re interrogative verbs + interrogative relators + indefiniteordinals

David Gil gil at EVA.MPG.DE
Wed Mar 28 20:48:16 UTC 2001

In response to Matthew Dryer:

> >>But let me try and salvage something of the intuition behind the
> >>conjecture.  In response to a question such as "Which book did you
> >>read?"  you could answer "This book" (thereby seemingly replacing
> >>"which" with "this"), but also "War and Peace", or "The book that John
> >>gave me yesterday", and so forth.
> >>
> >>So perhaps what licenses "which" in English is that the set of possible
> >>responses forms an open class.
> This response of David's is a reasonable one and one that I anticipated.
> >>And likewise for "how many", even in
> >>languages that have a finite and small set of lexical quantifiers.
> But I DON'T think that this case is analogous to the preceding one.  While
> any expression denoting a book is an appropriate response to a "which book"
> question, the only appropriate responses to "how many books" is a numeric
> expression (or a numeric expression followed by the word for "books"),
> which in a language with a finite number of numeric expressions will be a
> finite number of responses.  (It is true that in English a possible
> response might be something like "more than you could possibly imagine" and
> there is no limit to the set of responses of this sort, but I'm not sure
> how common this is, and it's not clear that there are not always an
> unlimited number of possible answers of this sort to any conceivable
> question.)

Seems to me that they ARE analogous, for the reason Matthew gives in the
above parenthetical comment, or, alternatively, responses such as "as
many as X", where "X" is any description of an amount.  Whether such
responses would actually be common in a languages with a closed set of
numerals is probably not relevant, and, if anything, would be a
reflection of general cognition rather than grammar.

I believe the crucial thing that languages CAN'T do (and I suspect that
this was behind Claude Hagege's original query) is have a WH word whose
appropriate responses are restricted to members of a closed grammatical
class, such as prepositions (as opposed to spatial or temporal
expressions which may belong to open classes of expressions),
tense/aspect markers (as opposed to time or other such adverbial
expressions), demonstratives, numeral classifiers, genders, case
markers, and so forth.  (Some of these strike me as potentially quite
useful things to have, but nevertheless, to the best of my knowledge,
they don't occur.)


David Gil

Department of Linguistics
Max Planck Institute for Evolutionary Anthropology
Inselstrasse 22, D-04103 Leipzig, Germany

Telephone: 49-341-9952321
Fax: 49-341-9952119
Email: gil at

More information about the Lingtyp mailing list