<HTML><HEAD>
<META http-equiv=Content-Type content="text/html; charset=utf-8">
<META content="MSHTML 6.00.6000.16945" name=GENERATOR></HEAD>
<BODY style="MARGIN: 4px 4px 1px; FONT: 10pt Tahoma">
<DIV>Dear colleagues,</DIV>
<DIV> </DIV>
<DIV>The last sentence of Gilbert Lazard motivates me to contribute: Linguists and people working on cognition, language philosophers etc. have found out so many interesting properties of language (and communication, and cognition) which are NOT obeyed by linguists themselves:</DIV>
<DIV>1. Languages are vague (in terms of exact reference either to entities or scenes): As a coseqeuence, we would have to take into consideration that also our, the linguists "languages" (Models, descriptions) are somehow vague. The Viennes school of philosophers e.g. tried to construct the "exact language" around 1900; they failed, naturally. This situation is the same for all sciences which cannot derive their work from clear axioms and absolutely formal derivations from those.</DIV>
<DIV>2. Languages - offering immense progress for cognition - also have "traps" built in. One is: An object term (= noun) seduces us to believe that there must be a referent for it. Only falling into this trap, we can - in a shortcut - assume, that there is a homogenous referent of "typology" because everyone uses the notion. The same is valid for all descriptive or model theoretical terms. e have to be aware of the fact that our linguistic models at least partially have the same (unexact) properties - like uncontrolled metaphors) which we can find (especially as "cognitive linguists") in texts of non-linguists, be they everyday texts or scientific ones. To give just one example: Is it true that a verb "governs" something? I don't think so.</DIV>
<DIV>3. Typology (not only in linguistics) is very much dependent on how much parameters we assume which are thought to distinguish the several types. It is a matter of rather simple mathematics to tell us that the number of parameters used correlates directly to the number of types we will find. A good example for that is the Greenberg typology. If you woul add, let's say two additional parameters to it, you would get a lot more of language types.<BR>4. The only thing to overcome the problems at least partially, is to work in a structuralist way (i.e. to use operational definitions and intersubjectively evaluable methods). But even here, the translation (=semantic/pragmatic adequacy) problem remains. And also the different pre-decisions/opinions/(impliicit axioms) influence descriptions or models.</DIV>
<DIV> </DIV>
<DIV>Best Regards</DIV>
<DIV> </DIV>
<DIV>Franz Dotter</DIV>
<DIV> </DIV>
<DIV>University of Klagenfurt<BR>Center for Sign Language and Deaf Communication<BR></DIV>
<DIV><BR> </DIV>
<DIV>May I add that, if the notion of typology seems to be problematic, it is because it lacks a clear theoretical basis, a question which is rarely considered ? </DIV>
<DIV><BR></DIV>
<DIV>Best.<BR>
<DIV><BR>
<DIV>
<DIV><SPAN class=Apple-style-span style="WORD-SPACING: 0px; FONT: 12px Helvetica; TEXT-TRANSFORM: none; COLOR: rgb(0,0,0); TEXT-INDENT: 0px; WHITE-SPACE: normal; LETTER-SPACING: normal; BORDER-COLLAPSE: separate; orphans: 2; widows: 2; -webkit-border-horizontal-spacing: 0px; -webkit-border-vertical-spacing: 0px; -webkit-text-decorations-in-effect: none; -webkit-text-size-adjust: auto; -webkit-text-stroke-width: 0">
<DIV>Gilbert Lazard</DIV>
<DIV>49 av. de l'Observatoire, F-75014</DIV>
<DIV><BR></DIV>
<DIV><A href="mailto:gilzard@orange.fr">gilzard@orange.fr</A></DIV>
<DIV><BR></DIV>
<DIV></DIV></SPAN></DIV></DIV></DIV></DIV></BODY></HTML>