1. The term ‘intertextuality’ was coined by Kristeva, J., ‘Bakhtine, le mot, le dialogue et le roman’, Critique 33 (1967), 438–65Google Scholar translated by Jardine, A., Gora, T., and Roudiez, L. S. as ‘Word, Dialogue and Novel’ in Desire in Language: a Semiotic Approach to Literature and Art (Oxford, 1980)Google Scholar, edited by Roudiez, L. S.. Bruce, D., ‘Bibliographic annotée écrits sur l'intertextualité', Texte: Revue de Critique et de The'orie Litteraire 2 (1983), 217–58Google Scholar contains a useful bibliography. Riffaterre, M., ‘Syllepsis’, Critical Enquiry 6 (1980), 625–38CrossRefGoogle Scholar has made an effort to distinguish between intertextuality and an intertext since an exclusive focus on the latter would really be source criticism under a different name.

2. Fragment 32 in Strzelecki, L., Cn. Naevii Belli Punici Carmen (Leipzig, 1964)Google Scholar. In some manuscripts the reading is ‘urit vastat populatur’, ‘burned, devastated and ravaged’, but for a full analysis see Barchiesi, M., Nevio Epico: Storia Interpretazione Edizione Critica dei Frammenti del Primo Epos Latin (Padua, 1962), 337–48Google Scholar. On this particular fragment see Mariotti, S., ‘Concinnat in Naevio Bellum Punicum Ft. 39.3 M’, 1–5 in Kontinuität und Wandel: Lateinische Poesie von Naevius bis Baudelaire (Hildesheim, 1986)Google Scholar, edited by F. Wagner, W. Maaz, and U. von Stache. For a general discussion of Naevius see Goldberg, S. M., ‘Saturnian Epic: Livius and Naevius’, 19–36 in Roman Epic (London and New York, 1993)Google Scholar, edited by Boyle, A. J.. Heubner, H., Kommentar: P. Cornelius Tacitus, Die Historien Volume 2 (Heidelberg, 1968)Google Scholar, lists as a parallel for Tacitus’ phrasing the Seneca, Elder, Controversiae 1.6.12Google Scholar: ‘vastari omnia ac rapi, conburi incendiis villas’, ‘universal devastation and plunder, villas consumed by fires.’

3. One further example is Tacitus, , Histories 1Google Scholar. 40.1, ‘non tumultus, non quies, quale magni metus et magnae irae silentium est’, ‘there was neither uproar, nor silence, but the sort of hush which is typical of great fear or great anger’, and Naevius, Bellum Punicum fragment 57 in Strzelecki, op. cit: ‘magnae metus tumultus pectora possidit’, ‘an agitation of great fear gripped their hearts.’

4. See Baxter, R. T. S., ‘Virgil's Influence on Tacitus Histories 3’, CP 66 (1971), 93–107Google Scholar for details. Kraus, C. S., ‘“No Second Troy”. Topoi and Refoundation in Livy 5’, TAPA 124 (1994), 267–89 notes that the fall of Troy is the literary archetype for all captured citiesGoogle Scholar. Josephus, Even at BJ 3. 319, 325, and 327Google Scholar in his description of the fall of Jotapata seems to echo phrases from Virgil's account of the fall of Troy, at Aeneid 2. 265 ffGoogle Scholar.

5. See Kenney, E. J., ‘Iudicium Transferendi: Virgil Aeneid 2.469–505 and its Antecedents’, 103– 20 in Creative Imitation and Latin Literature (Cambridge, 1979)Google Scholar, edited by A. J. Woodman and D. West. Note esp. 112–13 for the resonances between Virgil, Ennius, and Livy. Ogilvie, R. M., Commentary on Livy Books 1–5 (Oxford, 1965), 120Google Scholar, observes that Virgil owed much of his second book to Ennius' Annales and quotes Servius, Ad Aen. 2.486: ‘de Albano excidio translates est locus’, ‘the topic has been transferred from the destruction of Alba Longa.’

6. Conte, G. B., The Rhetoric of Imitation (Ithaca and London, 1986), 23 and 62Google Scholar. Worton, M. and Still, J., Intertextuality: Theories and Practices (Manchester, 1990), 24Google Scholar, use an identical image when they refer to ‘textual scars’. Riffaterre, M., ‘Compulsory Reader Response: the Intertextual Drive’, 56–78Google Scholar in the same volume, refers vividly to a ‘lexical Janus, one word with two faces’ (72).

7. See, for example, Woodman, A. J., ‘Self-imitation and the Substance of History: Tacitus Annals 1.61–65 and Histories 2.70, 5.14–15’, 143–55 in Creative Imitation and Latin Literature (Cambridge, 1979)Google Scholar, edited by Woodman, and West, , and Ginsburg, J., ‘In maiores certamina: Past and Present in the Annals’, 86–103 in Tacitus and the Tacitean Tradition (Princeton, 1993)Google Scholar edited by T. J. Luce and Woodman. Gibbon drew attention to the potential for history to become lessinteresting to readers as the subject matter became more remote: historical writings could thus come to lose ‘presque tout leur mérite, excepté celui que leur auteur a su leur donner par la manière dont il a traité son sujet’, ‘almost all their merit, except that which their author could give them through the manner in which he treated the subject’, ‘Remarques sur les ouvrages et sur le caractère de Salluste, Jules César, Cornelius Nepos, et Tite-Live’, 430–1 in Miscellaneous Works of Edward Gibbon Esq. IV (London, 1814)Google Scholar.

8. This is to assume here a model of historical writing where historians aim to scrutinize the facts and to reconstruct an event as it really happened. Martindale, C, Redeeming the Text: Latin Poetry and the Hermeneutics of Reception (Cambridge, 1993), 19–20Google Scholar offers a number of objections to this model. Cf. Momigliano, A., ‘The Rhetoric of History and the History of Rhetoric: On Hayden White's Tropes’, 49–59 in Settimo Contributo alla Storia degli Studi Classici e del Mondo Antico (Rome, 1984)Google Scholar. One key factor to consider is the tendency of ancient writers, including historians, to resort to rhetorical topoi. Kraus, C. S., Livy, Ab Urbe Condita Book VI (Cambridge, 1994)Google Scholar puts it in a nutshell on 16: ‘Linguistically, as well as ethically, history thus becomes a “knot of different times”, in which theoretically unique, non-recurrent events are built from other episodes, other stories.’

9. For figurae in the sense of ‘allusions’, see Suetonius, Vespasian 13 and Domitian 10, Macrobius, , Saturnalia 7.3Google Scholar, and Quintilian 9.2.65.

10. Cf. the arguments about where Lucan's Pharsalia fitted on the sliding scale between history and poetry. Servius, , Ad Aen. 1.382Google Scholar: ‘Lucanus namque ideo in numero poetarum non meruit, quia videtur historiam composuisse non poema’, ‘For Lucan does not deserve to be numbered amongst the poets, because he seems to have written a history and not a poem.’ Cf. Martial 14.194, Petronius, , Satyricon 118–24Google Scholar, and Quintilian 10.1.90. The debate extended beyond antiquity. In 1728 when Voltaire wrote his epic poem, La Henriade, he observed: ‘This poem is based on a known history, the truth of which has been respected in the principal events. The others, being less reliable, have been either omitted or rearranged according to the requirements of verisimilitude in a poetic composition. In this way, every effort has been made to avoid the weakness of Lucan, whose poem is nothing more than an overblown chronicle’, 39 in Oeuvres Complètes de Voltaire VIII (Paris, 1877–1885)Google Scholar.

11. On the Caudine Forks see Salmon, E. T., Samnium and the Samnites (Cambridge, 1967)Google Scholar, reviewed by Frederiksen, M., JRS 58 (1968), 224–9Google Scholar, and Lipovsky, J. P., A Historiographical Study of Livy Books 6–10 (Salem, New Hampshire, 1981), 140–51Google Scholar. There are references in Livy 9.1ff., Dionysius of Halicarnassus, , Roman Antiquities 16.1—2Google Scholar, Cicero, , De Officiis 2.75 and 3.109Google Scholar, Cicero, , De Senectute 12.41Google Scholar, Valerius Maximus 5.1.5 ext. and 7.2.17 ext., Appian, , Roman History 3.4.2–7Google Scholar, Quintilian 3.8.3, Columella 10.132, Lucan, , Pharsalia 2.137–8Google Scholar, Italicus, Silius, Punica 8Google Scholar. 567, and Florus, , Epitoma 1.16Google Scholar.

12. On whether these peace terms were in fact a feodus, ‘treaty’, or a sponsio, ‘promise’, see Crawford, M., ‘Foedus and sponsio’, PBSR 41 (1973), 1–7Google Scholar.

13. The italics correspond to echoes in two Tacitean passages which will be discussed below.

14. For focalization see Genette, G., Narrative Discourse, translated by Lewin, J. E. (Oxford, 1980)Google Scholar.

15. ‘Caecus’, ‘blind’, was frequently used to mean mentally or morally blind, as at Cicero, , De Amicitia 15.54Google Scholar, Sallust, , Jugurtha 25.7Google Scholar, Horace, , Satires 2.3.44Google Scholar, and Tacitus, , Agricola 43Google Scholar. The fact that the soldiers themselves (rather than Livy the narrator) make this observation vividly enables a reader to imagine that (s)he is seeing the shifting situation through the eyes of the participants.

16. Wild beast imagery is also used by Dionysius, of Halicarnassus, at Roman Antiquities 16.2Google Scholar but in a different context. The speaker (presumably Gavius Pontius) says that the Samnites ὥσπερ θηρία', ‘like wild beasts’, will probably tear him apart if he agrees to let the Romans go unscathed. Livy applies the animal imagery to the Romans instead of the Samnites, and uses it to convey not ferocity but vulnerability. Cf. the ‘slaughtering like animals’ topos highlighted as especially typical of historical writing by Woodman, A. J., Velleius Paterculus: the Tiberian Narrative (2.94–131) (Cambridge, 1977), 202Google Scholar.

17. Cf. Homer, , Odyssey 15.400Google Scholar, Euripides Fragment 133 in Nauck, A., Tragicorum Graecorum Fragmenta (Hildesheim, 1964)Google Scholar, Cicero, , De Finibus 2.105Google Scholar and Ad Att. 5.12.4, and Virgil, , Aeneid 1.203Google Scholar. Here Aeneas, despite personal misgivings, optimistically points to a moment in the future when he and his men will perhaps even be able to recall their current trials with some pleasure. The Caudine Forks soldiers pessimistically picture the impending humiliation, but it will certainly give them no comfort after the event to have survived.

18. Cf. Livy 1.28.11 where the Alban soldiers avert their eyes from the cruel punishment of their commander Mettius, who is symbolically ripped in two by horses after wavering in his loyalty between Fidenae and Rome.

19. Cf. Appian, , Roman History 3. 4.6Google Scholar where the soldiers already know that disgrace is worse than death beforethe surrender. Thus there is no process of humiliating discovery, as in Livy's account.

20. Perhaps the most memorable occurrence of eludo is at the opening of Cicero, , In Catilinam 1.1.1Google Scholar: ‘quam diu etiam furor iste tuus nos eludet?’, ‘how long will that frenzy of yours still continue to mock us?’

21. Cf. Davidson, J., ‘The Gaze in Polybius' Histories’, JRS 81 (1991), 10–24Google Scholar who observed that in Polybius' narrative ‘… events are always mediated through the gaze of the inhabitants of his history and that of his supposed readers… Instead of a single reality, we are given several…’ (10).

22. Cf. Walker, A. D., ‘Enargeia and the Spectator in Greek Historiography’, TAPA 123 (1993), 353–77Google Scholar who discusses the Thucydides passage in detail. Livy offers his own version at 1.25.2 where he registers the feelings of the armies on both sides as they watch the battle between the Horatii and the Curiatii. On enargeiasee Quintilian 6.2.31–2.

23. Cf. Duris, of Samos, , FGrHist 76Google Scholar F 1 on mimesis and its hedonistic function in a historical work of arousing emotion in a reader. This notion of mimetically orientated historiography is discussed by Gentili, B. and Cerri, G., History and Biography in Ancient Thought (Amsterdam, 1988), 7–33Google Scholar. On the other side of this coin is Polybius 2.56.10–12, the famous attack on Phylarchus' account of the fall of Mantinea. According to Polybius Phylarchus had tried to recreate the horrors before the audience's eyes and had thus written something closer to tragedy than to history.

24. See Andresen, G., ‘Tacitus und Livius’, Wochenschrift für Klassische Philologie (1916), 402Google Scholar and Syme, R., Tacitus (Oxford, 1958)Google Scholar, Appendix 34.

25. Keitel, E., ‘The Function of the Livian Reminiscences at Tacitus Histories 4.58.6 and 62’, CJ 87 (1992), 327–37Google Scholar, makes this comment on 327.

26. The description of the column as ‘silens’, ‘silent’, may be particularly apt given that ‘silentes’ was another way of referring to the dead, as at Virgil, , Aeneid 6.432Google Scholar, Ovid, , Metamorphoses 5.356, 15.772, 15.797Google Scholar, Seneca, , Medea 740Google Scholar, and Flaccus, Valerius, Argonautica 1.750Google Scholar. Tacitus and Livy were both sensitive to silentium and its nuances. On Livy's arresting use of silence see Ogilvie, R. M., op. cit. (n. 5), 486Google Scholar and Walsh, P. G., ‘The Literary Techniques of Livy’, Rh. Mus. 97 (1954), 97–114Google Scholar, who refers on 103 to Livy's ‘silentium’ device. For some eloquent silences in Tacitus see Histories 1.40.1 (with Husband, R. H., ‘Galba's Assassination and the Indifferent Citizen’, CP 10 [1915], 321–5)Google Scholar, Histories 1.55.1, and Histories 3.67.2.

27. Tacitus' phrasing, ‘procurrentes ex agris tectisque et undique effusi’, ‘running from the fields and houses and streaming from all sides’, recalls Virgil's description of the Italian youth admiring Camilla in the catalogue of heroes at Aeneid 7.812: ‘omnis tectis agrisque effusa iuventus’, ‘all the young people streaming from the houses and fields.’

28. A sense of pride (and hence the ability to feel shame) was an intrinsic component of being a good soldier. MacMullen, R., ‘The Legion as a Society’, Historia 33 (1984), 440–56Google Scholar talks about the tight-knit relationships which could develop between soldiers in one particular unit and describes how the focus of loyalty on one's own little society of fellow soldiers could become so intense that the whole purpose of a particular war might easily be forgotten. The proud ala Picentina perhaps fits this category. Birley, E., ‘Alae named after their Commanders’, Ancient Society (1978), 257–73Google Scholar suggests that the prefect of the cavalry L. Rustius Picens (CIL 10094 = ILS 2679) was the officer responsible for the name of the unit, which clearly had a strong sense of its own identity.

29. See Bruckmann, H., Die römischen Niederlagen im Geschichtswerk des T. Livius (Bochum-Langendreer, 1936), 1–31Google Scholar.

30. See C. Sallusti Crispi: Catilina, Iugurtha, Historiarum Fragmenta Selecta, Appendix Sallustiana (Oxford, 1991)Google Scholar, edited by L. D. Reynolds. For Sertorius’ injury, see too Plutarch, , Sertorius 1. 4 and 4.3Google Scholar. The biographer makes the observation that the most tricky and crafty generals, Philip, Antigonus, Hannibal, and Sertorius, were all one-eyed. As Aelian notes at VH 12.43 Antigonus' descriptive nickname, Monophthalmus, evidently took a turn for the worse when people began to call him ‘Cyclops’. Tacitus may have meant his readers to see Sanctus’ injury as being a marker of foreignness. All the most notorious one-eyed generals had been barbarians, apart from Sertorius, but even he could be considered an honorary Lusitanian. At Histories 4.13.2 we learn that the oneeyed Julius Civilis consciously modelled himself on Sertorius and Hannibal because of their common affliction. There is an interesting reversal of this pattern at Annals 2.9 where the barbarian Arminius confronts his Romanized brother Flavus across a river: Flavus has the eye missing, not Arminius.

31. Another commander who diverges from the ideal is Antonius Primus at Histories 3.24.1, who harangues his men during the night battle at Cremona, ‘alios pudore et probris, multos laude et hortatu, omnes spe promissisque accendens’, ‘stirring some through their sense of shame and through threats, many by praise and encouragement, and all by hope and promises’. For the more standard version of the commander's harangue of his men, seePlutarch, , Antony 44.3Google Scholar, Onasander, , Strategikos 33.6Google Scholar, and Dio 69.9.3 where promises do not feature: instead the commander simply praises the brave and reproaches the cowards.

32. Tacitus’ sources for the second campaign of Cremona certainly included memoirs written by the Flavian general Vipstanus Messalla, one of the interlocutors of the Dialogus and the temporary commander of the Legio VII Claudia in A.D. 69. See Histories3.25 and 3.28. At Dialogus30.1 Messalla partly blames the recent decline in oratory on the substandard education being offered to young men: ‘nee in auctoribus cognoscendis nee in evolvenda antiquitate nee in notitiam vel rerum vel hominum vel temporum satis operae insumitur’, ‘there is not enough care devoted to becoming thoroughly acquainted with authors, to unfolding the past, and to knowledge of deeds, people, and occasions.’ This sounds like the kind of man who would have known his Livy: perhaps the Caudine Forks echoes in the Vitellian surrender could have originated in Messalla's memoirs even though he was a Flavian general.

33. At Livy 6.3.3 the people of Sutrium are ‘inermis’, ‘unarmed’, as they leave their town ‘miserabili agmine’, ‘in a wretched column’, and at Caesar, , BC 2Google Scholar. 11.4 the people of Massilia are ‘inermes cum infulis’, ‘unarmed and wearing fillets’. Such examples show that these are common elements in most surrender scenes.

34. Note Livy 9.6.2: ‘Circumstabantarmati hostes, exprobranteseludentesque; gladii etiam plerisque intentati…’, ‘The armed enemy stood around, taunting and mocking them; swords were even brandished at many legionaries…’, and Tacitus, , Histories 3.31.3Google Scholar: ‘circumstiterant victores et primo ingerebant probra, intentabant ictus…’, ‘The victors stood around them and at first jeered and aimed blows at them.’

35. On topoi see Cairns, F., Generic Composition in Greek and Latin Poetry (Edinburgh, 1972), especially 98–124Google Scholar, ‘Originality in the Use of Topoi’. On the role of the intertext note Frow, J., ‘Intertextuality and Ontology’, 45–55 in Intertextuality: Theories and Practices (Manchester, 1990)Google Scholar, edited by M. Warton and J. Still, who observes: ‘The identification of an intertext is an act of interpretation. The intertext is not a real and causative source, but a theoretical construct formed by and serving the purposes of a reading.’ Authorial intention thus takes a back seat to effect of the text on the reader. For the difference between ‘influence’ and ‘intertextuality’ see Elam, H. R. s.v. ‘Intertextuality’ in The New Princeton Encyclopedia of Poetry and Poetics (Princeton, 1993)Google Scholar.

36. See Quintilian 10.1.19 for the liquefaction of food as a vivid metaphor for the process of careful reading: ‘Repetamus autem et retractemus, et ut cibos mansos ac prope liquefactos demittimus, quo facilius digerantur, ita lectio non cruda, sed multa iteratione mollita et velut confecta, memoriae imitationique tradatur', ‘We must return to what we have read and reconsider it and, just as we do not swallow our food until we have chewed it and reduced it almost to a state of liquefaction, so that it can be digested more easily, likewise our reading should not be committed to memory for imitation while it is in a crude state, but it should be softened and reduced to a pulp by frequent re-perusal.’ Tacitus had ‘digested’ the Livy passage carefully, enabling him to use it on more than one occasion.

37. On Flavian propaganda in general see Briessman, A., Tacitus und die Flavische Geschichtsbild, Hermes Einzelschriften 10 (Wiesbaden, 1955)Google Scholar, Ferrill, A., ‘Otho, Vitellius and the Propaganda of Vespasian’, CJ 60 (1964–1965), 267–9Google Scholar, Nicols, J., Vespasian and the Panes Flavianae, Historia Einzelschriften 28 (Stuttgart, 1978)Google Scholar, andRamage, E. S., ‘Denigration of Predecessor under Claudius, Galba and Vespasian’, Historia 32 (1983), 201–14Google Scholar.

38. On Tacitus' investment of the Vitellian soldiers with a German identity see Ash, R. E., ‘Individual and Collective Identities in Tacitus’ Histories', D. Phil, thesis (Oxford, 1996), 59–97Google Scholar.

39. The lictores were official attendants for consuls, magistrates, and priests whose duties included carrying the fasces. Originally they helped carry out punishments and thus embodied power and authority in a very literal way: see Livy 1.8 for Romulus’ twelve lictores. Caecina's consulship was rather short-lived: see Townend, G. B., ‘The Consuls of A.D. 69–70’, AJP 83 (1962), 113–29Google Scholar.

40. Conte, G. B., op. cit. (n. 6), 23Google Scholar. Certainly, Oakley, S. P., A Commentary on Livy Books VI-X (Oxford, 1997)Google Scholar, cites, Histories 4.62Google Scholar rather than Histories3.31 for echoes of Livy's Caudine Forks narrative. This should serve as a salutary reminder that some levels of borrowing are clearly more significant than others.

41. ‘Deformitas’, ‘wretchedness’, can mean either physical hideousness (as at Cicero, , De Officiis 3.29.105)Google Scholar or moral deformity of character (as at Quintilian 6.1.12). Frequently physical ‘deformitas’ could provoke powerful feelings of compassion in onlookers. For example at Livy 2.23.3—7 the squalid state of an old man who had been tortured by his creditors causes the plebs to feel such pity that a riot is triggered. Orators tried to recreate such feelings in the courtroom, as Quintilian notes at 6.1.30–3. Yet the only onlookers present at Histories 4.62 are gleeful Gauls: it is left to Tacitus' readers, the audience external to the text, to feel the pity.

42. It is also significant that the Legio XVI Gallica was present at both surrenders and thus forms a common physical link between the two capitulation scenes at 3.31 and 4.62.

43. Woodman, A. J., Rhetoric in Classical Historiography(London and Sydney, 1988)Google Scholar is crucial on this whole question. See his index under inventio and note his fundamental distinction on 90: ‘There is thus a distinction between a “singular factual statement about the past”, which will normally be one of the “public facts” of history… and will constitute the hard core, and the exaedificatio or superstructure which required to be built up around it.’ Cf. Cornell, T. J., ‘The Formation of the Historical Tradition of Early Rome’, 67–86Google Scholar in Past Perspectives: Studies in Greek and Roman Historical Writing (Cambridge, 1986)Google Scholar, edited by Moxon, I., Smart, J. D., and Woodman, A. J., and Wiseman, T. P., ‘Lying Historians: Seven Types of Mendacity’, 122–46Google Scholar in Lies and Fiction in the Ancient World (Exeter, 1993)Google Scholar, edited by Wiseman and C. Gill. We should perhaps still note the oddity of Tacitus describing surrender scenes for which (presumably) eye-witnesses existed through the filter of a poorly documented incident which took place almost four hundred years earlier.

44. Nonetheless there are still unresolved questions about the ‘truth value’ of Livy's Caudine Forks description. For example, the detail of Livy's narrative may itself have been influenced by Mancinus’ defeat in Spain in 137 Cicero, B.C., De Officiis 3.29.109Google Scholar, Tacitus, , Annals 15.13.2Google Scholar, and Fionas, , Epitoma 1Google Scholar. 34.7 explicitly set the two incidents side by side. The topographical value of Livy's description has also been challenged: see Horsfall, N., ‘The Caudine Forks: Topography and Illusion’, PBSR 50 (1982), 45–52Google Scholar. Livy's Caudine Forks passage certainly appealed to poets, as at Virgil, , Aeneid 11.522–9Google Scholar, the site of Turnus’ proposed ambush. Horsfall, , ‘Illusion and Reality in Latin Topographical Writing’, G&R 32 (1985), 197–208Google Scholar refers to ‘striking analogies of apparent detail’ between Livy's and Virgil's descriptions. Rehm, B., Das Geographische Bild des Alten Italien in Vergils Aeneis, Philologus Suppl. 24.2 (Leipzig, 1932), 79–81Google Scholar sees Virgil's passage as an example of a set piece, the ‘convallis’, ‘enclosed valley’, which was not necessarily derived from Livy.

45. See CassiusDio 62.17 and Tacitus, , Annals 15.41Google Scholar. Kraus, C. S., op. cit. (n. 8), 93–4Google Scholar discusses synchronism of notable events and cites Grafton, A. T. and Swerdlow, N. M., ‘Calendar Dates and Ominous Days in Ancient Historiography’, Journal of the Warburg Institute 51 (1988), 14–42CrossRefGoogle Scholar.

46. See Chaplin, J. D., ‘Livy's Use of Exempla and the Lessons of the Past’, unpublished dissertation (Princeton, 1993)Google Scholar and Bloomer, W. Martin, Valerius Maximus and the Rhetoric of the New Nobility (London, 1992)Google Scholar.

47. Fantham, E., Lucan De Bello CiviliBook 2 (Cambridge, 1992), 91Google Scholar notes that both Appian, , BC 2Google Scholar. 36 and Dio 41.5, 41.8, and 41.16 describe civilians at Rome recalling the wars of Sulla and Marius, firstly when Caesar crossed the Rubicon and again when he approached the undefended city. If this is rhetorical topos, it is surely one which plausibly reflects reality.

48. See Bartsch, S., Actors in the Audience: Theatricality and Doublespeak from Nero to Hadrian (Cambridge, Mass, and London, 1994), 88CrossRefGoogle Scholar, and in general Chapter 3 passim, ‘Oppositional Innuendo: Performance, Allusion, and the Audience’, 63–97. Cf. Iser, W., ‘Interaction Between Text and Reader’, 106–19 in The Reader in the Text (Princeton, 1980)Google Scholar, edited by S. R. Suleiman and I. Crosman. Iser observes on 106: ‘… the literary work has two poles, which we might call the artistic and the aesthetic: the artistic pole is the author's text, and the aesthetic is the realization accomplished by the reader.’