Is grammar derivable/motivated/explainable?

[Wolfgang Schulze]

Martin Haspelmath schrieb:

> The real problem with the passage cited by Oesten Dahl from Christer
> Platzack's textbook is the passage immediately before, highlighted by me
> in the following:
>
> "There are other current theories of grammar that DO NOT ASSUME an
> independent internal grammar, such as functional grammar, according to
> which grammar is derivable from language use..."
>
> But of course, most functionalists do agree that there is an internal
> grammar, and that this is precisely what we need to explain.

Both Christer's and Martin's wordings demonstrate a basic problem of
"functional" approaches, namely the use of a terminology that clearly
belongs to another scientific paradigm. The term "internal grammar" is
closely related to the MIT-Orthodoxy, but adopted by "functionalists" in
a rather uncritical way. In fact the term seems to be rather
desemantisized in the sense Martin uses it; it probably refers to some
kind of "system internal structuring" or so. But such a kind of
adaptation is dangerous and misleading: It also presupposes that
"functionalists" have a well-formulated theory of what "internal
grammar" is with respect to the frame work of "functionalism". If you
(Martin) say that "there is an internal grammar" do you mean that there
ALSO is an "external grammar"? Or is grammar always (per definitionem)
"internal"? To which kind of "internality" do you refer? A cybernetic
(self-determinated and self-regulating) system? If yes: Is this system
"independent" (organ-like) or embedded in some extra-linguistic
cognitive reality? Remember that the MIT-Orthodoxy also is based on some
kind of "derivationism": LAD (if not a functional organ) is related to
what once was called LOT (Language of Thought) or to other kinds of
cognitive prerogatives.
        I think the basic problem is that functional approaches often argue on
the basis of an unexplained multicausal derivationism which allows to
introduce "extralinguistic" motivations for given linguistic structures
ad hoc and rather randomly. The assumption of a "non-derived" (internal)
layer reminds me of the way Neo-Grammarians approached their data: What
we are looking for is the "system" (and "this is precisely what we need
to explain"), the search for external motivation is left to
Psycholinguistics etc. to which functionalism refers in case a
system-"internal" motivation cannot be established.
        Obviously the question remains what "internal" means. If you use it in
the MIT way then the term seems rather incompatible with "functionalism"
in its traditional sense. If you claim that "internal" means some kind
of reference to system-internal structures in the Saussurian sense then
this usage may be compatible with businees-a-usual functionalism (or
Neo-Grammar in Typology, if you want), but is is completely
a-theoretical and fails to motivate whatever is "internal grammar".
        Though I call myself a "functionalist" (or "derivationalist"), I cannot
subscribe to Martin's claim that "most functionalists do agree that
there is an internal grammar". Before doing so, we have to explain (in
terms of derivationalism, not in terms of the MIT-Orthodoxy) what is the
"derivational" (or functional) basis for this so-called internal grammar
[I tried to get closer to this in the lay-out of a "Grammar of Scenes
and Scenarios" (see my home page for the "Person, Klasse, Kongruenz"
volume (vol.1)[ http://www.lrz-muenchen.de/~wschulze/pkk_1abs.htm ]
which tells you a little bit more about this)]. Without doing so terms
like "internal grammar" are nothing but an a-theoretical but popular
adaptation of main stream terminology or labels for a linguistic program
that tacitly denies its own "functionalism".
Wolfgang

--
_____________________________________________________
| Prof. Dr. Wolfgang Schulze
| Institut fuer Allgemeine und Indogermanische Sprachwissenschaft
| Ludwig-Maximilians-Universitaet Muenchen
| Geschwister-Scholl-Platz 1
| D-80539 Muenchen
| Tel:  +89-21802486 (secr.)
|       +89-21802485 (office)
| Email: W.Schulze at mail.lrz-muenchen.de
|
http://www.lrz-muenchen.de/~wschulze/
_____________________________________________________
-------------- next part --------------
A non-text attachment was scrubbed...
Name: W.Schulze.vcf
Type: text/x-vcard
Size: 192 bytes
Desc: Visitenkarte für Wolfgang Schulze
URL: <http://listserv.linguistlist.org/pipermail/funknet/attachments/19990309/04149a41/attachment.vcf>