Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!uunet!munnari.oz.au!uhccux!lee
From: lee@uhccux.uhcc.hawaii.edu (Greg Lee)
Newsgroups: comp.ai
Subject: Re: Language Acquisition (was: Machines *CAN* think!)
Message-ID: <7263@uhccux.uhcc.hawaii.edu>
Date: 12 Apr 90 19:53:23 GMT
References: <66tJ025u9b3u01@amdahl.uts.amdahl.com>
Organization: University of Hawaii
Lines: 17

From article <66tJ025u9b3u01@amdahl.uts.amdahl.com>, by kp@uts.amdahl.com (Ken Presting):

>But for the investigator to be able to put his questions to the native
>speakers, their language must be learned, which requires a context
>dependent radical interpretation.

Except on a very literal level, there is essentially no truth to
this.  The native speakers may know or be taught the investigator's
language, or the investigator may be a native speaker.  As a practical
matter, if one's interest is in a language whose conversational
aspects are very different from one's own, teaching native speakers
linguistics has a lot to recommend it -- then they can devise the
theories and put the questions.  In any case, as a demonstration
of the crucial importance of non-linguistic context to language
learning, understanding, or analysis, this just doesn't make it.

			Greg, lee@uhccux.uhcc.hawaii.edu