2. So Kirk, ad loc., citing the same line at 20.159 where it introduces the duel between Aeneas and Achilles, and at 23.814 where it ominously precedes the clash between Diomedes and Telamonian Ajax.

3. As in the introductory study by Bowra, C. M., Homer (London, 1972)Google Scholar, where the scene between Glaucus and Diomedes ‘fits into the battle by providing a moment of delightful relaxation’ (p. 68).

4. Direct influence of the Iliad on War and Peace is very limited. Tolstoy didn't read the poem in Greek until the winter of 1870–1, a year after the publication of the last part of the novel. For Tolstoy's dedication to Greek at this time see Troyat, H., Tolstoy (Harmondsworth, 1970)Google Scholar: ‘[Tolstoy/ revelled in Homer, discovered Plato and said the originals were like “spring-water that sets the teeth on edge, full of sunlight and impurities and dust-motes that make it seem even more pure and fresh”, while translations of the same texts were as “boiled, distilled water”. Sometimes he dreamt in Greek at night. … “Now I firmly believe” he said to the poet Fet, “that I shall write no more gossipy twaddle of the War and Peace type.”’ (p. 454).

5. Redfield, J., Nature and Culture in the Iliad (Chicago, 1975)Google Scholar writes well on this (without reference to Owen or Tolstoy), but forces too sharp a disjunction between culture and nature: ‘Culture, which appears to us in our social lives so solid and enduring, reveals itself on the battlefield for what it is. For the warrior, culture appears as a translucent screen against the terror of nature’ (p. 103). But how else, except through the broader definition of culture, is our apprehension of ‘the terror of nature’ to be made articulate?

6. For an extended critical comparison between the Iliad and War and Peace see Steiner, G., Tolstoy or Dostoevsky (Harmondsworth, revised edition 1967), especially pp. 71–9Google Scholar. The comparison in Christian, R. F., Tolstoy, a Critical Introduction (Cambridge, 1969)Google Scholar is appallingly ill-informed about Homer (e.g., ‘An epic poem in the classical, Homeric sense is essentially an adventure story told for entertainment … It is not concerned with the psychology of its heroes’, pp. 71–2). There is a brief comparison in Greenwood's, E. B. chapter ‘Tolstoy and Religion’ in Jones, M. (ed.), New Essays on Tolstoy (Cambridge, 1978)Google Scholar.

7. See Nagy, G., The Best of the Achaeans: Concepts of the Hero in Archaic Greek Poetry (Baltimore and London, 1980), pp. 69–75Google Scholar and Peradotto, J., Man in the Middle Voice: Name and Narration in the Odyssey (Princeton, 1990)Google Scholar, especially chapter 4 ‘Polytropos: the naming of the subject’.

8. Kirk adduces this parallel in his note on 6.119, but makes nothing of the difference between the two episodes. For a brief comparison see Reinhardt, K., Die Ilias und ihr Dichter (Göttingen, 1961), p. 513Google Scholar.

9. For an analysis of the construction of the Diomedes episode in Book 6, with a different emphasis from that adopted here, see Andersen, Ø., Die Diomedesgeslall in der Ilias (Oslo, 1978), pp. 96–107Google Scholar.

10. To kill two such warriors would be a κῦδος conferred on him by Athene (5.260).

11. The scholiasts on 6.123 (ii.151 Erbse) typically go for narrative integrity here, supposing that the problem over Glaucus has arisen because the mist lifted by Athene at 5.127 has descended again.

12. e.g., θνητῶν ἀνθρώπων ending the lines at 1.339, 8.404, 20. 204, 220, 233; καταθνητῶν in this position is more characteristic of the Odyssey (see Kirk, ad loc).

13. An elaboration of Homer's version of the story can be found in Nonnus, , Dionysiaca 20Google Scholar. A different version emerges for us in the fourth stasimon of the Antigone (955–65) where it supplies the second of three examples of mortals who suffer incarceration. This may have been influenced by the treatment in Aeschylus's, Lycurgea triology (T.67–9 and p. 234 Radt)Google Scholar.

14. Cf. the analysis by Lohmann, D., Die Komposition der Reden in der Ilias (Berlin, 1970), p. 12CrossRefGoogle Scholar.

15. σεύω: cf. 20.189 and Ebeling, Lexicon Homericum s.v. I(b); θείνω: cf. 10.484, 20.481 and 21.21 where Lycurgus's ox-whip is replaced by the sword.

16. ἀνδρόϕονος occurs thirteen times in the Iliad, but 6.34 is its only application to a mortal other than Hector, the only other reference not to Hector being to Ares at 4.441.

17. Most recently at 5.862–3, but cf. also 7.215, 19.14, and 20.44.

18. A curious parallel is the nymphs' reception of Diomedes παρὰ πατρὸς ἄνακτος/δεξάμενοι κόλποισι in Hymn 26.4.

19. Kirk (p. 171 on 119–236) repeats the view he expressed in Songs of Homer (Cambridge, 1962), pp. 164–6Google Scholar, that the Diomedes/Glaucus episode is ‘inorganic’. I argue rather that the abbreviated narrative is evidence of how the poet selected from an already familiar story points to support the analogizing which binds the episode solidly into its context here.

20. For the possible meanings πεῖραρ here see Kirk on 143; the scholia may (as Kirk says) ‘remain silent on the topic’ here, but in discussing ἔριδος κρατερῆς καί ὁμοιίου πτολέμοιο / πεῖραρ at 13.358–9 (iii.468–9 Erbse) they interestingly suggest a μικτὴ ἀλληγορία combining the notions of τέλος and δεσμός. This is unfortunately not taken up in the one-sided entry in LSJ s.v. II.2, whose reasoning is explained and supported in Merry and Riddell's long note on Od. 12.51 (i.5O8). The shared meaning ‘bonds that tie us to our destruction’ for όλέθρον πείρατα is an especially powerful image in view of what I suggest will be Glaucus' misinterpretation of his situation here.

21. For this sense of σῆμα see LSJ s.v. 5.

22. Being aware of it is not, of course, the same thing as interpreting it correctly.

23. See, for example, Macleod's, C. W.Iliad, Book XXIV (Cambridge, 1982), p. 12Google Scholar. Also, Edwards, M. W.’ analysis of the simile in Homer, Poet of the Iliad (Baltimore, 1987)Google Scholar ultimately sees in it ‘the poet's own consciousness of the mortality of human greatness and of the consolation of the continuance of the human race’ (pp. 204–5). But see the important qualification of this view in Lynn-George, M., Epos: Word, Narrative and the Iliad (London, 1988)CrossRefGoogle Scholar: ‘[the simile's] image of survival is one of continuity without continuity. Where there is repeated renewal from the irrevocable, that renewal is tinged with futility in its reproduction of finality’ (p. 199).

24. Interestingly, when Apollo takes up the simile later (21.464–6) to illustrate to Poseidon the difference between the life of men and that of the gods, he emphasizes the anonymity of mortals (ϕύλλοισιν ἐοικότες, 464) and suppresses mention of recurring generations in predictable cycles, which might suggest that the human race could to a limited extent share collectively in the immortality the gods jealously guard for themselves.

25. Lynn-George (see n. 23), p. 200. For the wider application of the simile see Griffith, M., ‘Man and the Leaves: a Study of Mimnermus fr. 2’, CSCA 8 (1975), 73–88Google Scholar.

26. The same phrase describes the public status of Aeneas in an identity-σῆμα at 20.214.

27. For the narratival significance of the σῆματα on the Argive shields see Thalmann, W. G., Dramatic An in Aeschylus's Seven Against Thebes (Yale, 1978), pp. 107–23Google Scholar.

28. M. Lynn-George (see n. 23) gives a stimulating interpretation of the ambivalent significance of types of σῆματα in the Iliad (pp. 255–66).

29. For the identification see Aristonicus's note (ii.157 Erbse).

30. Homer suppresses mention of the move to Tiryns, because (I suggest) he wants to stress Bellerophon's continuing innocence, and so avoid reference to the miasma which drove him out of Corinth. Anteia is Sthenoboea in Euripides' play on the myth (cf. Loeb, D. L. Page'sGreek Literary Papyri, pp. 126–7Google Scholar).

31. Cf. the σῆμα which Deianira sends to authenticate her gift of the robe to Heracles (Soph, . Trach. 614Google Scholar).

32. ‘The only definite reference in Homer to writing’, Kirk on 168–9. Kirk also discusses the possible identification of the πίναξ or ‘folded tablet’.

33. Cf. 2.308, 353, 4.381 (and later 8.171, 9.236, etc.). The one exception is its use in the sense of ‘tomb’ at 2.814.

34. I take the suppression of any reference to Pegasus here (see West, M. L. on Hesiod, , Theogony 325Google Scholar) to be a narratival ploy to keep mirabilia to a minimum in the whole account.

35. Though Bellerophon had rashly been willing to take on the Chimaera, which had a θεῖον γένος (180).

36. Leaf's proposed transposition of 200–2 to follow 205 is discussed with qualified approval by Kirk, ad loc.

37. Macleod (see n. 23) considers Bellerophon's reversal of fortune to be the really significant point in Glaucus' story (pp. 11–12). I argue that the function of this reversal in the episode as a whole is a more complex matter.

38. See Lohmann (a 14), pp. 89–91.

39. For the death of Glaucus see Apollod, . Epit. 5.4Google Scholar (p. 203 Wagner), Hyg., Fab. 113Google Scholar, and Quint. Sm. 3.277.

40. The ‘foreshadowing’ of external events within the narrative of the poem is discussed by Edwards, M. W., The Iliad: a Commentary, Volume V: Books 17–20 (Cambridge, 1991), pp. 7–10CrossRefGoogle Scholar. One important feature which distinguishes Glaucus’ situation from that of Achilles is, as Edwards points out, that the latter ‘is unique as the only character who knows in advance that his death is imminent’ (p. 7, my italics).

41. We need to remember that Diomedes has already responded positively to his own cautionary tale by modifying his behaviour accordingly.

42. See Kirk's, analysis of his ‘modified ring-form’ on p. 188Google Scholar. For the ξενίη as ‘an integral part of the heroic institution of guest-friendship’ see Kirk on 218–19 and the important discussion by Donlan, W., ‘The Unequal Exchange between Glaucus and Diomedes in the Light of the Homeric Gift-Economy’, Phoenix 43 (1989), 1–15CrossRefGoogle Scholar (esp. 6–7).

43. ‘No critic, ancient or modern, has satisfactorily explained this bizarre incident and its unexpected change in ethos’ (Kirk on 234–6, unaware of Donlan's article). Kirk's view of the episode – ‘a fantasy which the audience will recognize as such and which has no particular bearing on the “real” roles and characters of the two participants’ (p. 191) – carries our understanding no further. The view is sustainable only if we accept the notion that Homer ‘withdraws for a moment from his regular narrative mode’ (Kirk, p. 191), precisely the opposite of what I suggest he is in fact doing. Donlan (see n. 42, pp. 1–2) points to the fatal flaw in the sociological theory of Calder, W. M. III (‘Gold for Bronze: Iliad 6.232–36’Google Scholar, GRBS monograph 10 [1984], 31–5Google Scholar) that Glaucus and Diomedes belong to a ‘Gift-Economy’ in which ‘superiority in gift-giving equates to superiority in social prestige’, and that Glaucus outgave Diomedes on purpose to display his superiority. As Donlan asks (p. 2), in such a context why should Homer present this as an incident of divinely inspired stupidity?

44. Donlan (see n. 42), 13–15. As Donlan states, the superiority of Diomedes in the episode had already been asserted by Craig, J. D., ‘xPƳΣΕΑ XAΛΚΕΙΩΝ’, CR 17 (1967), 243–5Google Scholar.

45. The significance of the parallel is noted by Donlan (see n. 42), 11.