Some Lexicalist Remarks on Incorporation Phenomena

An important theme in the study of language during the twentieth century has been the balance between the grammar of words (morphology) and that of phrases and sentences (Syntax). Some writers, such as Morris Swadesh (1939), have seemed to treat this as a merely terminological issue, referring to (much of) morphology simply as “internal syntax.” More recently, some (e.g. Selkirk 1982, Williams 1989) have argued explicitly that the mechanisms necessary to describe the internal organization of words are a direct extension of syntactic principles. Others, in contrast (cf. among others Anderson 1992, chap. 2; Zwicky 1992) have argued that there are substantive differences between morphological and syntactic principles which have important implications for the organization of linguistic knowledge. This discussion has largely centered on the analysis of a small number of constructions for which both syntactic and (more purely) morphological accounts seem to be available. If the syntactic account can be substantiated in these cases, the more general conclusion seems warranted that syntactic mechanisms are (or at least can be) responsible for combining pieces into complex words. If, on the other hand, there are reasons to prefer the autonomously morphological analysis in such circumstances (which are, after all, the best candidates for syntactic word formation prima facie), that conclusion casts doubt on the more general perspective from which structurally autonomous elements are manipulated syntactically to form unitary words, a notion that pervades much recent work.