Questions with ‘unbounded’ dependencies: A usage-based perspective
WH questions with long distance dependencies (henceforth LDD questions) have played an important role in the development of syntactic theory, especially in the generative framework. A critical property of such constructions is that the dependencies are ‘unbounded’, i.e. there can be (in principle) any number of intervening clauses, as demonstrated by the examples in (1).
a. What did Steve believe Chris needed?
b. What did Steve believe they thought Chris needed?
c. What did Steve believe they thought Maria imagined Chris needed?
However, real-life LDD questions are quite different from these constructed examples: they virtually always contain only one intervening clause, the main verb is almost always say or think, the auxiliary do or did, and the subject you, as in the following examples from the Manchester corpus:
a. What do you think that is?
b. What do you think you’re doing?
c. Where did you say you were going?
Moreover, if they differ from this prototype, they do so in only one respect: that is to say, they have a different subject or a different auxiliary or a different verb, but rarely all three (Dąbrowska 2004, Verhagen 2005). Dąbrowska (2004) suggests that speakers may store lexically specific templates such as WH do you think S-GAP? and WH did you say S-GAP?. These would enable them to produce and understand LDD questions by inserting new lexical material into the template and modifying it as necessary (e.g. by substituting a lexical NP for you). This paper tests this proposal using data from an acceptability judgement task.
Dąbrowska, E. (2004). Language, Mind and Brain. Some Psychological and Neurological Constraints on Theories of Grammar. Edinburgh: Edinburgh University Press.
Verhagen, A. (2005). Constructions of Intersubjectivity: Discourse, Syntax and Cognition. Oxford: Oxford University Press.