The frequently problematic and obscure text of Propertius has the ability to drive readers to distraction and editors to despair. Apart from the hundreds of individual textual difficulties which hinder accurate interpretation, one big question about the whole corpus has long troubled scholars: did Propertius write four or five books of poetry? It is not my intention to go through in detail all the evidence which has been thought to be relevant to the solution of this problem. In recent years, Butrica, Günther, Hendry, Heyworth, Lyne and Murgia, for example, have done so in expert fashion, but only to arrive at opposing views². For Günther, Heyworth, Lyne and Murgia, Propertius wrote five books; for Butrica and Hendry, he wrote only four. What I hope to do in this paper is to suggest that there may be some evidence which deserves more attention than it has received thus far, and which should be taken into account in any future discussion of this divisive issue.

From the editio princeps of 1472 up to the beginning of the nineteenth century, editors of Propertius presented his poems in four books, exactly as they are still presented in the editions currently in use today, perhaps with the exception of J. P. Postgate’s Select Elegies of Propertius³. In 1816, however, Lachmann suggested that the book traditionally thought of as the second book in fact contained the remains of two original books, and he went on to present the Propertian corpus in five groups, with Book 3 beginning with the poem usually called 2.10, the traditional Book 3 becoming Book 4, and Book 4 becoming Book 5⁴. For Lachmann, the transmitted text called Book 2 was corrupt and lacunose, and he found strong support for his division at 2.13.25f:

sat mea sat magna est, si tres sint pompa libelli

quos ego Persephonae maxima dona feram.

He argued that the words tres libelli could only make sense if they belonged in an original third book, and so they had to be preceded by a division which had at some stage in the transmission become obscured, a lost division which Lachmann restored between 2.9 and 2.10. He also pointed out other features in support of his argument. Alongside 2.13.25f he put 2.3.4, et turpis de te iam liber alter erit, which, he argued, could only belong in a second book of poems. He also pointed out that at 1362 verses the book traditionally called Book 2 was around 300 verses longer than any other extant Augustan poetry book⁵. Many have accepted Lachmann’s daring thesis, but many have also rejected it, particularly in recent years, which suggests that the question may be worth another look⁶.

It is well known that Augustan poets, following the example of their Hellenistic models in particular, developed an exquisite sense of structure and proportion in the construction of poetry books⁷. This aspect of their art is apparent in both small details and in large-scale design. In his Eclogues, Vergil places the key word iuvenem in the middle of the middle line of the opening poem, and scholars have traced complex structures underpinning the unity of the whole collection⁸. In his Georgics, he names Maecenas four times in a symmetrical pattern, at 1.2, 2.41. 3.41 and 4.2. Similarly, it has been pointed out that he names the River Euphrates at Georgics 1.509 and 4.561, and again at Aeneid 8.726, in each case six lines from the end of the book in question, doing so because Callimachus referred to “the Assyrian river” six lines from the end of his Hymn to Apollo⁹. Again in the Georgics, the four hundredth line of the fourth book has been shown to rework closely the four hundredth line of the fourth book of Homer’s Odyssey¹⁰. Highly elaborate structural patterns have been discovered in the Aeneid¹¹ and the appreciation of structural patterns has been shown to be vital to the understanding of Vergil’s imitative technique¹². Similarly, complex strucures have been traced in Horace’s Odes¹³ and of course in the work of Propertius himself¹⁴. One does not have to believe each and every proposition of this kind, or accept the validity of every aspect of all the complex diagrams drawn by scholars to trace numerous links and structural patterns within collections of poems, to agree that this approach to ancient texts has revealed an important element of the artistic sensibility of Roman poets. As Goold has put it in his Loeb edition of Propertius, “[t]hese are literary curiosities, but no less real for that¹⁵.”

It seems likely, therefore, that at the same time as Propertius took care to structure each individual poem and book of poems, he will also have paid attention to the over-arching structure of his œuvre as a whole. It is generally agreed that the poems we have were written over a fifteen-year period, from around 29 to 15 B.C.¹⁶ Each book was no doubt originally published separately, before they were all gathered together, and the habit of thinking of and referring to the four books we have as a complete collection is deeply ingrained. But there is in fact considerable disagreement about the form in which his poems circulated after their completion. For Williams Books 1-3 were published as a collection and Book 4 added later¹⁷. On the other hand, Butrica is convinced that Book 1 “circulated by its author’s choice as an autonomous work that never formed part of a larger collection”, and that “the remaining three books must have been published together in a three-book collection”¹⁸. Heyworth, on the other hand, has argued that it is “inconceivable that Propertius regarded his first book as an isolated publication, a Monobiblos”, and that “the five books of Propertius circulated together in antiquity”¹⁹. I agree with Heyworth that Propertius wrote five books, but I also agree with Butrica that Book 1 should not be lumped together with those which followed it. Therefore, I would argue, following Birt, Richmond and Skutsch²⁰, that the Propertian corpus was originally composed in two parts, a single book plus a four-book collection²¹, and I suggest that this argument is strengthened by the existence of an obvious structural similarity between the books usually referred to as Propertius 2-4 and two four-book works, Callimachus’ Aetia and Vergil’s Georgics²².

In 1983, following the publication of Parsons’s work on the papyrus fragments known as the ’Lille Callimachus’²³, Thomas demonstrated that the prologue of the third book of Vergil’s Georgics is closely modelled on the Victoria Berenices which opened the third book of the Aetia²⁴. There can be no doubt that the placing of Vergil’s imitation is important²⁵. The opening 48 lines of Georgics 3 famously look forward to the composition of a poem recounting the victories of Octavian (victorisque arma Quirini, 3.27; cf. pugnas / Caesaris, 3.46f), praise of Berenice by a Greek poet giving way to praise of the Roman victor over Cleopatra. Equally, there can be little doubt that Vergil’s Georgics as a whole is deeply influenced by Callimachus’ Aetia²⁶. Following his discussion of Callimachus and Vergil, Thomas goes on to treat their joint influence on two later Roman poets, Propertius and Statius. Concerning the latter, he shows how Silvae 3.1 resembles both the Victoria Berenices which opened Aetia 3, and the prologue to Georgics 3. Concerning the former, he points to Wimmel’s discussion of the similarities between Propertius 3.1 and Georgics 3.1-48. At this point it may be useful to provide evidence of the extent of the links between these two texts²⁷:

Vergil here describes a poetic triumph in which the Muses are led from Greece to the banks of the Mincius, where a temple will be founded and sacrifices and celebrations offered in praise of Octavian’s victories, another way of saying that he will compose an epic poem which will embrace the victory at Actium and the subsequent triple triumph of 29 B.C. In doing so he faces up to the challenge set by Callimachean criticism of post-Homeric epic, implying that he will write a poem worthy of both Homer and Callimachus. Propertius reacts by rejecting war as a theme and affirming his commitment to Philetan and Callimachean poetics, imagining a poetic triumph in elegiac terms and leaving Roman historical themes to others. But he also likens himself to Homer, predicting comparable poetic immortality. The two poets are in fact dealing with identical subject matter, and given the Callimachean background to the Vergilian passage it is striking that Propertius created this extraordinary collection of parallels for a poem of which the first word is Callimachi, a poem which is throughout deeply Callimachean in spirit and which even contains a chariot race (11-14), a poetic triumph to cap Berenice’s Nemean victory²⁹.

On the fact that Propertius opens his third book with such detailed imitation of Georgics 3.1-48, Thomas writes: “I follow Lachmann in the view that Book 2 of Propertius is in fact a conflation of two books, and I agree with Birt (1882) 422-26 that at least in terms of publication the Monobiblos is to be separated from the rest of the collection. If so, and few now have any doubts, then 3.1 is still to be considered the opening poem of the third book. Skutsch (1975) 229-33 has in fact removed any doubts on the matter, but for those who do not believe in a Monobiblos and in the fact that the second book is a conflation, 3.1 will still be 3.1”³⁰. Thomas may have been correct in 1983 to say that “few now have any doubts” about the division of Book 2, but since then important voices in Propertian scholarship have expressed disagreement with Lachmann³¹. Nevertheless, Thomas’s point about the structural parallel between the opening of Georgics 3 and Propertius 3.1 is impregnable. And further evidence in support of Lachmann’s thesis and of Thomas’s assessment of the relevance to it of the prologue of the third Georgic may be provided by the fact that in addition to the close links between Propertius 3.1 and Georgics 3.1-48, Propertius 2.34, the final poem of Book 2, contains close imitation of the closing section of Georgics 2.

In famous lines at Georgics 2.490-4 Vergil refers to the wisdom and happiness to be won from knowing and writing about the mysteries of nature and the countryside and its deities:

felix qui potuit rerum cognoscere causas

atque metus omnis et inexorabile fatum

subiecit pedibus strepitumque Acherontis avari:

fortunatus et ille deos qui novit agrestis

Panaque Silvanumque senem Nymphasque sorores.

At 2.34.71-4 Propertius, writing about Vergil’s poetic career, exclaims:

felix qui vilis pomis mercaris amores!

huic licet ingratae Tityrus ipse canat.

felix intactum Corydon qui temptat Alexin

agricolae domini carpere delicias!

I find it impossible to believe, given the massive similarities between Propertius 3.1 and Georgics 3.1-48, that this parallel can be considered both fortuitous and meaningless³². Further links support the idea that 2.34 is indeed related to the end of the second Georgic. At 2.475-8 Vergil desires to know about natural philosophy:

Me vero primum dulces ante omnia Musae,

quarum sacra fero ingenti percussus amore,

accipiant caelique vias et sidera monstrent,

defectus solis varios lunaeque labores.

Compare Propertius 2.34.27-30 and 51-4, describing the philosophical interests of Lynceus:

quid tua Socraticis tibi nunc sapientia libris

proderit aut rerum dicere possis vias?

aut quid Cretaei tibi prosunt carmina plectri?

nil iuvat in magno vester amore senex.

…

harum nulla solet rationem quaerere mundi,

nec cur fraternis Luna laboret equis,

nec si post Stygias aliquid restabimus undas,

nec si consulto fulmina missa tonent.

For rerum causae and caeli vias read rerum vias and ratio mundi. The amor of Vergil differs from that of Lynceus, but their philosophical interests are exactly the same. Each seeks to understand the workings of nature, represented by the interrelated movements of sun and moon, particularly the phases of the moon (lunae labores and Luna laboret). And just as Vergil in this section of the Georgics, albeit implicitly, is aligning his poem with his predecessors in the genre of didactic poetry, Orpheus, Hesiod, Empedocles, Aratus and Lucretius³³, so Propertius goes on to place himself at the end of a long line of elegists, Varro, Catullus, Calvus and Gallus (2.34.85-94):

haec quoque perfecto ludebat Iasone Varro,

Varro Leucadiae maxima flamma suae;

haec quoque lascivi cantarunt scripta Catulli,

Lesbia quis ipsa notior est Helena;

haec etiam docti confessa est pagina Calvi,

cum caneret miserae funera Quintiliae.

et modo formosa quam multa Lycoride Gallus

mortuus inferna vulnera lavit aqua!

Cynthia quin vivet versu laudata Properti,

hos inter si me ponere Fama volet.

Contextualising these similarities will help to delineate them more clearly, and to bring out further links between these two texts.

Propertius 2.34 is a puzzling and problematic poem in many respects, but some aspects are clear enough³⁴. First of all, it is indeed one poem, not two as suggested by some editors, and it was indeed the final poem of Book 2, the division between Books 2 and 3 being secure, on whatever reckoning of the nature and structure of the Propertian corpus³⁵. It is addressed to one Lynceus, who is accused of having designs on the poet’s puella, and is warned not to pursue her (1-24). Lynceus is presented as a poet who hitherto has been more interested in philosophy than in girls; the poet urges him to follow Philetas and Callimachus instead of more lofty and serious epic and tragic poetry (25-42). Write lighter love poetry, the speaker goes on to advise, following my example as a successful elegist, for it is my pleasure to loiter at banquets, unlike Vergil, who is writing a Homeric epic about Actium, Caesar and Aeneas, having completed his Georgics and Eclogues. In the latter love featured as an important theme and pleased both inexperienced lovers (like Lynceus) and experts (the poet; 43-84). Erotic themes have been the speciality of Varro, Catullus, Calvus and Gallus, and like them I, Propertius, hope to win immortality with my poems about Cynthia (85-94). This is clearly a poem about poetry and genre, about literary history and contemporary politics, with a special focus on the writing of Vergil’s Aeneid³⁶. And these are exactly Vergil’s concerns in the middle of the Georgics.

Thanks to the work of Buchheit and Hardie it has become clear that the prologue to Book 3 of the Georgics cannot be read on its own, but must instead be taken in close connection with the end of Book 2, the two passages together forming a diptych occupying the central section of the whole poem³⁷. The essential unity of the passage from 2.458 to 3.48 is important for our concerns, because I believe that Propertius’ imitation in 3.1 of the opening of Georgics 3 must be seen in the context of his reworking of the close of Georgics 2 in 2.34, and that together these two poems represent Propertius’ reaction to “[t]he most extended self-referential discussion of the poet’s task in Virgil’s œuvre”³⁸. Here, in the middle of the Georgics, Vergil canvasses poetic alternatives to writing poetry about the countryside, i.e. in his own Eclogues and Georgics, expresses his ambition to embrace the highest poetic theme, natural philosophy, and at the same time looks forward to the composition of an epic poem which will include Octavian, his Trojan origins and the development of a new Roman imperium. In doing so he looks back to both Homer and Ennius as possible epic models, but he also has in mind Aristotelian and Callimachean criticism of second-rate post-Homeric epic poetry. In 2.34 and 3.1 Propertius is dealing with exactly the same topics, Homer and the epic genre, Callimachus, contemporary history and Caesar, Vergil’s poetic career to date and the writing of the Aeneid. Furthermore, Propertius explicitly links his two poems by naming Philetas and Callimachus at 2.34.31f and 3.1.1, as if to recreate the double panel at the centre of the Georgics³⁹. And it is quite possible that this complex pattern of allusion to the middle of the four books of the Georgics, against the background of Callimachus’ Aetia⁴⁰, took place in the middle of the four-book collection of love poems, amores / Amores, which Propertius went on to write after the completion of his brilliant monobiblos, his Cynthia.

Given the state of the available evidence, we may never achieve a satisfactory understanding of the relationship between Propertius and his models, and in particular of the intertextual patterns linking his elegies to the Aetia of Callimachus and to Vergil’s Georgics and Aeneid, not to mention the relationship between Propertian poetics and the concept of Callimacheanism, however such a concept may have been defined in the mind of an Augustan poet. Propertius is a love poet who presents the Aetia as a key intertext while knowing that the Aetia was not a poem about love⁴¹. At the same time, he knows that his own elegies could never really be defined simply as love poems and he aspires to Callimachean aetiology as a higher poetic calling, as an alternative to epic⁴². And as a result, his four books of amores / Amores represent the first systematic reaction to and rewriting of Vergil’s epic and indeed of Vergil’s whole poetic career. In recent years, James Butrica has offered a powerful revisionist reading of Propertius’ three-book (i.e. the traditional books 2-4) collection of elegies, making little of Vergil and denying the importance of Callimachus before the fourth book, in which the poet explicitly describes himself as Romanus Callimachus⁴³. But if, from the beginning, a four-book collection could be read against the backdrop of the Aetia and the Callimachean Georgics, Butrica’s impressively argued and invigorating thesis will have to be challenged. DeBrohun has attempted to do so, and has set the question of the Callimachean and Vergilian nature of Book 4 in a new light⁴⁴. But it is only by paying attention to the interrelating themes and structures of the Propertian corpus as a whole that we will be able to grasp the complexities of his meditations on genre and literary history, on the relationship between Latin literature and Roman life, between poets and rulers, poetics and politics, Callimachus and Augustus. And without carefully contextualizing Callimachus and the reception of Callimacheanism in Augustan Rome (e.g. can Propertius’ use of the Aetia be separated from his reading of the Georgics and the Aeneid?)⁴⁵, and without further research into the complex intertextuality of all his poems, taking into account the dialogue between the single-book Cynthia and the four books of amores / Amores, we are unlikely to make much progress in our attempts to define exactly what he meant when he hoped for his native Umbria to swell with pride at his poetic achievement (5[sic].1.63-64):

Ut nostris tumefacta superbiat Umbria libris,

Umbria Romani patria Callimachi!