Fragmentary ancient texts are worth studying for many reasons, not least because they can compel us to re-examine what we think we know about classical literature. No matter what literary genre may be under discussion, the extant works represent only a tiny fraction of what once existed, and it would be unwise to treat them as typical. These surviving texts were preserved and transmitted for a variety of reasons, but they were not deliberately selected to provide a representative cross-section.Footnote ¹ Therefore, fragments always deserve careful scrutiny. By nature they are lacunose, evasive and frustrating. They leave us with many unanswered questions, and they tend to rule out definitive interpretations or clear conclusions. None the less, they can prevent us from reaching the wrong conclusions and instead encourage us to foster a more sceptical or open-minded attitude. They are valuable precisely because they make us question received wisdom, challenge unexamined preconceptions, and discourage overconfident generalizations. In their messiness and intractability fragments inevitably disrupt neat-and-tidy accounts of literary history.Footnote ²

The sort of methodological approach that I have just outlined might be characterized as ‘fragmentary thinking’ or ‘thinking fragmentarily’, to adopt the language of a provocative recent discussion by Hannah Čulik-Baird.Footnote ³ To ‘think fragmentarily’ is defined as the project of questioning mainstream narratives and challenging dominant ideas. In Čulik-Baird's words, ‘fragmentary thinking’ is an alternative to ‘canonical thinking’, in that it ‘moves us away from the acceptance of canons as natural or neutral by allowing small details to carry as much weight as large structures’. She also argues, more radically, that ‘fragmentary thinking’ can have bigger consequences for Classical studies as a discipline, especially at a time when many of us are engaging in debates about ‘decolonizing the curriculum’ or asking what it means to study ‘Classics’ in the twenty-first century. ‘In theory’, Čulik-Baird asserts, ‘the fragment should, by its very nature, stand for alterity, heterodoxy, “otherness”.’ In other words, we are not just dealing with narrowly literary concerns: there is potentially a great deal more at stake here, in intellectual, symbolic and ideological terms. Where do our ideas about the ancient world come from? How did ‘classic’ or ‘canonical’ texts and ideas come to acquire such value? How and why did the ‘non-canonical’ come to be devalued, neglected or suppressed? What types of human experience have been privileged or excluded along the way? To what extent are our own views of ‘Classics’ shaped by personal status, sexuality, gender, race, social hierarchies or contests in power? What does it mean to be a Classicist? ‘Fragmentary thinking’ may suggest ways of responding to these questions and challenging deeply entrenched modes of thought and behaviour.

While I generally accept and welcome Čulik-Baird's argument, it is not my intention here to engage with all of its aspects or broader political implications.Footnote ⁴ What I aim to do in this article is to demonstrate some of the benefits of ‘fragmentary thinking’ by applying it to one particular topic of interest: sex in Greek comedy. I want to narrow the focus down to a single fragment—Damoxenus, fr. 3 PCG—but I argue that this small piece of evidence compels us to ask some big questions. The problem for literary historians is that this fragment constitutes an inconvenient piece of evidence that does not fit the mainstream narrative. That is, it does not match up with the widely accepted view of New Comedy and its supposedly normative attitude to sexuality and human relationships. I suspect that it is for precisely this reason that it has been almost totally ignored.Footnote ⁵

This intriguing fragment comes from one of the comic genre's most obscure practitioners. We are told by the Suda that Damoxenus was an Athenian comic poet (though the spelling of his name suggests a non-Attic origin), and we know from epigraphic evidence that he won at least one first prize in the Dionysia, at some point during the early third century b.c.e., but otherwise Damoxenus has left little trace on the historical record.Footnote ⁶ Almost all our information about his work comes from Athenaeus, who preserves the titles of two of his comedies—Hauton Penthôn (‘The Self-Mourner’) and Syntrophoi (‘The Foster-Brothers’)—and quotes an extract from each of them. The second of these, a 68-verse portion of dialogue featuring an unusually philosophical cook, has been widely discussed both for its parody of Epicurean doctrines and for its culinary humour.Footnote ⁷ These titles and fragments are valuable and interesting, but (as far as we are able to judge) they both seem entirely ordinary and unremarkable, in the sense that they give us exactly what we would expect from a ‘normal’ New Comedy.Footnote ⁸

What concerns us here is a further extract (fr. 3) quoted by Athenaeus from an unidentified play:

Athenaeus supplies no framing material or any other information about the dramatic context of the fragment: he merely quotes it, without comment, in a section of the Deipnosophistai concerned with different types of ball games. The scenario seems to have taken place at the gymnasium, and the speaker is clearly an older man recounting how he and other men sat watching the younger men as they exercised in the open air.Footnote ⁹ Athenaeus ostensibly quotes the passage, alongside several other comic quotations, as an illustration of the importance of graceful movement (εὐρυθμία) in ball games; but the Damoxenus fragment stands out because here the game itself is obviously not the only or the main source of interest (either from the speaker's point of view or from our own).Footnote ¹⁰

What we are looking at here is unmistakably a description of love at first sight. In many respects its sentiments and language follow familiar, conventional patterns. Underlying the whole mise-en-scène seems to be the metaphor, familiar from archaic Greek lyric poetry, of love as a sort of game or sport: the eroticized ball game is especially reminiscent of the world of Anacreon.Footnote ¹¹ Other specific details here—the emphasis on the gaze of the desiring subject;Footnote ¹² the idealization of physical perfection;Footnote ¹³ the crucial importance of eye-contact as a conduit of emotion, together with the idea that a single glimpse of the beloved is enough to effect a miraculous transformation on the lover's heart and mind;Footnote ¹⁴ the power of beauty to make a man cry out loud as an involuntary reflex reaction;Footnote ¹⁵ the notion that the peak of beauty occurs around the age of sixteen or seventeen;Footnote ¹⁶ the uncontrollable or divine onslaught of passion;Footnote ¹⁷ the ‘illusion-breaking’ appeal to the audience to share or confirm the lover's own reactions;Footnote ¹⁸ the image of love as a sort of illness or madnessFootnote ¹⁹—all these features find parallels elsewhere in comedy and other types of literature.Footnote ²⁰ Greek and Roman New Comedy in particular is full of such descriptions, to such an extent that they can be regarded as generic formulas or clichés.Footnote ²¹ The speaker in the fragment is describing his feelings in precisely the same sort of way as other lovestruck men in comedy do—except in one respect. In every other attested work of New Comedy the beautiful young object of desire is female, not male.

It is this crucial factor that makes Damoxenus’ fr. 3 seem extraordinary. While it would not be at all surprising to encounter homoerotic content in any other genre of Greek literature (including works from roughly the same contemporary context, such as the Platonic dialogues, the Idylls of Theocritus, Hellenistic epigram, and so on), it is a big surprise to encounter this sort of scenario in a comedy of the third century. There is nothing quite like this fragment elsewhere in the surviving remains of New Comedy.

Almost all of our other available information about New Comedy—including the surviving plays and fragments of later Greek comedy and their Roman adaptations—would seem to suggest that these plays all conform to a single fixed generic master-narrative.Footnote ²² ‘What New Comedy does’, in Lowe's words, ‘is to propose a universe that is overwhelmingly systemic: a world in which individual relationships are apprehended as part of a collective process governed by precise and inviolate rules of play.’ As we read in numerous scholarly accounts of the genre, the material that survives invariably conforms to the same basic pattern, viz. the playing-out of a heteronormative marriage plot.Footnote ²³ The ‘rules of the game’ permit only heterosexual relationships, and the optimum outcome is the marriage of a citizen woman to a citizen man, accompanied by the birth (or retroactive legitimation) of citizen offspring. Mistresses are (sometimes) tolerated; boyfriends are not.

Time after time, as our standard textbooks remind us, the love-stories dramatized in New Comedy are not purely or primarily about love (whether we are talking about erotic passion or romantic love).Footnote ²⁴ Because of the plays’ relentless focus on marriage, respectability, the financial stability of the household, and the institution of citizenship, we have come to understand the characters’ relationships not as private concerns but rather as a codified expression of civic identity. Gender roles, sex and emotions are seen as fundamentally ideological, politicized phenomena.Footnote ²⁵ In this stylized comic universe, sexuality is—apparently—synonymous with heterosexuality. In such a context it is easy to understand why other types of sexual desire and other types of relationship might be downplayed or written out of the narrative altogether.Footnote ²⁶ As Dover put it, ‘the reader who turns from Plato to Comedy is struck not only by the consistent comic reduction of homosexual eros to the coarsest physical terms but also by its displacement from the centre to the periphery of sexual life; for comedy is fundamentally heterosexual.’Footnote ²⁷

These aspects of Menander's comedy were already explicitly noted by Plutarch, who praised it for what he perceived as its supreme respectability and decorum:Footnote ²⁸

Whether this is (in effect) an assessment of New Comedy as a whole or just Menander in particular is hard to judge. By insisting that Menander never featured on stage a man in love with a boy, is Plutarch implying that by contrast other Greek comedians of the period sometimes did? Perhaps so: the passage above could certainly be interpreted in that way. But at any rate it is clear that Menander's influence on later comedy and audience taste was formative and far-reaching.Footnote ²⁹ Even though we know that the Roman comic playwrights used other Greek authors as their models (including Philemon, Diphilus, Apollodorus and others), there is no sign that they ever chose plays with a significantly different type of plot-line. By sticking closely to the Menandrian formula of the heterosexual marriage plot, the Roman comic playwrights succeeded in establishing it as the norm. If there ever were any fourth- and third-century Greek comedies that diverged from this familiar pattern, almost no trace of them has survived.

But this is not to say that homosexuality was never mentioned at all in New Comedy. It is possible to cite a small number of passages from later Greek comedy (including so-called ‘Middle’ Comedy) which allude to men's love for other men or boys. What is striking about such allusions—and what differentiates them from Damoxenus’ fr. 3—is that almost all of them are obviously hostile or contemptuous (to a variable degree). They take the form of conventional ‘buggery’ jibes which make fun of the passive partner in sexual intercourse, or of references to perceived effeminacy, or of ribald jokes based on the assumption that any man deprived of sex with a woman for a long time will eventually resort to masturbation or anal sex with another man for want of a better alternative. Such jokes are discussed and analysed at length by Dover: widespread throughout Old Comedy, they are much less well attested thereafter.Footnote ³⁰ From the fourth century and later Dover cites only Eubulus’ fr. 118, which describes the sexual frustration of the Greek heroes after ten years’ fighting at Troy without the consolations of hetairai:Footnote ³¹

Additional comparable passages include Diphilus, Theseus fr. 49 (in which three Samian prostitutes attending the Adonia laugh as they imagine the groans of a sodomized blacksmith); Menander, Dyskolos 891–2 (in which Sikon uses double entendre to imply that the cook has been trying to bugger him);Footnote ³² Menander, Sikyonios 264–6 (in which Moschion is mocked for his smooth, beardless appearance and insulted as a ‘bugger’—λάσταυρος, a rare Menandrian obscenity);Footnote ³³ Anaxandrides, Odysseus fr. 35.9 (from which we learn that anyone caught gazing at beautiful young men might earn the obscurely mocking nickname Καινὸς Θεατροποιός, ‘New Theatre-Maker’);Footnote ³⁴ Alexis, Hypnos fr. 244 (a passing reference to a male prostitute who worries about bad breath when kissing his erastês);Footnote ³⁵ Alexis, Carthaginian fr. 106 (the use of the uncommon word βακήλος, ‘eunuch’, as an insult directed at dissolute or effeminate men); and Eubulus, Antiope fr. 10 (which uses the word κόλλοψ as an insulting term in relation to a certain Callistratus, a politician ‘with a big bottom’ who allegedly goes about prostituting himself to other men).Footnote ³⁶

Nevertheless, apart from this handful of coarse witticisms, there are also a few other small clues suggesting that homosexual love affairs might occasionally have been more central to the plot of comic drama, or (at least) that they may have been dramatized in a more nuanced manner. Dover points to the existence of lost comedies such as Diphilus’ Pederasts, Eubulus’ Ganymede,Footnote ³⁷ and Antiphanes’ Ganymede, Thamyras and The Pederast. These titles are suggestive; unfortunately, the fragments tell us nothing of relevance. It is impossible to tell exactly how these plays treated their subject matter, or what attitude was adopted by the mythological titles to their myths in relation to the contemporary Athenian world. This list might perhaps be extended to include a couple of titles by Anaxilas: Euandria (‘Male Beauty’Footnote ³⁸ or simply ‘Manliness’) and Hyacinthus Pornoboskos (‘Hyacinthus the Pimp’, a title suggesting a crudely sexualized burlesque of myth).Footnote ³⁹

We also have a short extract from Amphis’ Dithyramb (fr. 15), in which an unidentified speaker criticizes a man for claiming, unconvincingly, that he values a pretty boy's character more than his body:

Even though this unnamed erastês is undeniably being mocked, the primary target of the humour seems not to be homosexuality as such. This passage is considerably more interesting (and funnier) than the other comic fragments just quoted, because it alludes to the notion, familiar from contemporary philosophy, that there is more than one type of love. Specifically it makes fun of ‘Platonic’ love: the idea, famously encountered in the Socratic dialogues, that the consummation of one's physical desires is unnecessary, and that bodily beauty is much less important than the beauty of a person's soul or other attributes such as wisdom or aretê.Footnote ⁴⁰ Obviously there were many down-to-earth types (in comedy and in real life) who found it impossible to credit this sort of high-mindedness, just as there were those who questioned whether Socrates himself could genuinely possess as much sexual self-control as he professed. But whether or not there were many fourth-century Athenians practising ‘Platonic’ love in real life, one would hardly expect any comedian to advocate sexual abstinence in his work. According to comic logic, anyone who claims not to be interested in sex must be either lying or deluding himself.

There is—just about—enough material among the plays and fragments to make us question whether sex in later Greek comedy was exclusively or fundamentally heterosexual in nature. It is also clear that none of these other pieces of evidence exactly resembles Damoxenus, fr. 3. We have to be cautious when interpreting a short, decontextualized excerpt of this sort; but apart from its length and level of descriptive detail, the Damoxenus fragment has four definitely demonstrable features which mark it out. First, the erastês is not obviously being presented as a ludicrous or contemptible figure: nothing that he says is self-evidently silly, nor is there any prurient focus on the physical details of sex (at least, not in the lines that survive). Second, the passage seems to show a genuine interest in what it actually feels like to experience homosexual desire, in that it centres on the psychological effects of desire on the lover as well as the physical characteristics of the beloved. Third, as already observed, this lovestruck man conveys his feelings using precisely the same sort of language and formulaic expressions as any other lover in comedy. Fourth, and most telling, in all the other surviving works and fragments we see the character of the erastês only from the viewpoint of another person, being lampooned or stigmatized as it might be, often merely in the form of passing references or incidental jokes. By contrast, Damoxenus puts the lover on stage and allows him to describe his own desires.

All these features prompt us to ask who this character was and where this speech came within the plot of the lost play. The shortage of material and the lack of contextual information make it impossible to know for sure; nevertheless, it is possible, with due caution, to suggest plausible scenarios into which the fragment might fit. The fact that his emotions are being presented in this way might suggest that the speaker is one of the central characters and a major focus of our attention. Could it be that he is the main character, and that his feelings for the young man are at the heart of the story? Typically in other New Comedy plot-lines the coup de foudre is the big event that sets the plot in motion. Glimpsing the beautiful object of desire and falling in love tends to happen near the start of a play, or to be narrated in a prologue speech or monologue as having taken place before the play begins. Perhaps we are looking at a speaker in a prologue explaining how he came to fall in love and preparing the audience for a plot in which the consequences of this passion are played out during the five acts to follow. A close parallel would be provided by Moschion's prologue speech in Menander's Samia, in which Moschion (the main character) explains how he fell for Plangon, the girl-next-door, and what the consequences were. Moschion's speech also—as in line 9 of the Damoxenus fragment (ἄνδρες)—makes use of direct address to the spectators, to increase their level of emotional engagement and empathy with the speaker's experiences.Footnote ⁴¹

Even if the fragment does not come from a prologue, and even if the speaker was not the main character of the play, this erotically charged encounter in the gymnasium must have had some sort of sequel. Unless we are to imagine that the besotted speaker simply walked off stage after delivering these lines, never to be mentioned again, something must have happened next to develop the theme of homosexual love. It seems inevitable that Damoxenus must have dramatized not just the crucial moment of falling in love but also the consequences of this type of sexual passion for the speaker and those around him. But many questions remain. Who exactly was the speaker? What were his age and social status? Was he an upper-class Athenian citizen? Was he rich or poor? Was he already married or betrothed? Was homosexual love presented as a viable alternative to ‘respectable’ married life? Did the play suggest that men should suppress their natural desires and accept cultural conventions? Was homosexual love seen as a cause of chaos and confusion to a greater extent than heterosexual love? Is there any special significance in the fact that the love object is from Cos, or at any rate not an Athenian citizen, just as so many female lovers and courtesans in comedy are presented as non-Athenian?Footnote ⁴² Is the figure of the boyfriend to be seen as somehow analogous, in structural terms, to the figure of the mistress or courtesan, as a problematic figure to be neutralized, removed or rehabilitated via matrimony by the end of the play?

It is impossible to answer any of these questions, but there are, I suggest, essentially three potential outcomes, all of which would have provided ample scope for comic entanglement, misapprehension and misunderstanding. First, the speaker eventually (perhaps with some difficulty) ignored or stifled his desire for the young man and returned to his wife or female lover. Second, the speaker (perhaps aided by friends or slaves) eventually contrived a situation whereby he could have both a wife and a boyfriend. Third, the speaker eventually chose a boyfriend in preference to marriage. These three outcomes are presented in decreasing order of probability, and the last is so unlike what we have come to see as the typical New Comedy ‘happy ending’ that it may strain credulity. But all three find partial analogues in other Greek and Roman comedies (provided that we replace the word ‘boyfriend’ with ‘girlfriend’). No matter which one scenario came to pass, the fact that Damoxenus featured a same-sex love interest in a comic plot at all is remarkable. He explicitly wanted to play around with the possibility of an alternative to heteronormativity within the world of comedy; to write a play which tackled the sort of questions posed in the previous paragraph. This sharply contrasts with all those other plays which relentlessly concentrated on heterosexual love and marriage and simply ‘swept homosexuality under the carpet’.Footnote ⁴³

It would be nice to imagine an Athenian comedy that affirmed same-sex love and avoided a drearily predictable conclusion in the form of yet another wedding feast. But even if Damoxenus’ play did end more conventionally, with the lover's passion for the young man being decisively thwarted or suppressed, this would in itself be unusual and highly significant, since it would be an otherwise unattested example of a comedy which set out to distinguish overtly between acceptable and unacceptable forms of sexuality.

On the basis of current evidence, it would be easy to conclude that such a play was a one-off, designed to épater la bourgeoisie or to pose a provocative challenge to the generic rulebook of New Comedy. Perhaps we might treat it as the exception that proves the rule. Nevertheless, it is not necessary to imagine that in its own time Damoxenus’ play was a unique experiment. Maybe those lost comedies with titles such as Pederasts and Ganymede (mentioned above) were comparable; maybe numerous other Greek comedies, of which no trace now survives, were doing the same sort of thing. After all, many real-life Athenian males in the third century b.c. did fall in love with young men: when they went to the theatre, might they not want to watch plays that showed other men behaving in the same way? Why assume that comedy was so different from other literary genres? Was the heterosexual marriage plot really a fixed generic convention? Could it be that we just perceive it as such, because of the choices made in the process of transmission and reception, or because of the preoccupations and prejudices of scholars in the centuries following Damoxenus’ heyday? Perhaps there was originally another strand of later Greek comedy that centred on alternatives to marriage and a less monochrome view of human sexuality. It is even possible to speculate that this strand of comedy was deliberately suppressed and quietly written out of history.Footnote ⁴⁴ If so, it is all the more amazing that any of these fragments survived at all. In this respect it is worth stressing again that the sexual content of fr. 3 was not the reason for its preservation; Athenaeus quoted the lines to illustrate something quite different. But, as in so many other cases, the ‘incidental’ or ‘accidental’ details contained in book fragments turn out to be even more valuable than the details that the quoting author deliberately chose to preserve in the act of excerption.

Fragments seldom allow us the satisfaction of definite conclusions, and I have none to offer here. Nevertheless, I hope to have demonstrated that Damoxenus’ fr. 3 constitutes an important and overlooked source for the historian of Greek comedy. I have also argued that fr. 3 and perhaps some of the other fragments discussed in this article represent vestigial traces of an alternative, unfamiliar, uncanonical variety of New Comedy that might have survived and flourished but (for whatever reason) did not. Whether or not the reader finds this argument persuasive, the fact remains that the Damoxenus fragment seems extraordinary, if not unique—and this is precisely what makes it such a good case study for ‘fragmentary thinking’. I started off by calling this fragment an ‘inconvenient’ piece of evidence, in the sense that it does not fit the accepted narrative. But unless we simply ignore it, we have to try to account for it somehow, even if this makes us question our knowledge or rethink our fundamental assumptions about ancient comedy and society. Such a process can be unsettling, especially when (as in this case) most of the questions we raise cannot be conclusively answered. But if it encourages us to see the world differently, then it is worth the effort: ὁ δ’ ἀνεξέταστος βίος οὐ βιωτὸς ἀνθρώπωι.Footnote ⁴⁵