biling@ucscd.UCSC.EDU (Doug Rosener) (02/24/89)
In article <3322@uhccux.uhcc.hawaii.edu> lee@uhccux.uhcc.hawaii.edu (Greg Lee) writes: > >A reasonable way to rate the prospects of an analytic approach is to ask >(and answer) the question: what has it helped us find out? Looking at >the score for the last few years, and sticking to fundamental >discoveries, I make it syntax: 3, semantics: 0. The discoveries are: > >(1) Movement constraints (Haj Ross) -- constituents cannot occur > "too far" from where they belong, >(2) Cross-over (Paul Postal) -- nominals cannot come on the wrong > "side" of coreferents, >(3) One per sent (Charles Fillmore) -- when nominals are classified > by role (agent, patient, ...) one finds at most one of each > role represented per clause. > >(Disclaimer: probably few linguists would agree with my scoring.) > >My conclusion is that semantics as currently conceived has not >gotten us anywhere, and probably never will. > > Greg, lee@uhccux.uhcc.hawaii.edu I am presently taking a course in Montague Semantics and have had courses in syntax and phonology. The Montague material is by far the most impressive I have seen (o.k., that might not be saying much). If you have mastered this material, could you please explain what you think its shortcoming are? It seems to capture quite a bit of the problems of sense and reference (all that worrying about the present king of france on the net a while back), handles opacity ("I am looking for a unicorn" doesn't presuppose a unicorn exists but "I am petting a unicorn" does). I know there is a whole world of problems left to solve, but what direction do you see as a better one. Naively yours, Doug Rosener
lee@uhccux.uhcc.hawaii.edu (Greg Lee) (02/24/89)
From article <6451@saturn.ucsc.edu>, by biling@ucscd.UCSC.EDU (Doug Rosener): " " I am presently taking a course in Montague Semantics and have had " courses in syntax and phonology. " " The Montague material is by far the most impressive I have seen " (o.k., that might not be saying much). " " If you have mastered this material, could you please explain what you " think its shortcoming are? For one thing, Montague grammar is not a semantic theory in the current sense (grounding symbols), or in any real sense. Its models incorporate nothing about our perception of the world. It's just a theory of syntactic types. Yes, I know that it distinguishes syntactic types from "semantic" ones -- but except for an ad hoc notation for marking off, e.g., common nouns as different from intransitive verbs (with multiple slashes), the semantic types and syntactic types are notational variants. For another, the rule in natural language is that verbs do not create opacity. In MG they do, unless a special meaning postulate ad hoc for each verb makes them transparent. So MG gets that wrong. For a third, the compositionality assumption that is basic to MG is obviously wrong for natural language, since there are idioms in natural language. Finally, and crucially, there is no fundamental discovery about natural language that is incorporated into MG or that MG has helped us to discover. I hope that some of what I've just said is wrong, since I like MG -- I enjoy symbol pushing. I await instruction. Maybe further discussion ought to be restricted to sci.lang? Greg, lee@uhccux.uhcc.hawaii.edu