Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!tut.cis.ohio-state.edu!ucbvax!decwrl!sun!pitstop!sundc!seismo!uunet!mcvax!ukc!strath-cs!glasgow!gilbert From: gilbert@cs.glasgow.ac.uk (Gilbert Cockton) Newsgroups: comp.ai Subject: Re: Question on Chinese Room Argument Message-ID: <2464@crete.cs.glasgow.ac.uk> Date: 24 Feb 89 10:39:49 GMT References: <3305@uhccux.uhcc.hawaii.edu> Reply-To: gilbert@cs.glasgow.ac.uk (Gilbert Cockton) Organization: Comp Sci, Glasgow Univ, Scotland Lines: 33 In article <3305@uhccux.uhcc.hawaii.edu> lee@uhccux.uhcc.hawaii.edu (Greg Lee) writes: >From article , by harnad@elbereth.rutgers.edu (Stevan Harnad): >used, I have a limited sympathy with the argument. When we >know the mechanism behind the behavior, we don't usually speak >of 'understanding'. Hence the linguistic fact in English (and French, German, Chinese etc? - comments please) that any mechanical process cannot possess understanding. It is a central feature of "understanding" that mechanical processes are not involved. Thus for computer-based reasoning, one must choose another word or run the risk of been seen as ignorant or, more likely, disingenuous. >And we do develop new usages in the course of a conversation, as here when >some come to be willing to attribute understanding to the Chinese room. Some new uses are just plain deviant or mistaken and die with the conversation. Take the software 'toolkit' which contains a) unconfigured components and not tools b) some parts of a system, not all of them (a kit is complete). For first-rate linguists, I presume that all new meanings are valid and do not represent some form of verbal dyslexia on the part of anyone who uses them uncritically? The same is true of the new and creative meanings developed within the AI subculture. If a computer system has understanding, then where does it lie? Mine's in that still small voice within - why do AI types have to disown their's? Why insist on being 'scientific' when it's quite clear that you can't be on these issues? -- Gilbert Cockton, Department of Computing Science, The University, Glasgow gilbert@uk.ac.glasgow.cs !ukc!glasgow!gilbert