🚀 go-pugleaf

jwi@cbnewsj.ATT.COM (Jim Winer @ AT&T, Middletown, NJ) writes:

>Lyle Seaman writes:
>| ...
>| So I think the debate over "do machines think" is really just a debate 
>| over how we (fluent English speakers) use the term _think_ and not a 
>| discussion of what machines can do.  
>| 
>| Now, to confuse the issue.   Do dogs love their humans?  Does this question
>| raise the same problems as the question of whether machines think?

>Language is exactly the point. The French have 47 words for love, the Eskimos
>have 47 words for snow, and the English language speakers have one word for
>think -- obviously, thinking is not a large part of our culture. As a result,
>the only pragmatic way to determine whether a machine thinks is in terms
>of its function:
>	An airplane flies.
>	A submarine swims.
>	A machine that functions at a task that
>		normally requires thinking, thinks.
>	ergo,
>	A dog that behaves like it loves its human, does love its human.

Eskimos have 47 words for snow (I believe you) because they distinguish
between different types of snow.  Thus the English coverage of these topics
by a single word does not mean that these distinctions are unreal.

Therefore, first, you can't prove ANYTHING about reality by analysing
language.  This is a well known point.  See Popper.  Second, the above
points are simply thickly-applied behaviourism.  If you believe
behaviourism, you can keep on thinking of stupid examples like the above,
and keep on getting applauded by the other behaviourists out there.
If, however, you see that it is a real question whether other beings have an
actual consciousness like the one (I presume) you have in your head, then you
will realise that you won't decide the question by silly word analogies.

Regards,

Ron House.   (s64421@zeus.irc.usq.oz)
(By post: Info Tech, U.C.S.Q. Toowoomba. Australia.  4350)

From:	BUNYIP::root "System PRIVILEGED Account"  5-MAR-1990 12:52:19.69
To:	uqvax::newsmgr 
CC:	
Subj: