Newsgroups: comp.ai.philosophy
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!fas-news.harvard.edu!newspump.wustl.edu!crcnis3.unl.edu!news.mid.net!news.mci.net!news.cic.net!newsxfer.itd.umich.edu!gatech!swrinde!cs.utexas.edu!news.unt.edu!hermes.oc.com!internet.spss.com!markrose
From: markrose@spss.com (Mark Rosenfelder)
Subject: Re: Strong AI and consciousness
Message-ID: <D0Ct7r.D4r@spss.com>
Sender: news@spss.com
Organization: SPSS Inc
References: <D000q2.8pn@spss.com> <D01MoH.Ir9@cogsci.ed.ac.uk> <D03Gtp.1Ew@spss.com> <D075MI.AGA@cogsci.ed.ac.uk>
Date: Mon, 5 Dec 1994 20:07:02 GMT
Lines: 59

In article <D075MI.AGA@cogsci.ed.ac.uk>,
Jeff Dalton <jeff@aiai.ed.ac.uk> wrote:
>In article <D03Gtp.1Ew@spss.com> markrose@spss.com (Mark Rosenfelder) writes:
>>>>Jeff Dalton <jeff@aiai.ed.ac.uk> wrote:
>>>>>Suppose people counted "fool's gold" as gold in the past.  Surely it's
>>>>>possible to meaningfully say they were wrong.

>>You can't just say "it's gold if it's *like* that stuff there"; you have
>>to have some notion, even if you can't express it propositionally, of 
>>*how much* likeness is required. 
>
>Yes, and they could have had a notion that was sufficient for them
>to be wrong about fool's gold, I would say.  Exactly how this would
>work is something we might discuss.

OK, how about saying how it would work?  Since we agree that they would not
be wrong to call fool's gold "gold" if both substances were indistinguishable
by a set of properties that comprise the definition of the term, this 
"likeness" idea seems to be the only proposal on the table to make sense of 
your original assertion (quoted above) that they could be "wrong" about fool's 
gold.  So I'd like to see some details (or a better example if you've got one).

>Now, presumably you'd accept that a miner, 49er, might have thought
>some fool's gold was gold.  He may have lacked much in the way of
>relevant detailed knowledge.  Nonetheless, by "gold" he meant
>that valuable metal that he's see in rings, coins, etc; he may
>also have been able to list certain properties explicitly.
>He thought a bit of fools' gold was gold (the same stuff he's
>seen in rings, etc, that valuable metal), and he was wrong.

Why was he wrong?  I don't say he wasn't; I just want to see how your
criteria for a wrong attribution work.

>We can ask what continuity in the meaning of "gold" is required
>to extend the possibility of such mistakes further into the past,
>and we can consider whether that continuity obtained or not.
>That is my view, at least.

Some changes of meaning are gradual, but this one was surely something of
a revolution.  Nowadays "gold" is defined in terms of the theory of elements
(as Feynman put it, physicists have this funny way of counting; "gold" 
is their way of saying 79).  Before that theory existed it had to be
defined in a different way, and I can't see how a gradual evolution 
between the ancient and modern meanings could have happened.  (There was
continuity in terms of the prototypical *referent*, of course.)

>Now, I think "of the same kind as this <object>" is an objective
>property for some kinds.  For instance, for so-called natural kinds
>(as opposed to, say, nominal kinds).  So for gold, it's an objective
>property, since gold is a natural kind.  Many would disagree with
>this view.  I'm not firmly comitted to it myself.  It does have some
>problems, which might be discussed. 

That an atom is gold or not is about as objective a property as we can
imagine.  That a lump of metal is gold or not starts to get subjective.
For instance, is my wedding ring gold?  It's very likely an alloy; and
even if the percentage of other elements is small, I can't see that 
the class of rocks with a particular range of percentages of gold
is a natural kind.  
