Skip to content Skip to navigation

Full citation: 

Hallett, Judith & Marilyn Skinner, eds. 1997. Roman Sexualities. Princeton: Princeton University Press. ISBN 0-691-01178-8

Contents summary: 

Introduction by Marilyn B. Skinner

This collection applies feminist analysis to local versions of a common Mediterranean sex/gender system. This introduction [as is typical for collections of this type] synthesizes the material into an overall understanding. Rome represents a distinct and discrete system within that common system, whose features in turn speak to other intersectional fields. As in Greece, the sex/gender system involved patterns of dominance and submission that don’t always align with biological sex. The central organizing concept is an understanding of intercourse as bodily penetration of an inferior which, taking as its prototype a male-female relationship in which the female is considered inferior, means that the penetrated individual in any coupling is “feminized”. Forms of pleasure experienced outside that framework were stigmatized.

This general pattern is pan-Mediterranean, but this collection focuses on Rome and its provinces during the period from 70 BCE to 200 CE, roughly the late Republic to the middle Empire. Changes in the position of the elite in political power structures changed their preoccupations from genuine political responsibility and power to concern with personal honor and autonomy in the face of barriers to upholding those states. The increasing incorporation of “foreigners” and former slaves into Roman society escalated prejudice along ethnic and class lines. Outsiders could gain support through personal connections (amicitia) while the old hereditary nobility responded in literature, projecting their anxieties.

Literary texts tended to reflect the elite who weren’t always aware of (or concerned with) regional differences or religious/cultural shifts in the understanding of sexuality. Elite men represented one core sexuality system, then extended that system to their dependents. Departures from the system are used as a tool for social control and commentary, often motivated by forces other than the specific sexual behaviors involved. For example, accusations of “effeminacy” could be used against any non-conformist sexuality. The grotesque descriptions of “the tribade” reinforced male images of superiority. Poetry uses “wantonness” as an expression of counter-cultural resistance.

Roman concepts of sexuality are inseparable from those of gender. “Woman” is set up as an affirming opposite to “man” but the dividing line doesn’t follow physiology. There is a discussion of different Greek and Roman attitudes toward women and their status. The Greek image and ideal was for women to be connected with the interior, private space of the household, to be a transferred property from one male to another, and to be the passive vessel for the generation of a male heir.

Roman society elevated the male control over the family to include adult children (including married daughters, in some circumstances). Women were a part of the display of the family’s status and therefore needed to be active and visible in the public sphere. The care of infants, rather than being one of the primary concerns of wives, was left to servants.  Daughters could, in theory, be eligible to be their father’s heir (even if married). Women were expected to promote their birth family’s goals within their husband’s family. If the husband did not have manus (legal control) over his wife, he also had no legal control of her property.

As a class, Roman women might be “other” and inferior, but exceptional individuals might be praised as “masculine” for displaying virtue. Female sexuality was not inherently hostile to the dominant structures. Women’s ability to inhabit “masculine” characteristics reflected the gender permeability that also stigmatized any “feminization” of men.

Greek pederasty was, ideally, a relationship between men of the same class, distinguished by age and status. Roman attitudes, in contrast, considered any acceptance of a passive role to be incompatible with “manliness” (vir-tue). Therefore relationships with young male citizens were forbidden as involving shame (stuprum). Roman (male) sexuality partook more of violence and aggressiveness compared to the Greek concept. The discussion compares sexual dynamics to various types of public entertainments involving violence.

The prototypical man (vir) was the model from which all other categories diverged and by which they were defined. This took the adult male citizen as the central model. The word vir (man) was limited in application to this central model and was not applied to boys, to working class men, to slaves, or to “disreputable” people. Vir denotes not only a set of behavioral principles but also a defining set of privileges.

This central model can also be examined in terms of what lifestyles and acts were viewed as transgressive. In contrast to the Foucaultian view of sexual acts as licit or illicit, Parker’s article suggests a system of acceptable or perverse “sexual personas” based on whether the person’s status aligned with the role the played with respect to the penetrative hierarchy. Persons who acted counter to this alignment (the tribade or the cinaedus) are categorized as monstrous and stigmatized, and these are viewed as deviant preferences.

Several of the articles are concern with the instability of male status and identity. “Infamy” (infamia) is the participation in, or pursuit of, stigmatized sexual roles. But to what extend did these “infamous” roles represent actual personal identity, as opposed to being a social tool to shame or attack individuals and behaviors? Under what circumstances could a man adopt a “feminized” persona unscathed? For example, the authorial persona of a lyric poet complaining humorously of a controlling female lover? The “controlling mistress” could have “masculine” power attributed to her without becoming monstrous. This is compared to the relationship between client and patron (amicitia).

Men characterized women simultaneously as “other” within a dualistic system, but also as a source of “natural” power both generative and destructive.

Roman society was characterized by “perilously permeable” class, ethnic, and gender boundaries. These are conflated in various combinations to express elite anxieties, e.g., sometimes depicting foreigners as hyper-virile, sometimes as feminized. Imperial women seen to possess illegitimate political power were accused of sexual excess. Any female ambition could be seen as a desire to appropriate male sexual power.

The ultimate female transgression was being the active figure in penetrative sex. The tribas stood in for this inversion of sexual norms, but Roman society could only envision sex between women as pseudo-phallic, and expected it to be expressed physiologically (i.e., with a masculine physique and enlarged clitoris) as well as in behavior. The idea of the tribade was distanced from the Roman here-and-now by identifying her as Hellenic, as existing in the past, and/or as having masculine physiology. As a Roman reality, she is simultaneously empowered and negated. And yet, non-phallic sex between women is mentioned, as in Martial’s Philaenis, who must then be framed, not as monstrous for claiming male identity, but as a failed man for behaving against male sexual norms. Similarly, see Ovid’s depiction of Sappho in the Heroides who has rejected her previous love for women but fails at being a desiring (masculine) woman, no longer allowed even her talent as a renowned poet.

Finally we have a rare female voice in the poet Sulpicia who transforms traditional male-female tropes and imagines a different type of female sexuality as an agent, just as she claims poetic agency.

Part One: Unmarked Sexuality

1. Invading the Roman Body: Manliness and Impenetrability in Roman Thought by Jonathan Walters

Phallocentrism meant that Romans found it hard to conceive of sex between women that didn’t involve either a natural (clitoris) or artificial (dildo) analog. Walters points out that the concept of “woman” in Roman discourse was a male construct, used by men to talk about men, not to talk directly about women.

Part Two: Wayward Sexualities

2. The Teratogenic Grid by Holt N. Parker

This article discusses alternate sexual maps than a binary based on similarity and difference (i.e. homosexuality and heterosexuality). Parker explores the Roman system based on an active/passive distinction and the nature of the penetrated orifice. Non-penetrative activities such as kissing, fondling, etc. are recognized as sexual, but do not participate in the construction of a sexual persona or role.

If the sex and/or status of the receptive partner in intercourse aligns with the official social hierarchy, then the act is licit.

Misalignment is anomalous and abnormal: for example, a woman in any “active” role, a “passive” man penetrating a vir, or the inversion of vaginal penetration by the act of cunnilingus (by a man). The desire to perform oral sex was seen as both “passive” and degrading, and as gender-neutral with regard to the “active” (receiving) partner. This creates the “joke” in Martial’s epigram 7.67, where the tribade Philaenis is being so “manly” that she butt-fucks boys, fucks girls, considers fellatio to be too unmanly for her to perform...but then (absurdly) enthusiastically performs cunnilingus.

The active-passive axis results in any sexual activity that does not involve inserting a penis into something being considered “unmanly”. For both sexes there is a hierarchy of “humiliation” with respect to the orifice being entered, going from least humiliating to most, it is vagina > anus > mouth.

Passivity was so expected for a woman that an indication of enjoyment was suspect. But though this was the supposed ideal, it is contradicted in literature, especially poetry, where men desire a woman who actively enjoys sex. The “abnormal” sexual roles for women exist on multiple axes, with any degree of “active” making her masculine, even simply desiring to be penetrated (especially desiring the wrong sort of person to do the penetrating). A woman can perform active sex with another woman either by rubbing (where a penetrative clitoris is assumed), or with a dildo (one penetrates the other), or by receiving cunnilingus. There are no references to mutual masturbation, though the textual data alone can’t distinguish whether it wasn’t done, or whether it was ignored due to not being part of the conceptual system. Active women are imagined to be physically masculinized, with an enlarged clitoris (regardless of whether they are engaging in sex with men or women). Clitoridectomy is described in medical texts as a possible remedy for this excess desire.

But Parker asks whether figures such as the cinaedus or the tribade as depicted actually existed, or whether they were social stereotypes used for rhetorical purposes, where actual practice and identity may have been fuzzier.

3. Unspeakable Professions: Public Performance and Prostitution in Ancient Rome by Catharine Edwards -- No comments.

Part Three: Gender Slippage in Literary Constructions of the Masculine

4. Dining Deviants in Roman Political Invective by Anthony Corbeill - No comments.

5. Ego mulier: The Construction of Male Sexuality in Catullus by Marilyn B. Skinner - No comments.

6. The Erotics of amicitia: Readings in Tibullus, Propertius, and Horace by Ellen Oliensis - No comments.

7. Reading Broken Skin: Violence in Roman Elegy by David Fredrick - No comments.

Part Four: Male Constructions of “Woman”

8. Pliny’s Brassiere by Amy Richlin - No comments.

9. Female Desire and the Discourse of Empire: Tacitus’s Messalina by Sandra R. Joshel - No comments.

10. Female Homoeroticism and the Denial of Roman Reality in Latin Literature by Judith P. Hallett [originally blogged separately as entry #53]

This article looks at the disconnect between Roman literary considerations of female homosexuality and their everyday reality. The period covered is the 2nd century BCE through the 2nd century CE. Various mythic origins were attributed to homosexual desire. One example is the story of how a drunken Prometheus , when creating humans from clay, attached sexual organs to the “wrong” bodies, thus creating individuals whose internal preferences were counter to their external organs. One common theme in these literary discussions is to position homosexuality as foreign -- and especially Hellenic -- and as deriving from or belonging to an older era. These portrayals also represented female same-sex desire as being “masculine”, even to the point of involving male sexual organs (at least in symbolic form). This contrasts with the best-known Classical Greek reference to female homoeroticism: Plato’s myth of desire being based on originally dual-bodied individuals longing for their “other half” where the dual-bodied were composed of all possible combinations of male and female. The bulk of the article looks at the references to female same-sex desire in Roman literature that build up this “Hellenizing, archaizing, masculinizing” framing of the subject. With the understanding that these mentions are from male authors in a notoriously misogynistic culture, they provide a view -- though almost certainly a distorted one -- on everyday practice in that culture.

In the play Truculentus, a character puns on two similar-sounding words to suggest that a female character “fuck your mistress”, though the bit is a passing joke rather than a significant plot element. Seneca the Elder, in discussing how to speechify about unmentionable subjects, gives the example of a man who caught his wife having sex with another woman and killed them both, then needed to explain the matter when presenting his defense. Both of these examples draw from earlier Greek originals.

The normal word used by Roman writers for female homosexuals is tribas (meaning “one who rubs”), a Greek word and retained in its Greek form. But despite the root meaning of the word, Roman use typically implied masculine-framed activities such as penetration. The author notes that two of the three authors she cites who comment directly on contemporary Roman behavior do not use tribas.

Ovid also draws on earlier Greek material for his Metamorphoses in which the gender-disguised Iphis is in love with Ianthe and laments her desire as impossible and unnatural (until given a divine sex-change). Ovid has several passing references to Sappho and her love for women, but frames that love as shameful.

Seneca the Younger is one of the few Roman writers who comments on contemporary women who “rival men in their lusts” though the implication that it is lust for women is fuzzy. And in a longer passage, he similarly focuses on contemporary women taking on masculine habits and vices (such that they have “lost the privileges of their sex as a result of their vices”). The author notes that this “would seem to include same-sex love” but I’m not sure I’d take the allusions as conclusive.

Martial, famous as a writer of satiric epigrams, has three that address women identified as tribades. Two are addressed to a woman with the Greek name of Philaenis who participates in a great many masculine athletic activities and is rudely accused of being a sexual aggressor to both boys and girls. Martial says she correctly calls as “girlfriend (amica) the woman she fucks" and that she performs oral sex on women. Martial’s third example is similarly masculinized in his description of her sexual activity with women (even using a grammatically masculine noun for “fucker”).

Juvenal’s satires include one in the voice of a woman named Laronia who alludes to women performing mutual oral sex, but in the context of claiming that Roman women don’t do that.

Overall, it is questionable whether these examples accurately represent Roman women’s lives, but it can be assumed that they represent at least some men’s attitudes.

11. The Lovers Voice in Heroides 15: Or, Why Is Sappho a Man? by Pamela Gordon

Given the esteem in which the Greeks held Sappho, exhalting her in feminine terms as a muse, why were Romans less certain of her praiseworthy womanliness? This article looks at Ovid’s Heroides number 15 written in the voice of Sappho. The Heroides are a collection of elegaic poems attributed (probably correctly) to Ovid, which take the form of letters from abondoned heroines to the men who spurned them. [Note: while putting together this summary, it occurred to me that Catherynne M. Valente’s The Refrigerator Monologues, which takes the form of narratives of “fridged” girlfriends and wives of superheroes, is a direct homage to Ovid’s work.] In Sappho’s case, the letter is addressed to the young Phaon, who has abandoned the aging and no longer inspired poet for the girls of Sicily.

There are some questions about the provenance of the Sappho piece, given that it is not included in the earliest surviving copies of the collection, but begins appearing in 12-13th century manuscripts. (Gordon doesn’t come down solidly on the question, but does consider the work authentically Roman.) In the medieval context, the work is believed to be Ovid’s translation of an actual work by Sappho. Modern scholarship often focuses on the question of authorship but Gordon is more interested in what the work says about Roman attitudes toward sexuality.

Recent studies have suggested the poem offers alternative modes of sexuality to the traditional Roman model, while others counter that it follows the traditional model in the other poems of the collection in positing that “to be a heroine means to be abandoned.” Other scholars consider that the voice of the poem is a hybrid of Sappho’s poetic style with Ovid’s voice, upsetting the “male poetic order” but not entirely representing a feminine voice.

Considered in the context of the portrayal of female homoeroticism, the poem matches a recurring pattern in Roman (male) writing: the masculinization of tribades, both in their desires and their bodies. This “mythic mannish” Sappho (following the terminology of Newton 1984) is a product of Roman traditions, not Greek ones. How does this masculinization fit--not into general conceptions of gender--but into the gendering of other characters in the Heroides? In fitting Sappho into the category of “woman forsaken by a man” is this a deliberate “heterosexualizing” of her? (But see all the commentary on how the categories of heterosexual and homosexual are inapplicable in Roman culture.) Gordon feels that the gender elements are more complicated.

Unlike the abandoning lovers of the other Heroides, Phaon is not a hero or a king. He seems instead to be a sort of icon of youthful beauty, originally a mortal lover of Aphrodite. Phaon is mentioned in a fragment from Sappho, and Ovid may likely have had access to additional references. Scholars have reconstructed Sappho’s mention as touching on the motif of Aphrodite’s love for Phaon. In that context, the reference is part of Sappho’s unconventional approach to love between men and women. While her verses on love between women stress mutuality, the verse mentioning Phaon uses the model of a powerful goddess choosing a beautiful youth in a way that breaks free of usual male-female scripts. It imagines a dominant, desiring woman with a submissive but response man.

Ovid takes this scenario, maps it onto Sappho rather than the goddess, and sides with the dominant gender paradigm that Sappho rejects. Ovid didn’t invent the idea of Phaon being Sappho’s lover. The motif appears, for example, in the 4th century BCE plays of Menander. But Ovid takes this motif and turns it into a “conversion narrative” whereby Sappho doesn’t simply include desire for a man among the many women she desires such as Anactoria, but entirely rejects the love of women in favor of Phaon. There are also implications that she now considers her previous love for girls as disgraceful. Only in the context of 1st century BCE Rome does this framing of the story emerge.

This “conversion narrative” clashes with the Foucaultian assertion that Classical culture made no distinction in the sex of a sexual partner. Here Sappho’s love for girls and her love for Phaon are depicted as qualitatively different--as two entirely separate types of experience.

Although words derived from the name Lesbos were not yet used to denote sex between women in Classical Greek or Latin, there are a number of references indicating that the island was popularly associated with that practice. For example, Lucian’s choice of Lesbos as the origin for his gender-transgressive tribade Megilla/us.

Ovid’s other heroines are sorrowing not simply because of yearning for their absent lovers, but because their departure has created real social difficulties involving loss of status and honor. They are discarded “fallen women.” The other heroines describe their own beauty as a feature in their story, but don’t dwell on the appearance of their lovers. In contrast, Sappho describes Phaon’s beauty and then notes that at least she has skills to make up for her lack of the same. Phaon is passive compared to the other lovers: being, rather than doing. The other heroines don’t speak directly about sex, while Sappho boasts of her expertise.

The explicit language and desiring gaze of Sappho find their parallels in Ovid’s male voices. In Ovid’s stories, beauty is the cause of rape. The only male figures who are raped are similarly described as youthful and beautiful. Female rapists are punished by failure or loss, while male rapists go on their merry way.

Another aspect presented as masculinization of Sappho is in a passage where she describes an erotic dream about Phaon and admits “it all happens, it feels so good, I can’t stay dry (sicca).” But after a discussion of how sicca of a woman usually means “frigid or sexually unresponsive”, Gordon concludes that in this passige “not dry” refers to a masculine “wet dream” thus implying a penis. (This seems to overlook female secretions and emissions on arousal. But alternately, would that be something that Ovid would think to describe?)

Ovid’s Sappho has also lost the maternal devotion to Kleis that she shows in the Greek poetry and instead describes motherhood as a burden. These various features, taken together, align Ovid’s Sappho with his male characters, rather than his female ones. In this, she is reminiscent of the stereotype of the “mannish” lesbian of the early 20th century, who se identity is formed by attempting to imitate men.

Looking from the other side of the couple, why is Phaon a boy rather than an adult man? It’s because Sappho is required to be drawn to youthful beauty, just as any other vir would be. The active-passive alignment means that Sappho does not demand that Phaon love her, but rather that he allow himself to be loved. Phaon is the eromenos and Sappho the erastes. But as Phaon has grown to adulthood, he in turn is required to seek youthful beauty to love--to do, rather than to be. There can only be one “man” in a relationship, and as Phaon comes into manhood, he must reject man-Sappho.

Lucian’s female couple has one “man” (Megilla/us) and one “normal woman” (Demonassa). It is meaningful that Megilla/us is from Lesbos (the source of tribades) while Demonassa is from Corinth. Demonassa is not a tribade, only Megilla/us is.

Thus Ovid’s Sappho follows the pattern seen by Hallett (elsewhere in this collection) that women who engage in same-sex activity are seen in Roman literature as masculine, anachronistic, or Hellenic--she is made to be all three. While Greeks did not consider that a man must be “other” to love a man, Romans could only imagine same-sex love in terms of self and other. With regard to sex between women, it was not desire for a woman that made a woman a tribade, but rather the state of being a tribade made her desire women. Romans could not imagine that a woman could love a woman, therefore to love a woman, Sappho must have been a “man”.

The article concludes with a consideration of cinaei. Should we believe that they actually existed as an identifiable identity, given that the only evidence comes from hostile sources? Yet that evidence of existence can be retrieved from the textual silences. Similarly, the Roman attempts to deny the existence of women who loved women must be viewed critically. Gordon brings in comparisons from other eras when the “official” position was that to love women meant to be a man. If we disbelieve that assertion in other eras because we have counter-evidence, then why take the Roman claims at face value? Why accept the hostile definition of non-conforming women as necessarily “male” rather than understanding them as something else entirely. (The article ends with this question in the context of lesbian historical theory.)

[Note: I think there are some important cracks in the supposed “Roman sexual system” displayed here. If, as is argued for men, there is no moral distinction made with regard to the sex of a man’s partners, so long as he sticks to his approve role, then why should Sappho be depicted as seeing a qualitative difference between her relations with women and those with men? According to the “official system” perhaps Phaon should feel shame at being dominated by a woman, but there is no support in that system for Sappho feeling differently about desiring men than desiring women. To me, this leaves a lot of space either for doubting the ability of upper class Roman men to accurately depict women’s emotional lives, or for doubting that the Roman sexual system was a gender-neutral as some claim, at least when women were involved.]

Part Five: Female Construction of the Desiring Subject

12. Tandem venit amor: A Roman Woman Speaks of Love by Alison Keith

And, in fact, we have one solitary female voice, recording her own thoughts about sexual desire and her place in the Roman sexual system. This is a set of elegies by the poet Sulpicia, probably the daughter of Servius Sulpicius Rufus, writing at the very end of the 1st century BCE. Keith considers Sulpicia’s voice in parallel with the depiction of Dido in the newly-written Aeneid (and with which Sulpicia was almost certainly very familiar).

Male elegiac poetry presented no model to a female author, being concerned with men rejecting the standard male career path for the life of a poet-lover. Virgil’s Dido offers a closer parallel. The treatment of Dido in the Aeneid corresponds closely to Augustan policies on morality, and Keith suggests a possible direct connection between the representation of Dido’s sexual transgression and Augustan attempts to regulate female sexuality. Dido is an anomaly in the context of Roman cultural norms as a female head of state. But she is controlled in the narrative by restricting her story to her sexuality and its consequences. The love/desire for Aeneas converts her from a leader to a lover, and at that, a lover who has abandoned loyalty to her late husband. The article goes into some detail of how Dido’s actions went against Roman ideals, and how that transgression was depicted as the cause of her downfall and death.

The elegiac tradition set itself up in opposition to those cultural norms, but from the point of view of men rejecting cultural expectations. Sulpica expresses how the admission of desire is inherently “improper” by those cultural norms, but that she considers concealing her desire to be more shameful than proclaiming it openly. There is a discussion of the concept of pudor (“what is seemly”) that is central to the text. Sulpicia simultaneously rejects conventional pudor and invents a new standard that requires being true to love, in parallel with male elegaic poets embracing nequitia (depravity). Sulpicia equates traditional pudor with deception and concealment. “It isn’t what you do, but what you’re known to have done.” In various ways, she expresses how her beloved is “worthy” of her love, which in turn require that she honor that love with public transparency. The sequence of poems includes several that seem to unveil events around Sulpicia’s birthday. The first takes on the motif of the separated lovers, but where she has been required to retire to the country by the man who controls her movements. (There is a description of various country “delights” that are unpleasing because they mean separation from the beloved.) This is followed by joy at the journey being postponed enabling their meeting, but he is busy, distracted. She laments that the greatest grief for a woman is to “yield to an ignoble lover”. And then finally there is a happy reuniting.

Sulpicia regularly reverses the expectations of morality--her lapses are concealment, or “false pudor”, her virtues are honesty about her (socially inappropriate) passion. Thus we get a female take on expected behavior within a romantic relationship and how it might be rejected, similarly to how male elegiac poets rejected the strictures imposed on them.

historical