From Aaron Sun Mar  3 15:28:10 GMT 1996
Newsgroups: sci.psychology.consciousness

From Aaron Mon Mar  4 01:28:31 GMT 1996
Newsgroups: comp.ai.philosopy,sci.cognitive
Subject: Slides for a lecture on consciousness

I had to give a lecture on consciousness (actually "A systems
approach to consciousness") at the RSA on Monday 26th Feb.

I've made the slides available in postscript form for anyone who is
interested. (Two slides per A4 page: 20 slides, 10 pages). It's
postscript because there are diagrams.

There are two URLs, one for ftp (compressed postscript), one for WWW
access:

Version in the Cognition and Affect project FTP directory:
    ftp://ftp.cs.bham.ac.uk/pub/groups/cog_affect/Aaron.Sloman.consciousness.lecture.ps.Z

Version accessible via my Web page:
    http://www.cs.bham.ac.uk/~axs/misc/consciousness.lecture.ps

Both are in directories with other relevant papers and/or discussion
notes.

The slides slightly extend some of the points I've made in earlier
postings to sci.psychology.consciousness, regarding the notions of
reduction, functionalism, what it's like to be an X, etc. These are in
the following directory:
    http://www.cs.bham.ac.uk/~axs/misc/

The main point of the lecture was that the following two alternatives
which are often taken to exhaust all possibilities are both wrong,
namely:
EITHER
    Consciousness is something that is all or nothing
        (i.e. any thing either has consciousness or doesn't)
OR
    Consciousness is a matter of degree.

I offer a third alternative, namely that "consciousness" is a
*cluster* concept.

I suggest (it's too compressed to be an argument) that we
unwittingly use much of our mentalistic language to refer to a
rather ill-defined cluster of capabilities which may be present or
absent in different combinations, in various animals, in infants, in
people with brain damage, in normal individuals at various stages of
development, etc. I start to list some of the types of capabilities.

This sort of cluster concept is too ill-defined to be expressed as
any boolean combination of the individual capabilities. (E.g. it is
taken for granted that some combinations will always occur together,
but what should be said when they don't, e.g. in infants or in brain
damaged people, is not defined.)

Presence of absence of particular capabilities can correspond to a
discontinuity in design space. But there's no ONE major
discontinuity, merely lots and lots of important but different
discontinuities.

Talk about "degrees" misleadingly suggests a continuum, with no
important boundaries. Design space is FULL of important boundaries.
Most are still not understood.

We need to make significant progress in the analysis of such
capabilities and the architectures that make different combinations
of those capabilities possible, and what differences the presence or
absence of such capabilities can make in different circumstances.

We can can then clarify and extend or replace the concepts
("consciousness", "experience", "qualia", "awareness", "feeling",
"subjective", etc.) that are used to generate all the currently
fashionable but hopelessly ill-defined questions that generate so
much interminable discussion, with so little progress.

(This will be analogous to the way in which a new theory of the
architecture of matter led to clarification, extension, and
replacement of ordinary concepts of kinds of stuff, including the
revelation of conceptual spaces with gaps in them not filled by
current concepts: as happened with the periodic table of the
elements).

It's a pity the human brain is not better equipped to tell when its
owner doesn't know what he/she is talking about. But that's just one
of many ways in which our self-monitoring capability is limited.

Aaron
---