<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta http-equiv="Content-Type" content="text/html;charset=ISO-8859-1">
<title></title>
</head>
<body text="#000099" bgcolor="#ffffff">
<p class="MsoNormal"><span lang="EN-US"
style="font-family: "Arial Unicode MS";">Dear Colleagues,<br>
I think both Nick and Greville have addressed a very important point,
which is
also decisive for the design of descriptive grammars. If I am not
totally
wrong, usage-based analyses (hopefully) gain more and more ground in
Typology. As
a matter of fact, descriptive grammars should be much more usage-based
than they appear today. Naturally, that's not always easy to do.
Especially for
'smaller' languages without a written tradition, it is difficult do
establish a
corpus of sufficient significance. For instance, in the descriptive
grammar of
Udi (East Caucasian) I am currently writing I work with a corpus of
roughly 200.000 words (from native narratives, other types of text,
conversation etc.), a
little bit of nothing in the light of corpora usually referred to in
corpus
linguistics. Still, even such a small corpus allows retrieving
important
information on the actual significance and dynamics of grammatical
issues in
the given language, and - as a consequence - reconstructing the
linguistic
knowledge base and linguistic practice of its speakers. Often enough,
the corpus-based data go against what one has elaborated with the help
of more or less openly elicited data. So, it seems that
Bernhard's 'wish' perhaps comes a little bit too early. What we need
first are
more comprehensive descriptive grammars based on corpus analyses
(together with
the standard illustrative and analytic sections). Here, typologists
should
start to address analogous questions in order to make the corpus-based
data of
the individual grammars comparable. In a second step, cross-linguistic
surveys
could emerge based on the type of data Bernhard has called for.<br>
Best wishes,<br>
Wolfgang</span><span lang="EN-US" style=""><o:p></o:p></span></p>
-- <br>
<div class="moz-signature">
<meta http-equiv="content-type" content="text/html; charset=ISO-8859-1">
<title>schulze adr</title>
<small><font face="Arial Unicode MS">#############################<br>
<small><span style="font-weight: bold;"><big>Prof. Dr. Wolfgang Schulze
</big></span></small><br style="font-weight: bold;">
<span style="font-style: italic;">Institut für Allgemeine
und Typologische Sprachwissenschaft
</span>(IATS)<br style="font-style: italic;">
[General Linguistics and Language Typology]
<br>
Department für Kommunikation und Sprachen / F 13.14
<br>
Ludwig-Maximilians-Universität München
<br>
Geschwister-Scholl-Platz 1
<br>
D-80539 München
<br>
Tel.: ++49-(0)89-2180 2486 (secretary)
<br>
++49-(0)89-2180 5343 (office)
<br>
Fax: ++49-(0)89-2180 5345
<br>
E-mail: <a class="moz-txt-link-abbreviated"
href="mailto:W.Schulze@lrz.uni-muenchen.de">W.Schulze@lrz.uni-muenchen.de</a>
<br>
Web: <a class="moz-txt-link-freetext"
href="http://www.ats.lmu.de/index.php">http://www.ats.lmu.de/index.php</a></font></small>
</div>
</body>
</html>