Having considered in general terms the constraints on narrative inventio, I turn to examples of inventio in action to probe the question of how far from the truth truthlike takes us. I begin with a crime foisted on Lucius Quinctius Flamininus (consul 192 bce) by Valerius Antias, for which Livy claims to offer a more reliable report in the context of the census of 184 (39.42.7-13, plus 39.43.1-3). Livy opens with a summary of Cato’s speech against Flamininus, with particular attention to the charge of having executed a Gallic noble to compensate his scortum (‘‘catamite’’), one Philippus Poenus, for depriving him of Rome’s gladiatorial shows by taking him to Gaul. Livy then summarizes Antias’ version of the story, with its different details about Flamininus’ offense. To Livy it looks as if Antias wrote without consulting Cato’s speech and put his trust in a fabula with no auctor.
Antias’ story is the only example in Wiseman (1994) of the invention of an event in Rome’s history by an author of historical narratives; the other historical ficta discussed there have their origins in speeches or oratory. It is therefore important to note that in Livy’s version at least this story appears in a speech, not in the narrative of Flamininus’ command in Gaul in 192, which Livy had narrated in book 35 (20.1-7, 22.2-4, and esp. 40.3 for the capitulation of the Boii, which is the context for the alleged killing), while Livy’s description of Antias’ version (39.43.1-3) suggests that in Antias too the context is the censorial accusation, not the underlying events. That is, the existence of two incompatible versions of Flamininus’ offense in the historical record is not particularly good evidence for inventio in narrative, since in the surviving history the story does not impinge on the narrative of events. Wiseman shows that, despite Livy’s critique, Antias’ version appears as a historical factoid in later writers (Cic. Sen. 42; Val. Max. 2.9.3), but these are not historians. Compare Luce (1989b: 177): ‘‘The one extended case where we can see an ancient historian using his written source (Livy and Polybius) indeed shows inventio on Livy’s part, but the added touches are not extensive and do not impair the essential integrity of the Polybian narrative.’’
For an example of ficta in narrative I turn to Woodman’s best example of Tacitean inventio. He argues that the account of Germanicus’ visit to the German battlefield on which Arminius defeated Varus’ army (Ann. 1.61.2) shows a (very small) ‘‘hard core of information’’ developed into a substantial scene by the reuse of, among other things, material from his earlier account of Vitellius’ visit to the civil war battlefield on which his legates defeated Otho’s army (Hist. 2.70; Woodman 1979: 147-9; 1988: 176-8). Similarities of wording and a parallel deployment of eyewitness testimony link the passages. This analysis and the larger argument that Woodman builds on it have sometimes been taken (as Woodman intends!) to impugn Tacitus’ reliability as a historian (after one has given the ancient historical texts the literary analysis they deserve, he says, ‘‘there is precious little historical evidence left,’’ Woodman 1983: 120). But does it?
It is reasonable to suppose (and Woodman does suppose) that Tacitus had what he considered reliable evidence that Germanicus visited the battlefield. The senate voted triumphal insignia to Germanicus’ legates in 15 ce ob res cum Germanicogestas (‘‘on account of their deeds with Germanicus,’’ Ann. 1.72). One of the recipients, Aulus Caecina, played a large role in the visit to Teutoberg, including clearing the route for Germanicus’ army (1.61; cf. also 1.63). More discussions may have ensued in connection with the senate’s dedication of an arch in the Forum ob recepta signa cum Varo amissa ductu Germanici (‘‘on account of the recovery under the generalship of Germanicus of the standards lost with Varus,’’ 2.41). What Tacitus may not have had is eyewitness testimony about the details of the visit, who said what when, what the place looked like, and so on. But all he needed to write the passage in the Annals is the fact of the visit and the likelihood of the presence of survivors of the battle in Germanicus’ force (see M. Gwyn Morgan 1992: 24), along with a complicated mixture of previous descriptions of battlefields, Vergilian language, life experience, and so on, as Woodman shows. But even if, as I believe, these were Tacitus’ raw materials for his narrative of Germanicus’ battlefield visit, so that his account has to be considered truth-like rather than true, the extent of invention ficta is here again quite limited.
In assessing an ancient historian's reliability one of the important steps is to define the boundaries between event (for which a responsible historian, or at least one that a modern historian will be comfortable using as an authority, will have evidence she or he considers reliable) and description (for which there may well be no direct source, but which a narrative historian can, indeed must, provide out of ‘‘background material’’). In the passage just discussed, for example, inventio provides a description. Similarly, in the introduction to his recent translation of the Annals, under the heading of the historian’s ‘‘treatment of events,’’ Woodman illustrates the reach of inventio with a description of the senatorial attitude to Nero when he put off a trip to Egypt (Woodman 2004: xvii, on Ann. 15.36: in incerto erant, procul an coram atrocior haberetur, ‘‘[they] were uncertain whether he should be regarded as more frightening when at a distance or before them;’’ tr. Woodman 2004). But when Tacitus emulates Sallust's neatly paradoxical phrase about the frightening effect of powerful men (ut absens an praesens, pacem an bellum gerens, perniciosior esset in incerto haberetur, ‘‘so that it was not certain whether he [Jugurtha] was more destructive absent or present, at peace or waging war,’’ lug. 46.8) to convey an attitude from a century after Sallust's death, he is not treating a historical event by means of (inappropriate) invention but rather describing fear. That senators found Nero frightening is a well-attested historical fact, but Tacitus does not need a contemporary source to describe the nature of their fear. Once again, ‘‘background material’’ is sufficient. A modern narrative historian will have his or her own ‘‘background material’’ for describing that fear to modern readers.
I conclude this section with a passage that offers a glimpse of a Roman historian’s desktop. This particular historian, the elder Seneca, is better known as a rhetorician; his memoir on declaimers and declamation survives, albeit in a somewhat tattered form, but his ‘‘history from the beginning of the civil wars almost up to the day of his death’’ does not (the work is thus characterized by his son, the younger Seneca; see Peter 1914: 2.98). However, Seneca’s surviving memoir includes a discussion of four historical narratives of the death of Cicero, an event he himself must have treated in his history (Suas. 6.14-21).
Cicero was proscribed and killed in 43 BCE. The event occurred during the youth of Seneca and the earliest of his four rivals, Livy; for the other three - the Tiberian-era historians Cremutius Cordus, Bruttedius Niger, and Aufidius Bassus - it was an event of the recent past. The first question Seneca asks of the competing narratives is whether they agree as to the central fact, which they do. But rhetoric shows up in the very asking of this basic historical question, since in his view the central event is not the execution of Cicero but Cicero’s refusal to plead for his life. Seneca notes with approval the fact that his four authors ‘‘spin’’ the event similarly, and criticizes a fifth author, Caesar’s critic Asinius Pollio, for having Cicero abase himself before his execution. But he also reports that Pollio did not go so far as to include this nonsense (inepte ficta) in his history of the civil wars; he used it in a speech (Suas. 6.14).
After ascertaining the agreement as to the central event, Seneca turns to the individual accounts, quoting some of each. They agree on a number of subsidiary details as well: the litter in which Cicero died, the fact that he volunteered his neck, and, as Seneca notes (6.20), the display of his head and hand(s) in the Forum, the site of his oratorical successes. The individuating elements are quips (Livy, Aufidius Bassus), thoughts (Livy, Cremutius Cordus), feelings (Cremutius Cordus), and the description of the scene in the Forum (Livy, Cremutius Cordus, Bruttedius Niger); one author, Niger, names Cicero’s assassin, but the man is otherwise unknown to history. How Seneca bettered his predecessors in narrating this event we do not know, but he seems to have taken care to ascertain the relevant constraints on inventio.
World History
