Bryn Mawr Classical Review 02.01.18

John J. Winkler, The Constraints of Desire. The Anthropology of Sex and Gender in Ancient Greece. New York & London: Routledge, 1990. Pp. x + 269. ISBN 0-4159-0122-7 (hb). ISBN 0-4159-0123-5 (pb).

Reviewed by A.M. Keith, University of Toronto.

The publication of this thoughtful and thought-provoking collection of studies makes an important contribution to debates in several fields of classical study: gender and culture studies, literary interpretation and curriculum debate, social and religious history. The book falls into two sections with a central "Interlude": the first section, entitled "Andres", is androcentric in its examination of the protocols which constrain male behaviour in ancient society, while the second section, "Gunaikes", is gynocentric in focus. Winkler's interest in stimulating dialogue informs each individual study and is reflected in the very organisation of the collection, since the two sections pivot around the mediating "Interlude". Indeed, the single most exciting aspect of this book lies in its invitation to the reader to generate further discussion. Throughout, Winkler's prose style engages his reader as an active partner in the narrative exposition of his argument.

The introduction raises a number of theoretical issues which recur in subsequent chapters. Taking as his starting point a personal anecdote, Winkler proceeds to a scrupulous accounting of his own relationship to the material under study in a discussion that problematises the notion of scholarly objectivity as it reinstates it. Such a procedure owes a considerable debt to the feminist credo that the personal is political, and indeed, an important stratum of scholarship in Winkler's study is feminist anthropological theory. Thus he suggests that the task of the scholar is to "learn to see the various kinds of spin and misdirection that qualify the meaning of ... pronouncements [concerning sex and gender] in their full social context, the unspoken stage directions that are understood but not voiced by the social actor" (p. 5). Such an approach -- requiring that attention be paid to both text and the circumstances of textual production -- aligns Winkler within the framework of post-modern debate about sexuality and gender, and points, in particular, to Foucault's pronouncement that in studying a culture's sexual protocols, the silences may be just as important as (and possibly even more important than) the public statements.1 For this reason, Winkler tackles not only texts which explicitly treat issues of sex and gender, but ranges broadly over areas usually conceptualised by Classicists as widely disparate fields: "ancient Greek oratory, vases, politics, magic, poetry and religious rites" (p. 10). Winkler makes a cogent and compelling case for a holistic examination of materials "that together ... reproduce the variety of perspectives and experiences which single approaches tend to simplify and iron out" (p. 10).

The first three chapters -- which constitute Part I, "Andres" -- investigate the cultural construction of manhood and especially male sexuality in the ancient Greek world. Winkler is concerned to ask what kinds of texts and what parts of these texts we read in our attempts to interpret ancient attitudes towards sex and gender. These questions invite us to consider not only the limits which our own ideological positions impose on this canon, but also the limits imposed by the body of texts which accidentally survive from antiquity.2 Indeed, Winkler is concerned to interrogate the texts themselves as representative of a peculiarly narrow ideological position:

It cannot be said too strongly or too frequently that the selection of book-texts now available to us does not represent Greek society as a whole. The social and editorial conventions within which most public speaking and published writing took place tended to give voice to a select group of adult male citizens and to mute the others -- female, adolescent, demotic (working persons with a minimum of leisure), metic (non-citizen). (p. 19)3
Thus Chapter One, "Unnatural Acts", rejects the privileged position of philosophy in "reconstructing a picture of ancient society" (p. 19), and instead employs the evidence provided by Artemidoros' Dream Analysis to generate a broad picture of the varying social meanings of male sexual acts in antiquity. Chapter Two, "Laying down the Law",4 traces the characteristics of the citizen hoplite and the kinaidos, whom Winkler identifies as the two opposing poles within which Athenian upper class male sexuality was constructed and against which manhood was measured over the period 430-330 BCE. Of fundamental importance to Winkler's demonstration is the potency of gossip and oratory as instruments of social control and political manoeuvering, and in these findings he is in broad agreement with D. Halperin's essay "The Democratic Body".5 The final chapter of this section, "The Constraints of Desire", proposes a psychoanalytic interpretation of ancient erotic magic spells.6 Winkler suggests that such spells tell us considerably more about the psychology of the (mostly male) magic user than they can about the psychological state of the (mostly female) victim, though he notes that they do indeed provide evidence in an indirect way of the social constraints exercised over the (usually female) victim of such spells.

The mediating Interlude, entitled "The Education of Chloe", is structured as a dialogue between Longus' second century CE novel Daphnis and Chloe and Winkler himself in the role of "a visiting anthropologist, who notices problems which native experiences raise without directly addressing" (p. 104).7 He then proceeds to read the novel as an account of "the pain of sexual acculturation" (104), focusing on the violence associated with Chloe's sexual initiation. I was particularly impressed by Winkler's insight into the pedagogic strategies of ancient literature propounded in a brief interpretation of the myth of Pan and Syrinx as it is related and re-enacted in a festival at D&C 2. 34-35 (pp. 119-120). Noting that ancient literature has the pedagogical function of promoting and licencing violence directed towards women in the patriarchal societies of the ancient Greco-Roman world, Winkler argues that narrative reinscribes culturally sanctioned violence against women even as it denies and conceals it. Such a reading takes narrative seriously -- as humanities departments and university administrators all too rarely do -- as a powerful ally in the cultural construction of patriarchy.8 Winkler's conclusion to this chapter faces the issue squarely, and deserves quotation in full:

And what does it finally mean? I find it hard to determine whether the well-concealed Longus had a fundamentally patriarchal attitude to Chloe -- that she is to be simultaneously protected and made to undergo a painful rite of passage -- or the more critical stance I have outlined here. The former reading is implied by most of the modern critics who have noticed the violence at all... But the larger methodological issue is whether readers should simply be trying to reproduce the author's meaning (if he had one -- that is, if he had one) as the goal... If our critical faculties are placed solely in the service of recovering and reanimating an author's meaning, then we have already committed ourselves to the premises and protocols of the past -- past structures of cultural violence and their descendants in the bedrooms and mean streets and school curricula of the present. This above all we must not do. (p. 126; cf. p. 19)
Chapter Five is devoted to elucidating the central role played by Penelope in the plot of Odysseus' homecoming, which Winkler finds to be "stronger and more cunning ... than is often attributed to her" (p. 130). He suggests that Penelope's role in Odyssey 17-23 be interpreted in the light of evidence provided by anthropological studies of pre-industrial communities, which emphasise the constraints upon both male and female discourse as well as the constructive public use to which lies and deception are put in a scarcity economy (p. 134). Sensitive to the dynamics of Odysseus' disrupted household and the resulting restrictions which limit Penelope's manoeuvrability, Winkler argues that she is constrained by the presence within the great hall of the untrustworthy serving-women from explicitly admitting that she entertains the possibility that the beggar is Odysseus. This anthropological analysis of Penelope's actions is complemented by a narratological reading which makes use of modern theoretical discussion of narrative "secrecy".9 Winkler suggests that Penelope colludes with the Homeric narrator in the interview scene of Odyssey 19 to focus our attention on Odysseus' tests of family and retainers, with the result that the famous test of the bed in Odyssey 23 effectively tricks not only Odysseus but also the poem's audience (p. 156). Thus Odysseus polumêtis, the master-trickster ... is tricked -- not for the first time -- and so in a sense are we. For at that moment we realize that the entire telling has been one-sided, slanted in favor of Odysseus and his enterprises. (p. 158) These remarks lead, in the conclusion of this chapter, to a discussion of the phil-Odyssean stance of the poem (pp. 158-159; cf. 143),10 and the suggestion that a broadly androcentic bias, characteristic not only of the Odyssey but also of many subsequent readings of the Odyssey, has promoted inquiry into the behaviour of male agents even as it directs our gaze away from the actions of female characters.

The sixth chapter is a revised and expanded version of Winkler's previously published study, "Garden of Nymphs: Public and Private in Sappho's Lyrics".11 The newly revised examination of Sappho's poetry includes greater theoretical discussion in a lengthened introduction, and an interesting analysis of the sound effects of Sappho's poetry. It remains one of the most exciting interpretations of the nature of Sapphic poetry. The final chapter, "The Laughter of the Oppressed: Demeter and the Gardens of Adonis", reviews the ancient evidence about the "women's-only" [sic] festivals assembled by M. Detienne.12 Winkler identifies a "masculinist vision" (p. 199) in Detienne's study and rejects such unself-consciously phallocentric criticism. Instead Winkler undertakes to reconstruct a multiplicity of possible female experiences at these festivals, and his suggestions are always illuminating. Winkler consistently displays a great sensitivity to feminist politics and explicitly articulates his concerns in this area toward the conclusion of the chapter.

The resulting account may, I fear, still be overly preoccupied with phallic issues of interest to men: instead of claiming that "phallic men are central," as Detienne's account does, mine claims that "phallic men are peripheral and their pretensions amusing." In both cases the focus is on men. And, in a sense, the energy of this essay has been directed as much or more towards Detienne as towards Demeter, and it may appear to some to be an undignified male squabble rather than the feminist exploration we would really like. But each scholar must contribute what he or she can to the corporate enterprise. (p. 206)13

It seems to me that Winkler has indeed made a substantial contribution to Classical Studies -- without in any way compromising his feminist principles -- in a book which all Classicists, whatever their particular field of study, will find both useful and interesting. Winkler's commitment to uniting an interpretation of what is said about sex and gender with an understanding of how all such utterances are culturally constrained, consistently generates new and provocative readings even of much-discussed material. This exciting book will set the standard for debate in many areas of study and will undoubtedly stimulate a great deal of discussion in the future.


  • [1] P. Bourdieu, Outline of a Theory of Practice, translated by R. Nice (Cambridge, 1977) [originally published in French as Esquisse d'une théorie de la pratique (Switzerland, 1972)]; M. Foucault, The History of Sexuality, 2 volumes, translated by R. Hurley (New York, 1978 & 1985) [originally published in French as La volenté de savoir (Paris, 1976) and L'Usage des plaisirs (Paris, 1984).] On the significance of silence, see Foucault, History of Sexuality I, p. 27.
    Classicists have been quick to join in the debate: see P. Brown, The Body and Society: Men, Women and Sexual Renunciation in Early Christianity (New York, 1988); D. M. Halperin, One Hundred Years of Homosexuality and Other Essays on Greek Love (New York, 1989); D.M. Halperin, J.J. Winkler, and F.I. Zeitlin (edd.), Before Sexuality (Princeton, 1990); A. Richlin, The Garden of Priapus (New Haven, 1983); and A. Rousselle, Porneia: On desire and the body in antiquity, translated by F. Pheasant (Oxford, 1988) [originally published in French as Porneia (Paris, 1983)].
  • [2] The heated canon-debate which has polarised other literature departments in the Academy in the course of the 1980's has barely begun in the field of Classics, and will receive a welcome spur from Winkler's study. The problems involved in distinguishing competing ideological positions in the study of literature are well set out by T. Eagleton, Literary Theory (Minneapolis, 1983), pp. 17-53 and 194-217, and are pursued further in W. J. T. Mitchell (ed.), The Politics of Interpretation (Chicago, 1983).
  • [3] On "muted group" theory and its relevance to gender studies, see S. Ardener (ed.), Perceiving Women (London, 1975); M. Crawford & R. Chaffin, "The Reader's Construction of Meaning: Cognitive Research on Gender and Comprehension," in P. P. Schweickart & E. A. Flynn (edd.), Gender and Reading (Baltimore, 1986), pp. 3-30; and C. Kamarae, Women and Men Speaking (Rowley, Mass., 1981).
  • [4] A version of this study appears in Halperin-Winkler-Zeitlin (1990) at pp. 171 -209.
  • [5] Halperin (1989) 99-112; see now also the forthcoming study of V. J. Hunter, "Gossip and the Politics of Reputation in Classical Athens", in Phoenix 45.1 (1991).
  • [6] A shorter version of this study will appear in C. A. Faraone & D. Obbink (edd.), Magika Hiera (New York, forthcoming).
  • [7] A shorter version of this study will appear in B. Silver & L. Higgins (edd.), Rape and Representation (New York, forthcoming).
  • [8] Cf. Chapter Six, pp. 164-165. On the prescriptive pedagogical use of classical texts in the post-classical world, see further M. Homans, "Feminist Criticism & Theory: The Ghost of Creusa" Yale Journal of Criticism 1.1 (1987) 153-182; and cf. D. Pope, "Notes toward a Supreme Fiction: The Work of Feminist Criticism" in Women and a New Academy, edited by J.F. O'Barr (Madison 1989), 22-37.
  • [9] See, for example, F. Kermode on narrative "secrecy", in The Genesis of Secrecy: on the Interpretation of Narrative (Cambridge Mass., 1979), pp.15, 23-47, and idem, The Art of Telling (Cambridge Mass., 1983) pp. 133-155.
  • [10] Cf. J.S. Clay, The Wrath of Athena (Princeton, 1983), pp. 34-38.
  • [11] H.P. Foley (ed.), Reflections of Women in Antiquity (New York, 1981), pp. 63-89.
  • [12] M. Detienne, The Gardens of Adonis: Spices in Greek Mythology, translated by J . Lloyd (Atlantic Highlands, N.J., 1977) [originally published in French as Les jardins d'Adonis: La mythologie des aromates en Grèce (Paris, 1972)].
  • [13] D.M. Halperin, in a discussion of the relevance of Diotima's gender to the doctrine Plato has her propound in the Symposium [Halperin (1989) 113-151 = Halperin-Winkler-Zeitlin (1990) 257-308], concludes with a similar (though lengthier) critique of male scholarly appropriation of feminist theory: see Halperin (1989) 149-151 [= Halperin-Winkler-Zeitlin (1990) 294-298].