1. Genette, G., Narrative discourse (1980)Google Scholar, trans, from Figures III (1972) 161–211Google Scholar, Nouveau Discours du Récit (1983) 43–55Google Scholar: Bal, M., Narratology: introduction to the theory of narrative (1985, from the second Dutch, ed. of 1980) 100–18Google Scholar. There is a good survey of the older literature on ‘point of view’ in Conte, G. B., ‘Saggio di interpretazione dell'Eneide. Ideologia e forma del contenuto’ MD 1 (1978)Google Scholar = Virgilio, Il genere e i suoi confini (1984) 55–119Google Scholar = The rhetoric of imitation (1986) 141–84Google Scholar, at Virgilio 68 n.10 = Rhetoric 154 n.10. Amongst modern surveys, Pugliatti, P., Lo squardo nel racconto. Teorie e prassi del punto di vista (1985)Google Scholar is better than Lintvelt, J., Essai de typologie narrative. Le ‘point of view’, Théorie et analyse (1981)Google Scholar. Of particular interest for classicists are Segre, C., ‘Punto di vista e plurivocità nell'analisi narratologica’, Atti del convegno internazionale: Letterature classiche e narratologia, Selva di Fasano, 6–8 Ott. 1980Google Scholar (Ist. di Filol. Lat. dell'Univ. di Perugia 1981) 51ff. and Fusillo, M., Il romanzo greco (1989) 111–78Google Scholar (with much useful bibliography).

2. de Jong, I. J. F., Narrators and focalizers: the presentation of the story in the Iliad (1987)Google Scholar.

3. For some controversial aspects of Bal's use of focalisation see Bronzwaer, W., ‘Mieke Bal's Concept of Focalization’, Poetics Today 2.2 (Winter 1981) 193–201CrossRefGoogle Scholar, with her reply ‘The Laughing Mice or: On Focalization’ in the same issue (202–10). For a radical critique of the basic narratological division, see Smith, B. Herrnstein, ‘Narrative versions, narrative theories’ in Mitchell, W. J. T. (ed.) On Narrative (Chicago n.d., originally Critical Inquiry 7 (1980–1981)) 209–32Google Scholar with the reply by Seymour Chatman 258–65. Gian Biagio Conte points out to me one danger in the use of ‘focalisation’ rather than ‘point of view’, in explaining his own preference for the latter term:

… per me non è solo un procedimento di tecnica letteraria (più o meno alessandrina, più o meno apolloniana, più o meno di ‘stile soggettivo’) ma … è una ‘relazione di verità’ … Cosi il punto di vista diventa il centro stilistico-filosofico, ideologico-espressivo, del discorso virgiliano. Questa è la ragione per cui uso ancora il termino ‘punto di vista’ e non quella di ‘focalizzazione’, in quanto lo redefenisco in termini contenutistici e non formali, o meglio di ‘forma dei contenuti’ (cioe in termini che riguardano la forma che i contenuti assumono nel discorso virgiliano, vale a dire come essi sono strutturati). Dietro c'è più l'esperienza della critica russa (e postpraghese: Mukarowskj!) che non l'influenza dello strutturalismo francese …’

Certainly the wider connotations of ‘point of view’ are welcome, but its familiarity in English as a term of art paradoxically makes it easier for the unfamiliar ‘focalisation’ to achieve the recognition of the ideological dimension which I agree is vital.

4. Cf. Bal, M., ‘Notes on narrative embedding’, Poetics Today 2.2 (Winter 1981) 41–59CrossRefGoogle Scholar.

5. De Jong(n.2) 118–22.

6. De Jong (n.2) 136–45. Cf. her earlier article, ‘Homeric words and speakers: an addendum’, JHS 108 (1988) 188–9CrossRefGoogle Scholar.

7. De Jong (n.2) 121.

8. Hopkinson, N., Callimachus, Hymn to Demeter (1984) 157Google Scholar.

9. McKay, K. J., Erysichthon: a Callimachean comedy (1962)Google Scholar.

10. McKay (n.9) 95.

11. Chafe, W. L., ‘Givenness, contrastiveness, definiteness, subjects, topics, and point of view’, in Li, C. N. (ed.), Subject and Topic (1976) 25–55 at 54Google Scholar, quoted in Brown, G. and Yule, G., Discourse Analysis (1983) 148CrossRefGoogle Scholar.

12. See especially Kuno, S., ‘Subject, theme, and the speaker's empathy – a re-examination of relativization phenomena’, in Li, (n. 11) 417–38Google Scholar.

13. Kuno (n.12) 433.

14. Perutelli, A., ‘Registri narrativi e stile indiretto libero in Virgilio (a proposito di Aen. 4. 279 sgg.)’ MD 3 (1979) 69–82Google Scholar; see also now C. J. Mackie, The characterization of Aeneas (1988) 7,79–81. There is a considerable body of literature on FID in Latin: see the works cited in Hofmann, J. B. and Szantyr, A., Lateinische Syntax und Stilistik (1965) 362Google Scholar, especially Regula, M., ‘Streifzüge auf dem Gebiet der lateinischen Syntax und Stilistik; V. Besondere Darstellungsformen der Reproduktion’, Glotta 31 (1951) 90–2Google Scholar.

15. Ron, M., ‘Free indirect discourse, mimetic language games and the subject of fiction’, Poetics Today 2.2 (Winter 1981) 17–39, at 35–6CrossRefGoogle Scholar. I cannot resist the observation that one of those who read a draft of this piece where the boundaries of the quotation here were not clear took the second paragraph to be by me, and thought it the best part of the article: ‘who speaks?’ is indeed problematic. My own additions are in angle brackets: the square bracketed ‘[sic]’ is Ron's.

16. Conte (n.1).

17. A. Traina, Enciclopedia Virgiliana s.v.superbia.

18. Knight, W. F. Jackson, ‘Animamque superbam’, CR 46 (1933) 55–7, at 57 n. 6Google Scholar: ‘the thought is Juno's: and here too rex and superbus are associated. To Juno, Roman rule may well have seemed a tyrannical despotism.’

19. See e.g. Austin ad loc.

20. Henry, J., Aeneidea 1 (1873) 217–18Google Scholar: ‘Add to which, that the brief GENUS INVISUM, the hateful race, thrown in between the two more particularly detailed causes, expresses a virus, a concentration of feeling, which had only been weakened by particularization … [the] meaning being, not that the Trojan race was hateful to Juno, because descended from Electra or from Dardanus, but that the race was hateful to her, was an abomination to her (no matter for what reason), and that therefore in the Trojan war she took part against those who were of that race exactly as at present she takes part against and persecutes Aeneas and his companions because they are of that race, that GENUS INVISUM, that hated brood …’

21. Aen. 3.1–5.

22. Iliad 13.621; 21.414, 224.

23. Inf. 1.75.

24. See the scholia on Iliad 15.94 with Erbse ad loc. and on 5.881; Buttmann, P., Lexilogus (1837) 185–92Google Scholar; Ebeling, H., Lexicon Homericum (1885)Google Scholar s.v.

25. Iliad 13.621.

26. Cf. Seneca, , Troades 4–6Google Scholar (Hecuba) me uideat et te, Troia: non umquam tulit / documenta fors maiora, quam fragili loco / starent superbi, with Fantham on 1–4; Ovid, Met. 15.420–31 (where the discussion of the authenticity of lines 426–30 has taken turns not irrelevant to the question of the focalisation here in Virgil: see Bömer ad loc); in general, cf. Theognis 1103–4; Catullus 51 (with Ferrari, W., ‘Catulls Carmen 51’, in Heine, R. (ed.), Catull (1975), 241–61 at 253–6Google Scholar); Servius Sulpicius Rufus ap. Cic. Fam. 4.5. 4; Seneca, Ep. 91.9–12; Alexiou, M., The ritual lament in Greek tradition (1974) 83–101Google Scholar; Hall, E., Inventing the barbarian (1989) 131–2Google Scholar (who refers to Fittipaldi, M. F., ‘The fall of the city of Troy and its significance in Greek poetry from Homer to Euripides’, diss. Yale 1979Google Scholar, which I have not seen).

27. Bowie, A., ‘The death of Priam: allegory and history in the Aeneid’, forthcoming in CQ (1990)Google Scholar.

28. Servius on Aen. 2.557, Pompei tangit historiam.

29. Cf. Monaco, G., ‘Bonus Ancus‘ GIF 24 (1972) 245–50Google Scholar; Conte, G. B., ‘Il trionfo della morte e la galleria dei grandi trapassati in Lucrezio III 1024–1053’, SIFC 37 (1965) 114–32Google Scholar; Nisbet and Hubbard on Hor. Carm. 1.28.7.

30. Cf. Wankel, H., ‘Alle Menschen müssen sterben: Variationen eines Topos der griechischen Literatur’, Hermes 111 (1983) 129–54Google Scholar; Pattoni, M. P., ‘L'exemplum mitico consolatorio: variazioni di un topos nella tragedia greca’ SCO 38 (1988) 229–62 at 229 n. 1Google Scholar.

31. Cf. Trag. adesp. 372 ([Plut., ] Cons. ad Apollon. 15, 110dGoogle Scholar, with Hani ad loc); Marc. Aur. 6.47 with Rutherford, R. B., The meditations of Marcus Aurelius (1989) 128Google Scholar.

32. I stress ‘obviously’: it is of course possible to see signs in the Aeneid of a critical attitude to Troy, and to interpret these in various ways, e.g. as a sign that Troy is typologically transcended by Rome (cf. Cairns, F., Virgil's Augustan epic (1989) 127–8CrossRefGoogle Scholar, and contrast the interpretations of Numanus Regulus’ speech by Horsfall, N., ‘Numanus Regulus: ethnography and propaganda in Aen., ix, 598f’, Latomus 30 (1971) 1108–16Google Scholar and Lyne, R. O. A. M., Further voices in Vergil's Aeneid (1987) 202Google Scholar). But the obvious stress in the Aeneid is on the pathos of Troy's fall.

33. Bowie (n.27).

34. Cf. 1.697–8 aulaeis iam se regina superbis / aurea composuit sponda mediamque locauit. As always, more than one story can be told of the focalisation here.

35. Ennius, , Andromacha frr. 92–9Google Scholar Vahlen = 87–94 Jocelyn, who collects Cicero's references.

36. Serv. on 1.726 (quoted from Cicero), 2.241 ‘uersus Ennianus’.

37. 2.499–505.

38. I take examples like Plautus, , Asin. 11Google Scholar / Trin. 19 uortit barbare and Cicero, , Orat. 160Google Scholar to be examples of humorous shifts of focalisation. The usage in Greek tragedy is complex, but obviously use by Persians in Aeschylus' Persae is a special case and examples like Eur., Hec. 1200Google Scholar may involve focalisation switches: on the whole question, see Hall (n.26) passim. It may be that with Peerlkamp and Forbiger we should see the barbarico … auro here as the spoils taken from other barbarians (cf. Hall (n.26) 212 on Eur., Troades 477–8Google Scholar) but the reader cannot fail even so to think of the Trojans too as ‘barbarian’. Mackail rightly draws attention to Milton's allusion in P.L. 2.1–4, where Satan is depicted on a throne ‘which far / outshone the wealth of Ormus and of Ind / or where the gorgeous East with richest hand / showers on her kings barbaric pearl and gold’. The OED quotes this as the earliest instance of ‘barbaric’ in the sense ‘pertaining or proper to barbarians or their art; in the characteristic style of barbarians as opposed to that of civilized countries or ages’: it would be strange to take it completely ‘neutrally’, despite the italicisation as a proper name in the early editions (see Fowler ad loc). Pope, 's imitation in Temple of Fame 94Google Scholar tones down the force of the adjective in Milton.

39. Cic. Tusc. 3.45 exaggeratis … regiis opibus, quae uidebantur sempiternae fore.

40. Cf. OLD 1.c, 2.b; Traina (n.17) 1072.

41. Gillis, D., Eros and death in the Aeneid (1980) 138Google Scholar.

42. Traina (n.17) 1073.

43. Compare also the difference between 8.722 uictae longo ordine gentes and 2.766 pauidae longo ordine matres.

44. Austin on Aen. 6.468, comparing 3.344–5 and Georg. 3.516–17; Mackie (n.14) 133 n.1. But cf. TLL III.1055.26ff.; Seel, O., ‘Um einen Vergilvers (Aeneis, VI, 468)’, in Hommages à M. Renard (1969) 677–88Google Scholar.

45. For the term, see Levelt, W. J. M., ‘The speaker's linearization problem’, in Longnet-Higgins, H. C., Lyons, J., and Broadbent, D. E. (eds.), The psychological mechanisms of language (1981) 305–15Google Scholar.

46. Odyssey 6.232–5, 23.159–62. For the extensive bibliography on this repetition, see Heubeck on the latter passage; for its interpretation, note Hainsworth on the former, ‘The modern reader, who is trained in such matters, will probably recall the present use, with Nausicaa's reaction, on reading the second, and see in it a symbol of Odysseus as a bridegroom: it is possible that some such thinking unconsciously affected the poet's choice of imagery. However the Homeric audience were too thoroughly accustomed to repetition for any particular instance to have been significant to them; the repetition therefore cannot have been conscious.’ This modern reader would stress the way Penelope shows Odyssean self-control in not reacting. This is not irrelevant to the Aeneid.

47. For Prometheus and the Caucasus, see Roscher, Lex. Myth. s.v. Prometheus 3042; ‘il Caucaso gelato’ is the phrase of Tasso Gerusalemme Liberata 16.56. 3, quoted by Pease.

48. Pease on Aen. 4.366 duris.

49. Heuzé, P., L'image du corps dans l'oeuvre de Virgile (1985) 567Google Scholar.

50. Marpessa was a mountain on Paros, the main site of the quarries: cf. Steph. Byz. s.v. Despite cautes Norden curiously translates ‘starr wie ein Marmorbild’, referring to the common comparison of women to statues (on which see Nisbet and Hubbard on Hor. Carm. 1.19.6. and McKeown on Ovid Am. 1.7.51). This woman has not allowed herself to be fashioned: whether that is good or bad depends upon one's point of view.

51. It is instructive to note how Norden changed his mind on the sense of 6.469 between the first and second editions of his commentary. Having noted the erotic parallels for the line (Eur., Med. 27f.Google Scholar, Theocr. 2.112, Musaeus 160, A. P. 5.252) he continues: ‘Aber wenn ich aus dem formalen Anschluss V.s an solche Vorbilder in der 1. Aufl. folgerte, dass nach der Intention des Dichters Dido den Aeneas noch liebte (odit et amat), so trug ich durch diese Annahme, wie mir Heinze brieflich bemerkte, einen falschen Zug in das Bild hinein: sie ist mit ihm nur inimica (472), das Bild der Dido ist mit heroischer Grösse gezeichnet, alles Sentimentale fehlt.’

52. Heinze, R., Virgils epische Technik ed. 3/4 (1957) 370–3Google Scholar.

53. Rosati, G., ‘Punto di Vista narrativo e antichi esegeti di Virgilio’, Ann. Scuola Norm. di Pisa, Classe di Lett, e Fil. (1979) 539–62Google Scholar; Lazzarini, C., ‘Elementi di una poetica serviana. Osservazioni sulla costruzione del racconto nel commentario all Eneide, II’, SIFC 82 (1990) 241–60 at 248–9Google Scholar.

54. Otis, B., Virgil: a study in civilized poetry (1964)Google Scholar.

55. La Penna, A., ‘Sul cosidetto stile soggetivo e sul cosidetto simbolismo di Virgilio’, Dialoghi di Archeologia 1 (1967) 220–44Google Scholar; Conte (n.1.).

56. Perutelli, A., ‘Similitudini e stile “soggetivo” in Virgilio’, Maia 24 (1972) 42–60Google Scholar.

57. Bonfanti, M., Punto di Vista e modi della narrazione nell'Eneide (1985)Google Scholar.

58. See the works of Pugliatti and Segre cited in n.1.

59. La Penna (n.55) 227.

60. La Penna (n.55) 223.

61. Rosati (n.53) 540.

62. De Jong (n.2) 227.

63. Conte (n.1) 149.

64. Conte (n.1) 150.

65. Conte (n.1) 153.

66. Conte (n.1) 161–2.

67. There are of course traces of alternative ways of reading epic elsewhere in Augustan literature, and indeed in the Aeneid itself: and the tragic reading of Homer begins with tragedy itself. I do not want to save ancient authors from monologic writing by making ancient audiences monologic readers.

68. Feeney, D. C. rev. Conte (n.1), JRS 79 (1989) 206–7Google Scholar.

69. Lyne, R. O. A. M., Words and the poet (1989) 165Google Scholar.

70. Lyne, refers to his earlier Further voices in Vergil's Aeneid (n.32) 227Google Scholar.

71. For some recent reflection on (and deconstruction of) the apparent choice between a formalist view of the literary as ‘set over against power’ versus a putatively historicist view of the literary as ‘one of power's essential modes’ as formulated in New Historicism, see Porter, C., ‘History and literature: “After the New Historicism”’, New Literary History 21 (1990) 253–72CrossRefGoogle Scholar, with the reply of R. Fraden, 273–8. But I have been most helped in thinking about ‘oppositional’ criticism by Maslan, M.'s lucid attack on it in ‘Foucault and Pragmatism’, Raritan 10 (1990) 94–114Google Scholar.

72. I am indebted for help at various stages to Angus Bowie, David Cram, Peta Fowler, John Henderson, Patricia Johnston, Stephen Marsh, and Alessandro Schiesaro; and above all to Gian Biagio Conte and Oliver Lyne for paternal indulgence.