¹ For intertextuality in Latin literary studies, see Conte, G.B. and Barchiesi, A., ‘Imitazione e arte allusiva. Modi e funzioni dell' intertestualità’, in Cavallo, G., Fedeli, P. and Giardiana, A. (eds), Lo spazio letterario de Roma antica, vol. 1, La produzione del testo (Roma 1989) 81–114 Google Scholar. See also Fowler, D., ‘Intertextuality and Classical Studies’, MD 39 (1997) 13–24 Google Scholar, and Hinds, S., Allusion and Intertext: Dynamics of Appropriation in Roman Poetry (Cambridge 1998)Google Scholar; see particularly Edmunds, L., Intertextuality and the Reading of Roman Poetry (Baltimore and London 2001)Google Scholar, who focuses exclusively on Roman poetry and devotes considerable attention to both Catullan and Horatian texts. My emphasis on the agency of the author, rooted in a determination to situate the poems of Catullus and Horace in their contemporary historical contexts, differs sharply from his. For example, Edmunds defines intertextuality as ‘the various ways in which one text can signal its relation to another’ (xi). He also refers to ‘authorial intention’ as lately repudiated in the field of Classics, glossing over the important differences between exercise of creative agency in literary composition on the one hand and conscious, purposive textual control fully knowable by both author and readers on the other. For a challenge to such assumptions, see the arguments of Heath, M., Interpreting Classical Texts (London 2002)Google Scholar who also advocates a historically informed understanding of classical literary texts.

I recognise and respect efforts by scholars such as N. Holzberg, who express scepticism on the historical reality of Catullus' ‘Lesbia’; see Holzberg, , Cantil. Der Dichter und sein erotisches Werk (Munich 2002) 16–18 Google Scholar. But, as I note below, the evidence from ancient authors such as Ovid and Apuleius that she was an actual individual cannot easily be dismissed and is further corroborated, albeit in an indirect manner, by various other ancient sources, Cicero among them, as discussed below.

² Longinus quotes four stanzas (plus one line) of Sappho's poem at 10.1-3 of On the Sublime, which was written between the first and third centuries CE. These stanzas are ‘translated’ by the first three stanzas of Canili. 51 (which lack the final line of the second stanza). For Catullus' additions to Sappho's original, see Wiseman, T.P., Catullus and His World. A Reappraisal (Cambridge 1985) 153–4Google Scholar; for the dialogic relationship between Sappho and Catullus, see Miller, P.A., ‘Sappho 31 and Catullus 51: The Dialogism of Lyric’, Arethusa 26 (1993) 298–308 Google Scholar.

³ Sappho's words are quoted by the fourth century BCE Demetrius, On Style, 146 (see Radermacher, L., Demetrii Phalerei qui dicitur De Elocutione libellus [Leipzig 1901] 34)Google Scholar.

⁴ Rayor, D., Sappho's Lyre. Archaic Lyrics and Women Poets of Ancient Greece. Translations, with Introduction and Notes (Berkeley and Los Angeles 1991)Google Scholar.

⁵ On these lines in Canili. 11 and 62, see Jenkyns, R., Three Classical Poets: Sappho, Catullus and Juvenal (London 1982) 50 Google Scholar; Wiseman (n. 2) 120-1 and 144-6; and Ancona, R., “The Untouched Self: Sappho and Catullan Muses in Horace, Odes 1.22’, in Spentzou, E. and Fowler, D. (eds), Cultivating the Muse: Struggles for Power and Inspiration in Classical Literature (Oxford 2002) 161–86Google Scholar. For the relationship between this passage in 62 and Sappho frag. 105c, see Quinn, K., Catullus: The Poems (London 1970) 280 Google Scholar. All translations from the Latin are my own.

⁶ ‘[Poem 35] brings us into the world of poetry … and in particular the world of Sappho, to whom Caecilius' girlfriend is compared’ (Wiseman [n. 2] 148).

⁷ For references to Sappho as “The Tenth Muse’, see Palatine Anthology 9.506 and 7.14Google Scholar, as well as the discussion of J. Hallett, P., ‘Sappho and her Social Context: Sense and Sensuality’, Singns 4 (1979)447–64Google Scholar.

⁸ Lesbia is named in poems 5,7,43, 51, 58,72,75,79,83,86,87,92 and 107. As Wiseman (n. 2) observes, the name ‘occurs sixteen times in Catullus' poems, always in the nominative or vocative cases’ (130).

⁹ Apul. Apol. 10: eadem igitur opera accusent C. Catullum quod Lesbiam pro Clodia nominarit (‘For the same reason let them accuse Gaius Catullus, because he allegedly used Lesbia as a name for Clodia’). At Trist. 2.427–30Google Scholar, Ovid also claims that Catullus used Lesbia as a pseudonym for the woman with whom he was actually engaged in an illicit love affair: sic sua lascmo cantata est saepe Catullo / femina, cui falsum Lesbia nomen erat. / nec contentus ea, multos vulgavit amores, / in quibus ipse suum fassus adulterium est (‘Thus often his woman was sung about by sexually playful Catullus, a woman for whom the false name of Lesbia was employed. And not satisfied with her, he published many love poems in which he admitted to his own adulteries.’).

¹⁰ On this Clodia, daughter of Appius Claudius Pulcher, consul in 79 BCE, see Skinner, M.B., ‘Clodia Metelli’, TAPA 113 (1983) 273–87Google Scholar, and Catullus in Verona: A Reading of the Elegiac Libellus, Poems 65-116 (Columbus 2003) 81, 91 and 93 Google Scholar; Wiseman (n. 2) 15-53; and Hemelrijk, E.A., Matrona Docta: Educated Women in the Roman Elite from Cornelia to Julia Domna (New York and London 1999) 174–5Google Scholar.

¹¹ For Clodius as Catullus' Lesbius pulcher and at the Bona Dea rites, see, inter alios, Suet, . Iul. 46 Google Scholar; Plut, . Caes. 9 ff.Google Scholar and Cic. 38; Cic, . Att. 1.12.3 and 13.3Google Scholar; Dio 37.45.1. See also Skinner, , ‘Pretty Lesbius’, TAPA 112 (1982) 119–27Google Scholar and (n. 9: 1983) 80-1 and 89-90; Wiseman (n. 2) 22-53 and 66-91.

¹² For Cicero on Clodia Metelli, see Skinner (n. 9: 1983) and (n. 9: 2003) 81-2, 93-4, and Wiseman (n. 2) 54-91. For Catullus and Marcus Caelius Rufus, see Skinner (n. 9: 2003) 91, 93,125 and 179, and Wiseman (n. 2) 134,136,164 and 69-173.

¹³ Odes 1.22.24 Google Scholar, dulce loquentem (‘sweetly speaking’) also ἆδυ ϕωνείσας at lines 3-4 of Sappho 31 L-P; Catullus does not appear to render this phrase into Latin. For Horace's evocations of Sappho and Catullus in Odes 1.22 Google Scholar, see Ancona (n. 4); for the poem's debt to Lesbian lyric, see Davis, G., ‘ Carmina/Iambi: The Literary-Generic Dimension of Horace's Integer Vitae (C. 1.22)’, Quaderni Urbinati de Cultura Classica 27 (1987) 67–78 CrossRefGoogle Scholar.

¹⁴ See Santirocco, M.S., Unity and Design in Horace's Odes (Chapel Hill and London 1986) 18 Google Scholar, who also notes Horace's acknowledgments of poetic debt to Pindar and Callimachus as other Greek sources of inspiration; Horace, however, does not mention either of these two poets by name here.

¹⁵ For popularis as ‘of the country to which one belongs’, see the entry in OLD (1404) s.v. popularis.

¹⁶ For the possible interpretations of the term mascula in this context, see the comments of Porphyrio on this passage ( ‘mascula’ autem ‘Saffo’, vel quia in poetico studio est [incluta], in quo saepius viri, vel quia tribas diffamatur fuisse (“Masculine Sappho’, either because she is famous for her dedication to poetry, for which men more often gain fame, or because she is insultingly said to have been a tribad’); Hallett (n. 6); and Brooten, B.J., Love Between Women: Early Christian Responses to Female Homoeroticism (Chicago 1996) 341 CrossRefGoogle Scholar. For the date of this poem, see Fraenkel, E., Horace (Oxford 1957) 339 Google Scholar, who comments that ‘the word mascula is not dragged in from the chronique scandaleuse; it is essential to the thought of the passage. It is, Horace suggests, extraordinary for a woman to walk in the footsteps of that most virile poet Archilochus, but Sappho … could do so, for the power of her poetry was as great as any man's’ (346).

¹⁷ Some manuscripts have the masculine discipulorum rather than discipularum; the association, however, between female pupils and cathedrae, high-backed chairs, at Mart. 3.63.71 and Juv. 6.91, as well as the scholiast's comments would establish the reading discipularum. For the fictive or at least pseudonymous figures of Demetrius and Tigellius/Hermogenes, see Rudd, N., The Satires of Horace (Cambridge 1966) 123-34 and 292 Google Scholar; for the date of Sat. 1.10 Google Scholar, and the historicity of individuals named within it, see Rudd 136-59.

¹⁸ For the associations between the verb plorare and the wailing, indeed the wailing of women, portrayed as characteristic of Latin love elegy in its historical guise as a genre of lament, see, for example, Ov, . Am. 3.9.1–3 Google Scholar: Memnona si mater, mater ploravit Achillen, / et tangunt magnas frisila fata deas; / flebitis indignos. Elegia, solve capillos (‘If his mother wept for Memnon, if his mother wept for Achilles, and the sad fates touch the great goddesses, sorrowing Elegy, loosen your unworthy locks’). A lament for the recently deceased elegiac poet Tibullus, it defines the literary and thematic concerns of the elegiac genre, here personified as a mourning woman.

¹⁹ Lydia also figures in Odes 1.8 and 1.13Google Scholar; Chloe in 3.7 as well as 1.23. On the dramatic scenario of Odes 3.9 Google Scholar, see Johnson, T., ‘Locking-in and Locking-out Lydia: Lyric Form and Power in Horace's C. 1.25 and III.9’, CJ 99 (2004) 113–24Google Scholar, and Putnam, M.C.J., ‘Horace, Odes 3.9: The Dialectics of Desire’, in D'Arms, J. and Eadie, J.W. (eds), Ancient and Modem Essays in Honor of Gerald F. Ebe (Ann Arbor 1977)Google Scholar.

²⁰ Line 19, spoken by the male lover - si flava excutitur Chloe (‘if golden-haired Chloe is tossed out’) - may recall Canili. 65.19-22: malum, / procurrit casto virginis e gremio / quod miserae oblitae molli sub veste locatum / dum adventu matris prosila, excutitur (‘An apple has fallen out of the virtuous lap of a maiden, after it was placed under the soft garment of the poor girl, who forgot about it, when, at the arrival of her mother, she leaps up and it is tossed out’). Here the verb form excutitur, used by Horace to describe the potential rejection of a literarily talented female lover, is associated with literature, and with women who love as well: Catullus equates his inadvertent, ‘mental’ tossing out of a promise to write a poem with a maiden's forgetful tossing out of an apple secretly given her by her fiancé. For more on this passage, see below.

²¹ For Lydia and Chloe as names of former slaves, freedwomen of Greek birth who sold their sexual services as hetaerae, long-term companions for hire, see Garrison, D.H., Horace. Epodes and Odes. A New Annotated Latin Edition (Norman 1991) 215 and 310 Google Scholar; for the patronymic Tyndaris, and the paternity of the mythic Helen, see OLD (1998) s.v. Tyndareus.

²² Santirocco (n. 13): ‘Finally, even Tyndaris herself carries poetic associations. The wine she will drink, for instance, is a Lesbian vintage (21), and thus may conjure up Horace's models, Sappho and Alcaeus. More definite, however, is the fact that Tyndaris is herself a poet… It is interesting that both the musical instrument and the subject of the song are lyric: the Teian provenance of Tyndaris' lyre conjures up Anacreon, and though the persons of whom she sings are epic characters, their action here is not heroic but lyric: the struggle of two women for the affections of one man’ (52).

²³ On the Roman sexual double standard, see, for example, Treggiari, S., Roman Marriage. lusti Coniuges from the Time of Cicero to the Time of Ulpian (Oxford 1991) 221-2,281,299-319, 511–3Google Scholar.

²⁴ It is also worth noting that Horace portrays women as speaking in a few of his other poems. Odes 1.30 Google Scholar is an eight line invocation to Venus placed in the mouth of one Glycera, apparently another Greek freed woman. The final three and one-half lines of Odes 4.6 Google Scholar are assigned to a bride. Epodes 5, 12 and 17 feature female speakers too. Epode 5 is a dramatic dialogue between the witch Canidia and a young boy she is about to kill; Canidia also speaks in lines 53 ff. of Epode 17, lashing out at the poet for ridiculing her. In lines 13-20 of Epode 12 a nameless woman complains of the poet's sexual neglect; strikingly, in line 17 she curses one Lesbia for originally bringing them together. Garrison associates this ‘matchmaker’ with Catullus' Lesbia and characterises her as a ‘promiscuous married woman … another rich lady who goes in for midnight cowboys’; Horace may be offering a veiled critique of the erotic scenarios represented by Catullus' Lesbia poems as well.

While these representations of female speakers do not - like those of Chloe and Tyndaris -depict them as poets, Horace does assign these women words of their own. But lines assigned to women constitute only a small percentage of those in Horace's poetic corpus; as we shall now see, women's words occupy a far greater percentage of the lines bequeathed by Catullus.

²⁵ For the affinities between Catull. 45 and Hor, . Odes 3.9 Google Scholar, see, for example, Garrison (n. 20) 310, and Nielsen, R., ‘Catullus 45 and Horace, Odes 3.9: The Glass House’, Ramus 6 (1977) 132–8CrossRefGoogle Scholar.

²⁶ For these women's voices, see Hallett, , ‘Women's Voices and Catullus' Poetry’, CW 95 (2002) 421–4Google Scholar. As Skinner (n. 10: 2003) 202 observes, the speech of the door in 67 has been interpreted as similar to that of a respectable matron as well as that of a menial household attendant. Perhaps Catullus is deliberately characterising this female speaker as both: such literal ‘liminality’ bespeaks an ambiguity of status resembling that signaled by the words of Artis (a man rendered a woman) in 63 and the coma Berenices (a lock of hair rendered a comet) in 66. On the door, see also J. Deuting, ‘Catullus 17 and 67, and the Catullan Construct’, in this issue.

²⁷ See Cantull. 64.60 (quem procul ex alga maests Minois ocellis ), 135 (immemor a devota domum periuria portas), and 148 (nihil veriuria curant).

²⁸ For the poems in Tibullus Book 3, and M. Valerius Messalla Corvinus, see Skoie, M., Reading Sulpicia. Commentaries 1475-1990 (Oxford 2002) 9–10 Google Scholar; for Messalla's career, see Syme, R., The Augustan Aristocracy (Oxford 1986) 200–6Google Scholar.

²⁹ For Sulpicia's mother Valeria, see Jerome, , Ad Iovinianum 1.46 Google Scholar, and the discussions of Skoie (n. 27) 10, and Hallett, , ‘Sulpicia and the Valerii: Family Ties and Poetic Unity’, in Amden, B., Jensen, P.F., Nielsen, T.H. and Tortzen, C.G. (eds), Noctes Atticae: Articles on Greco-Roman Antiquity and its Nachleben. Studies Presented to Jorgen Mejer on his Sixtieth Birthday (Copenhagen 2002[a]) 141–9Google Scholar, and ‘The Eleven Elegies of the Augustan Poet Sulpicia’, in Churchill, L.J., Brown, P.R. and Jeffrey, J.E. (eds), Women Writing Latin. From Roman Antiquity to Early Modern Europe. Volume 1. Women Writing Latin in Roman Antiquity, Late Antiquity and the Early Christian Era (New York and London 2002[b]) 45–65 Google Scholar. For the relationship between C Valerius Catullus and these Valerii, see Wiseman, , ‘Sirmio, Sir Ronald, and the Gens Valeria ’, CJ 88 (1993) 223–9Google Scholar.

³⁰ On the identity of this Lygdamus, see Butrica, who also argues that Sulpicia's brother collaborated with her on [Tibullus] 3.8-18 and is the author of 3.8,10 and 12.

³¹ See Tibull. 3.4.89-91 [Lygdamus]: Scvllaque virgineam canibus succincta figurant, / nec te conceptam. saeva leaena tulit, / barbara nec Scythiae tellus horrendave Syrtis . See Catullus 64.154-156: quaenam te genuit sola sub rupe leaena. / quod mare conceptum spumantibus exspuit undis / quae Syrtis. quae Scylla rapax, quae vasta Charybdis. For other echoes of Catullus in Lygdamus' poetry, see Parca, M., “The Position of Lygdamus in Augustan Poetry’, in Deroux, C. (ed.), Studies in Latin Literature and Roman History IV, Latomus 196 (1986)461–74Google Scholar.

³² See Hemelrijk (n. 9) 175, who states: ‘Cicero mentions [Clodia's] ‘pantomimic [or balletic] interludes’, perhaps hinting at her marital interludes, her love affairs … His words should not be taken literally: they do not necessarily mean that she actually wrote mime or was a librettist of pantomime, as Wiseman believes; because of the strong sexual associations of the stage they must have been meant to blacken her.’ In a note on 337, which takes further issue with Wiseman (n. 2) 37-8, Hemelrijk then observes that a scholiast on this passage, Schol. Bob. 135 ( Stangl, T., Ciceronis orationum scholiastae [Hildesheim 1912])Google Scholar, interprets Cicero's remark as evidence that Clodia danced in pantomime: she claims that this statement refutes the possibility of Clodia's writing libretti.

But it is perfectly possible that Clodia not only performed in but also created the scripts and dance routines of pantomimes. By the same token, owing to the negative connotations of the theatrical involvements by respectable Roman women, Cicero's sneering assertions in a public forum that Clodia wrote pantomime libretti constitutes an insult to her reputation, even if Cicero is telling the truth.

It also warrants attention that the interpretation of embolia as ‘ballet interludes’ in this passage is far from certain. Cicero also employs embolium at QF 3.1.24 as a technical literary term, to describe his own insertion of a dramatic episode into his writing on contemporary events. Indeed, Cicero's use of embolium at Sesi. 116 and in this letter to his brother are the only two instances of this Greek word in Latin, quoted by the entry in OLD (602) s.v. embolium. Clodia may well have inserted her own dramatically-flavoured writings into other works, or works by others, prose or poetic.

³³ For poetria applied to Sappho elsewhere, see, for example, Ov, . Heroid. 15. 183 Google Scholar and Galen 4.771. On the Greek noun poetria in Latin generally, see OLD (1396) s.v. poetria; the entry cites only three instances of the word in Latin texts, those by Cicero and Ovid in the Pro Caelio and Heroides 15 respectively, and one by the late second century CE Terentius Maurus.

For the claim that Clodia could not have been a poet in her own right since Cicero ‘had much to gain by associating Clodia with the stage because of its bad reputation’ and therefore ‘puns on the double meaning of fabula, which means both ‘play’ and ‘fictitious tale’, in order to discredit her as a witness for the prosecution’, see Hemelrijk (n. 9) 175.

³⁴ For the literary connections of this [Caecilia] Metella, see Wiseman, , Cinna the Poet and Other Roman Essays (Leicester 1974) 188–91Google Scholar. To be sure, legitur may mean ‘read about' as well as ‘read’. But Ovid uses lego in the passive voice with a literary figure as the subject, in the sense of ‘have one's writings read’, elsewhere in Tristia 2: e.g. 370 (on Menander) and 463 (on Tibullus). Hemelrijk (n. 9) 320 points out that in his commentary on the Tristia, Georg Luck maintains that ‘Metella, celebrated under a pseudonym in the poetry of her admirers, now wrote poetry herself under her own name.’ Hemelrijk, however, concludes that ‘this hypothesis seems too shaky to include a Metella among our poetesses.’ For Propertius' portrayal of Cynthia as a poet at 1.2.27-30 and 2.3.21-22, see Hemelrijk, 79-80 and 274-5.

³⁵ See Janan, M., “When the Lamp is Shattered’ Desire and Narrative in Catullus (Carbondale 1994) 78 Google Scholar: ‘Catullus sets himself up as ‘reader’ of Lesbia-as-text, in his obsessive attention… to decoding her speech: he obsessively investigates her ‘true’ meaning and her true desire.’ See also Skinner (n. 9:2003) 66-8.

³⁶ So 7.1-2: quaeris quotmihi basiationes / tuae, Lesbia, sint satis superque (‘You ask, Lesbia, how many of your kisses are enough and more than enough’); 72.1-2: dicebas quondam solum te nosse Catulium, / Lesbia nec prae me velit tenere Iovem (‘You used to say, Lesbia, that you knew Catullus alone [or Catullus alone knew you], and that you did not want to hold Jupiter above me’); 109.1-2: iucundum, mea vita, mihiproponis amorem I nunc nostrum inter nos perpetuumque fore (‘My life, you propose that this love of ours will always be abiding among us’). Although scholars such as Skinner (2003) 66 regard the speaker of 72.7 as Lesbia, because Catullus has been addressing her so far, V. Pedrick - as Skinner notes on 207 - views the interlocutor as an ‘eavesdropper’ or member of the reading audience. See Pedrick, , ‘ Qui Potis Est, Inquis? Audience Roles in Catullus’, Arethusa 19 (1986) 187–209 Google Scholar.

³⁷ So 36.3-4: nam sanctae Veneri Cupidinique / vovit, si sibi restituais essem / desissemque truces vibrare iambos … (‘For she vowed to holy Venus and Cupid, if I were to have been returned to her, and to have stopped hurling fierce iambic verses . .’); 70.1-3: nulli se dicit mulier mea nubere malle / quam mihi, non si se Iuppiter ipse petal. / dicit (‘My woman says that she prefers to marry no one desiring her as much as me, not if Jupiter himself should seek her. She says this’); 83: Lesbia mi praesente viro mala plurima dicit: / haec illi fatuo maxima laetitia est. / mule, nihil sentis? si nostri oblita taceret, / sana esset: nunc quod gannii et obloquitur, / non solum meminit, sed, quae multo acrior est res, / irata est. hoc est, uritur et loquitur (‘When her husband is around Lesbia says a great many insulting things to me: this is the greatest source of happiness to that foolish man. Idiot, do you perceive nothing? If she, having forgotten about us, were to be quiet, she would be of sound mind: now because she snarls and insults, she does not merely remember, but, a situation that is much more striking, she is angry. This is the reality: she bums and speaks’); 92.1-2: Lesbia mi dicit semper male nec tacet umquam / de me (‘Lesbia always speaks ill of me and is never silent about me’).

³⁸ For Catullus' foregrounding of his own literary efforts, and the reactions of others to his poetry, see, for example, 1, 6,16,36,42,49, 50, 65,69,116.

³⁹ A shorter version of this argument about Catull. 2b and 51 may be found in Hallett (n. 25). The lively debate over the relationship between the representations of the female beloved in admittedly fictionalised Latin love poetry and historical Roman reality shows no sign of stopping. Hemelrijk's contribution (176 ff.) is worth quoting, since it is relevant to my discussion of the relationship between Catullus' poetry and that by Sulpicia: ‘Of course poetry is not the same as autobiography, and the realist approach that attempts to detect the ‘real’ woman behind the pseudonym and to describe her life and outward appearance on the basis of the poems seems naive. Yet, the opposite view, which reduces these women to symbols of the creative process of writing or metaphors for the author's poetic ideals, is also unsatisfactory. Though it rightly attacks the simple and direct relation posited by the ‘realists’ between the poems and reality, the assumption that the women figuring in Roman elegy bear no relation whatsoever to real Roman women is unconvincing … [It] must be taken into account that ancient readers never seem to have questioned the reality of the elegiac mistresses. This does not mean that we should believe that elegiac poetry comprises precise portrayals of identifiable women, but it does indicate that the women figuring in love poetry were sufficiently true to life to be convincing… [And as] the relation between literature and life is reciprocal … we should not simply ask how the puellae of love poetry were inspired by contemporary Roman women but also to what extent historical women were influenced by literary creations such as the puellae of the love poets.’

⁴⁰ See Fordyce, C.J., Catullus: A Commentary (Oxford 1961) 91 Google Scholar, for the separation of these three lines from poem 2; Wiseman (n. 2), who underlines the parallels with poem 61 as well as with 65, refers to it as ‘a poem of which only three lines survive’ (121). I translate the neuter adjective gratum as ‘this experience’, i.e. playing with the passer. On the possibility of Lesbia as the speaker in 2b and elsewhere, see Johnson, Marguerite, ‘Catullus 2B: The Development of a Relationship in the Passer Trilogy’, CJ 99 (2003) 11–34 Google Scholar.

⁴¹ Wiseman (n. 2), for example, regards these three lines as a separate poem. Similarities in vocabulary between Canili. 2 and Suet, . Tib. 43–8Google Scholar, however, also render the connection between 2 and 2b more plausible: in this passage Suetonius describes Tiberius' sexual activities and aids, including a painting of Atalanta engaged in oral sexual endeavour (Porrosi quoque tabulam, in qua Meleagro Atalanta ore morigeratur). These similarities include ut aspectu deficientis libidines excitaret (recalling 2.4: incitare); nedum credi fas sit (8: credo); quasi pueros primae teneritudinis, quos pisciculos vocabat, institueret, ut natanti sibi inter femina versarentur ac luderent lingua morsuque sensim adpetentes (2-5: quicum ludere, quem in sinu tenere / cui primum digitum dare appetenti / et acris solet incitare morsus: 9: ludere ): also striking is the use of an animal name - little fishes - or Tiberius' sexual playmates, and a diminutive one - like solaciolum in 7 - at that; solitus inludere (2: ludere . 4: solet. 9: ludere). Significantly, various words in Suetonius’ description - libidines and libidinis; naturarti ligurire - recall another Catullus passage, cited by Nonius (200 Lindsay): Catullus Priapo: de meo ligurrire libidost.

For the ‘obscene interpretation of the poem’, see, for example, Hooper and Thomas. Needless to say, the apparent echoes of Catullus 2 in Suetonius, not mentioned in these studies, renders such a reading - and a reading in classical Roman times, if not necessarily in Catullus' own day - more legitimate. Hooper, R.W., ‘In Defence of Catullus' Dirty Sparrow’, Greece and Rome 32 (1985) 162–78CrossRefGoogle Scholar, and Thomas, R.F., ‘Sparrow, Hares, and Doves: A Catullan Metaphor and Its Tradition’, Helios 20 (1993) 131–42Google Scholar.

⁴² For the similarities and other connections between the apple descriptions in 2b and 65, see Wiseman (n. 2), 121-2. See also Skinner (n. 9: 2003) 14-15 for a further discussion of this simile; as she observes, ‘it is not quite clear what is being compared: is it the tenor of the simile the carmina Battiadae Catullus sends Hortalus, insofar as both apple and poetry are gifts, or the dicta, the words Hortalus had previously voiced, which may appear to have been forgotten, like the apple … the ingredients of the simile do not correspond in simple one-to-one fashion with elements in the framing text.’

⁴³ See, for example, Wiseman (n. 2) 154-5, who proposes that ‘the change of direction in the final stanza is not a total surprise. The reader now sees that the version of Sappho in the first three stanzas was, as it were, in inverted commas: the ‘Catullus’ who addressed them to the woman is now himself addressed, and his malaise diagnosed. As in poem 8 (the last Lesbia-poem to contain the vocative Catulle), we must distinguish the persona, Catullus the lover, from Catullus the poet. The former is allowed to masquerade as Sappho, but the latter has the final comment, and it is a bitter one. These physical symptoms are the sign of a self-destructive emotional excess.’

While I agree with Wiseman that the final stanza provides a diagnosis of Catullus' malaise, there are substantial differences between poems 8 and 51. In 8 the puella is never identified by the name Lesbia, and Catullus also speaks of himself in the third person (13-14 obdurat, requiret, rogabit) as well as in the second person singular. Furthermore, while Catullus refers to himself with the pronoun tu at 8.7 and 9, and then applies the same pronoun to his puella at 14, his address at 12, vale, puella!, makes it clear that he is shifting second-person addressees from himself to the puella; there is no similar signpost in 51. It is, of course, possible that Catullus has modeled the final stanza of 51 on verses in Sappho 31 L-P now lost tous.

⁴⁴ For the thematic relationship between otiosi in 50 and ottum in 51, see, for example, the discussion of Finamore, J., ‘Catullus 50 and 51: Friendship, Love and Otium ’, CW 78 (1984) 11–19 Google Scholar.

⁴⁵ For some Homeric dimensions of poem 68, thematic and stylistic, see Skinner (n. 9: 2003) 160-1. Admittedly, the negative attitude that Catullus apparently adopts at 68.103-4 towards the extramarital affair between Helen and Paris - which he describes with the word otium - may seem inconsistent with his portrayal of the relationship between himself and Lesbia: in both 68 and elsewhere in his poetry he represents himself and Lesbia as engaged in a similar illicit liaison. But, as Wiseman (n. 2) 115-8 and 161-4 observes, Catullus not only idealises marriage but often describes his affair with Lesbia in words associated with marital unions.

⁴⁶ Skinner (n. 9: 2003) 100 adds: ‘We further observe that her language in poem 70, for all its amusing hyperbole, is glib: ostensibly professing love, she smoothly evades actual commitment. Verbal finesse glosses over a lack of ethical, as well as emotive depth: hence at 72.5-6 the speaker who at last understands her (nunc te cognovi) finds her multo… vilior et levior.’

⁴⁷ In this elegiac endeavour, the first two lines slightly revise the opening couplet of 70 (nulli se dicit mulier mea nubere malle / quam mihi, non si se Iuppiter ipse petat), drawing on the word cupido and present active participle amanti in 70.3 (sed mulier cupido quod dicit amanti) for the word capienti, and substituting a dactyl and spondee for the two dactyls in 70.1. The phrase praeterea est beginning the third line does not appear in Catullus' poetry, but he frequently uses the adverb praeterea and elides the final syllable at e.g. 99.11 (praeterea infesto miserum me tradere amori); est gratum appears at 107.2 and 3; solum me nosse Catullum slightly revises solum te nosse Catullum in 72.1. As for the final line, it is a slight revision of the final line of 68,160, lux mea, qua viva vivere dulce mihi est.

⁴⁸ What I have done in rewriting the hendecasyllabic 36, and putting together a pastiche from several Catullan elegies, is first and foremost an exercise in Latin verse composition, one which involves transforming and re-positioning words from actual Catullan texts, and one which could easily be adapted for students in secondary and university classrooms. But it is also a ‘textual operation’ comparable to that routinely performed by textual critics determined to ascertain what ancient Roman authors originally wrote, as opposed to what various manuscripts attest. As such, it represents a form of hypothetical thinking long accorded credibility and indeed intellectual prestige in the professional study of classical philology.

⁴⁹ It is also noteworthy that Catullus' descriptions of what is attractive about Lesbia in poems 43 and 86 emphasise her witty speech and sophisticated ways (presumably including ways of speaking). In the former, set up as an ironic contrast between the shortcomings of Mamurra's beloved and the appealing features of Lesbia, Catullus criticises the former for her nec sane nimis elegante lingua, ‘tongue [or language] that is not sufficiently sophisticated’, implying that Lesbia's lingua is extremely elegans. In the latter, contrasting Lesbia favourably to one Quintia, he faults Quintia for lacking what Lesbia possesses in abundance (3-6): nam nulla venustas, / nulla in tarn magno est corpore mica salis. / Lesbia formosa est, quae cum pulcherrima Iota est, / turn omnibus una omnis surripuit Veneres (‘For there is no erotic sophistication, no spark of wit in so great a body. Lesbia is beautiful, who is not only altogether most physically appealing, but also has, alone, stolen all the erotic charms from all women’).

⁵⁰ On the history of the ‘orthodox view’, and Sulpicia's ‘friend’ (amicus), sometimes known as the ‘author’ (auctor) or the ‘garland poet’, see the detailed study of Skoie (n. 27) especially 162-212.

⁵¹ See Hallett (n. 28:2002a) and Hallett (n. 28:2002b). It is, of course, impossible to prove that Sulpicia actually wrote any of these eleven elegies, or that Messalla's niece even existed. No ancient source independently attests to her existence; while echoes of her writing may be discerned in other Augustan writers, such indirect intertextual evidence is not the same thing as direct testimony. But rather than expend efforts on arguing that she did not write some or all of these poems, another ultimately unproveable proposition, I focus my energies on evidence which supports the view of a single author, an aristocratic and literarily learned young woman contemporary with Ovid and Tibullus, kindred to their patron Messalla. There is much, albeit indirect, evidence of this kind.

⁵² For the significance of the pseudonym Cerinthus - found in 3.9, 3.10, 3.11, 3.14 and 3.17 -in Sulpicia's poems, see Roussel, D., ‘The Significance of the Name Cerinthus in the Poems of Sulpicia’, TAPA 120 (1990) 243–50Google Scholar. As Skoie (n. 27) 127-8 observes, the German scholar Heyne refutes the possibility of Cerinthus being the Cerinthus mentioned by Horace in Satires 1.2 - or at least poses the question rather rhetorically (quis affirmet?). But the idea that Sulpicia chose the pseudonym to evoke this particular passage in Horace is not considered.

⁵³ It merits attention that Lygdamus, who pays homage to Catullus directly in 3.6.39-42 as well as through evocations of Catullus' words, also adopts a resistant pose in echoing Horace. For at 3.1.6 and 23,3.2.12 and 29,3.3.1 and 23,3.4.57 and 60, and 3.6.29 he refers to his beloved by the pseudonym Neaera, recalling - in Garrison's words (n. 20) on 191 and 317 – ‘the central figure in Pseudo-Demosthenes' In Neaeram', ‘a famous Greek hetaera of the fourth century’ BCE. Yet Horace had earlier used the same name, in Epodes 15 and Odes 3.14 Google Scholar, for a woman he portrays in the former poem as unfaithful to him, and in the latter poem a woman described as merely a potential sexual playmate with whom he might celebrate Augustus' return from Spain in 24 BCE. Furthermore, in bewailing Neaera's preference for non-stop nights (assiduas nodes) with a rival at Epodes 15.11-16, he speaks of himself in a highly personal way - as Flaccus, one of only four occasions in which he identifies himself by his own cognomen.

To be sure, Lygdamus portrays himself as at times unhappy in his love affair. Indeed, 3.4 centres on a nightmare in which Apollo, god of poetry, reveals to him that Neaera prefers to marry someone else; in 3.6 she cares for him no longer. Yet he depicts his Neaera as a woman of his own exalted social station, and as far more responsive to him, as both a lover and a poet, than Horace's woman (or women) with the same name. At 3.1.27-8, in fact, he depicts her as once his wife; at 3.2.29-30 he imagines that his tombstone will specify the cause of his death as his passion for ‘his wife, Neaera, snatched from him’ (Lygdamus hic situs est. dolor huic et cura Neaerae / coniugis ereptae causa perire fitit). Like Sulpicia, therefore, Lygdamus recalls Horace only to resist him: he has created an all-important beloved for himself who shares a name with a minor character in Horace's poetry, but differs strikingly from her Horatian namesake.

⁵⁴ Earlier versions of this paper were presented at St Joseph's University in February 2004 and at a summer institute on Catullus and Horace for secondary school Latin teachers, funded by the National Endowment for the Humanities, held at the College of Notre Dame of Maryland in July 200S. I would like to thank my hosts on both occasions, Maria Marsilio and Sister Therese Marie Dougherty, several of those who attended these presentations - most notably Henry Bender, Eleanor Winsor Leach, Barbara McManus and Ann Raia - as well as Sheila Dickison, Donald Lateiner and the anonymous referee for this journal.