¹ SeeN. G. Wilson,Scholars of Byzantium(London,1983),204–6,Google ScholarA., Kashdan,The Oxford Dictionary of Byzantium(New YorkandOxford,1991),1, 427–8 s.v.Google ScholarChoniates, Michael. His writings in prose and verse were published by S. P Lambros (2 vols., Athens, 1879 and 1880).Google Scholar

² See my Oxford (1990) edition of Callimachus’ Hecale, pp. 38^0. In ZPE115 (1997), 55–6,1 suggest that Michael may provide a clue to the link between Hecale frr. 1 and 2 H. It is odd that he never mentions Callimachus by name.

³ Vol. 2, p. 295, lines 20–2 ed. Lambros.

⁴ I am grateful to Dr Dirk Obbink for confirming this point after consultation with the computer. A glance at Michael's hexameter verse (2, pp. 375–93) should suffice to show that this elegant phrase is not his own composition.

⁵ SeeP. M. C. Forbes IrvingMetamorphosis in Greek Myths(Oxford,1990)139–48.Google Scholar

⁶ is probably not the word which Michael would have chosen for Niobe alone, but he needs a noun which will cover both his mythical and his scriptural () example (Lot's wife turned into a pillar of salt, from Genesis 19.26). Also, as Mr N. G. Wilson points out, there is a wordplay between and (‘posted’, or even ‘pilloried’). I learn from R. Atanassova that Niobe is linked with Lot's wife in Clement of Alexandria Protrepticus 103.4, and that the detailed transformation of Lot's wife in Prudentius (Hamartigenia 742ff.) strongly recalls Niobe.

⁷ Cf. Cicero, Tusc. Disp. 3.63 ‘Niobe fingitur lapidea propter aeternum, credo, in luctu silentium’. In Aeschylus' Niobe, the heroine sat silent by her children's tomb, uttering no word until the play was well advanced. See Aristophanes, Frogs 911–24 with Tragicorum Graecorum Fragmenta vol. 3, Aeschylus (ed. S. Radt [Gottingen, 1985]), pp. 265ff.; O. P. Taplin, HSCP 76 (1972), 60–2; W. S. Barrett in The Papyrus Fragments of Sophocles, an Edition with Prolegomena and Commentary by Richard Carden with a Contribution by W. S. Barrett (Berlin and New York 1974), pp. 171–235, especially pp. 171–4 and 223–35. Barrett is particularly concerned with Sophocles' Niobe, but also deals with other tragic accounts, and with the legend in general.

⁸ Nemesis speaks of Aura, who has insulted Artemis.

⁹ In Aeschylus fr. 158.2 Radt Niobe's father Tantalus says of his homeland .

¹⁰ As in Quintus of Smyrna 1.305 .

¹¹ As … in the Iliad scholion connected with Euphorion fr. 102 Powell (to be discussed below).

¹² M. van, der ValkEuslathii Commenlarii ad Iliadem Pertinentes,4(Leiden,1987),962.Google Scholar

¹³ 4.963.23 van der Valk. The epithet does not favour a pre-Hellenistic poet; nor does it exclude Quintus of Smyrna, to whom van der Valk (see his note Ibid.) would give the reference. The unnamed poet apparently stressed that the rock looked like a woman only from afar, not when viewed from nearby. That accords with Q.S. 1.302–4. But the poetic phrases which appear in Eustathius (see below) are not drawn from Quintus, and if (as I suspect) Eustathius and Michael have in mind the same poet, Michael's verbatim quotation shows that he is not Quintus of Smyrna.

¹⁴ At this point Eustathius is ostensibly summarizing Palaephatus, though no such phrase occurs in the surviving abridgment of N., Festa (ed.), Palaephatus 8 (Mythographi Graeci vol. 3, 2 [Leipzig, 1902]).Google Scholar

¹⁵ SeeF. VianRecherches sur les Posthomerica de Quintus de Smyme(Paris,1959),131–3.Google Scholar

¹⁶ E.g. in Vian's book (n. 15 above) though with varying degrees of confidence: p. 132 ‘sans doute Euphorion’, but p. 133 ‘Qu′il s‘agisse d’Euphorion ou d‘un epique plus ancien’.Google Scholar

¹⁷ is a regular designation of Euphorion in Athenaeus (e.g. 4.182e). As for Eustathius' knowledge of Euphorion, nearly all the references and quotations can be found in surviving secondary literature earlier than Eustathius; the exceptions are fix 52 and 171 Powell, though these too may be derived from secondary literature now lost rather than directly from the text of Euphorion. But the poetic colouring in Eustathius' accounts of the Niobe myth (above) does suggest direct access to a lost poetic source, whether Euphorion or some other writer.

¹⁸ I have not noticed any trace of Euphorion elsewhere in Michael's writings.

¹⁹ This phrase clearly includes Lot's wife.

²⁰ If Michael's source were Callimachus' elegiac Aetia, then the author of Eustathius' version of Niobe (described as ) must almost certainly have been a different poet.

²¹ See Forbes Irving (n. 5 above), pp. 19–32 (especially p. 27), and e.g. my edition of Ovid, Metamorphoses 8 (Oxford, 1970) on lines 719–20.

²² The passage o f Michael is o f considerable interest, and provides a good example of the way in which he mixes classical with Biblical allusions. I translate Mich. Chon. 2.206.5–11, where Michael conveys thanks to his young friend Georgios ‘both for making a copy of Nicander and for doing so from such an original, which seemed to be a closed book for reading, in my opinion, to everyone but him; it had been written in such very ancient letters, as if they belonged to a different language, and these blotted and faint, so that one could more easily discern the track of a serpent across a rock something that not even Solomon [Proverbs 30.9] knew than the hexameter treatise on snakes which the book contained. Nonetheless, treating my instructions as a further commandment from the Lord, which brings light to the eyes [Psalm 19.8], he accurately surveyed what could not be espied [, probably from Aratus, Phaen. 25], read the Nicandrean Theriaca which were beyond the range [, perhaps from Sophocles, OT111] of ordinary eyes, and made a copy, hunting down the reptiles as not even a deer can d o [Nicander, Ther. 141–4], or that most sharp-sighted of birds [the eagle see Ther. 448ff.].’ Later, however, Michael seems less pleased with Georgios, ending a letter curtly (without final salutation!): (2.242.24 it is unclear whether he refers to the original or to the copy). As for the manuscript written ‘in very old letters, as if they belonged to a different language’ (2.206.8, … ), one naturally thinks of an uncial text. Mr N. G. Wilson, however, is surprised that Michael should have had such difficulty with uncial script, and suggests that he may be referring to an early, experimental form of minuscule (perhaps c. A.D. 800), which can be very hard to read.

²³ Eustathius'; would suit Nicander too, since nearly all the Nicandrean poems are in hexameters. The Delphians in 254/3 B.C. (SIC? 452) granted privileges to Nicander of Colophon . It seems beyond doubt that there were two poets called Nicandei; the younger of whom flourished perhaps c. 200 B.C.;seeCameron, Alan, Callimachus and his Critics (Princeton, 1995), 199–205. I suspect, however, that the existence of two Nicanders was forgotten as early as the first century B.C, leaving scholarly squabbles about a single Nicander's date and parentage.Google Scholar

²⁴ I wonder (though there is no evidence in the ancient scholia) whether it might have sprung from scholarly interpretation of Odyssey 9.140–1 .The possibility of taking there as an internal accusative is mentioned byJ., Chadwick,Lexicographica Graeca(Oxford, 1996), 248–9, although he (like editors, who put a comma after line 140) prefers to take as nominative, with in apposition.Google Scholar

²⁵ Also in later Greek prose (LSJ s.v. II.2). The compound is more often used with internal accusative, as in H.H. 3.380 , Ap. Rh. 3.225 (see Malcolm Campbell. (A Commentary on Apollonius Rhodius, Argonautica III.[Leyden, 1994], 206–7), who adds [Orph.], Arg. 1132).Google Scholar

²⁶ Note, however, that ‘superba’, in cod. Paris. 8458 (saec. XV), is no doubt a conjecture (see n. 30 below); the main manuscripts have ‘superb(a)e’.

²⁷ Heyworth tells me that ‘the reading of the Poggian manuscripts is split; T and S (the most reliable) agreeing with N in reading "lacrimas"; the others following the Petrarchan tradition. Contamination is probably responsible; but this might be one of those cases where the archetype had alternatives.’ The variant ‘lacrimans’ for ‘lacrimas’ has been printed by several editors. Conjectures include ‘lacrimas depluit’ (Scaliger, printed in G. P. Goold's 1990 Loeb), ‘sollicito lacrimans defluit os Sipylo’ (Housman, with ‘Niobae. superbae’ in the hexameter), ‘lacrima sollicito defluit a Sipylo’ (Phillimore, also with ‘Niobae. superbae’). I would prefer the reading of N to any of these.

²⁸ For discussion of the two passages together, see S. J. Heyworth, ‘Some Allusions to Callimachus in Latin Poetry’, MD 33 (1994), 51–79 at pp. 56–9.

²⁹ And perhaps also for Ovid, Met. 6.312 ‘lacrimas [v.l. ‘lacrimis’] etiam nunc marmora manant’. The exact equivalent for defluo would be . Although LSJ do not recognize the use of that verb with internal accusative, note the phrases in Eustathius on Niobe (discussed above) and , which look as though they come from an ancient poem, perhaps even the poem from which Michael drew his quotation. It is possible that , whether or not in tmesis with , stood earlier in the first line, e.g. .

³⁰ Heyworth (as a future editor of Propertius) warns me that vindication of ‘lacrimas defluit’ would not remove all the difficulties which have been felt over the couplet 2.20.7–8: ‘Housman's point [ The Classical Papers of Housman, A. E., edd. J., Diggle and Goodyear, F. R. D. (Cambridge, 1972), (vol. 1) pp. 270–1] that we would expect in,Google Scholar not a Sipylo still seems to hold good; Shackleton, Bailey [Propertiana (Cambridge, 1956), pp. 102–3] produced reason for doubting the conjecture superba; and I have yet to see any cogent explanation of sollicito’ (S.J.H.).Google Scholar