But when we consider not just the disposition of ideas in the text but the very process by which words are interpreted, similar struggles between the transhistorical and the contingent come into play. Cicero provides a deceptively simple precept for reading history that points to some of these Sallustian complexities: “in historiography all things are referred to ‘truth’” (cum in illa (sc. historia) omnia ad veritatem, Quinte, referantur, Leg. 1.5).31 Although truth there is being defined specifically against pleasure as the criterion for evaluating history, when we apply this dictum to the interpretation of historical narrative,32 it possesses an element of ambiguity which Sallust’s procedures here make acute. What do we mean by veritas? For even this word might find a different signification in a philosophical treatise (not to speak of the debates among different philosophical schools) and in history, where it defines what actually was done against falsifications. If we imagine this formula describing the specific hermeneutic predisposition a contemporary reader would bring to his work, it could only complicate the referential dimension of Sallust’s own words, and this might immediately appear as the Catiline’s own universals (omnis homines, omnis vis) are immediately anatomized into warring factions (the energetic vs. the slothful; the mind vs. the body). Presenting the historical truth of events and absolute philosophical truth as the alternative referents of Sallust’s language would direct the reader equally to absolute intellectual formulations, potentially like Platonic forms whose truth precisely cannot be apprehended through earthly experience, and to that confusing experience itself, to the competing appropriations of language that would also have shaped a reader’s practical sense of what virtus was.
Applying this view of the semantics of historiography to the basic units of the text shows how Sallust’s words are not only instruments for representing the past but themselves synecdochic instances of temporal change. Indeed, a speaker within the text, Cato, makes the same point explicitly: iam pridem equidem nos vera vocabula rerum amisimus: quia bona aliena largiri liberalitas, malarum rerum audacia fortitudo vocatur, eo res publica in extremo sita est (52.11). This much studied passage itself seems an instance of the specific application of a universal truth, Thucydides’ famous generalization about the linguistic disruptions that result from political revolution, itself induced from what happened in the Corcyran revolution of 427 BCE.33 Cato on the contrary seems to be appropriating it as a description of a specifically Roman experience (“we have lost…”). What makes this appropriation so significant, and so crucial to the programmatic significance of the claim for Sallust’s work, is the subtle amplification of Thucydides’ point. As Batstone observes, Thucydides never said that any of the competing judgments expressed by words were true. Thus as Cato seems to be appropriating the impartial analysis of actions undertaken by the historian—and it is important to recognize that the revaluation Thucydides describes results from the polemics of both parties—he is also retrojecting ex post facto historical description into counterrevolutionary argument. This is signaled when he takes over the generic marker of historical discourse, its veracity, as a rhetorical ploy to declare that his own words are, like history, true. In fact, however, the reminder of the generic origins of his claim (viz. in Thucydidean historiography) potentially undermines such a move by reminding the reader that either side could assert that their own language was “true.”
But if Sallust, as historian, uses the allusion to Thucydides to point a distinction between the history that describes actions and the rhetoric that influences them, Cato’s own allusion to Thucydides pulls in the opposite direction by suggesting that history itself could become part of a political debate. And this blurring of the ability truly to name events when the naming itself becomes part of events points to another slippage in Cato’s own language that will also come to describe a key phenomenon of Sallust’s own text. Thucydides’ account of revolutionary language reveals the complexity of the relationship between words and things in part through the way it challenges interpretation. As Wilson (1982) argued, the most common and resonant rendering of the Greek words yields “nonsense”: the point is not that the words themselves have “changed their customary meanings,” as most translations would have it, but that the customary valuations placed on things by means of words have changed. Similarly, the main force of Cato’s complaint is that we no longer use the proper terms to designate actions: “giving away others’ possessions” is the “true” way of calling out what is going on, but we renounce that description in favor of a more positive sounding “liberalitas.” However, as the assonantial repetition of things (rerum ~ malarum rerum) and naming (vocabula ~ vocatur) in the next clause hints, this partisan euphemism simultaneously wrecks the possibility of describing things truly, and degrades language itself. For just as the term audacia is displaced as the true index of what Catiline is doing, so the replacing word “fortitudo” no longer uniquely refers to a brave action but also to a reckless one. In this sense, Cato’s phrase validates the “mistranslation” of Thucydides. If, in the real events of history, deeds are described by words that do not reflect their “true” nature (so that stealing becomes “generosity”), the ability of history as a genre that refers to reality to describe truth is similarly affected because the words it uses no longer retain a stable definition. Cato’s complaint about “the loss of words” might seem immediately falsified by their very persistence in the text: liberalitas and fortitudo are right before the readers’ eyes.34 But just as Cato makes Thucydides’ timeless analysis a way of winning a particular argument in history, so he creates a split within the very words of the historical text between the synchronic and the diachronic: the forms of words stay the same, but what they refer to changes.
In this respect too, Cato’s claim can inform a reading of Sallust’s text. For, as Sallust’s restrictive and somewhat idiosyncratic moral vocabulary might make more noticeable, crucial words like virtus and ingenium will recur unchanged throughout his text, and yet their significations will be transformed, both because of the historically inspired tendentiousness of the other voices that appropriate them and simply because of the growing complexity of Sallust’s discourse. This double sense of words as always balancing a static form against a continually changing pattern of references suggests how Sallust’s text as text possesses two temporal dimensions. Whatever virtue comes to mean in the represented past or in the discourse itself, it abides as an unchanging sign, perhaps as an eternal memorial of instances of individual virtue in the world. But the shifting meanings of virtue portray the linear flow of the text as a mimesis of the transience of experience, as the different meanings of virtue brings the text closer to the historical struggles it reproduces.
This way of relating Sallust’s language to the thematic dialectic between mortal experience and eternity, through the contrast between form and meaning, can be replicated yet again even at the level of their form alone. Cato’s interest in specifically vera vocabula turns his phrase into a virtual translation of a branch of language study that was central to several intellectual projects of the late republic, etymology. Varro’s treatise on the Latin language, the first third of which is devoted to etymology, appeared at virtually the same moment as Sallust began his career as a historian. Varro’s use of the term, as de Melo helpfully makes clear, is very different from Cato’s because he is not interested in semantics (de Melo 2019, 41–3), in how the meaning of words change, but rather in how their form has been altered over time from when it was originally imposed on its referents.35 This consciousness of words’ forms themselves as shaped by history forms a complement to the use, not least by Varro himself in his antiquarian works, of etymology as evidence for recovering the Roman past, and it suggests how the forms of words alone can be indices of change.
Not only does the abundance of archaism in Sallust’s diction and orthography give temporal depth to his diction, but he can exploit formal similarities and etymology to amplify the power of words to signify the processes of history. An example comes at a crucial moment in Rome’s moral history that also has profound semantic consequences, the advent of ambitio:
Sed primo magis ambitio quam avaritia