1 On my count, fifty-eight of the poems are wholly or partly invective in tone and content, while only twenty-six can reasonably be numbered as love poems, even on a broad definition (I have double-counted those, such as 11, 36, and 37, which both concern the speaker’s love for Lesbia and denigrate either his rivals or the beloved herself). For Catullus as a love poet, see esp. Prop. 2.25.3–4, 2.34.87–8; Ov. Am. 3.9.62; Ov. Tr. 2.427–30; Mart. 8.73.5–10, 12.44.5–6. For Catullus as an invective poet, see Quint. 10.1.96; Tac. Ann. 4.34.8; Suet. Iul. 73. Testimonia are conveniently assembled by Wiseman 1985: 246–62.

2 Infamously, Fordyce excluded from his commentary ‘a few poems which do not lend themselves to comment in English’ (Fordyce 1961: v). Notable exceptions to the prevailing climate of prudishness include Housman 1931 (translated 2001); Krenkel 2006b (a collected volume of essays originally published between 1963 and 2000); and Adams 1982. Richlin’s frank and un-euphemistic study The Garden of Priapus (Richlin 1992; first edition 1983) was an important milestone.

3 Quinn 1972: 23–6 is – or was – an influential statement of the view (dating back to Quinn 1969: 27–43) that many of the poems can and should be relegated to a lower ‘level of intent’ and treated less seriously than others. Accordingly, while insisting on the unity of the collection, Quinn devotes some 150 pages to the Lesbia poems and only 80 to what he calls the ‘poetry of social comment’, a heading under which he includes – for example – poems 10, 13, 35, and 45, alongside political and personal invectives.

4 E.g. Diog. Laert. 5.87 (Heraclides Ponticus’ treatise ‘On Archilochus and Homer’); Antipater of Thessalonica, Anth. Pal. 11.20.5; Dio Chrys. 33.11–12; [Longinus], On the Sublime 13.3.

5 Infamously, Archilochus unblushingly proclaims himself a rhipsaspis – one who drops or casts away his shield in battle to facilitate his own escape (fr. 5); Hipponax’s self-depiction as an impoverished beggar in frs. 32 and 36 seems to have been characteristic. For iambic poetry’s tendency to reinforce communal values, see e.g. C. G. Brown 1997: 41–2; Swift 2019: 24–8.

6 Newman 1990: 43–75; Wray 2001: 167–86; Cowan 2015, 2021. See also Heyworth 2001; Lavigne 2010; Hutchinson 2012: 75–8; Swift 2019: 48–9. For the distinction between Callimachean and Archilochean ‘code models’, see esp. Wray 2001: 167–203. Hill (2021: 166–7), in contrast, argues that Catullus differentiates between two Callimachean modes, the elegiac and the iambic.

7 See esp. Macleod 1973; also Forsyth 1977a; Newman 1990: 45–6; Davies 2000; Wray 2001: 189–90; Skinner 2003: 1–3, 21–2; Hill 2021.

8 The text of line 1 is problematic: some editors (e.g. Fordyce, Goold) retain the manuscript studioso and/or amend uenante, which is undeniably awkward without an object.

9 For the conjecture aucta (‘accumulated’, ‘stockpiled’), see McKie 2009: 173–4; the manuscripts offer the meaningless amitha (O) or amicta (GR).

10 The archaism and quotation marks are intended to reflect the striking reminiscence of Ennius, Annals fr. 95 Skutsch (Romulus speaking, nam mi calido dabis sanguine poenas, ‘you will pay me the penalty with your warm blood’): Catullus (self-)mockingly assimilates his poetic vengeance on Gellius to Romulus’ killing of his brother, Remus. The archaizing suppression (technically, ‘ecthlipsis’) of the final –s – a feature to be found regularly in Lucretius but otherwise avoided by Late Republican and later poets – underlines the Ennian echo, as does the holospondaic line 3 (the only instance in post-Ennian Latin). At the same time (pace Hill 2021), the archaic roughness of technique offers a self-consciously reflexive contrast with the Callimachean ideals of literary polish and refinement connoted by the adverb studiose, ‘assiduously’.

11 And/or as sexual assault: like other words for weapon, telum, ‘shaft’, can serve as a euphemism for the penis, as my translation is intended to suggest (Adams 1982: 19–20); in this context, caput is arguably suggestive of oral penetration (so Richlin 1992: 149–50; cf. Adams 1982: 192).

12 Alongside general similarities of theme and characterization: as we have noted, the iambic speaker, like Catullus (cf. Chapter 1, pp. 14–19, above), typically presents himself as a (disreputable) outsider, and both Archilochus and Hipponax thematize sexual misconduct (see n. 35 below), including fellatio and incest. Archilochus represents himself as a rejected lover, vilifying his former betrothed and her family, and mourns the loss of his brother; characteristically Hipponactean themes shared by Catullus include the speaker’s impoverishment (cf. poem 13), while frs. 115–117 West (variously assigned to Archilochus and Hipponax) combine accusations of theft and betrayal of friendship (cf. poems 12, 25 and 33 for the former and 30, 73 and 77 for the latter).

13 Archilochean invectives: Plut. Cato Min. 7; humourless sternness: e.g. Mart. 10.20.21 tum me uel rigidi legant Catones (‘then even strait-laced Cato and his like would read me’), 11.2.1 triste supercilium durique seuera Catonis/frons (‘the stern brow and severe expression of harsh Cato’). See further Buchheit 1961, Scott 1969, Skinner 1982b and esp. Cowan 2015.

14 Perhaps a speaking name (so Ingleheart 2014: 60–2), chosen to suggest the adjective rabidus, ‘crazed’.

15 Cowan 2021: 56–60, noting parallels in Horace’s Epode 5 and Ovid’s Ibis for the iambic victim as scapegoat, as well as the specific echo of Hipponax fr. 128.3–4 ὄληται / βοʋλῆι δημοσίηι (‘that he may die by the will of the people’; ∼ Catull. 108.1: populi arbitrio). For further possible echoes of Hipponax, see Koenen 1977; Vine 2009. It is significant, too, that the scazon or ‘limping iambic’, Catullus’ most often-employed metre after the elegiac couplet and the hendecasyllable, is supposed to have been invented by Hipponax.

16 The most plausible candidates are Publius and Gaius (or Lucius?) Cominius, known mainly from Asconius’ commentary on Cicero’s lost Pro Cornelio (see esp. Neudling 1955: 48; Cowan 2021: 60–2).

17 Syme 1939: 149. For invective as a matter of audience expectation, see Cic. Font. 37, with Craig 2004: 192–9. See also Krenkel 1990 = 2006b: 479–86.

18 For lists of invective topics, see Nisbet 1961: 192–7; Opelt 1965: 128–64 (esp. 148–59); Craig 2004: 189–92; W. J. Tatum 2007: 336. For Catullus and contemporary political/forensic invective, see esp. Koster 1980: 280–93; W. J. Tatum 2007.

19 References and discussion: Richlin 1992: 97–102 (rhetorical invective), 148–51 (Catullus); Corbeill 1996: 114–24 (os impurum), 125–73 (mollitia).

20 See e.g. Parker 1997; C. A. Williams 1999: esp. 15–28, 125–59; Skinner 2014: esp. 257–8, 280–2.

21 See esp. Edwards 1993: 63–97. For a nuanced study of the figure of the cinaedus in Greek and Roman culture, see also Sapsford 2022 (with brief discussion of Catullus 25, 16, and 57 at 147–8, 159–62).

22 On female stereotypes in (Catullus and) Cicero, see esp. Richlin 1992: 97; Dixon 2001: 133–56; Myers 2021: 85–6.

23 Cf. poems 78b, 97, and 98 (with an ironic reflex at 99.7–10, where the speaker is himself the victim of a kind of performative invective on Juventius’ part). For the os impurum and similar double entendres in Cicero, see e.g. Verr. 2.3.23; Dom. 25, 47, 83; Pis. 20; Cael. 78; Phil. 11.5, 7, with Richlin 1992: 99; Krenkel 2006c [1981]: 277–8, 286. See also Corbeill 1996: 104–24. Richlin (1992: 99) notes that Cicero – unlike Catullus – maintains generic propriety by keeping such accusations at the level of innuendo (e.g. Dom. 25: qui sua lingua etiam sororem tuam a te abalienauit, ‘who with his tongue stole even your sister away from you’, of Clodius’ henchman Sex. Cloelius), and generally reserving them for ‘lowly, less powerful victims’.

24 On these poems, see further pp. 15–16 above.

25 In addition to poem 16, see poems 21, 37, and 74, and cf. 15.16 (where caput lacessas may hint at [the threat of] irrumatio: see n. 11 above).

26 When used in the masculine, impudicus – as Craig Williams shows – often refers specifically to adoption of the receptive role in homoerotic sex; but Catullus also plays here on the broader sense, ‘a general lewdness or indecency’ (C. A. Williams 1999: 172–4; quoted phrase at 173).

27 For this reading of poem 16, cf. Selden 1992: 477–80; Batstone 1993: 150–5; Fitzgerald 1995: 49–53; Krostenko 2001: 277–82. Poem 25 is similar: here, the effeminate Thallus is threatened with a beating – a type of punishment usually reserved for slaves, and thus demeaning. At the same time, the verb conscribillent (literally ‘write all over’, ‘scribble on’), used of the marks to be made on Thallus’ body by the whip, hints at the poem’s illocutionary force, in ‘branding’ its victim as both mollis and servile. The punishment, moreover, neatly fits the crime, since one of the objects stolen by Thallus is Catullus’ catagraphi (7): whatever these may have been – perhaps embroidered fabrics, or writing tablets? – the literal meaning, ‘inscribed’, ‘filled with writing’, is picked up in a bilingual pun by the verb conscribillent (so Fitzgerald 1995: 103).

28 E.g. [Virg.], Copa 2; [Virg.], Priapea 19.4; Mart. 5.78.28; Apul. Met. 2.7.3.

29 Svenbro 1993: 187–216.

30 Relevant here is the (explicit or implicit) connection established in other poems between irrumatio and the silencing of the victim (most obvious in poem 74, where Gellius’ uncle will not say a word – necessarily so! – even should Gellius penetrate him orally; see also 21.13, where irrumatus resonates with the counterfactual tacerem, ‘I would keep quiet’, at 9, and poem 80, where the Victor’s exhausted state and Gellius’ whitened lips bear witness to the act of fellatio, which implicitly renders Gellius himself silent).

31 CIL iv.2360 (= Courtney 1995: 92 [no. 79]), iv.4008, iv.8230, xiii.10017.40.

32 On Catullus’ use of obscenity, Lateiner 1977 remains pertinent; see also Skinner 1992; Lorenz 2012. On obscenity in Roman culture more generally, see esp. Richlin 1992: 1–31. Catullus himself draws the reader’s attention to his appropriation of graffito formulae at 37.9–10, where he threatens to defile the front of the salax taberna with sexual graffiti (frontem tabernae sopionibus scribam): like the framing threats of poem 16, this can be understood as a self-fulfilling threat – the poem itself ‘defiles’ or ‘stains’ the house where Lesbia allegedly prostitutes herself.

33 Cf. CIL iv.1427: Saluia felat Antiocum, and numerous instances of the Rufilla felat type, e.g. iv.1651, 2259, 2275.

34 On graveyards as the haunt of prostitutes, see e.g. Mart.3.93.15; Juv. 6.O.16. On tundo as sexual slang, see Adams 1982: 148; on red-haired people as unattractive, see e.g. Ter. Haut. 1061–2. Alternatively, the near-juxtaposition with the verb fellat conceivably hints at a derivation from the Greek rophein (Ionic ruphein), ‘to suck’ (Michalopoulos 1996: 76). See also Nappa 1999; Ingleheart 2014: 62–3. For provincial unsophistication/lack of taste, cf. esp. 39.10–21 and 43.6, where Ameana’s reputation for beauty is attributed to ‘the province’, in contrast to Catullus’ own – implicitly more cultivated – preference for Lesbia.

35 Archil. frs. 42, 44 West; Hipponax frs. 17, 21 West (fellatio), 12, 70 West (the ‘motherfucker’ Bupalos). Cicero: e.g. Pis. 20, with Richlin 1992: 99 (fellatio); Cael. 32, 36; Att. 2.1.5; QFr. 2.3.2 (incest). For the prominence of fellator and cunnilingus as insults in the graffiti, see C. A. Williams 2014: 506–8. It is worth noting, too, that the allegations of adultery and (implicitly) sexual passivity laid against Caesar in Catullus 57 find contemporary counterparts in both literary and sub-literary contexts: Suetonius (Iul. 49), in reporting a persistent rumour that Nicomedes of Bithynia had been Caesar’s lover, cites Calvus fr. 17 Courtney alongside quotations from speeches of Dolabella and the elder Curio and one of the scurrilous soldiers’ songs traditionally chanted during the triumphal procession (Versus triumphales 1 Courtney); in another snatch of song quoted by Suetonius in a later chapter (Iul. 51), the populace are urged to ‘lock up their wives’, now that the ‘bald adulterer’ (moechum caluom) has returned. On the mutual interpenetration of popular and literary invective, see further Ruffell 2003: esp. 44–61.

36 For the identification of Mentula with the Mamurra of poems 29 and 57, see Chapter 1, p. 5, n. 21, above.

37 So Deuling 1999.

38 decoctoris…Formiani, ‘the bankrupt of Formiae’, 41.4 = 43.5; cf. 13.7–8 for Catullus’ lack of cash.

39 See e.g. Democr. fr. 283 D-K; Lucr. 5.1119; Sen. Ep. 9.20.

40 The resonant alliteration of the final line recalls Ennius, Ann. 620 Sk.: machina multa minax minitatur maxima muris, ‘many a huge, menacing siege-engine menaced the walls’. Syndikus (1984–90: iii.142) attractively suggests that Mamurra/Mentula is thereby figured as a ‘bad’ Priapus: whereas the garden god’s erect phallus (regularly troped as a weapon) acts a warning to potential thieves, Mentula’s insatiable appetites present a threat to the property of others. These literary resonances again coexist piquantly with echoes of the unvarnished obscenity of the graffiti: cf. esp. CIL iv.7089: imanis mentula es, ‘you’re a huge prick!’

41 For invectives against Caesar, a series that overlaps with the Mamurra/Mentula cycle, see Chapter 1, pp. 14–15, above.

42 For Catullus as miles gloriosus in poem 10, see Nappa 2001: 93; Bernek 2004; Gaisser 2009: 51–5; Polt 2021: 27–30.

43 For the ambiguous (ironical?) address to Furius and Aurelius as comites, ‘comrades’, in poem 11, see Chapter 2, p. 45 and n. 42, above. For the Furius and Aurelius poems as a carefully structured cycle, see Skinner 1981: 43–7; Beck 1996: esp. 224–35 (the latter, however, posits responsion between this group and the ‘Lesbia-libellus’ of poems 1–11, rather than with the Veranius and Fabullus poems).

44 For philosophical parody in poem 23, see Németh 1971; Macleod 1973: 299–300; Marsilio and Podlesney 2006: 170–1. O’Hearn 2020 argues for a more serious reading, and suggests that Catullus consistently exploits the ambiguity of the adjective beatus, in particular in the adjacent poems 9 and 10, and 22 and 23.

45 The identity of the unnamed iste of 24.5 with Furius is guaranteed by the ‘refrain’ cui neque seruus est neque arca (‘who has neither slave nor strongbox’; 24.5 8, 10), a quotation from 23.1.

46 The poem is perhaps in part a riposte to Furius Bibaculus’ poem (fr. 2 Courtney) on Valerius Cato’s Tusculan villa, which similarly turns on a pun. Cato’s scholarly acumen cannot help him save his property from the auctioneer: for all his ability to solve critical conundrums, this is one he cannot disentangle (soluere [5] and expedire nomen [6] can mean either ‘solve’ [a critical problem] and ‘supply a name [sc. for an unnamed literary character]’, or ‘pay off’ a debt). Cf. Hollis 2007: 139–40 (who, however, thinks it likely that Furius Bibaculus’ poem is the later). For the possibility of identifying Catullus’ Furius with the poet, see Loomis 1969; Courtney 1993: 200; Marsilio and Podlesney 2006; Hollis 2007: 126–7; Hawkins 2011.

47 The phrasing of lines 7–10 hints at a metapoetic dimension, alongside the sexual sense: compare qui in platea modo huc modo illuc/… praetereunt, ‘those who pass by on the street, this way and that’ with Callim. Epigr. 28.1–2, ‘I take no pleasure in the path that carries many people this way and that’ (where both the path and the promiscuous beloved are analogous to the hackneyed ‘cyclical epic’ [1]). Cf. Fitzgerald 1995: 47–9.

48 See Chapter 2, n. 56, above.

49 CIL xi.6721.34.

50 Richlin 1988, 1992: 148–51.

51 Richlin 1988: 359; cf. Richlin 1992: 26–30 for obscenity as figuratively ‘staining’.

52 Feeney 2009.

53 See Chapter 1, pp. 17–18, above. Gellius is perhaps to be identified with the L. Gellius Poplicola (consul 36 bc): see Neudling 1955: 75–7; Wiseman 1974: 119–29. For a political reading of the Gellius cycle, see Skinner 2003: 21–4, 88–91. Cf. Rankin 1976: 120–1; W. J. Tatum 1997: 499–500.

54 The obscure verb glubit clearly has a sexual reference, though the precise meaning in the present context is disputed: the literal sense is ‘skin (an animal)’, ‘strip the bark from’ (cf. Chapter 2, n. 34, above). The phrase ‘incantatory repetition’ is taken from Commager 1965: 98.

55 Cic. Cael. 38.