Henry Stapp's posting about qualia...

Date: Thu, 6 Jul 1995 11:21:06 -0700 (PDT)
From: STAPP@theorm.lbl.gov
To: psyche-d@rfmh.org
CC: STAPP@theorm.lbl.gov
Subject: science and ontology II

Science and Ontology II: Comments on some recent postings.

Suppose we have finally come up with a `satisfactory understanding' of
human consciousness, in the sense that we have found the brain correlate
of consciousness. I imagine that we then will have found that there is
some large (relative to normal) surge of energy in some particular mode
of activity, spread out over various parts of the brain, and that we will
also then understand how this surge of brain activity will: 1) cause a
`facilitation' of this mode of activity that will allow it to be
re-activated at later times by relatively small triggers; 2) initiate
some appropriate follow-up brain/body activity; and 3) contain the
structural or logical information that is present in the associated
conscious psychological surge. Let us suppose that the various lines of
evidence mentioned by  McKee have been followed through, and all converge
nicely to this explanation of human consciousness. And let us suppose
that we will see, in the overall organization of brain activity that
underlies this high-level brain correlate of consciousness, a tower of
functionally related levels of the kind described by Cooper. And suppose
this identification of the `brain correlate' accounts perfectly for when
brain processing generates consciousness and when it doesn't (cf. Baars),
and also what the quality or content of the associated conscious thought
will be.

The achievement of such an understanding might quite justifiably be
called the proper conclusion of the science of consciousness. What
further developments of this subject could be needed, or are possible,
within science?

Yet where does this leave the Chalmers-Klein classification? Is opinion
on those matters a part of science? How could one know, even then,
whether what was important was the logic of the computation per se or
its implementation in the biological substrate? Would the substitution
of silicon for carbon make a difference? Would the substitution of a
computer with copper wires and mechanical switches make a difference?
Is there any empirical evidence one way or another? Is there any
*possibility* of empirical evidence one way or another? Is not the correct
scientific posture simply that there is and can be no empirical evidence
on those matters, and that whereof there is no evidence, or possibility
thereof,  scientists should remain silent. For how can one pass from
evidence based on behavior (in the broad sense that includes measurable
behavior of neurons and conglomerates of them and other cells, up to
and including verbal reports) to conclusions about whether there is
actually an inner light, or inner darkness? The basic premise of the
science of consciousness is that each normal alert human being has the
inner light that most of us seem, on the basis of reports of others and
of our own personal experience, to possess. To extrapolate beyond
ourselves some further *assumption* is needed.

An assumption that most scientist would find congenial is that carbon
is important only because of what it *can do*! There is nothing magical
about carbon per se: if some other materials could be formed into a
machine that had all the real time functional organization,
capability, and activity of a human brain then I think most scientists
(and philosophers?) would be willing to grant as `plausible' that
this psuedo-brain would also generate consciousness.

But this leads to the basic question: What is the principle by which
functional organization, capability, and activity generates a reality
that seems absent from the physical matter in which those functional
properties are instantiated? How does *function* (in a broad sense)
generate inner light?

Of course, in science we have to start somewhere: there are always
assumptions that are justified in the end by the power of the
resulting whole formalism to organize our thinking about our
observations in a useful way. In the field of consciousness
research the basic assumption apparently needs to be that `function'
generates a kind of beingness, namely inner light, which appears to
be absent from the underlying physical matter.

Cooper's suggestion that the `abstraction', the `virtual machine',
*comes into being* seems to be an attempt to systematize this
general idea: to make it more precise and formalized. Since he seems
to grant that the virtual machines are instantiated in a physical
matter that obeys the deterministic laws of classical physics
(PSYCHE-D 95 07 05) the introduction of the notion that these virtual
machines *come into being* is, it seems, either just a way talking
that adds nothing essential, since all the necessary beingness is
already present in the physical matter itself, or it is an attempt
to prepare the way for the idea that these virtual machines do
*come into being* in the way similar to the way that our conscious
experiences seem to come into being, i.e., as an emergent quality of
beingness that arises from functional properties of the brain.

This latter approach is I think a proper one: it exploits the power
of the theorist to construct theories that are rationally honed to
problem at hand. Since consciousness is known to exist, and seems
connected to `functional' properties, a rational procedure is
formalize this connection. This tack seems to involve down-playing
the physics basis, since the new ontological type doesn't mesh
naturally with that of classical physics. The new ontological
type is superfluous within the context of classical physics:
nature, as conceived classically, doesn't need it. The new
realities do nothing beyond what matter alone does automatically
on its own. Hence adding superfluous actual realities in this
situation is unnatural: we add it, in spite of its unnaturalness,
to make a place for our consciousness.

However, the new action-based ontological type does fit in
perfectly with the concepts of an ontologically formulated quantum
mechanics (ala Heisenberg/von Neumann/Wigner), which also rejects
the classical conception of the static beingness of physical matter
in favor of a dynamical/functional one, based on events whereby
functional activities *come into being*.

The dispute over whether computation (i.e., function) or physics
(i.e., physical matter in motion) should be regarded of the
ontological basis of experiential reality lies at the root of the
Chalmers-Klein classification scheme, or at least of much of the
associated basic differences in opinion that pervade the field of
conscious research. It arises from the static or inert
classical-physics idea of the beingness of physical  matter.
This notion has always been a bit of a problem philosophically.
In quantum theory *being* is naturally connected to *doing*:
what is `actual' is the quantum event, which is naturally
an act by which a functional activity actually *comes into being*.
This quantum conception of what actually exists did not arise from
an effort to account for mental reality, but, on the contrary, from
Heisenberg's effort to return to the classical-physics ideal of
keeping human minds out of the content of basic physical theory.
Yet the functional properties associated with mind come in
naturally when that `purely physical' theory is applied to brains.

Although classical concepts will, of course, always continue to be
the basis of much of science, because of its broad domain of
applicability, one obviously ought to go to the more fundamental level
of physics if one wants to consider basic ontological issues, and in
particular to address adequately the computation/biology/physics issue.
At that level the issue seems to fade away, because the actual realities
are functional rather than material.

hpstapp@lbl.gov