Newsgroups: comp.ai
Path: cantaloupe.srv.cs.cmu.edu!das-news.harvard.edu!news2.near.net!MathWorks.Com!europa.eng.gtefsd.com!howland.reston.ans.net!agate!ames!news.Hawaii.Edu!uhunix.uhcc.Hawaii.Edu!roitblat
From: roitblat@uhunix.uhcc.Hawaii.Edu (Herbert Roitblat)
Subject: Re: Chips & chimps
Message-ID: <CwD6qw.1vr@news.Hawaii.Edu>
Sender: news@news.Hawaii.Edu
Organization: University of Hawaii
X-Newsreader: TIN [version 1.2 PL2]
References: <34nh0sINN4p@ADEN.AI.CS.YALE.EDU> <Cw0EAE.7JH@cee.hw.ac.uk>
Date: Mon, 19 Sep 1994 06:44:56 GMT
Lines: 41

greg@cee.hw.ac.uk wrote:
: From article <34nh0sINN4p@ADEN.AI.CS.YALE.EDU>, by mcdermott-drew@cs.yale.edu (Drew McDermott):
: > 
: > In article <Cvszup.A89@cee.hw.ac.uk>, greg@cee.hw.ac.uk (Greg Michaelson) writes:
: > |> 
: > |> Has anyone made any computer models of the sort of linguistic behaviour
: > |> exhibited by simians which have been taught "natural" language eg
: > |> Washoe & Nim Chimpsky. If so then is the behaviour of such a model
: > |> indistinguishable from that of the corresponding chimpanzee? If so
: > |> then does the model pass a Chimping Test?
: > 


: The corpus of chimp language is relatively small and mostly chimps do not
: seem to string more than three or four symbols together. Thus, it should
: be possible to build a complete semantic net/frame/script/Montague
: grammar/agent system (or whatever your favourite form of semantics mught
: happen to be) for chimp language, or the vast majority of it.

If you knew the entire corpus ahead of making your model you could
make some very  uninteresting models of chimp or any other form of
"language" production.  The question about chimps is whether they
employ such an exhaustive enumeration or a productive grammar.
Several papers are of interest to this discussion in LANGUAGE AND
COMMUNICATION: COMPARATIVE PERSPECTIVES (edited by Roitblat, Herman, &
Nachtigall) 1993, Lawrence Erlbaum Associates.  Of particular
pertinence is the chapter by Hunt.


: I think that it would be very impressive to see a system that could
: reproduce indistinguishably chimp linguistic behaviour: it would
: draw into sharp focus the adequacy of purely behavioural standards
: for attributing particular internal dispositions.

: Greg Michaelson

--
Herbert Roitblat                    roitblat@uhunix.uhcc.hawaii.edu
Department of Psychology            roitblat@uhunix.bitnet
University of Hawaii                (808) 956-6727  (808) 956-4700 fax
2430 Campus Road,                   Honolulu, HI 96822 USA
