Hostname: page-component-6766d58669-l4t7p Total loading time: 0 Render date: 2026-05-21T10:30:30.271Z Has data issue: false hasContentIssue false
  • English
  • Français

ONCE MORE UNTO THE BREACH: MESOSTIC AND EPITAPH IN CATULLUS POEM 60

Published online by Cambridge University Press:  27 February 2025

Shane Hawkins
Shane Hawkins*
Affiliation:
Carleton University, Ottawa
*
shanehawkins@cunet.carleton.ca
Rights & Permissions [Opens in a new window]

Abstract

In addition to the acrostic–telestic combination natu ceu aes ‘from birth like bronze’, Catullus poem 60 contains the earliest attested Latin mesostic (mi pia ‘dutiful to me’), which runs down its caesuras. The use of pius anticipates the language of aristocratic obligation that is used of Lesbia in the epigrams and is perhaps also a wordplay on the praenomen of Clodia’s father, Appius. The complex acrostics and the syntax of mi pia, along with the setting of poem 59 (in sepulcretis), suggest that poem 60 can be read as a literary epitaph. Additional closural elements in the poem include an allusion to Callimachus and a sphragis in the form of a play on the author’s name.

Keywords

CatullusLesbiaacrosticmesosticpiusepitaph

Information

Type
Research Article
Information
The Classical Quarterly , Volume 74 , Issue 2 , December 2024 , pp. 583 - 595
Check for updates
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution, and reproduction in any medium, provided the original work is properly cited.
Copyright
© The Author(s), 2025. Published by Cambridge University Press on behalf of The Classical Association

1. A MESOSTIC IN POEM 60

A few years ago, I put forward some arguments for reading poem 60 as the closing poem of Catullus’ polymetrics.Footnote 1 In this view, the contrasts between loyal friends and the betrayals of greedy and depraved politicians and lovers pile up in the final polymetric poems and create a series of related images marked by terminal features, such as the language of weariness, death, leave-taking and words meaning ‘last’. Together these build a sense of closure for the libellus that arrives in poem 60, which bids farewell to the polymetric Lesbia and clears the way for new representations of her in poem 68 and in the epigrams. I argued that the poem specifically calls to mind a passage of Jason’s farewell to Medea in Euripides’ play of that name, and that it invokes an infamous line from the play (ἔρρ’ αἰσχροποιέ, 1346) that had come to associate Medea with the same sexual proclivity implied in the use of the pseudonym Lesbia. The insinuation is perhaps present also in the mention of Scylla, if we are meant to think of Callimachus’ Hecale 90 Hollis/288 Pf., where Scylla is called a fellatrix. I noted that the sentiment of poem 60’s akroteleuton (a combination of acrostic and telestic), which reads natu ceu aes ‘from birth like bronze’, agrees with the depiction of the addressee in the rest of the poem. I further suggested that aes, in a playful and daring touch, is a wink at Clodia Metelli’s cognomen. In sum, it was a reading of the poem as a bitter and final repudiation of Lesbia/Clodia that brings the polymetric poems to a close.

But perhaps nothing is ever truly final in matters of love, for if we look at poem 60 once again, this time with attention to its caesuras, we find that there is yet more to read: mi pia ‘dutiful/devoted to me’.

While acrostics are more familiar and heavily outnumber mid-line phenomena, mesostics are indeed occasionally sighted in the wild.Footnote 2 If we accept this as one example, it would appear to be the earliest mesostic yet identified in Latin literature. Given the fact that the poem already contains an acrostic and a telestic, I suggest that it is unlikely to be mere happenstance.Footnote 3

The mesostic falls at the caesura of each choliambic line, and the elision in verse four creates a striking effect. There are one hundred and twenty-six choliambic lines in Catullus’ libellus. One hundred and nineteen of these, or 95 per cent, have a break after the fifth syllable. Lines with a break after the fourth syllable, but not after the fifth, all have a break after the seventh, where the line caesura falls. As Loomis has shown, all of these lines fill the space of syllables five to seven with a single verb that is central to the line and crucial to the poem.Footnote 4 For example, in his attack on Egnatius’ teeth, Catullus repeats the verb renidere five times, and in line 7 the verb falls in this emphatic place: quodcumque agit, renidet. hunc habet morbum. Unique among the choliambs of Catullus is 60.4, which does not follow this pattern. Semantically, the line falls into two clear halves, ut supplicis uocem and in nouissimo casu, with an elision in the sixth syllable. Loomis, who notes that the most common elisions in the choliambics are at syllables 2, 6 and 8, argues that the purpose of elision at the sixth syllable is ‘to preserve the important break after the fifth syllable’; that is, to preserve the initial colon.Footnote 5 Since the effect of elision in this line is to eliminate the -em of uocem, the suppliant’s ‘voice’ is not just held in contempt (contemptam haberes), but is in fact cut off as the line is read. This suggests that Catullus is playing here with the (false) etymology of (con)temno that connects it to Greek τέμνω ‘cut’.Footnote 6 The effect is striking and yet another indication of the poem’s artful construction. Furthermore, given the very common practice of signposting acrostic play by Hellenistic and later authors, we might consider whether it is plausible that contemno/τέμνω hints at the mesostic that runs down those ‘cuts’, the caesuras.Footnote 7 Possibly, supplicis uox is a further clue to this acrostic/telestic play. That is, we may suppose that supplex is here understood as a formation built on the root of plectere, as if meaning ‘to bend down or under’, or even, as an indication of the suppliant’s posture, ‘embrace’ (so amplect- and complect-).Footnote 8 In other words, perhaps both the acrostic/telestic and the mesostic are subtly hinted at by the poet, directing the readers to ‘turn’ their eyes up and down the lines.

Another noteworthy feature of the poem is the syntax of pius with the dative. The syntax is not problematic in itself, as it is merely an example, or an extension, of the freer use of the dative of reference, which is common in adjective/adverb + personal pronoun phrases and unexceptional in Catullus.Footnote 9 The point of interest is that nearly all of the attestations of pius + dative are epitaphic. There are at least fifteen inscriptional attestations of this phrase dating from the first century b.c.e. to the fourth century c.e. They are widely distributed geographically but appear most importantly in Rome and even at Como near Catullus’ Sirmio. With a single exception, literary attestations only appear in the fourth century or later.Footnote 10 This distribution and the fact that thirteen of the fifteen inscriptions are sepulchral seem significant. With this in mind, we may now consider whether there are further characteristics of poem 60 that might be considered epitaphic and, following that, propose some interpretations of the phrase mi pia.

2. POEM 60 AS EPITAPH

The use of names in acrostics and forms of pius/pia are also common in a context that seems suitable to what may be the final poem of the polymetric poems. In what follows I argue that poem 60 exhibits the characteristics of a literary epitaph and that reading it as an epitaph opens new possibilities for interpreting the poem. As noted above, it seems significant that almost all of the attestations of the pius + dative construction listed above come from epitaphs. The deceased are very often referred to with forms of pius and pietas (piissimus/pientissimus) in funerary inscriptions. According to a statistical analysis by Nielsen, these words are the most commonly used after bene merens and are slightly less frequent than dulcis and carus and their superlatives.Footnote 11 Catullus, too, is aware of this conventional way of speaking of the dead: si ad pii rogum fili lugetur, orba cum flet unicum mater ‘if one mourns at the funeral pile of a dutiful son, when the bereaved mother weeps for her only child’ (39.4–5). pietas was clearly an expectation in Roman conjugal relationships, and both pius and pietas are terms applied to spouses in epitaphs and, in even greater number, to sons or daughters on behalf of their parents.Footnote 12 One is reminded of the marital language Catullus uses of his relationship with Lesbia, but also his surprising use of familial and parental imagery.Footnote 13

Given these considerations, and the fact that epigraphical acrostics are by and large funerary,Footnote 14 it seems possible that the acrostics of poem 60 and the marked diction of the mesostic would suggest to an ancient reader an epitaph—the last word, as it were, and perhaps yet another strategy of closure in the poem.Footnote 15 What I mean to suggest is not that poem 60 takes the form of an actual epitaph, but that it does what Richard Reitzenstein once wrote of an epigram—namely, that it plays freely with the form of an epitaph without actually wanting to be one.Footnote 16 A look at some of the characteristics of the literary epitaph helps one to see how poem 60 can be read in this manner.

A great deal of attention has been given to the process by which epigrammatic memorialization moved, as Martin Dinter succinctly puts it, ‘from its inscriptional roots to become a literary genre and thus an abstract concept’ and which was thereafter open to reception in other genres.Footnote 17 This was a well-advanced process already in the Hellenistic period, and the remarkable flourishing of epigraphic material in the Augustan period, a so-called furor epigraphicus, corresponds with the incorporation of various kinds of epigraphic writing, including sepulchral epigrams and epitaphs, in the works of Augustan poets.Footnote 18 Dinter, again, argues that epigraphic markers, phrases or formulae ‘that bear the connotation of the inscriptional’, are used ‘to construct epitaphic gestures in Latin poetry’, which is to say that Latin poets adopted expressions associated with funerary epigram in its memorializing function while blurring inscriptional and narrative modes in their poetry, in order to echo or gesture toward the epitaphic. This is similar to what I would like to suggest of poem 60: while not itself an epitaph, certain clearly identifiable features of the poem gesture to that form, which was a common, even traditional, technique of closure.

To be sure, poem 60 does not at first glance resemble an epitaph. It seems to lack even the most common features of an epitaph, such as the name, parentage and birthplace of the deceased. It might be pointed out first, however, that a reading of the poem as literary epitaph is not altogether surprising since elsewhere in his corpus we find Catullus working within the tradition of epitaphs to dead pets (poem 3), employing the characteristic language of epitaphs used to address passers-by (poem 4.1), echoing the language of sepulchral inscriptions at the close of the first Lesbia cycle (uiuat ualeatque, 11.17) and elsewhere (si quisquam, 96.1), including the formal conclamatio (the threefold repetition of frater in poem 101) and the final farewell to the departed (aue atque uale, 101.10).Footnote 19 In poem 60 itself the nouissimo casu of line 4 signals the termination of the relationship and perhaps also life itself.Footnote 20 And, though its significance depends on one’s view of the authorial arrangement of the poems, we should also keep in mind that the immediately preceding poem is set in a graveyard (in sepulcretis, 59.2; rapere de rogo cenam, 59.3).

The vogue for literary epitaphs among Hellenistic authors is most clearly evident from the collection of sepulchral epigrams that form Book Seven of the Greek Anthology. Typically, these epigrams identify the name, patronym and birthplace of the deceased, but there developed a noticeable tendency among poets to play with the question of identity in which the ingredients necessary for identification were obscured, engaging the reader in a process of interpretation by creating a puzzle to be solved. Some of these are essentially riddles for the reader, who must piece together clues to derive the identity of the deceased.Footnote 21 Peter Bing writes about this tendency in epitaph as a product of the increasingly literary nature of a form that was sparked by the separation of the epitaph from its monument and by an expressive brevity that was exploited and refined into a self-conscious artistic device. The form called for a kind of poetic playfulness that engaged the reader and encouraged an act of supplementation that Bing dubbed Ergänzungsspiel.Footnote 22 As Bing and others have pointed out, Callimachus appears to have played an important role in pushing the type to its limits.Footnote 23

Among the strategies poets employed to engage the reader are what George Walsh describes as ‘poems that try to make the generic demands of epitaph seem to come from the reader’, wherein the reader questions the monument by reciting the questions the epitaph itself provides.Footnote 24 These typically work like a conversation in which the inscription includes the questions of the reader and the response from the tomb in the voice of the dead. But here also we find examples, such as Callim. Anth. Pal. 7.522 (40 G.–P./15 Pf.), in which, as Walsh puts it, ‘what might have been cast as a dialogue between the reader and the epitaph … has been taken over by the reader, who talks … to himself’.Footnote 25

Bing further observes how literary epigram plays with a conventional practice of sepulchral epigram in omitting the name of the deceased and inscribing it elsewhere, extra metrum, above or below the poem. This is the case, for example, with Callimachus’ epigram on the tomb of his (unnamed) father (Pf. 21 = 29 G.–P. = Anth. Pal. 7.525) and with the epigram on his own tomb (Pf. 35 = 30 G.–P. = Anth. Pal. 7.415), in which he is identified only by his patronymic (Bαττιάδης). The two poems must be read together in order to make the best sense out of either. Bing draws an important conclusion from this fact:

Callimachus can reckon with the reader’s ability to see through his game and realize that the poems supplement each other. For the reader knows about such family grave-plots, and so possesses the information necessary to play the game. One of the pleasures of Ergänzungsspiel, in this instance, is that the reader must translate the context of such real-life family-plots on to the very different landscape of the scroll: the Sitz im Leben becomes the Sitz im Buch. And if we do this, if we imagine the Callimachus family plot set on the papyrus, it follows with virtual certainty that his two epigrams (though separated in the tradition) were juxtaposed on the scroll.Footnote 26

As ‘the deixis of the old inscriptional poems lost its real point of reference’, literary epigrams filled the vacuum by inviting readers to supply those references not merely to people, objects or places but also to other works of literature.Footnote 27

There is a final development to consider with regard to precedents: some time ago Reitzenstein noted the tendency among Hellenistic epigrammatists to mark the conclusions of individual books in their collections with literary epitaphs.Footnote 28 He singled out the sepulchral epigrams of Nossis, Leonidas and Callimachus, and three by Meleager.Footnote 29 This practice was continued by the Roman elegists and, I suggest, by Catullus before them. The adoption of epitaphic language in poetic writing on death seems natural, almost inevitable, and death (real or metaphorical) is a common device of literary closure.Footnote 30 Unsurprisingly, then, we may find such epitaphic language placed at or near the end of books. As Teresa Ramsby points out, however, epigraphic passages in poetry are not simply closural, but appear at other key points—often between major themes—to provide structure in a work. Epitaphs, therefore, can themselves be ‘significant elements of the narrative structure’, both internally and in conclusion.Footnote 31

Now that we have seen how poem 60 might be read as an epitaphic poem, let me return to the words of the mesostic and propose some ideas for its interpretation. To begin, one notes immediately that, unlike the poem’s akroteleuton, the sentiment expressed in the mesostic mi pia appears to be one that is very different from that found in the poem. One could argue that the stark contrast is not so surprising in a poet who followed his odi immediately with et amo; expressions of conflicting emotions are found in other Lesbia poems (such as poems 8, 72 and 92) and seem typically Catullan.Footnote 32 Although there is some ambiguity in the word, since it can be applied to several different kinds of family and social relations, including amicae, meretrices and pueri dilecti, it is worth noting that pius, like certain other words Catullus applies to his affair with Lesbia (amicitia, fides, foedus, pietas), belongs to the terminology of aristocratic obligation (for example 76.2).Footnote 33 This language is almost entirely confined to the epigrammatic poems, and there the subject is without exception Lesbia, so its appearance in poem 60 is worth noting. One can only speculate about its use at this point; perhaps it signals a change in theme as foedus-related language becomes more common beginning with poem 64 and appears thereafter in the epigrammatic poems about Lesbia, where she is consistently portrayed as incapable of such devotion (for example desine … aliquem fieri posse putare pium, 73.1–2).Footnote 34

Another possibility, however, is that pia here has a somewhat nuanced meaning that is found on epitaphs and which we also find in Catullus. According to Nielsen, the use of pius and pietas often has a particular motivation in epitaphs: ‘Being no longer alive, the dead were unable to demonstrate their pietas and thus to fulfil the expectations inherent in the notion.’ As he notes, the dead son in Catullus poem 39, the pius filius, ‘would not be able to support his mother in old age and eventually bury her as he should have done. Instead, the roles are reversed. She is now doing for him what he should have done for her. In a situation where the son receives pietas from his mother, he is still the one called pius’.Footnote 35 If we may assume the same dynamic in poem 60, it resolves any sense of contradiction with the rest of the poem, as the duty that is being commemorated is not Lesbia’s but Catullus’.

There was possibly an additional and altogether different kind of motivation for the use of pia in the mesostic of poem 60. I have already suggested that the aes of the telestic was a play of words on Clodia’s maternal cognomen, Metelli.Footnote 36 Although proper names are the common stock of literary acrostics, they are usually authorial signatures such as in the well-known cases of Nicander, Aratus, Ennius and Virgil. As pointed out earlier, however, there is also some precedent for using the name of a lover.Footnote 37 I suggest, then, that pia is a nod to another family name—the praenomen of Clodia’s father, Appius (Claudius Pulcher). As David Ross points out, pietas as a matter of loyalty to ties of kinship and alliances of marriage is an important aspect of familial connection or propinquitas, ‘and Catullus can imagine the alliance [with Lesbia] … as one resulting from a marriage between families, a bond of pietas linking the father-in-law and sons-in-law’ (dilexi tum te non tantum ut uulgus amicam, | sed pater ut gnatos diligit et generos, 72.3–4).Footnote 38 We even have a precedent for the pun in Varro’s play on the name of Clodia’s older brother, also Appius Claudius Pulcher, and the noun apis ‘bee’ (Rust. 3.16.3).Footnote 39 There is no etymological connection between pius (or apis) and Appius, just as Metellus is not etymologically related to Latin metallum (or μέταλλον), but there need not be for the sake of a homophonic paronomasia.

3. EPITAPH AND CLOSURE IN CATULLUS

The use of epitaphic language as a closural or structural element is also a phenomenon we find in Catullus, and here we may set poem 60 in the larger context of his work as well as of Hellenistic and neoteric poetry. The funereal and epitaphic language in Catullus, already noted above, appears in the farewell to Lesbia that marks the supposed break in their relationship (poem 11). It also appears, albeit in a lighter mood, in poem 14 (perderes, 5; periret, 14), which forms the short transition (poems 12–14) between the Lesbia poems and the uersus proteruiores (poem 14B). Catullus’ address to his dead brother forms the core of the last of the longer poems (68).Footnote 40 Of poem 96, to Calvus on the death of Quintilia, Marylin Skinner writes that ‘Catullus speaks for one last time as a member of an artistic circle and affirms the value of major productions by neoteric colleagues’.Footnote 41 She notes that the juxtaposition of two thematically related poems, in this case 95 on Cinna’s Zmyrna and 96, is also a strategy used to structure epigrams near the conclusion of the libellus. Ettore Paratore, in particular, has argued that Catullus wished to close this poem celebrating the poetry of Calvus with an allusion to his friend’s own poetry (quantum gaudet amore tuo, 96.6; Calvus 28 forsitan hoc etiam gaudeat ipsa cinis ‘perhaps your very ash feels pleasure at this’), which functioned as his own proper σφραγίς.Footnote 42 Surveying the poem’s use of typical neoteric language and its allusive Alexandrian qualities, he characterizes poem 96 as a programmatic manifesto with connections to both the poetry of Calvus and Hellenistic epigram.Footnote 43 Skinner, again, remarks of poem 101, Catullus’ powerful lament over his brother, that ‘the closure it marks by its position in the elegiac libellus—involving the failure of art to bridge the chasm between life and death, the illusory nature of Callimachean poetic immortality, and the end of Catullus’ resolve to sing songs made poignant by his brother’s fate (65.12)—is definitive.’Footnote 44

All these considerations help us understand poem 60. The reader engages with the text and comes to ask a single connected question in the act of reading, much as the readers of certain epitaphs recite the often puzzling questions posed therein. That single question is a complex one, because it compels the reader to question the identity of te in line 1 and of the supplex in line 4. Moreover, the first word of the poem, num, suggests a certain incredulity in a question that ought to receive a negative answer. But if it was neither the mountain lioness nor Scylla who mothered the unnamed addressee, then who did? The poem cannot be read without leaving the reader to ponder both the identity and the parentage of the addressee.

The poem is thus something of a puzzle that, in the practice of literary epigram referred to above, invites the reader to supply the appropriate references in order to fill the vacuum it creates. The missing information, as we saw, was often extra metrum in inscriptions and could be found in acrostics. As the conventions of epitaph moved from burial plots to book scrolls, readers might also find that information in the juxtaposition of poems as they were arranged in individual collections. In keeping with that practice, which we also noted is found in Catullus, we may read 59 and 60 as juxtaposed poems.Footnote 45 As noted, the setting of poem 59 is in sepulcretis and the two poems also share their metrical form. Further, while poem 59 appears to invoke an epigraphic type, the graffito, the first line and a half reads very much like an epitaph, Bononiensis Rufa Rufulum … uxor Meneni, the effect of which is however upended as the first line concludes with fellat.Footnote 46 As Christopher Nappa points out, the ‘collocation Rufa Rufulum raises questions of genealogy—under what circumstances might there be a woman and a man whose names are a simple and a diminutive form respectively’.Footnote 47 Thus Catullus hints at incest and then proceeds to adultery and finally indiscriminate prostitution as ‘a bustuaria moecha, a prostitute of the lowest order, whose patrons find her in cemeteries’.Footnote 48 Nappa also argues that saepe … uidistis in lines 2 and 3 implicates the audience of the poem in a typically Catullan way: ‘For to have seen Rufa ply her trade among the tombstones often, we must have been there often’, that is, presumably, as customers. It is reminiscent too, however, of the epideictic address to the passer-by, common in sepulchral epigram and epitaphic poetry.Footnote 49

If poem 60 marks the conclusion of the libellus as a literary epitaph, it keeps in following with a number of Catullus’ predecessors mentioned above. For one, there was certainly a precedent for Catullus to follow in attempting a literary epitaph in choliambs. The revival of archaic limping iambics in the fourth century b.c.e. subsequently led to a number of poems on moralizing topics in choliambs, which included Theocritus’ epitaph for Hipponax, fittingly, and Aeschrion’s for Philaenis, the reputed female author of a sex manual.Footnote 50 Among the fragments of the choliambic poet Phoenix of Colophon is an epitaph for Ninus.Footnote 51 Callimachus’ Iambs 1–5 and 13 are in scazons, the first of which features the ghost of Hipponax.Footnote 52 Moreover, to Reitzenstein’s list of literary epigrams that marked the conclusions of individual books, we can add Meleager in Anth. Pal. 5.215, a poem that, according to Kathryn Gutzwiller, concludes Meleager’s erotic book and also happens to close with an epitaph: ϵἰ καί μϵ κτϵίναις, λϵίψω φωνὴν προϊέντα γράμματ’ Ἔρωτος ὅρα ξϵῖνϵ μιαιφονίην ‘if you slay me, I will leave letters speaking thus: Look, stranger, on the murderous act of Love.’Footnote 53 The invocation of the language of epitaph in both Meleager and Catullus poem 60 offers additional support to her proposal that allusions to the framing poems of the Garland structure the Catullan corpus.Footnote 54

4. SIGNING OFF

Finally, I would like to point out the possibility of two additional wordplays in this poem that have not been noticed. While neither is essential to the arguments above, both of them add to our impression that poem 60 is quite a recherché poem with strong closural elements. First, that a lion’s offspring might be described as having a mente taetra seems unexceptional; taeter is applied both to animals (including lions, Rhet. Her. 4.51) and men. In montibus Libystinis … mente dura, however, Catullus perhaps alludes to a learned etymology. The association of rugged mountains haunted by horrifying beasts and hard-heartedness is very old and already a Hellenistic trope.Footnote 55 What are the montes Libystini? Perhaps they are to be identified with the Atlas Mountains, traditionally understood as the home of the Titan Atlas. Scholars have pointed out that Virgil refers to Atlas durus (Aen. 4.247) in what has been recognized as a reference to the (folk) etymology of Atlas from Greek τλάω/τλῆναι, as suggested in the name Atlas Telemon ‘Atlas the Enduring’. Although it need not have been the case, perhaps he and Catullus were also aware of the claim, as Strabo reports, that the native name for the mountain was Duris.Footnote 56 At any rate, Catullus’ montibus Libystinis (= Atlas)/durus conceivably involves a wordplay that anticipates Virgil’s Atlas durus.Footnote 57

Another possibility, however, is that the phrase refers to the mountain range (Jebel Akhdar) surrounding Cyrene, the birthplace of Callimachus, and that we are to take this as a somewhat recherché allusion to a poet whose importance for Catullus is well known.Footnote 58 The first mention of this region in Catullus is the Libyan sands of Cyrene, identified as the location of the sacred tomb of the ruler Battus.Footnote 59 It is, of course, in the sepulchral epigram discussed above that Callimachus identifies himself as the descendant of this Battus (Bαττιάδϵω σῆμα). Catullus uses this patronymic in the opening and closing poems of the epigram collection (carmina Battiadae, 65.16 and 116.2). Perhaps, then, we are to read the lion in the hills of Libya alongside the Callimachean Scylla fellatrix, mentioned at the beginning of this paper, as invocations of Callimachus.

The second wordplay is in Scylla latrans, which clearly refers to the Homeric etymology implied by the description of Scylla’s voice as that of a σκύλαξ ‘puppy’ (Σκύλλη … δϵινὸν λϵλακυῖα | φωνὴ μὲν ὅση σκύλακος νϵογιλῆς | γίγνϵται ‘Scylla … barking fiercely, her voice was in fact such as that of a new-born puppy’, Od. 12.85–7). Previously I discussed how this folk etymology and others (for example σκύλλϵιν ‘to shave’ but also, sensu obsceno, ‘to skin’) were used to characterize Scylla and Scylla-like women, and I would add here that the possibility of identifying another wordplay is difficult to resist: a Scylla latrans, or σκύλαξ, by another name in Latin is a catulus or perhaps, as a matter of wordplay, a Catullus.Footnote 60 Clearly Catullus has no problem working his own name into his poetry when he wishes to do so, including in his choliambics (44.3), but textual sites with acrostics often contain other kinds of sphragis material, including allusions to, or puns on, authors’ names. One might object that there is not a particularly close connection between the context of Scylla latrans in the poem and the name Catullus, but it is not clear that such a connection would have been considered necessary. For example, perhaps one of the best-known puns on an author’s name was Aratus’ play on his own name ἄρρητον (that is, Ἄρητος) at Phaen. 2. This pun seems to have been widely known by ancient authors and perhaps even by Catullus.Footnote 61 But, as Bing noted, ‘The pun, if such it is, is entirely detached from the surface meaning of the sentence.’Footnote 62

On the other hand, one may need look no further than the canine imagery that frequently shows up in iambic poetry, where it is both applied to the poet’s enemies and is also assumed as one aspect of the iambic identity.Footnote 63 Consider, as an example of the latter, Callimachus’ characterization of Archilochus (fr. 380 Pf.): ϵἵλκυσϵ δὲ δριμύν τϵ χόλον κυνὸς ὀξύ τϵ κέντρον | σφηκός, ἀπ’ ἀμφοτέρων δ’ ἰὸν ἔχϵι στόματος ‘he drew the bitter bile of the dog and the sharp sting of the wasp, from both he gets the poison of his mouth.’ Such an allusive (cf. Il. 18.322 δριμὺς χόλος αἱρϵῖ) and playful line (iambos from ἰός or even ἰὸς ἀμφοτέρων) must have made an impression. As for Catullus, one thinks of the adeste hendecasyllabi of poem 42 and the moecha turpis who has the poet’s writing tablets. The flagitatio fails even to wring a blush out of the brazen-faced bitch (si non aliud potest, ruborem | ferreo canis exprimamus ore, 16–17). What is left to do except give such a woman the labels she supposedly truly fears: pudica et proba (24)? In lines 8–9 she is attacked with another canine reference and a possible wordplay: mimice … | ridentem catuli ore Gallicāni ‘laughing like a mime actress … with the maul of a Gaulish hound dog’. Skinner calls this ‘a self-referential pun … that, as often in iamb, equates poet and victim’.Footnote 64 The same might be said of Scylla latrans.

References

1 S. Hawkins, ‘Catullus c. 60: Lesbia, Medea, Clodia, Scylla’, AJPh 135 (2014), 559–98.

2 Examples in C. Luz, Technopaignia. Formspiele in der griechischen Dichtung (Leiden and Boston, 2010), 42–3, 74; D. Higgins, Pattern Poetry (Albany, 1987), 5–7, 19–53, 171. See C. Schubert, ‘Ein Zeugnis aus Neros Dichterkreis? Zu den Kryptogrammen der Ilias Latina’, WJA 23 (1999), 137–41 for the possibility of a mesostic at caesura alongside an acrostic in the Ilias Latina. For diagonal ‘intexts’, see W. Levitan, ‘Dancing at the end of the rope: Optatian Porfyry and the field of Roman verse’, TAPhA 115 (1985), 245–69; M. Hanses, ‘The pun and the moon in the sky: Aratus’ ΛEΠΤΗ acrostic’, CQ 64 (2014), 609–14; M. Squire, ‘“How to read a Roman portrait”? Optatian Porfyry, Constantine and the vultus Augusti’, PBSR 84 (2016), 179–240, 359–66. Anth. Lat. 214 Riese is a sixth-century acrostic-mesostic-telestic.

3 These phenomena travel in groups: V. Garulli, ‘Greek acrostic verse inscriptions’, in J. Kwapisz et al. (edd.), The Muse at Play (Berlin, 2013), 246–78, at 267 n. 32 notes that ‘both telestics and mesostics are rare and, when they appear, they are usually combined with acrostics.’ Cf. Luz (n. 2), 1. For some thoughts on intention and accident in acrostic wordplay, see M. Robinson, ‘Looking edgeways. Pursuing acrostics in Ovid and Virgil’, CQ 69 (2019), 290–308 and M. Robinson, ‘Arms and a mouse: approaching acrostics in Ovid and Vergil’, MD 82 (2019), 23–73.

4 J.W. Loomis, Studies in Catullan Verse (Leiden, 1972), 111–12. I omit 22.20, which has a strong break after the second syllable.

5 Loomis (n. 4), 115.

6 On elision, see M. Weiss, Outline of the Historical and Comparative Grammar of Latin (Ann Arbor and New York, 20202), 144–7. Etymology: Schol. Ter. Andr. 492 temnor autem Graecum est, i.e. caedor et reicior. One may find examples of poetic effects with elision or synaloepha in L.P. Wilkinson, Golden Latin Artistry (Cambridge, 1963), 20, 21, 55, 71, 77–9. For wordplay involving elision in Catullus (17.26, 84.8), see E. Vandiver, ‘Sound patterns in Catullus 84’, CJ 85 (1990), 337–40 and L. Morgan at https://llewelynmorgan.com/2018/11/03/when-words-collide/ (accessed June 24, 2024).

7 The metrical term caesura is first attested in Diomedes (late fourth century c.e.), but the terms incisio and incisum (‘cut’), calques on the earlier Greek κόμμα (Demetr. Eloc. 9; Dion. Hal. Comp. 26), were all in use in Catullus’ day (Cic. Orat. 211, 223; Quint. Inst. 9.4.22). Aristides Quintilianus (third century c.e.) refers to the caesura as τομή (1.24).

8 For example, J.J. O’Hara, True Names: Vergil and the Alexandrian Tradition of Etymological Wordplay (Ann Arbor, 19962), 227 notes the use of (genua) amplectens as a gloss on supplex in Verg. Aen. 10.523 et genua amplectens effatur talia supplex. In reality, supplex is formed either with the root of plectere ‘to plait’ or with that of placere ‘to please’ (Festus 206 L, 402 L; Accius, Epig. 290). The former is preferred by both A. Ernout – A. Meillet, Dictionnaire étymologique de la langue latine (Paris, 20014), 669 and M. de Vaan, Etymological Dictionary of Latin and the Other Italic Languages (Leiden, 2008), 472, ‘to bend upwards, beg’.

9 E.g. gratum est mihi (2b.1), non mihi tam fuit maligne (10.18), Quintia formosa est multis; mihi candida, longa, recta est (86.1–2).

10 The TLL records fourteen examples: 1) CIL I2 1259 (first century b.c.e., Rome) Brutia Q(uinti) l(iberta) Rufa pia patrono, dum uixsit, placuit; 2) CIL VI 15580 (first century c.e., Rome) piissimae sibi; 3) CIL XII 870 (first century c.e.?, Arles, France) pius sueis; 4) CIL VI 10229, line 7, ‘Testamentum Dasumii’ (108 c.e., Rome) pientissima mihi; 5) CIL II2 7, 878 (late first/early second century c.e., Mirobriga Turdulorum, modern Capilla, Spain) pius suis omni(bus); 6) CIL V 5268 (second century c.e., Como, Italy) pater naturalis | filio sibi pientissimo; 7) AE 1971 n. 320 (151–230 c.e., Komárom/Szőny, Hungary) D(is) M(anibus) pientissimae parentibus suis Amiciae Digniolae; 8) Inscr. Alger. I 32 (first to second century c.e., Annaba, Algeria) pius suis; 9) AE 1982 n. 508 (mid second century c.e.?, Salpensa, Spain) pia parentibus; 10) CIL XI 764 (second/third century c.e., Bologna) pientissima uiro suo; 11) Iugosl. Šašel 2877 (Nadin, Croatia; Late Principate) [matri filioque] sibi pientissimis; 12) CIL XII 2089 (563 c.e.?, Vienne, France) pauperebus pius; 13) CIL 2091 (566 c.e., Vienne, France) pauperebus pius. The TLL also includes CIL X 3049 (late second/early third century c.e., Pozzuoli) Marco Valerio Laoti … |… filio | dulcissimo ac super ceteros karos | mihi piissimo …, but this is an ambiguous case (S. Tuck, Latin Inscriptions in the Kelsey Museum [Ann Arbor, 2005], 141 reads karos mihi, piisimo). To this list we may add 14) CIL VI 34913 (first century b.c.e. to third century c.e., Via Pinciana, Rome) filis meis mihi piissimis and 15) an inscription on a woman’s gold ring, CIL XIII 10024.94 (third to fourth century c.e., Springiersbach, Germany) uiuas | mi pia Opptata. Literary examples, with the exception of the first of these, are only attested later: 1) Epicedion Drusi (Consolatio ad Liuiam), line 296 (first century?), infelix Druso sed pia turba suo; 2) Gregorius Iliberritanus, De arca 5 (fourth century) deus sibi fecerat pium; 3) Claud. Cons. Stil. 3.216–17 (370–404 c.e.) piam uictis; 4) Chromatus, In Matth. 33.6, 64.6.1 (ob. 407) (deus) omnibus … pius est; 5) Petrus Chrysologus, Serm. 123.10 (fifth century) cui pius, qui sibi sic impius?; 6) Ennodius, Carm. 2.13.2 (fifth/sixth century) fuit … supplicibus … pius; 7) Epistulae Austrasicae 9 (letter from Germain, bishop of Paris; ob. 576) nobis semper piissimae.

11 In a survey of 3,797 epitaphs Nielsen finds that 2,220 epithets are used to characterize the person commemorated and that about ten percent (226) of those epithets are forms of pientissimus/piissimus; H.S. Nielsen, ‘Interpreting epithets in Roman epitaphs’, in B. Rawson and P. Weaver (edd.), The Roman Family in Italy (Canberra and Oxford, 1997), 169–204, building on work by R.P. Saller, ‘Pietas, obligation and authority in the Roman family’, in P. Kneissl and V. Losemann (edd.), Alte Geschichte und Wissenschaftsgeschichte (Darmstadt, 1988), 393–410 and S.G. Harrod, Latin Terms of Endearment and Family Relationships (Princeton, 1909), 10–17.

12 In Nielsen’s study (n. 11) there are 43 examples vs 121 examples.

13 E.g. 58.3 plus quam se atque suos amauit omnes; 72.4 pater ut natos diligit et generos; 109.6 sanctae foedus amicitiae; see P. McGushin, ‘Catullus’ sanctae foedus amicitiae’, CPh 62 (1967), 85–93; D. Konstan, ‘Two kinds of love in Catullus’, CJ 68 (1972–3), 102–10; M.P. Vinson, ‘And baby makes three? Parental imagery in the Lesbia poems of Catullus’, CJ 85 (1989), 47–53; R.H. Simmons, ‘Deconstructing a father’s love: Catullus 72 and 74’, CW 104 (2010), 29–57.

14 E. Courtney, ‘Greek and Latin acrostichs’, Philologus 134 (1990), 3–13, at 6 notes that ‘the vast majority of epigraphical acrostichs are funerary, with the name of the deceased spelt out in the acrostich’. There are some fifty examples collected in F. Bücheler, Carmina Latina epigraphica (Leipzig, 1895), including telestics or akroteleuta (nos. 726, 727, 1615, 1616). Several of these draw attention to the name of the deceased in the acrostic (e.g. no. 109 n]omen si queris, iunge u[ersum exordia). See nos. 108, 109, 273, 511, 570, 651, 676, 696, 748, 797, 1366, 1745, 1814, 1829; A. Amante, ‘Gli acrostici nella poesia sepolcrale latina’, Athenaeum 1 (1913), 288–94. A total of eighty-six acrostic inscriptions are listed in G. Sanders, ‘L’au-delà et les acrostiches des Carmina Latina epigraphica’, Roczniki Humanistyczne 27 (1979), 57–75 = Lapides Memores (Faenza, 1991), 183–205, including (at 195) CE 535 = CIL VIII 15569 (pagan but late, 100–300 c.e., Africa Proconsularis), with an acrostic PIVS.

15 For closural elements in the poem, see Hawkins (n. 1), 581–8.

16 R. Reitzenstein, Epigramm und Skolion (Giessen, 1893; reprint Berlin, 2019), 139, on Nossis in Anth. Pal. 7.718.

17 M. Dinter, ‘Inscriptional intermediality in Latin literature’, in P. Liddel and P. Low (edd.), Inscriptions and their Uses in Greek and Latin Literature (Oxford, 2013), 303–16, at 303. See also R.F. Thomas, ‘“Melodious tears”: sepulchral epigram and generic mobility’, in M.A. Harder et al. (edd.), Genre in Hellenistic Poetry (Groningen, 1988), 205–23.

18 J. Nelis-Clément and D. Nelis, ‘Furor epigraphicus: Augustus, the poets, and the inscriptions’, in P. Liddel and P. Low (edd.), Inscriptions and their Uses in Greek and Latin Literature (Oxford, 2013), 317–47.

19 On poem 3, see N.I. Herescu, ‘Catulle, 3: un écho des nénies dans la littérature’, REL 25 (1947), 74–6; R.F. Thomas, ‘Sparrows, hares, and doves: a Catullan metaphor and its tradition’, Helios 20 (1993), 131–42. On uidetis, hospites, recondita quiete and the ship’s list of achievements in poem 4, see H.J. Mette, ‘Catull carm. 4’, RhM 105 (1962), 153–7. On poem 11, see S. Levin, ‘A dying man’s farewell, parodied by Catullus’, CW 69 (1976), 374–5 and J.C. Yardley, ‘Catullus 11: the end of a friendship’, SO 56 (1981), 67; On poem 96, see M. Citroni, ‘Destinatario e pubblico nella poesia di Catullo: i motivi funerari (carmi 96, 101, 68, 65)’, MD 2 (1979), 43–100, at 51 n. 9 with reference to examples collected in R. Lattimore, Themes in Greek and Latin Epitaphs (Urbana, 1942), 55–65. For poem 101, see H.P. Syndikus, Catull (Darmstadt, 1987), 3.107 and A. Feldherr, ‘Non inter nota sepulcra: Catullus 101 and Roman funerary ritual’, CA 19 (2000), 209–31, at 210: ‘the final phrase “ave atque vale” repeats the farewell to the dead that marked the end of the funeral service and also appears in funerary inscriptions.’

20 For the sense of nouissimus as ‘last’ and nouissimus casus as the finality of death, see Hawkins (n. 1), 563–4.

21 Thomas (n. 17), 213 remarks upon ‘the highly conscious disguising of typologies that characterizes the Hellenistic examples’. On the ‘puzzle of identity’ and the riddling nature of some epitaphs, see G.B. Walsh, ‘Callimachean passages: the rhetoric of epitaph in epigram’, Arethusa 24 (1991), 77–105 and R. Scodel, ‘Two epigrammatic pairs: Callimachus’ epitaphs, Plato’s apples’, Hermes 131 (2003), 257–68.

22 P. Bing, ‘Ergänzungsspiel in the epigrams of Callimachus’, A&A 41 (1995), 115–31. See also R. Hunter, ‘Callimachus and Heraclitus’, MD 28 (1992), 113–23, at 114 on Callimachus’ epitaph for Heraclitus and the genre as ‘a provocation to speculation’.

23 And not only in the epigrams: note the conceit of the speaking tomb (Aet. fr. 64 Pf. and Ia. 11) and the address from the dead Hipponax in Ia. 1. See Scodel (n. 21), 262 for Callimachus playing with the conventions of inscription; Walsh (n. 21), 97 on Callimachus having ‘turned the genre inside out’; Hunter (n. 22), 123: ‘Callimachus moves completely away from’ the traditional forms of funerary poem … ‘they remain, however, hovering over his poem, advertising its difference. In Callimachus the gradual shift from “real” epitaph to “literary” epigram has been taken a further, and decisive, stage’ away from its monumental origins to the realm of memory and poetic tradition.

24 Walsh (n. 21), 88. Examples include G. Kaibel, Epigrammata Graeca ex lapidibus conlecta (Berlin, 1878), 80–1 (no. 218, Paros, second century c.e.); W. Peek, Griechische Grabgedichte (Berlin, 1960; reprint Berlin, 2021), nos. 425, 426, 428, 430–4 = W. Peek, Griechische Vers-Inscriften: I. Grab-Epigramme (Berlin, 1957), nos. 1833, 1842, 1860, 1866, 1870, 1871, 1881, 1883. There is some analysis of this type of epitaph in W. Rasche, De Anthologiae Graecae epigrammatis quae colloquii formam habent (Munster, 1910), 25–34, 39–42, who includes several poems from Anthologia Palatina Book 7.

25 Walsh (n. 21), 97. For further discussion of this epitaph of Callimachus, see M. Hosty, ‘“But who art thou?”: Callimachus and the unsatisfactory epitaph’, GRBS 59 (2019), 202–14.

26 Bing (n. 22), 128; ‘some kind of sphragis at the end of Callimachus’ epigram book’, according to R. Höschele, ‘The traveling reader: journeys through ancient epigram books’, TAPhA 137 (2007), 333–69.

27 Bing (n. 22), 131, discussing Callimachus’ epigram 22 Pf. = 36 G.–P. = Anth. Pal. 7.518 and Theoc. Epigr. l.

28 Reitzenstein (n. 16), 139 n. 2; so also U. von Wilamowitz-Moellendorff, Sappho und Simonides (Berlin, 1913), 298–9 and U. von Wilamowitz-Moellendorff, Hellenistische Dichtung in der Zeit des Kallimachos, 2 vols. (Berlin, 1924), 1.135.

29 On Nossis 11 G.–P. = Anth. Pal. 7.718, see also A.S.F. Gow and D.L. Page, The Greek Anthology, 2 vols. (Cambridge, 1965), 2.442, 2.606 and K.J. Gutzwiller, Poetic Garlands (Berkeley, 1998), 85–8; Leonidas 93 G.–P. = Anth. Pal. 7.715; Callim. 30 G.–P. = Anth. Pal. 7.415; and Meleager 2, 3, 4 G.–P. = Anth. Pal. 7.417, 7.418, 7.419. So also M. Gabathuler, Hellenistische Epigramme auf Dichter (Leipzig, 1937), 48–9, 56, who adds Meleager in Anth. Pal. 7.421 = 5 G.–P. (a riddling epitaph) and points to similar closures among Latin poets.

30 J.C. Yardley, ‘Roman elegy and funerary epigram’, EMC 40 (1996), 267–73, at 273: ‘one should not be surprised to find the topoi and the language of epitaph in those places in elegy where the poets are discussing death, as they often do. But occasionally we also find expressions or words typical of funerary epigram in contexts which are not in themselves sepulchral, but where those expressions or words are still meant to be recognized as funerary, to have a sepulchral ring.’

31 T.R. Ramsby, Textual Permanence: Roman Elegists and the Epigraphic Tradition (London, 2007), 19.

32 The anonymous reader suggests that ‘one could argue that there might be further textual play here, as MI PIA is also an anagram of IMPIA, so the mesostic could perhaps capture the conflicted feelings Catullus has towards Lesbia.’

33 While this is widely agreed upon, the precise significance of this language is disputed. There are good discussions of opposing views in M. Vinson, ‘Party politics and the language of love in the Lesbia poems of Catullus’, in C. Deroux (ed.), Studies in Latin Literature and Roman History 6 (Brussels, 1992), 163–80; W. Fitzgerald, Catullan Provocations (Berkeley, 1995), 117–20.

34 foedus appears twice in poem 64; fides seven times in poem 64 and once in poem 67; pietas once each in poems 64 and 67; pius once in poem 68b.

35 Nielsen (n. 11), 196.

36 Hawkins (n. 1), 570–1, where I might have mentioned a precedent for wordplay with the name. Naevius, making a different pun, plays on metelli ‘wage-earners’ in the verse fato Metelli [or metelli] Romae fiunt consules, which was still well known in Catullus’ day (Cic. Verr. 1.29 and Ps.-Asc. ad loc.). See V. Ferraro, ‘Mai chiamare Metello un metello’, in L. Munzi (ed.), Forme della parodia, parodia delle forme nel mondo greco e latino (Naples, 1998), 73–84, after H.B. Mattingly, ‘Naevius and the Metelli’, Historia 9 (1960), 414–39; E. Peruzzi, ‘Dabunt malum Metelli’, PP 52 (1997), 105–20; with a different emphasis, D.N. Sánchez Vendramini, ‘Naevius’ Fehde mit Q. Caecilius Metellus’, Mnemosyne 62 (2009), 471–6.

37 Hawkins (n. 1), 571 n. 31.

38 D.O. Ross, Style and Tradition in Catullus (Cambridge, MA, 1969), 90.

39 The anonymous reader correctly points out that praenomina do not have the same onomastic specificity as cognomina, which are typically the targets of wordplay. We do find puns on praenomina, however, such as the Plautine pun on Quintus (M. Fontaine, ‘Reconsidering some Plautine elements in Plautus (Amphitryo 302–7, Captivi 80–4)’, CJ 111 [2016], 417–27), the excellent nickname Biberius Caldius Mero for Tiberius Claudius Nero at Suet. Tib. 42.1, and a pun on Tullus/Tertullus (HA 4.29). For various kinds of homophonic wordplay in Catullus, see J. Ingleheart, ‘Play on the proper names of individuals in the Catullan corpus: wordplay, the iambic tradition, and the Late Republican culture of public abuse’, JRS 104 (2014), 51–72, at 62–3.

40 Catullus’ disillusionment with Lesbia in poem 76 is also expressed in extreme terms. Cf. J.G.F. Powell, ‘Two notes on Catullus’, CQ 40 (1990), 199–206, at 199 n. 2: ‘It is hardly possible to escape the implication, in the appeal to the gods at the end of the poem, that Catullus is actually presenting himself as on the point of death and in the grip of a disease—even if that disease turns out to be the purely nervous malaise caused by disappointed love.’

41 M.B. Skinner, Catullus in Verona (Columbus, OH, 2003), 114.

42 E. Paratore, ‘Osservazioni sui rapporti fra Catullo e gli epigrammisti dell’Antologia’, in Miscellanea di studi Alessandrini in memoria di Augusto Rostagni (Turin, 1963), 562–87, at 583.

43 Especially Anth. Pal. 7.476, Meleager’s epigram on the death of his beloved Heliodora; Anth. Pal. 7.23, Antipater of Sidon’s sepulchral epigram for Anacreon; Anth. Pal. 7.643, Crinagoras on the death of Evander’s daughter; the repetition of forms of cinis and gaudeo at Calvus 27, 28 and Catullus 68b.95, 90, 98; 96.1, 6; 101.4, and σποδιή, τέρψις and ϵὐφροσύνα at Anth. Pal. 7.23–4. On the theme, see B. Lier, ‘Topica carminum sepulchralium Latinorum’, Philologus 63 (1904), 54–64, at 54–5.

44 Skinner (n. 41), 128.

45 Additional reasons are given in P. Claes, Concatenatio Catulliana (Amsterdam, 2002), 86–7, supported by N. Holzberg, ‘Catullus as epigrammatist’, in C. Henriksén (ed.), A Companion to Ancient Epigram (Hoboken, 2019), 441–58, at 454.

46 The first line has often been compared to Pompeian graffiti such as CIL 4.1427, 4.2402 and especially 4.2491 Rufa ita uale quare bene felas. A very different approach is taken by S.J. Heyworth, ‘Catullian iambics, Catullian iambi’, in A. Cavarzere et al. (edd.), Iambic Ideas (Lanham, MD, 2001), 117–40, at 122, who claims the ‘juxtaposition of two choliambic poems in 59 and 60’ is an oddity ‘in defiance of the regular search for variety’ in Catullus, a situation that some would explain as the accretion of patchy fragments at the end of the polymetric poems under the hand of an editor; similar objections are found in G.O. Hutchinson, Talking Books (Oxford, 2008), 111. As Holzberg (n. 45), 454 points out, however, not only does positioning two pieces with funeral motifs at the end of a book make ‘perfect sense’, one can point to other examples of books ending in two pieces of the same metre (Hor. Carm. 2.19 and 2.20, 4.14 and 4.15, Epod. 14 and 15 as ‘false ending’ before 17 and 18; [Verg.] Catalepton 14 and 15). D. Fowler, ‘First thoughts on closure: problems and prospects’, MD 122 (1989), 75–122, at 82–3 (reprinted in D. Fowler, Roman Constructions. Readings in Postmodern Latin [Oxford, 2000], 239–83): ‘The assertion of regularity after variation … which is the essential principle of metrical closure in verse or prose (the fixed ends of the hexameter or trimeter, the prose clausula) is also a principal operative at higher textual levels’.

47 C. Nappa, ‘Catullus 59: Rufa among the graves’, CPh 94 (1999), 329–35, at 331. See also Ingleheart (n. 39), 62–3.

48 Nappa (n. 47), 322: ‘This catalogue seems, in fact, calculated to implicate Rufa in nearly every possible form of sexual vice’; cf. C.L. Neudling, A Prosopography of Catullus (Oxford, 1955), 156–7.

49 E.g. CE 1173 si me uidisses aut si mea funera nosses, fudisses lacrimas, hospes, in ossa me; G. Vestrheim, ‘Voice in sepulchral epigrams: some remarks on the use of first and second person in sepulchral epigrams, and a comparison with lyric poetry’, in M. Baumbach et al. (edd.), Archaic and Classical Greek Epigram (Cambridge, 2010), 61–78. See also Thomas (n. 17), 213–14, 218 on the act of ‘seeing’ in epitaphs.

50 Theocritus on Hipponax: Anth. Pal. 13.3 = G.–P. 1.532; Aeschrion on Philaenis: Anth. Pal. 7.345 = G.–P. 1.3–5. On the invective character of this poem, see M. Kanellou, ‘The curious case of Philaenis in AP 7.345 = Ath. Deipn. 8.335B: an early fictitious mock epitaph by Aeschrion’, in C. Cusset et al. (edd.), Féminités hellénistiques: voix, genre, représentations (Leuven, 2020), 269–94.

51 On the ‘Epitaph of Lynceus’, see now L. Bettarini, ‘Un frammento della coliambografia ellenistica: il cosiddetto “epitafio di Linceo” (PStrasb WG 307v = Phoenix (?) fr. 4 Knox [Pack3 1349])’, QUCC 134 (2023), 145–62. Paus. 1.9.7 records that the ‘iambic poet’ Phoenix wrote a lament for Colophon.

52 After Catullus we find a choliambic epitaph by Apollonides (Anth. Pal. 7.693, first century c.e.). Martial 5.37 is an epitaph in scazons written to his young slave girl, Erotion. P. Watson, ‘Erotion: puella delicata?’, CQ 42 (1992), 253–68 convincingly shows how this poem combines elements of the erotic with invective. The five choliambic poems in the Anthologia Latina (nos. 212–16, Pars 2. Carmina Latina epigraphica Bücheler) all date to the second or third century c.e.

53 K. Gutzwiller, ‘Catullus and the Garland of Meleager’, in I. Du Quesnay and T. Woodman (edd.), Catullus: Poems, Books, Readers (Cambridge, 2012), 79–111, at 87.

54 For poem 60 as a programmatic poem with links to Meleager, see Hawkins (n. 1), 581–8. The embedded literary epitaph was also an important technique of closure among the later Latin elegists, whose individual books sometimes concluded with epitaphs. See Thomas (n. 17), 216–21 on sepulchral epitaph as a closing device in Virgil, Horace and Ovid. On Horace, see T. Woodman, ‘Exegi monumentum: Horace, Odes 3.30’, in T. Woodman and D. West (edd.), Quality and Pleasure in Latin Poetry (Cambridge, 1974), 115–28. Prop. 1.21, 1.22 and 4.11 as well as the elaborate closural sequence at the end of Book 1 of Martial, which includes two cepotaphia or poems from a garden tomb (114, 116), are all important examples; C. Henriksén, ‘Martial’s modes of mourning. Sepulchral epitaphs in the Epigrams’, in R.R. Nauta et al. (edd.), Flavian Poetry (Leiden and Boston, 2006), 349–67; L.B.T. Houghton, ‘Epitome and eternity: some epitaphs and votive inscriptions in the Latin love elegists’, in P. Liddel and P. Low (edd.), Inscriptions and their Uses in Greek and Latin Literature (Oxford, 2013), 349–64.

55 For examples of the theme, see the bibliography in Hawkins (n. 1), 561–2, to which add the large collection of material in A.S. Pease, Vergili Maronis Aeneidos liber quartus (Cambridge, MA, 1935), 315–18.

56 Strabo 17.3 ὄρος ἐστίν, ὅπϵρ οἱ μὲν Ἕλληνϵς Ἄτλαντα καλοῦσιν, οἱ βάρβαροι δὲ Δύριν. See R.W. Cruttwell, ‘Virgil, Aeneid, iv. 247: “Atlantis Duri”’, CR 59 (1945), 11; G. Doig, ‘Vergil’s art and the Greek language’, CJ 64 (1968), 1–6; O’Hara (n. 8), 55–6, 154.

57 For the Atlas wordplay, see P.R. Hardie, ‘Atlas and axis’, CQ 33 (1983), 220–8, at 223 n. 17, suggesting that Aratus (Phaen. 22) punned on Atlas as the world axis who ἔχϵι ἀτάλαντον γαῖαν ‘holds the earth in equipoise’. For another mountain wordplay in Catullus, compare the bilingual figura etymologica at 11.9 altas Alpes; Serv. ad Aen. 10.13 notes that Alpes was thought a Gaulish word meaning ‘mountain peaks’ (omnes altitudines montium licet a Gallis alpes uocentur).

58 I owe this idea to an anonymous author whose work has yet to be published.

59 Poem 7.3–6. Fierce lions giving birth in the hills is an image that also appears in Callimachus’ Hymn to Delos 120, Hymn to Demeter 50–3 (cf. Il. 17.133–6), Hymn to Zeus 10–14. Catullus also employs the motif of encountering a lion at 45.6–7 (cf. the ‘green-eyed’ lion at Il. 20.172).

60 Hawkins (n. 1), 572–81.There is not, of course, an actual etymological connection between catulus (or catellus) and Catullus. Another example of Catullus playing on his own name may be poem 56.3 Cato Catullum, as if Catullus were a diminutive of Cato (probably M. Porcius Cato [Uticensis]); see Fitzgerald (n. 33), 78 and R. Cowan, ‘On not being Archilochus properly: Cato, Catullus and the idea of iambos’, MD 74 (2015), 9–52, at 43. I suggest that in this regard Catullus may have been influenced by the model for his poem, Archil. 168W, where the addressee, Ἐρασμονίδης Xαρίλαος ‘Folkappeal Darlingson’, has a name that is practically an anagram of Archilochus (χαρι-λαος/αρχι-λοχος). Two lines later Xαρίλαος is called πολὺ φίλταθ’ ἑταίρων, which is essentially a gloss on his own name.

61 For arguments that it was known to Callimachus (Anth. Pal. 9.507 = 27 Pf. = 56 G.–P. 3–4), Leonidas (Anth. Pal. 9.25 = 101 G.–P.), Ptolemy Philadelphus (Suppl. Hell. 712), Apollonius of Rhodes and Virgil (G. 1.429–33), see P. Bing, ‘A pun on Aratus’ name in verse 2 of the Phainomena?’, HSPh 93 (1990), 281–5 and P. Bing, ‘Aratus and his audiences’, MD 31 (1993), 99–109; S. Stewart, ‘“Apollo of the shore”: Apollonius of Rhodes and the acrostic phenomenon’, CQ 60 (2010), 401–5; E.L. Brown, Numeri Vergiliani (Brussels, 1963), 102–4; T. Somerville, ‘Note on a reversed acrostic in Vergil Georgics 1.429–33’, CPh 105 (2010), 202–9, at 205; J.T. Katz, ‘Vergil translates Aratus: Phaenomena 1–2 and Georgics 1.1–2’, MD 60 (2008), 105–23. L. Kronenberg, ‘Seeing the light, part II: the reception of Aratus’ LEPTĒ acrostic in Greek and Latin literature’, Dictynna 15 (2018) (online), sections 46–51 offers a complex argument that it was known to Catullus, among others.

62 Bing (n. 61 [1993]), 105.

63 E.g. Semon. 7.12W; Solon 36.27W; Archil. 93aW, 196a.24, 41W; Hipponax 3aW, 66W, 165bW, 115W; Callim. Aet. 74.4–5, Ia. 1.82–3. See further J.N. Hawkins, ‘The barking cure: Horace’s “anatomy of rage” in Epodes 1, 6 and 16’, AJPh 135 (2014), 57–85 and J.N. Hawkins, ‘Anger, bile, and the poet’s body in the Archilochean tradition’, in L. Swift and C. Carey (edd.), Iambus and Elegy (Oxford, 2016), 310–39.

64 M. Skinner, ‘A review of scholarship on Catullus 1985–2015’, Lustrum 57 (2015), 91–360, at 257. Skinner is following Ingleheart (n. 39) and Heyworth (n. 46).