Dear Aaron,                                  Dec. 23, 1997

Your message came at an opportune time. I have received a number
of communications generally like yours, but less incisive,
and I am just in the process of making additions/amendments to my
paper for the purpose of clarifying these points. Your paper will be very
useful. As a warm-up, and to give to you a shorter and quicker answer,
I will answer your questions here, briefly.

>From:	SMTP%"A.Sloman@cs.bham.ac.uk" 22-DEC-1997 02:27:22.32
>To:	STAPP
>Subj:	Ontology paper

>Hello Henry (and everyone)
...
>Anyhow, I've at last started reading your "Quantum Ontology and Mind
>Matter Synthesis", and finding it mostly very clear and interesting.
>
>Mostly I don't find myself (so far) actually disagreeing, though I am
>not sure exactly what is being said in various places.
>
>E.g. I have no quarrel with the key notion that the building blocks of
>reality may have more to do with information than with features
>like mass, solidity, motion, etc. -- which is my way of saying what I
>think you express by saying:
>
>    "nature is built not out of matter but out of knowings"
>
>though possibly there's a subtle difference, on which more below. In
>general I don't think it matters what kinds of "stuff" the universe is
>made of as long as those kinds of stuff have the right causal
>interactions to explain everything else. But that, I mean that when
>various sorts of configurations of the stuff exist they should be
>capable of producing various sorts of states, processes, and
>interactions with other configurations.
>
>I can vaguely imagine that the universe might be built out of fragments
>of information (e.g. fragments of information about other fragments and
>configurations in the environment -- which of course raises questions
>about where the recursion bottoms out).
>
>This this might give it far more versatility and power than if it had
>been built out of something like newtonian point masses. Chunks of
>information (semantic content) can interact in far more interesting ways
>than chunks of newtonian matter.
>
>Don't ask me to explain all that: it's just a vague probably incoherent
>preamble explaining that at some level I have no objection to what you
>write, and would be happy to discover that the mathematical details of
>the theory actually provide a much deeper and more powerful version of
>these intuitions. (This is probably related to Leibniz' theory of
>monads which I am slowly reinventing?)

HPS: Yes! My ojective is basically to give the mathematical details
of how to build our theory of nature out of information-type stuff
rather than matter-type stuff. The point is that QT has that kind of
structure. That theory is basically different from CM (classical 
mechanics) because it has holistic/nonlocal events that are the 
actualizations of human knowings, but that also are represented
within the mathematical structure. These events drop out of the theory 
in the approximation that gives deterministic classical mechanics. 

>1. My main puzzle (so far) is a point that has come up often before, and
>which I think Pat also raised after reading the paper. Namely, it's one
>thing to say that in some sense the ultimate constituents of the
>universe are, or involve knowings, and that these are referred to in the
>equations of quantum mechanics.
>
>It's quite another to slide into talking about
>    "our conscious human knowings",
>as you frequently do.
>
>That apparently commits you EITHER to saying
>
>(a) that there were conscious human knowings before there were any
>    humans, and in places where there are no humans (e.g. the centre of
>    a distant supernova?),
>OR
>(b) that the totality of the universe is what happens to be known to
>    humans, and it's just a mistake to think that the physical world
>    existed before we did and might exist after we are gone.
>
>This raises all sorts of questions: which humans? Does it matter if no
>single human existed all the time and there's just a sort of wave of
>humanity whose "knowledge" constitutes the fabric of the universe?
>
>If we get rid of human knowings and are left with a more abstract and
>generic notion of elements of information which are transferred,
>transformed, transmitted, combined, etc. etc. then I have no problem
>about that, AS LONG AS the laws of interaction suffice to support all
>the phenomena to be found in the universe, including mental states and
>processes, poverty, crime, evolution, computers, and deadlocks in
>operating systems.
>
>This could even be close to what Leibniz thought about the nature of
>reality, though he ddin't come up with appropriate equations or laws of
>dynamics to do the required job.

HPS: The strict Copenhagen interpretation was in terms of human 
knowledge. But it did not aspire to be an ontology. My intention
is to extend it in a way that will retain all of the pragmatic
results of that theory, which encompass all the empirically validated
results of both classical and quantum theory, but allow all of that 
structure to be seen as part of a reasonable ontological description.

So I do retain the human knowings, but regard them as a special case
of a more general type of reality that is not confined to human beings.

The approach is basically different from that of classical mechanics
because it starts with, as given, this human-knowledge-based theory
as the launch point. 

Since we have human knowings as basic elements we have, it seems to 
me, a much better basis for including `crime','poverty', `computer 
programs', etc. 

The events that are analogous to human knowings, but that are occurring 
in simpler systems, are not experiencially equivalent to human knowings. 
I use the general term `knowings' to encompass them, since human 
knowings are the known paradigmatic case.. 

But my discussion of chords of vibrations was intended to give an idea 
of what a "knowing" associated with some inanimate system might be like. 
But of course our memories are so important to the quality of our 
experiences as to make the "experience" associated with the actualization 
of some chord of vibration in an inanimate system difficult to imagine.

The difficulty here, as regards science, is that the pragmatic quantum
approach that includes only human knowings, encompasses all empirical
findings yet discovered. So although rational thought demands that
in an ontological setting we must have non-human generalizations
of the human thoughts that enter into QM, still we have not yet
had any experimental confirmation. That is THE big problem.
In principle the analogs of the events of human knowings should be 
difficult, but not strictly impossible, to detect. 


>It's been obvious enough for a long time that even if the world is made
>of matter the matter cannot be anything remotely like our ordinary
>conception of physical stuff, whether solid, liquid or gaseous.

HPS: Are you refering to quantum phenomena?
And for me "matter" means, unless otherwise defined, the stuff 
that CM talks about.


>If by "knowings" you don't mean to refer to states, processes and
>abilities in human brains but something more primitive, not necessarily
>part of anything human, then presumably the concept of a "knowing" is
>defined entirely by the relevant equations in the theory? In that case
>our calling these things "knowings" or "information fragments" could be
>misleading because of the familiar connotations of those terms. I.e. the
>use of existing concepts to describe these new entities needs defending.

HPS: Yes! One might just ignore any "psychic" aspect of these events
except in human beings.

But if we succeed in finding the neural corrolate of, say, pain in
human beings we might be in a position to have SOME evidence about
whether various animals feel pain. And conceivably we might someday
be able to say more about some connection between the circumstances
under which collapse events occur and some theoretical idea of the
"experiential" aspect of these events. But I generally agree
that the attempt to discuss the experiential aspects of the nonhuman
analogs of human knowings might not be scientifically useful.


>(I am not impressed by quotations from great physicists who have not
>been trained in conceptual analysis!)

HPS: My quotes were meant to confirm what the scientists themselves
thought they were doing: what the Copenhagen interpretation is?
But I do think that these words do signify something important:
they signify that the founders did in fact consciously formulate 
the theory in terms of their descriptions of what they might do
and might observe instead of properties that they thought could be
adequately described in a more traditional-in-physics way.
This at least opens the door to a conceptualization of physical
theory that brings knowlege/information in in a different way than in
classical mechanics.


>Anyhow, assuming we can get round the main conceptual difficulties, I am
>left with the question whether your knowings do or do not require
>humans (or other animals). Mine (and Leibniz') don't.

HPS: Nor do mine!

>2. I have another problem, which perhaps will be resolved when I've read
>more, concerning your hypothetical experiments in which decisions are
>taken in two regions of space-time which are too far apart for any
>signal to link the decisions.
>
>My problem has to do with your description of what happens in each
>region, i.e. an observer does or does not make an observation, and you
>require this to be "freely chosen" i.e. ``without there being any cause
>*from within that physical system* of this act''. Now if the two
>events are outside each others' cones of communication there's no
>problem about neither influencing the other.
>
>But each region must be part of a larger history with waves of influence
>coming from possibly shared causal precursors far enough back in time.
>
>This shared history could make it impossible for the two events to be
>totally unrelated and independent. Even if each involves a scientist
>taking a decision triggered by decay of a particle in the brain, it
>could be that there are constraints on possible combinations of
>decisions which come from all the way back to the initial configuration
>of the universe (or some intermediate structure).
>
>Now I know that individual state changes in sub-atomic particles are
>supposed to be inherently non-deterministic, i.e uncaused. But according
>to my dim understanding the laws of QM have pretty rigid statistical
>implications. This means that if you think about sufficiently large
>numbers of space time regions in which similar non-determined events
>occur, then certain patterns of events will be ruled out with an
>astronomically high probability. If this didn't happen QM would be
>refuted, e.g. if predicted interference patterns were not found when
>lots of electrons are shot through two slits.
>
>This implies that there has to be some sort of causal link between the
>distinct events even if no signal can pass from one to the other. The
>causal link goes back via some possibly large configuration, possibly a
>long time ago.
>
>All of this, to my mind, casts doubt on whether it really is possible to
>have the two regions containing experimenters who are TOTALLY causally
>disconnected.
>
>Of course, if you believe in some kind of magical free will which isn't
>constrained by the laws of physics then we probably have no more to
>discuss. But I don't think you do, in which case it is not obvious to me
>that your hypothetical experiment can really involve all the degrees of
>freedom which you claim. In fact, isn't that kind of non-local
>interaction exactly part of what characterises QM?

HPS: There are two different issues here: One is whether we can
treat the *choices made by the two experiementers* as if they were free,
and not entwined in some subtle way with the system they are examining.
Even if we bring the experimenters in from another planet, or fix their
decisions to be controlled by combinations of temperature readings in
Chicago, New York, and Bombay, combined with cosmic ray arrival times
from a distant galaxy, you could say that one cannot treat that
decision is if it were not connected to the details of the experiment
because everything comes from the big bang. Indeed, I-ching seems to be 
based on such an idea. So if you take that position the proof breaks down.
But most scientist seem to think that that is pretty far-fetched.
and are willing to accept that choices that can by fixed in such whimsical
ways can be TREATED AS IF they were unconstrained.  

The fact that the system being examined has a past is fully taken into
account: it is only the whimsically chosen choices of the experimenters 
that must be treated as if they were disconnected from the rest of the
experiment.

>3. I have various other worries e.g. when you compare QM with CM:
>
>    "Quantum stuff can do something very different: it can come into
>    existence in a way that classical matter cannot. This is because the
>    quantum stuff is knowledge, which can truly grow, not matter, which
>    can only rearrange."
>
>This seems to me to miss the point that rearrangements can amount to
>growth of knowledge. That's what happens in brains and computers for
>instance. Thus "only rearrange" is misleading if it implies that
>rearrangements never constitute or implement changes in knowledge.
>(Another facet of the "nothing buttery" debate.)

HPS: I'll change my wording here. The point is that in QM one has
sudden nonlocal comings-into-beingness of whole complex structures.
Nothing comparable can happen in CM.


>4. Finally a point of curiosity. In my recent postings to psyche-b
>(saved in http://www.cs.bham.ac.uk/~axs/misc/binocular-rivalry )
>I've been trying to undermine the Baars' theory that the contents of
>consciousness must be internally consistent, by showing how an
>appropriate architecture, e.g. a biologically useful visual
>architecture, could contain inconsistent elements. (Stan seemed to agree
>that Baars had got this wrong.)
>
>Does the QM theory of consciousness say anything supporting or
>contradicting Baars?

HPS: I have been taking as a basic premise the idea that each event 
has to actualize a coherent idea. But sometimes we do have fuzzy
ideas, which we strive to make clear and coherent. So I quess I'll
have to think more about your question.

>Cheers
>
>Aaron
>====
>Aaron Sloman, ( http://www.cs.bham.ac.uk/~axs/ )
>School of Computer Science, The University of Birmingham, B15 2TT, UK
>EMAIL   A.Sloman@cs.bham.ac.uk
>Phone: +44-121-414-4775 (Sec 3711)       Fax:   +44-121-414-4281