Published online by Cambridge University Press: 01 October 2015
As is well known, the work of Aristarchus on Homer is not preserved by direct tradition. We have instead many fragments preserved mainly in the Homeric scholia, the Byzantine Etymologica and the Homeric commentaries by Eustathius of Thessalonica. These fragments go back to the so-called Viermännerkommentar (abbreviated VMK), the ‘commentary of the four men’, a commentary that is dated to the fifth-sixth century c.e. and collects the works of Aristonicus, Didymus, Nicanor and Herodian. In the first century b.c.e. Aristonicus explained the meaning of Aristarchus’ critical signs in a treatise called Περὶ τῶν σημείων τῶν τῆς ᾿Ιλιάδος καὶ ᾿Οδυσσείας, while in the Περὶ τῆς ᾿Αρισταρχείου διορθώσεως Didymus studied Aristarchus’ Homeric recension. In the second century c.e. two more scholars, Herodian and Nicanor, dealt with Aristarchus while analysing questions of prosody in the Homeric language (Herodian) or the punctuation of the Homeric text (Nicanor). Not all of these four ‘men’ are equally important, however, as sources for Aristarchus. In fact, Herodian and Nicanor had aims that were quite independent of Aristarchus’ enterprise: the former was concerned with problems of prosody, accentuation and aspiration in Homer, whereas the latter had developed a new system of punctuation to elucidate the Homeric text from a syntactic point of view. Although both Herodian and Nicanor did take an interest in Aristarchus, their focus was thus different from that of their Alexandrian predecessor. The goal of Aristonicus and Didymus, on the other hand, was specifically to reconstruct Aristarchus’ work on Homer; it is for this reason that they are considered the most trustworthy witnesses for Aristarchus’ fragments.
I would like to thank the anonymous referee for CQ, as well as Monica Negri, for reading this article and providing useful feedback. Casey Dué helped me with accessing the digital images of Venetus A.
1 The existence of the VMK and the titles of the works it contained can be inferred from the subscriptiones at the end of each book (with some minor variants in the wording) in the Venetus A (Marcianus Graecus Z. 454 = 822), the main codex with scholia derived from the VMK: παράκειται τὰ ᾿Αριστονίκου σημεῖα, καὶ τὰ Διδύμου περὶ τῆς ᾿Αρισταρχείου διορθώσεως, τινὰ δὲ καὶ ἐκ τῆς ᾿Ιλιακῆς προσῳδίας Ἡρωδινοῦ καὶ ἐκ τοῦ Νικάνορος Περὶ στιγμῆς (‘[The work on] critical signs by Aristonicus and the work by Didymus on Aristarchus’ recension are here added; there are also some [excerpts] from the treatise on Iliadic accentuation by Herodian and from the one by Nicanor on punctuation’). On the tradition of the VMK and Aristarchus’ sources, see H. Erbse, Scholia Graeca in Homeri Iliadem (Scholia Vetera), vol. 1 (Berlin, 1969), xlv-lix; S. Matthaios, Untersuchungen zur Grammatik Aristarchs: Texte und Interpretation zur Wortartenlehre (Hypomnemata 126) (Göttingen, 1999), 38–43; F. Schironi, I frammenti di Aristarco di Samotracia negli etimologici bizantini. Introduzione, edizione critica e commento (Hypomnemata 152) (Göttingen, 2004), 11–25; F. Pontani, Sguardi su Ulisse. La tradizione esegetica greca all'Odissea (Sussidi eruditi 63) (Rome, 2005), 96–100 and 148–50.
2 In particular, Lehrs and Ludwich considered Didymus the best witness for Aristarchus. Cf. K. Lehrs, De Aristarchi studiis Homericis (Leipzig, 1882³), 16–28, esp. 27–8; A. Ludwich, Aristarchs Homerische Textkritik nach den Fragmenten des Didymos dargestellt und beurtheilt, 2 vols. (Leipzig, 1884-5), 1.23, 41–3, 64–7. Recently, however, G. Nagy, Homer the Classic (Washington, DC, 2009), 35–7 has thrown Didymus’ authority into question, claiming that Didymus might not have had access to the final stage of Aristarchus’ work on Homer, as Aristarchus’ ‘definitive base text’ and ‘definitive hypomnēmata’ were preserved not at Alexandria but at Rome; Aristonicus, who consulted them at Rome, is therefore considered in Nagy's view to be a more reliable source on Aristarchus. The most detailed analysis and assessment of Didymus’ sources and method is carried out by M.L. West, Studies in the Text and Transmission of the Iliad (Munich and Leipzig, 2001), 46–85.
3 I will limit my analysis to the scholia to the Iliad because only for the Iliad does a manuscript close to the VMK—and so rich in Aristarchean fragments—exist: the Venetus A. On the other hand, the scholia to the Odyssey are much more heterogeneous and thus much more difficult to analyse; see Pontani (n. 1), 148–50. Moreover, it is only recently that a modern edition—one which clearly identifies the different origins of the scholia—is finally being prepared by Filippomaria Pontani; when writing this article, however, only the scholia to Od. 1–4 were available. The scholia to the Iliad are quoted according to Erbse's edition: Scholia Graeca in Homeri Iliadem (Scholia Vetera) Recensuit Hartmut Erbse, 7 vols. (Berlin, 1969–1988).
4 On Aristarchus’ editions quoted by Didymus, see West (n. 2), 61–7.
5 The plural ἐν ταῖς ἐξητασμέναις (ἐκδόσεσιν?) is problematic; already Lehrs (n. 2), 23 suggested to read a neuter ἐν τοῖς ἐξητασμένοις (ὑπομνήμασιν), which could be identifiable with the ἠκριβωμένα ὑπομνήματα mentioned elsewhere (see below). Cf. also West (n. 2), 61 n. 46.
6 Cf. Lehrs (n. 2), 21–7, followed by Ludwich (n. 2), 1.22–7, and Cohn, L., s.v. ‘Aristarchos’ (22), RE 2.1 (Stuttgart, 1896), 862–73Google Scholar, at 863–5; Erbse, H., ‘Über Aristarchs Iliasausgaben’, Hermes 87 (1959), 275–303 Google Scholar; van Thiel, H., ‘Zenodot, Aristarch und andere’, ZPE 90 (1992), 1–32 Google Scholar; id., ‘Der Homertext in Alexandria’, ZPE 115 (1997), 13–36 Google Scholar, criticized by Schmidt, M., ‘Variae lectiones oder Parallelstellen: Was notierten Zenodot und Aristarch zu Homer?’, ZPE 115 (1997), 1–12 Google Scholar.
7 R. Pfeiffer, History of Classical Scholarship. Vol. 1: From the Beginnings to the End of the Hellenistic Age (Oxford, 1968), 214–17. See also F. Montanari, ‘Zenodotus, Aristarchus and the Ekdosis of Homer’, in G.W. Most (ed.), Editing Texts. Texte edieren (Aporemata 2) (Göttingen, 1998), 1–21; F. Schironi, The Best of the Grammarians. Aristarchus of Samothrace on the Iliad (Ann Arbor, forthcoming), chapter 1.2.
8 So Montanari (n. 7), 18–19. See also F. Montanari, ‘Ripensamenti di Aristarco sul testo omerico e il problema della seconda ekdosis’, in M. Cannatà Fera and S. Grandolini (edd.), Poesia e religione in Grecia. Studi in onore di G. Aurelio Privitera (Naples, 2000), 479–86.
9 Cf. Pfeiffer (n. 7), 213.
10 Schol. Il. 12.258a (Ariston.) κρόσσας μὲν πύργων ἔρυον: ὅτι ‘κρόσσας’ ἐν μὲν τοῖς ὑπομνήμασι κεφαλίδας, ἐν δὲ τοῖς Περὶ τοῦ ναυστάθμου κλίμακας. […]. A (‘because in the commentaries [Aristarchus says that] κρόσσας [are] stepped copings, but in On the Camp [he says that they are] ladders’).
11 See, in particular, Lehrs (n. 2), 17–18, 27-8; Ludwich (n. 2), 1.66, 204-6; Erbse (n. 6), 282-4; M. van der Valk, Researches on the Text and Scholia of the Iliad, 2 vols. (Leiden, 1963-4), 2.223; F. Montanari, I frammenti dei grammatici Agathokles, Hellanikos, Ptolemaios Epithetes: in appendice i grammatici Theophilos, Anaxagoras, Xenon (SGLG 7) (Berlin and New York, 1988), 98–100; West (n. 2), 55-6, 175; Nagy (n. 2), 23-4. The scholia on Il. 2.111 have also been discussed for other reasons by K. Nickau, Untersuchungen zur textkritischen Methode des Zenodotos von Ephesos (Berlin and New York, 1977), 24, 64-6, 201 (Zenodotus had also athetized Il. 2.111-18) and by Matthaios (n. 1), 233-5 (on the meaning of ἐπίθετον).
12 The Aristonicus scholium ad loc. (Schol. Il. 9.169a) does not touch on this question but discusses another problem. However, the quotation from the ‘perfected commentaries’ recalls another note by Aristarchus discussing the similar Αἴας δ’ ὁ μέγας at Il. 16.358: Schol. Il. 16.358a1 (Ariston.) Αἴας δ’ ὁ μέγας: ἡ διπλῆ, ὅτι συγκριτικῶς λέγεται πρὸς τὸν ἕτερον Αἴαντα ὁ μέγας, ὁ Τελαμώνιος, ἐπεὶ ἐκεῖνος μείων. A (‘the diplē because Ajax son of Telamon is called ‘the great’ [ὁ μέγας] in comparison to the other Ajax since that one is lesser’).
13 E.g. Schol. Il. 4.456a (Ariston.) <ἰαχή τε φόβος τε>: ὅτι τὴν μετὰ δέους φυγὴν φόβον εἴρηκεν (‘because he calls the flight with fear φόβος’) and Schol. Il. 4.456b (Did.) <ἰαχή τε φόβος τε>: ᾿Αρίσταρχος ‘ἰαχή τε πόνος τε’· οὐ γὰρ γέγονέ πω φυγή (‘Aristarchus writes ἰαχή τε πόνος [instead of ἰαχή τε φόβος]; for the flight has not happened yet’).
14 It is uncertain whether or not μέγας was present in the second ‘revised’ ekdosis. The fact that Aristarchus’ pupils, such as Dionysodorus and Ammonius, knew this reading (cf. Schol. Il. 2.111b) may suggest that the variant μέγας was present in the second edition; yet, if so, why did Didymus omit to mention this ekdosis in his analysis of Il. 2.111?
15 If so, it is difficult to accept the suggestion by A.R. Dyck, The Fragments of Comanus of Naucratis (SGLG 7) (Berlin and New York, 1988), 224 n. 13 that syggrammata such as Πρὸς Κομανόν, Πρὸς Φιλίταν and Πρὸς Ξένωνος παράδοξον were a sort of preliminary work, composed by Aristarchus at the beginning of his career. Indeed, all the evidence (see also below) seems to suggest that the Homeric monographs represent a later stage in Aristarchus’ career, probably after the ‘perfected’ hypomnēmata.
16 As suggested by Lehrs (n. 2), 28; Ludwich (n. 2), 1.206 and 297; Erbse, ad Schol. Il. 9.18.
17 According to the reconstruction proposed here, if indeed Aristarchus considered μέγας as a reading for Il. 2.111 and 9.18 when discussing Il. 9.169, Aristonicus’ fault resulted in failing to notice such a development in Aristarchus’ analysis within the ‘perfected’ commentaries, even if it pertained to a different entry. However, Aristonicus did a good job overall in ‘summarizing’ Aristarchus’ ‘perfected’ hypomnēmata. On the other hand, Didymus confirms both his own fame as a bibliophile and his own nickname χαλκέντερος (‘brazen-guts’) by being able to quote all the available evidence where Aristarchus mentioned this variant.
18 So also West (n. 2), 52.
19 This is evidence that Aristarchus consulted manuscripts in preparing his Homeric edition(s). Yet the extent to which Aristarchus used manuscript evidence to determine his Homeric text is another hotly debated issue. See in particular Ludwich (n. 2), 1.3-8; Erbse (n. 6), 280; Pfeiffer (n. 7), 94; R. Janko, The Iliad: A Commentary Volume 4: Books 13–16 (Cambridge, 1992), 26; id., ‘Seduta di chiusura’, in F. Montanari and P. Ascheri (edd.), Omero tremila anni dopo. Atti del congresso di Genova, 6–8 luglio 2000 (Rome, 2002), 653–66, at 659; A. Rengakos, ‘Review of M.L. West, Studies in the Text and Transmission of the Iliad (Munich and Leipzig, 2001)’, BMCR 2002.11.15; G. Nagy, Homer's Text and Language (Urbana, IL, and Chicago, 2004), 87–109, and id. (n. 2), 9–21; West (n. 2), 36–7, 46–85 (in particular 69–71), and id., ‘West on Rengakos (BMCR 2002.11.15) and Nagy (Gnomon 75 , 481–501)’, BMCR 2004.04.17; Schironi (n. 7), chapter 2.2.
20 These two sets of scholia have been analysed by, among others, Erbse (n. 6), 276–7, 296–8; Montanari (n. 7), 11–18; id. (n. 8), 479–83; Nagy (n. 2), 21–33. I agree with most of Montanari's solution, even if, as will become clear, my final conclusions are slightly different from his.
21 Cf. also Suda α 1641: Ἀμμώνιος: Ἀμμωνίου, Ἀλεξανδρεύς, Ἀλεξάνδρου γνώριμος, ὃς καὶ διεδέξατο τὴν σχολὴν Ἀριστάρχου πρὸ τοῦ μοναρχῆσαι τὸν Αὔγουστον (‘Ammonius: [son of] Ammonius, of Alexandria, acquainted with Alexander, who also took over the school of Aristarchus before the reign of Augustus’). On the meaning of Ἀλεξάνδρου γνώριμος (was Ammonius a friend of Alexander Polyhistor? Or was Alexander the predecessor of Ammonius at the head of the School of Aristarchus? Or is Ἀλεξάνδρου γνώριμος a mistake for Ἀριστάρχου γνώριμος?), see A. Blau, De Aristarchi discipulis (Diss., Ienae, 1883), 6 with n. 2.
22 For the reconstruction of Aristarchus’ editions on the basis of these titles by Ammonius, see the bibliography cited in nn. 6 and 7 above.
23 If so, Aristarchus kept the first repeated line (line 396) because it was necessary after Il. 10.395: ἀνδρῶν δυσμενέων σχεδὸν ἐλθέμεν, ἔκ τε πυθέσθαι | ἠὲ φυλάσσονται νῆες θοαὶ ὡς τὸ πάρος περ (‘[Hector commanded me] to go close to the enemies and find out | whether the swift ships are guarded as before’).
24 On Nemesion's tetralogy, see Lehrs (n. 2), 31 n. 15; Erbse, ad Schol. Il. 10.397-9b (with bibliography); van der Valk (n. 11), 1.107-9; Nickau (n. 11), 260–3 (both van der Valk and Nickau also discuss this scholium and Aristarchus’ solutions).
25 This hypothesis seems to be suggested also by Montanari (n. 7), 16.
26 The stigmē is the second sign of the couple antisigma + stigmē; in that combination the stigmē can be conceived as having the function of ‘highlighting’ the problematic part in a passage because it signals lines which are transposed (as in Il. 10.397-9 ‘transposed from’ Il. 10.310-12) or tautological.
27 ‘To eliminate (ἐξελεῖν) a line’ technically is different from ‘athetizing (ἀθετεῖν) it’, as the first operation involves either the complete omission of the line from the text or its physical cancellation (with a horizontal stroke?) in the manuscript. Athetizing, on the other hand, means to mark the line with an obelos without deleting it from the manuscript, and thus represents a lighter intervention by the scholar on the text. See Schironi (n. 7), chapter 4.1, § 1.6. The only way to reconcile these two pieces of evidence is to assume that by τελέως ἐξελεῖν Ammonius meant ‘to athetize’, as often happens; see Montanari (n. 7), 16–17; in this way, Ammonius would be referring to the obeloi in the ‘revised’ ekdosis for which the scholar speaking in Schol. Il. 10.397-9b (Didymus?) does not find any reason in the hypomnēmata.
28 Montanari (n. 7), 17–18, on the other hand, thinks that the scholium of Aristonicus can be reconciled with the notes of Didymus and Ammonius as describing the same state of the text; to do so, however, we must suppose that a great part of Aristonicus’ note (discussing the athetesis and a possible solution by emending the verbs in the text) was lost.
29 That this was the original version seems to be confirmed by the fact that those lines had been athetized previously by Aristophanes (see Didymus in Schol. Il. 10.397-9a), and the only reason for this must have been grammatical, since repetition alone is not per se enough of a reason for an athetesis in the practice of the Alexandrian grammarians. Indeed, βουλεύοιτε, ἐθέλοιτε and μετὰ σφίσιν are still the readings of the vulgate.
30 Didymus’ note in Schol. Il. 10.398 (ἐν ἄλλῳ ‘φύξιν βουλεύουσι μετὰ σφίσιν οὐδ’ ἐθέλουσι’) seems to suggest that the reading was indeed present ‘in another manuscript’ (ἐν ἄλλῳ). Perhaps Didymus saw the reading in a text going back to Aristarchus but did not think it was Aristarchus’ choice in the first diorthōsis. Alternatively, we could speculate that ἐν ἄλλῳ is the result of bad epitomizing for ‘in the other edition of Aristarchus’ (ἐν τῇ ἑτέρᾳ τῶν Ἀριστάρχου).
31 Aristarchus certainly accepted the use of the third person instead of the second or the first in ‘reported orders’, as is shown by Schol. Il. 16.496a (Ariston.); see R. Nünlist, The Ancient Critic at Work. Terms and Concepts of Literary Criticism in Greek Scholia (Cambridge, 2009), 324–5.
32 Cf. Erbse, ad Schol. Il. 19.365-8a2; for a completely different (and unconvincing) interpretation of this scholium, see van der Valk (n. 11), 1.424–5.
33 Montanari (n. 7), 12.
34 The attribution of scholiastic notes, presumably derived from the VMK, to any one of the four scholars is mostly based on the content of those notes: thus scholia discussing a reading are attributed to Didymus, those treating the reasons for critical signs or for an athetesis to Aristonicus; notes focussed on prosody and accentuation of specific words most probably derive from Herodian; and those concerned with punctuation from Nicanor. Even though in the majority of the cases the content provides reliable attributions, sometimes (such as in this case) the origin of a comment in a scholium is not so easy to identify. In the present case I suggest that the content of these notes indicates a different attribution, which in addition solves the problems concerning the contradiction about Didymus’ and Aristonicus’ relative knowledge of Aristarchus’ opinions on Il. 19.365-8.
35 Even if Aristonicus is our main source for Aristarchus’ atheteseis, Didymus also mentions them occasionally; see Didymus in Schol. Il. 5.906b1; 7.443-64b1.2; 8.235b1.2; 9.694a2; 11.179-80b; 11.356b2; 13.658-9b; 17.404-25; 21.130-5a1; 24.30a.
36 On this criterion, see F. Schironi, ‘Theory into practice: Aristotelian principles in Aristarchean philology’, CPh 104 (2009), 279–316, at 297–300.
37 This would not be a unique case, as there are other scholia apportioned in this way by Erbse himself: e.g. Schol. Il. 5.700a (Did. | Ariston. | Did.); Schol. Il. 6.510-11a1 (Ariston. | ex. | Ariston.); Schol. Il. 10.306b (Did. | ex. | Did.); Schol. Il. 13.191d (ex. | Did. | Hrd. | Did. | Hrd.).
38 In particular, the term συνέπεια is always used by Aristonicus in the scholia to the Iliad (Schol. Il. 3.395, 3.423a, 9.26-31, 16.89) with only two exceptions: it is used once by Herodian (Schol. Il. 11.51b) and once in a scholium partly attributed to Didymus (Schol. Il. 7.256-7), in which, however, Aristonicus is also quoted: (Did. | Ap. H.) τοὺς στίχους τούτους οὐ προσίενται ἔνιοι, ὥσπερ οὐδὲ Ζηνόδοτος, ἀλλὰ τὸ τῆς συνεπείας οὕτως ἔχει παρ' αὐτῷ, | ὥσπερ καὶ ὁ Ἀριστόνικος (p. 132 Friedl.) ἐκτίθησιν, ἣν περιττὸν ἐνομίσαμεν γράψαι (‘some do not accept these lines, as Zenodotus does not either, but in his text the syntactic connection reads in this way, | as Aristonicus also explains, which we considered superfluous to write down’). It is obviously very difficult to decide what to attribute to Didymus and what instead is the scholiast's rephrasing of Aristonicus’ original note (Didymus’ words could actually end at ὥσπερ οὐδὲ Ζηνόδοτος and the section on the συνέπεια could be part of Aristonicus’ note excerpted by the scholiast). This is another case of a ‘composite’ VMK scholium with a difficult attribution, like Schol. Il. 19.365-8a1.
39 Indeed, an athetesis certainly did not prevent Aristarchus from commenting upon interesting features of a ‘rejected’ line, as we can see elsewhere (e.g. Aristonicus in Schol. Il. 11.767a1).
40 It is impossible, on the other hand, to suggest a relative chronology for Aristarchus’ monographs, except for one, On the Camp, which was written certainly later than the second ‘perfected’ hypomnēmata. However, I am tempted to suggest that all the monographs were composed later, after the second commentary, as if they were a sort of ‘summary’ of Aristarchus’ past exegetical activity now focussed on specific topics (but see above, n. 15).
41 In principle, both Aristonicus and Didymus could have consulted the autograph works of Aristarchus, since bookrolls could last over two hundred years; see G.W. Houston, ‘Papyrological evidence for book collections and libraries in the Roman Empire’, in W.A. Johnson and H.N. Parker (edd.), Ancient Literacies: The Culture of Reading in Greece and Rome (Oxford and New York, 2009), 233–67, at 248–51.
42 This seems to be suggested by Ammonius’ works On the Fact That There Were No Multiple Editions of Aristarchus’ Recension and On the Re-edited Recension. Cf. Erbse (n. 6), 277; Pfeiffer (n. 7), 216–17.
43 For a discussion on the relationship of critical signs, ekdosis and hypomnēma, see F. Schironi, ‘The ambiguity of signs: critical σημεῖα from Zenodotus to Origen’, in M.R. Niehoff (ed.), Homer and the Bible in the Eyes of Ancient Interpreters (Jerusalem Studies in Religion and Culture 16) (Leiden and Boston, 2012), 87–112, at 88–100.
44 See West (n. 2), 47–8 and, in particular, 48 n. 8. Cf. also Nagy (n. 2), 35 (though he thinks that at Rome there was also ‘the definitive base text of Aristarchus, marked with the master's marginal signs’, namely, Aristarchus’ final ekdosis).
45 Other works by Aristarchus reached Rome at a later point. According to Galen (PA 13), the Palatine Library had ‘the Aristarcheia, which include the two Homers’ (Ἀριστάρχεια οἵτινές εἰσιν Ὅμηροι δύο); the latter could be either the commentaries on the Iliad and on the Odyssey, or the two ekdoseis of Homer. Nicholls, M.C., ‘Galen and libraries in the Peri Alupias ’, JRS 101 (2011), 123–42Google Scholar, at 130–1 suggests that Aristarchus’ books might have come to Rome either after Octavian took Alexandria in 30 b.c.e. or (more likely in my view) around 80 c.e., when Domitian sent copyists to Alexandria to copy books in order to refurbish Roman libraries (Suet. Dom. 20.1).
46 West (n. 2), 49–50.
47 Schol. Il. 8.535-7: … τὰ αὐτὰ δὲ λέγει περὶ τῶν στίχων τούτων ὁ Δίδυμος (p. 115 Schm.), ἃ καὶ ὁ Ἀριστόνικος· διὸ οὐκ ἐγράψαμεν τὰ Διδύμου (‘about these lines Didymus says the same things as Aristonicus; therefore, we did not write down Didymus’ words’); and Schol. Il. 15.86c: ταὐτὰ ὁ Δίδυμος (p. 116 Schm.) τῷ Ἀριστονίκῳ λέγει περὶ τῆς γραφῆς τῆς δέπασσι (‘regarding the orthography of δέπασσι Didymus says the same things as Aristonicus’).
48 On this scholium, see above, n. 38.
49 See West (n. 2), 46–85.
Full text views reflects PDF downloads, PDFs sent to Google Drive, Dropbox and Kindle and HTML full text views.