Response by

The ultimate assessment of my book that Prof. Dyck offers in the final paragraph of his review is fair enough: if I am any judge, the contribution of the book is on the side of semantic explication and close reading. Since language and culture move on not entirely parallel tracks, illuminating cultural development by way of historical semantics is a vexed business with the best of evidence and in the surest of hands. Thus it was very welcome to read that Prof. Dyck found some of my analyses, particularly those of Catullus, illuminating and that he found the project a worthy one.

I write here only to respond to a few points that come up in D.’s review. Readers of the review may have the impression that the intellectual project of the book is a word study aiming at Catullus’ polymetrics; but it is not, or is not only, such a study. I also propose a pragmatic (in the non-technical sense) model of semantic change and its relation to social change and ideological construction. The book is one case study in the use of a model which I hope may prove of some value beyond the confines of my own project.

Perhaps I may sketch out the model, described in the book on pp. 6-10 and 77-87 and central to its entire argument. I see the distribution of a word—that is, its applications to various referents—as the outcome of the interaction between the semantic structure of the word and prevailing ideologies. The semantic structure of a word, I argue, consists of a potentially enumerable number of semantic features.1 These features, in turn, pass through the ‘filter,’ as it were, produced a given ideology. This ‘filter’ is the critical determinant of the referents upon which a given lexeme can alight in the act of labeling.

To take one example: ‘bellus’ contains, as I argue, two elements: I. ‘positive evaluation’ and II. ‘reference to evaluator.’ That is, ‘bellus,’ quite in accord with its origin as a diminutive of ‘bonus,’ represents the act of positive evaluation as the subjective response of a given party. This is different from ‘bonus’ itself, which represents the evaluation as objective. Hence the difference between ‘bene habere’ ‘to be well situated,’ as of a political position, vs. ‘belle (se) habere’ ‘to feel fine,’ as of health or mood (pp. 51-52).

‘Bellus’ is applied to different referents according to the way ideologies of subjective evaluation are constructed at a given time or in a given text. In Plautus there was generally a ‘filter’ running which defined the evaluator as a single party representing his own response. By contrast in certain passages of Varro, Cicero, and Catullus, it is clear that a different ‘filter’ was in operation, giving ‘bellus’ a different sense, one referring to an evaluation acceptable to a smallish body of like-minded, typically elite evaluators. Thus Varro’s ‘belli homunculi,’ of convives (Men. fr. 335) or Cicero’s ‘hominis… et belli et humani,’ of a person who keeps track of birthdays (Fin. 2.102).2

When the distribution of an entire set of words appears to change in some specifiable direction, then one may well wonder whether some new ideology has not drawn them in, affixing them to the referents with which they now consistently appear. One may wonder, in other words, whether the ‘filters’ through which a given a set of lexemes is run have not been altered by a new governing ideology or a new discursive formation. In my estimation, the lexemes I consider in this book—’bellus,’ ‘elegans,’ ‘lepidus,’ ‘facetus,’ ‘festiuus,’ and ‘uenustus’—did indeed undergo such a change. Scattered across a variety of referents in Plautus, they appear in the late Republic applied almost exclusively to a very narrow and very particular set of referents: virtually all aesthetic practices that figured in the self-presentation of the social elite. (Hence the label ‘language of social performance,’ hereinafter LSP.)3 These aesthetic practices, to the extent that it is possible to tell, seem to have come to full bloom first in the second century. That, of course, was a period when the social elite was increasingly flush and experimenting with new forms of cultural practice and new modes of self-expression.4 Hence my own belief that the filters which govern the use of the LSP in the late Republic developed or coalesced in response to the social changes of the second century.

Of course it need not be the case that the development happened in the second century; it may well be that a different speech register, already existent or nascent at the time of Plautus but never used by him, simply comes to dominate later sources.5 The cardinal point is that the ‘filters’ which govern the use of the lexemes by Plautus on the one hand and Cicero et al. on the other are different filters, generated by different ideological systems. The book is not an attempt to trace the history of a set of lexemes per se, but to trace through certain applications of that set the construction of the ideological category with which the set is standardly associated in the late Republic and which seems to find its primary lexical expression in that set—a category something like “the social performance of Hellenism or aestheticism as a means to individualize oneself” (p. 9).6

In chapters 5-8 I aim not merely to offer close readings of passages where the LSP occurs but also to understand the social construction of the category that the LSP represents. Passages from Cicero’s speeches prove to be especially illuminating, since they reveal the variety of stances that were possible towards the category of “the social performance of aestheticism” and demonstrate not only how contentious that category was but also some of its characteristic properties. The use of the LSP in de Oratore, in which received uses of the lexemes are subtly, or not so subtly, altered reveals Cicero’s partly strained attempt to harness a category of obvious contemporary power to his own nostalgic construction of aestheticism. Catullus’ poetry, by contrast, cares nothing for harnesses; instead of subordinating attractiveness to social worth, as do most authors who use the LSP, Catullus’ use conflates attractiveness and social worth. Thence the LSP becomes the keywords for his polymetrics.7 Chapter 4, which treats the application of the LSP to the technical language of rhetoric, suggests that the contentiousness which clearly surrounds the use of the LSP in later texts may be imagined as having figured in the original application of that language to the nascent Latin rhetorical tradition.8 From the Augustans forward the category—if indeed it continued to exist in quite the same way—seems to lose its special connection to the lexemes.9


1. These features are not necessarily stable over time. ‘Venustus,’ for example, seems to have lost the element ‘erotic’ in certain applications by the time of the late Republic, while retaining it in others. Thus Cicero’s ‘nihil uenustius quam tua pegmata’ (Att. 4.8.2), where ‘uenustus’ has its occasional connection to ‘well-arranged’ shapes (cf. pp. 44-47) and does not seem to be erotic in any meaningful sense, whereas all of Plautus’ ‘uenusti’ are (pp. 42-43). D. has reversed my argument, suggesting that I claim ‘uenustus’ acquired the element ‘erotic’ by the late Republic; rather, I argue that it sometimes retained it, sometimes lost it (cf. e.g. “…in all those applications where ‘uenust(us)’ had to do with practices of elite self-definition, it has been de-eroticized by the late Republic,” p. 50).

2. Plautine Latin does not always provide a neat contrast with post-Plautine Latin. D. misreads my mention of Plautus’ ‘benigne’ (Mil. 729, p. 33 n. 42) depicting me as “keen to separate” his use from Lucilius’ (1269M). In fact I cite the Plautine passage with the exact opposite intention: to illustrate one of the points which it is the purpose of that section of the argument to establish, namely that certain words and phrases describing elite display behavior have apparent precedents in the language of New Comedy, a fact that complicates their interpretation.

3. D.’s label ‘language of socio-aesthetic appraisal’ is very apt.

4. D. represents me as having supported only by three pieces of evidence the claim that “all manner of aesthetic forms began to be used in the second century not to express Roman civic ideals…but also to advertise the status of a particular social class.” This misrepresents the argument. The three pieces of evidence in question describe only possible instances of *active* participation by the social elite in certain forms of performance; the preceding paragraph enumerates other aesthetic forms which the elite consumed, possessed, or observed, as it were *passively*. But D. ibid. may be right that it is better to speak of individualism within a class rather than “advertis[ing] the status of a particular social class”—although that view also requires speculation about the social and psychological economies of the Roman elite.

5. One might have the impression from the review that I was not aware of the latter possibility, but cf. e.g. p. 22: “The difference between Plautine and later Latin suggests two, not wholly exclusive, possibilities: either something happened after Plautus that forged the language of social performance into a coherent set; or circumstances changed sufficiently to allow the appearance of a set that was already coherent, perhaps only partly or incipiently so.” It is true, however, that I do use the language of “development” and “change” throughout the book.

6. Hence my omission of the terms ‘bonus, decorus, honestus, praeclarus, pulcher, suauis,’ which D. suggests I should have considered: these adjectives and their derivatives typically appear under the sway of a different ideological category than the one which governs ‘lepidus, bellus’ et al., viz. that of substance as against style; and, with the very occasional exception of ‘suauis’ (cf. pp. 190-91 on Cicero Pis. 93), these words are standardly ameliorative, not sharing with ‘lepidus, bellus’ et al. the curious property of being now ameliorative, now pejorative. I attribute the alternation of ‘lepdius’ et al. not to any inherent property of the individual lexemes but to the variable attitudes towards the ideological category that they commonly represented.

7. I am glad to find D. in agreement with my claim that Catullus’ evaluative language did not spring full-grown from his head, but owed much to the use of the LSP in contemporary Roman culture.

8. Regrettably I handled less than adroitly the precise contribution of Stoicism to the ideological contention that I imagine for this period, as D. has helpfully pointed out. The question of the influence of Stoicism aside, it is indubitable that in Roman culture it was possible to construct aestheticism both negatively and positively, and that, while there were strands that preferred one construction to the other, it was also possible (certainly in the late Republic) to switch preferences (or rather, to represent oneself as having switched preferences). To call the strands that constructed aestheticism positively, as D. does, a “philhellenic coterie”seems to me, by conjuring up the image of the Scipionic circle and of intellectual adherence, as against cultural sensibility, to minimalize the probable scope and cultural effects of Hellenism-cum-aestheticism in Roman culture. The consumption and display of the aesthetic products of Hellenic culture did not require “Greek erudition” per se. A shared cultural shift, if it be widely enough distributed in a given culture or subculture, can indeed have a real effect on semantics in a way that the usage of a single author may, ceteris paribus, be thought less likely to do; hence to reject, with D., the possibility that such a shift can have palpable effects in the use of a set of lexemes on the grounds that Catullus’ creative usage of those same lexemes did not have such effects is to mix apples and oranges.

9. The review may leave the impression that I argue that the association of the language of social performance to Marc Antony was the sole cause of this broken connection. In fact I hang my hat on the different claim that “one choice among the many made by Augustan poets in bounding their characteristic idiom, the choice to reject the language of social performance, can be explained precisely, by referring to the cultural model that that language commonly brought to mind, and *perhaps* to a particular set of political associations” (p. 304, emphasis added here). The point of this section is to speculate on two unanswerable, because negative, questions: why, when Antony’s personal life seems tailor-made to be attacked by an idiom that expressed “aesthetic performance,” does Cicero not use it in the Philippics, save for a single instance, and that in an apparently stereotyped phrase (Phil. 2.20, p. 296 n. 26)? And why, when the language of social performance provided such an attractive idiom for Catullus to express his poetic ideals (ch. 7 passim), do the Augustan poets, whose poetic ideals are partly similar, avoid it almost entirely?