The Third Road to Quantum Gravity

  • #31
A recent paper that may be of some interest:

Fibered Manifolds, Natural Bundles, Structured Sets, G-Sets and all that: The Hole Story from Space Time to Elementary Particles
J. Stachel, M. Iftime
http://xxx.sf.nchc.gov.tw/abs/gr-qc/0505138

They don't get into any heavy category theory. Stachel is a well known Einstein historian. From the conclusion:

"Therefore, the following principle of generalized covariance should be a requirement on any fundamental theory: the theory should be invariant under all permutations of the basic elements out of which the theory is constructed.

Perturbative string theory fails this test..."
 
Last edited by a moderator:
Physics news on Phys.org
  • #32
a little bit of history...

If numbers, powers, the mathematical infinite, and suchlike are to be used not as symbols but as forms for philosophical determinations and hence themselves as philosophical forms, then it would be necessary first of all to demonstrate their philosophical meaning, i.e. the specific nature of their Notion. If this is done, then they themselves are superfluous designations; the determinateness of the Notion specifies its own self and its specification alone is the correct and fitting designation. The use of those forms is, therefore, nothing more than a convenient means of evading the task of grasping the determinations of the Notion, of specifying and of justifying them.

...

It has therefore been freely admitted that the cognition that stops short at the Notion purely as such, is still incomplete and has only as yet arrived at abstract truth. But its incompleteness does not lie in its lack of that presumptive reality given in feeling and intuition but rather in the fact that the Notion has not yet given itself a reality of its own, a reality produced from its own resources.

...

Continuous and discrete magnitude can be regarded as species of quantity, provided that magnitude is posited, not under any external determinateness, but under the determinatenesses of its own moments; the ordinary transition from genus to species allows external characteristics to be attributed to the former according to some external basis of classification. And besides, continuous and discrete magnitude are not yet quanta; they are only quantity itself in each of its two forms. They are perhaps, called magnitudes in so far as they have in common with quantum simply this: to be a determinateness in quantity.

The Science of Logic, Hegel (1770-1831)

Apparently Hegel is undergoing a bit of a renaissance in Philosophy. I find this fascinating. In the latter 19th century and later he was not particularly admired by Westerners, partly I guess because of the association with Marx.

Does anyone know who coined the term quantum in its physics context?
 
Last edited:
  • #33
Kea said:
[.

Does anyone know who coined the term quantum in its physics context?

I read somewhere that it was Planck. The source remarked that during Planck's student days, (this would have been the 1860s), "quantum" was the University slang word for an indefinite chunk or slab, as in "Pass me a quantum of cheese."
 
  • #34
selfAdjoint said:
The source remarked that during Planck's student days... "quantum" was the University slang word for an indefinite chunk or slab...

Hi selfAdjoint

Interesting. Most sources seem to attribute it to Planck's 1900 paper, but they claim that he borrowed the term directly from the Greek. It's interesting to see that there may be a more sophisticated evolution of the term throughout the 19th century - and not surprising really. Humans have a bad habit of treating their ancestors like complete ignoramuses.

Cheers
Kea :smile:
 
  • #35
to Urs

Urs asked me about Non-standard analysis and topos theory over on Woit's blog. I just noticed the question because the discussion has gone onto the archive now. Non-standard analysis is definitely not my subject, but I do remember Ross Street talking about the Robinson topos (don't google that - you won't find anything), and I found this paper

http://www.math.ucla.edu/~asl/bsl/0403-toc.htm (the first paper)

by Palmgren on constructivism and Non-standard analysis. Maybe you'll find this interesting, if you're about Urs.

Must go. All the best
Kea :smile:
 
  • #36
Robinson Topos

Without worrying about what it is, if \mathbf{E} is the Robinson topos then there is a functor

\mathbf{E} \rightarrow \mathbf{Set}

such that the image of the 'reals in \mathbf{E}' is the set of non-standard reals including the infinitesimals.

Ross Street says we should take non-standard analysis to be the study of the reals in \mathbf{E} rather than the study of the more contrived 'reals plus infinitesimals' in the usual topos \mathbf{Set}.
 
  • #37
Urs Schreiber's Dissertatiom

Kea have you seen the draft of Urs' http://www-stud.uni-essen.de/~sb0264/SchreiberDissPartI.pdf ? Loop space, categorification, gerbes and all. Urs has been discussing these topics here and there for some time, but this now is an impressive construction, and all in Urs's ultra-clear explanative style, too!
 
Last edited by a moderator:
  • #38
selfAdjoint said:
Kea have you seen the draft of Urs' dissertation?

Hi selfAdjoint

Yes, he mentioned it on the String Coffee Table, amongst other forums. Thanks for posting it. Part I, which is supposedly the outline, is 90 or so pages. John Baez will be talking about this subject at the Streetfest http://streetfest.maths.mq.edu.au/ in July.

Cheers
Kea :smile:
 
  • #39
the real importance of the Large Hadron Collider

setAI said:
I have been thinking lately about where the search for Quantum Gravity may be headed in the near future and it has struck me that the LHC is going to be a major pivot point in the future of research- if they are able to observe micro black-holes with the LHC- it seems to me that such a tremendous achievement will capture the imaginations of everyone [...]

Sure. But, they won't find tiny black holes at the LHC.

The only reason anybody talks about this possibility is that some string theorists made up a far-out theory where quantum gravity effects could show up at a more or less arbitrary energy scale, and then - for publicity reasons - picked an energy scale slightly bigger than what anyone has been able to study so far, to get people excited about discovering tiny black holes at the LHC. It might happen... but it won't.

The real importance of the LHC is that string theorists have been saying for years that supersymmetry is right around the corner, detectable at an energy scale slightly bigger than what anyone has been able to study so far. (Notice a pattern?) And, lots of them claim that some effects of supersymmetry will be detected at the LHC. They might be... but we'll see.

If they're not, government funding for string theory will drop, so interest in other approaches to quantum gravity will increase.

If they are, string theory will have some actual data to chew on, and progress should accelerate.

Oh, and the Higgs. If that works as expected, the Standard Model will be confirmed - great, but ho hum. If not, things will get really exciting.
 
  • #40
Robinson topos

Kea said:
Without worrying about what it is, if \mathbf{E} is the Robinson topos then there is a functor

\mathbf{E} \rightarrow \mathbf{Set}

such that the image of the 'reals in \mathbf{E}' is the set of non-standard reals including the infinitesimals.

Ross Street says we should take non-standard analysis to be the study of the reals in \mathbf{E} rather than the study of the more contrived 'reals plus infinitesimals' in the usual topos \mathbf{Set}.

Where can one read about the Robinson topos?

The book Synthetic Differential Geometry talks about a number of topoi with infinitesimals, following Lawvere's ideas on differential geometry (which I sketched in week200 ). But, I haven't looked at this for a long time, so I don't know exactly which topoi they consider.

Personally I don't think any of these different approaches to calculus are sufficiently different to be worth worrying about, unless one is fascinated in them for their own sake. Things would be very different if one could make real progress on some hard math or physics problems in one of these alternative approaches. People have tried - I've read papers about quantum field theory that use infinitesimals - but nothing much has some of this so far.
 
  • #41
john baez said:
Where can one read about the Robinson topos?

Good question. I don't know. I heard about it from Ross Street. Apparently Princeton University Press have reissued Robinson's original 1960's book Non-Standard Analysis but I haven't yet seen it.

If one plays around with Google one can find licorice allsorts, such as
http://arxiv.org/PS_cache/quant-ph/pdf/0303/0303089.pdf
with an interesting list of references...but there are only so many hours in the day! I find myself wandering over to the Philosophy library sometimes for classic papers by topos theorists.

Kea :smile:
 
Last edited by a moderator:
  • #42
Robinson topos

Kea said:
Good question. I don't know. I heard about it from Ross Street. Apparently Princeton University Press have reissued Robinson's original 1960's book Non-Standard Analysis but I haven't yet seen it.

I'm pretty sure Robinson didn't think he was inventing a topos; I think he was working solidly within a more old-fashioned tradition in logic.

There's a similar example in Sheaves, Geometry and Logic by Mac Lane and Moerdijk. Cohen proved the independence of the axiom of choice using a technique called "forcing" to create nonstandard models of the Zermelo-Fraenkel axioms. Moerdijk and Mac Lane simplify the idea behind this by constructing a topos like the usual topos of sets, but in which the axiom of choice fails. I've never understood quite how close their construction comes to Cohen's original result - I don't think it instantly implies his result.

Maybe Street is just smart enough to realize that Robinson, like Cohen, was also subconsciously creating a new topos.

But, we're drifting from physics here. And, as you point out,

...there are only so many hours in the day!
 
  • #43
Kea said:
Good question. I don't know. I heard about it from Ross Street. Apparently Princeton University Press have reissued Robinson's original 1960's book Non-Standard Analysis but I haven't yet seen it.
...

I actually bought Robinson's book in the (notorious) late 1960s and probably still have it in one of the boxes up in the attic----unless it has been donated to some library.

At least my impression of it was that it is grounded in old fashioned logic and focused on old fashioned analysis: calculus, the Reals... In line with what JB says. I read (in) it with a mixture of hope and disappointment, but didnt see how to take it anywhere.

It is interesting that Princeton UP has reissued it and that Ross Street has been talking about it.

Anais Nin has a wonderful passage that starts "Nothing is lost, but it changes..." almost a poem.
 
  • #44
john baez said:
Maybe Street is just smart enough to realize that Robinson, like Cohen, was also subconsciously creating a new topos.

It's possible that the detailed definition is due to Ross Street, perhaps unpublished. He went into the details: let me reproduce a little...

Let Ev be the topos of evolving sets, that is the functor category [ N , \mathbf{Set} ] from the ordinals into the topos Set. This is the topos that Markopoulou studied as a Newtonian causal set theory. The terminal object is the set sequence of one point sets.

Now one needs the notion of an ultrafilter \nabla. Firstly, a (proper) filter on a Heyting algebra is a collection of subobjects ( for sets see http://mathworld.wolfram.com/Filter.html ) such that

1. 1 \in \nabla and 0 is not in \nabla

2. x , y \in \nabla \Rightarrow x \wedge y \in \nabla

3. x \leq y , x \in \nabla \Rightarrow y \in \nabla

One of the characterisations of an ultrafilter is that the quotient of the Heyting algebra by \nabla is isomorphic to 2, as a lattice. It turns out that for an ultrafilter, and a topos E, then \mathbf{E} \backslash \nabla is a 2-valued topos (the terminal has 2 subobjects).

What is \mathbf{E} \backslash \nabla ? It has the same objects as E. The hom set (A,B) is a colimit (over U \in \nabla) of (A \times U, B).

This idea is used on Ev to define the Robinson topos. \nabla is something called a non-principal ultrafilter on N (doesn't have a least element). The Robinson topos is an example of an elementary topos that is not a Grothendieck topos.

Kea :smile:
 
Last edited:
  • #45
Kea said:
A recent paper that may be of some interest:

Fibered Manifolds, Natural Bundles, Structured Sets, G-Sets and all that: The Hole Story from Space Time to Elementary Particles
J. Stachel, M. Iftime
http://xxx.sf.nchc.gov.tw/abs/gr-qc/0505138

They don't get into any heavy category theory. Stachel is a well known Einstein historian. From the conclusion:

"Therefore, the following principle of generalized covariance should be a requirement on any fundamental theory: the theory should be invariant under all permutations of the basic elements out of which the theory is constructed.

Perturbative string theory fails this test..."
If it's any consolation [or perhaps a curse:smile:] I am in complete agreement.
 
Last edited by a moderator:
  • #46
A new third way paper

This paper on http://www.arxiv.org/PS_cache/gr-qc/pdf/0509/0509089.pdf just appeared on the arxiv. After a lot of entertaining generalities on iconoclasm in science, he gets down to an account of Absolute Differential Calculus (ADC) a sheaf-based theory and its approach to QG. A tangy Greek salad; enjoy!
 
Last edited by a moderator:
  • #47
selfAdjoint said:
This paper on http://www.arxiv.org/PS_cache/gr-qc/pdf/0509/0509089.pdf just appeared on the arxiv.

Thanks, selfAdjoint. I see it's by I. Raptis. He's one of few people who seem to have been looking at toposes in physics for quite a while now. One of the Isham school, I think.

Must finish pulling the splinters out before I read it...

Right. That's done. Mmmm. Raptis is rather enthusiastic about the Mallios approach. He mentions its possible connection to Category Theory approaches but he doesn't seem to have learned much about categories yet. For instance, he talks about algebra replacing geometry when all category theorists know that categories do both.

Of course, I enjoyed reading it. Anyone that throws Hegel, Prometheus and Einstein into the same blender is probably a buddy of mine.

:smile:
 
Last edited by a moderator:
  • #48
...and the correct form of the proverb is

nothing venture, nothing gain
 
  • #49
Kea said:
...and the correct form of the proverb is

nothing venture, nothing gain


Probably just a typo, but it's Nothing ventured, nothing gained.
 
  • #50
selfAdjoint said:
...it's Nothing ventured, nothing gained.

No! It's not. It's a 14th century English proverb, originally from the French. It states Nothing venture, nothing gain.

http://www.worldofquotes.com/proverb/French/18/

I learned this when Edmund Hillary's autobiography came out a couple of years ago. :smile:
 
  • #52
Hot off the presses!

Geometry from quantum particles

From: David Kribs
Date: Tue, 11 Oct 2005 02:18:17 GMT (19kb)

We investigate the possibility that a background independent quantum theory of gravity is not a theory of quantum geometry. We provide a way for global spacetime symmetries to emerge from a background independent theory without geometry. In this, we use a quantum information theoretic formulation of quantum gravity and the method of noiseless subsystems in quantum error correction. This is also a method that can extract particles from a quantum geometric theory such as a spin foam model.

http://www.arxiv.org/abs/gr-qc/0510052
 
  • #53
setAI said:
Geometry from quantum particles

http://www.arxiv.org/abs/gr-qc/0510052

This paper by Fotini Markopoulou and David Kribs was one of those listed by Smolin today, in a post mentioning some highlights from the Loops '05 conference so far.

I copied Smolin's list of QG advances here:
https://www.physicsforums.com/showthread.php?p=784856#post784856

The Kribs/Markopoulou paper is #4 in a list of 7 that he highlighted. (And still two days more to go in the conference!)

To see the Smolin's post that I exerpted in context, scroll down to comment #5 here:
http://www.math.columbia.edu/~woit/wordpress/?p=279#comments
 
  • #54
also today ...

Spacetime topology from the tomographic histories approach II: Relativistic Case
I. Raptis, P. Wallden, R. R. Zapatrin
21 pages
http://www.arxiv.org/abs/gr-qc/0510053

One might think this subject is becoming a bit more popular!
 
  • #56
Peirce's Existential Graphs

Over a cup of coffee recently, mccrone was telling me about the semiotics of Charles Sanders Peirce and how it fitted into a modern context of biological thinking which he was sure was of great importance to physics. I think he's probably right. In return, I naturally tried to convince him that Category Theory was the right modern language to discuss these sorts of things. Anyway, the conversation prompted me to do one of those things that is always on the to do list about half way down the page: go to the philosophy library and get out the collected works of Charles Sanders Peirce. When I saw how many volumes there were I modifed this resolution and chose just a few, including the wonderful reference 3 (see below) which became my introduction to Peirce's Existential Graphs.

Naturally, Louis Kauffman has already written a beautiful article on this subject (reference 2).

Hopefully you will know by now that diagrammatic techniques are endemic to categorical computation. What Peirce did was develop a surface diagram notation for basic logic. So for braided monoidal categories we have knots, and for logic we have Existential Graphs. Moreover, he did this over 100 years ago!

For example, how does one express the notion of not X? If X is a symbol on a page, one simply draws a circle around it. This cuts X off from anything else on the page. Two rings, one inside the other, act as an identity (this is Boolean logic). The identity can be deformed so that the two circles are joined at a point...and this naturally looks like one loop with a kink in it.
Conjunction of two terms X and Y is represented by simply writing them both down, with no extra symbols. The empty picture is the statement true. Exercise: what is the diagram for false?

This all fits into a fantastical philosophical scheme...but must go now.:smile:

References:

1. Nice webpage: http://www.clas.ufl.edu/users/jzeman/

2. L. H. Kauffman The Mathematics of Charles Sanders Peirce
in Cyber. Human Know. 8 (2001) 79-110, available at
http://www.math.uic.edu/~kauffman/Papers.html

3. Semiotic and Significs: The correspondence between C. S. Peirce and
Victoria Lady Welby
ed. C. S. Hardwick, Indiana University Press (1977)

4. Collected Papers of Charles Sanders Peirce vol IV,
ed. C. Hartshorne and P. Weiss, Harvard University Press (1933)
 
Last edited:
  • #57
...

Next, Peirce introduces lines of identity. That is, terms X and Y may be joined by a line. A whole lot of terms may be joined by a network of lines. It is OK for the lines to cross the circles. And now, without any further ado, the rabbit appears ... quantification can be expressed without any symbols by saying that if the outer end of a line is enclosed by an even number of circles then the term represents something definite, and if an odd number then anything at all of that type.

I must figure out how all of this can be modifed to quantum logic. We have Coecke et als diagrammatics, but that just comes from monoidal category theory and the logic seems to be a bit of an afterthought. Note that drawing a line of identity from X to X and putting it beside a line of identity from Y to Y is exactly how one represents X \otimes Y in a monoidal category.

Perhaps we could alter Peirce's not not X = X rule and substitute a Heyting not not not X = not X rule, which would be an allowance of deletion of two circles but not the last two.
 
Last edited:
  • #58
...

The law of the excluded middle looks like (from Kauffman)
 

Attachments

  • PFpeirce.jpg
    PFpeirce.jpg
    3.3 KB · Views: 499
  • #59
...

and the proof of this fails in the Heyting case because at the last step the deletion of two circles about the left hand Q is not permitted.
 
  • #60
Hi Kea

Thanks for http://charlotte.ucsd.edu/users/goguen/pps/nel05.pdf and other links on Peirce above. I see reading from this thread that you have also taken in Hegel.

Your theme for this thread is: The third road says "get the logic right, and you'll see how computational the universe is". And the right logic is category theory – it is a general enough theory of logics to “eat” any more particular ones that may have been suggested in the past such as Peirce’s organic/semiotic/triadic approach.

I still have no feel whatsoever for the substance of category theory despite having read a bit more about it now. Perhaps I can provoke you into some jargon-free explanation which gets at its essence.

I understand set theory is based on collections of crisp, discrete, bounded, located, persistent objects. So “atoms with properties”, a mechanical view in which all action and organisation and systemhood is emergent (thus is does not need to be represented at the most fundamental level – the object and its properties).

Category theory seems to take the correct step in saying, no, reality has both locations and motions, stasis and change, form and substance, local and global – the whole gamut of standard metaphysical dichotomies. So to define a basic something, you need both an object and its actions.

A mechanical view here would say that category theory is just accounting for an object and its properties in more distinct fashion. But a holistic or background-independent view points out that all atoms exist in a void. And the void is a thing with properties. The void has crisp spacetime structure. And even the freedoms that the void permits, such as the inertial motions of particles, are essential properties of the void.

So perhaps a more organic view of category theory is that it breaks reality into its most natural dichotomy - that which is semiotically constrained and that which is semiotically not visible, thus free to happen. An object such as a particle (or a void) is produced by a system of self-constraint acting on a ground of pure potential (Peircean vagueness, Anaximander’s apeiron). A particle gains a crisp identity as all the other things it might be become constricted to near impossibility (in simple terms, a cold and expanded Universe steadily robs an electron of its chances to be a quark or tau, etc). But within every system of constraint there are also emergent freedoms. A crisply made particle (that cannot freely transmute and which now has mass and cannot fly at light speed) can now wander about in an “empty” void with weak gravity, in fairly unconstrained inertial fashion.

Peircean logic – as outlined in that Kauffman paper – is seeking to describe a figure~ground breaking in which both figure (object, or atom) and ground (context, or void) are simultaneously developed. This is indeed a background independent approach – or rather it depends on “vagueness” as the unformed, and insubstantial, ground that then divides to make crisp atoms in a crisp void. Or in category theoretic terms(?), crisp objects and their crisply permitted contextual properties, their various possibilities for action.

Or using x and not-x terminology, we would start in a realm where x-ness and its antithesis are mere unformed possibility (like perhaps order and disorder, atom and void, chance and necessity – absolutely any dichotomy that makes metaphysical sense). Then in creating the crisply not-x, we create the x. Or with equal emphatic-ness, if we create the crisply x, it creates the crisply not-x. As in relativity, the choice of reference frame – “who moved first?” – becomes arbitrary.

I think as you get deeper into Peirce, problems start to arise. For one thing, I don’t think he considers the issue of scale and so his position on hierarchies remains fuzzily developed.

However his semiotic approach as applied to modern physics might read something like this. The Universe has a “mind” – a set of interpretative habits that we know as Newtonian/relativistic mechanics. This generalised mind (a Peircean thirdness) looks into the well of quantum potential (pure vague Peircean firstness) and interprets it into particular physical events or occasions – the classical realm of particles having interactions.

The mind of the Universe never sees a naked quantum realm, only the kinds of events and regularities it has come to expect. This is the famous irreducible triadicity of semiosis. There is the interpreter and the thing in itself. And then the joint production that is the construction of particular signs – particles whizzing about hither an thither in a disinterested void.

Peircean logic contains everything and the kitchen sink. You have the monadic principle of vagueness. You have the dyadic principle of dichotomous separations (or phase transitions or symmetry breakings we might call them). And you have the triadic principle of semiosis (or hierarchical complexity).

Again, what is category theory about at root and does it really map to the whole of Peirce’s organic framework or just perhaps to the dyadic part?

Cheers – John McCrone.
---------------------------------
 
Last edited by a moderator:

Similar threads

  • · Replies 15 ·
Replies
15
Views
5K
  • · Replies 7 ·
Replies
7
Views
3K
  • · Replies 29 ·
Replies
29
Views
6K
  • · Replies 58 ·
2
Replies
58
Views
9K
  • · Replies 105 ·
4
Replies
105
Views
15K
  • · Replies 7 ·
Replies
7
Views
5K
  • · Replies 8 ·
Replies
8
Views
3K
  • · Replies 37 ·
2
Replies
37
Views
8K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 28 ·
Replies
28
Views
5K