There’s a new paper out on the arXiv this evening, advertising a new concept called “string universality”. The authors argue that in six dimensions, by use of appropriate compactifications, any consistent 6d supergravity theory has a string theory realization. They go on to conjecture that the same might be true in four dimensions. As for the implications of “string universality”, they write:
If it is correct, or even close to correct, that string universality holds in six dimensions, then in this case we seem in some sense to be in the worst possible situation vis a vis low-energy predictions. If every possible consistent theory can be identified from low-energy considerations, and all of these theories can be realized in string theory, then string theory would seem to have no predictive power for low-energy physics…
Not being able to predict anything sounds bad for string theory. But wait, they go on to explain why not only is this not a deadly problem, it’s actually a “strength of the theory”. You see, there’s “symmetry and elegance” to a principle that is consistent with absolutely everything and constrains nothing. Some worrywarts might have problems with such a principle since it can’t be tested, but, just because something can’t be tested doesn’t mean it’s not right, no?
This may seem like a very awkward situation for string theory. It should be emphasized, however, that there is no reason a priori why a theory of quantum gravity relevant at the Planck scale of 1019 GeV should make any prediction for physics at the scale of 1 TeV, 16 orders of magnitude below the quantum gravity scale. String theory is valuable as a framework for describing quantum gravity. If in fact, string theory can be used to provide a UV completion of essentially any low-energy theory whose coupling to quantum gravity does not violate some basic principle like unitary via anomalies, this can be seen as a strength of the theory. There is a certain symmetry and elegance about the notion of a quantum gravity theory which provides for the production of essentially all possible low-energy behaviors in some regime of the theory or region of the metaverse.
If indeed, string theory can give rise to such a wide range of low-energy behavior that predictions at the TeV scale cannot be made precisely, it may bother some scientists that this makes the theory difficult to test. But, on the other hand, this does not make the theory any less likely to be correct. It just makes it more difficult to verify.
I actually think this would be great, this is a good place for String theory, not as a unique theory that gives immidiately testable predictions but as a generic framework like QFT.
Just as in QFT you specify gauge groups and representations to obtain a theory, in ST you specify a geometry/compactification.
It’s not a theory of everything but a theory of every theory.
I think it’s a beautiful idea: Strings form a completion of effective, low-energy field theories, and (hopefully) offer the correct way to make the loop diagrams finite order by order.
Why stop there? String Theory should not be satisfied until it has subsumed all of the inconsistent theories as well.
MathPhys,
But that’s not the idea the authors are claiming is beautiful. Their new idea “string universality”, for which they claim “symmetry and elegance”, is that of a “theory which provides for the production of essentially all possible low-energy behaviors in some regime of the theory or region of the metaverse.”
In the past, a physical theory was considered to be “symmetric and elegant” if it explained a large number of observable phenomena with a small number of constructions. What is new here is the discovery that you can get “symmetry and elegance” at the opposite extreme: a theory which explains no observable phenomena using extremely complicated constructions. They couple this with a very powerful new principle of how to do science: just because a theory can’t ever be tested doesn’t make it any less likely to be correct. This principle has wide possible applications, promising to bring back together science and religion, healing the terrible split which occurred with Galileo.
If this in fact can be proven (or at least shown to the satisfaction of the string theorists), this would be a significant advance; it would also put a stop to the incessant “test of string theory” press releases.
They couple this with a very powerful new principle of how to do science: just because a theory can’t ever be tested doesn’t make it any less likely to be correct.
Yeah. I’ll bet that the IDers will be all over this, not to mention the astrologers and diehard believers in psychic phenomena. Talk about riding coattails…
Notice (yet again) the way the author skates right by the essential emptiness of “verification” without testability. This is precisely what galvanized Karl Popper as a young man, as he observed the depressing contrast between Freudian psychology and Marxism on the one hand, and what he saw happening at the same time in theoretical physics, where people actually seemed to care about checking for the possibility that their ideas might be wrong, ie, might imply assertions that contradict observations and results of experiments. More precisely, they cared enough to formulate ideas that were logically capable of contradicting observations and results of experiments.
Richard Feynman died in February of 1988, but even then he smelled a large rat.
If this in fact can be proven (or at least shown to the satisfaction of the string theorists), this would be a significant advance….
Not to mention a great piece of mathematics.
Peter Shor,
There’s no more claim here than anywhere else to be able to show “string universality” in 4d. What is remarkable in this paper is the claim that “string universality” would not be a reason to give up on string unification, but instead would be evidence of “strength”, and a “symmetric and elegant” scientific picture of the world, with the fact that it could never be checked not a problem. Most string theorists I talk to claim the argument for string theory unification these days is “we still don’t understand string theory”, that when we do, maybe it will somehow work out. They acknowledge that if, once string theory is completely understood, it has nothing to say about particle physics, then the game is over, you can’t claim string unification as successful science.
I don’t think there’s any prospect of press releases about “tests of string theory” stopping. Even now, these pretty much all just ignore what is known about string theory.
One other peculiar thing in this paper is the wording of the implication of “string universality” as “predictions at the TeV scale cannot be made precisely”. The problem being discussed is whether ANY predictions, no matter how crude, can be made, not the question of precise predictions. It’s unclear to me why the authors use this language that doesn’t correspond to what they are talking about.
why the 6d case is even worse than the 4d case? Naively I would guess that to get 4d one needs to compactify more extra dimensions so there is more arbitrariness about how to choose the extra dimensions: one M-theory in 11d, 5 string theories in 10d, …, 10^500 not-even-wrong theories in 4d.
fh just repeated the far too common retort on this blog…
“just as in QFT you specify gauge groups and representations to obtain a theory, in ST you specify a geometry/compactification.”
NO! There is the SUBJECT “Field Theory” about the properties of different field theories. This is not the standard model, which is a, single, not family of, not class of, not generic, not a googol of, but ONE field theory. That’s the theory.
It seems that the distinction between possibly beautiful math, which doesn’t require empirical confirmability; ultimately, only ‘logical proof’ – and science, which requires empirical testability, no matter how beautiful a model (or ‘theory’) – is still not appreciated by many of the posters above!
Well, i do not think that it is important for a theory of quantum gravity, to make predictions at the TeV scale. Someone who is interested in predictions at the TeV scale should work on ordinary phenomemology, and not on any quantum gravity.
The aim of theories of quantum gravity are just to give possible models that describe a possible quantum version of gravity, and mayme come up with additional fields.
Research on quantum gravity is mostly purely foundational research that deals more with mathematical consistency and mathematical innovations than phenomenology.
Benni,
I don’t disagree with you that string theory has nothing to say about particle physics (other than as a possible calculational tool in QCD). However, many of the people who work on the subject seem to think otherwise and disagree with you.
‘What is new here is the discovery that you can get “symmetry and elegance” at the opposite extreme: a theory which explains no observable phenomena using extremely complicated constructions. They couple this with a very powerful new principle of how to do science: just because a theory can’t ever be tested doesn’t make it any less likely to be correct. This principle has wide possible applications, promising to bring back together science and religion, healing the terrible split which occurred with Galileo.’
This humorous sarcasm is just not funny anymore, just sad.
Nigel, your sadness surprised me, so I had to react. That passage made me laugh out loud. “healing the terrible split which occurred with Galileo” is perfect!
The tone is controlled for maximum effect. On occasion Peter gets it just right. Perhaps if we can laugh at the way this branch of physics is going, things are not so bad. I have the impression that you are sad because you are pessimistic about the decline of empiricism in theoretical particle physics. The decline in standards.
But other branches of science are quite healthy and that includes other types of physics (does it not? You may know better than I about that.) So be of good cheer. If I may venture a prediction, things will all turn out fine.
This humorous sarcasm is just not funny anymore, just sad.
Not even fun.
Benni said: Well, I do not think that it is important for a theory of quantum gravity, to make predictions at the TeV scale. Someone who is interested in predictions at the TeV scale should work on ordinary phenomemology, and not on any quantum gravity.
Consider a rephrasing of this for an earlier era:
Well, I do not think that it is important for a theory of relativity* that is consistent with Maxwell’s electrodynamics, to make predictions at relative velocities small compared to the speed of light. Someone who is interested in predictions at these velocities should work with ordinary Newtonian kinematics and dynamics, and not with [Lorentz-invariant] relativity.
I trust that the problematic nature of the second formulation is apparent. 🙂
———————
(* of necessity, an alternative to Galilean relativity)
Well, I do not think that it is important for a theory of relativity* that is consistent with Maxwell’s electrodynamics, to make predictions at relative velocities small compared to the speed of light. Someone who is interested in predictions at these velocities should work with ordinary Newtonian kinematics and dynamics, and not with [Lorentz-invariant] relativity.
Most models of quantum gravity — including the string theories Benni was talking about — predict that the effects of quantum gravitational phenomena at the TeV scale are negligible, just as relativity predicts minimal deviation from Newtonian mechanics.
If you want easy access to TeV scale phenomenology, quantum gravity isn’t likely to help much, for the same reason that QCD doesn’t do much for people studying solid state physics.
Nigel Cook, if you look for research into Graphene in arxiv, you will see SciFi materializing in front of you.
NO! There is the SUBJECT “Field Theory” about the properties of different field theories. This is not the standard model, which is a, single, not family of, not class of, not generic, not a googol of, but ONE field theory. That’s the theory.
Okay, so here’s what I don’t get.
Let’s say the 4D string universality conjecture above turns out to be true, and every 4D SUSY particle theory has a compactified string theory equivalent. Let’s say that we ignore the infinite landscape of theories this opens up, and pick a single one of these compactifications– let’s say we arbitrarily pick a compactification that precisely corresponds to the MSSM.
It does seem basically impossible to go from this point to any new predictions about low-energy particle theory– our MSSM-string model will only give us the same predictions that the MSSM would if we were using field theory tools, by definition.
But, there is one difference, which is that our MSSM-string model will incorporate supergravity. Our MSSM-string model will necessarily include a graviton; and it’s understood to be impossible to calculate anything concerning gravitons with field theory, whereas string theorists claim that under their formulation it is possible.
This means our MSSM-string model will allow us to do supergravity calculations in a setting where all the MSSM particles exist: we will have the ability to do all the low-energy things we could in field theory, since as 4D universality would tell us we can convert any field theory into a string compactification, and we will have new abilities at the SUGRA scale. Is this not useful?
Our MSSM-string model will necessarily include a graviton; and it’s understood to be impossible to calculate anything concerning gravitons with field theory, whereas string theorists claim that under their formulation it is possible.
Nonsense. It’s perfectly possible to calculate things involving gravity in field theory — that’s what GR, or SUGRA, is. The difficulty sets in only (a) near the Planck scale or (b) in subtle situations like the correlations that encode information in Hawking radiation. We also know that super-Planckian scattering doesn’t do much except make black holes, so there’s some sense in which the only regime where quantum gravity could be probed is right in the vicinity of, but not much above, the Planck scale. (This is a little cavalier.)
Coin,
The standard argument that string theory is “better” because it provides a consistent “ultraviolet completion” of a QFT has several problems with it:
1. It’s not true. Recall that we don’t know what non-perturbative string theory is, and we do know that perturbative string theory is not fully consistent. What is being sold as the new piece that will provide consistency is not a known theory, but an unknown one that is just conjectured to exist.
2. Such a supposed completion is not unique, you probably can come up with 10^500 of them. Not only do you get no predictions at 1 TeV, you get no predictions at any scale, since you’ve always got lots of possibilities.
3. Claiming that adding a complex superstructure that can’t be tested somehow is an improvement is just bad science.
What I really don’t understand about some string theorists these days is why they just can’t face up to standard scientific practice: if you pursue conjecture X and find that it is basically empty and can’t be tested, you admit failure and move on. You don’t start writing about “X universality”, promoting X as a great scientific success, an improvement over other ideas.
I enjoy lurking here and thought I’d venture a question:
Peter Woit says:
June 8, 2009 at 7:53 am
…a “theory which provides for the production of essentially all possible low-energy behaviors in some regime of the theory or region of the metaverse.”
I looked up “metaverse,” and found no physics references to this term, only references to a “cyberpunk fiction virtual world,” that is presumably emulated in Second Life, and such applications. Yet, the authors of the paper in question seem to be using it in a physical sense. Can anyone clarify? Are “metaverse” and “multiverse” being used interchangeably by some physicists?
hrt,
In the usage of people interested in this, “metaverse” just seems to be a less popular variant of “multiverse”. I don’t know why these authors decided to use it.
Off the topic of this post but perhaps related. Have these papers
been causing any kind of stir around the theoretical physics community? They seem to be new or perhaps renewed approaches to unification with interesting fallout like inflation. Looking at these with limited understanding it seems like there are some promising approaches to larger scale theories. Hat tip to Ars Technica for their article. To this layman’s eye this seems vaguely related to things like the double relativity and such that have been around.
Nature Physics, 2009, DOI: 10.1038/nphys1298
PRL, 2009, DOI:10.1103/PhysRevLett.102.161301
PRL, 2009, DOI: 10.1103/PhysRevLett.102.221301
Markk,
Yes, lots of people have started writing papers about this kind of quantum gravity model. Presumably it will soon get sorted out whether there’s a serious inconsistency in the idea. But even if it’s consistent, this looks to me like just one more untestable quantum gravity theory. It explains nothing about observable particle physics. For more discussion of it, best to find a blog run by someone with more interest in quantum gravity than me.
Pingback: String Universality Reloaded « Not Even Wrong
Pingback: The end of science and art « Jim’s Blog