On the Inexistence of Specifiers and the Nature of Heads
Michal Starke
May 2001

This papers argues that there is no such thing as "specifier" (as distinct from "head"). Put differently, all merges are first merges and there is no such thing as "second merge" (with different properties from first merge). What we (mis)took for specifiers are in fact heads, in the sense that they project the label of the merger. As a result, all syntactic relations reduce to head-complement relations. It is argued that this simplification makes the theory both more elegant and more integrated. The independently needed theory of labelling now does much more work, and extraneous assumptions such as 'checking domains and mechanisms' are unnecessary.
Format: [ pdf ]
Reference: lingbuzz/000004
(please use that when you cite this article)
keywords: syntax, elegance
Downloaded:1355 times


[ edit this article | back to article list ]