BMCR 1998.10.16

Roman Sexualities

, ,

This collection makes an important contribution. Its laudable aim is, in Skinner’s introductory words, to demonstrate that “Roman constructions of sex should constitute a discrete research area within the general field of ancient sexuality” (p. 3). Thus Roman Sexualities brings together essays by twelve scholars by way of contributing to the ongoing process of evening out the balance in the scholarship on Greek and Roman sexuality. Some of the essays are extremely interesting and useful; others have their weaknesses; but on the whole the collection is certainly worth reading. Here I offer my reactions to the individual essays seriatim.

Marilyn Skinner’s introductory piece exposes some of the assumptions unifying the collection, and at the same time Skinner articulates her own perspective on the larger picture of Roman sexuality, candidly admitting that when she does so “neither my co-editor nor any of my fellow contributors would agree with every claim I make” (p. 3). While, understandably enough, it is not always clear when Skinner is speaking for herself or for all, she does mark one basic assumption shared by all the contributors. Having begun by following Halperin, Winkler, and Zeitlin in their definition of “sexuality” as “the cultural interpretation of the human body’s erogenous zones and sexual capacities,” and having noted that Roman sexual ideologies share with those of the Greeks a marked tendency to represent sexual practices in terms of the antithesis between active and passive, dominant and subordinate, penetrating and penetrated (pp. 3-4), Skinner continues: “The present collection presupposes that … historical and social contingencies were projected onto the dominance-submission grid of Roman sexuality, creating documents in which ostensibly crude sexual narratives serve as an ordered semantic system for articulating social anxieties” (pp. 4-5). As for what some of these social anxieties might be, Skinner makes an argument reminiscent of Paul Veyne’s 1978 “La famille et l’amour sous le Haut-Empire romain”: that between the first century B.C. and the beginning of the third century A.D. “changed conditions of political existence are mirrored in the literary record of the educated elite, who grew more and more preoccupied with preserving personal autonomy and honor in an atmosphere of constraint” (p. 4).

Here my first queries arose. Does not the proffered definition of “sexuality” leave unexplored important questions, namely, what we mean by “sexual capacities” and what ” the erogenous zones” are? Next, the opposition between the terms “sexual” and “social” seems to imply that the two are ontologically distinct spheres of reality, only put in contact by metonymic means (i.e., by the agency of human discourse); but it is an argument at least worth considering that the concepts of the “sexual” and the “social” are themselves terms unique to human discourse, and that the tendency to identify them as distinct spheres of thought and experience, susceptible to being brought into relation by various conceptual and linguistic means, is itself an effect of specific ideological traditions.

Still, as working definitions, these form a useful starting point. Skinner’s subsequent overview of the scholarship on ancient sexuality is helpful and balanced, as is her bibliographically annotated sketch of some differences between Greece and Rome on questions of gender and sexuality. Finally, after offering her perspectives on the contributions made by the individual essays, Skinner concludes with some questions for further exploration.

The collection’s twelve essays are organized into five thematic units: “Unmarked Sexuality,” “Wayward Sexualities,” “Gender Slippage in Literary Constructions of the Masculine,” “Male Constructions of ‘Woman’,” and “Female Construction of the Desiring Subject.” The first category is represented by one essay, Jonathan Walters’ “Invading the Roman Body: Manliness and Impenetrability in Roman Thought,” whose central argument is that “the Roman sexual protocol that defined men as impenetrable penetrators can most usefully be seen in the context of a wider conceptual pattern that characterized those of high social status as being able to defend the boundaries of their bodies from invasive assaults of all kinds” (p. 30). Although Walters does not signal the fact, much the same thesis has already been presented for Greece (above all by Halperin 1990 and Winkler 1990). Indeed, Walters’ essay is marked by a noticeable lack of reference to previous scholarship on any number of points; even when one finds acknowledgment of predecessors, it is spotty. Thus, for example, we read of “the characterization, widespread in the public discourses of the Greco-Roman world, that sex is a one-way street, something one person does to another” (p. 30), but we find no citation either of ancient evidence or of modern scholarship on this “widespread” conceptualization, although Halperin 1990 and Winkler 1990 again deserve citation here. On p. 31 Walters writes that “in Roman society, the only culturally condoned male homosexual relationship was that between a sexually active adult citizen man and a sexually passive male, usually younger, who was a slave, an ex-slave, or a noncitizen” and adds in a footnote, “See, for example, Gonfroy 1978 and Verstraete 1980,” neglecting the considerable amount of work that has been done on this point since the publication of those two articles. Here a reference to Richlin’s and Cantarella’s books at the very least is called for. Walters’ brief discussion of the controverted lex Scantinia (pp. 34-5) cites only two works of scholarship (Berger’s 1953 Encyclopedic Dictionary of Roman Law and Lilja 1983), but there have been several important discussions of the law both before and after Lilja, including Boswell’s 1980 Christianity, Homosexuality, and Social Tolerance, Dalla’s 1987 Ubi Venus mutatur: Omosessualità e diritto nel mondo romano, Cantarella’s 1988 Secondo natura: Bisessualità nel mondo antico, and Richlin’s 1993 “Not Before Homosexuality.” When Walters goes on to stress the crucial difference between free-born and slave boys in connection with the law (pp. 35-6), his discussion, citing no previous scholarship, might leave non-specialist readers with the impression that he is proposing a novel hypothesis. That is hardly the case: see Veyne 1978, Cantarella 1988, Fantham 1991, and Williams 1995 (all cited in the collection’s bibliography and most in Skinner’s introduction, but none in this essay). Finally, Walters’ discussion of masculine impenetrability fails to cite a basic term of Roman discourse on this point, pudicitia (for which see, e.g., Fantham 1991 and Richlin 1993), and leaves unexplored the nuances of another important term, stuprum (for which see especially Fantham 1991 and Williams 1995), glossing it simply as “illicit sexual activity” (p. 34).

This lack of focus on the Romans’ own vocabulary of sexuality is paralleled by the essay’s tendency to offer broad comments on Roman sexual discourses without the benefit of support from the ancient sources. Thus on p. 31 we read of “the penetrator-penetrated relationship as ‘naturally’ involving a more powerful individual wielding power over a less powerful one,” and similarly on p. 33 we read that the praetextatus is “a young male who is not a full-fledged man, and is therefore ‘naturally’ an object of sexual desire to adult males.” In both instances we look in vain for any textual evidence, and the quotation marks in both formulations betray a basic problem: the rhetoric of “nature” is precisely that, a complex rhetorical tradition in need of careful interrogation. Which ancient texts identify which sexual practices as “natural” and “unnatural”? And why? These crucial questions are passed over here; the reader ought to be referred at least to Boswell 1980, Foucault 1986, and Winkler 1990 for some helpful introductory discussion of “nature” and ancient sexual discourse.

On p. 32 Walters makes a crucial distinction between “males” and “men”: “Not all males are men, and therefore impenetrable.” In particular, he refers to the special nuance of the term vir, which “does not simply denote an adult male; it refers specifically to those adult males who are freeborn Roman citizens in good standing, those at the top of the Roman social hierarchy” — those who are “sexually impenetrable penetrators” (p. 32). Again, though, one misses references either to the source material or to previous scholarship on this point (see, e.g., Richlin 1993). Moreover, while Walters is certainly on to something here, if one turns to the attestations of the noun vir, one will find it occasionally applied to non-Romans, to low-born citizens, and, albeit in a qualified way, to some outstanding embodiments of all that is contrary to Roman masculinity (cf. Phdr. App. 10.20, where a cinaedus is called, perhaps sarcastically, illum virum, and note the almost oxymoronic phrase virum mollem at Mart. 1.96.10, 3.73.4); even the castrated priests of Cybele are called steriles viri at Mart. 3.91.5. Moreover, on p. 33 we read that “both in the modern West and in ancient Rome, homosexual activity between men is commonly described in public discourse in a framework that assimilates it to heterosexual activity, reading one partner as ‘male’ and the other as ‘female’ …” Here Walters seems to ignore his own warnings about the language of masculinity: does he mean sexual activity between “men” or between “males”? And is it “male” and “female” or “masculine” and “feminine”? Walters’ formulations, while offering helpful glimpses into Roman ideologies of masculinity, seem insufficiently nuanced; and the generalization regarding “the modern West” ignores the noticeable tendency in American popular discourse, at least, to view all gay men as effeminate or at best of questionable masculinity.

Walters’ final arguments on the Roman relationship between beating and sexual penetration (pp. 37-42) constitute the most original contribution of this essay. He first points to an interesting conceptual parallelism: the bodies of Roman citizen males were ideally exempt both from being beaten and from being sexually penetrated, while slaves were susceptible to precisely those two things, as were citizen males found guilty of, for example, adultery. “Sexual penetration and beating, those two forms of corporeal assault, are in Roman terms structurally equivalent” (p. 39). This is a helpful insight, although again it ought to be noted that Winkler 1990: 48 and Halperin 1990: 96 make these very points about Athens. Walters goes on to make the interesting observation that the Roman soldier was liable to being beaten by his superiors in the interests of military discipline, although only with a vine staff and not, for example, a birch rod. Walters then points to the problem of the soldier’s wounds, which, although constituting a violation of his body, could be construed as a badge of honor and mark of masculine prowess. But Walters’ attempt to bring together these insightful observations creates some confusion. “A man,” he writes, “can without losing his superior status be penetrated by a sword or a vine, but not by a penis or a birch” (p. 40). It seems to me one thing to detect a symbolic parallelism or even equivalence between beating and penetrating, but quite another to speak of being “penetrated” by a vine or birch. Still, Walters’ final suggestion is helpful indeed: he argues that the anecdote regarding Marius’ soldier, who was held to have justifiably killed his superior when the latter made a sexual advance on him, reveals an anxiety about the sexual penetrability of soldiers that was especially intense precisely because soldiers could be subject to corporal punishment by their superiors.

Holt Parker’s essay “The Teratogenic Grid” introduces the section entitled “Wayward Sexualities,” and begins with a point that an ever-increasing number of scholars, following in the footsteps of Dover, Veyne, Foucault, and others, have insisted upon: that the contemporary Western division of “categories of individuals” into “heterosexuals” and “homosexuals” is “a rather parochial affair and a comparatively recent development even in the culture of the West” and that, by contrast, “the ancient world, both Greek and Roman, did not base its classification on gender, but on a completely different axis, that of active versus passive” (p. 47). In this regard Parker reminds his readers that “it is very difficult for us to ignore our own prejudices and realize that what may be literally a matter of life and death in our culture would have been a matter of indifference or bewilderment to the Romans” (p. 48). Here, as with Walters’ assertion regarding “the modern West,” there is what seems to me an overly broad generalization about what “our own prejudices” are. In any event, by way of support for the argument that what mattered in antiquity was not the distinction between homosexual and heterosexual but the opposition between active and passive, Parker offers the grid (p. 49) from which his essay takes its title. In it he presents key terms of the Latin sexual vocabulary, both verbs ( futuere, pedicare, irrumare, etc.) and agent nouns ( fututor, pedicator, irrumator, etc.), organized around two axes: the opposition between active and passive on the one hand, and the mode of penetration (vaginal, anal, and oral) on the other. I find the notion of schematizing the vocabulary in this way extremely helpful, though not “novel” (as Skinner characterizes it in her introductory essay, p. 15), as I suggested the creation of a similarly organized chart in my 1992 dissertation.

But, while the construction of such a chart reveals a great deal about the nature of Roman sexual discourses, it seems to me that Parker does some peculiar things with his grid. First, on p. 49, Parker argues that “the puella or femina (i.e., the normal/passive female) has open to her exactly three possible sexual passivities: to be fucked in the vagina, the anus, or the mouth.” And yet in his grid the terms femina/puella appear in the slot for the vaginally passive female, while the anally passive female is identified as pathica and the orally passive female as fellatrix; Parker concludes that “a woman is defined as ‘one who is fucked in the vagina'” (p. 49). Apart from the jarring juxtaposition of obscene and technical language in this reproduction of a Roman “definition,” I am struck by the inconsistent usage of the terms femina and puella : they designate both the “normal/passive” female, penetrable in three ways, and specifically the vaginally penetrated female. Nor is it clear what exactly “a fellatrix” or “a pathica” is. Do these terms refer to unchanging, essential identities? How does a woman who is penetrated in more than one mode fit into Parker’s grid?

Second, I find especially peculiar Parker’s decision to fill the slot “vaginally passive male” with the term cunnilinctor. Apart from the fact that this is not a Latin word (the Roman term for a man who performs the act is cunnilingus), Parker’s argument that “for a man to commit cunnilingus is to be fucked by a woman” (p. 51) is not supported by the sources he cites. There is certainly an abundance of ancient evidence condemning cunnilingi, and it is selectively cited by Parker (who does not, however, refer to the authoritative compilation in W. Krenkel, “Tonguing” [ WZR 30 (1981): 37-54]). As Krenkel, Richlin, and I have argued, such men are in the first instance condemned by Roman writers on the grounds that they are filthy or impure ( impuri, a term discussed in some detail by Richlin); another factor that I would emphasize (and alluded to by Parker, pp. 51-2) is the fact that, in the relentlessly phallocentric sexual ideologies of classical antiquity, this act, not requiring the intervention of a penis, is by its very nature anomalous. Yet Parker wants to show that cunnilingi are in Roman terms vaginally passive men. What is his evidence? First, he cites “Martial’s outrage in 11.61, where he attacks a man so passive that his mouth becomes a cunt for a cunt” (p. 52). While this is an intriguing image, it is nowhere to be found in Martial’s epigram, where, on the contrary, we read of Nanneius’ lingua fututrix (11.61.10). Far from being passive, then, Nanneius is inappropriately active, using his tongue to penetrate his female partners. Next, Parker cites Martial 7.67 (further discussed by Hallett in her contribution to this volume), where the tribad Philaenis is so masculine that she penetrates boys and girls (1-3), refuses to perform fellatio because she deems it “unmanly” ( parum virile, 14), and yet paradoxically performs cunnilinctus, thinking it “manly” ( cunnum lingere quae putas virile, 17). The narrative voice roundly condemns Philaenis, and Parker explains the narrator’s perspective thus: “What Philaenis fails to realize is that cunnilingus is equally passive, equally an act of being penetrated” (p. 53). Now cunnilinctus is indeed implicitly “unmanly” in this text, but in what sense is it represented as “passive”? One looks in vain, for example, for the language of patientia in this poem. In fact, the text describes Philaenis’ behavior with a verb in the active voice: sed plane medias vorat puellas (15). It is hard to see how Philaenis’ practice of “devouring girls’ middles” constitutes in Roman terms an “act of being penetrated.” One thinks again of Nanneius’ lingua fututrix.

Parker’s insistence that “vaginally passive” men are those who perform cunnilinctus leads him to an especially infelicitous formulation. On p. 59 he claims that “a woman cannot fuck a man except by forcing him into cunnilingus,” ignoring a passage from Seneca, discussed by two other contributors to Roman Sexualities, where we read of noticeably masculine women who penetrate men ( viros ineunt, Sen. Epist. 95.21); we might also compare Martial’s epigram on Philaenis that is cited, as we have just seen, by Parker himself: Philaenis anally penetrates boys ( pedicat pueros, 7.67.1; Parker inexplicably speaks of her “attempts at” penetration [p. 53]). Similarly peculiar is Parker’s assertion that the pathicus will “delight not only in being the victim of men, being penetrated by a penis in the anus or mouth, but also in being the victim of women, primarily by providing cunnilingus but also by being used, as it were, as a dildo with his penis (Mart. 5.61)” (p. 57). The reading of Martial 5.61 seems strained (certainly there is no hint of the imagery of a dildo), and in any case, as far as I can tell, there are no extant sources in which a man identified as a pathicus is said or even implied to enjoy performing cunnilinctus. Parker cites Martial 4.43 and 10.40, but these epigrams speak of cinaedi, not pathici, and we cannot assume that the terms are interchangeable; in any case it is not at all clear that 10.40 alludes specifically to cunnilinctus, while the point of 4.43 seems to be that the addressee is not a cinaedus but is instead something worse, namely a cunnilingus; there is no hint that he is both simultaneously.

Towards the end of his essay, Parker reiterates a basic and important point: “the cinaedus simply does not correspond to our construction of the ‘homosexual'” (p. 58). I am in full agreement with this argument, as I am with Parker’s next point: “Nor does the Roman hatred of the pathicus correspond to ‘homophobia,’ though again they do share some features. Each is an expression of the fear and hate that the dominant group in any society expends on the ‘Other’ …” (p. 58). Here Parker is largely arguing against Richlin’s 1993 “Not Before Homosexuality,” where it is argued “first, that men identified as homosexuals really existed at Rome and, second, that their existence was marked both by homophobia within the culture and by social and civil restrictions” (p. 530). Thus, for example, we might contrast Parker’s description of Suetonius’ reference to the sexual experiences of the emperor Claudius (p. 55: “Suetonius has no word for ‘heterosexual’; it is not a category he can readily label. Instead he has to describe this peculiarity of taste to his readers.”) with Richlin’s claim that “it would really be fair to say that Suetonius describes Claudius as a ‘heterosexual'” (p. 532). And yet Parker does not acknowledge his divergent reading of the material, indeed giving the appearance that he is fundamentally in agreement with Richlin’s arguments: “Richlin (1993b: 524) rightly defines cinaedi in normal Roman terms as ‘those who like to be sexually penetrated by other men,’ and, I would add, sexually used by women (cf. ibid.: 533)” (p. 61). On p. 533 Richlin indeed mentions the fact that cinaedi could be imagined as having sexual relations with women, but she notes that “we have only invective as evidence for this” and does not integrate this evidence into her explanatory paradigm, which clearly remains informed by the premise that cinaedi were “identified as homosexuals” (p. 530) and were men who “like to be sexually penetrated by other men” (p. 524). In short, when Parker adds his qualification to Richlin’s definition, he seems to be fundamentally altering it even while deferring to its authority.

The collection’s third essay is Catharine Edwards’ “Unspeakable Professions: Public Performance and Prostitution in Ancient Rome,” a careful discussion of a far-reaching question: why were actors, gladiators, and prostitutes so often treated alike in both moralizing and legal texts? Edwards’ answer, in short, is that “they lived by providing sex, violence, and laughter for the pleasure of the public-a licentious affront to Roman gravitas” (p. 67); they “were the objects of other people’s desires” who “served the pleasure of others” and “were tarnished by exposure to the public gaze” (p. 68). Overall, Edwards’ discussion of the evidence is as illuminating as it is convincing. There are only a few points on which one might wish for a little more clarity. First, with regard to male prostitution, Edwards notes the reference in the Fasti Praenestini to “a holiday for male prostitutes” (p. 82); but if we take seriously (as we should) Walters’ warning that we should be careful about distinguishing “men” from “males,” we should observe that the text speaks specifically of boy prostitutes ( pueri lenonii). Next, with regard to gladiators, Edwards nicely summarizes their ambiguous image: they are hypermasculine heroes, imagined as being highly desirable to women, yet often despised (pp. 77-8). But it seems to me that Edwards does not fully integrate into her analysis those sources that disparage gladiators not as hypermasculine but rather as effeminate (e.g., Sen. N.Q. 7.31.2-3; Juv. 2.143-8). She alludes to the problem in n. 47, where she suggests that only the retiarius was liable to being represented as effeminate, and yet in a statement summarizing the overall image of all three tarnished professions she makes the following argument: “All these bodies are the objects of uninhibited public gaze. Subordinated to the desires of others, these infamous persons are assimilated to the feminine and the servile, unworthy to be fully Roman citizens” (p. 85). How do we reconcile this with the statement that gladiators are “emblems of an aggressive masculinity” (p. 78)?

The next essay brings us into the section on “Gender Slippage in Literary Constructions of the Masculine.” Anthony Corbeill writes on “Dining Deviants in Roman Political Invective.” This essay sets out to describe the ways in which Roman images of decadent feasting relate to questions of gender and sexuality: these images are informed not only by notions of “moral profligacy and extravagant feasting,” not only by xenophobic stereotypes regarding Greece and the East, but also by “Roman concerns about the nature of the masculine self,” for “the effeminate male actively participates in the banquet’s debauchery” (p. 99). Corbeill makes a number of good points, discussing, for example, the way in which the Roman vocabulary for financial profligacy draws on the language of decadent feasting ( decoquere, comesse and the like), and in his discussion of Roman antipathy to dancing stressing that the word cinaedus originally denoted a type of dancer.

Once he starts theorizing about effeminacy and sexuality, though, Corbeill is rather less convincing. He begins (p. 108) by looking for an explanation for “why Roman society fixated on the fear of effeminacy” (perhaps an overstatement in itself), and he turns to ancient scientific and medical thought, discussing Greek medical writers’ theories about the dangers of men becoming soft and effeminate. Yet, while medical theories of gender may well have exerted influence in many spheres, it seems inappropriate to seek a causal explanation for the problematization of effeminacy in so limited a body of material. Highly polarized gender systems such as those characterizing ancient Greek and Roman cultural traditions clearly derive their strength from fundamental structural oppositions (the masculine is what it is because it is not feminine), and the relevant ideological systems just as clearly exert tremendous pressure in all areas of public discourse in the interests of maintaining relevant boundaries and distinctions. It seems to me that we should not turn to any one area of discourse (e.g., medical writings) for the explanation for hostility to those who violate gender categories.

With regard to Roman references to effeminate males, Corbeill makes a rash assertion: “In late Republican oratory, effeminate qualities imply passive homoerotic activity” (p. 109). It is a pity that Corbeill fails to summarize in one place precisely which qualities are represented as effeminate in Roman texts: a reference to the encyclopedic overview in H. Herter, “Effeminatus” ( RAC 4 [1959]: 620-50) would have been welcome. Even more problematic is the fact that among the stereotypes handed down to us by the Roman oratorical tradition is that of the effeminate womanizer (as is noted by, e.g., Richlin 1993 and Edwards 1993). One thinks of a Republican orator’s description of an effeminate young man sashaying about “more softly than women in order to please women” ( incedentem ut feminis placeat femina mollius, Sen. Contr. 2.1.6, cited by Corbeill in n. 42 as an unexplained “interesting possibility”); one also thinks of Cicero’s description, likewise cited by Corbeill (p. 119), of the effeminate followers of Catiline at Cat. 2.22, explicitly said to include adulteri (shortly thereafter, in a passage Corbeill does not quote, Cicero tweaks them for their dependence on their mulierculae [2.23]); and finally one thinks of a passage most apposite to Corbeill’s discussion but unfortunately not cited by him, where Cicero directs withering scorn at an effeminate Verres and son in the company of loose women at decadent dinner parties (Cic. Verr. 2.5.81). To be sure, Corbeill does note the existence of ancient sources that “depict men engaging in both dominant and subservient sexual roles without any notion on the accuser’s part that the charge seems to raise logical problems” (p. 112), but he does not integrate them into his explication, apart from noting that “the theoretical possibility that a man could lose his gender has opened up a legitimate space for invective,” a fairly unclear explanation (and does an effeminate man really “lose his gender”?). In the end, Corbeill seems committed to the view that “effeminacy connotes a specific sexuality,” and what that “specific sexuality” is becomes clear from the next sentence, where he alludes to “the precision with which the Romans defined the sexually passive male” (pp. 116-7).

Corbeill seems intent on arguing against Foucault on the question of “homosexuality,” but again he fails to convince. He summarizes Foucault’s outlook thus: “Hence for the Greeks, at least, it is claimed that activity we in the twentieth century would label ‘homosexual’ was simply a ‘set of acts’ in which the sex of the person penetrated is irrelevant” (p. 116). There seems to be an oversimplification of Foucault’s ideas here, made all the more ironic by the fact that Corbeill refers to a review of Halperin and Winkler which claims that those scholars “seriously oversimplified Foucault’s ideas” (p. 126 n. 49). Pace Corbeill, Foucault and others like him are not as much interested in the “activity we in the twentieth century would label ‘homosexual'” as they are in identities (e.g., the cinaedus as opposed to “the sodomite” as opposed to “the homosexual”). Moreover, when Corbeill claims that according to Foucault for the Greeks “the sex of the person penetrated is irrelevant,” he blurs the crucial questions: irrelevant for whom, to what, and in which contexts? Corbeill continues: “Certainly [Foucault’s] assertion that, until recent centuries, the ‘sodomite’ was not defined by ‘a kind of interior androgyny, a hermaphrodism of the soul’ finds clear refutation in the exegesis of Seneca quoted above, as well as in many of the texts I have cited throughout this section” (p. 114). Again the criticism falls flat. Seneca ( Epist. 52.12) does not and cannot provide such a refutation: neither he nor any other ancient author writes about the “sodomite,” a category of sexual identity that did not exist until later in European history. Furthermore, while it is certainly true that Seneca is interested in behaviors that reflect a person’s moral constitution (he speaks of an argumentum morum), he makes no allusion to a “hermaphrodism of the soul” (and one wonders which Latin phrase could be so translated in any case). Finally, Corbeill’s note on this passage (n. 52) adduces another quotation from Foucault: it is “completely incorrect to interpret [these traits] as a condemnation of … what we generally refer to as homosexual relations.” Although Corbeill implies that he is still refuting Foucault, it seems to me that Corbeill has actually supported Foucault’s argument. The ancient sources he has surveyed do not criticize homosexual relations per se but rather those men who desire to be penetrated. Indeed, writing of same-sex practices in Roman invective, Corbeill himself earlier argues that “the dominant partner does not ever seem to have been the direct object of abuse for playing the active role” (p. 110).

Corbeill’s apparent belief that Seneca writes about “sodomites” and “hermaphrodism” brings me to the problem of translation. On p. 115 Corbeill translates impudicus as “effeminate” and on p. 113 as “shamelessly effeminate.” I am curious about the justification for adding the adverb in the second instance, but above all I question the translation of the adjective itself. How is impudicus different from effeminatus (which is also translated as “effeminate” on p. 115) or mollis or semivir or cinaedus or pathicus ? As for the last two, it is odd to see impudicus and cinaedus both translated elsewhere not as “effeminate” but as “pathic” (pp. 127 n. 58 and 114 respectively); are impudicus, cinaedus, pathicus, “effeminate,” and “pathic” all identical? It seems to me that explorations of Roman ideologies of gender and sexuality need to be more attentive to the native linguistic framework within which those ideologies were propagated. Still, Corbeill’s article does us the service of demonstrating one way in which the Roman textual tradition links effeminacy with a variety of disapproved behaviors, not all of them sexual.

Marilyn Skinner’s essay ” Ego mulier : The Construction of Male Sexuality in Catullus” offers a discussion of the ways in which Catullus’ poetry creates for the male reader a “sympathetic engagement with fictional portrayals of the other sex, to a degree that allows even the sacrosanct values of epic to be called into question” (p. 130). Rightly cautioning against extrapolating from the complicated stance of the Catullan persona to an “actual shift in emotional attitudes” among young Roman men of his day (p. 130), Skinner aims to contribute to the ongoing discussion of the ways in which the personae of Catullus and other Roman poets flirt with the feminine, with an eye to seeking “analytic clues to the symbolic structures” of Roman culture. By way of highlighting the dangers of collapsing the Catullan persona into the historical figure Catullus (dangers complicated by the ancient tendency to identify the narrative voice with the author himself [p. 132]), she begins with a poem that has not even the veneer of autobiography, namely the Attis poem.

First, she revives an interpretation suggested by Quinn, namely that the story of Attis, by treating a much-courted young man who ends up castrating himself Veneris nimio odio (Catull. 63.17), underscores the necessity for young men to make the normative transition from passive, pursued boy to active, pursuing man: an interesting hypothesis which, if correct, would have important implications for our understanding of the psychosocial dynamics of Roman masculinity. Next, in line with her interest in conceptual relations between the sexual and the social, Skinner argues that another way of reading the Attis poem is to perceive “that the monstrous inversion of gender relations contained in the asymmetrical partnership of minax Cybebe… and her emasculate consort Attis reflects elite alarm over perceived restrictions on personal autonomy and diminished capacity for meaningful public action during the agonized death throes of the Roman Republic” (p. 142). Through this interesting but admittedly speculative reading (“let us try,” she suggests on p. 141), Skinner sees in Catullus 63 “a contemporary narrative of political impotence … retold as a myth of self-destructive estrangement from the male body” (p. 142). Turning to the Lesbia poems, Skinner invites us into another interesting “thought experiment” (p. 143) bringing together the political and the sexual. Since Lesbia is a highborn Roman matron and Catullus only a member of the provincial elite, “their affair is tantamount to a client relationship, with Lesbia playing the de facto role of patron and Catullus occupying the subordinate place of lesser amicus…. One thematic objective of the Lesbia poems, it would seem, is to connect the frustration of ambition among Catullus’s peers with aristocratic venality by encoding exploitative manipulations of the patronage system within the most privileged circles as a noblewoman’s greedy pursuit of sexual adventures” (p. 144).

Finally, Skinner asks the big question: What we are to make of this poetry in which a male poet toys with a feminized persona? Here she explicitly parts ways with scholars like Richlin who tend to view such co-opted images of the passive, abandoned woman as pornographic objects put on display for the male gaze; instead, Skinner invokes Zeitlin’s phrase “playing the other” to suggest that we might rather view these feminine personas “as alternative subject positions permitting scope for emotive fantasy,” and thus that “although their passions were gendered ‘feminine,’ a male reader was expected to discharge his own repressed feelings through sentimental involvement in the character’s predicament” (p. 145). Although some might be uncomfortable with the psychologizing, and I for one would rather say “invited” than “expected,” the line of inquiry strikes me as both interesting and fruitful.

Ellen Oliensis’ “The Erotics of amicitia : Readings in Tibullus, Propertius, and Horace” provides equally thought-provoking discussion of two themes in those three poets: the relationship between the persona of the client and that of the lover, and “the cross-fertilization of friendship by sexual love, amicitia by amor” (p. 155). There are any number of valuable insights to be found in this essay. For example, observing the central importance of obsequium to both lover and client, Oliensis reminds us that the slavery of the lover at least is in the end a fiction: “The ease with which the poet-lover typically deploys this rhetoric suggests that his abjection is less than absolute. The poet plays the slave within a fiction of his own masterful making. In this sense, the power of the beloved is always his gift, and dependent on his willingness to keep playing” (p. 152). And when it comes to the “erotics of amicitia,” Oliensis is especially sensitive to the ways in which bonds are created between male amici in such a way as to construct a shared erotic experience, whether by way of a triangle (Messalla, Tibullus, Delia; Gallus, Propertius, and a beloved boy or girl) or by way of an “erotic/lyric exchange” between two male friends (Calvus and Catullus, Horace and Maecenas). This essay, among the best written in the collection, is notable for its careful argumentation and for its wealth of insights into the works of these poets.

David Fredrick’s “Reading Broken Skin: Violence in Roman Elegy” begins by noting that “this essay situates elegiac violence in the context of two distinct ways of representing the mistress: erotic description, which fashions an incomplete but aesthetically perfect body as poetic metaphor ( candida puella); and jealous suspicion, which produces a degraded body liable to verbal or physical aggression ( dura puella)” (p. 172). Observing the associations between epic (as opposed to lyric) and masculinity, Fredrick proposes that any “recuperation” of masculinity in elegy is “parodic, but not simply funny; for its male authors, elegy’s wounds are ambiguous metaphors for the transformation of elite masculinity into text” (pp. 172-3). His discussion ends by establishing an explicit link, like Skinner’s, between erotic text and social and especially political context: “When it says of the wound, ‘This too is text,’ elegy exposes the semiotic dilemma of the male body defined by a vanishing capacity for political action” (p. 190). In between, Fredrick offers readings of Catullus, Propertius, and Ovid in an essay that needs, and generally repays, a second reading, as it works on a fairly high theoretical level and the language is often dense and allusive — although sometimes perhaps excessively so, as on p. 178: “Callimachus’s echo, ‘Some other man has him,’ is inscribed on Lesbia’s anatomy: some other man has her qua‘her.'” In short, Fredrick’s essay offers a strong dose of theorizing that, while it might put off some readers, will surely appeal to those accustomed to the discourse.

Amy Richlin’s essay on “Pliny’s Brassiere” begins by noting the Hellenocentric bias of much scholarship on ancient medicine, and does us the service of searching the Roman sources (especially Pliny the Elder, Festus, and Columella) for evidence concerning beliefs regarding women’s health issues. The essay is filled with fascinating details, and is informed by Richlin’s characteristic sensitivity to the contemporary political implications of what she is studying. She takes as a test case the elder Pliny’s comment that headaches may be helped by tying a breast-band to the head. Following up on some earlier theorizing, Richlin suggests that we can read Pliny’s recommendation in both a pessimistic and an optimistic way: both as a sign of the ways in which women’s bodies can be appropriated by patriarchy and patriarchs, and (if we ask who gave Pliny the breast-band [p. 205]) as a reminder that even in highly oppressive patriarchies women have acted as agents, “working out their own strategies to deal with whatever system they find themselves in” (p. 200). In the latter connection, she reminds us that certain folk-remedies involving women’s saliva or menstrual blood seem to require women’s active participation (pp. 205-6), and then proceeds to a useful attempt to “recover from Pliny’s Natural History some idea of women’s own health practices in first-century C.E. Rome” (p. 207). There is much food for thought here.

Sandra Joshel’s essay on Messalina begins with a reminder of the ways in which Claudius’ wife has been invoked as “the proper name for uncontrolled female sexuality that can be specified by an adjective or place or race” (p. 222). Joshel is not interested so much in the historical facts about the woman herself as she is in the figure of Messalina as “a fiction for the terms through which the Romans experienced their own present and its history” (p. 222). Thus in Tacitus’ narrative, she argues, “Messalina functions as a sign in a discourse of imperial power that simultaneously informs, if not determines, her image,” and “this paradox is the result of a double move made by Roman imperial discourse: woman figures empire, and yet, at the same time, empire becomes the woman” (p. 223). Although the essay’s attempt to establish an equivalence between Messalina and empire (“woman figures empire” and “empire becomes the woman”) may not convince everyone, there are a number of helpful insights to be found in her discussion of “Tacitus’s construction of [Messalina’s] violence, her excessive desire that produces chaos and emasculates, and her ambiguous voice that moves the narrative but is essentially mute” (pp. 222-3). Certainly, as Joshel argues, Tacitus’ Messalina is a handy test case for the ways in which “the sexual clings to the political” (p. 226).

Judith Hallett’s “Female Homoeroticism and the Denial of Roman Reality” is an unrevised reprint of her ground-breaking 1989 article. It is a pity that the discussion was not updated to reflect the important scholarship in the field that has appeared since 1989; still, it is helpful to have this review of Roman depictions of sexual practices between females made more accessible. Hallett’s thesis is that Roman images of sexual practices between females consistently betray an anxiety not evident in Roman images of sexual practices between males (a significant point indeed), and that Roman images of female homoeroticism are marked by a tendency to “masculinize,” “Hellenize,” and “anachronize” (see, e.g., p. 266 for the triadic formulation), that is, to deny the reality of sexual experience between women as a contemporary Roman phenomenon. That tribades are represented as masculine women is certainly true: indeed, in view of the fact that Martial, for one, imagines a woman he calls a tribas to penetrate both boys and girls, I would argue that the tribas is not defined as a “lesbian” (or, as Hallett suggests, “bull dyke,” p. 256; cf. Gordon on p. 275: tribades are “women who desire other women”) but rather precisely as a masculine woman: a gender-deviant rather than a homosexual.

But the argument that Roman texts deny tribads contemporary Roman reality by associating them with the distant past and with Greece rests on shaky foundations. For example, Hallett observes that in his tale of a drunken Prometheus mistakenly applying penises to some women and vaginas to some men when shaping our species out of clay, “by crediting the origin of tribades to a Greek figure from the remote past, Phaedrus further dissociates females who engage in same-sex love from the actual and contemporary human scene” (p. 256). But the same could be said of molles mares (a phrase here referring to males who enjoy being penetrated) and indeed of the entire human race, likewise created by a Greek figure from the remote past. Next, alluding to a Plautine joke revolving around sexual practices between females, Hallett pointedly alludes to “the attribution of [female homoeroticism] to the Greek world and to an earlier point in time,” (p. 258) but of course Plautine comedy offers jokes on all sorts of subjects, from sexual practices between males to sexual practices between males and females, from the scheming slave to the braggart soldier to the hooker with the heart of gold. All of these, too, are attributed to the Greek world and an earlier point in time. On p. 259 Hallett observes that when Scaurus, quoted by the elder Seneca, makes his naughty remark about the man who found his wife in bed with another woman, he switches to Greek. But a characteristic feature of Seneca’s Controversiae is precisely that it quotes not only Latin but also Greek orators on a variety of topics (e.g., at 1.3.12 on an unchaste woman being thrown from the Tarpeian Rock, at 1.5.9 on a man who raped two women in one night, at 1.6.12 on a pirate chief’s inheritance). Finally, on p. 260 Hallett notes Ovid’s allusions to female homoeroticism, “all involving Greeks from the past (and largely the remote, mythic past),” but the same case can be made for the story of Iphis and Ianthe in the Metamorphoses and the Epistula Sapphus included among the Heroides as can be made for Phaedrus and Plautus: these texts invoke an extremely wide range of human behaviors — homoerotic and heteroerotic (note that in the Epistula Sapphus Sappho focuses on her passion for the beautiful young man Phaon), sexual and otherwise — in such a way as to involve “Greeks from the past.” Still, Hallett’s article reminds us of an important fact: the Roman textual tradition seriously problematizes sexual relations between females in ways that stand in revealing contrast to its treatment of sexual relations between males.

The title of Pamela Gordon’s “The Lover’s Voice in Heroides 15: Or, Why Is Sappho a Man?” alludes to Marilyn Skinner’s 1993 article “Why Is Sappho a Woman?”, which in turn alludes to David Halperin’s 1990 essay “Why Is Diotima a Woman?” Each of these three very different questions is well worth asking; Gordon’s comes in response to the distinction between flattering Greek images of Sappho as “the tenth Muse” and Roman anxieties about her masculinity and her erotic relationships with women (although Gordon herself admits in a footnote that the distinction is not all that absolute: “the Greece/Rome dichotomy is, of course, not always clear-cut; there is much variation between regions and periods, and Sappho’s lyrics about women had already become problematic in Hellenistic society” [n. 2]).

Gordon begins by asking an important question indeed: is Sappho’s voice in this poem “a woman’s voice” or merely “Ovid the male ventriloquist” (p. 275)? Her answer, recalling Hallett’s discussion of the way in which Roman texts masculinize tribads, is this: “The Ovidian Sappho writes so much ‘like a man’ that the poem would work well as a parody of what French psychoanalytic critic Luce Irigaray has called ‘hom/m/osexuality’ (a play on l’homme), or what Esther Newton has dubbed the ‘Mythic Mannish Lesbian'” (p. 275). Now, the last phrase brings Gordon into some difficulty, as she acknowledges the problems raised by applying the term “lesbian” to an ancient woman, but nevertheless wishes to see “some striking similarities between Ovid’s fictional Sappho and Esther Newton’s formulation without claiming that the Mythic Mannish Lesbian is necessarily a transhistorical or cross-cultural phenomenon” (p. 276). When Gordon says that in Heroides 15, “as in the other Roman texts cited by Hallett (1989a) and Richlin (1993b), homoerotic behavior is construed as something unnatural or peculiar at best” (p. 279), another problem arises. The texts that Richlin cites have to do with sexual behavior between males, not females, and in any case Richlin does not argue that they represent homoerotic behavior per se as “unnatural or peculiar,” but only the preferences and practices of men who seek to be penetrated. This is a crucial distinction. Moreover, as in other essays in this collection, there is here a certain imprecision with regard to the language of “nature”; in Heroides 15, Sappho says nothing about natura and makes no claims about animal behavior (Iphis in the Metamorphoses does precisely this, but Gordon does not cite that passage here).

In her discussion of “The Mannish Muse” (pp. 280-3), Gordon makes a helpful comparison between Sappho’s persona and that of the male lovers in the Heroides who have abandoned their girlfriends. But she oddly interprets Sappho’s physical reaction to her passionate feelings for Phaon ( siccae non licet esse mihi, 134) to mean that Sappho has a “wet dream” and thus, “like other tribades, has acquired a phallus” (p. 283; cf. ibid.: “her apparent membrum virile“). In view of the fact that Sappho is not otherwise identified in this text as a phallic tribas, and in view of the fact that her phrase most readily invites interpretation as a reference to abundant vaginal secretions, it seems a stretch to speak of phalluses here.

In the section entitled “Why Is Phaon A Boy?” (pp. 284-6), Gordon reminds us that Sappho’s beloved is not a fully mature man but a beautiful youth, and thus (p. 285) that “here the Roman vantage point for this view of Sappho emerges more fully. In pursuing a pretty boy, Sappho conforms to the Greek stylistics of male sexual behavior …” But in both Greek and Roman ideological traditions, women as well as men were represented as being capable of displaying erotic interest in beardless youths or boys: consider the myth of Eos and Tithonos (see Robin Osborne, “Desiring Women on Athenian Pottery,” in Natalie Boymel Kampen, ed., Sexuality in Ancient Art [Cambridge, 1996], pp. 65-80), and cf. Theocr. 15.129-30, Tibull. 1.8.31-2, and Mart. 4.28.7. This consideration challenges Gordon’s argument to the effect that Sappho’s interest in a boy rather than a man is a further sign of her “otherness.” Still, problems aside, Gordon’s discussion raises thought-provoking questions.

With Alison Keith’s essay on Sulpicia (” Tandem Venit Amor : A Roman Woman Speaks of Love”) we come to the final section of the collection: “Female Construction of the Desiring Subject.” Taking as her starting point a scholarly consensus that Sulpicia wrote during the 20s B.C., and considering the evidence to the effect that during the same decade the Aeneid was being discussed and circulated at least in parts, Keith sets forth her thesis:

“I shall argue that Vergil’s portrayal of the love of Dido and Aeneas provides Sulpicia with a framework in which to articulate a woman’s love for a man, and that Sulpicia’s imaginative engagement with the Dido episode in the Aeneid exposes the inadequacy of our standard either/or model of ‘pro-‘ or ‘anti-‘ Augustanism for investigating the relations among female subjectivity, agency, and sexuality in Augustan Rome” (p. 296).

A number of interesting comments on the two texts follow, such as this comparison between Dido and the persona of Sulpicia’s elegies: “Like the Vergilian Dido, who does not conceal her ‘fault’ ( culpa) but calls it marriage ( Aen. 4.170-2), ‘Sulpicia’ delights in her fault ( peccasse, [Tib.] 3.13.9) and revels in its publication. Where Dido is the object of rumors purveyed by Fama, however, ‘Sulpicia’ publishes her transgressive passion to gain a literary reputation” (p. 301-2). Like a number of her fellow-contributors, Keith finds political resonances in the language of the erotic. Thus, for example, she notes that with the shift from the third person epic to the first-person elegiac narrative, we have a “loss of political resonances”: “Paradoxically, however, this reorientation of the Dido episode from public to private allows Sulpicia to disrupt the logic of subsuming the transgressive political ambitions of a woman in her deviant sexuality. Sulpicia thereby challenges, however slightly, the protocols regulating the sexual conduct of Roman women” (pp. 302-3). This is an interesting point, although it seems somewhat excessive to speak of “disrupt[ing] the logic of subsuming the transgressive political ambitions of a woman in her deviant sexuality” solely on the basis of the use of the first-person narrative technique and the absence of explicit political references in the text.

One final quibble. Keith summarizes Dido’s sexuality thus: “Viewed through this androcentric elite Roman perspective, Dido’s nationality and gender (ideo)logically render her socially disruptive and therefore sexually deviant” (p. 299). It seems to me, though, that we need some mention of the fact that Venus causes Dido to desire Aeneas, in other words that the text explicitly portrays Dido as a victim of forces beyond her control, and thus not simply as “disruptive” and “deviant.” Now, Venus’ intervention does not necessarily contradict Keith’s argument (one could easily make the case, for example, that it is hardly coincidental that the victim of passion is a woman, whereas the man Aeneas is characterized by a proto-Stoic tendency to subjugate his passions), but to ignore this consideration is to invite accusations of oversimplification. In general, though, Keith’s strategy of bringing together Dido and Sulpicia helps shed light on two very different texts that have something to teach us about Roman ideologies of gender and sexuality.

Much the same could be said of this collection as a whole. While some of the essays have their problems (imprecise formulations and insufficiently clear conceptualizations above all), all of them raise interesting questions, and there is much to be learned from this volume.

On p. 33, Walters does cite Sen. Epist. 95, observing that Seneca aims to “contrast what ‘nature’ intended for women with a nightmare vision of them as sexually penetrating and thus dominating men,” in n. 9 offering a translation of Epist. 95.21: “though nature made them sexually passive.” But Seneca’s Latin, unfortunately not quoted here, is pati natae : not quite as bald an evocation of the concept of “nature” as the Latinless reader might be led to think by Walters’ translation. It is a pity, too, that Walters ignores Sen. Epist. 122, where Seneca offers a lengthy discussion of behaviors that he characterizes precisely as contra naturam (see Winkler 1990 and my forthcoming book [below, n. 2] for discussion of this and other ancient allusions to “natural” and “unnatural” sexual practices). Craig A. Williams, “Homosexuality and the Roman Man: A Study in the Cultural Construction of Sexuality,” Ph.D. diss. Yale 1992, pp. 263-4. A revised version of my dissertation is forthcoming Roman Homosexuality: Ideologies of Masculinity in Classical Antiquity (Oxford University Press, 1998). In private correspondence, Marilyn Skinner has apologized for the omission of any reference to my dissertation in Roman Sexualities. See OLD s.v. Interestingly enough, there is no attested Latin nomen actionis; Parker follows English usage in referring to the act as “cunnilingus” (p. 51). We might do better to take our cue from the existence of the noun linctus (see OLD s.v.) and posit a term cunnilinctus. There is one source that might be cited in support of the notion that cunnilinctus can be conceived in Roman terms as “passive”: Seneca’s narrative of Hostius Quadra, who, simultaneously being anally penetrated and performing cunnilinctus, exclaims simul et virum et feminam patior (Sen. N.Q. 1.16.7; cf. the discussion at Williams 1992: 255-9). But Parker fails to refer to this passage, thereby leaving the reader without any direct textual support for his notion of cunnilinctus as “an act of being penetrated.” When Parker observes (p. 57) that in 4.43 “Martial … insists he did not call a man a cinaedus; he called him something worse, a cunt-licker ( cunnilingum)” he oversimplifies. The speaker does not represent his use of the term cunnilingus as “something worse,” but in fact just the opposite: hoc leve et pusillum (9). To be sure, the characterization is ironic, and the point of the epigram must be that the insult cunnilingus is actually harsher than cinaedus, but Parker’s description elides the irony. When Edwards translates Juv. 6.110 ( facit hoc illos Hyacinthos : i.e., their status as gladiator makes them beautiful) as “That makes them all Adonises” (p. 78), she obscures an intriguing point, namely that Hyacinth, the beautiful youth loved by Apollo, could be evoked as the paragon of a youthful male beauty imagined to be attractive to women. I discuss this point further in my forthcoming book (above, n. 2). sic vixit ut muliebria cotidie convivia essent, vir accumberet nemo praeter ipsum et praetextatum filium — etsi recte sine exceptione dixeram virum, cum isti essent, neminem fuisse. There is another error of omission when, by way of illustrating the point that accusations of effeminacy could not be leveled against just anyone, Corbeill observes that “Cicero’s corpus contains no evidence that he was himself so charged” (p. 112). Apart from the fact that we should hardly expect to find within Cicero’s corpus evidence for slurs of effeminacy against himself, Corbeill ignores the existence of Dio Cassius’ report of a verbal assault on Cicero made in 43 B.C. by Quintus Fufius Calenus, who maliciously depicted the orator as, among other things, effeminate (e.g., 46.22.3: γύννις). In this connection Oliensis proposes that “in their different ways, Propertius’ poems to Gallus and Horace’s poems to Maecenas enter into a kind of exchange that circumvents the seemingly rigid opposition between penetrating and being penetrated, having the phallus and lacking the phallus. When two friends trade poems, the one-way act of penetration yields to a series of interactions in which each man plays each part alternately” (pp. 168-9). I find this line of argumentation (for which see also Williams 1992: 331-67 and John Boswell, Same-Sex Unions in Pre-Modern Europe [New York, 1994], pp. 75-80) very interesting. Some readers may be put off by arguments like the following: “She functions as a sign in a Roman imperial discourse that fragments woman, distributing her across the tensions between the private and public spheres, between the city and the provinces, and between the sovereignty of the emperor and the freedom of the senator. As such, Messalina figures empire both good and bad, present and past. The discourse of empire and the work of history effect a further displacement. The desiring woman signals the exercise of imperial power, but the exercise and always threatening excess of imperial power become the woman. In the process of displacement, her body becomes what she signifies” (p. 247). I present these arguments, together with analogous claims concerning the cinaedus, at greater length in my forthcoming book. See also Bernadette J. Brooten, Love between Women: Early Christian Responses to Female Homoeroticism (Chicago and London, 1996) with Halperin’s comments in BMCR 97.12.3. Gordon briefly discusses in her epilogue the question of the authorship and date of Heroides 15; she argues that the text “look[s] quite at home in first-century Rome,” but acknowledges that others may wish to give it a later date, and thus refers to its author as “Ovid” only for convenience’s sake (p. 4). In her note on this passage (n. 6), she cites Richlin 1993 on the word “homosexual,” but as it is Richlin’s argument that the Romans did identify cinaedi as “homosexual” (see the discussion of Parker above), Gordon’s unqualified citation of Richlin only confuses matters. Is Ovid’s Sappho a “lesbian”? A “homosexual”? Neither? Cf. p. 286: “This ‘other’ nature is so integral to her character that, although she may turn her gaze to a male, Sappho must remain, absurdly, a hom/m/osexual.” Keith discusses Venus’ and Amor’s actions only in her later analysis of the Vergilian background to Sulpicia’s first elegy (p. 301).