1. The true author of the Invectio has been variously regarded as L. Piso, Clodius, and others. For a summary see Syme, Ronald, Sallust (Berkeley and Los Angeles 1964), 314–18CrossRefGoogle Scholar. Syme dates the invective to ‘the beliefs and practices of rhetoricians in the age of Caesar Augustus’ (318).

2. Voss, J., De Historicis Latinis Libri III (Leiden 1651), 75f Google Scholar.

3. Schwartz, E., ‘Die Berichte über die catilinarische Verschwörung’, Hermes 32 (1897), 554ff Google Scholar. = Ges. Schr. (Berlin 1956), ii.275ffGoogle Scholar.

4. The inspection came from Boissier, Gaston, La conjuration de Catilina (Paris 1905 Google Scholar); Drumann–Gröbe, , Geschichte Roms V² (Berlin and Leipzig 1919), 463ff.Google Scholar; Tolkiehn, J., PhW45 (1925), 1404f.Google Scholar; Baehrens, W. A., ‘Sallust als Historiker, Politiker, und Tendenzschriftsteller’, Neue Wege zur Antike 4 (1927), 38ff.Google Scholar; Seel, O., Sallust von den Briefen ad Caesarem zur Coniuratio Catilinae (Leipzig 1930 Google Scholar); Schur, W., Sallust als Historiker (Stuttgart 1934), 183ff.Google Scholar; Broughton, T. R. S., ‘Was Sallust Fair to Cicero?’, TAPA 67 (1936), 34ff Google Scholar.; Last, Hugh, ‘Sallust and Caesar in the Bellum Catilinae ’, Milanges de philologie, de littérature et d’histoire anciennes offerts à J. Marouzeau (Paris 1948), 355ff.Google Scholar; Steidle, W., Sallusts historische Monographien, Historia Einzelschriften 4 (1958), 15f Google Scholar., 93f.; LaPenna, A., ‘L’interpretazione sallustiana della congiura di Catilina’, SIFC 31 (1959), 18ff.Google Scholar; Earl, D. C., The Political Thought of Sallust (Amsterdam 1966 Google Scholar).

5. As always there are voices of dissent. The anti–Ciceronian bias is maintained by G. Funaoli, R–EIA 1922f.; Lämmli, F., ‘Sallusts Stellung zu Cato, Caesar, Cicero’, MH 3 (1946), 112 Google Scholar; Laistner, M. L. W., The Greater Roman Historians (Berkeley 1947), 56 Google Scholar; Löfstedt, E., Roman Literary Portraits (Oxford 1958), 100f.Google Scholar; and Hands, A. R., ‘Sallust and dissimulatio ’, JRS 49 (1959) 56ff Google Scholar. As recently as 1976, Renehan, R. argued for ‘disrespect’ and ‘parody’ in Sallust’s portrait of Cicero: ‘Imitation in Sallust and his Sources’, CP (1976), 99f Google Scholar. The pro-Caesarian bias is maintained by Baehrens, Last, Schur (all cited in n.4 above), Lämmli and Leeman, A. D., ‘Le genre et le style historique à Rome: théorie et pratique’, REL 33 (1953), 208 Google Scholar. The voices of dissent, however, have been in the main quiet since Syme.

6. Syme(n.l above), 94.

7. McGushin, P., C. Sallustius Crispus, Bellum Catilinae: A Commentary (Leiden 1977), 230 CrossRefGoogle Scholar.

8. I do not mean to imply that there is any such thing as ‘face value’ or that any reader can in fact read without reading between the lines. Recent literary theory makes it abundantly clear that reading is and must be a process of reading between the lines. I only grant, for the sake of discussion, that it is easier to agree about what some lines mean (or about what is between some lines) and very difficult to know what the ‘face value’ of other lines could possibly be. Here I address lines of which the interpretive history seems to clearly indicate that we cannot agree about what is ‘in them’ or between them.

9. Syme (n.l above), 104.

10. History as mimesis is not common in antiquity, although it is always some form of rhetorical endeavour. History’s mimetic nature is addressed in a rudimentary form in Diodorus. He complains (120.43.7) that the writing, the suggraphē, of events did not have the pathos of the events themselves, because the written document could not imitate events which happen simultaneously. The concern with pathos is literary and rhetorical; the failure of mimesis with regard to simultaneity suggests the importance of mimesis, the creation of the feel of events, elsewhere. See also Duris of Samos, FGH, ed. , Jacoby (Berlin, 1926 Google Scholar) 2A, 76, fr. 1. For this study, the most important analysis of ancient historiography is that of Fornara, Charles W. in The Nature of History in Ancient Greece and Rome (Berkeley 1983 CrossRefGoogle Scholar): history begins as mimesis, and, from Herodotus on, the historian’s use of direct speech was intended ‘to create the illusion that [the reader/listener] was the observer of the deeds under description’, 30–32. My discussion extends Fornara’s observation to the report, direct or indirect, of rumours, charges and beliefs. The text creates for Sallust’s reader the complex interactive network of belief and suspicion that surrounded events both at the time and later, or, in Fornara’s terms, the illusion that the reader is present at the vague and contradictory explanations of and reactions to unsubstantiated charges. I am also arguing that Sallust does this because those contradictions are the interpretive problem which Sallust returns to the reader.

11. 1 am trying to demonstrate that the assumption that, if a conflict in the text is unresolved, it is because it is, at least for Sallust, unresolved, is an efficient, consistent and coherent assumption. My emphasis, therefore, will fall on the surface of the text, and I will try first to show that the conflicting strands of this narrative do not lead out of the labyrinth, and second to demonstrate that the experience of pursuing the conflicting strands is a valuable one that suggests ‘facts’ about the inherent obscurity of the events. To use the terms of S. Langer, Sallust creates a virtual labyrinth because the obscurities and impasses of the text are expressive of the life of the times (pathos); see ‘The Great Literary Forms’, Feeling and Form (New York 1953), 280–305Google Scholar. While this philosopher is not in fashion these days, her notion of ‘presentational forms’ is still useful, in part because it may mark off a symbolic territory which resists current deconstructive ambitions. It will be seen in what follows that my method is eclectic, but that it owes most to ‘reader–response criticism’. I use the American designation because this analysis is little influenced by the formal and categorical concerns of Wolfgang Iser and the German school of Rezeptionstheorie. The theoretical underpinnings for this kind of analysis were first laid out by Fish, Stanley E. in his Appendix to Self–Consuming Artifacts (Berkeley 1972 Google Scholar), ‘Literature in the Reader: Affective Stylistics’, now available in a useful anthology, Reader–Response Criticism: From Formalism to Post–Structuralism, ed. Jane P. Tompkins (Baltimore and London 1980), 70ff Google Scholar.

12. Fornara (n.10 above), 30 and 142, although the general idea is developed throughout.

13. The phrase come from Fornara (n.10 above), 73, although the next paragraph seems to contradict this sense of a non–analytic Sallustian mimesis. There is, however, no clearer example of Sallust’s tendency to prefer mimesis to analysis than his own description of the difficulties which face an historian (3.2): dein quia plerique quae delicta reprehenderis malevolentia et invidia dicta putant, ubi de magna virtute atque gloria bonorum memores, quae sibi quisque facilia factu putat, aequo ammo accipit, supra ea veluti ficta pro falsis ducit. Secondly because what you say in criticism of misdeeds is thought by most to be the product of malice and envy; when you record the significant virtue and fame of the great, everyone is happy to accept what he thinks he could easily do himself, but anything beyond that is treated as fiction and therefore false. The commentaries often note that this is an adaptation from Thucydides 2.35.2; and McGushin (n.7 above ad loc), following Büchner, further comments that in Sallust’s adaptation the invidia is ascribed to the historian, rather than the audience. One should, however, go further and note that Sallust here creates the perspective of the invidious reader when he assigns invidia to the historian. Supra ea veluti ficta pro falsis ducit is a fair paraphrase of Thuc. 2.35.2: tōi de huperbailonti autōn (‘to what goes beyond that’ = supra ea); phthonountes ēdē kai apistousin (‘they react with envy and disbelief’ = flcta pro falsis ducit). The concept not translated, phthonos (‘envy’), appears not as analysis in Sallust but as another objection the reader might make. The echo of Thucydides provides Sallust’s reader with the reason for the hypothetical cries, ‘ficta’ and ‘invidia’, and for the reader who remembers his Thucydides we have the rich image of the invidious reader crying ‘invidia!’. But Sallust himself only surrounds his hypothetical historian with those who attempt to distort his fama (‘reputation’) and fides (‘trustworthiness’). This is, then, an image or picture of the problem the historian faces, an echo of the voices of his critics, not an analysis of the forces at work. The technique is different from Thucydides’ (or Pericles’) technique in its greater personal concern with fama and in its mimetic and subjective form.

14. The provocativeness of Sallust’s text is more than evident in the commentaries it has produced. Vretska is indicative (ii.485): ‘Wieso — muss man fragen — fiel dieser unbekannte Mann uberhaupt auf seinen Weg nach Norden auf, wenn nicht eine Durchstecherei — absichtlich? — dahinter war?’ When I argue that Sallust ‘intends’ to provoke just such questioning, I do not grant that he ‘hints’ at any answers. In fact, I am arguing that when the truth is not known from later evidence, the text does not so much create an illusion of suspicion and uncertainty, but the effect of suspicion and uncertainty. Consequently, the history of the interpretation of Sallust’s text is a history of critics who act like one or another actor within Sallust’s text: some, like Crassus, blame Cicero; some, like Tarquinius, blame Crassus; some like the equites, find the charges against Caesar believable. If this is true, Sallust’s text should, at a more sophisticated level, create a ‘meta–reader’ who stands above quod dictum est aut credi potest (‘what is said or what can be believed’).

15. Since ‘facts’ are ipso facto already interpretations, it is important to stress that, by my interpretation, Sallust found the parallels significant and the response to the accusation significant. These, then, are the (interpreted) ‘facts’ to which he draws attention. The mimesis is the appearance of these provocative parallels and this dubious charge without the intrusion of Sallust’s own analysis. The picture created is suspicious but inconclusive. Thus this mimesis, like any mimesis, wavers between what an ‘implied observer’ would have noticed at the time and what seems significant to Sallust when he comes to the events later. Mimesis is always (un)grounded somewhere between imitation of what the thing itself is and imitation of how the maker feels about it.

16. One may think of these explanations as what was thought, as what Sallust imagines the senators whould have said or muttered, or as what they said to him later. They are the (internal or external) voices of interpretation, and represent variations on fama. The mimesis, however, is not in any direct representation of speech, but in the way that Sallust’s words are suited to the actors and represents their subjective responses. The first and third groups, for different reasons, are blunt and direct. The second group, admittedly duplicitous, is also self–justifying: magis leniunda quam exagiianda (‘better to mollify rather than provoke’) is an effort to put a ‘good face’ on their own misrepresentation. I can see no other explanation for (1) why the second group offers a rhetorical justification and the third group refuses to announce its opinion about Crassus, or (2) why the groups are not, as we shall see, mutually exclusive, and, therefore, cannot be a complete and objective analysis of the vote.

17. It is the received opinion that Sallust gives us three groups; cf. Vretska (n.14 above), ii.481: ‘Dreifach sind die Meinungen gespalten.’

18. There are, of course, precedents in Sallust for both interpretations. B.J. 58.2: alii… alii …magna pars…; here, magna pars is a great part of the totality of the other two. B.J. 31.13: pars … alii … plerique; it is unclear here whether the perpetrators of the caedes are a subgroup of those responsible for the quaestiones iniustae, or a third group. B.C. 61.8; B.J. 32.3; B.J. 69.2 all offer three distinct alternatives; however in these examples there is a single operative differentia.

19. Note that existumabant (‘they thought’) in group two is already redolent of the accounting of group three.

20. In n.ll above, I referred to Langer’s ‘presentational forms’ as part of the theoretical background to this discussion. Her notion of ‘expression’ derives in part from Pierce’s ‘icons’, which, according to Pierce, are significant in that they may reveal more about their signatum than was necessary to create them: ‘For a great distinguishing property of the icon is that by the direct observation of it other truths concerning its object can be discovered than those which suffice to determine its construction.’ (Pierce, , ‘Logic as Semiotic: The Theory of Signs’, in Semiotics: An Introductory Anthology, [Bloomington 1985]Google Scholar, 11 [ = ‘Sign’ in Baldwin’s Dictionary of Philosophy and Psychology 1902].) In a sense, then, they are like a good scientific theory which becomes acceptable when it explains or predicts more than the problems it was designed to explain. See Kuhn, Thomas S., The Structure of Scientific Revolutions (Chicago 1962; repr. 1970), 154f Google Scholar. If we apply this model to Sallust’s text here, we may (and should) extend our contemplation of Sallust’s presentation in ways for which we do not have to assume self–conscious intention or that this was the problem Sallust specifically designed this particular text to explain. For instance, the interaction of forces presented here explains both why the accumulation of wealth and power by men like Crassus was dangerous and destroying the Republic and why, at some point, it became necessary, because it was the only way to protect oneself against a certain kind of attack.

21. Cf. Romae omnia venire (‘[Jugurtha believed] that everything could be bought at Rome’), B.J. 28.1.

22. Videri is a direct, precise, and double–edged equivocation. It is a technical term for the decisions of an investigative body, and, as such, may be translated ‘be deemed’. However, just as the American jury operates under an admitted legal fiction (that it is the finder of fact), so the Latin term videri reflects its origin as a judgement or opinion about matters of uncertainty. In other contexts, and in general, it means only ‘seems’, ‘appears to be’, or, in its literal sense, ‘is deemed’. Thus, while Sallust reports the Senate’s vote with the lex legitima, indicium falsum videri, just two lines after all had cried out, indicium falsum est, he reminds his more subtle readers of the essential uncertainty. The Senate’s judgement was at best an opinion about uncertain appearances.

23. For a survey of the possible occasions, see Vretska (n.14 above), ii.488. On the unlikelihood of Cicero being behind Tarquinius’ accusation, see Syme (n.l above), 104; McGushin (n.7 above), 232.

24. ‘Could’ is the important word, here. Sallust does not say that Cicero named Crassus, only that it was believed that Cicero did; nor does he say that Autronius named Crassus.

25. See Vretska’s (n.14 above) careful structural analysis of this contrast, ii.482.

26. Sed may by resumptive or transitional as well as contrastive; in fact, it may be both contrastive and resumptive. It is the context that determines how we shall name the usage; the same is true of the English word, ‘but’. Sallust here uses sed in a passage where the content is clearly contrastive. If there had been no contrast, and if Sallust were ‘merely’ moving to a new topic, sed would still mean ‘but’, but it would be a rhetorical ‘but’, dismissing part of the narrative or argument in order to focus attention on a new point, rather than a discursive ‘but’, which shapes a proposition with contrasting parts. Even in English ‘but’ can mean ‘and’ in this rhetorical sense. That possibility is not in dispute. The point is that the content, at least in part, is contrastive.

27. It is typical of the passage as a whole that this vignette of Cicero’s virtuous resistance does not allow the readers to reject Crassus’ accusation. Just as the differentia above changed, here the objects of inimicitia and so the purposes of (false) accusation have changed. In rhetorical terms, Sallust offers an ethos argument which may or may not be relevant to the particular accusation.

28. Res autem opportuna videbatur, quod is privatim egregia liberalitate, publice maxumis muneribus grandem pecuniam debebat (‘The opportunity seemed to be there, because he [Caesar] had incurred enormous debts due to his exceptional private generosity and his lavish public entertainments’, 49.3).

29. The circumstances in which rumours become common and the type of characters who believe them are outlined with specific reference to Tacitus by Shatzman, I., ‘Tacitean Rumours’, Latomus 33 (1974), 549ff Google Scholar. They are the same for Sallust: the rapid and uninformed talk of the unempowered under unusual circumstances, usually caused by and producing fear and irrationality. See AC. 29.1; 30.2–3; B.J. 13.1; 40.4–5.

30. Didaxis is a well recognized element of the later Graeco–Roman tradition. See Fornara (n.10 above), 108–10. Just as I argue that Sallust extended the ‘interpretation of the oral tradition’ to include the obscurities caused by rumour and ‘what could be said’, so I would like to suggest that the problem of misinterpretation, not the character of an individual, can be an exemplum. In a sense the misinterpretation the equites represent is a negative exemplum in the same sense that Catiline himself and the conspiracy are a negative exemplum.

31. ‘This incident … probably occurred after Caesar’s oration in the debate of Dec. 5 … S., through forgetfulness or by design, has shifted the chronology …’ McGushin (n.7 above), 234. My account gives the didactic design. It is, however, not necessary to align a detail like this with the accounts of Plutarch, Caes. 8, or Suetonius, Div. Jul. 14.2. It is, in Sallust’s account, exemplary of the dangers that surrounded suspicion, and the efficacy of rumour.

32. The mimesis here is the same as that described above when Sallust transfers invidia from its position as the cause of objections to the objection itself. There Sallust created the (invidious, ‘spoken’ only by allusion to Thucydides) reader crying invidia. Here, Sallust creates the (unstable) equites thinking animi mobilitate … quo studium [nostrum] in rem publicam clarius esset (’through instability of character … to make [our] loyalty to the republic all the clearer’). This interpretation requires acceptance of (1) subjective mimesis in Sallust, (2) Sallustian cynicism about actors and causes, (3) Sallustian psychological sophistication. This essay is about all three.

33. McGushin (n.7 above), 234. One must modify this distinction at least to read ‘one based on the situation as the equites understood it, and they were wrong’.

34. The only possibility that could make periculi magnitudo an explanation based upon the situation is that the equites were randomly drawing their swords at everything that moved because of the extent of the danger. This, of course, makes it a non–explanation for why they drew on Caesar. The only possible explanation ‘based upon the situation’ for their drawing on Caesar is: a mistake! wrong danger!

35. Fornara (n.10 above), 53f.

36. Cf. ex aliis rebus magis quam quod cuiquam id compertumforet haecfama valebat (‘this report gained acceptance for reasons other than that anyone had found evidence to support it’, 14.7); nobis ea res pro magnitudine parum comperta est (‘I do not have sufficient evidence to make a judgment on a matter of such gravity’, 22.3); and quam verissume potero, (‘as truthfully as I can’ 4.3). See in general Buchner, K., ‘Das verum in der historischen Darstellung des Sallust’, Gymnasium 70(1963), 231ff Google Scholar.