¹ The traditional view is represented by, among others, Schmid, W., RE vi. 1Google Scholarcol. 1121, PLRE i.296, Paschoud, F., Clinq études sur Zosime (Paris, 1975), p. 169Google Scholar. Goulet, R., JHS 100 (1980), 60ffCrossRefGoogle Scholar., has argued with some probability that the formula τελ⋯ν εἰς ἒκτον κα⋯ δ⋯κατον ἒτος (V. Soph. 10.1.2) means that Eunapius was still in his sixteenth year, had not yet reached his sixteenth birthday, when he arrived at Athens. For a thorough study of the problems involving the dating-formulae the reader should consult Baker, A. E., Eunapius and Zosimus (Diss. Providence, 1987), pp 1 ff., 103ffGoogle Scholar. Goulet further argued that Eunapius commenced study with Proaeresius after the recision of the decree against the Christians passed by Julian on 11 January 364 (CTh. 13.3.6), and he has been followed by Blockley, R. C., The Fragmentary Classicising Historians of the Later Roman Empire (Liverpool, 1983), i. ix (Addendum) with p. 1Google Scholar, and Booth, A. D., AHB 1 (1987), 14fGoogle Scholar.

² The text is that of J. Giangrande (Rome, 1956).

³ The approximate absolute dates are supplied by the critical intersection of Eunapius' career with the ban from teaching suffered by Proaeresius in consequence of Julian's edict of 17 June 362 (CTh. 13.3.5 with Jerome, , Chron. s.a.362)Google Scholar.

⁴ e.g. PLRE i.731; Goulet's argument (see n. 1 above) that Eunapius commenced his association with Proaeresius after the latter's exclusion and return is, if possible, even more repugnant to the sense of 10.8.3 and the weight we may reasonably allow πετ⋯ πέμπτος ἒτος to bear.

⁵ That Proaeresius literally was on his deathbed when Eunapius took his leave, is, of course, improbable. We must distinguish, however, between the actual sequence of events (whatever it may have been) and Eunapius' presentation of it, and we should assume that he anticipated the probable effect of what he wrote. To link the five years with Proaeresius to his decision to leave Athens and then to connect this sequence with the notice of Proaeresius' death ‘not many days later’ ineptly and gratuitously conveys the impression of heartlessness.

⁶ Eunapius has ἦσαν μετ⋯πεμπτοι in 7.4.4, ⋯τ⋯γχανε…μετ⋯πεμπτος ὑπ⋯ in 10.7.1. The usage suggested here conforms with that of Herodotus 8.67.2, ⋯πε⋯ δ⋯ ⋯πικ⋯μενος προ⋯ζετο, παρ⋯σαν μετ⋯πεμπτοι οἱ … τ⋯ραννοι κα⋯ ταξ⋯αρχοι ⋯π⋯ τ⋯ν νε⋯ν. Compare Thucydides 6.74.1: Ἀλκιβι⋯δης γ⋯ρ ⋯τ᾽ ⋯πῄει ⋯κ τ⋯ς ⋯ρχ⋯ς ἢδη μετ⋯πεμπτος, where (of course) the clause beginning with ⋯κ attaches to ⋯πῄει, not μετ⋯εμτος (Classen-Steup ad loc). The need for ὐπ⋯ (as in V. Soph. 10.7.1) is obviated by the logic of the sentence. I take it that in the (reconstructed) sentence of Eunapius the μ⋯ν and Se contrast ἠπε⋯γτο with ⋯ξεβι⋯σαντο while μετ⋯πεπτος qualifies and orients the sentence as a whole.

⁷ Medicine was virtually professio altera to Eunapius (see, e.g. V. Soph. 23.6.3–7, and Oribasius, , Libri ad Eunapium: Nuncupatio, ed. Raeder, , CMG vi.3, pp. 317fGoogle Scholar ., reprinted by Giangrande, pp. xxxvii–xxxviii). Eunapius' interest perhaps commenced as a result of the miraculous recovery of his health under the treatment of Aeschines, whom he tells us that he revered (10.1.11–14). The association of Maximus and Chrysanthius ultimately was based on their possession (in common with Proaeresius) of Aedesius as their teacher. If the study of medicine attracted Eunapius at this point in his career (now that his hopes regarding Proaeresius had collapsed), study with one other member of this tightly-knit group was an obvious possibility.

⁸ A family connection (7.4.5) tied Eunapius to the profession through Chrysanthius.

⁹ The words ‘a sophistic career had been prescribed for him’ appear to indicate that the trip to Egypt involved the choice of some other profession, an idea his parents rejected. But other explanations are conceivable.

¹⁰ cf. 23.3.15: τ⋯ν δ⋯ τα⋯τα γρ⋯ϕοντα ⋯κπαιδεὺσας ν⋯ον ἒτι ὃντα, ⋯ν⋯κα ⋯παν⋯λθεν Ἀθ⋯νηθεν, οὐκ ἒλαττον ἠγ⋯πα κτλ. The expressions ⋯κπαιδε⋯σας and ν⋯ον ἒτι ὃντα, though they do not speak decisively for any view, nevertheless better suit a brief interlude in Athens than five years of study under one who, in that case, should properly have been regarded by Eunapius as the διδ⋯σκαλος of record. The fact must be faced that Eunapius does not designate himself as the student of Proaeresius when μετ⋯ π⋯μπτον ἒτος would not only justify it but should require it. Diophantus was the διδ⋯σκαλος (V. Soph. 16.1.3) although Libanius studied in Athens for only three years and, as Eunapius attests, rarely approached his master. Piety toward Chrysanthius is not compatible with ingratitude to Proaeresius. Eunapius' sweeping assertions of the primacy of Chrysanthius are by themselves enough to sink the notion of a five-year-long residency in Athens.

¹¹ Chrysanthius had studied with Aedesius as a ν⋯ος (23.1.5), spending what does not appear to have been many years with him (23.1.10). 23.3.15 (quoted in the preceding note) establishes the fact that Chrysanthius had instructed Eunapius prior to the latter's trip to Athens as well as after it.

¹² This passage would be rather more susceptible to the usual interpretation if Eunapius had pointed it differently. If he had, for instance, told us that Priscus returned to Athens (not to Greece), we might well infer a different train of ideas, e.g. that after Priscus was vindicated, he came to Athens and was met there by Eunapius who at that time was studying in the city. This, however, is precluded by the logic of the passage, which it would be arbitrary to constrain on the basis of the subsequent and unconnected passage of 10.8.3.

¹³ That the Athenian ephebate is irrelevant was recognized by Goulet, , JHS 100 (1980), 62CrossRefGoogle Scholar, though his own view of Eunapius' age at this juncture is unacceptable. In general, see A. Baker on this matter (cited in n. 1 above).

¹⁴ PLRE i. 296.

¹⁵ JHS 100 (1980), 63Google Scholar n. 30.

¹⁶ Giangrande would read ⋯σ⋯γε, ‘nam inter Eumolpidas inscripsit dicere voluit auctor’. But the sense of ἂγω required is one which will contrast the credentials of the (Eumolpid) hierophant with those of his successor, who was not even an Athenian, much less a Eumolpid. Cf. κατ⋯γω LSJ s.v. 8. But ⋯ξ Eὐμολπιδ⋯ν ἤχθη, the least violent correction, for ⋯ς Eὐμολπ⋯δαν ἦγε (the reading of Laurentianus) does not satisfy.

¹⁷ I shall discuss this elsewhere in connection with the date and contents of the ‘first installment’ of Eunapius' historical work.

¹⁸ See, e.g. Blockley, op. cit. (n. 1 above), i.4.

¹⁹ Eunapius saw the famous thaumaturge and philosopher, Maximus, when he was ‘still young’ (ν⋯ος ἒτι, the same phrase he used to describe his own age after his return from Athens to Lydia: 23.3.5, quoted above in n. 10) and Maximus was an old man (7.1.1). He further implies (7.1.2–3) that he observed Maximus when engaged in philosophical discussion. The likely context for such scenes was Constantinople between 367 and 371 (7.6.1–2), a period of rare tranquillity in Maximus' otherwise turbulent life and one, moreover, which harmonizes with Eunapius' self-description.

²⁰ For the destruction of the temples (about 385) see Jones, A. H. M., The Later Roman Empire (Oxford, 1964), i. 167Google Scholar , ii. 1101 (nn. 75, 76).

²¹ A similar conclusion would follow if it were certain that Eunapius' hierophant, the penultimate in the series, was that Nestorius whom Zosimus 4.18.3 describes as ὐπ⋯ργηρων and active in Athens shortly after the death of Valentinian in 375. For then we would be impelled to date the prophecy no earlier than the seventies. The precision with which the hierophant describes his successor requires that he was privy, directly or indirectly, to deliberations concerning his replacement. The natural occasion for such discussions can only have come very late in the life of the aged hierophant. But the evidence is much too uncertain to support any such conclusion, accepted though it is by Blockley, op. cit. (n. 1), i.4. Zosimus 4.18 is at least in part a doublet of 5.6; and on Zosimus' own authority (4.18.4) we know that he inserted this material into his own account because he thought it ‘suitable to the context’. That he has correctly interpolated Syrianus' ὔμνος is impossible to believe.

²² The Greek is quoted above p. 520. If Eunapius was precluded from meeting Himerius because Himerius left Athens to orate before Julian it follows inevitably that their paths would have crossed in Athens if Himerius had remained there instead. The fact that Himerius approached Julian when he was αὐτοκρ⋯τωρ confirms what otherwise could be safely presumed, namely, that Himerius proceeded to Constantinople. For the date of Julian's arrival at that city see Ammianus 22.2.4; for his departure see Ammianus 22.9.2 with Seeck, , Hermes 41 (1906), 514Google Scholar .

²³ It is noteworthy that Eunapius employs the particle που in 10.8.3, when he gives his age κατ⋯ τουτον⋯ τ⋯ν χρ⋯νον. Since one is either in one's sixteenth year or one is not, που must have been intended to cover a slight imprecision. Such usage would be justified if Eunapius celebrated his sixteenth birthday not long before his departure from Athens.

²⁴ It is a pleasure for me to express here my immense gratitude to the John Simon Guggenheim Memorial Foundation for the generosity which allows me to spend the current academic year (1988–9) in research while on leave of absence from Brown.