¹ Friedman, N., ‘Point of View in Fiction: the Development of a Critical Concept’, PMLA 70 (1955), 1160–84CrossRefGoogle Scholar; Booth, W. C., The Rhetoric of Fiction (1961), 149–65Google Scholar; Hägg, T., Narrative Technique in Ancient Greek Romances (1971), 112–37Google Scholar; Beck, R., ‘Some Observations on the Narrative Technique of Petronius’, Phoenix 27 (1973), 42–61CrossRefGoogle Scholar; Smith, W. S. Jr, ‘The Narrative Voice in Apuleius’ Metamorphoses’, TAPA 103 (1972), 513–34Google Scholar; Paardt, R. Th. van der, ‘Various Aspects of Narrative Technique in Apuleius’ Metamorphoses’, in Aspects of Apuleius’ Golden Ass, edd. Hijmans, B. L. Jr and Paardt, R. Th. van der (1978), pp. 76–9Google Scholar. Scholes, R. and Kellogg, R., The Nature of Narrative (1966), pp. 242–7, consider ancient narratives.Google Scholar

² Booth, preface: ‘I am aware that in pursuing the author’s means of controlling his reader I have arbitrarily isolated technique from all of the social and psychological forces that affect authors and readers. For the most part I have had to rule out different demands made by different audiences in different times.’ Hägg, pp. 17–18: ‘The function of the narrative is described, in the first stage, without regard to whether a technique used is intentional or unintentional, original or traditional’.

³ Like, I think, most classicists, I am concerned to apply to the work of art only a criticism which the author might have recognized and in terms of which one may presume he composed the work. Those who are concerned about the ‘Intentionalist Fallacy’ or who practise ‘Rezeptionsästhetik’ will think differently, but run the risk of changing the meaning of the work of art from generation to generation.

⁴ ‘Point of View’ analysis is concerned with the manipulation of the supply of information from the events narrated via witness and narrator to the reader; the manipulation consists primarily in restrictions; but if the narrator is omniscient, then the supply of information is not restricted and is relatively unmanipulated. Consequently, it is authors who deviate more than most from omniscient narration who attract most attention from this form of criticism.

⁵ Plato, Tht. 189e, with Scholes and Kellogg, p. 180.

⁶ Modern novels, notably those of Kafka, sometimes display a device known as the ‘third-person reflector’, where the supply of information to the reader is diminished in the same way as in a first-person narration, but instead of a first-person narrator there is a third-person he/she through whose eyes the events are seen. This device leads, I think, to a rather negative treatment of the choice of person in which to tell events in some modern theorists: ‘Perhaps the most overworked distinction is that of person. To say that a story is told in the first or third person will tell us nothing of importance unless we become more precise’, Booth, p. 150.

⁷ Complete conspectus of narrators: Hijmans, B. L. Jr in Aspects, p. 113Google Scholar. Role of Lucius: Riefstahl, H., Der Roman des Apuleius (1938), p. 21. Protagonist and witness: Friedman 1174–5 and van der Paardt 77.Google Scholar

⁸ Junghanns, P., Die Erzählungstechnik von Apuleius’ Metamorphosen und ihrer Vorlage (1932), p. 3.Google Scholar

⁹ cf. Junghanns, pp. 118–20.

¹⁰ Outstanding example in van der Paardt, p. 89 n. 16.

¹¹ A nice example of desperate measures in B. Callison, Flock of Ships: the dying first-person narrator writes the novel, somewhat implausibly, on his death-bed.

¹² Junghanns, p. 10 n. 8.

¹³ Historia Apollonii Regis Tyri, init.

¹⁴ This approach may explain some, but not all (not Lucius or Psyche), late naming in Apuleius; cf. Brotherton, B., ‘The Introduction of Characters by Name in the Metamorphoses of Apuleius’, CPh 29 (1934), 36–52Google Scholar; Tatum, J., Apuleius and The Golden Ass (1979), p. 27 and n. 5.Google Scholar

¹⁵ Observed in this connection by van der Paardt, p. 77.

¹⁶ Only ‘likely’ because a third-person narrator can, if he wishes, surrender his omniscience temporarily for effect: in particular, it is a favourite move of Heliodorus, e.g. 1. 8 init., I am aware that the distinction to be made is not so much between two meanings of scilicet (they probably felt the same to a Roman) as between two uses, but have overlooked philosophical precision in the interests of clarity.

¹⁷ It would be stretching the reader’s confidence too far not to list the examples, but too tiresome to argue at length. Speculation to preserve perspective: 3. 26; 4. 3, 10, 12 (ter), 23; 6. 29; 7. 25, 28; 8. 7, 16, 25, 26; 9. 11, 42; 10. 1, 5, 17, 34 (bis); 11. 1, 16, 20, 27. Irony: 1. 12, 15; 3. 27; 4. 19; 5. 23, 24, 25, 26, 29, 31; 6. 1, 8, 9; 7. 16, 18, 27; 8. 24, 26 (bis), 28, 30; 9. 3.

¹⁸ At 4. 9; cf. my comments in CR n.s. 29 (1979), 70.

¹⁹ In compiling this table, I have used J. Segebade & E. Lommatzsch, Lexicon Petronianum (1898), for the instances of scilicet but not for their classification and have accepted (without complete conviction) the deletion of two examples under the heading ‘Part of Speaker’s Knowledge’. I have used the Loeb as an easily comparable format for the two authors.

²⁰ On irony, see Riefstahl, pp. 64 ff.; Apuleius employs it down to the smallest detail, e.g. the word bonus, cf. Scobie, A., Apuleius, Metamorphoses I, a Commentary (1975), p. 96. Tatum, pp. 164–5, clearly thinks parrot glossectomy is typical of Apuleius.Google Scholar

²¹ Hägg, pp. 321–2.

²² Flor. 2. Apuleius, ‘Lucius of Patras’ and emotions: Junghanns, pp. 26 n. 32, 28 f., 33 and index s.v. Psychologie; Riefstahl, pp. 23, 61 ff.

²³ Substantiality of the narrator Lucius: Riefstahl, p. 22.

²⁴ ‘He thinks of one thing at a time in its close-up aspect, ever intent on his showmanship’, Perry, B. E., The Ancient Romances (1967), p. 254.Google Scholar

²⁵ Useful, if brief, observations in Scholes and Kellogg, p. 247; more developed account in Hägg, pp. 114, 120, 124–6.

²⁶ Presentation of the novel as history is clearly a choice of point of view readily available in a culture of which historiography was so great a part; but it scarcely has the genetic implications for the origins of the novel that used to absorb scholars’ attention: see Trenkner, S., The Greek Novella (1958), pp. 180–1, with bibliography, p. 180 n. 2Google Scholar. Heine, R., Untersuchungen zur Romanform des Apuleius, diss. Göttingen (1962), p. 29, is to my mind right to see it as a formal question and, in the case of Chariton, as a misplaced attempt at realism.Google Scholar

²⁷ Fr. 1, fr. 27 Habicht.

²⁸ Photius, Bibl. cod. 129, with Apul. Met. 1. 1, compared well by Scobie (1975), p. 65 (except that δι⋯πoρoι means ‘several’, not ‘various’); see also H. van Thiel, Der Eselroman (1971), i. 44; M. Molt, Ad Apulei Madaurensis Met. Librum Primum Commentarius Exegeticus, diss. Groningen (1938), ad loc.; P. Vallette, Apulée, les Métamorphoses (1940), i. xvii and esp. p. xvii n. 3, which neatly counters the starkly negative attitude of B. E. Perry, CPh 18 (1923), 229 ff., esp. 230 n. 2.

²⁹ Was it done in the manner of Achilles? If so, it would be difficult to explain how the name of the hero of the ass story, Lucius of Patras, became that of the author – cf. R. Helm, Apulei Opera II. 2 (1910), p. vi and n. 1, H. J. Mason in Aspects (1978), p. 2.

³⁰ This has been a source of great difficulty: Helm (1910), pp. vi-vii with Perry (1923), p. 230. Attempts to save the plurality of Apuleius’ title, Metamorphoses: Perry (1923), pp. 234–8, Vallette, pp. xxiv-xxv, van Thiel, pp. 26–8; modern scholarship has spiritualized the meaning of ‘metamorphoses’: Tatum, pp. 28–33, Scobie in Aspects (1978), pp. 45 ff.

³¹ Tatum, p. 26: for a possible explanation of these similarities, see p. 435 below.

³² Oudendorp ad loc.; Scobie (1975), p. 77Google Scholar; Smith, W. S. Jr, Lucius of Corinth and Apuleius of Madaura, diss. Yale (1968), p. 106; Smith (1972), pp. 513 ff., esp. 519; Tatum, pp. 24–6.Google Scholar

³³ Rohde, E., RhM 40 (1885), 66–91Google Scholar; Bürger, K., ‘Zu Apuleius’, Hermes 23 (1888), 489–98. In the following I have not discussed Calonghi's strange, diffuse and isolated position that the prologue is a dialogue between Lucius and Apuleius. Other bibliography: van Thiel, p. 43 n.Google Scholar

³⁴ Molt 4, Scobie (1975), pp. 71–2, both inadequate discussions without reference to Leo (v. infra).

³⁵ Leo, F., ‘Coniectanea’, Hermes 40 (1905), 605–6, taken up by Helm (1910), vi, xi-xii.Google Scholar

³⁶ It is helpful to regard all authors inasmuch as they appear in ‘their‘ books as fictional to a degree: see Booth, pp. 71–6, 156 on the ‘implied author’.

³⁷ Friedman, p. 1168: ‘Since the problem of the narrator is adequate transmission of his story to the reader, the questions must be something like the following: (1) Who talks to the reader?…’

³⁸ Riefstahl, p. 22, van Thiel, pp. 44–5, Heine, p. 121 (citing Hammer, S., ‘L’etat actuel’, Eos 39 (1926), 235).Google Scholar

³⁹ Vliet, J. van der, ‘Die Vorrede der Apuleischen Metamorphosen”, Hermes 32 (1897), 79–85, Tatum, p. 88, Junghanns, p. 14.Google Scholar

⁴⁰ Smith (1972), p. 520, (1968), p. 106, works to which my debt will be obvious.

⁴¹ To an extent this is a typical Apuleian false directive to the reader: the reader who understands Cupid and Psyche will already be wondering about argutia Nilotici calami and Taenaros Spartiaca; for false directives, cf. especially tragoediam, non fabulam at 10. 2 (237. 13 Helm) with Smith (1972), pp. 522–3.

⁴⁸ Helm (1910), pp. vi, xii.

⁴³ Riefstahl, p. 20.

⁴⁴ ‘Inepte’ Helm (1910), p. vii, cf. ‘inepte’ on 10.33 at p. xi; ‘jocularly’ A. Scobie, More Essays on the Ancient Romance and its Heritage (1973), p. 44, although he looks too for a deeper purpose – an attempt to preserve the greater realism of the frame story after Cupid and Psyche.

⁴⁵ cf. van der Paardt, p. 80; by ‘insignificant’ I mean such grumblings as ‘sed agilis atque praeclarus ille conatus fortunae meae scaevitatem anteire non potuit’ (4.2). Cf. also Smith (1968), pp. xvii-xx.

⁴⁶ Intrusion and illusion: Booth, pp. 16,205 f.; Riefstahl, pp. 22–3, cf. Smith (1972), pp. 526–7. Some similar implied communication between author and reader occurs in the Satyricon, cf. Beck, pp. 49, 56–7, who, however, is bent on eliminating the author and creating a narrator-Encolpius in his likeness (on which procedure, see Booth, p. 151).

⁴⁷ On the other hand, Scobie (1973), p. 42 could be right that it is simply pleonasm.

⁴⁸ Notably, references to the reader at 1. 1, 10. 2, 11. 23.

⁴⁹ Here Riefstahl, pp. 22–3, must be right that the emphasis on the narrator distances the reader from the events; Helm (1910), p. vii observes the loose connection of the stories late in the novel; Hijmans, p. 121 n. 42, observes that names are rarer late in the novel.

⁵⁰ Vixque enarratis cunctis at 10. 26 serves the same purpose, with the vix a very nice final touch.

⁵¹ Obtrusion (or ‘intrusion’): Allott, M., Novelists on the Novel (1959), pp. 192 f., 269 ff.; Don Quixote, part II, chapter 2 with Scholes and Kellogg, p. 255; Don Quixote, intrusion and display of the author’s genius: Booth, p. 212.Google Scholar

⁵² As Helm rightly observes, cf. van der Paardt, p. 78.

⁵³ Inst. 4. 2. 94.

⁵⁴ A conspectus of the variety of narrators: Hijmans, p. 113.

⁵⁵ This is one way amongst others of giving a narrator special access to information, cf. Friedman, p. 1174.

⁵⁶ Tatum, p. 72 (cf. 74) rightly speaks of ‘rapid and unexpected transformations’.

⁵⁷ Junghanns, p. 146.

⁵⁸ On these scenes, see for instance Riefstahl, pp. 71, 74–5, Junghanns, pp. 136, 179, Smith (1968), p. 50, van der Paardt, p. 6 and especially Tatum, p. 42.

⁵⁹ Even allowing for the Spielraum (Riefstahl 1) which exists within the genre for an individual author.

⁶⁰ In this context I think it is fair, despite the reservations of Hägg, p. 13, to emphasize their similarities: see Heine, p. 6 and Junghanns, p. 7.

⁶¹ Tatum emphasizes well Apuleius the sophist. Reardon, B. P., Courants Litteraires Grecs des II^e et III^e siècles après J. C. (1971), pp. 106–7 shows in how real a sense the sophistic had a ‘qualité théâtrale’.Google Scholar

⁶² It will be seen that here I part somewhat from the emphasis of Hägg, p. 18, who sees rhetoric as something not designed to describe the novel and therefore as a distorting medium; my thesis is that if the author of the novel is conditioned by this ‘distorting’ medium, then the critic who fails to take account of it himself distorts his appreciation of the novel by insisting on a supposed (and non-existent) novelistic purity. But in the end, Hägg is simply attempting a-historical description (cf. p. 419 above).

⁶³ I am not convinced that the rhetoricians before Apuleius had already classified a genre of novel and described it as a species of narratio; cf. E. Rohde, Der griechische Roman und seine Vorläufer² (1900), p. 377 n. 1, Trenkner, pp. 182–5.

⁶⁴ Lysias and the novel: Trenkner, pp. 154 f.; ‘Lucius of Patras’ had a clear and simple style, cf. Phot. Bibl. cod. 129 and Junghanns, p. 2. n. 2.

⁶⁵ Quint. Inst. 4. 2. 53; plausibility in fiction also, e.g. Hermog. Prog. 1. 12.

⁶⁶ Inv. 29.

⁶⁷ Cael. 62.

⁶⁸ 3.26, 4.22, 8. 7, 8. 25, 4.12, 11.20; Apuleius likes inferential particles too, cf. Scobie (1975), pp. 90–1.

⁶⁹ Inst. 11. 3. 162.

⁷⁰ ad loc.: ‘proponit quid sit toto in opere edissertaturus. Docilitas autem captatur et attentio colligitur.’ More broadly, a similar technique is employed when the whole Aristomenes story is presented so as to disarm the reader's disbelief at the outset, ‘in sharp contrast with the almost “take-it-or-leave-it” attitude’ of the Greek original (Scobie (1973), p. 41).

⁷¹ Initium narrationis a persona: Quint. Inst. 4. 2. 129.