Active/Passive, Acts/Passions: Greek and Roman Sexualities

Greek and Roman antiquity once impinged on the consciousness of modern historians for the models they provided of democracy and empire, political theory and military strategy. Today, if modern historians think about ancient Greece and Rome, they are quite likely to do so in the context of the history of sexuality. One reason for the relatively high visibility of the ancient world in this subdiscipline is the apparent Greek acceptance of (some) male same-sex relations, which makes such an apparent contrast with the contemporary situation. Since 1978, when Kenneth Dover broke the taboo on discussion of ancient same-sex relations in polite scholarly circles with Greek Homosexuality, scholarly works have followed one another fast and furiously.[1] The level of discourse in this discussion has been both extremely scholarly—whole theories often turning on the interpretation of certain words or individual passages in fairly arcane documents—and extremely vehement, revealing that these are issues about which scholars care even more than most historians usually care about their subjects.[2] The reasons are not far to seek: scholars’ approaches to issues of gender and sexuality often have real-world political antecedents or ramifications, which they see no reason to hide.

Recent scholarship on ancient sexualities (or the lack of ancient sexualities, since some scholars deny the applicability of “sexuality” to the period) has reached consensus, if not complete agreement, on two main points: the social construction of sexuality (that “sexuality” is not a thing that can be found in all cultures but is created by the various discourses of particular societies), and the active/passive dichotomy (that the ancient world, both Greek and Roman, categorized sexual behaviors or identities not by the gender of the participants but by the sexual role each played). These ideas are still relatively novel, especially to nonspecialists, and some of the works discussed here are concerned with establishing them; other works, however, take these two points as the new orthodoxy to be demolished.

The concept “the ancient world” is problematic in itself, implying as it does that cultures from archaic Greece to imperial Rome may be lumped together. These cultures do seem to share some fundamental features with regard to sexuality, notably a focus on activity versus passivity. These features, however, characterize many other pre-modern and modern Mediterranean (and other) cultures as well. Other differences—a greater acceptance of pederasty by Athenians, a lower valuation of moderation among Romans—depend a great deal on what sort of texts one reads. The books under review all speak to particular historical moments and groups of texts, and most do not make claims about ancient culture generally. Because my purpose is largely to identify trends in classical scholarship that may interest historians of other periods, I have not attempted to build on these individual studies a broad, comparative overview of Greek and Roman sexualities. After discussing some of the theoretical issues that currently exercise the field, I will move roughly chronologically, dealing first with two books on Greece, then two on the Hellenistic/Roman world.

Although the theoretical position that sexuality is socially constructed has become dogma in the field, much scholarly writing still presents it as a new and radical idea. This is in part because it is counterintuitive and has had little effect on attitudes outside the academy. Fundamentally, social constructionism argues that the categories in which we think about sexuality—like homosexuality and heterosexuality—are not universal but are a creation of our culture.[3] Its opposite, called by social constructionists “essentialism,” argues that there are fundamentally different kinds of people, that in every culture there are those with homosexual, heterosexual, and various other orientations—although of course different societies may think about them and treat them quite differently. The popular lists of “famous gays in history” are essentialist in conception. Although humanities scholarship in the 1990s has almost universally rejected this view,[4] such essentialism is implicit in the contemporary search for genetic markers or biological corollaries of a predisposition to homosexuality. It is also congenial to many gay activists who believe that society will be more tolerant if it understands homosexuality as something inborn, not chosen. Social construction does not imply that individuals choose their own identities—it is the discourses of the broader culture, for example medical, legal, or religious systems, that construct systems of sexual identities—but nevertheless it is often misunderstood as implying that these identities are not “real.”

It is helpful to distinguish two versions of the social constructionist argument.[5] The more moderate of the two states that the meanings that people put on sex acts or desires are different in different cultures, that we cannot assume that the categories in other times and cultures are the same as ours (that a man who has sex with both men and women is “bisexual,” for example), and that as historians we must examine the attitudes and mentalities of any given society to see how that particular society constructed sexual identities. It may be the case that we find roles (active or passive) more important than object choices, but the opposite may also be true. It may be the case that we find no deeply felt personal identity based on sexual preference, but the opposite may also be true. While we cannot assume congruence with modern categories, neither can we assume dissonance with them.

The stronger version of social constructionism, relying on the insights of Michel Foucault, states that not only the particular categories familiar to us but also the very notion of a sexual orientation are creations of bourgeois capitalism.[6] Only in nineteenth-century Europe and North America did people come to view their sexual preferences as part of what constituted them as individuals. People in other societies may have had preferences for a particular type of partner, role, or act, but these preferences did not define them as a type of person. An “identity” based on sexuality was a categorization of convenience in earlier eras but was not psychically deep. The founding exponents of this school of thought include Jeffrey Weeks for the nineteenth century and David Halperin for ancient Greece.

It may be impossible to delve far enough into the psyches of ancient people to understand what sexual subjectivities they may or may not have had. This is true for many other aspects of their lives and mentalities besides the sexual. When Halperin argues that the very concept of “a sexuality” was unknown to ancient Greeks, he understands “sexuality” as a system of discourse. If we define “sexuality” as a creation of nineteenth-century medical discourse, we can all agree that the Greeks did not have such a concept, but it is an open question whether they may not have had something else that could also be called “sexuality” by someone using a less restricted definition. Halperin is careful in his recent work to note that he is not denying the existence of sexual identities before the modern period but of sexual orientations, which are quite different.[7] Other scholars, however, have used Foucault’s ideas without the careful reading that characterizes Halperin’s work, and have used it simply to say that sexuality is not worth studying before the modern era.

Most modern scholars who draw on Foucault rely mainly on the first, introductory volume in his History of Sexuality. He further discussed the ancient world in the second and third volumes, The Uses of Pleasure and The Care of the Self, dealing with fifth-century Athenian culture and Hellenistic culture respectively.[8] In these latter volumes, he attempted to apply the analytical program he had laid out in the first, using what amounts to a quite traditional historical method. He cited selected, mainly canonical texts, in support of the argument that, in classical Greece, the regulation of the aphrodisia (a term that, he notes, means something like “sensual pleasures,” but cannot be translated precisely)[9] was a matter of the health of the body and mind rather than a matter of morality. In the Roman world, the link between power over the self and power over others became attenuated, and a sense of danger and imperfection came to characterize the ethics of sexual activity.

The volumes of The History of Sexuality focusing specifically on the ancient world met with mixed reviews. Many historians and other classical scholars have long been critical of Foucault on specifics, arguing that he was no historian. Others have responded that the power of the ideas is such that they are far more important than the details.[10] It is of course true that one can be wrong about the sources and still present ideas of such force that they are very fruitful in the hands of those more prepared to deal accurately with the sources. But classicists have other grounds, too, to criticize Foucault.[11]

In Rethinking Sexuality: Foucault and Classical Antiquity, edited by David H. J. Larmour, Paul Allen Miller, and Charles Platter (1998), a group of classicists, theoreticians, and philosophers discuss Foucault’s views on classical antiquity. Several of these articles are primarily concerned with the place of The History of Sexuality in, and its contribution to the understanding of, Foucault’s oeuvre. Others, however, comment more directly on Foucault’s contribution to the study of classical antiquity.

The most sustained critique of Foucault has come from feminist scholars, but they have not all critiqued him in the same way. Three feminist essays in this book, by Page duBois, Amy Richlin, and Lin Foxhall, illustrate the range of views. DuBois points out that, by beginning a history of sexuality with the Greeks, Foucault assumed “the inevitable primacy of masculine subject-formation, of women’s subjection and submission.”[12] He accepted the interpretation of Greek culture that made the history of the self the history of the male self, with women always subordinated. By focusing only on classical Athenian culture, rather than the lyric culture of Sappho, he wrote women out of the history of sexuality. And yet duBois’ essay is far from an attack on Foucault. She draws from his work the revelation of the defamiliarization of the past, the notion that the ancient world was radically different, which allows a fresh view of Sappho, among other figures. The discipline of classics, she claims, is fundamentally ahistorical, because “one employs the philological method not to uncover the strangeness of antiquity but rather to dwell in a culture assimilated to our own.”[13] Foucault gives it back that strangeness.

An emphasis on the strangeness and otherness of the past is important to a historian, who should not assume that people have always thought the same way as she does. But she should not assume, either, that they thought differently. DuBois critiques the position that would say of the ancient Greeks, “If, for example, we are homophobic, so must they have been.”[14] But it would be equally wrong to say, “We are different from them, so if, for example, we are homophobic, they must not have been.” A healthy respect for the alterity of the past need not make us deny all similarity or continuity.[15] Yes, the context in which women existed in antiquity was quite different from that in late capitalist postmodernism, but this does not mean that we must deny continuity when the evidence points to it and we have carefully queried whether our own preconceptions cause us to interpret it that way.

Amy Richlin treats Foucault as a historian much more directly than duBois does. For duBois, it is the theoretical force of his historicist stance that is important, not his particular interpretations of the ancient world. For Richlin, it is what he has to say about that world: not the details of particular texts but his historical method. She notes that Foucault’s antiquity lacks not only women but also Jews, Africans, Egyptians, Semites, Northern Europeans, children, babies, poor people, and slaves. The absence of women is dictated by his choice of genres and authors, “so that the text replicates the omissions of the history it documents.”[16] Richlin makes a great deal of the use of “the Greeks” to mean specifically male (and elite) Greeks. DuBois comments on this, too: both note that the “experience” discussed in Foucault’s “four great axes of experience” of the “person” include “the relation to one’s wife,” although not all persons may have wives.[17] But for duBois, this does not invalidate Foucault’s important insights; for Richlin, it does. Foucault is dangerous because he is such an icon and perpetuates such a traditional misogyny. This may seem to be throwing out the baby with the bathwater. Certainly, feminist historians have used the tactic, “I am only discussing here the experience of white middle-class women, because that is the only group for which there is evidence.” Why should Foucault not do the same, as long as he notes that he is focusing on the experience of men? Richlin points out that he often fails to note it; his disclaimers are not sufficient to balance out the naturalization of the upper-class masculine. Discussing women only as wives, discussing only male-male love under the rubric of “Erotics,” and then packaging the whole as authoritative replicates a historical imbalance.

Richlin presents a number of useful directions for a history of ancient sexuality that includes women. Lin Foxhall takes up a part of this agenda. She stresses the different way in which women and men in ancient Greece experienced time and the fact that women controlled men’s access to “the three-generation time scale which framed most of everyday life.”[18] She also discusses the importance to social reproduction of women’s ritual activity in which they displayed their sexuality to each other in the absence of men. By dominating the oikos (household), men attempted to monopolize civic life, but women controlled not just physical but also social reproduction into the next generation. Further, she argues, women were not just seen as passive sexually; although moikheia (adultery) was an offense against the man who controlled the woman involved, the implication is that the woman had taken control of her own sexuality, subverting her husband’s control. By ignoring the household context of feminine sexuality, Foxhall shows, Foucault has ignored a big part of the picture. A number of the articles in Roman Sexualities, discussed below, also work to place women at the center of a history of ancient sexuality.

Scholars of Greece, in particular, have tended to accept Foucault’s theoretical insights, if not his research on Athenian society. The prominent 1990 collection of essays on the ancient Greek world, Before Sexuality, recognized the stronger and weaker forms of social constructionism: the title, the editors explain, could mean “before our sexuality,” suggesting merely that “sexual meanings and practices in the ancient Greek world were constituted differently from our own,” or it could “suggest that the very category ‘sexuality’ is a specifically modern construction.”[19] One of the editors, John J. Winkler, went on elsewhere to stake out a moderate social constructionist position, for example in a discussion of the kinaidos (Latin cinaedus), often interpreted as a “passive homosexual”: “The kinaidos, to be sure, is not a ‘homosexual’ but neither is he just an ordinary guy who now and then decided to commit a kinaidic act. The conception of a kinaidos was of a man socially deviant in his entire being.”[20] It is, however, the view that sexuality and sexual identity did not exist in the ancient (or medieval, or early modern, or non-Western) world that has gained currency among non-classicists.[21]

Both moderate and strong social constructionists have tended to agree that gender roles—masculine or feminine, active or passive—were more important than object choice in the ancient world, although they disagree on whether this means that the Greeks and Romans had sexualities very different from ours or that their classifications were based on gender roles rather than (not as a part of) sexuality. Key to the distinction of gender roles was the concept that men are active and women passive, or that men are penetrators and women penetrated.[22] Thus anyone who is penetrated (or is in other ways passive) is gendered feminine, and anyone who penetrates is masculine. For the Romans, to penetrate other men could be a sign of masculinity (hence Valerius Asiaticus’s taunt, “Question your sons, Suillius, they’ll say that I’m a man,” whereas a modern taunter might be more likely to say, “Ask your mother”).[23] Women who penetrate (with dildos or large clitorises) and men who are penetrated are seen not primarily as sexual deviants but as gender transgressors. The primary example of such a deviant man is the kinaidos or cinaedus, but the exact meaning of these terms and the exact way in which such a person deviated from accepted gender roles is the subject of some dispute.

This way of understanding sex as something someone does to someone else seems fairly common in ancient Mediterranean culture.[24] Although this is a long way from modern understandings of homosexuality as related to the gender of object choice, not the gender of act performed, the idea that it is only the passive man or active woman who is perverted, not the man who penetrates another man or the woman who is penetrated by another woman, certainly survived well into this century.[25] This governing paradigm of ancient sexuality may be very different from the scholarly construction of sexuality today but not so far from a view widely held among the North American public, that gay men are effeminate and lesbians masculine.

The fact that there is more than one way of understanding homosexual behavior in contemporary culture should remind us that the ancients did not have a unitary view of it, either. Attitudes varied from archaic to classical Greece to Rome, and varied, too, within a given polis, as David Cohen stresses for Athens, with its internally contradictory expressions.[26] T. K. Hubbard has argued recently, as Kenneth Dover did in 1978, that same-sex relations were far more acceptable among elites than among the mass of the Athenian people and that the latter condemned both the active and the passive partner.[27] He argues that the Greeks did categorize by gender of object choice rather than role. Such a view is not incompatible with a moderate idea of the social construction of sexuality, since the argument is made on the basis of Greek texts and the categories that emerge from them, rather than on the imposition of modern categories.

The mainstream or majority view of a given culture, whether that of today or antiquity, is created in large part by a dominant masculine discourse. Did everyone in ancient Greece or Rome understand sex as something that a penetrator does to a penetrated, and if so, did those who were penetrated still see themselves as active participants? Answering these questions requires examining a wide range of texts, beyond the canonical, in innovative ways. In order to do so, a historian must understand something about how a given text helped construct the mental world of its contemporaries: who read it, and how did they understand it? Did it represent a dominant or subversive point of view?

Bruce Thornton’s Eros: The Myth of Ancient Greek Sexuality (1997) is critical of much recent work on the history of sexuality, which he dismisses with terms like “fashionable.” He renounces the attempt to answer these questions: “This book, then, is not about what the Greeks ‘really’ thought or felt or did about sex. It is about what the literary remains from 700–100 B.C. say about sex.” Many historians, however, retain a belief that texts constitute, if only in an indirect way, evidence about the contexts that they were created by or helped to create. A narrow focus on the texts also risks losing sight of how those who did not write surviving texts might have felt and lived. Thornton, to judge from his dismissal of attempts to do so, does not think that studying the mentalities of non-elites through the use of “so-called nonprivileged data” is a worthwhile or possible project.[28] To dismiss the whole field of social history because everyday attitudes are not encapsulated in canonical texts is just as reductive as dismissing canonical texts because they were written by dead white European males.

Thornton positions his book as a brave and honest voice crying in the wilderness of cultural-theory cant. This rhetorical move allows him to ignore the questions that social constructionism raises. Thus he writes of Greek attitudes toward “homosexuality” without asking whether the behaviors and identities to which he refers are really the same or fundamentally different from modern “homosexuality.” He notes that “a Greek would not categorize as ‘homosexual’ a man who has penetrated another,” but then goes on to discuss Aristotle referring to “homosexuals,” when in fact, as his own discussion shows, it is not a category based on object choice but the taking of pleasure in passivity that Aristotle criticizes.[29] Thornton does raise substantive questions about whether pederasty was relatively unproblematic to the Greeks, as Halperin, Winkler, and Dover would have it.[30] He argues that many texts speak of boys and women in the same ways, indicating that, for male citizens, sex with women was the norm, and pederasty followed a “heterosexual paradigm,” but this could as easily be taken as evidence that gender roles (penetrator/penetrated), not object choice, were what mattered.[31] His discussion of the kinaidos is not in deep disagreement with Winkler’s, at least in terms of who the Greeks thought the kinaidos was.[32] Indeed, he also shares with social constructionists the fundamental premise of the alterity of the Greeks: unlike our contemporaries, according to Thornton, the Greeks viewed eros as a volatile, chaotic, dangerous, and uncontrollable force, inextricably linked to violence. He suggests that we would do well to learn from the Greeks in this regard and follow the lead of Camille Paglia, who alone has recognized that it is just as dangerous a force today.[33]

James Davidson, in Courtesans and Fishcakes: The Consuming Passions of Classical Athens (1998), also contests the current orthodoxy about the active-passive dichotomy, in an engaged and engaging manner. He takes a moderate constructionist position, arguing that appetites exist in all societies; they are shaped by “historical, social, economic, cultural, intellectual, ideological, etc.” contexts but are not created by them.[34] The book, as its title implies, is about more than sex; it is about the desires generally, or at least men’s desire.[35] Based on his evidence, a case could be made that Greek men defined their identity not by sexual desires but by their eating habits.[36] He describes classical Athenian culture as one where the control of one’s desires and the fight against passion were of crucial importance—a view that, despite Davidson’s renunciation of Foucault, echoes the argument of the latter in The Uses of Pleasure. The passions on which he focuses are the love of fish and the love of men for courtesans. As in the study of same-sex desires among men, women’s desires have only a minor role to play here, but in his discussion of distinctions among those women often lumped together as “prostitutes,” he does provide a useful discussion of the nature of their experience and agency. He attempts to recover the highly cultured courtesan, the hetaera, as an independent and desiring woman, different from the pornai or common whores, and he challenges the view that women fell into only two categories, the secluded, private wife or the public prostitute.

Davidson’s focus on passion leads him to attack the widely held view that it was passivity and effeminacy that made the kinaidoi (and the similar katapugones) such figures of opprobrium.[37] It was not, he argues, because they abdicated the masculine role of penetrator that they were considered shameful but rather because they were unrestrained in their desires. Indeed, he claims, the evidence rarely speaks of their being penetrated. Their status had nothing to do with physical integrity but was a function of self-control. Yet the kinaidos/katapugon clearly was understood as someone who enjoyed being anally penetrated, whether or not this was the core of his identity. Even if it were the immoderacy of his desire, rather than his passive role, that was important, he was still fundamentally a gender transgressor, feminine in that very immoderacy. The kinaidos comes in for far more censure than a man with an immoderate desire to penetrate, although the latter also can meet with disapproval.

Ultimately, Davidson does not undercut the characterization of the Athenian view of sex as something someone does to someone else, a hierarchical act, rather than something two people do together. He argues that those who support the “power-penetration” theory in which the kinaidos is dominated and emasculated by being penetrated are actually applying not a Greek but a Victorian view. Given these attitudes, which have developed over a long Western tradition, “It is small wonder that classicists have interpreted rear-entry penetration in the classical world in terms of aggression and power. But in classical Athens the penetrated were not seen as the inert objects of someone else’s gratification. Women certainly did not lie back (or bend forwards) and wait for things to be done to them . . . Even passive sodomites are shown joining in at every level . . . The kinaidos/katapugon is not a sexual pathic, humiliated and made effeminate by repeated domination, he is a nymphomaniac, full of womanish desire, who dresses up to attract men and has sex at the drop of a hat.”[38] Here, Davidson misinterprets the claim some scholars make when they speak of sexual passivity. The one who is penetrated does not have to be inert or apathetic in order for intercourse to be understood as one person doing something to someone else. Indeed, as Halperin has recently stressed, what distinguishes the kinaidos is not that he is penetrated but that he desires to be penetrated.[39] The passivity is anatomical rather than affective. To say that historians of sexuality have denied that women or kinaidoi took pleasure in sex is to misinterpret the use of the term “passive.” Davidson’s own formulation of the kinaidos as “full of womanish desire” acknowledges the nature of the gender transgression involved.

Although Davidson claims that in classical Athens sexuality was not characterized by power and domination, he contrasts it with other societies where it was, including ancient Rome.[40] A volume edited by Judith Hallett and Marilyn Skinner, Roman Sexualities (1997), certainly supports the Roman end of that contrast.[41] The articles in the book do not agree on all points, but several important themes emerge: the importance of the active/passive dichotomy in Roman culture; the critical importance of sexuality to Roman masculinity; and the existence of different discourses, including women’s discourses about their own sexualities.

There is a tendency among historians to lump sexuality together with the study of women and gender; this is in part because of the way men in the historical periods in question treated women as controlled by their sexualities, but it has the pernicious effect of pushing us and our students to think of women as sexual beings (or sexual objects) first and foremost, to conflate the study of homosexuality with the study of women, and to erase male heterosexuality as a subject of study. A number of articles in this volume show the importance of sexuality to masculine identity regardless of what in the modern era would be called sexual orientation.

Holt Parker and Anthony Corbeill’s articles both focus on the way the active/passive distinction affected ideas about masculinity. Parker constructs a chart to show how the Romans characterized various sexual acts, but his discussion of cunnilingus (equating it with a man’s penetration by a woman) pushes the idea of a sexual “system” beyond the evidence.[42] He also omits the active woman from the grid intended to summarize the possibilities for sexualities as the Romans saw them. Parker’s conclusion, “Tacitus in Ohio,” contains a wonderfully lucid social constructionist explanation of the difference between Roman sexual categories and contemporary North American ones. Parker uses the conceit of imagining Tacitus’s field work in the United States to point out how little we know of how those Romans vilified as cinaedi would have felt about it, or behaved.[43] One could, of course, say the same about women, whose sexuality is depicted mainly in hostile sources.

Most of what survives about cinaedi is the vilification, which shows us how intensely at least one segment of Roman society equated masculinity with penetrative sexual behavior. Corbeill’s article is particularly enlightening on this point. Criticism of banquets as a Greek or Asian import was closely tied to the invective against submissive sexuality.[44] Some scholars have questioned whether the passive male or cinaedus actually existed as a recognized type, characterized by what Foucault called “a kind of interior androgyny, a hermaphrodism of the soul” rather than as a literary construct based on men who merely engaged in specific acts.[45] Both Parker and Corbeill, following on important work by Maud Gleason and Amy Richlin, suggest that cinaedi did exist.[46] The question Parker raises, although Corbeill does not, is how the individual so identified felt about his own identity. Others may have seen inner characteristics expressed through outward behavior, but did the cinaedi? Did the availability of a literary type for use in self-fashioning mean that this was really a sexual identity? If the sources do not exist that could tell us about the subjectivity of the cinaedus, some modern historians would have us take this as evidence that there was no such subjectivity or sexual identity. Even leaving aside the vicissitudes of survival of source material, however, if people did not write about certain types of feelings in the Roman period as they do today—if it was a less confessional age—does that mean that they did not have those feelings? It would be a mistake to assume that because our contemporaries tend to identify themselves by their sexualities, the Romans did, too; but equally it would be a mistake to assume that the Romans could not have.

If we do not know how those labeled cinaedi viewed themselves, we know more about the mainstream Roman elite’s fear of being so labeled. Marilyn Skinner, in an article originally published in 1993, points out that masculinity in ancient Rome was a very fragile condition but necessary for public manifestations of rank and authority.[47] It was not achieved simply by being born male and growing up; it required constant vigilance to avoid being feminized by a loss of social status, a loss of control over one’s family and slaves, or inappropriate sexual behavior. Jonathan Walters discusses the importance of impenetrability to Roman concepts of manhood. Not all adult males were impenetrable, and not all were considered real men (viri). To be the passive partner in a sexual relation was muliebria pati, “to have a woman’s experience,” but to be penetrated was not just the experience of a woman but also that of a slave or freedman. He discusses why the corporal punishment of a slave, but not that of a free child or the penetration of a soldier by a sword, was equivalent to sexual penetration.[48]

Walters’s argument raises a problem common to historians of gender who refer to biological females being constructed as masculine, and vice versa, or of gender being constituted by performance.[49] This, however, is a use of language not taken literally outside the academy, and perhaps too literally within it. Ask a hundred people off the street (either in ancient Rome, the modern United States, or anywhere else) to explain their concept of masculinity, and you will get a wide variety of answers. Ask them instead to define the term “man” and you will get a small range of variation. On one level, no matter how devoutly we believe that real men don’t eat quiche, cry, or engage in violence against women, we still use the term “man” in common parlance to mean an adult male human. An adult male in ancient Rome who was penetrated might become an object of scorn, and might be castigated as unmanly or not really a man, but there would be no question that laws or medical texts that applied to viri applied to him. His gender transformation is metaphoric or symbolic, but no one thinks that having a woman’s experience has “really” made him a woman.

About the sexual subjectivity of biological females, like that of feminized males, surviving texts reveal relatively little. Thus Catharine Edwards writes about prostitution without any consideration of how the prostitutes may have felt about their infamy.[50] Sandra Joshel, writing about a woman of the elite, Messalina, who is far better documented than women generally, still does not attempt to discover the “real” Messalina’s experience; she is “concerned with the writing of history, not with history as a set of events.”[51] Even the study of female homoeroticism focuses on men’s constructions of women’s sexuality. Pamela Gordon argues that Ovid’s Sappho is masculinized and “simply acts out a charade of male sexuality.” She suggests that we “imagine the masculinized lesbians of the Roman texts not as monsters or fools (as their creators intended), but as dauntless rebels.”[52] The modern reader may, despite the hostility of these authors to women, be able to infer from them something of how the women may have seen themselves and their sexualities.

Amy Richlin draws on “a jumble of encyclopedias and agricultural handbooks” to more nearly approach women’s sexual experience. Her title, “Pliny’s Brassiere,” refers to Pliny’s use of a female undergarment tied around his head to relieve headaches; this brings her to the “medicinal uses of the female human body,” which in turn tell us something about women’s sexualities.[53] Women used various forms of herbal medicine and sympathetic magic to control (mainly to promote) their own fertility.

Judith Hallett’s “Female Homoeroticism and the Denial of Roman Reality in Latin Literature,” originally published in 1989 but still important enough to fully warrant its reprinting in the Hallett and Skinner volume, also attempts to disentangle women’s actual sexual lives from masculine literary constructions of them. Hallett notes that Roman writers discuss tribades (women who engaged in sex with each other) as something from the remote Greek past, not as part of their own contemporary society, and also masculinize them, denying their resemblance to normal Roman women. They do so despite evidence in their own writings that they knew full well that women could give each other pleasure without penetration. “But for Roman males who wrote about tribadism, it was evidently easier to deny the actual and avow the unlikely than to abandon assumptions about how, according to biological nature and Roman culture, women ought to behave.”[54]

Although Hallett argues that same-sex female erotic behavior existed and that Roman writers who did not acknowledge its existence except as a monstrosity were engaging in deliberate self-deception, she is not able in the space of an article to discuss in detail that experience. A major and important attempt to do so is Bernadette Brooten’s Love between Women: Early Christian Responses to Female Homoeroticism (1996). The second part of Brooten’s book, and that which will be of most interest in the contemporary debate about the morality of homosexuality, focuses on a biblical text, Romans 1: 18–32. Brooten’s discussion of this text will be disconcerting to many Christians: to members of the religious right, because she shows that Paul’s homophobic attitudes were reflections of his time (and therefore, although she does not say this explicitly, not to be taken as eternal truths), and to those on the religious left who would like to explain or interpret away Paul’s apparent homophobia, because she shows that he meant it. Historically speaking, however, her argument makes a great deal of sense: Paul was not particularly innovative in either a progressive or reactionary way but rather was part of a complex web of texts on homoeroticism that circulated in the Hellenistic Mediterranean.

While Brooten has made an extremely important contribution to the study of early Christian attitudes toward women’s sexualities, our concern here is with the first half of the book, in which she sets the context for her exegesis of Paul. In meticulous detail, she gives the evidence for female homoerotic relationships in the Roman world and comes as close as is possible to how women themselves experienced those relationships. She notes that “[t]he sources bear witness to male constructions of female homoeroticism, rather than to lesbians’ perceptions of themselves,” but she goes as far as she can in reasoning from those constructions to the experience.[55] Like other scholars, she points to the importance of the active/passive distinction in Roman mentalities: men were supposed to be active, women passive, and to transgress these roles was unnatural. Yet there is evidence that women who loved other women did not necessarily accept these attitudes. Brooten also questions the strong social constructionist view that there can have been no sexual identities, orientations, or sexualities in the ancient world: “I present non-Christian material in this book for a category of persons viewed in antiquity as having a long-term or even lifelong homoerotic orientation,” a category that included both men and women.[56]

Some of the evidence Brooten presents, though sketchy, casts doubt on the traditional Roman view described by Hallett that female homoerotic practices involved penetration in imitation of men. For example, a wall painting from Pompeii depicts, among other things, one woman performing cunnilingus on another, who fellates a man. Although this can hardly be taken as typical of Roman sexual practices, it is noteworthy that the female same-sex practice depicted is not phallic.[57]

The four chapters of the first section of the book focus on four types of sources. Greek erotic spells from Egypt actually state the names of women who commissioned the spells and the women they loved. Brooten cautions that “the individual women probably did not compose their own spells. Rather, these spells contain highly formulaic language that reveals more about cultural ideology than about individual women’s lives.”[58] Very true; but it still reveals more about individual women’s lives than other sources do, namely that real, non-elite women, not literary fictions, did feel strongly enough about other women to commission these spells, and the cultural atmosphere was such that they were able to go public in doing so. That Brooten takes forty pages to present and analyze three brief spells indicates something about the level of detail in this book; she has taken pains to preempt the criticisms of specialists.

The chapter on astrological texts does not get as close to the lives of actual women, but it does present female homoeroticism as among the range of sexual possibilities the astrologers recognized, although they considered it an exception to the norm since it called for women to be active rather than passive. By explaining the astrological reasons for women’s desire to have sex with other women (and also to have sex with many men), these texts show that it was a recognized part of everyday life, not just something literary authors wanted to present as an example of Greek perversion.

In her chapter on medical texts, Brooten suggests that Foucault’s placing of the medicalization of homosexuality in the nineteenth century was about 1,700 years too late. “The patient suffering from such a disease [certain forms of same-sex desire] did have an identity, apparently a lifelong one, characterized by behavior considered unnatural (i.e., appropriate for the opposite sex), unless treatment effected a cure.”[59] This treatment might include clitoridectomy, and Brooten raises the unanswered question of “at whose request such brutal surgery took place”—husband? brothel owner? The medical texts accept the view that female homoeroticism requires penetration.

Finally, the literature of dream interpretation tells us something of the generally accepted meanings of sexual activity. Artemidorus Daldianus (fl. second century ce) placed woman-woman intercourse in his category of dreams about unnatural sexual acts and distinguished between dreams of being the active and passive partner. Brooten suggests that these acts were perceived as unnatural because they failed to recreate patterns of social dominance; they perpetuated the penetrator/penetrated dichotomy, without that dichotomy corresponding to positions in the social hierarchy, as it was usually assumed to do.

The move by feminist and other scholars, Brooten among them, to recover people’s experiences of what we might tendentiously call their own sexualities goes against Foucault’s argument that what is important to study is sexuality, which is a discursive phenomenon, rather than sex, which is unreal and ahistorical. The important contribution of much of this recent work is that it does both: it attempts to reconstruct lived experience while recognizing the substantial methodological and theoretical problems that stand in the way of that reconstruction. It analyzes both the discourse and its potential relation to practice.

The theoretical sophistication that the study of sexualities in ancient Greece and Rome has acquired makes a familiarity with this field necessary to anyone who studies or teaches the history of sexuality in other times or places. These new works teach us that it is possible to recover some aspects of women’s sexualities even from male-dominated and male-documented cultures, that multiple and contradictory discourses can exist in one society, and that continuities as well as discontinuities are historical. They reveal some broad differences between Athenian and Roman sexual norms, but the differences and contradictions within each culture stand out more than the contrasts. We may never reach agreement on the exact meaning of figures like the kinaidos/cinaedus for ancient societies, but this should be no surprise. There is no agreement on the nature of gay and lesbian identities today, even among gay men and lesbians. When all we have are texts that were, after all, not composed in order to answer the questions we would like to ask, the discourse may seem more unified, but it is entirely possible that opinions among the ancient Greeks and Romans about sexuality were as varied and as vehement as the opinions of scholars today.

Ruth Mazo Karras is a professor of history at the University of Minnesota. She is author of Common Women: Prostitution and Sexuality in Medieval England (1996) and other works dealing with gender and sexuality in medieval Europe, as well as Slavery and Society in Medieval Scandinavia (1988). Her current research project examines competing constructions of masculinity in the later Middle Ages. In addition to medieval history, Karras teaches courses in the history of sexuality, so her interest in ancient sexualities grows out of her teaching.


I wish to thank Christie Balka, Martha Davis, Ralph Hexter, Chris Karras, Janice Siegel, and Dan Tompkins, as well as three anonymous readers for the AHR, for their thoughtful readings of earlier versions of this article. All opinions, of course, are my own.

1) Kenneth J. Dover, Greek Homosexuality (Cambridge, 1978).

2) As an example of the latter, see David Halperin’s attack on Amy Richlin in “Forgetting Foucault: Acts, Identities, and the History of Sexuality,” Representations 63 (1998): 91–120, esp. 104–05. Slightly less vitriolic but similarly reflecting a heavy personal involvement in the argument is Richlin’s “Zeus and Metis: Foucault, Feminism, Classics,” Helios 18 (1991): 160–80. Marilyn B. Skinner, “Zeus and Leda: The Sexuality Wars in Contemporary Classical Scholarship,” Thamyris 3, no. 1 (1996): 103–23, also available on the World Wide Web at, provides an overview of the controversy.

3) Edward Stein, ed., Forms of Desire: Sexual Orientation and the Social Constructionist Controversy (New York, 1990), unites some of the early texts in this discussion.

4) Practically no one claims to be an essentialist. Eva Cantarella, Bisexuality in the Ancient World, Cormac Ó Cuilleanáin, trans. (New Haven, Conn., 1992), by its title and continuing use of the terms “homosexuality,” “heterosexuality,” and “bisexuality,” would seem to fall into the essentialist camp, but Cantarella seems to be using “homosexuality” as synonymous with “homosexual behavior” and rarely writes of people as being “homosexuals” or “bisexuals.” When she does do so, in the case of “passive homosexuals” (p. 113), the context indicates that she means those engaged in homosexual practices, rather than orientation: she is discussing a law that punished people for their actions.

5) John Thorp, “The Social Construction of Homosexuality,” Phoenix 46 (1992): 54–61, calls these the “weak” and “strong” forms of social constructionism.

6) Michel Foucault, The History of Sexuality, Vol. 1: An Introduction, Robert Hurley, trans. (New York, 1990).

7) See, for example, Jeffrey Weeks, Against Nature: Essays in History, Sexuality, and Identity (London, 1991), 92: “these social identities and intimacies based on sex are relatively new. They have not, could not have existed throughout the mists of time, because the conditions that gave rise to them just did not exist”; and David Halperin, One Hundred Years of Homosexuality (New York, 1990), 27: “Before the scientific construction of ‘sexuality’ as a supposedly positive, distinct, and constitutive feature of individual human beings—an autonomous system within the physiological and psychological economy of the human organism—certain kinds of sexual acts could be individually evaluated and categorized, and so could certain tastes or inclinations, but there was no conceptual apparatus available for identifying a person’s fixed and determinate sexual orientation, much less for assessing and classifying it.” Besides Foucault, Mary McIntosh’s article originally published in 1968 was also influential on Weeks: McIntosh, “The Homosexual Role,” in The Making of the Modern Homosexual, Kenneth Plummer, ed. (Totowa, N.J., 1980). Halperin’s more recent formulation is in “Forgetting Foucault,” 101–04, 108–10.

8) Michel Foucault, The History of Sexuality, Vol. 2: The Uses of Pleasure, Vol. 3: The Care of the Self, Robert Hurley, trans. (New York, 1985–86).

9) Foucault, Uses of Pleasure, 35.

10) David Halperin, Saint Foucault: Towards a Gay Hagiography (New York, 1995), 5–6. On 189–90 n. 9, Halperin lists some of the scholars he thinks have been inappropriately hostile to Foucault.

11) See especially Richlin, “Zeus and Metis,” for a feminist critique of Foucault.

12) Page duBois, “The Subject in Antiquity after Foucault,” in Rethinking Sexuality: Foucault and Classical Antiquity, David H. J. Larmour, Paul Allen Miller, and Charles Platter, eds. (Princeton, N.J., 1998), 85–103, quotation at 86.

13) DuBois, “Subject in Antiquity after Foucault,” 93.

14) DuBois, “Subject in Antiquity after Foucault,” 94.

15) Thus, when duBois criticizes Amy Richlin for seeing a continuity of misogyny, calling this ahistorical and an erasure of historical difference due to a failure to question the category “women” (89), she is tilting the balance too far in the direction of alterity.

16) Amy Richlin, “Foucault’s History of Sexuality: A Useful Theory for Women?” in Larmour, Miller, and Platter, Rethinking Sexuality, 138–70, quotation at 139.

17) Richlin, “Foucault’s History of Sexuality,” 143; duBois, “Subject in Antiquity after Foucault,” 187.

18) Lin Foxhall, “Pandora Unbound: A Feminist Critique of Foucault’s History of Sexuality,” in Larmour, Miller, and Platter, Rethinking Sexuality, 122–37, quotation at 126.

19) “Introduction,” in Before Sexuality: The Construction of Erotic Experience in the Ancient Greek World, David M. Halperin, John J. Winkler, and Froma Zeitlin, eds. (Princeton, N.J., 1990), 5.

20) John J. Winkler, The Constraints of Desire: The Anthropology of Sex and Gender in Ancient Greece (New York, 1990), 45. See now Halperin’s discussion of the kinaidos in “Forgetting Foucault,” 100–04.

21) See, for example, Martha Nussbaum, “Therapeutic Arguments and Structures of Desire,” differences: A Journal of Feminist Cultural Studies 2 (1990): 46–66, esp. 49.

22)  Eva Keuls, The Reign of the Phallus: Sexual Politics in Ancient Athens (Berkeley, Calif., 1985), is a strong, perhaps too strong, statement of the centrality of this phallic view not only to ancient sexuality but to all of Athenian culture.

23) Tacitus, Annales 11: 2, John Jackson, ed., Loeb Classical Library (Cambridge, Mass., 1986), 250.

24) Halperin recently complained that Bernadette Brooten, in the book under review here, Love between Women, failed to attribute the idea to him, Winkler, or Foucault; she, however, replied that she had made this argument earlier. “The GLQ Forum: Lesbian Historiography before the Name?” GLQ: A Journal of Lesbian and Gay Studies 4 (1998): 560 and 627 n. 1. This point is so pervasive in the literature that it would seem to be impossible to attribute it to a single author. Richlin, “Zeus and Metis,” 172–73, also demonstrates that feminist scholars made this point before the publication of Halperin’s and Winkler’s major books in 1990.

25) George Chauncey, Jr., Gay New York: Gender, Urban Culture, and the Making of the Gay Male World, 1890–1940 (New York, 1994), 119, suggests that this idea that the active partner was “normal” was characteristic of working-class culture in the 1920s. See also Chauncey, “Christian Brotherhood or Sexual Perversion? Homosexual Identities and the Construction of Sexual Boundaries in the World War I Era,” in Hidden from History: Reclaiming the Gay and Lesbian Past, Martin Bauml Duberman, Martha Vicinus, and George Chauncey, Jr., eds. (New York, 1989), 294–317. The idea that the passive partner in a female same-sex relationship was not a lesbian was certainly characteristic of the relationships depicted in Radclyffe Hall, The Well of Loneliness (New York, 1990); see Elizabeth Lapovsky Kennedy and Madeline D. Davis, Boots of Leather, Slippers of Gold: The History of a Lesbian Community (New York, 1993), 323–71, on the persistence of the “gender-inversion” model in working-class Buffalo in the 1940s and 1950s. Some fems considered only the butches to be lesbians.

26) David Cohen, Law, Sexuality, and Society: The Enforcement of Morals in Classical Athens (Cambridge, 1991), 171–202.

27) T. K. Hubbard, “Popular Perceptions of Elite Homosexuality in Classical Athens,” Arion, 3d ser., 6 (1998): 48–78; Dover, Greek Homosexuality, 148–51.

28) Bruce J. Thornton, Eros: The Myth of Ancient Greek Sexuality (Boulder, Colo., 1997), xii.

29) Thornton, Eros, 194.

30) Thornton, Eros, 193–202; Cohen, Law, Sexuality, and Society, 171–202.

31) Thornton, Eros, 194.

32) John J. Winkler, “Laying Down the Law: The Oversight of Men’s Sexual Behavior in Classical Athens,” in Halperin, Winkler, and Zeitlin, Before Sexuality, 171–209. This similarity comes despite his praise of Camille Paglia for her “brutal demolition” of Winkler. The first half of the article he cites, Paglia, “Junk Bonds and Corporate Raiders: Academe in the Hour of the Wolf,” Arion, 3d ser., 1 (Spring 1991): 139–212, is indeed structured as a demolition of Halperin’s and Winkler’s work, though not all will share Thornton’s view of its success; the second is an ad hominem et feminam diatribe against contemporary academia.

33) Thornton, Eros, 218.

34) James Davidson, Courtesans and Fishcakes: The Consuming Passions of Classical Athens (New York, 1998), 312.

35) When Davidson notes with disapproval (xxiii) that “Foucault’s study of Greek sexuality has very little on women at all and gives the impression that the Greeks were very much more interested in boys,” he clearly means women as partners for men, since it is Greek men who may or may not have been more interested in boys.

36) Ironically, Halperin has used the counterfactual example of identities based on eating habits as an analogy to explain his view of social construction: “Sex before Sexuality: Pederasty, Politics, and Power in Classical Athens,” in Duberman, Vicinus, and Chauncey, Hidden from History, 41–42.

37) Davidson, Courtesans and Fishcakes, 167–82. There is some dispute about the meaning of “katapugon”; Davison argues that it refers not just to a passive male but to a lascivious person.

38) Davidson, Courtesans and Fishcakes, 179.

39) David Halperin, in “GLQ Forum: Lesbian Historiography before the Name?” 568.

40) Davidson, Courtesans and Fishcakes, 169.

41) So does Amy Richlin, The Garden of Priapus: Sexuality and Aggression in Roman Humor, rev. edn. (New York, 1992), which Davidson does not cite on this point.

42) Holt Parker, “The Teratogenic Grid,” in Judith P. Hallett and Marilyn B. Skinner, eds., Roman Sexualities (Princeton, N.J., 1997), 47–65, 51.

43) Parker, “Teratogenic Grid,” 62–63.

44) Anthony Corbeill, “Dining Deviants in Roman Political Invective,” in Hallett and Skinner, Roman Sexualities, 99–128.

45) Foucault, History of Sexuality, 1: 43.

46) Maud Gleason, “The Semiotics of Gender: Physiognomy and Self-Fashioning in the Second Century C.E.,” in Halperin, Winkler, and Zeitlin, Before Sexuality, 399–415, now incorporated in Gleason, Making Men: Sophists and Self-Presentation in Ancient Rome (Princeton, N.J., 1995), 55–81; Amy Richlin, “Not before Homosexuality: The Materiality of the Cinaedus and the Roman Law against Love between Men,” Journal of the History of Sexuality 3 (1993): 523–73; Parker, “Teratogenic Grid,” 60–62; Corbeill, “Dining Deviants,” 112–17.

47) Marilyn B. Skinner, “Ego mulier: The Construction of Male Sexuality in Catullus,” in Hallett and Skinner, Roman Sexualities, 129–50.

48) Jonathan Walters, “Invading the Body: Manliness and Impenetrability in Roman Thought,” in Hallett and Skinner, Roman Sexualities, 29–43.

49) On gender as performance, see Judith Butler, Gender Trouble: Feminism and the Subversion of Identity (New York, 1990), 135–41.

50) Catharine Edwards, “Unspeakable Professions: Public Performance and Prostitution in Ancient Rome,” in Hallett and Skinner, Roman Sexualities, 66–95.

51) Sandra Joshel, “Female Desire and the Discourse of Empire: Tacitus’s Messalina,” in Hallett and Skinner, Roman Sexualities, 221–54, quotation at 222.

52) Pamela Gordon, “The Lover’s Voice in Heroides 15: Or, Why Is Sappho a Man?” in Hallett and Skinner, Roman Sexualities, 274–91, quotations at 283, 288.

53) Amy Richlin, “Pliny’s Brassiere,” in Hallett and Skinner, Roman Sexualities, 197–220, see 200.

54) Judith P. Hallett, “Female Homoeroticism and the Denial of Reality in Latin Literature,” in Hallett and Skinner, Roman Sexualities, 255–73, quotation at 268.

55) Bernadette J. Brooten, Love between Women: Early Christian Responses to Female Homoeroticism (Chicago, 1996), 25.

56) Brooten, Love between Women, 9. A number of scholars have questioned Brooten’s conclusions on this point, charging that she applies a modern concept of egalitarian lesbian relationships that is not present in the ancient sources; for these and other critiques, and Brooten’s response, see “GLQ Forum: Lesbian Historiography before the Name?”

57) Brooten, Love between Women, 60.

58) Brooten, Love between Women, 73.

59) Brooten, Love between Women, 144.