Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site spar.UUCP
Path: utzoo!watmath!clyde!bonnie!akgua!whuxlm!harpo!decvax!decwrl!spar!ellis
From: ellis@spar.UUCP (Michael Ellis)
Newsgroups: net.philosophy
Subject: Re: Sc--nce Attack (self-awareness)
Message-ID: <516@spar.UUCP>
Date: Fri, 13-Sep-85 07:46:50 EDT
Article-I.D.: spar.516
Posted: Fri Sep 13 07:46:50 1985
Date-Received: Sat, 14-Sep-85 17:07:50 EDT
References: <45200016@hpfcms.UUCP> <1605@pyuxd.UUCP> <491@spar.UUCP> <1635@pyuxd.UUCP> <1128@ames.UUCP> <1667@pyuxd.UUCP>
Reply-To: ellis@spar.UUCP (Michael Ellis)
Organization: Schlumberger Palo Alto Research, CA
Lines: 64

>> 	I think you missed the catch in the question. In order to equate
>> conscious awareness to any physical mechanism(s), you'd have to be able
>> to distinguish between actual self awareness and a perfect counterfeit
>> of it. Suppose I build a computer which can act *exactly* as though it
>> is self-aware. Suppose, further, that I give you complete access to the
>> machine's internals, and complete documentation, plus a staff of experts
>> to help you understand its operation. You would soon understand quite
>> well how the machine managed to act as if it were self-aware, and might
>> even jump to the conclusion that the machine was *not* self-aware, since
>> you could follow the chain of events from stimulus to response in perfect
>> detail, and no procedure called "self-awareness" was in the chain anywhere.
>> [Kenn Barry]
>
>Why would I do that? Can you define the way in which this is a "counterfeit"
>and not the "real thing". [Rich Rosen]

    Kenn made no suppositions whatsoever about the machine's awareness.
    Rather, he is asking "How can we tell if a machine that perfectly
    simulates human awareness actually possesses awareness?"

>>...
>Or maybe we are just "simulating it perfectly" ourselves.  Can you define the
>difference?
----
>> For all I really *know*, I am the only self-aware creature in the universe;
>> all you other folks are just mindless automatons.
>How did you know?

    Self-awareness is direct experience of self, just as pain is direct
    experience of bodily or psychic harm. 

    Self-aware entities have feelings, incorrigible knowledge of one's own
    private internal world. Such knowledge differs from objective knowledge
    in several ways -- it is directly experienced and totally certain.  In
    the same way that imaginary pain is, nonetheless, pain, so imaginary
    awareness is still awareness. 

    I would gladly `torture' a car, for example, by running the engine far
    beyond its proper operating level until broke down, if that were somehow
    to my benefit -- say, if somebody paid me.  But to torture a cat would
    result in intense guilt and nightmares for the rest of my life --
    because I believe that cats are aware beings and `really' feel pain,
    unlike cars. 
    
    Anthropomorphism, you say? No doubt about it. 

    Subjective delusion or not, the question has been put to you, Rich:

	Can you describe what objective evidence there might be for
	conscious awareness?

    I am convinced that you cannot.

>> 	So, how could you *ever* determine, scientifically, whether my
>> machine was truly self-aware, or only simulated self-awareness?
>Again, aren't you just creating a bogus differential between "simulated"
>self awareness and "real" selfawareness?  What is the difference?

    An entity with real self-awareness knows of its own existence.

    An entity with simulated self-awareness does not know of its own
    existence; however, the aware beings around it think it is self-aware.

-michael