Stephen P. King (email@example.com)
Thu, 01 Jul 1999 16:58:45 -0400
This is very good! :=) I will make my responses without too much proofs,
as I wish to be brief and give continuity to this discussion.
Matti Pitkanen wrote:
> On Wed, 30 Jun 1999, Stephen P. King wrote:
> > Hi All,
> > I found an interesting quotable in david Chalmers' book The
> > Conscious Mind:
> > "So the suggestion is that the information spaces required by
> > physics are themselves grounded in phenomenal or protophenomenal
> > properties. Each instantiation of such an information space is in fact a
> > phenomenal (or protophenomenal) realization. Every time a feature such
> > as mass and charge is realized, there is an intrinsic property behind
> > it: a phenomenal or protophenomenal property, or a microphenomenal
> > property for short. We will have a set of basic microphenomenal spaces,
> > one for each fundamental physical property, and it is these spaces that
> > will ground the information spaces that physics requires. The ultimate
> > differences are these microphenomenal differences.
> > Of course, this view again requires a variety of "outrageous"
> > panpsychism, but I have already argued that such a panpsychism is not as
> > unreasonable as commonly supposed. Given that I have already suggested
> > that there may be phenomenal properties whenever there is information,
> > we might as well press these properties into service in a useful role.
> > The ontology that this leads us to might truly by called a
> > double-aspect ontology. Physics requires information states but cares
> > only about their relations, not their intrinsic nature; phenomenology
> > requires information states, but cares only about the intrinsic nature.
> > This view postulates a single basic set of information states unifying
> > the two. We might say that internal aspects of these states are
> > phenomenal, and the external aspects are physical. Or as a slogan:
> > Experience is information from the inside; physics is information from
> > the outside."
> I have probably read these lines a couple of years ago.
> This is nice formulation for dualism. Experience and physics as same
> thing seen from different sides. I believe that Chalmers defines physics
> what I would call objective reality. And experience as conscious
> information about it. There are several counterarguments.
> a) Why and how universe decomposes into several regions having inside and
> outside. Why no only single huge conscious experience representing all
> possible information about Universe meaning drowning into details as
> *duality* would suggest?
"Why and how universe decomposes into several regions having inside and
1) The Universe U as an actually infinite totality it All that Exists,
(not in the limit of an infinity which defines a potential infinity as these
define mere subsets of U). It is neither matter nor information in it-self. It
is All inclusive, there is Nothing that it is not.
2) U admits all possible decompositions relative to any indexing
(lexicographic or otherwise) by any subset of itself. In other words, the
Whole can be divided into as many parts as the dividing part can count. This
is related to the usual ideas implicit in set theory but we are talking about
dynamical systems that "become" not static "beings". This distinction is well
defined by Hitoshi's definition of "scattering" vs. "bound" states of local
quantum systems (LSs), considered as well modeling the subsets of U. Thus
subsets of U are seem as evolving systems in that their information content
and physical complexity is not constant. This idea is clearly modeled by
Peter's SIMs and MIMs.
3) The subsets of U have by definition "insides' and "outsides" given by
the basic notions of n-ary set membership. We can easily see that the
classical binary membership functions are special cases of n-ary set
4) See Hitoshi's discussions of the decomposition properties of LSs both in
his papers and on the Time List!
5) The p-adic relations are implicated in the hierarchical ordering that the
"inside" and "outside" entail iff we can show a non-Archimedean metric...
"Why no[t] only [a] single huge conscious experience representing all
possible information about [the] Universe mean[s] ... drowning into details
as *duality* would suggest?"
To posit a single huge conscious experience for U implies tacitly a
subject-object dichotomy, thus does not eliminate duality. The set of "all
possible information" describing U is not knowable by a single consciousness
since it is obvious that this set is not uniquely enumerable by a subset of U.
Put another way, there exists more than one disjoint convex (?) partially
ordered set of information, since the order in which the bits of information
is essential to the 'meaning' of the information. Information that is not
"about something other than itself" is not information. We can not posit U as
being conscious since it is ALL and as such can not have it-self as an object.
> b) Why conscious experiences seem to give so little information about the
> objective reality?.
Simple, conscious experiences are only possible to be had by finite
systems, the evolutionary behavior of such can span an nonenumerable number of
almost disjoint experiences, but only one instance at a time, this speaks to
the irreducibility of experiences! The subtleties of NP-Completeness of the
computational aspect of this is the proof.
Think of the possible minimum distance routes that a traveling salesman
can take, given N cities, such that she visits a given city only once. The
amount of computing resources needed to compute the route increases at an
exponential function of N. If we associate each city with an experience or
observations and the path joining them with a particular poset of observations
which constitute a given observer's experience to a solution of the
minimization problem, the needed computational resources rapidly exceed the
possible configurations of the physical realities that are being observed.
Thus we can not assume that these posets of observations exist a priori in
an experienciable sense. Yes, they "exist" in an ontological sense as subsets
of U, but they can not be said to be knowable prior to the experience of them.
They might be considered as bound states?
The notion that EPR espoused in their celebrated discussion with Bohr that
"If, without in any way disturbing a system, we can predict with certainty
(i.e., with probability equal to unity) the value of a physical quantity, then
there exists an element of physical reality corresponding to this physical
quantity." implicitly assumes that objects has finite definite properties
independent of an observation. This notion is severely at odds with the facts
of the Uncertainty Principle!
We can consider the set of observations to be isomorphic with the set of
possible experiments that can be performed on an arbitrary quantum system. The
results of such experiments are not a priori knowable, one must perform the
experiment -make the observation - and the results are definite only for the
moment of the measurement.
In metaphorical terminology, we can say that "the Universe can only
experience it-self via the interactions of it's subsets"; there are an
infinity of such interactions possible and they can not be performed
simultaneously, since that would imply that all the interactions are not
sensitive to the order in which they are taken. We know that the order in
which observations are made of a quantum system's canonically conjugate
aspects are sensitive to order, thus the NP-Complete problem of ordering
prevents all possible observations to occur simultaneously. The notion of
computational concurrence is implicit in this!
One of the consequences of the idea discussed here is that the prior
computational results are "available" to all of the systems involved in the
finite interactive computation. We see a good discussion of this in Rupert
Sheldrake's books: A New Science of Life : The Hypothesis of Morphic Resonance
and The Presence of the Past : Morphic Resonance & the Habits of Nature. (
http://www.sheldrake.org/experiments/ ) This idea is discussed in my paper
http://members.home.net/stephenk1/Outlaw/forbiden.html . This notion is
implicit in the notion that "logical implication goes backwards" that Pratt
discusses in http://boole.stanford.edu/chuguide.html#ratmech Thus we say "A
causes B if and only if B* implies A*"; where A and B are the physical
configurations and A* and B* are the information structures encoded. Also,
Peter's hypothesis that secondary observers are the cause of the irreducible
randomness of quantum systems is related! :)
> c) How it is possible to have wrong information, make mistakes, in the
> framework of strict duality?
A good question! "Wrong information" is synonymous with incomplete
information or noisy information. We generalize Pratt's strict binary duality
(as explained in http://boole.stanford.edu/chuguide.html#ratmech ) to show
that the involution transformation is inexact for all Chu transforms that
involve incomplete and/or noisy information. We might be able to use the
Poisson brackets (?) to model how the Mind -> Body -> Mind (Body -> Mind ->
Body) involutions are not exact in the sense the the original is recovered.
The exact recovery of such can be considered as an algebraic identity, related
to a single fixed point. Upon generalizing the involution transformation we
see that the "greatest fixed point" that Peter discusses appears as a natural
consequence of the "inexact" duality involution.
This illustrates that the original Cartesian duality is just plain wrong
as Pratt points out. We have a dualism that is well know in the mathematical
world, only its usefulness in solving the Mind-Body interaction problem has
not be understood by philosophers.
> d) Isn't consciousness is epiphenomenon? Is free will illusion?
That is consciousness an epiphenomenon of? Even if free will is an
illusion, we must be able to explain the existence of the illusion! Since we
are proposing that Logical entailment (implication) is dual to physical
causation and we are allowing to error to introduce inexactness in the
mappings, we generalize from strict onto isomorphism mappings to fuzzy almost
isomorphic mappings between instantiations of experience, or more
mathematically, from strict (invertible) Cauchy evolutions of observables to
non-Hausdorf Gelfand (?) evolutions ( http://xxx.lanl.gov/abs/quant-ph/9712038
) of observations. Thus the classical notion of causal determinism is seen to
be the illusion due to the smallness of the Planck constant and not free will!
> e) Doesn't this lead to panpsychism as Chalmers himself admits?
Panpsychism, properly understood, is not necessarily a bad thing!
We must consider exactly what we mean by "panpsychism"! The identification
of "soul or spirit" (as in the above reference) with a poset of information is
not too difficult nor "counterintuitive".
> The manner to save the day is to introduce quantum jump between objective
> realities and define conscious information as information difference.
But is it not necessary and sufficient that these "objective realities" are
not a priori to particular instantiations of observations? We have a
coinductive relationship (http://lex.ucsd.edu/links/demos/doc/coind.htm)
between the (finite) physical configurations of the LS (modeling observers)
and the (finite) information structure (? n-ary lattice (Boolean for n = 2) ?
). The "information difference" is given relative to the commonalties ( ?
intersection ? ) between the posets. We should say we have a shared "reality"
instead of postulating an "objective reality" for all possible observers!
Onward to the Unknown,
This archive was generated by hypermail 2.0b3 on Sun Oct 17 1999 - 22:36:55 JST