Lexicalism, periphrasis and implicative morphology1

Several other chapters in this book analyze grammatical phenomena according to the architectural assumptions and constraint-based representational apparatus of lexicalist theories of grammar; in the present chapter, we ask some fundamental questions concerning what it means for a theory to be lexicalist. While critically assessing some consensus beliefs among lexicalists, this exploration will also serve as a corrective to certain pervasive misunderstandings about lexicalism and its limitations as propagated by some of its more vigorous detractors.2 In particular, we provide a careful overview of certain lexicalist assumptions concerning the relation between lexical representations, morphology, and syntax, and we propose a lexicalist framework incorporating a much richer conception of morphology than has customarily been assumed in lexicalist theories. According to this conception, periphrasis (multi-word expression) is as much a mode of morphological realization as synthesis is. In our discussion, we assume the familiar distinction between derivational and inflectional morphology: while principles of derivation define systematic relations between lexemes, principles of inflection define a lexeme’s inventory of grammatical words and their phonological realizations; traditionally, an inventory of the latter sort is referred to as an inflectional paradigm. Recently, a number of researchers have argued for the morphological status of periphrasis in the inflectional domain (Ackerman & Stump 2004; Börjars et al. 1997; Börjars & Donohue 2000; Sadler & Spencer 2002; Sells to appear***; Spencer 2001***, 2003; and Stump 2001, 2002, among others, but see Kiparsky 2005 for a dissenting view); by contrast, the incidence of periphrasis in the derivational domain has, so far, received relatively little attention among proponents of word based realizational morphology. Here, we develop a morphological perspective on periphrasis in the inflectional domain, and propose an extension of this perspective into the domain of derivation. We accordingly focus our discussion on two kinds of periphrastic constructions: in the inflectional domain, we focus on compound tense constructions; in the derivational domain, on phrasal verb constructions. As a preliminary to our discussion, we propose (§2) a simple taxonomy of lexicalist approaches to periphrasis which will enable us to situate our approach within the landscape of alternatives. We then examine some specific instances of the relevant complex predicate constructions, identifying the theoretical problems that they present for standard lexicalist assumptions and developing realizational analyses which resolve