2

Contextualising Catullus: a re-examination of 66.1–14

I. M. Le M. Du Quesnay

Introduction

The detailed and scholarly commentaries of Nino Marinone, Giulio Massimilla and Annette Harder have made an immense contribution to our understanding of Callimachus’ Coma Berenices.¹ It is possible now to appreciate more clearly than previously the relationship of Catullus’ translation (Poem 66) to its source text. The main part of Callimachus’ text has been recovered from a papyrus that preserved the first version he wrote, which was an occasional poem composed in the summer of 245 BC and circulated independently. The text translated by Catullus, however, was a revised and expanded version of the original, and it was incorporated into the Aetia as the final poem in the final book.²

By Roman standards, Catullus’ poem is a remarkably sustained and accurate rendering of the source text. There is nothing incontrovertible, in the way of evidence or arguments, to suggest that Catullus has altered the paraphrasable sense of Callimachus’ text by inserting his own material or by incorporating material from elsewhere with the intention of creating something which is seeking to say something significantly different from the source text. When reading those parts of Catullus’ poem for which the source text does not survive, caution is necessary. Where direct comparison is possible, the differences between Catullus and Callimachus mostly affect tone and style rather than content. When the source text survives, it is easy to spot and reflect upon the stylistic embellishments, the occasional expansion of a point or the significance of an omission of word or phrase. But we cannot simply assume that we can reconstruct the source text from Catullus’ translation.

The prologue with which Catullus’ poem opens constitutes the introduction for everything that follows: namely, a mix of emotional scenes of separation and departure; the fantastical and wondrous account of the catasterism of Berenice’s Lock; and the subtle and serious representation of the royal pair, Berenice II and Ptolemy III. Ptolemy came to the throne at the end of January 246 BC. His campaign against Syria – the so-called Third Syrian War – began in the summer of the first year of his reign and he returned to Alexandria in the summer of 245 BC.³ (This was the occasion of the first version of Callimachus’ poem, which provides a retrospective (and selective) account of the events of the previous year.) Ptolemy is succinctly portrayed as a decisive and successful commander who swiftly adds to the extent of his kingdom. But Berenice is the centre of attention: her piety towards the gods wins their support and favour for her husband; her personal courage; and her passionate devotion to her new husband. The new royals are represented as worthy successors to their adoptive parents, Ptolemy II and Arsinoe II, the Theoi Philadelphoi, and to their adoptive grandparents, Ptolemy I and Berenice I, the Theoi Soteres. At some point during 244/243 BC Ptolemy III and Berenice II were included into the dynastic cult as the Theoi Euergetai. Their inclusion during their lifetime and so early in their reign seems to have been a significant innovation. The poem may have been revised for inclusion in the Aetia in anticipation or celebration of this.

In this chapter I shall be concerned principally with the prologue (lines 1–14) of Catullus’ poem:⁴

Omnia qui magni dispexit lumina mundi,

qui stellarum ortus comperit atque obitus,

flammeus ut rapidi solis nitor obscuretur,

ut cedant certis sidera temporibus,

ut Triuiam furtim sub Latmia saxa relegans

dulcis Amor gyro deuocet aërio,

idem me ille Conon caelesti in limine uidit

e Bereniceo uertice caesariem

fulgentem clare, quam cunctis illa deorum

leuia protendens brachia pollicita est,

qua rex tempestate nouo auectus hymenaeo

uastatum finis ierat Assyrios,

dulcia nocturnae portans uestigia rixae,

quam de uirgineis gesserat exuuiis.⁵

He who has (learned to) distinguish all the luminaries in the vast firmament, who established the rising and settings of the stars, how the fiery gleam of the devouring sun is obscured, how the constellations move at constant and regular times, how sweet Love summons from her heavenly circuit She of the Three Ways, removing her clandestinely to a place concealed beneath the Latmian Rocks, this very same man, the famous Conon, saw me on the threshold of heaven, [me] the Lock [cut] from the head of Berenice,⁶ [now] shining clearly, which she promised to all the gods while stretching out her smooth arms [in prayer], at the time when the king, sailing away from his recent nuptials, went to lay waste the Assyrian lands, taking with him the traces of the nocturnal fight which he had waged for the maiden’s spoils.

I Conon

Catullus opens with a carefully balanced sentence of seven couplets. The first three, which teasingly misdirect the reader, are devoted to quasi-hymnic praise of the earlier discoveries of the astronomer who has now ‘discovered’ a new constellation. The pivotal central couplet identifies the speaker as the new constellation (the Coma) and the astronomer as Conon, who was a contemporary of Callimachus and a distinguished mathematician who enjoyed the respect of Archimedes.⁷ His standing gave credibility and authority to his discovery.⁸

Only the first line of Callimachus’ text survives:

πάντα τὸν ἐν γραμμαῖσιν ἰδὼν ὅρον ᾗ τε φέρονται

Having come to know on the basis of diagrammatic representations the sky in its entirety and (?worked out?) how (the stars?) move …. ⁹

The precise sense of Callimachus’ fragmentary Greek is disputed. Harder prefers to take ἐν γραμμαῖσιν as attributive with τὸν … ὅρον and translates: ‘Observing the whole sky as divided by lines and the movements …’.¹⁰ Massimilla takes the prepositional phrase rather as adverbial with ἰδὼν and, following Pfeiffer, understands the point to lie in the contrast between ἐν γραμμαῖσιν ἰδὼν and ἔβλεψεν ἐν ἠέρι (7): Conon had studied the previously identified stars in diagrammatic representations, probably on an astronomical globe, and had learned to recognise them all in the night sky, with the result that he was able to identify the ‘new’ (i.e. previously unidentified) constellation that he called the Lock of Berenice. If this view is right, then Catullus has captured well the substance of Callimachus’ text even though he has noticeably omitted the technicality of ἐν γραμμαῖσιν.

Catullus’ editors accept dispexit, which is the conjecture of Calphurnius. The transmitted reading is despexit, which has been defended on the grounds that Conon ‘gazed down’ on the star charts or at the astronomical globe. But it is difficult to see how the prefix alone can have been expected to bear the weight of meaning carried by Callimachus’ ἐν γραμμαῖσιν, even if Catullus expected his readers to have in mind the wording of the Greek text.¹¹ While omnia deceptively, but only briefly, appears to match πάντα, omnia … magni … lumina mundi taken together evoke the night sky teeming with stars and dispexit puts the needed emphasis on observation: where others see only countless spots of light, the astronomer can identify the named stars and the constellations. In translating τὸν … ὅρον by lumina mundi, Catullus may have in mind the etymology given by Varro: appellatur ... graece ab ornatu κόσμος, latine a puritia mundus (Men. 420 Astbury) and he may also have been prompted by an etymology in the scholia such as that found at [Aristot.] Mund. 400a6–7: ἐν καθαρῷ χωρῷ …, ὃν ἐτύμως καλοῦμεν οὐρανὸν … ἀπὸ τοῦ ὅρον εἶναι τὸν ἄνω (‘in a pure region, which we rightly call heaven (οὐρανός) from its being the upper boundary’).¹² If dispexit is accepted, then Catullus has shifted the emphasis in the first line from Conon’s geometric skills to his expert observations.¹³ Conon’s ability to recognise a new constellation depends on his having learned which stars and constellations had been identified previously.¹⁴

The Callimachean original which lies behind 66.2–4 is lost. From Catullus, we can see that Callimachus referred to a variety of astronomical topics on which Conon had worked.¹⁵ In 66.2 comperit probably refers to his calculations of the orbits of the named stars and constellations. These calculations will have involved the use of line-drawings to demonstrate how the stars moved in relation both to each other and to the earth to produce solar eclipses (66.3), the regular procession of the signs of the zodiac that mark out the sidereal year (66.4) and the regular phases of the moon (66.5–6). There is some evidence that Conon was close to Ptolemy III and he may have dedicated some of his works to him.¹⁶ The investigations of both the sidereal year and the lunar month will have been of particular interest to Ptolemy, who attempted to introduce a leap-year calendar in 238 BC so that the religious and monthly calendar stayed in line with the rising of the named stars and constellations and with the seasons.¹⁷ If this is correct, then we have to suppose that Catullus might have removed other technical vocabulary which specified the nature of Conon’s activities.¹⁸ He has instead aimed at elegant paraphrase, which allows the reader to at least follow the key role of these lines in Callimachus’ argument: namely, that the astronomical and mathematical expertise of Conon serves to ‘guarantee’ the truth of Conon’s latest discovery, the catasterism of the Lock of Berenice.

In 66.5–6, Conon’s discoveries about the movements of the moon are presented in mythical form with reference to the well-known story of Selene’s love for Endymion, conveyed through furtim and dulcis Amor.¹⁹ The verb deuocet shows that Amor is here playing the role of a practitioner who uses magic spells to summon down the goddess to her lover.²⁰ The myth is specified allusively through the reference to Latmia … saxa.²¹ Since very similar phrasing is found also in Apollonius’ Argonautica (4.57 Λάτμιον ἄντρον) and in [Theocritus] (20. 38 Λάτμιον … νάπος), it is reasonable to suppose that Catullus in this couplet is following Callimachus closely.²² Triuia is Diana (here Selene) and alludes to the etymology found in Varro, Ling. 7.16: Diana est ab eo dicta Triuia, … quod Luna dicitur esse, quae in caelo tribus uiis mouetur, in altitudinem et latitudinem et longitudinem. Plutarch (De fac. 937f) provides a very similar etymology for Τριοδῖτις as a name for the moon, and specifically notes the interest of οἱ μαθηματικοί in the complexities of the three movements. The reference is probably to Conon’s work on the phases of the moon and especially of the new moon (when the moon is not visible): see Etym. magn. 153.6–9 ‘when Selene slept with Endymion the rest of the world went moonless’.²³ Possibly Callimachus was here alluding to a use of the myth by Conon himself.²⁴ Since Endymion was held to have been the first astronomer,²⁵ Callimachus may also have been implying an elegantly encomiastic comparison with Conon, whose discovery of the new constellation is proclaimed in the next couplet.

II The catasterism of the Lock, Berenice’s vow and dedication

The central couplet of the opening sentence for the first time identifies the speaker (me … caesariem), and the laudandus of the previous couplets (Conon), and transfers the focus to Berenice (66.7–10):

idem me ille Conon caelesti in limine uidit

e Bereniceo uertice caesariem

fulgentem clare, quam cunctis illa deorum

leuia protendens brachia pollicita est,

The corresponding lines of Callimachus are cited by the scholia to Aratus 146:

Κόνων δὲ ὁ μαθηματικὸς Πτολεμαίῳ χαριζόμενος Βερονίκης πλόκαμον ἐξ αὐτῶν

κατηστέρισε. τοῦτο καὶ Καλλίμαχός πού φησιν·

†η† με Κόνων ἔβλεψεν ἐν ἠέρι τὸν Βερενίκης

βόστρυχον, ὃν κείνη πᾶσιν ἔθηκε θεοῖς.

Conon the mathematician, seeking favour with Ptolemy, marked out from these [sc. unnamed stars] the Lock of Berenice. Callimachus also says this somewhere: ‘Conon saw me in the sky, the Lock of Berenice which she dedicated to all the gods’.

These lines of Callimachus have been the subject of much debate.²⁶ The main problem is that ἔθηκε cannot mean ‘vow’ or ‘promise’. It must mean ‘dedicated’; and, in Catullus, the vow is made at this point and the dedication is mentioned explicitly only at 66.37–8 reddita … | … uota (cf. 62 deuotae). It seems to me that the simplest solution is that proposed by Massimilla, which is to extend the sense into the following couplet and read: ἔθηκε θεοῖς | εὐξαμένη (‘she dedicated to the gods, having vowed …’).²⁷ If that is correct, then Callimachus moves from Conon’s sighting of the Lock to the dedication that necessarily preceded the transformation of the Lock into a constellation, and then back further in time to the vow which had preceded the dedication.²⁸ All that is required is to allow the aorist in the relative clause to have the force of a pluperfect: ‘which she had dedicated to all the gods, having (previously) vowed …’.

The text of Catullus 66.7–10 is to a degree uncertain. The transmitted reading in 7 is caelesti numine. All editors agree that in is necessary in view of Callimachus’ ἐν ἠέρι. Most prefer Vossius’ in lumine. But Heinsius’ in limine seems to me preferable.²⁹ If it is correct, then in limine uidit may deliberately refer back to Callimachus’ ἰδὼν ὅρον (1) to underline that Catullus has chosen to frame the account of Conon’s geometric modelling with references to his observations rather than with the contrast between geometric calculation and observation evident in Callimachus. On any view Catullus has elaborated and elevated the simplicity of Callimachus, changing the emphasis without significantly altering the sense.

It is clear in line 9 that Catullus has also reordered the material; and the Callimachean source for fulgentem clare is lost. The phrase is interesting because the Coma Berenices notoriously consists of faint stars, a point remarked on elsewhere.³⁰ The point seems to be that, once Conon had defined the shape and location and provided an aition, the new constellation could be readily identified and became famous and so shone ‘distinctly’. A similar point is made by Callimachus in Hymn 4 about Asterie, formerly an obscure wandering island, to be renamed Delos (οὐκέτ’ ἄδηλος, 53) following the birth of Apollo.³¹ There is, however, a more difficult problem in 9, where the transmitted reading is multis … dearum. This is retained by many editors, although it seems to be inexplicably inconsistent with both Callimachus’ πᾶσιν … θεοῖς and with Catullus’ own recapitulation at 33–4, cunctis … diuis … pollicita es, where cunctis diuis precisely matches Callimachus’ πᾶσιν … θεοῖς, although the tone is more recherché and elevated. No satisfactory explanation is available for either multis or for dearum. The simplest solution seems to be to accept Haupt’s cunctis … deorum: substantival cunctus used in the masculine plural with a genitive is very rare and so the style is even more elevated than that of cunctis … diuis. The text of Callimachus at 33–4 is missing but he might have used a more elevated variation at that point; if so, Catullus might have thought the more exotic form more appropriate to the opening of the poem.³² Line 7 (idem me ille Conon caelesti in limine uidit) corresponds closely in sense to Callimachus’ hexameter with some elaboration. idem emphasises that the serious and high-minded mathematicus just described was also capable of playful wit and human empathy: Conon used his skills to ‘find’ the lost sacrifice and managed as a result to console the distraught Ptolemy.³³ But line 8 transforms the simplicity of Callimachus’ τὸν Βερενίκης | βόστρυχον so that it occupies the entire pentameter. In place of the genitive of Berenice’s name, Catullus uses the more elevated adjectival phrase (Bereniceo uertice, 8). At line 62, Callimachus has Βερενίκειος καλὸς ἐγὼ πλόκαμος. Catullus seems to have wanted to have the more impressive phrasing for the first mention of Berenice and of the lock. The less familiar form perhaps encourages the reader to recognise that Berenice means ‘Victory Bringer’: through her pious prayers, her love for her husband and her personal offering, Berenice has secured the support of the gods and so helped to bring about her husband’s victory. That Catullus’ decision was deliberate is suggested by the self-conscious way in which his own very different version of 62 (deuotae flaui uerticis exuuiae) looks back to these lines (8–10): uerticis picks up uertice, and deuotae formally defines pollicita est. Catullus translates Callimachus’ βόστρυχον by caesariem: both are relatively rare and thus striking words for long flowing hair, but Catullus probably expected his readers to be aware of the ancient etymology caesaries a caedendo.³⁴ He may have been prompted by Callimachus’ νεότμητον (51), or by whatever exegetical material he read alongside the text of Callimachus, which probably specified the nature of the vow in a similar way to Hyginus Astron. 2.24.1 se [sc.Berenice] crinem detonsuram; Ps. Nonn. Hist. I ad Greg. Naz. Or. V (In Iulian. II) 5. 10–11 (Nimmo Smith): τὸν πλόκαμον ἀποκαρεῖσα τὸν ἑαυτῆς. While it is clear that quam … illa corresponds to ὃν κείνη, pollicita est cannot be a translation of ἔθηκε which, as we have seen, means ‘had dedicated’. If Massimilla’s conjecture is correct, then Catullus has simply omitted the reference to dedication at this point; and his pollicita est corresponds to εὐξαμένη. His choice of polliceri is not a standard part of religious language and is used of ‘making a vow to a god’ only here and at 66.34;³⁵ perhaps significantly, however, the brief prose summary of Callimachus’ poem (the Callimachean Diegesis), which preserves the first line of the poem, reads:³⁶

φησὶν ὅτι Κόνων κατηστέρισε τὸν Βερενίκης βόστρυχον, ὃν θεοῖς ἀναθήσειν ὑπέσχετο κείνη, ἐπειδὰν ἐπανήκῃ ἀπὸ τῆς κατὰ Συρίαν μάχης.

He [Callimachus] says that Conon marked out as a constellation the Lock of Berenice, which she [Berenice] promised to dedicate to the gods when he [Ptolemy] returned from the war against Syria.

So polliceri may have been suggested by a similar use of ὑπέσχετο in an exegetical text available to Catullus. However that may be, Catullus moves directly from Conon’s sighting of the constellation to Berenice’s ‘promise’ to cut off the hair from her head, if her prayers, implied by leuia protendens brachia (66.10),³⁷ are granted.

The meaning of 7–10 (me…quam cunctis … deorum … pollicita est) is recapitulated and clarified at 33–5:

atque ibi me cunctis pro dulci coniuge diuis

non sine taurino sanguine pollicita es,

si reditum tetulisset.

and then it was that you [sc. Berenice] promised me, together with the blood of bulls, to all the gods in return for your beloved husband, if he returned home.

The addition of pro dulci coniuge (33) emphasises the very personal nature of her offering and her now passionate longing for her new husband.³⁸ In her public role as Queen, Berenice had also offered a bull sacrifice, the lavish scale of which is hinted at in the litotes non sine taurino sanguine (34). The substance of her prayers is summarised with economical simplicity (si reditum tetulisset, 35): the archaic reduplicated form lends appropriate dignity and evokes the language of prayer.³⁹ It is only when the prayers have been granted and Ptolemy has safely (and successfully) returned that the Lock is removed and dedicated (quis ego pro factis … reddita, 37) and the vows are discharged (uota … dissoluo, 38). The logic of 35–8 clearly carries the implication that the sacrifices were made to precisely the same gods as the prayers in 9–10: they had given what had then been requested. And the natural inference must be that, after the return of the King, the sacrifice was made in the same place as the propemptic prayers had been made on his departure.

Although some commentators have taken a different view, that is also the interpretation that is supported by the evidence of Hyginus, who explains the ‘origin’ (aetion) of what he calls the crinis Berenices by providing the following summary of Callimachus’ account (Callimachus dicit) at Astronomica 2.24.1:

cum Ptolemaeus Berenicen, Ptolemaei et Arsinoes filiam, sororem suam, duxisset uxorem et paucis post diebus Asiam oppugnatum profectus esset, uouisse Berenicen, si uictor Ptolemaeus redisset, se crinem detonsuram; quo uoto damnatam crinem in Veneris Arsinoes Zephyritidis posuisse templo eumque postero die non comparuisse.

when Ptolemy [III] had taken Berenice, the daughter of Ptolemy [II] and Arsinoe [II], his own sister, as his wife, and had set out, a few days after, to attack Asia, Berenice vowed that, if Ptolemy returned victorious, she would cut off her lock; and that, when she was obliged to fulfil this vow,⁴⁰ she dedicated it⁴¹ in the temple of Aphrodite Arsinoe Zephyritis and that on the following day it had disappeared.⁴²

The identity and date of ‘Hyginus’ are controversial, but the most recent editors of the Astronomica have persuasively argued that at least the author of the Astronomica is identical with C. Julius Hyginus, the freedman of Augustus, sometime librarian of the Palatine Library.⁴³ If that is correct, then the sources available to Hyginus will have been similar to those available to Catullus and his readers. For the main part, Hyginus in the Astronomica seems to follow the Catasterismi attributed to Eratosthenes of Cyrene, which provided prose summaries of mythical accounts of the various constellations,⁴⁴ probably in a fuller version than the one surviving.⁴⁵ But Hyginus cites Conon and Callimachus rather than Eratosthenes as his sources for his account of the crinis (i.e. Coma) Berenices:⁴⁶ it is more likely that Hyginus took his summary from a fuller and better-informed version of the Callimachean Diegesis than that he personally summarised the texts of Conon or Callimachus.⁴⁷ At a number of points Hyginus’ prose summary clarifies and makes explicit what is in the texts of Callimachus and Catullus. Hyginus says that Berenice dedicated her lock of hair in the temple of Aphrodite Arsinoe Zephyritis.⁴⁸ There is no reason to doubt this.⁴⁹ If we assume, as seems likely, that the dedication was made to the same gods and in the same temple as the vow was made, then the scene described in 66.9–12 is set in this same temple. That seems entirely appropriate, since the temple was constructed on Cape Zephyrion by the Ptolemaic admiral Callicrates. Aphrodite Arsinoe is one of the forms in which the deified Arsinoe, ‘mother’ of Ptolemy III and of Berenice II, was worshipped. As a goddess of the seas she protected travellers and sailors; and as a goddess of marriage she had care of young girls seeking marriage and the newly married (Posidippus 116 A–B = 12 HE). Her temple would be the obvious setting for the prayers and vows which Berenice, the King’s chaste new wife, made for him as he set off with his fleet to war. So far as we can tell from Catullus, Callimachus did not make the setting explicit: since he was writing about a major event that would have been fresh in the minds of his readers, he may have felt no need to do so. Catullus has not chosen to make the point more explicit. But the special role of Aphrodite Arsinoe can be assumed to be intensified in relation both to the new King leading out his navy on his first campaign and to the recent wedding of Berenice to Ptolemy, since they were her ‘adoptive’ children and the successors to herself and her sibling husband.⁵⁰ Recognising the setting in the temple of Aphrodite Arsinoe is crucial to understanding the coherence of the deceptively simple account of these events (7–14).

III Ptolemy’s departure and the Syrian campaign

qua rex tempestate nouo auectus hymenaeo

uastatum finis ierat Assyrios,

The text of line 11 is uncertain. In the major manuscripts of Catullus (OGR), line 11 ends: nouo auctus hymenaeo. The hiatus is problematic and almost certainly points to the need for emendation. If auctus is retained, then the solution would be to read nouis … hymenaeis.⁵¹ But to find a meaning for auctus that suits this context is not straightforward, and none of its defenders has provided a convincing explanation. Basically, auctus means ‘increased (in size or number), strengthened or enriched (by additional resources), enhanced or boosted’. Here many have taken it to mean, rather vaguely, ‘blessed’: an empty line-filler.⁵² But it is far from obvious in what sense Ptolemy could be said to have been ‘increased in power, strength or importance’ by his ‘recent wedding’.⁵³ The usual comparison is with 64.25, where Peleus is addressed as taedis felicibus aucte. But the sense there is unproblematic: the standing of the mortal hero is enhanced by his marriage to the goddess Thetis and by the facts that the union is sponsored by Jupiter and that it will result in his being the father of Achilles. A simple solution to both the problem of the hiatus and the problem of sense is to read auectus. It appears in one manuscript (c. 1460) and was independently conjectured by Peiper (1875) and Birt (1904).⁵⁴ The words nouo … hymenaeo anticipate the account of the wedding provided in 13–14, and auectus introduces the theme of Berenice as abandoned newlywed which dominates the following section (66.15–34). It is entirely appropriate to its context:⁵⁵ when Ptolemy left for Syria he sailed from Alexandria.⁵⁶ The word might also suggest an appropriate unwillingness to leave his beloved new bride: Catullus uses it with similar emotional force at 64.132 in describing Ariadne, who was carried off in his ship by Theseus and then deserted far from home (me patriis auectam … ab aris).⁵⁷

The succinct account of the King’s departure, which provides in a compressed form the background to the vow, is expressed in language that is elevated by the use of archaic language: qua … tempestate; the pluperfect form ierat; and the use of ire with supine which here governs the accusative.⁵⁸ The purpose of the campaign is indicated by the verb uastare, which is normally used to describe punitive attacks on enemy territory undertaken to deter or avenge aggression.⁵⁹ Ptolemy certainly brought back a huge amount of booty from his campaign (Porphyrios, BNJ 260 F 43):

plundering the kingdom of Seleukos, [Ptolemy] carried off forty thousand talents of silver and two thousand valuable vessels and statues of gods, among which were those that Cambyses had taken from Egypt and carried off to Persia.

And this is confirmed by sources written very shortly after his return. The contemporary Adulis inscription says (OGIS 54.20–2):⁶⁰

he sought out all the sacred objects which had been taken away from Egypt by the Persians and brought them back with the rest of the treasure from the provinces [of Syria] to Egypt.

As there is no particular reason to suppose that Catullus is misrepresenting Callimachus here, it is at least possible that Callimachus also used the adjective Assyrios to define Ptolemy’s area of operations and did so to refer to the area around the Euphrates (as in Hymn 2.108), the area in which he carried out his punitive campaign. Catullus and, presumably, Callimachus have aligned Ptolemy’s original purpose for setting out with what was represented as the successful outcome of his campaign. This is confirmed by the recapitulation at 66.35–6: is haud in tempore longo | captam Asiam Aegypti finibus addiderat (‘He, within a period of no great duration, had taken Asia and added it to Egypt’). The simple but sweeping language renders the Callimachean original (whatever it was) in the language and style traditionally used to describe the achievements of Roman military commanders.⁶¹ The claim is in line with the way the achievements of Ptolemy III were reported in the period after his return, as is confirmed by the more detailed account given in the Adulis inscription (OGIS 54.1–20):

The Great King Ptolemy (1) … having inherited from his father the kingship (τὴν βασιλείαν, 5–6) of Egypt … Syria and Phoenicia … Lycia and Caria …, launched a campaign against Asia (ἐξεστράτευσεν εἰς τὴν Ἀσίαν, 8) … Having taken control of all the territory west of the Euphrates (κυριεύσας δὲ τῆς τε ἐντὸς Εὐφράτου χώρας πάσης, 13–14) … and having made all the rulers in the provinces subject to him he crossed the river Euphrates and having made … all the rest as far as Bactria subject to him …’

A similar account is provided by Porphyrios of Tyre (BNJ 260 F43):

Ptolemy came with a great army and invaded the province … he conquered so much that he captured (caperet) Syria and Cilicia and the upper parts beyond the Euphrates and almost the whole of Asia (prope modum uniuersam Asiam). … And indeed Ptolemy himself held Syria but he handed over Cilicia to his friend Antiochus to govern and to Xanthippus, the other general, the provinces across the Euphrates.

Catullus’ emphasis on the extent of Ptolemy’s victories and his claim to have brought Asia under the control of Egypt both clearly reflect alignment with the official view promoted at the time of Ptolemy’s return.⁶²

The original reason for Ptolemy’s abandoning his new bride and attacking Syria was, according to our sources, very different. Berenice, the daughter of Ptolemy II and Arsinoe I and the full sister, by birth, of Ptolemy III, had married the Seleucid king, Antiochus II, to seal the peace that had concluded the Second Syrian War in 252 BC. Ptolemy II had provided her with a famously large dowry consisting of ‘countless thousands of gold and silver’ and it was agreed that her son would succeed Antiochus II as King of Syria. Antiochus II died in the summer of 246 BC while he was staying at Ephesus with his former wife Laodice and their sons (Seleucos II and Antiochus Hierax), and, as will be seen later, news of his death was reported in Babylon on 19 August 246. Some sources say he was poisoned; that was presumably the Ptolemaic version. Whatever the truth, Laodice claimed that the final wish of Antiochus II was that Seleucus should succeed him. As soon as Berenice, wife of Antiochus, who was based in Antioch with her young son (also called Antiochus), heard the news, she will have recognised the danger to herself and her son. She sent an urgent request to Ptolemy III for support and protection. According to Justin (27.4–8):

When she learned that men had been sent to kill her, Berenice [i.e. wife of Antiochus] barricaded herself in Daphina [near Antioch]. When it was reported to the cities of Asia that she was being besieged there with her tiny son (paruulo filio), they all sent help to her (auxilia), feeling pity for such an undeserved change of fortune (casum tam indignae fortunae) in memory of the high standing (dignitatis) of her father [Ptolemy II] and his ancestors. Her brother Ptolemy [III], terrified by the danger his sister was in, left his kingdom and flew [to her aid] with all his forces (Ptolomeus periculo sororis exterritus relicto regno cum omnibus uiribus aduolat). But Berenice, before the relieving forces arrived (ante aduentum auxiliorum), since she could not be defeated by force (ui expugnari), was caught by treachery and murdered.

Justin’s account is supported generally by other sources.⁶³ The Gurob Papyrus (BNJ 160), which appears to be Ptolemy III’s official report of the initial campaign of the Third Syrian War, seems to show Berenice still alive and in control of Antioch when Ptolemy first arrives with a substantial naval force.⁶⁴ It is also clear from an inscription (SEG 42.994) that there was time for Ptolemy III and his sister Berenice to declare her son Antiochus (the paruulus filius) as King.⁶⁵ But, whatever the truth of the matter in detail, it is clear that Berenice and her son had both been killed soon after Ptolemy arrived, if not before. Following their deaths, Ptolemy had continued his campaign in Syria: his motive now was to avenge his sister and her son; to inflict major disruption on Syria and secure at least some strongholds in Syria; and to demonstrate his military power and claim to be a worthy heir to Alexander the Great in Egypt.⁶⁶ It is clear from an unpublished Babylonian cuneiform tablet (BCHP 11) that Ptolemy was involved in besieging Seleucia on the Euphrates (Zeugma) and then Babylon from December 246 to March 245.⁶⁷ And the recently published Alexandria Decree provides some evidence that he went at least as far as Susa.⁶⁸ The expectation that Egypt would exercise control over Asia presumably seemed plausible enough (Ptolemy had appointed his governors) and the war seems to have continued for two or three more years.⁶⁹

Catullus, like (presumably) Callimachus, has removed all reference to Ptolemy’s failed attempt to protect his sister and nephew and preserve their claim to the Seleucid throne. On one level that may be seen as a matter of tact: when celebrating the king’s return, it was more appropriate to emphasise his successes. But we may also suppose that Callimachus was participating in a deliberate rebranding of the very reasons for the campaign and a reshaping of history. Interestingly enough, although the evidence is less clear, it appears from Catullus that there was a similar move shaping the narrative of his return (66.35–6). Catullus makes no reference to the seditio (‘uprising’) which several sources said was the reason for his swift return.⁷⁰ His recapitulation of the motive for departure (11–12) suggests only that that purpose has been achieved and that the prayers made by Berenice included (unsurprisingly) a prayer for victory, as implied by Hyginus 2.24.1: uouisse Berenicen, si uictor Ptolemaeus redisset. And Catullus wittily suggests, by having haud in tempore longo (36) pick up amantes | non longe a caro corpore abesse uolunt (31–2), that the desire of Berenice and Ptolemy to be reunited contributed something to his swift return. So here again, though less explicitly than in 66.11–12, the king’s actions are related to his recent wedding (nouo … hymenaeo, 11). Again, from Catullus, it appears that a difficult historical reality has been simply removed from sight. There is no sign that Catullus was aware of the more complex historical realities lying behind Callimachus’ text, although he potentially had access to better accounts of the period than the modern reader. It seems most likely that Callimachus also omitted reference to the failed attempt to protect Berenice, wife of Antiochus II, and her son.

IV The wedding night

The closing couplet of the opening sentence continues the description of the departing king, but the focus switches from the war in prospect (the proelia torua, 20) to the struggle (rixa) with Berenice for her virginity on their wedding night (66.13–14):

dulcia nocturnae portans uestigia rixae

quam de uirgineis gesserat exuuiis.

The text is unproblematic and, although Callimachus’ text does not survive at this point, parallels between Catullus’ phrasing and that of two Greek epigrams strongly supports the view that he is following Callimachus closely. It seems highly probable, as Harder suggests, that the military language used to describe Ptolemy’s struggle to take Berenice’s virginity was present in Callimachus.⁷¹

Catullus’ translation is expressed with stylised elegance: 13 is a ‘golden’ line with adjectives and nouns arranged around the verbal element (portans) in abAB pattern. The couplet picks up and expands on the implications of nouo … hymenaeo (11), vividly evoking the wedding night of Berenice and Ptolemy and emphatically underlining how close in time the wedding was to Ptolemy’s departure for war. The king leaves still bearing the marks of Berenice’s nails, inflicted in her fight to defend her virginity.⁷² But this is a ‘battle of love’ and dulcia both looks back to dulcis Amor (6) and anticipates pro dulci coniuge (33). The novel expression quam [sc. rixam] gesserat is clearly based on the standard phrase bellum gesserat, and in combination with exuuiae (14) belongs to the erotic topos of militia amoris, which is found in epithalamic poetry to describe the taking of a bride’s virginity as well as in other forms of love poetry.⁷³ The phrase de uirgineis … exuuiis is a vivid but euphemistic reference to Ptolemy taking the virginity of Berenice on the night of their wedding.⁷⁴ The military language of the couplet points up the contrast between this scene of passionate lovemaking and the grim realities of the war which has taken Ptolemy away. Ptolemy’s wedding night becomes an omen of success for the war to come. His victory in the battle with Berenice de uirgineis … exuuiis anticipates the spoils he would bring back from his campaign (with exuuiae looking back to uastatum, 12).

The closeness in time of Ptolemy’s departure for war and his wedding to Berenice is not only implied in 13–14 but repeated clearly in nouo … hymenaeo (11), nouis nuptis (15) and nouo … uiro (20).⁷⁵ The reference to the wedding is sustained and given further prominence by the use of familiar epithalamic topoi in 15–18 (the bride’s fear of her first sexual experience and her tears for her loss of her virginity). There can be no doubt that the emphasis is Callimachean and thus contemporary with the events described. There is no evidence for the precise date of the wedding, or even for its timing in relationship to the Third Syrian War, apart from Hyginus’ statement that Ptolemy departed for the war just a few days after the wedding (Astron. 2.24.1 paucis post diebus). Catullus’ description of Ptolemy (nocturnae portans uestigia rixae, 13) suggests that he left the day after the wedding but Hyginus’ more specific phrasing is more realistic: some time would be needed to prepare even a small expeditionary fleet. But what Hyginus says is consistent with Catullus’ repeated insistence on the fact that Ptolemy left immediately after the wedding. Since Hyginus is clearly drawing on early summaries of Callimachus or on earlier scholia on his text, the information he provides should not be lightly dismissed. The modern orthodoxy is however different. Some date the wedding to around the accession of Ptolemy III (at the end of January 246 BC); others place it even some years earlier.⁷⁶ The Third Syrian War arose from the events immediately following the death of Antiochus II, which was announced in Babylon on 19 August 246.⁷⁷ Antiochus cannot have died much before the end of July 246 BC. Catullus (presumably following Callimachus) implies that the wedding celebrations of Ptolemy and Berenice were unexpectedly interrupted by the need for Ptolemy to depart. That means that the wedding cannot be earlier than July 246 BC. There is no reason why Callimachus should have misrepresented so emphatically the relationship between two recent important and doubtless memorable public events: the King’s wedding to Berenice and his departure on his first campaign.⁷⁸ A gap of six months or more simply does not fit with what Catullus says.

But 66.13–14 (reinforced by 15–18) raise an even more serious historical problem. The couplet represents Berenice as a virgin bride: the words de uirgineis … exuuiis make the point vividly. The same point is made again, more playfully, at 66.77–8 and in the corresponding lines of Callimachus: there the Lock complains about having had no opportunity to enjoy the scented perfumes used by married women and having ‘drunk’ only plenty of the cheap olive oil used by virgins. The Latin text is very uncertain but in Catullus Berenice is described as uirgo; and in Callimachus as παρθενίη, which is contrasted with γυναικείων.⁷⁹ In spite of this, it is commonly said that Berenice had been previously married to Demetrius the Fair at the insistence of her mother following the death of her father Magas. The only evidence for this episode in the life of Berenice is Justin’s epitome of the Augustan historian Pompeius Trogus (26.3.2–8). It deserves careful consideration.

The passage runs as follows:

per idem tempus rex Cyrenarum Magas decedit, qui ante infirmitatem Beronicen, unicam filiam, ad finienda cum Ptolomeo fratre certamina filio eius desponderat. (3) sed post mortem regis mater uirginis Arsinoe, ut inuita se contractum matrimonium solueretur, misit qui ad nuptias uirginis regnumque Cyrenarum Demetrium, fratrem regis Antigoni, a Macedonia arcesserent, qui et ipse ex filia Ptolomei procreatus erat. (4) sed nec Demetrius moram fecit. itaque cum secundante uento celeriter Cyrenas aduolasset, fiducia pulchritudinis, qua animis placere socrus coeperat, statim a principio superbus regiae familiae militibusque inpotens erat studiumque placendi a uirgine in matrem contulerat. (5) quae res suspecta primo uirgini, dein popularibus militibusque inuisa fuit. (6) itaque uersis omnium animis in Ptolomei filium insidiae Demetrio conparantur, cui, cum in lectum socrus concessisset, percussores inmittuntur. (7) sed Arsinoe audita uoce filiae ad fores stantis et praecipientis ut matri parceretur, adulterum paulisper corpore suo protexit. (8) quo interfecto Beronice et stupra matris salua pietate ulta est et in matrimonio sortiendo iudicium patris secuta.

During the same period the king of Cyrene, Magas, died, and he, before his final illness, in order to bring to an end his conflicts with his [half-]brother Ptolemy [II], had betrothed Berenice, his only daughter, to the son of the latter. (3) But after the death of the king the girl’s mother, Arsinoe, in order that the marriage, which had been agreed against her will, should be annulled, sent people to fetch from Macedonia to be married to the girl and to be ruler of Cyrene the [half-]brother of king Antigonus [Gonatas], Demetrius, who also had himself been born to a daughter of Ptolemy [I]. (4) And Demetrius wasted no time. And so, when with a following wind he had quickly sailed to Cyrene, trusting in his good looks, with which he had begun to win the heart of his socrus, he, right from the start, behaved haughtily towards the royal family and without due respect to the military and had switched his eagerness to find favour from the girl to her mother. (5) This behaviour was first viewed with distrust by the girl, and was then rejected by the people and the military. (6) And so, once the minds of all had turned to the son of Ptolemy [II], plots were hatched against Demetrius and, when he had retired to the bed of his socrus, assassins were sent in to [kill] him. (7) But Arsinoe, on hearing the voice of her daughter standing by the doors and urging that her mother be spared, protected her lover for a little while with her own body. (8) Through his assassination Berenice exacted retribution for her mother’s disgrace without violating her own filial duty and, in her choice of marriage, complied with the decision of her father.

Justin paraphrases Pompeius Trogus’ account of the death of Demetrius.⁸⁰ Although the account provides important insights into the events following the death of Magas, Justin is not primarily concerned to give an accurate summary of the facts. Most strikingly he twice wrongly names Berenice’s mother and the wife of Magas as Arsinoe.⁸¹ But there is now no doubt that her birth mother was Apama, as is clearly stated by the scholia to Callimachus 110.45 (= 110e Harder): Ἀρσινόης μητρός· κατὰ τιμὴν εἶπεν, ἐπεὶ Θυγάτηρ Ἀπάμας καὶ Μάγα.⁸² Apama was related by birth to the Seleucid kings of Syria and by marriage to the Antigonids who ruled in Macedon; and, as Justin clearly shows, her motivation was to thwart the decision of her husband that Cyrene should after his death return to Ptolemaic control. Justin has decided to convert an episode from Trogus’ account of how the usurper Demetrius met his death to a moral tale of a pious daughter (Berenice) seeking to uphold her father’s wishes against the machinations of her impious mother. In naming Berenice’s mother Arsinoe, he reveals his lack of concern with the political dimensions of the narrative and, whether through conscious decision or inattention, he gives her mother the name which reflects Berenice’s official style following her marriage to Ptolemy III, and which was probably the way she was generally distinguished from others of the same name.⁸³

In Justin’s account, Demetrius exhibits the hallmarks of the tyrant: superbia (his lack of respect for the regia familia), impotentia (his lack of self-control in his dealings with the milites), both of which relate to his rejection by ‘good’ men (presumably the supporters of Berenice), sexual depravity (his affair with Apama) and secrecy (implied by suspecta).⁸⁴ Justin says that Apama promised or contracted to give Berenice in marriage to Demetrius (misit qui ad nuptias uirginis … Demetrium … arcesserent); but he never says that the marriage took place. Instead he says explicitly that Demetrius had immediately (statim a principio) switched his attentions from Berenice to her mother Apama (studiumque placendi a uirgine in matrem contulerat). At this point Berenice is still described as uirgo and there is no suggestion that the wedding has happened. Justin twice refers to Apama as Demetrius’ socrus; but that word encompasses those who promise their child in marriage and not only those who go on, as most would, to become mothers-in-law.⁸⁵ Justin uses it to point up Demetrius’ disregard for morality, law and decency. The first occasion is soon after his arrival, where the attraction of Apama to Demetrius’ charms is used to explain his arrogant behaviour (superbus). The implication is that Demetrius had assumed the role of king with Apama’s support.⁸⁶ Nothing is said which suggests that he had at this point married Berenice, who is twice referred to as uirgo before the word socrus appears for the second time, when the scene is set for the assassination of Demetrius while he was in bed with Apama. When, in the same context, Demetrius is described as adulter, that is with reference to his disgraceful sexual relationship with Apama, the widow of Magas, to whom she still owed wifely loyalty, a sexual relationship subsequently stigmatised as stupra matris.⁸⁷ It is not evidence that he had married Berenice. When referring to Berenice, Justin does so twice by name, once at the beginning and once at the end to frame his summary account. He twice refers to her as filia, once when her father betroths her and once when her mother hears her urging the assassins to spare her, both to emphasise her filial piety. He also refers to her four times as uirgo; but he never refers to her as nupta, coniunx or uxor. There are also positive indications that the marriage never happened. Once Demetrius transferred his affections to Apama, Justin says that everyone turned their minds to the son of Ptolemy II (uersis omnium animis in Ptolomei filium) and the omnes clearly include not just Berenice but also the populares and the milites. The implication is that, at this point, it was still not resolved whether Berenice should marry Demetrius or Ptolemy. The closing words in matrimonio sortiendo iudicium patris secuta, most naturally mean that, as a result of the removal of Demetrius and her participation in the assassination, Berenice had made her choice in favour of her father’s candidate for her hand rather than her mother’s. It cannot be reconciled with the view that Berenice had been persuaded by her mother to marry Demetrius and subsequently changed her mind and had him killed, thereby freeing herself to marry the person preferred by her father.

V Conclusion

The opening lines of Catullus 66 provide a close translation of the opening of Callimachus’ Coma Berenices as it stood at the close of the Aetia. Callimachus had first composed a version of the poem to celebrate the return of Ptolemy III from his campaign in Syria (probably in the late summer of 245 BC) and the dedication of the lock of hair by Berenice in the temple of Arsinoe Aphrodite in fulfilment of the vow she made in the previous summer for his safe and victorious return. When the poem was revised for inclusion in the Aetia, probably two or more years later, Callimachus added details of the cult offering to be made to the catasterised and deified Coma (66.79–88) by faithful and happily married women. In the years between the King’s return and the publication of the Aetia both Ptolemy III and Berenice II had been incorporated into the dynastic cult. For all its wit and elegance and for all the emphasis given to the moving account of Berenice’s distress at the departure of her new husband and the fantastical description of the ascent of the Coma to its position among stars, Callimachus’ poem has a clearly serious contribution to make to the promotion of such novel honours. The opening praises of Conon, perhaps the most famous and genuinely outstanding mathematical astronomer of his time, serve to establish the credibility of the claim that the lock of hair sacrificed by Berenice has found favour with the gods as demonstrated both by their assistance to Ptolemy in his successful campaign in Syria and by the presence of the new constellation in the heavens. If it is correct to assume that Berenice made her vows as well as her dedication on their fulfilment in the temple of Arsinoe Aphrodite, then Berenice’s worship of Arsinoe Aphrodite and her special relationship with her divine adoptive mother marks the beginning (especially 9–10), the middle (especially 37–8) and the end of the poem (especially 89–90).

While Callimachus’ intentions to celebrate and promote Berenice and Ptolemy III are generally accepted, there are two widely held and deeply embedded beliefs that obscure the nature of Callimachus’ representation of their relationship. The first is that their wedding took place either shortly before or shortly after the accession of Ptolemy III or even earlier, following the intervention of Ptolemy II in Cyrene after the death of Magas. The uncertainty stems from the lack of evidence other than the text of Catullus. If, however, we accept that Catullus 66 is a generally reliable guide to what Callimachus wrote, then we have a witness to Callimachus’ almost contemporary account that the wedding took place immediately before Ptolemy’s departure in July or August 246. The passionate physical relationship of Berenice and her new husband is depicted at length with significant emphasis. It is worth noting that Berenice’s first child was born either just before (or, possibly, just after) Ptolemy’s return and the second and third children, at least, were born before the revision of the Coma and its inclusion in the Aetia. The future of the dynasty was secure for at least another generation.

The second problematic belief is that not only did Apama, the widow of Magas, seek to thwart her husband’s wishes that Cyrene should be re-united with Egypt through a marriage of his daughter Berenice to the son of Ptolemy II but that she had successfully persuaded Berenice to marry Demetrius. Justin’s account, our only evidence for this episode, does not, as we have seen, support this view. It is not easy to understand why Callimachus (or anyone else) should have been so concerned to represent Berenice as a virgin bride if it were not true: both of her dynastic predecessors, Berenice I and Arsinoe II, had had husbands before marrying, respectively, Ptolemy Soter and Ptolemy Philadelphus. Although a virgin bride was an ideal, for Callimachus to have given such prominence to her virginity if it was well known or widely believed in court circles that Berenice had been previously married would have been to risk ridicule and undermining the credibility of his carefully constructed image of the new king and queen.

Of course, Callimachus avoids all reference to some crucial details of the underlying events, concentrating instead on Berenice’s private display of distress at the prospect of separation from her husband and on the fantastical detail of the experiences of the Coma and her transformation into a constellation. No detail is provided about the reasons that necessitated Ptolemy’s urgent departure. Callimachus assumes that his audience will be all too well aware of the death of Berenice, wife of Antiochus, and her young son at the hands of the Seleucids and the violation of the solemn agreement made between Antiochus and Ptolemy II. Ptolemy III’s campaign is represented by Catullus, presumably following Callimachus, as punitive, swift and decisive and establishes him as a worthy successor to Alexander the Great. There is no suggestion that his swift return has any cause other than the successful completion of his mission, apart from a witty and knowing suggestion that he was longing to return to his new bride. The seditio domestica mentioned by the historical sources is totally absent. But so too is any overt reference to the enormous value of the treasure and other booty which he brought back to Egypt which financed the euergetism that earned for him and Berenice the epiclesis Euergetai (the Benefactors). While the poem is deeply embedded in its contemporary context, so far as we can judge from the fragmentary evidence we have, Callimachus incorporates current propaganda with a light touch. The King is treated with the respect his position deserves; and Berenice is treated with the fondness and pride in her qualities and achievements that are appropriate to her lifelong companion, the Coma who is the narrator throughout.

Appendix: Berenice’s bonum facinus

In 66.19–32, the Coma recalls Berenice’s extreme distress on learning of her new husband’s departure so soon after the wedding and how she had rallied on being reminded of her earlier bravery (25–8):⁸⁸

at ego certe

cognoram a parua uirgine magnanimam.

anne bonum oblita es facinus, quo regium adepta es

coniugium, quo nil fortius ausit alis?

And yet I knew for certain that you were stout-hearted from being a slip of a maiden. Or have you forgotten the noble deed through which you gained your marriage to the King, than which another could dare nothing braver?

The quality to which the Coma appeals is Berenice’s magnanimitas, as unforgettably demonstrated by the bonum facinus through which regium adepta es | coniugium.⁸⁹ Catullus’ language here gives little away about the nature of Berenice’s bonum facinus: Berenice’s bold courage was something for which she was well known. Polybius (5.36.1) says that, when Sosibios was plotting the murder of her and her son Magas, they had to exercise great caution μάλιστα διὰ τὴν Βερενίκης τόλμαν (‘especially because of Berenice’s boldness’). It has to be assumed that Catullus has chosen not to expand for the benefit of his Roman readers but is following Callimachus closely. The adjective magnanimus, which is conventionally used of kings and heroes, is found first in Plautus (Amphitruo 212): magnanimi uiri freti uirtute et uiribus, which well illustrates its essential manly associations.⁹⁰ It is found applied to a woman only here (and Hyginus, Astron. 2.24.2 Callimachus eam magnanimam dixit). The implication is that Berenice had demonstrated a courage more typical of a man than a woman, and a courage that would have seemed heroic even in a man.⁹¹ The juxtaposition with uirgine and the polar contrast between parua and magnanimam gives further emphasis: her courage is not only remarkable for a woman but, given her youthfulness, even more so.⁹²

In the phrase bonum facinus, the adjective serves not just to rule out the negative connotations facinus had come to carry but also to indicate that the deed was one which showed the kind of noble qualities that might be expected of a magnanimus uir.⁹³ The bonum facinus is then further defined by two relative clauses, each introduced by quo. In the first, regium looks back to rex (11), the king whose departure has caused the display of womanly grief and distress described in the preceding lines: it is not simply that the bonum facinus secured for her a royal marriage or marriage with a king; it was the deed that won or secured her marriage with Ptolemy III. The second clause reasserts that the facinus was one that demonstrated exceptional courage: fortius looks back to and confirms magnanimam.⁹⁴ And that implication is further strengthened by the archaic ausit alis which suggests a demonstration of audacia (in its positive sense).

Since it is now generally assumed that the reference is to Berenice’s role in the assassination of Demetrius, it should be noted that the phrase a parua uirgine suggests that for Catullus (and presumably Callimachus) the bonum facinus which demonstrated her magnanimitas was performed while she was a uirgo (unmarried).⁹⁵ Justin’s account of the assassination is used to illustrate the pietas of Berenice rather than her bold courage. But it can, perhaps, be assumed that the story was told in various ways and that at least some of them emphasised Berenice’s boldness and courage.⁹⁶ The removal of Demetrius was necessary to free Berenice to follow through the marriage contracted between Magas and Ptolemy II. But it seems that Ptolemy saw no reason to see the marriage happen immediately. He was slow to respond to the unfolding events in Cyrene and it was at least two years before he intervened to take control of Cyrene. For reasons unknown, several more years passed before Ptolemy III finally married Berenice.

An alternative explanation is given by Hyginus (Astronomica 2.24.2):

hanc Berenicen nonnulli cum Callimacho dixerunt equos alere et ad Olympia mittere consuetam fuisse. alii⁹⁷ dicunt hoc amplius Ptolemaeum Berenices patrem, multitudine hostium perterritum, fuga salutem petisse; filiam autem saepe consuetam insiluisse in equum et reliquam copiam exercitus constituisse et complures hostium interfecisse, reliquos in fugam coniecisse; pro quo etiam Callimachus eam magnanimam dixit.

This Berenice some, including Callimachus, have said was in the habit of rearing horses and sending them to the Olympic Games. Others say in addition that Ptolemy, the father of Berenice, was [once] terrified by a large enemy force and sought safety in flight; but that his daughter, who had often practiced [this], leapt onto a horse and rallied the remaining military forces; and that she killed most of the enemy and put the rest to flight; and because of this Callimachus also called her μεγάθυμος.

Although the account of Berenice saving her ‘father’ lacks precision and historical context, and, no doubt, simplifies and exaggerates her role in killing and routing the enemy, it is plausible in itself.⁹⁸ Her status in Cyrene as Basilissa, her youth and her gender will have shamed her troops into rallying to support her. The detail about her leaping onto the horse is distinctive.

Since Magas, the King of Cyrene, was the birth-father of Berenice, the prevailing modern view is that this incident actually involved Berenice and Magas.⁹⁹ If that is the case, it has nothing to do with Berenice winning the hand of Ptolemy III, and Hyginus is wrong to relate it to Catullus’ magnanimam and so to the bonum facinus. But it is far from clear that the incident relates to Magas (her birth-father) rather than to Ptolemy II (her adoptive father).

Hyginus introduces Berenice as the daughter of Ptolemy II (Astron. 2.24.1 Berenicen Ptolemaei et Arsinoes filiam). In doing so, he is using her official genealogy and style. That is, importantly, also the way Callimachus refers to Berenice at the beginning of Aetia 3; and in the Coma (45) Callimachus also refers to Arsinoe explicitly as the mother of Berenice. As we have seen, this style of reference was also adopted by historians, and Justin, even when talking about Magas and Apama, gave her mother’s name as Arsinoe. The scholia were aware of the realities but perhaps they and the prose summaries related to them did not always feel the need to spell out the entire situation. If, as seems most likely, the story was taken by Hyginus from an early version of the Callimachean Diegesis or from scholia, then Catullus and his readers will have known the same explanation of Callimachus’ reference. There seems to be no reason to believe that the story does not relate to Ptolemy II and that patrem here is not simply the normal, official way of referring to her adoptive father.

At some point before his death in January 246, Ptolemy II had taken back control of Libya and Cyrene. It seems highly likely, though direct evidence is lacking, that Ptolemy intervened perhaps a year or two after the assassination of Demetrius the Fair, either on his own initiative or at the request of Berenice (or her friends in Alexandria).¹⁰⁰ An intervention in person during the turbulent period that followed the removal of Demetrius would provide an occasion for the incident related by Hyginus.¹⁰¹ Hyginus does not identify the hostes but, if the event did happen at this time, they could have been either disaffected citizens of the Greek cities wishing to claim their freedom or the native tribes whom the King of Cyrene was expected to keep under control. Ptolemy II may have proclaimed his adoption of Berenice (or his intention to adopt her) as a justification of his intervention; or it may have followed her demonstration of loyalty and bravery in the incident described by Hyginus. If this was what happened, then the adoption would be a prelude (and certainly no obstacle) to her future marriage to the son of Ptolemy II Philadelphus when he eventually succeeded to the throne.

The Hippika of Posidippus may provide some clue to what happened next. At least four of the epigrams (78–9; 81–2 A–B) refer to victories of a Berenice at various pan-Hellenic Games: the Olympian, Nemean and Isthmian Games. The identity of this Berenice is contested but, in my view, she is probably Berenice II.¹⁰² If this is correct, Ptolemy II used the opportunity of the pan-Hellenic Games to advertise both his adoption of (and, by implication, military backing for) Berenice, βασίλισσα of Cyrene, and his regaining of Cyrene (Libya). That would also explain why Hyginus or his source mention Berenice’s participation in the Games along with the rescue of her (adoptive) father.

Although much remains uncertain and problematic, Hyginus’ explanation of the reference implied by bonum facinus has much in its favour. As he describes the action, it indubitably qualifies Berenice for the description magnanima and is an unambiguous demonstration of her courage. If we assume that the incident does relate to Ptolemy II, as Hyginus says, in some such context as that suggested, then it is not difficult to see how it might have led, first, to her being immediately adopted (in 250 or 249 BC) and incorporated into the Ptolemaic family in order to consolidate the position of Ptolemy II in Cyrene; and subsequently, following his death and the succession of her ‘brother’ Ptolemy III in January 246, to her marriage to her ‘brother’ in July 246 BC. On the other hand, it is far from clear why Callimachus would have wanted to remind his readers so forcefully of the Demetrius episode in this poem: even if it is accepted that Demetrius never married Berenice, the nature of story and the complex historical background suggest that there will have been various versions in circulation and, while some may have emphasised Berenice’s bravery, others (reflecting an Antigonid or Seleucid perspective) may have equally been more hostile. When dealing with such potentially embarrassing events as the failed attempt to save Berenice (wife of Antiochus) and her son or the seditio that caused Ptolemy to return early from his Syrian campaign, Callimachus’ strategy – so far as we can discern it from Catullus’ poem – was silently to omit them.

This chapter is a substantially revised version of the paper given at the 2014 colloquium in memory of David West. Through his participation in the Boreas seminars (and other such occasions) he brilliantly exemplified how serious commitment to coming closer to an historical understanding of Latin poetry could be conducted convivially, informally and purposefully; and that honest disagreement about texts, if based on evidence and careful argument, can continue to contribute to that ultimately elusive goal. I warmly thank Tony Woodman for all his support and encouragement, which has, this time, gone far beyond anything one might reasonably expect of an editor. Responsibility for any errors and misjudgements that remain is mine.

¹Marinone (1997), Massimilla (2010), Harder (2012).

²For the two editions, see Massimilla (2010) 464; Harder (2012) 2.799–800. For the impossibility of reconstructing precisely the text of Callimachus from that of Catullus, see Bing (2009) 65–82, but I do not share his doubts about the reliability of Catullus as a general guide. On lines 79–88 of Cat. 66 see Du Quesnay (2012) 162–75, an essay to which the present chapter is a companion.

³The war continued under his generals until 242/241 BC.

⁴In an Appendix there is also some discussion of lines 25–8.

⁵The text of Catullus is problematic: that presented here is eclectic. For details, see Marinone (1997); Ramírez de Verger (2005); and, especially, Kiss (2013). Massimilla (2010) follows Marinone (1997) but retains the transmittted reading in 9 (multis … dearum) against Marinone’s (multis … deorum). Harder (2012) follows Thomson (1997) and retains the transmitted auctus (11); but reads iuerat (12). In line 7 I prefer in limine, the reading of four fifteenth century manuscripts (see Kiss (2013) ad loc.), to Vossius’ in lumine; and in 9 Haupt’s cunctis … deorum (9): see Courtney (1985) 92 and Syndikus (1990) 204 n. 25.

⁶There is no persuasive reason to think that Catullus, when faced with Βερενίκειος (62) and Βερενίκης (7), wrote Beroniceo, the late form transmitted by the MSS: see Marinone (1997) 15, 88; Kiss (2013) ad loc. Trappes-Lomax (2007) prefers Bereniceio (14–15, 208).

⁷For Conon, see below, n. 15. For his relationship with Archimedes, see Netz (2009) 179.

⁸The ‘misdirection’ and the purpose of the opening lines are recognised by Harder (2012) 2.801.

⁹I have based this translation on Netz (2009) 179. His important discussion of Greek mathematics is extremely helpful in illuminating the way Conon worked.

¹⁰Harder (2012) 1.293; 2.802.

¹¹For defence of despexit, see Barrett (1982); for arguments against see Marinone (1997) 79; Thomson (1997) 450. McKie (2009) 13–9 argues against both despexit and dispexit and for Bentley’s descripsit. But descripsit seems difficult without radio and appears to imply that Conon defined all the constellations, which he did not.

¹²See Maltby (1991) 396; and for related etymologies of ὅρος, see Massimilla (2010) 467–8; Harder (2012) 2.802.

¹³For the importance of observation in ancient astronomy see Lehoux (2007) 55–69; it was used in conjunction with higher-level geometry. Catullus may be reflecting Roman prejudices: the Greeks held mathematici in the greatest esteem but the Romans valued its practical application rather than its more abstract manifestations in geometry: see Cic. Tusc. 1.5 in summo apud illos [i.e. the Greeks] honore geometria fuit, itaque nihil mathematicis inlustrius; at nos metiendi ratiocinandique utilitate huius artis terminauimus modum; De or. 1.10 quis ignorat, ei, qui mathematici uocantur, quanta in obscuritate rerum et quam recondita in arte et multiplici subtilique uersentur?

¹⁴See Syndikus (1990) 202–3.

¹⁵For information on Conon, see Fraser (1972) 1.400–1; 2.580–1 nn. 188–96; 582 n. 203; 584 n. 221; Marinone (1990) 101–5; Marinone (1997) 85–6; Massimilla (2000) 467–9; Harder (2012) 2.804.

¹⁶See [Probus] on Virg. Ecl. 3.40 Conon mathematicus, Aegyptius natione, Ptolemaeo adsiduus, libros de astrologia VII reliquit; Schol. Bern. on Virg. Ecl. 3.40 Conon, Samius genere, mathematicus, stellarum peritissimus magister, uel ut quidam, Alexandrinus, qui cum Ptolemaeo fuit. Also the scholia to Aratus 146; Hygin. Astron. 2.24.1.

¹⁷See Hölbl (2001) 108; Pfeiffer (2004) 131–5; 140–4; 249–57; Hannah (2005) 88–91.

¹⁸See Syndikus (1990) 203; Harder (2012) 2.801.

¹⁹The myth took various forms and the location of their meetings was variously given: see the scholia to Apollonius 4.57–8, where it is reported that Sappho (199 L-P) and Nicander in his Europeia (F 6–7; 24 G–S) told the story of their love. furtim primarily suggests concealment but it also hints at the topos of furtiuus amor (see Maltby (2002) 158; Moreno Soldevila (2011) 54–5). Adverbs in -tim were common in early Latin, and furtim here has an elevated and slightly archaic flavour: see Woodman (2016) 129.

²⁰Cf. Lucian, Dial.deor. 233.1 ὦ τέκνον Ἔρως … τὴν Σελήνην καθαιρεῖς.

²¹Ovid Ars am. 3.83 Latmius Endymion, Nonn. Dion. 4.196 Λατμιὰς … Σελήνη, 48.668 Λάτμιον Ἐνδυμίωνα, Cic. Tusc. 1.92 Endymion … in Latmo … qui est mons Cariae; Paus. 5.1.5 (the shrine of Endymion on Latmos); Marinone (1997) 83–4.

²²See Acosta-Hughes (2010) 80–1; Harder (2012) 2.802.

²³Etym. magn. 153.6–9 (reporting Nicander in the Aitolika where the myth is not set in Latmos) τῆς Σελήνης τῷ Ἐνδυμίωνι συγκαθευδούσης, συνέβαινε τοὺς ἄλλους τόπους ἀσελήνους εἶναι. The scholia to Apollonius 4.60 explain: σκοτίῃ ἐνὶ νυκτί as ἐν ἀσελήνῳ; and φιλότητος as τῆς τοῦ Ἐνδυμίωνος διὰ τῶν σῶν ἐπῳδῶν (i.e. Selene’s love (cf. dulcis Amor) was caused by the incantations of Medea).

²⁴For the use of mythology by the mathematical writers, see briefly Netz (2009) 56–7, where he quotes the opening of a work of Eratosthenes addressed to ‘king Ptolemy’.

²⁵E. g. the scholia to Apollonius 4.57–8 Ἐνδυμίων πρῶτος ἐπεχείρησε τῇ περὶ τὰ μετέωρα φιλοσοφίᾳ, παρασχεῖν δὲ αὐτῷ τὰς ἀφορμὰς τὴν σελήνην ἔν τε φωτισμοῖς καὶ κινήσεσιν (‘Endymion was the first to undertake the study of astronomical phenomena and took as his starting points the moon in its phases and movements’); Plin. Nat. 2.43. Cf. Cairns (2016) 90–1.

²⁶See Marinone (1997) 85–9; Massimilla (2010) 468–70; and Harder (2012) 2.802–3, 806–7, who thinks that the scholiast is quoting ‘Callimachus only very loosely’ and from memory.

²⁷See Massimilla (2010) 155; 469: he supports his conjecture by reference to Callim. epigr. 55. 2–3 (Pf. = 16 HE = AP 6.148) ἔθηκε θεῷ | εὐξαμένα. His proposal is more persuasive than that of McKie (2009) 142–5 (ὃν κείνη πίστιν ἔθηκε θεοῖς).

²⁸Massimilla’s suggestion also sits very neatly with Ps. Nonn. Hist. I ad Greg. Naz. Or. V (In Iulian. II) 5. 10–11 (Nimmo Smith) ηὔξατο ὅτι, εἰ ὑποστρέψει ἄτρωτος, τὸν πλόκαμον ἀποκαρεῖσα τὸν ἑαυτῆς, ἀναθήσει ἀνάθημα ἐν τῷ ἱερῷ (‘[Berenice] vowed that, if he returned unharmed, she would cut off her lock and would dedicate it in the temple’).

²⁹For discussion, see Marinone (1997) 86–8; for details of conjectures and manuscript readings, see Kiss (2013). Both limine (Heinsius) and lumine (Vossius) are also found in a few fifteenth-century manuscripts. Trappes-Lomax (2007) 208 argues unpersuasively for in numine. For in limine, compare Virgil Ecl. 5.56–7 miratur limen Olympi | sub pedibusque uidet nubes et sidera Daphnis; and Catullus 66.69 me … premunt uestigia diuum.

³⁰fulgentem clare seems to stand in conflict with the scholia to Germanicus (72.19 Breysig) stellae obscurae septem, quae uocantur crines Berenices Εὐεργέτιδος, and Eratosth. [Cat.] 12 ἀμαυροὶ ἑπτά, οἵ καλοῦνται Πλόκαμοι Βερενίκης Εὐεργέτιδος (cf. Ptol. Alm. 7.5). For the astronomical detail, see Pàmias i Massana-Zucker (2013) 39; 188–9; Oppen de Ruiter (2015) 73–5; 182–3 nn. 4–14.

³¹See Lanzara (1990) 85.

³²A sixteenth-century manuscript has deorum which is adopted by Marinone (1997): see Kiss (2013). The main argument against cunctis deorum, which is not fatal, is the rarity of the construction: see McKie (2009) 140–2. McKie proposes uotis illa deorum. cunctis deorum is accepted most recently by Ramírez de Verger (2005). Marinone (1997) discusses all earlier proposals and reads multis illa deorum. For cunctus, see Adams (1973) 129–31; Malloch (2013) 155; and 327 (defending cunctisque ciuium).

³³So it happened according to Hyginus (Astron. 2.24.1 eumque [i.e. crinem] postero die non comparuisse. quod factum cum rex aegre ferret, … Conon mathematicus cupiens inire gratiam regis dixit crinem inter sidera uideri collocatum et quasdam uacuas a figura septem stellas ostendit quas esse fingeret crinem). Catullus was probably familiar with some similar account. For this use of idem, see OLD 10; Woodman (2016) 58; 297.

³⁴See Marinone (1997) 89; Maltby (1997) 94.

³⁵See TLL 10.1.2548.28–37, where there are further examples of polliceor from Christian writers. It cannot mean ‘to hold forth’ in the sense of ‘to offer’, as suggested by Oppen de Ruiter (2015) 77.

³⁶F.110a Harder, from the Callimachean Diegesis. For this type of material, see Cameron (2004); Dickey (2007) 66; and Harder (2012) 1.68–70.

³⁷protendens bracchia is the standard prayer gesture: see Nisbet-Rudd (2004) 263; Horsfall (2006) 160.

³⁸The significance of the sacrifice of the lock of hair has been the subject of much discussion: see Marinone (1997) 19–21; Harder (2012) 2.798–9, 803–4; Oppen de Ruiter (2015) 86–7, 187–8 nn. 71–83, with extensive bibliography. It was a traditional offering, in a wide variety of contexts: one of the most pertinent antecedents is Iliad 23.140–51, where Achilles refers to a vow made by his father Peleus that Achilles would cut a lock of hair and sacrifice it to the river-god of his homeland along with animal sacrifice in the event of his safe return home. It may be that the combination of coniuge and caro corpore (32) in close proximity should remind the reader of Calvus Fr. 6 Courtney = 31 Hollis [Ceres] cara iugauit | corpora conubiis.

³⁹For the archaism, see Skutsch (1985) 211. Cic. Att. 15.11(= 389 SB).4 etenim erat absurdum, quae si stetisset res publica uouissem, ea me euersa illa uota dissoluere, suggests that such formulations were standard and familiar.

⁴⁰That is ‘when he had returned victorious’: for damnatus uoti, see Oakley (1998) 2.268.

⁴¹See OLD pono 8c; TLL 10.1.2644.65–2645.35.

⁴²For non compareo, see OLD compareo 2; TLL 3.2010.10–62.

⁴³See Le Bœuffle (1983) xxxii–xliii; Viré (1992) iii–iv; Winterbottom (1994) 50; Reeve (2011) 375–6, who argues that, unless very compelling arguments to the contrary are adduced, the identification should be accepted; Pàmias i Massana-Zucker (2013) lxxxvii. But not all are convinced, and the Fabulae attributed to him appear to have been abbreviated: see Kaster (1995) 208; Boriaud (1997) vii–xiii; Cameron (2004) 33–45; B. M. Levick and T. J. Cornell, in Cornell (2013) 1.476 with n. 15.

⁴⁴See Le Bœuffle (1983) ix–xviii; Pàmias i Massana-Zucker (2013) lxxxvii–xciii.

⁴⁵See Kidd (1997) 44–5; Cameron (2004) 33–4, 38–41, 103–4; Pàmias i Massana-Zucker (2013) xx–xxi.

⁴⁶It appears from his final sentence (2.24.2) that Eratosthenes gave a different account: see Geus (2002) 211–23; Pàmias i Massana-Zucker (2013) 39–40, 188–90.

⁴⁷See Cameron (2004) 65, 183; cf. 52–6. That Hyginus is writing in the same tradition of prose summaries as the Diegesis is clear from the similarity to the much later Ps. Nonn. Hist. I ad Greg. Naz. Or. V (In Iulian. II) 5. 7–16 (Nimmo Smith): ‘A certain Beronice was the wife of the Ptolemy called “The Benefactor” in Alexandria. So, when her husband, Ptolemy, was at war, she vowed that, if he returned unharmed, she would cut off a lock (of her hair) and dedicate it in the temple (ηὔξατο ὅτι, εἰ ὑποστρέψει ἄτρωτος, τὸν πλόκαμον ἀποκαρεῖσα τὸν ἑαυτῆς, ἀναθήσει ἀνάθημα ἐν τῷ ἱερῷ); Beronice then made the dedication. A certain Comon (i.e. Conon) was an astronomer (ἀστρονόμος) in her time and in flattery of her (πρὸς κολακείαν αὐτῆς) he said that the gods had placed (ἀνέθηκαν) this lock among the stars. And now there is, indeed, a cluster-shaped arrangement of stars (τις βοτρυοειδὴς θέσις ἀστέρων) in the sky which they call the Lock of Beronice.’ For earlier views, see Marinone (1997) 22 n. 28. Cameron (2004) 40 is noncommittal about the date of Hyginus: ‘no later than second century’. For Hyginus’ use of Latin translations, see Cameron (2004) 40–1. For what is known of the exegetical material on Callimachus, and bibliography, see Harder (2012) 1.68–70.

⁴⁸For Aphrodite Arsinoe, see Gutzwiller (1992); Stephens (2004); Barbantani (2005); Bing (2009) 234–52; Carney (2013) 98–100.

⁴⁹See Fraser (1972) 1.239–40; 2.1023–5; Massimilla (2010) 480–1. Harder (2012) continues to favour the more complicated reconstruction of Pfeiffer (1965) 2.805–6, 817, 821–4, 828–9.

⁵⁰In Posidippus 119 A–B (= 12 HE) Aphrodite Arsinoe has the epithet Philadelphus; and Arsinoe Philadelphus appears in warlike guise, with spear and shield, in Posidippus 36 A–B.

⁵¹For clear discussion of the hiatus issue and the arguments in favour of this solution (first proposed by Avantius 1495), see McKie (2009) 145–9. Goold (1983) reads auctatus. While that deals with the hiatus, it is no improvement in terms of sense.

⁵²See OLD augeo 6c: ‘made happy, “blessed” (with a marriage or children; also, by an omen)’. If that is the meaning, it is difficult to see how it explains why he had left for war. It is difficult to take it as concessive: ‘although blessed with a recent marriage’. The basic sense is also strong when it is used as an adjective: see OLD auctus¹. Some have seen a reference to the restoration of Cyrene to Egyptian control (see Marinone (1997) 95); but that does not fit with the contemporary evidence (245 BC) of the Adulis inscription (OGIS 54), which represents Cyrene as inherited by Ptolemy III from his father.

⁵³Contrast Ov. Fast. 3.601–2 iam pius Aeneas regno nataque Latini | auctus erat populos miscueratque duos.

⁵⁴It is the reading adopted and well defended by Marinone (1997). It was advocated by Goold (1958) 107 and printed in his 1973 text. Goold later preferred auctatus, supported by Trappes-Lomax (2007) 209; but arguments against are well put by McKie (2009) 147.

⁵⁵Pace Fordyce (1961) who says ‘it hardly fits this context’. McKie (2009) 146–7 admits ‘there is little to fault in the reading’.

⁵⁶Cf. Grainger (2010) 158. That Ptolemy went by ship to try and protect his threatened sister is consistent with Justin (27.1.6 aduolat; cf. 26.3.4 secundante uento celeriter Cyrenas aduolasset); and it is confirmed by the Gurob papyrus (BNJ 160): for full discussion of the problems raised by this document (with bibliography) see Gambetti (2011).

⁵⁷The verb (auehi) is often used of ‘departing’ without specification of the means of transport, which is usually clear from context and can as readily be assumed to be by ship (as here) as by horse or chariot: see TLL 2.1303.63–1304.58: cf. e.g. Plaut. Cist. 579–81; Virg. Aen. 2.43.

⁵⁸For qua tempestate, see Cic. De Or. 3.153; Goodyear (1972) 115–16; most editors prefer iuerat to the transmitted ierat, which is defended by Marinone (1997) 96, who compares, for the long i, Plaut. Amph. 401 and Ter. Ad. 27 and takes the form as an archaism, which would be appropriate after qua … tempestate; for ire with the supine to express purpose, see Wackernagel (1920) 1.278–80; Horsfall (2008) 133. For this ‘background’ use of the pluperfect, cf. Kraus (1994) 95–6; Oakley (1997) 1.127.

⁵⁹See OLD s.v. 2.

⁶⁰This is now further confirmed by the decree from Alexandria (243 BC): see El-Masry–Altenmüller–Thissen (2012) 164–7.

⁶¹Note especially Cic. Rep. 3.24 illa laus in summorum imperatorum incisa monimentis finis imperii propagauit (introductory text uncertain); and, e.g., Woodman (2014) 16. For similar praise of Ptolemy II, see Theoc. Id. 17.86–94; with Hunter (2003) 159–70.

⁶²The decree from Alexandria (243 BC) supports the claim that he went at least as far as Susa: see Altenmüller (2010); El-Masry–Altenmüller–Thissen (2012) 101–2, 162.

⁶³Gambetti (2011) provides a full introduction to and commentary on the Gurob Papyrus, with full discussion of problems and bibliography. For earlier accounts of the Third Syrian War, see especially Will (1979) 1.248–54; Hölbl (2001) 48–51; Huß (2001) 338–54. More recent discussions are Grainger (2010) 133–6, 149–52, 153–70; Altenmüller (2010); El-Masry–Altenmüller–Thissen (2012), especially 151–67.

⁶⁴See Gambetti (2011) n. 6.

⁶⁵See SEG 42.994 (A.7–8) βασιλέως Ἀντιόχου | τοῦ ἐγ βᾳσιλέως Ἀντ|[ιό]χου καὶ βασιλίσσης Βερενίκης, A.7–8; cf. D. 10–13.

⁶⁶For the Third Syrian War as revenge for the murder of Berenice and her son Antiochus, see Justin 27.1.8–9 universae civitates … in ultionem eius, quam defensuri fuerant, Ptolomeo se tradunt; 27.3.4 Ptolomeus, rex Aegypti, sub specie sororiae ultionis Asiae inhiabat; Polybius 5.58.10; Appian Syr. 65.

⁶⁷Also in El-Masry–Altenmüller–Thissen (2012) 155–9: it appears (p. 156) that Xanthippus arrived from Egypt in mid-January 245.

⁶⁸As also claimed in the Adulis inscription (OGIS 54) but previously often doubted: see Altenmüller (2010); El-Masry–Altenmüller–Thissen (2012) 101–2, 162.

⁶⁹Seleucus II is recognised as king in Babylon from mid-July 245 BC. It seems reasonable to suppose that Ptolemy had left Syria by then: see Altenmüller (2010) 30. But peace was not concluded until 242/1: see Grainger (2010) 167–70.

⁷⁰Justin 27.1.9 qui nisi in Aegyptum domestica seditione reuocatus esset, totum regnum Seleuci occupasset; Porphyrios (F 43) [Ptolemy] audisset in Aegypto seditionem moueri …; Cf. P Haun 6 14–15 Ε]υφρατου … καὶ εἰ μὴ τότε Αἰγυπτίων ἀπόσ[τασις ἐγένετο] … ἐπανῆλθεν εἰς Ἀλεξά[νδρειαν, if correctly restored: see Bülow-Jacobsen (1979). For discussion of the seditio with bibliography, see Huß (2001) 373–5; Grainger (2010) 163–4; Clayman (2014) 128, 214 nn. 34–5; Oppen de Ruiter (2015) 76, 184 n. 26: the details are obscure and interpretation is difficult.

⁷¹The most important parallels are Hedylus 2.4–6 HE (AP 5.199) παρθενίων … λάφυρα πόθων | … μαστῶν ἐκδύματα, μίτραι, (cf. uirgineis … exuuiis); | … σκυλμῶν … μαρτύρια (cf. uestigia rixae); Agathias AP 5.294. 17–18 τὰ λάφυρα … | σύμβολον ἐννυχίης … ἀεθλοσύνης (cf. exuuiis and nocturnae uestigia rixae). Hedylus (probably of Samos) was a contemporary of Callimachus, and another of his epigrams (4 HE) is about a dedication in the temple of Aphrodite-Arsinoe. Agathias wrote much later (sixth century AD) but was very familiar with Callimachus: his poem goes on to talk in military language about taking the girl’s virginity. In both poets the spoils are to be dedicated to Aphrodite (in Agathias AP 5.294.24 Κύπρι τροπαιοφόρε). See Massimilla (2010) 470 for further parallels; and Harder (2012) 2.807–9 for a careful discussion.

⁷²For good discussion of 66.13–18, see McKie (2009) 149–55. He cites Claudian 14.29 nocturni referens uulnera proelii in support of this interpretation of nocturnae portans uestigia rixae. See also Nisbet-Hubbard (1970) 89.

⁷³See especially Catull. 62.24 (quid faciunt hostes capta crudelius urbe?); 59 (et tu ne pugna cum tali coniuge, uirgo). For militia amoris, see now Soldevila (2011) 275–86.

⁷⁴The reference is probably to the zona uirginea, for which see Catull. 67.27 with Portuese (2013) 240–1; cf. 2.13; and 61.52–3 with Fedeli (1983) 50. See also Adams (1982) 196 (on nodum uirginitatis eripere); Fedeli (2005) 433 (exuuiae); cf. Ovid Her. 5.140–3 (spolium uirginitatis); 17.114 (spolium … pudoris).

⁷⁵For the language, see Navarro Antolín (1996) 317–8, 403–4.

⁷⁶For various suggestions, see Laronde (1987) 382; Ogden (1999) 80–1; Hölbl (2001) 46; Huß (2001) 333–5; Criscuolo (2003) 325–6; Clayman (2014) 38–9; Oppen de Ruiter (2015) 30–3.

⁷⁷See Huß (2001) 339.

⁷⁸If the wedding took place in July 246, it is perhaps worth noting that this was the month in which there were a number religious ceremonies which served as preludes to the rising of Sirius (21/2 July), which was associated with the rise of the Nile: see Pfeiffer (2004) 121–31. By 238 BC there were regular celebrations of the Theoi Euergetai on the 5^th, 9^th and 25^th of each month; those on the 5^th and 25^th marked the birthdays of the King and Queen respectively, as is now known from the Alexandria Decree; possibly that on the 9^th celebrated their wedding (p. 124). If, as generally thought, the seditio domestica that caused Ptolemy to end his personal campaign was caused by a flood and poor harvest in 246/45 BC (see Huß (2001) 373–5), it may have been held that his departure before the rising of Sirius had been in some way responsible. But so much is uncertain that this can only be speculation.

⁷⁹For the text of Catullus 66.77–8, see Ramírez de Verger (2005); Trappes-Lomax (2007) 216–17; Kiss (2013); Courtney (2013) 209. It seems to me that Catullus’ text should be amended to convey the sense of Callimachus: precisely how best to do that is less clear.

⁸⁰[Pomp. Trogus] prol. 26. ut frater Antigoni Demetrius occupato Cyrenis regno interiit (‘how the (half-)brother of Antigonus, Demetrius, having seized the kingship at Cyrene, perished’): Plut. Demetrius 53.3 τὸν … ἄρξαντα Κυρήνης ἐκ Πτολεμαίδος (‘the ruler of Cyrene [son] of Ptolemais [daughter of Ptolemy I Soter]’). Demetrius the Fair was the half-brother of Antigonus II Gonatas and the son of Demetrius Poliorcetes and Ptolemaïs, daughter of Ptolemy I Soter.

⁸¹Paus. 1.7.3 Μάγας δὲ ἤδη γυναῖκα ἔχων Ἀπάμην Ἀντιόχου τοῦ Σελεύκου Θυγατέρα; see Oppen de Ruiter (2015) 11.

⁸²Apama often appears also as Apame in modern accounts. For what is known of her, see Clayman (2014) 35–9; Oppen de Ruiter (2015) 10–13.

⁸³Harder (2012) 1.198 F.54.2 νύμφα, κασιγνήτων ἱερὸν αἷμα θεῶν, with scholia ad loc. 1.218 (= F. 60d) θυγάτηρ τῶν θεῶν ἀδελφῶν, οἱ εἰσιν Πτολεμαῖος καὶ Ἀρσινόη ὧν ἀνηγόρευον τὴν Βερενίκην. For examples of the official nomenclature, see Huß (2001) 343 n. 38; Pfeiffer (2004) 10, with nn. 191–2; Massimilla (2010) 475; Harder (2012), 2.397–8; El-Masry–Altenmüller–Thissen (2012) 218 (22–4); 219 (31–2); Oppen de Ruiter (2015) 10–11, 175 nn. 90, 91.

⁸⁴For these various hallmarks see e.g. Woodman (2014) 288, 294; (2016) 319 s.v. ‘tyrant’.

⁸⁵And that is its meaning here: see Ulp. dig. 38.10.6.1 socri et socrus appellatione sponsorum parentes contineri uidentur (‘the parents of the betrothed are deemed to be included under the designation father-in-law and mother-in-law’).

⁸⁶See Clayman (2014) 38, 197 n. 88.

⁸⁷For the terms, cf. Papin. Dig.48.5.6.1 proprie adulterium in nupta commititur … stuprum uero in uirginem uiduamue committitur. But both words here are used primarily to convey moral disapproval and to mark the behaviour as that of a tyrant. For stuprum, see Adams (1982) 200–1, 223; Woodman (2016) 88–9.

⁸⁸I follow Goold (1983) and Ramírez de Verger (2005) in reading quo in both 27 and 28: that the syntax of the two ablatives is different seems to be outweighed by the gain in rhetorical effectiveness; and fortius in 28. But, like Trappes-Lomax (2007) 211, I prefer Owen’s nil to the pointless awkwardness of non. For full details, see Kiss (2013). With the encomiastic comparison in 28, compare Theoc. Id. 17.15 (of Ptolemy II) βουλάν, ἅν οὐκ ἄλλος ἀνὴρ οἷός τε νοῆσαι (‘a plan such as no other man could have devised’).

⁸⁹An interesting variation on the phrase at 62.57 par conubium … adepta est (see Agnesini (2007) 339; Ter. Andr. 332). The contrast between par and regium arguably supports the need to amend auctus (11), if it also implies that Ptolemy had the higher status.

⁹⁰It probably occurred in Ennius: see Christenson (2000) 183; Skutsch (1985) 685–6; Norden (1957) 222–3; Horsfall (2013) 447.

⁹¹The facinus of Cloelia (Livy 2.13.6–11) is perhaps the best known example of uirtus uirginis. According to Livy (2.13.11), the Romans honoured the exceptional courage she exhibited as a uirgo with an equestrian statue: Romani nouam in femina uirtutem nouo genere honoris, statua equestri, donauere; in summa Sacra uia fuit posita uirgo insidens equo.

⁹²It is generally assumed that Callimachus had μεγάθυμος in the equivalent text: Hyginus may have taken his translation from Catullus (see Cameron (2004) 40–1). Athena has the epithet μεγάθυμος in Hom. Od. 8.520, 13.121: see Harder (2012) 2.810. Callimachus and Catull. 66.77–8 recall Pallas Athena as described in Callimachus, Hymn 5.5–32, when she is cleansing herself from the dust and gore of the battle with the Giants, in which she had fought in defence of her father, Zeus: see Koenen (1993) 107–9. This would have pointed relevance if Hyginus’ explanation of the bonum facinus were the right one. See, for discussion of 77–8, Marinone (1997) 197–204; Massimilla (2010) 502; Harder (2012) 2.845. For the battle against the Giants, see Bulloch (1985) 118–19; for Athena’s use of unscented oils as a sign of manliness, see Xen. Sym. 2.3; Bulloch (1985) 124–5.

⁹³The combination is probably elevated, if not in itself archaic: on facinus, see Oakley (1998) on Livy 8.24.9. Contrast the similar language of Sall. Cat. 25.1 (Sempronia, quae multa saepe uirilis audaciae facinora commiserat).

⁹⁴The adjectives are also paired at Cic. Tusc. 4.61; Off. 1.63, 65, 88. The point is not affected by the choice between fortior and fortius.

⁹⁵For a full and interesting account of how Hyginus’ account came to be replaced by Justin’s as the key to understanding the reference in 66.25–8, see Benedetto (2008). Marinone (1997) 112–13 (who assumes that Hyginus is referring to Magas rather than Ptolemy II) sees a reference to both accounts in these lines but his suggestion has not found favour (but see Thomson (1997) 453).

⁹⁶As Justin tells the story, Berenice’s role is similar in some respects to that of Electra (save for her pietas in sparing her mother): cf. Vell. Pat. 1.1.3 hunc [Aegisthus] Orestes matremque socia consiliorum omnium sorore Electra, uirilis animi femina, obtruncat.

⁹⁷The contrast between nonnulli and alii refers to two different authorities or sets of authorities Such imprecise plurals are characteristic of scholia and related materials: see Dickey (2007) 110–12. They need not refer to more than a single source: see Lightfoot (1999) 248; Cameron (2004) 106; Skutsch (1985) 371. Here Callimachus is expressly included in the nonnulli; and then (with pro quo Callimachus etiam) also associated with the contrasting group denoted by alii. If the text is sound, then Hyginus seems to be saying that Callimachus and others wrote about Berenice rearing horses and sending them to the Games (a claim now partially vindicated by the Victoria Berenices (Aetia 3) and, probably, by Posidippus’ Hippika, especially 78–9; 81–2 A–B); and that others tell of her rescuing her father; the latter may not have been an episode treated explicitly by Callimachus but it was what he had in mind when he called Berenice magnanima. For very fragmentary scholia possibly relating to this episode, see Massimilla (2010) 149; 471–2.

⁹⁸For comparable stories, see Stephens (2005); Pillonel (2008); for plausibility, see Oppen de Ruiter (2015) 118. Note especially Val. Max. 9.10 ext. 1 (the revenge taken by Berenice, wife of Antiochus, on the assassin of her son).

⁹⁹See in particular Pfeiffer (1965) 1.321–2, where he suggests that F388 is related to this story; Hollis (1992); Chiesa (2009).

¹⁰⁰Probably 250/49 BC: see Grainger (2010) 148, who rightly emphasises (p. 146) that the date usually given for Magas’ death (250 BC) is only approximate. The basic discussion is Chamoux (1956); the key text is Agatharchides (BNJ 86 F 7): ‘In the sixteenth book of the European Histories Agatharchides says that Magas, who had been king of Cyrene for fifty years, was unwarlike and lived in luxury (τὸν Κυρήνης βασιλεύσαντα ἔτη πεντήκοντα ἀπολέμητον γενόμενον καὶ τρυφῶντα). At the end of his life he became excessively obese and was suffocated by his fat because of the idleness of his body and the mass of food he consumed’. Oppen de Ruiter (2015) 8–9 argues for 252/51 as the date of Magas’ death, and any date later than that seems to create an impossibly crowded timetable: it could even have been 253/52.

¹⁰¹Polybius (10.21.2–4) refers to the liberty of Cyrene being under threat; Plutarch (Phil. 2–3) mentions that Cyrene was full of ‘confusion and (political) fever’; see also Paus. 8.49.2. There was a brief period during which there was a koinon (a league of the city states) in Libya. Callimachus AP 13.7 (epigr. 37 Pf. = 17 HE) refers to fighting at Euesperides which may belong to this period. Solinus says that Euesperides was replaced by a new city named Berenice (27.54): hanc [sc. Berenicen ciuitatem] Berenice muniuit quae Ptolemaeo tertio fuit nupta et in maiori Syrti locauit. Much is unclear. See Hölbl (2001) 46; Clayman (2014) 37–40; Oppen de Ruiter (2015) 26–7.

¹⁰²For Berenice as participant in the Games, see Harder (2012) 2.389. For Berenice II as the laudanda in these poems, see Bastianini-Gallazzi (2001) 205–11; Huß (2008); Massimilla (2010) 223, 227, 237; Clayman (2014) 147–58. Posidippus 79A–B.1 (παρθένος ἡ βασίλισσα) and 82A–B.4–6 (παῖδα … σὺν πατρὶ … μόνη βασιλίς) seem strikingly similar to a parua uirgine and to support the view that Berenice reigned, at least briefly, alone as Basilissa in Cyrene, supported by Ptolemy II. The alternative is Berenice, wife of Antiochus II and full sister of Ptolemy III, for whom the case is made by Criscuolo (2003) 328–31 and Thompson (2005).