Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!tut.cis.ohio-state.edu!gem.mps.ohio-state.edu!ginosko!uunet!munnari.oz.au!csc!ccadfa!csadfa!rim
From: rim@csadfa.oz (Bob McKay)
Newsgroups: comp.ai
Subject: Re: What's the Chinese room problem?
Message-ID: <2245@csadfa.oz>
Date: 4 Oct 89 05:18:28 GMT
References: <2281@uceng.UC.EDU>
Organization: Dept. of Computer Science, University College, UNSW, ADFA, Canberra, Australia
Lines: 37

From article <2281@uceng.UC.EDU>, by dmocsny@uceng.UC.EDU (daniel mocsny):
> Two languages aren't even necessary. Two people who speak the "same"
> language can misunderstand each other. Two translation steps are already
> going on there---from the speaker's thoughts into a serial symbol
> string, and then from the string to the hearer's thoughts......

The conception here of speaking and listening as translation is a little 
worrying: it has somewhat of the flavour of the homunculus theories of 
sensation, with the same dangers of an infinite regress.

> ......If the hearer's
> thoughts differ substantially from the speaker's, then the translation 
> has failed.

It's almost never true that we would even wish the hearer's thoughts to be
substantially the same as the utterer's: I grant you that we usually do wish
there to be a correspondence of some sort between their thoughts, but the
problem lies in defining that correspondence. I doubt if it can be defined
without some prior understanding of the relationship between syntax and
semantics, which is where this discussion started off.

> 
> However, I think absolute translation *must* be possible in principle,
> unless we believe that the human mind has an infinite information
> content.

Is it clear that the human mind has any information content (in Shannon's sense)
at all? We don't yet understand how the human brain holds knowledge, or even if
there is any level at which it is discretised: if not, it's not at all clear 
that information theoretic measures apply,
      Cheers
	Bob McKay
-- 
Bob McKay		   Phone:	+61 62 68 8169	    fax: +61 62 68 8581
Dept. Computer Science		ACSNET,CSNET:	rim@csadfa.cs.adfa.oz
Aust. Defence Force Academy	UUCP:	...!uunet!munnari!csadfa.cs.adfa.oz!rim
Canberra ACT 2600 AUSTRALIA	ARPA:	rim%csadfa.cs.adfa.oz@uunet.uu.net