ABSTRACT
Halliday (1994) correctly observes that the current opposition between theories of language is no longer “structuralists” versus “generativists” rather between those that are primarily formal and syntagmatic in orientation (i.e. syntacticians), and others that are paradigmatic (i.e. functional) and semantic in orientation. Our approach is a crosscurrent for it makes use of insights borrowed from one to the other. Thus, while generative syntacticians have been clearly advocating a maximally general phrase structure analysis of nonlexical/functional categories such as tense, aspect, modality, mood, negation and so on (Chomsky 1986, 1995; Pollock 1989; Ouhalla 1989, 1991; Whitman 1989; Bowers 1991; Carstens 1991; Benmamoun 1992, 2000; Fassi Fehri 1993 among several others), which used to be generated under a single structural node, namely Inflectional Phrase (IP), enunciativists and discourse analysts have been trying to advance systematic and uniform accounts for those same verbal categories, especially the ones expressing aspect, tense, and modality (Guillaume 1964; Benveniste 1966, 1970; Culioli 1970, 1976, 1982; Waugh 1975, 1987, 1990; Adamczewski 1982a,b, 1986, 1991; Givón 1982; Dahl 1985, 2000; Pica 1985; Delmas 1987; Joly and Roulland 2001 among many others) on the basis of their salience in any particular speech production. Despite their different theoretical constructs, they appear therefore to share a common objective the essence of which is to do justice to the structure and function of these verbal categories, which, for decades, have been put aside and considered the least salient items in language. As investigators have realized the extent to which the presence of these categories shapes the syntactic representation of the clause and governs its semantic interpretation, more research has recently focused on the behavior of these categories, and even on their universal nature.1 In addition, the study of such verbal categories as ATM within two different theories of language has theoretical and empirical implications. On the theoretical level, this move enhances the cognitive status of these verbal categories as a fundamental component of human thinking. On the empirical side, it has had a major effect on clausal syntactic representations, on the one hand, and their semantic interpretations, on the other hand.
