From: THEORM::STAPP 23-AUG-1997 14:30:34.60
To: @KLEINDIS.DIS
CC: STAPP
Subj: Re: Quantum...synthesis: Reply to Aaron
Pat Hayes writes:
------
Henry- re. your recent reply to Aaron. OK, current physics does not allow
us to retreat into a comfortable assumption of Newtonian regularity.
However, given the following range of options, I know which I find the
'spookiest':
1. Action at a distance which violates special relativity.
2. Small breakdowns of past-future causality, so that the future might
indeed partly determine the past.
3. A cloud of subjectivity which infuses the entire cosmos throughout space
and time and yet is only made apparent, mysteriously and briefly, in the
neural organs of a few species of carbon-based organisms inhabiting a small
rocky planet orbiting a minor sun in an undistinguished galaxy.
Best wishes
Pat Hayes
PS. I wonder, could you briefly explain the following:
2. The *part* of the quantum dynamics that is expressed in terms of the
physical aspects is incomplete in a way that provides a perfect place
for dynamically efficacious conscious experiences.
Where is that perfect place, exactly? One of my difficulties with the New
Quantum Mysterians is that I can find no such place, since the predictive
power of QED depends on the collapse of the wavefunction obeying exactly
the probability distribution provided by the Schrodinger equations. That
'incompleteness' doesnt seem able to account for subjective experience,
since if a subjective agent were to work any influence over these
distributions, the result would be soon observable as *violating* the
predictions of QED.
----
Welcome back Pat! Missed your challenging presence.
About your options 1 and 2 I should make it clearer that 1 is the only
one: I expressed option 1 also in terms of action of future on past only
to try to put it in its strongest starkest form, or at least what appeared
to me to be strongest and starkest. Precisely stated, the condition that
cannot be reconciled with the predictions of quantum theory is the demand that
[given two spacelike separated regions A and B (i.e., two regions such that
no point of either region can be reached from any point in the other without
going faster than the velocity of light), and an experimenter in each
region free to choose and perform---all completely within his region---one
or the other of two possible experiments, each of which will yield one
or the other of two possible outcomes]
any statement that:
1), involves conditions pertaining to only one of the two
regions ; and
2), can be proved to be true provided only that the experimenter in the
other (say far-away) region freely chooses to perform, say, the `first' of
his two mutually exclusive possible experiments,
must continue to be true also under the condition in which that far-away
experimenter makes (in his far-away region) the other choice.
If the idea that no influence can travel faster than light is strictly true
then this demand MUST be satisfied because
the information about which choice the far-away experimenter makes
does not have enough time to get to the nearby region to influence
the outcome there. [No condition is placed on the outcome of the
measurement in the far-away region: that is not mentioned.]
The two regions can be very far apart (a recent experiment in Switzerland
is supposed to have confirmed the fact that the relevant connections
hold over distances of at least 14 km), and the relevant velocity can made
much larger that the velocity of light (no limit in principle), and the
effects are large (e.g., violations of the nondependence postulate in 25%
of the instances), and (in principle) the effect does not fall off with
distance!
Your option 3 (cosmos infused with subjectivity) is not an option. The
only option is 1 (or a many-minds kind of universe where "everything happens"
and all possible---mutually incomplatible---outcome are experienced as
happening, but "we" are not aware of all the other parallel "we's" who are
having the other experiences, e.g., the `other' possible outcome of the
experiment.)
[Actually, I believe that this many-worlds option is not really allowable,
but my argument needs to be rigorized.]
The introduction of consciousness in my proposal was not supposed to
`get around' or `cope with' the nonlocality problem: it does no such thing,
and was not meant to. I stressed the nonlocal character of nature (modulo,
perhaps, the many-minds scenario) only to justify my claim that the
classical conception of nature are "profoundly" wrong: these
nonlocalities are large, pervasive, large-scale (e.g., kilometers) effects.
These gross breakdowns of classical ideas about causality has, amazingly
enough, not destroyed physics. There is a "physical theory" that does
codifies all our empirical findings.
Because this theory codifies EMPIRICAL FINDINGS it is naturally
expressed in terms of our experiences.
Another reason for expressing it in terms of our experiences
is this: what else can we safely do when the entire normal idea of
the causal structure is known to be fundamentally false.
How do you keep your sanity in the face of such a
finding, which seems to jeopardizes the meaning of everything we do,
i.e., of every interpretation of every experiment we do?.
Keeping science afloat amidst such a huge breakdown of normal ideas
about causality was not an easy matter. This difficulty is why so much
of the discussion about QM is about causality.
My ontological proposal does not, as you seem to think, put human
beings in any special place of honor. Exactly the opposite!
I stressed [See my "Companion Posting August 19"] that my proposal
was a extension/elaboration of Heisenberg's idea that there are "events"
out there associated with inanimate systems. These events are not associated
with any human observer: almost all events are undoubtedly
of this non human-related kind.
The reason for talking about nonlocality was just to justify going over
to this "event" type ontology recommended by Heisenberg, i.e., to provide a
powerful, and I think compelling, reason for giving up the comforts of the
intuitively congenial ontology based on substance.
Once one accepts the idea that nature has these collapse events associated
with various physical systems, it would be unnatural not to allow events
associated with our own body/brains. But these events play no bigger a role
in nature than we do.
Thus your option 3 is not only no way out of the nonlocality
problem, it is completely contrary to what I have proposed.
It is sheer lunacy.
However, these special human-based events are important to "us":
"our" science is based on them, even though the basic ontology is not.
Our most basic science is quantum theory. I take it in its most
orthodox form, which associates certain collapse events with our conscious
experiences. The structure of quantum theory allows these collapse events to
be associated in a special way with our body/brains, although they are not
*confined* to our body/brains.
[The projection operator associated with the collapse is built out of
the field operators associated with spacetime points lying in the
persons body/brain, but the action of this operator on the state
vector of the universe can change properties far away from this
body/brain.]
You asked:
-----
PS. I wonder, could you briefly explain the following:
2. The *part* of the quantum dynamics that is expressed in terms of the
physical aspects is incomplete in a way that provides a perfect place
for dynamically efficacious conscious experiences.
Where is that perfect place, exactly? One of my difficulties with the New
Quantum Mysterians is that I can find no such place, since the predictive
power of QED depends on the collapse of the wavefunction obeying exactly
the probability distribution provided by the Schrodinger equations. That
'incompleteness' doesnt seem able to account for subjective experience,
since if a subjective agent were to work any influence over these
distributions, the result would be soon observable as *violating* the
predictions of QED.
-----
By the "physical aspects" I mean the part specified by the Schroedinger
-directed state vector of the universe. This part is a generalization of
classical mechanics: it is expressible over spacetime and is deterministic.
And it is governed by a local process.
But the whole second part of the dynamical process is left out: there is no
hint within the "physical" formal structure of any collapse events. (That is
the whole basis of the many-world thesis that there are no such events). So
if we are going to seriously pursue Heisenberg's event idea ontologically
then a whole new machinery, yet to be developed, is needed.
The idea is too daunting even to consider. But logic demands that we do so.
Once identified, it is clear that this part of the dynamics is the real
actual ball game. This part is what is really in charge: the part described
by the quantum formalism is some lesser statistical shadow.
Of course, that is what Einstein was saying. But he was unwilling
to swallow the bitter pill that is inseparable from it: the failure of the
relativity-theory idea of no faster-than-light influence.
And if we include (as a very special case) those collapse events associated
with our human experiences that orthodox quantum theory is built upon,
naturally associating them with those particular physical systems we call
our body/brains, then our experiences *become involved with* collapse events
that are controlling the activities of our body/brains.
But what, exactly, is the nature of this "become involved with"
relation? That is the key question!
But the answer is obvious. What is required by the conditions already
set forth is that the needed new machinery take the "raw data" that
is represented by the wave function of the body/brain before the
collapse event, which according to Heisenberg's intuition is to be regarded
as "potentia" (of Aristotle), and create from this "raw data"
the single coherent experience that the orthodox quantum theory demands,
and that appears in the realm of experience associated with that body/brain
in conjunction with the corresponding physical (collapse) event
associated with that body/brain.
I assume that the full process is fully causal: I find absurd the idea
a definite choice can pop out of nothing at all. But in contrast to
the Schroedinger process, this selection process is nonlocal and
nonunitary.
Of course, there are a multitude of conceivable ways to go, all of which can
be explored.
But let me just boldly pick one that strikes me as
natural/plausible/reasonable/pleasing.
You would like to say that the conscious experience IS
some aspect of the body/brain: that it is not merely INVOKED
by some complex pattern of neural activity.
I, and many others' find this pill too hard to swallow, if the brain is
conceived classically: for in that case the two concepts are too
disparate. I am certain that it will never be possible to explain how
those two things can be the same thing.
[One reason is that experiences really exist but the particles
postulated in classical certainly do not. Another reason is that
classically conceived matter is by definition nothing but a
disjunctive aggregation of infinite tiny ontologically independent
bits, whereas conscious experiences are ontologically more complex:
the information they carry cannot be broken down into such tiny
disjunctive bits. I do not expect you to feel the force of this argument,
which I elaborated upon in my "nothing buttery" exchanges with Aaron]
But let me explain why I think the functionalist position may
very well be "right", once the quantum character of the body/brain
is recognized and exploited.
A quantum mechanical system is represented by a vector in a Hilbert space.
But what is a vector on a Hilbert space?
Well, any mathematician can tell you. He will give a set of mathematical
properties and say that a vector in a Hilbert space is anything that
satisfies these properties. Fine!
But what is the nature of the reality that is represented by the state of
the universe?
From the pragmatic point of view no answer is needed or called for: the
vector is just our computational tool.
But when pressed the pragmatic physicist will have to say that it represents
his knowledge.
But now we are groping for an ontology. But what IS the reality represented
by the state vector of the universe, as an ontological beast?
I can tell you that some vector in some calculation that I am performing;
or contemplating, represents a line going from this point on my desk to
that point on my desk: THAT physical vector is the REAL vector that
the vector in my thoughts represents.
So what is the real thing that the state vector of a quantum
ontology represents?
It is, I think, in an ontological context, not enough to say that it
IS nothing at all but the abstract mathematical construct that
appears in our computations, or imagined computations. Or at least it is
logically possible that this vector that we imagine represents something
ELSE that really exists, and has the mathematical properties of a vector
in Hilbert space.
This line may seem excessively metaphysical to you, but please bear with
me for a moment.
Let us just go along, for a moment, with what the founders were
saying, but now from an ontological standpoint.
The said the state vector represented "our knowledge", and that
the nonunitary, nonlocal changes in the state vector represented
"experiential changes in our knowledge". I suggest that to pass from the
pragmatic to the ontological stance we drop the "our".
Mathematically, the changes generate by a unitary transformation
(or evolution) is basically no change at all: it can be regarded as
merely a shift in point of view of something that has not changed at all.
Indeed, that is how it is often interpreted in quantum theory.
A key feature of the unitary evolution generated by the Schreodinger
equation is that evolution operator does not depend on the state:
there is, in this sense, NO SELF-REFERENCE!
But in the nonunitary transformation that is associated with experience
the evolution operator does depend on the state: THERE IS SELF-REFERENCE!
And there is a real change: a change that is more than merely a change
in point of view. I the classical limit only the trivial kind of change
survives.
The idea that the realities behind the physics formalism are more like
knowledge and experience that like classical substance does not seem
absurd to me. The notion that our experiences are realities having
a mathematical structure does not seem completely ludicrous. For
they are closely connected to brain activity, and brain activity
has mathematical structure. And Hilbert space vectors can represent
extended whole structures: whole structures in brains, and, by going
to momentum space, whole patterns of activity.
I do not know whether these kind of considerations resonate with any
part of your intiution, but I throw them out for your consideration.
I'll have to think some more about this crazy idea. But it does give
one a toe-hold on the idea that we should eventually get rid of duality,
with things on one level INVOKING things on the other level, and see
mind and matter as aspect of one and the same.
But let me get back to your question.
I am supposing that the distinguished vector(s) that NEED SOMEHOW TO BE
DEFINED in connection with each quantum event (i.e., the basis vector(s)
associated with the event) are defined by their experiential
qualities---they are not defined by any known physical quality, and
in the only cases where collapse events are "known" to occur, the
distinguished basis vectors are in fact characterized by their
experiential qualities (that was Bohr's and Heisenberg's and Dirac's
main point). This choice lies outside the basic rules of the quantum
formalism, and hence no disruption of the predictions of quantum theory
are engendered by saying that the selection of the distinguished
basis vectors is governed by the experiential-type qualities of the
various possible basis vectors.
By controlling which questions are asked---i.e., which basis vectors
are being used to specify the framework for the collapse---the
experiential aspect can exercise great influence over the course of brain
events, staying completely within the statistical rules
of QM.
For example, of one asks again and again always the same question,
giving the brain chances to come at the affirmative answer in many ways,
then the affirmative answer (which is the one associated with an
experience) is likely to come up eventually. A quantum system that
is empowered, by virtue of the self-referential property of the
evolutionary process that governs the collapse event, to
choose which questions it will ask itself, on the basis of the
experiential qualities of the possible answers, can, while strictly
conforming to the quantum statistical rules, move ahead in a direction
controlled by the coherency of the experiences it generates.
Your questions were good one's: I hope they have moved us ahead.
Best regards,
Henry
---------------------------------------------------------------
Artistic Director
Festival Opera Walnut Creek
(510) 944-9610
--------------------------------------------------------------
--MimeMultipartBoundary--