Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!uwm.edu!uakari.primate.wisc.edu!ginosko!uunet!mcsun!ukc!icdoc!syma!aarons
From: aarons@syma.sussex.ac.uk (Aaron Sloman)
Newsgroups: comp.ai
Subject: Re: Discovering What Nature Wants
Summary: Start from a theory, not verbal intuitions
Keywords: Consciousness and Intentionality.
Message-ID: <1443@syma.sussex.ac.uk>
Date: 21 Oct 89 07:16:42 GMT
References: <357@massey.ac.nz> <2376@munnari.oz.au> <2394@uceng.UC.EDU> <1090@oravax.UUCP> <2461@munnari.oz.au>
Organization: School of Cognitive & Computing Sciences, Sussex Univ. UK
Lines: 73

ok@cs.mu.oz.au (Richard O'Keefe) writes:

> Date: 18 Oct 89 05:51:58 GMT
>
> It's time I got some useful work done, and it has become evident that
> trying to play Socrates in this group isn't getting me anywhere.  Can
> we leave it at this:
>
>     some people use language in such a way that when they say
>     "Nature can form intentions" they _mean_ "some parts of Nature
>     can form intentions", and for them the truth of the former
>     necessarily follows from the truth of the latter.
>
>     other people do not use language in that way.

Philosophical discussions frequently degenerate into a succession of
non-terminating repeitions of "X means so and so" "No it doesn't".

I find most such disputes about as silly as arguing about whether
a circle really is an ellipse or not. Clearly there is the precisely
defined mathematical sense of "ellipse" and and older, less well
defined sense and clearly a circle is an ellipse in the first sense
and (probably) not in the second (though it's clearly a limiting
case).

The philosophical situation changes if you can provide a theoretical
framework within which to locate a new, precisely defined, technical
concept, say "quasi-X", AND THEN ask whether a so and so is an
instance of it. You can then make progress investigating whether a
so and so satisfies the conditions for being a quasi-X (though that
may turn up further concepts requiring rational reconstruction).

So anyone who wants to argue about whether nature (or anything else)
has intentions, should provide a theory about the architecture and
functions of systems capable of having quasi-intentions and other
quasi-mental states, and then investigate whether Nature has the
appropriate kind of architecture, etc. (One day I shall try to
publish my version of such a theory - it requires the system to be
divisible into a number of co-existing, independently variable,
causally interacting sub-states, whose causal interactions
correspond to the causal roles that seem to be characteristic of
beliefs, desires, etc. We then find there are many different
architectures with more or less different functional
differentiation, for which we have to define different concepts of
quasi-intention, quasi-belief, etc. I.e. the concepts are
mechanism(or architecture)-relative. No one architecture is the
RIGHT one. Cats and dogs and very young humans have
quasi-intentions, but their functional architectures are different
from a typical adult western human, so you can't assume their
quasi-intentional states to have all the properties that your own
do. Similarly for beliefs, desires, feelings, etc.).

If we can find an appropriate functional architecture in Nature (the
biosphere?), I think it will be very different from that of
individual humans, so if it has quasi-intentions (beliefs, hopes,
fears, etc.) those states will have very different properties from
ours. (Just as some ellipses have properties circles don't and vice
versa.)

Some philosophers will argue against this that there is an "essence"
to the concept of "intention" that we somehow learn without knowing
any theory about functional architectures of intentional systems.

My reply is that such pre-theoretical conceptual essences are often
riddled with deep ambiguity and internal incoherence, despite strong
convictions people have that they know what they mean.
"Consciousnesss" is an extreme case. (The pre-einsteinian concept
of a persisting location, independent of any framework of reference,
is another example: after all you can POINT at a position, so it
MUST have an identity.... similarly you can, you think, point at
your own consciosness.....)

Aaron