Passive and Passive-like Constructions in Hmong

The pairing of linguistic form to discourse function has been much discussed in the literature over the last several years. Prince (1996) phrases the question thusly: what is it exactly that discourse functions are associated with? Particular form-function pairings do not appear to be cross-linguistically universal; a given linguistic – that is, syntactic – form may be associated in different languages with distinct discourse functions. Indeed, language change often occurs when language learners treat an L2 form as though it bears the function associated with a related form in L1 (see Prince 1996, 1998 for a discussion of topicalization in Yiddish and English; Snyder 1999 for a discussion of reflexivity in Tahitian and Tahitian French). Chomsky (1993) states that in a principles-and-parameters approach, the notion of grammatical construction is eliminated. However, as Prince (1996) and Goldberg (1995) note, it is difficult to explain form-function pairings without recourse to the notion of form. One approach might be to claim that either the syntax or the discourse function is basic and that the non-basic element is derived from the primitive. On the syntax-as-basic side, consider Safir’s definiteness effect; in the function-as-basic camp, Givón, Chafe, and other functional linguists, who argue that form is an iconic reflex of function. Another possibility is to argue that it is not that either form or function is basic, but rather that it is the form-function correlation which is basic. This position is taken by linguists working within Construction Grammar (Zwicky 1987; Goldberg 1992, 1995). Advocates of this position are not necessarily committed to the functionalist iconicity view, but rather they believe that linguistic form is somehow “motivated” by discourse function (Goldberg, p.c.). A final possibility is to argue that the form-function pairing is essentially arbitrary. This approach has been taken by Prince, Birner, and Ward, among others. Prince (1996) posits an abstract universal Construction-template which may be mapped onto one syntactic form in one language and to another syntactic form in another language or onto more than one syntactic form in a sinWe are grateful for many helpful comments from several people. We would especially like to thank Robin Clark, Tom McFadden, Ellen Prince, and Maribel Romero for their remarks and suggestions. We are also indebted to our Hmong informant, Reolee Lor. Any errors which remain are our own.