I've been attempting to put together an overview of the various theoretical motivations that have been proposed for do-support in the literature, but the topic has been frustratingly difficult to research.

What I'm looking for specifically: analyses that address the universals of syntax and morphology that give rise to do-support. Typologies of do-support, essentially, rather than analyses of do-support in particular languages. Grimshaw's Last Resorts: A Typology of Do-Support is the best example I know of.

Can anyone here point me in the right direction?

  • 2
    There are no "universals of syntax or morphology that give rise to do-support". Whether one posits it or not depends on what one's description of English is like. You'd do much better to look at data from English (which is after all the only language where "do-support" actually occurs) and figure out whether you need it in your description. Whatever that is.
    – jlawler
    Jul 17 '12 at 20:49
  • @jlawler, It's simply not true that English is the only language with do-support (in the same way that wh-words don't begin with the letters "wh" in every language). Jeremy, the "universals of syntax and morphology" differ vastly depending on the theoretical framework being employed; there is no guarantee that "do-support" as a descriptive phenomenon has a uniform analysis in a given theory. Can you say more about the background of your question? Are you unsatisfied with the OT framework of the Grimshaw article, or is there some other improvement on it you'd like to see?
    – Aaron
    Jul 22 '12 at 17:26
  • 1
    @Aaron: Using "Do-Support" and "Wh-Words" as technical terms to apply to all languages is simply arrogance. There may be processes in other languages that are similar, but they're never identical, except to those who share the Faith.
    – jlawler
    Jul 22 '12 at 17:50
  • @jlawler, Nothing is identical, except to itself. However, there is good evidence that a Romance dialect displays behavior very similar to English do-support in questions (on which see Beninca and Poletto's 2004 NLLT article, also discussed in the Grimshaw link in the question). This is also claimed of Korean, although I don't really know that evidence very well. And given the robust literature on their common properties (such as island constraints), the claim that wh-words don't form a crosslinguistically coherent class is surprising. Do you have something specific in mind?
    – Aaron
    Jul 22 '12 at 18:14
  • Just curious, how do you understand "universals"?
    – Alex B.
    Aug 3 '12 at 21:04

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.