Stephen P. King (stephenk1@home.com)
Tue, 24 Aug 1999 15:44:53 -0400
Dear Matti et al,
Matti Pitkanen wrote:
snip
> > > [MP] Also cognitive spacetime sheets contain classical fields and energy
> > > but energies are extremely tiny. Cognitive spacetime sheets
> > > take small sample of material spacetime sheet: they are measurement
> > > instruments.
> >
> > Umm, how is the "sample" related to the "whole"? This looks like the
> > css "tastes" the mss...
> >
> [MP] Basically it tastes. For instance, extremely tiny current through
> potentiometer allows to measure the potential difference associated with
> circuit. Frequencies are mapped to frequencies (radio forecasting).
I did not think that it is necessary to have direct "connections"
between the space-times of LSs, e.g. worm holes, but the main idea of
TGD, starting from Wheeler's thought, makes me wonder about it. We
really need to have an understanding of how Pratt's causality works. We
should call it "coinductive causality". The mapping of frequencies to
frequencies, does not require "touching", the mechanical means of
causation, we can instead think in terms of "entrainment" and
"bisimulation".
> > > [MP] Without geometric time one loses practically all known physics.
> > > For instance, there would be no concept of energy
> > > (invariance of laws of physics with respect to time translations),etc...
> > > Main stream physics do not even mention the notions of
> > > psychological time or subjective time: as far as predictions are
> > > considered only geometric time is needed. Of course this all
> > > is just play with formulas. They are not understood but they work
> > > magically. For instance, scattering rates involve integrals over entire
> > > geometric spacetime: taking literally this is nonsense from standard
> > > physics view but the formulas work.
> >
> >[SPK]I have a way of defining energy! I just do not know how to explain
> >it mathematically... It is roughly the potential difference between the
> > LSs. If the LSs are at equilibrium with respect to each other, there is
> > no energy definable. The use of "integrals over entire geometric
> > spacetimes" to calculate scattering rates makes sense!
> >
> [MP]
> Defining energy withouth introducing time: I am not sure whether you mean
> this. But again there are *very* strong purely group theoretic constraints
> on energy. Energy and momenta must transform in quite specific manner in
> Lorentz transformations: this is empirical particle physics fact.
> To summarize it: if one gives up time, one must still be able to realize
> Poincare group as at least approximate symmetries, not in spacetime
> anymore but in statespace.
Umm, this characterization is incorrect. The LSs that are at
equilibrium with each other have neither energy nor time relative to
each other. What we really need to to understand how the time and energy
quantities are canonical conjugates of each other. I completely agree
with your point about the group theoretic constraints! I hope to have
the Schommers papers to you by the end of the week, I would like to
discuss his formalism of time-energy operators.
> Also Maxwells equations, etc.. are essentially 4-dimensional construct.
> E and B form components of 4-dimensional tensor. Again one should
> end up with situation in which everything is apparently 4-dimensional.
> Time appears as a parameter which behaves as a time coordinate
> in spirit of Einstein. In this kind of situation it is not useful
> to pretend that geometric time is not there.
I agree. :-) This is one of the reasons why I think that Weyl's scale
invariance theory has something to tell us! His theory seems to explain
why we experience objects in terms of 4-dimensions. And this action term
gives a simple principle of greatest probability. "The law of nature is
that the actual state of the world [that the observer experiences] is
that which is statistically most probable." (Space, Time and Gravitation
: An Outline of the General Relativity Theory (Cambridge Science
Classics Series) by Arthur S. Eddington, pg. 178) [my note] The
relationship between the vector potential and scalar potential, sources
of EMF, that seem to me to be intimately related to the dimensionality
and group theoretic behavior of space-times and that it is possible to
relate these to variations of geometry is significant.
> > > The problem is to understand the triplet subjective time-psychological
> > > time-geometric time: how they relate to each other.
> >
> [SPK] Yes! I would very much like to understand what you are thinking on
> > this! :-)
>
> [MP] It seems that I am poor communicator or that these ideas are not
> communicable(;-). Could it be that your dualistic philosophy is what
> makes it so difficult to understand my point of view?
I don't think so! In the dualistic mode, a "third" is understandable,
it is the relationship between the dual opposites. Thus, I think, that
"psychological time" is a relationship between subjective and geometric
"times". The problem I have is that you use the term time in a confusing
manner. To me, time is a measure of change. Applied to geometry, it is a
measure of the variations of the geometry. Applied to subjectivity, it
is a measure of the variation in the perceptions of such, this would be
a variation either continuous or discrete of the frame of observation,
the "picture seen". This translates into a variation in the space-time
that the LS observes.
This seems to conflict with the idea that geometry is independent of
observation, but perhaps not! It we consider that the space-times of all
LSs form an equivalence class ST, we should be able to consider ST to
have its own "time" (it would be similar to Wheeler's "superspace"), but
there would be some subtleties that we need to look at carefully.
Were psychological time fits in seems to be a cultural/language
construction. The interactions of persons seem to select for certain
modes of perception. I must, though, protest since this is taking us
into an area of study that is far to anthropocentric for my taste. I am
interested in the philosophy of physics for observers in general, not
just those confined to the Earth!
snip
> > [SPK]
> > > > Umm... why then do people identify the reals with temporal events? Your
> > > > idea of "values seem to increase" is usually associated with "the
> > > > thermodynamic arrow" in the literature...
> > >
> > > [MP] Perhaps because Einstein talked about events. Just thinking
> > > carefully what Riemann geometry really is one realizes that
> > > there are no time arrow involved. Time and space are in same position.
> > > There is no preferred time coordinate to measure. If one introduces
> > > clock measuring geometric time, one should introduce also clocks measuring
> > > all three spatial coordinates.
> >
> [SPK] We really do not need to assume that space and time are actual
> >outside of experience.
>
> [MP] As I told above: if you give up time you have hard time in
> reproducing existing precision tested particle physics and the end result
> is the conclusion that everything behaves as if geometric spacetime
> exists and the simplest thing to do is to allow its 'real' existence.
I apologize that I did not explain myself adequately. I do not want to
"give up time"! I am very interested in a physics that has a time that
"flows"! But in order to do so, we need to understand and model how this
"flow" occurs and what is the "potential" that "drives it". The fact
that the Universe can be decomposed into disjoint "parts" is an
indication that there exists a state of non-equilibrium among them, and
I propose that it is the "equilibration" of this non-equilibrium that is
doing the "driving".
I think that it is because the Universe in itself can not "know" what
"it is like to be itself" in zero time. The potentially infinite flow of
time is the manifestation of the Eternity required for the Universe to
experience all modes (finite and otherwise) of itself. This notion is
very "mystical" and thus inherently problematic. I am happy though,
since it appears that we can formalize it, and thus create a good
quantitative physics model. :-) The key idea is that there is a
consistent way of mapping information to material configurations (and
its inverse) and this is the foundation of the duality that I am
proposing.
One thing that I believe with help would be to discuss what we mean by
"knowledge"!
[MP]
> > > The map assigning definite temporal cm coordinate of cognitive spacetime
> > > sheet maps subjective time to geometric time and then we make the error of
> > > projecting the irreversiblity and clockability of subjective time
> > > to properties of geometric time.
[SPK]
> > It is just that you use the term "time" so liberably. I am confused.
> [MP]
> I am not using time liberably. I have only done a lot of work to
> discover that it has very many meanings which usually are not realized.
> I carefully distinguish between subjective time, geometric time and
> psychological time.
Ok. :-) I just would like to understand your point of view.
> Of course, general coordinate invariance allows define infinite number of
> variants of geometric time: Minkowski time, lightcone proper time, time
> coordinate of spacetime surface, etc... This increases the confusion
> unless one is very careful with the basic definitions.
Yes, this is why I like Hitoshi's model.
> Subjective time is something absolute: it just ticks, you cannot apply
> Lorentz transformations to it. There is no subjective space.
Umm, I am not so sure! I see that we could define a "subjective space"
as being to "subjective time" what an ensemble is to a time series. I do
not how to say this in mathematics. :-(
> > > [MP] I would have agreed about symmetry for half a year ago but I
> > > do not know anymore. Strong NMP did not
> > > tell whether it is subsystem or its complement which is measured.
> > > The notion of self forced different interpretation: experiencer
> > > is self whose subsystem and its complement
> > > inside self define the quantum measurement.
[SPK]
> > Could you please give us an explicit explanation of "strong NMP"? Is
> > there a problem with the subject-object symmetry?
>
> [MP]
> Strong NMP fixes the subjective time evolution inside each self.
> Quantum measurement occurs and corresponds to measurement of density
> matrix for some subsystem of self (or its complement inside self, there is
> complete symmetry). The happy subsystem is the one for which
> entanglement negentropy gain is largest in quantum jump.
> Entanglement negentropy goes to zero in quantum jump.
> Its p-adic counterpart is given by SUM(n)p(n)Log_p(p_n)
> where p(n) are entanglement probabilities. Log_p(x) is
> the integer valued p-adic counterpart of logarithm function.
> Real counterpart S_R of S is obtained by canonical identification
> and it is S_R which is the largest one for the winner of NM race.
The symmetry that I am thinking of is that that see would see when we
consider all subsets of the Universe as having "selves". This symmetry
is not "perfect" as we are usually dealing with a one to many situation.
The maximizing of "entanglement negentropy" is, I think, that I am
talking about when I say that the interaction of LSs is a (quantum)
computation. The bisimulation model involves this notion, since the pair
participants are acting to predict each others behavior, thus the term
bi-simulation...
The fact that what there is a relationship between the "amount" of
information simulated and the "amount" of free energy dissipated by the
simulation. I see this "race" as a tournament occurring between the
"possible" subsystems. I have a hunch that the periodic gossiping model
in graph theory is equivalent to the tournament and it gives us a way of
modeling this behavior in networks of LSs.
(http://www.cs.wvu.edu/~chif/cs418/1.html)
> > [SPK]
> > > >If any classical system's behavior can be modeled by a UTM and a UTM
> > > > can be implemented in a classical system, I see a connection between
> > > > time and computation! (Noting that this particular model assumes unique
> > > > initiality conditions) Now that we have good models of quantum
> > > > computation, such as Peter's, we can understand better how time works in
> > > > a quantum context... It is just a model, a way of thinking after all...
> > > >
> > >
> > > [MP]
> > > Problems are caused by noncomputability in classical context.
> >
> > Could you explain?
> >
> [MP] I referred to the fact that most classical systems are chaotic and
> one cannot calculate the predictions in finite time. I think you mentioned
> the case in which simple linear wave equation fails to be predictable
> by computing.
What you are pointing out is that it is impossible to compute the
behavior "faster" than it takes "in real time". Ah, but this is what I
use as an axiom! The key notion is that all behaviors of the subsets of
the Universe are computational acts in themselves (bisimulations of the
LSs).
I am suggesting that what we observe as the "flow of time" *is* the
Universe's simulation of "what is it like for the behavior to be
experienced"! The failure of the computation of a simple linear wave
equation to be *exactly* predictive is, I believe, why we must use
statistical inference!
The way that Hitoshi's model sets up a subject-object relationship that
is defined in terms of quantum - inside and classical - outside lends
itself beautifully to this notion. That must be understood of my
hypothesis, is that the "particulars" of experience, the set of "what is
is like to be X" are not a priori given (synthetics). What *is a priori*
is the Universe of all possible objects. Umm, the word "object" here is
pretentious since the term tacitly implies that such have definite
properties in themselves. This is not the case in my thinking.
I am saying that the subsets of the Universe, its parts are Local
Systems and have dualistic aspects. The interactions among them
generates a "third", so my model is compatible with your tripartite
idea. Umm, it would help if we used the formalism of Category Theory to
discuss this, but I am a bit weak in my knowledge of it. :-( Is there
any one in the Time List group that is familiar with Category theory?
> > > >
> > [SPK]
> > > > Let me see... "Any finite sequence [of symbols] can be unambiguously
> > > > coded in binary (or decimal) and thus corresponds exactly to some
> > > > rational number." What I am making noise about is that "symbols" are
> > > > matter codings of information. (Without matter it would be impossible to
> > > > make records, thus disproving Idealism...) Umm, another thing about
> > > > Calude's Lexicons are like programs, they need material configurations
> > > > to be read, e.g. reading heads - tape, etc.
> > >
> > > [MP] This is perfectly understandable if symbols carry finite bits of
> > > information to us. What fascinates me that characterization
> > > of quantum jump by bit sequence, integer leads to characterization
> > > of infinite sequence of quantum jumps by real and since most
> > > reals are lexicons this means that every quantum jump (classified
> > > in pinary resolution) appears infinitely many times in sequence.
> > > This means complete information about all possible quantum jumps.
> > > The infinite p systems with infinitely long subjective memory
> > > have all the data needed to build a physical theory and communicate
> > > it to us(;-)!
[SPK]
> > Can you think of why it is that "symbols [would] carry finite bits of
> > information to us"?
> >
> [MP] I see this as follows. The letters which I read now, wake-up
> sub-selves inside me. These subselves I experience thoughts with meaning
> and information. The only problem is to establish standard rules. Standard
> sensory input wakes-up stand subself. [Note that though is just lower
> level thinker in this picture: economical!]
Sure, but that is a very anthropic way of putting it. We need to be
more secular! I believe that the work of Barwise and Seligman's
Information Flow is very helpful toward this end! (and guess what, they
use a generalization of Chu spaces!)
> It is somewhat difficult to say how many bits of information single letter
> actually contains. It can be coded to recognizable pattern using finite
> number of binary digits in computer memory but the number of digits
> needed to code alphabet depends only on the number of letters in alphabet:
> it does not tell about real information.
And how exactly do we figure out "how many bits of information a single
letter *actually contains*"? All we have are the Shannon and Fisher
concepts... We forget that what is meaningful information to one is not
necessarily to another! This is why we need to think about this very
carefully! How many possible alphabets *exist*? How many ways are
possible to render or encode them in matter? Infinite! So we are dealing
with infinite equivalence classes that, I say (with Pratt), are dual to
each other. It is when we look carefully at how the dynamic behavior of
each equivalence class act to select particular elements from the other
that we see residuation at work!
I believe that the cardinality of these equivalence classes is
"undecidable", but I do not know how to prove this rigorously. :-( My
clues come from the work of Calude and Chiatin...
> [People believing in I Chin migh be right: perhaps higher level
> selves are desperately working to establish standard language
> with which to communicate with us. Linguistic conventions are
> always randomly choosen and the problem is that we, as rational
> westerners, simply cannot take seriously that some random rules
> would make sense.(;-) ]
Why should these "higher selves" be concerned with us? Are we wholly
concerned with the individual workings of the cells that make up our
bodies? It is obvious that there is a communication going on between the
"selves" in the hierarchy, but it would best be categorized as
"unconscious".
snip
> > [MP]
> > > Chalmers represented in his books arguments stating that the basic
> > > problem of dualism is that consistency with the determinism of physics
> > > leads to conclusion that mind is one-one image of matter and can be
> > > eliminated as un-necessary so that one has materialistic
> > > monism. Or that in interactive dualism there is no
> > > basic difference between matter and mind like degrees of freedom
> > > and one could quite well call them just matterlike. Have you studied
> > > these arguments?
> >
> > [SPK]
> > > > ...I do not understand Sarfatti's version of dualism.
> > >
> > > [MP] The idea is to regard some fields as matter like and some fields
> > > as mind like obeying determistic dynamics. The dynamics of matter fields
> > > looks nondeterministic if one forgets the presence of mind like fields.
> > > The problem is that the decomposition into matter and mind like fields
> > > is completely ad hoc.
> > >
> > > Pilot wave would be mind like field and classical particles would
> > > represent matter.
> >
> > This looks like a material monism to me! Ad hoc indeed!
> >
> [MP] The basic objection against dualism consistent with physics is
> that it reduces to material monism. Determinism tends to lead
> to the conclusion that mental is mere mirror image of material.
> Be cautious, this objection is dangerous creature!
Yes, I need to be more specific! I question how Sarfatti's "pilot wave"
is dual to "classical particles" and how they interact. I am positing
(with Pratt) that the behavior of information is dual to the behavior of
matter. The "interaction" is the same as that of how information is
"signaled" by a particular configuration of matter, say the symbol
"tree".
We just take it for granted in the usual models of physics! To say that
the "mental is mere[ly a] mirror image of [the] material" is avoiding
the problem. This is the usual hand waving refutation of monists... So,
I agree with your characterization! When we look closely at the notion
of determinism we see very serious problems! The inability to define
Cauchy hypersurfaces (e.g. unique initial conditions) makes the concept
of absolute determinism collapse! We see that the classical equations of
motion are mere "cartoons".
With Pratt's model we have a way to interpret the relationship between
material causality = time (thermodynamic entropy) and logical causality
= logical chaining of inference. We can invert the relationships (Mind &
Body) and recover the same relations! I suspect that the fermi-bose
duality of supersymmetry is closely related!
snip
> > [SPK]
> > What if every system in the ensemble has "experiences", each just
> > slightly different...
>
> [MP]
> In this case self experiencse all of them as separate. Next level self
> experiences their average. In retina it seems necessary to assume
> that color experiences of rods and cones are statistically averaged
This "experiencing their average" is what I understand as the "center
of mass" that Hitoshi identifies with the "outside" of an LS. What I
think is important is that the averaging process is not simple! We need
to look carefully at how weights (projection, orthogonality,
non-Hausdorffness, etc.) and commutator properties (Fermi vs. Bose
statistics) play their roles!
Onward,
Stephen
This archive was generated by hypermail 2.0b3 on Sat Oct 16 1999 - 00:36:30 JST