*!*Ghomeshi, Jila and Diane Massam. 1994. To Appear, Linguistic Analysis
tải về 302.24 Kb.
|
|
a background feature, do we need to say TELOS is optional?): it can be telic. Raising predicates take a TELOS feature but do not specify a value for it: it is not picky about its result state (seems odd after just saying that paint has an OPTIONAL TELOS feature). Subjects of non-VPs must be expressed locally: the predicate and its subject must be cocomplements [what about depictives?]: A non-VP predicative complement XP must be predicated of a direct o-commander of XP [i.e., if A directly o-commands B iff A is immediately to the left in the subcat list.] SUBCAT: items from the list are cancelled from right to left as they are combined with the verb in phrase structure [procedural defn?] Control resultatives: paint: [SUBCAT $<$ NP, NP$_1$ (AP[SUBCAT $<$ NP$_1$ $>$])$>$] laugh: [SUBCAT $<$ NP, [1]NP, (PP[SUBCAT $<$[1]NP$>$])>] [this isn't quite right, since the PP isn't optional if the NP object is
"agents must be expressed as subjects...leaving only the object position for the raised NP" pg 8 Compares with unaccusative accounts.
*!*Willems, DOminique. 1997. On Linking semantics and syntax through the lexicon. A macro-lexical approach of French verbs. Gent, Belgium. Talk at Stanford U. April 1997. Collected 3000 French verbs and categorized them by construction [Constructional polysemy:]
Change of verb meaning in different constructions: \eenumsentence{ Il glisse. (He slid) Il glisse un livre a Marie (He slid a book to Marie) \eenumsentence{ la branche plie le vent plie la branche Pierre apprend l'espagnol Pierre apprend l'espagnol aux 'etudiants. \eenumsentence{ remplir une dinde de farce farcir une dinde farcir un 'ecrit de citations (metaphorical)(farcir = remplir) farcir une dinde d'ail *!*Zadroxny, Wlodek and Alexis Manaster Ramer. 1995. Idioms as Constructions and Universal Grammar. I think it's an incredibly worthwhile thing to do to try to capture what's "normal" vs. what's marked cross-linguistically. That's been a deficiency in Construction Grammar, so I'm really happy you're working on the topic. But your criteria for whether something is considered an idiom seems to change in an important way in the paper. Initially (e.g. in discussion of put the cart before the horse), even a slight difference in *any* language is sufficient to decide something is idiomatic. And being "inappropriate" can count as a sufficient enough difference (would that make "How old are you?" idiomatic, since in Sp, this would be inappropriate in same contexts? Instead you would use, lit, ``how many years do you have?"). But there are languages that don't have passives or at least don't have a construction that resembles the English passive in form. You state on page 15, ``we cannot say that passives are idioms because *many* lngs have passives with the same meaning as in Eng." That's a different thing that saying that the passive has to exist in *all* lngs, which is what seems to be implied in the early discussion. Also, if it's the function that exists cross-linguistically, then it's not clear how you actually decide put the cart before the horse or even kick the bucket are idiomatic, since presumably, every lng has some way to express those concepts. It seems to me that deciding what's idiomatic in a particular lng might in fact be a different thing than deciding what's marked cross-linguistically. For example, say a lng at one point has productive passives, but through historical change, begins to lose them. At a certain point, a few instances are sure to remain, likely learned as fixed phrases, after the productive uses are lost. Are those fixed phrases idiomatic or no? Cross-linguistically, presumably not, since you want to say that passives are not idiomatic. But language internally, you might want to say yes, since they must be learned very idiosynractically. Conversely, imagine a lng which has a cross-linguistically marked way of designating possession, or topicalization or what have you. You might want to say that this is an idiomatic construction cross-linguistically, but it's not clear you would say that it's idiomatic to speakers of the lng, in the sense that it's fully productive, and probably highly motivated by other things in the lng. Nunberg, Wasow and Sag have a recent paepr in Language on idioms in which they try to list several properties associated with what are intuitively idioms. Not all properties hold of each expression they'd like to call idiomatic, but it's arguably a prototype category, defined by some of those properties. Which raises the final thing I wanted to mention which is that it's not clear to me why you want to adhere to a binary distinction: idiomatic vs non-idiomatic. Why not have degrees of markness (or conversely, motivation)? ****
Your defn of "open idioms" isn't complete (it isn't distinguished
pg 4: I'd add an example to (6) to clarify what expression you're talking about. pg 4: check the {\em on "car" and "voiture" pg 8: I think you mean "Talmy presents a catalogue ...which are *grammatically*
pg 9: Change, "John thinks that" to "I think that"? pg 16: the black humor of Der Fewer, der Better is
My book is 1995. (I appreciate all the refs to it) In case you want a written reference for the Michaelis and Lambrecht article: it is going to be published in Conceptual Structure, Discourse, and Language 1995. A. Goldberg (ed). Stanford, CA: CSLI Publications. {document tải về 302.24 Kb. Chia sẻ với bạn bè của bạn:
|