The occurrence of be with do in (33b) is compatible with the idea that the addressee is being encouraged to take the leadership, to perform an act, and hence the context is directive. For the (c) sentences, the existence of negative polarity in (32c) corresponds to the presupposition that the negative sentence (concerning which an explanation is being sought) is true; the positive polarity form of (33c) is compatible with the notion that the sentence is a call to action.
This chapter has discussed the need for incorporating pragmatic interpretation principles into the description of certain grammatical constructions, and hence has served as an argument in favor of a model of language which does not begin with the assumption that questions of language use and linguistic interaction can only be seriously studied if we first establish a pragmatics-free account of grammar.
Three distinguishable notions of pragmatics emerge from these considerations. First, there is the pragmatics which builds on reasoning about why such-and-such a message was presented on such-and-such a situation. (Mom asks Junior what time it is to make Junior realize that it's time to go to bed.) Second, there is the pragmatics which is based on conventions about what sorts of ideas people can express in given contexts. (One doesn't inquire into a stranger's age as a matter of small talk.) Third, is the pragmatics that belongs in the description of the language's grammatical structures.
The construction grammar principle according to which pragmatic interpretation features are intimate parts of grammatical description does not relieve the analyst of the need to distinguish what is "compositional" and what is "idiomatic." For a structurally complex structure, we need to ask whether its interpretation can be analyzed as a compositional product of its constituent parts, or whether it is an instance of a complex structure with its own status as a separately functioning grammatical construction. The construction grammarian, fortunately, has reason to be pleased however things come out. If a complex structure can be seen as derivable from its component parts, then one can be satisfied that the grammatical structures one already has are sufficient to deal with the newly examined data. On the other hand, if it seems clear that there are special properties attending the whole of a complex construction, one is pleased to be working within a model in which such results are not an embarrassment.
Aronoff, M. ( 1976). Word formation in generative grammar. Cambridge, MA: MIT Press.
Bolinger, D. ( 1977). Meaning and form. London: Longman.
Cole, P. (Ed.). ( 1978). Syntax and semantics 9: Pragmatics. New York: Academic.