The last carouse marks the onset of Alexander’s terminal illness; it was seen in some way to have brought about his death. This article examines how those few hours were transformed in memory and written text. It may be read as a commentary on the five words quoted above (from Arrian’s rephrasing of the Ephemerides), on what they fail to say, and on why there was so little to say.
Certain earlier banquets at Alexander’s court have become set-pieces in ancient and modern narratives: this one, more consequential than any other, usually slips by in a few words. The silence has been occasionally broken. Fifty years ago Schachermeyr, in Alexander in Babylon, gave a bird’s-eye view of Alexander’s movements on this day, while Bosworth’s ‘The death of Alexander the Great’ focussed on the rumours that followed. In 1988 Heckel explored the origins of the Liber de morte Alexandri Magni; Bosworth again, in a chapter in From Arrian to Alexander, redirected attention to the Ephemerides, and in the same year his Conquest and Empire allowed more space to this incident than any other recent Alexander biography.Footnote 1 Since then much has been written about the historical context and about each of the sources, and a new analysis is needed.
THE NARRATIVE SOURCES
Sources listed in this initial paragraph will reappear below with precise references. Several surviving narratives, written long after Alexander’s time, are of use: through them primary sources are fitfully visible. Diodorus Siculus (first century b.c.) named no source here. Quintus Curtius’ narrative of this event (first century a.d.) is lost in a lacuna, though his discussion of the question whether Alexander was poisoned survives. Both authors cited Cleitarchus (third century b.c.) elsewhere, and almost certainly used him here.Footnote 2 Justin (second/fourth century a.d.) abbreviated Trogus’ Philippic Histories (first century b.c.), whose ultimate sources are not named. Plutarch’s Life of Alexander (c. 100 a.d.) cites Aristobulus and the Ephemerides, and implies knowledge of other accounts. Arrian’s Anabasis of Alexander (mid second century a.d.) generally relies on the memoirs of Ptolemy and Aristobulus, contemporaries of Alexander, but on this event cites only the Ephemerides, alluding to other unnamed authors. The Itinerarium Alexandri (fourth century a.d.), largely based on Arrian, adds details from some other source. The Liber de morte Alexandri Magni and Alexander Romance insert into their narratives a list of those who drank with Alexander; the former enigmatically mentions Onesicritus, early author of an Alexander memoir. Additionally two passages in the Deipnosophists of Athenaeus (third century a.d.) quote Nicobule, an almost unknown writer, on Alexander’s last dinner and last carouse.
Texts quoted in this section are numbered [1] to [11] for quick reference. All translations are mine. First, narratives written soon after the event.
[1] καὶ αἱ βασίλϵιοι ἐϕημϵρίδϵς ὧδϵ ἔχουσιν· πίνϵιν παρὰ Μηδίῳ αὐτὸν κωμάσαντα· ἔπϵιτα ἐξαναστάντα καὶ λουσάμϵνον καθϵύδϵιν τϵ καὶ αὖθις δϵιπνϵῖν παρὰ Μηδίῳ καὶ αὖθις πίνϵιν πόρρω τῶν νυκτῶν· ἀπαλλαχθέντα δὲ τοῦ πότου λούσασθαι· καὶ λουσάμϵνον ὀλίγον τι ἐμϕαγϵῖν καὶ καθϵύδϵιν αὐτοῦ, ὅτι ἤδη ἐπύρϵσσϵν.
And the royal Ephemerides have it thus: that after carousing he drank with Medius; then, withdrawing and bathing, slept, and again dined with Medius, and again drank the nights through; after leaving the drinking party took a bath; after the bath ate a little and slept there, because he was already feverish. (Arr. 7.25.1 = BNJ 117 F 3a)
Arrian never quoted verbatim: we have what he chose to give us, rephrased here in oratio obliqua. Footnote 3 His use of the Ephemerides on Alexander’s last days can usefully be compared with that of Plutarch (Alex. 76 = BNJ 117 F 3b), who quoted selected extracts verbatim, but beginning on the following day. The expression translated ‘drank the nights through’ was surely not in the original: it was inserted here by Arrian to recall his own previous use of this Platonic phrase (see [10]).
Working from Plutarch’s quotation it is seen that Alexander’s last carouse and the drinking with Medius occurred on the night of 17th/18th Daesius.Footnote 4
Each scholar takes a view on whether daybooks were kept at Alexander’s court, whether the surviving Ephemerides fragments stem from a genuine document or from forgery, and why the text was published. Eumenes of Cardia, named as author, served as Alexander’s secretary and kept his archives; he fought among the Successors, though despised as the only Greek; he forged documents.Footnote 5 He could have published the text after Alexander’s death; he could have made a selection; he could have falsified it.Footnote 6
Plutarch elsewhere alluded to ‘the royal Ephemerides, in which is written repetitiously and very frequently that “this day [Alexander] slept after the drinking party”, sometimes “and the next”’.Footnote 7 It has been argued that a genuine daybook would not insist on drinking parties, but this argument is weak. In the fragments that survive there are records of sacrifices and of discussions with commanders, as would be expected. If the king dined ceremonially, or dined out, that fact required recording. As to sleeping it off, if the irregularity of the king’s daily life affected the conducting of court events it would have been reasonable to note this.Footnote 8
In any government that keeps records a daybook will eventually be wanted as aide-memoire and key to the archives. This text is in the style of such a document, and the few words directly relevant here represent what a secretary would record. If something is lost in Eumenes’ editing and Arrian’s rewriting, if the text therefore embodies a point of view, it remains an early source on the same footing as the others. They contribute to our understanding of it; it contributes to our understanding of them.
[2] Νικοβούλη δέ ϕησιν ὅτι παρὰ τὸ δϵῖπνον πάντϵς οἱ ἀγωνισταὶ ἐσπούδαζον τέρπϵιν τὸν βασιλέα· καὶ ὅτι ἐν τῷ τϵλϵυταίῳ δϵίπνῳ αὐτὸς ὁ Ἀλέξανδρος ἐπϵισόδιόν τι μνημονϵύσας ἐκ τῆς εὐριπίδου Ἀνδρομέδας ἠγωνίσατο καὶ τὸν ἄκρατον προθύμως προπίνων καὶ τοὺς ἄλλους ἠνάγκαζϵν.
Nicobule says that during dinner all the performers strove to amuse the king, and that at the last dinner Alexander himself, recalling an episode from Euripides’ Andromeda, performed it and, pledging eagerly in unmixed wine, compelled the others [to reciprocate].
(Ath. Deipn. 537d = BNJ 127 F 2)[3] Νικοβούλη δὲ ἢ ὁ ἀναθϵὶς ταύτῃ τὰ συγγράμματά ϕησιν ὅτι παρὰ Μηδϵίῳ τῷ Θϵσσαλῷ δϵιπνῶν ὁ Ἀλέξανδρος ϵἴκοσιν οὖσιν ἐν τῷ συμποσίῳ πᾶσι προὔπιϵ, παρὰ πάντων τὰ ἴσα λαμβάνων, καὶ ἀναστὰς (ἐκ) τοῦ συμποσίου μϵτ᾿ οὐ πολὺ ἀνϵπαύϵτο.
Nicobule, or the one who ascribed that text to her,Footnote 9 says that Alexander, when he dined with Medius of Thessaly, pledged every one of the twenty in the symposium, received the same from each, and, having risen from the symposium, soon afterwards went to sleep.
(Ath. Deipn. 434c = BNJ 127 F 1)Footnote 10In addition to these citations by Athenaeus Nicobule is mentioned once by Pliny.Footnote 11 Nothing else can be said of her. Many scholars have imagined that her work was an attack on Alexander: Gregor Weber’s view, that ‘in reports such as those of Nicobule one can discern the essence of royal daily life’, is closer to reality.Footnote 12 The two fragments are not phrased as verbatim quotations; it has been plausibly suggested that Athenaeus is summarizing a longer text.Footnote 13
A fragment by Ephippus of Olynthus,Footnote 14 placed just before Nicobule [3] by Athenaeus, is often drawn into modern narratives: ‘Alexander once asked for a two-chous cup [ποτήριον δίχουν] and drank from it, pledging Proteas, who took the cup, praised the king fulsomely and drank it off. A little later Proteas asked for the same cup and drank from it, pledging the king. Alexander took it and drank bravely, but could not finish and lay back on the pillow, letting the cup fall from his hands, and, sickening thenceforth, he died, because, says [Ephippus], Dionysus was angry with him for having captured and destroyed Thebes, his native city’.Footnote 15 The pace has suddenly changed: this last sentence is a summary by Athenaeus of Ephippus’ theme.Footnote 16 Lionel Pearson sawFootnote 17 what others have missed,Footnote 18 that this is an undated anecdote (ποτέ ‘once upon a time’) not connected to the last carouse.Footnote 19 Proteas is not otherwise listed as present at Medius’ party. The competitive pledging or toasting described by Nicobule is not easily combined with Ephippus’ anecdote into a single event, although, placed side by side in the Deipnosophists, they illustrate one another.Footnote 20
[4] Ἀριστόβουλος δέ ϕησιν αὐτὸν πυρέττοντα μανικῶς, διψήσαντα δὲ σφόδρα, πιϵῖν οἶνον· ἐκ τούτου δὲ φρϵνιτιᾶσαι καὶ τϵλϵυτῆσαι τριακάδι Δαισίου μηνός.
Aristobulus says that he had a raging fever, became extremely thirsty, drank wine, hence suffered a frenzy, and died on the last day of Daesius. (Plut. Alex. 75.6 = BNJ 139 F 59)
Plutarch’s citation of Aristobulus, like Arrian’s of the Ephemerides [1], is rewritten in oratio obliqua: the original might have employed the same words, but we cannot know whether it was longer and more informative.
After Alexander’s death Aristobulus became a citizen of Cassandreia, the city founded in 316 by Antipater’s son Cassander (to whom Alexander was anathema).Footnote 21 It has been suggested that Aristobulus served with Antipater and his son, helping to plan the new city:Footnote 22 in any case he was freer to honour Alexander’s memory after Cassander’s death.Footnote 23 His memoir, reflecting scientific interests, was indeed written in old age, after about 295,Footnote 24 probably using the already-published work of Chares (see below) as a source on court events.
Although Aristobulus’ Alexander narrative and Ptolemy’s are lost, they can be compared in those few cases where Arrian chose to cite both by name. Aristobulus, against both Plutarch’s source and Ptolemy, briefly told a third version of the court historian Callisthenes’ death, distasteful but not to be dismissed—the same version that Chares told more fully.Footnote 25 To Alexander’s sudden illness in Cilicia Ptolemy and others assigned a romantic cause (he swam in the ice-cold river Cydnus); Aristobulus diagnosed exhaustion.Footnote 26 At Gordium how did Alexander unfasten the historic cart? Ptolemy is not cited, but according to others Alexander drew his sword and, famously, cut the Gordian knot; according to Aristobulus he pulled out the pin.Footnote 27 There is a pattern: Aristobulus ‘did not indulge in rhetoric or exaggeration and did not aim at sensation’.Footnote 28 Quite the contrary: if he saw a heroic story approaching he walked the other way; yet his hero was Alexander.
[5] As he left, intending to sleep, Medius approached him, inviting him to join his own friends, who were, he said, the king’s closest friends. Persuaded by Medius’ invitation Alexander went to the symposium. Those present numbered twenty: Perdiccas, Meleager, Python, Leonnatus, Cassander [Asander?], Peucestes, Ptolemy, Lysimachus, Philip the doctor, Nearchus of Crete,Footnote 29 Stasanor, Heracleides of Thrace, Europius [Seleucus?], Ariston of Pharsalus, Philip the engineer, Philotas, Menander. Of these Perdiccas, Ptolemy, Olcias, Lysimachus, Eumenes and Asander did not know what was to happen. All the others were part of the plot, had conferred with Cassander and Iollas and exchanged oaths […]. Alexander reclined and Iollas brought him a cup […]. Suddenly Alexander gave a loud cry as if an arrow had pierced his liver.
(Historia Alexandri Magni 3.31)Footnote 30[6] eum diem festum agentem amicorum conuentum habere uellet. iam non alienum uidetur qui fuerint demonstrare, quorum Onesicritus fugiens simultatem mentionem facere noluit. fuit Perdiccas, Medius, Leonnatus […] quorum quid ageretur nemo fuit ignarus praeter Eumenem et Perdiccam et Ptolemaeum et Lysimachum et Asandrum et Holciam.
[Medius], celebrating a festival, wanted to gather his friends. Thus it seems not irrelevant to list those whom Onesicritus, fearing enmity, chose not to name: Perdiccas, Medius, Leonnatus […]. Of these none was ignorant of what would happen except Eumenes, Perdiccas, Ptolemy, Lysimachus, Asander and Holcias. (Liber de morte 97 = BNJ 134 F 37Footnote 31 )
In the case of the Alexander Romance (the tradition to which both texts are assigned) it is even more delicate to extract early material from its later context than it is with the preceding citations transmitted by classical authors. The Romance was already coalescing in the third century b.c., Footnote 32 though surviving recensions were written many centuries later; details of the present episode originated earlier still, very soon after Alexander’s death.Footnote 33 Quoted above is the Greek manuscript A with a variant from the Latin Liber de morte. A story had been told, in preceding sections of both, of a conspiracy to poison Alexander: its protagonists were Cassander, his brother Iollas who was Alexander’s cupbearer, and Medius of Thessaly who was Iollas’ lover. A theatrical competition was described; then follow the quoted texts. With help from other recensions the list of names can be rendered coherent, as reflected in my translation: the validity of the list will be discussed below.
Now come the surviving historical narratives.
[7] reuersus igitur Babyloniam multis diebus otio datis intermissum olim conuiuium sollemniter instituit; totusque in laetitiam effusus cum diei noctem peruigilem iunxisset, recedentem iam e convivio Medius Thessalus instaurata comisatione et ipsum et sodales eius inuitat. accepto poculo media potione repente uelut telo confixus ingemuit elatusque conuiuio semianimis tanto dolore cruciatus est, ut ferrum in remedia posceret.
He returned to Babylon and passed many days in amusements: he reverted to his interrupted habit of a regular banquet. Full of happiness he prolonged this into a wakeful night. He was already returning from the banquet when Medius the Thessalian, a carouse having begun, invited him and his friends. He was given a cup. While drinking he suddenly groaned, as if pierced by a spear; carried from the party half conscious, he was tortured by such severe pain that he demanded a sword to end it. (Just. 12.13.6–9)
[8] τούτων δὲ παρακϵλϵυομένων θυσίας ἐπιτϵλϵῖν τοῖς θϵοῖς μϵγαλοπρϵπϵῖς μϵτὰ πάσης σπουδῆς παρϵκλήθη πρός τινα τῶν ϕίλων Μήδιον τὸν Θϵτταλὸν ἐπὶ κῶμον ἐλθϵῖν· κἀκϵῖ πολὺν ἄκρατον ἐμφορηθϵὶς ἐπὶ τϵλϵυτῆς Ἡρακλέους μέγα ποτήριον πληρώσας ἐξέπιϵν. ἄϕνω δὲ ὥσπϵρ ὑπό τινος πληγῆς ἰσχυρᾶς πϵπληγμένος ἀνϵστέναξϵ μέγα βοήσας καὶ ὑπὸ τῶν φίλων ἀπηλλάττϵτο χϵιραγωγούμϵνος.
They advised him to make generous sacrifices to the gods with all speed. He was then invited by one of his friends, Medius the Thessalian, to come to a carouse. There, having already drunk much unmixed wine in honour of the death of Heracles he filled a large cup and drained it.Footnote 34 Suddenly he gave a great cry, as if struck by a violent blow, and his friends led him away by the hand. (Diod. Sic. 17.117.1–2)
[9] ἀποθέμϵνος τὸ πένθος αὖθις ἦν ἐν θυσίαις καὶ πότοις. ἑστιάσας δὲ λαμπρῶς τοὺς πϵρὶ Νέαρχον, ϵἶτα λουσάμϵνος ὥσπϵρ ϵἰώθϵι μέλλων καθϵύδϵιν, Μηδίου δϵηθέντος ᾤχϵτο κωμασόμϵνος πρὸς αὐτόν· κἀκϵῖ πιὼν ὅλην τὴν ἐπιοῦσαν ἡμέραν, ἤρξατο πυρέττϵιν, οὔτϵ σκύϕον Ἡρακλέους ἐκπιὼν οὔτ’ ἄφνω διαλγὴς γϵνόμϵνος τὸ μϵτάϕρϵνον ὥσπϵρ λόγχῃ πϵπληγώς …
Putting grief aside he was at sacrifices and drinking parties again. But after feasting Nearchus’ people splendidly, then bathing as usual before going to sleep, he went at Medius’ invitation to carouse with him, and there, drinking all the next day, he became feverish. So it was not that he drank dry the bowl of Heracles, nor that he suddenly felt pain in his back as if pierced by a spear … (Plut. Alex. 75.3–76.1)
[10] καὶ τϵθυκὼς τοῖς θϵοῖς τάς τϵ νομιζομένας θυσίας […] ϵὐωχϵῖτο ἅμα τοῖς φίλοις καὶ ἔπινϵ πόρρω τῶν νυκτῶν […]. ἀπὸ δὲ τοῦ πότου αὐτὸν μὲν ἀπαλλάττϵσθαι ἐθέλϵιν ἐπὶ κοιτῶνα ϵἰσὶν οἳ ἀνέγραψαν· Μήδιον δὲ αὐτῷ ἐντυχόντα, τῶν ἑταίρων ἐν τῷ τότϵ τὸν πιθανώτατον, δϵηθῆναι κωμάσαι παρὰ οἷ· γϵνέσθαι γὰρ ἂν ἡδὺν τὸν κῶμον.
Having made the usual sacrifices to the gods […] Alexander was celebrating alongside his friends and drank the nights through […]. There are those who write that he himself wanted to leave the drinking party for his bedroom, but that Medius, at that time the most trusted of his Companions, met him and asked him to carouse with him, saying that it would be a good party. (Arr. 7.24.4–25.1)
[11] morem sibi proceres exercitus fecerant, uti mutuo sese conuiuiis acciperent. enim cum forte apud Medium conuiuarent, comesatione sese illis conuiuam Alexander facit. statim denique Herculis Medius scyphum offert ad uina, neque is honorem dei nomine aspernatus poculum complet stringitque continuum.
The army commanders had made a habit of reciprocal invitations to banquets. On one occasion when they were banqueting at Medius’ Alexander joined them at the carouse. There suddenly Medius offered him the bowl of Hercules for his wine. Accepting the honour in the god’s name, he filled the vessel and drank it off. (It. Alex. 118)Footnote 35
Justin’s narrative [7] is in this episode not very different from the Romance [5, 6]. He went on to develop the poison story, taking it as true, and attributing the plot to Antipater and his sons Cassander and Iollas, who trusted no one else in Babylon but Medius. This, Justin explained, was why ‘the party was revived at the Thessalian’s’ (12.14.6–8). By Diodorus [8], Plutarch [9] and Arrian [10] the story was narrated without mention of poison, but all three authors discussed and rejected the poisoning story later. Plutarch’s quoted text is followed directly by the citation of Aristobulus [4], then by an extract of the Ephemerides that takes the story to Alexander’s death. Arrian’s quoted text leads into the citation of the Ephemerides [1]. The brief Itinerarium Alexandri [11] is not based on Arrian here, and its version of the bowl of Heracles story cannot derive from any text cited above.
THE COMMENTARY
κωμάσαντα: ‘having caroused’
The verb κωμάζω is a technical term of classical conviviality: in the austere style in which the Ephemerides were written it was used here in its technical sense. A κῶμος ‘kômos, carouse, serenade’ was a release from the formality of a dinner and symposium. One set out, perhaps in a group, for home or some other haven, as, famously, in the case of Alcibiades (Pl. Symp. 212c–223d) whose carouse brought him and a flute-player to Agathon’s symposium, where he was to stay all night: a fictionalized occasion, unique and in some sense typical.
A Macedonian drinking party, πότος, might transform in the Greek way into a κῶμος. Philip’s carouse after the battle of Chaeronea is an example;Footnote 36 Alexander’s most famous carouse was at Persepolis, during which the palace of the Persian kings was destroyed by fire.Footnote 37 On the present occasion (the sources are unanimous) the κῶμος began simply: Alexander set off for bed.
ἐν τῷ τϵλϵυταίῳ δϵίπνῳ: ‘at the last dinner’
Before considering how the last carouse ended we must know where it began. The Ephemerides do not help: whatever they reported is lost.
Each surviving narrative offers an answer. Plutarch (Alex. 75.3–4) has Alexander recovering slowly from the loss of Hephaestion: eventually returning to sacrifices and drinking parties, on one occasion he feasted Nearchus’ men and Medius met him on his way to bed [9]. Justin’s immediate context (12.13.6) makes no reference to Hephaestion: there were days of amusement and a revival of the custom of a formal dinner; Medius caught him as he was leaving this [7]. Diodorus has ‘amusements and a life of festivity’; portents were offset by sacrifices (17.116.1, 117.1), after which came Medius’ invitation [8]. Arrian (7.24.4) is close to Justin and Diodorus: after sacrifices Alexander celebrated with friends and drank late; some wrote of a meeting with Medius [10]. There is a common story: sacrifices and drinking parties in Plutarch and Diodorus; a dinner for Justin, a celebration with friends for Arrian; on the way to bed, a chance meeting.
The text of the Alexander Romance is prone to disturbance at this point, the Liber de morte also, but recension A tells of Alexander enjoying himself among friends and with the ‘artists of Dionysus’, of whom many were holidaying in Babylon: they had ‘wreathed Alexander and competed against one another’ (3.31.6): on leaving this celebration Alexander encountered Medius [5, 6]. Nicobule [2] said of a dinner or series of dinners that ‘the performers strove to amuse the king’ and that at ‘the last dinner’ Alexander himself performed a piece from Euripides:Footnote 38 this focus on performers fits the context in the Deipnosophists and must represent something real in Nicobule’s text, else Athenaeus would not have cited it.
The absence of the ‘artists of Dionysus’ in certain sources might represent the plain truth, but it seems more likely that the Alexander imagined by Diodorus, Plutarch and Arrian had no time for undignified pleasures (uoluptatibus comioribus: It. Alex. 118) and certainly no time to perform himself: hence those authors give us dinners with friends unspecified and divine observances without Dionysus, while the Romance and Nicobule have him performing, competing and wreathed.Footnote 39 Several authors concur that this ‘last dinner’, as Nicobule called it, morphed into a πότος (Arrian), a nox peruigilis (Justin) and a κῶμος in fourth-century Greek usage (Ephemerides). If we may thus harmonize dissonant sources, Nicobule, the Romance, the Itinerarium and Justin allow partial reconstruction of an event to which no surviving text gives more than a few words. Modern biographies, if mentioning this occasion at all, generally select details from either Arrian or Plutarch exclusively.
πίνϵιν παρὰ Μηδίῳ: ‘he drank with Medius’
The last carouse led Alexander to his meeting with Medius. The terminology is variable for the event that now followed and the variation is informative. πίνϵιν ‘drink’ is the simple verb chosen in the Ephemerides [1]. Nicobule [3] had συμπόσιον ‘symposium’ twice (perhaps recalling the symposium which Alcibiades interrupted), but why, in this same fragment, do we find the participle δϵιπνῶν ‘dining’? The Itinerarium [11] suggests that this would have been an easy term to choose, albeit misleading: there had indeed been a dinner at Medius’ and Alexander arrived as it ended.
Meanwhile Arrian’s reader [10] might want to know why Alexander accepted the invitation. The claim that MediusFootnote 40 was currently the most trusted Companion appears to be an improvised response: nothing supports it. Medius according to Plutarch (Quomodo adul. 65C) by spreading rumours about Callisthenes, Parmenion and Philotas had some responsibility for the deaths of all three.Footnote 41
Arrian (7.24.4, 7.27.2) and Diodorus (17.117.1) called the event at Medius’ a κῶμος; Arrian (7.24.4) and Plutarch (Alex. 75.4) used the verb κωμάζω: Justin (12.13.7) and the Itinerarium [11] offer comisatio, the same term as a loanword in Latin. Cleitarchus, a source for Diodorus and potentially for others, perhaps employed this word. It would be a natural usage for an author distant from the event, since a κῶμος was what classically followed a δϵῖπνον and συμπόσιον. This word choice differs sharply from that of the Ephemerides, conceivably drafted on the day after the event: in its author’s mind κωμάζϵιν was what Alexander began to do when he set off for bed and ceased to do when he accepted Medius’ invitation.
The claim that there were twenty guests is made by Nicobule [3] and, with a list of names added, in the Romance [5].Footnote 42 Twenty meant little in a Greek context but was appropriate in Macedonian conviviality, given that there are said to have been twenty guests at the dinner that celebrated the marriage of Caranus (Hippolochus at Ath. Deipn. 128c); the number twenty has archaeological support in Macedonia, as Carney observes.Footnote 43
Another reason for Nicobule to have mentioned this number emerges in the light of the extract from Ephippus (Ath. Deipn. 434a–b = BNJ 126 F 3) quoted above. In Macedonian usage, when one pledged or toasted another, the other drank off the wine in the cup, and, sooner or later, returned the pledge: thus Alexander ‘compelled’ the twenty guests to drink (ἠνάγκαζϵν: the word is Nicobule’s from another context [2]), not as an absolute monarch but as a well-brought-up member of the Macedonian court. Under the same rule the twenty guests in turn would have compelled Alexander to drink off twenty cups, as he of course foresaw.Footnote 44
σκύϕον Ἡρακλέους ἐκπιών: ‘draining the bowl of Heracles’
Aristobulus promoted a viewpoint on such events: ‘Those drinking parties [πότοι], so Aristobulus says, were prolonged not for the sake of the wine but out of affability to his companions’ (Arr. 7.29.4 = BNJ 139 F 62). Aristobulus’ readers dismiss this impatiently, but it may have been true at first: as Badian observed, Alexander as a new monarch ‘would dine, drink and hunt with his hetairoi … this was how he would earn the confidence of his men’.Footnote 45 The young Alexander’s diplomacy was conducted παρὰ πότον ‘over a drink’, said Ptolemy (BNJ 138 F 2), as his father’s had been.
Aristobulus was certainly responding to common opinion: Alexander drank too much, as hinted by Menander below; it disordered his life, as seen in the Ephemerides; it contributed to his last illness, as diagnosed by Aristobulus [4], and led to his death, as Ephippus said (BNJ 126 F 3). However keen to dismiss the wine question, Aristobulus could not ignore it. Elsewhere he told of a Syrian soothsayer who advised Alexander to drink through the night (BNJ 139 F 30); he retailed Callisthenes’ famous joke when refusing wine (οὐδὲν δέομαι […] ᾽Αλϵξάνδρου πιὼν τοῦ ᾽Ασκληπιοῦ δϵῖσθαι ‘after drinking Alexander’s I don’t want to have to call for Asclepius’’: BNJ 139 F 32).Footnote 46
The use of the bowl of Heracles on this occasion is clearly attested in Seneca’s Letters to Lucilius (83.23), in the Itinerarium [11],Footnote 47 and in Plutarch’s dismissal [9]; it is noticed allusively in Diodorus [8]. Since all these knew of it we may suggest Cleitarchus as source.
Diodorus’ μέγα ποτήριον ‘big cup’ was evidently Plutarch’s σκύφος Ἡρακλέους ‘bowl of Heracles’, a particular specimen of the σκύϕος ‘bowl’,Footnote 48 fundamentally different from the modest, polite κύλιξ ‘goblet’. Pownall reasonably identifies the ποτήριον δίχουν ‘three-litre cup’ from which Proteas and Alexander pledged one another (BNJ 126 F 3) with the bowl of Heracles at Medius’ party.Footnote 49 She finds awareness of it in a play of Menander performed soon afterwards: κοτύλας χωροῦν δέκα ἐν Καππαδοκίᾳ κόνδυ χρυσοῦν, Στρουθία, τρὶς ἐξέπιον μϵστόν γ’. – Ἀλϵξάνδρου πλέον τοῦ βασιλέως πέπωκας. – οὐκ ἔλαττον, οὐ μὰ τὴν Ἀθηνᾶν. – μέγα γϵ (‘In Cappadocia I drank off three times a full gold kondy of ten pints.’ ‘You drank more than King Alexander!’ ‘As much, by Athena.’ ‘That’s mega!’, Kolax fr. 3).Footnote 50
Having already compelled himself to drink more than any of his companions, Alexander thus drank dry the bowl of Heracles, a challenge he could scarcely refuse: it was the annual festival of the death of Heracles on Mount Oeta in Thessaly;Footnote 51 Medius was a Thessalian and was celebrating, as the Liber de morte [6] says; Alexander himself was the descendant of Heracles.
μϵτ᾿ οὐ πολὺ ἀνϵπαύϵτο: ‘soon afterwards he went to sleep’
Plutarch again denied [9], as did Arrian (7.27.2), that immediately after drinking Alexander cried aloud and was carried or led away. The detail of the ‘great cry’ might have originated in the same reality as Aristobulus’ statement that Alexander ϕρϵνιτιᾶσαι ‘suffered a frenzy’ [4], or it might have arisen long after the event. It is mentioned by the authors of all surviving narratives except the two texts conceivably written by eye-witnesses: it would be out of scope for the Ephemerides; Nicobule said nothing of it, offering the less exciting fact that Alexander ‘left the symposium’ [3].
In the Ephemerides [1] ‘withdrawing and bathing [he] slept’. To this Nicobule added an unexpected scrap of knowledge: how long it took Alexander to go to sleep [3]. How could she know that? An answer is suggested by a citation that follows closely in Athenaeus: ‘Polycleitus of Larissa in Book 8 of his Histories says that Alexander slept on a gold couch, and that flute-players female and male always followed him to camp and drank till dawn’ (Deipn. 539a = FGrHist 128 F 1). If Nicobule had been one of the artistes at court during these last days,Footnote 52 she could indeed have observed, or learned from colleagues on κῶμος duty, how long it was before Alexander fell asleep.
There may seem to be a timeslip dividing the Ephemerides from other sources as regards the day following the last carouse: Alexander again dined with Medius, again drank, and, now feverish, slept in the bathhouse. Was there, then, a dinner and drinking party on the next day? A report compiled day by day of the king’s acta differs from screenshots of rumour that show a semi-conscious Alexander, helped to bed, never to recover. It made no difference if he took some food and drank with his host on the next day: to Nicobule [2] his ‘last dinner’ was his recital from the Andromeda and celebration with the artistes on the day before.
πολλὰ δὲ καὶ ἄλλα οἶδα ἀναγϵγραμμένα: ‘I know that much else has been written’
This was the last day on which Alexander played the normal part of a Macedonian monarch; after about eleven days he died.Footnote 53
There was a silence: Justin (13.1.1) and Curtius (10.5.7) said exactly that; it may have been Arrian who called it ‘a silence deeper than any desert’ (Suda s.v. σιγηλός; Arrian, τὰ μϵτὰ Ἀλέξανδρον fr. 9 Roos). Yet the silence was broken and the causes of Alexander’s death were discussed. Justin alone, rarely credited, states that two explanations were in play: ‘His friends put it about that his immoderate drunkenness was the cause, but in fact there was treachery, covered up by the power of the Successors’ (12.13.10, cf. 12.16.12).Footnote 54
As to immoderate drinking, Ephippus indeed wrote of the revenge of Dionysus, but he is not numbered among Alexander’s friends. If Justin is to be taken at his word, we must rather look to Nicobule and to Eumenes. The latter, secretary and friend, might have published the Ephemerides soon after the king’s death.Footnote 55 In this apparently official text evidence is offered that would support the claim of excessive drinking,Footnote 56 also of death from fever; but additionally a handle is given to possible rumours of ‘treachery’, that is, of poisoning. A time and place are offered—a private party, a night of heavy drinking—where poisoning could have occurred; it is placed on record that thenceforth the king lay at Medius’ or in a temple or a bath-house, increasingly ill, eventually unable to speak, soon to die (BNJ 117 F 2–3).Footnote 57
Plutarch said that ‘no one was suspected of poisoning at the time’ (Alex. 77.2).Footnote 58 That was certainly not true. ‘A great many believed that he was killed by poison […]. Believed or not, the story was soon extinguished by the power of those whom it implicated’ (Curtius 10.10.14–18); ‘[Some of the writers say] that after [Alexander’s] death, [Antipater] being all-powerful in Europe and after him his son Cassander, many authors did not dare to write about the poison’ (Diod. Sic. 17.117.7–118.2). Now it is asserted by an unnamed but often reliable author that the Athenian orator Hypereides proposed a decree honouring Iollas ‘who was thought to have given Alexander the poison’ ([Plut.] X orat. 849F);Footnote 59 this would have occurred in late summer 323 and would help to explain the special enthusiasm shown by Iollas’ father Antipater in tracking down, torturing and killing Hypereides late in the following year (admittedly he had other reasons for revenging himself on Hypereides).Footnote 60 All this reinforces the impression that Plutarch was wrong and that Justin, Diodorus and Curtius were right: there were rumours of poison and they were silenced.Footnote 61
The statement of the Liber de morte [6] that Onesicritus abstained from listing those who were at the last carouse offers further support. Onesicritus wrote one of the earliest accounts of Alexander’s adventures: native of Astypalaea, well within reach of Antipater and Cassander, it is likely enough that he feared reprisal.Footnote 62
Ephippus, too, wrote very soon after Alexander’s death. Given his title, one would have expected him to say much more than that (as quoted above) Alexander ‘sickened and died’. To Nicobule no word on the days following Medius’ party is attributed. These points may, however, mean nothing: both authors were little read and rarely quoted.
μηνύσϵως γϵνομένης: ‘when information was revealed’
The presence at Medius’ of Alexander’s personal cupbearer, Iollas, is well attested. Olympias thought he was there (Plut. Alex. 77.2); he was there according to Justin (12.14.6) and the Romance [5]. Arrian outlined the story clearly, disbelievingly, in indirect speech: ‘I know that much else has been written about Alexander’s death: that Antipater sent him a drug and that he took it and died; that Aristotle, already fearing Alexander because of Callisthenes,Footnote 63 had created the drug for Antipater and that it was given him by Cassander’s younger brother Iollas; that Iollas was the royal cupbearer and had been in some way abused by [Alexander] shortly before his death; and others say that Medius, who was Iollas’ lover, had a part in it—for it was he who brought Alexander into the kômos’ (7.27.1–2).
Οnce he had taken control in Egypt, Ptolemy had no reprisals to fear. Yet Arrian observes at the moment of Alexander’s death that ‘neither Aristobulus nor Ptolemy narrates otherwise than this’ (7.26.3), meaning that he found nothing in them to contradict the Ephemerides.Footnote 64 Now Aristobulus is quoted by Plutarch [4] for an account so brief that a dispute between him and the Ephemerides can scarcely arise: we may gather from Arrian’s cryptic statement that Ptolemy was similarly brief.Footnote 65 In general Ptolemy (as reported by Arrian) gave few insights into events at court, unlikely as they were to contribute to his own credit or to Alexander’s.Footnote 66 If he was present on this occasion (as claimed in the Romance) it was nothing to be proud of, and he let the incident drop from his memoir.
Yet it was in Egypt, in Ptolemy’s time and after, that the silence would be breached. Estimates of the date of Cleitarchus’ Histories of Alexander (BNJ 134) have varied greatly. Early dates, before 300, have been widely accepted,Footnote 67 although according to Diogenes Laertius (2.113) ‘Cleitarchus’, not further identified, was a student of Stilpo the Megarian philosopher, who was contemporary with Ptolemy I and may have taught at Alexandria.Footnote 68 It now appears that W.W. Tarn and others were right to place Cleitarchus’ work later,Footnote 69 if, as P.Oxy. 4808 affirms, he was in government service in Alexandria and was tutor to the future Ptolemy IV Philopator: he would then have died not before 235;Footnote 70 the sensational material that Cleitarchus offered, disdained by Arrian (who never named him) and Plutarch, but enjoyed by others, was not after all owed to his presence on Alexander’s expedition but to his access to memoirs and documents. Plutarch cited him once (Alex. 46 = BNJ 137 F 15) and probably used him more often.Footnote 71 As noted at the outset, Cleitarchus’ work probably underlies the narratives of Diodorus and Curtius. If so, it would then be Cleitarchus’ version of the drinking at Medius’, featuring the bowl of Heracles, that Plutarch knew [9] and Diodorus accepted [8].Footnote 72
It would be through another source, not Cleitarchus, that Trogus heard of the story circulated by ‘Alexander’s friends’ that excessive drinking caused the king’s death. Through this source Trogus learned that Medius was the only man in Babylon that the poisoners could trust (Just. 12.13.10, 12.14.7). This source asserted that the poisoning story was true, and Trogus and the compiler of the Liber de morte accepted it. Iollas, Alexander’s cupbearer beloved by Medius, plays no part in the narratives that mention the bowl of Heracles: he is in the Romance [5], the Liber de morte (96–9), Justin (12.14.6, 9) and Arrian (7.27.2).Footnote 73
ϵἴκοσιν οὖσιν ἐν τῷ συμποσίῳ: ‘there being twenty at the symposium’
‘Trogus’ source’ (if I may usurp that term),Footnote 74 or at least Justin, by asserting that Medius, alone of those in Babylon, was told of the poisoning plot (12.14.7), contradicts the claim in the Romance [5, 6] that it was known to fourteen of the twenty men at the party.
Of the alleged symposiasts a few recur in the ‘Testament of Alexander’, a few in other episodes of the Romance, but the majority occur there only once; yet all can be identified, confidently or doubtfully, in historical sources. A list of real names most of which would mean nothing to the intended reader suggests, not necessarily that the list was valid, but that it originated at an early date when those names had relevance. ‘All twenty men are known independently to have had reason to be at Babylon’, wrote Lane Fox; Samuel rightly jibbed at this,Footnote 75 but none of them can be shown not to have been in Babylon. To create even such a credible list as that was a rare achievement.Footnote 76
In considering the question whether the list is valid we can observe what the twenty men have in common. There is no justification for a general claim that they were Alexander’s closest friends or particular friends of Medius. They appear to be a random sample of Companions and others who were in Babylon—until one looks forward, with Heckel, at what they would be doing in 317/315.
Among those who ‘did not know’, Perdiccas is dead; the fate of Olcias is unknown. Still alive are Ptolemy, Lysimachus and Asander, relatively unattached. Eumenes, fighting for the legitimate succession, close to Olympias, opposed to Antigonus Monophthalmus and Demetrius Poliorcetes, is to die in 315.
Among those who ‘knew’, Leonnatus has turned against Perdiccas and Eumenes and is now dead, as is Meleager, executed long ago by Perdiccas. Peithon, having abandoned Eumenes, is soon to die. Alive are Medius, Philotas, Nearchus and Menander, certainly attached to Antigonus; Ariston, Philip the engineer and Heracleides the Thracian, attached to Antigonus if they have been correctly identified. Seleucus (tentatively restored to the list), Stasanor and Peucestas, unattached in 317, are destined to submit to Antigonus in 316.Footnote 77
On this basis Heckel dated to the years 317/316 the construction of the list, testament and letter to the Rhodians that are combined in the Liber de morte, proposing that its purpose was to support the Polyperchon faction:Footnote 78 in 318/317 Polyperchon, newly in power, recalled Olympias to Macedon (Diod. Sic. 18.57.2), whereupon ‘information was revealed’, said Plutarch cryptically, and Olympias, among other acts of retribution, desecrated the grave of Iollas ‘who had served the poison’ to Alexander, thus ‘avenging his death’ (Plut. Alex. 77.2; Diod. Sic. 19.11).Footnote 79 So the same political intrigue entailed the stirring of Olympias, the ‘revealing of information’ to her and, if Heckel was right, the circulation of a document that would become the Liber de morte and would form part of the Romance.
For reasons deriving from the associated texts Bosworth preferred to date that document to 309/308 and to name Ptolemy as responsible. Baynham, writing in the same volume, agreed, starting from an analysis of the gruesome portent also described in the Liber de morte. If this dating is accepted then Ptolemy inspired the adoption of the poison story and the forging of the ‘Testament’ in what I may call an unattributable appendix to his memoir.Footnote 80
Samuel had argued that the list of those who drank with Medius was a separate document in origin, ‘independent of politics’, ‘going back to accurate knowledge of those present’;Footnote 81 we know that the poison story was already spreading as Olympias returned to Macedon; the longer the time lapse between Alexander’s last carouse and the creation of the list, the more difficult its creation would have been. We may postulate that if a list existed in memory from the beginning, it was radically revised to become part of the poison story in a document of about 317/315; a few years later this was attached, with Ptolemy’s approval, to the forged testament of Alexander. Heckel has shown, more fully than others, that the list as we have it is weighted with later significance. Too many of the guests who ‘knew’, scarcely linked together in 323, are aligned with Antigonus or against Olympias and Eumenes in 317/316. Medius can hardly have been so prescient.
The question whether there was poison in 323, and whether at least Medius knew of it, is not my subject:Footnote 82 I add only that any who knew of it would have laid plans for a future beyond the king’s death. There is little sign that anyone had.
πυρέττοντα μανικῶς: ‘with a raging fever’
It is probable that Aristobulus used Chares of Mitylene (FGrHist 125) as a source on events at Alexander’s court:Footnote 83 on several occasions the two agreed, and Chares was among the earlier of Alexander’s contemporaries to publish a memoir, while Aristobulus was among the latest. Yet no one quoted Chares on the last days. A native of Mytilene, within reach of Antipater and his son, perhaps Chares was one of those who decided to say nothing.
As for Aristobulus [4], it cannot be certain that he was describing an incident of the day in question, but it hardly mattered. The connection would be made, since the party at Medius’ was already prominent in the record and linked with Alexander’s terminal illness. Admittedly the Ephemerides [1] first noted Alexander’s fever on the following day, but this is because the fever first affected his movements on that day. Feverishness at a drinking party would have been out of scope.
Alexander’s pathological thirst was not plucked out of the air: Aristobulus had used it to explain an immediately preceding incident (Arr. 7.24.1–3 = FGrHist 139 F 58).Footnote 84 A physician would probably have employed the expression ‘raging fever’ of a later stage in an illness, after the ‘frenzy’, closer to coma and death;Footnote 85 but much depends on the cause of fever, and Aristobulus was not a physician. None the less he was right in humoral theory: however thirsty, a feverish patient must not assuage that thirst with wine.
The writers of the surviving narratives, three to four centuries later, were at a loss. Arrian had based himself throughout on Aristobulus and Ptolemy, but Ptolemy, though often an eye-witness, wrote of court events only what was public knowledge, while Aristobulus, probably not present, swept them aside. Here Arrian did not trouble to cite Aristobulus, who apparently covered the last days in sixteen words, or Ptolemy, who apparently had nothing to add. Diodorus, alone, gives us [8] the probable Cleitarchus version, including a couple of details that Plutarch had read and explicitly rejected [9]. Another source, reflected in Justin [7] and in the Romance [5, 6], was known to Arrian (7.27.1–2), but he did not trust it. Onesicritus, Ephippus and Chares were apparently unhelpful. Perhaps no later writers except Pliny and Athenaeus ran Nicobule [2, 3] to earth. Plutarch fortunately cited Aristobulus [4]; beyond him, distrusting Cleitarchus and disliking the poisoning narrative, Plutarch and Arrian were left with no source worth naming except the Ephemerides [1].
Excessive drinking and poison having been the popular theories at first, Aristobulus has the honour of being the earliest writer to state plainly a third cause of death: a fever to which ill-considered drinking contributed. The fever is now identified most often as Plasmodium falciparum malaria,Footnote 86 though Salmonella typhi is a serious contender.Footnote 87 If Aristobulus is right, the last carouse was not the underlying cause of anything, but it was the moment at which Alexander himself, a courtly Macedonian monarch who had become too divine to listen to reason,Footnote 88 was encouraged to bring about his own death.
