I am not a linguist but I want to know if there are any parsing tools in existence that can help me do the following kind of thing:

Given a sentence like the one below I would like to generate many possible grammatical sentences via mechanism like (but not necessarily restricted to) wh movement.

"The Rolling Stones were formed in 1962"

Wh movement could generate sentences like

"When were the Rolling Stones formed?" "Where were the Rolling Stones formed?"

So take a sentence and be able to produce a number of questions which need not be the proper question. E.g in this case the proper question is when not where.

I am guessing there are ways of taking a question and using wh movement to generate the form of possible answers.

In general I'm interested in producing from a sentence many other sentences that are grammatical and with syntactic variety but maintaining something of the core sense of the original sentence. I realize that's vague but it doesn't have to be perfect.


1 Answer 1


I suggest you look at the account of English wh-expressions given in Generalized Phrase Structure Grammar by Gazdar, Klein, Pullum and Sag. It's very thorough going and explicit. It takes the approach of classifying those expressions containing a wh-word that can be moved (or rather, those which correspond to declarative counterparts, since GPSG doesn't have movement transformations).

Consider for instance how the initial wh-expression in this example is built up: "In the opinion of which politicians is paying down the national debt important?". The property of being a wh-expression that can be fronted in a question starts with "which" (though the Left Branch Condition prevents moving this alone) then is inherited by containing expressions, working up step by step up the tree structure: to "which politicians", then to "of which politicians", then to the N' "opinion of which politicians" (which again can't be moved), then to the NP "the opinion of which politicians", then to the PP "in the opinion of which politicians".

This classificatory approach to a complex phenomenon in English is very powerful.

  • There's also a thorough analysis by Sag of English "filler-gap" constructions, including wh-expressions, in Language 86.3:485-545, Sept 2010. It has a complete formal statement in HPSG, plus an extensive bibliography.
    – jlawler
    Commented Aug 26, 2016 at 17:52

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.