Pragmatics 7:4.461-497. International Pragmatics Association TROPIC AGGRESSION IN THE CLINTON-DOLE PRESIDENTIAL DEBATE Asif Agha 1. Introduction This paper attempts to answer a rather basic question about the nature of linguistic aggression: How do we recognize aggression in utterances? Or, more particularly: How do we construe any given utterance-event as an instance of an ‘aggressive’ act? Questions like this are examples of a more general kind of question, one with which any attempt to account for utterance meaning must contend; this is because all such attempts face the problem of the relation between events and actions. Spoken utterances are acoustical events. Yet when these spoken events register as ‘actions’, they do so because we can ascribe a significance to them. So, the more general question is: How? It is almost a truism to say that the construal of an event as an action occurs always from some situated perspective on the event, a perspective from which significance can be ascribed to it. Posed this way, the problem seems daunting. Yet, as I hope to show here, different perspectives on the same event do not differ randomly from each other. They can be scaled or calibrated one-to-another. The materials of the Clinton-Dole presidential debate which I consider here are very well suited for this demonstration, for reasons which will become evident below. But let me first say something about ‘tropes’ and why I think they are relevant to the problem of action. The modern view of tropes and figures in language is shaped most profoundly (and, in the most unfortunate way) by the period of the decline of rhetoric into mere ‘tropology’ (Ricoeur 1979, ch. 2), i.e., by the attempt to classify tropes into inventories and types, ever more detailed, by giving each a distinct name. Yet this attempt has some of the qualities of a rearguard action; it is an attempt to deal with a problem by containing it, in names. Today, it seems reasonably clear that the attempt exhaustively to name and enumerate ‘the tropes’ of language is futile; certainly, to every inventory so far given, we are always able to add some more. Yet the tropic qualities of language have interested everyone who has considered the problem of utterance meaning. This is partly because tropic utterances appear to have more than one meaning; they appear to be instances of more than one act. From this standpoint, the problem of ‘tropes’ has a natural connection to the problem of ‘action’: Our attempts to classify ‘events’ as ‘actions’ become mired, inevitably, in the grey areas between prototypical cases. Consider aggression. We may be glad of the pure instances of aggression that we find—for science, if not in life!—but only until we come upon cases where the aggressive qualities of an act are intertwined with something else, some quality which is the very opposite of aggression. Since such cases are routine in everyday life, the DOI: 10.1075/prag.7.4.02agh