In this paper I argue that Propertius’ careful cultivation of three as a canonical number throughout Elegies 4.10 is critically undermined at the end of the poem by the distinct lack of a third etymology for the epithet of Jupiter Feretrius. This noteworthy absence of an expected third etymology, backed by the homophonic similarity between Feretrius and the phrase fere tria (“roughly three,” Welch 2005: 163), invites the reader to question the fixity of three that so strongly shapes the poem both structurally and linguistically. Like Virgil, Propertius appears to adhere to the idea that there were exactly three dedicators of the spolia opima, a number determined by Augustus’ authority on the matter after he personally looked into it (Livy 4.20.5–11). In Propertius, however, the accuracy of that number is challenged by the eventual failure of the rule of three, which is destabilized throughout the poem via the figures of Cossus, Crassus, and the poet himself.
I begin by showing that the number three plays a far more fundamental role in 4.10 than has previously been explored. In addition to the three Roman victors, their three foes, and the three stanzas that frame the poem (an introduction, interjection, and conclusion), the intervening narrative is also defined by descending multiples of three: 18 lines are dedicated to Romulus, 12 to Cossus, and 6 to Marcellus. Moreover, the language itself is marked by triplets, not least of which is trina, tribus, and tria. In some cases the poet repeats the same word three times (e.g. nunc, Feretrius, sanguis, and spolium). He also uses sets comprising three synonyms (e.g. exuvium, spolium, praeda; and lecta corona, prima palma, ultima praeda), two instances of a word paired with an example (e.g. a description of Romulus’ arms between two appearances of arma), and even three examples (e.g. Acron, Veii, and Virdomarus’ torque, all spoils). By the end of the poem, where Propertius offers two alternative etymologies for the epithet Feretrius, neither convincing, the reader is wholly conditioned to expect a third, but that expectation is subverted.
This “missing” etymology invites the reader to question what other sets of three might actually be fere tria, most importantly the very group that dictates the number in the first place: the dedicators themselves. First, Propertius challenges Cossus’ status as dux, which had been determined by Augustus himself, in two ways: 1) by overlaying the three historical dedicators with the three Roman generals whose standards had been lost to the Parthians and recovered by Augustus (RG 29), the second of whom (L. Decidius Saxa) had not been supreme commander but a legate; and 2) by refraining from using the term dux of any of the Romans in his historical narrative while applying that crucial term to each enemy leader, thereby implicitly rejecting Augustus’ re-definition. (Indeed, line 46 is a parodic parroting of Augustus’ position as expressed by Livy [4.20.6]). Second, Crassus’ claim to have won spolia opima in 29 B.C.E., which was denied as a direct result of Augustus’ assertion that Cossus had been supreme commander while Crassus had not (Dessau 1906, Flower 2000: 49–52), goes unmentioned but nevertheless casts a pall on the poem in light of the recent timing of the controversy and Propertius’ subtle impugnation of Augustus’ authority (Harrison 1989, Garani 2007, Ingleheart 2007). Finally, Propertius presents himself very much as a triumphator-dedicator in the same vein as the Roman victors whose feats he describes (Pinotti 2004, Welch 2005), becoming with delicious irony the third challenge to the ostensibly canonical number of three. The glaring failure of three in the final lines of the poem prompts the reader to realize that the fixity of this number is merely a literary fiction constructed by one auctor to reflect the historical fiction of another, and the rule of three itself is eventually revealed—if only in hindsight—as a prime example of authorial artifice.
Problems of Triumviral and Augustan Poetics