BMCR 1999.11.05

Roman Homosexuality: Ideologies of Masculinity in Classical Antiquity

, Roman Homosexuality: Ideologies of Masculinity in Classical Antiquity. Oxford: Oxford University Press, 1999. xii, 395; 12 pls.

In Rome of the early Empire, there were many men who threw off the conventions of traditional Roman manhood and instead assumed an “effeminate” appearance and manner, thereby, in the usual case, advertising their eagerness for sexual encounters with other males. These were the “softies” ( molles), the cinaedi. 1 Their numbers cannot even be guessed, but, in a city of a million persons, they might easily have numbered in the tens of thousands; Juvenal saw them flocking into Rome on every available transport (9.130-133). And golly, were they ever conspicuous. The Apostle Paul, with his usual provincial primness, adduces overt homosexual behavior as his chief example of the capital’s decadence ( Rom. 1.26-27).2

Roman sources on cinaedi, though abundant, are also almost invariably hostile, and they illuminate only a fraction of the social landscape. Who were these men? In typical cases, what was their background? How did they survive in an adverse environment? What did they think about their world, and what were their aspirations? We have some tantalizing clues. By far the most singular is the speech that Juvenal gives to the jaded libertine Laronia: the cinaedi stand together, arrayed like a phalanx of soldiers. “Great is the accord among the soft ones,” she complains: magna inter molles concordia (2.45-47). Is Laronia’s remark intended seriously? If so, what exactly does it signify? How did the cinaedi achieve their accord, and how exactly did it serve them?

A book could be written about the meaning of Laronia’s remark. Unfortunately, it isn’t this book; don’t come to Craig Williams expecting much in the way of historical reconstruction. As Williams admits (p. 5), his title is somewhat misleading; Roman Heterosexuality would have served him as well. In fact, his subtitle, for all its dour academicism, captures the substance of the book more exactly, so gird your loins for another romp through the theme park of social constructionism.3 Nonetheless, there is considerable truth in Williams’s observation that the cinaedi constituted “a necessary negative” to the standard Roman concept of masculinity, illustrating, almost in comic book fashion, “what a real man must not be” (p. 183); and, for the rest, Williams writes with a calm and clarity uncommon in contemporary cultural studies.

The early chapters are the best. On the basis of literary sources Williams contends (pp. 18-19) that Roman male sexuality centered on three “traditional protocols governing sexual practices”: first, “a self-respecting Roman man must always give the appearance of playing the insertive role in penetrative acts, and not the receptive role”; second, “apart from his wife, freeborn Romans were officially off-limits [as] sexual partners for a Roman man”; and third, “less of a rule than a tendency pervading the ancient sources: a noticeable proclivity toward smooth young bodies.”

Insofar as homosexual behavior is concerned, the importance of these “protocols” is that they do not formally discriminate between heterosexual and homosexual acts. As Williams puts it (p. 4), “[H]omosexuality turns out not to have been an important issue for the Romans.” This is, in fact, a major theme of his book. But some qualification is required before accepting such a conclusion, since Williams rarely goes further than assessing whether a given form of sexual conduct was reproachable. Comparison with Latin America, which has a code of masculinity strikingly similar to Rome’s, suggests the possibility of a more subtle sexual etiquette in which insertive homosexual behavior, although not necessarily censured (especially if it is not flaunted), is still accorded appreciably lower social esteem than extramarital sex with women.4 Williams does not try to measure the degree of approval for various forms of sexual expression, but the sources may well be too elastic to permit this.

More significantly, the “traditional protocols” were not deeply prescriptive, and in fact were often trumped by what Williams calls the “Priapic model of masculinity” (p. 18): a hyperaggressive male sexuality manifested most obviously in the distinctive Roman fascination with oversized penises (pp. 86-95). Thus, for example, although social mores may have inclined males in the direction of “smooth young bodies,” Plautus already provides evidence for sexual advances toward mature males (pp. 77-82). In a later age, such “real men” would furnish the cinaedi with their clientele (pp. 181-183, 215-218).

In any case, it is clear — and, amazingly, such a conspicuous fact has gone virtually unrecognized in earlier scholarship5 — that native Roman morality did not reproach homosexual behavior itself, so long as it was correctly directed and performed. Williams is surely right about this, and his discussion of early sources on sex with male slaves and prostitutes (pp. 30-47) is wholly convincing. Equally so, I think, is his treatment of a more troublesome question, the Roman “reception” of Greek pederasty (pp. 63-67): the Romans had little difficulty incorporating pederasty within their sexual model, except that they firmly resisted, at least in principle, homosexual advances toward free Roman youths. The result is a form of pederasty normally displaced toward male slaves, though with some notable exceptions (Catullus comes to mind).

Throughout his discussion, Williams insists on melding homosexual behavior into a more general treatment of male sexuality, by including, for instance, broader discussions of male extramarital sex in relation to ideals of marital fertility (pp. 47-56); here his handling of Musonius Rufus’ Stoic strictures is especially helpful (cf. also pp. 139-140, 239-244). However, in the last chapters of his book, when Williams turns to examine the cinaedi more closely, he encounters greater difficulties. There is no thornier historical task than using hostile sources to investigate a socially marginal group. Williams’s working methods are fairly obviously inadequate to this task, but inadequate in an interesting way that indicates the limits of his approach.

On matters methodological, Williams is fastidious, usually to a fault. His method rests on an initial distinction between representation and reality, a distinction “crucial to this inquiry” (p. 9). He explains the distinction thus: “By reality I mean the actual day-to-day experiences of real Roman men (for example, that on a certain day one man anally penetrated another), whereas representation refers to the ways in which Roman men as a group publicly portrayed the experiences, whether real or imagined, both of themselves and of others …” (p. 9).6

However, this coarse distinction is deficient in both its heads. First, if what Williams says is to be intelligible, “reality” has to embrace not only human conduct as such, but also repetitive and recursive patterns of behavior, direct and indirect psychological perceptions of it, the elaborate social and political institutions that provide it with context, the technological and economic base that sustains it, and so on — in short, an immensely rich world, alive with incongruity and conflict, reaching well beyond immediate action and sensual experience. On occasion Williams urges the importance of “reality” (e.g., p. 10), but he clearly has little taste for the subject. Instead, he regularly employs a flattened concept of reality in order to bestow artificial privilege on cultural “representations”: “culture shapes individual experience [in that the latter is] given public meaning precisely through the representational medium of language” (p. 6).

But this formulation is, at best, profoundly misleading. How many of us, for instance, would seriously claim that “culture” (the “discourse” of ideas) plays more than a secondary role in bestowing “public meaning” on our sexuality or any other vital element of our individual personalities? Under normal circumstances, when it comes to constituting the psychology of our daily lives, does undifferentiated intellectual “discourse” have more than a small fraction of the influence of our families, friends, communities, schools, workplaces, and churches?7 And even then, from a sociological standpoint, Williams’s methodology is drastically underdetermined, since he fails to supply the necessary causal links tying cultural “discourse” to actual historical experiences of homosexual behavior — an exceptionally challenging project even today. There is no particular difficulty, I think, in maintaining that the internal and external meaning of much individual conduct is socially constructed, although cashing out this intuition is still laborious. But such an assertion is extremely far removed from the more radical view that the meaning of conduct is culturally constructed.8

I stress this point because, in my opinion, Williams is consistently unfair, and often extremely so, to the cinaedi, those adult males who “ceded their masculinity” (p. 183) in large part by flaunting their desire for sex with other males. For obvious reasons, this group is of particular modern interest. Amy Richlin and Rabun Taylor have recently maintained that the cinaedi formed a distinct “subculture,” a population segment readily identified by shared values and “cultural identifiers” that sharply distinguish them from mainstream society.9 The ancient sources on the cinaedi make this view attractive; they highlight a flamboyant repertory of distinctive dress, hair styles, and other easily recognized mannerisms, as well as customary cruising grounds, occupational clustering, and (if Laronia can be believed) even intra-group support.

Predictably, however, Williams scouts the idea of a subculture, arguing that since “these men were fundamentally identified as effeminate rather than as ‘passive homosexuals’ or even as penetrated men, any group identity they might have shared was based first and foremost on their effeminacy, their flouting of the rules of masculinity in general, and not on the specific sexual practice of being penetrated” (p. 222). To put it bluntly, this argument betrays a deep conservative partiality. The cinaedi are “fundamentally identified as effeminate” by unfriendly sources speaking for a prevailing paradigm of male sexuality, a paradigm emphatically rejected by the cinaedi. But, as the past century has made abundantly clear, sexual identity is defined not simply through a dominant paradigm, but also by individual will and choice, often developed through the agency of subcultures and other social groupings. The issue, in short, is not just what others thought of the cinaedi, but also what they thought of themselves.

Williams takes conflicting stances on this core issue. For example, on page 159 he observes: “All that remains of these people are men’s impersonations of them. Their own voices have been silenced, and it is up to us to imagine what they might have sounded like.” How true, how very true; but the project is left to languish. Far more common is the position on page 216, where he asserts that his “main interest” is to “understand how [the cinaedi ] were represented by others in the public discourse of their cultural environment”; accordingly, “we must … describe them as gender deviants rather than as homosexuals.”

As several colleagues have remarked to me, this latter formulation raises a further difficulty: Williams operates without hard concepts of sex, gender, sexual orientation, sexual identity, and sexuality, and hence he has problems articulating their interrelations. For example, according to Williams the word cinaedus was used to describe “any man … who displayed an effeminacy that was most noticeably embodied in a penchant for being anally penetrated” (p. 176, cf. 175-178 and 209-215). In this definition, gender deviance (“effeminacy”) is the prime determinant, but sexual orientation (“penchant”) is smuggled in, so to speak, as a main component of gender deviance. But Williams does not convincingly explain why sexual orientation must always be subordinated to gender deviance, and one might more easily believe that context determines which factor predominates; for instance, in Catullus 16 (“Pedicabo ego vos et irrumabo, / Aureli pathice et cinaede Furi”), the sexual element is foregrounded while effeminacy is ignored. Late in his book, at page 215 (cf. 215-218), Williams finally admits that “these deviant males were … usually imagined to desire to be penetrated not by women but by men.” The reader of these breathtaking words searches in vain for the tiniest trace of Laronia’s knowing smile.

To be sure, all scholarship makes use of brackets, and Williams’s cultural conservatism would represent no more than the exercise of a methodological option, had it not also led him to repeatedly undervalue potentially pertinent information. For example, many Roman writers attribute to cinaedi the practice of scratching their heads with one finger, evidently as some sort of signal. But Williams rejects the obvious inference that this was a sexual signal, or indeed any signal at all. Instead, he conjectures that “the gesture was associated with a desire to avoid disturbing one’s carefully coiffed, and perhaps artificially curled, hair,” hence that it betokened mainly effeminacy (p. 223). Roman sources stoutly repulse this bland notion. Thus a famous fragment of Calvus, directed at Pompey, is unambiguous in its logic: “From this act, you should deduce he’s after a man.”10 Williams turns cartwheels over this text: “Calvus’ ditty on Pompey assumes that a man who makes this gesture, being a pseudo-woman, might ‘want a man,’ like any ‘woman’ ought to. But that is merely an effect of the underlying cause, which is not homosexuality but effeminacy” (p. 223). The intermediate step is transparently superfluous. And in fact the practice survives today in parts of Italy, where (as a friend of mine inadvertently discovered some decades ago) it is still used to solicit homosexual partners — a startling testimony in itself to the number and social strength of the cinaedi in ancient Rome.

Or take same-sex weddings. Martial and Juvenal both describe marriage ceremonies between males, ceremonies complete in every detail down to wedding veils and dowries. Their purpose is obscure, of course, although ( contra John Boswell) there is no question of valid same-sex marriages in Roman law. Williams reserves the awkward subject for an appendix (pp. 245-252), where he concedes that such weddings were apparently not uncommon in the High Empire. But what did the participants think they were doing? Williams doesn’t even conjecture. Fruitful lines of inquiry beckon. For example, Williams still relies on the ancestral Roman concept of marriage as closely tied to the bearing and raising of children (p. 247); he entirely ignores the now abundant scholarship pointing to the emergence, in the early Empire, of more individualistic ideals of marriage built around a couple’s social and sexual companionship.11

Finally, many early imperial sources mention that some cinaedi, allegedly for reasons of deceit or hypocrisy, were cultivating an insistently shaggy, austere appearance. Williams, barely noticing the phenomenon, takes these sources at face value (p. 188: “some cinaedi were able to conceal their identity”), and immediately hares off on an interesting account of recent scholarship on male “appearances and reputation” (pp. 188-193). But evidence for the “remasculinization” of the effeminate cinaedi is potentially of great moment, particularly when it takes the form of hypermasculine costume and demeanor; for this could signal a subculture fundamentally repositioning itself in relation to the dominant paradigm.12 Hypermasculine behavior, in any case, is hardly reconcilable with the goal of concealment.

So it goes: a community denied, its voices hushed, and all in the name of “culture.” The problem with this book is a chronic poverty of imagination. The problem arises already in the introduction, where Williams announces: “[W]hat never changed is the ideology [of manhood], and that is precisely the main concern of my study” (p. 13). To be sure, some alert readers may well surmise that his working methods predispose Williams to reconstruct a static ideology, however intrinsically improbable such a result may seem (what, no change at all, in half a thousand years? how to explain such torpor?).

Still, for the sake of argument, grant the point: the ideology of manhood held steady. Nonetheless, during this very long period the social context certainly did not remain unchanged, and in particular the city of Rome did not. In the period that Williams studies, Rome grew from a town of several tens of thousands into a sprawling world capital, by far the largest Western city before the nineteenth century. The unswerving ideology thus had to confront enormous social challenges. Within this teeming and diverse metropolis, how likely is it that a dominant culture or ideology had any necessary impact on individual conduct, especially in the absence of systematic social reinforcement through organized religion, general educational institutions, and the law?13

Rome provides, one might have supposed, an almost ideal test case for examining the part that large cities have historically played in the emergence of homosexual communities and subcultures. Is it correct to believe that the inevitable interstices of urban life at Rome afforded social space for “lifestyles” radically at variance with moral tradition? In any case, the cinaedi were scarcely alone in their challenge; thus, Williams discusses at length the notorious sexual ambiguity of the great Roman knight Maecenas (pp. 157-159), but fails to adduce the rich scholarship on otium equestre, the deliberate choice of wealthy men to forgo the ordeal of public ambition.14 Conversely, is it fanciful to suggest that the golden age of the cinaedi may have occasioned a perceptible sharpening in the Roman ideology of public manhood, with consequence especially in the squalid prescriptive codes of the second century A.D.?15

Four centuries earlier, the cinaedi had arrived on the Roman scene as a deep shock to orthodox sensibilities (see just the outrage of Scipio Aemilianus, p. 23); and, as the diatribes of a Seneca or a Juvenal show, the shock never wholly abated. Whether these “gender deviants” are more properly esteemed as “culture heroes” must forever remain a matter of individual taste, although I, at least, would be honored to plead their case; for, on virtually anyone’s reckoning, the cinaedi are just far more interesting than those stolid cretins they regularly offended.16 We require, therefore, an account of why the cinaedi so assiduously pursued their confrontational course of behavior, and also of why we should not accept the easy inference that their conduct expressed a vital element of what we would call their sexuality. In any event, it is inherently implausible that their social and cultural impact was anywhere near as slight as Williams implies.

This book, which will be widely consulted and cited, has much to recommend it. Williams provides a good bibliography,17 and his index of sources is helpful; by contrast, the general index is inadequate (e.g., no entry for Laronia). Beyond this, his forthright deliberation is welcome, even as he clears away much of the prejudicial dead thought that has long encumbered scholarship on this difficult theme. Nevertheless, it is unlikely that Williams has provided the final word on any subject he discusses. Rather, more modestly, he has prepared the ground for much future research, although this in itself is a signal service. But what remains to be seen is whether Roman social historians will soon or ever be able to dispense with the concept of homosexuality.18


1. A caution here: the cinaedi are not described as transvestites, and “effeminate” just means non-masculine (on the assumption of a sharp gender dichotomy), though they derive elements of their accoutrement from women. See H. Herter, “Effeminatus,” RAC 4 (1959) 620-650.

2. See B.J. Brooten, Love Between Women: Early Christian Responses to Female Homoeroticism (Chicago and London, 1996) 232-259, 264-266. Jewish and early Christian views are not considered in the book under review.

3. Social constructionism, a form of antirealistic relativism, holds that the objects of our knowledge are either wholly or partly constituted by our coming to know them in the way we do. See J. Searle, The Construction of Social Reality (New York, 1995); F. Collin, Social Reality (London and New York, 1997); I. Hacking, The Social Construction of What? (Cambridge, Mass., 1999). A good example is money: a ten-dollar bill is an intrinsically worthless piece of paper, but it acquires value (“social meaning”) because we collectively experience money as having value and so come to attribute value to it. Is “homosexuality” (the psychological predisposition toward sexual relations with persons of one’s own sex) socially constructed in exactly the same way as money? The usual answer to this question is no; although the term “homosexual” is modern and the associated identity largely modern, the predisposition itself is a factual substrate of uncertain but non-voluntary (and probably non-social) origin. See M. Ruse, Homosexuality: A Philosophical Inquiry (Oxford, 1988); E. Stein, The Mismeasure of Desire: the Science, Theory, and Ethics of Sexual Orientation (Oxford and New York, 1999). To be sure, historians find it hard in practice to observe these distinctions. In any case, time has not dealt kindly with John Boswell’s “essentialist” claims that large numbers of ancient males were “gay.”

4. S.O. Murray, “Machismo, Male Homosexuality, and Latino Culture,” in Latin American Male Homosexualities (Murray, ed.; Albuquerque, 1995) 49-70.

5. An important exception is Paul Veyne, “L’Homosexualité à Rome,” Communications 35 (1982) 26-33; see also C.A. Williams, “Greek Love at Rome,” CQ 45 (1995) 517-539.

6. Dichotomies of this sort are common in social constructionism. Compare D.M. Halperin, “Historicizing the Subject of Desire: Sexual Preferences and Erotic Identities in the Pseudo-Lucianic Erôtes,” in Foucault and the Writing of History (ed. J. Goldstein; Oxford, 1994) 19-34, supporting “Foucault’s proposition that sexuality is not lodged in our bodies, in our hormones, or in our genitals, but resides in our discursive and institutional practices as well as in the experiences which they construct” (p. 33). But we are scarcely obliged to choose between these stark alternatives, each of which, in real life, contributes to creating a complex social world; and in any case “institutional practices” rapidly fall away in Halperin’s writings, as earlier in Foucault.

7. Compare the methodological observations in S.O. Murray, American Gay (Chicago and London, 1996) 5-9. As Murray observes (p. 6), “[W]hile ideas matter (especially conceiving possible alternatives to social orders masked as ‘natural’ ones), they don’t matter all that much.” Of course, cultural discourse can and does play other important social roles.

8. A similar point is made by J. Escoffier, American Homo: Community and Perversity (Berkeley, Los Angeles, and London, 1998) 173-185, who notes that recent “queer theory,” after deconstructing and evacuating the “social,” has then reinserted the “cultural” in the same place.

9. A. 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) 523-573; R. Taylor, “Two Pathic Subcultures in Ancient Rome,” Journal of the History of Sexuality 7 (1997) 319-371. On subcultures, see the survey in S. Thornton and K. Gelder, eds., The Subculture Reader (London, 1996); classic is D. Hebdige, Subculture: The Meaning of Style (London and New York, 1979).

10. Seneca, Contr. 7.4.7, quoting Calvus: “digito caput uno / scalpit. quid credas hunc sibi velle? virum.” Compare Plutarch, Pompey 48.7 (Clodius, also of Pompey), and Juvenal, 9.130-133. (Other sources at p. 357 n.347.) For what it’s worth, Pompey’s hairdo was rather ordinary.

11. See esp. S. Treggiari, Roman Marriage: “Iusti Coniuges” from the Time of Cicero to the Time of Ulpian (Oxford, 1991) 205-228; S. Dixon, The Roman Family (Baltimore, 1992) 61-71.

12. I am thinking here, of course, of the emergence of the “clones” in the 1970s. See L. Humphreys, “New Styles of Homosexual Manliness,” Transaction (March, 1971) 38-65; M.P. Levine, Gay Macho: The Life and Death of the Homosexual Clone (New York and London, 1998).

13. On law, see pp. 119-124, 193-195. Beyond noting that legal sanctions against homosexual behavior went generally unenforced (no “witch-hunts,” p. 195), Williams has little new. Against the sources, he interprets the mysterious Lex Scantinia as directed also at heterosexual forms of stuprum (p. 120), apparently only because he refuses to believe that the Romans ever recognized homosexual behavior as a distinct category. D. Dalla, Ubi Venus Mutatur: Omosessualità e Diritto nel Mondo Romano (Milan, 1988), is much preferable on legal matters. Williams makes a sharp point, however, in suggesting (p. 224) that the absence of sustained persecution may actually have slowed the social development of the cinaedi.

14. C. Nicolet, L’Ordre Equestre à l’Epoque Républicaine vol. I (1966) 457-464, esp. 459; J. Bleicken, Cicero und die Ritter (Göttingen, 1995) 63 n.128, with further bibliography; and generally J.-M. André, L’Otium dans la Vie Morale et Intellectuelle à Rome des Origines à l’Epoque Augustéene (Paris, 1966).

15. See M.W. Gleason, Making Men: Sophists and Self-Presentation in Ancient Rome (Princeton, 1995), and esp. “The Semiotics of Gender: Physiognomy and Self-Fashioning,” in Before Sexuality: The Construction of Erotic Experience in the Ancient Greek World (ed. D.M. Halperin, J.J. Winkler, and F.I. Zeitlin; Princeton, 1990) 389-415 (an important article that Williams misses).

16. This is also the point of Hortensius’ famous retort when accused of effeminacy (Gellius, NA 1.5.2-3, discussed at pp. 155-157).

17. But add esp. E. Meyer-Zwiffelhoffer, Im Zeichen des Phallus: Die Ordnung des Geschlechtslebens im Antiken Rom (Frankfurt and New York, 1995); L. Hermans, Bewust van Andere Lusten: Homoseksualiteit in het Romeinse Keizerrijk (Amsterdam, 1995); and H.P. Obermayer, Martial und der Diskurs über Männliche “Homosexualität” in der Literatur der Frühen Kaiserzeit (Tübingen, 1998).

18. See J. Thorp, “The Social Construction of Homosexuality,” Phoenix 46 (1992) 54-61; J. Black, “Taking the Sex Out of Sexuality: Foucault’s Failed History,” in Rethinking Sexuality: Foucault and Classical Antiquity (ed. D.H.J. Larmour, P.A. Miller, and C. Platter; Princeton, 1998) 42-60.