From Aaron Sloman Sat Apr  5 22:11:21 BST 1997
Newsgroups: comp.ai.philosophy
References: <3341B635.7704@netcom.com> <5hufbl$465@usenet.srv.cis.pitt.edu> <5huoe3$pr@ux.cs.niu.edu>
From: N.o.SpamSloman@cs.bham.ac.uk (Aaron Sloman See end for reply address)
Subject: Re: More of Aaron Sloman's words on Searle's argument

I haven't had time recently to read stuff in this group, but when
scanning subject lines in an idle moment I saw this one and couldn't
resist the narcissistic curiosity it provoked.

rickert@cs.niu.edu (Neil Rickert) writes:
> Date: 2 Apr 1997 17:00:19 -0600
> Organization: Northern Illinois University

Quoting
> andersw+@pitt.edu (Anders N Weinstein):
> >But that's exactly what I meant: the lunatic idea that a disembodied
> >*computer* all by itself is the right sort of thing to have
> >intelligence

There's nothing lunatic about it, and you should resist the
temptation to use such rhetorical devices as substitutes for
argument.

As I've pointed out more than once in the past in this group and
elsewhere, there is no logical reason why it should be impossible
for a disembodied mathematician to be passionately concerned with
investigations in pure number theory.

Its time could be fully occupied exploring conjectured theorems,
trying to produce counter examples to conjectured non-theorems,
searching for more general, more elegant, and deeper theoretical
frameworks, etc. Highly motivated mathematicians will be able to
give many more examples.

Such a disembodied mathematician could be delighted with successful
proofs, disappointed or even depressed on discovering fallacies in
its arguments, hopeful about lines of investigation that look
promising, surprised at unexpected relationships, frustrated at
repeated failures, and so on and so on.

None of this logically requires eyes, ears, legs, relationships with
other people, etc, etc.

I remember long ago when I was a mathematics student the
days I spent lying flat on my back with my eyes closed trying to
find a proof of a theorem I had read about, namely that if A and B
are two sets and there's a one to one correspondence between A and a
subset of B and a one to one correspondence between B and a subset
of A then there's a one to one correspondence between B and A. (The
Cantor/Schroeder/Bernstein theorem??? It looks trivial but isn't.)

The fact that I had to occasionally get up and eat, etc. was just a
distraction. I fear that readers who have never had such experiences
will not have any idea what I am talking about. Sounds as if Anders
is one of them.

Of course, in my case there was a history of being embodied that led
to my exposure to the problem and the language and concepts and
techniques required for its formulation and solution. But that's
just a contingent fact about me.

There's no logical reason why a disembodied fanatical mathematical
researcher should not be created all ready to go: as long as it has
the right architecture to support the process. (Whether any
architecture implemented in a digital computer could suffice, or
some Penrose-type quantum gravity machine is required is a separate
question which I'll ignore for now.)

[AW]
> > -- an idea one finds in the idea of the Turing Test,

The possibility of a disembodied intelligence has nothing specific
to do with the Turing test. I do not regard the turing test as a
test or criterion of intelligence, and neither did Turing,
incidentally.

He thought "intelligence" was too ill defined to be worth arguing
about. He merely put forward his test as a fairly modest
technological challenge which he thought could be met before the end
of the century and he was right. (Other versions of the test are
less modest and harder to achieve.)

But nothing much of interest followed from Turing's claim, except
that he put forward and refuted arguments attributed to some of his
contemporaries who claimed such things were impossible, usually
because of their own ignorance or lack of imagination).

The quote from Anders continues...
> > .... -- an idea one finds in the idea of the Turing Test, say,
> >in which not the slightest difficulty is detected in the idea of an
> >always blind, always entirely disembodied machine discoursing calmly
> >about colors and the meaning of "a summer's day" -- stems from deeply
> >held prejudices which certainly have philosophical roots.

Ignoring the fact that that looks like rhetoric substituting for
argument, I take it you have never read Turing's original article?

In any case there are specific reasons why a system (like my
disembodied mathematician) which is causally detached from a
particular environment is not capable of referring to things in that
environment. E.g. there's the simple (and trivial?) reason that
without causal linkage there can be no identification of one object
rather than another exactly like it (i.e. the usual symmetric
universe or "twin earth" argument).

In other words a disconnected engine cannot achieve semantic
reference to spatio-temporal particulars except via the accident of
uniqueness.

Whether that should be called "successful reference" is a moot
point. (I suspect that's all WE have as a basis for some our
thoughts about remote particulars, but that's not relevant now.)

Anyhow that's no argument against the possibility of a disembodied
intelligence inventing the idea of a *type* of universe with which
it has had no acquaintance or causal linkage and having all sorts of
thoughts about hypothetical universes of that type.

You can then quibble about the applicability of phrases from *our*
language, e.g. "summer's day" to express the thoughts of *that* sort
of disembodied individual (or even an individual from another part
of the universe who has never had any contact with our culture) but
nothing follows about the impossibility of such an agent having
structurally similar thoughts.

Moral: don't be hide-bound by normal ways of thinking about thinking
and conservative linguistic philosophy. Expanding one's mind can be
exhilarating and lead to fascinating (but disciplined) explorations
of possibilities which would otherwise be pronounced incoherent (or
"lunatic") on the basis of nothing more than philosophical
prejudice.

This message will be stored, alongside others, in
    http://www.cs.bham.ac.uk/~axs/misc/disembodied.minds

Cheers
Aaron

From Aaron Sloman Tue Apr  8 08:36:22 BST 1997
Newsgroups: comp.ai.philosophy
References: <3348C8E8.796F@netcom.com> <5ic46q$jop@aurora.cs.athabascau.ca> <5icgad$f4e@ux.cs.niu.edu>
From: N.o.SpamSloman@cs.bham.ac.uk (Aaron Sloman See below for reply address)
Subject: Re: More of Aaron Sloman's words on Searle's argument

rickert@cs.niu.edu (Neil Rickert) quotes an article, possibly from
burt@cs.athabascau.ca (Burt Voorhees), that has not yet reached this
site, apprently responding to my comment:

[AS]
> >>> >As I've pointed out more than once in the past in this group and
> >>> >elsewhere, there is no logical reason why it should be impossible
> >>> >for a disembodied mathematician to be passionately concerned with
> >>> >investigations in pure number theory.
> >>>
> >>> >None of this logically requires eyes, ears, legs, relationships with
> >>> >other people, etc, etc.

[BV]
> >>> I'm not at all sure that this would be the case.  Where would this
> >>> disembodied mathematician have gotten the various internal conceptual
> >>> structures involved in thinking about mathematics, if not originally
> >>> from sensory and kinesthetic inputs?

This is the old old ``concept empiricist'' thesis that no concepts
are possible except those that come via some sort of abstraction
from sensory experience.

It is blindly accepted as a truism by large numbers of people,
especially first year philosophy students and non-philosophers, but
also many professional philosophers, though there is not a single
good argument for it. It appears to be accepted by people who

    (a) assume (usually without any argument) that ALL concepts have
        to be created from something or other

    (b) have powers of imagination and design that are too limited
        to enable them to think about any other way concepts might
        exist.

Concept empiricism was refuted very clearly in Immanuel Kant's
Critique of Pure Reason many years ago. His argument was that in
order to be any kind of experience you need to have concepts.
Therefore any individual needs concepts that are prior to all
its experience.

If I remember correctly, he claimed that concepts of space, time,
geometrical relationships and number had to be innate in this sense:
just the sort of stuff a disembodied mathematician might enjoy
working with. (I don't think he was right in detail: he did not know
enough to design a mind that could bootstrap itself in the way human
infants do. I doubt that he knew much about developmental psychology
either).

Concept empiricism is also refuted by the fact (not acknowledged by
philosophers of science who think everything can be defined in terms
of pointer readings and measurements) that sciences often make
massive advances by introducing new concepts that cannot be defined
in terms of anything that was previously understood but which play a
powerful organising role thereafter. (E.g. Newton's concepts of
"mass" and "force", and the concept of "gene" in biology.)

In any case whether Kant was right or wrong people like Burt seem to
have trouble imagining the possibility of creating a robot that is
ready to run, fully armed with a collection of concepts, motives,
beliefs, and an architecture for processing these things, without
having any prior experience of its own. Limited imaginative powers
cripples much philosophical thinking: people think everything has to
be like what they can imagine.

The issue is discussed at length by Roger Young in a paper he
presented in 1994, "The Mentality of Robots", in Proceedings of the
Aristotelian Society. I recommend it to anyone whose imagination
needs a bit of help on this issue.

Incidentally the particular sort of disembodied mathematician I was
talking about, when responded to the unsubstantiated claim by Anders
Weinstein that it was a "lunatic idea", would be better described as
"disconnected" rather than disembodied.

I.e. it might have a physical implementation in some kind of
information processing engine, but without normal sensory
apparatus linking it to the environment.

Whether a really disembodied information processing system is
possible (e.g. minds of angels processing mental stuff not
implemented in any physical stuff) is a totally different question
on which I did not intend to express a view, partly because the
science we call physics (like our conception of matter) changes over
time, so that the notion of what is physical (or material) is not a
very clear concept.

Aaron