¹ Dionysius, writing at Rome in the last decades B.C., discusses historiography both in the prologue to his own history, Roman Antiquities, and in various essays, most notably Thucydides and Letter to Pompeius. All are most accessible in Loeb editions. Lucian wrote How to Write History in A.D. 166; it too is available from Loeb, and is included in Macleod, M. D., Lucian: A Selection (1991)Google Scholar.

² So most recently and most explicitly, Rebenich, S., ‘Historical prose’, in S. E. Porter (ed.), Handbook of Classical Rhetoric in the Hellenistic Period (1997), 265–337Google Scholar, who describes historians as following the trends of the rhetorical schools. Cape, R. W., ‘Persuasive history: Roman rhetoric and historiography’, in W. J. Dominik (ed.), Roman Eloquence (1997), 212–28Google Scholar, discusses relevant issues. Michel, A., Laparole et la beauté (1982)Google Scholar gives a survey of the aesthetic dimensions of rhetoric from antiquity to modernism, but ancient historians do not feature. Stimulating is E. Mattioli, ‘Retorica ed estetica’ in G. Fenocchio (ed.), Le ragioni della retorica (1986), 151–62.

³ It is a commonplace of the critical literature that the interests of the historian and those of the rhetorician are incompatible. See e.g. Woodman, A. J., Rhetoric in Classical Historiography: Four Studies (1988)Google Scholar; Gabba, E., Dionysius and the History of Archaic Rome (1991), 74Google Scholar (who confesses errors in his earlier attitude, but side-steps the problem); Moles, J. L., ‘Truth and untruth in Herodotus and Thucydides’, in C. Gill and T. P. Wiseman (eds), Lies and Fiction in the Ancient World (1993), 88–121CrossRefGoogle Scholar, 116ff.

⁴ In the German the emphasis upon history's visual quality is noticeably stronger: ‘Das wahre Bild der Vergangenheit huscht vorbei. Nur als Bild, das auf Nimmerwiedersehen im Augenblick seiner Erkennbarkeit eben aufblitzt, ist die Vergangenheit festzuhalten’, Schriften I (1955), 496Google Scholar. The emphatic placing of the words at the start of the second sentence, and wiedersehen, aufblitzt, even Augenblick, all reinforce history's visual quality.

⁵ On the role of the sublime in the evolution of the modern discipline of history, see White, H., ‘The politics of historical interpretation: discipline and de-sublimation’, in The Content of the Form (1987), 58–82Google Scholar.

⁶ For a masterful summary of Benjamin's position, see A. Schmidt, ‘Walter Benjamin und die Frankfurter Schule’, in R. Buchholz and Kruse, J. A. (eds), Magnetisches Hingezogensein oder schaudernde Abwehr (1994), 122–34Google Scholar. See too Kittsteiner, H. D., ‘Walter Benjamins Historismus’, in N. Bolz and B. Witte (eds), Passagen: Walter Benjamins Urgeschichte des XIX Jahrhunderts (1984), 163–97Google Scholar; Docherty, T., Alterities (1996), 7–12CrossRefGoogle Scholar.

⁷ Titles are abbreviated as in Liddell and Scott, Greek English Lexicon ⁹. References to Dionysius are to the Loeb editions (1968 etc.); those to the critical essays (trans. Usher) include page references if necessary. Analysis of the relationship of Dionysius' various works of historical theory can be found in Sacks, K. S., ‘Historiography in the rhetorical works of Dionysius of Halicarnassus’, Athenaeum 61 (1983), 41–67,Google Scholar and on his theories generally see Gabba, op. cit. (n. 3), 60–90; Fox, M., Roman Historical Myths (1996), esp. ch. 3Google Scholar.

⁸ Unspecified references to Lucian are to Πῶς δεῖ ‘Ιστορίαν συγγράΦειν,in the OCT, ed. Macleod, M. D., vol. 3 (1980), and include line references to that edition where necessary. The standard accounts of the work remain G. Avenarius, Lukians Schrift zur Geschichtschreibung (1958)Google Scholar and Homeyer, H., Lukian: wie Man Geschichte schreiben soll (1965)Google Scholar. See too Baldwin, B., Studies in Lucian (1973), 75–95Google Scholar.

⁹ On the wider context of ancient historians' self definition, see Marincola, J., Authority and Tradition in Ancient Historiography (1997)CrossRefGoogle Scholar; Strasburger, H., Die Wesensbestimmung der Geschichte durch die antike Geschischtschreibung (1966)Google Scholar.

¹⁰ See Ant.Rom. 1.1; Th. 8; Hist.Conscr. 9, 11. 16–17; 39,11, 14ff.

¹¹ See Th. passim. At 9 Dionysius shifts from a general discussion of Thucydides' qualities to focus upon his organization of his material. Hist.Conscr. 34 and 40–9 is where Lucian is most explicit about his ideal. Lucian is less concerned with style, but see 46–7; 51. One of the central virtues for both authors is ἀκρίβεια which must be understood as the clear expression of correctly selected material; see Hist.Conscr. 51; Ant.Rom 1.5.4.

¹² See Pomp.3, p. 384 where Dionysius refuses to apologize for referring to the ποιήσεις (poetic creations) of Herodotus and Thucydides; Th. 7 and Pomp. 6, pp. 392–4 both express Dionysius' indulgence for historians who include mythical material. Hist.Conscr. 8 explicitly rules out poetic elements, defined as τὸ μῦθον καὶ τὸ ἐγκώμιον; but advocates poetic grandeur for battle narrative, 45, and finally allows both for myth and personal praise or blame within very closely denned limits, 59–60.

¹³ 34.

¹⁴ Th. 1; Pomp. 3, p. 370; 6, p. 398; Hist.Conscr. 6. For the mock-serious aim of Lucian's prologue see Branham, R. B., Unruly Eloquence (1989), 56–7CrossRefGoogle Scholar. Cf. Robinson, C. P., Lucian and his Influence in Europe (1979), 32Google Scholar.

¹⁵ See Baldwin, op. cit. (n. 8), 80–1.

¹⁶ 28.

¹⁷ cf. esp. 12–13; 63.

¹⁸ Lucian's spelling, 42,1. 19; cf. 5,1. 16; 61.

¹⁹ See below, pp. 87–9.

²⁰ On bias, see especially Hist.Conscr. 7; 13; 39–41; 61; 63. Cf. Dionysius, Th. 8.

²¹ 7.

²² 12.

²³ 10.

²⁴ 10.

²⁵ 10, and 53 on prologues: an orator's prologue should have three parts, the historian's two.

²⁶ 11. Likewise Morgan, T., ‘A good man skilled in politics: Quintilian's political theory’, in Y. L. Too and N. Livingstone (eds), Pedagogy and Power (1998), 245–62CrossRefGoogle Scholar, 258–61, points out that Quintilian was unconcerned about the different effects of reading or listening.

²⁷ But generic boundaries, and the corresponding expectations, vary: the veracity of Homer is used as an example for inappropriate use of mythologizing praise: 40.

²⁸ For the question of bias, and the difference between contemporary and ancient history, see Luce, T. J., ‘Ancient views on the cause of bias in historical writing’, Classical Philology 84 (1989), 16–31CrossRefGoogle Scholar.

²⁹ There is a reflection here of Lucian's own ambiguous political position. See Swain, S., Hellenism and Empire (1996), 312–29Google Scholar, and Millar, F. G. B., The Roman Near East 31BC to AD337 (1993), 454–56Google Scholar. A similar context is carefully explored for Dio Chrysostom by T. Whitmarsh, ‘Reading power in Roman Greece’, in Too and Livingstone, op. cit. (n. 26), 192–213.

³⁰ See the preface to Oratt. Vett.; Isoc, passim; Ant. Rom. 1.1–5. See Hidber, T., Das klassizistische Manifest des Dionys von Halicarnass: die praefatio zu “de oratoribus veteribus” (1996)CrossRefGoogle Scholar. On the values of Isocrates himself, see Too, Y. L., The Rhetoric of Identity in Isocrates (1995), esp. 205–21Google Scholar, and on the basic dualistic quality of Isocrates' values, clearly passed on to Dionysius, N. Livingstone, ‘The voice of Isocrates’, in Too and Livingstone, op. cit. (n. 26), 263–81, 272.

³¹ On Dionysius' audience, see Schultze, C., ‘Dionysius of Halicarnassus and his audience’, in I. S. Moxon et al. (eds), Past Perspectives (1986), 121–41Google Scholar, and Gabba, op. cit. (n. 3), 213–16.

³² Pomp 3.372; Th. 41, p. 590.

³³ Th. 8.

³⁴ Ant. Rom. 1.1.1.

³⁵ Ant. Rom 1.1.2–3; Pomp. 3.372.

³⁶ Hist.Conscr. 38, referring int. al. to the Sicilian expedition.

³⁷ Th. 26 = Thucydides 7.69.4–72.1.

³⁸ Note how Usher, by introducing the term rhetoric, also introduces a false distinction which is not present in the Greek. His use of the term aesthetic earlier in the passage is likewise part of a characteristic misrepresentation of Dionysius' position. See below, p. 91.

³⁹ Th. 37–41. For a fuller discussion, see Fox, op. cit. (n. 7), 63–74.

⁴⁰ Th. 11 and especially Pomp. 3.374.

⁴¹ As Dionysius says, πασῶν ὲν λόγοις ἀρετῶν ἡκυριωτάτη τό πρέπον Pomp. 3, which should be read in full awareness of the richness of the word λὸγος. For to prepon, see M. Pohlenz, ‘τὸ πρέπον ein Beitrag zur Geschichte des griechischen Geistes’, Kleine Schriften 1 (1965), 100–39; G. Kennedy, The Art of Persuasion in Greece (1963), 67; 273ff.; M. Perniola, ‘Retorica e decoro’, in G. Fenocchio (ed.), Le ragioni della retorica (1986), 103–12; Mitchell, A. C., ‘The use of πρέπειν and rhetorical propriety in Hebrews 2:10’ Catholic Biblical Quarterly 54 (1992), 681–701Google Scholar.

⁴² On Pre-Platonic ideas of stylistic morality, see K. J. Dover, Aristophanes: Frogs (1991), 14–18, and more generally, G. Nagy, ‘Early Greek views of poets and poetry’, in G. Kennedy (ed.), The Cambridge History of Literary Criticism (1998), 18–50.

⁴³ To employ development as a term would suggest a teleological change, possibly reflecting the changing conditions of the Greek intelligentsia under the Empire; to confirm such a view is beyond the scope of this paper, but it should be noted that Hermogenes, Id. 404–13, treats historical writing in a way which closely recalls Dionysius. Hermogenes even includes the historians in his discussion of the style appropriate to panegyric.

⁴⁴ As the analogy progresses, it becomes clear that Lucian is envisaging a chryselephantine sculpture; perhaps a sculpture of a divinity did not require a model, and as such had no external point of reference, although, given ubiquitous divine anthropomorphism, Lucian's lack of interest in a model is striking.

⁴⁵ So Dionysius, Th. 9. See B. Cardauns, ‘Zum Begriff der “oeconomia” in der lateinischen Rhetorik und Dichtungskritik’, in T. Stemmler (ed.), Ökonomie: sprachliche und literarische Aspekte eines 2000 Jahre alten Begriffs (1985), 10 (cited by W. Wuellner in Porter, op. cit. (n. 2), 51–2).

⁴⁶ See H. White, ‘The value of narrativity in the representation of reality’, in The Content of the Form (1987), 1–25, for an examination of the notion that there is a minimal pre-narrative element in chronicle. For Lucian on chronicle, see 16.

⁴⁷ These techniques are emphatically expressed as additions: ἐπιθεὶς τὴν τάξιν ἐπαγέτω τὸ κάλλος καὶ χρωννύτω καὶ σχηματιζέτω καὶ ῥνθμιζέτω (48) (adding arrangement, let him introduce beauty and add colour and shape it and give it rhythm).

⁴⁸ An interesting point of comparison is Barthes', R. essay from 1967, ‘Le discours de l'histoire’, reprinted in Le Bruissement de la langue: essais critiques IV (1993), 163–77Google Scholar (translated as The Rustle of Language (1986)). His account of how historiography produces a version of the past is remarkably reminiscent of Lucian's. Historical texts work by disavowing their linguistic basis, by claiming to reproduce the past directly, although with certain markers which signify the labour of the historian and his sources, so that ‘l’histoire semble se raconter toute seule' (168). Interestingly, in his related essay ‘L'effet de réel’, ibid., 179–87, he sustains the conventional view that this was a modern rather than an ancient phenomenon, antiquity characterized as generally content with vraisemblance (186). I suggest below that we can take Lucian as a significant exception to this trend, and a pre-echo of the modern way of thinking.

⁴⁹ At 35 Lucian asks Ἀλλὰ πὸυ τὸ τῆς τέχνης κχὶ τὸ τῆς συμβουλῆς χρήσιμον (but what is the role of technique and advice?), suggesting indeed that his version of the ideal historian depends upon personal qualities much more than upon technique.

⁵⁰ Woodman, op. cit. (n. 3), 83-9 (cf. 203-4) examines the res/verba distinction in Cicero and Quintilian, while B. Vickers, In Defence of Rhetoric (1989), 38, points out that the division of Quintilian's work into inventio and elocutio enacts that same distinction. Lucian's view approximates Woodman's claim that inventio is not what we expect from a historian. Clearly for Cicero and Quintilian, as for Dionysius, there was no theoretical contradiction between historical research and inventio.

⁵¹ Plutarch's appeal to Peitho at the opening of malign. Herod. (855A) implies, albeit extremely briefly, a similar relationship between rhetorical interests and personal bias.

⁵² See Vickers, op. cit. (n. 50), 26–7, 62–3; M. Heath, ‘Invention’, in Porter, op. cit. (n. 2), 89–119.

⁵³ See 12; 13; 17; 40.

⁵⁴ 20–3.

⁵⁵ 24 presents the transition from stylistic to factual errors; and in 25 Lucian applies the standard of probability to accounts which he believes are exaggerations and distortions. 27 and 32 are clear statements of criticism for erroneous vision resulting from ignorance and lack of reading. The ideal historian is experienced: 37.

⁵⁶ 8.

⁵⁷ See F. Brommer, Herakles II: Die unkanonischen Taten des Helden (1984) and Lexicon Iconographicum Mythologiae Classicae, s.v. ‘Omphale’.

⁵⁸ We must distinguish here between the lion-skin and other clothes. See Fox, M., ‘Transvestite Hercules at Rome’, in Cleminson, R. and Allison, M. (eds), In/visibility: Gender and Representation in a European Context. Interface 3 (1998), 1–22,Google Scholar esp. 10–11.

⁵⁹ The same polarity between physical essence and external attribute can be found in earlier literary representations of Heracles. See N. Loraux, ‘Herakles: the super-male and the feminine’, in D. M. Halperin et al. (eds), Before Sexuality (1990), 21–52.

⁶⁰ 35.

⁶¹ 23. Such images of incongruity recall the opening of Horace, Ars Poetica. C. O. Brink, Horace on Poetry: The Ars Poetica (1971), 246, points out the resemblance between Horace's grandiose prologue producing a mouse, and the hyperbolic prologue of Hist. Conscr. 23. Perhaps Lucian knew his Horace, or perhaps the resemblance is no more than a manifestation of an established comic trope. On Lucian's Latin, see Swain, op. cit. (n. 29), 319 n. 75.

⁶² 27.

⁶³ 13.

⁶⁴ 7,11. 5–7.

⁶⁵ On enargeia see Brink, op. cit. (n. 61), 246; Fowler, D. P., ‘Narrate and describe: the problem of ekphrasis’, JRS 81 (1991), 25–35,Google Scholar esp. 26–7; J. Eisner, Art and the Roman Viewer (1995), 25–8; R. Webb, ‘Poetry and rhetoric’, in Porter, op. cit. (n. 2), 339–69, 344–5. This material is rhetorical and poetic. Compare the absence of explicit discussion of enargeia where one might expect it (re the Polybian tradition of autopsy) in Marincola, op. cit. (n. 9), 79–81.

⁶⁶ 47 cf. § 6.

⁶⁷ 16. Compare Brutus' praise of Caesar's naked style in Cicero, Brutus 75.262. It is noteworthy that Brutus is the speaker here, but the notion of the unadorned style as a hallmark of the good historian was familiar. See references collected by Marincola, op. cit. (n. 9), 10, n. 42.

⁶⁸ Gadamer, H.-G., Wahrheit und Methode ⁶ (1990), 47–87Google Scholar. See also Eagleton, T., The Ideology of the Aesthetic (1990), 70–101Google Scholar.

⁶⁹ That defence is conventionally seen as beginning with White, H., Metahistory (1973)Google Scholar, a work whose reception and influence have been very varied.

⁷⁰ As well as foreshadowing, Lucian also draws out a thread in both Thucydides and Polybius. Davidson, J., ‘The gaze in Polybius' Histories’, JRS 81 (1991), 10–24,Google Scholar characterizes Polybius as a historian who exploits the idea of immediate vision to organize his own disappearance. There is a close resemblance here to Lucian's ideal, particularly if Davidson is right about Polybius' visualizing technique.

⁷¹ So Davidson, op. cit. (n. 70), 24: ‘Polybius is invisible; he had long ago arranged for himself to disappear.’ Jameson names this minimized historian the vanishing mediator in his account of Weber's ambitions for sociology: Jameson, F., ‘The vanishing mediator’, The Ideologies of Theory. Vol 2, Essays 1971–1986 (1988), 3–34Google Scholar.

⁷² See D. A. Russell, ‘Longinus’ On the Sublime (1964), xlii–xlviii; Martano, G., ‘II saggio dul sublime’, ANRW 32.1 (1984), 364–403Google Scholar. Others bring the legacy up to date: G. Lombardo, Hypsegoria: studi sutta retorica del sublime (1988), 13–34, describes [Longinus] as ‘tardo moderno’, while N. Hertz, in similar vein, sees Walter Benjamin as a modern [Longinus], ‘Lecture de Longin’, Poétique 15 (1973), 292–306,Google Scholar at 301–2, his Aura recalling hypsos.

⁷³ Lombardo suggests that for [Longinus] the sublime is a modus vivendi rather than a modus scribendi (op. cit. (n. 72), 18–19). Lucian's comments on the need for historians to display the right attitude suggest he may be thinking on similar lines.

⁷⁴ Certainly both Lucian and [Longinus] reinterpret mimesis; see above, pp. 86–7 and cf. Too, op. cit. (n. 42), 210. A. Battistini suggests that Tacitus Dialogus is pointing in the direction of a shift in the meaning of inventio, parallel to a move towards greater literality. He depicts inventio as ‘non più investita di finalità argomentative e persuasive ma ristretta a nuclei tematici riproducibili da testo a testo’, ‘Ornamenta e scrittura’, in G. Fenocchio (ed.), Le ragioni della retorica (1986), 71–90, at 72, cf. above, p. 86. See too W. J. Dominik, ‘The style is the man’, in Roman Eloquence (1997), 50–68, at 62–6. Possibly relevant is Webb's observation that [Longinus] is the only ancient theorist to distinguish between vividness in poetry and rhetoric; in the former it produces ekplexis, in the latter, persuasion, art. cit. (n. 65), 345.

⁷⁵ cf. Too's reinstatement of the conventional rhetorical element in [Longinus'] thought; op. cit. (n. 42), 188–94. cf. Michel, op. cit. (n. 2), 112.

⁷⁶ Too, op. cit. (n. 42), 215–17. One could read these signs of greater individualism as part of the much larger phenomenon described by Veyne, Foucault, et al. That would over-reach the scope of this paper, and is in itself problematic; see the detailed discussion of Swain, S., ‘Biography and biographic’, in M. J. Edwards and S. Swain (eds), Portraits (1997), 1–37Google Scholar, esp. 5–22.