Hostname: page-component-74656f9884-rdgjk Total loading time: 0 Render date: 2026-11-07T05:49:30.077Z Has data issue: false hasContentIssue false
  • English
  • Français

A STICHOMETRIC ALLUSION TO CATULLUS 64 IN THE CULEX*

Published online by Cambridge University Press:  20 November 2014

Dunstan Lowe
Dunstan Lowe*
Affiliation:
University of Kent
*
d.m.lowe@kent.ac.uk
Rights & Permissions [Opens in a new window]

Extract

In a recent note, I collected instances of ‘stichometric allusion’, the technique in which poets allude, in one or more of their own verses, to source verses with corresponding line numbers. The technique existed in Hellenistic Greek poetry, but seems more prevalent (or at least, detectable) among the Latin poets of the Augustan era, who applied it to Greek and Latin predecessors alike, as well as internally to their own work. New illustrations of each type may be added here to those previously brought to light. Further examples, detected in an unsystematic fashion, no doubt lie dormant in published discussions and commentaries. Callimachus is still the only known Greek practitioner; perhaps his Roman successors considered the technique not merely Hellenistic but Callimachean. Authors of later ages employ the same techniques in equally haphazard fashion, although this does not mean that they had necessarily noticed examples from antiquity.

Information

Type
Shorter Notes
Information
The Classical Quarterly , Volume 64 , Issue 2 , December 2014 , pp. 862 - 865
Copyright
Copyright © The Classical Association 2014 

In a recent note, I collected instances of ‘stichometric allusion’, the technique in which poets allude, in one or more of their own verses, to source verses with corresponding line numbers.Footnote 1 The technique existed in Hellenistic Greek poetry, but seems more prevalent (or at least, detectable) among the Latin poets of the Augustan era, who applied it to Greek and Latin predecessors alike, as well as internally to their own work. New illustrations of each type may be added here to those previously brought to light.Footnote 2 Further examples, detected in an unsystematic fashion, no doubt lie dormant in published discussions and commentaries. Callimachus is still the only known Greek practitioner; perhaps his Roman successors considered the technique not merely Hellenistic but Callimachean. Authors of later ages employ the same techniques in equally haphazard fashion,Footnote 3 although this does not mean that they had necessarily noticed examples from antiquity.

Here I propose a new case in the pseudo-Virgilian Culex, which is more exact than any previously noted because it contains verbal parallels in two consecutive verses. The passage in question is 131–3:

posterius cui Demophoon aeterna reliquit
perfidiam lamentandi mala, perfide multis,
perfide Demophoon et nunc deflende puellis
Later Demophoon left to her the endless agonies
Of lamenting his treachery – Traitor!
Traitor, Demophoon! Even now, you're wept for by many a girl.

The main model for these lines, directly or indirectly, may be the lost verses of Callimachus in which Demophoon was reproached for his faithlessness.Footnote 4 However, another, less commonly remarked modelFootnote 5 is Catullus 64.130–3 (the identical elements appear in the same lines, 132 and 133):

atque haec extremis maestam dixisse querellis,
frigidulos udo singultus ore cientem:
‘sicine me patriis avectam, perfide, ab aris,
perfide, deserto liquisti in litore, Theseu?’
And in her last complaints she spoke these grieving words,
her damp mouth keening, shivering, gulping:
‘Is this how you've carried me from my homeland's altars – traitor!
Traitor! – deserting me on the lonely shore, Theseus?’

The features which these passages share – perfidy, repetition, vocatives – are typical of the so-called ‘neoteric’ style.Footnote 6 This is enough to explain why looking near the 132-line mark yields similar features in four other poems, two Ovidian and two pseudo-Virgilian, which incidentally all involve betrayal and reproach.Footnote 7 But it is very unlikely that the stichometric correspondence in the Culex is accidental. It involves the same repeated word, both times falling in the same metrical sedes (the author of the Culex inserts an additional perfidiam), and spanning the same – apparently arbitrary – line numbers. The authorship and date of the purportedly Virgilian Culex are still debated, though all agree that it postdates Catullus.Footnote 8 It is therefore a valuable case study for this recently identified compositional technique.

Few would deny that Roman readers of the Culex were equipped to detect such an allusion; a better question is whether they would consider it ‘allusion’ at all. The close relationship between these two poems is already well known. Overall, they are similar in length, and each contains a reproachful lament, also of similar length.Footnote 9 Ross has argued that the Culex is a parody of neoteric epic narrative technique, for which Catullus 64 is our prime example.Footnote 10 Furthermore, the Culex contains several incidental borrowings from Catullus 64;Footnote 11 it is no surprise to find Ariadne's epanaleptic cry among these, since it had inspired many imitations, including the two Ovidian examples which gave our author the phrase perfide Demophoon.Footnote 12 Wills traces its evolution, noting that perfide … perfide owes ‘a debt to Catullus’, just as perfide Demophoon owes one to Remedia Amoris 597 and ultimately to Callimachus.Footnote 13

For a reader who is highly sensitive to the relationship between Catullus 64 and the Culex, there are ample opportunities for finding ‘point’, which makes the correspondence an allusion. The Catullan passage is the opening couplet of Ariadne's impassioned reproach to Theseus, and thus very prominently placed, whereas the imitation in the Culex seems tangential to the main narrative, as if the author took pains to include it.Footnote 14 A catalogue of trees leads on to the almond tree, which is obliquely named as ‘Phyllis’, the girl who transformed into one; unexpectedly, as Seelentag points out, the poet then digresses even further from the context by apostrophizing her fiancé rather than Phyllis herself.Footnote 15 Yet Demophoon was the son of Theseus, and Roman poets frequently identified him as the heir of his father's deceitfulness;Footnote 16 thus, as Seelentag again observes, he constitutes an echo of Ariadne's lament.Footnote 17 From a Bloomian point of view, Demophoon – the infamous latecomer (posterius, 131) who failed to retrace his steps – is thus the ideal vehicle for an allusion to Catullus, a gesture of literary filiation tinged with anxiety of influence.Footnote 18

Yet the rich thematic possibilities of the Catullan allusion (upon which a psychologized reading rests) are far from explicit, and the stichometric correspondence can still be treated as a formal or visual phenomenon. Downplaying the allusive sensibilities of the Culex poet and his readership opens up the intriguing possibility of a method of allusion not inspired by thematic relevance. The overlap of only one syllable between the unrelated words deserto and Demophoon (and between the final syllables of aris and multis) is purely auditory or visual, with no real semantic import, implying that the author of the Culex engaged with the source text in a meticulous but localized manner. One possibility is that frequent recitation of the source text led to subconscious echoes in composition. Another is that the Culex author kept Catullus 64 physically present during the act of composition, drawing not only upon its content and stylistic properties but also upon its visual layout.Footnote 19 In both cases we would expect more extensive imitation of the model, though the preceding and following lines show no further similarities, and a smattering of other potential parallels are far less convincing.Footnote 20 As technology provides us with new ways to process and analyse text, further examples of stichometric allusion like the one in the Culex – whose thematic dimension is subtle and indirect – will come to light, and our understanding of ancient reading practices may consequently change.

Footnotes

*

I am grateful to Donncha O'Rourke and William Levitan for stimulating discussions which improved this article, and to CQ's anonymous reader for a reminder that meaning is everywhere.

References

1 Lowe, D., ‘Women scorned: a new stichometric allusion in the Aeneid’, CQ 63 (2013), 1–3CrossRefGoogle Scholar.

2 The following list extends that of Lowe (n. 1); several items are drawn from the valuable discussion by O'Rourke, D., ‘Intertextuality in Roman elegy’, in Gold, B.K. (ed.), A Companion to Roman Love Elegy (Malden, MA, 2012), 390–409CrossRefGoogle Scholar. A compelling Callimachean example is when the rare participle τεθυωμένον (‘sweet-smelling’) appears at Hom. Hymn Aph. 63 and Callim. Hymn Ath. 63, in the same sedes (Hadjittofi, F., ‘Callimachus' sexy Athena: the Hymn to Athena and the Homeric Hymn to Aphrodite’, MD 60 [2008], 9–37Google Scholar, at 32). In Roman comedy, there is a close resemblance between Plaut. Capt. 800 and Ter. Eun. 801 (Fontaine, M., ‘Dynamics of appropriation in Roman comedy: Menander's Kolax in three Roman receptions [Naevius, Plautus, and Terence's Eunuchus]’, in Olson, S.D. [ed.], Ancient Comedy and Reception: Studies in the Classical Tradition of Comedy from Aristophanes to the Twenty-first Century [Berlin, 2013]Google Scholar). Schiesaro, A., ‘Ibis redibis’, MD 67 (2011), 79–150Google Scholar, suggests that Tristia 2.7–8 supplies the name suppressed at Ibis 7–8, ‘Caesar’. Augustan allusions to Greek include Virgil's translation of Theocr. Id. 11.43–4 at Ecl. 9.43 (Keeline, T., ‘Virgil and the Theocritean scholia’, in Thomas, R.F. and Ziolkowski, J.M. [edd.], The Virgil Encyclopedia [Malden, MA, 2013]Google Scholar). Augustan allusions to Latin include Horace's sole use of anhel– to describe Paris on Ida at Carm. 1.15.31, which recalls Catullus' sole use of anhel– to describe Attis on Ida at 63.31 (Putnam, M.C.J., Poetic Interplay: Catullus & Horace [Princeton, NJ, 2006], 154Google Scholar n. 21), and Propertius' use of exsequiae at 4.7.5, which was a Virgilian hapax at Aen. 7.5 (O'Rourke [this note], 401). A more complex relationship exists between Ovid's description of Elysium beginning at Amores 3.9.59, Tibullus' at 1.3.59, and Propertius' at 4.7.59 (Solmsen, F.W., ‘Propertius in his literary relations with Tibullus and Vergil’, Philologus 105 [1961], 273–89CrossRefGoogle Scholar, at 274, 281–9; O'Rourke [this note], 402). Ovid's line 60 (restat, in Elysia valle Tibullus erit) resembles Propertius' line 60 (mulcet ubi Elysias aura beata rosas) and to a lesser extent Tibullus' line 58 (ipsa Venus campos ducet in Elysios). One intratextual example is Propertius' use of the yoking of bulls as a metaphor for desire's burden at both 2.3.47–50 and 2.34.47–50 (O'Rourke [this note], 397). Although Lowe (n. 1) included incipit allusions (e.g. Calp. Sic. 5.1: ‘forte …’, imitating Verg. Ecl. 7.1), these are clearly a special case and not properly ‘stichometric’.

3 In Milton's Paradise Lost, the array of hanging lamps at 1.726–30 recalls those at Aeneid 1.726–7. Intratextually, Dante describes Geryon and Lucifer as ‘ladders through Hell’ at Inferno 17.82–6 and 34.82–6 (Durling, R.M. [tr.], The Divine Comedy of Dante Alighieri: Volume 1: Inferno [Oxford, 1996], 18Google Scholar), and uses the ‘self-rhyme’ of Cristo in Cantos 14 and 19 of Paradiso, both times in lines 104, 106, and 108 (Hart, T.E., ‘The Cristo-rhymes, the Greek cross, and cruciform geometry in Dante's Commedia: “giunture di quadranti in tondo”’, ZRPh 106 [2009], 106–34Google Scholar, at 116).

4 Νυμϕίε Δημοϕόων, ἄδικε ξένε (‘Bridegroom Demophoon, unjust visitor’): Callim. fr. 556 Pfeiffer.

5 Wills, J., Repetition in Latin Poetry: Figures of Allusion (Oxford, 1996), 134Google Scholar, seems alone in noticing a Catullan debt here. The most recent commentary on the Culex (Seelentag, S., Der Pseudovergilische Culex: Hermes Einzelschriften 105 [Stuttgart, 2012]Google Scholar) does not mention the parallel.

6 ‘A lack of fides is the classic accusation against the deserting man’ (Gibson, R.K., Ovid: Ars Amatoria Book 3 [Cambridge, 2003], 286Google Scholar, with citations). There are repetitions of proper names at Cat. 64 19–21 (Thetidis … Thetis … Thetidi), 37 (Pharsaliam … Pharsalia), 285–6 (Tempe, | Tempe); Culex 131 and 133 (Demophoon … Demophoon), 271 and 273 and 275 (Ditis … Ditis … Ditis); Ciris 66 (Crataein … Crataeis), 105–6 (Alcathoi … Alcathoi), 130–1 (Scylla … Scylla), 295–6 (Britomarti … Britomarti), 498 (Iovis … Iovis) and 540 (Nisus … Nisus). The closest parallel to Demophoon … Demophoon is Ciris 295–6, where the repeated name is a four-syllable non-Latin vocative (Britomarti). On the stylistics of repeated nominal forms, see Wills (n. 5), 124–86.

7 The vocative perfide appears at Ov. Ib. 130 (speque tuae mortis, perfide, semper alar), and repeated proper nouns appear at Ov. Her. 5.127–8 (illam de patria Theseus – nisi nomine fallor – | nescio quis Theseus abstulit ante sua) and [Verg.] Ciris 130–1 (nec fuerat, ni Scylla novo correpta furore, | Scylla, patris miseri patriaeque inventa sepulcrum). Epanalepsis of a more common, non-Greek name appears at [Verg.] Dirae 129–30: fabula non vana est, tauro Iove digna vel auro | (Iuppiter, avertas aurem) mea sola puella est.

8 Seelentag (n. 5), 9–17, provides a convenient status quaestionis and accepts the dominant view (introduced by Güntzschel, D., Beiträge zur Datierung des Culex [Munich, 1972]Google Scholar) that the Culex is later than Ovid, but still early imperial.

9 Catullus 64 is 408 lines long, with at least one lacuna, and the Culex is 414 lines long. Ariadne's lament is 70 lines long (Cat. 64.132–201) and the gnat's is 73 lines long (Cul. 210–382).

10 Ross, D.O., ‘The Culex and Moretum as post-Augustan literary parodies’, HSPh 79 (1975), 235–63Google Scholar.

11 According to Drew, D.L., Culex: Sources and Their Bearing on the Problem of Authorship (Oxford, 1925)Google Scholar, esp. 68, 84, Catullus 64 is the major influence upon the Culex – after Lucretius. Although the majority of Drew's parallels are loose, there are several specific verbal echoes. These include gaudia vultu (Cat. 64.34; Cul. 120), omnia … omnia (Cat. 64.186–7; Cul. 348–9), impia … impia (Cat. 64.403–4; Cul. 124–5), late … omnia frangens (Cat. 64.109) with obvia … carpens … late (Cul. 166–7), and myrtus at the line-end, in a catalogue of flowers and linked with Laconia (Cat. 64.89; Cul. 400).

12 The direct link between Catullus' perfide … Theseu (64.133) and the Culex poet's perfide Demophoon (133) was Ovid's intervening perfide Theseu (Fast. 3.473; cf. [Verg.] Aetn. 583). Jacobson, H., ‘The date of the Culex’, Phoenix 58 (2004), 345–7CrossRefGoogle Scholar, suggests that the transformation of Phyllis into the almond tree is a variant which postdates Ovid, making the entire premise for the trope part of its ongoing evolution.

13 Wills (n. 5), 132–8, esp. 134.

14 Alternatively, since the poem's main project is a plangent accumulation of allusions, the imitation is central to the poem. I thank CQ's anonymous reader for this observation.

15 Seelentag (n. 5), ad loc.

16 For Demophoon as the heartbreaker son of the heartbreaker Theseus, see Prop. 2.22a.1–2, 24.43–4; Ov. Her. 2.75–8, Ars Am. 3.35–8, 459–60 (see also Ov. Her. 4.65–6, where Phaedra hurls the same charge at Hippolytus). The apostrophe to Demophoon may also make the Culex an ‘heir’ to Ov. Her. 2, which is addressed to Demophoon in its entirety: both poems have funerary inscriptions as their closing couplets.

17 Seelentag (n. 5), ad loc.

18 Just as the initial ‘p's at Cul. 131–3 echo Ariadne's sobbing patriis … perfide … perfide, the words multis … et nunc signal the allusion as the sequel to many prior Phyllis-laments. They also activate the etymology of Demophoon as ‘Killer of the Many’, just as the Culex itself ‘culls’ the whole sub-genre of elegiac and epyllionic soliloquy. I thank CQ's anonymous reader for these observations.

19 Using what we know of how classical Latin poems were presented in written form, O'Rourke, D., ‘Paratext and intertext in the Propertian poetry book’, in Jansen, L. (ed.), The Roman Paratext: Frame, Text, Readers (Cambridge, 2014)Google Scholar, suggests that the placing of text, e.g. in adjacent columns, formed the basis for intratextual and paratextual allusions.

20 There are superficial visual or aural resemblances between words in the same sedes, especially at line endings: aequor/auctor (12), tuetur/petuntur (53), fluctibus in/talibus in (98), labello/capellae (104), pectus/per artus (138), femina/cacumina (143), tauro/torvo (173). The only exact verbal parallel is the unremarkable quis at the start of lines 145. On two occasions the same word appears in different positions in the line: sensus (189) and mente/mentem (200). These effects seem coincidental.