Robert Levine (levine@ling.ohio-state.edu)
Tue, 17 Oct 95 00:13:37 EDT
With respect to Steve's comment that we don't want to consider mass and plural nouns to be lexical phrases---why don't we? Would this outcome be any more embarrassing for the theory than the fact that bare stems are simultaneously words and morphemes, in the vast majority of cases in English? The class of words in question---personal names, pronouns, mass nouns and plurals---either select specifiers only optionally or not at all; they are thus either necessarily or optionally saturated, and their distribution should therefore be parallel to that of other saturated expressions, i. e. phrasal signs. I don't see what conceptual difficulties this raises or why any technical infelicities should result from such treatment. There was never congruence between the bar zero/bar higher-than-zero distinction and the word/non-word distinction; in GPSG, for example, a conjunction of bar zero categories was still a bar-zero category, as was a verb+particle construction like _look up_; conversely, pronouns were entered in the directly under NP nodes. Does the HPSG treatment need to be any different? Bob Levine
This archive was generated by hypermail 2.0b3 on Fri Dec 18 1998 - 20:33:04 PST