《Mark C. Baker Rutgers University》.pdfVIP

* Degrees of Nominalization: Clause-like constituents in Sakha Mark C. Baker Rutgers University 1. Background The generative theory of categories is characteristically discrete. Something is a noun, or it is not. It has the binary feature value +N, or it does not. This is especially true within the category theory of Baker 2003, where the feature +V, characteristic of verbs, is understood as the ability to license a specifier, while the feature +N, characteristic of noun, is taken to be the ability to bear a referential index. Either a category has a referential index, or it does not; either it has a specifier of it does not. It makes little or no sense to say that a category has 50% of a referential index, or two thirds of a specifier. This discreteness seems well-justified for basic, morphologically simple lexical categories; see Newmeyer 1998 and Baker 2003 for some discussion. But in the realm of nominalization/complementation, one can observe some continuity, it seems. Some embedded clauses seem to have many nominal features and few clausal features; others have fewer nominal features and more clausal features. An early version of this observation was John Ross’s (1973) “Nouniness Squish”, where he investigated a seeming continuum of phrases that are more or less nominal in English. More recently, it is common to see discussions of degrees of nominalization in much functionalist-typological literature (Comrie 1976, Croft 1991, Koptjevskaja-Tamm 1993, Croft 2001, Cristofaro 2003, Malchukov 2004). The question, then, is how to reconcile this seeming-continuity of nominalization with the generative theory of discrete lexical categories. The Sakha language presents an interes