Speakers often answer a question with what appears to be merely a phrase, a fragment of a sentence, rather than with a full sentence. Merchant (2004) offers an analysis of fragment answers in which the new information/answer is fronted to a clause-peripheral position and the remainder of the sentence is not pronounced. Two written acceptability judgment experiments are reported that tested predictions of this analysis. The first, in English, tested the prediction that clausal fragment answers should only be fully acceptable (...) when the clausal answer is introduced by an overt complementizer (What did May deny? That Josh left.). This is because clauses may front only when an overt complementizer is present (That Josh left, May denied, but not *Josh left, May denied). The second study was conducted in German, a language that does not permit prepositions to be stranded, left behind, when a noun phrase is moved in overt syntactic structures such as questions or topicalizations. Consequently, when the object of a preposition is questioned, only a prepositional phrase fragment answer, not a noun phrase fragment answer, is predicted to be fully acceptable. Both predictions were confirmed. The results support the claim that syntactic structure is present in unpronounced constituents, and tells against theories of syntax that eschew such structures. (shrink)
VP ellipsis generally requires a syntactically matching antecedent. However, many documented examples exist where the antecedent is not appropriate. Kehler, 533–575. 2002, Coherence, Reference and the Theory of Grammer, CSLI Publications. Stanford.) proposed an elegant theory which predicts a syntactic antecedent for an elided VP is required only for a certain discourse coherence relation, not for cause-effect relations. Most of the data Kehler used to motivate his theory come from corpus studies and thus do not consist of true minimal pairs. (...) We report five experiments testing predictions of the coherence theory, using standard minimal pair materials. The results raise questions about the empirical basis for coherence theory because a syntactically-matching antecedent is preferred for all coherence relations, not just resemblance relations. Further, strict identity readings, which should not be available when a syntactic antecedent is required, are influenced by parallelism per se, holding the discourse coherence relation constant. This draws into question the causal role of coherence relations in processing VP ellipsis. (shrink)
Carminati (2002) shows that the existence of both phonetically full and phonetically null pronouns (pro) in Italian reﬂects a division of labor with respect to anaphora resolution. Pro prefers to link to prominent antecedents more than its phonetically overt counterpart does (where prominence is determined by syntactic position in intrasentential anaphora cases).
Responds to comments by G. Kempen (see record 1996-00289-001) regarding L. Frazier's (see record 1995-31821-001) comments regarding G. Kempen's (see record 1995-31826-001) comments on Frazier et al's (see record 1994-32229-001) article on processing discontinuous words. There are serious problems with Kempen's account of the data of Frazier et al. These problems involve the principle of uncertainty invoked by Kempen. It is unclear that all control items in the original Frazier et al study were open to a complex verb analysis. Kempen's (...) alternative account of the results is not only incomplete, but it is inconsistent with the English data and with half of the Dutch data, in addition to relying on a principle of parsing that is otherwise unmotivated. (PsycINFO Database Record (c) 2000 APA, all rights reserved) (unassigned). (shrink)