On bridge requirements in English

 

Joanna Śmiecińska

Poznań

 

 

It has been traditionally assumed within the generative framework that wh-movement is impossible or degraded across a range of structurally analysable configurations, such as complex NPs, adjuncts, subjects and wh-islands, and acceptable elsewhere. Structures meeting the configurational requirements but nevertheless having a degraded grammaticality status would be banished from the scope of syntactic research proper. Such was the case with the so called non-bridge constructions described by Erteschik in her (1973) work and largely ignored thereafter. The paper offers a reinvestigation of Erteschik’s original ideas on bridge requirements from a minimalist perspective.

Erteschik (1973) observes that the allegedly unconstrained long wh-movement across finite indicative complement clauses is subject to limitations; only semantically dominant (roughly not presupposed) complements allow for extraction while the non-dominant ones block it. The dominance of a given clause can either depend on the context or on the semantic complexity of the main predicate, with the semantically more complex verbs selecting non-dominant complements. It is this alleged correlation between semantic complexity of verbs and their extraction propensities that led Chomsky (1977) to a conclusion that the unacceptability of non-bridge constructions like (1) below is of a lexical nature and, thus, of no interest for syntax.

 

(1) "(42)* what did John complain that he had to do this evening?" Chomsky (1977:85)

 

Chomsky’s conjecture shaped the manner of investigation into long wh-movement in EST and GB, where the theoretical apparatus would impose a parallel treatment of sentences like (2a) and (2b) below:

 

(2a) ** What do you wonder how she cut?

(2b) ? What do you wonder whether she likes?

 

In both cases the intermediate CP specifiers are occupied by wh-elements. Both (2a) and (2b) are assumed to violate subjacency, the difference in grammaticality between them remaining largely unexplained.

 

On the other hand, a clear discrepancy in the acceptability of sentences like (3a) and (3b) below must be ignored, as no intervention effects are present:

 

(3a)*What do you know (that) he likes?

(3b) What do you think he likes?

 

The same is true of CNPC violations, whose severity can vary without apparent structural reasons:

 

(4a) **What did you discuss a claim that he would want?

(4b) ?What did you make a claim that he would want?

 

The paper attempts to resolve some of the above inconsistencies, among others, by re-examining the nature of bridge requirements. Offering a critical view of Erteschik’s verb typology (backed up by a survey on over 100 verbs and an experimental study), it aims at showing that the bridge properties of complement clauses of certain verbs do not correlate with these verbs’ lexical complexity, but depend on the, mainly contextual, semantic dominance of the whole clause, this being frequently manifested by the absence of that complementiser in the dependent clause. Thus, examples (3a) and (4a) are not grammatical because we are trying to extract a wh-element from a non-dominant clause. On the other hand, the (3b) and (4b) examples involve extraction out of dominant (not presupposed) complements and are considerably better in terms of acceptability.

 

It is claimed in the paper that the possibility of long wh-movement is closely related to sentence semantics, and that, in fact, the dominance condition responsible for the ungrammaticality of (3a) and (4a) accounts for the highly degraded (if not unacceptable) status of (2a), standardly attributed to intervention effects and subjacency. This claim has, among others, serious consequences for the validity of the argument that the intervention effects constitute an empirical piece of evidence for a successive cyclic derivation of long wh-questions in English. Couching bridge requirements into the theoretical vestiges of the pre-minimalist P&P must inevitably fail, as no modules exist in this framework that would constrain the non-bridge expressions.

 

Dealing with the same problem in minimalist terms is much more promising, for, being strictly derivational, this theory pays more attention to the timing and types of operations involved than to strictly configurational representational limitations. It is not surprising, then, that factive/non-dominant islands as well as the related locality problems have gained some interest in recent minimalist works, among others in Fanselow (2000), Stepanov (2001) or Uriagereka (1998). What is lacking however in these authors’ approaches is the unification of island conditions and bridge requirements, which the present paper seeks to achieve.

 

 

Selected References

 

Chomsky, N. 1977. "On wh-movement", in: Peter Culicover - Thomas Wasow - Adrian Akmajian (eds.), 71-132.

 

Chomsky, N. 1999. "Derivation by phase". MIT Working Papers in Linguistics 18. Cambridge, Mass.: MITWPL.

 

Epstein, S. et al. 1998 A derivational approach to syntactic relations. Oxford: Oxford University Press.

 

Epstein, S. and N. Hornstein (eds.). 1999. Working Minimalism. Cambridge Mass.: MIT Press.

 

Erteschik, N.S.. 1973. On the nature of island constraints. [Unpublished Ph.D. dissertation, MIT.]

 

Fanselow, G. and A. Mahajan. 2000 "Towards a minimalist theory of wh-expletives, wh-copying, and successive cyclicity", in: Uli Lutz – Gereon Müller – Arnim von Stechow (eds.), 195-230.

 

Franks, S. 2000. "A PF-insertion analysis of that", Syntaxis 3.

 

Frazier, L. and C. Clifton. 1989. "Successive cyclicity in the grammar and in the parser", Language and Cognitive Processes 4: 93-126.

 

Lutz, U., G. Müller and A. von Stechow (eds.). 2000. Wh-Scope Marking. Amsterdam/Philadelphia: John Benjamins.

 

Ross, J.R.. 1967. Constraints on variables in syntax. [Unpublished Ph.D. dissertation, Cambridge, Mass.: MIT.]

 

Stepanov, A. 2001. Cyclic domains in syntactic theory. Ph.D. dissertation, Cambridge, Mass.: MITWPL.

 

Uriagereka, J. 1998. Rhyme and Reason: An Introduction to Minimalist Syntax. Cambridge: MIT Press.

 

 

Home | Abstracts