From: Mitchell J Porter (mjporter@U.Arizona.EDU)
Date: Sun Apr 15 2001 - 21:51:22 MDT
A week from now, the Foresight Institute will be running
a big brainstorming session on the future (see
http://www.foresight.org/SrAssoc/spring2001/program.html).
There's no way I can be there, but I'm sure a number of
people on this list will be, so I'm going to say here
what I'd like to have said there, and maybe it will make
its way there and make a difference.
My aim would be to get people to consider, as a serious
possibility, that the mind is essentially quantum in
some regard. This is not a wild contingency or an idle
speculation; it's only a few steps away from things we
already know, and if true it is likely to have profound
consequences for the shape of things to come.
First: We already know that quantum mechanics is a basic
feature of the world. We also have a rudimentary theory
of quantum computation, and a growing roster of
experimental achievements that involve quantum
information (creation of entangled states, quantum key
distribution, quantum teleportation). To find that
some natural process essentially involves quantum
phenomena should not be surprising at all.
Second: There are now several decades' worth of debate
over how one can understand thought or consciousness
in terms of computation. There is no consensus about
how to do this, and there are a number of outstanding
problems. For example:
- Computational processes involve 'coarse-grained'
states of physical systems. But a multitude of
different coarse-grainings is possible for a given
physical system. If thought is a coarse-grained
neural computation, which one is it? Which coarse
graining is the 'real' one?
- Words have their meanings by convention, not
objectively or inherently. We have no evidence
that bit patterns in a computer are any different.
For a computational state to represent something,
it has to have a certain structure, but it doesn't
go the other way - the same structure can represent
different things. So if thoughts are computational
states, what feature determines their actual
referential content? What makes them 'about' anything?
- If thinking is algorithmic execution, does the
implementation matter? Could a Giant Look-Up Table
be conscious?
Third: It really is an open question as to whether
quantum coherence or quantum computation might
occur in living matter. There is a well-known paper
(http://xxx.lanl.gov/abs/quant-ph/9907009) arguing
that mesoscopic quantum states would rapidly
decohere under biological conditions; there is
another paper (http://xxx.lanl.gov/abs/quant-ph/0007020)
arguing that the first paper's argument does not
apply to quantum-*field* effects. The matter will
have to be decided by experiment, and biophysics
is still a young discipline.
I don't think that quantum computation per se is
any better at solving the problems in philosophy
of mind than classical computation. However, the
ground rules will be changed if quantum processes
are shown to be relevant to cognition, since the
novel ontological features of quantum theory (in
particular entanglement) thereby become potentially
relevant to ontology of mind. Wholly new approaches
will be opened up. I don't want to argue for a
a particular quantum theory of mind here, just that
it is a significant possibility that consciousness
is in some way intrinsically quantum-mechanical.
So what would be the consequences for the future?
I don't know, this is precisely *why* I would like
to see serious futurists thinking about this.
A new paradigm of mind would have to have a major
impact on human culture. And as far as technological
futurism is concerned, it seems that a quantum theory
of mind *must* have serious consequences for
artificial intelligence, mind uploading, and perhaps
cryonics. If thought and consciousness require
quantum implementations, they can only truly occur
in quantum AIs, and 'mind uploading' would have to
be into quantum computers. Many of today's most
radical visions of a posthuman Mind Age would need
to be thoroughly rethought.
So, that's my message for Palo Alto. Thanks for
your attention.
This archive was generated by hypermail 2.1.5 : Sat Nov 02 2002 - 08:07:00 MST