describe the structure of the reciprocal relation and the ‘strategic’ situations of the forms of power within the apparatus as a whole (and no more!) means to decode the ‘logic’ or the ‘intention’ of power relations. The apparatuses of alliance and sexuality are good examples of global (structural) power relations: their development could not be described system-theoretically, but rather they are global structures into which specific forms of power ‘fit’. Their emergence is not explained causally, but their structure is simply described and the history of their shifts, ruptures and tensions narrated. This history can be explanatorily fruitful and can illuminate interesting facts, but neither the narrative nor the history of the global structures has theoretical priority over the description of the local power forms. Rather the relations between local forms of power and global strategies is ontologically and epistemologically reciprocal. Ontologically they are like the relation between species and genus in Aristotle: genera are encompassing, comprising several species below them, but they exist only potentially: the species are the actualisations of the genus. Accordingly, the individual, local relationships of power actualise the forms of power, and these forms of power, in turn, actualise the global structures and apparatuses – there can be no local power relation without the power form and a global power structure, and no power form or global power structure without local power relations. Historically, power never occurs in a space without order, but the social orders and structures are always tied to their actualisation by means of local power relations.40 Epistemically it 40 Rouse (1994) 106 ff. describes these relations similarly, but in a different form derived from Wartenburg’s (1990) suggestion of the concept of ‘social alignment’: each individual exercise of power is dependent on the fact that other persons and groups act in harmony with the person exercising power (i.e. within the framework of an analogous ‘social alignment’, which is often created by sub-institutional or institutional social rules). But, vice versa, each local exercise of power also represents a re-enactment of the social alignment.

22

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

seems clear that the global apparatus can be ‘derived’ from local forms of power, but that inversely the understanding and interpretation of local power forms will also be transformed and sublimated by their relationship to the global apparatus. On all three levels we can render power intelligible and at the same time show how it is contingent. Forms of power, their dynamic, their global strategies and their shifts represent intelligible structural patterns, even if their emergence and development can only be represented in narrative form. Further, we have seen that all forms and strategies of power rely ontologically on individuals who instantiate them, but that these individuals are theoretically irrelevant for the analysis of power because their specific characteristics do not play a systematic role in it, either in a narrative or an explanatory sense. From this perspective the thesis that power is not tied to subjects, not even individual subjects, could be stated more precisely as follows: the analysis of power can regard power theoretically as if it were not tied to subjects. The same result can also be derived from Foucault’s discursive analysis, and so this may be the place to include a remark on the Foucauldian concept of discourse. One often gets the impression that Foucault wants to trivialise the concept of discourse – perhaps in order not to restrict the discursive analysis unnecessarily from the outset. We are often told, for instance, that ‘the forms of discourse about sex’ multiply or change; in such claims ‘discourse’ appears to mean nothing more than an actual form of language-use. However, the analysis of discourses relates these forms of language-use to a predicate, or a description which itself has a rich structure, e.g.‘confession ritual’ or scientia sexualis. For Foucault this predicate or description apparently also belongs to a particular discourse. As a first approximation one may perhaps say that discourses are forms of language-use seen relative to a certain predicate or under a certain description – for example, forms of language-use in so far as the description ‘confession ritual’ or scientia sexualis applies to them. This description, however, is only given theoretical content by the analysis; hence it could be argued that certain forms of language-use only become discourses as a result of the discursive analysis. Forms of language-use can, of course, be analysed in a variety of different general ways – syntactic, semantic, grammatical, logical or pragmatic. These different forms of analysis subsume historical forms of languageuse to a specific description. In the same way the Foucauldian discursive analysis seeks to subsume particular historical forms of language-use to specific kinds of descriptions that are characterised, among other things, by the following incomplete list of aspects:

morals, knowledge and power – – – – – – – –

23

which kind of things are being talked about; which kind of predicates are used to talk about things; which kinds of sentences include truth values; which forms of power are intrinsically related to which forms of language; which forms of knowledge and standards of rationality are used; which kind of relation to self is connected with this language-use; into which global apparatus of power the form of language fits structurally; which ideas of the good life are associated with it.

Inasmuch as particular historically attested statements or texts allow an answer to this list of questions, in the context of the discursive analysis they are regarded as discourses. For example, at a particular point in the history of European culture people start to explore, and to talk about, their sexual fantasies, or at another point they begin to describe sexuality as a causal mechanism. From the nineteenth century on psychoanalysts acquire new forms of power in their conversations with patients that are intrinsically related to the specific forms of language they use in these conversations – it is a power based on competence in the interpretation and diagnosis of the patients’ reports and observations of themselves. This competence then takes on a scientific aspect that leaves its mark on the language and theory of psychoanalysis in the form of standards of rationality and method. In this way the Foucauldian discursive analysis turns out to be a specific analysis of language-use that establishes, and at the same time explores, historically specific discourses. The notorious e´nonc´es, ‘statements’, which are neither sentences nor propositions in the ordinary sense but historically formulated utterances, are the elements of discourses, in so far as they are examined specifically with respect to one of the aspects mentioned above – not explicitly, of course, but as part of the interpretation involved in the discursive analysis. Just as it is the discursive analysis which constitutes a particular common form of language-use in its entirety as a discourse, so the e´nonc´es are revealed by the discursive analysis. More precisely, both levels are closely related to each other: to reveal the e´nonc´es and their orders means nothing else than to highlight particular discourses. The discursive analysis is especially concerned with large-scale, contingent ruptures or transformations, that is with forms of practised speech in which suddenly new kinds of objects are discussed and endowed with new kinds of attributes; in which speech is related to new forms of power; which employ new forms

24

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

of knowledge and standards of rationality and reveal a new kind of relation to self, etc.; in short, in which new e´nonc´es emerge. The discursive analysis assumes that from the internal perspective of the form of language-use these general discursive aspects are not themselves the objects of speech, but are present only as implicit structures or rules in a particular use of language – however, these rules or structures can be made explicit from the external perspective of the discursive analysis, namely within a theoretical framework which aims not so much at the analysis of meaning but at the analysis of rules and structures. It is through this kind of analysis that certain forms of language-use become discourses, i.e. can be regarded as discourses. The formation of discourses, despite the contingency of their internal structure and history, apparently represents one of the most important intelligible aspects to which Foucault’s analysis can be applied. However, the relationship between the analysis of power and the analysis of discourse is somewhat precarious. On the whole the latter is restricted to the archaeological level of the forms of knowledge. As far as technologies of the body, practices of power or techniques of the self are concerned, discourses and discursive analysis are rarely mentioned, and even then not precisely. The generally accepted picture seems to be that Foucault studies language-use through discursive analysis, forms of power through the analysis of power and techniques of the self through ethics – but this is unsatisfactory. On the one hand those forms of language-use, which on the standard interpretation are studied by discursive analysis within the framework of archaeology, describe particular – e.g. epistemic – practices, and on the other hand forms of power and techniques of the self can only be accessed in the historical analysis through documents, that is descriptions and forms of language-use. For this reason the concept of discourse must be broadened so that we can also talk of discursive analysis within the context of the analysis of power and ethics. This must be achieved by extending the depth-structural aspects through which the discourses are defined. The aspects mentioned (questions, rules) can then be understood as examples of epistemic abstractions, through which epistemic discourses are constituted. Historical forms of language-use can also be examined from other aspects, which we can call depth structural. Descriptions or documents about historical forms of penal practices or of laws can, for example, be examined from the aspect of which fundamental forms of power (repressive, judicial, structural, intended, productive?) they reveal. If we call these abstractions dynamic, then for particular forms of language-use they constitute dynamic discourses. Or else descriptions

morals, knowledge and power

25

and documents about designs for a good life, about forms and means of self-transformation or about yearnings and desires can be studied from the aspect of which moral problematisations, which teleology and which forms of subjugation are visible in them. If such abstractions are called ethical, then for particular forms of language-use they constitute ethical discourses. Therefore we can say that when conducting a discursive analysis Foucault always analyses historical documents and texts of language to find the depth-structural aspects or rules which the discursive analysis reveals to be particularly interesting and relevant. However, the process is partly an epistemic, partly a dynamic or ethical discursive analysis – and perhaps sometimes all of them at the same time. Forms and strategies of power on the one hand, e´nonc´es and discursive formations on the other, are thus structures which show up in the theoretical analyses of power and of discourse. It should be clear from what has been said that these structures – viewed ontologically – exist only in their individual substrates: in individual relations of power and specific speech acts. The structures of power and of discourse are the forms, and to some extent also the effects, of concrete relations of power and languageuse. Yet, together with their individual substrates they combine to form an extensive, complex grid, about which it is no longer meaningful to say that particular individuals plan its existence and form, or even – in the ordinary sense – ‘know’ it. Rather, it might be said from a theoretical perspective that grids of this kind continually establish or transform themselves in a contingent way (what is meant by saying that these grids establish or transform themselves is that their formation and transformation is a – generally unintended – cumulative result of concrete power relations and historical forms of language-use). One important aspect of this could be put even more pointedly: the analyses of power and of discourse, as Foucault construes them, are primarily concerned with dissecting and specifying the historically contingent depth structures of the relations of power and speech acts observed in each case. Perhaps it is not entirely clear which of these structures are deep and why; but at a first approximation it may be argued that the global strategies and ‘apparatuses’ represent depth structures within the framework of relations of power, whereas the respective specific e´nonc´es and the corresponding discourse formations constitute the depth structures of speech acts. Such structures are undoubtedly not deep in a semantic or hermeneutical sense – the Foucauldian analytic does not search for deep meanings; rather, these structures represent only the most general forms of the respective discursive and non-discursive practices. Hence it is more than doubtful whether the ‘depth’ of these

26

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

forms should be construed in traditional philosophical concepts, such as ‘constitution’ or ‘transcendence’ or the a priori (in a historicised sense), as is often the case in modern literature. It should be recalled that these ‘structures’ – e.g. certain strategies of power or certain standards of truth – can also be represented as fundamental rules through which concrete relations of power and speech acts are organised (it is precisely for this reason that Foucault likes to speak of ‘technologies’). But just as the depth structures or rules of power cannot be planned, intended or manipulated by single individuals, so the depth structures of speech acts that are geared towards the ‘production of truth’ cannot be discovered, conceptualised or transformed by a single individual. What is more, single individuals can only follow them as rules implicitly, not explicitly, and they are normally brought to conform to these rules in common practice. However, the analyses of power and discourse can render these rules explicit – and this is one of their most prominent theoretical tasks. Hence it is not only the complex and global character of power and discourse, but also their depth-structural aspect, which exclude any explanatory role for individuals in the analysis of power and discourse. Foucault’s concepts of le pouvoir and le savoir are meant to refer precisely to these forms of power and knowledge ‘which have no subject’ – a power that is not exercised by subjects and a knowledge that is not known by subjects. Yet this is not a sociological or epistemological mystery, but a clear and comprehensible concept. Apparently there is another interesting perspective which allows Foucault to claim that power is not tied to subjects, without regarding it as a free-floating entity or structure. As has already been pointed out,41 in his later works, and above all in the Autobiography, Foucault assumes that the relationship between the processes of subjectivisation and objectification are of central importance for his study of ‘the historical ontology of the self’. But power relations are also decisive as forms of ‘government’ of individuals: it is only the double process of subjugation to power relations and the reflexivisation of self-reference that makes of individuals ‘subjects’ and at the same time ‘objects’. In other words, power relations are already constitutive for the process which sees individuals become subjects. Thus power relations remain ontologically linked to individuals, but at the same time they are by no means tied to subjects – in fact it is the other way round: in their subjectivisation subjects are dependent on power relations. However, it should be clear that this independence of power from 41 See above, pp. 9–10.

morals, knowledge and power

27

subjects is purchased at the expense of a concept of the subject which, when compared with the status and weight of the concept within traditional ‘subject philosophy’, has lost a great deal of its explanatory power and has thus become quite trivial. Hence the fashionable talk of the ‘death of the subject’ proves to be a scientifically comprehensible and ontologically harmless concept that is also not particularly far-reaching. It consists of no more than the following claims: – that in the context of the specific methodical orientation of the analyses of power and discourse neither individuals nor subjects play an important explanatory role; – and that from the perspective of a subjectivisation which is bound to subjugation and reflexivity, power is always constitutive for the formation of the subject. At the same time it is presupposed that, ontologically speaking, in its most fundamental aspect the power relationship remains a relation between real individuals. This demonstrates how power can consistently be both nominalistically conceptualised and regarded as being independent of subjects. One could say that in his archaeology and analysis of power Foucault proclaims the explanatory death of the subject. The analyses of power and discourse aim at a purely formal, structural approach. They are concerned with the analysis of particular historical forms of nominalistic power relations, and texts and statements which have been handed down. The analysis is also quite compatible with the thesis that the historical dynamic of these forms and complexes of forms is essentially contingent. This demonstrates how power is at the same time both intelligible and contingent. Thus, two of the four general systematic conditions mentioned above have been given a consistent interpretation. The discussion of the other two conditions, that is the relation between positive, non-repressive power and negative, repressive power, and the relation between local power bases and ‘consolidated’ institutional or political power structures, requires a more profound understanding of the concept of power which Foucault develops through the relation between power and knowledge. It is the intrinsic interconnection with knowledge that reveals the fundamental character of power. For Foucault this is not only a systematic statement, but a historical one, as becomes quite clear in HS, where he discusses the power–knowledge complex before the analysis of power is introduced.

28

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

My discussion here will follow his arrangement. My prime concern is to show that Foucault extracts a specific concept of non-repressive power from the analysis of the power–knowledge complex, and distinguishes it as the basic form of power. By adding further conditions this ‘positive’ concept of power can be transformed into the concept of repressive power: how strong the resulting concept of repressive power is will depend on what further conditions are added.42 We will see that in his historical analyses Foucault often also employs repressive concepts of power – despite the systematic fundamental role of the non-repressive concept. 2 If Foucault’s proposal of an intimate connection between power and knowledge is supposed to be part of an interesting theory that contributes to a systematically fruitful concept of power, then it should be possible to show that this thesis goes beyond the best-known standard models that connect power and knowledge. The most important of these models regards power as domination or as a dominant system of norms, and knowledge as science or scientific theory-formation. It is then maintained, for instance, that power is usually connected with the beginning and the end of theoryformation, because power often influences the choice of fields of research and the application of proven theories (where power is based on a legitimate or even just system of norms, it is even reasonable to demand that this be the case). However, on the other hand, the critical verification of theories, i.e. the core of scientific activity, is neutral in value and free from power – or at least it should be. It is in exactly this sense that power is opposed to ‘truth’. Other models consider knowledge and its products (data, theories) as means to the intended end of securing and developing personal power, or they construe knowledge in terms of its effects on the stability of the conditions which are necessary to secure and develop comprehensive systems of power. Foucault, in contrast, aims at an interconnection of power with intrinsic aspects of theory development in the human sciences, or reflexive ‘games of truth’. In doing so he proceeds, as I have already mentioned, on the assumption that the establishment of fields of knowledge with their respective structures of power can neither be described in terms of a theory of action as the intended goal of influential agents, nor system-theoretically as the differentiation of systems of power and knowledge that stand over and above the individuals. Hence, an 42 See in more detail below, pp. 49–50.

morals, knowledge and power

29

interpretation of the power–knowledge complex and the corresponding fundamental concept of power has to be found which cannot be reduced to the standard models outlined here.43 The ‘repressive hypothesis’ which Foucault carefully reconstructs at the beginning of HS using the example of sexuality is a variant of the form of the connection between power and knowledge which Foucault attacks and seeks to overcome. A careful study reveals that this variant contains all the elements of the well-known standard models outlined above. According to the repressive hypothesis, from the seventeenth century onwards sexual behaviour was increasingly regimented through new methods of repression practised by particular persons who employed, among other methods, the application of knowledge – for example, sexuality was now restricted to core families and the marriage bed; its form and aim were standardised by means of pleasureless utilitarianism and the sober teleology of production, and were directed at procreation. The kinds of repression grew ever more subtle and employed sub-legal mechanisms (as a safety valve a few social islands, such as clinics and brothels were created, where, for a time, freedom from this repression could be bought). But at the same time these repressive mechanisms – which were often intended by the particular agents involved in them – fit in smoothly with the imperatives of preservation and development of rising capitalism: that is, of the new comprehensive system of power in modern Western societies. The imperatives of capitalism imply the application of a work ethic and discipline (for which the restriction of sexual activity and fantasy was clearly helpful); the securing of jobs (which was served by concentrating sexual behaviour on reproduction and hygiene); the untroubled acceptance of the system (for which brothels and clinics are necessary) – as well as the objective, scientific investigation of the mechanisms of power (through a new science, (political) economics). Thus the most important forms of 43 Among the literature on Foucault I know of no detailed and conceptually clear analysis dealing with this problem. Rouse’s article (1994) on the power–knowledge complex is certainly the most interesting recent study, but does not refer to any relevant work. Initially Rouse restricts himself to the remark that epistemological practices of surveillance, of enticing out confessions and of documentation also restrict and control behaviour and action (96), but his conception of knowledge is too instrumental, and that of power too repressive. More interesting, however, is his analogy between the concepts of political and epistemic power, which are seen as defining from a God’s-eye view non-partisan standards and forms of legitimacy in the political and epistemic realms respectively (102). According to Rouse this analogy and the struggle against these two concepts of power and their implications are the central motive for Foucault suggesting an intrinsic relation between power and knowledge. But with regard to the forms of this relation one finds little enlightenment in Rouse.

30

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

sexual repression, which clearly do indeed have an intentional element, can be explained functionally or system-theoretically within the framework of the repressive hypothesis. As for sexuality itself, speech, knowledge and the truth about the manifold forms of sexual activity and desire are prohibited and repressed, and repression proves to be the basic form of the connection between power and knowledge. Knowledge and truth can only establish themselves in a space free of power (‘truth’ here is understood in terms of the correspondence theory and is interpreted in the framework of metaphysical realism). Foucault’s reconstruction of the repressive hypothesis thus confirms that he is clearly aware of the best-known standard models of the connection between power and knowledge, but seeks to go further. However, it is only possible to decide whether he succeeds in an interesting way if some of the more important forms of the power–knowledge connection which he dissects and specifies can be explicated in a reasonably clear manner. As an example, I would like to attempt to outline such an explication of two of the most important discursive practices in the area of sexuality that Foucault analyses in HS: the confession ritual and the scientia sexualis. In doing this it will be important to pay equal attention to the restrictiveness and the productivity of the accompanying forms of power. If Foucault argues against the repressive hypothesis, this by no means implies that he denies that power does often appear in repressive form – even vis-`a-vis knowledge and truth. Rather, he denies only that the fundamental character of power is repressive. Perhaps this position is trivial on the level of fundamental concepts, because if the explication of the individualistic power relation suggested at the beginning of this chapter is acceptable, then it is immediately obvious that the repressive character of certain forms of power is not part of the fundamental explication. It would have to be introduced conceptually by adding that the action to which a person is to be induced through the influence of power frustrates important preferences of that person, and that an act of refusal would be sanctioned (only then – if one gave up Foucault’s neutral discursive standpoint – could the sophisticated terminological machinery of critical theory be linked to an account of power which included, for example, legitimate and illegitimate interests, objective needs and their ideological distortions). In short, the explication of the fundamental notion of power already designates the original relation of power as non-repressive. But neither is the non-repressive element of power characterised in any more detail nor is attention paid to how it is productive, in particular

morals, knowledge and power

31

of knowledge and truth. For Foucault it is quite clear that at this point historical analysis is indispensable. For example, looking at the seventeenth century Foucault proceeds on the basis of the well-known fact that sexual behaviour was standardised and restricted in a multitude of ways, through the frustration of preferences and the threat of sanctions in both the ecclesiastical field and that of secular jurisdiction. But he also points to the explosive multiplication of ‘epistemic’ discourses in which exactly those sexual phenomena and practices which were to be suppressed were investigated (irrespective of whether they were forms of deviant behaviour or fantasies). In the course of the next century this tendency was intensified. The epistemic discourses on sex, sexual behaviour and sexual fantasies increasingly accentuate the relevance and the mystery of every variety of sex, so presenting sex as interesting and both in need of and capable of being analysed. These discourses no longer regard their subject matter, the sanctioned sexual phenomena, as being just existent, but as being ineradicable; and they take the most important variations of sexual behaviour to be essential properties of individuals, in the traditional technical philosophical sense of ‘essential properties’ as defining their identity and personality and being causally relevant for many of their most important characteristics.44 In a still quite superficial sense it could be said (and, in fact, Foucault does say) that the form of power involved here acts repressively, but does not really seek to eradicate the repressed phenomenon. Rather it takes very seriously the necessity of its existence, its relevance, its constitutive role for the personality and its intelligibility. Maybe the fact of taking seriously or acknowledging even constitutive functions is already a weak form of ‘productivity’. For if we join Foucault in denying the crude variant of metaphysical realism, we will have to admit that in non-empirical objectdomains epistemological and theoretical attitudes contribute – perhaps in a way that has not yet been fully clarified – to the ‘existence’ of the phenomena considered. But then we would not have got beyond a philosophical position that is widely acknowledged, and the exact details of the relations between forms of power and epistemic discourses would still remain obscure. How are we to describe methodologically, for instance, the situation that representatives of the church are actively involved in the regulation and repression of sexual practices, but at the same time actively make these practices the subject of epistemic discourses and thereby instantiate

44 On this ‘stimulation to discourses’ see HS 17–35.

32

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

(with ‘weak’ productivity in the sense mentioned above) new forms of power? It may be confusing that on some occasions Foucault quite freely employs causal language in this context, and on other occasions a ‘means– ends’ language, and that the latter operates partly on an intentional and partly on a functional level (there were certain ‘incitements’ to launch particular discourses, but certain ‘goals’ were pursued as well, and all this could ‘have served the development of power’, etc.). But probably one comes closest to Foucault’s intention by saying that persons, as representatives of repressive structures of power, regarded certain epistemic discourses as a means of optimising the intended repression, but that the employment of these means led (causally) to a ‘production’ of the sanctioned phenomena (in the weak sense mentioned above), and hence also to the unintended establishment of a new form of power. The methodological structure of this account is of course a variant of the ‘invisible-hand’ model, and so of another standard action-theoretical model, although perhaps with the qualification that Foucault would have to downplay the ‘causality’ involved in the production of non-intended side-effects in a nominalistic manner (e.g. a` la Davidson). In other words, he would have to separate it from the epistemic relevance of natural laws or social rules. The invisible-hand model still does not offer anything new, but it has the advantage that it permits one to speak about intentions and causality which are compatible with the epistemic contingency of the phenomena in question (e.g. new ‘productive’ forms of power and their connection to epistemic discourses), as is required for Foucault’s technique of analysis. But above all, this model also allows a more detailed enquiry into the power–knowledge interconnection which enriches the standard model in a fruitful and interesting way. However, this can only be recognised through a closer study of individual epistemic practices – for instance the confession ritual, which plays a vital theoretical role in Foucault. In the Christian practice of confession it would seem that in the specific situation of togetherness to which interrogation leads, the pursuit of truth is directly connected with acts of the (repressive) exercise of power. Foucault does not attempt to differentiate exactly how closely, but what we can at least say is that, in the context of the ritual of the Christian confession, interrogation is also pursuit, enquiry is also surveillance, finding out is also raising the alarm and explanation is also the establishment of guilt. The same questions, descriptions and admissions can be construed within the confession ritual in two different ways, or be interpreted as different speech acts – as the search for

morals, knowledge and power

33

and disclosure of the truth, or as the establishment of relations of power. These two levels can no longer be sharply separated and assigned to different modes of speech or ‘discourses’ – as in the ritual of investigation and confession in secular jurisdiction. But it has still not yet been established exactly enough to what extent the confession as a form of truth production ‘is imbued with relations of power’, as Foucault sometimes puts it. First of all the act of confession, as seen by the person who confesses, must be more subtly differentiated. In the act of investigation the enquiry by the person questioning is also pursuit, surveillance and a distribution of guilt which, however, often ends in the granting of absolution and can lead to an approving intimacy. At the same time admission by the penitent is not only a revelation of the truth, but also an admission of guilt and the expectation of damnation and reward – damnation of the act or thought which has been admitted, reward for confessing to it. In other words it is the subjugation of the person confessing, not just to the accepted norms, but also to the competence of the interrogator’s judgement. This interpretation of the knowledge– power connection suggests a performative model: the same utterance can be described performatively as an epistemic assessment and at the same time as an act of submission to a relation of power or an exercise of power. Thus so far two models have been developed for Foucault’s theory of the intrinsic relationship between power and knowledge; a specific ‘invisiblehand’ model and a performative model. Foucault does not specify more precisely the relationship that exists, both on the side of the investigating and of the confessing person, between the level of the production of truth on the one hand, and the level of establishing an individual relation of power and of subjugation on the other. Given the individualistic analysis of the relation of power suggested at the beginning of this chapter, one might suspect a causal relation here. The priest is in a position to induce the penitent to confess the truth because, thanks to the background of given social structures and religious convictions, he has the potential to threaten and constrain: power causally brings about the production of truth, and in the confession ritual it is through the form of the causal relation that it becomes clear that truth does not stand outside power, but is ‘imbued with it’. This description may occasionally be correct, but probably it is often not complete because it is also important to take the intentional aspect of the context into account. Within the framework of the confession ritual the investigating person will often have worthy intentions – he intends to find out the truth and thus save the soul of the penitent. Conversely the penitent submits consciously and

34

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

deliberately to an imperative of truthfulness in order to achieve her own psychic healing and personal deliverance, salvation and forgiveness. On the other hand, the establishment of the power relation, the threat, the coercion and the subjugation within the framework of the confession ritual are rarely intended by the persons involved; they represent an aspect which plays a very minor role from the inner perspective of the ritual, but which is used to describe the confession ritual when it is viewed from an external perspective. This does not rule out the possibility that this (external) perspective can be adopted by the persons involved, but normally this will not be the case. The relationship can be better illustrated using a well-known actiontheoretical situation type: Oedipus killed the beggar at the crossroads, but at the same time the beggar could, from an external perspective, correctly be described as Oedipus’ father, so that had Oedipus known and accepted this description, then the act of killing would have been patricide. Similarly, the priest investigates the conduct and thought of the penitent within the framework of a seriously conducted truth production, yet the confessing person can at the same time be described, against the background of her religious convictions and the prevailing social rules, as a being who is anxious about the salvation of her soul, and who depends on the priest for absolution. If the priest and the penitent were to know and accept this description, then the act of investigation would be the establishment of a relation of power and an act of subjugation. It is the possibility of this description, as well as the possibility that the persons involved accept it, which specifies the form in which in the case of the confession ritual the relations of power and the production of truth seem to be intimately connected. This insight is essentially a counterfactual action-theoretical model of the intrinsic relationship between power and knowledge. What appears as the production of truth when viewed from the internal perspective of a social relationship could also be described from the external perspective as a relation of power. Not just a speech act, as in the performative model, but more generally a whole complex of acts can be described differently, although the differences are defined by a change of perspective and different levels of information. Production of truth in this context, of course, means something very specific. The only methodological rule which plays a part here is the imperative of truthfulness for the confessing person. Furthermore, an important role is played by a precondition which dominated philosophy in the seventeenth and eighteenth centuries, and which today is often regarded as part of the core of the myth of subjectivity – that all persons have a privileged access

morals, knowledge and power

35

to the truth of their own mental states and experiences. Whoever believes this, and pre-supposes that the imperative of truthfulness is fulfilled, must hold the ritual of confession to be a thoroughly reliable procedure in the production of truth. Yet it is obviously decisive for Foucault that not only the intentions geared towards the production of truth which are at work in this procedure, but also the power-relation which is often only visible from the external perspective, are not just side-effects but constitutive factors in this kind of truth production. Hence the relation of power is often not the cause of the production of truth in confession, but is part of the system of rules which – it seems – guides and directs the production of truth in the form of the confession ritual. In so far as this is the case, an intrinsic connection of power and knowledge within the Christian pastoral practice can be discerned directly. Thus a number of types of relations between power and knowledge have been outlined that undoubtedly go beyond conventional standard models, for example a specific form of the invisible-hand model, a performative model, and a counterfactual action-theoretical model. Following up some occasional remarks by Foucault one might be inclined to think that these examples reveal one of the most important aspects of the productive character of power. To put it in short: in the confession ritual power produces truth. But although this description is certainly not wrong, it is clearly superficial. The reason why the power which is at work here seems to be repressive is primarily that it works with sanctions and frustrates the subjugated person. But under such conditions the assertion of its productivity amounts to the trivial thesis that within the framework of the confession ritual repressive mechanisms rely on the possibilities of threats and compulsion to extract true self-descriptions. It would be more interesting to point out, as Foucault does, that the intimate togetherness of confession, and the detailed investigation of sexual activities and fantasies, incite and differentiate rather than repress and obscure these fantasies and activities – in other words, although it is repressive, in this case the relation of power keeps the phenomenon which is to be repressed alive. It even helps it to new vitality, and this despite the fact that power now also tries to subject the mental realm to its control. But this form of the productivity of power is an empirical effect that seems to be linked to particular forms of repressive power in certain circumstances. This account is not particularly helpful in elucidating the productivity of the fundamental, non-repressive power which is the target of Foucault’s analysis. This is not surprising, for the character of non-repressive power has not yet been explicated either.

36

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

Possibly this elucidation cannot be provided by an analysis of the confession ritual within the historical framework of Christian pastoral activity. Presumably those forms of confession would be more suitable which fall outside the scope of repressive religious structures, and which Foucault mentions only very rarely in passing – forms through which we, for example, confess our love to another person, or admit a truth to ourselves that we have long kept hidden away. Here too the confessing person normally follows an imperative of truthfulness and submits to the judgement of a listener, who is important to her – important perhaps in the context of an attempt to lead a good life. In such cases it is by no means misleading to speak of a relation of power which produces truths – but not by means of repressive structures. This relation of power is still asymmetric (and obviously partly reflexive), but is constituted in the context of personal relations between the confessing person and an opposite number who is at least virtually present; and these personal relations are usually specified by conventional performative rules – rules, for instance, which transform a statement into a confession of love or into a confession of a hitherto concealed truth about oneself. Only by, and as a result of, following these rules, can we confess the truth. However, the form of the interconnection between power and knowledge, the character of non-repressive power, and the form of its productivity as revealed by these examples are illustrated by Foucault with reference to discursive practices which appeared historically later than the Christian practice of confession, but which proved to be extremely successful: Foucault labels this intricate complex of practices of discourse scientia sexualis. 3 From a theoretical point of view it is presumably no coincidence that at the beginning of his historical analysis of the scientia sexualis Foucault briefly refers to that other ‘great procedure for producing the truth of sex’ which had evolved in many non-European societies: the ancient ars erotica.45 Viewed superficially, the practices of ars erotica and scientia sexualis are undoubtedly meant to be distinguished from each other, but on closer inspection the ars erotica may well contain aspects which indicate a form of power and a manner of its connection with knowledge that are of systematic interest, and are also apparent within the framework of scientia sexualis. 45 HS 57 f.

morals, knowledge and power

37

The scientia sexualis of the nineteenth century was still aimed at a phenomenon that was regarded as suspicious and problematic, and was to be sanctioned. It was embedded in an imperative of public utility and optimal reproduction which were separated from pleasure. On this basis the scientia sexualis sought to establish a theory which was both explanatory and also partly quantitative, and which was intended to discover mechanisms and structures beneath the surface of various sexual practices and to render these practices causally comprehensible. This does not apply to the ars erotica, the subject of which are sexual practices which are regarded as being the source of sublime sexual lust, and can lead to ecstatic enlightenment. Just like an old ars or techne, the ars erotica is based on ancient experience, and at the epistemic level remains a kind of knowledge of rules that direct sexual practices towards the goal of optimising sexual communication and reaching the height of pleasure. As a knowledge of rules, ars erotica is propositional and can be related unproblematically to a set of true sentences. This knowledge can be articulated in speech, and can therefore be passed on verbally and in writing, as well as being the object of instruction and learning. However, the knowledge which is handed down remains qualitative and reflexive, for it is concerned with the quality and intensity of the pleasure experienced under certain rules within the framework of reflexive self-observation. At the same time, the knowledge involved in the ars erotica, in so far as it is a techne, has a non-propositional aspect in which it is more like a skill, a knowledge how, which is directly expressed in the performance and mastery of sexual practices. This nonpropositional technical knowledge can also be taught and learned – but only through the joint activity and practice of teacher and disciple. Unfortunately Foucault does not address the question of whether the passing on of a technelike the ars erotica establishes or presupposes relations of power, although the question is by no means absurd. Intuitively it is quite appropriate to regard the process of introducing a pupil to a rule-governed practice – especially if this practice is venerable and sophisticated – as a form of disciplining, or regulating, within which the master has the means and the desire to induce the disciple to perform certain forms of sexual practice which the disciple, without the influence of the master, has neither the will nor the ability to carry out. Furthermore, it also makes sense to say that the disciple, once initiated, improves reflexively by her own efforts. This specific relation of the master to the disciple, or of the disciple to herself, obviously fulfils the conditions of an individualistic relation of power as sketched out at the beginning of this chapter. It is equally obvious that this form of power relation is not repressive, because

38

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

the practice it ‘produces’ is not negatively sanctioned but highly valued, and does not lead to frustration, but to the highest communicative and sexual satisfaction of the preferences of all those ‘subjugated’. If it is acceptable to speak of a relation of power in this context, then this relation could initially be explicated by means of a simple additional specification to the account of an individualistic relation of power given above: the power involved consists primarily in the ability to induce other persons to follow certain social rules. This is certainly not a completely satisfactory explication; nevertheless the form of power involved may be specified terminologically with the concept of regulative power. In the case of a technelike the ars erotica, the existence and exercise of regulative power is constitutively connected with the passing on of knowledge of the rules and the practice they regulate. It could probably even be argued that in many cases the exercise of regulative power is a constitutive element of passing them on. Hence regulative power produces the practices which it encourages in a nominalistic sense over and over again. In the case of scientia sexualis, if one considers the ‘means’ by which, according to Foucault, ‘this immense and traditional extortion of the sexual confession came to be constituted in scientific terms’,46 it is soon clear that what is at issue is a catalogue of criteria of scientific rationality like those that are still defended in modern philosophy of science. Scientia sexualis has to satisfy five methodological conditions of adequacy: the data (and also the subjective confessions) are to be objectified in the light of accepted background theories through techniques of observation and interrogation; the central object, that is sex, is assumed to have a comprehensive potential to explain all possible forms of deviant behaviour, and this potential also has a systematising function within the context of the scientia sexualis; sexual mechanisms are posited operating on a level far beneath observable superficial phenomena, and so have to be described in non-empirical terms; a scientific community of sex experts is constituted who, in contrast to the ‘patients’, have relevant background knowledge and a hermeneutical-explanatory capacity at their disposal; and finally, in the domain in which proposed theories are tested in therapies and clinical trials, the employment of normative or moral criteria is rejected, and only the ‘descriptive’ language of the pathologist is allowed.47 To put it in short, it is a question of: 46 HS 65. 47 HS 65–7 characterises these conditions in an unusual and unclear way (as so often when Foucault tries to create his own terminology): (a) clinical codification of the inducement to speak; (b) postulate of a general and diffuse causality; (c) principle of a latency intrinsic to sexuality; (d) method of interpretation; (e) medicalisation of the effects of confession.

morals, knowledge and power

39

– the objectification of the data, – the postulate of causal mechanisms in the field of sex, – the assumption of theoretical depth of hypotheses on sex relative to observable phenomena, – the requirement that background theories be used in the evaluation of data, – the exclusion of normative vocabulary from the science of sex. It goes without saying that Foucault does not want to argue that with the introduction and acceptance of these conditions a science of sex – and therefore sexuality as a scientific subject matter – was constituted, which according to present standards was successful or at least scientifically rational. Rather, from the internal perspective the discursive analysis declares only that the relevant traditional epistemic discourses on the practice of confession in the realm of sexual behaviour in the nineteenth century were suddenly enriched with these demands, suppositions and assumptions. At times Foucault claims that ‘the production of discourses is valorised with a truth value’.48 This mode of expression ought not to be misunderstood. It too is of course a description from the internal perspective: the claims to truth which assert themselves in the valorisation of the production of discourse with truth values are always equivalent to particular methodological rules as they occur historically. Foucault does not commit himself affirmatively to one particular conception of truth, but from the external perspective of present mainstream philosophy it could be argued that Foucault designs his historical studies as if he were using a historically relativised, verificationist conception of truth. Obviously far-reaching philosophical questions emerge at this point, but for the time being it may be noted merely that, in the special case of the scientia sexualis, this was not the first time that the production of discourses on sex was valorised with a truth value. It would be better to say that this had already taken place within the framework of the confession ritual, and that with the establishment of the scientia sexualis the form of this valorisation, that is the form of claims to truth and consequently the form of knowledge, changed dramatically. Because power structures are hardly mentioned, Foucault’s analysis of the scientia sexualis in chapter 3 of HS tends to be read as primarily ‘archaeological’, but a closer look soon reveals that this approach is inappropriate. At a systematic level much depends on recognising just how the intrinsic power structures of the scientia sexualis have changed from those of the confession ritual. For example, it is notable that with the establishment of 48 Foucault 1983, 8.

40

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

a culture of experts there is a new epistemic gap between experts and nonexperts which is structurally different from the relationship between priest and penitent. Thanks to their background knowledge it is the experts, the epistemic inspectors alone, who decide what is to be asked and investigated, and who know why, and with what consequences, it is investigated. And if the phenomenon which is to be analysed is also regarded as mysterious, explanatorily deep and dangerous, then elementary psychological mechanisms on the part of the patient lead to a ‘game of deception and concealment’ as a reaction to the ‘epistemic power’ of the experts. At the same time the phenomenon produces in the experts an increase in the desire for knowledge and in the incentive to observe and control. In his analysis of the confession ritual Foucault had already pointed out that here power conquers new areas and extends its control, for instance to the mental realm, and to the smallest and most secret desires and fantasies. Similarly, the five conditions which are constitutive for the scientia sexualis can be understood as forms of a new and perhaps even broader expansion and sublimation of power. For seen in this way the conditions appear as a bundle of arrangements which further disenfranchise the persons investigated. One of the most important aspects of this disenfranchisement becomes clear if one realises that, according to Foucault’s description, the investigated persons in the scientia sexualis lose the privileged position which they had held within the framework of the confession ritual. There penitents were assumed to have – at least when they followed the central imperative of truthfulness – a privileged relationship of truth to their own mental realm (that is through the authority of the first person). In the scientia sexualis, on the other hand, the content of the admissions is first of all subjected to the interpretation and causal orientation of the experts; it is de-subjectivised and objectified; it is regarded as a problematic indication of deeper causal mechanisms, assigned its place in a comprehensive context of causes and effects and finally described in terms of medical pathology. Even resistance to investigation will be explained theoretically, and the failure of the theory can be attributed to this resistance. One of the most interesting aspects of this transformation is surely the fact that this disenfranchisement at the same time represents the start of the attack on the myth of subjectivity, an attack which today seems to have been brought to a successful conclusion with the total demystification of the myth. But it is far more important not to overlook the fact that the conditions which establish the scientia sexualis at the same time represent the rules and regulations that specify, for instance, how sex should now be talked about,

morals, knowledge and power

41

what kind of object sex is within the framework of the scientia sexualis, which standards of rationality and what evidence decide whether confessions can pass as truths – truths both for the experts who listen and for the patients who observe themselves. According to Foucault, by jointly submitting to these rules, experts and patients produce a new kind of truths. It remains open, and indeed even irrelevant, whether or not these truths are socially sanctioned, and whether or not their production frustrates the preferences of the subjects involved. The ‘subjugation’ to these rules may perhaps take place in different ways, but in the most important cases the future ‘experts’ are induced by the masters to follow these rules within the framework of a collective scientific practice – in this way the decisive scientific community is constituted and reproduced, and in the investigative situation the unsuspecting patients are then induced by the experts to follow these rules and directives. The disciples and patients would certainly never have submitted to this discipline by themselves, nor without this inducement. These relations of power are by no means necessarily, or even usually, repressive. However, they instantiate methodological or operational rules and are anchored in scientific as well as in other institutions, and whether or not they are followed is what makes the distinction between rationality and irrationality, truth and falsity – and in the process the historical contingency of this set of rules is systematically veiled. To be sure, the mechanisms and forms through which these discourses (in the technical sense) come to have the standing they do can be examined – in scientific literature, for instance, or in popular presentations, at congresses and in the educational system. But what is crucial is that the forms of power relation that these discourses represent are quite simply identical with the methods of the production of truth. This form of power is ‘productive’ in the sense that it induces people to follow those rules which open up room and possibilities within which their truths can be produced. This type of power is obviously a specific variant of regulative power: certain persons have the means, opportunity and will to induce other persons to submit to certain rules in their discursive practices which essentially define the possibilities of establishing true sentences, and thereby above all draw a line between rationality and irrationality or reason and madness, as well as between truth and falsity or acceptance and rejection. The ‘inducement’ usually occurs not through explicit prescription, but through common action and practical training in an instructional context. Foucault’s discursive analysis of the confession ritual and the scientia sexualis, together with the references to the ars erotica, seems to seek to delimit this specific

42

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

technological and regulative type of power. The interconnection of this kind of power with deep knowledge in the sense of le savoir is the most intimate that can be imagined, for the historical forms of this specific regulative power are identical with the structures (and so also with the e´nonc´es) of the corresponding historical discourses which aim at the production of truth. Foucault develops this concept of power with reference to very specific ‘truth-producing’ discourses – the discursive practices in which persons are induced to say something about themselves, to admit and to confess. But it is easy to see that this restriction is not necessary. The forms of self-confession and self-observation that Foucault investigates perhaps document the complex interplay of repressive and non-repressive forms of power particularly clearly. But it seems obvious that this regulative power which subjects certain persons to a particular historical machinery for the production of truth is also at work outside the context of the human sciences. The concept of regulative power could obviously be extended so that one could speak of a disciplined subjugation to social rules in general – not only specifically to truth-producing rules – and hence to rules that are central to non-discursive practices. In fact, some of Foucault’s works could indeed be interpreted as developing this more general concept of regulative power. With regard to the general concept of regulative power one could also speak of a more general ‘productivity’ of power; it is the forms or rules to which this type of power subjects the persons of a particular society that open up the space for social action in a very general way in the first place, inasmuch as the forms or rules are constitutive for this action in the usual action-theoretical sense. Yet it does seem to be reasonably clear that Foucault’s interest in general forms of regulative power is secondary and derivative. This is because he is primarily concerned with a history of truth and thought, that is with those transformations which saw the emergence of new truth-producing discourses although, of course, more general forms of regulative power will also be pertinent. At times Foucault describes himself as being mainly concerned with clarifying in particular how certain ‘fields of knowledge’ emerge in the realm of human self-reflection, and this seems to reveal an important aspect of the ‘productivity’ of power: power ‘produces’ fields of knowledge. As far as regulative power is concerned, this description can be easily explained. Obviously this productivity of power ought not to be misunderstood in a strong ontological or even idealistic sense – as if power ‘produced’ domains of objects or ‘worlds’. Rather, if persons accept, or subjugate themselves to, ‘truth-producing’ discursive rules under the influence of regulative power, we should take this to mean simply that persons who have been

morals, knowledge and power

43

disciplined in this way have established a field of knowledge, i.e. a scientific perspective of a particular object-domain. This consideration, which is undoubtedly of great importance for the systematic reconstruction of the analysis of power and Foucault’s programme, should be underpinned with a more exact explication of the concept of regulative power. If (a)–(g) in explication (M*) of the individualistic relation of power49 are acceptable as the set of sufficient conditions for the existence of a power relation between individuals, then a more precise definition of the concept of regulative power could be added to (M*) in the following way: (MR) Person P1 has the regulative power to induce a person P2 to do X if: (1) conditions (a)–(g) in (M*) are fulfilled; (2) X is the act of following a social or methodological rule R; (3) complying with R is constitutive of act X; (4) the form in which person P1 induces person P2 to comply with rule R typically consists in a practice in which R is implicitly included being handed down by P1 to P2 . Clause (3) is intended to provide a differentiation of ‘regulative’ and ‘constitutive’ rules which was suggested by Searle. A rule such as ‘if you want to stay healthy, then don’t eat too much’ is regulative as far as the act of eating is concerned since this act can be carried out and understood independently of following the rule. On the other hand, chess rules are an example of constitutive rules – that is rules which are constitutive for the act of playing chess, since the act cannot be performed without following and (in a certain sense) understanding the rules of chess. Thus constitutive rules open up particular spaces or possibilities for actions (and at the same time delimit the space by implicitly forbidding violations). In the handing down of implicit constitutive rules within the framework of particular practices through the exercise of regulative power, disciplining and the opening up of new liberties are linked in a highly intimate manner. This is how regulative power ‘produces’ possibilities of action and actions, and here the essence of the ‘productivity of power’ becomes apparent. It should be noted that the concept of regulative power as defined in (MR) is still neutral as far as positivity and repression are concerned. But we could consider introducing the following specification:

49 See above, pp. 17–18.

44

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

(MRP) Power in the sense of (M*) or regulative power in the sense of (MR) is positive and non-repressive when: (1) the conditions in (M*) or (MR) are fulfilled; (2) carrying out action X or following rule R furthers the realisation of the desires, needs and interests of P2 by extending the corresponding possibilities of action. (MRR) Power in the sense of (M*) or regulative power in the sense of (MR) is repressive when: (1) the conditions in (M*) or (MR) are fulfilled; (2) carrying out action X or following rule R – frustrates the desires, needs and interests of P2 – or even prevents P2 fulfilling her own desires, needs and interests; (3) the refusal by P2 to carry out action X or follow rule R is sanctioned socially (by the production of feelings of guilt, through judicial punishment or physical injury). Of course these concepts of power can be further differentiated – for example by distinguishing between legitimate and non-legitimate repressive power, between violence, compulsion and influence, or by specifying the means of power (e.g. personal characteristics of P1 such as education or charisma; resources to which P1 has access, such as money or credit, or social elements such as authority or influence; possibilities of threatening physical violence or rational rhetorical abilities). Many of the standard theories of power in fact consist mainly of a classification or hierarchy of the means of power and do not bother to provide a basic explanation of the concept of power itself. Foucault’s analysis, on the other hand, avoids discussing the means of power, but instead has the advantage that it indicates how we can work out basic explanations of the concept of power itself. The semantic distinction between (MRP) and (MRR) perhaps contributes something to the differentiation of positive and negative power, but it is obvious that the concepts of non-repressive and repressive power not only require the difficult explication of desires, needs and interests, but also pre-suppose the subjugation to moral standards which are incompatible with a strict ethical neutrality or relativism – unless one restricts oneself to determining relevant historical standards. But at any rate, the linking of the concept of repressive power to that of regulative power would be one of the hinges connecting the Foucauldian analysis of power to critical theory.50 50 See recently on this subject Sch¨afer (1995), 154–98.

morals, knowledge and power

45

I would like to suggest that Foucault’s analysis of power should be interpreted as being concerned with dissecting and proposing regulative power as the fundamental non-repressive and productive type of power – particularly that specific variant of regulative power that is geared towards the subjugation to methodological, truth-producing rules of discourse. This specific form of regulative power exhibits a general form of epistemic productivity: it opens up space for those speech acts or discursive practices that aim at the production of true sentences. All other forms of this productivity which, as it was shown above, Foucault outlines in his analyses of the confession ritual and the scientia sexualis are only aspects, or partial descriptions, of this more general form – for example the quantitative proliferations of discourses, the recognition of the existence and relevance of the phenomena examined, the causal production of knowledge, the stimulation of fantasies and curiosity, or the establishment of fields of knowledge. Above all, it can be shown that this specific form of regulative power is particularly intimately connected with deep knowledge in the sense of le savoir. All the other kinds of interconnection between power and knowledge which, according to the above, Foucault outlines in his analyses of the confession ritual and the scientia sexualis – e.g. the invisible-hand model, the performative model, the counterfactual actiontheoretical model or the epistemic power of experts over clients – are simply kinds of this most intimate form of interconnection. If this suggestion is acceptable then it would obviously be inaccurate and misleading to follow Foucault in talking of the ‘relations of force’ as the sum of individualistic relations of power in the sense explicated above, as if these relations of force constituted the basis of all forms of power. The asymmetry of every relation of power, even of a non-repressive, regulative, productive relation, doubtless signifies an asymmetrical ‘relation of force’, an imbalance of the possibilities of effect and reaction. But although this structure is necessary, it is in no way sufficient for the form of power that Foucault’s analysis presents as being theoretically fundamental. This is because the existence of social or methodological rules is always conceptually or ‘grammatically’ a precondition of the effectiveness of regulative power. The fundamental form of power which Foucault assumes is therefore not located in a state of nature, devoid of rules and order and in which a brutal war game of interests and forces is being played out, but rather it is to be found within the conceptual framework of a social space. Consequently Foucault’s analysis of power does not have to face the Hobbesian problem of social and political order and it is inappropriate (although suggested by some of his own formulations) to expect a solution to this

46

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

problem from his analysis of power. Rather, in HS Foucault tries to identify a fundamental, non-repressive, productive, ‘regulative’ form of power beyond the traditional repressive, judicial concept. According to Foucault it can be shown that one of the most important variants of this fundamental regulative form of power is connected in a particularly intrinsic way with truth-producing discourses – such that the historical contingency and the historical transformations of these forms of power, of these discourses and their interconnection are comprehensible. Consequently, if Foucault argues that local relations of force can ‘consolidate’ themselves so as to become powerful institutions, social systems or even states, then this does not mean that these entities emerge from a Hobbesian state of nature and play of forces devoid of rules, but only that the social or methodological rules to which the basic form of regulative power is always related can also organise themselves into more extensive and more influential structures. However, even in the context of this more modest problem, the analyses of power and discourse do not appear to me to be intended to be explanatory (even if Foucault expresses himself at times in a way that might suggest this). They do not seek to explain how social orders develop, but to analyse what forms of power exist, how they are connected to the production of truth and how these complexes of truth and power transform themselves historically. They do not claim to be historical or causal explanations, but are structural analyses. However, Foucault appears to take up a remarkably uncompromising attitude on this point: it is not that the analyses of power and discourse simply have no explanatory goals, but at the same time admit that such an explanation might make sense in the framework of other theoretical approaches. Rather for Foucault the possibility of there being fruitful general historical or causal explanations in this context endangers the fundamental contingency of forms of power and knowledge – and therefore his central message of human liberty in the face of apparently all-encompassing complexes of power and knowledge. Hence, for fundamental systematic reasons Foucault would have to reject any demand for an explanatory clarification of the ‘process of consolidation’ by which institutions or states arise. An adequate critique of this point could only be accomplished by discussing these systematic reasons. This completes the treatment of the last two of the four conditions identified above (p. 18) for an adequate interpretation of Foucault’s analysis of power. It has also been shown how the relationship between positive and negative power can be thought of consistently, and to what extent for Foucault positive, productive power can be the fundamental

morals, knowledge and power

47

phenomenon in all networks of power. Furthermore, the central concept of regulative power allows us not only to conceptualise the most intimate interconnection of power and knowledge imaginable, but also to formulate a consistent position on the phenomenon of sub-institutional and institutional or state power. 4 lf one inspects the systematic perspectives that follow from this reading of Foucault’s analysis of power and of discourse, then what is immediately striking is that one of the most brilliant and severe critiques of Foucault’s view of power – expressed some years ago by Honneth – appears in a new light. The general line of the interpretation of Foucault’s view of power which Honneth suggests, and which forms the basis of his criticism, is obvious: Foucault’s analysis of power is reconstructed in terms of the underlying concept of social struggle (as Honneth himself explicitly notes).51 Honneth begins his interpretation with the assertion that within the framework of the Foucauldian genealogy discourses are ‘systems of social knowledge’; that ‘knowledge formations assume the special function of augmenting power’; and that ‘the theory of knowledge becomes the theory of domination’.52 It is only by means of these displacements that, in Honneth’s eyes, Foucault comes close to a theory of society which could be taken seriously. Perhaps one can indeed interpret Foucault in a certain sense as a social theorist, but if the above considerations are correct then Honneth seems to pre-suppose precisely what Foucault wants to avoid – namely that we should conceive of power as domination, or of the analysis of power as a 51 See Honneth (1993), 157. 52 See Honneth (1993), 152–3; more generally chs. 5–6. In the afterword to his influential and stimulating work Honneth does not retract anything substantial in his critique. Some of the questions concerning the systematic reconstruction of Foucault’s analysis of power discussed above are touched on by Taylor (1985 and 1992, 210–20), but unfortunately without any detailed discussion. Taylor considers for instance the nominalistic lack of a subject of power, and various models of its relation to human agents (including the invisible-hand model; he also suggests, at least at one point (175/211), that there might be an intrinsic relation between power and knowledge. But on the whole he seems to be of the opinion that no systematically consistent model of power is apparent in Foucault. Yet, like Honneth, Taylor mistakenly starts out with the two premises that Foucault relates strong explanatory claims to his analysis of power (patterns of explanation are discussed throughout part iii) and that Foucault’s concept of power ‘is closely linked to domination’ (174/220). These premises also induce Taylor to address inappropriate questions to Foucault. In particular Rouse’s (1994) criticism of this is valid.

48

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

theory of domination; that we should subsume the systematic positioning of discourses or social systems of knowledge to the aspect of their function in the augmentation of power; and that we should explain discourse as a ‘social system of knowledge’ and thereby cloud rather than clarify its specificity. In the light of these pre-suppositions it is hardly surprising that Honneth claims that the relation between power and knowledge is ‘not very clear’ in Foucault. Even more significant, and – at least from the perspective of the above reflections – more problematic is Honneth’s interpretation of power (in Foucault) as a social struggle of interests. It is easy to see that this assumption leads to a number of problems: – commonly accepted norms and moral orientations appear to be mere cultural deceptions for Foucault, with no systematic place in the framework of the theory; – the various kinds of regulation (e.g. discipline, control) appear to be mere instruments used by power; – regulation as a decisive means employed by modern power can only be conceived of as a way of imposing a repressive, rigid uniformity, and cannot be connected with the productivity of power in an interesting way; – the assumption that knowledge serves power functionally leaves no systematic space for the orientation of scientific endeavour towards objectivity and truth; – the stabilisation of purely local relations of power into institutions remains implausible and unexplained; – the explanation of the way in which discourses function to increase power unexpectedly appeals to institutions, which contradicts the fundamental role of the local theory of power; – the analysis of power oscillates unstably and unclearly between an actiontheoretical and a system-theoretical approach. These are the essential points of criticism which Honneth brings against Foucault’s analysis of power, and undoubtedly one should concede that the phrasing of some of Foucault’s texts definitely fits Honneth’s interpretation, and is reason enough to raise critical questions. But on the other hand, it is clear that throughout Honneth assumes a repressive concept of power, and does not seriously attempt to determine the productive character of the Foucauldian concept. If one bears in mind the concept of regulative power and the level of forms of power, it is immediately recognisable that norms and forms of regulation are an integral part of power

morals, knowledge and power

49

which at the same time clarify its productive character: making human behaviour conform to given social rules, government, produces particular actions and freedoms, but always pre-supposes the existence of systems of social rules. As has already been suggested above, from Foucault’s perspective the (typical) question (of origin), that is how social rules and institutions that ensure societal stability can arise from conflicts of interest, is hardly systematically appropriate. He is more interested in seeing how through conformity to systems of rules certain actions become possible in the first place, what kinds of systems of rules these are, and what ruptures and transformations are historically observable. One might perhaps ask how sub-institutional and sub-state forms of regulation relate to the institutions and apparatuses of the state, but for Foucault this is not a central systematic question. Although the discursive analysis occasionally recognises the existence of causal mechanisms in this context, it remains sceptical about whether they can be discovered, and whether this would be fruitful. The discursive analysis is primarily interested in other aspects of the power–knowledge complex and it is, at the very least, misleading to specify this question as being one about the relation between struggles of interest and institutions. Thus when Foucault maintains that the ‘moving substrate of force relations’ is the basis of all relations of power, then this does not mean – in terms of regulative power – that the rules and institutions coordinating action arise only from struggles of interest in the first place, and that the analysis of power would have to explain this. Rather, force relations ensure conformity to the respective given social rules, and trivially such rules will continue to hold only when they are thus enforced. But at the same time these rules are productive in opening up and defining new spaces for action. In this respect it is not accurate to separate the conception of power in Foucault derived from the theory of force as ‘a network of antagonistic force relations’ from the allegedly later ‘action-theoretical’ conception of power as a ‘structuring of possibilities of action’.53 Rather, it has to be assumed that both these apparent alternatives can and must be considered together. Social rules can ‘consolidate’ themselves so as to become institutions (institutions are standardly construed as systems of rules enforced by sanctions), but it must be remembered – and this is Foucault’s main concern – that institutions and apparatuses of state are also based on sub-institutional rules and their enforcement in relations of force. What is more, as we have seen, it is incorrect to attribute to Foucault the view that discourses within which truth values are assigned 53 See Fink-Eitel (1978).

50

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

are to be understood exclusively as functioning to preserve or intensify power. Yet this attribution is the only support that can be cited for the rationalist reservations that have been expressed about Foucault’s alleged system-theoretical debasement of truth and objectivity. One of the historical forms of power is rather the conformity to particular standards of rationality and thereby the production of particular objective truths – even if this is also part of a global strategy of power. What is more, this often takes place within the framework of institutions. But this should not conceal the fact that here too the fundamental regulative power is always at work, more or less clearly and undiluted and that, according to Foucault, this form of power is becoming increasingly prominent in the institutions of modern societies. At last we can see that Foucault’s supposed oscillation between theory of action and system theory is merely the consequence of a particular reading of his texts. To get closer to Foucault’s thoughts one must presumably see, and also specify, that and to what extent the analyses of discourse and power are neither action-theoretically, nor system-theoretically coloured, as I have attempted to demonstrate. One of the positive aspects of the interpretation that I have suggested is that this account of the fundamental concept of regulative power may allow us, in the context of Foucault’s later work, to connect the analysis of ancient techniques of the self and modern ‘political technologies of individuals’ to the analysis of power and the archaeological analysis of discourse. Undoubtedly in Foucault’s eyes these domains see a considerable shift in the theoretical questions and problems involved in the analysis of power; but it is more than doubtful whether this theoretical development can be totally separated from the perspective of the analysis of power, and whether this would be desirable from a systematic point of view. If one considers, for instance, Foucault’s remarks on the close interconnection of subjectivisation and objectification, and also assumes that for Foucault the formation of the subject includes the twin aspect of subjugation and reflexivity (see above pp. 9–10, 27), then subjectivisation and objectification can be placed on three different levels: – the individual person submits herself to the rules of the games of truth, and thereby experiences herself as a being that observes itself and decodes. In this way the subject becomes a scientific object; – the individual person submits herself to singular, asymmetrical relations of power, their global strategies and disciplining, and experiences herself reflexively as an active (overcoming resistance) or at least passive

morals, knowledge and power

51

(resisting) being. In this way the subject at the same time becomes an object of standardising bodily discipline; – the individual person submits herself voluntarily to the practical techniques of a good life committed to happiness and perfection, and thereby experiences herself as a being that examines and cares for itself and organises its own self-transformation. In this way the subject becomes the object of a self-transformation through practical techniques. These three levels address the epistemic, the dynamic and the ethical formation of both subject and object, and these can at first sight be attributed to the three Foucauldian axes of archaeology, genealogy and ethics. But on a closer look it is evident that these three levels are no more than an artificial distinction. This becomes particularly clear when we consider that relations of power can also play an important role in the epistemic and ethical formation of subjects and objects, and that conforming to games of truth will be important for the dynamic and ethical formation of subjects and objects. Furthermore it is conceivable that conforming to games of truth that involve the human sciences (for example psychoanalysis), as well as conforming to certain bodily disciplines, is often also linked to a technique of the self or a self-transformation of the subjects. To be sure, Foucault’s general criteria for subjectivisation and objectification are not only fulfilled in the realm of ethics, but are also valid for the context of power and knowledge. However, at the same time the archaeological, genealogical and ethical axes are intertwined in a particularly intimate manner in all three realms. This complex interconnection can be plausibly reconstructed from the analytical perspective of the concept of regulative power.54 For mechanisms of regulative power which can be attributed to the realm of pedagogical disciplining are effective not only in the epistemic, but also in the dynamic and ethical, formation of subjects and objects. Young persons must be made to follow the rules of science if they wish to become scientists, the rules of political strategy if they want to become politicians or the rules of ethics if they wish to develop excellence of character and live a happy life. However, this subjugation is productive: it opens scope for the production of further statements that define fields of knowledge, of 54 Foucault’s increasing tendency to see his various works and ‘programmes’ from the consistent standpoint of the ‘history of truth’, i.e. of particular processes of subjectivisation and objectification in the variants of which different forms of power are effective, is particularly apparent in Martin, Gutman and Hutton (1988) and his Autobiography.

52

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

political decrees that delimit spheres of power and of moral recommendations that structure tested models of a good life. What is more, the subjects who produce these statements, decrees and recommendations can, at the same time, be their addressees. But although it may be possible to separate the three forms of regulative power and the three corresponding types of discursive production analytically, they can, and often are, closely intertwined historically. The rehearsing and learning of scientific methods and games of truth can at times be part of the moral formation of the subject, and may pre-suppose the regulative enforcement of political power relations; or, vice versa, this enforcement may rely on conformity to particular games of truth and characterise the regulative model in dealing with desires; and practical techniques of the self which are intended to produce a self-transformation may themselves represent a particular game of truth, and define the heart of acceptable political behaviour. This interconnection is possible, indeed likely, for these are related forms of the mechanisms of regulative power that contribute, with their different orientations, to a comprehensive context of disciplining. This connection also seems to be emphasised in the last area of research that Foucault dealt with shortly before his death – that is the analysis of the ‘political technologies of individuals’. Foucault no longer had the time to present any detailed historical studies on this subject, but his remarks seem to indicate that he wanted to investigate how the modern idea of reason of state developed against the background of the formation of a ‘police science’ which was concerned with producing a ‘technology of administration’.55 This technology had the task ‘of promoting civil respect and public morality’ in order to ensure law and order. According to this conception, the police was supposed to monitor closely the relations of people to each other, and to encourage them to lead a life that would make them happy – but primarily in order to construct a strong, powerful state capable of development, since the happiness of the subjects was considered to be an important element in this. This new ‘political rationality’ prescribes that the police intervenes in a positive way to organise individuals’ lives – it has to ‘regulate’ social life so as to produce something new that promotes the happiness of the individual and the strength of the state. Foucault explicitly emphasises that this positive government does not employ arms and laws, or protection and prohibition. It is immediately apparent that the power which the police is meant to exercise according to this new political rationality is a variant of regulative 55 See Martin, Gutman and Hutton (1988), 16–49 and 145–62.

morals, knowledge and power

53

power which is, however, now again connected with specific ‘political’ goals, that is with goals that belong to a sphere of power which stands above the individuals. It is no accident that under the heading ‘bio-politics’ Foucault returns in this context to phenomena (for instance population policy) which had already been objects of study in the analysis of power. And it is even less of an accident that the ideas of the ‘new’ political rationality about ‘beneficial’ administration show a strikingly close connection to central perspectives of the political philosophy of classical antiquity – for instance that it is the goal of political science to promote the state (and not the prince or the individual subject); that ruling requires a particular competence, a specific ‘political’ knowledge; that every state (every polis) is in permanent competition with other states (poleis), and that attempts to reach agreement between them cannot prevent this; that every state represents an end in itself which stands in a complicated relation to the happiness of the male citizens (female citizens did not count); and that in general one of the tasks of the state police is to promote public morality and human happiness (a dramatic deviation from classical liberal ideas). Of course Foucault noticed how close modern technologies and ‘police science’ were to central ideas of ancient political philosophy; but he is of the opinion that under the surface of this common ground something new is in the air in the modern age. He emphasises, for instance, that in modern ‘police science’ people and their happiness are only relevant in their social relations, that is as building blocks of a strong state. These remarks are not very convincing, but even if Foucault is right there seem to be a number of interesting features in the political technologies of individuals which strongly resemble ancient ideas. Thus under the heading ‘political technology’ it could be helpful to investigate the ways in which in antiquity individual technologies of the self, epistemological games of truth and the implementation of political technologies were related to each other, and so how the investigation of ethical work can be aligned with archaeology and the analysis of power. It should by now be clear that the concept of regulative power should also play an important role in this context. Seen from the perspective of this reading of the analyses of power and discourse, some of the most important critical questions concerning Foucault’s programme can be put differently from the way they are formulated in much of the current discussion about Foucault. It should be obvious, for instance, that Foucault did not seek to develop a societal theory with explanatory pretensions. But above all we should not assign too much systematic significance to the controversial ‘neutrality’

54

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

of the analyses of discourse and power. Rather, we should regard it as a provocative kind of intellectual coquetry that seeks to oppose a form of moral and rational universalism which presumably nobody any longer seriously proposes. Obviously not even Foucault, as the analyst of discourse and power, can look through God’s eyes, and the analyses of discourse and power are developed from a particular perspective which they cannot leave. This means that not even Foucault himself – or anybody else who carries out interesting cultural analyses – would be able to specify the exact point which identified his view as a specific perspective. Rather their own perspective is, for Foucault as for every other writer, always the universal perspective – universal in the sense that all other specific perspectives must be translatable into it, and Foucault is particularly successful in translating different problematisations, trains of thought or practices into the theoretical and conceptual framework of his own analytical approach. At the same time it has to be assumed that ‘our’ general perspective is not immune to revisions, but remains fallible, and it cannot be ruled out that the study of other historical forms and perspectives, studies of the kind Foucault pursued, could provide good reasons for such a revision. The same holds in particular for the games of truth. Foucault can perhaps show that the historical concepts of rationality, that is the historical forms of ‘games of truth’, differ significantly from the present form (at least as far as the standards of logically valid argumentation or the careful presentation of evidence for historical analyses are concerned); and he can perhaps cite good reasons to doubt that these forms have passed through an evolutionary historical development towards the present form. But at the same time it can, and has, to be conceded that the dominant current form of games of truth is also central to Foucault’s own analyses of power and discourse. It is perfectly consistent with the analyses proposed by Foucault to apply our standards of morality and rationality, those that we have good reasons to defend and therefore can and must regard for the time being as universal, to other historical forms of morality and reason and to critically assess these historical forms from the perspective of the present. But are our standards of morality and reason the general standards? We have good reasons to assume this, and we should continue searching for even better reasons. Yet this does not rule out that our standards may change together with the good reasons (perhaps under the impression of historical analyses or intercultural discussions) – but to an extent that nobody can foresee. This is because at the same time we also have good reasons to assume that our standards of reason and rationality are not immune to revision, and will remain

morals, knowledge and power

55

fallible. Studies of historical forms of games of truth certainly provide some of the reasons for this assumption, and maybe even some indications for a possible revision of our standards. Neutrality and relativism with respect to standards of morality and reason cannot be defended consistently as a philosophical position, but this does not imply a naive universalism. Furthermore, in the specific case of the Foucauldian programme neutrality and relativism would directly contradict his general approach, because the God’s-eye view, which is a precondition of the neutral attitude, is itself evidently incompatible with a sensitivity to the specific nature of the historical forms of moral and epistemic ‘games’. Therefore the problem of neutrality is better played down in discussions of Foucault.56 The analyses of discourse and power may perhaps lose some of their fashionable appeal, but nothing of their cognitive content, if their neutral distance is given up. Instead other questions arise. If neutrality is abandoned, and if the analyses of power and of discourse 56 Foucault’s thesis of neutrality is the element of his thought that is most disturbing for Charles Taylor (1985), 174–8/(1992), 220–39 as for many scholars of Foucault (on this subject see the recent study by Sch¨afer (1995), 103–53. Sch¨afer’s list of objections to Foucault’s ‘antinormativism’ (see 103–5) does not take into account the strongest lines of argument against relativism which are based on an extensional theory of meaning). Taylor’s analysis of Foucault exemplifies the extent to which an overtly strong (and probably unjustified) systematic evaluation of the problem of neutrality threatens to thwart a more detailed reconstruction of other important intuitions by Foucault. Gutting (1989), 262 ff., points out (with good reasons, it seems to me) that Foucault in no way advocated a general relativism because – Gutting argues – Foucault’s analyses are historically always extremely local, and he would never hold a universalistic thesis (not even a relativistic one). What is more, he concerns himself only with ‘dubious’ human sciences and seems to accept the objective truth of the hard natural sciences, and objective truth is not even generally denied for the human sciences. Finally, despite all the admiration for Nietzsche there is no evidence that Foucault has adopted the latter’s extreme perspectivism (as Gutting understands it). Of course Gutting’s remark that the theory-ladenness of acts and the connection of theories to power do not provide a sufficient basis for a relativistic position should be taken into account as well (with regard to Nietzsche, some recent works point out that his philosophy, and especially his concept of ‘genealogy’, perhaps need not involve perspectivism at all: see for instance Geuss (1994), 274–92). Rouse (1994) attempts to counter concerns about Foucault’s supposed neutrality or relativism with the thesis that Foucault was not concerned with formulating true insights into human society or a normative critique of political conditions – and that it is therefore inappropriate to have such expectations of him. Rather, according to Rouse, Foucault wants to present a new interpretation of what it actually means to formulate claims to truth or political criticism. Doubtless Rouse is right in accusing some of the most prominent critics of Foucault of themselves assuming precisely what Foucault attacks when they propose that for normative and relative standards there is either a non-partisan, God’s-eye view, or else that complete relativity holds. Rouse, however, does not discuss exactly what it would mean to give up a God’s-eye view nor to what extent Foucault’s own analyses evade particular (and in some cases local) rational and normative standards.

56

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

regain the confidence to make their own evaluative judgements, then presumably the relation of the Foucauldian programme to traditional critical theory would have to be considered anew.57 In particular the relation of the specific methodological to the general social form of power should be examined more closely, or perhaps reinterpreted in the light of Foucault’s historical analyses. Important questions are also raised by one of the central messages implicit to Foucault’s work, the message that the possible extent of the difference between our and other conceptions of morality, reason, scientific rationality and other central ideas – and therefore the extent of the particularity of our ideas about these things – must and can to a great extent be determined by means of historical and cultural analyses, even if these analyses themselves remain committed to our ideas. In this context we must ask not only how we can determine and assess the extent of these differences, but above all what it means to assert such differences. Foucault does not always manage to demonstrate that the great ruptures and transformations which he describes reveal a degree of cultural difference that goes beyond mere differences of opinion. But even if Foucault were right, the very notion of sweeping cultural or scientific transformations is quite vague, as the discussion of Thomas Kuhn’s work has clearly shown. One of the strongest and most important arguments that has been put forward against a neutralistic relativism, regardless of whether it is meant coquettishly or is seriously defended, holds that the very possibility of successfully understanding even extremely strange ideas and problems – a comprehensibility and interpretability that nobody presumes more decidedly than Foucault – implies a large number of common beliefs and shared forms of morality and reason.58 But where exactly is the boundary between what is common and what is different, and in what terms should this boundary be described? It should be plain that these questions cannot be discussed without reference to that great debate which currently plays a dominant part in the philosophical scene, and which comes under the heading ‘theory of meaning’. Questions of similar importance follow from Foucault’s characterisation of depth structures of power and of knowledge, whose relation to each other in the context of their historical forms is decisive for the messages of the analyses of power and of discourse. Clearly Foucault dissociates himself both from a philosophy of the subject and from theories of 57 See in particular McCarthy (1990) and (1991). 58 See Davidson (1993).

morals, knowledge and power

57

ontological constitution, but then the characterisation of depth structures becomes even more of a notorious problem – not only because it is difficult to determine the historical forms of these depth structures, but above all because it is difficult to determine what exactly depth could mean in this context. Thus, from a systematic point of view it would be interesting and helpful to consider Foucault’s analyses not so much from the abstract point of view of neutrality and relativism, but rather to see how far his analyses are fruitful for, and can be connected to, the current debate about problems of meaning, interpretation, ‘grammatical’ structures and the distinction of scientific from everyday cultures.

2

THE ETHICAL TELEOLOGY

Foucault begins his analysis of classical ancient philosophy by looking at the ethical discourses that deal with sexual pleasure and its problematisation. This he does on the same four levels that he had previously identified as the central aspects of every ethical investigation – ethical substance, the mode of subjugation, the teleology and ethical work (i.e. the practice of the self that the subjects must be engaged in to transform themselves into moral persons).1 The textual basis he employs consists almost exclusively of passages from the works of Plato, Aristotle and Xenophon. The main results of his analyses of these texts can be summarised as follows. The crucial passages studied by Foucault identify the aphrodosia – the dynamic of the interconnection of desires, sexual activity and pleasure which threatens to exceed all confines and destroy all order and reason – as the central subject of moral concern, that is as the ethical substance. Two main parameters are at the heart of the analysis of this dynamic: the quantitative degree of activity and the sexual polarity. It is not the forms of love and its practices that are important for the moral problematisation of pleasure, but the number and frequency of sexual acts, as well as the differentiation between the subject and the object of appetite – that is 1 UP part 1, 33–93; at the beginning of this part (UP 37) the sub-division of Foucault’s four levels suggested in ch. 3 of the introduction (UP 25–32) is explicitly applied to the first four chapters of the first part.

58

ethical teleology

59

matters of moderation and greed, and of activity and passivity, in both heterosexual and homoerotic relationships. Appetite and pleasure are considered to be natural and necessary, and so are not intrinsically bad. Nevertheless, they are a source of concern as they belong to the field of lower activity which humans share with animals, and they are based on forces which have an inherent tendency to excess and to bursting all bounds. Thus, in the use (chresis) of pleasure it is not so much what is allowed or forbidden that is important, but a control and prudence that rely on a rational calculation to determine the degree of yielding, as well as the correct time for the activity, while at the same time taking into account the status of the actors involved. This is the manner in which the robust dynamic of sexuality is to be subjugated, while at the same time it is accepted as necessary and natural. But this subjugation must take the form of an agonal struggle with herself on the part of the subject – for there is no other way of taming the intrinsic immodesty of sexuality. The subject’s soul has several parts, exhibiting a heautocratic structure; the internal struggle implies a self-reference which should end in victory, and lead to the settled state (hexis) of continence (enkrateia), although this must repeatedly be fought for. Domination and mastery are the form of ethical work which classical authors recommend in this context, not investigation and decoding as in Christianity. This form of ethical work has a clear end (telos): temperance (sophrosyne) that is a stable disposition of the control of pleasure, freedom from the enslavement to pleasure, the leadership of reason (logos) and prudence (phronesis), reflexive knowledge and self-control that also legitimates power over others. At the same time this emphasises the ‘virile’, i.e. active, aspect of the ethical teleology, which gains an ‘aesthetics of existence’ inasmuch as it takes on ‘the brilliance of a beauty that was revealed to those able to behold it or keep its memory present in mind’.2 A further, far-reaching consequence is deduced from this characterisation of the teleology: Putting it schematically, we could say that classical antiquity’s moral reflection concerning the pleasures was not directed towards a codification of acts, nor toward a hermeneutics of the subjects, but toward a stylisation of attitudes and an aesthetics of existence.3

It is conspicuous that in the context of this diagnosis Foucault almost completely ignores the intimate relationship between subjectivisation, 2 UP 89. 3 UP 92.

60

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

objectification and power – even when a discussion of this relationship seems to be almost inevitable, for example in the domain of teleology. In this domain, sophrosyne is conceptualised as the result of agonistic ethical work and moral subjectivisation, which at the same time is based on knowledge of the self and practical prudence as forms of a particular game of truth, and is an essential precondition of political power. This deficit is of particular systematic impact as aphrodisia, chresis and enkrateia can hardly be adequately described, and certainly not be comprehended, independently of the teleology. It is the teleology as the model of a good life that determines the central concern about sexuality, the main strategies for reducing this concern and the way in which these strategies are carried out. But Foucault does not focus on this significant approach since he reflects neither systematically, nor by using examples from ancient texts, upon the relationship and the connection between the four different levels of ethical discourse. It is this restricted perspective of the early ethics programme that allows him to trim the ancient teleology into an aesthetics of existence in the first place – an aspect which in many interpretations is quite incorrectly regarded as the core of his programme of ethics. A more precise presentation of how the teleologies which Foucault deals with are presented in the classical sources could help to redress this deficit and to provide a historical basis for the later, extended – and from a systematic point of view more interesting – perspective of the Foucauldian ethics. I intend to expound this using Aristotle’s ethics as an example. First of all I shall (1) present an outline of the ‘teleology’ which attempts to do justice to the complexity of this concept within the framework of Aristotle’s ethics, before (2) proceeding to prove the intrinsic interconnection of this teleology with domains of knowledge and a specific form of ‘games of truth’, and then (3) investigating how, for Aristotle, ethical activity was closely linked to political demands, requirements and relations of power, and (4) how the central concept of ethical work or moral subjectivisation can be more precisely differentiated, not least by considering various forms of regulative power. 1 Foucault is quite right to emphasise that ‘[t]he desires that led to the aphrodisia . . . and the pleasures that could be obtained from the aphrodisia had their cause, according to Aristotle, in necessary things that concerned the body and the life of the body in general’, and as such are natural

ethical teleology

61

and important.4 But Foucault does not take into account the fact that Aristotle embedded reflection on the function of aphrodisia, that is of matters concerned with sexual activity,5 systematically into his metaphysics and ethics. As for the metaphysical background to this reflection, Aristotle emphasises that in the context of aphrodisia all humans experience a pleasure (hedone) of extraordinary intensity (if not always in the manner they should), and that there is a scientific explanation for this intensity that reveals the exceptional tension connected with sexual activity.6 Of necessity desire (epithymia) is connected to this pleasure, for epithymia can be defined as pursuit (orexis) of the pleasurable – it is a deficiency (endeia) in pleasure, or – more precisely – it is caused by this deficiency, and seeks the ‘replenishment (anaplerosis) of this deficiency’.7 Thus, together with pleasure and desire, the aphrodisia form a unity the necessity of which can be described and explained scientifically. Nevertheless, it is instructive that Aristotle presents the core of this explanation in his tract De Generatione Animalium, and thereby indicates quite clearly its deep, metaphysical background. According to Aristotle, the function and the necessity of the unity of aphrodisia, pleasure and desire belongs in the context of the perpetual reproduction of natural species, which in turn refers to the highest end towards which all animals, as well as humans, naturally strive. Procreation ensures the survival of the natural species, which according to Aristotle’s metaphysics are paradigmatic cases of ousiai, that is of those entities which exist to the highest degree.8 It is precisely because they belong to an eternal and natural species that each and every mortal, and finite, animal can participate in the eternal and the divine which, in turn, is the highest end of all living creatures.9 However, the perpetual cycle of birth and death is itself embedded in the most universal of structures, and the deepest of dynamics, which rule the cosmos. For the cosmos, as Aristotle sees it, is ruled by perpetual cycles of which each stage is ‘absolutely necessary’ (anankaion haplos), and which exist partly in the field of perishable things, 4 UP 48, with reference to Ar. NE vii 6, 1147 b23–7, where no more than this is stated. 5 In spite of all the difficulties of translation to which Foucault refers in UP 35–7, the aphrodisia are those aspects of sexual intercourse which are related to the sexual act in a closer sense – conception, penetration, orgasm in men and women. 6 Ar. NE vii 14, together with GA i 18, 723 b3–724 a3. 7 See Ar. An. iii 11, 434 a3; Top. vi 3, 140 b27; An. ii 3, 414 b5 et passim; Pseud.-Ar. PP 930 a28; Ar. NE iii 13, 1118 b19. 8 Ar. An. ii 4, 415 a23–8 and Met. Z. 9 Ar. An. ii 4, 415 a29–b9; see on this Detel (1997a).

62

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

but above all in the field of the perpetual individual things, in other words the divine stars. The perpetual cycles of perishable things, for example the cycle of birth and death, ‘imitate’ the perpetual, mathematically exact changes of the heavenly spheres and try to emulate them as best they can. And vice versa, it is the exact dynamic of divine motion, with the ‘immovable mover’ as it deepest fundament, that is the scientific cause for the perpetual cycle of mortal beings.10 Aristotle identifies pleasure in a very general manner: animals and humans all seek pleasure in the same way,11 and essentially pleasure is a perceptible process towards a natural state, that is a motion through which the soul (in humans consciously) is brought to its natural state.12 Accordingly, desire (epithymia) as a pursuit of pleasure must be an aspect of the will to live.13 Thus, Aristotle felt that according to its function the scientifically observable unity of aphrodisia, pleasure and desire, a unity that so powerfully drives all animals to procreation, is indeed embedded in the deepest cosmic mechanisms. It is only against this background that its naturalness and necessity can be fully understood. From Aristotle’s standpoint embedding the dynamic of aphrodisia, pleasure and desire in the highest ends of all animals, and in the deepest mechanisms of the cosmos, is incompatible with the assumption that this dynamic is intrinsically destructive and a threat to life, for he certainly did not think that the deepest cosmic mechanisms were destructive. So we have a first indication that Foucault failed in his attempt to categorise the function of desire, sex and pleasure adequately. There are many other indications which point in the same direction when we start to consider that aspect of the Aristotelian ethics which Foucault calls ‘teleology’, and which he characterises as the settled state of temperance (sophrosyne). It is important to see that Foucault interprets the teleology in terms of the same model of domination and restriction that is to be found throughout his entire interpretation of classical Greek medical and philosophical texts: sophrosyne is a stable disposition of the control of pleasure and the self. It establishes the leading role of reason (logos) and prudence (phronesis), but this must continually be achieved and secured anew in the agonistic struggle of the parts of the soul, for only so can it lead to freedom, to being free of enslavement of the pleasures: 10 See Ar. GC ii 11; GA ii 1, 731 b24–732 a2; Phys. ii 9, 192 a16–18; GC ii 10, 336 b25–337 a8; Phys. viii 10. 11 NE vii 14, 1153 b24–6. 12 NE vii 12, 1152 b12–14; Rhet. i 11, 1369 b32 ff. 13 NE x 2, 1172 b35 ff.; 4, 1175 a10–21; ix 9, 1170 a19.

ethical teleology

63

Sophrosyne was a state that could be approached through the exercise of self-mastery and through restraint in the practice of pleasures; it was characterized as a freedom . . . Its polar opposite was . . . an enslavement – the enslavement of the self by oneself. To be free in relation to pleasures was to be free of their authority; it was not to be their slave . . . What one must aim for in the agonistic contest with oneself, and in the struggle to control the desires was the point where the relationship with oneself would become isomorphic with the relationship of domination, hierarchy, and authority that one expected, as a man, a free man, to establish over his inferiors.14

It is apparent that on this interpretation the aspects of the teleology (sophrosyne) and of the form of the ethical work (enkrateia) are intrinsically interconnected: for Foucault understands enkrateia, the struggle against and victory over the pleasures, as a central feature of moderation and temperance as a teleological disposition. However, Aristotle carefully distinguishes enkrateia and sophrosyne in the systematic framework of the ethical virtues, and relates them precisely to each other. Ethical virtue is directed at the field of irrational feelings. It is a settled basic disposition, and in action and choice its aim is the mean; for example, with respect to feelings of fear and trust, courage is a mean state of character between cowardliness and recklessness. But this talk of the mean state of character does not imply, as Foucault and many other Aristotelian scholars claim, a doctrine of permanent moderation, but rather being capable of reacting appropriately to individual situations. According to Aristotle this mean state of character does not exclude the possibility of hefty reactions when the situation demands them.15 The parameters on which Aristotle’s concept of ethical virtue is based allow us to differentiate four typical states of character: – Virtue (arete) is a mean state of character with respect to feelings, actions and choice; it consists of a stable basic disposition to choose and do with pleasure what is right. – Continence (enkrateia) is a mean state of character with respect to actions and choice, but not feelings; it consists of a stable basic disposition to choose and do what is right, but with pain and effort. – Incontinence (akrasia) is a mean state of character with respect only to choice, not actions and feelings; it consists of a stable basic disposition to choose what is right, but to do what is wrong, and this with pain. 14 UP 78, 79, 83. 15 See e.g. Ar. NE ii 6, 1106 b8–28; ii 9, 1109 a28.

64

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

– Finally, badness (kakia) is not a mean state of character in any respect; it consists of a stable basic disposition always to choose and do with pleasure what is wrong. This rough characterisation demonstrates the difference between virtue and temperance (arete and sophrosyne) on the one hand and continence and agonistic victory (enkrateia) on the other. The virtuous person (in the ethical sense) experiences pleasure and joy in choosing and carrying out actions which comply with a mean state of character, and therefore focuses her feelings on this mean. However, the person who has to keep her self under control must force her self (and is also capable of doing so) to choose and act in accordance with the mean state of character; in doing so she experiences no joy or pleasure, but pain and disquiet. Aristotle does not regard the latter as virtuous and prudent; to use a Foucauldian concept, the manner of subjugation, the practice of the self as victorious, agonistic enkrateia, must be clearly distinguished from the telos of sophrosyne. For Aristotle enkrateia is a deficient, sophrosyne a perfect basic disposition. Of course this view does not mean that in the field of teleology there is no longer any practice of the self or ethical work. But what is clear is that the model of domination and restriction of enkrateia as a deficient ethical state cannot easily be applied to ethical work within the framework of teleology, as Foucault attempts to do. Seen from the perspective of Aristotle’s texts, Foucault underestimates the central concept of ethical work and assigns it an incorrect systematic role. This is because he does not locate ethical work in the telos of a good life, but in the deficient state of enkrateia, and thus improperly generalises the model of domination and restriction. This in turn reveals another question that Foucault cannot even formulate within the concept of his interpretation – that is the question of what specific form Aristotle thinks that the ethical work that leads from enkrateia to sophrosyne, from self-control to virtue, has. This question can only be appropriately addressed, and the teleology viewed in a genuinely Aristotelian sense, if we free ourselves of Foucault’s focus on pleasures in the realm of sexuality and nutrition. For Aristotle approaches this problem from a remarkably general and subtly differentiated angle; for example, he feels that virtue and continence – each in its own way – generate a relationship to particular pleasures (in the case of continence a relationship of domination), but are at the same time accompanied by pleasure or pain which are not the object but the effect of self-restraint. This already suffices to indicate that as far as the teleology of the ethical life is concerned, Aristotle sees various forms of pleasure,

ethical teleology

65

while Foucault’s special historical interest in the problem of sexual pleasure, i.e. the way in which sexual pleasure is problematised, threatens to shackle and restrict his analytical access to classical texts. On its most general definition, pleasure (hedone) according to Aristotle is a perceptible process towards a natural state – and this is true of every creature that can experience pleasure and pain. Humans and animals thus all seek pleasure in the same way, and the pursuit of pleasure or desire (epithymia) is a basic aspect of the will to live.16 The differences between Aristotle’s analyses of human pleasure in books vii and x of the Nicomachean Ethics are a much-discussed topos of scholarly research: in NE vii pleasure is identified with unrestrained activities, but according to NE x pleasures supervene on the activities which correspond to them.17 Yet, notwithstanding all the differences, both interpretations share the view that just as there are according to Aristotle a variety of activities, so too there are many different forms of corresponding pleasures, not all of which are bad.18 Because human nature is not simple but complex, we will not always find the same things pleasurable, and in particular those forms of pleasure that supervene on good activities are themselves good, while those forms of pleasure that supervene on bad activities are bad.19 For Aristotle our concept of the pleasurable is not external to our model of a good life and of happiness (eudaimonia), for the good implies a picture of pleasure that is essentially linked to excellent, virtuous action, even if even good pleasure is naturally not the ultimate end of our action, but merely supervenes on these ends. The good life does not involve the complete avoidance of pleasure, but a correct view of what is good and what is bad pleasure. If we learn to act excellently, then at the same time we learn – and this is intrinsically connected with it – to experience pleasure in doing so. Remarkably enough, Aristotle also claims that this applies to bodily pleasure, for there is also a correct and an incorrect form of experiencing pleasure in eating and sex.20 These ‘necessary’ forms of pleasure also allow a more or less, an excess or a deficit. So ‘it is the pursuit of this excess, not that of the necessary pleasures, that makes a man bad; everyone enjoys 16 See Ar. NE vii 12, 1152 b12–14; Rhet. i 11, 1369 b32 ff.; NE vii 14, 1153 b25 ff.; NE x 2, 1172 b35 ff.; 4, 1175 a10–21; ix 9, 1170 a19. 17 See on this Festugi`ere (1946); Lieberg (1958); Ricken (1976); Owen (1971–2); Gosling (1973–4); Taylor (1987). 18 See e.g. Nussbaum (1986), 294 f.; Annas (1988), 285–300, as well as the classic article by Owen (1971–2). 19 Ar. NE vii 14, 1154 b20–3; x 5, 1175 b25–32. 20 This interpretation is excellently developed by Annas (1988) (see note 18), who also discusses the problems of the relevant texts in detail.

66

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

tasty food and wine, and sex in some degree, but not everyone to the right degree’.21 And on the other hand: ‘There is also a type of person who finds too little enjoyment in bodily pleasures.’22 Thus, it is not the victorious, agonistic suppression of a threatening urge that is worthy of praise, but finding the right measure in the positive attitude to the physical pleasures that all humans find particularly desirable, and which they experience intensely.23 The right measure excludes not only lack of restraint, but also abstinence. It can only be achieved when the relationship of physical pleasure and sexual activity to other forms of pleasure is correct, and when pleasure and sex can be incorporated into a properly thought out model of the good life as a whole. The species and the values of pleasures are dependent on the species and values of the respective activities to which they are linked. Their purity and stability, their anthropological specificity and their relation to virtue, are examples of criteria for determining the quality of different forms of pleasure. Thus, pleasures that we as humans share with animals are inferior to pleasures that are specific to us; a pleasure that is not mixed with reluctance, and can be long-lasting, is superior to impure, shortlived pleasure; and of course pleasure that is related to virtuous action is superior to pleasure in bad action.24 According to these criteria sexual pleasures are undoubtedly inferior, but this alone is not sufficient reason for Aristotle to be concerned about them, or to problematise their use. The exact form in which he does problematise them only becomes clear when the concept of sexual desire is embedded in the functional, conceptual framework of eudaimonia that Aristotle develops in the Nicomachean Ethics – a relation that is painfully absent from Foucault’s analysis. For reasons of brevity, I would like to illustrate the functional terminology that for Aristotle represents the conceptual framework for the determination of eudaimonia in the form of three explications.25 (A) (i) Where X is an organism, then F is a functional characteristic of X if: (a) F is necessary for the specific end (telos, normal state) of X; (b) X has F as long as X exists; (c) F is explanatory of other characteristics of X. 21 22 23 24 25

Ar. NE vii 14, 1154 a16–18. Ar. NE vii 11, 1151 b23 f. Ar. NE vii 15. Ar. NE x 1–5; see also i 3; esp. NE x 5, 1175 b37 f.; in addition NE vii 12–14. See above all Ar. NE i 6.

ethical teleology

67

(ii) (a) The specific achievement of an organism X is the ability of X to realise its functional characteristics. (b) The ability mentioned in (a) is the specific excellence or virtue of X. (c) The structure that within the context of the specific achievement of an organism X is revealed in the hierarchy of the functional characteristics of X, is the soul of X. (iii) The specific achievement of humans consists, in outline, of: (a) employing growth and perception for the exercise of reason; (b) guiding growth and perception through the exercise of practical reason; (c) when possible occasionally also exercising reason independent of growth and perception; whereby all of this is optimised and occurs as constantly as possible. The ‘reason’ in (A) refers to a very specific ability – contemplative rather than investigative, practical rather than productive. In other words it is the active, practical reason that is characterised by a peculiar correspondence between pursuit and truth – a mixture of theoretical reflection about genuine needs, and a practical wish for full personal development. (B) The active, practical reason; (i) unlike investigative reason, is not directed at what is necessary, but at that which can be different; (ii) unlike practical ability, is not directed at an end outside action (i.e. it is not production), but at an action that is an end in itself; (iii) is connected to the contemplative part of the soul, i.e. it includes deliberation (about the action); (iv) is a kind of unity of pursuit and thought, or a correspondence between correct pursuit and thought; (v) is more closely directed at a concept of the good, happy life as a whole, i.e. (a) it must recognise the really good life (truth); (b) it must also pursue the good life earnestly (correct pursuit); (c) it must deliberate about the role of individual actions in the overall plan of the good life (specificness); (d) it is an end in itself, i.e. it produces action as a constituent of happiness (autarky). Happiness can now be more exactly defined:

68

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

(C) Happiness is an activity of the soul, which leads to action in such a way that the activity is (i) directed by practical reason in the sense of (B); (ii) an excellent specific achievement of humans in the sense of (A); (iii) kept up throughout an entire lifetime. Thus, according to Aristotle we lead a good, happy life when we have transformed ourselves into stable, mature personalities, that are well developed intellectually, physically, socially and emotionally, and make consistent decisions. In particular, the good life is as free and autarkic as possible, it is active and complex, consistent and stable, as well as being unrestricted by extreme personal or financial misfortune. Three aspects must be taken into account when considering the systematic integration of the various types of pleasure into the functional concept of eudaimonia, or the best life. First, for Aristotle all motions and activities that maintain or restore a natural state in conscious organisms26 are incidentally accompanied by feelings of pleasure (in the former case without the addition of pain, in the latter case accompanied by a decreasing feeling of pain).27 Feelings of pleasure effectively supervene on the realisation of the specific achievements of all conscious organisms, and in each case the feelings are appropriate and ‘right, and the same is true, without qualification, of the specific achievements of humans, as outlined in (A)– (C).28 Furthermore, the conception of eudaimonia according to (A)–(C), as well as according to NE i and NE x, is to be understood inclusively: the successful, flourishing life of the happy person is not satisfied with merely scientific-theoretical activity (this would be the exclusive, intellectualistic interpretation of eudaimonia), but includes pleasure, honour, 26 All conscious organisms, that is all animals, experience pleasure and pain, and so also desire see Ar. An. ii 2, 413 b23; 3, 414 b3; iii 7, 431a 10–12. 27 See e.g. Ar. NE vii 13, 1152 b33–1153 a15. 28 As Kenny (1992), 21 is right to emphasise, in Aristotle’s mature ethics as laid out in the Eudemian Ethics, the concepts of the good, the pleasant and the virtuous coincide in their application (see e.g. Ar. EE 1237 a8–9). Thus a truly happy person, who leads the best life, also optimises his pleasures (see e.g. EE 1249 a18–21). According to Kenny (1992), the thesis that a happy life is not directed towards a single end, but should consider different ultimate ends according to the respective current situation and circumstances, and that in life these ends should be adapted to each other reflectively, is more clearly formulated in the EE than in the NE. At any rate, Aristotle generally believed that pleasures make the unrestricted specific human activities (energeiai) complete (NE x 4–5, 1174 b14–1175 b1; 1175 b23–35), or are even identical with them (NE vii 10, 1153 a1–5; this passage is also to be found in EE iv); see on this the classic studies by Festugi`ere (1946) and Owen (1971–2).

ethical teleology

69

reason, autarky, good relationships with family, friends and fellow citizens, and good health as well as external goods, and modest prosperity. Put more precisely, these elements are not means of obtaining happiness, but are constitutive elements of it, just as dance steps are not the means of dancing but constitute it.29 And finally, from (A)–(C) it is clear that happiness more precisely consists of optimising the order of the various constitutive elements of a successful life under the leadership of active reason (practical wisdom), and this order does not optimise just particular individual elements, but an entire life to the ‘good’ or ‘best’ life. To sum up, we can say of Foucault’s comments on the ‘teleology’ with respect to the Aristotelian ethics that he is correct in recognising the intrinsic unity that the aphrodisia forms together with desire and pleasure. However, he underestimates the scientific and metaphysical foundation of this unity which has as its aim a natural normal state in sentient beings, and places sexual pleasure in the broader context of the forms of pleasure. This mistake has serious consequences: Aristotle certainly does not attribute to the dynamic of aphrodisia, desire and pleasure an inherent lack of moderation that can also lead to weakening and over-exertion, and so is to be regarded fundamentally with concern and mistrust. In fact, for Aristotle sexual activity is not only natural and necessary, but the nature 29 The question of whether the Aristotelian concept of eudaimonia is to be understood inclusively or exclusively (that is intellectualistically), has been the subject of lengthy scholarly debate. For example, Cooper (1990) provides a weighty defence of the intellectualistic interpretation, while Ackrill (1980) is an influential proponent of the inclusive standpoint. On the whole Ar. NE i seems to be closer to the inclusive, NE x (esp. ch. 7) closer to the exclusive reading. However, we should note that in NE x – the focus of attention is the restriction of the political and the contemplative life, and therefore particularly the intellectualistic components of the contemplative life are emphasised; – so too some external conditions of the best life are mentioned (necessities of life: 8, 1178 a25–6; food, health, a few external goods: 9, 1178 b33–1179 a9; relaxation, amusement, games: 6, 1176 b28–35); fellow workers (7, 1177 a35; friends, ix 9, 1170 b14–19); – the purely intellectual life is described as being divine and too high for human attainment (7, 1177 b27–1178 a8; 8, 1178 b8 ff.). Thus the best life that humans can achieve will, on the whole, include purely intellectual activities, but also consist of other constitutive elements. One of the most recent and exhaustive discussions of this problem is to be found in Kraut (1989) (esp. ch. 5 (‘Inclusivism’, 267–311)). He successfully defends a highly differentiated variant of the inclusive interpretation against the ‘exclusivists’. In his influential book Kenny (1978) presents the view that in NE the exclusive, in EE, on the other hand, the inclusive aspect dominates, but emphasises in Kenny (1992) that at least NE ii–iv and to some extent also NE i, support the inclusive view (see esp. 6, 28, 93).

70

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

of its dynamic is such that it can be integrated into a consistent and proven good life; it is only immoderate, excessive and threatening in those of a bad mental constitution. Thus the object of Aristotle’s moral concern is not so much sexuality itself, that is the dynamic of aphrodisia, desire and pleasure, but rather that constitution of humans that directs them to a more or less good life, and above all decides the moral quality of their life. Thus the fourth and last of the aspects which Foucault wishes to distinguish in ancient ethics as a whole, the teleology, has been determined within the framework of the Aristotelian ethics. It is only on the basis of an appropriate interpretation of the teleology that the other three aspects and their internal relationship can be correctly understood – an understanding that Foucault renders impossible because he only discusses the teleology after the ethical substance, the manner of subjugation and the ethical work. Therefore I wish to make a few points about these three aspects on the basis of the Aristotelian ethics. From the perspective of the teleology, Aristotle finds the aphrodisia and sexual pleasure morally interesting only in the context of all those forms of pleasure that are important for the good life and eudaimonia. This conclusion is of particular importance for the identification of the ‘ethical substance’ that is at the heart of moral concern about sexual pleasure. If the specific human happiness in the functional sense is to be understood inclusively, then it involves a number of different activities – from nutrition and sexual activity, to taking care of friendships and other social relationships, and political and scientific work – which are all accompanied by specific feelings of pleasure. This means that there is a series of different forms of pleasure which the practice of the self involved in the good life must deal with. The inclusive form of happiness, as Aristotle sees it, quite definitely implies that none of these forms of pleasure should be completely suppressed or given far too much weight. This also applies to sexual pleasure: neither a life of sexual abstinence nor one that is mainly devoted to sexual satisfaction can be a good or best life. The good life is in fact characterised: – by a sensitisation to all forms of pleasure; – by the ability to create in long-term self-interest a stable hierarchy of the pleasures by means of practical reason, and to integrate them into a consistent concept of the good life.

ethical teleology

71

Certainly this involves suppressing or mastering some pleasures, but on the whole as far as pleasure is concerned the telos in fact involves an optimisation of the experiences of pleasure. To be more precise, Aristotle maintains that it is an empirical consequence of the best life that even our most diverse experiences of pleasure must be optimised in the long term.30 Thus, any behaviour, any pursuit, any choice that could endanger our route to complex happiness or its preservation is a cause for concern and a reason for problematisation. Undoubtedly sexual activity is one of them – but only in a particular sense: not because of its inherent, worrying, immoderate dynamic, but because it is intense. Humans of a poor mental constitution could succumb to it and pursue it excessively, or else be revolted by it and choose an ascetic life. Thus sexual activity cannot be inherently worrying or immodest, since in a fundamental sense it is natural and constitutive for happiness. This is the deeper systematic reason why Aristotle does not regard sexual activity as essentially bad or sinful, but instead is mainly concerned with its quantitative aspects, with moderation and reasoned calculation. Overall it is the correct, and rational, relationship between sexual pleasure and the maximisation of all our hierarchically ordered forms of pleasure that are important. But if sexual activity thus only provides a reason for moral problematisation under certain circumstances, and this has less to do with its nature than with the mental condition of humans, then on the other hand it is certainly not the only cause for moral concern. Every type of badness destroys or impedes the highest happiness – for example, cowardice, miserliness or dullness; excessive love of fame leads to a hectic political life that seeks only external recognition, and that will certainly not be the best life; and those who only want to earn money confuse the end and the means, and thereby block their own view of the human telos. In this extensive catalogue of morally problematic activities and attitudes, sexuality is only one of many elements. Hence Foucault’s identification of the ‘ethical substance’ of the Aristotelian ethics, and so of the most influential of all classical ethics, is 30 This is a direct consequence of the wording with which Aristotle claims to explain what pleasure is (in each case) – this is at the end of the long discussion of pleasure in NE vii 11–14 and x 1–5: pleasure is the ‘unimpeded activity of our natural state’ (anempodistos energeia tes kata physin hexeos; see vii 13, 1154 b32–3 with 1153 a14–15: here hexis is the term for a stable disposition of an activity specific to humans and the explication refers to each such specific activity). Owen (1971–2) shows that in principle Aristotle stands by this in EE also.

72

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

both imprecise and one-sided – imprecise because he attributes to sexual activity too great an implicit threat and dynamic, one-sided because he improperly separates the moral problematisation of sexuality from moral concern for the model of the best life. What is more, it is hard to see that the ‘ethical substance’ presents a new analytical category for ancient ethics. The Aristotelian ethics are a theory of the good life that appeals to the clarification of our rational self-interest – a clarification that, among other things, pre-supposes a certain amount of empirical and scientific knowledge. But above all this theory produces a complex, terminologically advanced picture of human happiness, and nothing is more natural than that anything that endangers this happiness should be grounds for moral concern. Any satisfactory interpretation of Aristotle’s ethics that meets normal standards requires determining what Foucault calls ‘ethical substance’, and so Foucault does not seem to introduce a fundamentally new method. Much the same can be said about the ‘use of pleasures’, that is the ethical form of ‘subjugation’. If we take this as meaning the use of sexual pleasures, then it is clear that we can not only claim, but with regard to Aristotle’s complex teleology also justify, and explain, why immodesty is the decisive improper behaviour, and why the strategy of pursuing temperance, of giving way when necessary and stabilising the balance of the pleasures, is morally desirable. This follows directly from the Aristotelian teleology of the good life. We could go one step further and claim that the use of pleasures cannot even be separated from the teleology of the Aristotelian ethics. This becomes immediately clear when one takes into account that the teleology in the Aristotelian sense does not describe a static telos as a structure, but explicitly characterises an activity guided by reason – activity, and not passivity and lack of motion, is a characteristic of happiness. For this reason the correct use of sexual pleasures is a constitutive element of the good life, and the description of, and instructions for, its use are part of Aristotle’s ethical teleology. Thus, in this context the category of the form of subjugation becomes part of the category of the teleology. At this point it is certainly not wrong to point, as Foucault does, to the aspect of subjugation; but it is more appropriate to emphasise that for Aristotle the achievement of happiness, of autarky and of competence in ethical judgement includes an act of subjugation – subjugation to the end of a consistent concept of a good life, of which the appropriate hierarchy of pleasures and their maximisation are part. The telos of the good life itself includes this subjugation, and therefore does not stand beyond

ethical teleology

73

the act. But at the same time it is clear that the term ‘subjugation’ can be misunderstood, at least in this context. For Aristotle it is rather a reasoned decision to integrate sexual activity into the model of a good life in a consistent manner. From his standpoint the category of the use of pleasures is defined too narrowly if only the sexual pleasures are considered. Nor can it be restricted to just the physical pleasures, but must include all forms of pleasure and their appropriate use. In his eyes the use of sexual or physical pleasures cannot be appropriate when they are not set in the correct relation to the use of the other pleasures, and vice versa. Finally, attention must be drawn to the fact that for Aristotle neither the ‘identification of the correct moment’ nor the social status of the actors plays a part in the use of (sexual) pleasures. It is not the person who seeks to preserve fame, respect or authority who should use her sexual pleasures appropriately, but the person who wishes to be happy in the best sense. From the Aristotelian perspective the form of ethical work or practice of the self must also be described in a more complex way than Foucault suggests. It was demonstrated above that the model of domination and restriction of enkrateia, which is a central concept for Foucault, is in Aristotle a stage of human development that in the ethical sense is deficient, and is at best part of the process of one’s transformation into a good and happy person. But it is important to remember that for Aristotle the good life as the ethical telos also includes a particular practice of the self. For eudaimonia is also a form of activity – but one which is not so much work on the transformation of ourself, as a constant endeavour to fully realise our intellectual, physical and social capacities. This endeavour undoubtedly demands reflective self-control – but not in order to restrict and conquer the threatening dynamic of sexual pleasure, rather so that we repeatedly optimise anew the functional structure of our specific achievements, and thus also the sublime balance of all of our pleasures, in the light of a rational model of the best life, and in the face of changing circumstances in our lives.31 This form of the practice of the self no longer consists of an agonal, painful and ultimately successful struggle against pleasure, for the reasoned telos of the good life has already become the driving motivation 31 Nussbaum’s (1986) influential study focuses on the ancient insight into the contingency and fragility of happiness and competence, as expressed not only in philosophical texts, but also by many poets (above all the tragedians). In particular ch. 10 (‘Non-scientific deliberations’) shows how according to Aristotle practical reason should constantly adapt anew the way we deal with the changing contingent circumstances of our lives to the rational model of the good life.

74

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

for action. Rather it is a reflective and controlled integration of concrete practical decisions and actions into the telos of life – a goal that is not easy to achieve given the contingency of the condition humaine. 2 Thus, from the perspective of the Aristotelian ethics, Foucault does not do justice to the conceptual complexity and systematic relevance of an ethical teleology. As could be shown, this not only leads to a distorted identification of the problematisation of the pleasures (ethical substance), of the use of pleasures (the form of ethical subjugation) and of work on the pleasures (the ethical work), but the close relationship of the ethical model of the good life to knowledge and power that Aristotle proposes is also neglected. Thus Foucault denies himself the chance of studying interesting systematic relationships between archaeology, genealogy and ethics on the basis of classical philosophical texts and nor can he develop a richer concept of ethical work that goes beyond the narrow and implausible boundaries of the aestheticisation and stylisation of our own existence. I would like to support this view with a few additional remarks on Aristotle’s ethics and its systematic environment. The functional definition of eudaimonia, which distinguishes the best life, is essentially concerned with the concept of practical reason (phronesis).32 According to Aristotle practical reason is neither a practical knowledge (for it is not directed instrumentally at ends outside of action) nor an investigative prudence (for it is not directed at that which is necessary). But although its most valuable fields are the contingent, which can also be different, and action, which contains its end in itself, it is nevertheless an important form of knowledge. The Aristotelian ethics describe phronesis as a peculiar unity or correspondence between pursuit and thought: on the one hand it contains the impulse and the motivation to realise the truly good life earnestly, on the other hand it requires an acceptance and understanding of our reasonable self-interest and the pattern of the truly good life, as well as the ability to consider and incorporate particular situations and decisions into this pattern.33 Thus, according to Aristotle, phronesis is an important cognitive component of the ethical teleology that is concerned with a complex field of knowledge. Foucault was 32 See NE i 6 and (A)–(C) above, pp. 66–8. 33 On the individual tasks of practical reason see above all Ar. NE vi 5, 1140 a24–b7; vi 2, 1139 a3–15; a29–31; b4–7.

ethical teleology

75

certainly aware of this component, and of its historic roots in the Socratic– Platonic doctrine of the unity of virtue and knowledge – sophrosyne ‘cannot be understood without reference to truth’. But it is significant that his understanding of practical reason is extremely restrictive and abstract. Foucault’s interpretation is heavily coloured by his model of the control involved in enkrateia: moderation and control of pleasure imply a sovereign and regulative position of the logos with regard to appetites (the ‘structural form’ of practical reason), as well as the decision what one should do, and how and when one should do it (the ‘instrumental form’ of practical reason). According to Foucault, only in Plato does a third element play a role – knowledge of the self, that is knowledge of one’s own soul (the ‘form of the ontological recognition of the self by the self’).34 But this tells us more about the function than the content of the cognitive component of practical knowledge, and as for the form of the relevant ‘game of truth’, in other words the appropriate method, Foucault says nothing at all. At the beginning of the Nicomachean Ethics Aristotle emphasises that knowledge of the greatest good, that is of the appropriate conception of eudaimonia, ‘is of great importance to us for the conduct of our lives’, and that is it should be an integral part of eudaimonia, since this knowledge allows us ‘like an archer to hit better’ the targets in life which we all have in mind in one way or another.35 However, this is not just everyday, common, trivial knowledge, but political science of a general kind, which aims – and is also necessary – to make humans (as citizens) virtuous. Ethics is a branch of political science, just like other forms of knowledge such as property management or public speaking.36 But to understand the important role of political science, it is necessary to bear in mind that Aristotle regarded living in political communities as a distinctive feature of humans, and thought that a good, justifiable life was only possible for us if the political constitution of the state in which we live was also good. Thus, for him, practical reason refers to a large number of different cognitive contents – in fact to all those fields which he himself covers in his ethics and in the Politics. At the beginning of the Rhetorics he remarks that the art 34 UP 86–9 esp. the summary 88 f.: ‘be it in the form of a hierarchical structure of the human being, in the form of a practice of prudence or of the soul’s recognition of its own being, the relation to truth constituted an essential element of moderation. It was necessary for the measured use of pleasures, necessary for controlling their violence.’ 35 Ar. NE i 1, 1094 a22–5; see also 1095 a3–6 (the end of ethics is not knowledge, but action); i 10, 1099 b23–32 (the end of political science is noble action on the part of the citizens); ii 2, 1103 b27–9 (ethics contributes to virtuous action); x 10, 1179 a34–b31 (ethics is necessary, but not sufficient for virtuous action). 36 Ar. NE i 1, 1094 a27–b12.

76

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

of public speaking is related to dialectics and political science (including ethics), so that just like the good citizen and happy person in general, so too the good speaker must have an appropriate view of what constitutes a happy life and excellence of character. He must be able to distinguish and assess the various types of political constitution, to recognise the characteristics of voluntary and involuntary action, as well as to name the motives for wrongdoing, to appreciate the value of friendship, to know the various complex definitions of what is pleasurable and just, as well as to have gathered a comprehensive knowledge about feelings.37 Effectively this is a description of the main subject matter of Aristotle’s writings on ethics and politics. Comprehensive, complex and sound knowledge of humans and the polis are included in the ethical teleology that he develops: the ‘relationship to truth’, which sophrosyne and phronesis also have on Foucault’s interpretation, is certainly not vague or abstract. On the contrary it is quite specific and concrete. This is particularly true of the methodological aspects of this knowledge, that is – to use Foucault’s terminology – of the game of truth that is constitutive for practical reason. At the beginning of the Nicomachean Ethics Aristotle not only makes quite clear that this is a knowledge that is guided methodically (ethike, politike), he also refers in some detail to a number of methodological peculiarities: we cannot expect exact knowledge in this context, but must be satisfied with a smaller degree of certainty which at best allows us to comprehend what is for the most part so (to hos epi to poly); we should always start with that which we know, and attempt to move on from there to more general principles, while at the same time being satisfied with determining the facts, and not trying to discover the causes.38 In a famous passage at the beginning of book vii of the Nicomachean Ethics Aristotle explains the methods of ethics more exactly: 37 Much of the first book of Aristotle’s Rhetorics sketches the contents of this knowledge, with the exception of the ‘theory’ of feelings presented in Rhet. ii 1–19. His arguments in the Protrepticus, passages of which are preserved in Iamblichus’ exerpts, are more forceful and general. The Protrepticus is a riposte to the Antidosis written by the influential orator Isocrates, a fictional defence speech, which, among other things, complains about the uselessness of academic philosophy. In the Protrepticus Aristotle not only pointedly refers to the great value of ‘useless’ activities that do not have a specific purpose, for example philosophical activity (see frag. B11–B14, B25, B27, B42–3, in the edition by During ¨ (1969)), he also points out that even theoretical philosophical knowledge of nature is important for adequate action: ‘This knowledge may be in itself theoretical, but it provides us with the possibility of modifying our action in accordance with it’ (frag. B51, see also B38, B41, B46–7). 38 Ar. NE i 1, 1094 b11–22; i 2, 1095 a31–b13.

ethical teleology

77

Here . . . we must first set out the evidence [phainomena], and after calling attention to the difficulties [aporiai] proceed to establish [deiknynai], if possible, all the received opinions [endoxa] about these affections, or failing that as many as we can of those that are best supported. For if the discrepancies are resolved, and received opinions left validated, the truth will be sufficiently demonstrated.39

According to this description the ethical method consists essentially in considering the quite different, often even incompatible, accepted opinions regarding a particular question, examining them on the basis of empirical evidence, and then rendering them as consistent as possible by means of a suitable interpretation that is not at odds with the evidence, so arriving at plausible opinions. Quite clearly the methodological game of truth that is employed in Aristotle’s ethics is dialectics (in the ancient sense).40 But he certainly does not restrict it to ethics and politics: rather the dialectic method, much like logic and the theory of science, is not limited to a specific subject, for one of its specific characteristics is that it can help us in the discussion of any problem we are presented with.41 On the other hand, it is true that the dialectic method is particularly prominent in ethics, and together with the subject matter of ethics constitutes a ‘game of truth’ that is specific to it.42 Perhaps we can follow Foucault in interpreting the Aristotelian ethics as a recommendation for a ‘subjectivisation’, by means of which persons as actors constitute themselves as moral subjects – in the rather trivial sense that we should lead a life that is being examined constantly in such 39 Ar. NE vii 1, 1145 b2–7; see also Phys. iv 4, 211 a7–11. 40 It is frequently stated that dialectics is related to the endoxa, assesses the problems and seeks to render the endoxa consistent; see e.g. Ar. Top. i 1, 100 a30–b22; An. Prior. i 1, 24 a22–4; i 30, 40 a9–10; SE 11, 172 a21–6; Met. iii 1, 995 b23–4; Rhet. i 8, 135 b35; An. Post. i 6, 74 b21–6; i 11. The literature on Aristotle’s ethics is immense. The recent studies by Nussbaum (1986) and Reeve (1992) are interesting and influential, while Detel (1993) ii, 243–5 reviews a number of important works. Nussbaum and Reeve attempt to describe the close connection between teleology and ethical (dialectic) method in different ways: Nussbaum tends to separate the ethical method and ethical knowledge more clearly from demonstrative science, while Reeve prefers to emphasise their common ground. At any rate, their work is impressive proof of the intrinsic interconnection of ethical teleology with both the methodology and the content of a specific practical knowledge in Aristotle. 41 See Ar. An. Post. i 12, 78 a10–13; Top. i 1, 100 a18–20; i 11, 172 a12; Met. iv 4, 1004 b19–20; Rhet. i 1, 1354 a1–3 on the relationship of dialectics and (demonstrative) science see Detel (1993) i, 151–7 (with a list of texts in Aristotle with a non-ethical, dialectical argumentation). 42 See on this Barnes (1980); also Mueller-Goldingen (1988), 461–82 (with further literature); Irwin (1981), as well as the classic Owen (1967).

78

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

a way that by a permanent process of self-control and self-reflection we subjugate ourselves to the demands that a rational model of the best life makes of us. But if this is correct, then according to Aristotle the subjectivisation is intrinsically connected with an ‘objectification’ that opens up a rich and complex field of knowledge in the form of ethics and the science of the polis, and within which a clearly defined and surprisingly complex ‘game of truth’ is played in the guise of the dialectic method.43 Persons themselves belong to this field of knowledge – as rational actors and good citizens. Thus at the same time, within the context of the objectification connected with ethical subjectivisation, we make ourselves the objects of a knowledge that of necessity includes knowledge of the self in the objective sense. According to Aristotle a particular feature of this knowledge and its associated field is that in principle they should, under favourable circumstances, be accessible to all humans – in the environment of the classical polis at least to all male citizens who are involved in political processes. The general applicability of the dialectic method corresponds to this universal epistemic accessibility of the field of knowledge, that is of the relevant game of truth which is indeed not the topic of a special science, but of education (paideia).44 According to Aristotle a universal ethical and political objectification is linked to ethical subjectivisation, inasmuch as this objectification is a constitutive element of the ethical teleology. Quite clearly Foucault dramatically underestimated the factual content, the methodological specificity and the systematic weight of this objectification within the context of the ethical teleology. The ethical work of the ancient world is at the same time educational work in a much more comprehensive and complex sense than Foucault is prepared to admit. 3 As Aristotle sees it, ethics as the theory of and instruction in the good life is closely connected with the ‘theory of the polis’, that is ‘politics’ or ‘political technology’. Politics in this sense is identified as the supreme science for the accomplishment of the best life in the polis, and as such pursues the same end for the polis as a whole as ethics does for each individual.45 Sometimes Aristotle seems to talk about the happiness (eudaimonia) of the polis as a whole, but this should not be interpreted holistically: Aristotle 43 Of course it is not just the general comments cited in notes 38–42 with which Aristotle characterises dialectics. It is above all in the Topics that he provides a comprehensive and complex arsenal of dialectic argumentation. 44 See Ar. PA i 1, 639 a5–12; Met. iv 4, 1006 a5–9; NE i 4, 1044 b12–27. 45 Ar. NE i 1, 1094 a 27–b10; i 13, 1102 a5–13; ii 1, 1103 b3–6.

ethical teleology

79

does not presuppose that the polis is a supra-individual entity to which we can meaningfully attribute happiness. Rather the polis as such is ‘happy’ when its citizens are – and the polis is happy to the same extent as its citizens. What is more, although the ‘political technology’ is a science (or ‘technology’), which is used and employed by the rulers, this should not be interpreted in a totalitarian sense: Aristotle does not assume that the rulers prescribe what the individual citizens should understand as the good life. Instead the polis should have a just constitution which enables all citizens to lead a good and proven life. The ‘political technology’ serves to establish and stabilise such a constitution (politeia).46 Foucault only refers to a single line connecting ancient ethics as a practice of the self and political conditions of power in the ancient polis, and this line only represents an indirect relationship: only those who achieve the state of self-control and government of pleasure (in the form of sophrosyne, or at least in the form of enkrateia) can be regarded as suitable for a political career and to be trained in political rule.47 But if the state (polis) and its constitution (politeia) primarily serve to enable its citizens to enjoy a good, happy life as laid down in the ethics, then the power structure of the state must be intimately tailored to the structure and the conditions of eudaimonia. This connection can indeed be found in the Aristotelian Ethics and Politics, above all within the framework of the difficult relationship between the contemplative and the political life. When he first introduces the distinction between four ways of life (the contemplative, the political, the pleasure-seeking and that which revolves around money) in the first book of the Nicomachean Ethics, Aristotle takes the opportunity to point out that apart from the contemplative way, none of them fulfils the criteria for a happy life.48 His argument that this is 46 According to Pol. iii 1–3 each state (every polis) can be identified criteriologically on the basis of its constitution (politeia), but it is ontologically identical with the sum of its citizens. In Pol. vii 1–3 a state is good if it enables and encourages a good life in its citizens and the criteria for a good life and the good state are identical; finally, it is possible, although not desirable and therefore no contradiction, that the citizens as a whole are honest (in the form of their majority decisions), but that at the same time individual citizens are not. Thus, the happiness of the citizens is not identified with the happiness of the state as a supra-individual entity – Aristotle did not propagate a metaphysical holistic concept of the state. Furthermore, the state does not decide the happiness of the citizens, but rather the eudaimonia as a plan is explicated independently of state institutions (e.g. in the Aristotelian ethics), but requires a good state in order to be realised – Aristotle does not propagate a state totalitarianism of values. The discussion of the best constitution in Pol. vii explicitly pre-supposes the concept of eudaimonia and of the good life that are developed in NE (see Pol. vii 1, 1323 a14–1324 a4). 47 See UP 69–77. 48 Ar. NE i 3, esp. 1095 b22–31 (on the political way of life).

80

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

particularly true of the political way of life is of enormous significance: the end of the political life is honour, but ‘honour is felt to depend more on those who confer it than on him who receives it, and we feel that the Good is something proper to its possessor, and not easily taken from him’. Honour and recognition as values are not compatible with autarky, which is an essential characteristic of eudaimonia – in our chase for these values we humans are dependent on the opinion of other humans.49 A good life should be so structured that it is as independent of contingent factors as possible, and this is definitely not the case for a life of which the highest end is the maximisation of recognition and honour, for these are permanently influenced by external circumstances and judgements.50 For Foucault one, if not the main, characteristic of ancient ethics is that the ethical subjectivisation and objectification ‘open onto an aesthetics of existence’, in that the recommended good life ‘took on the brilliance of a beauty that was revealed to those able to behold it or keep its memory in mind’. It is through this ‘stylisation of attitudes’ that ‘one would be able to give one’s conduct the form that would assure one of a name meriting remembrance’.51 Perhaps for Aristotle recognition, honour, glory, beauty, remembrance, or an aesthetic way of life can often be side-effects or consequences of a proven life guided by eudaimonia. But his arguments against the political life as a form of the best life imply that these consequences cannot also be the main goal of the best life, as they involve more contingency, dependence and hectic activity than would be compatible with the autarky of the happy life. However, Aristotle’s position on this matter is more complex than the above comments suggest. At the beginning of the Nicomachean Ethics he already makes quite clear that the autarky of the happy life by no means requires absolute independence from social relationships, nor the solitude of the hermit’s existence: The same conclusion seems to follow from another consideration. It is a generally accepted view that the perfect good is self-sufficient. By self-sufficient we mean not what is sufficient for oneself alone in a solitary life, but something that includes parents, wife and children, friends, and fellow-citizens in general, for man is by nature a social animal.52

49 See also Ar. NE x 7, where he also points out that the political life requires power and does not guarantee sufficient leisure. Both increase the dependence of political activities. 50 See NE i 10, 1100 b33–1101 a8; also x 8, 1178 a34–b3; i 7, 1097 b14–16. 51 UP 89, 92 f. 52 Ar. NE i, 1097 b7–11.

ethical teleology

81

As far as recognition is concerned, Aristotle regards friendship as certainly the most important of all the social relationships that a developed, happy person can engage in, particularly if the friends are similarly developed, happy persons. The mutual respect of free and equal persons is a constitutive element of the flourishing and happy life, but in this case the respect of the other person coincides substantially with confident and justified respect for oneself. For the developed and free citizen respects exactly the same things in his similarly developed and free friend as he respects in himself – namely the realisation of eudaimonia in the course of a complex and stable ethical work with all its ethical and cognitive elements.53 The stable assessment of the self, together with the calm respect of the developed other person, the coincidence of love of the self and love of the friend, are for Aristotle the form that recognition of a good life in the reactions of others assumes as an essential element of eudaimonia. This reciprocal relationship between the free and equal is not in danger of losing autarky (in the correct sense of the term); it remains independent of contingent, changing circumstances and it is beyond criteria such as glory, admiration and remembrance, which have more to do with the pursuit of honour (thymos) than with the realm of reason (logos). These criteria have their roots in the judgement of lesser humans, and, as Greek tragedy so dramatically illustrates, can often swiftly and cruelly turn into their opposite. 53 The emphatic discussion of friendship in NE viii–ix accentuates that friendship is the greatest of all external goods (ix 9, 1109 b9–10), that the happy and excellent life necessarily requires friendship (viii 6, 1157 b19–21; ix 9, 1169 b17–19) and that humans are by nature created to live together, and thus friends are by nature close to each other (ix 9, 1170 a13–14, b13–16). Of course, it is not primarily a friendship for the sake of benefit, or of what is pleasant, but that form of friendship that has as its end excellence (viii 3). This form of friendship is well balanced and reciprocal: it is characterised by mutual respect and goodwill and unselfish action in the interest of the other (viii 2, 1155 b31–4); viii 4, 1156 b6–12; ix 4, 1165 a2–9; see Rhet. 1380 b36–1381 a2). But at the same time all forms of friendship are cultivated for the sake of self-advantage (NE viii 4). But this is not a contradiction, for the relationship of the excellent person to himself is structurally identical with his relationship with the excellent friend (we wish the friend all the best; we wish the same as the friend and want to be together with him always; and we share pleasure and pain with the friend – and all of this is also true when we ourselves are the friend; see viii 4). The excellent person acts towards his friend, who is ‘another himself’, in the same way as he does towards himself (ix 9, 1170 b6–9), and this respectful relationship is justified because that which we love in both ourselves and our friend is objectively and reasonably lovable (ix 8). Furthermore, recognising and respecting excellence in a friend also contributes towards the development and stabilisation of our own excellence (see ix 9, 1169 b22–1170 a12). See on this the important article by Cooper (1980); also Nussbaum (1986), 354–72; Kraut (1989), ch. 2 (78–154); Reeve (1992), 114–22 and 173–82; Kenny (1992), 43–55.

82

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

The political life, that is the life that is dedicated to politics, is for Aristotle certainly not the best life. But the inclusive form of eudaimonia, and the fact that social relationships should be consistent with the autarky of the good life, are important indications that political activity could nevertheless be a significant element of eudaimonia. Indeed, it can be shown that Aristotle defends this view in a complex series of arguments.54 At the same time these arguments allow a closer examination of the way in which the Aristotelian teleology is intimately interconnected with the propagation and support of political relations of power – an interconnection, incidentally, that goes well beyond anything that Foucault is prepared to accept, namely that self-control and stylisation of conduct are important conditions for the political career of a full citizen. So how does Aristotle think that the sphere of politics is interwoven with the ethical teleology? First of all we can say that for him a publicly organised education, guided by good political laws, is an important condition for the development of a good character;55 for the unity and cohesion of the polis is in danger when education is left to the private discretion of parents. Even after the completion of education, moral practice in the public sphere of the polis remains an important means of maintaining and further developing the good character, and without this a good life is impossible.56 Thus, education and moral practice in the public political arena are for Aristotle certainly necessary for the development of a good character, which is in turn a precondition of eudaimonia. But beyond that, political activity is also an intrinsic telos and a constitutive element of the best life. Perhaps not all rational plans of the best life will include the demand that every opportunity for political action should be realised, but the right to hold political office must be upheld, for without this right life remains alienated and uprooted – a state that would be incompatible with eudaimonia.57 After all, we humans are ‘political creatures’ – political activity is part of our nature. Here Aristotle not only points out that we cannot survive outside political communities, but stresses that we cannot lead a good and 54 The best summary of these arguments is to be found in Nussbaum (1986), ch. 12 (343–77). Kraut (1989), ch. 1 (15–77) and ch. 5 (267–311) offers a detailed discussion of the relationships between the contemplative and the political life. 55 NE ii 1, 1103 b23–24; x 10, 1179 b23–1180 a24; Pol. viii 1, 1337 a21–9. 56 NE x 10, 1180 a1–4. 57 MM ii 12,1212 a34 ff.; Pol. iii 5, 1278 a34–8; accordingly, the best politeia should be organised so that all full citizens have a lifelong right to hold public office; see MM i 33, 1194 b5–23, Pol. i 6, 1255 b20, v 2, 1261 a39, iii 4, 1277 b7, iii 17, 1288 a12, ii 12, 1274 a22 ff., iii 4, 1276 b38 ff., vi 2, 1317 b2–3, b32 ff.

ethical teleology

83

happy life outside them either – a life without political and social activity could hardly be regarded as fully human.58 From this perspective we can also understand that justice as a fundamental ‘rational’ value is the ‘complete’ virtue under which all other merits of character can be subsumed: a self-centred non-political life would lead to the loss of all other merits of character.59 More interestingly, Aristotle believes that this close interconnection between ethical teleology and political activity implies very specific political structures and relations of power. It is a well-known fact that according to his political theory the state is ‘a community of equals aiming at the best possible life’; the perfect state with the best constitution can best realise eudaimonia, and a state is excellent when its citizens are excellent.60 What is more, the state is so complex that it not only consists of parts, but also pre-supposes necessary conditions and structures that are required for its survival.61 The parts of which a political community consists are precisely its full citizens, whose complete development in relation to eudaimonia is to be guaranteed by the best constitution (politeia). Thus, the full citizens are free and equal, that is they should be able to choose how they lead their life prudently and independently, and be each others’ equals in intellectual and social self-development.62 But how can this fundamental freedom and equality, as a constitutive element of the ethical teleology, be consistent with the distinction between rulers and ruled, or the politically dominant and dominated? Aristotle feels that there is only one political condition that ensures this consistency, namely the rotation of political power among the full citizens: One principle of liberty is for all to rule and be ruled in turn . . . For equals the honourable and the just consist in sharing alike, as is just and equal. But that the unequal should be given to equals, and the unlike to those who are like, is contrary to nature . . .63

To put it more precisely, the political rulers will always be members of the group of full citizens, always be voted for by the full citizens or chosen from among them by lot, and give up their office to other full citizens by See NE ix 9, 1169 b16–19; Pol. i 2, 1253 a1–7 (on this Nussbaum (1986), 349–51). NE v 3, 1129 b26 ff. Pol. vii 8, 1328 a37–8; vii 13, 1332 a3–6, 31–3. Pol. vii 8, 1328 a21–36. Pol. vii 14, 1332 b11–12. The citizens participate in a ‘life of free choice’ (Pol. iii 9, 1280 a32–4) and, in contrast to slaves, are ‘their own’ (Pol. i 4, 1254 a14–15); what is more in their eudaimonia and excellence they are ‘equals’ (see Pol. vii 8). 63 Pol. vi 2, 1317 b1–2; vii 3, 1325 b6–9.

58 59 60 61 62

84

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

regular rotation. A direct, non-representative, grassroots democracy of all full citizens is thus the only form of political rule that is consistent with the ethical teleology and its fundamental role in the best politeia. Of course this picture must be expanded by two further basic assumptions that bring with them far-reaching consequences. On the one hand, any state necessarily includes a functional division of labour of which the most important elements are securing food, the production of tools, military protection, the provision of financial resources for public services such as building programmes and the police, as well for war, the cultivation of religion and finally the public procedure of deciding what is appropriate and just.64 On the other hand, the development and maintenance of excellence – including political activity – demanded a degree of leisure and necessary wealth which went far beyond what the ancient economy could provide for everybody who lived in and around a particular polis.65 Consequently, if the functional division of labour within the state requires a division of the classes into farmers, craftsmen, soldiers, the rich (!), priests and magistrates, and judges, then it is clear that farmers, craftsmen and above all slaves, as ‘philistines and labourers’ do not have sufficient leisure to attain excellence and eudaimonia, and thus must not only be excluded from political power, but also by definition do not belong to the parts of the state (only to its necessary preconditions).66 Much the same applies to all those additional persons who for other reasons cannot attain excellence, such as children, adolescents and women. Thus, if the highest human end, a state consisting of citizens living happily, is to be realised then not only must it have a non-representative, grassroots democracy, but, given the economic conditions of the ancient world, it must also have a sufficient number of persons who work to achieve this end, yet who at the same time must be excluded from the privileges of political activity and from the opportunity of conducting ethical work. In other words, in the best state there will of necessity be people who have reached very different stages of development with regard to the ethical teleology, and if social value is to be assessed on the basis of the degree of this development, then the fundamental social relations of recognition such as love or respect will also include asymmetrical relationships,67 and 64 Pol. vii, 8–10. 65 Pol. vii 9, 1329 a1–2; see NE x 7, 1177 b4–6; Pol. ii 9, 1269 a34–6; Pol. ii 11, 1273 a32–b7; Pol. vii 14, 1333 a37–1334 a10; Pol. viii 2, 1337 b4–14; 1338 a1–13; on this Nussbaum (1986), 347. 66 Pol. iii 5, 1278 a15–21; vii 9, 1328 b39–1329 a2, 1329 a34–9. 67 See e.g. NE viii 8.

ethical teleology

85

these can only be transformed into reciprocal forms of recognition of equal value among the fully developed (male) full citizens.68 Beyond these central political and social structures of the best state, a number of additional, more trivial and external preconditions can be deduced from the demand that the ethical teleology be realised for all male full citizens. The state should be large enough for the underprivileged working class to be able to provide the economic conditions that the full citizens require in order to pursue intensive ethical work and prudent political activity. On the other hand, the number of full citizens should be small enough to guarantee a non-representative, grassroots democracy that is based on its members knowing each other. Its territory should be sufficiently large and suitable to ensure economic autarky, but small enough to be defended properly, so that the ethical work of the full citizens is not restricted by economic or military dependence on less perfect states. Finally, the state should, of course, be peace loving, and seek only modest prosperity, for too much martial glory and luxury pervert even the best character.69 In this way instruction in moral ‘subjectivisation’, which can sometimes lead to an aesthetic stylisation of conduct (although this is certainly not its chief purpose), is intimately linked in the Aristotelian ethics to a number of definite and very concrete political relations of power, and these are the direct result of the close systematic interconnection of ethical and political ‘science’ as a specific field of knowledge with a uniform game of truth. By making themselves the objects of a self-transformation towards the ethical telos, the individuals must at the same time become the objects of political human science and strategies of power. This process includes the subjugation of the individuals to teleological ideals and ethical rules of ethical work, to methodological-dialectical rules and games of truth, as well as to political strategies and ends. What is more, this process is accompanied by continual self-observation, self-control and self-interpretation. In short, the process of objectification is at the same time – on all three levels – also a process of subjectivisation in the Foucauldian sense of ‘subject’. Perhaps we can follow Foucault in interpreting the Aristotelian ethics as instruction in ethical subjectivisation, or the formation of the moral subject in the guise of particular practices of ethical work. But it is only when we look at the complex ethical teleology that the intrinsic interconnection of ethics, archaeology and genealogy (in Foucault’s sense) becomes 68 NE viii 3–7. 69 Pol. vii 4–7.

86

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

visible. Yet this, paradoxically, is missing from Foucault’s own interpretation of ancient ethics. At the same time, it is precisely this interconnection that, at least in the case of the Aristotelian ethics, requires that the central concept of ethical work be differentiated, with far-reaching systematic consequences, above all for the concept of power. The methodological status of Aristotle’s ethics as a discipline of the human sciences assumes a complex and interesting relationship with the forms of ethical subjectivisation which it itself recommends.70 One of the main indications of this is the application to ethics of the difference between the facts (the ‘that’) or what is ‘known to us’, and the causes (the ‘why’) or what is ‘known absolutely’ – a difference that is to be found in Aristotle’s theory of science. According to Aristotle’s view, in ethics we should also proceed from the ‘that’ (the phenomena which are known to us), and move on to the ‘why’ (the explanation of the phenomena),71 even though the ‘that’ and the ‘why’ will not have exactly the same meaning here as in the natural sciences. What exactly Aristotle understands to be the factual starting-point of ethical science and instruction becomes clear in those passages that emphasise that apart from a natural disposition, the ethics also pre-supposes an ethical habit and conduct. This precondition has volitional, cognitive and practical aspects: whoever wishes to study and apply ethics in the Aristotelian sense successfully, must already pursue the end of becoming excellent in the ethical sense; he must also know that particular actions are excellent and he must already be accustomed to act in an excellent sense.72 The Aristotelian ethics is not intended to recommend virtue and excellence to all humans. The Aristotelian ethics does not address those who despise virtue and excellence; it does not want to turn bad people into humans who act and think excellently, for arguments and instruction are wasted on people who do not have both feet firmly on the ground of the ethical ‘that’, or the ethical phenomena. Rather, ethics is intended to reveal to those who already have experience of the ethical habit and the capacity for factual moral judgement, exactly what they do thereby, what further ethical development would amount to, and why they should achieve it. This 70 On the following, see above all the excellent article by Burnyeat (1980). 71 On the scientific distinction between ‘that’ and ‘why’ see Ar. An. Post. i 13, 78 a22–8; i 27, 87 a31–2; ii 1, 89 b23–31; ii 8, 93 a16–18; Met. vii 17, 1041 b14–16 (in PA i 1, 639 b6–11; 640 a13–16 Aristotle says ‘phenomenon’ instead of ‘that’; see in general Detel (1993) i, 244–7; ii, 547–9); on the application of this distinction in the ethics see NE i 2, 1095 b2–13; i 7, 1098 a33–b4 (on this subject see Burnyeat (1980), 71 ff.). 72 See e.g. NE i 1, 1094 b29–1095 a11; i 3, 1095 b6–13; x 9, 1179 b20 ff.; Pol. vii 13, 1332 a39–b10; vii 15, 1334 b6 ff.; viii 3, 1338 b4–8. Of particular relevance in this context is the much-discussed chapter NE ii 1.

ethical teleology

87

knowledge is for its part an essential precondition for the development of that stable, mature and complex character that should distinguish free and equal citizens, and guarantee their eudaimonia within the framework of a rational plan for the good life. 4 Thus in Aristotle’s view ethics as the theory and instruction of the formation of moral subjects is based on far-reaching preconditions, about which a great deal can be said. The systematic place for this context is apparently the discussion of upbringing and education (paideia), the role of which in the framework of classical ethics cannot be underestimated. What is of importance here is not the structure, function and content of elementary education (grammar, gymnastics, music and arithmetic), but rather the form and direction of moral upbringing. For Aristotle this must make use of an inborn part of our animalistic nature – the pursuit of pleasure. Thus our discussion of the Aristotelian teleology returns to the topic with which, in harmony with Foucault, we started our thoughts, and takes the form of an investigation of its preconditions. According to Aristotle there are processes of learning that primarily develop the human ability to enjoy something, or to experience pleasure from it. In this way we can sometimes learn that something is pleasant or agreeable. In particular it is possible to learn to enjoy something in an appropriate manner, to experience pleasure in it in an appropriate manner and above all to find the right activities pleasant. The upbringing and education of children and the young should guide this process of learning.73 This ‘learning’ is to be understood in an emphatic sense: it is not only a matter of knowing that certain actions are excellent, but also of carrying out these actions for the sake of their intrinsic value, and experiencing enjoyment and pleasure for the sake of this value, in doing so – and to achieve this requires time and practice, in short ‘habit’. The life of the young may essentially be characterised by feelings and the pursuit of immediate pleasure,74 but many of them experience shame (aidos) in the process, and therefore have a sense of what is appropriate and excellent which can be made use of in their education.75 As a feeling for noble and excellent action, shame goes well beyond plain fear of punishment, and is 73 See NE ii 3, esp. 1104 b3–13; also i 8, 1099 a17–21; ii 9, 1109 b1–5; iii 4, 1113 a31–3; iv 1, 1120 a26–7; x 1, 1172 a20–3. 74 NE i 3, 1095 a4–8; viii 3, 1156 a31–3. 75 NE iv 9, 1128 b10–12, 15–21.

88

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

an important condition for the transformation of the passionate pursuit of immediate pleasure into an intensive effort to experience pleasant feelings when performing actions of intrinsic value: ‘It is a regrettable fact that discussion and instruction are not effective in all cases . . . the mind of the pupil has to be prepared in its habits if it is to enjoy and dislike the right things.’76 Thus for Aristotle there are two steps in the transformation of the child into a mature moral subject, which he distinguishes terminologically with the concepts of ‘habit’ (ethos) and ‘teaching’ (didache). From this follows that the formation of moral subjects and the corresponding ethical work take on three different forms in Aristotle – depending on whether we are in the phase of habit, teaching or maturity. The phase of habit is concerned with more than just disciplining childlike appetites. Certainly the bad manners of children are a kind of ‘licentiousness’ that ‘passes all bounds’ and thus must be ‘rendered docile and submissive’, and certainly it is the aim of this that ‘the child ought to live in accordance with the directions of his tutor’, and that ‘the desiderative element’ aligns itself with reason.77 However, as was mentioned above, this obedience to the rules and directions of the tutor means that the child’s will and impulse are directed at a suitable end, which in turn is conceptualised by particular opinions about correct and excellent action. But the appetition (orexis) and the concept of the end (logos ho heneka tinos) are the origin of the choice (prohairesis), which itself is the origin of action (praxis).78 Thus by specifying appetition, end and choice, the pedagogical disciplining of children and adolescents during development of the ethical habit opens up to them the scope for moral action. In so far as children and adolescents are actually capable of ethical work, the main concern during this phase is to employ observation of the self and selfcontrol, aided by shame, to follow the ethical rules set by the tutor, and thus to gain access to the field of moral action. But the stable and excellent character that is an essential element of moral maturity and eudaimonia is still a long way off. For ‘the virtues we do acquire by first exercising them, just as happens in the arts’: ‘[a]nything that we have to learn to do we learn by the actual doing of it’, and so ‘we become just by performing just acts, temperate by performing temperate ones, brave by performing brave ones’.79 It is only by constantly acting morally that we form ourselves as moral subjects, and so we must direct 76 77 78 79

NE x 9, 1179 b23–6. NE iii 12, 1119 b1–17. NE vi 2, 1138 a31–1139 b5. NE ii 1, bes. 1103 a30 ff.

ethical teleology

89

our entire life according to a conscious and examined decision to act in this way. This decision requires the help of ‘teaching’, and initiates the second phase of moral subjectivisation. At this stage it is necessary to gain an appropriate concept of the best life and eudaimonia, and this is only possible within a framework of complex, extensive reflections, such as are presented in the Nicomachean Ethics. Part of these reflections consist of participating in the dialectic game of method which guides ethical reflection; but the main concern is constantly to relate individual decisions on courses of action to the concept of eudaimonia that dialectics produces and, vice versa, to test this concept on the basis of important types of such decisions. This procedure still demands discipline and self-control, but now at a higher level that is characterised by a higher degree of autonomy, and by its alignment to an overall model of how to live. Thus, both these phases of the moral formation of the subject include not only various forms of ethical work, but also interconnect the ethical work of the subjects that are being formed with forms of productive, regulative power. According to the Aristotelian view, at the most elementary levels of moral subjectivisation, that is of the moral formation of the subject, this regulative power should not just consist of private education and personal instruction, but also be conducted in the public, political sphere. The pedagogical disciplining of children and adolescents should be carried out in public educational institutions, and above all take into account the legislation of the polis.80 Here it is in effect the political rulers who exercise this regulative power that is inextricably enmeshed with growing ‘accustomed’ to excellent action. On the other hand, in the second stage of moral subjectivisation it is primarily the dialecticians and moral philosophers who introduce those individuals who pursue eudaimonia earnestly and at leisure to the rules of the theory of argument which are to be employed. It is by means of these rules that a complicated network of regulations is developed from the complex concept of a good, proven life, and the initiates must submit themselves to these regulations if they wish to practise those actions which finally lead to the stabilisation of a good character.81 However, within the systematic framework of the Aristotelian ethics there is a third stage of moral subjectivisation – the final stage in which subjectivisation is complete and the excellent character is irreversibly stabilised; this is the stage of moral maturity as an essential element of 80 See above all NE x 9, 1179 b20 ff.; Pol. iii 1. On this Nussbaum (1986), 346 f. 81 These can, of course, only be basic rules, which do not anticipate all details; see NE i 7, 1098 a21–5.

90

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

eudaimonia. This phase poses delicate questions about the categories of ethical work and productive power which interested Foucault, for it is reasonable to assume that the eudaimon (the happy person) has completed the ethical work, finished the stylisation of his conduct and has even freed himself from productive relations of power. However, this assumption would be an over-simplification. First of all we must remember that even according to the inclusive interpretation of eudaimonia scientific contemplation played an important part in Aristotle’s plan of the best life. To achieve this it is necessary to educate not just the ethical, but also the intellectual virtues;82 to play not just the game of truth of dialectics, but also that of the demonstrative sciences which Aristotle develops in a particularly complex and subtle fashion in the Analytica Posteriora.83 It is probably impossible to decide whether Aristotle could accept that a gradual education as a scientist, with its specific ethical work, and all its methodological forms of regulative power, was an element of eudaimonia; if so, then it would be easy to specify a particular case of ethical work that even covered the stage of the fully developed and formed subject. But even if we had to assume that, at the level of eudaimonia, scientific and contemplative education was complete, the precarious and unfinished business of scientific work could nevertheless be described within the framework of the category of a practice of the self. Support for this view is provided by the fact that Aristotle – contrary to a commonly held view – was not an epistemological fundamentalist, but always assumed that scientific results were ‘vulnerable’ and that scientific research was never finished.84 Even after the completion of a thorough scientific education, in the changeable business of scientific work, in the ups and downs of scientific success and failure, and in his discussions with his colleagues and friends, the scientist will have to be careful to maintain self-control, find the right measure and constantly act in accordance with the ethos of scientific contemplation.85 In the process particular attention must be paid to abiding by the methodological rules that have been laid 82 NE x 7 and ii 2, 1103 a14–17. 83 See on this Detel (1993) i, especially the summary 289–334; on the relationship between dialectics and demonstrative science see 151–7. 84 This anti-fundamentalist interpretation of Aristotle’s theory of science is defended in detail in Detel (1993). 85 In the Posterior Analytics Aristotle emphasises the point that ‘it is hard to be sure whether one knows or not; for it is hard to be sure whether one’s knowledge is based on the basic truths appropriate to each attribute’ (An. Post. i 9, 76 a26–8). Furthermore, there are numerous references to the various errors and mistakes which can be made in the course of scientific research (see on this Detel (1993) i, 328–32 with references).

ethical teleology

91

down, for although there will no longer be any clearly asymmetrical regulative relations of power, the circle of scientists involved or the community of cooperating researchers will regularly insist on following methodological rules in the form of reciprocal regulative power. This sort of argument also applies to the life of the eudaimon as a whole. Eudaimonia is an activity that may be based on a stable, excellent character, but nevertheless has to rely on external and relational goods, and is thus essentially vulnerable and fragile.86 For action takes place within the context of contingent events that cannot be fully controlled, or foreseen by humans. Accordingly the corresponding game of truth of practical knowledge is not as precise as demonstrative scientific knowledge.87 For this reason Aristotle criticises the ethical ‘theory of good conditions’ which maintains that the eudaimonia of morally mature humans who have a stable character cannot be restricted by any contingent events. In particular, he recognises that even in the field of excellent activity there can be moral conflicts that can only be solved with great difficulty, as is so often described in classical Greek tragedy. If we want to lead a rich and complex life, then according to Aristotle we must accept and deal with the contingent events and risks connected with our practical activity.88 The ethical work of the eudaimon consists mainly of dealing with this contingency without seriously endangering his excellent character. The delicate deliberative balancing act that constantly mediates between the various intrinsic values of practical activity, as well as the contingent events of life, is in principle never complete, and requires the highest level of practical and theoretical attention. But this effort generally helps the excellent character to stabilise itself in the tempest of life, and constantly to optimise the balance of pleasures. For Aristotle free and equal friends are the most important aids for the ethical work of the eudaimon.89 Not only do they contribute to reducing the contingencies of life by helping to maintain the external goods, or aspiring to ensure the justice of political activities; the excellent character of the eudaimon can also be reflected in them. We recognise ourselves in a friend, and so friendship is an important precondition for the eudaimon’s 86 The ‘vulnerability of the good character’ is the subject of Nussbaum’s (1986) excellent and successful book on the ‘fragility of goodness’, which is the source of much of what follows (on Aristotle see in particular chs. 11–12). 87 See NE i 1, 1094 b11–13; ii 2, 1103 b31–1104 a11; i 7, 1098 a21–35. On the uncertainly of life see e.g. NE i 11, 1100 b22–1101 a13 (further passages are referred to by Nussbaum (1986), 330–6). 88 For details see Nussbaum (1986). 89 On friendship see above all NE viii–ix.

92

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

reflexive knowledge.90 This self-reflection is particularly valuable when the eudaimon is faced with extreme misfortune, or confronted with moral conflicts that endanger his character and eudaimonia. The stabilising work that he must perform in this situation is made all the easier if his dealings with his friends can constantly remind him of the intrinsic value of his and their characters. As the scholarly community in the special case of scientific activity contributes essentially to maintaining the stability of good scientific research, so in the general case of practical activity the community of free and equal friends contributes essentially to maintaining the stability of eudaimonia. Reciprocal regulative relations of power exist between friends, and they constantly help them to renew their pledge of excellence of character in the face of the contingency of life and the fragility of eudaimonia. To sum up, we have seen that the model of domination and restriction that Foucault develops for the concept of ethical work in the ancient world does not go far enough, at least not in the case of the Aristotelian ethics. Enkrateia only plays a role in the first phase of ‘habit’, but even then this role is deeper than Foucault realises – opening up the broad and complex scope for moral activity. On the other hand, during the phase of ‘teaching’ and in the state of eudaimonia, it is the establishment and maintenance of the stability of the good character that matter. At the same time the attitude to pleasure should be rationally transformed, and the overall balance of pleasures optimised. There are various forms of regulative power that have an influence on this. From the perspective of Aristotle’s ethics it is not so much a case of a model of domination and restriction, as one of stabilisation and transformation that characterises ancient concepts of ethical work. The stylisation and the glory of conduct, which can sometimes accompany the stabilisation of the excellent character under a wide variety of contingencies, is more of a side-effect than an intrinsic goal which is to be sought earnestly. 90 One of the texts that bring this point out most clearly is to be found in the Magna Moralia (1213 a10–26).

3

THE SCIENTIFIC REGIMEN

According to Foucault, classical medical texts on the guidelines for an ideal regimen provide further instructive illustrations of the ways in which sexuality was problematised in the ancient world. One of the most important results of his analysis of these texts is that ancient dietetics develops a comprehensive perspective of bodily activity: it regulates nourishment, physical exercise, sleep, bathing and sometimes sexuality too. Foucault points to a number of important parameters that ancient dietetics relies on – the quantity, the overall state of the body, the length of time, the season, the general climate, general human nature; nor should the relationship of all these aspects to each other be forgotten.1 As Foucault sees it, ancient dietetics does not just suggest isolated rules for good health, but an overall regulation of life that covers both the mental and the moral aspect. Ancient dietetics pre-supposes not only that the care of the body and of the soul are mutually dependent on each other, it also maintains that mental discipline is a precondition of a successful regimen; above all, the primary aim of dietetics is not the unqualified prolongation of life, but the optimisation of the quality of life, that is the maximisation of pleasurable health within natural bounds. This goal depends on a concept of 1 The author of De Victu (the names of most of the authors of the works in the Corpus Hippocraticum are unknown) presents a particularly long and impressive list at the beginning of book i; see ch. 2.

93

94

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

the good life, for following rational dietetic regulations is an indication of the mental self-control which is an intrinsic part of a good life. Finally, in Foucault’s view ancient dietetics represents a reflective technique of existence, since essentially it relies on reflexive vigilance and the patient’s concern for her own body. The regulation of daily life prescribed by dietetics not only demands the observation of external factors, but also a ‘serial’ vigilance, a conscious and active application of the rules that goes beyond passive internalisation and open discussion with doctors.2 For this reason his discussion of ancient dietetics is of particular significance for the development and the details of Foucault’s programme of ethics. However, seen from the perspective of texts on dietetics of the classical period, Foucault’s diagnosis proves to be violent and one-sided, a thesis that I would like to justify and explain in detail. First of all (1) I shall demonstrate the ambivalent systematic role that dietetics and its associated practices of the self played in ancient health care, before (2) investigating the relationship between medicine and ethics in its historical dimension; then (3) I shall reconstruct how in the history of ancient medicine dietetics was reduced from being a concept of an entire art of life to physiological regulation as just one of many elements of the good life, and how this historical development was interlinked with the formation of a new domain of knowledge, a new scientific discipline and a new game of truth; finally, (4) I will show how, from the perspective of this ‘archaeological’ process, the attitude of ancient dietetics to sexual activity and its consequences can be identified. 1 In ancient medicine dietetics played a prophylactic as well as a therapeutic role. Although as Plato, for example, often emphasises, mental discipline may be required of the sick in following dietetic rules as part of a therapy in order to get healthy again in the long term, it is above all prophylactic dietetics that aims to regulate life as a whole and to maintain and optimise health. If Foucault understands ancient dietetics as an ‘art of existence’ or an ‘art of life’ that demands a serial reflexive vigilance and represents the ancient form of the constitution of the subject, then this applies above all to prophylactic dietetics, although this specific form of dietetics does not play a particularly important part in surviving classical medical texts.3 2 UP 101–8. 3 An example of purely prophylactic dietetics is the short treatise De Victu Sano in the Corpus Hippocraticum.

the scientif ic regimen

95

Certainly there were forms of prophylactic dietetics that aimed to provide a comprehensive regulation of the way of life; the treatise Hygieina, which is attributed to the doctor Diocles of Carystus, is a prominent example. But even in the ancient world there was a great deal of criticism of this branch of the discipline, and interestingly enough this criticism was based on the relationship of this form of dietetics to rational models of the good life. One of the leading sceptics was Plato. Foucault reads his reservations as an objection to exaggerated and excessive forms of regimen,4 but on closer inspection we can see that Plato rejects the (then) new form of dietetics as an art of life lock, stock and barrel and describes it (and not only its excessive application) as petty regulation for rich hypochondriacs who do not even bother to find a decent job.5 Plato seems to recommend a modest (vegetarian) diet,6 but that is because it corresponds to a natural life and not as the result of a practice of the self as an art of life. Plato’s main objection, though, is that the ‘petty regulations’ recommended by dietetics are not consistent with a good life directed at prudent dealings and reasonable political aims. Not only does personal subjugation to a tightly woven set of dietetic rules unacceptably restrict the economic and political activities necessary for the state, it also prevents the reflexive learning that was supposed to encourage the development of virtue: fussiness about one’s health, in excess of normal physical training . . . makes any kind of study or thought or private meditation difficult . . . you will always be prevented from exercising and proving your talents, and never stop worrying about your health.7

For Plato a regimen guided by medicine was diametrically opposed to moral subjectivisation as part of a rational model of the good life and is a clear symptom of lamentable and decadent political structures. The author of De Victu defends himself indirectly against such accusations, for he explicitly specifies dietetic regulations that apply to the majority, ‘the great mass’ who could only afford to feed themselves normally. He assumes that ‘the great mass’ must go about their business and so do not have ‘the chance of neglecting everything to concentrate on taking care of their health’. But beyond this he also provides instructions for 4 5 6 7

UP 105 f. Plato Polit. iii 405c–408d. Plato Polit. iii 372a–b; Tim. 89d. Plato Polit. iii 407 b–c. Aristotle warns against excessively intensive and exhausting athletic training in children and adolescents; see Pol. vii 16, 1335b; viii 4, 1338b–1339a.

96

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

the regimen of those who are ‘thus favourably situated’ because they are ‘convinced that neither wealth nor anything else is of any value without health’.8 These passages clearly reflect a critical debate on the function and value of prophylactic dietetics. But what is more surprising is that the author seems to accept that dietetics either – in the life of ‘the great mass’ – only plays a subordinate role, or else – in the life of ‘the free’ – pre-supposes privileged social status. At this point the political dimension of prophylactic dietetics as a comprehensive art of life is evident. However, Foucault does not seem to regard this aspect of dietetics as being particularly important, for he only mentions it briefly in connection with excessive forms of diet. But it is unclear how the excessive diet is to be differentiated from diet as an ‘art of life’, if the latter has to take into account ‘numerous elements in the physical life of a man, or at least of a free man’ and follows the course of an ordinary day, moment by moment, from waking up on through to the evening meal and the onset of sleep, with attention given along the way to the very first exercises, the ablutions and massagings of the body and the head, the walks, the private activities and the gymnasium, lunch, napping and another round of walking and gymnasium activities, oiling and massage, dinner. At all times, and encompassing all of a man’s activities, regimen problematized the relations to the body.

Thus it is no wonder that ‘dietetics . . . required what might be called a “serial” attention’ that the individual must constantly maintain.9 The dietetics which Foucault describes as an art of life or of existence is – it would seem – the excessive prophylactic form which Plato castigates with such intensity and derision, and to which the author of De Victu alludes. On the other hand, dietetics for ‘the great mass’ appears as an art of life neither in its prophylactic nor in its therapeutic form. Yet it is not only the role and function of dietetics as a whole that is ambivalent – the same is true of its status as a ‘practice of the self’. Certainly an excessive diet may have required constant observation of the self and reflection, and anyone who could assume that the dietetic directions of doctors were sensible and successful may indeed have regarded ‘hygienic’ observation of the self as a stylised practice of the self; but Foucault fails to see that – mainly non-medical – texts that propagate reflexive concern for one’s own body also have a social background that is not always particularly pleasant. The fact is that the reason why the system of ancient health 8 See De Victu iii 68 (beginning), iii 69. 9 UP 101 f., 106.

the scientif ic regimen

97

care – if indeed we can talk about one – required everyone to have a practical interest in medical knowledge and in self-observation of their physical state was that often no doctors were available. What is more, access to the medical profession was not controlled, there was no education in medicine and there were no professional licences. Generally doctors had a miserable reputation and were widely mistrusted, for their successes were on the whole less than modest; serious experts could often not easily be distinguished from quacks whose only interest was money.10 Thus much of ancient dietetic and medical techniques of the self had the form of a desperate programme of self-help rather than an aesthetic stylisation of life as a whole.11 In other words, when Foucault describes classical ancient dietetics as an art of existence, a practice of the self and an aesthetic stylisation, this description has an extremely restricted application. It is only true of the excessive prophylactic diet which was regarded as a reliable form of knowledge and regulated all aspects of life – but which was therefore of interest to free and affluent male citizens, and thus pre-supposed those ‘unequal, hierarchical and authoritarian institutions’ that were generally required for scientific activity in the ancient world.12 2 In order to justify his understanding of ancient dietetics, Foucault should be able to prove that dietetics saw itself in general as a doctrine that was not only concerned with physiological balance, but also with psychological equilibrium, mental discipline and moral formation. Significantly, Foucault does not rely on many texts to support this interpretation, and the few that he refers to are drawn exclusively from non-medical sources – Plato, Xenophon and the Pythagorean tradition.13 This is no coincidence, for not a single passage supporting his view is to be found in the texts that Foucault otherwise normally relies on, i.e. the De Victu and the De Victu 10 The Hippocratic oath (see The Oath in the Hippocratic corpus) can be seen as a reaction to this; it obliges doctors not only to pay attention to the well-being of the patient, and not only forbids abortion, euthanasia and the surgical removal of stones, it also regulates the passing on of medical knowledge, pensions for doctors, medical confidentiality and the social behaviour of doctors towards their patients; that is, it attempts to establish medicine as a profession (in the literal sense). 11 See on this Frede (1986); also Horstmannshoff (1990). 12 See Lloyd (1991b), 370. 13 See UP 101–5 with notes. Foucault seeks here to establish a broader understanding of ancient dietetics.

98

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

Sano in the Hippocratic corpus and the Hygieinia of Diocles of Carystus. Only the author of the De Priscina Medicina (probably Hippocrates himself) notes at one point that those who have nothing other than immediate pleasure in mind (and so who are completely without any control, we may add) have no use for a regulated diet – a statement that borders on the tautologous.14 This lack of textual evidence reveals far-reaching historical and systematical difficulties in Foucault’s work on ancient dietetics, and thus it must be commented on. It is true that Foucault’s strategy could be supported by pointing out that in the classical world doctors and philosophers – above all in the pre-Socratic period – always assumed the unity of body and soul, and thus there was traditionally a close connection between medicine and philosophy.15 The study of the soul was generally taken to be an analysis of the various life functions of humans, such as procreation, nourishment, growth, breathing, perception and thinking. Thus it was not unusual for doctors to write treatises on the soul, or to base the diagnosis and therapy of psychological disorders on physiological aspects.16 On the other hand philosophers regarded it as their task to treat the principles of illness and health,17 and from the fifth century onwards there was a widespread tradition that saw philosophy as a therapy of the soul – a therapy that, for the reasons indicated, cannot be clearly separated from concern for the body. In the Platonic dialogue Charmides, Socrates claims that Thracian doctors were forbidden by their divine King Zalmoxis to treat the body without the soul.18 Certainly the ‘material’ interpretation of the soul – for example the assumption of the pneumatic nature of the soul, or that it 14 See Prisc. Med. ch. v. 15 See on this and what follows Wehrli (1951); Frede (1986), 225 ff.; also Carella (1981). 16 For example, Asclepiades of Bithynia, Sopranus and Sextus Empiricus wrote works on the soul. Hippocrates is reputed to have defended physiognomics (see Gal. Progn. 1 xix 530 K; Corp. Hipp. Epid. ii 5, 1; ii 6, 1), and in the Corpus Hippocraticum mental states are often interpreted as the expression of physical ones (e.g. Aphor. ii 33; De Victu i 35, iv 86; Epid. iii 14, 98; an instructive example is the attempt to refute the thesis that hydrophobia is a purely mental phenomen in Cael. Aur. Acut. iii 13). 17 See Ar. Sens. 436 a17–22; Resp. 480 b28–30. Diogenes Laertius reports that Pythagoras was the first philosopher to be interested in medicine (viii 12) and that Alkmaion’s medical writings mainly took the form of natural philosophy (viii 83; see Ar. Met. 986 a22 ff.). But it was probably Empedocles who played the most important philosophical role in the history of ancient medicine. He attempted to build a physiology as a systematic basis for medicine on his doctrine of the four elements. Two of his most famous students (Pausanias and Acron of Acragas) were doctors, and all important pre-Socratic philosophers after Empedocles worked on physiology, in particular Anaxagoras, Diogenes of Apollonia and Democritus. 18 Plato Charm. 156 d–e; see on this Coolidge Jr (1993).

the scientif ic regimen

99

was to be found in the blood – will have contributed to these views and strategies.19 In the fourth century the opinion was still widespread that black bile caused madness. The compiler of the Hippocratic aphorisms talks of mental illnesses that are caused by climatic factors,20 while the author of the De Victu, who only works with the elements of fire (heat) and water (cold) in his dietetics, talks of mental and physical health as a balance between the two elements, and develops on the basis of the relative strength of the two elements a typology of human intelligence.21 Some Hippocratic doctors rely on physiological methods to treat particular psychic disorders, above all hellebore (elleboros), the effect of which was proverbial in the Hellenistic period,22 and in the De Victu of course a change in the fire–water relationship is recommended as a therapy.23 Thus medical concern for the body was at the same time philosophical concern for the soul. But it is going one step further to see the traditional concern for of the body–soul complex as concern for the control of pleasure and an ethical way of life – and it is this additional step that Foucault would have to prove in order to provide satisfactory support for his extended interpretation of ancient dietetics. As Foucault quite rightly points out, this particular complex can be traced back historically to the Pythagoreans,24 but he pays no attention to the route that the development of ancient medicine then took from these beginnings. This is unfortunate, for it is this route that puts ancient dietetics and its approach to the regulation of sexual pleasure in a very different light from that which Foucault would like to convey. Due to problems of the (non-) survival of ancient texts, it is hard to determine the content of early Pythagorean therapeutics, but the doxographic evidence provides enough proof to support the close connection between dietetics and ethics that Foucault also propagates. For instance, in Porphyrius’ and Iamblichus’ biographies of Pythagoras there is a close, almost automatic, parallelisation of body and soul,25 while Diogenes Laertius reports that the simple way of life (which was regulated by 19 The soul as pneuma in the orphic texts (1B11), in the Pythagoreans (58B1), in Xenophanes (21A1), Anaximenes (13B2), Diogenes (64A19, B5) (DK); located in the blood in Empedocles (31B105, A86, 10) (DK). 20 Aphor. iii 20; see Aphor. vi 56, Epid. ii 6, 1 and Pseud.-Ar. PP xxx 1 on melancholy. 21 See De Victu i 35. 22 Epid. vii 45; De Victu i 35; see Hor. Sat. ii 3, 82; Epist. ii 2, 137; on this also Pseud.-Ar. Physiognomica ii 35. 23 De Victu i 35–6. 24 UP 102. 25 See e.g. in the Pythagoras-Vitae Porph. 22 and 35; Iambl. 7, 34 and 31, 96.

100

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

dietetics) leads to both physical health and a sound mental state.26 Music, which was an important element in Pythagorean therapeutics, was also supposed to have an effect on body and soul.27 But above all we hear that the Pythagoreans assumed that correct dietetic conduct had an educational and a moral effect: the correct mixture (eukrasia) of eating and drinking had an important contribution to make to excellence of the soul.28 So the Pythagorean way of life as a conglomerate of dietetic and ethical behaviour was subjected to rigorous self-control and self-reflection, which demanded of the members of the school that they constantly gave account of their daily life.29 If a ‘serial vigilance’ over one’s own physical state and moral conduct as an integral part of dietetics and ethics is to be found anywhere, then it is in the Pythagorean bios. The close connection between medicine and ethics was probably not restricted to the Pythagorean circle, although evidence of this connection can hardly be found in the few pre-Socratic fragments that have survived. Democritus is an exception; his interest in the Pythagorean school is confirmed by his writing a treatise on Pythagoras, and he probably wrote a Dietetics of his own that was presumably of an ethical nature30 – to judge by his references to the analogy between medicine and ethics, to the dangerous effects of immodesty and to the application of a general ethics of the mean.31 In any case, the effects of the interconnection between ethics and medicine can still be clearly found in the classical philosophical texts of the fourth century, but they now take on a specific form that is quite instructive and allows us to draw important conclusions about the development of medical dietetics. In a famous passage in Plato’s Phaedrus Socrates draws an analogy between the art of medicine and philosophical rhetoric, before going on to maintain that Hippocrates’ method is exemplary for any investigation of the ‘nature’ of things. Just as medicine relies on its skills in making bodies healthy by using appropriate food and drugs, so rhetoric employs fine speeches and rules of conduct to impart convictions of excellence to the soul. In both cases the ‘nature of the whole’ must be taken into account, and generally one should follow Hippocrates in analysing whether things are simple or complex, what effects they have on other 26 Diog. Laert. Pythagoras 13. 27 See the praise of music in Iambl. 115, 64 and his report on ‘good health through music’ 25, 110–111. 28 See Gal. i 768; Iambl. 3, 13; also Plato Tim. 86b ff., Polit. 600a; Diog. Laert. viii 19; Porph. Vit. Pyth. 34 (see Frede (1986), 231). 29 See Iambl. 29, 165; 35, 265; Diog. Laert. Pythagoras 22; Seneca, De Ira iii 36. 30 See Dem. 68 A33, 68 R26 c (DK). 31 See Dem. 68 B31, 68 B159, 68 B234, 68 B191, 68 A167 (DK).

the scientif ic regimen

101

things, and what effects other things typically have on them.32 Thus it is above all in methodological aspects that medicine has a paradigmatic function for ethics as a therapy for the soul, but at the same time both deal with clearly separate matters: the division of labour of medicine and philosophy now becomes increasingly evident, even though as technai both disciplines have a great deal of methodological common ground. Plato says much the same in many other passages of his dialogues.33 Plato’s most influential student – and at the same time the most famous philosopher child of a doctor – continues this tradition of emphasising the analogy between medicine and ethics, though with much sharper methodological acumen. In the Nicomachean Ethics (particularly in books i, ii, vi and x) Aristotle draws a number of parallels between the methods of medicine and ethics: – they are practical, not theoretical forms of knowledge; their aim is not explanation but above all action; – they are concerned with individual, contingent situations; practical knowledge depends not only on knowledge of what is general, but also above all on knowledge of the specific circumstances of the concrete case; – they both pre-suppose universal knowledge of their field (the soul and the body); – they must be satisfied with a lesser degree of accuracy than mathematical and theoretical sciences.34 In a number of important cases Aristotle even uses medical analogies and illustrations to justify and support ethical theses; in particular the doctrine that humans have a specific function (ergon), the description and definition of excellence (arete) and finally the doctrine of the moral mean (meson).35 But perhaps a passage in the discussion of voluntary and involuntary action in the third book of the Nicomachean Ethics comes closest to Foucault’s interpretation (it is disconcerting that Foucault does not quote this passage): It is not only the vices of the soul that are voluntary; physical defects too are voluntarily incurred by some people, and we blame them for it. Nobody blames those who are naturally ugly, but we do blame those that become so through lack of exercise and care for their appearance. Similarly too in 32 Plato Phaedr. 270 b–d. 33 See e.g. Plato Gorg. 464a ff., 55e ff., esp. 502a–b; Nom. 857c–d. 34 See e.g. Ar. NE ii 2, 1104 a4–10; 5, 1106 b2–7; vi 7, 1141 a28–32; x 9, 1180 b8–10; b23–8; Pol. i 10, 1258 a10; iii 11, 1281 b38–1282 a4. 35 See in more detail Lloyd (1968); also Wehrli (1951), 83–7.

102

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

the case of physical weakness and disability. Nobody would criticise a person who is blind by nature or as a result of disease or injury – but anyone would blame a person whose blindness is due to heavy drinking or some other selfindulgence. Thus physical defects for which we are responsible are blamed, but those for which we are not responsible are not blamed.36

Aristotle seems to regard it as a moral rule that we should do all we can to remain handsome and healthy – or at least as much as is consistent with eudaimonia as a whole in so far as it includes physical well-being. But this does not mean that medicine becomes a branch of ethics, if only because much of what affects health is not within our power. The same applies to (therapeutic) dietetics, which was a constitutive element of medicine in the ancient world. What is more, Aristotle knows perfectly well that his – preferably methodological – analogies between ethics and medicine are no more than that, analogies.37 Not only are the aims of both disciplines different, in spite of their methodological common ground their scientific status is also quite different. If we follow Aristotle in dividing the sciences into the practical, productive (poietical) and theoretical,38 then ethics is a practical, but dietetics more of a productive (poietical) science;39 and regimen is only one of the means of achieving its end (health).40 But above all medicine is linked to (theoretical) sciences in a way that ethics certainly is not. A comprehensive account of theoretical scientific knowledge will conclude with a presentation of the principles of health and illness; and medicine – at least when it is ‘philosophically formed’ – will begin with a justification of its principles that is drawn from more fundamental sciences. In this way the general scientific theory of the universe and its principles arrived at by science coincides with the beginning of medicine as a theory.41 This relationship is partly due to the fact that one element of medicine deals with the general and the causes, and the other with the individual case. The doctor knows that for this type of fever (today we can say that the many fevers described by ancient authors are different forms of malaria) this drug is typically helpful (and he may even know why), but it is this fevered Socrates who is to be cured, and about whom the doctor therefore needs to know as Ar. NE iii 5, 1114 a21–30; similarly Plato polit. iii 405c–d. This is also emphasised by Lloyd (1968), 251. See Met. vi 1. See e.g. Top. i 3, 101 b8–10; 110 b18; i 5, 143 a3–8. An. Post. 12, 77 a36–41 makes it clear that medicine has its specific principles. 40 Top. i 3, 110 b8. 41 See Sensus et Sensibilia 1, 436 a17–b1; De Iuv. 480 b21–30.

36 37 38 39

the scientif ic regimen

103

many individual details as possible. The former is a matter for science, the latter of experience, and often highly experienced doctors have more success than doctors who have concentrated on obtaining more general knowledge.42 But general medical knowledge certainly has its foundations in theoretical contemplative science.43 The origins of ancient healing, and in particular of dietetic medicine, rooted as they are in the early Pythagorean school, are characterised by an intrinsic fusion of ethics and medicine. By the end of the fifth century only a faint echo of this remained, in the form of methodological comparisons and analogies between ethics and medicine. Perhaps dietetic concern for one’s own body and health from the philosophical perspective can be regarded as a component of reasonable models of the good life within the boundaries laid out by the structure of eudaimonia, and is thus an element of reflexive moral subjectivisation; but at the same time the division of the domains of ethics and medicine, and the separation of the disciplines, had finally taken shape. In the process not only had prophylactic, non-excessive dietetics assumed a particular and defined systematic role within the context of the pursuit of excellence and eudaimonia, but this also reflected a historical process which had led to the re-shaping of dietetics. Thus it is both an over-simplification and an exaggeration to follow Foucault in describing the diet that was recommended by the regimen of the new dietetics as an ‘art of life’, ‘that . . . was a fundamental category through which human behaviour could be conceptualized. It characterized the way in which one managed one’s existence.’44 Perhaps Pythagoras would have agreed with this phrasing, but a Hippocratic doctor, or even Plato or Aristotle, would have objected to such a simplistic assessment. 3 Foucault quotes the authority of Plato and of the author of the muchdiscussed Hippocratic treatise De Priscina Medicina (Ancient Medicine, probably written by Hippocrates himself),45 to arrive rapidly at the opinion that in classical antiquity regimen was regarded as an art of existence and life.46 In spite of their different accounts of the history of medicine 42 See Pol. iii 11, 1281 b38–1282 a7; Met. i 1, 981 a2–b9. 43 See also the first book of the early peripatetic Problemata Physica, which deals only with medical questions of a general kind in a causal sense. 44 UP 101. 45 See Pichot (1991), 501. 46 See UP 99–101.

104

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

and their opposing attitudes towards dietetics,47 both authors indeed talk of the dominance of dietary rules in contemporary medicine (that is in the first half of the fourth century). But it requires some sophisticated implicit arguments to progress from this factual basis to Foucault’s interpretation of dietetics as an art of life. First of all we would have to claim that Plato and Hippocrates were talking about the same kind of dietetics, and secondly pre-suppose that both authors saw this dietetics as a set of rules governing the way of life in the ethical sense. But unfortunately we can do neither. Plato was interested in the excessive prophylactic (or merely life-prolonging) dietetics, which he saw in stark contrast to the foundations of the good life, whereas Hippocrates was mainly concerned with non-excessive, therapeutic dietetics, which on the whole does not dominate the way life is conducted, and which therefore Hippocrates did not in any way link to the art of life. More important is the fact that in his violent fixation with a dietetics governing existence Foucault neglects an important systematic aspect of the domination of dietetics, one which is immediately obvious from an impartial reading of De Priscina Medicina. The very first sentence of this treatise is informative and programmatic: All who, on attempting to speak or write on medicine, have assumed for themselves a postulate as a basis for their discussion – heat, cold, moisture, dryness, or anything else that they may fancy – who narrow down the causal principles of diseases and of death among men, and make it the same in all cases, postulating one thing or two, all these obviously blunder in many points even of their statements, but they are most open to censure because they blunder in what is an art, and one which all men use on the most important occasions, and give the greatest honours to the good craftsmen and practitioners in it.

Hippocrates – let us assume that he wrote De Priscina Medicina – opens with a hard methodological attack on proponents of a new abstract, axiomatically organised medicine. He is less interested in factual problems of medical theory than in the methodological status of medical research and theory formation. From the second chapter on he propounds his main methodological thesis: the ‘old’ medicine had already found the suitable empirical methods for a rational development of medical therapies and theories. The starting point was age-old experimental experiences with 47 As Foucault (UP) indeed noticed, in Polit. iii 405c–408d Plato regards dietetics as a trendy, decadent form of medicine, and rejects it categorically, while Hippocrates sees dietetics as the heart of old medical science, which should be preserved and perfected for a variety of reasons.

the scientif ic regimen

105

various kinds of nourishment; the old doctors collected individual case histories, which they then sought to link causally with nutritional deficits. Empirical discoveries which were subsequently made about these causal links Hippocrates defines as the theoretical heart of medicine, which apparently includes immediate therapeutic implications (ch. 3). Above all, it is Hippocrates’ view that it is important for medicine as a professional discipline (techne) to find the right measure, or a good mean and balance, between excess and deficit in nourishment – the aim of medical dietetics was not the control of excessive appetites, but the equilibrium of the physiological forces. But since in each individual innumerable and disparate forces and counter-forces are at work, this equilibrium cannot be found abstractly and deductively, but only empirically and taking into consideration individual cases (ch. 9). For humans have individual natures (physeis), which can lead to different reactions (ch. 12). The old and tested medical science had already discovered that it is not the effect of simple, abstract materials that cause illness, but a specific complex imbalance of the physiological forces in relation to the particular nature of the individual patient. And a suitable diagnosis and therapy must be empirical, based on detailed case studies (ch. 14). To be sure, it cannot be denied that this old medicine had been shown to be deficient in many details, but this is not sufficient reason to change to a different method based on abstractions and axiomatisations (ch. 12). The empirical dietetics of the old medicine is in principle the correct factual and methodological approach, and the new medicine (the school of Cos, from which Hippocrates came) had proved that this methodological strategy could be developed further with considerable success (chs. 6, 8). One of the most important aspects of this development was the inclusion of the ‘exercises’ (ponoi), that is gymnastics and athletics, as well as baths, purges and emetics in the empirical methodology of medicine (ch. 4). Of course, empirical medicine should also include more general scientific theories about the nature of humans as a whole, but the axiomatists are wrong to believe that these elements of medical science can be deduced from non-medical scientific theories. On the contrary, such general scientific theories about humans (‘philosophy’) must be based on the sound foundation of empirical medicine (ch. 20). Above all the axiomatists proclaim heat and cold as simple, abstract medical principles, but it is easy to see that this does not get them anywhere, and can hardly explain anything (ch. 16–23). Thus the fact remains: the heart of medicine as a scientific techne and therapeutics is empirically based dietetics (ch. 24). Historically this techne had its origins in a doctrine for the sick, and had developed into a dietetic teaching for all humans. In

106

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

the course of this it had become structurally more complex, and was now applied more widely (ch. 5). It is only by looking at De Priscina Medicina as a whole that we can understand the extent to which it concentrates on analysing the scientific nature of medicine as a techne and can see how fascinating the problems that Hippocrates raises are. It is understandable that classical scholars have long been electrified by this treatise,48 in particular when we consider the fact that Hippocrates uses methodological termini technici (in particular hypothesis and eidos) that were only introduced into philosophy several decades later by Plato. But above all we see that Foucault’s diagnosis of dietetics fails to register what is a significant historical process. De Priscina Medicina documents the way in which dietetics became the heart of scientific medicine, its formation as a domain of knowledge, and how the empirical-casuistic methodology was introduced as the specific game of truth of scientific dietetics. The author of the main dietetic work in the Corpus Hippocraticum – the three books of the De Victu – also opens his treatise with comments that could be interpreted as methodological. However, he is a proponent of the new axiomatic medicine that Hippocrates attacks so virulently in De Priscina Medicina, and places particular emphasis on the view that scientific dietetics must assume a general theory of the nature of humans, which ‘narrows down the causal principle of diseases and of death among men, and makes it the same in all cases, postulating one thing or two’49 (as Hippocrates had so critically put it), that is fire and water, from which all things are constituted, including humans.50 He demands knowledge of the natural effects of these principles, as well as the study of the artificial effects which can be regulated by nourishment and exercises as part of the techne of medicine. The inclusion of exercises (ponoi) and their relationship to nourishment are explained in a typically abstract, axiomatic manner: nourishment replenishes deficits, exercises use up material – these are their basic effects, and so it is by skilfully mixing the two that a doctor can find the right therapy. But the author of De Victu does also grant 48 See e.g. Lloyd (1991a), who refers to numerous other works. 49 See De Victu i, ch. 2 and De Priscina Medicina i. 50 When talking about the abstract hypotheses of axiomatic medicine, Hippocrates always mentions heat, cold, dryness and moisture, never the four elements water, fire, air and earth. Thus it is doubtful whether his attacks were specifically aimed at the author of De Victu (this would only be the case if the elementary qualities were identical with the elementary substances). Hippocrates seems instead to have Alcmaeon and Philolaus, in other words the Pythagorean tradition of medicine, in mind. But it still remains unclear ‘who is attacked in De Victu’ (see Lloyd (1991a)).

the scientif ic regimen

107

empirical observations a place in the techne of medicine – the stars, the seasons and the individual constitution of the patient must be taken into consideration, and he even notes that in specific cases the medical therapy must often be inexact as the doctor cannot keep the patient under permanent observation. Clearly at this time the scientific status of dietetics was still disputed. In the classical period dietetics established itself as the dominant element of medical science, and thus – within the narrow limits defined above – a programme of subjectivisation was created that may not have represented a universal art of existence, but certainly did constitute at least one of the components of the moral subjectivisation that was connected to contemporary views about the pursuit of eudaimonia. But this process was also closely linked to the formation of a new domain of knowledge that included all the essential aspects of the analytic level that Foucault calls the ‘archaeology (of knowledge)’. In the case of ancient dietetics the levels of ethics and archaeology are inextricably linked. This should have been of particular interest for Foucault, and so the fact that he neglected it so completely and did not notice the good textual evidence for it is a clear indication of the restrictive nature of his early ethics programme. The formation of dietetics as the heart of scientific and empirically based ancient medicine can be seen as the separation of a scientific theory of regimen from its old intimate connection to ethics as propagated by the Pythagoreans. It is this development that is reflected in the faint echoes of the old connection found in Plato and Aristotle, who, as we saw above, basically restricted themselves to talking about analogies and already recognised the division between the disciplines of ethics and medicine.51 But 51 Wehrli (1951), 102, for example, is quite right to point out: here [sc. in the Pythagorean school] the internal bond between medical science and ethics, that we have postulated as a historical pre-condition of Aristotle’s medical analogies, seems to be realised. The origins of Pythagorean dietetics lie in the famous taboos on certain foods, which are themselves deeply rooted in magical beliefs. They then became part of the comprehensive regimentation of life which Plato knows as the “Pythagorean way of life” (Polit. 600b), in which superstition and ethical views are hard to separate. At this level following the old rules forbidding certain foods can be reinterpreted as the symbolic conformation to general self-control, or else it can lead to the ethically orientated dietics which Democritus also propagated. The process of secularisation is complete when all that is left is the mere attention to health, as is the case with the De Victu. If this picture of the process is correct, then it is one aspect of the separation of an independent science of medicine from what was originally a magical therapy with much broader aims. According to Wehrli the formation of regimen as a scientific domain of knowledge was linked to the rejection of dietetics as an art of existence and life, and its restriction to one

108

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

above all, the development was embedded in a broader historical process that could be described as the secularisation of the old Pythagorean and pre-Pythagorean medicine.52 The works of Homer and Hesiod tell us something about medicine in the eighth and seventh centuries. The gods are mainly responsible for plagues and illness, and so they can be appealed to for a cure, mainly using ritual incantations. This religious aspect is visible, for example, in the cult of Asclepius, which can be traced back to the great doctor of the same name who was later made a god, and is already mentioned in Homer. The sick flooded into the numerous temples of Asclepius in hoards. The main element of the treatment was a dream that the patients had while sleeping there, which provided advice on how they could be healed – advice that had to be interpreted by religious healers. But apart from prayers and sacrifices, the temples of Asclepius also offered regimen plans, purges, gymnastic exercises and diversions, and Homer’s accounts of numerous treatments (e.g. for wounds) confirms the specific professionalism of the healers. However, all of these methods could only be successful with the help of the gods, and they had to be thanked with prayers and offerings. Hippocratic medicine, on the other hand, gives the impression of being thoroughly secular – in spite of all the different approaches and methods that are to be found in the Corpus Hippocraticum. Incubation and the divine origin of disease are completely absent – even for the ‘sacred’ disease (epilepsy). Exorcisms and incantations are searched for in vain, and medicine is no longer part of the priest’s office, but a profession whose services are paid for privately, or by the polis. Aetiology and therapy are now described in a purely natural-philosophical vocabulary, and medicine seeks to establish its status as a scientific discipline. During the course of two or three centuries ancient medicine had undergone a thorough process of secularisation, in which famous doctors and scientists such as Alcmaeon, Empedocles or Philolaus and the south Italian schools of medicine will certainly have played an important role.53 But probably the most important were the Ionian schools of Cnidus and Cos, where the theoretical heart of ancient medicine which was to have such significant historical consequences was developed – the theory of the humours, which defined good health as an equilibrium of the humours and illness as a disturbance of this balance. of the important factors of eudaimonia; this process was part of the general secularisation of medicine in the ancient world. 52 E.g. Pichot (1991), 408–11, 481–2, 497–533. 53 Andriopoulos (1990).

the scientif ic regimen

109

The creation of dietetics based on scientific theory was doubtless a part – according to Hippocrates the main part – of the process of secularisation. During the course of this process dietetics may have developed from therapeutics for the sick to prophylaxis for everyone (as Hippocrates reports), and may on occasions have taken on excessive forms (as we hear from Plato and others); and certainly in the classical period dietetics included not just theoretical, methodological, empirical and therapeutic components, but also prophylactic recommendations which can be interpreted as the propagation of corporeal discipline and self-control. But that did not make it an ethically inspired universal art of existence (as it was at the time of the Pythagoreans). On the contrary, the removal of the intimate interconnection between religion, ethics and medicine, and the reduction of prophylactic dietetics to just one aspect of the eudaimonistic art of life, were a constitutive element of the secularisation of medicine. All in all the daily submission to dietetic rules covered only a narrow section of the domain of dietetics, and represented only one specific aspect of the moral subjectivisation that had as its aim the good life. However, at the same time the establishment of the canon of the rules of subjection was closely connected with the archaeological formation of a new scientific domain of knowledge, and a new scientific discipline. 4 Most of Foucault’s analysis of classical dietetics deals with the attitude to sexuality,54 although in the relevant texts little space is given to dietetic rules for sexual activity. It is above all in De Victu and the dietetics of Diocles that Foucault finds confirmation for what he had already found in De Priscina Medicina: it is not the sexual act itself or its forms that are problematised, but its context, its quantity and its circumstances. There is no universal boundary between what is allowed and what is forbidden, and recommendations oscillate between ‘more’ and ‘less’ according to the situation. Sexual activity is placed at the interface between individuals and circumstances, between temperament and climate, and it should follow a more or less restrictive regimen. It is a question of ‘how best to calculate the opportune times and the appropriate frequencies’, and thus requires a great deal of reflection and prudence.55 It is this specific orientation 54 See UP 97–108 on dietetics as a general guide to living, 109–39 on the dietetics of the aphrodisia. 55 UP 109–16, esp. 116.

110

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

of ancient dietetics that leads Foucault to the conclusion that this techne ‘created the possibility of forming oneself as a subject in control of his conduct; that is the possibility of making oneself . . . a skilful and prudent guide of himself, one who had a sense of the right time and the right measure’.56 But, as we have already seen, there is not the slightest shred of evidence for this ethical interpretation of dietetics in the medical texts of the classical period. The specific direction of dietetics that Foucault describes belongs in a completely different context. An analysis of this context also demonstrates that the domination and restriction model of the attitude to sexual pleasure that Foucault apparently believes he can recognise, not only in ancient ethics (in the usual sense) but also in ancient dietetics, is in reality of only secondary importance, just as was the case with the Aristotelian teleology. This can be confirmed in detail if we look at the dietetic texts in the Corpus Hippocraticum. In the dietetics for the healthy (De Victu Sano) sexual activity is not mentioned at all, and even in the four long books of De Victu short dietetic comments on sex are to be found in just five chapters.57 This relative lack of evidence, which Foucault elegantly ignores, reveals that compared with rules concerning nourishment, exercises, baths and sleep sexual regimen played a subordinate role in Hippocratic medicine. In many therapeutic and prophylactic contexts the Hippocratic doctors saw no reason at all to take action to regulate sexual activity. Of greater interest and significance, however, is a brief look at all the passages in De Victu that do deal with sexuality. The short description of the general effect of sexual activities in the second book of De Victu is particularly instructive.58 This book deals almost exclusively with the different effects and ‘powers’ (dynameis) of various dietetic tools, and does not contain any dietetic rules but presents generalisations from the sciences.59 The dynamis of sexual intercourse, the author of De Victu succinctly notes, consists of making our bodies thin, moist and hot – thin as a result of the emptying it leads to, moist as a result of the moist remains of sexual ‘exertion’ or ‘exercise’ (ponos) in the body 56 UP 133 f. 57 See De Victu i 35, ii 58, iii 68, iii 73, iii 85. In UP part 2, ch. 2 (The diet of pleasures) he only refers to ii 58 and iii 68. 58 De Victu ii 58. 59 The topics discussed include the following: the physis and dynameis of climate (ii 37), of the winds (ii 38), of food and drink (ii 39–56), of baths (ii 57), of oilings and sexual activity (both in the short ch. ii 58), of vomiting (ii 59), of sleep (ii 60), of exercises (ii 61–5); only in the last chapter (ii 66), which covers pain caused by exercise, are a few dietetic rules to be found, though not for sexuality).

the scientif ic regimen

111

(but which soon dry) and hot from the exertion and the secretion of moisture. The background to these comments would seem to be elementary empirical generalisations, such as: when X is emptied, X becomes thin, or when X moves and secretes moisture, then X grows hot. But what is of greater importance is that the dynamis of sexuality, like the analysis of the effects of all other dietary methods, is integrated into the theoretical framework of the two medical principles of fire and water that the author of De Victu pre-supposes (in stark contrast, in terms of both content and method, to the humour theory propounded by Hippocrates and the Cos school).60 The correct balance of the mixture of these principles in the body is health, any disturbance of this equilibrium is an illness and dietetic regulations take into account the effects of the individual dietetic tools, the circumstances and constitution of the patient in order to restore or stabilise the equilibrium. All other dietetic recommendations in De Victu that refer to sexual activity are to be found solely in this context. The most prominent example is the seasonal diet of sexual exertions.61 In winter it is of course good for our health to fight cold by activating the hot, and since the old feel the cold more than the young, this is particularly important for them. A whole range of measures is recommended: we should only eat once a day, and our food should preferably be dry and warming (wheat bread, roast meat, little fluid and few vegetables); a lot of different sports are important, from running to wrestling and intensive walks (even at night); we should sleep on a hard bed and take warm baths. And since we know that sexual exercises are warming (and after a while also dry), we should increase the frequency of our sexual activity in winter, the more so for the old. Similar considerations apply in the mild months of spring and the heat of summer; fire gradually gains the upper hand over the cold, and we must fight the heat in and around us. Clearly we should then in particular increasingly restrict our sexual activity – to a minimum in the hot summer months. Similar instructions are to be found in the first book of De Victu, but here it is our mental state that is the focus of attention.62 Our soul also consists of (particularly pure) components of fire and water, and when these components have the right mixture, then our intelligence is optimal: keen and active, but at the same time concentrated and persevering. If 60 See De Victu i 3–4, where the elementary dynameis of the two principles are also described: fire makes hot and dry, water makes cold and moist. 61 De Victu iii 68. 62 De Victu i 35.

112

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

the psychological component of fire is too dominant, then we are active, but unconcentrated; if water is dominant, then we are persevering, but ponderous. Fortunately a number of dietary measures can help here too, reducing the excess of either fire or water in our soul; accordingly the psychologically ponderous among us should, among other things, increase our sexual activity, while those of us who are unconcentrated, on the other hand, should reduce it. Finally there are two references in De Victu to sexual intercourse as part of dietetics for the rich, who can afford to think about regimen throughout the day.63 One of the typical problems of the wealthy is surfeit, which is discussed in detail.64 This is the result of a disturbance of the balance between food and exercise, i.e. an excessive dominance of food. Sexual activity brings only brief respite to the problem; the unpleasant symptoms very soon return stronger than before, as sexual intercourse in a state of an excess of nourishment blocks the intestines even more. Among other measures it is therefore sensible in this case to abstain completely from sex.65 Occasionally the rich also suffer from shivering fits and sleepiness during and after sport. Here too the cause is an imbalance between food and exercise, but this time resulting from too much exercise. In this case we should, of course, spend less time in the palaestra and running, and more time relaxing by taking warm baths, sleeping on a soft bed, getting drunk now and then, and – so that we do not lose too much moisture – only indulging in sex after the consumption of a modest and pleasant amount of wine.66 It is important to remember that this short review covers all that the author of De Victu has to say about sexual regimen. It seems quite clear that the parameters that, according to Foucault’s analysis, characterise ancient dietetics of sexual pleasure – the quantity, the circumstances, the right time – are extremely vaguely phrased. What is more, they have little to do with the imperatives of a mental and ethical discipline; rather they are the direct results of the factual and methodological demands of a dietetics that claims to be scientific. Furthermore, it is not a matter of the dividing lines between what is morally allowed and forbidden, since in Hippocratic medicine it is the balance of the physiological constituents of the body that is the telos of prophylaxis and therapy. But most importantly, the sexual dietetics of De Victu – if indeed we can talk of such a thing given 63 64 65 66

De Victu iii 69 ff. De Victu iii 72–7. De Victu iii 73. De Victu iii 85.

the scientif ic regimen

113

the sparseness of the scattered comments about it – does not follow the model of the domination and restriction of the pleasures that has the most dangerous and insatiable dynamic of all. Instead ancient dietetics employs sex as a means of stabilising or restoring the physiological balance, based on scientific investigations of the dynameis of sexual activity within the framework of theoretical medical principles – and this only occasionally. Not a single word is given over by the author of De Victu to consideration of whether its employment – or that of the other dietetic tools that are recommended – requires special moral or mental discipline. In fact he restricts himself to a descriptive, conditional statement: if humans wish to remain or become healthy, then they must follow particular dietetic rules – and this for reasons that can be deduced from theoretical medical principles and are the results of the way the dynameis of the various dietary methods are analysed. Health does not consist in the carefully calculated release of pleasures, but in their integration into efforts to optimise the physiological equilibrium – just as for Aristotle sexual pleasure should be stably integrated into the overall structure of eudaimonia, and thus into the orientation of our general balance of pleasures. As we have seen, Foucault points out that classical Greek authors saw considerable risks and dangers in sexual activity. According to his diagnosis both the act of sex and its effects were problematised:67 the form of the act was seen as a violent and threatening mechanism, which by the expenditure of an important life-giving fluid was to mediate between the death of the individual and the eternity of the species. Furthermore, the effect of the act was felt to be problematic because of its pathological consequences for individual organs and for the overall state of the body, as well as with regard to successful reproduction. This made the strict regulation of sexuality a necessity – in order to ensure the stability of one’s own body as well as the survival and optimisation of the species. Foucault’s argumentative strategy is easy to follow: the more obvious the awareness of the risks of sexual activity is in dietetics, the more plausible the domination and restriction model of sexual regimen within the framework of medical dietetics becomes. But this interpretation finds no support in the ancient dietetic texts; they say nothing about the metaphysical or the eugenic role of sexuality. Certainly a few passages in the Hippocratic writings De Victu, De Morbis and Epidemics deal with some of the problematic consequences of sexual activity, but it is revealing to look at the way in which they do so. The author 67 See UP 117–24 (Risks and dangers) 125–39 (Act, expenditure, death).

114

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

of De Victu, for example, notes that in patients who do not digest their food properly and have diarrhoea, but suffer no pain and are not really ill, too frequent indulgence in sexual activity can make their condition worse, and even lead to illness. Or that sexual intercourse can alleviate the symptoms of those suffering from surfeit – as noted above – for a short while, but in the medium or long term will make them worse.68 One of the illnesses discussed in De Morbis is ‘dorsal phthisis’: patients eat normally and do not have fever, but have a tingling sensation crawling down their spine, have nocturnal ejaculations, feel weak and have a heavy head and ringing in their ears. These symptoms occur frequently among newly weds and those who are sexually very active, and so the therapeutic recommendations include abstaining from sex for one year (no more exact aetiology is provided).69 Two of the sixteen cases in the third book of the Epidemics involve sexual activities (again without any aetiology):70 a certain Nicodemus in Abdera got a fever because he made love when drunk; the fever became worse over the next few days, the poor man got shivering fits, vomited, suffered from terrible thirst and pain all over, until the fever and all the symptoms disappeared after twenty-four days in a krisis. The case of an adolescent from Meliboea, who became overheated after he had indulged in drunken and sexual excesses over a long period, was even worse. He was punished with shivering fits, he felt sick and could not sleep, had difficulty breathing and his urine was oily. On the tenth day his skin was dry and taut and delirium set in. His general condition got worse on the fourteenth day and he was totally exhausted by the twentieth, before dying on the twenty-fourth. That already exhausts the few references in the Corpus Hippocraticum on which Foucault relies. Quite plainly the act or the effect of ‘normal’ sexual activity by healthy humans is not problematised in any of these passages. What they actually describe are always the consequences of dissipated sexuality, generally in connection with some other irregularity. No detailed aetiology is presented in these texts, but it is made clear that dissipated sexual activity significantly increases the excess of heat, and so can contribute to fever, to thirst or to dorsal phthisis. It is not sexual activity as such that is a dangerous source of risk (as Foucault implies), but immodesty and dissipation. This picture seems to fit the theoretical framework of Hippocratic medicine well: it is a lasting disturbance of the balance of 68 See De Victu iii 80, iii 73. 69 See De Morbis ii 51. 70 See Epid. iii 17, cases 10 and 16.

the scientif ic regimen

115

the constitutive elements of the body that is cause for concern, and such disturbances are generally, and in particular in the case of sexual activities, caused not by normal exercise, but by debauchery, or occur under pathological circumstances. There is no suggestion that due to its intrinsic nature sexual activity always tends to dissipation and therefore must be restricted by mental discipline. Nor are there any indications of this view in the early peripatetic writing, Problemata Physica. The majority of the thirty-two problems about sexual activity that make up the fourth book of this collection are composed and answered from the perspective of empirical science – such as how an erection, or feelings of hunger after sexual exertion, occur; why in summer women tend to sexual intercourse more than men, but in winter on the other hand men more than women; or why one has an orgasm more quickly when sober.71 As for the subject of the sexual act and its consequences, most of the explanations in book iv of Problemata Physica remain neutral, although in a few cases the effects on the general constitution are discussed. Some of these consequences are positive, others are negative. For example, sexual activity is only conducive to the healing of diseases that are caused by too much mucus, and similarly people who are naturally strong and well fed should be sexually active, as otherwise they will suffer from bitter bile.72 On the other hand sex is bad for the eyes, as it draws moisture out of them and those who indulge in sexual intercourse smell unpleasantly.73 These are the only specific negative consequences of sex that are mentioned in the Problemata Physica. Why sexual intercourse is generally exhausting and weakening is described and explained on only two occasions:74 because semen is a small, but very concentrated part of the body, and since it is distributed throughout the body, its discharge threatens to disturb its harmony. But since the way the problem is phrased says that this ‘mostly’ happens (‘why does one mostly become exhausted and weak during the sexual act?’), it is apparently a temporary problem that occurs immediately after sex, not a problem of general expenditure and danger. What is more, the moral level is virtually excluded in the Problemata Physica – even when the subject matter would lead us to expect moral comments. A prominent example of this is the discussion of the reasons for the tendency to active and passive sexual behaviour, which

71 72 73 74

See PP iv 9, 23, 25, 28. See PP iv 16, 29. See PP iv 2, 3, 12, 24, 32. See PP iv 6, 21.

116

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

although it is unusually long does not contain a single moral remark.75 However, there is a discussion of the shame felt not at the actual practice, but at confessing to a tendency to passive sexual behaviour (presumably only on the part of men): it is the result of those affected knowing that they have sexual wishes that go beyond the neccessary.76 Taken as a whole the aetiological discussions in book iv of Problemata Physica hardly give the impression of constant mistrust of sexual activity tout court. Even the explanation of why it is the most pleasurable of all activities does not contain any suggestion that this could be the source of a threatening dynamic that bursts all bounds.77 Thus in this context there is no need for mental or moral discipline that could keep the dynamic within limits. All told these observations produce a very different picture of ancient dietetics from that which Foucault proposes. Ancient dietetics as an art of life that was connected to moral regulation and the aesthetic stylisation of life as a whole, can be proved to have appeared in the pre-classical period in the broader context of its Pythagorean origins. But in the classical period only the dissipated, decadent prophylactic variety had survived, important preconditions of which were considerable political privileges and asymmetrical power relationships. On the other hand, dietetics for normal people, and for the privileged whose efforts to lead a good, proven life in general were earnest, had broken away from ethics in the course of the secularisation of medicine and forged links with the sciences. In this form it represented, on the one hand, just one of several characteristic aspects of moral subjectivisation as a whole that were intended to lead to eudaimonia; on the other hand, it formed itself as the heart of scientific medicine, the specific game of truth which became an express element of the methodological debates of ancient doctors and which was to transform dietetic therapeutics into a professional techne. Seen from this perspective sexual regimen played only a modest role as one of many dietetic tools, and its specific parameters were not so much the result of the special orientation of a moral subjectivisation as of its embedding in the scientific framework of medical principles; principles which excluded not only the moral problematisation of sexuality as an insatiable, menacing dynamic, but even its physiological problematisation as a weakening, health-threatening mechanism. Only excess and imbalance were dangerous, but this immodesty was not an intrinsic element of the sexual dynamic. The scientific dietetics 75 See PP iv 26. 76 See PP iv 27. 77 See PP iv 16.

the scientif ic regimen

117

of the classical period was not concerned with a model of the mental domination and restriction of sexual activity, but with a model of its sovereign embedding in the physiological equilibrium of the body, and thus with its empirically determinable contribution, not to the good life as a whole, but only to good health. In its structure the medical ‘teleology’ is certainly reminiscent of the Aristotelian teleology of the good life. Its end is the optimisation of the overall balance of the pleasures, as is prescribed in both dietetics and in the Nicomachean Ethics; and this optimisation is regarded as an important, constitutive element of a physiologically and scientifically describable ‘normal state’ of the body (and the soul), the main feature of which is the equilibrium of the constitutive elements of the body. Like the moral concern of Aristotle, so too the dietetic concern of ancient doctors is for any kind of threat to this balance that is the result of a pattern of behaviour that is inappropriate to the situation. Excesses of pleasure are merely a special case of inappropriate behaviour. It is only from this perspective that we can see to what extent the extremely limited subjectivisation that ancient dietetics recommended was not only intrinsically linked to an objectification within the human sciences, but also involved the formation of a productive form of power. For the scientific dietetics of the classical period had two demands: that doctors as teachers of medicine were to pass on the methodology of the medical techne to their pupils, and above all that doctors should ensure that within the framework of therapeutic practice their patients followed the scientifically founded physiological regulations of regimen. Apparently these are exemplary forms of productive regulative power, the application of which was intended ever to produce anew both the techne of medicine and a healthy physiological equilibrium.

4

T H E A S Y M M E T R I C A L R E L AT I O N S H I P

A crucial part of Foucault’s ethics programme consists in looking at the forms of moral regulation which were involved in the behaviour of the husband towards his wife, and of the pederast towards his young beloved, in classical antiquity. Foucault has good reasons for thinking that the way in which the role of dominant partners in asymmetrical relationships is problematised should provide particularly revealing insights into moral attitudes to sexuality. According to Foucault, when moralists of the classical period prescribe sexual moderation for the husband or the pederast in spite of his dominant social position, then this is an act of disciplining which has little to do with respect for the interests of an equal partner in a relationship, i.e. with the adoption of a ‘moral position’ in the Kantian sense. Rather it is a question of demonstrating a self-control that gives one’s life an aesthetic splendour and recommends one for a political career. ‘Govern yourself no less than your subjects and consider that you are in the highest sense a king when you are a slave to no pleasure, but rule over your desires more firmly than over your people’ is Isocrates’ advice in a speech delivered to the young monarch Nicocles.1 Nicocles’ own speech to his people (which Isocrates also wrote) is a detailed proof of his ‘self-mastery as a moral precondition for leading

1 Isocr. To Nicocles 29.

118

t h e a s y m m e t r i c a l r e l at i o n s h i p

119

others’, as Foucault puts it:2 political leadership can and should be unhesitatingly entrusted to a monarch who can demonstrate that he is master of himself. I would like to engage in a critical discussion of this interesting line of argument and look at the cases of both homosexual relationships and heterosexual marriage. We shall see that the Foucauldian interpretation cannot be applied universally, that further distinctions must be made as far as its content is concerned and above all that its white areas must be filled in. First of all I shall (1) look at the moral stylisation of pederasty, and prove that its interconnection with the aesthetic and political imperatives of self-discipline, as diagnosed by Foucault, remained limited to the narrow circle of the political elite, the public speakers. Then I shall (2) point out that there are good reasons to claim that in the case of ‘normal’ full citizens the moral problematisation of pederastic relationships was based on other ethical coordinates, and that these coordinates are related to the fundamental standards and criteria that determine the difference between the free and slaves. Even in classical antiquity this perspective led to deep-seated reservations about pederasty from which the philosophical authors of the period drew quite clear consequences, making a case for a general ban on physical love in pederastic relationships and suggesting that the structure of sexual dominance should be transformed into a form of regulative power. Next (3) I shall turn to concern about relationships of marriage, above all about the conduct of the husband and with particular reference to how this is articulated in classical texts on ‘economics’. In this context, we should remind ourselves of the fundamental and all-pervading social status of the oikos, and of marriage as the heart of the oikos within the ancient polis, as well as of the dominant role of the scientific tradition that sought to prove the inferiority of women to men. It is also illuminating at this point to study how the mythical figures of cheated wives were dealt with in contemporary poetry. Only by taking into account all these aspects of classical Greek culture does the complexity of moral concern for the conduct of the husband in Xenophon, Plato and in the Peripatos become fully comprehensible. Finally (4) the form of moral concern about pederastic relationships and relationships of marriage in classical antiquity also reveals that at the time it was not yet possible to consider how an interactive, relational sexuality could and should be regulated; sexuality was always seen as an asymmetrical relationship of power that could only be freed from its repressive structure, 2 UP 172.

120

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

and transformed into a regulative power relationship, at the cost of the complete renunciation of all physical contact. For moral authors of the classical Greek period the technological sublimation of relational, or even reciprocal, sexual relationships that characterises the development of both Far Eastern and Roman ars erotica, and has made such a lasting impression on modern discussions of sexuality, was totally unacceptable. 1 Foucault’s reflections on homoeroticism, and in particular on relationships with boys in classical Greek culture, are certainly among the most stimulating passages in UP – not least because these reflections involve far-reaching, provocative theses.3 Foucault assumes that during the classical period desire, and in particular sexual desire, was essentially seen as being directed towards everything that was beautiful – depending on the respective contingent circumstances and inclinations. As far as morality is concerned, sexual desire and love (eros) are all the more honourable the more beautiful, desirable and respectable the objects of desire are; and they are all the more depraved the more easily their demands can be given in to without reflection and without consideration of the object. From this point of view, the honour of the object and the control of appetite are important criteria and limits for the assessment of and the concern about eros, whereas the act itself, the specific tendencies and the sexual inclination of the appetite are of very little relevance. Under these circumstances the question of whether homoerotic relationships were legal or natural was unimportant and the differentiation between heterosexuality and homosexuality was more or less irrelevant. Hence homosexuality was not regarded as erotically deviant or even perverted. Seen against this cultural background Foucault is surprised by the proliferation of discourses on the love of boys and its reflective problematisation in the relevant classical texts. He seeks an explanation that goes beyond the unsatisfactory thesis of bisexuality or of the tolerance of homophile practices among the ancient Greeks. Of course like all other experts Foucault is aware that erotic discourses and the problematisation of homosexual relationships in classical antiquity concentrate on the special case of pederasty, that is the liaison between an older lover and a boy. But what is more important and informative, and Foucault is right to point this out, is the cultural regulation of this particular relationship: the manners which 3 See generally UP part 4.

t h e a s y m m e t r i c a l r e l at i o n s h i p

121

were to be observed, the strategies of courtship or the rights and duties of both parties. Above all it is a sort of institutional and spatial openness that shapes homophile practices: the lovers met in a common public space (the street, the market or the gymnasium) where they could both move freely and encounter numerous rivals; the boy is not restricted in his reaction to being courted and there is no institutional power that guarantees that the suitor will be successful. According to Foucault all these circumstances result in a substantial difference between pederastic relationships and relationships of marriage; they produce a completely different network of activities and practices of courtship, rivalry, persuasion or defence from those that were usual in the establishment or maintenance of marital or heterosexual relationships. At the same time, the complexity of their regulation demonstrates the enormous social concern that existed about pederastic relationships. Another important factor is time: the boy becomes a man, the lover an old man. The ravages of time threaten the beauty of the homoerotic partner. Certainly this was also true of heterosexual and marital relationships, but because of her thoroughly underprivileged position in classical antiquity the female partner had no possibility whatsoever of reacting to the changing situation. Due to their spatial and institutional openness homoerotic relationships were different in this respect, and so it is hardly surprising that in connection with pederasty the ancient moral texts, which were all written by men, should also include uncertainty, transitoriness and fragility among the matters of concern: this particular kind of relationship is threatened, and of limited duration, it often changes rapidly, and so moral reflection also considers how this threat can be avoided, how freedom from uncertainty and contingency can be gained – in short, how a pederastic relationship can be transformed into a stable symmetrical friendship. But Foucault does not regard all this as the core of the problematisation of pederasty and concern about it in the ancient world. The specific object of this concern – as one of the most important extant texts quite clearly shows4 – is the honour and the disgrace of the boy. In this situation the young man maintains his honour by making use of his physical and intellectual qualities to ward off rivals, by choosing lovers who promote these qualities and by not giving in to his suitors too quickly or unconditionally. He was certainly not always expected to reject advances; according to the 4 See the Erotikos, which is attributed to Demosthenes and was probably written about 330 bce.

122

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

Erotic Essay it was not considered shameful to submit to them occasionally. But it was important that the boy ensured that his own virtue was fostered, that his lover had suitable moral and intellectual qualities and that his own conduct was guided by these criteria. It would be shameful to be too submissive, or even to agree to degrading sexual practices for reasons of physical softness, sexual lust or out of desire for financial gain. It was of central cultural and moral importance that the boy himself experienced no sexual pleasure when once in a while, and in a manner that was carefully controlled, he gave in to the demands of his lover. He should present a picture of a handsome, modest being without any erotic needs of his own, who does not in any way establish a close relationship with his lover. He knows that he is physically and morally attractive, he will smile sweetly at his lover and value his services and advice. Occasionally he will allow him – his head bowed – to touch his face, his body, his genitals and very occasionally to satisfy his sexual desires between his thighs. But he will never allow his lover to penetrate him, never during any of these practices will he show sexual feelings, he will certainly not have an erection, and under no circumstances will he develop close physical contact with his lover or even allow it.5 For Foucault the cultural and moral background to this restrictive problematisation and regulation of pederastic relationships is clear. In the classical period sexual practices were discussed in the same categories as social practices and relationships in general: as relations between high and low, dominant and dominated, active and passive, penetrating and penetrated. The situation of the young beloved in pederastic relationships was so precarious because the boys and adolescents would soon become free men, and as full citizens free men – since they were members of the social elite in their polis – had a dominant and important role to play in the political and private domain. Hence the function of youths as attractive objects of desire in pederastic relationships should under no circumstances lead to the internalisation of passive, submissive and lascivious behaviour; rather it should remain consistent with their up-bringing to manful conduct and arete, that is to activity, competitive action and self-control. The moral problematisation and concern practised by ancient moralists with regard to pederasty and the role of the young beloved were quite clearly directed at the problem of this consistency, and the regulations and restrictions they propose are meant to ensure it. It was not the pederastic relationship as such that was regarded as unnatural, but rather the passiveness and 5 This cultural image is mainly drawn from vase paintings, as Dover (1978) has shown.

t h e a s y m m e t r i c a l r e l at i o n s h i p

123

lack of discipline of young men, and seen from this perspective it is easy to understand that ancient texts maintain an embarrassed silence about homoerotic relationships between free men of the same status and age. To fully appreciate this picture it may be helpful to point out that the historical and social background of homoerotic relationships between an older, active lover and a younger, passive beloved is today generally seen in the old Doric rites of initiation that introduced adolescent males into the society of men. The myth of the rape of Ganymede by Zeus reflects the Doric custom in which an older man abducted a chosen young man in a pre-announced, ritualised rape. He then spent two months with him in absolute isolation, instructing him in hunting and other male skills. Homoerotic sexual contacts undoubtedly accompanied this introduction to the rules and virtues of grown men, and were thus also socially tolerated. At the end of this rite of initiation the young man received from his mentor the weapons of a warrior, a cup and a bull. He now belonged to the warrior class, had to live in the men’s house and to celebrate his transformation in a ceremony during which he sacrificed the bull. When they went to war the mentors probably stood next to their charges in the line of battle in order to stabilise the virtues into which they had initiated them.6 The picture that Foucault presents of the moral problematisation of pederastic relationships in the classical period is certainly not inappropriate, and there is a great deal of evidence to support it. However, he often tends to overestimate the cultural implications of the evidence he relies on.7 For example, Aeschines’ long speech Against Timarchus, which Dover had already identified as one of the most important texts for the analysis of ancient homosexuality, hardly includes any evidence for the precarious 6 See Dover (1978); Patzer (1982). Heitsch provides a summary with textual references and other literature in the appendix to his translation and commentary of Plato’s Phaidros ((1993), 234–6). However, it should not be overlooked that Bethe had already justified and defended this historical explanation in an article that was quite remarkable for its time (he published it in 1907). Bethe also refers to an old tradition whereby the man’s sperm was regarded as the seat of the soul and of character, or at least as a psychological and ethical element, and hence the act of homoerotic love could be seen as the direct passing on of male arete (see the reprint of this paper in Siems (1988), 17–57). Sometimes classical Greek homosexuality is described as pseudo-homosexuality, because it does not fulfil particular characteristics and aetiological criteria which psychiatry has defined for homosexuality; see e.g. Devereux (1988) (Devereux’s analysis – he was a trained psychiatrist and had a good knowledge of classical Greek literature – suffers from the fact that it seriously considers homosexuality as a ‘perversion’; it is hard to understand how such an article could have been included in a volume of papers representative of work on the subject; Siems (1988)). On homosexual practices as part of male rites of initiation in other societies; see e.g. Herdt (1982) and (1984); Godelier (1987). 7 This is also emphasised by Cohen (1991), 171–3.

124

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

role of the boy in pederastic relationships, which is what Foucault wishes above all to prove. Aeschines accuses his opponent Timarchus of sexual prostitution on a wide scale, and this in spite of the fact that he is a grown man. Furthermore, Aeschines also regards Timarchus’ sexual conduct as a specific expression of his general lasciviousness and lack of discipline; he pursues every kind of luxury and so unscrupulously squanders his father’s fortune. But Aeschines’ main aim is to deny his political opponent Timarchus the opportunity of speaking in the assembly. This political dimension of the speech points to an aspect of the regulation of pederastic relationships that Foucault tends to play down, but which at the same time leads to a radical cultural restriction of the domain to which this regulation is applied. In general it is important not to forget how difficult it is when dealing with cultural questions to find representative texts that provide reliable information about what an entire society, or even just a social group, thinks and practises. This is particularly so of the views and practices of the classical period about matters of sexuality and eros. In classical Athens it was not Socrates, Plato or Aristotle who were the authorities people consulted on cultural questions, but rather Homer, Hesiod, Tyrtaeus, Solon, Sophocles and Euripides. The extant political speeches that were made in the assembly or the law-courts make this quite clear.8 However, the social regulations suggested by rhetorical texts only apply to a small elite of full male citizens, and possibly they tell us little about actual social practices.9 In forensic speeches the moral code of active manliness, which Foucault is right to regard as one of the central ideas of many classical texts dealing with ethical problems, features particularly prominently in the context of the antithetic concepts hoplites–kinaidos. All full male citizens, and only they, were well-equipped soldiers, and one of the main concepts of the ideal ancient constitution (politeia) obliged the political elite as the upholders of democratic rights to defend their polis – in other words to be prepared to sacrifice their life and their person in honourable combat for their city. The citizen soldiers received regular military training between the ages of eighteen and sixty, and no matter how great the political rivalry between them, on the battlefield they had to support each other. Military 8 Lycurgus, Against Leocrates 82–110; Aeschines, Against Timocrates 141–53; Demosthenes 19, 243–56 are good examples: it is also significant that Aristophanes generally pokes fun at the ‘philosophers’, rather than regarding them as intelligent interpreters of political and moral events (see e.g. The Clouds). Similar views are also known from other comedians (for example in Ameipsias’ Komoi, or Eupolis’ The Flatterers). For details see Gailly (1946). 9 All of this is emphasised by Winkler (1990) (see esp. 171–4); similarly Cohen (1991).

t h e a s y m m e t r i c a l r e l at i o n s h i p

125

service demanded above all extreme self-discipline. Hunger, thirst, sexual abstinence, lack of sleep and fear of death all had to be endured honourably and without complaint. Warfare was the main area in which the citizen-soldier (hoplites) of the ancient polis had to exercise and demonstrate self-control (enkrateia) and it was above all in this respect that he had to prepare and prove himself as the ‘master of his self’.10 However, the real significance of disciplined physical training as the foundation of defence and self-defence only becomes clear when it is seen in connection with the moral of the physical inviolability of the hoplites. According to Athenian law the body of a hoplites – in contrast to that of women, metics and slaves – was absolutely sacrosanct. At worst his property could be confiscated, but as a physical person he was not to be attacked under any circumstances.11 This inviolability was one of the main characteristics that separated the free citizen from slaves and others who were not free; to assault a male full citizen was regarded as offering the worst insult imaginable, for it meant treating a free man like a slave.12 The moral of physical inviolability also implied a ban on sexual subjugation, for sexual aggression was regarded as a particularly dramatic form of enslavement. Thus sexual dominance and activity were intimately connected with the fundamental ideal of the free citizen-soldier, and at the same time impinged on the disciplined physical training and mental self-control which were an essential precondition of this ideal. In classical antiquity not only was the scientific discussion of nature dominated by pairs of predicative opposites, so too was the public moral debate. In the domain of social practices it has been pointed out correctly that a zero-sum game was played: ‘People do not like to honour others, for they assume that in doing so they are themselves deprived of certain things.’13 In moral and political competition the winner achieves domination at the direct cost of the loser. Hence the compliant (kinaidos) or soft man (malakos) is regarded as the absolute opposite of the citizen-soldier (hoplites): he is cowardly and undisciplined, allows his body to be touched or even penetrated and refuses to participate in political competition. Thus deep down in his nature the kinaidos is a slave, and when a freeborn 10 See Winkler (1990), 178 f. 11 See e.g. Demosth. 21. 178; 21. 180; 22. 53–5. 12 The enslavement of free women and youths naturally implied the possibility of sexual assault and was therefore punishable by death; see Deinarchus 1.23 for three such cases. In war sexual aggression was the most powerful expression of victory. 13 Anonymus Iamblichi (sophistic excerpt), see Fragmente der Vorsokratiker (Diels and Kranz ii, 89 (82) p. 400); see also e.g. Plat. Nom. i 626b.

126

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

citizen shows even the remotest sign of such behaviour, then this is a cause for the greatest moral concern.14 Thus giving way to pleasure, the loss of self-discipline and the abandonment of competitive behaviour were not only damaging to a political career – as Foucault is quite right in noting – but were at loggerheads with the fundamental model of the citizen-soldier and his own unique manliness. On the other hand, steadfastness in the face of desire, the stabilising of self-control and the acceptance of political competition not only involved the stylisation of life, but also the fulfilment of a suitable life as a free full citizen of an ancient polis. However, strictly this only applied to the full male citizens, for the contrast between hoplites and kinaidos is at the same time the contrast between the masculine and the feminine man, and is thus based on the polarity between man and woman: ‘The contrast between hoplite and kinaidos is a contrast between manly male and womanly male, and therefore rests on a more fundamental polarity between men and women.’15 Like the kinaidos, the woman also desires to be submissive – a view that is at the heart of ancient misogyny.16 Thus an inability to exercise self-control, or the loss of self-discipline by a free full male citizen, were the equivalent of a sex change. In the classical period the sexes were construed mainly socially. The pair of opposites, hoplites–kinaidos, is originally no more than a cultural leitbild that was sometimes employed in discussion or public speeches in the ancient polis. But this says nothing about the exact way in which the leitbild was interpreted or applied.17 In this context it is worth taking a closer look at dokimasia, the procedure by which the lifestyle of all full male citizens who wished to run for public office, and therefore to partake in the lottery for office, was investigated. A similar procedure was followed when one became a full citizen (when young men were accepted as ephebes into the body of full citizens or when full citizenship was granted). The number of people to be investigated each year presumably prevented a detailed enquiry into the individual candidates’ private lives, but it was accepted practice that anyone present could make an accusation.18 More significant is the fact that there are a number of indications that above all the ‘public speakers’ (rhetores) were subjected to a particularly detailed See Xen. Mem. 1.5.5; 1.3.11. Winkler (1990), 182. See Halperin (1990a), 271. This question is dealt with in Winkler’s excellent study (1990), from which much of what follows is drawn. 18 See Ar. Ath. Polit. 55,4: ‘Would anyone like to accuse this man?’ On routine cases of dokimasia see Demosth. 40.34, 59.3, 59.72; Aisch. 3.31.

14 15 16 17

t h e a s y m m e t r i c a l r e l at i o n s h i p

127

interrogation into their private lives. The public speakers did not form a specific official body, for in ancient Greek democracies all full male citizens could in principle speak in the popular assembly or the law-courts. But it was de facto always the same small circle of people who regularly submitted motions and delivered speeches – ‘the usual and established rhetores who manage the political establishment of the polis’.19 Demosthenes draws an explicit contrast between the public speakers and the ‘majority’ that consists of ‘private persons’ (idiotai).20 Hyperides replies to an accuser: ‘you treat Euxenippus, who is a private person (idiotes), as if he had the rank of a public speaker’.21 Whoever entered the political arena like a boxer in the ring played a dangerous game: ‘The life of a private person is secure and free from risk, but that of a politician (politeuomenoi) is precarious, open to attack, and full of trials and misfortunes every day.’22 In his famous speech against Timarchus Aeschines notes: ‘The law does not investigate those living their private lives (idioteuontes), but those engaged in politics (politeuomenoi).’23 Timarchus had prostituted himself sexually and squandered his paternal inheritance, but such accusations were only so serious because he was one of the public speakers.24 Demosthenes traced the regulation back to Solon: it was not Solon’s intention to punish all persons who led a debauched life, but he did pass a law that forbade men who had prostituted themselves sexually to speak in the popular assembly (and thus engage in political activity), ‘for he knew – I say, he knew that of all states the most antagonistic to men of infamous habits is that in which every man is at liberty to publish their shame. And what state is that? A democracy.’25 When Xenophon portrays the Epicurean Aristippus as a man who has a thoroughly uncontrolled character, and ‘with regard to appetite for food, drink, sex, and sleep, as well as avoiding cold and heat, and also hard work, has absolutely no enkrateia’, it is significant that at the same time Aristippus openly declares that he does not wish to be one of those who have political ambition and wish to rule the city. In doing so he also accepts the division See Demosth. 22.37; Deinarch. 1.71. Demosth. 22.37; see also Aisch. 1.165. Hyper. 4.30. Demosth. 10.70; see also Hyper. 4.9: although for public speakers the reward for their efforts was often fame and honour, nevertheless they permanently exposed themselves to great risk. 23 Aisch. 1.195. 24 The speech Against Timarchus is the only surviving dokimasia of a public speaker. See Dover (1978), 29 on this. 25 Demosth. 22.30–1. 19 20 21 22

128

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

suggested by Socrates’ into those who have self-control and can therefore rule others, and those without self-control who have no wish to exercise domination over others. For Xenophon Aristippus’ undisciplined character is a completely satisfactory explanation of his refusal to participate in political contest and the competitive life of the ‘public speakers’ and ‘wish for a life of the greatest ease and pleasure that can be had’.26 Aeschines refers to a large group of full male citizens to whom the same seems to apply: they ‘commit a sin against their own body’ and like Cephisodotus they waste their overwhelming beauty in a life without honour. Aeschines refers to Solon, who thought that ‘anyone who when young stood aside from the ambition for noble honours because of his shameful pleasures . . . should not be eligible to share in honours when older’.27 All in all the picture is quite clear and consistent. No texts claim that in the common people lack of discipline, and in particular sexual lasciviousness, was per se grounds for criticism or was considered to be problematic. Even in the dokimasia the ideology of the hoplites–kinaidos was only applied to the very small circle of those full male citizens who claimed to belong to the political leaders, the ‘public speakers’, or who already played a leading role in the political life of the polis. Within the framework of this ideology and its application sexual lasciviousness was regarded as only one of several factors, and it was problematised above all in its extreme form of sexual prostitution, which itself had more to do with money and power than with sexual desires. The problematisation was expressed ‘not in terms of sexual behaviour but in terms of political participation’.28 2 In the previous section it was shown that the moral regulation of pederastic relationships stimulated by concern for the threat to an active, disciplined manliness was restricted to the elite of the ‘public speakers’; in this context it was primarily used as a weapon in conflicts over political participation. The restriction to just this function clearly weakens Foucault’s thesis that the moral regulation of pederasty was at the core of a universal stylisation of life for all full male citizens of an ancient polis – for most of them never made use of their political possibilities to assume influential administrative posts, or to participate actively and table motions in the public assembly; 26 See Xen. Mem. 2.1.9; see also 2.1.1, 2.1.2, 2.1.7, 2.1.8. 27 Aisch. Against Timarchus 155–60. 28 E.g. Winkler (1990), 195.

t h e a s y m m e t r i c a l r e l at i o n s h i p

129

and if they did then it was only at a very subordinate level. What is more, it would seem that on the whole the moral references to enkrateia in pederastic relationships were instrumentalised politically by the elite circle of influential public speakers, and were not regarded as a goal in their own right in models of a good life. But there are indications that there was also a social regulation of pederasty independent of the imperatives of a political career – in other words also for the large group of young men who did not strive for political influence. This suggests that there were other interesting forms of the problematisation of homosexual or pederastic relationships which, regrettably, Foucault did not pursue. It would seem to be the case that in Athens the law protected free boys and young men from homosexual molestation as long as they still attended the gymnasium. This law already had an ancient tradition in the classical period, and was attributed to Solon and Dracon. Schoolboys were always accompanied by paidagogoi, who looked after them in public; they were equipped with a rod and one of their tasks – which is illustrated on a number of classical vases – would have been to ward off sexual approaches.29 It was illegal to arrange relationships for youths,30 and there is some evidence that actual sexual satisfaction in a pederastic relationship was not only considered inappropriate for the young beloved, but also for the older lover.31 Nevertheless, it was considered fundamentally natural and understandable that grown men should feel physical sexual desire for attractive young people, irrespective of whether it was for young girls or young boys.32 For example, when an Athenian admitted that he was in love, it was normal to ask whether he was in love with a boy or a girl.33 Given the segregation of the sexes, and the fact that most women were kept indoors, free citizens encountered many more young men than young women in public.34 But there were significant moral and legal reservations about concrete sexual acts in pederastic relationships, in other words not about the desire but about its satisfaction. It is not without reason that the situation has been compared to the attitude towards pre-marital sex in heterosexual See Aisch. Against Timarchus i 6–12; on this Lacey (1968), 157 ff. Aisch. Against Timarchus i 13–14. See e.g. Lysias iii 4, 26, where the lovers feel ashamed after their victory. See on this Dover (1974), 213 f. See e.g. Aristophanes, Frogs 55–7; Theocritus ii 44, 150 (a young girl is unsure whether her lover has left her for another girl or a boy). 34 Above all of course in gymnasia and wrestling schools; see Dover (1978), 54–7.

29 30 31 32 33

130

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

relationships in the eighteenth and nineteenth centuries. Here too the man’s desire was regarded as quite natural, but its satisfaction as a reprehensible Pyrrhic victory for the man and as morally disastrous for the young woman (in both cases it was usual to brand the all too compliant young beloved as a prostitute, even when it was obvious that she had indulged her lover’s sexual desire not for money, but out of love).35 Given that, as Foucault also emphasises, the young beloved was not to experience any sort of sexual arousal in pederastic relationships, classical Athenian society seems to have been sensitive to the fact that at least boys and schoolboys should not simply be sexually instrumentalised by the natural sexual desire of older men. This sensitivity was apparently related to the accepted norm – which was mentioned in the previous section – of the physical inviolability of free full citizens as a criterion destinguishing the free from slaves. Thus, beyond the imperatives of a political career it was not so much the rule of active manliness and self-discipline that was the basis of the moral problematisation of actual sexual acts in pederastic relationships, but the ban on the physical instrumentalisation and exploitation of young full citizens.36 A number of ancient texts reveal that these moral reservations about pederasty also made their mark on a moral zero-sum game that employed the dichotomy between honour and disgrace.37 A man’s sexual ‘victory’ over his beloved brought the pederast honour, and so was celebrated publicly; but at the same time submission brought the beloved an equivalent loss of honour and disgrace. Since just what sexual practices the pederast and the beloved enjoyed in private often remained secret, even the smallest indication of victory or submission provided by the public behaviour of the couple assumed enormous importance, and could decide between honour and disgrace.

35 See on this Dover (1974). Classical comedy never criticises men who prefer beautiful young men to beautiful young girls, but is always damning of men who subjugate themselves to the sexual wishes of other men, see Dover (1978), 137. 36 The penal code treated the physical assault of a full citizen as hybris, that is as an attack on the community, and it was often punished by death (see Ar. Rhet. 13784 for a definition of hybris; see also NE 114b29). Demosthenes (xxi 180) reports that a certain Ctesicles, who had assaulted a personal enemy with a whip, was executed in spite of the fact that in his own defence he pointed out that he had been drunk. In fact, on the contrary, the judge took his drunkenness to be an additional reason to sentence him to death precisely because this indulgence led him to commit such a crime (for more details see Dover (1978), 35). More recently Cohen ((1991), 174–82) has reviewed the entire complex of legal concerns about pederasty. He reaches the same conclusion, and adds further interesting examples. 37 For more details see Cohen (1991), 186–201.

t h e a s y m m e t r i c a l r e l at i o n s h i p

131

The moral and philosophical authors of the fourth century were quite explicit and radical about the attitudes and views that were at the heart of these social mechanisms. One of the most influential texts on the subject is Xenophon’s Symposium, in which – just as in Plato’s dialogue of the same name – there is an atmosphere of homoerotic flirting, and there are occasional insinuations of actual homosexual relationships.38 The central speech, Socrates’ long address about eros, makes no explicit or implicit reference to the dangerous situation of the young beloved in a pederastic relationship. But in the context of this ‘problematic relationship’ that Foucault emphasises, it is significant just how clear cut the distinction is that Socrates makes between ‘vulgar’ and ‘heavenly’ Aphrodite. Love of the body or love of the character, submission or chastity, lasciviousness or shame, lack of discipline or self-control – these are the conceptual opposites that Socrates uses when explaining the difference between the vulgar and the heavenly eros. Love of a beautiful body, as Xenophon’s Socrates emphasises, is only interested in immediate gratification, is quickly satisfied physically and so changes constantly; it diminishes with time, it does not help the beloved in difficult situations and is not interested in his advancement – it exploits and takes advantage and so is not reciprocated but leads to hectic activity and loneliness. In the context of the zero-sum game love of the beautiful character must be exactly the opposite; it is longlasting and permanent, it is never fully satisfied, it increases with time and aims to promote and support the beloved – thus it can be reciprocated and mutual and so free the lover from loneliness and psychological slavery. The Socrates of Xenophon’s dialogue explicitly restricts most of the regulation of pederasty of the kind that Foucault identifies to the heavenly Aphrodite and the divine eros. Only he who renounces physical pleasure can love well-trained and strong-willed instead of lascivious and effeminate young men, will pay attention to the perfection of his beloved’s character and gain the public trust that is constructive for a political career – for neither can a lascivious lover help his beloved to honourable behaviour nor can an honourable lover love lascivious men. The analogy between the theory of eros presented by Socrates in Xenophon’s Symposium and Pausanias’ address in Plato’s dialogue of the same name cannot be denied.39 But as we shall see in more detail in the next chapter, Plato’s own philosophy of eros was much more complex and subtle. At the same time, it is immediately apparent that Plato not 38 See above all Xen. Symp. viii. 39 See Plato Symp. 180c–185c.

132

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

only shares these reservations about pederastic relationships that involve physical satisfaction, in fact he takes them further. The message that welltrained young men are physically attractive, but that the natural receptiveness to such physical beauty should be philosophically sublimated, as well as the ambivalence of shimmering homoerotic tension and of intellectual withdrawal from this tension, run through many of Plato’s dialogues.40 To be sure, Plato admits that not everyone can be expected to exercise the sexual abstinence demanded of the perfect philosopher, that sexual desire is a strong impulse and that sometimes it is necessary to fulfil one’s sexual wishes.41 But there can be no doubt that at least in his late work on legislation in the polis (Laws) he comes to the conclusion that the physical satisfaction of sexual desire is morally acceptable and politically desirable only within heterosexual marital relationships. In the Republic he already dismisses sexual pederasty as ‘unmusical’ and ‘common’, and demands that such relationships must be kept free of even the slightest sexual connotations and should be transformed into well-meaning father– son relationships.42 The Laws goes even further, describing pederasty – and homoerotic relationships in general – as ‘unnatural’ (para physin).43 Sexual intercourse between heterosexual married couples for the purpose of conceiving children is the only kind of relationship upon which no restrictions are placed in the law on erotic relations; otherwise even husband and wife are to be as abstinent as possible.44 If it is impossible to avoid having an extra-marital affair, then it should be kept secret, and most certainly should not be homosexual.45 Certainly Plato’s philosophical theory of eros, above all as presented in the Lysis, Symposium and Phaedrus, has more depth and richness and this may be particularly true of its treatment of pederasty. But when it comes to making philosophical laws for the average full citizen of an ancient polis, then Plato damns all forms of sexual practice, and in particular sexual satisfaction within pederastic relationships. Yet at the same time he seems, like Xenophon, to be one of those authors who demonstrates an understanding of homoerotic wishes, fantasies and dreams. However, whereas Xenophon and the Attic orators 40 Two of the best examples of this are the opening of the Charmides, and the end of the Symposium; see also the programme of sublimation in Polit. ix, 589b1–3; vi, 485d6–12. 41 Plato Polit. iv, 432a6–9; vii, 558d11–c3; v, 458d2–6. 42 Plato Polit. iii 403a7–12; b4–6. 43 Plato Nom. i, 636 b1–c1; viii, 841d4–8. 44 Plato Nom. viii, 838e–839a; 840c–d. 45 Plato Nom. viii 841c–842a.

t h e a s y m m e t r i c a l r e l at i o n s h i p

133

only applied this view at most to the close circle of the (future) political elite, in his later works Plato extended it to cover all full citizens.46 It was Aristotle who then specified this attitude towards homosexuality, and in particular pederasty, more precisely. He was of the same opinion as Plato in the latter’s later works, that homosexual activity was an ‘unnatural’ form of sexual pleasure,47 and at the heart of his description of pederasty as an ‘unhealthy state of enjoyment’ we can perhaps see the belief that this state is often the result of years long homosexual abuse of children and boys.48 But above all Aristotle demonstrates a sober lack of sympathy and a high degree of realism when he integrates pederastic relationships into his classification of friendships. Friendships can exist between equals and unequals, and since that which is worthy of love has three forms (the pleasant, the useful and the excellent), there are accordingly three different goals and three different forms of friendship: for the sake of pleasure, advantage or excellence.49 Friendships which have as their goal pleasure or advantage are ‘lower’ forms of friendship and are characterised by contingent unreliability.50 But at the same time friendships of whatever form are a central feature of a successful life, and their goal is both self-advantage and the friend’s benefit (although both should be ‘justly’ distributed).51 The fundamental reason why friendships of a lower form are contingent and unreliable is that the friends want each other ‘merely as providing some benefit or pleasure’, and not for the sake of the other.52 In the case of friendships between unequals, and in particular in the case of pederasty, these goods are often different, so that a pederastic relationship is in fact a mixture of the two lower forms of friendship – it is a friendship in which pleasure is exchanged for advantage. The lover’s goal is sexual pleasure (in whatever form), whereas the young beloved is looking for gifts and attention.53 Thus it is difficult to apply the criterion of justice to it, for 46 That Plato was well aware of the implications of this stricter attitude (Nom. viii, 836b) is confirmed by the ‘Athenian’ in the Laws who claims: ‘Even so, I could put up quite a convincing case for supposing that the difficulties are not beyond human powers, and can be overcome’ (Nom. viii, 839d). 47 Ar. NE vii 5, 1149b14. 48 Ar. NE vii 5, 1148b25–30. 49 Ar. NE viii 1–3. 50 Ar. NE viii 3, 1156a6–21. 51 Ar. NE viii 6, 1157b19–21; ix 9, 1169b17–19; but see above all the emphatic opening of NE viii, which introduces the lengthy discussion of friendship in NE viii–ix. 52 Ar. NE viii 3, 1156a11 f. 53 See Ar. EE vii 3, 1238b32–39; 10, 1243b15–20; NE viii 4, 1157a1–10; 8, 1159b10–19; ix 1, 1163b32–1164a15. It cannot be emphasised enough that in the classical world pederasty

134

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

that would require weighing a gain in pleasure against material advantage. On the other hand lovers often believe (unjustly) that their beloved loves them in the same way as they love their beloved, or at least they hope so. But this exposes them to ridicule.54 Thus for Aristotle, as soon as a pederastic relationship includes a sexual component it is a sober exchange of pleasure and advantage – and the pleasure is predominantly sexual while the advantage is mainly material. If we then go further and include Aristotle’s opinion that the normal form of pederasty is unnatural and unhealthy, unreliable and contingent, often even hypocritical and ridiculous, then we arrive at an overall picture of sexually inspired pederasty that is extremely unfavourable, and its problematisation is based on criteria other than the norm of active manliness. Aristotle’s texts contain absolutely no hint whatsoever of concern for a dramatic loss of manliness, of self-control, or the typical conditions for a political career on the part of the young beloved in a pederastic relationship.55 Rather such a relationship is presented simply as an unhealthy, elitist, and at the same time as a primitive, form of human relationship that has nothing to do with friendship or love in the best sense. The highest form of friendship is based on the recognition, optimisation and stabilisation of excellence of character, and love does not have as its goal pleasure or advantage, but belongs to a different part of the soul to appetite and pleasure. The fact that it holds its object in high esteem is the result of reason, and it values sexual activity not out of desire, but at best because it is an expression of affection.56 It is also remarkable that Aristotle does not pay particular attention to the non-sexual form of pederasty that aims at the perfection of character separately. This form could certainly be integrated into his classification of friendship, for all three forms of friendship can exist between equal as well as unequal persons (and men).57 But this form of pederasty has no systematic role to play in either the discussion of friendship in the Nicomachean Ethics, or the section on upbringing in the last book of the Politics (though this book is probably incomplete).

54 55 56 57

was regarded as being a very expensive amusement. Therefore it was regarded as being a habit of the rich, the upper class, and in the law-courts it could be used to support political accusations (of oligarchy) (see Lacey (1968), 158). See Ar. EE vii 10, 1243b15–20 (on the problem of justice); EE vii 3, 1238b32–39 and NE viii 8, 1159b10–19 (on the ridiculousness of many lovers). It is significant that in his entire discussion of the moral problematisation of pederasty in UP part 4 he does not make even a single reference to Aristotle. See Ar. NE vii 6, 1149a32–b2; An. Prior. ii 22, 68a39–b6. See Ar. NE viii 15, 1162a34–b5.

t h e a s y m m e t r i c a l r e l at i o n s h i p

135

All in all the references to and discussions of homoerotic relationships in the classical world presented so far suggest that Foucault’s diagnosis must be subjected to restrictions and distinctions, and once again power and knowledge in a particular sense play a more important role than Foucault is prepared to admit to in The Use of Pleasure. As he emphasises, pederasty was problematised morally out of concern for the threat it posed to manly activity and discipline within the context of the imperative of a political career in the polis. But this form of problematisation was mainly concerned with the small restricted circle of full male citizens who were active as public speakers and wished to play a permanent leading role in political business. It was within the framework of this specific context that the political landscape of the ancient polis played a leading theoretical role in the moral interpretation of pederasty. Certainly the picture of the effeminate, lascivious beloved in a pederastic relationship was ridiculed from the perspective of a normative concept of manliness. But what disturbed ancient authors about it was the disastrous danger of a change of sex, a view that was based on a far-reaching distinction between masculinity and femininity that proves to be of fundamental importance for homoerotic relationships. This specific form of problematisation was mainly inspired by misogyny and discrimination against women in the ancient world, which will be examined in more detail in the next two sections of this chapter. However, there was a moral problematisation of pederasty in the classical period which went beyond the imperatives of political activity and the image of the effeminate male object of desire. Seen from the perspective of the bodily inviolability of free citizens, whether young or old, politically active or not, male or female, which was one of the fundamental distinctions between the free and slaves, the sexual assault of boys and young men was regarded as being essentially worrying and in some cases was forbidden by law. Pederastic appetite could be regarded with a degree of understanding, but its fulfilment with criticism. Philosophers such as Xenophon, Plato and Aristotle expressed this attitude in a more radical form, and consequently were only prepared to accept pederastic relationships in their non-sexual form. Thus they attempted at least theoretically to put an end to the ancient tendency to sexually abuse boys and youths. But what is perhaps more important is that this more general problematisation of pederasty – which Foucault does not discuss – expresses the view that sexuality can only be interpreted as a relationship of physical power and dominance that is completely divorced from all acceptable forms of reciprocity. When we take into account that sexual assault on children and youths was purchased with expensive material considerations, then this

136

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

sexual power relationship also represents an unpleasant form of repression.58 The exchange of pleasure and advantage as the central aspect of sexually oriented pederasty does not escape Aristotle’s impartial analysis, but is passed over in silence by Foucault. This theoretical and moral attitude towards pederasty completely excludes the possibility of considering a technical refinement or physical sublimation of homoerotic practices. Thus compared with later ancient techniques of love, such as are described by Ovid for example, the ‘art’ of ‘erotics’ had to lead a truly pathetic existence as a mere technique of courting and flirting, a play for admiration, honour and recognition. Seen archaeologically, in the Foucauldian sense, this picture confirms that homoeroticism as a relational love relationship could not yet become the object of knowledge. The ‘game of truth’ that accompanied pederasty must therefore be defined very generally and indirectly: homosexual eros was to be presented as a preliminary phase of the pursuit of the good and the truth which was concerned with very different objects from the erotic phenomena themselves. Thus the main reason why in the classical period sexually oriented pederasty was problematised outside the narrower political sphere was related to the fact that it established a physical relationship of dominance that ultimately was inconsistent with the status of the object of pleasure as a free person – an inconsistency that could not in any way be reduced by the presence of material presents; on the contrary, it brought the young beloved dangerously close to prostitution. The moral and philosophical countermodel of a non-sexual pederastic relationship that includes important pedagogic components shows in all clarity what was expected morally of the lover. Certainly it was also expected that desire should be tamed, but above all the sexually oriented relationship of dominance was to be transformed into an educational one that was based on friendship, and apparently could be described as an exemplary instance of the exercise of regulative power. Once again in the moral and philosophical suggestions for the exercise of sexuality we encounter a fundamental structure similar to that already recognised in the models of the good life, or in dietetic recommendations: it is not so much a model of domination and restriction, as a model of transformation and sublimation that is being recommended. In the case of pederasty the lover is to transform his physical, 58 It is significant that in old and middle comedy (the extant plays of Aristophanes and Menander) heterosexuality is generally the norm. This is understandable when we consider that on the whole the comedies deal with everyday people who did not have sufficient wealth and leisure to be able to afford a pederastic relationship (see Dover (1978), 147–51).

t h e a s y m m e t r i c a l r e l at i o n s h i p

137

repressive power over his youthful beloved into a positive, regulative one that aims to increase the opportunities that young men enjoyed to perfect their own life. 3 Foucault assumes that ‘Xenophon’s Oeconomicus contains the most fully developed treatise on married life that classical Greece has left us’,59 and so he devotes a lengthy analysis to the account of Isomachus’ Household.60 After an accurate and sensitive presentation of the text he comes to a surprising conclusion: In the ethics of married life the ‘fidelity’ that is recommended to the husband is therefore something quite different from the sexual exclusivity that marriage imposes on the wife. It has to do with maintaining the wife’s status and privileges, and her pre-eminence over other women . . . The husband’s self-restraint pertains to an act of governing – governing in general, governing oneself, and governing a wife who must be kept under control and respected at all times, since in relation to her husband she is the obedient mistress of the household.61

According to this diagnosis the sexual self-restraint of the husband is not intended to prevent him from being unfaithful, but to ensure that the wife ‘keeps the pre-eminent place that marriage has assigned to her: not to see another women given preference over her, not to suffer a loss of status and dignity, not to be replaced at her husband’s side by another’.62 For – as Foucault reads the text – Isomachus’ marriage is threatened not by the occasional extra-marital sexual pleasure the husband enjoys, but at most by the rivalry between his wife and other women for the leading position in the household. This summary is surprising because it cannot be supported by Xenophon’s text. At no point does Xenophon even vaguely discuss the problem of marital fidelity, and certainly not the motive for the husband’s sexual self-restraint. More generally, the text does not contain any hint of a moral problematisation of sexual behaviour on the part of the husband, nor of any moral demand whatsoever for self-restraint. If the text presents 59 See UP 152 (apart from the treatise Pseud.-Ar. Oeconomica iii, of course, which is only extant in a Latin translation, and of which the author and exact date are unknown). 60 See UP 152–65. 61 UP 165. 62 UP 163.

138

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

a message concerning the sexual and emotional attitude of the husband towards his wife, then it is the statement that the wife will be all the more attractive to her husband the more thoroughly she combines a natural and healthy physical beauty with carrying out the work and tasks allocated to her in a professional and independent manner. However, this also applies to the husband, for when a man and a woman live together in a marital relationship then they cannot deceive each other about their physical appearance or their professional estate management abilities.63 Certainly Xenophon maintains that a good kyrios who manages his oikos well must keep his passions and desires under control, but this has nothing to do with his behaviour as a husband, but is concerned with the classical – and at the time much-discussed – phenomenon of weakness of the will within the context of necessary ‘economic’ tasks. As Socrates points out at the beginning of the dialogue, in many areas of life there are people who are extremely knowledgeable but cannot put this to advantage as they are unable to put into practice what they theoretically know to be correct.64 Isomachus goes on to demonstrate to Socrates in great detail that this is particularly true of agriculture. In this case weakness of the will is a particularly important factor, for it is easy to learn what a farmer should do, and in fact almost everybody has this knowledge; but whether or not a farmer or the administrator of an estate is successful, whether he grows rich and is respected, or becomes poor and loses his reputation, depends solely on his strength of will and self-discipline in daily carrying out diligently what he knows to be correct – whether he is prepared to perform his tasks punctually and to work hard.65 The r´esum´e which Foucault believes he can extract from Xenophon’s Oeconomicus is both spectacular and speculative, and is clear proof that Foucault’s identification of the systematic place of moral concern for the behaviour of the husband in this ‘reflective practice of married life’66 is inappropriate. What lies at the heart of this concern can only be revealed 63 See Xen. Oec. x. Xenophon (Isomachus) cleverly points out that both elements are closely linked: professional work in the house ensures physical fitness and optimises natural physical beauty. It is surprising that Foucault says that this way of talking about women’s attractiveness may astonish us (UP 162), unless of course he followed the antiquated view that professionalism in their work contributes nothing to women’s beauty. As for the context of Xenophon’s remarks about the beauty and attractiveness of women in the framework of the intellectual scene in classical Athens, they are to be located in the milieu of Socratic and Platonic thought. 64 See Xen. Oec. i, 16–23. 65 See Xen. Oec. xx, 16–25; i, 16–18. 66 UP 163.

t h e a s y m m e t r i c a l r e l at i o n s h i p

139

when the connection between the institutional weight of the oikos within the ancient polis, the classical doctrine of the inferiority of women to men, and the status of the Oeconomicus in scientific theory is taken into account – in other words when the rules for the husband’s behaviour, as they are presented in the text, are seen not only from an ethical point of view, but also as part of power relations and archaeological structures in a Foucauldian sense. If Foucault’s analyses of ancient moral views of marital and pederastic relationships are compared, then it is easy to gain the impression that he felt that the social weight of both practices in classical Greek society was more or less the same.67 However, in the ancient polis pederasty was always restricted to a small circle whose prosperity secured them the leisure and the financial means to engage in pederastic relationships and their competitive erotics. On the other hand, as the social core of the oikos as an institution, marriage was a fundamental and widespread element in ancient society. Thus moral concern for conduct within marital relationships cannot be adequately identified without taking into account its intrinsic connection with the maintenance of the oikos and the oikos family. In the ancient world the large household, the oikos, was quite literally patriarchal. It was run by the kyrios, a grown-up, male full citizen who was head of the family, and had the right of disposal (kyrieia) of the entire oikos and all parts of it – in other words of his wife and her dowry, of his legitimate children, of the slaves, livestock, property and money. The kyrios conducted all legal and financial transactions that concerned the oikos and was in charge of all productive work that secured its income – mainly agricultural produce for the requirements of the oikos and any surplus which was then sold. Since the kyrios had access to extremely cheap labour in the form of slaves, who normally repaid their costs within two years, a well-run oikos with decent land was generally a profitable organisation. Under no circumstances were women allowed permanently to assume the position of the kyrios; although wives did often administer their husbands’ estates temporarily when they were absent for a length of time, they were not allowed to sell off any part of them or to pledge them as security.68 The property which a wife brought into the husband’s oikos as her dowry – and which was often substantial – remained legally separate from her spouse’s estate. It became subject to the husband’s kyrieia, was 67 UP parts 3 and 4. 68 See Penelope in the Odyssey, who ran the oikos for many years, or Demosthenes, speech xlvii.

140

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

controlled and administered by him and on his death was passed on to his male heir in the oikos. But he himself or his heir remained responsible to the wife’s family and was liable to repay the dowry to her family in full if she died or was divorced from him without leaving any children (in principle wives could successfully file for divorce, but their prospects after divorce were bleak). A woman never had property in the sense that she could dispose of it as she wished, but only in the sense that it could never be legally separated from her, and so could be used to support her. Daughters who had no brother had an absolute right to inherit their father’s oikos. In the ancient polis a whole network of legal measures covered this explosive eventuality. A kyrios who was punished with only having daughters – the phrase ‘this man died without issue and left x daughters’ was common – could adopt a son (normally from among his relatives); if he did not do this in time then his closest male relative on his father’s side, and who was not already kyrios of another oikos, was bound to marry the heiress, and so to take over the oikos as kyrios. If he was already married he even had to divorce his wife. However, the thoroughly patriarchal structure of the oikos family and its servants, which among other things also included the sexual availability of all female slaves in the oikos to the kyrios, does not reveal the full extent of the social weight of the oikos in an ancient polis. That requires an understanding of the primary function of the oikos. Each oikos was tied to property, in most cases arable land (kleros), and together these farms or large estates made up the agrarian foundation on which all ancient poleis depended. Furthermore, as the polis raised only few taxes (in particular there were none on property or income), all public expenditure and the entire military apparatus in both war and peace were financed on the basis of donations (‘liturgies’) from the profits made by the oikoi. Since Solon’s laws (for example the introduction of personal wills) had freed the oikoi from the influence of the family clans (the gene), the large estates and farms of all kinds constituted the economic backbone of the polis and were its main elements. Thus a basic assumption of ancient political philosophy was the succinct observation that the oikoi were the smallest units and elements that constituted the polis.69 Understandably the polis did everything in its power to protect all oikoi in its own domain, and to help them ‘flourish’ (the decline of even a single oikos, for whatever reason, was regarded as a political disaster). Just how vital the political importance of the oikos was, for example, is demonstrated 69See Ar. Pol. i.

t h e a s y m m e t r i c a l r e l at i o n s h i p

141

by the fact that legally all matters involving the oikos were considered to be public rather than private.70 In many poleis the sale of agricultural land belonging to an oikos was forbidden, and even when it was allowed (as in Athens) it was felt to be morally extremely disturbing – from a number of law suits in the Athenian courts it is clear that the claim that somebody had sold land out of greed for money was regarded as a particularly serious accusation, and the demise of an oikos because its land had to be sold off out of necessity meant a dramatic loss of face for the kyrios and the end of his self-respect. One of the main political aims of this strict legal and moral attitude towards the sale of land belonging to an oikos was to prevent individuals from accumulating a number of oikoi; it was the task of each kyrios to focus all his attention and resources on administering just one oikos. Thus moral and legal sanctions on the sale of arable land or other important elements of the oikos restricted the kyrios’ otherwise comprehensive right of disposal of the oikos. The kyrios was not regarded as the private owner of his oikos in the modern capitalist sense, but as the guardian and warder of the family property which could generally look back on a long tradition, and was to be secured and consolidated for the future. This in turn points to an internal aspect of the oikos that corresponds to these moral, legal and political regulations that applied to it. The oikos family saw itself permanently at the juncture between the past and the future, between tradition and its continued existence. From the internal perspective of the family, the primary task was certainly to ensure that the oikos was productive enough to feed all its members including slaves decently, to clothe and house them, as well as to provide adequate liturgies to finance its public responsibilities. As the same time, respect for the past meant that it was an elementary moral duty to support and care for those members of the oikos who could no longer work, as well as to honour the dead and the family’s ancestors by caring for their graves and performing the required religious ceremonies. And with regard to the future, it was expected that the continued existence of the oikos was to be secured by good economic management, carefully planned marriages, and an excellent up-bringing for children and descendants. As many philosophical writers emphasise, this internal perspective and the tasks that the oikos set itself enabled the members of the oikos family to partake in one of the forms of immortality 70 This meant that accusations could be submitted not just by the injured party (as in private law) but by anybody, and that the plaintiff did not (as in private law) have to win the votes of at least one-fifth of the judges in order to avoid punishment.

142

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

that mortal beings can enjoy. Thus the oikos also had a significant religious aspect, and constituted a religious union dedicated to the cult of Zeus Hecataeus; but for its living members it was their social shelter, their ‘social net’. It is only against the background of the enormous economic and social weight of the oikos within the ancient polis that the fundamental significance of marriage between male and female full citizens, and the legitimacy of their children, can be fully appreciated. A functioning marriage was indeed an important prerequisite for the oikos’ property, and certainly in most cases this functionality was guaranteed by the merciless legal, cultural and sexual oppression of the wife. Girls, young women and wives were kept in almost total seclusion in their houses – so much so that in some court cases doubts could even be cast on their very existence. There was not to be even the slightest suspicion of pre- or extra-marital sex – but not so much for moral as for legal and political reasons; given the clear distinction between citizens and non-citizens, and because of Pericles’ law that only the children of two full citizens were themselves full citizens, the legitimacy of children had to be free of all doubt. What is more, it is common knowledge that in the ancient polis women had no political rights whatsoever and enjoyed only an elementary education. They were also generally married off very young (at about the age of fifteen), and their husband, whom they often never even saw before the wedding, was chosen by their father or their kyrios alone. However, it should not be overlooked that in spite of this repression women and wives did have a certain – on the whole morally defined – scope for personal development. The oikos offered, of course, all free women belonging to it complete protection from sexual harassment by men – and thus free women at least were in a much better position than young men vis-`a-vis pederasts. It was here that the husband’s sexual freedom was restricted by law: all sexual contact with another free woman in another oikos – that is not only with the spouse, but also with the mother, daughter or sister of the kyrios – was treated as adultery and in all ancient states was without further ado punished by death. It amounted to the worst possible offence, an attack on the property of the kyrios’ oikos and the legitimacy of its descendants, and was therefore regarded as an attack on the oikos as a social institution, which the polis felt a particular need to protect. Domestic activities, in particular administering provisions and income, as well as overseeing the servants, remained the domain of the wife within the oikos, but this was on the whole a demanding job for which women had to prepare themselves. For this reason it was normal for them to be

t h e a s y m m e t r i c a l r e l at i o n s h i p

143

educated in their parents’ house (not in public schools) in housework, running the household, looking after and bringing up children, reading and writing and sometimes in the basics of music; and after marriage they were usually further instructed by their husband how to run his specific oikos. Thus when classical authors problematise conduct within marriage (above all of the husband, the kyrios), then they are not dealing with a simple, or relatively unimportant matter; what they are articulating is moral concern for the preservation and protection of the fundamental, most important social institution in the polis – the oikos. If a husband misused his sexual freedom and his social authority in a manner which posed a threat to his marriage, and so to the integrity of the oikos, then such behaviour could only be regarded as in every practical respect alarming – economically, politically, legally and morally. The problem was made even more serious by the fact that on the whole the wife played an important and demanding role in the oikos, in spite of being oppressed both legally and politically. These are the essential conditions upon which Xenophon bases his Oeconomicus.71 Another condition was the firm conviction that the social subjugation of women was the result of their natural inferiority to men, and so was morally justifiable. Although this is widely understood today, it is still worthy of mention in this context. Economic tracts of the classical period base their conclusions on the ‘nature’ of men and women,72 and the same theoretical fundament was just as indispensable to political theories.73 To point out just one example, it is both interesting and amusing to take a closer look at the argument between those two giants of classical philosophy, Plato and Aristotle, over a few nuances of the relevance of the nature of women. Plato always demonstrates his conviction that men were superior to women in all fields,74 and naturally wants to restrict access to higher, or even philosophical, education to boys.75 But at the same time he dared suggest that in principle men and women had identical aptitudes;76 that women should work not just in, but also outside, the house;77 that women should learn all skills, should also be educated in 71 These conditions – as well as the relevant texts – can be read in chapters i, iii, iv, v and vii of Lacey’s classic work (1968). 72 See Xen. Oec. vii 23–32; Pseud.-Ar. Oec. i, 3, 4. 73 Ar. Pol. i. 74 Plato Polit. v, 455c–d. 75 Plato Polit. vi, 536d ff. 76 Plato Polit. v, 454d–457b. 77 Plato Polit. v, 451c.

144

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

music and gymnastics and should do military service;78 and that the political elite should ensure that not just men but women too should be able to lead a happy life.79 Confronted with such provocation, Aristotle felt he had to reply to his old teacher and emphatically restated the generally accepted view that women’s virtues were basically suited to carrying out commands;80 that women were by nature emotional and unstable,81 so that their deliberative abilities were extremely limited and required the guidance of a man’s deliberative decisions;82 that dominant women were contrary to nature;83 and that the ability to remain silent was a woman’s supreme virtue.84 No doubt Socrates speaks from Aristotle’s heart in Xenophon’s Oeconomicus when Cristobulus poses the rhetorical question ‘Is there anyone to whom you commit more affairs of importance than your wife?’85 It is worthwhile remembering the embarrassing and desperate argumentative manoeuvre with which Aristotle attempts to justify the natural inferiority of women.86 He opens with a definition: a person rules by nature when he has the ability to exercise deliberative thought, and serves by nature when he does not have this ability. He then claims that men rule over women by nature. This he justifies by stating that an efficient cooperation between ruling males and serving females can be observed in all higher animals. Whether this also applies to the specific case of humans is either implied by this justification or not. In the latter case it is an example of a petitio principii, in the former case proof is lacking. Even if most women serve by nature in the sense of Aristotle’s initial definition, this does not mean that they do so as a result of biological circumstances, and thus in this sense by nature; for the fact that many women are indeed incapable of deliberative thinking does not mean that they are by nature so, i.e. because of biological circumstances – everybody in the ancient world knew how badly educated women were. It is possible that here Aristotle pre-supposes purely natural-scientific arguments for the inferiority of women that were well established at the time, and treated female physiology as the subject of scientific medical 78 79 80 81 82 83 84 85 86

Plato Polit. v, 451a–452a, 457b ff.; Nom. vii, 804c–805b. Plato Nom. vii, 805d–806c. Ar. Pol. i 5, 1254b13 f.; i 13, 1260a21–4. Ar. HA ix 1, 608b8–11; NE vii 9, 1151b15. Ar. Pol. i 13, 1260a13; NE vi 13, 1144b26 f. Ar. NE viii 11, 1161b2–3; viii 10, 1161a1–3. Ar. Pol. i 13, 1260a30; iii 4, 1277b23. Xen. Oec. iii 12. See Ar. Pol. i.

t h e a s y m m e t r i c a l r e l at i o n s h i p

145

theory.87 Such theories generally regarded women as cold and moist, men as warm and dry. Women are like fields that are sown with seed; the regulation of moisture and of the various fluids is of great importance for both. The female body contains more extra blood because it draws more moisture out of food, and because its flesh is softer and more porous it can store more blood and other fluids; this is why the female body excretes more fluid than the male body (for example blood during menstruation and milk when nursing). The flesh of the male body, on the other hand, is hard and dense; it absorbs much less fluid, so that men are drier and warmer.88 In other words, the Hippocratic physiology of women was based on the assumption that the female body had an excess of fluids, and so it was important for their health that these fluids were regularly excreted. This could be achieved through menstruation, orgasms and breast-feeding – thus regular sexual activity and frequent pregnancies were important factors of the good health of women and were recommended as therapy. This meant that marriage was a medical necessity for free women, but it was also important that women experienced an orgasm during sex because, according to Hippocratic theory, they ejaculated just as much fluid as men. This explains why, for example, the peripatetic tract Problemata Physica claims that in summer women have a greater inclination to sexual activity than men – because in the heat of the summer beings of a warm nature lose the balance of their temperature, while beings of a cold nature optimise it: ‘the man is warm and dry, but the woman cold and moist’.89 However, according to this theory the moisture of women also posed a dangerous threat to men; for what is moist tends to overflow, to burst its banks and to lose control. A classic example of this is drunkenness, and the theory can, of course, easily explain why women burst into tears more often than men. Generally speaking emotions and feelings produce moisture – they melt, loosen and dissolve.90 This is particularly true of sexual appetite, and helps us understand why, when women’s sexual desires are awakened, they are insatiable and endless,91 and why women are more susceptible 87 See on this Carson’s excellent and informative article, Carson (1990); also Hanson (1990), esp. 317–19. 88 See e.g. Hipp., On Female Diseases, and Hanson (forthcoming). 89 Pseud.-Ar. PP iv 25, 879a31–35; see also Hipp. De Victu i 27. 90 Carson (1990), 137 ff. presents an number of texts supporting this. 91 Classical Greek literature is full of warnings about overpowering female sexuality. For example, as early as Hesiod we hear (with reference to the precarious balance of qualitative powers) that a bad woman ‘rusts and dries out her man’ with her insatiable appetite

146

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

than men to emotional outbreaks.92 It is the dry, warm constitution of their body and mind that enables men to resist appetites of all kinds, and to exercise self-control and intelligence – while it is the physiological constitution of women, their moisture and coldness, that determines their physical, psychological, moral and intellectual inferiority to men. Thus it is the man’s task to restrict and to control women’s overflowing exuberance and lack of self-control.93 If we are to identify the moral concern for the conduct of the husband in the economic writings of Xenophon and pseudo-Aristotle correctly, then we must bear in mind both the fundamental social importance of the oikos, and the powerful poetic, philosophical and scientific doctrine of the basic inferiority of women to men, as well as its threatening consequences. But this does not by any means cover all of the social factors that influenced moral concern for the conduct of the husband in classical Greece. Paradoxically, one of the most important additional factors in this context was moral concern for the behaviour of the wife. In classical Athens, as in many other Mediterranean societies, heterosexual relationships were organised and regulated by a zero-sum game of honour and disgrace. An important part of this was the view that the main element of a woman’s honour – not just a wife’s, but also a virgin’s or a widow’s – was her chastity and faithfulness, and this had to be secured by her good reputation; a husband’s or a kyrios’ honour consisted primarily in all the women within the domain of his kyrieia and oikos maintaining their female honour. Sexual assaults by other men on any of these women not only constituted an assault on the property of the kyrios, but above all on his honour. Thus excluding women from the public sphere, and tying them to the private domain of the house, also protected the kyrios’ honour. But the connection between moral concern for the wife’s behaviour and moral concern for the husband’s behaviour is only fully revealed when we realise just how manifold the threat to the woman’s honour in classical society was. It was not just the menace of female sexuality that constantly threatened to break down the woman’s resoluteness; there were a number of other social factors which could cause danger.94 If as part of the zero-sum game of the husband’s or the kyrios’ honour and disgrace he had to accept (Hes. Works and Days, 702–5); Aeschylus talks of the ‘flaming eye’ of women as soon as they ‘have tasted’ a man (frag. 243 Nauck); women’s lasciviousness is a favourite topic in comedy (see Aristoph. Thesm. 504 ff.; Ekkl. 468–70; 618–20; Nub. 553 ff.; Lysistr. 553 ff.). 92 See e.g. Ar. HA ix 1, 608b. 93 On the influence of these views on poetry see Dover (1974), 98–102. 94 See on the following the excellent, informative analysis by Cohen (1991), ch. 6 (133–70).

t h e a s y m m e t r i c a l r e l at i o n s h i p

147

a dramatic loss of honour if a woman in his oikos had a sexual relationship with another man, so too the intruder gained honour to the same degree. In spite of the risks that adultery involved for the adulterer, the increase in his honour was an important social motive. Furthermore, given the considerable difference of age between husband and wife involved in arranged marriages, pre- or extra-marital liaisons were often the only opportunity for both men and women to experience a romantic relationship. But although there were strong motivating forces for adultery in the classical world – the insatiable sexuality of women, the increase of honour for the adulterer and the experience of passionate, romantic love – the traditional view was that the complete seclusion of women in the home virtually banished such dangers. But then we must ask how under such circumstances adultery could actually take place. The fact that it is mentioned so often in the sources, and that in the ancient polis there were comprehensive laws that dealt with the matter, are a clear indication that the total seclusion of women was in fact a social ideal rather than actual practice. Certainly it was a credit not only to women but also to the kyrios when women could claim that they lived by this ideal – as for example Andromache, Hector’s wife, in Euripides’ tragedy The Trojan Women.95 But we also hear of women who were involved in important financial decisions in the oikos – not only the administration of money but also the drawing up of wills; wives took an interest in how their own dowry was employed, or in the fortune of their deceased spouse, and had an important word to say in the education of children, even when they grew older. But above all there was a wide range of activities which required women to leave the house. In classical Athens financial necessity meant that many women had to work outside of their house – at a stall in the market, in the fields, in wine making or as wet nurses, and in particular when their husbands were away fighting in a war, were on military service or could not earn a living themselves. And there were many opportunities to go out other than when working; women fetched water, visited neighbours, attended funerals and family celebrations, visited public baths, appeared in court and above all had an important role in religious ceremonies; they did indeed show themselves in public, sometimes dressed festively. What is more, life in the ancient polis often involved close contacts with relatives and neighbours, and maintaining such contacts was on the whole the woman’s task – in short, free women in classical antiquity had numerous opportunities to 95 See Eur. Troj. 634–83; in Aristophanes husbands regularly get angry when they find out that their wives are out of the house, or have been: see e.g. Thesm. 397 ff., 410 ff., 519, 785–91.

148

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

meet other men and form pre- or extra-marital relationships. And this was made all the more possible by the almost complete separation of wives which meant that husbands frequently did not and could not know what their wives got up to in or out of the house.96 It is only when we realise that adultery could completely ruin a husband’s honour, and bring down lasting disgrace on his family and descendants, but that on the other hand there were many chances and powerful motives for forming adulterous relationships, over which the husband had little control, that we can fully appreciate the extent of the social burden, the social relevance and the moral concern involved in the husband’s pursuit of his goal of upholding his own honour and his wife’s faithfulness. However, a number of texts97 confirm that this goal could be endangered by unfaithful or negligent conduct on the part of the husband. Many passages bear eloquent witness to the deep-felt emotional injury that such behaviour could cause in the wife, and in Euripides’ Elektra Clytaemnestra openly speaks of the danger inherent in such a situation: But when, this granted, ‘tis the husband errs, Slighting his own true bride, and fain the wife Would copy him and find another love.98

In the classical world sexual moderation and respect for his wife was recommended to the husband not just because he had to ensure the survival of the oikos and his family, but also because under the prevailing social circumstances any gross or dissolute behaviour could easily lead to adultery on the part of the wife. And he himself had more to lose from that than the adulteress, for it was a question of his honour and his self-respect. 4 Bearing all this in mind, let us have another look at Isomachus’ Household. Of particular significance is the question with which Xenophon has Socrates open the entire dialogue: ‘Tell me, Cristobulus, is estate management the name of a branch of knowledge [episteme], like medicine, smithing and carpentry?’ 96 See on this Lacey (1968), ch. vii; Cohen (1991), 150–65; Dover (1974), 69 with numerous references. 97 See below, pp. 158 ff. 98 See Eur. Electr. 1036–40.

t h e a s y m m e t r i c a l r e l at i o n s h i p

149

Xenophon claims to be dealing with a scientific subject: the administration of an estate can and should be the subject of scientific thought. Aristotle felt that ‘economics’ (the teaching of the administration of the oikos) was subsidiary to political science, the aim of which was to make the members of the polis community good and happy by means of a good constitution and good laws.99 Given the social relevance of the oikos for the ancient polis, it is hardly surprising that at a time when science was being divided into new separate branches the oikos should form a new field of knowledge. At the beginning of the Oeconomicus Socrates and Cristobulus discuss several questions of scientific theory concerning the status of economics as an episteme or a techne – for example to what extent it can be taught independently of experience and property, what the relationship is between knowledge and its application in this case, and under what conditions this knowledge can achieve its practical aims.100 These explicit reflections serve to reinforce the claim that the author of the Oeconomicus has scientific ambitions. But what exactly is the subject of the art and science of the administration of an estate? What are its most important fields and concerns? Subsequent to their more general theoretical discussions about the nature of the science, Socrates and Cristobulus go through the most important fields: building suitable houses, the correct use of implements, the punctual organisation and arrangement of all material and equipment, the proper treatment of subordinates, agricultural techniques in a narrower sense, and – emphasised by a short interpolation, the length of the discussion of it, and its place at the end of the list – the treatment of the wife.101 After a traditional exhortatio, which praises economics as the most important and finest of all skills,102 there is a change of scene in the dialogue. The conversation between Socrates and Cristobulus comes to an end, and Socrates continues with a description of how he himself was instructed in the details of economics by Isomachus, an experienced, respected, and successful kyrios and administrator of his household.103

99 Ar. NE i 1, 1094a27–b1. 100 Xen. Oec. i, 1–23. In particular it requires discipline in following scientific guidance, and hard work in order to apply economic knowledge successfully. In this sense self-control is an important condition of successful economic practices (see esp. Xen. Oec. i, 16–23). 101 Xen. Oec. iii, 1–5 for the list as far as agricultural techniques; there follows the interpolation (6–10), and instructions for the treatment of the wife, whose central function in running the household is then emphasised, (10–16). 102 Xen. Oec. iv–v. 103 Xen. Oec. vi, 11–17.

150

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

The structure of this description is of particular significance, for it reveals what Xenophon regards as the main focus of economics. It is within the systematic and scientific context of this focus that the behaviour of the kyrios in his role as husband is also discussed. First, however, it is useful to see how Socrates was introduced by Isomachus to that part of economics which at first glance is the primary task of the administrator of the oikos: the art of agriculture, which the kyrios is to oversee himself. At this point Isomachus repeatedly emphasises how easy it is to learn these things,104 and that anyone with any common sense already understands the most important aspects of agriculture. Isomachus demonstrates this in an imitation of the famous geometry lesson in Plato’s dialogue Meno,105 at the same time ironically applying Socrates’ midwifery skills on Socrates himself to prove that with clever questions he can get Socrates to ‘give birth’ to the knowledge of the most important aspects of agriculture.106 However, this does not take up an excessive amount of space in Isomachus’ teachings;107 it forms part of the second section describing the kyrios’ tasks,108 which also include physical training and rhetorical exercises,109 but above all instruction of the head manager,110 the stabilisation of self-discipline as a condition of success,111 as well as learning to command, i.e. the correct treatment of his subordinates in the oikos.112 The entire first section of the Oeconomicus deals with the administration of matters within the house and of the goods produced on the estate, in other words with the domain of the wife; but significantly Isomachus’ description takes the form of a lesson that he gives his young wife.113 Thus one of the most important tasks for the kyrios in scientific ‘economics’ is the instruction of his wife in the tasks and skills that she will have to master. If we consider the overall structure and context of the Oeconomicus, then we can identify three main elements of economic science: the skills 104 Xen. Oec. xv, 4, 10; xvi, 2; xx, 1. 105 See Plato Meno 82e–85a. 106 See Xen. Oec. xvi, 8 (Isomachus wishes to refresh Socrates’ memory, for he is aware that Socrates will already know much of what he has to say); xviii, 10 (Socrates realises that he understands a great deal about sowing, but was not aware of this until now); xix, 14–15 (posing questions is a form of teaching, answering a form of learning). 107 See Xen. Oec. xv, 4–xix, 19. 108 See Xen. Oec. xi ff. 109 Xen. Oec. xi, 11–25. 110 Xen. Oec. xii, 1–xv, 1. 111 Xen Oec. xx, 1–29. 112 Xen. Oec. xxi, 1–12. The importance of this task is also emphasised in Oec. iii, 4, v, 14–17, ix, 5, 12–13. 113 See Xen. Oec. vii, 4–x, 13.

t h e a s y m m e t r i c a l r e l at i o n s h i p

151

of agricultural production (the art of earning a living), the skill of looking after provisions (the art of the internal administration of the estate) and the skill of exercising power (the art of leading people) – the latter itself consists of the skills of ruling oneself, ruling one’s wife and ruling one’s subordinates (the managers and male and female slaves). These skills should merge in order to stabilise the oikos as the most important and fundamental institution of ancient society, and their combination apparently reflects an internal interconnection of the three levels at which the processes of the formation of fields of knowledge, the exercise of power and moral subjectivisation can be described. As was already suggested above, Xenophon’s moral concern for the conduct and the role of the husband was undoubtedly inspired by the higher goal of the upkeep and maintenance of the oikos. When the enormous weight which he assigns in this context to optimal, professional work by the wife within the house is taken into account, then it is easy to understand why he demands that the husband and kyrios treats his wife with respect and moderation. But this is not just some abstract, impersonal aspect of marriage, and most certainly not the mere instrumentalisation of mutual respect; rather, Xenophon, just like Plato and Aristotle, feels that character, intelligence and the ability to performs one’s tasks in a professional manner are all integral parts of the ‘beauty’ of individuals: an experienced wife, who keeps herself physically fit and carries out her specific tasks and duties well, is more attractive as a complete person than some slave or hetaira. If a husband injures his wife’s feelings through dissolute behaviour, and so reduces her commitment, then he not only endangers the existence of the oikos and his honour, he also loses an attractive partner and this double price is surely too high.114 But that still does not complete our assessment of the moral problematisation of the husband’s conduct as Xenophon sees it. Another important aspect of this concern only becomes apparent when more weight is given to the level of archaeology and of the analysis of power within marriage, and more generally within asymmetrical human relationships. According to Xenophon’s account, part of applying the art of economics by the kyrios is ensuring that he enjoys the loyalty of all his subordinates – for the continued existence of the oikos and his honour depend on a functional, hierarchically organised cooperation. In this context Xenophon recommends treating all subordinates in the oikos fairly in accordance with their performance – for example, providing male and female slaves with adequate 114 See Xen. Oec. iii, 10–16.

152

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

food, clothing, housing and medical care, and granting them permission to have a family. Commanding those subordinates who themselves have administrative duties – the head estate manager outdoors, the head housekeeper indoors and above all the wife – is an altogether more delicate skill. In these cases the skill of leadership is partly reflexive, for one aspect of its duties and aims is to teach the skill itself (and so to teach what one does) – commanding a wife or a head manager thus takes the form of a regulative power, the exercise of which passes on the science of economics to leading subordinates. As for the head manager, Xenophon emphasises that his knowledge of economics must be just as profound as that of the kyrios, and that if he performs his job well then the kyrios should make him rich, and treat him as a free man and a gentleman.115 In this way he can learn to give orders to his workers and will be virtuous, i.e. he will remain honest and loyal, and respect the property of others.116 Thus a good head manager must in all respects be as able as the kyrios himself. The situation vis-`a-vis the wife seems to be similar. Although in his introduction Socrates says – perhaps tongue in cheek – that there is nobody with whom the kyrios speaks less than his wife, at the same time there is nobody to whom he entrusts more important tasks and matters.117 A good wife contributes just as much to the prosperity of the oikos as a good kyrios,118 and as the mother of his legitimate children she ensures that the kyrios will be looked after in his old age;119 as ‘queen’ of the oikos she enters into a union with the kyrios as king in order ‘to form a perfect partnership in mutual service’;120 and although god or nature has equipped her physically so that she is more suited to work in the house, nevertheless she can have a reliable memory, be diligent and exercise self-control just like her husband.121 If she does not tend, keep and administer what the kyrios earns, then his work is in vain, and so the wife can complement her husband perfectly122 and even ‘prove herself better than the kyrios is, [and] make him her servant’.123 Thus when an estate is administered according to scientific principles, then care will be taken to ensure that the work done by the wife and the 115 116 117 118 119 120 121 122 123

Xen. Oec. xii, 4; xiv, 9. Xen. Oec. xiii, 3–4; xiv, 2–7. Xen. Oec. iii, 12. Xen. Oec. iii, 15. Xen. Oec. vii, 12–13. Xen. Oec. vii, 17–18. Xen. Oec. vii, 27. Xen. Oec. vii, 39–40; 28. Xen. Oec. vii, 42.

t h e a s y m m e t r i c a l r e l at i o n s h i p

153

manager receives the recognition it deserves, and that both of them are acquainted with the basics of economic science. In doing so the kyrios will assume that his wife and his manager have moral capabilities equivalent to his own. According to the science of economics, recognition, teaching and the assumption of reason in subordinates are important aspects of the art of leadership – but at the same time the strict hierarchy between husband and wife, kyrios and manager, and wife and housekeeper must be upheld. Yet this interface gives rise to a far-reaching problem: how can rulers stabilise their rule if in order for their rule to be profitable to themselves they have to grant their subordinates such competence, or even instruct them in it? In the special case of marriage this problem appears to assume a particularly extreme form: on the one hand a powerful tradition held that women were clearly inferior to men, but on the other hand the wife had a very important and influential role to play in the oikos. Xenophon had a clear answer to the question at hand: it is by treating their subordinates decently and justly, and regarding them as fully reasonable, that rulers ensure their subordinates’ loyalty, and secure their own rule. The kyrios’ position as absolute leader is sufficiently secured by his legal authority and his property rights; the recognition, the respect and the love which he shows above all for his wife are the best way of stabilising his rule over her, and making it successful. At this point another crucial aspect of moral concern for the conduct of the husband that Xenophon brings to bear becomes apparent: it is only within the framework of scientific economics, to which the art of leadership belongs, that dissolute behaviour on the part of the husband can be shown to be only superficial proof of his power; on closer study it is seen to be in fact a threat to the warranted relation of power between husband and wife in an oikos marriage. Xenophon recommends moderation to the husband as the best strategy for securing his rule, and this rule he should regard as being quite natural. The art of leadership, which already accounts for a significant, lengthy section of the Oeconomicus, then assumes a leading role in Aristotle’s economics as presented in the first book of his Politics. Essentially Aristotle distinguishes three elements of economics – the relationship between master and slave, between father and children and between husband and wife.124 This means that the art of leadership is the essence of all important aspects of economics, and Aristotle can sum up as follows:

124 Ar. Pol. i 12, 1259a36–b3.

154

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

Thus it is clear that household management attends more to men than to the acquisition of inanimate things, and to human excellence more than to the excellence of property which we call wealth, and to the excellence of freemen more than to the excellence of slaves.125

But within the framework of economics Aristotle not only attaches particular importance to the art of leadership, he also states with all possible clarity the problems involved that had already occupied Xenophon: A question may indeed be raised, whether there is an excellence at all in a slave beyond those of an instrument and of a servant – whether he can have the excellences of temperance, courage, justice, and the like; or whether slaves possess only bodily service. And whichever way we answer the question a difficulty arises; for, if they have excellence, in what will they differ from freemen? On the other hand, since they are men and share in rational principle, it seems absurd to say they have no excellence. A similar question may be raised about women and children, whether they too have excellences; ought a woman to be temperate and brave and just, and is a child to be called temperate, and intemperate, or not?126

For Aristotle this problem is of great political significance: if rulers give their subjects free rein they become unrestrained, but if they feel unwell they will pester their masters. In particular the unbridled behaviour of women, who after all account for one-half of the population, can have fatal consequences for the state, as the example of Sparta shows. If, as in Sparta, women ‘live in every sort of intemperance and luxury’, may even openly enjoy lesbian relationships, and take numerous decisions, then wealth is valued highly, greed is promoted and property is unevenly distributed.127 Thus it is more than justified when men restrict and limit the lack of discipline shown by women, though without claiming that they are completely without reason, for their cooperation is of vital importance.128 Aristotle’s solution to this theoretical and political problem is well known. There are various categories of forms of rule: the kyrios rules over slaves like over implements, over children as a king and over his wife like a 125 Ar. Pol. i 13, 1259b16–21. There is also an art of making a living for the estate manager that is part of economics, but which Aristotle only deals with en passant. However he explicitly states that this art does not aim to acquire riches (Pol. i 10, 1258a30 f.). Economics is not identical with making a living (Pol. i 8, 1256a10 f.) and has nothing to do with the art of earning money (Pol. i 9); on the practical art of making a living which includes agricultural skills, see Pol. i 11. On this see also Pol. i 3, 1253b5–13. 126 Ar. Pol. i 13, 1259b21–33. 127 Ar. Pol. ii 9, 1269b13–1270b6. 128 See Ar. Pol. i 13, 1260b15–19.

t h e a s y m m e t r i c a l r e l at i o n s h i p

155

statesman, ‘for the slave has no deliberative faculty at all; the woman has, but it is without authority, and the child has, but it is immature’. Thus a woman’s virtues – of bravery, justice and prudence – do not have the same quality as the corresponding virtues in a man; because of women’s lack of decisiveness female virtues are generally of a ‘servile’ kind:129 women can only be virtuous if they are constantly guided by men, just as appetition must be guided by intellect.130 This model of marriage as a kind of government has its drawbacks, however, for in most constitutional states the citizens rule and are ruled by turns, for the idea of a constitutional state implies that the natures of the citizens are equal, and do not differ at all. Nevertheless, when one rules and the other is ruled we endeavour to create a difference in outward forms and names and titles of respect . . . The relation of the male to the female is always of this kind.131

Thus this continual statesmanlike rule of men over women is threatened by the equality and lack of difference in this form of rule. This is all the more true as women have a central part to play in various communities: guaranteeing reproduction, the preservation of life and even a good life. It is a fact of nature that without women men can neither receive life, nor reproduce their sex nor draw up the model of a good life.132 It is precisely these propositions that lead to the demand for marital fidelity by the husband in the early peripatetic Oeconomica, a series of writings which were attributed to Aristotle, and are quite clearly influenced by the first book of his Politics and Xenophon’s Oeconomicus.133 With remarkable clarity and brevity the author points out that the treatment of women should enjoy a prominent position in economics: – because the union of men and women is necessary for reproduction, and is therefore more natural than any other kind of community; – because men and women need each other not only in order to secure the survival of human life, but in order to lead a happy life; 129 130 131 132 133

Ar. Pol. i 13, 1260a8–b7. Ar. Pol. i 5, 1254b3–9. Ar. Pol. i 12, 1259b4–9. Ar. Pol. i 2; see also NE viii 12, 1162a. To be more precise, this is only the case for book 1 of this tract. Book 2 reveals stoic influence, while book 3 is only known from a Latin translation, and is of uncertain date and authorship.

156

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

– because marriage serves to provide security in old age; – because men and women, thanks to their different natural aptitudes, which complement each other perfectly, can work together ideally as far as ‘economics’ are concerned.134 More remarkable, however, are new nuances which are to be found in the tone of this account of the Aristotelian conditions. There is no longer any reference whatsoever to the inferiority of women to men in this context. Instead emphasis is laid on cooperation, complementariness, goodwill and mutual help within marital relationships.135 For this reason the husband must have a vital interest in being well treated by his wife and ‘suffering no injustice or misfortune’. But this can only be achieved when he in turn does his wife no wrong, and ‘a man does his wife wrong when he gets involved with another woman’.136 Foucault believes that he can assume that this ‘involvement’, the ‘connections’ (synousiai) with other women, refer to permanent, serious affairs rather than occasional infidelity, and the comparison of women with supplicants who may not be wronged is a reference to the inferiority of women.137 But the former is pure speculation, while the latter is an explicit reference to one of Pythagoras’ doctrines. Thus when Foucault sees the justification for the demand for sexual moderation on the part of the husband in the early peripatetic Oeconomica partly in impersonal, partly in egoistic, motives (economical prosperity, the aesthetic stylisation of existence, a political career),138 then this is the result of an over-interpretation of the text that does not take sufficient account of the small but revealing differences between Aristotle, Xenophon and pseudo-Aristotle. Aristotle assumes an unambiguous, terminologically defined, scientific justification of a hierarchy between husband and wife, but at the same times articulates in all clarity the strategic problem for the husband’s dominance that this leads to. Xenophon attempts to solve the difficulty by suggesting that wives be treated in a humane, friendly manner, although his motive for this is the stabilisation of the husband’s superiority. But he also emphasises an intrinsic and not just an instrumental value that wives have who take care of their natural beauty and fulfil their professional tasks satisfactorily. This tendency is taken further by pseudo-Aristotle, who pointedly avoids any 134 See Pseud.-Ar. Oec. i, chs. 1–4. the clear reference to mortal human beings partaking in eternity through marriage is remarkable, (ch. 2). 135 Pseud.-Ar. Oec. i, ch. iii, 2. 136 Pseud.-Ar. Oec. i, ch. iv, 1. 137 Pseud.-Ar. Oec. i, ch. iv, 1. 138 See UP 180–4.

t h e a s y m m e t r i c a l r e l at i o n s h i p

157

reference to a social hierarchy in marriage, and goes even further than Xenophon and Aristotle had before him in accentuating the biological, social and economic weight of the wife in marital relationships and the oikos. If there was a tendency among the moralists of the classical period to recognise that women were equal to men, then Plato is naturally an important witness. Perhaps Foucault is right when he states that Plato’s later legislative suggestions in the Laws may have emphasised the symmetry between men and women, and in this context demanded the husband’s sexual faithfulness,139 yet this symmetry in no way implies that husband and wife are held to ‘sexual fidelity’ by a personal bond . . . The symmetry is not based on a direct and reciprocal relation between the two, but on an element that dominates both of them: principles and laws to which they are both subjected in the same way.140

But what Plato suggests was in many ways a special case – as ancient authors after him always thought. The oikos has no part to play in his model state; he dissolves the family and attaches the greatest importance to putting the polis into its best possible state. But above all, the husband’s sexual moderation is for Plato just one element in an ethical strategy that is part of a more general restriction of bodily pleasures; it propagates the avoidance of physical sexuality in homophile relationships, and limits heterosexuality to reproduction. It is precisely this systematic framework that enables Plato to formulate the (limited) symmetry and reciprocity of marital relationships as an ethical and political demand, one aim of which for husbands was that it ‘inspires men with affection for their own wives’.141 According to Foucault’s interpretation the realignment of the moral regulation of marital relationships towards the equality of wives only began in the Hellenistic period.142 One of the most important philosophical testimonies to this development is the surviving fragment of a treatise On Marriage, which was written by the last great representative of the Old Stoa, Antipator of Tarsus (c. 200–130 bce). In it Antipator defends marriage as a community which more than any other can lead to a full, flourishing and happy life – but only when the attraction between husband and wife is equal and mutual.143 One of the most touching witnesses to this 139 140 141 142 143

See Plato Nom. vi, 784d–e; viii, 838a–841b. UP 169. Plato Nom. viii, 839b. More generally Nom. viii, 839d–842a. See UP part 3. See SVF iii, frag. 63 (v. Arnim); also Canick-Lindemaier (1988).

158

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

attitude – beyond all moral and philosophical treatises – is the epitaph that the doctor Olycon composed for his wife about 20 bce. Remarkably he praises her not only for her role in the house, but also for her professional work (as a doctor): Greetings to you, Pantheia my wife, from your husband who, Since your passing on, has suffered unbearable pain at your mortal death. Hera Zygia never saw a wife so fine to behold, Neither in body, nor prudence nor manners. It was you who bore me the offspring who all have my features, And you looked after both your husband and your children. Your steady hand guided the rudder of life at home, You furthered the fame of the art of medicine, And although a woman your skills rivalled mine. So Glycon, your husband, has built you this tomb, That also hides the body of the immortal Philadelphos, And here I too will lie when I depart this life. Just as I was fated to share the marriage bed with only you, So too the earth of this grave will clothe us in the same shroud.144

However, it should not be overlooked that the ground for this attitude to marriage had already been prepared in the classical period, if in an inconsistent and fragmentary manner.145 In this context several great female figures in tragedy are rightly pointed out – not only Antigone or Electra, whose tragic lot did not have any immediate connection to the relationship between the sexes, but also the adulteress and murderess of her husband, Clytaemnestra, who confidently and provocatively justified her conduct,146 or Deianeira, who brought on her husband Heracles an agonising and ignominious death. The chorus may have disapproved of these deeds, but the central motive for the murderesses is impressively demonstrated: the emotional and mental injury suffered by the wives as a result of the ruthless behaviour of their husbands. Agamemnon had not only sacrificed his daughter Iphigeneia as if she were a mere animal, he had also slept with Chryseis and Cassandra, and Heracles even wanted his concubine Iole to move into his house; Deianeira’s desperate cry forebodes doom: ‘But then to live with her, sharing the same union – what woman could endure it?’147 144 145 146 147

Kaibel, Epigramm. Graec. 243b. (from the German by the translator). See on this the informative article by Vogt (1988), 118–67. Aeschylus, Agamemnon v. 1415 ff. Soph. Trach. 459 ff. Soph. Frg. 524 (p. 257 Nauck) preserves a complaint about the passive role women have to endure (‘the nothing of the female physis’).

t h e a s y m m e t r i c a l r e l at i o n s h i p

159

Euripides in particular thematised the conflict between the wife and her husband’s young lover in different variations. His Medea and Phaedra rebel against their merciless demotion by their husbands in a manner that was truly provocative in Athens at the time. Euripides even had to revise the first version of Hippolytus in which Phaedra – with an explicit reference to the intense homoerotic relationship of her husband with Peirithous – actively offers her services to her lover Hippolytus and suggests that he overthrow his stepfather.148 Certainly Euripides, the ‘dialectic’ tragedian, regularly provided a contrast to these tragic figures in the figure of the faithful, considerate wife, who patiently tolerated all her husband’s infidelities. But he has to justify this attitude, and in so doing reveals that it is no longer taken for granted, or natural.149 A few additional traces of this debate are to be found outside tragedy. The sophist Antiphon employs the ideal of the equality of spouses (‘think the same, breathe the same’) as an argument against marriage, since it is beyond a man’s power to fulfil these demands.150 This contrasts with Socrates, who according to Xenophon claimed that men and women were moral equals (the same arete) in a positive sense.151 One of Socrates’ most faithful students, Aeschines of Sphettos (an Athenian deme), vigorously defends the ethical equality of men and women in his dialogue Aspasia – so much can be reconstructed from the few surviving fragments – and proposes a new moral ideal for marital relationships founded on this equality.152 There is a remarkable statement in this direction by Isocrates, who was probably the most famous intellectual of the first half of the fourth century bce. In To Nicocles he reminds his ideal ruler: Furthermore I had no patience with the perversity of men who take women in marriage and make them partners in all the relations of their life [koinonian poiesamenoi pantos tou biou], and then are not satisfied with the compacts which they have made but by their own lawless pleasures bring pain to those whom they expect never to cause them pain; and who, though honest in all other partnerships, are without conscience in the partnership of marriage, when they ought to cherish this relationship the more faithfully inasmuch as it is more intimate and more precious than all others.153 On the accusation against Euripides see Aristophanes, Frogs v. 1043 ff. See e.g. Euripides frag. 909 and 822 (p. 653, 628 f. Nauck). See Antiphon frag. 49 DK. Xen. Symp. 2, 9; see also Diog. Laert. 6, 12, who attributes the same thesis to Antisthenes, a student of Socrates: andros kai gynaikos he aute arete. 152 See above all the fragment preserved in Cicero Iuvent. 1, 52 (see on this Dittmar (1912)). 153 Isocrates, Nicocles 40. 148 149 150 151

160

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

The utopian community of women and children which Plato defends in the Republic was probably a topic of discussion in Athens some twenty years before Plato finished writing it (about 375 bce). An indication of this is the fact that Aristophanes, who at the time still articulated the trends of contemporary debate in his comedies like no other, repeatedly deals with the theme – ‘the question of women’ in general. The disparagement of men’s deeds and the rejection of the deprecation of women in The Thesmophoriazusae, the love-strike to achieve peace in Lysistrata, and above all the description of the organisation of a women’s state and the proclamation of communism, and the community of goods and women in The Ecclesiazusae, are ridiculed with all the tools available to comedy, but for this very reason must also be appreciated as a reflection of contemporary debate.154 It would seem that this debate had an effect on the attitudes to marriage and the conduct of the husband which we find in Plato’s Laws, Xenophon’s Oeconomicus, Aristotle’s economics in Politics book i, and pseudo-Aristotle’s Oeconomica. Only when the variety of scientific, ethical and political thought on the value of marriage and the status of the wife in classical antiquity are taken into account is it possible to identify correctly their systematic place in the framework of contemporary moral writings. The moral problematisation of the sexual behaviour of the husband has its roots at the cultural interface of three powerful traditions; these emphasise the inferiority of women to men from the perspective of natural science, stress the economic, ethical and religious importance of the wife in the context of the oikos and the polis from the social standpoint and highlight the emotional vulnerability of wives from the viewpoint of poetry. It goes without saying that it is difficult to integrate these three traditions in a consistent fashion; indeed they lead to considerable moral and intellectual tensions. It is only within the framework of the new science of economics that an attempt is made to solve this complex problem – not always successfully, and with varying emphasis on the individual traditions.155 The regulative power which 154 Women claim ‘Is it not plain then that women are better than you men? We say women are best’ (Thesm. v. 799 f.); they have ‘a fine nature, grace, courage, intelligence and a wise and good love of our city’ (Lysistr. v. 545–7), and Praxagora is even able to persuade the men in the popular assembly that the rule of women and new social relationships within marriage are necessary (Ekkl. v. 170 ff.). 155 If book 3 of the pseudo-Aristotelian Oeconomica was written towards the end of the fourth century (which seems now to be accepted scholarly opinion – see Flashar (1983), 292), then it could be said to present the best balance of the three traditions. The work clearly has close links to Xenophon’s Oeconomicus and pseudo-Aristotle’s Oeconomica; but in contrast to the latter, Pseud.-Ar. Oec. iii concentrates exclusively on marital relationships, which it praises highly, while emphasising the economic role of the wife and the pedagogical guidance of the husband.

t h e a s y m m e t r i c a l r e l at i o n s h i p

161

these models recommend that the husband should still exercise over his wife is at the same time an expression of his natural superiority, his scientific knowledge, his respect and his self-control. The husband rules his wife at a juncture between political patriarchy, scientific economic knowledge and ethical self-discipline. Foucault’s three separate analytical levels of archaeology, genealogy and ethics are interwoven in the husband’s regulative power, although Foucault himself does not notice it. The threat to this regulative power was the subject of the ethical concern of moralists about the consequences of sexual dissipation on the part of the husband. At the same time it should not be overlooked that precisely this systematic framework restricts moral discussion of both homosexual and heterosexual relationships in classical antiquity to an extremely limited understanding of sexuality. They are thematised exclusively as relationships between a mature, educated, older man and an immature, unwilling, passive younger person, who is either not allowed to experience sexual feeling, or whose sexuality was immaterial, at least at the emotional level. As a result classical antiquity had no way of dealing with the concept of a reciprocal art of love in marital relations, much as was the case with homosexual relationships.156 This is all the more surprising when we consider the fact that contemporary scientific thought attributed to men and women, to young and old alike, the ability to experience sexual feelings – for example in an explicit analogisation of the sexual organs and mechanisms in men and women; 157 although this is always done with reference to individual, isolated persons and never from the viewpoint of a ‘relational’, interactive or even reciprocal sexuality. To put it bluntly: sexual, physical eros as a social phenomenon, as an interactive relationship between two (almost) equal partners, is never even taken into consideration by classical Greek authors. When physical eros is discussed, then it is localised in each case in single, isolated persons; and when eros represents a social relation in the full sense, then its sexual and physical dimension is completely ignored. ‘Relational’ sexuality is accepted as a topos worthy of discussion under the condition that the subject can be considered from a more general, non-specific sexual viewpoint – prudence, the securing of power, the demonstration of activeness, the upholding of freedom, the production of the good. In the language of a theory of knowledge this can be put by claiming that relational sexuality established in the form of a social relationship had not yet been constituted as a field of knowledge in the classical period. Classical Greek 156 See above, section 2 of this chapter, pp. 128 ff., esp. pp. 135–7. 157 See on this Laqueur (1992).

162

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

authors are a long way from openly considering sexual, and in particular marital or homoerotic relationships, and instead develop an extremely one-sided and uncomfortable attitude to relational sexuality. It is quite possible to draw the conclusion that in fact they want to distract from the actual social phenomenon of sexuality. It is easy to see that this attitude can indeed be consistent with Aristotle’s, and even Plato’s, acceptance of sexuality as a necessary, important and natural element of a good and flourishing life. For this acceptance is always approached from the perspective of the free, superior, active, penetrating man, and never from the perspective of the young male beloved or the young woman. The moral problematisation of homosexual and marital relationships in classical antiquity takes care to ensure that an appropriate ethical discussion of them does not break with this taboo – probably in the apt belief that abandoning it would bring about the collapse of important elements of the social organisation of the ancient polis.

5

THE EPISTEMIC EROS

Foucault’s analysis of Platonic erotics brings him to an extremely complicated domain of classical ancient philosophy, and one that can be discussed from a number of angles.1 In the ‘relation to truth’ he sees the decisive new perspective Plato introduces into the classical Greek theory on love. Erotics, as a purposeful art of love . . . will be our topic in this section as well. But this time it will be treated as a developmental context for the fourth of the great austerity themes that have run through the ethics of pleasure over the entire course of the history of the Western world. After the relation to the body and to health, after the relation to wives and to the institution of marriage, and after the relations to boys . . . I would now like to consider the relation to truth . . . in the form of an inquiry into the nature of true love.2

Within this new framework of erotics Foucault correctly diagnoses a series of fundamental shifts in the problematisation of love in the relevant Platonic dialogues, in particular the transition – from the question of behaviour in love, to the question of the nature of love; – from the question of the honour of the boy, to the question of the love of truth; 1 UP, part 5, 229–46. 2 UP 229.

163

164

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

– from the question of the asymmetry of erotic relationships, to the question of the convergence of mutual love; – from the question of the virtue of the beloved, to the question of love for the master and his virtue. With these shifts Plato’s erotics not only seeks to give – in respect of homoerotic relationships – a new and complex answer to the problem that the beloved boy is an object of desire; for the first time it also introduces the enquiry into truth – in the shape of an enquiry into true love – into erotics. Thus, according to Foucault, the form and the goal of ethical work are redefined in Plato: its aim is now to reveal the relationship of the subject to truth as the hidden basis of love, and to show that the soul must struggle against the power of appetition in a double relation to truth – within the framework of a search for knowledge of the nature of its own appetite, as well as in the form of a search for the true object of its appetite. In these transformations and shifts Foucault finally sees the origin of demands for abstinence, symmetry and knowledge which later were to be characteristic of the Christian ethics of sexuality.3 But Plato is not the only author of the classical period to introduce the perspective of truth and knowledge into the discussion of eros. Soon Aristotle was to enquire into the ‘essence’ or the ‘nature’ of love; but above all, as we have already seen, various ‘games of truth’ were involved in the way the ancient world approached erotic phenomena – dialectics in Aristotelian ethics, empirical methods of natural science in ancient dietetics and the politico-psychological art of leadership in ancient economics. Thus it will not be a question of studying at the ‘archaeological’ level the very fact that Plato related erotic behaviour to truth and knowledge, but rather how exactly the specific Platonic game of truth that is to be played in erotics is to be determined. However, from the perspective of the results of the erotic transformations apparent in Plato, Foucault reaches a conclusion that can indeed be linked to a widely held interpretation of erotics in Plato’s dialogues. According to this reading Plato demands that human beings should only be loved if they are beautiful and good. It is not individual, unique humans, with all their weaknesses and qualities that are the objects of Platonic love, but abstract versions of such people – versions that only consist of complexes of their best qualities. The absolute fulfilment of love as understood by Plato is the complete opposite of the overwhelming affection for 3 See above all UP 240–6.

the epistemic eros

165

concrete individuals that we generally identify with love today.4 So too Foucault claims: that Plato transforms the traditional characterisation of the person loved into an enquiry into the lover and love itself, the nature of which primarily identifies the true object of love; that Plato saw the essence of love in the pursuit of eternity and pure beauty, which could never be instantiated in perceptible things in the full sense; and that according to Plato it is in their yearning for this abstraction that the lover and the beloved can be united in complete harmony. Thus the aim of the ethical work of Platonic erotics is supposed to be a transcendence of relationships of love for individual persons, even if the relationships themselves must be the starting-points for this transcendence. However, Martha Nussbaum has recently produced a vigorous refutation of this view. Her stimulating interpretation of Plato’s Symposium places particular emphasis on the setting at the beginning of the dialogue, the speech by the comic poet Aristophanes in the middle of the banquet, and the description of Alcibiades and his sudden appearance at the end of the symposium and of the Symposium.5 The crucial claim of Nussbaum’s interpretation is that two incommensurable attitudes to life are linked to Socrates and Alcibiades: both men are portrayed as exemplifying types of personalities whose ways of life can never converge. Socrates’ attitude leads to philosophical knowledge, but at the same time turns us into insensitive beings who can no longer feel their bodies; Alcibiades’ attitude preserves and stabilises the body’s passion and feelings, but does not achieve abstract knowledge, leading instead to disorder, tumult, violence and death. For Nussbaum Plato’s main message is that we must choose between these two attitudes, that both have their advantages and values, but that we must also take into account the sacrifices that both make inevitable. To be sure, in the end Plato recommends Socrates’ way, but Nussbaum finds it remarkable how critical he is of it, as well as the dramatic way in which he so precisely sketches a viable alternative. 4 One of the most influential proponents of this interpretation is Vlastos (see Vlastos (1981); see also e.g. Irwin (1977)). According to Vlastos, in the Symposium beauty is only the vehicle for the actualisation of erotic energy that is to lead to the highest love of the forms. Platonic eros is deliberately applied not to persons, but to impersonal objects as markers for the predicates ‘useful’ and ‘beautiful’: to structures, political programmes, mathematical objects, scientific theories or ‘the good’. In Vlastos’s view Plato was the first to realise what deep feelings (even as far as an ‘erotic fixation’) we can experience for such wonderful impersonal objects, and how these feelings can grow into an authentic mystical spirituality. But the price for these advances in the theory of eros is extremely high, as Vlastos stresses: persons and individual subjects are only important as means and images and are of no intrinsic value. 5 See Nussbaum (1986), 165–99.

166

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

The very setting of the dialogue, which is always carefully constructed in Plato and is of particular relevance, connects events in the Symposium to the political environment in which Alcibiades actually lived. The dialogue presents a conversation between Apollodorus and a friend recounting an earlier discussion in which Apollodorus was involved, and in the course of which he remembered a speech by Aristodemus that, among other things, described an address by Socrates in which Socrates reported on a speech by Diotima, who in turn talked about some secrets of the mysteries. This complicated chain of reports underlines the importance of the event that is described. But what is more significant is the fact that Apollodorus’ earlier discussion had taken place several days previously, and was with Glaucon, a businessman. Glaucon was so eagerly curious that he had searched all over Athens for Apollodorus, in spite of the fact that he was a practical man who apparently knew nothing of philosophy and literature, and he did not realise that the symposium had taken place more than ten years before. Glaucon thought that the rumour about a banquet at which love was the subject of discussion, and at the end of which Alcibiades had appeared, was an indication that Alcibiades, the much-loved, much-hated general and son of Athens, who impressed everyone with his intelligence, his charm and his beauty, was back in the city. This setting is a reference to the dramatic situation in Athens in 404, when the democrats’ victory over the Thirty Tyrants was still fresh and unsure, the Spartans’ final victory over Athens was imminent and Alcibiades’ return to Athens would have helped the democrats’ cause enormously.6 Aristophanes is the only participant not to praise Socrates’ report of Diotima’s views,7 and in his speech Nussbaum sees a description of an unSocratic and unphilosophical approach to eros which is otherwise missing in the conventional interpretation of the Symposium. Aristophanes regards love as a relationship between two unique persons that can lead to absolute fusion and harmony, but at the same time is subject to luck, chance and risk: the risk of losing what is most important in the life of an individual is highest. The lovers are incomplete and unsatisfied on their own, but when they have found each other and are joined together, then the restlessness of their feelings and their affection come to an end. Eros is overcome.8 In stark contrast to this, according to Socrates Diotima preaches the submission of individual eros, the killing off of erotic physical feelings, and the erotic devotion to the eternal structure of beauty and of the good 6 See Nussbaum (1986), 168–71. Alcibiades was murdered in 404 on the orders of the Spartan general Lysander. 7 See Plato Symp. 212c4–5. 8 See Nussbaum (1986), 171–6.

the epistemic eros

167

itself; this move is supposed to free the transformed eros from the influence of chance and the threat of loss.9 Of course, Plato recommends us to endeavour to climb up this ladder of love, right up to the highest rung where the forms are visible, but according to Nussbaum the provocative interruption of the carefully constructed speeches by the sudden appearance of the drunken Alcibiades reminds us all too clearly of the price to be paid. For in spite of being drunk Alcibiades wants to tell the truth – indeed he speaks the truth, for he challenges the truth fanatic Socrates to stop him if he does not say what is true, yet Socrates does not interrupt him once. It is the relation to truth that places Alcibiades’ speech on such an important systematic level. But Nussbaum sees this claim and relation to truth not only in Socrates’ account and his relationship with Alcibiades, but in the form and the direction of Alcibiades’ speech. Alcibiades does not employ abstract concepts to tell the truth about eros, but images of an individual relationship that is lived and experienced with an individual, indeed unique human being. In this way Plato – according to Nussbaum – wants to suggest that particular experiences and knowledge of eros can only be gained through individual, physical and emotional episodes – pathonta gnonai, learning through experience and suffering is the ancient Greek proverb that is the heart of Alcibiades speech – a leitmotiv that is frequently to be found in ancient poetry. With this perspective Alcibiades takes up points of view that had already been hinted at in Aristophanes’ speech. The epistemological side of the orientation that he reveals consists of emphasising the particular knowledge that comprehends the universal not in the form of general definitions, but in individual, concrete images. In Nussbaum’s view, the two-world theory which on the traditional interpretation Plato always held in the Symposium, takes the form of a twoworld theory of eros; it is personified in the examples of Socrates and Alcibiades in a manner that not only demonstrates their incompatibility, but also illustrates the importance of each of them, and that they are equally worthwhile.10

9 See Nussbaum (1986) 177–84. 10 See Nussbaum (1986), 184–99, esp. 197 f.: Plato shows us, through Socrates and Diotima, how, despite our needy and mortal natures, we can transcend the merely personal in eros and ascend, through desire itself, to the good. But we are not yet persuaded that we can accept this vision of selfsufficiency and this model of practical understanding, since, with Vlastos, we feel that they omit something. What they omit is now movingly displayed to us in the person and the story of Alcibiades. We realize, through him, the deep importance unique passion has for ordinary beings; we see its irreplaceable contribution to understanding. But the story brings a further problem: it shows us clearly that we cannot simply add the love

168

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

In the following I would like to show that neither Foucault’s nor Nussbaum’s interpretation does full justice to the Platonic theory of eros – basically because their view of the epistemological game of truth connected with the philosophy of eros in Plato’s philosophy is not precise enough. I shall attempt to explain this criticism by (1) looking in detail at the systematic and dramatic relevance not only of Aristophanes’ speech, but of all the speeches in the Symposium before Diotima’s, and (2) on the basis of this explain the relationship between the individual speeches and Socrates’ and Alcibiades’ behaviour; next (3) I shall extend the horizon of the discussion by looking at the two other Platonic dialogues that deal with eros, the Lysis which is early and the Phaedrus which is late, and finally (4) I shall bring the political (‘genealogical’) and epistemological (‘archaeological’) dimension of Plato’s theory of eros into sharper focus. 1 In accordance with the traditional interpretation of the Symposium, Foucault tends to assume that Diotima’s lengthy speech presents the Platonic theory of eros, while the other speeches deal with current beliefs about love. ‘They illuminate the background of the Platonic doctrine, the raw material that Plato elaborates and transforms.’11 Diotima identifies eros as a demon, not as a god – an intermediate being between the mortal and the divine, that in spite of its inconstancy and restlessness strives for the better. According to the Platonic definition, love is the desire for the good, more precisely the wish to be in permanent possession of the good, and its characteristic work (ergon) is the production of the good and the beautiful. From a different perspective eros represents a finite mortal being’s striving for eternity and immortality; a movement up a ladder of love for the beautiful, that is also the good, to its next form: from love of a single beautiful body, the love of many beautiful bodies, the love of a single beautiful soul via the love of all souls, of fine manners, endeavour and knowledge, to the love of the Platonic form of the beautiful and good itself. Thus philosophy, as love for knowledge and truth, is one of the most important special forms of erotic striving. The transformations and shifts in the enquiry into eros that Foucault recognises in Plato’s philosophy are apparently a direct consequence of of Alcibiades to the ascent of Diotima . . . Socrates was serious when he spoke of two mutually exclusive varieties of vision. We see two kinds of value, two kinds of knowledge; and we see that we must choose. One sort of understanding blocks out the other. 11 UP 230.

the epistemic eros

169

Diotima’s speech. The discussion focuses on the essence of love and love of the truth; and in the practice of the self which works its way up to the highest, philosophical form of love, the loving master leads the way for his beloved pupil, eros joins together with knowledge to become the goal of an erotic episteme, and the lover and beloved converge in the form and the direction of their ethical work – in their erotic endeavour to achieve symmetry, knowledge and abstinence. An alternative interpretation of the Symposium should assume that not just Diotima’s speech, but all the other speeches on the subject of eros present important aspects of Platonic erotics, and so introduce an interesting corrective to Diotima’s speech.12 These aspects are given a dramatic setting by the different interests and attitudes to life of the characters involved: there is the young, versatile Phaedrus, who is interested in cosmology and mythology, but who also enthuses about thrilling orators like Lysias – his sceptical attitude to traditional mythology resulted in him being accused of blasphemy and sent into exile for over ten years;13 Pausanias, who had attended lectures by the famous sophist Prodicus, was knowledgeable in linguistics, semantics and rhetoric,14 while Eryximachus, a doctor and pupil of the influential sophist Hippias presents the natural-scientific side of the discussion;15 finally comedy and tragedy are represented by Aristophanes and Agathon, while Alcibiades stands for the emotional and political approach and Socrates (or Diotima) for the philosophical attitude to eros. What is more, nearly all of the symposiasts have a homoerotic relationship with one of the others, and these relationships are of carefully chosen types. There is the flagrantly erotic relationship between Eryximachus and Phaedrus,16 who as a couple suggest the topic of the speeches. On the other hand the relationship between Pausanias and Agathon seems already to have been sublimated into 12 Warner (1992) also assumes this thesis. He reaches the conclusion that the speeches prior to Diotima’s successively express in the vocabulary of feelings and psychology what Diotima then explains intellectually and abstractly. Rosen had already suggested a similar view; he sees the first speeches as ‘coherent steps in the gradual unfolding of the central question of the dialogue’ (Rosen (1968), 197 as well as the short synopsis 198–201). However, Rosen tends to emphasise the defects and shortcomings of the first five speeches from the perspective of Diotima’s speech. Nevertheless, his book on the Symposium is justifiably still regarded as one of the best-informed and most sensitive analyses of it. 13 See Heitsch (1993), 74 with references. 14 See Plato Prot. 315c. 15 See Plato Prot. 315c. 16 Socrates calls Eryximachus Phaedrus’ ‘companion’ (Plato Phaedr. 268a9), and the two of them are often mentioned together by Plato (Prot. 315c2–3, Symp. 176b5–c2, 177a; Phaedr. 218a7, 223b7).

170

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

friendship, whereas the attraction between Alcibiades and Agathon is full of tension and problems. Alcibiades is also involved in a subtle, complex and changing relationship with Socrates. The sole exception is Aristophanes – just as the subject matter of Attic comedy confirms that it was on the whole heterosexually oriented, so Aristophanes, in a speech about a fantastic metaphysics of the origin of eros, is the only symposiast who claims that heterosexual and homosexual (homophile and lesbian) love are of equal value;17 furthermore he is the only one of the guests not to acclaim Socrates’ account of Diotima’s speech.18 Thus in Plato’s Symposium the different aspects, and the complex manifestations of eros at various levels, are given a dramatic backdrop through the choice of symposiasts and the various kinds of relationships they have with each other, and then presented in more detail in the different speeches presented by the symposiasts. Phaedrus, a specialist in cosmology and mythology, presents eros as a primeval cosmic power,19 and quotes Hesiod’s Theognis as proof, according to which eros came into being from the very first division of the undivided first cosmic principle (Chaos in Hesiod). As an all-embracing cosmic power love is irresistible: it produces the closest personal ties between humans which even survive beyond death, and at best seem to allow sublimation but never complete asceticism; as a personal tie eros is directed towards ethical excellence, political virtue and a glorious life – a first indication of the correct form of erotic sublimation. This perspective turns the traditional view of pederastic relationships around: the relationship remains asymmetrical, but divine, cosmic eros as the goal of erotic striving is more likely to be found in the lover than in the beloved, while eros as an active power will be more common in the beloved than the lover,20 for generally the beloved sacrifices himself for the lover and not the other way round.21

Plato Symp. 191d–e. Plato Symp. 212c. Plato Symp. 178a–180b. The gods ‘are even more amazed, delighted and beneficent when the beloved shows such devotion to his lover, than when the lover does the same for his beloved. For the lover, by virtue of Love’s inspiration is always nearer than his beloved to the gods’ (Plato Symp. 180b). 21 Rosen (1968), 39–59 interprets the three most important aspects of Phaedrus’ speech only from the standpoint of their deficiency: the reference to eros as a cosmic power is inspired by an inappropriate demythologisation of the gods, and from an excessively radical materialisation of physics; the reversal of the pederastic relationship is indebted to the much despised image of the effeminate homosexual with physical erotic feelings for the lover; and the pederastic relationship itself is described mainly in terms of usefulness

17 18 19 20

the epistemic eros

171

But whereas Phaedrus praises eros from the perspective of the beloved in a passionate relationship, the next speaker, Pausanias,22 talks from the standpoint of the lover whose earlier physical love for the young, handsome Agathon has now been sublimated into a deep friendship, and this experience – as well as his training in semantics and rhetoric – colours his speech. Thus it is Pausanias’ job to suggest a conceptual distinction between ‘earthly’ and ‘divine’ eros23 – in other words a two-world theory of eros that assumes paramount importance in a variety of ways in the speeches and discussions that follow. Earthly, physical eros is both insatiable and unreliable in its exclusive fixation on the sexual satisfaction of the lover; the divine, spiritual eros, on the other hand, seeks out rational souls in a modest manner and aims at the beloved’s improvement. However, earthly and divine eros can also be distinguished on the part of the beloved and not just of the lover. It is from this perspective that the conditions are to be defined that govern when the beloved should submit and allow touching, hugging, kissing or the lover to occasionally satisfy himself sexually between his thighs: it is disgraceful, and an expression of earthly eros, if the beloved submits too quickly, is influenced by improper means such as money or power or acts out of an unworthy desire for sexual satisfaction. On the other hand, it is beautiful, and an expression of divine eros, if the beloved only submits after prolonged resistance to courting and pleading, having been persuaded by good speeches and services, and thus with the intention of attaining virtue and education.24 At first sight Foucault seems to be right when he sees such distinctions and regulations as part of current views on eros which are to be transformed and transcended later through the philosophical theory of eros.25 However, a closer look reveals that Pausanias’ two-world theory already goes far beyond current views on erotics. For although at the beginning of his speech he draws a sharp, deliberate distinction between heavenly and earthly eros with regard to their object, form and aim, at the same time he does not claim that earthly and physical eros are incommensurable, but in fact outlines the conditions under which physical eros can be part

22 23 24 25

and personal pleasure. Thus Rosen sees Phaedrus as representing a vulgar or material eros. This interpretation overlooks the – from the Platonic viewpoint – positive moments in the speech and also runs contrary to the flair of the symposium that Plato describes. Plato Symp. 180c–185c. Plato Symp. 180d–181a. Plato Symp. 183d–185b. See above, pp. 121–2, 131–2. Similarly Ferrari (1992) also assumes that Plato ‘diverts certain received opinions about love to his own peculiar philosophic ends’, but that in the Lysis, the Symposium and the Phaedrus he does not present a ‘comprehensive theory of love’ (248).

172

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

of divine eros – after all ‘submission’ can be ‘beautiful’. What matters is not that but how physical eros is related to the body. If it is inconsistent, quick, unfounded, insatiable and purely physical, in other words is generally egocentric and concentrates on the body, then it is disgraceful and to be criticised; in other words it is truly ‘earthly’. Divine eros, in contrast, is characterised by the moderate and rational integration of physical eros, rather than its exclusion, as defined by the conditions of ‘beautiful’ submission which directly follow from the nature of divine eros. It is not so much a question of giving up physical eros for divine, spiritual love, or of sublimating physical to spiritual eros, but of an insight into the proper use of physical pleasure that is aware of the functional role of sexuality in attaining virtue.26 Physical eros is problematic and gives cause for concern only when it remains isolated and does not have any further goal, but not when it is incorporated correctly into the pursuit of virtue and knowledge and subjugated to the power of divine eros. Pausanias develops a subtle form of the two-world theory that manages to avoid the simplification often encountered in other approaches, and reveals a systematic direction to which Plato’s philosophical erotics always remains true.27 The seating arrangement of the symposiasts requires the comic poet Aristophanes to speak next after Pausanias, but he gets such a bad attack of hiccups that he asks Eryximachus to help, which he happily does in two ways; as a doctor by giving him medical advice and as a friend by declaring that he is prepared to speak in his stead. A bodily disturbance that draws attention to several decisive weaknesses in the first speeches about eros raises its head in the shape of Aristophanes’ hiccups: order and disorder of both the soul and the body must be related to the difference between earthly and divine eros. In addition, it must be acknowledged that it is the specific task of skilled people like doctors to produce order and get rid

26 See above all Plato Symp. 183d–e; the shortest and most succinct statement is to be found at 185b: ‘it is right to let the lover have his way in the interests of virtue’. 27 These aspects are completely absent on Rosen’s critical-moralistic interpretation. In Rosen’s opinion Pausanias’ theoretical and conceptual arguments are only aimed at justifying his ‘selfish’ wish that as a pederastic lover he should also gain physical satisfaction. This is why Pausanias – according to Rosen – does seriously attempt to distinguish between vulgar and sublime eros, but to develop a theory of style in approaches to erotics that aims at a relaxation of legal and moral reservations in classical society about pederastic relationships involving physical experiences (see Rosen (1968), 60–89, esp. 86: for Pausanias ‘intelligence and refinement is a secondary or epi-phenomenon of caution or efficient selfishness. His real commitment to efficiency leads him to praise vice in the act of honoring virtue.’). This view is clearly influenced by a two-world Platonism that itself is based on an excessively restrictive interpretation of Plato.

the epistemic eros

173

of disturbances.28 The speech delivered by the doctor Eryximachus deals with these standpoints and so aims to integrate, state more precisely and generalise the speeches of both his predecessors.29 Indeed the eros of the primeval cosmic power with its twin aspects is now defined more exactly by Eryximachus – in two ways, in that the might of eros is explained by the expansion of its domain and its twin aspect by the nature of its ‘work’ (ergon). Without doubt it is primarily the physical dimension that Eryximachus touches on first: at the beginning of his speech he emphasises that eros has power not only over the human soul, but also over the body. What makes this perspective so significant is the fact that – as soon becomes clear – we are indeed dealing with divine eros in this context. In this way Eryximachus corrects Pausanias’ oversimplified view that divine eros is directed at the soul and earthly eros at the body. In fact both forms of eros are directed at both the human soul and the human body – and what is more, at all important domains of the cosmos.30 This universality reveals eros’ truly cosmic dimension; but this is also the reason why its twin aspects cannot be distinguished by reference to their domains. In Eryximachus’ view divine and earthly eros as cosmic powers can instead be distinguished on the basis of their ‘work’ (ergon), and the production of this work. As a doctor he reaches this conclusion by considering the paradigm of his profession. The goal of the doctor’s art is to achieve – with the aid of dietetic regulations, etc. – a physical equilibrium of the basic qualities, which is to be understood as the proper order of the body. He connects the aims and the work of divine eros as a natural cosmic power with the aims and the work of medical science for human bodies, and so links spiritual erotics with dietetics of the body. On the other hand, the exclusive fixation of earthly eros on physical desires leads to a disturbance of the qualitative equilibrium in the domain of the body, and thus its work can be described as bodily disorder. From this perspective the dynamic of eros can also be identified more exactly. As a dynamic process the divine eros of doctors uses its expertise to guide the body from disorder to order, whereas due to its ignorance earthly eros brings physical chaos.31 28 Plato Symp. 185c–e. The hiccups episode is often interpreted as Plato’s brief, subtle revenge for the way Aristophanes had ridiculed Socrates in the Clouds. However, this interpretation clearly does not go far enough. 29 Plato Symp. 186a–188e. 30 Plato Symp. 186a–b. 31 Plato Symp. 187b–e.

174

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

According to Eryximachus this viewpoint can easily be generalised: just as eros is directed at all parts of the cosmos, so thanks to its expertise and skill it produces in its divine aspects order everywhere, but in its earthly aspect disorder. Thus every skill (techne) represents an important form of erotic striving in the cosmos – expert human activity instantiates eros as a cosmic power.32 In this way for the first time in the Symposium not only is the goal of human eros, that is the promotion of practical reason and rational order in the beloved, but also human eros as an activity and motion, intimately linked with that activity which is guided by knowledge and which as a variety of regulative power aims at passing on education and science. Thus an ‘erotic’ interpretation is provided for the Socratic–Platonic connection between virtue and knowledge.33 It is as though Eryximachus sees a razor-sharp distinction between divine and earthly eros that makes the two aspects of eros absolute opposites. But this picture is too simple, for Eryximachus quite clearly points out that both forms of eros are present in many human endeavours: for he is the fair and heavenly one . . . But as for that other, the earthly Love . . . whatever we have to do with him we must be very careful not to add the evils of excess to the enjoyment of the pleasure he affords . . . And so in music, in medicine and in every activity, whether sacred or profane, we must do our utmost to distinguish the two kinds of Love, for you may be sure that they will both be there.34

It is certainly not the case that human activities are either an exclusive expression of earthly, or an exclusive expression of divine eros, for in fact they often integrate both forms. What is more, earthly eros can also be part of expertise. This is an indication of the tension in the definition of earthly eros that is never completely resolved within the framework of Eryximachus’ speech; earthly eros would seem to be partly expert, partly 32 Plato Symp. 186c–187b. 33 When Rosen (1968), 90–119 stresses the scientific ‘technicism’ of Eryximachus’ speech in particular, he does this primarily in order to point out serious shortcomings of this approach: it implies that the soul is corporeal, and overlooks the fact that physical health simply is not a good paradigm for spiritual well-being. Eryximachus plays down the irrevocable distinction between nature and humans that cannot even be expressed in terms of scientific medicine. Thus, according to Rosen, Eryximachus’ defence of pederasty contains contradictions, for scientific medicine better supports a preference for heterosexual relationships. This harsh moralistic viewpoint completely excludes the important positive moments in Eryximachus’ speech, an indication that it is inappropriate, especially from Plato’s perspective. 34 Plato Symp. 187e.

the epistemic eros

175

inexpert. However, there are discreet hints at a possible solution when Eryximachus considers the relationship between medical dietetics and cookery as a paradigm:35 as a representative of earthly eros cookery is also accompanied by expertise, but its aim is a skilful sublimation and maximisation of physical pleasures, which themselves threaten to destroy the bodily order that is pursued by medical dietetics as a specific form of divine eros. Thus Eryximachus identifies the anthropological representative of divine eros in its cosmic dimension as an activity guided by knowledge, and which aims at the production and the stabilisation of order in all domains, and in doing so he would seem to touch on all the essential elements of philosophical erotics that are dealt with later in more detail in Diotima’s speech. It is then all the more surprising that the poets Aristophanes and Agathon, who at the end of the symposium are the only symposiasts who can still keep pace with the hardened drinker Socrates, talk before Diotima.36 This dramatic setting gives considerable weight to their speeches, and must be understood as a literary device emphasising their connection with what Diotima has to say. If we recall that the first three speeches are closely connected by their subject matter, then we cannot help gaining the impression that the six speeches about eros in the Symposium are divided into two blocks of three speeches each. And this is all the more significant given that Aristophanes and Agathon put eros in a completely new light. Both writers pass over its cosmic dimension in their speeches, and avoid the distinction between divine and earthly eros; instead they concentrate on intricate and subtle aspects of erotic relationships between humans that are not accessible to the enthusiastic youth Phaedrus, or the trained orator Pausanias or the professional scientist Eryximachus, but are only visible to the fine sensibility of a poet. One of the most important aspects of this new approach that features in the last three speeches is, as was already mentioned above, the fact that Aristophanes, Agathon and Diotima now interpret eros as a single, indivisible search for the truly beautiful and good. It is this perspective that reveals the full dramatic relevance of Aristophanes’ attack of hiccups, for it was this indisposition that led to the speeches in fact being delivered in the correct order; because of the subject of his speech Eryximachus may not speak after Aristophanes.37 35 Plato Symp. 187e. 36 Plato Symp. 223c. 37 See Ferrari (1992), 250–2.

176

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

It is understandable that in her search for evidence for an alternative to Socratic erotics in the Symposium Martha Nussbaum draws attention to Aristophanes’ speech. He is the only symposiast who does not expressly praise Socrates’ report of Diotima’s speech,38 and at the beginning of his speech he stresses that he intends ‘to take quite a different line’ from those who have spoken before him.39 Thus Aristophanes distances himself in two ways, and in Nussbaum’s view one of the levels at which he does this refers to the comic element of the myth that he tells. The original eightlimbed, ball-like humans are strange enough figures; but their division makes of them grotesque creatures with their heads and genitals in the wrong place. Even after these limbs have been relocated, humans remain, as Nussbaum stresses, creatures that wish to be autarkic and divine – but in truth hectically charge around in desperation, searching for their other half in order to fit together the strange holes in their bodies and their external members.40 But nevertheless, the most important points in Aristophanes’ speech are founded on precisely these abstruse circumstances, and he relies on them to distance himself from the other speakers. The crucial aspect of this fundamental difference from the previous speeches follows directly from the metaphysics of the origin of humans that Aristophanes presents: eros is based on an old loss, on an irreversible deficiency not only of unity, but of strength and autarky such as the original eight-limbed, ‘round’ human creatures had possessed. Separation, need, weakness, in short: fragility, as Martha Nussbaum is quite right to stress, is the pre-condition and the foundation of eros, the reversal of this fragility is its aim. Thus for Aristophanes eros is a relationship between two unmistakable, irreplaceable human beings, who joyfully seek to re-establish their specific original unity. Successful eros is unique in each case: it depends on each individual finding exactly the right other individual, the other half that fits. The central theme of Aristophanes’ speech is individual love; he is not interested in explaining what it is that we love so excellently in others or in how we identify eros as a power; he does not define the work of eros as a structural order of the body or the soul that is attained with the help of the lover, but nevertheless primarily concerned with the beloved as an individual; what eros really aims at is a togetherness that is so intimate 38 Plato Symp. 212c. 39 Plato Symp 189c; the entire speech 189c–193d. 40 See Nussbaum (1986), 171–6.

the epistemic eros

177

and unique that it comes close to the original unity and fusion that was realised in the mythical, ball-like humans.41 This togetherness is based essentially on the ‘relocation’ of the genitals after the division of the unity, and thus requires the satisfaction of physical sexual needs, which in turn can itself be conducive to conducting other business.42 But above all, erotic togetherness is seen here as a reciprocal relationship that can be heterosexual, homophile or lesbian: Aristophanes’ eros breaks the tight cultural shackles of pederastic relationships. It is surely no coincidence that Aristophanes is the only speaker who is not involved in a pederastic relationship in one way or another. But most importantly this broad spectrum of possible relationships underpins the impression that in spite of its manifold forms love is a single, general aspiration.43 But the work of eros as presented by Aristophanes must also be studied from a different perspective. Only on the basis of erotic satisfaction can humans go about their business and lead a rational life; but when the lovers have found each other, then they wish to stay together and no longer do anything else – their life gets out of control and all other goals and wishes are neglected.44 What is more, erotic relationships are contingent in the extreme, and yet seek the reversal of this contingency in eternity and immortality, just as the lovers’ deficiency and need find their end in the autarky of successful eros. In other words, eros is imbued with a deepseated tension that manifests itself in various guises, which, however, have a common structure based on the fact that eros is a striving for the reversal of eros, and that therefore the existence of eros is incompatible with its telos. The highest goal of eros is the destruction of eros, the transformation of the lovers into an existence that no longer needs eros.45 41 ‘And so when this boy lover . . . is fortunate enough to meet his other half, they are both so intoxicated with affection, with friendship and with love, that they cannot bear to let each other out of sight for a single instant’ (Symp. 192b–c), and what the two of them always wanted is ‘to be rolled into one, so that (they) could always be together, day and night, and never be parted again’ (192e). 42 Plato Symp. 191b–c. However, the production of this intimacy and the satisfaction of eros are beyond the power of humans and cannot be redirected or distracted by education or expert knowledge. 43 So too Ferrari (1992), 252. 44 Plato Symp. 192d–e. 45 Rosen (1968) interprets Aristophanes from the perspective of a political poet who was extremely sceptical about contemporary new developments in the natural sciences, the sophistic movement and philosophy, as can be seen, for example, in the Clouds. According to Rosen, Aristophanes’ ‘key premise’ is that ‘the rational investigation of the cosmos leads to atheism and the destruction of political stability. By replacing the gods with natural

178

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

The speech by the second poet among those present, the host Agathon,46 seems to offer hardly any new distinctions and during its course becomes more and more of an encomium ending in a climax of poetic ecstasy. The brilliant rhetoric and the rich ornamentation of its composition remind Socrates of Plato’s deadly enemy Gorgias. But whether the speech tells the truth seems more than doubtful and so it is no wonder that Agathon does not pass Socrates’ subsequent test.47 But nevertheless, in spite of its rhetorical stylisation, the erotic enthusiasm that the speech generates is an important aspect of eros that is introduced by Agathon and was to be resumed in the Phaedrus, where eros is identified as a mania. But in his over-enthusiasm Agathon is moved to describe eros as a perfect god – as a ‘beautiful’ god, who is young, gentle and soft, who loves flower-filled, sweet-smelling places and gives splendour to life; and as a ‘good’ god who belongs to a flourishing life that leads not to violence but to creativity, not to restriction but to autonomy, and who is accompanied by virtue, is himself virtuous and is a level-headed ruler over pleasure and appetite. This characterisation is intended to identify eros as a reasoned striving and a rational desire, but neither is Agathon aware of this hidden agenda nor does he see the incompatibility of this identification and the perfection of divine eros. It is exactly this point that Socrates refutes in Agathon’s argument, while Agathon’s suggestion that eros should be praised not only for its effects, but also for its true nature, and his attempt to interpret the nature of eros as a beneficial combination of the soul and the body, of reason and desire, of virtue and sensuality, are not touched on by Socrates in his elenxis.48 This is all the more relevant when we take into consideration the fact that the refutation is based on a problematical conclusion that Agathon principles, this investigation removes the sole criterion for distinguishing between the just and unjust logoi ’ (122). His opposition is inspired by the muses, and so it is quite logical for him to introduce a myth, albeit with what Rosen sees as a pessimistic message: the separation of the ball-shaped humans is irreversible, the split sexual nature of humans also leads inevitably to restless, unstable harmony between opposites in the political sphere, and this can at best only be reduced by law, convention and tradition. However, Rosen’s interpretation of Aristophanes’ speech in the Symposium is one-sided, drawing too much from the viewpoint of comedy, and it overlooks the most important philosophical message of the speech. 46 Plato Symp. 194e–197c. 47 Plato Symp. 198b–201c. 48 Thus Agathon does not present a radical individualism inspired by Gorgias’ nihilism, as Rosen suggests. The implicit tendency of Agathon’s speech to which Socrates reacts goes beyond the price of youth and innovation, and beyond the danger of political atomism which Rosen sees as its central topic (see Rosen (1968), 159–96). Ferrari (1992), 252 f., in contrast, praises the positive aspects of Agathon’s speech in a more appropriate manner.

the epistemic eros

179

does not recognise because of his lack of an adequate philosophical education. From the thesis that when X loves Y, then X is lacking in respect of Y, Socrates concludes – and Agathon agrees with him – that eros is lacking in respect of the beautiful, and therefore is not itself beautiful. But this conclusion is inadmissible, since clearly it implicitly intensifies the ‘is’ of predication to an ‘is’ of identification. The step from ‘Y is beautiful’ to ‘Y is the beautiful’ is of course of enormous philosophical importance, and is spelled out plainly in Diotima’s speech. It is this unrecognised but dubious step that trips up Agathon, yet it remains a critical and problematic point in Socrates’ argument. Diotima’s speech is Plato’s answer to this problem. If we follow the speeches by the two poets, then eros is a unique, intimate, reciprocal relationship between two unmistakable individuals based on loss and deficiency, which nevertheless is also at the heart of an emotionally subtle, flourishing, splendid life, and combines itself with virtue and knowledge to produce an attitude of lively, enjoyable rationality – a relationship that thanks to its highest goal at the same time has a tendency to point beyond itself to an existence for which neither finality, nor need, nor emotion, nor eros are of importance. According to this picture the conceptual distinction between earthly and divine eros, as well as between the anthropological and the cosmological dimension of eros, is hardly relevant for the finite life itself; instead it is shifted to distinctions and tensions that are rephrased from the perspective of the contrast between life and immortality, between finality and godliness and between needy individuality and higher autarkic unity. It should be clear by now that the speeches of the five symposiasts, who each in their own way pay homage to eros before Socrates and Diotima do so, draw attention to a series of standpoints which are to be of importance in Plato’s philosophical erotics. The interpretation of eros as an irresistible cosmic power, which is nevertheless exemplified in personal human relationships; the distinction between divine and earthly eros which can be overcome to some extent by the integration of the correct use of philosophical eros into the love of beautiful souls; the technological identification of the work of eros and the structural definition of the erotic object, which, however, is then relativised by the poetic description of an erotic relationship between unmistakable individuals; and finally the dramatic tension between the existence and the goal of eros, which is incompatible with a perfect form of existence – all of these aspects are taken up by Diotima, who integrates and states them more precisely in her speech. Most importantly, Diotima removes the latent inconsistencies that seem to be inherent in these elements. Thus the first five speeches in the

180

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

Symposium by no means merely represent the current views on eros, as Foucault, following many other scholars, assumes; rather they introduce some of the most important aspects for philosophical eros. This also means that the speeches do not develop an alternative, opposite standpoint to Diotima’s speech, as is proposed by Nussbaum, but that the suggestions made by the symposiasts are worked into philosophical erotics and integrated systematically. If this conclusion can be upheld, then it means that Foucault, Nussbaum and the traditional reading of the Symposium are wrong in assuming that Plato’s theory of eros recommends the model of a life of asceticism and impersonal intellectualism.49 2 In several succinct definitions Diotima presents a systematic summary of the main standpoints of the previous speeches: – Eros is a demon that is situated between mortals and gods, and oscillates between the two; it is by virtue of its origin an intermediate being, that nevertheless strives towards the better.50 49 In her sensitive and deep article Arlene Saxonhouse introduces an interesting aspect (Saxonhouse (1984)). In Plato’s Republic women are allowed to participate in politics, but the price is the elimination of distinctions between the sexes, and of the erotic sphere. The Symposium poses the question of what happens to the female when eros returns. According to Saxonhouse it is significant that music, women and flute-players are excluded at the beginning of the dinner. The men want to be alone and the first five speeches concentrate primarily on politics as a sphere of the male. However, the central sixth speech is by a woman, and so Diotima makes the political sphere vanish. As an intermediate being eros is now essentially unpolitical, and so belongs to the domain of metaphysical-spiritual tension between the finite and the eternal. But above all, although it is neither male nor female, the effects of eros are described by means of an essentially female category: eros makes humans fertile, pregnant and capable of giving birth, and it is this female characteristic that makes them capable of intellectual ascent and produces the connection with the immortal and divine. At the end of the meal, after Diotima’s speech, music and women return together with Alcibiades, who is drunk. Alcibiades begins his speech with images of flute-players, and associates Socrates with a Silenus, a flute-player whose speech has the power of music. For Alcibiades, Socrates is a combination of the male and the female, of speech and music. He is a hermaphrodite who proudly presents his female side in his skills as a midwife. 50 Plato Symp. 202a–204c (this passage also stresses the philosophical element of eros: eros is ‘a lifelong seeker after truth’ (203d), for it floats between wisdom and ignorance, ‘But secondly, he brings his father’s resourcefulness to his designs upon the beautiful and good’ (203d), and so is ‘a lover of wisdom’ and stands ‘between wisdom and ignorance’ (204b)). At first sight there seems to be a parallel to Diotima’s refutation of Socrates in Socrates’ refutation of Agathon (Symp. 201e). But a closer look reveals an important difference: Diotima does not demonstrate that eros is not beautiful, but that it is not a god; it has needs, whereas gods are complete (happy).

the epistemic eros

181

– Human love is desire of the good, or to be more precise, a desire for permanent possession of the good, that is for happiness.51 – The work of eros is the production of the good in the beautiful, in all the different forms of the good and the beautiful.52 Thus eros is indeed an old and venerable cosmic power, but it is not a god, nor is it perfect, but in fact mediates between deficiency and perfection, between human finality and divine eternity. The proof of this intermediate status receives a great deal of attention in the Symposium, for it is of some systematic significance – not only because it facilitates the analogy between erotic and philosophical attitudes, but because it demonstrates that eros oscillates between individual and universal, between contingency and necessity and between finality and eternity, and in a certain sense is directed at both domains at the same time. Seen from this perspective, the claims that in Eryximachus’ speech eros is oriented towards order and universal structures, and in Aristophanes’ towards the individual and specific, are both justified and one-sided: Agathon may have found the right approach when he tries to characterise erotic endeavour as a combination of desire and practical reason, but he fails to realise (as was shown above) that when this characterisation is fully understood it is incompatible with his own definition of eros as a magnificent god. Diotima’s definition of eros as a striving for permanent possession of the good would seem to confirm well-known reservations about Plato’s theory of eros; eros is supposed to be possessive, egoistic and impersonal. But in fact this picture gets dismantled in various respects, and so is too simple. The good stands for eternity and divinity,53 and thus for the destruction of the interests of mortal beings: the aim of eros is to overcome egoism, and this process is an intrinsic erotic tendency. Furthermore the good stands generally for all specific forms of the good: no instance is excluded, and thus it is necessary for it to focus on an impersonal good. But, above all, it is the definition of the ‘work’ of eros that implies a significant clarification. The production of the good in the beautiful 51 Plato Symp. 204c–206a. In the chapter on Plato in her quite remarkable book (Eros, the Socratic Spirit) Catherine Osborne (1994) presents these statements in the form of the two theses that eros is a lover, not a beloved, and that it has knowledge of what it lacks and yearns for – in structural analogy to Socrates’ knowledge of his lack of knowledge. She concludes from this that for Plato eros had primacy over the object of love: we must always be inspired by eros if we want to recognise that the beautiful is worthy of love (see Osborne (1994), 86–116; also note 86 below). 52 Plato Symp. 206b–207a. 53 Plato Symp. 207a–208b.

182

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

not only means an instrumentalisation and devaluation of splendour and aesthetics, it also represents the decisive mediation between finality and eternity that erotic striving achieves. This mediation is the form in which our eros seeks to overcome its internal tension between existence and telos. Fundamentally what is involved is the production of the good in another beautiful, which is intended to improve the other (mainly other humans). This is the primary way in which eros is productive, directing eros beyond the lover towards the other, and overcoming its apparent egoism. Eros transforms ethical work into work on the other. This is particularly so in that the creation of the good in a beautiful person is already connected to the production of better endeavours and knowledge that are properly the object of higher forms of eros. However, it is precisely the capacity for epistemic generalisation of the beautiful that coincides with the production of these good things in beautiful humans: the beloved turning his eyes towards the open sea of beauty, he will find on such contemplation the seed of the most fruitful discourse and the loftiest thought, and reap a golden harvest of philosophy, until, confirmed and strengthened, he will come upon one single form of knowledge, the knowledge of beauty I am about to speak of.54

The lover is not yet capable of seeing the forms, but rather develops himself and attains higher levels only in his dealings with humans he loves. In Diotima’s speech the intermediate status of eros is manifested above all in the ladder and hierarchy of erotic attitudes that lead from our love of an individual beautiful body, via the love of many beautiful bodies, of a beautiful soul, of all beautiful souls, of beautiful morals, customs and endeavours as well as beautiful knowledge, finally to love of the form of the beautiful.55 This image provides a double corrective to the twoworld doctrine of eros as proposed by Pausanias and Eryximachus: on the one hand there are not just two, but many different, erotic attitudes that form a virtual continuum between an upper and a lower limit, and on the other hand the ‘lower’ forms of eros are not branded as ‘earthly’, but recognised as a vital, necessary starting-point for ‘higher’ erotic attitudes. It is stressed that striving for the good guides all kinds of eros – no type of eros is so ‘earthly’ that it would be excluded, at least not once its true conduct has been recognised. In this sense Pausanias and Eryximachus were right in their attempt to distinguish different erotic orientations, 54 Plato Symp. 210c–d. 55 Plato Symp. 210a–212a.

the epistemic eros

183

but wrong with their suggestion that eros can be divided into two aspects with normative boundaries; and Aristophanes and Agathon were right when they refused to adopt Pausanias’ and Eryximachus’ over-simplified distinction, but wrong with their strategy of not allowing any distinction whatsoever. Diotima makes it clear that those who wish to climb to the very top of the ladder of erotic attitudes must be capable of appreciating and using each of these attitudes as a rung leading up to the next orientation. This is at the heart of erotic practice, but it also has an epistemological side to it: it is vital that anyone who wants to be successful in this practice is capable of understanding that the objects of this practice on each of the rungs – beautiful bodies, beautiful souls, beautiful knowledge, etc. – are entities that show the way to other kinds of entities on the next rung up. Thus the erotic practitioner should possess the capability of identifying the good in individual bodies with the good in all bodies, in souls and in attitudes, and finally of recognising the many beautiful things as instantiations of the form of the beautiful (and good).56 Diotima provides no answer to the question of the extent to which the rungs of the erotic ladder differ, nor of how the hierarchy of the ladder is defined in detail; those are technical aspects of philosophical ontology, and a symposium following an ecstatic victory celebration would hardly be the appropriate occasion to discuss them. She is satisfied with drawing attention to the ‘pluralistic’, intermediate status of eros, which consists of eros in the best sense being able on each rung to see two steps of the erotic episteme, and being able to mediate between them each time. Inappropriate, and perhaps ‘earthly’, eros is then to be recognised through its inability to perceive its own status as mediator, by firmly remaining on just one rung and not wanting to understand this rung as a connection and step up to the next one.57 56 Similarly Patterson has recently noted that according to Diotima’s speech on each rung the lover is inspired by the object of love to produce a logos. This logos reveals what is so wonderful and beautiful about the respective object of love, but at the same time states that what the lover found so attractive is in fact not exactly what it had seemed to be. According to Patterson this means more precisely: the logos points out that the lover was originally attracted by object X, but that X is only an image of Y, that the lover finds X attractive as an image of Y, and so it is in fact Y that he really finds so attractive (see Patterson (1991). Lawrence’s (1991) critical comment is not convincing; he doubts that there is a connection between eros, theory and philosophy in Plato, and sees here instead a systematic tension). 57 ‘And this is the way, the only way, he must approach or be led toward, the sanctuary of Love. Starting from individual beauties, the quest for the universal beauty must find him ever mounting the heavenly ladder’ (Plato Symp. 211c); in contrast ‘what makes their case

184

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

However, Diotima certainly does not seem to be of the opinion that all humans are so well initiated in ‘the more elementary mysteries of love’ that they reach the highest rung of erotic practice and episteme. She is not even sure that Socrates can,58 and Socrates himself does not say of Diotima’s speech that it has led him to the top of the ladder, merely that she provided him with the highest possible motivation to climb up it as far as he can with the help of eros – and that he tries to pass on this motivation to other humans.59 It is significant that Diotima describes reaching the highest rung as coming as close as one can to attaining ‘perfection’ and ‘immortality’, and thus transcending the finite life of humans60 – a final step that even Socrates cannot accomplish, although of all the speakers in the Symposium, and of all the speakers in Plato’s dialogues, he seems to lead the best kind of philosophical life. Thus Diotima, Socrates and Plato assume that many people will get stuck on one of the lowest rungs of erotics, while acknowledging that they are all engaged in erotics. It is not the fact that many people do not reach the highest form of erotics that is cause for moral concern in this context, but that many humans do not recognise that the rung of the ladder of erotic practice on which they are is only part of an entire ladder, and that it would be worthwhile endeavouring to climb up to the next rung. This motivation, as well as the correct assessment of the status of their current erotic practice in the continuum of all erotic practices, is the most that many people can achieve. Plato’s approach to erotics can be put more succinctly when we take into account that what had until then been regarded as the highest form of erotic practice is in fact better characterised as the transcendence of eros, the overcoming and ending of the erotic attitude. At this perfect, almost divine, level the soul embarks on peaceful contemplation that knows no erotic striving, for eros has achieved its goal. But this rung certainly does not represent the condition humaine, and thus it cannot be assumed that Platonic erotics recommends to everyone reaching this rung as the ‘teleology’ of their ethical work – but only to those few who possess almost superhuman discipline and talent. However, this elitist viewpoint also has a more gentle face: it recognises all forms of erotic practice, which include at least the motivation to reach the next rung and an appropriate so hopeless is that, having neither beauty, nor goodness, nor intelligence, they are satisfied with what they are’ (204a). Ferrari’s (1992) claim that each step up to the next rung of the erotic ladder is characterised by a change of perspective from the aim to the product of erotic attitude misses an important epistemic point of erotic ascent. 58 See Plato Symp. 210a. 59 See Plato Symp. 212b. 60 See Plato Symp. 211b, 212a.

the epistemic eros

185

appreciation of one’s own status, and defines these forms as an achievable human teleology. The correct ‘erotic’ attitude on each rung is all that most humans can attain, and the ‘philosophical’ attitude is no more than an explicit description of this orientation – motivation and striving for wisdom combined with an awareness of one’s lack of knowledge, but not wisdom and knowledge themselves. What the erotic, philosophical attitude does not consist of is peaceful contemplation and unrealistic abstinence. As Diotima so impressively explains, it is a constant swaying and oscillation between the bodily and contingent on one side, and the abstract and universal on the other; it alternates between flourishing and fading, it is in a permanent state of practical and theoretical restlessness – for only god is in the proper sense wise.61 Diotima’s speech leads subtly to the realisation that praise of love, that is of the love of love, is inconsistent. Love is directed at an object that has not yet been realised, but it should at the same time comprehend its object. Thus only the object of love can be praised, and not love itself. What is more, the cosmic dimension claims, among other things, that each thing in the world that can strive for something at all, strives to become what it really, or by nature, is. In this sense everything that strives is ‘auto-erotic’. The true object of love is not that which lovers do not have and so want – but that which they are in need of and require in order to become what they really are or should be. Therefore the beautiful and the good as the object of love cannot be separated from that in which it is instantiated, and it must specify its true nature: the formulae ‘X is beautiful’ and ‘X is good’ in fact say that X is a beautiful or good thing Y with respect to its nature N. To love a good or beautiful structure X always means to love the realisation of X in an individual Y as an approximation to N. Thus individuals are always involved in Platonic love; and by loving our true self we can also love others – as that which in truth they are or can be. This is the context in which we can more generally even love the striving parts of the cosmos and the cosmos itself – in their respective auto-erotic structures. In this sense too Platonic eros is neither egocentric nor impersonal in the normal sense. However, as an ability to perceive erotic phenomena it has to be learned – the Platonic capacity of loving is a virtue with an epistemic dimension, not an obsessive or possessive emotion. The miracle is not the realm of the forms, but the 61 Plato Symp. 203b–204c. When eros is correctly defined we can see that ‘all humans love’ (and so also philosophise), that eros can reveal itself in various forms (205a–b; 208e–209c) and that only one form of the Socratic knowledge of a lack of knowledge is necessary for it (204a).

186

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

cosmos and its parts as an image of the forms. As an intermediate being eros sees the visible cosmos correctly – as this miracle.62 Drinking and its etiquette are a leitmotiv in the Symposium that is employed as a dramatic symbol for the order of the six speeches about eros, and the systematic structure of their content. At the beginning of the dialogue the symposiasts agree to follow the advice of the doctor Eryximachus and to drink only modestly, for the day before they had drunk far too much wine at a celebration of Agathon’s first victory in a tragic festival.63 In the closing scene describing the end of the gathering in the light of dawn, nearly all of the guests have either fallen asleep or left; only Agathon, Aristophanes and Socrates are still awake, discussing the problem of the relationship between comedy and tragedy, and are still drinking out of a large beaker, passing it to the right as etiquette demands, the same direction as the order of speaking. Alcibiades suddenly bursts noisily into this orderly procedure; he is already extremely drunk himself and demands that the others should all drink more wine. After his speech praising Socrates, which he insists should not be compared with the previous speeches about eros because it is the speech of a drunken man, further uninvited guests join them and all etiquette – in particular of drinking – breaks down.64 Of course, the provocative disruption of etiquette by the drunken Alcibiades and his noisy followers is a dramatic device indicating an interruption of the speeches on eros, and an attack on the scholarly definitions of the cosmic and personal power of eros that they propose – an attack that is started by a man who stands for a life of passionate relationships and creative political work. At first sight the setting seems to support Martha Nussbaum’s thesis that the two ways of life followed by Socrates and Alcibiades are incommensurable; after all, in his speech Alcibiades returns to a relationship of individual love,65 and this rejection of the philosophical content of Diotima’s speech is emphasised literarily by the jealous exchange between Alcibiades, Agathon and Socrates.66 But the relationship between Socrates and Alcibiades is more than just a conventional homoerotic relationship between two individuals; Nussbaum herself points out that Alcibiades’ speech shows how in philosophical discourse the roles of lover and beloved begin to be reversed to some extent, until 62 63 64 65 66

See Kosman’s (1976) excellent article. See Plato Symp. 176a–e. See Plato Symp. 223b–d. Plato Symp. 215a–222b. Plato Symp. 222c–e.

the epistemic eros

187

a complex situation arises in which both partners in homoerotic relationships are part lover, part beloved, and the goals of their endeavours in the relationship seem to converge. This convergence, which in the end means that Nussbaum’s thesis of incommensurability is implausible, is not just a question of Alcibiades the beloved becoming the lover, and Socrates the lover becoming the beloved, as Foucault emphasises. There is in fact another subtle sense in which their relationship remains asymmetrical: lover and beloved not only exchange roles, they also exchange domains. For whereas in a traditional homoerotic relationship the lover desires the individual body of the beloved, and the beloved values the lover’s orientation towards the good, virtue and practical reason, in this case Alcibiades, who seems to be the beloved, loves the individual Socrates, while Socrates, who seems to be the lover, loves the universal beauty in Alcibiades. But perhaps that is still not a sufficiently exact description of the situation; for in fact we should assume that both of them, Alcibiades and Socrates, love an individual human being together with his reference to virtue, truth and the universal – but each in a different way. Alcibiades loves Socrates mainly because of his uniqueness in regard to the truth; on the other hand, Socrates above all loves the universal beauty and the universal eros in Alcibiades as a succinct representative instantiation of these structures. It is Socrates’ uniqueness that is the basis of the individual relationship that Alcibiades has with him, and at the same time this uniqueness is the reason why what Socrates mainly sees in Alcibiades is not the personal but the universal, and why he ironically ridicules Alcibiades every time Alcibiades seeks to draw attention to Socrates’ specific personality with its unique charisma. But Socrates’ uniqueness as described by Alcibiades has other important aspects. Guided by Diotima, Socrates has been able to see the forms, but remarkably in spite of this he remains a seeker of truth.67 It is as if he only began along the path of earnest searching in the full sense once he had seen the forms – it is a path that may be accompanied by crises, and can stray from the straight and narrow, or even lead to a dead end, but a particular view of the forms can again and again provide the motivation to follow 67 When properly understood, the description of the vision of the forms in Symp. 210e–211e is itself sufficient to give those who hear or read this description (in this case Socrates) the ability to see the forms; for it is not possible to do more than provide such a description in order to stimulate somebody to a vision of the forms. Thus Socrates actually says ‘images . . . so godlike, so golden, so beautiful’ (216e–217a), that even Alcibiades has once seen (Symp. 217a; 222a); nevertheless just like Diotima (Symp. 210a) Socrates (Symp. 212b) does not describe himself as already having such wisdom either; rather he remains a thoroughly ‘erotic’ person.

188

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

it. A view of the forms that remains consistent with the state of epistemic searching, and does not yet lead to wisdom and the highest knowledge, would consist of the realisation that forms or optimal structures exist to which we seem to refer in mathematical, ethical, or scientific discussions; and this realisation would also include an understanding of what characterises forms qua forms. But this humble kind of view of the forms would not imply complete knowledge of the content of each individual form – the search for the knowledge of their content, and the correct way to attain it would still have to continue. This is the kind of view of the forms that Socrates must have enjoyed if he continues in his Silenuslike state of knowledge of his own lack of knowledge; for without this humble kind of vision of the forms we would not even know that we are ignorant. So Socrates continues his search, and this is exactly how he is like Alcibiades, who – inspired by Socrates – does not want to remain where he is either.68 This erotic striving is typical of both men, and it is this common ground that defines the convergence of their relationship in spite of all the differences. On the other hand, in his uniqueness Socrates stands beyond almost all other humans; for even in a situation in which Alcibiades, as the apparent beloved, categorically feels that all the usual conditions for decent submission are fulfilled,69 Socrates remains unresponsive. And when Socrates is gripped by philosophical thoughts, then he will stand somewhere all day long and forget everything that is going on around him.70 This is why Socrates is ‘absolutely unique; there’s no one like him’.71 Thus Plato does indeed emphasise the difference between Socrates’ and Alcibiades’ attitudes, but this difference would not seem to be sufficient proof for Nussbaum’s theory of their two incommensurable worlds. For their worlds can also be seen to converge – they have a similar erotic restlessness, and share the universally good and universal truth in the individually beautiful as their domain. And as for Socrates’ uniqueness that puts him beyond all other humans, the Socratic way of life is not so much a serious alternative that could be recommended to all ‘erotic’ persons, as a vanishing point that reveals the correct orientation to normal humans and their erotics. In this way it would seem that central messages of the speeches of Diotima and Alcibiades converge from two different 68 69 70 71

Plato Symp. 216a. Plato Symp. 218c. Plato Symp. 220c–d. Plato Symp. 221c.

the epistemic eros

189

starting-points. Every basic erotic attitude is worthy of recognition on every rung of the ladder, as long as it keeps the Socratic orientation in sight as a vanishing point, and thanks to an appreciation of its own deficiency strives for the higher; that is, it is prepared to accept whatever state that it has attained as a step up to the next erotic rung. However, this endeavour requires the acceptance of the theory of forms as a formal ontology, and so the capability of observing the domain of individual things which are structured by forms from various epistemological aspects, thus developing oneself epistemically as far as one’s education and ability allow. This, and not Socratic asceticism, abstinence and transcendence, remains the teleology of Platonic erotics; it can be recommended to all humans as a practice of the self, and its loss, or even the lack of it, is the prime cause for moral concern in Plato. He does not problematise not reaching the highest rung on the ladder of erotic states, but instead the loss of all appreciation of the ladder and of practical dissatisfaction with whatever state one has attained – in other words the loss of the erotic attitude that is fundamental to the ethical practice of the self. The main object of Plato’s moral concern about eros is – the loss of eros. 3 If we want to get to the heart of the Platonic problematisation of erotic behaviour, above all in pederastic relationships, then the Lysis and the Phaedrus may not be neglected. For the discussion in the Lysis and the first two speeches on eros in the Phaedrus illuminate Plato’s original approach to the problem of eros, while Socrates’ speech in praise of eros in the Phaedrus represents Plato’s final and most mature word on the phenomenon. The diagnosis of Plato’s moral concern about erotic behaviour that was developed above in the previous two sections as part of the interpretation of the Symposium must pass muster when applied to the Lysis and the Phaedrus. The dialogue Lysis deals with the discourse of the love-struck lover. While strolling beneath the city walls of Athens, Socrates meets Hippothales and several friends.72 As an expert in the art of love, Socrates soon realises that Hippothales is head over heels in love, and wants to hear more about it. But while Hippothales is busy being coy, his companion Ctesippus makes fun of Socrates’ wish: ‘And do you really, Socrates, set any value on what this fellow says?’ In fact Ctesippus is already thoroughly 72 Plato Lys. 203a–206c.

190

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

fed up with having to listen to the lover’s outpourings about his beloved, and for this reason thinks it necessary to point out that Hippothales not only enthuses endlessly about his darling Lysis, in the process he also talks only childish nonsense and drivel, not to mention the peculiar poems and songs he recites. Ctesippus is convinced that Hippothales is making a fool of himself. But even if this is a hint at the lover’s mania, his possession by an erotic demon, Socrates does not need long to explain that actually in spite of his confusion Hippothales is pursuing a simple and understandable strategy – all he is doing is praising everything about his beloved and being at his service in every way possible in the hope that in the end he will ‘conquer’. During the course of the dialogue Socrates criticises such talk and attitudes on the part of the person in love as being selfish and imprudent – selfish because their goal is not really the beloved’s good, but the lover’s own satisfaction and recognition, and imprudent because they make his beloved even more arrogant and unattainable. Socrates also points out that one of the lover’s most important aims must be to promote knowledge and understanding in his beloved in order to maximise his own enjoyment. But this describes the interconnection of love, the art of love and knowledge only superficially and incompletely, that is only as the ‘work’ (ergon) of love in the object that is loved. However, above all a series of additional important questions are formulated and discussed which all deal with the problem of how, what, why and with which goal we love, and in which way love and friendship differ. But all these questions remain unanswered in the end. The infamous central Socratic question is not even posed: what are love and friendship? Yet without an answer to this question the relationship between love (eros) and friendship (philia) cannot be explained either. Nevertheless, the catalogue of problems and the aporetic discussion in the Lysis clearly point the way to the content of the speeches in the Symposium; but most significant of all is that the theme of the dialogue oscillates between being in love, true love and fully developed friendship, and thus already suggests that these three forms of behaviour are different rungs or aspects of the ‘erotic’ attitude, and that all of them are to be included in eros.73 73 The question of how the Lysis treats the relationship between love and friendship has long been a topic of scholarly discussion. A typical debate took place at the beginning of the last century between Max Pohlenz and Hans von Arnim: while von Arnim propagated the thesis that friendship according to the Lysis is a relationship that has been completely purified of sex and eros, Pohlenz claimed that friendship in the Lysis develops from being in love and love, and always contains an element of desire and yearning (see von Arnim (1914) and (1916); Pohlenz (1914) and (1917). In his definitive interpretation of the Lysis Bolotin (1989) is right to side with Pohlenz (see esp. 201–26).

the epistemic eros

191

The first two speeches about eros in the Phaedrus strengthen, supplement and state this philosophical diagnosis of being in love more precisely.74 The subject matter of the dialogue as a whole is the ambitious project of a philosophically acceptable rhetoric. As an illustration and background for theoretical discussions that follow on the form, technique and aim of rhetoric, three speeches on the uses and disadvantages of eros are presented. The first is introduced by Phaedrus as a rhetorical exercise by the famous orator Lysias, who is glowingly admired by Phaedrus;75 it is based on a phenomenology of being in love that is much more insistent than in the first discussion in the Lysis, but otherwise points in the same direction. It diagnoses being in love as an unstable passion that changes its object quickly and contingently, and always seeks immediate pleasure. Those who are in love regard their services and favours to the beloved as no more than a means of achieving this aim; they are ill, barely of sound mind, incapable of controlling themselves and so soon break their word; most of their actions are carried out as if under compulsion, completely dominated by passion, and so they are unable to enter into earnest obligations. When they are in love, lovers say and do only what pleases their beloved, and are so obsequious that their beloved is incapable of having a high opinion of them and admiring them. And if they are successful, then they tend to publicise their happiness, much to the detriment of the beloved’s reputation, so behaving vainly and inconsiderately. But soon they grow jealous or suspicious, and attempt to restrict the beloved’s freedom of action and life as much as they can. Thus the youths must expect a number of serious disadvantages from lovers who are in love, as well as undignified behaviour, but hardly anything positive as a counterweight. This diagnosis explains the – only at first sight paradoxical – central thesis of Lysias’ speech: that youths should submit to someone who is not in love rather than someone who is, and should keep their distance from the latter. Socrates’ reaction to this speech is remarkable: he accepts that it is a fine piece of rhetoric, and admits that it makes a number of points that are acceptable.76 But he doubts whether it is correct,77 in spite of being of the opinion that there are better ways of defending the paradoxical thesis at 74 See Plato Phaedr. 230c–234c (first speech); 237a–241c (second speech). 75 Whether the speech really was delivered by Lysias is still hotly debated. Heitsch (1993) presents a summary of research to date, as well as reasons for assuming that the speech is genuine (see 77–80), but otherwise does not relate it to the Lysis. 76 Plato Phaedr. 235e. 77 Plato Phaedr. 234d–235a.

192

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

the heart of it.78 This complex interplay of right and wrong is then defined more closely in Socrates’ first speech about eros, to which Phaedrus quite naturally presses him after his grandiose comments, and which, as already mentioned, is intended to defend the same thesis as Lysias’ speech. Socrates’ speech is much more systematic than Lysias’; he starts with a fundamental concept of eros, and derives from this most of the problems that Lysias and Phaedrus have already seen to be connected with being in love: if eros is the irrational desire that completely overwhelms the striving for good in whoever is affected by it, and is only interested in physical stimulation and the immediate pleasure it leads to, then every lover must indeed harm his beloved in the name of this eros. For the lover can only maximise his immediate bodily enjoyment when the beloved is spiritually dependent, ignorant and rash, physically soft, weak and cowardly, socially isolated and economically reliant. And as far as the long-term perspectives of a relationship guided by eros of this kind are concerned, as the lover gets older he will, assuming that the relationship lasts, become increasingly repulsive physically to the beloved, and so will tend to be more and more jealous and suspicious; but if the relationship comes to an end, then the lover will regard his obligations as also being finished and will become unfaithful and break his word. Clearly all of the important points that Lysias mentions also appear in Socrates’ first speech, but by employing a more exact structure in his derivation Socrates demonstrates that they are only conditionally correct – only when a particular concept of eros is employed that itself is open to attack; this is the way in which correctness and falsehood are distributed in Lysias’ speech. What is also significant is that hardly any specific restriction to homoerotic relationships is made in Socrates’ speech, for almost all of the aspects that he lists could also be applied to the standard ancient marriage between an older man and a younger woman. But above all, the speech decodes the central reason for moral concern about eros that bothers Plato: the restriction to just that form of eros (whether in homoerotic or heterosexual form) which, according to Socrates’ explicit diagnosis, is assumed by Lysias. Whoever believes, as Lysias does, that as an irrational desire eros is only directed at physical stimulation and pleasure, and so is incompatible with striving for the good, will in Plato’s view behave in a way that gives every cause for manifold moral concern. In other words Plato’s moral problematisation of eros goes much further than the concern for the active manliness of the youth in his 78 Plato Phaedr. 235b–c.

the epistemic eros

193

role as an object of pleasure for an older lover that was emphasised by Foucault.79 This concern plays absolutely no role in the moral problematisation of pederasty in the Lysis and the first two speeches about eros in the Phaedrus. Socrates is not concerned about the activity or the manliness of the youth in a pederastic relationship, but with his up-bringing, his virtue and his education; it is not so much a question of the effects on the youth of active, or even penetrative, erotics on the part of the pederast, but of the pederast’s disturbingly limited and distorted understanding of eros and erotics. On the other hand the pederastic lover should not, as Foucault claims, exercise self-control in order to present himself as a self-controlled person leading a glorious life, and so to recommend himself for a political career; instead the pederastic lover should have altruistic reasons for his self-control – in order to promote his beloved’s true self-interests and his genuine well-being. It is only from this perspective that the pederast can be persuaded at all to reflect seriously on his erotic behaviour and to subject himself to philosophical criticism. This attitude is presented in Lysis in a few succinct steps:80 – If A really loves B, or is his true friend, then A must be interested in B’s true happiness. – B can only enjoy and profit from an activity H if he has an understanding of H. – In respect of this activity H, B then only partakes in true happiness if he enjoys and profits from H. – In respect of H, B only partakes in true happiness when he has an understanding of H. And so: – If A really loves B, or is his true friend, then A must – rein in B in respect of all activities of which B has no understanding; – be submissive to B in respect of all activities of which B has an understanding; – ensure that B has an understanding of as many activities as possible. But perhaps of more importance is the clear-cut outline of theoretical problems involving love that makes up the main part of the Lysis. This 79 Heitsch (1993), 84–7 also stresses that Socrates assumes certain basic principles in his first speech, but otherwise makes no exact comparison between it and Lysias’ speech, and essentially restricts himself to a paraphrase of it. 80 Plato Lys. 206c–210d.

194

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

early dialogue is ‘aporetic’ – it states questions, but presents no answers. Three complexes of problems can be distinguished. First: how does love (eros) behave with respect to friendship (philia)?81 A number of questions arise here. For example, is the beloved a friend to the lover? Is the lover a friend to the beloved? Or are they both friends to each other, although only one loves the other? These questions introduce important structural problems, problems of the asymmetry and reciprocity of eros, as well as of the sublimation of eros to friendship. A second complex of problems is addressed by the question:82 for which characteristics F and G of persons A and B respectively does a relationship of friendship establish itself between A and B? Several possibilities are discussed: when F and G are similar or identical? When F is the good and G is the bad? When F and G are opposites? Or when F and G are the good? No conclusion is reached, except that a tendency to claim that G is good and F is a deficit is apparent. If A is an intermediary being, that is neither good nor bad but knows this, and when B exemplifies the good, then this may describe a basis for the friendship between A and B. What is under discussion here is apparently the explanation of friendship, that is the rationality of eros and philia. Finally, the causes and aims of friendship can be problematised:83 for example, is desire on the part of a deficient person an important cause for the formation of a friendship? Is becoming good an aim of friendship? But is there not a hierarchy of goods? And this touches on the problem of the ‘work’ (ergon) of eros and philia. In this context the discussion in the Lysis leads to a carefully contrived thesis: if B is good for A in the sense that B is the nature of A, and if A does not possess B but is not in a state that excludes attaining B; and if, furthermore, A knows this and desires B; and finally, if B as the supreme good for A is also beneficial – then A is a friend of B and of anybody who comes close to exemplifying B. In this case the lover is good, while the beloved has the chance of becoming good (and knows it); the beloved desires to become good, and the lover’s aim is that his beloved does so; what the beloved loves in the lover is that he is good, and what the lover 81 Plato Lys. 210c–213d. On closer inspection this passage examines the implications of nonreciprocal love and non-reciprocal friendship. Does one-sided love imply reciprocal friendship, or one-sided friendship reciprocal love? Does one-sided love imply one-sided friendship, or vice versa? Can there be one-sided friendship without one-sided love in return? Plato deals with non-reciprocal love (as seems to be required in a pederastic relationship) against the complex background that love is often one-sided, or should be; that love is a kind of friendship; and that friendship is normally reciprocal. 82 Plato Lys. 213d–218c. 83 Plato Lys. 218cff.

the epistemic eros

195

loves in the beloved is that he has the opportunity to become good and wishes to do so; and just as the beloved does everything in his power to become good, so too the lover does everything in his power to ensure this. For what the beloved loves in the lover because he exemplifies it, and what the lover loves in the beloved because it is an opportunity that can be realised and is an intended goal, are identical. In this contrived sense love and friendship are neither clearly reciprocal nor clearly symmetrical, neither clearly egocentric nor clearly altruistic. This is, among other things, because love and friendship as exemplified by the lover and the beloved are committed to a common goal, but at the same time assume different forms which correspond to the basic asymmetry of the pederastic relationship. In the early philosophical writings of Plato these complex problems in relationships already lead to moral concern about pederasty. It should by now be clear that the Symposium deals with all three complexes of ‘erotic’ problems, and answers the most important questions raised by them – if in a subtle and complex manner.84 These questions arise if, on the one hand, we follow the phenomenology of being in love (which at times is called ‘earthly eros’ in the Symposium) developed at the beginning of the Lysis and the Phaedrus, but at the same time follow Plato’s Socrates in seeing the identification of eros with being in love as the prime reason for moral concern. For then it is important to clarify to what extent this identification is inadmissible, and how ‘true’ eros differs from being in love. Clearly it is this clarification that is served by the three complexes of problems that are presented in the Lysis and dealt with in the Symposium. The relationship between being in love and friendship (that at times is called the ‘divine eros’ in the Symposium) is one of the matters that is of great importance. But above all the Symposium teaches us that it is not being in love itself that is cause for moral concern, but rather remaining on this rung, identifying this rung with eros and the lack of any endeavour to transform and overcome, or to integrate being in love in an appropriate manner into the broad spectrum of erotic attitudes. Clearly the outline of the problems and the questions in the Lysis shows the way towards this diagnosis and so also confirms it. 84 There is a literary reference to this at the beginning of the Symposium describing how Socrates reflects on his forthcoming visit to Agathon – he wants to arrive as the beautiful at the beautiful, as the good at the good, although it is also admissible to come as the bad to the better. This disperses Aristodemus’ worries about appearing uninvited as a simple person at the house of an artistic man. This is a blatant joke and subtle play on the themes involved in the Lysis (see Symp. 174a–c).

196

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

In the Phaedrus Socrates veils his head when he delivers his first speech on eros – out of shame at having to deliver in the style of Attic rhetoric a cynical defence of a ridiculous thesis that must brand eros as an irrational appetite for immediate enjoyment. He then removes the veil to deliver a serious philosophical speech in praise of eros that is the most important and revealing testimony to the orientation of Plato’s erotics that we have. Martha Nussbaum interprets this speech as a ‘recantation’ and ‘countercharm’ to the supposed intellectual and ascetic theory of eros in the Republic and the Symposium that had only accepted non-philosophical eros as a costly alternative:85 In the Phaedrus, however, philosophy itself is said to be a form of madness or mania, of possessed, not purely intellectual activity, in which intellect is guided to insight by personal love itself and by a complex passionengendered ferment of the entire personality. Certain sorts of madness are not only not incompatible with insight and stability, they are actually necessary for the highest sort of insight and the best kind of stability. Erotic relationships of long duration between particular individuals (who see each other as such) are argued to be fundamental to psychological development and an important component of the best human life.86

85 See Nussbaum (1986), ch. 7 (see, however, her qualification of this on p. 217). 86 Nussbaum (1986), 201. Thus only being in love, ‘earthly’ eros, is egocentric and is interested purely in sexual fulfilment, and so Plato’s distinction between being in love and true love refutes one of the central charges that Vlastos levels at his theory of eros (see n. 4 above). This is particularly true of the classic and highly influential study of the relationship between ancient and Christian views of love that Anders Nygren presented more than six decades ago (see Nygren (1993)). Nygren’s starting point is the fact that the New Testament refers almost solely to agape, rarely to eros (agape is seldom found in ancient literature as a word for a kind of love). According to Nygren this change in terminology indicates that a totally new idea of what love is enters the stage: whereas the ancient world, and in particular Plato, could never completely free eros of egoism, possessive desire and physical appetite, it was Christianity with its understanding of agape as loving one’s neighbour that arrived at a view of love that is both altruistic and non-intellectual, and based on unselfish emotions. It is only now that a form of love appears that also enables us to love God, and the Christian God to love us. The Symposium in particular is read by Nygren as the ultimate expression of a view of love as egocentric, possessive, intellectual and impersonal. Apart from Nussbaum, Rist (1964) has also argued against this interpretation, though his criticism is half-hearted (see esp. p. 26). Above all Osborne has criticised Nygren in a brilliant monograph that analyses texts about love from more than seventeen centuries (see Osborne (1994)). She concludes that a consistent view of love is to be found in all of the texts she analysed, from Plato through Aristotle, the New Testament, to Augustine and Thomas Aquinas: love is unexplainable in the sense that there is no rational reason why we love somebody; rather love is an attitude that opens a positive overall perspective on somebody and so initially puts particular characteristics of the person in an endearing light. But this does not mean that love is subjective, or based on deception – love is indeed a reaction to what a

the epistemic eros

197

This sketch of the philosophical theory of eros in the Phaedrus may indeed be accurate, yet it is neither a recantation nor a countercharm, but – at least assuming that the reading of the Lysis and the Symposium suggested above is correct – the development and enchantment of an erotic theory that Plato had always held, and is now presented in its final form by Socrates and Phaedrus under the most beautiful external conditions possible, [u]pon my word, a delightful resting place, with this tall spreading plane, and a lovely shade from the high branches of the agnos. Now that it is in full flower, it will make the place ever so fragrant. And what a lovely stream under that plane tree, and how cool to the feet! . . . And then too isn’t the freshness of the air most welcome and pleasant and the shrill summery music of the cicada choir! And as crowning delight the grass, thick enough on a gentle slope to rest your head on most comfortably.87

At the beginning of his weighty speech Socrates suggests that eros should be identified as a particular form of mania – that it is a kind of effusive enthusiasm and obsession.88 He attaches great importance to persuading Phaedrus that it would be an over-simplification to assume that mania is the opposite of controlled prudence, and for this reason is thoroughly bad. On the contrary, as prophetical, medical and poetic mania clearly show, the greatest goods are bestowed upon us through mania. Priests, doctors and poets who are level-headed but not possessed or inspired generally achieve little good.89 As Nussbaum rightly emphasises, this is doubtless a reminder that certain non-intellectual elements in us are required for our motivational energy, and are important for our aspiration to practical reason and knowledge; but for the philosophical interpretation and justification of this initial definition, which then goes on to link philosophical activity with erotic mania, Socrates deploys heavy philosophical weapons: the doctrine of the division of the soul, the localisation of the soul in the ontological hierarchy of entities and – an almost natural-scientific – proof of the immortality of the soul.90 This lays

87 88 89 90

person really is, or can be, but as an overall perspective it is often only love that lets us see just what is good and endearing about a person. This is why love is linked to an unselfish orientation towards the truly good and to an ambition to eliminate deficits. The physical desire that can be associated with love is not to be identified with it, but seen as one of its consequences. Plato Phaedr. 230b–c. Mania is generally translated as ‘madness’, e.g. Wahnsinn by Heitsch (1993) in his commentary on the Phaedrus ; but this is somewhat misleading. Plato Phaedr. 244a–245b. See Plato Phaedr. 245c–246b; 253c–d; 247c–248b.

198

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

the foundation for the mythical metaphor of the soul as a chariot and charioteer drawn by two horses: now it strives upwards to the forms, now downwards to the mutable things. But no human chariot can climb right up into the realm of the forms, for that would be a divine state; instead each is drawn by one horse that strives upwards and has to fight and struggle with the second horse. No human is a god, but each and every human struggles fiercely with his bad elements as he strives permanently towards the good. It was precisely this oscillating state, with its theoretical and practical restlessness, and its implicit tendency to climb up the ladder of spiritual attitudes, that was identified as eros in the Symposium. And it is exactly this point that Socrates formulates more precisely in his speech in the Phaedrus, leading to an important and interesting consequence: it is no longer the oscillating attitude between above and below in general, nor the unqualified endeavour of ascension and transformation that should be regarded as the state of eros, but the winged attitude and winged endeavour in the intermediate ontological and psychological realm.91 For the soul as a charioteer has wings, with plumage that is nourished by beauty and truth, but that fades and rots in the face of ugliness and badness.92 Thus the fourth and most important form of mania, inspired possession by the beautiful such as is also revealed by those in love, strengthens the plumage of the soul93 and so is an important requirement for the chariot of the soul to be able at least occasionally to look beyond the vault of the heavens into the realm of the forms and true reality – the highest achievement of the philosophical man.94 This physical inspiration is no longer divorced from sensuality. However, whoever has rotten plumage and can hardly remember his last ‘initiation’ into the realm of the forms, is slow to turn away (if he does so at all) from the physical enjoyment of pleasure to other forms of beauty; in other words he understands sensual beauty as no more than one among many forms of beauty, he has lost sight of the hierarchy of the beautiful, and so remains self-righteously in the domain of mere sensual pleasure, like an animal, in unbridled dissipation without shame or fear, heeding only earthly eros.95 But in all other cases it is primarily the sight of beautiful and restless humans that strengthens the soul’s plumage and sets an 91 Plato Phaedr. 252b; 249e. 92 See the moving (but also pseudo-scientifically phrased) description of ‘winged inspiration’ in Plato Phaedr. 251a–252a. 93 Plato Phaedr. 249d–e. 94 Plato Phaedr. 248a. 95 Plato Phaedr. 250e–251a.

the epistemic eros

199

erotic transformation in motion. Even an unphilosophical life, as long as it struggles for recognition, is inspired by eros and lets itself be guided to a stable, reciprocal friendship, in spite of the fact that from time to time lovers satisfy their sexual appetites in ecstatic frenzy.96 And even if lovers succeed in leading a thoroughly ‘philosophical’ life, and incompletely renouncing the orgiastic satisfaction of their sexual appetite, nevertheless they will still embrace and kiss passionately, will still lie tenderly beside each other and be prepared to enjoy sex.97 Without doubt there is a continuum of erotic behaviour and attitudes between two archetypal forms of the philosophical, and the non-philosophical but honourable eros; and in this broad spectrum ‘victory (is) won by the higher elements of mind guiding them into the ordered rule of the philosophical life’,98 although on occasions they may also perhaps fall victim to the attacks of the unruly horse of the soul. To sum up, on the basis of his fully developed philosophy of the mind and the soul Plato uses the image of the chariot of the soul and the charioteer’s plumage in the Phaedrus to provide a more precise identification of divine eros: as the inspiration of erotic attitudes in their attempts to achieve a self-transformation within their own realm that lies between mutability and eternity. This modification leads to an even closer interdependence of knowledge, eros and sensuality than was developed in the Symposium. No other text in Plato’s dialogues goes as far as the weighty speech in praise of eros in the Phaedrus in attaching physical feelings of pleasure to the erotic attitude of philosophy; but without giving up the more general position of the Symposium which had asserted this standpoint within the framework of the speeches by the poets and in the figure of Alcibiades, and integrated it into the philosophical erotics of Diotima’s speech. According to the Phaedrus physical pleasure, sensual awareness of the beautiful, tender reciprocal erotic relationships and personal friendships are not only necessary preliminary stages, they are also unrenounceable elements of philosophical eros, and so of all philosophical knowledge.99 It is this 96 97 98 99

Plato Phaedr. 256c–d. Plato Phaedr. 255e–256b. Plato Phaedr. 256a. Understandably most of the commentary by Heitsch (1993), 90–121 on Socrates’ second speech about eros deals with the details of the philosophical psychology and its cosmological background that are presented in the speech. However, his short notes on the exact formulation of the erotic phenomena, and in particular on the relationship between physical and epistemic eros, are ambivalent (see 120 f.): he assumes an ‘internal discord’ in the soul that is expressed in the fact that the instrumentalisation of love for memory and knowledge accompany its readiness to despise the rules of decency and good behaviour.

200

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

perspective which gives Plato’s problematisation of eros its sharpest and most precise edge. If pleasure and erotics are not only a preliminary stage, but also an intrinsic element of the philosophical life and of philosophical knowledge, then the false orientation of eros and an incorrect relationship between pleasure and eros leads to the collapse of the highest and best that humans can achieve. For this reason such false orientation and incorrect relationships must be the object of moral concern. Yet we are not freed from this concern by a model of life that attempts to employ widespread prudence and self-control to keep the intellectual lid on pleasure, sex and tenderness as best it can, and to make them vanish; according to the Phaedrus this model of life would in fact hamper the philosophical life and philosophical knowledge, and destroy them. The Phaedrus is further away than any other of Plato’s texts from the purely restrictive model of eros that remains functionally orientated to the glory of life and political ambitions, and which Foucault proclaims was the case for ancient ethics. And when Plato demands moderation on the part of the pederastic lover, then he is not concerned about the precarious position of the boy in the framework of a social and political context that reduces male sexuality to no more than penetrating activeness and domination; what is at stake is the possibility and the perspective of a common, inspired philosophical life that must be based on an enthusiastic erotic attitude supported by sensuality as well as reciprocal friendship and recognition. 4 Several aspects of the reading of Plato’s erotics developed in the previous sections can indeed be found in Foucault’s reflections on ‘true love’ in Plato.100 In particular he praises the role that Plato attributes to physical love in philosophical improvement and the philosophical life in both the Symposium and the Phaedrus: ‘For Plato it is not the exclusion of the body According to Heitsch the image of the chariot of the soul allows a ‘direct but decent’ illustration of all the powers set free in eros that are gathered together in the ‘common striving for perfection by those in love’; but no clear interpretation seems to be attached to this formulation. Unfortunately Heitsch pays no attention to either Nussbaum or Foucault, whose interpretations of Plato’s doctrine of eros he does not seem to feel are serious, and so are not even mentioned in his bibliography. 100 The perspective of the ‘erotics of narrativity’, that was suggested by Halperin (1992) as a post-modern reading of the Symposium that follows Foucault, and interprets the Symposium as a narrative text, does not seem to me to be particularly fruitful philosophically. Indeed I see good reasons for thinking that Plato in fact wants to criticise ‘narrative’ philosophy; see on this Bowery (1995).

the epistemic eros

201

that characterizes true love in a fundamental way; it is rather that, beyond the appearance of the object, love is a relation to truth.’101 The significant shifts in the question of erotics that Foucault identifies in Plato are clearly direct consequences of this fundamental relation to truth. For when it is to be a question of the truth of eros, then it is understandable why – the question of behaviour in love changes into the question of the nature of love; – the question of the honour of the boy changes into to the question of the love of truth; – the question of the asymmetry of erotic relationships changes into the question of the convergence of love; – the question of the virtue of the beloved changes into the question of love for the master and his truth. Unfortunately Foucault does not go as far as producing a precise identification of the moral problematisation of sexual and erotic activities that is specific to Plato’s approach. Above all his talk of the ‘relation to truth’ of Platonic eros remains strangely undefined, and this in spite of the fact that Plato endeavours to establish an extremely close relationship between ‘erotic’ stance and our epistemic view. Exactly which game of truth is interconnected with Platonic eros? Which ‘erotic methodology’ guides the lovers to higher rungs of philosophical knowledge? These remain open questions in Foucault – yet precise and interesting answers can be provided to them. To do so, however, it is necessary to arrive at an understanding of what is at the heart of Plato’s epistemology (which, of course, can only be done in a few rough steps here). In doing so we shall see that there is an exact analogy to Plato’s rejection of the simplistic two-world theory of eros in his epistemology. It is only from this perspective that we can fully appreciate the significance of Platonic erotics – not least because it is only 101 UP 239. This assessment follows, among others, the line of argument that Kruger ¨ (1948) so forcefully presented in what is still an important book: Plato’s philosophy has little to do with the rationalistic, abstract and enlightened view of the world of modern times. Rather on the one hand it is derived from the wishes and passions of the concrete life of pleasure and happiness, and on the other hand it is inspired in its highest goals by a religious ardour that is mainly concerned with attempts at a precise identification of the relationship between finite beings and the eternal and divine. However, just as Foucault overplayed aesthetic aspects, so too Kruger ¨ puts too much emphasis on the mythical and demonic: in Plato the erotic attitude is more closely related to rational epistemology and reasoned games of truth that are accessible for finite beings than Kruger ¨ and Foucault are prepared to admit from their respective viewpoints.

202

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

this viewpoint that allows a more exact identification of the relationship between his erotics and the sphere of power.102 It is a widely held opinion that not only in his philosophical erotics, but also in his ontology and epistemology, Plato supported a two-world doctrine. A typical account of this sums up the doctrine as follows: The so-called two-world doctrine is at the heart of Plato’s thought. It differentiates between a spatio-temporal world of becoming, and a world of being that is beyond time and space. The world of being is accessible only to thought, and represents immutability, idealness and normality; the world of becoming, which is subject to perception, presents itself as a perishable image of the eternal structures beyond time and space.103

This means that Plato’s two-world theory is taken to imply two fundamental distinctions that stand in a precise relationship to each other: a distinction between two kinds of entities – immutable, eternal forms and mutable, spatio-temporal objects – and a distinction between two forms of cognition – knowledge and perception. The forms and eternal structures are the sole domain of knowledge; knowledge is conceived in terms of the 102 ‘Post-modern’ thought is frequently committed to the ‘deconstruction of metaphysics’ – although without explaining what exactly is understood as the ‘metaphysics’ of classical authors. One of the most chaotic examples of this as far as Plato is concerned is Derrida’s attack on Plato’s supposed metaphysics of presence, which is based on a superficial use of Heidegger’s extremely problematic interpretation of Plato (see Derrida (1972)). In spite of Derrida it has become fashionable in post-modern thought to read Plato as an anti-metaphysical thinker, who had no systematic philosophy but who propounded a practice of open dialogue that is far more rhetorical, literary, ironical and self-reflective than has been appreciated to date, and that, for example, draws attention to the contingencies and inherent problems that are involved in any model of mastering something or the disciplining of practices (for instance, this view colours most of the contributions to the collection of essays by Shankman (1994); one of the criticisms of Derrida’s deconstruction of Plato by Berger Jr (1994) is the accusation that Derrida confuses Plato’s anti-metaphysical philosophising with later metaphysical Platonism). However, this anti-metaphysical reading of the Platonic dialogues is extremely one-sided; in fact it should be quite clear that in particular a good understanding of an archaeological, genealogical and ethical analysis of Plato in the Foucauldian sense requires recourse to the systematical aspects of Platonic philosophy. For example, in his introduction to the volume cited above Shankman correctly stresses that in the Symposium Plato emphasises the ‘intermediate status’ of eros and philosophy between non-knowledge and knowledge, between imperfection and perfection in order to stress that philosophy is not doctrinaire, but includes open discussions, and that in the Sophistes Plato does not, as Derrida follows Heidegger in assuming, equate reality with the self-sufficient ‘presence’ of immutable being (that reveals no relation to that which is), but in fact criticises this equation in the form of eleatic philosophy. But from this it does not follow that Plato did not subscribe to any systematic philosophy, but only that he did not subscribe to a doctrinaire, or fundamentalist, or eleatic systematic philosophy. 103 See e.g. Graeser (1993), 129.

the epistemic eros

203

model of non-propositional ‘knowledge by acquaintance’, and is always true (‘vision of the forms’). Perception, on the other hand, is directed at only the spatio-temporal objects, and is unstable and unreliable. The final elements of the two-world doctrine are a further clear-cut distinction between soul and body, and the doctrine of the recollection of the vision of the forms in all processes of learning. However, the picture of the Platonic ontology and theory of knowledge provided by this reading is certainly too simplistic.104 For example, it makes no useful contribution to a description of the ‘game of truth’, that is the methodology that is to lead to knowledge of the forms. Yet Plato developed an interesting and complex theory about the game of truth that is organised around knowledge of the forms, and which on closer inspection refutes the reading of the two-world doctrine. In order to justify this claim, it is first necessary to take a brief look at Plato’s early theory of the forms, above all as outlined in the so-called arete dialogues (Laches, Euthypro, Charmides). The following theses provide a systematic summary of this doctrine:105 F1

F2

F3

F4

F5

(i) It is possible to talk about universals. (ii) If F is a universal, then F is distinct from, i.e. not identical with, any individual things that are F. (i) If F is a universal, then it is the same in everything that instantiates F. (ii) If F is a universal, then it can be identified by a definition. (i) If F is a universal, then F is always identical with itself (in the strict sense, i.e. it never changes in any respect). (ii) If F is a universal, then F always has the same finite number of properties G1 . . . Gn , (G is a property of the universal F if every F is also G). (i) An entity F is identical with itself in the strict sense if and only if F always has the same, finite number of properties G1 . . . Gn . (ii) An entity F is a form if and only if F is identical with itself in the strict sense. (i) If F is a form, and G1 . . . Gn are properties that define it, then the meaning of Gi is always the same. (ii) If a form is to be defined, then the meaning of the universals that define it must be kept constant.

104 One of the cleverest and fiercest attacks on the traditional interpretation is presented by Ebert (1974) – a book that has received far too little attention in recent work on Plato. What follows is based to a great extent on Ebert’s analyses. 105 See on this Detel (1973) and (1974).

204 F6 F7

F8

F9

F10 F11

F12

f o u c au l t a n d c l a s s i c a l a n t i q u i t y (i) If F is a form, and G1 . . . Gn are properties that define it, then Gi are also forms. (i) If S is a science or knowledge of F, then F exists always, and not just in a limited space or time. (ii) There is a special science or knowledge about moral things in the sense of (i). (i) If S is a non-philosophical science, then S is not the object of S. (ii) To know F (in the scientific sense) is not only to be acquainted or familiar with F, but also to know the definition of F. (i) If S is scientific philosophy, then science or knowledge is an object of S. (ii) Philosophical knowledge is formal knowledge (cf. F7–F9 (i)), and implies the theory of forms (cf. F1–F6). (i) In using a general terminus ‘F’ we normally refer to a form F. (ii) A form F is the ‘cause’ of F’s being F. There are forms F1 . . . Fn , F, such that (i) each Fi is F, but not every F is Fi (for all i); (ii) F is the same in all Fi ; (iii) some Fi are logically contrary to each other. If we define a form F by F1 . . . Fn , then (i) the definition of the defining expressions cannot be infinite; (ii) we must content ourselves with the standard meanings of a few words.

This catalogue of theses is – although it may at first sight seem surprising – one of the main foundations for an account of a first element in the game of truth involved in the theory of forms. We must remember that in the early ‘aporetic’ dialogues all attempts to define particular forms fail miserably. Plato’s early doctrine does little to strengthen epistemological trust in the possibility of ultimate knowledge or ‘vision’ of particular forms. But the course of the discussions in the arete dialogues reveals that all the participants always assume the theses F1–F12, not only in their hypothetical suggestions for universal definitions, but even when they realise and accept that their own proposals are refuted. These theses describe not the definition of particular individual forms, but the characteristics which distinguish the forms qua forms and the knowledge of forms. The central thesis of this early Platonic doctrine can be put as follows: F (i) People generally believe that universal sentences (a) make good sense, (b) refer to something in the world in so far as they are true,

the epistemic eros

205

(c) can be discussed rationally and may then be confirmed or refuted. (ii) If a person believes that the claims (a)–(c) in (i) are valid for universal sentences, then P must also believe, and believe implicitly, that theses on the theory of forms F1–F12 are true. From the central thesis F it is easy to arrive at a trenchant interpretation of the infamous knowledge of his ignorance that Socrates is always aware of from the outset, and which in the early dialogues his fellow discussants arrive at only at the end of the discussion after Socrates has refuted their proposals. This ignorance refers to the identification of the content and the definition of universal forms; but an insight into it requires knowledge of F1–F12, knowledge that even Socrates regularly claims for himself in his refutations. The starting-point of the game of truth that is organised around knowledge of the forms is the insight that any rational, communicative discussion about the truth of universal sentences pre-supposes the acceptance of the early doctrine of forms in the guise of F1–F12. It is only on the basis of F1–F12 that it is possible to state more precisely what a ‘rational’ discussion in the sense of F(i)(c) is. At this point inductive techniques for which Socrates was extolled by Aristotle come into play:106 I If the sentence ‘all F’s are G’ is true, then it is the case that (i) every x that is an F is also a G; (ii) for every H of which it is the case that all H’s are F, it is also the case that all H’s are G. In the early dialogues we find Socrates constantly working with (i) and (ii) in I. Naturally certain principles of refutation result directly from I, and Socrates sometimes employs: W The sentence ‘all F’s are G’ is false if (i) an x can be found that is an F, but not a G; (ii) an H can be found such that all H’s are F, but some H’s are not G. But Socrates often also makes use of an artful schema of refutation: W* The thesis ‘F:=G’ (‘G defines F’) implies that all F’s are G, and all G’s are F; and this definition is incorrect if an H can be found such that all H’s are F, but some H’s are not G. 106 See Ar. Met. A3.

206

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

So we can now say: apart from the assumption F, the game of truth that is part of the theory of forms also involves at least the rules I, W and W*. Without doubt this is true not only for Plato’s early dialogues, but for all of them. The rules I, W and W* form part of the basis of a rational discussion of universal sentences – and that is still the case today. At this point the systematic connection between this elementary game of truth and the problematic of the two-world doctrine with respect to epistemology and erotics may not yet be sufficiently clear. However, an illuminative passage in the Phaedrus can help us here. In the middle of the enthusiastic and mystical allegory of the chariot of the soul and the vision of true being, the description of the transmigration and incarnation of the souls and the law of adrasteia, we find a remarkably sober statement: seeing that man must needs understand the language of forms, passing from a plurality of perceptions to a unity gathered together by reasoning – and such understanding is a recollection of those things which our souls beheld aforetime as they journeyed with their god, looking down upon those things which we now suppose to be, and gazing up to that which truly is.107

This passage shows that the objects which we constantly examine for their properties by playing the truth game sketched in F, I, W and W* are spatio-temporal entities and are located in the epistemological domain of perception. And indeed, Socrates does constantly talk of fishermen, farmers, anglers, rods, beds and other perceptible objects. The game of truth involved in the theory of forms is based on perceptions, among other things, and so is interconnected with each of the two worlds which Plato supposedly distinguishes so clearly. In particular, according to the Phaedrus the mysterious recollection (anamnesis) of the transcendental vision of eternal being also relies on perception. But anyone who would like to attribute a two-world doctrine to Platonic epistemology will draw attention to those passages – above all in the Phaedo – in which Plato seems to dramatically downgrade the role of perception in the theory of knowledge: Then when is it that the soul attains to truth? When it tries to investigate anything with the help of the body, it is obviously led astray. – Quite so. – Is it not in the course of reflection, if at all, that it should get a clear view of the facts? – Yes.108

107 Plato Phaedr. 249b–c. 108 Plato Phaed. 65b–c; see also 66e–67c, 83c–d.

the epistemic eros

207

In the context of these passages philosophers are also required to divorce themselves as far as possible from their bodies. However, the argumentative role of such statements must be taken into consideration, and they must be more carefully analysed as regards their content. For example, in connection with Socrates’ statement quoted above reference is made to three different criteria for the evaluation of perceptions as part of a theory of knowledge: to their importance or unimportance, to their reliability or unreliability and to whether they promote or prevent knowledge of the forms. But only the third criterion is applied explicitly: for Socrates observes that perceptions bring about various bodily needs – in other words distraction and restlessness – which tend to prevent knowledge of the forms; but this is a temporary, social circumstance,109 rather than a permanent epistemic state. Reference is made to the other two criteria elsewhere in the Phaedo. The claim that, according to the first criterion, perceptions are of little epistemological importance, comes down to the thesis that, as a result of excessive influence on the part of the body, the soul ‘has been so beguiled by the body and its passions and pleasures that nothing seems real to it but those physical things’.110 It is not perception in a general sense that is supposed to pose an epistemological problem here, but a particular use of perception that fails to recognise the physical as representative of structures, and does not even accept that there is a difference between perceptible things and structures, but merely regards only the physical as the real being. It is only a particular metatheoretical attitude towards the perceptible that is odious, not perception itself.111 The criterion of reliability is also applied less strictly than seems to be the case at first sight. Certainly Plato admits the possibility that the senses can be deceived, but the criterion of reliability must be seen primarily against the background of a concise definition of knowledge: knowledge of X is precise, or always true, when X is the case and X is always the case. Thus by definition the object of knowledge is an eternal form, and for this reason perceptions can never be precise or always true.112 But of course that is not to say that perception – when properly understood and employed – cannot play an important part in knowledge of

109 110 111 112

See Plato Phaed. 65a–67c, in particular the summary 66b ff. Plato Phaed. 81b. See also Plato Phaed. 83c–d; 73c; 74a–76a. See Plato Phaed. 78e–79c.

208

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

the forms. When we talk about structures, such as the identical, and their deficient instantiations in perceptible things as, then at the same time we are agreed also upon this point, that we have not and could not have acquired this notion of equality except by sight or touch or one of the other senses . . . So it must be through the senses that we obtained the notion that all sensible equals are striving after absolute equality but falling short of it.113

In fact perception is necessary for knowledge of the forms, and can contribute not only to knowledge of particular structures but also to insight into the distinction between perceptible things and structures – as long as we understand that perceptible things generally contain a pattern recognition, which in turn points the way to the non-perceptible forms. Thus what Plato claims in the Phaedo, the dialogue that seems to contain the most forceful epistemological criticism of perception can be stated more precisely: W1 Perception is imprecise and confusing, deceives us, and leads us away from truth inasmuch as it (i) produces physical needs, and so distracts us from the knowledge of forms; (ii) regards only the physical as real, and does not see it as an instance of perfect structures; (iii) is not knowledge in the sense that it is directed at immutable, eternal entities. W2 Perception is useful, even necessary for knowledge of the forms, inasmuch as when it is correctly applied and appreciated (i) the formal distinction between perceptible things and perfect forms becomes clear; (ii) particular forms cannot be defined without the inductive use of perceived facts; (iii) perception has a theoretical component, i.e. represents a pattern recognition. It is important to realise that W1 and W2 are compatible. Plato always believed in the conjunction of W1 and W2,114 and it should now be clear 113 See Plato Phaed. 75a; see also 103c ff. 114 This can be observed above all in the Theaetetus, that discusses and refutes the thesis that knowledge is perception. In particular the Theaetetus cleverly leads the concept of a perception that does not include a pattern recognition ad absurdum.

the epistemic eros

209

that they are the first indications of a close analogy between the philosophical assessment of perceptions in the theory of knowledge, and physical love in the art of love: just as in the erotics of the Symposium and Phaedrus it is not ascetic renunciation that is called for, but the correct employment of physical love; so too in the theory of knowledge it is not the renunciation of perception that is required but that it is employed correctly. In both cases correct employment consists of regarding physical phenomena as representations and starting-points for higher rungs of erotics or knowledge, as well as in using them as such; the incorrect or problematic employment of physical phenomena consists of missing, or not even perceiving, this possibility for transformation. As far as methodology is concerned, the philosophically appropriate application of perception is apparently characterised by the game of truth that consists of the rules F, I, W and W*. Plato also describes the correct application of perception as ‘recollection’ (anamnesis) of the vision of forms enjoyed by our souls before birth,115 and once again systematically this would seem to be embedded in a two-world doctrine. It certainly cannot be denied that Plato talks of a vision of pure forms by pure souls, and that he even uses anamnesis or pattern recognition in an attempt to prove that the soul is immortal, but there are good reasons for doubting whether he was prepared to acknowledge that the assumptions involved in such a mythical background had any philosophical relevance. The manner in which these assumptions are first introduced in the Meno are extremely revealing – as a doctrine for priests, saints and divine men who preach the immortality and transmigration of the soul.116 This, of course, is an allusion to the Orphists and Pythagoreans, but in logical terms the doctrine of anamnesis is derived from this religious doctrine. Thus the religious teachings about the transmigration of the soul are sufficient but not necessary for the doctrine of anamnesis – so possibly there should be independent philosophical reasons for it. Plato’s account of transmigration is certainly not without irony. This is indicated by a literary device, namely a dramatic change of style: the Greek text now consists of dignified prose that quotes venerable authorities, and contains rhythms, rhymes and short colas, in other words all the unmistakable hallmarks of Gorgias, whose rhetoric Meno admired but Plato constantly opposed.117 It is also significant that the priests describe the vision of the pure forms by pure 115 See e.g. Phaed. 74a–76a. 116 See Plato Meno 81a–e. 117 Gorgias is often mentioned in the Meno, see 70b, 71c, 73c, 76b, 78c, 79e; on the context see Ebert (1974).

210

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

souls in terms of a model of knowledge by acquaintance that Socrates had already so sharply criticised in the last part of the early dialogue the Charmides.118 What is more, in the Meno Socrates admits quite openly that for him the importance of the priests’ teachings lies in the fact that it makes us ‘active and inquisitive’ about knowledge of the forms – but that otherwise he is not sure at all whether the doctrine can be justified philosophically.119 Thus Plato has doubts about just how convincing the doctrine of the transmigration of the soul and its pure vision of the pure forms is. In fact, the philosophical plausibility of anamnesis is demonstrated in the famous geometry lesson which Socrates gives to a slave with no previous mathematical education. The question that is to be answered is: given a square Q, how long is the side of a square with twice the area of Q? The demonstration of the slave’s gradual ‘recollection’ of the correct answer has often been felt to be ridiculous because Socrates appears to put the answers into the slave’s mouth with a series of leading questions. This may indeed be the case, but it is of absolutely no philosophical relevance whatsoever. It is only from its context within the dialogue that the significance of the geometry lesson can be understood. Meno has just failed in his attempts to define virtue (arete) and as a pupil of the ambitious Sophist Gorgias he regards this as a personal intellectual defeat. It is at this point that Socrates introduces the geometry lesson and the doctrine of anamnesis in order to remind Meno how important an insight into one’s ignorance is for the transition from belief (the ignorance of ignorance) to knowledge. So in the course of the discussion in the geometry lesson the most important steps are then also explicitly described as the result of epistemic states: from the ignorance of ignorance via the knowledge of ignorance, to knowledge.120 Thus from a philosophical perspective the process of the recollection of the forms is a transition from belief to knowledge, and this transition is the real problem of the theory of knowledge that the doctrine of anamnesis is meant to help solve. There is a detail in the geometry lesson in the Meno that is particularly instructive for the theory of knowledge. At the beginning of it Socrates defines a square as a rhombus with diagonals of equal length and draws a square with diagonals in the sand. The slave is meant to solve the geometrical problem that Socrates sets him by looking at the square, in other 118 See Plato Charm. 167a–171e. 119 See Plato Meno 81e, 86b–c; also the comment on the function of this myth in Plato Phaed. 114d–e. 120 See Plato Meno 82b–e, 83a–84a, 84d–85b.

the epistemic eros

211

words by employing perception. But it soon becomes obvious that they are not talking about the square that has been drawn, but about ‘the square as such’, that is the form of the square.121 But above all the slave ‘sees’ the solution from the beginning, for he looks at the square and sees its diagonal which is, of course, the side of the square with twice the area of the original square. But initially he ‘sees’ the diagonal not as the side of the square in the solution, but as the diagonal of the original square. The transition from belief to knowledge does not involve a radical change in the domain or the ‘worlds’, but rather consists of perceiving the same object in a new way, in other words a change in the pattern recognition of the same object. If we explain perception on the basis not of the model of direct seeing, but the ‘qua model’ or the propositional model instead, then we will always assume that when a person P perceives an X, then P perceives X as a Y, and can claim that X is a Y. As far as Plato is concerned we are now on the right track, for we are now employing perception in the philosophically appropriate manner. There are other indications to support this diagnosis in the Meno. For example, Socrates also describes the transition from belief to knowledge as follows: true beliefs have a dream-like quality in us, and only become knowledge after we have questioned and tested them in a variety of ways;122 furthermore, the beliefs in the soul come and go, and become knowledge only when they have been firmly bound to the soul (although they are just as useful for actions when they have the status of belief as when they have the status of knowledge).123 And it is assumed here that the corresponding beliefs and states of knowledge have the same objects; the way to Larissa, for example, can be the object of belief as well as of knowledge – depending on whether we know it from hearsay, or because we have seen it ourselves.124 It is always the same entities that are now the object of belief in us, now the object of knowledge; the transition from belief to knowledge, that is the process of anamnesis, is not so much a transition from one kind of object to another, as a change between epistemic states that deal with the same objects. This epistemological diagnosis of the way belief is transformed into knowledge presented in the Meno is by no means Plato’s last word on the problem. The appropriate epistemological assessment of perception that 121 Plato Meno 82b–c; note the formulation ‘a square is a figure like this’, i.e. the form of the square is a figure of the kind that is drawn here in the sand. 122 Plato Meno 85d. 123 Plato Meno 98a–c. 124 Plato Meno 96e–97c.

212

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

is at the heart of this transformation is discussed elsewhere, for example in the Phaedo where a particularly informative example is used to illuminate it.125 The example concerns the possibility of using the perception of things that are equal in order to get a glimpse of (to ‘recollect’) the form of the equal. We perceive that two things ‘a’ and ‘b’ are sometimes almost equally large, but never exactly so. Apparently this perception includes an idea of exact equality of size. We can thus distinguish two theses: (a) for any two individuals a and b it is always the case that: a > b or b > a; (b) a = b if and only if neither a > b nor b > a. Plato then claims in the Phaedo that (a) can be substantiated by propositional perception and pre-supposes (b), but that (b) cannot be the object of perceptions. A possible metaphorical description of this connection could be that we recall (b) by means of (a). But a more sober observation is that (b) is a logical construction made possible by a universal concept that we employ in order to illustrate a propositional perception of type (a). Given (b), it is also possible to say: (c) a > b if and only if neither a < b nor a = b; (d) a < b if and only if neither a > b nor a = b. In other words the relations and = are conceptually connected with each other, and this connection enables us to say that (b) is presupposed by (a); nevertheless, the relations < and > between individuals are the object of propositional perception, whereas this is not the case for the relation of equality. This can be put as follows: (e) when we seem to perceive that a = b, then it is the case that in fact a > b or b > a, and a − b or b − a is very small. The deficiency thesis (e) is also based on propositional perception, but presupposes both definitively and explicitly the logical construction (b). Thus perceptions can be employed propositionally in a simple unreflected manner; but in philosophical reflection they can serve as the basis for logical constructions, which themselves can be interpreted as descriptions of perfect structures. And when these descriptions are presented explicitly, then propositional perception can be sharpened and employed to obtain an insight into the differences and connections between forms and perceptible things. If we read the ‘recollection’ of ‘equality itself’ as 125 See Plato Phaed. 74a–c.

the epistemic eros

213

outlined in this key passage of the Phaedo in this way, then of course it does not follow that our soul must have seen the form of the equal before we were born, or before our earliest perceptions; it merely follows that we possess, a priori, an epistemological capability to see our propositional perception as the basis of all possible logical constructions, to perform these constructions, and finally to use them to sharpen our perceptions. Perhaps it also follows that the structures on which we endeavour to reflect with our logical constructions also exist in a certain sense. However, we surely cannot draw the conclusion from these considerations that our soul had a pre-natal existence, or is even immortal. In this connection it is interesting that in the Phaedo Socrates deduces the pre-natal existence and immortality of the soul from the philosophical reflections presented above, together with the thesis that the perceptible F-things ‘strive to be F’, that is have a mimetic relationship to the form F.126 Mimesis is most probably not the appropriate philosophical term to characterise the relationship between F-things and the form F – Plato himself already uses the term ‘participation’ (methexis) in the Phaedo; but historically the concept of mimesis in this context is a clear indication of a Pythagorean background. And there are in fact a number of indications that in the Phaedo – and only in the Phaedo – Plato portrays Socrates as a Pythagorean.127 This suggests that Plato feels the doctrine of the immortality and transmigration of the soul, including its specific two-world doctrine, does not follow from philosophical reflections on the appropriate philosophical employment of perceptions alone, but rather from religious premises that are generally held by priests, divine men and Pythagoreans. Paradoxically Plato establishes and develops his epistemological position on the correct philosophical approach to the realm of physical bodies in the very passage that is traditionally regarded as one of the most important pieces of evidence for his supposed two-world theory: in the simile of the divided line in the Republic.128 It is the middle of the three great philosophical similes presented at the end of the sixth and the beginning of the seventh book of the Republic. The idea is that the sections of a line illustrate a division into four domains of existence, and four epistemic capacities that correspond to them. The line and its sections on which the simile is based can be illustrated diagrammatically. 126 See Plato Phaed. 74e–75b. 127 On the detail see Ebert (1994). 128 See Plato Polit. vi, 509c–511e.

214

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

The simile of the line A forms B

↑ image

↓cause > intellection (original) (nous)

C

↑ image

↓ cause > understanding (original) (dianoia)

D

↑ image

↓ cause > belief, (original) propositional perception (doxa, aisthesis)

mathematical objects

perceptible objects mirror images and shadows E

Normally the simile of the line is taken as evidence for Plato’s two-world theory because at point C on the line there appears to be a sharp division between the domain of the intelligible and the domain of the visible, that is between physical bodies and structures, and between perception and thinking. But a closer look reveals that this interpretation is too simple. For example, the bottom section of the line, DE, has always been difficult to understand because it includes shadows and mirror images, but not all kinds of images of perceptible objects – artefacts such as paintings are not among them. However, natural images are more closely linked to their originals, follow the movements of their originals and are causal effects of their originals. These characteristics open up an extremely important epistemic possibility for Plato: they facilitate recognition of the originals through their images, for on the one hand the connection between them is sufficiently close, but on the other hand their differences are quite clear. But recognition of originals through images presupposes that the images are regarded as images of originals, and not just observed as perceptible phenomena. Furthermore, apparently the same epistemic faculty is applied to both shadows and mirror images on the one hand, and the originals on the other; this faculty is propositional perception, which can of course be ‘correctly’ or ‘incorrectly’ employed. Its ‘correct’ employment is related to the bottom two sections of the line; it recognises the images

the epistemic eros

215

as images of originals, and through indirect knowledge climbs to insight into the originals in section CD of the line. The same form of indirect knowledge also exists between the two central sections of the line, CD and BC. Here Plato uses the example of geometry which construes its definitions and proofs by using perceptible drawings, but they are about mathematical structures, not perceptible things. At this point Plato’s language becomes epistemologically clearer: Consider then again the way in which we are to make the division of the intelligible section . . . By the distinction that there is one section of it [i.e. B C] which the soul is compelled to investigate by treating as images the things imitated in the former division [i.e. C D] . . . For I think you are aware that students of geometry and reckoning and such subjects . . . make further use of the visible forms and talk about them . . . The very things which they mould and draw, which have shadows and images of themselves in water, these things they treat in their turn as only images, but what they really seek is to get sight of those realities which can be seen only by the mind.129

Finally, just as geometry makes use of perceptible drawings as pictures in order to attain knowledge of the originals in section BC of the line, so too ‘dialectic science’ or philosophy uses geometrical objects ‘as pictures’ in order to understand them as images of the forms in the top section of the line AB, and so climb to knowledge of the forms themselves.130 From an ontological perspective this triple indirect knowledge is possible because for any two adjacent sections of the line the objects in the lower section carry in themselves the structures of the objects in the upper sections, and so are ontologically dependent on their ‘existence’. Two further aspects of the simile of the divided line must also be emphasised. The simile does not so much stress clear-cut boundaries between domains of existence and epistemic faculties, as the relationships and connections between them that facilitate the path to knowledge and the possible climb to higher knowledge, perhaps even to the highest knowledge of all. But above all the line is less a simile than a visible illustration of the path to knowledge, and thus represents a subtle self-application, for we are to make use of the perceptible picture of the line in order to grasp the abstract theory of the path to knowledge.131 What is more, 129 Plato Polit. 510b, d, e. 130 Plato Polit. 511b–c. 131 This reading of the simile of the line is defended by Ebert (1974) with excellent arguments and further persuasive texts.

216

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

the illustration of the line itself is still an incomplete picture of the late Platonic ontology that is first hinted at in the simile of the sun in the Republic, and is only outlined fragmentarily in the later dialogues. This ontology makes a further distinction between the sections of the line, and conjoins this distinction with a dualism of principles that assumes a principle of the one with the dynamis of structuring, and a principle of the many with the dynamis of pluralisation.132 Everything that exists has a mixture of structure and plurality, although to varying degrees: in the hierarchy of that which exists, plurality increasingly gains the upper hand the further down the scale we go. The transition from structures to perceptible things is only one of the transitions to more plurality. The original–image relation of the illustration of the line, which is itself perceptible, is replaced in the exact ontology by the relations: x forms y, and x is the existence condition of y: if x belongs to a higher level of existence than y, and y is an image of x, and x the original of y (to describe the relation between x and y in the visual language of the line-‘simile’), then – in terms of the exact ontology – x forms (structures, orders) y, and the existence of x is a condition for the existence of y. An interesting aspect of this ontological dualism is that something exists exactly if it can be described propositionally or predicatively – and that neither of the two principles belongs to the domain of what exists nor can it be described propositionally or predicatively, but is ‘beyond that which exists’133 and can only become the object of the highest form of knowledge ‘like a blaze kindled by a leaping spark’.134 The following ontological hierarchy can be identified in Plato’s late dialogues and later doxographical writings on his ‘unwritten doctrine’: (1) the principle of the one; (2) various forms of the principle of unity (highest genera): identity, good, rest, limited; (3) forms in conceptual hierarchies; A B D

C E

F

G

132 See Gaiser (1963). 133 Plato Pol. 509b. 134 Plato Epist. vii 341d.

the epistemic eros

217

(4) numbers (mathematical objects without dimension): – natural numbers; – rational numbers; – irrational numbers; (5) geometrical objects (mathematical objects displaying dimensions): – zero-dimensional: points; – one-dimensional: lines; – two-dimensional: areas; – three-dimensional: stereometric objects; (6) – geometrically formed surfaces; – stereometric structures consisting of the four fundamental material elements; (7) – perceptible basic elements; – mixtures of the basic elements; – perceptible single things; (8) various forms of the principle of plurality (further highest genera) difference, bad, motion, unlimited, space; (9) principle of the many. The levels (3)–(7) comprise the domain of that which exists, and each of these levels is in itself ordered further hierarchically. This is a much more complex and comprehensive picture than that presented by the line with its four sections: the transition from (3) to (4) corresponds to point B on the line, the transition from (6) to (7) to point C, while point D is hidden in (7); the principles and all further distinctions in (3)–(7) are not to be found on the line. According to this richer ontology, the correct philosophical approach to the realm of physical bodies is distinguished and generalised into the correct philosophical approach to a complex ‘realm of existing things’, with the aim of climbing epistemically from any given ontological level to the next highest containing less plurality and more unity than the lower one. It is now possible to see how the ladder of erotic attitudes, or more precisely the propositional context of these attitudes as developed in Diotima’s speech in the Symposium, can be projected on to the more complex ontology of Plato’s late philosophy. Quite clearly the simile of the sun in book vi of the Republic arrives at a comprehension of the principle of unity (the ‘form of the good’) with extreme care and reservation; but above all the simile of the cave at the beginning of book vii can only be understood on the basis of the ‘connectionist’ interpretation of the line-‘simile’, which emphasises not the divisions between ontological and epistemological domains, but the dynamics

218

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

of the process of cognition. It is important to see that all three similes stress particular characteristics of seeing compared with other forms of perception, and that these characteristics represent sensory illustrations for thought: the necessity of light in the simile of the sun; the possibility of pattern recognition and indirect knowledge in the illustration of the line; the directionality of vision and the epistemic restriction to a directional field of vision in the simile of the cave. Normally this directional restriction is compensated for by moving the head and body, but the prisoners in the cave are bound in such a way that they do not have precisely this possibility of compensation. It is for this reason that they cannot recognise the shadows and images as just that – as they would be able to if they could turn their heads or move their bodies. And they lack other possibilities of compensating for the directionality of their vision, for example hearing and speaking. The point is not that the inhabitants of the cave only see shadows, but that they do not perceive the shadows as shadows and images. Conditions in the cave prevent the correct approach to the realm of physical bodies – or more generally – to the realm of images, and so they destroy truly philosophical erotics. Thus what was supposed to be unproblematic in the ‘simile’ of the line – that is recognising images as images – and was at the same time a precondition for climbing to a higher level of knowledge, is then problematised in the simile of the cave: Plato demonstrates that the possibility of indirect knowledge which was assumed in the ‘simile’ of the line is in fact dependent on special conditions which normally are not fulfilled. The simile of the cave operates one level lower down: it analyses conditions for the rise to ever higher knowledge illustrated by the line. One of the most important of these conditions is later described as the ‘conversion of the soul’135 – that is as the transition to the ability to see images as images. But this conversion is a process of education – and it is not without reason that the simile of the cave employs the antithesis ‘education – lack of education’. Converting the soul is an educational process that must assume a pedagogical form, and this is why Plato follows the simile of the cave with a comprehensive programme of philosophical education. Only now can we see the game of truth that is associated with philosophical erotics in Plato, and defines the ‘reference to truth’ of eros that Foucault so often conjures up. Put in methodological terms, in this game of truth inductions, logical conclusions, universalisations and mathematical abstractions with respect to structured, perceptible things play a 135 Plato Polit. vii, 518d.

the epistemic eros

219

decisive role; put philosophically, it is an epistemic dynamic that essentially depends on an ‘educated’, philosophical approach to physical bodies and images. The educated and suitable erotic approach to beautiful physical objects and other beautiful images is only a special case of this dynamic, even if it is of particular importance pedagogically, and can only unfold its full power in a sublimated pederastic relationship. But at a more general level this epistemic–erotic dynamic demands a restless spirit that oscillates between plurality and unity and is the essence of ‘demonic’ eros. Not surprisingly the political dimension is hardly apparent in the speeches about eros in the Symposium and Phaedrus, and when it is then it is half-hidden. It is therefore truly remarkable that nearly all of the speakers in the Symposium were famous for their involvement in politics. Of course this is truest of Alcibiades, but of Socrates too if in a different way, while Eryximachus and Phaedrus were also involved in the political disturbances of 415; Phaedrus was accused of blasphemy and had to go into exile, while in his comedies Aristophanes discussed nearly every current political topic more openly and concisely than almost any contemporary poet. Incidentally Aristophanes lampooned Agathon (whose plays are not extant) in the Thesmophoriazusae, and portrayed him as an effeminate, vain aesthete who was influenced by so-called late Attic dithyrambs, and made extensive use of grand language, verbiage and music in his tragedies. Aristotle also notes that Agathon had begun to turn the chorus’s songs into choral interludes.136 This is an indication of the depoliticisation of tragedy, a tendency that became stronger in the fourth century, for in classical tragedy the chorus’s songs reflected on moral, political and theological aspects of the plot. Finally, in the Phaedrus it is the orator Lysias who represents the political background, for the public speakers had enormous political influence in the Athenian democracy. Thus the choice of speakers in the Symposium and the Phaedrus would seem to be an indirect, literary reference to a political dimension of the ‘erotic’ orientation. There are also occasional hints at this aspect in the speeches. For example Phaedrus claims in the Symposium that no state can achieve greatness without eros;137 Pausanias notes that rulers do not like beautiful and good erotic relationships between their subjects, because rulers and tyrants regard virtue, insight and friendship among subjects as being dangerous.138 Even Eryximachus, although a doctor, makes an indirect reference to the art of politics when he points out 136 Ar. Poet. 18, 1456a30. 137 Plato Symp. 178d. 138 Plato Symp. 182c.

220

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

that erotic activity as a skill seeks to produce concord, order and harmony in all fields, and this applies a fortiori to the art of politics and the organisation of the polis;139 Aristophanes proposes the thesis that above all young men with homoerotic tendencies are ‘excellently suited to the affairs of state’;140 and even Agathon recounts that Zeus was instructed in the art of ruling over the gods and men by Eros, as were the other gods in their own particular skill.141 Much of Diotima’s speech is too abstract to concern itself with specific fields to which eros can be applied. But when she defines eros generally as love of the good, as striving to produce the good in the beautiful, then love for a good constitution (politeia) and striving to produce a good constitution are doubtless an important special case of erotic activity. In the passage in which Diotima goes on to discuss specific forms of eros,142 ‘the greatest and by far the most beautiful wisdom’ that is connected with eros and its desire to produce the good is indeed described as that ‘which is visible in the organisation of states and the household’.143 This is why in Athens Solon was honoured, ‘because he bore (good) laws’.144 The political sphere finally enters the stage in the Symposium personified by Alcibiades, just as he does not praise eros as an abstract power but in the individual form that eros has assumed in Socrates. Only at one point in his eulogy of Socrates as the personification of eros does he mention Socrates’ accusation that he, Alcibiades, neglects himself and concerns himself only with the affairs of the Athenians. But this accusation is serious enough for apparently it implies the assumption that ethical work is an important precondition for political activity. In other words, according to the Symposium the correct eros is just as important for the ruled in their political function as it is for the rulers, for in the political sphere its aim is a stable and well-ordered polis. The Phaedrus, on the other hand, concentrates solely on the level of individual couples of lovers and their interaction, and so contains no explicit reference to the ordered polis as the aim of ‘political’ eros. Nevertheless, the dialogue alludes to the paradoxical situation that it is public speakers like Lysias who have the most political influence, although they completely fail to recognise the essence of eros and appropriate care of the self. This paradox is closely linked to the traditional view of rhetoric, the criticism 139 140 141 142 143 144

Plato Symp. 187c. Plato Symp. 192a. Plato Symp. 197a–b. Plato Symp. 208b–209c. Plato Symp. 209a. Plato Symp. 209d.

the epistemic eros

221

and corrective of which is the central theme of the Phaedrus: according to this view rhetoric aims to win over the audience to the speakers’ intentions, and to successfully manipulate them emotionally, but without paying attention to standards of expertise, truth and truthfulness. Speakers who employ this kind of rhetoric are apparently not possessed by the right kind of ‘erotic mania’, for their goal is not that which is truly good for the polis and its citizens, but whatever flatters the masses and serves their own interests. This is why such speakers cannot discuss eros appropriately. It is the reference to this context that provides the link between eros and the political sphere in the Phaedrus. But that still does not provide a sufficiently exact definition of this link and the ‘erotic form of power’. When Socrates says that everyone must first take care of themselves before indulging in the affairs of the polis, then this can have a variety of meanings. Perhaps it merely means that only men who can exercise self-control will meet with the approval that they need for a successful political career (this is Foucault’s suggestion). But in fact Plato seems to assume that only men who have the right approach to their eros, one that aims at the production of the good, are suitable to assume political responsibility. It is not external approval or a successful career that are at stake here, but being truly suitable for the production of the truly good in the polis. An important indication of the connection between care of the self and political activity is, of course, Plato’s thesis that the well-ordered soul and the well-ordered polis have analogous structures. But this thesis must also be seen in the light of the aims of Plato’s political philosophy. He had a deep distrust of the democratic constitutions of the city-states, and this was directed mainly at the democratic mechanisms of control of the rulers by the people in the popular assembly and the courts. This external control was damned to failure in Plato’s eyes because ‘the people’ knew little about the matters that were under discussion and so often reacted irrationally or let themselves be influenced by skilful but unscrupulous speakers. Seen from this perspective Alcibiades’ appearance in the Symposium is also significant because the way he was treated by the Athenian demos was for Plato one of the classic examples for the irrational and unstable politics of the popular assembly, and so for the inappropriate control of the ruled by the rulers. It is only from this viewpoint that the specific political dimension that Plato’s erotics assume as care of the self is comprehensible. When external democratic control of the rulers fails, then all that remains is to erect a consistent anti-democratic politeia in which external control of the rulers from below must be replaced by the rulers’ self-control. This is exactly the path that Plato goes down in the

222

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

Republic; he combines the construction of a strictly anti-democratic class state with the most rigorous programmes of education and ethical work that had ever been proposed for political rulers. This up-bringing was intended to ensure the enormous degree of self-control that a political elite which was concerned for the truly good had to possess in order to be good and stable rulers. Thus in the context of Plato’s political philosophy self-control had to be the decisive element of political power, given the absence of any form of external control. ‘Political eros’ provided the motivation for the ethical work that led to the politically relevant forms of self-control. But at the same time a great degree of self-control was demanded not just of the rulers, but of the ruled too, for this was the only way of securing rational acceptance of their subordinate social position since they had no chance of influencing the rulers. Thus for Plato individual ethical work in which we care for ourselves was at the same time also political work in which we care for the polis, as well as epistemic work in that we care for the truth (especially the truth about humans and ourselves). The convergence of the archaeological, genealogical and ethical axes – to use the parlance of Foucault – is organised around the phenomenon of regulative power involved in those educational processes which not only demand ethical and political work145 but thus at the same time also lead to an epistemic game of truth, turning erotic phenomena into a philosophical field of knowledge. The political hierarchisation that is part of this picture is applied to heterosexual and homophile sexual relationships: the form of regulative power and of the game of truth through which, according to Plato, eros becomes an object of knowledge prevents the development of a concept of relational and reciprocal sexuality. In this way the regulative power that is interconnected with the Platonic game of truth contributed to the stabilisation of asymmetrical relations of power in pederastic, and above all in marital, relationships, and in its application to erotic phenomena assumed repressive features. 145 Wartenburg calls this form of power ‘transformative’ – correctly, for it aims at removing the power differential, and thus destroying itself. Wartenburg regards the Socratic method that Plato demonstrates itself as an exemplary expression of transformative power (see Wartenburg (1990), ch. 10).

6

G E N D E R , N AT U R E A N D R E F E R E N C E

Foucault’s analysis of ancient philosophical texts as well as the divergent interpretations that I have proposed in earlier chapters both make use of the distinction between the male and the female. But this distinction could refer to either biological sex or social gender. The relationship between the two, and a reference to biological sex differences, has long been the subject of heated controversy in feminist theory. This debate is not only of systematic interest, it also compels scholars who interpret ancient texts dealing with moral attitudes towards sexuality to reflect explicitly on the form of their historical reference to the male and the female. The notion of gender would appear to be dependent on and relative to historical and cultural contexts and is therefore often assumed to be socially constructed. This means, however, that in historical studies, for instance on ancient culture, this notion cannot be applied without further qualification, in particular where such studies are intended to contribute to an understanding of historical attitudes towards sexuality, and thus more generally towards gender. Post-modern thought as influenced by Foucault has seen an increasing tendency to refer solely to the concept of social gender, and to brand any reference to biological sex as naive and patriarchal. Influential feminist approaches have interpreted Foucault in a manner that threatens to deprive his historical studies on sexuality of an important, if trivial, foundation – the reference to the sexes beyond cultural boundaries and specifications. 223

224

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

In a recent exhaustive study of Foucault’s relationship to feminism, McNay has quite correctly drawn attention to systematic difficulties that Foucault’s doctrine of docile bodies has for feminist theory.1 Although Foucault demonstrated how relationships of power form bodies, he paid no attention whatsoever to the way in which the disciplined body is related to sex or gender. This ‘gender-blindness’ prevents him from making use of the theoretical potential of his genealogical method for a study of the social construction of gender. Above all, according to McNay Foucault’s doctrine of the docile body restores the ideology of the passivity of the body, in particular the female, which must bow in submission to the influence of power relationships: ‘Thus, Foucault’s understanding of individuals as docile bodies has the effect of pushing women back into a position of passivity and silence.’2 However, in Foucault’s later programme of ethics McNay sees the possibility of ascribing the ability to actively form and style the self not just to the ‘self’ in general, but above all to individuals of a particular gender. Although not even in the last two volumes of The History of Sexuality does he address the problem of gender formation explicitly, nevertheless ‘Foucault’s later idea of practices of the self implies an understanding of gender as an active and never completed process of engendering or enculturation’.3 But when McNay, following Reither, sees Foucault’s thesis of the unstable and off-centre nature of identity as the starting-point for a feminist criticism of a ‘politics of identity’ that relies on identity theory to root the concept of ‘woman’ in a stable heterosexual matrix, then it is unclear whether she is referring to biological sex, social gender or both.4 Foucault never explicitly turned his attention to these problems. But clearly historical studies of the moral problematisation of sexuality as discussed in the preceding four chapters of this book, and above all as conducted by Foucault himself, can among other things help us to see more clearly how genders were formed in a particular historical period. If such studies already employ a reference to the male and the female, then clearly 1 McNay (1992). A similar view is expressed by Deveaux (1994). On the present situation see also Sawicki (1994). For a comparison of the positions of McNay and Sawicki see Still (1994). Dean (1994) offers an explanation and criticism of Foucault’s ‘gender-blindness’, while Faith (1994) draws attention to Foucault’s analysis of the relationship of power and resistance, which she then uses as a basis for a description of processes of feminist resistance. 2 McNay (1992), 47. 3 McNay (1992), 71. 4 See McNay (1992), 112–15 on ‘Difference and Sexual Difference’. From now on I use the notion of sex for biological sexes, and the notion of gender for social or socially constructed gender.

g e n d e r , n at u r e a n d r e f e r e n c e

225

it cannot be the genders of the historical periods that are meant, for those are what are to be reconstructed in the first place; and it would be anachronistic to employ genders from other periods. Thus it must be differences of sexes to which Foucault’s analyses in volumes 2 and 3 of The History of Sexuality, as well as the interpretations that I laid out in the earlier chapters of this volume, refer. Yet as I have already pointed out, post-modern feminism attempts to deconstruct the concept of biological sex and thus to deny any possibility of a reference to differences of sexes. In this final chapter I wish to suggest a way in which this possibility can be secured, while at the same time doing justice to important feminist reservations about making use of the notion of biological sex. To do this I shall first of all (1) take a critical look at the main reasons for the destruction of the notion of sex and then (2) attempt within the framework of one of the leading current theories of meaning to develop a thin concept of sex; next I shall (3) defend this concept against essentialist objections raised by feminist theory and (4) outline its usefulness for important aspects of current feminist discussions.

1 As was shown above,5 the earliest reflections on the ‘nature’ of man and woman at the beginning of Western culture were not a particularly good start to conceptualising the relationship between women and men in the history of Western thought. ‘The man is hot and dry, the woman cold and moist’ is how the pre-Socratic and Hippocratic doctrine of the elements characterised the relationship; a revealing statement that was still influential in the Peripatos.6 It was the physiological constitution of women, their moisture and cold, that was the reason for their physical, psychological, moral and intellectual inferiority to men, and this was why it was the man’s job to limit and control women’s dissolution, excess and loss of control.7 Consequently, it is no surprise when women are not only linked to the cold and wet but also to the dark and bad, while men are associated with not just the dry and warm but with the light and the good.8 Thus from the beginning prominent protagonists of Western culture have attempted to fill physical differences between the sexes with social 5 See above pp. 145–6. 6 Hipp. De Victu i 27; Pseud.-Ar. PP iv 25, 879a31–5. See also the brilliant study by Carson (1990), as well as Hanson (1990) (esp. 317–19). 7 Dover (1974), 98–102 describes how these views are reflected in poetry. 8 See Lloyd (1983), 82.

226

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

content, or to apply normative conceptions that correspond to a patriarchally organised society directly to physical categories of the sexes; and this strategy is still successful today. Thus it is no surprise that twentieth-century feminist theorists have constantly been sceptical and reserved about the concept of biological differences between the sexes. The widespread acceptance of the distinction between sex and gender in the first wave of feminism9 was mainly a result of trying to propagate the claim that there was no causal connection between differences in the two, so that social sexual inequality could not be the result of differences of the sexes.10 However, it is possible to identify a controversy about difference and equality within this debate that has led to two contrasting positions: equality feminism reduces gender differences to mainly social and cultural aspects, and demands their removal; difference feminism maintains that there are fundamental, even essential differences between men and women; it suspects that the demands of equality feminism in fact disguise an adaptation to androcentric values, and requires instead accepting that specific female abilities, achievements and virtues be of equal relevance to male ones.11 In a second wave of feminism the emphasis of discussion shifted from the difference between the sexes to differences within the female sex. Klinger has quite rightly drawn attention to the fact that this change in the international feminist debate has two sources – the articulation of the experiences of women who are neither white, nor middle class nor heterosexual; and post-modern gender deconstruction.12 Disciples 9 Lerner (1986), 301 puts the matter succinctly: ‘sexual gender [sexuelles Geschlecht] is a biological reality for men and women. Gender specific role expectations [geschlechtsspezifische Rollenerwartungen] of men and women are a culturally related definition of behaviour that is regarded as appropriate for the sexes in a particular society at a particular time. This culturally specific definition of the role of the sexes is a historically dependent product.’ 10 See e.g. Maihofer (1995). Some critical notes on current attempts to arrive at a new understanding of ‘gender’ are to be found in Pu¨ hl (1994), 173. 11 See Klinger (1995), 801 f. Like Fraser, ‘Multiculturalism and Gender Equity: The US Difference Debates Revisited’, Typoscript, 2 (1996), Klinger talks of two ‘phases’ of feminism, Nicholson (1994) of two ‘waves’ (I personally prefer waves to phases). Klinger notes that equality feminism explains gender difference purely in social and cultural terms (801), but it seems to me that the equality position normally also accepts a certain degree of sexual difference which is then developed into full-scale essentialist differences (in the feminist sense) by difference feminism. 12 See Klinger (1995). Several authors support the view that most work in the second wave still implicitly pre-supposes the difference between sex and gender, in spite of its actual aims (see e.g. Nicholson (1994)). Klinger also follows Fraser in diagnosing an internal systematic tension between ‘post-colonial’ and ‘post-modern’ aspects of the new feminist position presented by the second wave.

g e n d e r , n at u r e a n d r e f e r e n c e

227

of this new position accuse the old viewpoint of essentialism, dualism and identity-centred thinking, since it is the result of an unjustifiable universalisation of the position of white, middle-class, heterosexual women; it defines in dualistic terms a monolithic difference between two sexes that is not otherwise distinguishable, and so is over-hasty in identifying gender categories without analysing their historical development and the way in which they are intrinsically linked to other categories. This new position, which shares with post-modern thought a tendency to social constructivism, has strengthened the movement to completely abandon the concept of sex or else to incorporate it into the notion of gender and to introduce it as no more than a social construct.13 This strategy was able to draw on a whole series of theoretical objections to the concept of sex that were already well established in feminist theory. For example, there is a widely held belief that any reference to the sexes supports an essentialist explication of gender categories that is quite unacceptable, because it is historically rigid and implies exclusion. This concern has been summed up by Young; the difficulty of the essentialist approach is that all attempts to define the essential attitudes of men and women can only have two possible consequences – either the reduction of the category ‘woman’ to purely biological attributes devoid of social meaning, or the determination of essential social attributes of women that are doomed to failure because of the multiplicity and diversity of life as actually experienced by women in different social contexts.14 However, a sufficiently thin definition of sex could perhaps avoid being branded as essentialist, and the fact that it would be of no social relevance might be a significant theoretical advantage. What Young disapproves of in the concept of sex may in fact be something that is theoretically promising. A further common theme in this context is the fear that maintaining the distinction between sex and gender reinforces some of the most important conceptual dualisms which for many authors (not just female) are one of the main expressions of patriarchal thought, and which the feministic

13 The works of Butler, Young, Nicholson, Maihofer, Honnegger and Laqueur come to mind. In Bodies that Matter, for example, Butler (1993), says that biological sex is not a simple fact or static condition of a body, but a process whereby regulatory norms materialize sex and achieve this materialisation through a forcible iteration of those norms. Thus Butler employs her performance theory in order to demonstrate the static content of the category of gender. 14 Young (1995), 119 ff.

228

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

perspective therefore seeks to attack or deconstruct.15 Perhaps some of the conceptual dualisms that have been so prominent throughout the history of Western thought do indeed pose problems, but there are very good philosophical reasons for the particular dualisms of body and mind, or of nature and culture, that form the basis of the distinction between sex and gender. Perhaps today hardly anyone would wish to defend ontological forms of dualism, but it seems hard to deny, for example, that the physicalistic description of neurophysiological states of the brain and the understanding of propositional attitudes, or the nomological explanations of events and the practical syllogisms of explanations of actions, each represent different approaches that are not mutually reducible. The reasons that have led authors from such different backgrounds as Kant, von Wright, Habermas or Davidson to adopt similar views on this matter are weighty indeed. Certainly there are at present a few extreme variants of schools of thought with a monistic alignment of some kind; thus the cognitive sciences would like to eliminate mentalistic vocabulary,16 while for social constructivists physicalistic vocabulary has no specific status.17 But the 15 See e.g. Maihofer (1995), 19–21 with further literature. 16 See for example the work of Lycan or Patricia and Paul Churchland; however, adherents of a strict monism in the cognitive sciences are in the minority: certainly some monists such as Searle and Roth do not support eliminative materialism, while others propose an emergentist materialism, for example Bunge and Ardila as well as Searle. Gardner (1992) offers a summary of the state of thought that is also readable and informative for nonexperts; see also Roth (1994); Searle (1996). 17 Social constructivism in the broadest sense assumes that (a) there are no clear differences between facts and the contents of our beliefs, and that (b) the way in which we arrive at beliefs should be explained in social and causal (i.e. scientific) terms (in the process there are various justifications for (a) – for example transcendental, objective-idealistic, historical-materialistic, in terms of the theory of science and the theory-ladenness of our language, or through the description of the ‘ways of world making’ as demonstrated by Nelson Goodman). On the whole this position is strengthened by the following two theses, that (c) the way in which we arrive at beliefs can only be explained in social and causal terms, and that (d) there are no non-contingent criteria for distinguishing between true and false or rational and irrational beliefs. Furthermore, it is often also claimed that (e) standards of reason together with beliefs are mostly historically mutable and that (f) the explanations of how we arrive at beliefs allow no clear distinction between natural and social facts. Clearly these theses belong to the sphere of the sociology of knowledge; one of the most widely discussed variants of this position today is the ‘strong programme in the sociology of knowledge’ that is propounded by Barnes and Bloor (see Barnes and Bloor (1982) with a detailed bibliography). My impression is that most post-modern variants of social constructivism accept theses (a)–(f), and add that (g) any serious attempt to define categories or beliefs is incompatible with (a)–(f), and that (h) such an attempt is pragmatically unavoidable, but at the same time a form of repressive power, and so should always be deconstructed. Because of (a), the social and causal generation of beliefs according

g e n d e r , n at u r e a n d r e f e r e n c e

229

arguments for these monistic approaches are far from convincing, and the prospect of being sucked down by the Scylla of cognitive science or seized by the claws of the Charybdis of social constructivism is not an attractive one – at least not without further convincing reasons that do not, however, seem to be available at present. Another important motive for wanting to replace the notion of sex with that of gender is the pertinent impression that statements and theories about biological sexes have changed considerably during the history of Western thought, and that these changes are probably closely linked to other theoretical and social shifts.18 For this reason we may be entitled to assume that, like other scientific theories, theories of biological sexuality are social constructs. And of course we cannot deny that the historical dynamics of scientific theory have been remarkable, or that, for example, biological theories involve suppositions, hypotheses and constructed claims just as much as social theories and alignments do. But it would be wrong to conclude from this that the domains of these different kinds of theories are also social constructs. Nothing forces us to draw this conclusion; there is no problem, and certainly no contradiction, in admitting that all human assumptions and theories are hypothetical constructs, while at the same time insisting that there is a distinction between the realm of nature and the realm of society, between natural laws and social rules or between physical facts and intentional actions. The rationality that is a constitutive element of actions, rules, roles or institutions has no counterpart in the realm of nature. However, the strongest and most influential motive for consistently rejecting the category of sex is probably the result of the impression that it is connected systematically in one way or another to biological determinism. Within the framework of patriarchal societies a reliance on the notion of sex can often be misused for the political and ideological justification of the social discrimination of women, even though a number of feminist theories have long proved that there can be no causal relationship between sex and gender. But even these theories still seem to involve hidden associations between sex and determinism. Nicholson has coined the phrase ‘biological foundationalism’ for these veiled remnants to (b) and (c) under conditions (d)–(f) is a social construction of facts and worlds; from the post-modern standpoint it is an expression of repressive power given (g) and (h) that should always be the subject of attempts at deconstruction, and above all because of (i) gives rise to radical relativism and scepticism. 18 Laqueur (1992) is often called as a witness for the prosecution on this point; see e.g. Maihofer (1995), 28–34; Nicholson (1994), 195–202.

230

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

of patriarchal thinking in feminist gender theory, and has quite correctly pointed out that in a concept of gender that implies these assumptions, physical distinguishing features ‘remain constant over long historical periods, and would be suitable for justifying and explaining cross-cultural similarities of personality and behaviour’ – in other words that such a concept of gender incorporates a weak biological foundationalism’ which ‘leads to the same cultural projections as biological determinism’. For this reason Nicholson also pleads for an alternative perspective that reconstructs the body and physical gender not as a constant but as a historically variable factor that is described in social terms.19 But significantly, Nicholson enriches the concept of sex just enough to be able to attack it successfully, for she understands the biological foundation of gender as part of the category of sex in the first place. Of course post-modern feminism is not in the least bothered by possible criticism of its objections to the concept of sex; on the contrary, it has intensified them dramatically under the flag of anti-essentialism and the influence of the Foucauldian analysis of power. It claims not only to be able to prove that naturalistic and scientific reference to sex and the heterosexual difference between the sexes are embedded in a constantly changing theoretical dynamic, but also to demonstrate that even the gendered body itself, including the difference between sex and gender, is a social construct or can at least be interpreted as a complex of established social and linguistic practices.20 However, it is difficult to say whether the leading proponents of postmodern feminism wish to defend a strict social-constructivist position that is related even to natural phenomena. Haraway’s work on the ‘construction of nature’ seems to be very closely related to social constructivism; 19 See Nicholson (1994), 189. 20 Butler could not have put it more clearly: I would argue that any effort to give universal or specific content to the category of women . . . will necessarily produce factionalization, and that ‘identities’ as a point of departure can never hold as the solidifying ground of a feminist political movement. Identity categories are never merely descriptive, but always normative, and as such exclusionary . . . if feminism presupposes that ‘women’ designates an undesignatable field of differences, one that cannot be totalized or summarized by a descriptive identity category, then the very term becomes a site of permanent openness and resignifiability. I would argue that the rifts among women over the content of this term ought to be safeguarded and prized . . . To deconstruct the subject of feminism is . . . to release the term into a future of multiple significations, to emancipate it from the maternal or racialist ontologies to which it has been restricted, and to give it play as a site where unanticipated meanings might come to bear (Butler (1992), 15–16). See also Butler (1990), ch. i.

g e n d e r , n at u r e a n d r e f e r e n c e

231

authors such as Flax and Butler do not discuss the way entities postulated by natural sciences exist, but at least as far as gender is concerned their approach is predominantly social-constructivist. For example, Butler explains that what is at stake in her thesis of the ‘construction of sex through a ritualised iteration of norms’ is understanding the material of bodies as the effect of a dynamic of power, that is of understanding the assumption of sex as a process within which the power of discourse forces the adoption of a cultural norm; and she also points out that inasmuch as sex is postulated as an entity that precedes gender and the social realm, this postulate is subject to the social conditions for constructing postulates. Clearly, this approach appears to fulfil the basic assumptions of social constructivism.21 This shift within the development of feminist theory of gender difference leads to a number of far-reaching theoretical disruptions and tensions which feature prominently in recent feminist debates. Two problems would seem to be of particular systematic relevance here. On the one hand, social constructivism runs the risk of losing the central linguistic and political object of reference for feminism – ‘the women’ whose crosscultural, transhistorical discrimination is the subject of feminist thought, and whose liberation the feminist movement aims to achieve through political means.22 Recent social-constructivist reflections on the ‘materiality’ of bodies belonging to a specific gender are connected with this problem. On the other hand social constructivism leads to such a fine-grained analysis of gender that the category is in danger of losing its unity, for example by postulating multiple gender identities as a reaction to the criticism by black and non-Western feminists of gender constructions which are too one-sidedly liberal or bourgeois.23 It is significant that in particular Derrida’s attempt at deconstructing social gender that was developed in the course of his debate with Lacan has been the subject of increased attention.24 This deconstructivist approach rejects naive talk about subjects striving for emancipation, because this talk implicitly pre-supposes identifications 21 See Butler (1993), 1 ff. and above note 15, conditions (a)–(c). Butler herself is almost the only author who does not regard her work as ‘post-modern’. 22 See e.g. Young (1995). In a broader systematic framework all discussions of the relationship between feminism and post-modernism, in particular of the theoretical role of the subject, are relevant, see e.g. Benhabib et al. (1993) and Klinger (1995), with interesting comments and a large bibliography. 23 See on this e.g. Alcoff (1988a) and (1988b); Spelman (1988); Ferguson (1991). 24 See Derrida (1988); on the relationship between Lacan and Derrida regarding this question see Cornell (1993).

232

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

that are themselves exclusionary. On this view the dissolution of a general reference to women turns out to be a theoretical advantage. And a feminist theory that is based on the analysis of power and the theory of discourse is likely to appreciate the dissolution of the category of social gender into a historically changing setting of manifold interlinked sexual and social practices. This variant of feminism can perhaps provide political, strategic or pragmatic, but certainly not theoretical, reasons for talking about or for women in a universalistic sense. Seen theoretically there is no longer anything that could connect talk of particular women in a specific context of practices and strategies of power with talk of women in another such context. However, this far-reaching consequence is apparently not acceptable to all feminist theorists, even when they are otherwise sympathetic towards post-modern feminism. As a result several attempts have been undertaken to secure feminism’s political foundation, but without fuelling the identityphilosophical and essentialist residues of traditional feminist philosophy. For example, Young’s suggestion that gender should be construed as serial collectivity comes to grips with both of the problems outlined above; she presents them in the most concise of manners as a fundamental dilemma of contemporary feminist theory and politics: The logical and political difficulties inherent in the attempt to conceptualize women as a single group with a set of common attributes and shared identity appear to be insurmountable. Yet if we cannot conceptualize women as a group, feminist politics appears to lose any meaning.25

Young’s sociological concept of serial collectivity is intended to be tight enough to secure the reference of feminist theory and practice, while at the same time being sufficiently open to avoid exclusionary identifications of the concepts of women and gender. A serial collective groups people around a ‘practical-inert milieu’ (e.g. a factory), the objects and structures of which restrict goals and scope for action and so loosely connect the people involved – that is through their alignment to one and the same milieu. But when it comes to describing gender or women Young enriches the concept of milieu and its structures significantly. Thus, for example, the way women treat their own body is taken to be constitutive of the ‘series woman’; but this definitely does not include only biological processes such as menstruation and pregnancy, but also all specific social 25 Young (1995), 105. Haraway ((1991), 155–61) seems to follow a similar systematic strategy with her concept of ‘effective affinities’. Another attempt is undertaken by Maihofer (1995).

g e n d e r , n at u r e a n d r e f e r e n c e

233

practices and rules that are related to the biology of the female body, as well as many purely social practices and rules, for example those related to clothing and make-up. As a result it is hard to avoid the impression that a significant part of the enrichment and of the diversification of the concept of gender is instantiated by, and therefore pre-supposed by, serial collectivity, and so must reproduce the problems of reference and unity anew. On the other hand, the general strategy pursued by Young seems to be of some systematic importance – irrespective of whether one finds it successful or not. For this strategy comes down to weakening the category of gender as a unified structure significantly, in the hope of solving the reference problem. However, Young’s attempt to keep a theoretical balance between weakening and securing reference is probably doomed to failure; at least as far as the questions presented above are concerned, social-constructivist feminism seems to do little to solve the problems involved. As a result it is hardly surprising that the post-modern school of feminism has been subjected to a great deal of criticism, even in internal feminist debate. For example, in her review of Butler’s Gender Trouble Nagl-Docekal has quite clearly identified the two central problems of social-constructivist feminism – the destruction of women as the subject, and the idealistic implications of a deconstructivist approach involving the analysis of power and discourse.26 Klinger has also pointed out the danger that standards of rational criticism will be relativised, and is worried that post-modern feminism could become completely disassociated from the discourse of emancipation – a concern that is shared by Benhabib and is linked to an implausible interpretation of Foucault’s analysis of power and discourse.27 Finally, Nussbaum, followed by Nagl-Docekal, vehemently opposes a social-constructivist contextualisation of scientific

26 See Nagl-Docekal (1993), esp. 143 f., 145 f. In particular Nagl-Docekal points out that it is the ‘patriarchal gesture’ that distracts attention from differences between women by indiscriminately suppressing women as women; she is also of the opinion that on closer inspection Butler in fact propounds an inverted biological determinism. See also e.g. Grosz (1990). 27 See Klinger (1995), Benhabib et al. (1993), Benhabib (1993a), and above all the thoughtprovoking article by Alcoff (1990); similar views are expressed by Christian (1987) and Hartsock (1987). A balanced assessment of the state of the discussion, which at the same time pleads for critical reflection of the continuation of the ‘collaboration’ between feminism and Foucault, is presented by Sawicki (1994) (see also the article by Deveaux (1994), who points out that not only Foucault’s early doctrine of ‘docile bodies’, but also his late analysis of power veils a number of specifically female experiences).

234

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

and philosophical reason and sees in it a cementation of traditional female roles.28 However, these and similar positions do not appear to demand explicitly that the concept of gender be rehabilitated in feminist theories of gender difference and assume its due systematic place within feminist theory. Rather it is the new feminist essentialism with its neo-Aristotelian origin as propagated by Nussbaum, Annas and others that has regained the confidence to talk of the nature of humans. It is this sort of talk that is often seen as the most radical systematic criticism of post-modern feminism yet.29 To be sure, feminist essentialism is a variant of the ethical objectivism that has always pre-supposed a reference to the sexes. This becomes clear, for example, when Annas maintains that ‘the biological sex differences between men and women bring with them, in all known societies, enormous cultural divisions’ and that ‘Your sex may close some options to you entirely, or merely make them more difficult’.30 At the heart of feminist essentialism is a definition not of sex, but of general human nature;31 this definition should be ‘thicker’, for example, than the list of primary goods in liberal theories, for it is meant to contain the most important characteristics, functions and abilities that all humans possess and that at the same time represent (taken separately) the necessary and (taken together) the sufficient conditions for a creature to be human. But this does not involve a physicalistic characterisation of general human nature – feminist essentialism is not a branch of biology or of scientific anthropology, but claims to be an objectivist philosophical ethics. From the outset, the concept of general human nature and its essential elements is a normative one that has developed from intercultural forms of recognition, from historically documented forms of human understanding of the self, from reflections on what

28 See Nussbaum (1995a). This article is essentially a review of the volume of essays by Anthony and Witt (1994), whose contributions are basically coloured by similar attitudes to post-modern feminism as those formulated by Nussbaum herself; see also Nagl-Docekal (1996). 29 On neo-Aristotelian essentialism in the ethical sense see e.g. the volume of essays by Nussbaum and Sen (1993), in particular the contributions of Nussbaum (1993a); Putnam (1993); Sen (1993); also Nussbaum (1995); Annas (1988); Annas (1993); Nussbaum (1993); Nussbaum (1988), and furthermore Scherer (1993). The antipost-modern slant of feminist essentialism becomes clear in Nussbaum (1993) for example. 30 Annas (1993), 279. 31 An exception is the approach found, for example, in Wolgast (1980). Wolgast is of the opinion that humans belong ‘essentially’ to one of the two sexes.

g e n d e r , n at u r e a n d r e f e r e n c e

235

constitutes the good life in different cultures and from the preconditions for commonly held moral judgements, e.g. on the worldwide repression of women.32 The objection raised by feminist essentialism to the relativism inspired by post-modern thought seems to me to be on the right track. However, the essentialist school of feminism makes no contribution towards solving the difficult problem of how the reference to sex or gender can be established and secured. For in order to preserve its own systematic integrity it must blatantly refuse to define the ‘nature’ of the man or the woman, since it is rather our conceptions of general human nature that are supposed to provide the normative potential for a globalised moral criticism of the repression of women. To be sure, the method of making explicit the general human nature, which feminist essentialism has so far only very vaguely sketched, gives rise to concern. The definition of general human nature is to follow from cross-cultural forms of recognition, of understanding of the self and of moral judgements; yet it is by no means clear which theoretical criteria are to be used to choose those persons whose opinions and judgements are held to be valid. When reference is made to the fact that ‘we’ feel that men and women have unequal opportunities or that many ‘texts’ and ‘statements’ contain a particular understanding held by humans as to what constitutes their ‘nature’, then the question arises of who exactly ‘we’ are, and which texts and statements are at stake – and above all why it is these persons and documents that provide the data that are to be taken into consideration. For historical and ethnographic evidence could certainly be found to validate completely different attitudes. What is more, it should not be overlooked that the characteristics of general human nature as understood by feminist essentialism are biological features filled with social content. But since patriarchal attitudes are a worldwide and cross-cultural phenomenon, there is no guarantee that the same methodology that is meant to provide a definition of the nature of 32 Nussbaum relies mainly on the first three of these four principles for the identification of general human nature, whereas Annas emphasises above all the fourth and final aspect. In Women and the Quality of Life (Annas (1993)) she argues along the lines that we may not regard the distribution of opportunity among men and women in either ‘traditional’ or ‘liberal’ societies as unjust because it frustrates our preferences or does not correspond to even our informed desires, or because it leads to an unequal possession of goods or demotes women’s rights and autonomy – but because this distribution of opportunity places radical restrictions on the scope for action and the formation of life that women as human beings enjoy. However, according to Annas our general human nature does not clearly define a particular ideal way of life, but rather limits in a generally negative way the number of possible good lives.

236

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

man and woman could not produce an essentialist definition that was still patriarchally coloured.33 There is every indication that as far as gender difference is concerned feminist theory is at present in a fragile and explosive situation. I would therefore like to suggest a theoretical strategy that it is hoped can help to reduce the tension. In doing so I shall follow the internal feminist criticism of post-modern feminism and argue in defence of a nominalist, nonrelativist interpretation of Foucault’s analysis of power and discourse, so implying that post-modern feminism is based on an implausible relativistic reading of Foucault’s thought. At the same time I would like to take up the dissolution of the category of social gender entertained by some feminist theories, and treat it as a promising and productive effect of post-modern and post-colonial feminism. However, I suggest radicalising this effect and pleading for the complete abandonment of the concept of gender as a theoretical category. A thin concept of sex established within the framework of a contemporary theory of meaning can secure the reference to the sexes that is a vital foundation of the unity of the subject woman, and of the struggle for the social and political emancipation of women. This strategy separates the thin concept of sex from the theoretical dynamics of scientific concepts of gender and thus can take into account justified feminist concerns about how content-rich essentialism tries to define gender. My suggestion is intended to remove almost all semantic content from our concepts of ‘woman’ and ‘man’. According to this utopia, when we try to clarify what it means to us to be a woman or a man, or perhaps to belong to another thin sex, then we would have almost nothing to say on the matter – just as we would have nothing to say in answer to the question of what it means to us to be German or French. On the other hand, our heads and bodies most certainly should react when we ask ourselves what it means to us to love a man or a woman, to be a mother or a father, to work in a hospital, in an office or a university, to have particular sexual preferences, to appreciate certain forms of art, to have particular eating habits, to lead a religious life or to live in a country with a particular constitutional form. One of the most plausible explanations for the rejection of even the thinnest category of sex in post-modern feminism appears to be that the 33 In her informative and important contribution Scherer (1993) above all draws attention to the danger of metaphysical assumptions and paternalistic strategies in Nussbaum’s essential feminism. Scherer prefers an ‘autonomy-accentuated reading’ (919) that weakens the essentialist claims but allows a clearer differentiation between the human and the good life.

g e n d e r , n at u r e a n d r e f e r e n c e

237

post-modern reconstruction of the crucial categories of subject, power and discourse has been extremely relativistic. The most fundamental, as well as the most dangerous, misinterpretation concerns the concept of power. According to this reading power is seen as a strategic complex that permeates all domains of society and should be regarded as a historically mutable, transcendental a priori for all social phenomena, including languages and scientific theories. Judith Butler regards this thesis as the core of the post-modern approach:34 But if there is a point, and a fine point [sc. in the post-modern position] . . . it is that power pervades the very conceptual apparatus that seeks to negotiate its terms, including the subject position of the critic . . . To establish a set of norms that are beyond power or force.35

Since the subject is always embedded in established organisational principles of material practice, and these principles are themselves defined in terms of the strategy of power, the subject is constituted by power strategies; and so each subject is established by a process of differentiation and exclusion that cannot be suppressed.36 Beyond power strategies there are no subjects – this is the infamous claim about the death of the subject. Subtle variants of post-modern feminism emphasise that this neither excludes the use of categories such as ‘subject’ or ‘women’ nor implies normative anti-fundamentalism; rather what is at stake is that while admitting that the use of identificational categories is unavoidable, we should always examine the exclusions they imply, and deconstruct their definitional power and keep in mind their mutability. This picture is completed by an inflationary talk of discourses and discursive practices. The network they form is supposedly linked to omnipresent strategies of power that permeate and constitute knowledge and the subjects. In this sense the web of power can perhaps be regarded as a network of texts upon which subjects and social domains are just as dependent as they are upon the strategic complex of power. It is then a short step to suggesting, for example, that genders should be reconstructed as a ‘hegemonic discourse’.37 The conclusion this story arrives at is that all domains of society are permeated by an unavoidable web of repressive power structures that can perhaps be changed, but never completely got rid of (not even when it is uncovered by the analysis of power), and so renders all talk of emancipation obsolete. 34 35 36 37

Butler herself talks of ‘post-structuralism’ (see also n. 13). Butler (1992), 6–7. See e.g. Butler (1992), 10–16. See Maihofer (1995).

238

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

In chapter 1 I argued that the situation is in fact less depressing. The starting-point of this sober picture was the observation that particular rules are constitutive of many types of action, and that following these rules is what opens up scope, options and freedom for actions in the first place.38 It is at this point that Foucault’s interesting concept of power can be introduced. In the opening chapter it was demonstrated that we must follow Foucault in seeing power relations as fundamentally nominalistic: it is always human individuals who exercise power over other human individuals. That is to say, particular individuals bring other individuals to do something that they would not otherwise have done without this influence, and this power relationship is asymmetrical. The specific form of power that Foucault was mainly concerned with was described earlier as regulative power: it is the power specific persons have to bring other persons to follow certain social rules. When these social rules are constitutive of particular types of action, then clearly the power in question is productive in the fundamental sense that it is only persons who are subjected to this regulative power who have the opportunity and the freedom to act in a particular manner – and perhaps even have the opportunity and freedom to violate particular rules in cases where this violation is itself to be regarded as an action. Regulative power need not, but can, be free of repression – apparently it is not repressive when certain persons are brought to follow particular rules, and when following these rules does not frustrate the actors’ preferences, or even satisfies them. A trivial consequence of this is that actions that are constitutive of particular rules are always subject to regulative power, and that rules of action have an element of exclusion if only because their enforcement implies that they may not be violated. But this certainly does not mean that all specific forms of regulative power are unavoidable, and all social rules are to be criticised purely because they are exclusionary. Rather we should make use of our best standards of evaluation to decide which social rules we should criticise for being exclusionary, and which we want to accept for precisely the same reason. In addition, the post-modern demand that we should be aware of the implications of exclusion in all social norms and rules is not particularly original. On closer inspection it turns out to be no more than the reasonable appeal to take a closer look at the valid norms and rules in each individual case and to treat them as fallible – on the assumption that neither action nor freedom is possible to the desired degree in a space that is 38 This line of argument is also followed, for example, by Raz (1994), esp. 160–2.

g e n d e r , n at u r e a n d r e f e r e n c e

239

free of norms and rules.39 On this view it does indeed make sense to criticise certain mechanisms of regulative power – those that are involved in people following social rules which we should reject according to our best standards of evaluation; or else those forms of regulative power that, without good reasons, condition particular groups of people to follow certain social rules. This reconstruction of the concept of power allows us without further ado to draw up a new version of the vague claim that is so common in the texts of post-modern authors and feminists – that power intimately permeates even our speech and rationality. This claim should be understood as meaning that specific mechanisms of regulative power are required if humans are to speak natural languages, and to understand or develop scientific theories. For in order to achieve such abilities humans must, among other things, be persuaded either explicitly or implicitly to accept grammatical and terminological rules and to adopt certain sets of methodological rules, and thereby to participate in a particular game of truth. This is an unavoidable case of regulation and discipline that opens up the possibility and freedom of participating in linguistic or scientific activity, and at the same time draws the boundary between rationality and irrationality, between reason and madness. At this level the connection between power and knowledge is apparently of the most intimate kind. But even in this important specific case of regulative power the mechanisms and rules of power cannot only be changed, they can also be evaluated to assess whether or not according to the best reasons available to us they are repressive, unnecessarily restrictive and exclusionary, or successful and productive in our best interests. This is how it would be possible to demonstrate that the analysis of power was compatible with the basic principles of critical theory, even if some problems must perhaps be solved first before the two approaches are profitably combined. One of these problems is certainly posed by post-modern criticism of the concept of the subject. This criticism could be defused by pointing out that the nominalist approach to the analysis of power propagates a distinction between humans as individuals and humans as subjects, and uses this distinction to insist that relations of power pre-suppose the ontological existence of human individuals with particular preferences, and are a precondition only for specific processes of subjectivisation. Thus it is 39 According to most relevant texts the deconstructivist position cannot be reduced to this terminological fallibilism but demands the permanent problematisation of any categorial definitions. But it is precisely this generalisation that is highly unlikely.

240

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

a mistake to conclude from the fact that mechanisms of regulative power are always involved in the constitution of the subject that there are no individual subjects as relata for power relations that exist independently of these relations. It may not always be clear how a description of these relata can be found that is not at the same time infected by the terminology of power relations, but at any rate it is clearly here that one of the theoretical foundations of talk of repression and emancipation is to be sought. We must also take into consideration that Foucault and the most influential post-modern feminists contribute little to clarifying the process of subjectivisation. Thus, we are told that subjects are ‘constituted’ by regulatory and identificatory practices, but it is not clear that this means anything more than that certain persons are subjected to particular forms of regulative power, and so are brought to adhere to certain types of rules. Certainly they are ‘subjugated’, that is ‘subjects’ in the ancient sense of subjectum, but this concept of a subject is weak and trivial, even when we assume that particular practices can have a deep influence on our attitudes. As a result talk of the death of the subject is a truism.40 On the other hand, we arrive at a much deeper sense of subject and subjectivisation if we assume that self-ascriptions are constitutive of, or at least necessary for, metarepresentational consciousness. If we can only be conscious of our mental states in the metarepresentational sense if they are individuated and accessible to us in second-order thoughts, and if our mental states are accessible and individuated only if they can be rendered linguistically,41 then any form of regulative power that leads us to develop our lingusitic abilities is constitutive of, or at least necessary for, subjects as conscious beings. But this connection between subject and power would not in any way affect our ability to denounce repression and to dream of emancipation. The consequence of all this is that we have reason to be sceptical of the post-modern feminist reconstruction of the category of gender as a mutable bundle of manifold social practices that is permeated with repressive 40 What is more, the post-modern thesis of the so-called processes of subjectivisation or of a stylisation of life suggests that historical proofs have been presented that these regulations have been enforced generally in particular societies. But this does not seem by any means to be always true. For example Foucault’s own analyses of the situation in antiquity in UP and CS are based on only a few authors and texts. Even if we assume that these texts have been interpreted correctly (for which there is no immediate guarantee), then all they offer are the suggestions of a few authors from a small circle among the intellectual elite on the correct way to lead one’s life, and few conclusions can be drawn about the broader cultural or social situation. Much the same is true of Foucault’s late works on ‘technologies of the self’ and on ‘government’ in Martin, Gutmann and Hutton (1988). 41 See on this Becker (1995).

g e n d e r , n at u r e a n d r e f e r e n c e

241

power mechanisms, or even as a hegemonic discourse. This reconstruction is fundamentally flawed, both systematically and theoretically, for a number of reasons: it makes any reference to women and to the struggle for women’s interests impossible; it assumes that it is impossible to escape from omnipresent networks of repressive power; and it proclaims deconstructivism towards the use of identificatory categories which must pretend that it is itself free of normative principles. Certainly post-modernism has proved that gender indeed dissolves into manifold mutable bundles of practices and social rules that are stabilised by the mechanisms of power, and this is an important theoretical step as long as we employ a suitable reconstruction of the notion of power. Yet on the whole the systematic price that a post-modern gender theory must pay is probably too high. If this strictly social-constructivist reading of post-modern feminist theory is correct, then the foundations for the post-modern feminist attack on all categories of sex are weak and hardly sustainable. Perhaps some of the most influential post-modern feminist texts are compatible with a softer interpretation in that they accept a non-repressive concept of productive power, thus do not regard standards of reason and rules of speech as necessarily repressive, restrict deconstructivism to a justifiable fallibilism regarding terminological definitions and see the justification for the constructivity of gender solely in the fact that this category is an element in a dynamic of scientific theory. However, this weak version of social constructivism then offers almost no satisfactory reason for rejecting all categories of sex. It is hoped that the reasons for thinking this will become clear in the next section. 2 The background for the line of argument presented in this section is Donald Davidson’s theory about the way in which we understand sentences – for instance, sentences in which women or men are talked about. Davidson claims that at the most elementary level this understanding is based on our relation to those facts which, in a situation of radical interpretation or triangulation, we can imagine will produce in both the interpreters and the interpretants comparable linguistic reactions referring to precisely those facts. These causally relevant facts then determine the content of the linguistic reactions.42 In a metalanguage in which we formulate an extensional interpretation theory we can assume a distinction 42 See Davidson (1967) and (1973).

242

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

between a physicalistic and a mentalistic vocabulary, and so we can use this theory to formulate the assumption that in most cases of radical interpretation the first metalinguistic sentences which we use to try to understand statements in the object language will be phrased in physicalistic vocabulary. This is particularly true of evidence used to test the interpretation theory: the first T-theorems that can serve as evidence will refer on the right-hand side of the equivalences to facts in the physicalistic vocabulary of the metalanguage. One of the reasons for believing this is that the interpretation theory may demand that in elementary situations of understanding it is assumed that the interpretants will produce sounds that are elements of a language; that they adopt mental states that are propositional attitudes; and that their behaviour can be interpreted as intentional actions. However, this assumption is neither an infallible nor a necessary condition, but can only be recognised from the extent to which the statements in question can be translated and the corresponding actions understood within the framework of a given metalanguage. Therefore the first facts to which we turn our attention if this insight is to bear fruit may not be previously deducible as social states of affairs or as intentional actions, but be describable only in a physicalistic vocabulary. For the individuation of these facts must be based on classifications that as human organisms we can execute at the level of perception as a consequence of our natural evolution.43 Only if we manage to interpret our elementary physicalistic descriptions of human behaviour in attempts at radical interpretation more or less consistently in the light of our fundamental standards of rationality can we obtain empirical linguistic evidence that can serve to justify a mentalistic or intentional interpretation of statements or actions phrased in an object language. This can be supported by pointing out that evidence for the attribution of actions must refer to the attribution of propositional attitudes (most prominently beliefs and intentions). However, propositional attitudes produce intentional contexts that must be reconstructed in extensional terms within the framework of a Davidsonian interpretation theory. Therefore, ultimately this reconstruction can itself only be based on physicalistic vocabulary and assumptions of rationality, for there is no other way of avoiding the intentional ‘conspiration of intentions, beliefs and actions’ (as Rosenberg puts it).44 43 See Davidson (1990). 44 See Rosenberg (1988), 30–6. This book offers an excellent introduction to the philosophy of the social sciences that is both analytical and dualistic.

g e n d e r , n at u r e a n d r e f e r e n c e

243

If, in a situation of radical interpretation, we pay particular attention to events which we can correctly describe metalinguistically as a person’s following a social rule, assuming a certain role, or acting within a particular institutional framework, then in the most important cases we are talking about rules that are constitutive of the corresponding actions. To say that in his observable conduct Fritz behaves according to the rules of chess is already to say, among other things, that Fritz performs the action of playing chess, and that the rules of chess are constitutive of this action. Stating the social rules, roles or institutions that are embedded in specific actions generally enriches the explanation of them. But at the same time we cannot determine the instantiation of social rules by particular people without a physicalistic description of the behaviour of these persons (for example, that Fritz moves a small piece that we call a ‘bishop’ across a board consisting of thirty-two white and thirty-two black squares which we call a chess board). Thus the evidence for being able to understand social facts which consist of particular people following certain social rules will be more complex than the evidence for the description of simple actions, and it will contain additional physicalistic descriptions. At the same time it is clear that in a situation of radical interpretation a T-theorem referring to a social fact is far less determinate than a T-theorem that refers to physicalistically describable events; for social rules vary greatly between cultures, much more so than the biological processes involved in the processing of information in humans. But we often find ourselves involved in situations of interpretation that are far less radical, for we can already assume that the linguistic structures that we wish to understand are elements of a language, and that we already understand certain parts of the language (an interesting special case of this situation arises when we attempt to understand words or sentences of our own natural language that we have not previously understood). The dependence of the metalanguage on a physicalistic vocabulary is certainly not vital in such situations of less radical interpretation. But the more socially infected the facts on which we align our attempts at interpretation in these situations are, the less determinate the reference to these facts will be. Thus the more the reference of certain words or sentences varies in different specific vocabularies, the more important it is to arrive at an intercultural understanding. And the greater the differences are in the expressions and languages that we wish to translate into our language, the narrower the common basis for their simultaneous understanding will be, and the more important it will be to refer the postulates of meaning of these expressions and sentences to basic facts that can only be described

244

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

physicalistically. This follows from the fact that the difference between physicalistic and mentalistic vocabulary that we may assume for our own metalanguage implies the assumption that nature is more constant than culture. Social constructivists and post-modern feminists are right in assuming that in all social and cultural spheres for which we wish to study the practices, power mechanisms and statements, natural languages are spoken, and what they talk about includes women and men. This is clearly a vital precondition of a discourse-theoretical approach. However, feminists will wish to retain the possibility of referring to men and women in spite of the historical, social and cultural differences. It is the social constructivists who have taken on the task of revealing the intrinsic connections between various social practices, rules and norms, and the conception of women and men, and where necessary of denouncing them ideologically or deconstructing them radically.45 However, as we have already seen, if we are to translate a specific historical discourse about women and men into the metalinguistic framework of, say, a feminist theory, and so try to understand it, then meaning theory requires that the views of what women and men are that are involved in both frameworks have common elements. We must follow Davidson and say that the common elements on which the reference for ‘woman’ and ‘man’ in all languages and discourse is based refer to facts about men and women than can only be described in physicalistic vocabulary. Viewed from Davidson’s perspective it is precisely the social, historical and cultural dissolution of the category of gender so convincingly demonstrated by post-modern feminism that makes it necessary to reduce the basis of a common reference for concepts of gender to physicalistically describable aspects of the sexes. This applies not just to the situation of radical interpretation in the strict sense, but also for any attempt at interpretation that varies simultaneously through a multitude of different languages, some of which we already understand. The crucial point of all this is that the expressions in foreign languages that we generally translate as ‘man’ or ‘woman’ in English must be increasingly divorced from social content as we attempt to translate them from more and more different languages that are elements of more and more 45 Even the cleverest of arguments, such as those presented by Butler in Butler (1990), 1–25 cannot refute this thesis, as long as a suitable plain concept of discourse is assumed. For if discourses are types of spoken or practised speech within the framework of particular abstractions, then the fact remains that it is, of course, always particular individuals with a particular gender who speak the discourses outside of all subjectivisations; and their existence in the simple nominalistic sense must be assumed for discourses to be ontologically instantiated.

g e n d e r , n at u r e a n d r e f e r e n c e

245

different cultures; for the social content of concepts of gender varies from language to language and from culture to culture. Only if the concepts of man and woman shared certain social content in all accessible languages would it be impossible to completely separate this social content from the physical aspects of the sexes. It might seem fair to object that patriarchal structures are widespread and infect all languages, so that this separation seems in fact to be impossible. But on close inspection this objection cannot be upheld, for there are certainly some languages that are not plagued by patriarchism – for example, the languages of matriarchal cultures or of the best feminist theories, and we can understand them well and translate them if we make the effort. So we can envisage an extremely weak notion of biological sexes. The definition of the male and the female that relies on a reference to those biological facts which in all languages and vocabularies that we understand induce linguistic reactions that we can correlate in T-theorems extensionally with our words for women, men or members of other sexes provides the thin concept of biological sex. The thin concept of biological sex ensures that it is possible to understand even quite divergent discourses about women and men within the framework of very different social practices and cultural domains, and to translate them into our languages and in particular into the language of feminist theory. It is, by the way, obvious that even social constructivists and post-modernists implicitly make use of such translations. Maintaining a thin concept of sex, as required by a Davidsonian meaning theory, can thus offer theoretical advantages for feminist theories. This observation also applies without restriction to those special natural-scientific theories that in many cultures are intended to expand, deepen and state more precisely the physicalistic description of the difference between the male and the female. But it is of extreme systematic importance to emphasise that the thin concept of sex must be quite clearly distinguished from scientific theories about the difference of the sexes. Such theories have undoubtedly often enjoyed a truly dynamic development during the history of human thought, yet neither discourse analysis nor feminism could trace this development or demonstrate its historical shifts; and the various natural-scientific theories of sex differences could not refer to each other either affirmatively or critically without sharing those beliefs and ways of reference that are provided by a thin concept of biological sex.46 This 46 A good example is the work of Laqueur (1992), which is frequently quoted in recent feminist literature. Even if we assume that Laqueur’s historical theses are correct, they are in fact a study of the history of anatomy as a ‘scientific’ discipline, and cannot be used

246

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

is particularly so of psychological research on sex differences, a research which has been increasingly pursued over the last two decades and which claims to be a branch of natural science. It would be incorrect to understand this sort of research as an attempt to add just a few marginal distinctions to the thin notion of biological sex; rather it forms part of the development of natural-scientific theory, and is therefore to be included in the latter’s historical and social dynamic.47 Seen from this perspective many feminist reservations about the category of biological sex can in fact be described as a criticism of any essentialist definitions produced by natural-scientific theories of sex difference, and on this interpretation the criticism is justified. But it may not be expanded to cover the thin concept of biological sex, and it should be clear that special results produced by the dynamic of natural-scientific theories on the question of sex and gender are completely irrelevant for the alignment of any feminist theory to the thin concept of biological sex that is required by meaning theory.48 The distinction between the thin notion of biological sex and scientific theories of sex differences can also be described by employing concepts drawn from the modern essentialist metaphysics of natural kinds. This approach makes a useful distinction between the empirical stereotype and the microstructure of natural kinds. The stereotype is defined in empirical terms, but its reference to a microstructure is assumed only hypothetically; it is the job of confirmed scientific theories to identify the corresponding microstructure precisely. In essentialist metaphysics as propounded, for example, in Putnam’s early work,49 a sharp distinction was maintained between the stereotypes that are dependent on specific possible worlds, and the microstructures to which we rigidly refer in all possible worlds. against a thin concept of sex (even if Laqueur is right in claiming that there was only one sex in the ancient world, and this perhaps still applied into the eighteenth century – which I doubt – the ancient world, of course, still distinguished a woman or a girl from a man or a boy – even in medical texts – and this is important). 47 Pool (1994) provides a popular review of recent research on sex differences in psychology (the notes contain numerous references to academic literature). 48 For example, when Donna Haraway talks of an ‘invention and reinvention of nature’ and says that ‘by constructing the category nature, natural science imposes limits on history and self-formation’ (see Haraway (1991), 1, 43), then she is clearly referring to naturalscientific theories of particular aspects of nature. I find her studies on primatology brilliant documents of ideology-critical feminism that reveal the patriarchal structures of accepted approaches in it. But even Haraway, as well as the research she criticises, implicitly bases her work on a thin biological difference of the sexes than cannot without further ado be incorporated into the process of the ‘reinvention of nature’. 49 See Putnam (1975).

g e n d e r , n at u r e a n d r e f e r e n c e

247

This approach restricts the causal theory of reference to natural kinds, and bases this theory on the empirical similarity of things that is provided by the identity of their microstructures. However, Davidson is right in assuming that the causal theory of reference should not be restricted to natural kinds, and that the similarity of things and situations that is necessary for the causal reference and its role in the theory of meaning probably relies on classifications that we undertake by processing perceptions, and which are therefore accessible to us as stereotypes in a physicalistic vocabulary.50 Nevertheless it remains for scientific theory to conduct further studies of the referents that have been defined in this way, and where applicable to postulate natural laws, theoretical elements or microstructures about them. The separation of scientific theories of sex differences from a thin concept of biological sex based on the theory of meaning perhaps helps us to recognise that this thin concept can escape feminist essentialist criticism. Not only is it important to bear in mind that the thin concept of sex is open in the classificatory sense, and so does not exclude any human being extensionally from the distinction of the sexes. It is also particularly important to appreciate that in feminist criticism of essentialism the notion of ‘exclusion’ would seem to have a normative or even political connotation, while the thin concept of sex is of course systematically divorced from all normative or even evaluative connotations. The various thin classes of biological sex do not enter into an evaluative relationship towards each other, and thus the extensional ‘exclusion’ of a person from the set of all men and women is no more than a physicalistic assessment devoid of evaluative content, and given this it would be quite literally meaningless to complain about it. Perhaps we should add that a thin concept of biological sex is obviously not rendered obsolete by the observation (e.g. as made by Butler) that not only the category of biological sex, but also the distinction between sex and gender, is a social construction. This statement is clearly empty and is in no way helpful. For example, the theory of meaning on which part of the considerations presented in this section are based is, of course, a social construction in the broadest sense. But then social constructivism or postmodern feminism are also social constructions in this sense, for apparently they are serious, if fallible, attempts to develop the best possible theories or suggestions. But this circumstance cannot be employed as evidence for or against these and other theories and concepts. 50 See Davidson (1990), 197–8.

248

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

It does not require much theoretical subtlety to state the biological aspects of the difference between the sexes that would induce humans in all cultures to react at the most elementary level with statements that mean the same as our terms ‘woman’ or ‘man’, ‘girl’ or ‘boy’. Some are characteristics that have always been used to determine the sex of new-born babies – i.e. the presence of a vagina or a penis; others are distinguishing features of grown-up women and men. This list is perhaps in principle an open one, but for sensible theoretical reasons it should be kept as short as possible in order to avoid making it too acute and turning it into a set of necessary and adequate conditions. It is sufficient to refer to trivialities that are common knowledge. Not all, but only, women become pregnant and bear children; not all, but only, women breast-feed children or else manifest sexual practices, for example, that involve the vagina. As with any conceptual differentiation that can be suggested, a biological difference of the sexes will not be unacceptable just because there are cases that cannot be subsumed under it, as long as there are sufficient cases that can be. Furthermore, in the specific case of thin sex differences it is also important to make quite clear that theoretically the physicalistic vocabulary that is used to describe these differences does not include a restriction to heterosexuality. There are no theoretical reasons why particular biological phenomena could not exist that led us to introduce one or two additional sexes, for example the hermaphrodite, to use the vocabulary of the ancient world. What is more, the thin form of a biological difference of the sexes hardly implies any exclusions; for it is conceptually weak and therefore extensionally broad and by means of additional specifications of the categories can be extended to cover contentious cases. 3 A comparison of societies and cultures reveals the full extent of their social variety apparent in the numerous and quite different maxims of action, rules, roles, norms and institutions they manifest. The structural differentiation apparent in the history of many societies has led to a remarkable specification of social practices that is manifested in a rapid sublimation of a wide variety of preferences, practices and experiences of life. The recent development of feminist theory has shown that it is virtually pointless to try to construct a theory that can transform the relationship between many of these practices and the difference of the sexes into a unified category of social gender. On the other hand, using meaning theory to establish a category of sex based on the thinnest possible concept opens up the possibility

g e n d e r , n at u r e a n d r e f e r e n c e

249

of reintroducing in a new and more radical form an important insight that was part of earlier feminist theory. The dissolution of the notion of social gender obviously makes any claim about connections between sex and gender senseless, even if it were the negative claim that there are no reasonable connections; but nevertheless current feminist theory could at least lay claim to the assertion that apart from a few exceptions maxims of action, rules, roles, norms and institutions in the social space of randomly selected societies are in no way connected to a thin biological difference between the sexes. The irreversible rift between the thin biological sex differences and social practices with their guiding social rules is already implied by the impossibility of reducing our mentalistic vocabulary to the physicalistic. Above all, since there are no psychophysical laws, postulating a thin biological difference between the sexes based on meaning theory has absolutely no implications for a possible connection between what we must understand by women and men – and maybe further sexes – in the universal sense, and social or sexual practices of whatever kind.51 This view would allow feminist theory to argue seriously in favour of freeing the difference between the sexes of all evaluative, social or normative connotations, apart from a few exceptions. This conceptual suggestion could, and indeed should, be accompanied by the political demand and utopia that our understanding (including our self-conception) of what it means to belong to a particular sex should be radically divorced from all sexual or social preferences and practices that, for whatever reason, we wish to manifest; and that all mechanisms of regulative power that lead to an intimate interconnection between biological difference of the sexes and social practices, rules, norms, expectations or roles should be rejected as unjustifiable ideology.52 The realisation of this utopian demand would 51 Although intentional or mentalistic descriptions cannot be deduced from physicalistic sentences, physicalistic characteristics of sex can perhaps be used to justify certain social roles according to the following well-known scheme: (a) only humans with the physicalistic characteristic X fulfil role R optimally; (b) only men (in the biological sense) have X; (c) only men fulfil R optimally (‘men’ can be replaced with ‘women’). This (simplified) scheme is apparently used more often when the concept of sex that is employed is richer. I claim that the thin concept of sex in the sense outlined above can only be used for a very few marginal roles. And in particular, I deny that this scheme may be employed for those physicalistic characteristics of the sexes that are attributed to the sexes by natural-scientific theories. Such attributions are the result of the dynamic of theory and are often themselves ideological. 52 Authors who argue in favour of androgyny state similar demands, but also believe that all forms of heterosexual identity can be dissolved. But this possibility is apparently removed

250

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

obviously broaden the scope for sexual and social practices significantly, and so makes this utopia seem highly rational.53 As has already been shown above, few social practices, rules or roles should be excluded from this demand. These exceptions are the result of the obvious circumstance that, even on the thinnest concept of sex, a limited number of social practices and rules are preferably connected with women – for example, practices that revolve around pregnancy and breast-feeding,54 for breast-feeding is an eminently social and communicative activity. Nor can we deny that these few temporary roles affect the experience of life of those women who exercise them for even a short time. But we can, and should, assume that the influence of these exceptions is marginal compared with the effects of all the other social practices and rules to which in all known societies the discipline of strategies of regulative power have subjected humans, and to which women in particular have been subjected as members of a particular sex. After all many women have no children and many mothers breast-feed their children for only a short time, if at all; it would be most certainly wrong to claim, for example, that the experience of breast-feeding has a deep effect on mothers’ attitudes and sense of life, or on the way they think. On the other hand, if fathers have a close and intensive relationship with their babies and young children, then in most cases this will have a life-long and decisive effect on their own preferences, values, attitudes and sense of life. It is probably self-apparent that the theoretical and political demand for the dissolution of the category of social gender, and for the radical desocialisation of a thin concept of sex, is perfectly consistent with admitting that, as a matter of fact, probably in all known societies there are indeed sets of maxims of action, expectations, social rules, roles, norms and institutional structures that are linked to thin sex differences. And inasmuch as women, men and perhaps members of another sex (in the thin biological sense) have made these maxims of action their own, conformed to these expectations, followed the social rules, played the roles, adopted the norms and fitted in with institutional demands, they have in a profound sense if we stick to a thin concept of sex (see on this e.g. Ferguson (1977), esp. 45 and 56). Nagl-Docekal seems to arrive at a similar radical de-socialisation of the concept of genders, but from the perspective of a feminist critique of reason, when she demands that the idea that sexes have (social) features should be totally rejected (see Nagl-Docekal (1996), esp. 201 f.). 53 In his work on the theory of the media Vogel (2001) has produced a thorough exposition of the concept of rationality that on closer inspection goes beyond the purely linguistic notion of rationality in current theories of rationality. 54 This connection follows the scheme presented above in n. 51.

g e n d e r , n at u r e a n d r e f e r e n c e

251

effectively aligned – or had to align – their self-conception as women, men or members of another sex in a social fashion. Thus of course different and varied social gender identities have in fact existed; they have had a deep, life-long effect on the self-conception of all humans. But apart from a few marginal exceptions this connection is totally contingent and utterly unfounded. This is why powerful social mechanisms are required to veil this contingency, and to constantly re-produce social content in the category of gender and stabilise it in the long term. These mechanisms would seem to manifest themselves on at least three levels. On the one hand an ideological strategy ensures that the thin concept of sex is filled with ‘thick’ social content within the framework of prevalent intentions and beliefs. On the other hand subtle enforced relations of regulative power ensure that humans discipline themselves socially in the desired and prescribed manner as members of a particular social gender and adopt the social content of the categories of gender in their actions. Finally, evaluative lines of arguments and sanctions produce a normative utilitarian hierarchy between particular social practices which gives them a widely varying moral and social value. For men it has always been extremely pleasant that their political dominance ensured that socially defined roles and maxims of male action have generally had a higher place in this hierarchy than the socially established roles of women. Were feminists to adopt this line of argument, they would in effect be analysing existing sets of social practices, roles, etc., that in certain societies have typically been instantiated or followed by women, men or members of other sexes;55 but they would no longer want to transfer the historical and cultural variability of these sets into a unified category of social gender, and at the same time regard actual ties between any of these sets and thin sexes as contingent ideological constructs, and demand that they be removed from the theoretical and political stage. As a result feminism could then return to one of its most important and principal tasks – cutting away at all levels and in the darkest corners of society the ideological veil beneath which the thin category of sex had been linked with social content; investigating the normative arguments that have led to the establishment of an evaluative hierarchy of various social practices and roles; and uncovering the regulative mechanisms of power that consistently resort to discipline and sanctions in order to establish ideological links between the biological difference of the sexes and social contents. 55 It is in this field that the analyses of post-modern and post-colonial feminism have produced important contributions – perhaps because some of the analytical tools of post-modern thought are extremely well suited to observing even the finest of distinctions in this context.

252

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

From this perspective it would be reasonable for feminist theorists to exclude from their own theories all gender categories that have any sort of social content – even when conducting a critical diagnosis. Rather it must be in the best systematic and political interests of feminism to refrain from all theoretical definitions of women, men or other sexes that involve any sort of social practices, rules or roles; for otherwise feminist theory would itself be based on the very ideology and regulative mechanisms of power that it in fact wishes to combat. Yet current feminist theory does not always seem to pay attention to this interest. In my opinion the trend to social constructivism and the abandonment of even a thin biological concept of the sexes, combined at the same time with recognition of the category of social gender, often leads to the concept of gender being filled with social content, either implicitly or explicitly, only women’s roles are now of course evaluated quite differently from the way they were in patriarchal thinking; and this concession probably means that the battle against the patriarchy is already half lost. Post-modern feminism in particular abandons all theoretical weapons that could be used to divorce sex differences from social content.56 This leads to the suspicion that at least some versions of feminist theory are ideologist, and must live with the charge that at the theoretical level they play into the hands of the patriarchy. Post-modern feminism has consistently used anti-fundamentalist arguments to remove such suspicion and rebut the charge, and these arguments reveal the relativist foundations of post-modern thinking. It regards talk of ideology critique and normative investigation as quite naive, because such talk pre-supposes the ultimately illusionary charge that the normative foundations of critique can be secured universally, and disguises the fact that all critique and investigation is itself always involved in local power games.57 From this perspective the only reasonable strategy is constantly to replace the involvement with power with another kind of involvement, and repeatedly to water down the unavoidable creation of identities and categorical definitions. But we can reject this counter-argument simply because relativism is an untenable philosophical position, and this is so of all its varieties, whether moral, rational or world view.58 However, this does not mean that we have 56 If, as in Butler (1990), the active subjects are only constituted in the course of a performance of social practices and (regulative) power strategies, then the very essence of the theory precludes this separation. 57 See e.g. Cornell (1993) and Butler (1992). 58 See Davidson (1974).

g e n d e r , n at u r e a n d r e f e r e n c e

253

to subscribe to naive universalism or fundamentalism. Even post-modern feminism cannot see through God’s eyes, but is itself caught in a specific perspective from which it cannot escape, and this is of course true for all theoretical approaches – including the thoughts presented in this chapter. The decisive point is that to state this is merely to state a triviality, devoid of interesting content; for apparently within our perspective we have no chance to define an Archimedean point from which we can seriously describe our perspective as a specific world view. What is more, as was suggested above,59 it should be clear that we must always claim that our perspective is a universal one into which all other specific views that we can understand in their own specificity and otherness can be translated. But the demand for, and necessity of, translatability implies a great deal of common ground between ‘our’ and other perspectives, and this renders the relativistic assumptions obsolete. At the same time we must admit that although we regard ‘our’ perspective as universal, this is only a hypothesis that is by no means immune to revision. In particular it is not unlikely that sensitive and careful historical or cultural studies, as well as intercultural dialogue, will result in certain revisions, even if nobody can predict how extensive these revisions will be or what form they will take. But above all, after such a revision we are back in the same situation again: we cannot see through God’s eyes, we must be able to translate all comprehensible specific perspectives into our new language and view, and so be able to continue claiming hypothetically that our new view is universal. This soft foundation is completely sufficient to pursue normative arguments as we please, with no danger of circularity, or to indulge in the critique of ideology. This is all the more true since this ‘soft’ universalism contains a ‘hard’ core. Davidson was right in pointing out that the assumption of a basic form of rationality must be taken as constitutive of all living creatures to which we attribute propositional attitudes and whose behaviour we interpret as actions. This basic rationality which is constitutive of all acting beings, and thus must be assumed to apply to the actions of any person in all cultures and at all times, includes elementary rules of logic, the most important principles of decision theory (e.g. the transitivity of preferences), Carnap’s demand for total evidence (accept all hypotheses that all in all support your reasons), Quine’s principle of conservation (under otherwise equal circumstances changes as few expectations as possible when dealing with unmanageable phenomena), as well as Davidson’s principle 59 See above ch. 1, pp. 53–6.

254

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

of continence, which demands that everyone should act in accordance with the recommendations of the best available reasons they have.60 From this standpoint it is meaningless to ask whether particular people subscribe to these principles of rationality or not: as actors with intentions and beliefs they always instantiate these principles, even if they have no explicit knowledge of them. Of course this does not mean that all actors always act rationally according the above-mentioned principles – irrational action is common enough. Although, to be more precise, it is not individual patterns of action that are irrational, but merely larger sets of inconsistent beliefs, intentions or actions, perhaps combined with principles. Above all, however, a person’s inconsistent behaviour patterns can only be described as irrational against the background of the general rationality of that person: we would be unable to understand creatures that as a rule do not follow the principles of rationality listed above, or to describe them as rational or irrational. For example we can recognise that a person does not wish to get wet and believes that it is about to rain because ceteris paribus that person takes suitable action in order not to get wet; were that person never, or rarely, to act in this way then we would be right in doubting whether they had such an intention or belief at all. And if we could never rationalise the behaviour of a particular creature in terms of intention and belief, then we would begin to question whether this creature had propositional attitudes at all. Or let us assume that we think we can recognise preferences in a creature such that it always prefers A to B, and B to C but never A to C; then we are justified in asking if such a creature really has genuine preferences. Yet if we can never attribute transitivity of preferences to a creature’s behaviour, it will be difficult to interpret it as an acting person.61 Davidson is probably right in basing this thesis – that particular creatures only have propositional attitudes and are acting persons if in their behaviour they normally instantiate the most common principles of rationality – on the argument that only this instantiation of basic rationality allows the individuation of propositional attitudes. For propositional attitudes can only be individuated by the relationship in which they stand to other propositional attitudes and this system of propositional attitudes must be coherent. We can only have a particular single propositional 60 See Spitzley (1994) on this, as well as for the relevant evidence in Davidson. 61 Thus some of our deepest standards of rationality can also be mapped on to the usual form of rationalising actions in terms of intentions and beliefs: it must be assumed that the form of the practical syllogism is irreversibly rational, the intentions must satisfy the transitivity of the preference, and the beliefs must be logically consistent.

g e n d e r , n at u r e a n d r e f e r e n c e

255

attitude if we also have a series of other coherent propositional attitudes, even if in each case it may be unclear exactly which further intentions or beliefs are correlated with a particular intention or belief. This ‘meaning’holism – is closely connected to the linguistic competence that we possess as speakers, and would have at our disposal if we had access to an extensional theory of interpretation. This is why Davidson can say: Here we have reached an aspect of rationality so fundamental that we cannot make sense of an agent who does not generally reason in accord with it. And so we have reached a point, where the distinction between the standards of rationality of the agent himself and of his critic merge. It is an objective, though normative, judgement that someone whose reasoning is on some occasion not in accord with it has reasoned irrationally.62

Quite clearly this poses a serious challenge to relativistic strategies, and if it is right, as I assume it is, then it renders these strategies obsolete. 4 One reason for recommending particular theoretical strategies is that they are of relevance to current debates on central problems in their specific domain. Thus I would like to conclude with a brief postscript that is intended to at least give some indication that the suggestions presented in the first three sections of this chapter can lead to attractive positions on some of the most hotly debated questions in feminist theory. One of these questions concerns the precarious relationship between feminist theory and psychoanalysis. From what has been said so far it follows that Freud’s theory and the subsequent psychoanalytical tradition must be regarded as one of the most influential and dangerous strategies to have unscrupulously filled the relationship of the sexes in the thin biological sense (i.e. within the framework of the difference that is connected with the phallus) with strong social content. One of the main reasons for this unfortunate result is the lack of any clear methodological 62 See Davidson (1985), 346. All told his article is particularly relevant in this context. Dennett (1993) presents similar arguments, for example in his stimulating article ‘Intentional Systems’. But Dennett goes even further than Davidson in his claim that we must attribute rationality to intentional systems; for Dennett’s notion of an intentional system is weaker than Davidson’s concept of humans who have propositional attitudes – Dennett regards a system as intentional if we can successfully explain and predict its actions by attributing opinions and wishes to it, but without claiming that it is conscious or moral, ‘really’ thinks or can communicate by means of speech (like mice or chess computers which we can certainly encounter in an intentional stance, and to which we must therefore attribute rationality).

256

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

status to Freud’s theory, which oscillates as it does between nomological explanations and hermeneutic interpretations. Feminist theory is right to criticise the patriarchal alignment of Freud’s approach, but this criticism disavows the concept of sex as a whole, even if this is not surprising when we consider how unashamedly Freudians have misused this concept for the interpretation of sexual and social practices. The sceptical reaction of feminists to any use of the concept is an understandable theoretical strategy, but ultimately the incorporation of sex in social gender demonstrates the enormous influence exercised by psychoanalytic theory, even on positions that are critical of psychoanalysis. For after all psychoanalysis was the first influential theory to push forward the dissolution of the distinction between sex and gender.63 Another problematic area is connected to the claim that there are two distinct forms of morality, a female and a male one. This distinction has earned a great deal of respect and acceptance in feminist circles thanks to Gilligan’s influential studies.64 In the light of the above discussion we must first ask how this claim is to be understood. One common reading is that a specific female moral attitude associated with standards of care and attention is attributed to women, while men align their moral position according to abstract criteria and universal principles. Proponents of this reading are doubtless right in complaining about the different normative evaluation in liberal Western industrial societies of care and attention on the one hand, and of a universalistic sense of duty and justice on the other. But their attempt at tying the two heterosexual categories of sex (in the thin biological sense) to different moral content is unfounded and smacks of ideology. If we only characterise women and men in terms of the thin 63 For an informative and comprehensive review of Freud’s attitudes to female sexuality in his various works see ‘Drei Abhandlungen zur Sexualtheorie’ (1905); ‘Einige psychische ¨ Folgen des anatomischen Geschlechtsunterschieds’ (1925); ‘Uber die weibliche Sexualit¨at’ (1931); ‘Die Weiblichkeit’ (1932); see Chasseguet-Smirgel (1979), 7–67. The volume also documents some of the most important affirmative and critical reactions to Freund’s theory of female sexuality within the psychoanalytic tradition that were already being voiced in the 1920s and 1930s, and since then the variety of the reactions has, of course, increased enormously. But in spite of their differences my impression is that all commentaries make use of Freud’s deepest terminology. Since Freud notions such as ‘castration complex’ or ‘penis jealousy’ have been unchallenged in the psychoanalytical tradition, and applied to the reactions of young girls and boys to ‘pure’ observations of those differences represented by the thinnest of concepts of sex. But this theoretical step is totally unfounded, and suggests strong social reactions on the part of young children to physical facts. This is simply absurd, for in this respect small children are still in a situation of radical interpretation. 64 The excellent and interesting works of Baier (1994) and Honneth (1994) are good examples of this respect.

g e n d e r , n at u r e a n d r e f e r e n c e

257

biological concepts, then it is hardly likely that women are particularly caring and attentive while men are particularly likely to indulge in abstract moral reasoning. On this reading the claim about two sorts of morality in fact merely reproduces artificial links between the heterosexual difference of the sexes and normative attitudes and practices.65 Care and attention, and an abstract sense of justice, are in fact two different attitudes that are of equal value; in a liberated society they are distributed contingently in spite of the difference of the sexes, and ideally each member of any sex should assume them to some extent on certain occasions. We should at least expect that as far as the content of valid social rules and laws is concerned these should take both attitudes into account. On closer inspection the data provided by interviews with people who are subjected to some of the most influential induced mechanisms of relative power can only lead us to another reading of the claim about two sorts of morality. This says that all we can observe (if we can indeed observe it) is that in the social space of our society women and men often have different moral attitudes, but that these differences are often the result of unjustifiable, ideologically based attributions and conditioning. On this reading, Gilligan’s studies, assuming that they are based on valid data, could be regarded as part of ideology-critical feminism, and would be thoroughly instructive and welcome.

65 This is true, for example, of the influential studies and lectures by Irigaray. In my opinion Irigaray’s claim about the duality of sex and the characterisation of the female as ‘the sex that isn’t one’ is increasingly overfilled with social content. Some sections of her early book Ce sexe qui n’est pas un (1977) can indeed be read as critical analyses of sexual attitudes and emotions that are implied by a thin biological difference of the sexes, and this line of argument must be of interest and relevance to feminist theory. For example, in the colloquium for her habilitation in philosophy she emphasises that she is a women, a female sexual creature, but that it is, strictly speaking, impossible for her to make such a statement (sc. within the framework of the dominant, male and phallocratic discourse), and that what is important would be to practise a difference of the sexes outside the male discourse. But this implies that she envisages a specific female ‘way of reading, writing, interpreting and assertion’ (Irigaray (1977), 155, 165 f.) that incorporates some central social contents into the difference of the sexes that is to be lived anew. In her later book Sexes et parent´es (Irigaray (1987), 130) this tendency becomes stronger. For example, Irigaray fears that feminist demands for equality ‘carry in them the danger of the destruction of the female sex’ (Irigaray (1987), 130), and this specific fear apparently also pre-supposes a social interpretation of the female sex outside patriarchal discourse. A further indication of this is the strong link Irigaray propounds between a difference of the sexes that is not coloured by patriarchal attitudes and mythical contexts (e.g. the heterosexual cosmology of most myths) that systematically undermine the differences between sex and social gender, and more generally between nature and society (cf. above all Irigaray’s lecture ‘Le genre f´eminin’ in Irigaray (1987), 119–38).

258

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

For some time transsexuality and transvestism have been the object of increased attention in feminist theory. Generally they are used to demonstrate that the heterosexual difference of the sexes is a patriarchally constructed dualism that represents a violation of sexual self-description, and fails to take the self-conception of transvestites and transsexuals into account. Recently Landweer has quite rightly voiced doubts about this interpretation, but she too remains committed to general social constructivism, and so has no plausible interpretation of her own to offer.66 From the standpoint I have presented here it seems sensible to regard the typical self-descriptions of transvestites and transsexuals not as exceeding the bounds of ideologically constructed patriarchal difference of the sexes with its social content, but on the contrary as its reproduction and ultimate perfection. Thus when, for example, people who are men in the sense of the thin concept of sex feel that they are women and describe themselves as having a sexual identity for which their body in the sexual sense is the wrong container, then the category of ‘woman’ that is expressed in this sort of language is a purely social concept and refers to a complex of sexual preferences, practices and social rules that has been condensed into a category of gender. Transvestites or transsexuals have apparently fully internalised the ideological-patriarchal construct of social gender, and as a result of this internalisation can only describe themselves as a person whose (social) gender identity corresponds to a false (biological) sex. In particular, these self-descriptions apparently pre-suppose that it is reasonable and admissible in the first place to fill the concept of sex with social content. But it is precisely this move that should be called into question. In a free society persons of whatever sex could have virtually any sexual preferences and try out any sexual practices whatsoever, and in the process follow random social rules without feeling that this contradicted their thin sex.67 For to feel a contradiction here would rest on the false assumption that biological sex was tied to social categories. It is even imaginable that in such circumstances the desire of transsexuals for surgical measures would diminish. Admittedly, this diagnosis probably does not apply to those sexual practices that are still tied to a thin biological category of sex, and it may be that these practices are particularly important for some transsexuals or transvestites. The difficult relationship between equality and difference has long dominated feminist discussion, and although it seems to have featured 66 See Landweer (1994). 67 This freedom is of course limited by biological boundaries that are the result of the definition of thin sex. But this limitation is not unduly restrictive.

g e n d e r , n at u r e a n d r e f e r e n c e

259

less prominently in recent years, it is still a central problem. Feminist attacks on the principles of equality in bourgeois liberalism have rarely been based just on the historical fact that the legal enforcement of equality in bourgeois societies was at first restricted to men, and has done little to change the – undeniable – massive political and social repression of women. Instead feminism has searched for reasons for this discrepancy, and at least two lines of argument can be identified here: some authors have attempted to prove that the root cause of this evil is to be found in the principles of liberal theory itself, above all in the atomistic, individualistic anthropology that emphasises the struggle for power, envy and rivalry as the essential features of ‘the’ human; features which are, however, soon revealed to be the central characteristics of male behaviour in patriarchal societies. Other feminists have brought difference-theoretical arguments to bear, either in support of or independently of anti-liberalism. They insist on the specificity of the female, and demand that specifically female competence and experience are fully taken into account and recognised. It is my impression that neither of these lines of argument is particularly promising. The anti-liberal strategy is based on an inappropriate interpretation of liberal theory (at least in its best modern variants, for example as propounded by Rawls), because liberalism does not seriously rely on anthropology. Rather, as part of an ambitious programme that aims to derive theses with a strong normative content from preconditions that are only very weakly normative, it has to rely on abstractions that are methodologically useful for the realisation of the programme: for example the ‘original position’ in Rawls’s theory does not define anthropological constants, but the methodological structure of a rational decision process and of the justification of a meaningful concept of justice based on weak normative premises. A further difficulty encountered by the anti-liberal feminist strategy seems to me to be that it has a dangerously close affinity to communitarian arguments that often smack of theoretical restoration. Thus I feel that it is rather recent attempts at combining feminist theory with the best variants of the liberal programme that should be rigorously pursued, for at present the latter offers the most attractive justification for a strong, meaningful concept of justice of the kind that must, for example, be involved in the context of ideology-critical arguments.68 68 Two feminist authors who sympathise with this strategy are Jean Hampton and Susan Okin, see e.g. Hampton (1991), 31–55; Hampton (1993); also Okin (1989) and (1989a). Recent convincing examples for this line of argument are O’Neill (1996) and Pauer-Studer (1996).

260

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

The ideas presented here naturally lead to even stronger reservations about the role of difference theory. For in all its variants it apparently tolerates, or even expressly aims at, filling the difference of the sexes with social content, and then demands that the various social contents of the categories of gender are assigned equal values. Quite clearly our approach must reject this strategy, and denounce it as being too ideological. Undoubtedly of more importance is the question of whether our approach can also lead to a more positive attitude to the problem of equality and difference. I think that this is indeed the case. We can accept the principles of equality by demanding the unrestricted application of the principles of justice in their best liberal guise, that is unrestricted political and legal equality under conditions of a fair dynamic to all members of whatever sex in the thin sense. And we can also integrate arguments of a particular kind from difference theory, although the ‘difference’ involved must no longer be connected to the difference of the sexes, but be interpreted as the distinction between different social practices and between the different corresponding experiences. Thus we can indeed fill the concept of difference with social content, but must divorce it completely from sex differences. What is then at stake is not only accepting the existence and justification of quite disparate practices and experiences, but also demanding that these disparate practices and experiences are taken fully into account according to normative principles – although quite independently of whether it is men or women (in the thin biological sense) who instantiate these practices and have made these experiences. For example, it cannot be denied that a life which is mainly dedicated to caring for other people leads to experiences that enrich life, and as such this should be taken into account at both the political and the theoretical level. However, it is not because they are specifically female experiences that we should take them into account, but because they are good and important; it may be the case that in our society it is unfortunately above all women who are ideologically conditioned to these practices and experiences, but in a free society we could also imagine most men partaking in them. This leads on to a point of some interest, namely whether particular O’Neill outlines an interesting universalisable conception of justice that avoids the ‘idealisation’ of patriarchally biased liberal theory, and so does justice to feminist critique of liberal contract theory while at the same time maintaining the most important principles of liberalism. Pauer-Studer also goes into Sen’s and Nussbaum’s critique of Rawls’s ‘weak’ theory of the good life, and produces an extremely informative suggestion for the integration of both approaches into liberal contract theory.

g e n d e r , n at u r e a n d r e f e r e n c e

261

social practices and the corresponding experiences can lead to different epistemic attitudes. If this were the case, it would not of course mean that there was a specific female or male mode of knowledge or even rationality, but merely that different modes of knowledge or notions of rationality are linked to different social practices. The connection between different epistemic worlds and the difference between women and men would certainly be contingent and politically relevant, but of no theoretical interest whatsoever.69 The final problem that I would like briefly to touch upon has been at the fore of recent feminist discussion, and it is probably no exaggeration to 69 Just as with Gilligan’s studies, many studies of feminist critique and theory of science can in this context be read in two different ways – so that they either assume gender-specific experiences that women as women and men as men will experience and which will have certain epistemic effects, or so that it is specific social practices that may be linked to specific epistemic orientations, but that these practices may themselves in effect be intimately but are in fact only contingently connected with the sexes (a typical example of this is the study that in some respects continued Gilligan’s work: Belenky, Goldberger and McVicker Clinchy (1986)). This ambivalence is the result of the fact that what these studies understand by ‘women’ already include those typical complexes of social practices that are regulatively associated with women in patriarchal societies. Among the three most important positions of the current feminist theory of science that Sandra Harding for example distinguishes, this is particularly true of the so-called ‘standpoint theory’ (as presented by e.g. Hartsock and Hilary Rose; Harding (1991) herself defends a particular (post-modern) version of this position, see e.g. her discussion on p. 119 ff.). Emphasis is constantly placed on the belief that however this difference is defined the ‘difference’, the ‘being other’ of women leads to other, sometimes richer epistemic attitudes (e.g. as Harding explains, as a difference of personality structure, competence or emotions): ‘Whatever the kind of difference identified, the point of these arguments is that women’s ‘difference’ is only difference, not a sign of inferiority. The goal of maximizing the objectivity of research should require overcoming excessive reliance on distinctively masculine lives and making use also of women’s lives as origins for scientific problematics, sources of scientific evidence, and checks against the validity of knowledge claims’ (Harding (1991), 122–3). Such statements have a dangerous and unnecessary social-essentialist undertone. It is the ‘living-conditions’ themselves (which are only linked with women contingently and ideologically) that are perhaps a source of richer and better knowledge, and care is taken to avoid the impression that particular living-conditions are necessarily associated with women as women. For even if it could be proved that the social practices to which women have been ideologically subjugated in the past were not a source of better scientific knowledge (but are useful in another way), then feminist science would still have another point – to demand that the ‘living-conditions’ of women and men must be restructured simply because knowledge is also power. See the recent informative article by Nagl-Docekal (1996), who provides a comprehensive summary of the present state and the most important variants of feminist critique of reason together with critical comments. In spite of a degree of approval, Nagl-Docekal finds this criticism is exaggerated, completely avoids reason-essentialism, and sees in the socially constructed ‘masculinisation of reason . . . the real stumbling-block for feminist critique of reason’ (Nagl-Docekal (1996), 199).

262

f o u c au l t a n d c l a s s i c a l a n t i q u i t y

claim that this is where the hardest battles are being fought at present. The question is: should feminist theory regard itself as primarily a post-modern or as a critical theory? The approach presented here provides good reasons for agreeing with Fraser when she says that this question rests on a ridiculous, unnecessarily radical dispute. On the other hand, I find that Fraser fails to arrive at a clear theoretical diagnosis of this dispute; in particular it is more than just a question of unnecessary misunderstandings between the leading protagonists in the debate, e.g. Butler and Benhabib, for any solution to the problem that is suggested implicitly depends on an interpretation of the post-modern analysis of power. In the first section of this chapter I have argued for the diagnosis that the main proponents of post-modern feminism, e.g. Flax, Cornell or Butler, have based their work on an exaggerated and inappropriate interpretation of the analysis of power. But on the other hand, I cannot see that the proponents of critical theory, such as Benhabib and perhaps to some extent Fraser, have adopted the more interesting or productive insights provided by the analysis of power.70 The sober reading of the analysis of power that I outlined above abandons the fashionable relativism of post-modern thinking, but without committing itself to a naive universalism. From a theoretical viewpoint it is in fact a ‘weak’ Davidsonian universalism that allows the systematic integration of a sober, nominalistic reconstruction of the analysis of power with the principles of a critical theory. Such attempts at integration would involve among other things a careful reconstruction of the intrinsic connection between power, knowledge and subjectivisation in Foucault; a rejection of the aesthetic stylisation of his programme of ethics; and the sharpening of his concept of productive regulative power to a concept of repressive power. These moves would provide the critique 70 In one of her most recent studies Fraser uses a differentiation of the concept of justice to suggest an interesting systematisation. When an (un)just distribution of (economic) recognition can be distinguished from an (un)just distribution of (cultural) recognition, then it is soon clear that there is an unjust distribution between the (social) genders in both respects. According to Fraser the removal of this double injustice can be either affirmative or transformative. At the economic level the more radical transformation should lead to a ‘feminist-socialist democracy’, and on the cultural level to the ‘destabilisation and deconstruction of the dichotomies of gender’. This may sound interesting, but is in fact merely recounted in typical post-modern jargon: as a constant re-modelling of changing gender-identities and differences. This is not radical enough – there should not be any gender-identities or gender-differences any more; instead the concept of gender should be completely eliminated; it is not the gender-identities that should flow freely, but sexual and social practices should be able to pour freely across the differences of sexes (see Fraser (1995)).

g e n d e r , n at u r e a n d r e f e r e n c e

263

of ideologies and feminist theory with the chance to use the analysis of power for their own goals. In this way a theoretical connection between Foucault’s analysis of power, critical theory and modern feminism could be established that would be in a position to exploit the theoretical potential of all three approaches in order to develop a powerful new critical theory of society.

REFERENCES

Ackrill, J. (1980): ‘Aristotle on Eudaimonia’, in Oksenberg-Rorty, 15–34. Alcoff, L. (1988a): ‘Cultural Feminism versus Post-Structuralism: The Identity Crisis in Feminist Theory’, Science 13, 405–36. (1988b): ‘Poststructuralism and Cultural Feminism’, Signs 13, 4–36. (1990): ‘Feminism and Foucault: the Limits to a Collaboration’, in Dallery, A. and Scott, C. (eds.) (1990), Crises in Continental Philosophy, New York, 69–86. Andriopoulos, D. (1990): ‘Alcmaeon’s and Hippocrates’ Concept of Aetia’, in Nicolacopoulos, P. (ed.), Greek Studies in the Philosophy and History of Science, Dordrecht, 81–90. Angehrn, E. and Fink-Eitel, H. (eds.) (1992): Dialektischer Negativismus, Frankfurt am Main. Annas, J. (1980): ‘Aristotle on Pleasure and Goodness’, in Oksenberg-Rorty, 285–99. (1989): ‘Naturalism in Greek Ethics’, in Cleary, J. J. and Shartin, D. C. (eds.), Proceedings of the Boston Area Colloquium in Ancient Philosophy 4, 149–71. (1993): ‘Women and the Quality of Life: Two Norms or One?’, in Nussbaum and Sen, 279–96. Anthony, L. and Witt, C. (eds.) (1994): A Mind of One’s Own: Feminist Essays on Reason and Objectivity, Cambridge. Arendt, H. (1958): The Human Condition, Chicago. (1980): Macht und Gewalt, Munich. Arnim, H. v. (1914): Platos Jugenddialoge und die Entstehungszeit des ‘Phaidros’, Leipzig. (1916): ‘Platos Lysis’, Rheinisches Museum f¨ur Philologie 71, 15–29. Aron, R. (1986): ‘Macht, Power, Pussissance’, in Lukes, 253–78. Backrach, P. and Baratz, M. (1962): ‘Two Faces of Power’, American Political Science Review 56, 451–460.

264

references

265

Backrach, P. and Baratz, M. (1963): ‘Decisions and Non-Decisions: an Analytical Framework’, American Political Science Review 57, 641–51. Backrach, P. and Baratz, M. (1970): Power and Poverty, Oxford 1970, esp. part i. Baier, A. (1994): ‘Wir brauchen mehr als Gerechtigkeit’, Deutsche Zeitschrift f¨ur Philosophie 42, 225–36. Barnes, B. (1978): The Nature of Power, Urbana/Chicago. Barnes, B. and Bloor, D. (1982): ‘Relativism, Rationalism and the Sociology of Knowledge’, in Hollis and Lukes, 21–47. Barnes, J. (1980): ‘Aristotle and the Methods of Ethics’, Revue Internationale de Philosophie 34, 490–511. Barry, A., Osborne, T. and Rose, N. (eds.) (1966): Foucault and Political Reason: Liberalism, Neo-Liberalism and Rationalities of Government, Chicago. Becker, A. (1995): Sprachen der Erfahrung, unpubl. manuscript, Frankfurt am Main. Belenky, M. F., Goldberger, N. R. and McVicker Clinchy, B.(1986): Women’s Ways of Knowing, New York. Benhabib, S. (1993a): ‘Feminismus und Postmoderne. Ein prek¨ares B¨undnis’, in Benhabib et al., 9–30. Benhabib, S. (1993b): ‘Subjektivit¨at, Geschichtsschreibung und Politik’, in Benhabib et al., 105–21. Benhabib, S., Butler, J. and Cornell, D. (1993): Der Streit um Differenz. Feminismus und Postmoderne in der Gegenwart, Frankfurt am Main. Berger, H. (1994): ‘Phaedrus and the Politics of Inscription’, in Shankman, 76–114. Bernauer, J. W. (1988): ‘Beyond Life and Death: On Foucault’s Post-Auschwitz Ethic’, Philosophy Today 32, 128–42. (1990): Michel Foucault’s Force of Flight: towards an Ethics of Thought, Atlantic Highlands. Bernauer, J. W. and Mahon, M. (1994): ‘Foucaults Ethik’, Deutsche Zeitschrift f¨ur Philosophie 42, 593–608. Bernauer, J. W. and Rasmussen, D. (eds.) (1988): The Final Foucault, Cambridge/ London. Blundell, M. (1992): ‘Commentary on Reeve’s “Telling the Truth about Love: Plato’s Symposium”’, Proceedings of the Boston Colloquium of Ancient Philosophy 8, 115–33. Bolotin, D. (1989): Plato’s Dialogue on Friendship: an Interpretation of the “Lysis”, with a New Translation, Ithaca/London. Bowery, A. M. (1995): ‘Plato Visits Postmodernity: Reflection on Narrative Philosophy’, South West Philosophical Review 11, 135–42. Brieler, U. (1998): Die Unerbittlichkeit der Historizit¨at. Foucault als Historiker, Cologne. Buchwald, W. (1964): Platon.“Phaidros”, Munich. Burger, R. (1980): Plato’s “Phaedrus”, Birmingham. Burnyeat, M. F. (1980): ‘Aristotle on Learning to Be Good’, in Oksenberg-Rorty, 69–92. Bury, R. C. (ed.) (1932, reprint 1966): The “Symposium” of Plato, Cambridge. Butler, J. (1990): Gender Trouble: Feminism and the Subversion of Identity, New York/ London. (1992): ‘Contingent Foundations: Feminism and the Question of “Postmodernism”’, in Butler and Scott, 3–21.

266

references

(1994): ‘Phantasmatische Identifizierung und die Annahme des Geschlechts’, in Puhl, ¨ 101–38. Butler, J. (1993): Bodies That Matter, New York/London. Butler, J. and Scott, J. W. (eds.) (1994): Feminists Theorize the Political, New York/ London. Canick-Lindemaier, H. (1988): ‘Ehe und Liebe. Entw¨urfe griechischer Philosophen und r¨omischer Dichter’, in Siems, 232–63. Carella, M. J. (1981): Matter, Morals and Medicine: the Ancient Greek Origins of Science, Ethics and the Medical Profession, New York. Carson, A. (1990): ‘Putting Her in Her Place: Woman, Dirt, and Desire’, in Halperin et al., 135–70. Cartwright, D. (1959): Studies in Social Power, Ann Arbor. Chasseguet-Smirgel, J. (1979): Psychoanalyse der weiblichen Sexualit¨at, Frankfurt am Main. Christian, B. (1987): ‘The Race for Theory’, in Cultural Critique 6, 51–63. Cohen, D. (1991): Law, Sexuality, and Society: the Enforcement of Morals in Classical Athens, Cambridge. Connolly, W. E. (1974): The Terms of Political Discourse, Princeton. Coolidge Jr, F. P. (1993): ‘The Relation of Philosophy to Sophrosyne: Zalmoxian Medicine in Plato’s Charmides’, Ancient Philosophy 13, 23–36. Cooper, J. (1980): ‘Aristotle on Friendship’, in Oksenberg-Rorty, 301–40. Cornell, D. (1993): ‘Gender, Geschlecht und gleichwertige Rechte’, in Benhabib et al., 80–104. Dahl, R.(1961): ‘The Behavioral Approach in Political Science’, American Political Science Review 55, 763–72. (1969): ‘The Concept of Power’, in R. Bell, Wagner, R. H. and Edwards, D. V. (eds.), Political Power , New York, 79–93. Davidson, A. (1986): ‘Archeology, Genealogy, Ethics’, in Hoy, 221–34. (2001): The Emergence of Sexuality: Historical Epistemology and the Formation of Concepts, Cambridge. Davidson, D. (1967): ‘Truth and Meaning’, in Davidson (1984), 17–36. (1973): ‘Radical Interpretation’, in Davidson (1984), 125–40. (1974): ‘On the Very Idea of a Conceptual Scheme’, in Davidson (1984), 183–98. (1984): Inquiries into Truth and Interpretation, Oxford. (1985): ‘Incoherence and Irrationality’, Dialectica 39, 345–54. (1990): ‘Epistemology Externalized’, in Davidson (2001), 193–204. (1993): Der Mythos des Subjektiven, Stuttgart. (2001): Subjective, Intersubjective, Objective, Oxford. Dean, C. J. (1994): ‘The Productive Hypothesis: Foucault, Gender, and the History of Sexuality’, History and Theory 33, 271–96. Deleuze, G. (1987): Foucault, Frankfurt am Main. Deleuze, G. (1992): Foucault, Frankfurt am Main. Dennett, D. (1993): ‘Intentionale Systeme’, in Bieri, P. (ed.) (1993), Analytische Philosophie des Geistes, Bodenheim, 162–83. Derrida, J. (1972): ‘La Pharmacie de Platon’, in Derrida, J., La Diss´emination, Paris, 69–198.

references

267

(1988): ‘Choreographies’, in Derrida, J., The Ear of the Other: Autobiography, Transference, Translation, Lincoln/NA. Detel, W. (1973): ‘Zur Argumentationsstruktur im ersten Hauptteil von Platons Aretedialogen’, Archiv f¨ur Geschichte der Philosophie 55, 1–29. (1974): ‘Die Kritik an den Definitionen im zweiten Hauptteil der platonischen Aretedialoge’, Kant-Studien 66, 122–34. (1993): Aristoteles: Analytica Posteriora, translation and comments, 2 vols. (Aristote¨ les, Werke in deutscher Ubersetzung, 3.2, H. Flashar (ed.)), Berlin. (1996): ‘Foucault on Power and the Will to Knowledge’, European Journal of Philosophy 4, 296–327. (1997a): ‘Why All Animals Have a Stomach: Demonstration and Axiomatization in Aristotle’s Parts of Animals’, in Kullmann, W. and F¨ollinger, S. (eds.), Aristotelische Biologie, Bad Homburg. (1997b): ‘Ein wenig ‘Sex’ muß sein. Zum Problem der Referenz auf die Geschlechter’, Deutsche Zeitschrift f¨ur Philosophie 45, 63–98. (2003): ‘Eros and Knowledge in Plato’s Symposium’, in Detel et al. (2003). Detel, W., Becker, A. and Scholz, P. (eds.) (2003): Ideal and Culture of Knowledge in Plato, Stuttgart. Deveaux, M. (1994): ‘Feminism and Empowerment: a Critical Reading of Foucault’, in Feminist Studies 20, 223–47. Devereux, G. (1988): ‘Greek Pseudo-Homosexuality and the Greek Miracle’, Siems, A. K. (ed.) (1988), 206–31. Dews, P. (1987): The Logic of Disintegration: Post-structuralist Thought and the Claims of Critical Theory, London. Diels, H. and Kranz, W. (1964): Die Fragmente der Vorsokratiker, Z¨urich/Berlin. Dittmar, H. (1912): ‘Aischines von Sphettos’, Philologische Untersuchungen 21, 1–23. Dover, K. J. (1964): ‘Eros und Nomos’, Bulletin of the Institute of Classical Studies 10, 31–42. (1974): Greek Popular Morality in the Time of Plato and Aristotle, Oxford. (1978): Greek Homosexuality, Cambridge, Mass. Dowding, K. M. (1991): Rational Choice and Political Power, New York. Dreyfus, H. and Rabinow, P. (1983): Michel Foucault: Beyond Structuralism and Hermeneutics, Chicago. Dumm, T. (1996): Michel Foucault and the Politics of Freedom, Thousand Oaks. D¨uring, I. (1969): Der Protreptikos des Aristoteles, Frankfurt. Ebert, T. (1974): Meinung und Wissen in der Philosophie Platons, Berlin. (1994): Sokrates als Pythagoreer und die ‘anamnesis’ in Platons ‘Phaidon’, Mainz. Erdmann, E., Forst, R. and Honneth, A. (eds.) (1990): Ethos der Moderne. Foucaults Kritik der Aufkl¨arung, Frankfurt am Main. Ewald, F. and Waldenfels, B. (eds.) (1991): Spiele der Wahrheit. Michel Foucaults Denken, Frankfurt am Main. Faith, K. (1994): ‘Resistance: Lessons from Foucault and Feminism’, in Radtke, H. and Lorraine, E. (eds.) (1994), Power/Gender , Newbury Park. Fay, B. (1986): Critical Social Science, Ithaca. Ferguson, A. (1977): ‘Androgyny as an Ideal for Human Development’, in VetterlingBraggin, M., Elliston, F. and English, J. (eds.): Feminism and Philosophy, Ottawa, 45–69.

268

references

Ferguson, A. (1983): ‘On Conceiving Motherhood and Sexuality’, in Trebilcot (1983), 152–82. (1991): ‘Racial Formation, Gender, and Class in US Welfare State Capitalism’, in: Ferguson, A. Sexual Democracy, Boulder 110–30. Ferrari, G. R. F. (1992): ‘Platonic Love’, in: Kraut, R. (ed.) (1992): The Cambridge Companion to Plato, Cambridge. Festugi`ere, A. J. (1946): Aristote: Le plaisir, Paris. Fink-Eitel, H. (1978): ‘Michel Foucaults Analytik der Macht’, in Kittler, F. (ed.), Die Austreibung des Geistes aus den Geisteswissenschaften, Paderborn, 38–78. (1989): Foucault zur Einf¨uhrung, Hamburg. (1992): ‘Dialektik der Macht’, in Angehrn and Fink-Eitel, 35–56. Flashar, H. (1983): ‘Aristoteles’, in Die Philosophie der Antike, vol. iii, H. Flashar (ed.), Basel/Stuttgart, 175–458. Flax, J. (1990): Psychoanalysis, Feminism and Postmodernism in the Contemporary West, Berkeley. Flynn, T. (1981): ‘Sexuality, Knowledge and Power in the Thought of Michel Foucault’, Philosophy and Social Criticism 8, 329–48. (1988): ‘Foucault as Parrhesiast: His Last Course at the Coll`ege de France’, in Bernauer and Rasmussen, 102–18. (1991): ‘Foucault and the Spaces of History’, Monist 45, 165–86. Foucault, M. (1970): The Order of Things: an Archaeology of the Human Sciences, London. (1976): Mikrophysik der Macht, Berlin. (1982): The Archaeology of Knowledge and the Discourse on Language, New York. (1983): Der Wille zum Wissen. Sexualit¨at und Wahrheit, vol. i, Frankfurt am Main (=HS). Frank, M. (1983): Was ist Neostrukturalismus? Frankfurt am Main. Fraser, N. (1981): ‘Foucault on Modern Power: Empirical Insights and Normative Confusions’, Praxis International 1, 272–87. (1985): ‘Michel Foucault: a ‘Young Conservative’?’, Ethics 96, 165–84. (1993): ‘Pragmatismus, Feminismus und die linguistische Wende’, in Benhabib et al., 145–60. (1994): Widerspenstige Praktiken. Macht, Diskurs, Geschlecht, Frankfurt am Main. (1995): ‘From Redistribution to Recognition? Dilemmas of Justice in a “PostSocialist” Age’, New Left Review, 212, 68–93. (1996): ‘Multiculturalism and Gender Equity: the US Difference Debates Revisited’, Typoscript, 2. (1997): ‘Gluck ¨ und Glas . . . Martha Nussbaum uber ¨ die Zerbrechlichkeit des Guten im menschlichen Leben’, Philosophische Rundschau 44, 1–20. Frede, M. (1986): ‘Philosophy and Medicine in Antiquity’, in Donagan, A., Perovich Jr, A. N. and Wedin, M. V. (eds.), Human Nature and Natural Knowledge: Essays Presented to Marjorie Grene, Dordrecht, 211–32. Freud, S. (1905): ‘Drei Abhandlungen zur Sexualtheorie’ (1905), in Freud, S., Gesammelte Werke, vol. v, Frankfurt am Main (1942/repr. 1968), 27– 145. (1925): ‘Einige psychische Folgen des anatomischen Geschlechtsunterschieds’, Internationale Zeitschrift f¨ur Psychoanalyse, 11, 401–10.

references

269

¨ (1931): ‘Uber die Weibliche Sexualit¨at’, Internationale Zeitschrift f¨ur Psychoanalyse 17, 317–32. (1932): ‘Die Weiblichkeit’, in Freud, S., Gesammelte Werke, vol. xv, Neue Folge der Vorlesungen zur Einf¨uhrung in die Psychoanalyse, London (1940/repr. 1949), 119–45. Gailly, L. (1946): Philosophes et philosophie dans la com´edie gr`eque, Li`ege. Gaiser, K. (1963): Platons ungeschriebene Lehre, Stuttgart. Gallie, W. B. (1955): ‘Essentially Contested Concepts’, Proceedings of the Aristotelian Society 56, 167–98. Gardner, H. (1992): Dem Denken auf der Spur. Der Weg der Kognitionswissenschaften, Stuttgart. Gerhard, U., Jansen, M., Maihoter, A., Schmid, P. and Schultz, I. (eds.) (1990): Differenz und Gleichheit, Frankfurt am Main. Geuss, R. (1994): ‘Nietzsche and Genealogy’, European Journal of Philosophy 2, 274– 92. Godelier, M. (1987): The Making of Great Men, Cambridge. Goldstein, J. (ed.) (1994): Foucault and the Writing of History, Oxford. Gosling, J. C. P. (1973–4): ‘More Aristotelian Pleasures’, Proceedings of the Aristotelian Society 74, 15–34. Gosling, J. C. P. and Taylor, C. C. W. (1987): The Greeks on Pleasure, Oxford. Gould, T. (1963): Platonic Love, London. Graeser, A. (1993): Die Philosophie der Antike, vol. ii, Munich. Gros, F. (1996): Michel Foucault, Paris. Grosz, E. (1990): ‘A Note on Essentialism and Difference’, in Gunew, S. (ed.), Feminist Knowledge: Critique and Construct, London/New York, 147–75. Gutting, G. (1989): Michel Foucault’s Archaeology of Scientific Reason, Cambridge. (1990): ‘Foucault’s Genealogical Method’, Midwest Studies in Philosophy 15, 327– 44. Gutting, G. (ed.) (1994): The Cambridge Companion to Foucault, Cambridge. Habermas, J. (1981): ‘Hannah Arendts Begriff der Macht’, in Habermas, J., Philosophisch-Politische Profile, Frankfurt am Main, 228–48. (1985): Der Philosophische Diskurs der Moderne, Frankfurt am Main. (1986a): ‘Taking Aim at the Heart of the Present’, in Hoy, 103–08. (1986b): ‘Genealogical Writing of History: on Some Aporias in Foucault’s Theory of Power’, Canadian Journal of Political and Social Sciences 10, 1–9. Hackforth, R. (1952): Plato’s ‘Phaedrus’, commentary, Indianapolis. Hacking, I. (1979): ‘Michel Foucault’s Immature Science’, Nous 13, 39–51. (1982): ‘Biopower and the Avalanche of Printed Numbers’, Humanities in Society 5, 279–95. (1986a): ‘Self-Improvement’, in Hoy, 235–40. (1986b): ‘The Archeology of Foucault’, in Hoy, 27–40. ¨ Hadot, P. (1991): Philosophie als Lebensform. Geistige Ubungen in der Antike, Berlin. Halperin, D. M. (1990a): One Hundred Years of Homosexuality and Other Essays on Greek Love, New York. (1990b): ‘Why is Diotima a Woman? Platonic Eros and the Figuration of Gender’, in Halperin et al., 257–308.

270

references

(1992): ‘Plato and the Erotics of Narrativity’, Oxford Studies in Ancient Philosophy, Suppl. vol., 93–130. Halperin, D. M., Winkler, J. J. and Zeitlin, F. I. (eds.) (1990): Before Sexuality: the Construction of Erotic Experience in the Ancient Greek World, Princeton. Hampton, J. (1991): ‘Two Faces of Contractarian Thought’, in Vallentyne, P. (ed.), Contractarianism and Rational Choice, Cambridge, 31–55. (1993): ‘Feminist Contractarianism’, in Antony and Witt, 227–55. Hanson, A. E. (1990): ‘The Medical Writers’ Woman’, in Halperin et al., 309–38. (forthcoming): ‘Gynaikeia: Greco-Roman Gynecology’, in J. Scarborough (ed.), Ancient Medicine, Chicago. Haraway, D. (1991): Simians, Cyborgs, and Women: the Reinvention of Nature, New York/London. Harding, S. (1991): Whose Science? Whose Knowledge? Thinking about Women’s Lives, Ithaca. Harr´e, R. (1970): ‘Powers’, British Journal for the Philosophy of Science 21, 81–101. Harr´e, R. and Madden, E. (1975): Causal Powers: a Theory of Natural Necessity, Oxford. Hartsock, N. (1983): Money, Sex, and Power: towards a Feminist Historical Materialism, New York/London. (1987): ‘Rethinking Modernism: Minority vs. Majority Theories’, in Cultural Critique 7, 187–206. (1990): ‘Foucault on Power: a Theory for Women?’ in Nicholson, L. (ed.) (1993): Feminism/Postmodernism, New York, 87–109. Heitsch, E. (1993): Platon, ‘Phaidros’. Translation and commentary, G¨ottingen. Herdt, G. (1984): Ritualized Homosexuality in Melanesia, Berkeley. Herdt, G. (ed.) (1982): Rituals of Manhood, Berkeley. Hindess, B. (1996): Discourses of Power: from Hobbes to Foucault, Oxford/ Cambridge. Honneth, A. (1993): The Critique of Power: Reflective Stages in a Critical Social Theory, transl. by K. Baynes, Cambridge, Mass. (1994): ‘Das Andere der Gerechtigkeit. Habermas und die ethische Herausforderung der Postmoderne’, Deutsche Zeitschrift f¨ur Philosophie 42, 195–220. Honneth, A. and Saar, M.(eds.) (2003): Michel Foucault. Zwischenbilanz einer Rezeption, Frankfurt am Main. Horstmannshoff, F. I. (1990): ‘The Ancient Physician: Craftsman or Scientist’, Journal of the History of Medicine 45, 176–97. Hoy, D. C. (1981): ‘Power, Repression, Progress: Foucault, Lukes, and the Frankfurt School’, Triquarterly 52, 43–63. Hoy, D. C. (ed.) (1986): Foucault: A Critical Reader, Oxford. Ingram, D. (1986): ‘Foucault and the Frankfurt School: a Discourse on Nietzsche, Power and Knowledge’, Praxis International 6, 311–27. Irigaray, L. (1977): Ce sexe qui n’est pas un, Paris. (1987): Sexes et parent´es, Paris. Irwin, T. (1977): Plato’s Moral Theory, Oxford. (1981): ‘Aristotle’s Methods of Ethics’, in O’Meara, 193–224. Isaac, J. C. (1987): Power and Marxist Theory: a Realist View, Ithaca. Janicaud, D. (1991): ‘Rationalit¨at und Macht. Foucault in der Kritik von Habermas’, in Ewald, and Waldenfels, 251–76.

references

271

Keat, R. (1986): ‘The Human Body in Social Theory: Reich, Foucault, and the Repressive Hypothesis’, Radical Philosophy 42, 24–32. Keenan, T. (1987): ‘The “Paradox” of Knowledge and Power: Reading Foucault on a Bias’, Political Theory 15, 5–37. Kenny, A. (1978): The Aristotelian Ethics, Oxford. (1992): Aristotle on the Perfect Life, Oxford. ¨ Klinger, C. (1995): ‘Uber neuere Tendenzen in der Theorie der Geschlechterdifferenz’, Deutsche Zeitschrift f¨ur Philosophie 43, 801–14. Kosman, L. A. (1976): ‘Platonic Love’, in Werkmeister, W. H. (ed.), Facets of Plato’s Philosophy, Phronesis Suppl. II, Assen, 53–69. Kraut, R. (1989): Aristotle on the Human Good, Princeton. Krips, H. (1990): ‘Power and Resistance’, Philosophy and Social Science 20, 170–82. Kr¨uger, G. (1948): Einsicht und Leidenschaft, Frankfurt am Main. Kuykendall, E. (1983): ‘Toward an Ethics of Nurturance: Luce Irigaray on Mothering and Power’, in Trebilcot, 263–74. Lacey, W. K. (1968): The Family in Classical Greece, London. Landweer, H. (1994): ‘Jenseits des Geschlechts? Zum Ph¨anomen der theoretischen und politischen Fehleinsch¨atzung von Travestie und Transsexualit¨at’, in, P¨uhl, 139–67. Laqueur, T. (1992): Auf den Leib geschrieben. Die Inszenierung der Geschlechter von der Antike bis Freud, Frankfurt am Main. Laslett, B., Brenner, J. and Arat, Y. (eds.) (1995): Rethinking the Political: Gender, Resistance and the State, Chicago/London. Lasswell, H. D. and Kaplan, A. (1950): Power and Society, New Haven/London. Lawrence, J. P. (1991): Commentary on Patterson’s ‘The Ascent in Plato’s “Symposium”’, Proceedings of the Boston Colloquium of Ancient Philosophy 7, 215–25. Lerner, G. (1986): Die Entstehung des Patiarchats, Frankfurt am Main. Lieberg, G. (1958): Die Lehre von der Lust in den Ethiken des Aristoteles, Munich. Lloyd, G. E. R (1968): ‘The Role of Medical and Biological Analogies in Aristotle’s Ethics’, Phronesis 13, 68–83 (also in Mueller-Goldingen (1988), 237–52). Lloyd, G. E. R. (1983): Science, Folklore and Ideology, Cambridge. (1986): review of UP, The New York Review of Books 33 (4 March), 24–8. (1991a): Methods and Problems in Greek Science: Selected Papers, Cambridge. (1991b): ‘Who Is Attacked in “On Ancient Medicine”?’ in Lloyd (1991a), 49–69. (1991c): ‘Science and Morality in Greco-Roman Antiquity’, in Lloyd (1991a), 352–71. Luhmann, N. (1973): Macht, Stuttgart. Lukes, S. (1974): Power: a Radical View, London. Lukes, S. (ed.) (1986): Power, New York. Mahon, M. (1992): Foucault’s Nietzschean Genealogy: Truth, Power, and the Subject, Albany. Maihofer, A. (1994): ‘Geschlecht als Existenzweise. Einige kritische Anmerkungen zu aktuellen Versuchen zu einem neuen Verst¨andnis von “Geschlecht”’, in P¨uhl, 168–87. Maihofer, A. (1995): Geschlecht als Existenzweise, Frankfurt am Main.

272

references

Martin, L. H., Gutman, H. and Hutton, P. H. (eds.) (1988): Technologies of the Self: A Seminar with Michel Foucault, London. McCarthy, T. (1990): ‘The Critique of Impure Reason: Foucault and the Frankfurt School’, Political Theory 18, 437–69. (1991): Ideals and Illusions: on Reconstruction and Deconstruction in Contemporary Critical Theory, Cambridge, Mass. McNay, L. (1992): Foucault and Feminism: Power, Gender and the Self, Cambridge. Mills, C. (1956): The Power Elite, Oxford. Morarcsik, J. M. E. (ed.) (1967): Aristotle: a Collection of Critical Essays, Garden City. Mueller-Goldingen, C. (ed.) (1988): Schriften zur aristotelischen Ethik, Hildesheim. Nagl-Docekal, H. (1993): review of Butler (1991), L’Homme. Zeitschrift f¨ur Geisteswissenschaft 4, 141–9. (1996): ‘Feministische Vernunftkritik’, in Apel, K.-O. and Kettner, M. (eds.) (1996): Die eine Vernunft und die vielen Rationalit¨aten, Frankfurt am Main, 166– 205. Nagl-Docekal, H. and Pauer-Studer, H. (eds.) (1996): Politische Theorie, Differenz und Lebensqualit¨at, Frankfurt am Main. Nicholson, C. (1994): ‘Interpreting Gender’, Signs: Journal for Women in Culture and Society 20, 79–105. Nicholson, L. (ed.) (1993): Feminism/Postmodernism, New York. Nussbaum, M. (1986a): The Fragility of Goodness, Cambridge. (1986b): ‘The Speech of Alcibiades: a Reading of the Symposion’, in Nussbaum, (1986a), 165–99. (1988): ‘Nature, Function, and Capability: Aristotle on Political Distribution’, Oxford Studies in Ancient Philosophy, Suppl. vol., 145–84. (1993a): ‘Menschliches Tun und soziale Gerechtigkeit’, in Brumlik, M., and Brunkhorst, H. (eds.) (1993), Gemeinschaft und Gerechtigkeit, Frankfurt am Main, 323–61. (1993b): ‘Non-Relative Virtues: an Aristotelean Approach’, in Nussbaum and Sen, 242–69. (1995): ‘Aristotle on Human Nature and the Foundation of Ethics’, in Altham, J. and Harrison, R. (ed.) (1995), World, Mind, and Ethics, Cambridge. (1995a): ‘Feministinnen und Philosophie’, Deutsche Zeitschrift f¨ur Philosophie 43, 375–87. Nussbaum, M. and Sen, A. (eds.) (1993): The Quality of Life, Oxford. Nygren, A. (1993): Agape und Eros, Chicago (1st edn 1930). O’Neill, O. (1996): ‘Gerechtigkeit, Geschlechterdifferenz und internationale Grenzen’, in Nagl-Docekal and Pauer-Studer, 417–50. O’Meara, D. J. (ed.) (1981): Studies in Aristotle, Washington. Okin, S. (1989a): Justice, Gender, and the Family, New York. Okin, S. (1989b): ‘Reason and Feeling in Thinking about Justice’, Ethics 99, 229–49. Oksenberg-Rorty, A. (ed.) (1980): Essays on Aristotle’s Ethics, Berkeley/Los Angeles/New York. Osborne, C. (1994): Eros Unveiled: Plato and the God of Love, Oxford. Owen, G. E. L. (1967): ‘Tithenai ta Phainomena’, in Morarcsik, 167–90. Owen, G. E. L. (1971–2): ‘Aristotelian Pleasures’, Proceedings of the Aristotelian Society 72, 135–52.

references

273

Parsons, T. (1967): ‘On the Concept of Power’, in Parsons, T., Sociological Theory and Modern Society, New York/London, 297–320. Patterson, R. (1991): ‘The Ascent in Plato’s Symposium’, Proceedings of the Boston Colloquium of Ancient Philosophy 7, 193–214. Patzer, H. (1982): Die griechische Knabenliebe, Wiesbaden. Pauer-Studer, H. (1996): ‘Geschlechtergerechtigkeit, Gleichheit und Lebensqualit¨at’, in Nagl-Docekal and Pauer-Studer, 54–98. Pichot, A. (1991): Die Geburt der Wissenschaft. Von den Babyloniern zu den fr¨uhen Griechen, Frankfurt/New York/Paris. Plochmann, G. V. (1972): ‘Supporting Themes in the Symposium’, in Anton, J. and Kustas, G. (eds.), Essays in Ancient Philosophy, Albany, 328–44. ¨ Plumpe, G. and Kammler, C. (1989): ‘Wissen ist Macht. Uber die theoretische Arbeit Michel Foucaults’, Philosophische Rundschau 27, 185–218. Pohlenz, M. (1914): review of H. v. Arnim, G¨ottingische gelehrte Anzeigen 5. Pohlenz, M. (1917): ‘Nochmals Platos Lysis’, in Nachrichten von der K¨oniglichen Gesellschaft der Wissenschaften zu G¨ottingen, Philologisch-historische Klasse. (1994): Eve’s Rib: The Biological Roots of Sex Differences, New York. Poster, M. (1980): ‘Foucault’s True Discourses’, in Humanities in Society 2, 153–66. Poster, M. (1982): ‘Foucault and History’, in Sociological Research 49, 116–42. Price, A. W. (1989): Love and Friendship in Plato and Aristotle, Oxford. P¨uhl, K. (ed.) (1994): Geschlechterverh¨altnisse und Politik, Frankfurt am Main. Putnam, H. (1975): ‘The Meaning of “Meaning”’, in Gunderson, K. (ed.), Language, Mind and Knowledge, Minneapolis, 131–93. (1993): ‘Objectivity and the Science-Ethics-Distinction’, in Nussbaum and Sen, 143–57. Rabinow, P. (ed.) (1986): The Foucault Reader, Harmondsworth. Rajchman, J. (1985): Michel Foucault: the Freedom of Philosophy, New York. (1991): Truth and Errors: Foucault, Lacan and the Question of Ethics, New York/ London. Raz, J. (1994): ‘Multiculturalism: A Liberal Perspektive’, in Raz, J., Ethics in the Public Domain, Oxford, 155–75. Reeve, C. D. C. (1992): ‘Telling the Truth about Love: Plato’s Symposium’, Proceedings of the Boston Colloquium of Ancient Philosophy 8, 89–114. Reeve, C. D. C. (1992): Practices of Reason, Oxford. Ricken, F. (1976): Der Lustbegriff in der Nikomachischen Ethik, G¨ottingen. Rist, J. M. (1964): Eros and Psyche: Studies in Plato, Plotinus and Origin, Toronto. Rorty, R. (1986): ‘Foucault on Epistemology’, in Hoy, 41–50. Rosen, S. (1968): Plato’s ‘Symposium’, New Haven. Rosenberg, A. (1988): Philosophy of Social Science, Boulder. Rossetti, L. (ed.) (1992): Understanding the ‘Phaedrus’, Sankt Augustin. Roth, G. (1994): Das Gehirn und seine Wirklichkeit. Kognitive Neurophysiologie und ihre philosophischen Konsequenzen, Frankfurt am Main. Rouse, J. (1987): Knowledge and Power: toward a Political Philosophy of Science, Ithaca. (1994): ‘Power/Knowledge’, in Gutting, 92–114. Ruddick, S. (1983): ‘Maternal Thinking’, in Trebilcot, 213–30. Sawicki, J. (1994): ‘Foucault and Feminism: a Critical Reappraisal’, in Kelly, M. (ed.) (1994): Critique and Power , Cambridge, 347–64.

274

references

Saxonhouse, A. (1984): ‘Eros and the Female in Greek Political Thought: an Interpretation of Plato’s Symposium’, Political Theory 12, 5–28. Sch¨afer, T. (1995): Reflektierte Vernunft. Michel Foucaults philosophisches Projekt einer antitotalit¨aren Macht- und Wahrheitskritik, Frankfurt am Main. Scherer, C. (1993): ‘Das menschliche und das gute menschliche Leben. Martha Nussbaum u¨ ber Essentialismus und menschliche F¨ahigkeiten’, Deutsche Zeitschrift f¨ur Philosophie 41, 905–20. Schmid, W. (1990): Die Geburt der Philosophie im Garten der L¨uste. Michel Foucaults Arch¨aologie des platonischen Eros, Munich. Searle, J. R. (1996): ‘Das R¨atsel des Bewußtseins’, Lettre International 34, 34–43. Seitler, W. (ed.) (1993): Michel Foucault. Von der Subversion des Wissens. Frankfurt am Main. Sen, A. (1993): ‘Capability and Well-Being’, in Nussbaum and Sen, 30–53. Shankman, S. (ed.) (1994): Plato and Postmodernism, Glenside. Sheridan, A. (1984): Michel Foucault. Will to Truth, London/New York. Siems, A. K. (ed.) (1988): Sexualit¨at und Erotik in der Antike, Darmstadt. Simon, H. (1957): Models of Man, New York. Smart, B. (1982): ‘Foucault, Sociology, and the Problem of Human Agency’, Theory and Society 11, 121–41. (1991): ‘On the Subjects of Sexuality, Ethics and Politics in the Work of Foucault’, Boundary 2, 201–25. Smart, B. (ed.) (1995): Michel Foucault: Critical Assessments, 7 vols., London. Spelman, E. (1988): Inessential Woman, Boston. Spitzley, T. (1994): ‘Zur Rationalit¨atsannahme bei Davidson’, in W¨ustehube, A. (ed.) (1994): Pragmatische Rationalit¨atstheorien, W¨urzburg, 205–22. Still, J. (1994): ‘What Foucault Fails to Acknowledge . . . : Feminists and the History of Sexuality’, History of Human Sciences 7, 150–7. Taylor, C. (1985): ‘Connolly, Foucault, and Truth’, Political Theory 13, 377–86. (1986): ‘Foucault on Freedom and Truth’, in Hoy, 69–102. (1992): Negative Freiheit, Frankfurt am Main. Trebilcot, J. (ed.) (1983): Mothering. Essays in Feminist Theory, Ottowa. Veyne, P. (1992): Foucault. Die Revolutionierung der Geschichte, Frankfurt am Main. Visker, R. (1991): Michel Foucault, Munich. Vlastos, G. (1981): ‘The Individual as Object of Love in Plato’s Dialogues’, in Vlastos, G., Platonic Studies, Princeton 1981, 1–34. Vogel, M. (2001): Medien der Vernunft, Frankfurt am Main. Vogt, J. (1988): ‘Von der Gleichwertigkeit der Geschlechter in der b¨urgerlichen Gesellschaft der Griechen’, in Siems, 118–67. Walzer, M. (1983): ‘The Politics of Michel Foucault’, Dissent 30, 481–90. Warner, M. (1979): ‘Love, Self and Plato’s Symposium’, Philosophical Quarterly 29, 329–39. (1992): ‘Dialectical Drama: The Case of Plato’s Symposium’, Apeiron 25, 157– 75. Wartenburg, T. (1990): The Forms of Power, Philadelphia. Wehrli, F. (1951): ‘Ethik und Medizin’, Museum Helveticum 8, 36–62 (also in MuellerGoldingen (1988), 79–106).

references

275

White, F. C. (1989): ‘Love and Beauty in Plato’s Symposium’, Journal of Hellenic Studies 104, 149–57. Winkler, J. J. (1990): ‘Laying down the Law: the Oversight of Men’s Sexual Behavior in Classical Athens’, in Halperin et al., 171–210. Wolgast, E. (1980): Equality and the Rights of Women, Ithaca. Young, I. (1995): ‘Gender as Seriality: Thinking about Women as a Social Collective’, in Laslett, Brenner and Arat, 99–124.

INDEX

ability, specific (of humans), 67 abstinence, sexual, 114, 132 abuse, sexual, 135 act, sexual, 35, 130 action/activity, 13, 29, 42, 60, 63, 65, 66, 67, 68, 74, 75, 76, 88, 89, 91, 101, 122, 174, 211, 251, 253 action-theoretical model, 34 activity, sexual, 1, 61, 66, 69, 70, 71, 73, 109, 110, 111, 112, 113, 114, 115, 134, 145 adultery, 142, 147, 148 aesthetic stylisation, 3 aestheticisation, 74 aesthetics of existence, 6, 7, 59, 60, 80 akrasia, 63 Annas, J., 234, 235 anamnesis, 206, 209, 211 aphrodisia, 58, 60, 61, 62, 69, 70 apparatus, 22 appetite, 58, 120, 127, 134, 135, 145, 196, 199 Arendt, H., 13 arete, 63, 64, 101, 122, 123, 159, 210 ars erotica, 36, 37, 38, 41, 120 asceticism, 71, 170, 189 asymmetry, 16, 45, 164, 194, 195, 201 autarky, 67, 69, 72, 80, 81, 82, 85, 176, 177

badness, 64, 71, 198 Backrach, P., 12 Barnes, B., 13, 228 Benhabib, S., 231, 233, 262 bisexuality, 120 Bolotin, D., 190 boys, 122, 124, 129, 133, 135, 143, 163, 192, 200, 201 love of, 120 Butler, J., 227, 230, 231, 233, 237, 244, 247, 252, 262 Carson, A., 145, 225 character, mean state of, 63, 64 choice, 88 chresis, 59, 60 christian pastoral, 34, 40 Cohen, D., 123, 124, 130, 146, 148 concern, moral, 58, 70, 72, 119, 126, 138, 146, 148, 153, 184, 189, 192 confession non-religious, 36 practice of, 32, 36 ritual, 22, 30, 32, 33, 34, 35, 36, 39, 40, 41, 45 conflicts of interest, 49 construction, social, 223, 229, 247 continence, 59, 63, 64

276

index control, 75, 99, 105 critical theory, 13, 44 Davisdson, A., 8 Davidson, D., 56, 228, 241, 242, 244, 247, 252, 253, 254 Derrida, J., 202, 231 desire, 2, 3, 4, 6, 25, 30, 58, 61, 62, 65, 68, 69, 120, 129, 130, 132, 168, 173, 178, 181, 190, 192, 194, 196 Devereux, B., 123 dialectics, 76, 77, 78, 90, 164 diet (see regimen) dietetics, 3, 93, 94, 95, 96, 97, 99, 100, 102, 103, 104, 105, 106, 107, 109, 110, 112, 113, 116, 117, 164, 173, 175 prophylactic, 94, 96, 97, 109 therapeutic, 104 difference vs. equality, 226, 258, 260 discourse, concept of, 22, 24, 244 discursive analysis, 7, 8, 22, 23, 24, 25, 26, 27, 39, 46, 47, 49, 50, 53, 54, 56, 233, 236 disease (see illness) disgrace, 121, 130, 146, 148 divine, 61 doctrine, 75, 98, 101, 105, 107, 168, 182, 200, 202, 203, 204, 205, 206, 209, 210, 213, 216, 224 dominance/domination, 28, 29, 47, 117, 128, 156 sexual, 119 domination, model of, 3, 92, 110, 113, 136 Dover, K., 122, 123, 127, 129, 130, 136, 146, 148, 225 Ebert, T., 203, 209, 213, 215 economics, 4, 29, 149, 150, 152, 153, 154, 155, 160, 161, 164 education, 44, 78, 87, 143, 171, 174, 179, 193, 218 emancipation, 231, 236, 237, 240 enkrateia, 59, 60, 63, 64, 73, 75, 79, 92, 125, 127, 129 e´nonc´es, 23, 25, 42 enslavement (see also slave), 59, 62, 63, 125 episteme, 148, 149, 169, 183, 184 epistemology, Platonic, 201, 206 epithymia, 61, 62, 65 equilibrium, physiological, 97 eros, 194, 200

277

cosmic, as a cosmic power, 170 and its destruction, 177 divine and earthly, 171 as an intermediate being, 181 and order, 172, 173, 174, 175, 176, 181, 220 Platonic, 200, 201 Plato’s theory of, 4, 168, 181 and politics (in Plato), 220 and reciprocal relationships, 81, 177, 179 and well-being, 102 winged, 198 work (ergon) of, 176, 177, 181 erotics, 136, 139, 163, 164, 165, 169, 172, 173, 175, 176, 179, 184, 188, 190, 193, 196, 199, 200, 201, 202, 209, 218, 221 erotics, philosophical, 172, 179 Platonic, 163, 165, 189, 196, 200, 201 essentialism, 227, 230, 234, 235, 236, 246, 247 estate management, 138, 148, 149, 152, 154 eternal/eternity, 61, 156, 165, 177, 181 ethical work, 53, 58, 59, 60, 63, 64, 70, 73, 74, 78, 81, 84, 85, 88, 89, 90, 91, 92, 164, 182, 184, 220, 222 ethics in Foucault, 6, 7 and medicine, 94, 99, 100, 101 relationship to archaeology and genealogy, 7, 9 eudaimon, 90, 91 eudaimonia, 65, 66, 68, 69, 70, 73, 74, 75, 78, 79, 80, 81, 82, 83, 84, 87, 88, 89, 90, 91, 92, 102, 103, 107, 108, 113, 116 excess, sexual (see lasciviousness) Fay, B., 13 feminism, 4, 224, 225, 226, 230, 231, 232, 233, 234, 236, 237, 239, 244, 247, 251, 252, 253, 259, 262 difference feminism, equality, 226 and Foucault, 224 Ferrari, G., 171, 178, 184 Fink-Eitel, H., 12, 13 fidelity, sexual, 137

278

index

finality, 179, 181, 182 Flax, J., 231, 262 Flynn, T., 8 force conditions, 17 formation, 8, 10, 15, 25, 27, 50, 51, 52, 85, 87, 88, 89, 94, 97, 107, 109, 117, 151, 216, 224 forms, 2, 6, 8, 9, 10, 13, 14, 15, 18, 19, 22, 24, 25, 26, 27, 29, 30, 35, 36, 37, 40, 41, 42, 45, 51, 52, 54, 56, 58, 60, 66, 73, 81, 85, 86, 89, 90, 92, 95, 96, 102, 109, 117, 118, 133, 134, 135, 154, 168, 174, 177, 181, 182, 185, 189, 195, 198, 202, 203, 204, 205, 208, 209, 210, 212, 214, 215, 216, 220, 222, 234, 238, 240 theory of, 189 vision of, 167, 187, 203 Fraser, N., 226, 262 freeborn and slaves, 4, 119, 135 freedom (see also liberty), 6, 46, 59, 62, 63, 83, 142, 162, 238 sexual, 142 friends, 81, 166, 172, 194 friendship, 76, 81, 91, 121, 133, 134, 170, 171, 190, 194, 195, 199, 219 games of truth, 9, 28, 50, 51, 52, 53, 54, 60, 75, 76, 77, 78, 85, 90, 91, 106, 116, 136, 164, 168, 201, 203, 204, 205, 206, 209, 218, 222, 239 gender, 223, 224, 226, 227, 229, 231, 232, 235, 237, 240, 247, 249, 251, 256, 257, 258 God’s eyes, 54, 253 good, the, 136, 165, 168, 181, 187, 194 theory of, 68, 162, 166, 168, 181, 182, 217, 220, 221 good life, 3, 23, 25, 36, 51, 52, 60, 64, 65, 66, 67, 70, 72, 73, 74, 78, 79, 81, 82, 83, 87, 94, 95, 103, 104, 109, 117, 129, 136, 155, 235, 260 government (see also rule), 26, 52, 118 Gutting, G., 10, 55 Habermas, J., 7, 13, 228 habit, 86, 87, 88, 89, 92 Hacking, I., 10 Halperin, D., 200 happiness, 51, 52, 53, 65, 67, 68, 69, 70, 71, 72, 78, 79, 191, 193, 201

Haraway, D., 232, 246 Harding, S., 261 Hartsock, N., 233, 261 health, 69, 95, 98, 99, 100, 102, 103, 107, 108, 111, 113, 117, 145, 174 health care, ancient system of, 96 hedone, 61, 65 Heitsch, E., 123, 191, 193, 197, 199 heterosexuality, 120, 146, 248 hexis, 59, 71 hiccups (Aristophanes’), 175 Hobbesian problem of social and political order, 45 homoerotic relationships, 4, 159, 162, 192 homoeroticism, 4, 120, 135, 136 homosexuality, 120, 123, 133 Honneth, A., 47, 48, 256 honour, 4, 68, 80, 104, 121, 125, 127, 128, 130, 136, 146, 148, 151, 163 hoplites, 124, 125, 126, 128 human sciences, 10, 85 husband, 4, 118, 138, 143, 146, 148, 150, 151, 153, 156, 157, 161 ideology critique, 252, 253 illness, 98, 102, 104, 105, 106, 108, 111, 114, 115 immortality, 141, 168, 177, 179, 184, 209, 213 imperative of truthfulness, 34, 36 incontinence, 63, 138 institutions, 13, 20, 41, 46, 48, 49, 97, 229, 243, 248 interpretation, radical, 241, 242, 243, 244, 256 interpretation theory, 241, 242 inviolability, physical, 125, 130, 135 Irigaray, L., 257 Isaac, J., 12 justice, 83, 91, 154, 155 kakia, 64 kinaidos, 124, 125, 126, 128 Klinger, C., 226, 231, 233 kyrieia, 139, 146 kyrios, 138, 139, 140, 141, 142, 143, 146, 147, 149, 150, 151, 152, 153, 154 Kruger, ¨ G., 201

index Lacey, W., 143 ladder, erotic, 167, 168, 182, 183, 184, 198, 217 lasciviousness, 114, 124, 128, 131 Landweer, H., 258 Laqueur, T., 227, 229 leadership, 4, 151, 152, 153, 154, 164 Lerner, G., 226 liberty (see freedom) life contemplative, 79 flourishing and happy, 81, 157 model of, 83, 89, 180, 186, 200 political, 80 logos, 59, 62, 75, 81, 88 love, 36, 58, 84, 114, 119, 120, 130, 131, 134, 147, 153, 157, 163, 164, 165, 166, 167, 169, 170, 171, 172, 176, 177, 179, 181, 182, 183, 185, 186, 190, 194, 195, 196, 200, 201, 209, 220 human, 181 individual, 176, 186, 220 Platonic, 164, 185 true, 163, 164, 200, 201 love-struck, in love, 189 lover, 4, 120, 121, 129, 131, 133, 134, 136, 170, 171, 172, 187, 189, 192, 193, 194, 200 luck, 166 Luckes, S., 13 Maihofer, B., 226, 227, 228, 232 male and female, 5, 126, 129, 155, 158, 223, 225 reference to, 232 mania, 178, 190, 196, 197 manliness, 124, 126, 128, 130, 134, 135, 192, 193 marriage/marital relationships, 4, 119, 132, 137, 138, 139, 142, 143, 151, 153, 155, 156, 157, 158, 159, 160, 161, 163, 192 mastery, 37 maturity, moral, 88, 89 McNay, L., 224 medicine, ancient, 3, 93, 94, 97, 99, 112, 116 medicine and philosophy, 62, 98, 101 men and women, equality of, 159 Mills, C., 11

279

misogyny, ancient, 126, 135 moderation (see also temperance), 59, 63, 75, 118, 137, 148, 153, 156, 157, 200 sexual, 118, 137, 148, 156, 157 moral subject, 77, 88 myth of subjectivity, 34, 40 Nagl-Docekal, H., 233, 234, 250, 261 nature, human, 65, 235 neutrality, 30, 53, 55, 57 Nicholson, L., 226, 227, 229, 230 norms, 9, 33, 48, 227, 231, 238, 244, 248, 250 Nussbaum, M., 165, 166, 167, 176, 180, 186, 196, 197, 200, 233, 234, 235, 260 Nygren, A., 196 object, 9, 50, 51, 58, 64, 120, 134, 164, 165, 181, 183, 185, 190, 191 of desire, 120, 122 oikos, 4, 119, 138, 139, 140, 141, 142, 143, 146, 147, 148, 149, 150, 151, 152, 153, 157, 160 O’Neill, O., 259 ontology, Plato’s later, 217 orators (see rhetores) order, 172, 173, 174, 175, 176, 181, 220 Osborne, C., 181, 196 paideia, 78, 87 passivity, sexual, 59 Patterson, R., 183 Pauer-Studer, H., 260 pederastic relationships, 3, 4, 119, 133, 134, 135, 136, 139, 174, 177 perception (in Plato), 206, 207, 208, 209, 212, 213 performative model, 33, 35 phronesis, 59, 62, 74, 76 pleasure, 37, 58, 59, 61, 62, 63, 64, 65, 66, 68, 69, 70, 71, 73, 81, 87, 99, 110, 112, 117, 122, 126, 131, 133, 134, 136, 172, 174, 192, 198, 199, 201 pleasure, forms of, 64, 65, 66, 69, 70, 71 pleasure, object of, 193 Pohlenz, M., 190 police science, 52, 53 polis, 3, 53, 76, 78, 79, 82, 84, 89, 108, 119, 124, 126, 128, 132, 135, 139, 140, 142, 143, 147, 149, 157, 160, 162, 220, 221, 222

280

index

political career, 51, 79, 82, 118, 131, 134, 156, 193, 221 political rationality, 52 pouvoir, 26 power analysis of, 2, 6, 7, 8, 9, 10, 16, 19, 20, 22, 24, 25, 27, 35, 43, 44, 45, 46, 47, 53, 54, 56, 230, 233, 236, 239, 262 basic concept of, 15, 16, 43 contingency and intelligibility, 27 forms of, 10, 17, 19, 20, 21, 23, 24, 25, 30, 31, 42, 45, 46, 48 and individuals, 8, 10, 12, 16, 19, 20, 22, 25, 26, 27, 31, 43, 50, 52, 53, 85, 89, 109, 179, 185, 186, 212, 238, 239, 244 and knowledge, 2, 10, 11, 26, 27, 28, 29, 30, 33, 35, 36, 46, 135, 239 levels of, 19, 50, 151, 251 means of, 44 positive, 15, 137 productive, 11, 15, 32, 45, 46, 90, 117, 238, 241 regulative, 38, 42, 43, 44, 45, 46, 49, 51, 89, 90, 92, 117, 119, 161, 174, 222, 238, 239, 240, 250, 251 relations of, 25, 153 repressive, 14, 15, 27, 28, 35, 36, 44, 137, 228 and subject, 18, 22, 26 theories of, 11, 12, 14, 16, 44 practice, erotic, 183, 184 problematisation, moral, 58 property, 139, 152 prudence, 59, 60, 62, 75, 109, 155, 162, 197, 200 psychoanalysis, 6, 51 public speakers (see rhetores) Putnam, H., 234, 246 rationality, 5, 18, 38, 41, 52, 54, 56, 194, 229, 239, 250, 253, 261 reason, 2, 5, 8, 41, 54, 56, 58, 59, 62, 67, 68, 69, 70, 73, 74, 75, 76, 88, 134, 154, 155, 174, 178, 181, 187, 197, 228, 234, 239, 241 reason-essentialism, 261 reciprocity, 16, 135, 157, 194 recollection, 203 reference, causal, 247 regimen, 93, 95, 96, 98, 102, 103, 106, 107, 109, 110, 111, 112, 113, 116 excessive, 96, 104

regulation, 18, 31, 37, 48 relations of force, 45, 46, 49 relativism, 44, 55, 56, 57, 229, 235, 252, 262 restriction, model of, 64, 73, 200 rhetores, 3, 119, 127, 135, 169 Rist, J., 196 Rorty, R., 10 Rosen, S., 169, 170, 172, 174, 177, 178 Rosenberg, A., 242, 246 Roth, G., 228 Rouse, J., 21, 29, 47, 55 rule (see also government), 10, 59, 60, 75, 79, 83, 151, 153, 154, 155, 162 ruled, 83 ruler, 83 satisfaction physical, 132, 172 sexual, 171 savoir, 26, 42, 45 Saxonhouse, A., 180 Sch¨afer, T., 7, 44, 55 Scherer, C., 234, 236 scientia sexualis, 22, 30, 36, 37, 38, 39, 40, 41, 45 Searle, J., 43, 228 self, practice of (see also ethical work), 64 self-control, 62, 63, 64, 78, 79, 88, 89, 94, 100, 107, 109, 118, 122, 125, 126, 131, 134, 146, 149, 152, 161, 200, 221, 222 self-discipline, 119, 125, 126, 130, 138, 150, 161 self-interest, reasonable, 74 self-restraint, 64 self-transformation, 51, 52 Sen, A., 234 sensuality, 178, 198, 199 sex, 22, 31, 38, 39, 41, 65, 127, 142, 190, 200 biological, 223, 224, 225, 226, 227, 229, 230, 231, 234, 235, 236, 244, 245, 247, 248, 249, 250, 251, 256, 257, 262 sexuality, 1, 5, 7, 8, 23, 29, 30, 38, 39, 59, 60, 64, 71, 72, 93, 109, 110, 111, 113, 114, 116, 118, 119, 124, 135, 136, 145, 146, 147, 157, 161, 162, 172, 200, 222, 223, 256 dangers of, 113 dynamis of, 111

index reciprocal, 120 relational, 119 shame (see disgrace) slavery/slaves (see also enslavement), 63, 118, 125, 131, 154, 210 social constructivism, 228, 230, 241, 252 society, theory of, 13, 20, 47, 53 sophrosyne, 59, 60, 62, 63, 64, 75, 76, 79 soul, 33, 34, 62, 67, 68, 75, 93, 98, 99, 100, 101, 111, 117, 123, 164, 168, 172, 173, 174, 176, 178, 182, 184, 197, 199, 203, 206, 207, 209, 211, 213, 215, 218, 221 chariot of, 198, 199, 200, 206 striving, erotic, 182, 184, 188 reasoned, 178 stylisation, ethical, 74 of existence, 74 subject, 9, 50, 51, 58, 59, 85, 88, 110, 230, 237, 240 creation of, 2 and power, 18, 22, 26 subjectivisation, 9, 10, 26, 27, 50, 51, 59, 78, 80, 85, 89, 95, 103, 107, 109, 116, 117, 151, 262 subjugation, 10, 26, 27, 33, 34, 41, 42, 44, 45, 50, 59, 64, 72, 74, 85, 95, 125, 200 submission, 109 substance, ethical, 58, 71 Taylor, C., 7, 47, 55 teaching, 78, 88, 89, 92, 149, 150, 153 technique of existence, 94 technology, political (of the self), 52 teleology ethical, 59, 61, 72, 74, 76, 78, 82, 83, 84, 85 and politics, 82, 180

281

temperance (see also moderation), 59, 63, 64, 72, 154 transcend/transcendence, 184 transsexuality, 258 transvestitism, 258 truth, 5, 7, 8, 9, 28, 30, 33, 35, 36, 41, 48, 50, 51, 55, 67, 76, 136, 163, 164, 167, 169, 176, 178, 185, 187, 188, 198, 201, 205, 206, 208, 221, 222 truth, concept of (in Foucault), 9, 39 production of, 26, 33, 34, 41, 42 reference to, 218 relation to, 164, 167, 187, 201 relationship to, 40 universalism, 54, 55, 253, 262 virtue, 63, 64, 66, 75, 83, 86, 95, 122, 144, 154, 164, 170, 171, 172, 174, 178, 179, 185, 187, 193, 201, 210, 219 Vlastos, G., 165, 167, 196 Vogel, M., 250 Warner, M., 169 Wartenburg, T., 12, 13, 14, 15, 21, 222 well-being, 102 wife, 4, 118, 119, 137, 138, 139, 142, 143, 146, 147, 148, 149, 150, 151, 152, 154, 156, 157, 158, 159, 160 Winkler, J., 124, 125, 126, 128 Wolgast, E., 234 women discrimination against, 135 inferiority of, 139, 143, 144, 146 liberties of, 147, 148 nature of, 145 subordination of, 139, 140, 144, 146, 245