autonomous syntax

Ellen F. Prince ellen at CENTRAL.CIS.UPENN.EDU
Fri Jan 10 16:45:37 UTC 1997


Jon Aske <jaske at abacus.bates.edu> wrote:

>That is very interesting and I would like to know more about your
>specific cases, but my experience with language contact and grammatical
>change, though it sounds similar to yours, has led me to very different
>conclusions.
>
>I have found that Basque seems to be increasing the number of clauses
>with postverbal elements, and the way it seems to be happening is that
>minor, marked (and ‘optional’) constructions which, for a variety of
>reasons, place the verb in rheme-initial position, and thus
>superficially look like the unmarked constructions of Romance languages,
>are being used more and more by those speakers which are most "under the
>influence" of a Romance language.
>
>The ‘overuse’ of these constructions results in a change in the contexts
>in which these constructions are used, i.e. in the pragmatics of those
>constructions.  In other words, the constructions are becoming
>relatively less marked than we would expect them to be.  This, of course
>is the well known phenomenon of convergence, a type of transfer that is
>rather common in language contact situations. Thus syntactic change is
>really pragmatic change in the constructions.  As I see it, these
>pragmatics are not drafted onto otherwise formal constructions as an
>afterthought, but are an intrinsic part of them.  The constructions in
>question do not make sense without reference to functional categories
>and the ordering relations are extremely iconic.

wow, THAT is very interesting to me! i will d/l your paper and read it with
interest -- but, from my understanding of what you say here, it sounds like
what i've been finding: a particular form in one language (here the basque
form with the verb in 'rheme-initial' position -- which i'm assuming is
describable syntactically, without reference to 'rheme'?) is 'matched up'
with an analogous form in another language (here canonical order in french),
with the discourse function of the 'matchee' coming to be associated with
the original form (here the LACK of discourse function of french canonical
order bumping out the df of the basque form). i would venture that it is
this lack of df that accounts for the increased frequency, not vice versa --
after all, a substantive discourse function constrains what contexts a form
may felicitously occur in; an 'unmarked' form can occur in any context and
should thus have a higher frequency. (this then would provide a different
database to the children acquiring the language in terms of frequency, which
may then result in their hypothesizing a somewhat different grammar.)

if that is a reasonable description of what you've found, then i'd think
you'd agree that form and function are NOT inextricably combined...
otherwise how could a form ever change its function -- and how could
analogous forms with different functions ever be seen as analogous by
speakers?

thanks for the post and i look forward to reading your paper!



More information about the Funknet mailing list