From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!utgpu!cs.utexas.edu!uunet!ogicse!news.u.washington.edu!milton.u.washington.edu!forbis Tue Apr  7 23:24:10 EDT 1992
Article 4920 of comp.ai.philosophy:
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!utgpu!cs.utexas.edu!uunet!ogicse!news.u.washington.edu!milton.u.washington.edu!forbis
>From: forbis@milton.u.washington.edu (Gary Forbis)
Newsgroups: comp.ai.philosophy
Subject: Re: A rock implements every FSA
Message-ID: <1992Apr4.192419.21684@u.washington.edu>
Date: 4 Apr 92 19:24:19 GMT
Article-I.D.: u.1992Apr4.192419.21684
References: <1992Apr2.164457.24191@oracorp.com>
Sender: news@u.washington.edu (USENET News System)
Organization: University of Washington, Seattle
Lines: 119

Some stuff deleted without notice.

In article <1992Apr2.164457.24191@oracorp.com> daryl@oracorp.com (Daryl McCullough) writes:
>forbis@milton.u.washington.edu (Gary Forbis) writes:
>>When I use the term "functionally equivalent" I mean that all of the functions
>>I consider important are equivalent.  Behavior is a function.  Qualia is a
>>function.  If some system is to be functionally equivalent to the mind it must
>>have qualia.  If a physical device is to be functionally eqivalent to the
>>brain it must support a mind.
>
>Okay, the way that you define it, functionalism is true by definition.
>I think that when people argue about functionalism, they are talking
>about whether two systems being equivalent in certain well-defined
>structural ways implies that they are also equivalent in mental properties
>such as qualia, understanding, etc. That is, the question is whether being
>equivalent as state machines implies being functionally equivalent, in
>your sense.

I think it is more productive to start with a definition of functionalism
where one can argue about the functional equivalence of two systems than
to argue over whether two "functionally eqivalent" systems have the same 
properties.  I could be wrong.

>We haven't come to a consensus about what it means for two state
>machines to be equivalent, except that I have proposed that they are
>equivalent if there is a mapping between corresponding states that
>preserves the transition relation. The questions are: is such an equivalence
>sufficient to imply equivalence of mental properties, and is such an
>equivalence necessary to imply equivalence of mental properties.

I think I was proposing that two state machines can only be said to be
equivalent within a context.  Typically we are only concerned about 
subdomains rather than the universal domain so we don't have to decide
if any mental properties supervene on a given physical system.  This
is an extention of the idea that boundaries are arbitrary.

>>As I understand behaviorism only the outside behavior is considered and not
>>the behavior of supervenient systems.  The HLT is behaviorally equivalent
>>to a human doing human things but the HLT is hollow.  It has no underlying
>>subsystems on which mind can supervene.  It is as if the brain did not
>>exist as a separate entity from the ear and the mouth.
>
>I think that is right. The question is whether it is permissable to
>treat brain, ear, eyes, and mouth as one big system, or is it necessary
>to consider the pieces as separate systems. My inclination is to think
>that boundaries are arbitrary; there is no natural dividing line inside
>the brain where you can say: "On this side of the line is merely input
>devices and output devices. On that side of the line is the real person,
>the seat of consciousness."

I think it is because the boundaries are arbitrary that for a system to
be said to be functionally equivalent all defining subsystems must also
be functionally equivalent.  I think that if we can draw a line around
any defining subsystem in one system and cannot find a funcitonally 
equivalent subsystem in a second system the two are not functionally 
equivalent.

If one chooses to define a person in non-physical terms then one needn't
be looking at physical subsystems to determine functional equivalence
except to the extent that a physical system exists upon which the 
non-physical subsystem can supervene.

Let's look at a rock implementing an particular FSA.  The FSA's state
is not defined by the physical state of the rock but rather then FSA's
state is defined to supervene on a particular physical state of the rock.
Further, the change in physical state of the rock does not define the
change in the FSA's state but rather the FSA's transitions are defined to
supervene upon particular physical changes of the rock.  In the case
where an FSA has inputs and output there must exist physical systems
external to the rock upon which they can supervene and there must exist
a mapping between inputs/outputs and these physical systems such that
when input A causes transition of the FSA from state B to C and upon finding
the rock in a physical state which maps to the FSA state B and the external
physical system wich maps to input A the physical system mapping to the
input must cause the rock to enter a physical state which maps to C.

>>Now if the HLT were broken down to interacting parts, I might begin to 
>>question its ability to enable the underlying hardware to support a mind. I
>>feel the mind exists because several physical devices are interrelated in the
>>right way.
>
>That's possible, but as I said, it seems to me that the breakdown of the
>brain into communicating pieces is arbitrary. It may be useful in describing
>how the brain works, but I don't think that the breakdown is objectively
>there.

I think it is a matter of definition.  I think a mind is defined in such a way
that different aspects map to different physical systems.

>>If someone could tell me why my experience of qualia ends at my body's surface
>>I would be grateful.  I'm not sure why I experience a desk's surface 
>>differently when I touch it with my finger and when I touch it with 
>>an imagined extention of my finger.
>
>Dennett, in _Consciousness Explained_, says a lot about this. He
>claims that if you use a pencil to feel with that eventually you come
>to think of your experience of the qualia of touch as occurring at the
>tip of the *pencil*.  If you tap or rub with a pencil, you will find
>(contrary to what you might think) that you can "feel" the textures
>and shapes of things through the pencil. He would say that you
>actually feel everything inside your head, but that the illusion of
>feeling with the tips of your fingers is, in a sense, a logical
>reconstruction. The feeling *must* have originated at the tip of the
>finger to account for the particular pattern of sensations. Of course,
>this reconstruction doesn't take place consciously; we are only aware
>of the results, and not the process.
>
>(I hope I got that approximately right.)

Because the definition of a system is arbitrary, one could just as well
define the feeling as having originated at the tip of the pencil.  I think
it is because we define ourselves as existing within our body that we say
our sensations of touch originate with our fingers.

>Daryl McCullough
>ORA Corp.
>Ithaca, NY

--gary forbis@u.washington.edu


