String theory ~ the theory of physical theory?
|May15-08, 03:37 PM||#1|
String theory ~ the theory of physical theory?
In http://www.physicsforums.com/showthread.php?t=235006 there is a reference to Peter Woit's blog, which has one entry regarding the paper
“So what will you do if string theory is wrong?” by Moataz H. Emam
I didn't want to inject the above thread with this sidetrack of mine, so I start a new thread.
This paper is a brief reflection that string theory may have a life of it's own regardless of physical relevance. But the the paper contains the following interesting and IMO ambitious view on string theory:
"I can imagine that string theory in that case may become its own new discipline; that is, a mathematical science that is devoted to the study of the structure of physical theory and the development of computational tools to be used in the real world. The theory would be studied by physicists and mathematicians who might no longer consider themselves either."
If I read the author right, his view of string theory, is that string framework is in a fundamental way more fundamental than physical theory itself, and thus implicitly of higher generality? And somehow that the study of physical law, in a larger context (say such as evolving theories), would imply studying string theory.
Somehow that phrasing is very appealing to me and right in line with some of my thinking, but it's paradoxal that I can't see how the string framework could be a fundamental framework and strategy of sufficient generality to study physical law?
The fact that I want to understand the physical law in context, is why I find string theory speculative. I don't see how the string framework is the solution to the expressed quest?
Does most string theorists share the basic sentiment of this as expressed by Moataz H. Emam or is he in minority? or is he trying to make string theory something it's not?
|May15-08, 04:08 PM||#2|
I grabbed Kea's comment off of Woit's blog. But I didn't mention Woit's blog or the article by M.H.Emam. That article must have been discussed in another entry, as you say. I don't recall any connection. So it really is a fresh topic.
As i understand it, Moataz Emam is a string theorist himself and presumably a devoted one. He seems enthused about string research. He may be unusual in that he is willing to contemplate hypothetically that string thinking might eventually not turn out to have much to do with nature.
Even then, says Emam, it is worthy to be pursued for its own sake. This represents, to my way of looking at it, an admirably dedicated commitment.
Personally I think i might have cautioned you not to broach the subject of Emam's paper because it is likely to spawn contention, or (if not two-sided contention) at least some one-sided expressions of outrage and contempt. A lot of people, it seems to me, simply don't care to contemplate as possible what he assumes hypothetically.
|May16-08, 12:49 AM||#3|
It's rather to reflect over what string theory is. Is it a normal theory, or is it something else (in a way that is also discussed before, and IMO it's a kind of framework that constraints theory construction). I certainly don't mean it in a bad sense. If it was a theory or theories, then that would be excellent IMO. Sometimes I find some of the critics according to poppian thinking to be unfair. The poppian decsription seems to me slightly simplified and out of date, since the notion of falsify is more complicate when it comes to a learning strategy. Beeing wrong isn't a failure - failing to learn is, or failing to learn _fast enough_ in competition is (IMHO at least)
So I was curious on the logic within which this is seen as obvious, because I agree that it is an admirably dedicated commitment, in a deeper way, because IMHO history has taught us that things change. And theories change, therefor to study theories as evolving in larger context seems nice IMO.
|May16-08, 02:34 AM||#4|
String theory ~ the theory of physical theory?
I guess what I did was, to find the most positive interpretation (relative to my view) of that paper.
If we try to make up a new discipline, which are to study the structure and evolution of physical law and physical theory, and from such an endavour try to learn something on practical models and strategies that can be used for real computational predictions, that sounds very nice IMO. Instead of considering theories that described the dynamics of observations, we add the level of self-reference that we are considering the dynamics of the theories themselves.
1) what is a physical theory, and what is physical law?
2) How do they emerge and what are their physical representation (informationwise)?
3) What is the distinction between physical law, and the evolution of physical law, if there is a higher level law of laws?
Maybe I read too muhc out of that paper, but this is questions I ask, and if some string theorists think like this then perhaps they have failed to argue in favour of it?
I personally expect that questions like the above, considers also the physical nature of information, and information processing. And considering things like confidence, and howto distinguish a random lucky guess from a skill, we unavoidably touch the foundations also of statistics and probability theory.
Now if some string theorist share the same visions, and perhaps could elaborate the connection here I think it would have the potential to defend string theory from some of the poppian style critics. I also think these question may be relevant in trying to find a conceptual connection between the various approaches.
|May16-08, 10:16 AM||#5|
I recently read a bit of Nancy Cartwright. Heard her talk, a year or so ago, and was impressed, but didn't immediately follow up until now:
see if that will get you the introduction to her book "The Dappled World"
It might interest you.
I suppose it could be argued that Cartwright presents a more practical and realistic view of physical law than Emam seems to have.
I cant say this is a special interest of mine or that I know much about it, but since you think generally about physical law you might get something out of her introduction.
|May16-08, 10:58 AM||#6|
I found something else that might interest you
The disunity of the sciences
BTW the subtitle of the Cartwright book is "A study of the boundaries of science"
She's a department chair at the London School of Economics and recipient of a Macarthur (one of those so-called "genius awards").
A smart articulate empiricist. (that word even more than realist, pragmatist, practical real-world...is descriptive).
Empiricism making a comeback look at the program of Strings-2008 talks. Look at the 8 onehour review talks. Two of them are by Jos Engeler and Lyn Evans, who aren't string theorists-----the CERN CSO (chief scientific officer) and the head of the LHC project.
It's all quite natural given the time and context but nevertheless you can see the pendulum swinging.
|May16-08, 12:58 PM||#7|
I took a very quick peak at the first pages. Both of them seem to contain in some sense logical reasoning. But I couldn't see if they come to a "constructive suggestions" rather than only arousing sensations :) - did you read them?
I am very philosophical in my style of reasoning in that I am guided more by soundness of reasoning that beauty of mathematics (Although there may be good reasons why they sometimes coincide).
In despite of my philosophical angle, I am definitely looking for a mathematical formalism, that is computable. A theory can fails to come up with a computable (in reasonable time) algorithm is of limited utility. Also I rarely read philosophy litterature as such.
|May16-08, 01:22 PM||#8|
Cartwright is a worldclass expert who has directly addressed the issue of a physical theory-deciding machine, or to put it more vaguely a "theory of theories", and whether or in what sense such a thing is possible. Is there always a social element? Consensus as to what constitutes evidence? Actual personal battles and struggle at some level? Ultimately does science depend essentially on the functioning of communities, with ethos ethic status-ranking etc. Or in what way can it be abstracted and objectified?
I'm not saying that Cartwright is more sophisticated. I don't know enough to judge. It does seem to me however that she has made a brilliant career thinking about exactly what you are proposing to discuss (theory of physical theories) and that she has thought and written and talked and argued with a lot of other smart people about this for decades of year almost nonstop (that, and economic theory too). Intuitively there should be something you couild learn from her, if you had the time. But maybe not.
|May16-08, 03:08 PM||#9|
Fra, you deserve some kind of response to your question and even though I may not be the most qualified person, I will, in the absence of other attempts, try. First notice that this forum is for Professionally researched theories that go beyond established models, it is not just limited to string brane loop---they just are symbolic examples of the kind of topic.
Nowadays Noncommutative geometry (NCG) is very strong because it reproduces the standard particle model simply from a picture of spacetime. That picture is NOT a differential manifold. It is an algebraic substitute for a smooth diff. geom manifold of the sort that Riemann invented in 1850.
String theory is based on differential geometry, on manifolds. That is another way to go.
Branes are differentiable manifolds, like Riemann invented.
Another strong contender is Causal Dynamical Triangulations (CDT) which is based NOT on differential manifolds but on a different idea of spacetime---piecewise linear, simplicial complexes, a limit of them, a Feynman path integral thru the set of them.
Then there are CATEGORIES AND TOPOI and things like that, that people like Grothendieck invented say around 1950 or roughly a hundred years later, and THEY give you a way of describing spacetime and geometry.
And there is whatever Witten is working on. In 2006 when I heard him give three 1.5 hour talks it wasnt string/M it was Geometric Langlands Program. He seemed to think that was more interesting. For the entire 4.5 hours he didn't mention string (somebody in the audience had to ask him about it at the end of the third talk.) So it is possible that THAT could form the basis of a new physical theory, or a machine for searching for theories. I shouldn't single Witten out. There are a lot of creative people nurturing new mathematical formalism and new approaches to physical theory. But he's one example.
So what you are seeing is a battle of ideas to determine what will be the next mathematical formalism by which humans depict space time geometry motion and matter.
Some theories are still based on vintage 1850 differential geometry----smooth manifolds, worldsheets, branes, whatever.
some theories are based on other things, categories, topoi, simplexes, algebra like NCG uses to represent geometry, and other post-1950 mathematics.
the struggle between theories is essentially one to determine what is going to be considered the right formalism, for the time being.
what comes out of this rowdy scuffle may have something to do with string formalism
or it may conceivably have nothing whatever to do with string formalism
we can't tell
research is the hardest thing to predict that humans do.
my own opinion is that it is foolish for someone to put all their bets on one particular mathematical formalism, at this point in the game
so I find what Emam proposed to be ridiculous. it is far too limited in scope.
but that is just my personal opinion.
|May16-08, 11:31 PM||#10|
By coincidence, Fra, the same day I posted the above I saw this by John Baez
since 1850 the european concept of continuum has tended to be the smooth manifold that Riemann invented, or something even simpler like Rn
what I'm telling you is that we are now seeing a brawl between different ways of advancing beyond that idea of continuum----we dont know which mathematical formalism will win out.
string worldsheets and branes are oldfashion smooth manifolds arising in a conventional Diffy Geom context like so much else in older physics, that formalism might prevail
or some other might prevail. it would be naive at this point to commit
BUT BAEZ HAS SOME REASONS WHY the oldfashion diffy geom. continuum is bad!
He shows us there are different ways to define what is a continuum or a smooth space or spacetime and he gives reasons why the vintage 1850 idea is unsatisfactory.
This is good. It is what you train and pay mathematicians for----to be able to see things like this.
Take a look.
So he and his buddy have joined the crowd of people looking for a better mathematical formalism for the continuum, and trying out various alternatives to the oldfashion smooth space idea.
what I'm saying is that ultimately if you want to theorize about the next physical theory, then you need a mental picture that includes alternative concepts of the continuum (not just the preserved Riemannian one you find pickled in so much conventional physics today)
David Gross occasionally says this. It comes across as a desperate outcry sometime: "We don't know what string theory is! We need a fundamentally new idea---fundamentally new ideas of space and time!...." and so on.
That is what they sound like when they need a new mathematical formalism. So take a look at what Baez writes in n-category cafe. He mentions Grothendieck (as I did earlier today). Maybe Baez doesnt have the right answer but he has something like the right SCOPE.
|May17-08, 01:47 AM||#11|
Thanks Marcus for all your comments and sharing of views.
But I agree with alot of what you say here. I have expressed my opinion in several posts that I fail to see the physical basis of the continuum. That's not to say that a discrete model can't be embedded in a contiuum model, but just that I see contiuum models containing uncontrollled unphysical degrees of freedom and that embeddding is thus non-physical, and we should shave off those degrees of freedom and get a more compact representation. My main objections are from the information point of view. You need alot of information to specify a continuum , and well there is something that isn't right there IMHO. If I am not mistaken Smolin have argue along similar lines. So in short, I need no further compelling to question the concept of contiuum
Now I know some may interject here and point out the difference between a physical continuum and a continuum model. Many who work with continuum models seem to agree that ultimately it's a property of the model but not necessarily nature. That is true, but IMO that is not taking the models serious enough. I expect a better connection. From the effective view we already know continuum models are great in many cases. So that isn't the question.
This is very much related to the concept of counting we discussed before. Wether we are counting states of matter of states of the geometry. I think we should bring back a concept of distinguishability. We should count observationally distiniguishable things. One can still understnad why the continuum model is a good approximation to this, as a smoothed version of the real thing.
What I am trying to formalise is instead of a generator of say time on the level of universal law, I am considering a local(local with respect to the observers information) strategy for producing an expectation of the future. And this will have a complex feedback, that is sort of applied inductively. And somehow this induction and other feedback is processed in parallell, and the progress of all this is identified with local time evolution. The my aim is thus to put the notion of physical law and the notion of physical states, on the same level, with the only difference that they live at different levels of the induction and processing. And the notion of law have more inertia than the physical states - so although in principle - both are dynamical objects, the relative change of physical law in a small "time" window is small enough to distinguish between them.
Problem I have are how to represent information. There is a nonlinear feedback means that there is no universal way to distinguish the feedback from what it relates to. And this feedback is a process, that defines time. I've been thinking about this for some time now and I am making progress in small steps.
Meanwhile I like to see if other approaches distinguish the same questions, and how their solutions are lined out.
Anyway, this is why I found Emamns paper paradoxal. He gave a small hint of a grand vision - good. But then he suggests that string theory is the solution to that. I do not have to agree, but it would be enlightning to see if anyone that agrees with him, could expand on how you reach this conclusion.
|May17-08, 03:51 AM||#12|
I see clear similarities with theory building in physics, as well as interactions in social, economical and biological systems. Game theory is one perspective where the above connections is there. This is all in line with my thinking, and the basic conceptual level here is somewhat clear to me. I am looking for how to exploit this.
The question of a physical theory-deciding machine, and wether such a machine can be universal? And what universal means, are interesting questions. My opinion is that each subsystem and observer is in an abstract sense a "theory-deciding machine". And I don't think there can be a universal one. From that point on, which seems rather hopeless, I ask how and why a local rules still emerge. And then to save us from chaos, these rules themselves are rated, which give them a kind of relative intertia.
From my experience with communicating with others, there seems to be two issues in communicating this.
The first is to convey the basis spirit of intent and the vision. I think this is best done in plain english, complemented by examples and analogies. But to understand this one needs a somewhat open mind.
The next issues is, once you are working within this spirit, to actually take it another step, towards something more formal, where you can induce a choice of mathematical or logical formalism, that will allow you to more constructively make quantitative models rather than conceptual models in words.
I have a conceptual idea on my own, and I sense alot of that in the philsophical writings of Rovelli but also Smolin. So I really appreciate their world. But I don't quite understand or accept all of their current solutions to step 2.
Right now, I don't think I would benefit much from reading a step 1 book. I am looking for step 2 suggestions. So I have currently looked into rovellit en penrose and like parts of it, but still looking. My impression is that those(edit: the books you suggested above - rovelli is more precise but I feel he is jumping to fast into the chocies, without properly reflecting over it. I found his line or reasoning to be broken somewhere in his relational QM argumentation, although he started out nice) books doesn't contain stuff at that level? or does it?
If it does, I would definitely want to read it. Let me know what you think. If you think it contains such constructive formal ideas, I might take your advise and order it, and give it a closer look.
|May17-08, 04:22 AM||#13|
It seems a problem is to bridge the a sound somewhat philosophical basis for theory building and science with the more computational stringent formalism.
I find many papers to start right in the middle of some somewhat formal context, and the argue on. But that is lacking the continous line of reasoning that has selected that formal system to start with. And if you think that the induction of something as per a particular line of reasoning, and also the choice of line of reasoning is part of the problems, such an approach which ignores the fundamentals and more or less arbitrarily chooses a formal framework is IMO a high risk one.
If one reads some of the original texts of founders of new disciplines, like Heisenberg and dirac and einstein, it's easy to see the significance of line of reasoning in the development of those theories. It is a clear guide. Extreme formalisations and axiomatisations are i think often post-constructions once the theory is mature. It's not always how real life progress is made as far as I can see.
|Jun3-08, 12:19 PM||#14|
|Jun3-08, 12:27 PM||#15|
The last lines from the TOE (physics) page
|Jun3-08, 04:11 PM||#16|
Pointy I have two comments on your two posts here.
One is not necessarily unfriendly, you may agree with it:
TOE is a really stupid term. It is media hype made up by John Ellis at CERN, who likes to talk to the media and TV and stuff. It is a distracting red herring.
I don't think anyone who is serious about the philosophy of science should bother discussing the TOE idea.
In naive credulous people it creates the fantasy of a UNIQUE ULTIMATE physical theory.
In fact over the centuries physical theory does become gradually more fundamental, more unified, more comprehensive in explanatory power.
But just because some joker on a TV program says TOE doesn't mean we are suddenly near the end of this process . We are nowhere near an end, and maybe there is no end.
So. I dont know, you may agree with this.
So why would serious people even be having a conversation about this phony idea?
My other reaction is critical, but not to make you feel bad. In your posts you dont SAY anything. You just copy/paste in stuff from like Wikipedia talk.
Quotes are supposed to illustrate something you yourself are saying. Or be quotes from some authoritative source or important scientist, that you are showing that this or that wellknown person had a certain opinion. And quotes can give context, so you arent just quoting ambiguously out of context. And you can use links.
But just pasting in anonymous Wikipedia talk text doesn't do anything. I find it hard to force myself to read a post that just pastes in verbiage by nobody in particular from nowhere special.
You know that Wikipedia physics articles themselves are not free from unreliable rubbish. So I would imagine that Wikipedia talk pages have even more stuff that one would naturally just ignore.
So I encourage you to say in brief what you are trying to say, and not merely paste in stuff.
|Jun3-08, 05:06 PM||#17|
The only candidate for a ``theory of everything'' is string theory, which is perhaps why you hate the term, marcus. The efforts I've seen to get the standard model + a theory of quantum gravity fail pretty spectacualry---for example, Smolin's recent work in trying to get the standard model predicts 4 neutrinos (that, naively, are related by some symmetry operation), something that has been experimentally ruled out for a long time (neutrino mixing).
Inasmuch as I can tell, only string theory can claim to have a quantum theory of graviy + the types of mathematical structures needed to get real low energy physics out. The exact low energy physics depends on the compactified directions, of course, but quite generally string theory gives family replication, non-abelian gauge symmetries, chiral matter, etc. And note that you get all of this stuff for free---it's not lumped in to some ad hoc stress energy tensor.
|Similar discussions for: String theory ~ the theory of physical theory?|
|Applied mathematics of Game Theory overlooked as a representaiton in string theory?||Beyond the Standard Model||0|
|Does String Theory Satisfy Einstein's Conditions for a Physical Theory?||Beyond the Standard Model||44|
|String theory as a generalization of gauge theory||Beyond the Standard Model||2|
|Representation theory of the groups appearing in String Theory?||Beyond the Standard Model||4|
|Matrix string theory, contact terms, and superstring field theory||Beyond the Standard Model||2|