Pre-Quantum theory talk at PI, Wednesday 11 October

[marcus]

I just checked the Perimeter site and Smolin had posted the title of his Wednesday talk

===quote===
Speaker: Lee Smolin
Title: Could quantum mechanics be an approximation to another, cosmological, theory?
Date: Wednesday October 11, 2006, 2:00 PM
Abstract: We consider the hypothesis that quantum mechanics is an approximation to another, cosmological theory, [and that the approximation is] accurate only for the description of subsystems of the universe. Quantum theory is then to be derived from the cosmological theory by averaging over variables which are not internal to the subsystem, which may be considered non-local hidden variables. I will explain the motivation for this view, give some examples of theories of this kind and investigate general conditions for such an approach to succeed.

===endquote===
http://www.perimeterinstitute.ca/activities/scientific/seminarseries/alltalks.cfm?CurrentPage=1&SeminarID=825

that means he will be illustrating and explaining stuff from a paper he just posted on arxiv.org.

http://arxiv.org/abs/quant-ph/0609109
Could quantum mechanics be an approximation to another theory?
Lee Smolin
10 pages

"We consider the hypothesis that quantum mechanics is an approximation to another, cosmological theory, accurate only for the description of subsystems of the universe. Quantum theory is then to be derived from the cosmological theory by averaging over variables which are not internal to the subsystem, which may be considered non-local hidden variables. We find conditions for arriving at quantum mechanics through such a procedure..."OK, so the idea is that Quantum Mechanics might be NOT FUNDAMENTAL but instead might be a superficial approximation of a deeper theory.
Quantum Mechanics might apply only to SMALL PIECES of the universe, where intuitively one can put a box around it and look at it as an outsider observer.

To the extent that Quantum Mechanics seems coo-coo, it seems that because you can't really put the box around----there are always some outside partners that seemingly shouldn't be connected to events inside, but are anyway. the T-shirt term for it is "disordered locality".

 

[marcus]

With luck, the talk tomorrow will be available video at the streamer site

the streaming media resources are posted here
http://streamer.perimeterinstitute.ca:81/mediasite/viewer/
in the left menu, click on "seminar series" to get the list of seminar talks
which have been posted.

there was a 1912 silent movie by the famous director D.W. Griffith called
http://www.imdb.com/title/tt0002497/
So near, yet so far

this is a efficient elegant way to get the equivalent of "hidden" variables
just by graph-theory modeling of the expansion of space.
it has to happen, in fact, unless you take pains to rule it out

if you start with a random-connected graph and do appropriate expansion/reconnection moves then you get some emergent approximate 3D space with a RESIDUE of surprise adjacencies----this just happens of its own accord and you need to put in some extra structure by hand if you want to AVOID a sprinkling of long-distance adjacencies

And maybe you shouldn't want to rule them out! It is a bother to rule them out, and if you leave them in then you get the moral equivalent of "hidden" variables, for FREE.

The 1912 movie SO NEAR, YET SO FAR had Mary Pickford and Lionel Barrymore,
and Lillian Gish too---and since it was a serious feature-length movie of its day, it lasted a full 17 minutes.

 

[marcus]

there is a quote from this paper we could consider:

http://arxiv.org/abs/quant-ph/0609109
Could quantum mechanics be an approximation to another theory?

bear in mind that reference [7] is to FINKEL's paper in which the expansion of space is seen to lead to nodes which are adjacent "mates" to a given node but which, in terms of the emergent macro metric, are extremely FAR from it. this comes naturally from graphtheoretical modeling of expansion of space by local moves. We can call these extemely distant immediate neighbors Finkel far-mates for lack of better term.

The quote from Smolin i want us to consider is from page 2 and it goes
==quote==
In fact, there is independent motivation for believing that non-local interactions arise from microscopic theories of spacetime These arise from the observation that locality can disordered in quantumstates of geometry in loop quantum gravity and similar theories[5, 6, 7]. This is possible because in such theories the classical metric is an emergent, approximate quantity, meaningful only at low energies. As a result, there can be a mismatch between the notion of locality which controls the dynamics of the fundamental microscopic theory and a different notion of locality that follows from the emergent metric which controls the approximate, low energy, dynamics. This can happen in both quantum and stochastic dynamics of discrete geometries[7]. Given the latter ... it is possible to conjecture that a stochastic theory of pre-geometry is the cosmological theory which quantum theory approximates for subsystems. ...
==endquote==

to repeat a key part of the conjecture:
" ...a stochastic theory of pre-geometry is the cosmological theory which quantum theory approximates ..."the Finkel far-mates are the "hidden variables" which create the illusion of quantum mechanics

 

[wolram]

I have an idea that pre geometry is another word for back ground, this
may be far from the touch line.

 

[marcus]

I have an idea that pre geometry is another word for back ground, this
may be far from the touch line.

that's a perceptive way of putting it. Let's make a distinction between two different uses of the word "background"----restrictive and unrestrictive

In the restrictive sense, a background is a shorthand expression for a "fixed pseudo-Riemannian metric on a vintage-1850 continuum".

This is the kind of antique rigid background that particle physicists are accustomed to and which string uses, in a version with curled extra dimensions.

And this is the kind of fixed background geometry that non-string QG people are interested in being INDEPENDENT of.

After all, Einstein's 1915 Gen Rel theory already had that kind of Background Independence---it was not committed to a having some preconceived metric or distance-function.

===============

in the unrestrictive sense, you could say the background is just the fundamental degrees of freedom used to model spacetime---and it can be completely free from preconceived geometry. it does not have to be vintage-1850 continuum (a Riemannian or pseudoRiemannian manifold) equipped with a lot of gear giving it rigid structure. it's shape can be dynamic or it doesn't even have to be a 1850s-style continuum: it couild just be a set of points with some more primitive form of structure.

in the unrestrictive sense, a background can be dynamic, with no prior commitments to some particular shape----and it can have the absolute bare minimum of structure, for instance a set of points with nothing but a list saying which points are adjacent to which others---basically that is what a graph or network is.

having that kind of unrestrictive background (for example: one without even a preconceived notion of distance, where even the idea of distance has to emerge as the story unfolds) is what QG folk call being background independent

 

[Mike2]

The quote from Smolin i want us to consider is from page 2 and it goes
==quote==
In fact, there is independent motivation for believing that non-local interactions arise from microscopic theories of spacetime These arise from the observation that locality can disordered in quantumstates of geometry in loop quantum gravity and similar theories[5, 6, 7]. This is possible because in such theories the classical metric is an emergent, approximate quantity, meaningful only at low energies. As a result, there can be a mismatch between the notion of locality which controls the dynamics of the fundamental microscopic theory and a different notion of locality that follows from the emergent metric which controls the approximate, low energy, dynamics. This can happen in both quantum and stochastic dynamics of discrete geometries[7]. Given the latter ... it is possible to conjecture that a stochastic theory of pre-geometry is the cosmological theory which quantum theory approximates for subsystems. ...
==endquote==

It seems that the holographic principle would be applicable here as well. Here the entropy of the cosmological event horizon (a global property) is linked to a horizon temperature within (a local property). The statistical property of the boundary is linked to the quantum property of particle creation. There does seem to be an assumption that the expansion of space is homogenous and isotropic - at least on the average. Perhaps bits of spacetime pop into existence on a random basis just like the particles would. I wonder if you'd have bits of spacetime also poping out of existence like virtual particles do, with an average more towards creation than annihilation? Can you have space and antispace annihilating each other?

Since spacetime curves around matter/energy, how then it would seem that matter effects the creation of spacetime/antispacetime(?). But that might mean that the vacuum energy of particles pairs is also effected by matter. If the vacuum energy of spacetime/matter quantum field were effected by permanent matter, then perhaps there is more vacuum energy around matter, and this might be dark matter. And if quantum spacetime bit production was effected around matter, then perhaps that explains spacetime curvature around matter. Perhaps a spacetime/matter QFT could be built around these assumptions without having to go through the LQG exercise. That's all the blithering I have for now, thank you very much.

 

[marcus]

Hi Mike, thanks for the comment. Before I forget, I should copy in the succinct argument that motivates the present paper.
One should be on the lookout for a non-local cosmological pre-quantum theory, suggests Smolin, because of the following considerations (given in the Introduction section).

===quote===

In spite of much progress clarifying foundational issues in quantum mechanics, there remains persistent evidence that quantum mechanics is an approximation to a deeper theory.

Among the reasons for this belief are;

• The unresolved difficulties in extending quantum theory to cosmology. If this cannot be done then one possible explanation is that quantum mechanics does not in fact extend to the whole universe. It must then be an approximation (which applies only for small subsystems of the universe) to a more fundamental cosmological theory.

•The difficulties in solving the measurement problem in the context of a theory with a realistic ontology.

•The success of quantum information theory, which reinforces the viewpoint that the quantum state represents the information that observers have of a system.

•The experimental evidence against the Bell inequalities tells us that any theory quantum mechanics is derived from must be non-local. It is then natural to hypothesize that this non-local theory is a cosmological theory, which is more adequate than quantum mechanics for the investigation of cosmological problems. These assertions raise issues about which there is disagreement and lively discussion. However, apart from taking the fact that these issues are unresolved as motivation, the aim of this paper is not to address that debate directly. Instead, we ask whether there are conditions that must be satisfied by any non-local, cosmological theory, so that quantum mechanics can be derived from it as an approximate description of subsystems.

===endquote===

remember this is MOTIVATION FOR A CONJECTURE, and people will state conjectures to get other people to help examine them, without necessarily believing. so we should not already go and start BELIEVING in this conjecture, but we can study it and consider possible tests and consequences

for one thing, the conjecture that there is a more fundamental NON-quantum theory that can cope with the whole universe (whereas quantum theory seems only to work for an isolated part of the world inside a box) MAY BE AT ODDS with some very nice ideas of John Baez that we were discussing here. Remember "star categories" and the application of higher algebra to quantum gravity? In that discussion it was revealed that the categories Hilb and Cob have remarkable similarities---the categories underlying Quantum Mechanics and General Relativity may be twin siblings, and things that appear puzzling in the case of one may seem natural in the case of the other. The general tenor of JB discussion was that we can get a deeper understanding of QM and why it seems to work by connecting it to a deeper understanding of spacetime.
Maybe the point where all the seeming paradoxes or puzzles go away is the moment where you understand QM and the fundamental degrees of freedom describing spacetime as the same thing.

Well that approach is very appealing and interesting, but it seems on the face of it immediately incompatible with the idea of a PRE-QUANTUM theory which is based on a model of the cosmos that isn't even quantum mechanical! Also this pre-quantum conjecture---which remember is only a conjecture---is very primitive. It is nuts-and-bolts and old-fashioned---it does not have elegant new algebra. Instead of category theory, it has Hamiltonians and Brownian Motion.

So we have to keep poised between these unmixable conjectures.

But for the time being I am imagining that the universe expanded from a random-connected graph (by local moves) and gradually smoothed out to an APPARENT simple 4D orderly thing but remained infested with these unexpected adjacencies which function only in microscopic extremis and join our familiar nodes with what we can call their Finkel far-mates. And by AVERAGING OVER THE WHOLE REST OF THE UNIVERSE we somehow induce uncertainty in our local affairs---producing the illusion of quantum mechanics.

 

[marcus]

Day before yesterday's seminar talk is now posted video!

click on "seminar series" in the left hand menu here
http://streamer.perimeterinstitute.ca:81/mediasite/viewer/FrontEnd/Front.aspx?&shouldResize=False

"Could quantum mechanics be an approximation to another, cosmological, theory?"

Wednesday 11 October at perimeter

 

[wolram]

I know i can not explain what i need to very well but i will try with this question.
Was the oscillation pre, pre geometry or did pre geometry (create) the oscilation, or what did start the elemental oscilations we see in every thing?

 

[marcus]

Hah! as usual you are way ahead of everybody else, wolram

including Smolin.

maybe you can help me with finding the right word: "pre" is not quite right.

Smolin conjectures the existence of a deeper "theory to which quantum mechanics is an approximation"

the best analogy I can think of is that QM is like NEWTONIAN MECHANICS (which is a handy superficial approximation to something more accurate and fundamental).

In the analogous case of Newtonian mechanics, there is a deeper level of theory made up of special and general relativity (among other stuff) that explains better how things move
and to which Newtonian mechanics with its inversesquare law is just a useful approximation which works well in familiar solar system and cannonball trajectories and pendulums and such.

So "pre" was a misleading word. I don't mean that this new theory he is conjecturing COMES BEFORE IN TIME. I mean that if it actually exists and is valid it is a MORE ACCURATE theory from which QM derives as a useful approximation.

Should I call it a "sub-Quantum" theory, because it is deeper and more fundamental?

I think of it as *ancestral*----the more fundamental picture of the world OUT OF WHICH quantum mechanics grows, or on which it rests.

maybe I should call it the "root-Quantum" theory---abbreviated "root-Q"

or simply the root theory

 

[wolram]

may be th (turtle egg) theory

 

[marcus]

and yes, I know I did not answer your question.

if there is a more fundamental theory, which avoids some of the chronic problems of QM, and to which QM is an approximation

then it will make predictions and will eventually be checked and possibly falsified, or it will pass the test and be cautiously credited,
and it MAY OR MAY NOT support the idea of a big bounce----the idea you get in Loop Quantum Cosmology. it is too early to speculate about what the beginning of expansion (14 billion years ago) might have looked like according to a new theory.

so, sorry wolram, I can't think of any words on the topic of beginnings---as you say, might as well be turtle egg

 

[wolram]

I have a theory that one can not have a rassberry ripple without the rassberry, or a mexican wave without the mexican, such nonsence,
how can the mill pond ripple with nothing to perturb it? even if the mill pond exists yet? from this i hold that whatever we came came from is two not one, now for the maths-----------