Timeline for Explanations for mathematicians, about the falsifiability (or not) of string theory
Current License: CC BY-SA 3.0
18 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jul 5, 2017 at 7:13 | comment | added | Samuel Monnier | Fair enough, but that's not the point. I don't think that the anomalous precession of Mercury was an ingredient useful to build or even hint at the correct theory. His work was driven by consistency considerations. He was lucky that there happened to be an experimental setup where relativity was differing significantly from Newtonian gravity. We are not as lucky with quantum gravity/string theory. | |
| Jul 4, 2017 at 6:37 | comment | added | Jonas Sourlier | Not sure whether your comment about Einstein's theory of general relativity is strictly true. Einstein successfully calculated the anomalous precession of Mercury before he published the theory. It was a huge hint that he was on to something big. So strictly speaking, GR already had experimental support when it was published. | |
| Jun 10, 2013 at 17:52 | comment | added | Urs Schreiber | a) String perturbation theory is a perfectly well defined theory. b) If anyone points out mistakes on nLab pages, they will be corrected. | |
| Jun 10, 2013 at 15:41 | comment | added | Peter Woit | I guess your argument is that we just need to get a list of string theory vacua and the ability to calculate detailed predictions from them, do this, then falsify the theory if we find no match. The problem with this is that we don't actually know what the theory is, so even in principle cannot do what you suggest. So, the theory is unfalsifiable by this line of thinking by lack of being well-defined. You're free to conjecture that a well-defined theory exists where these calculations can be done, but until someone produces it, the current version of "string theory" is unfalsifiable. | |
| Jun 10, 2013 at 15:35 | comment | added | Peter Woit | The models described in the paper you link to don't "reproduce convincingly the standard model". Even the authors refer to them as only "quasi-realistic". These are very complicated geometric constructions which are claimed to only give the the correct SM gauge group and fermion representations, a small amount of information about the theory (in particular, none of the continuous parameters of the theory are calculated). As typical for these constructions, more complexity goes in than comes out. No one would claim these could convince anyone this is an explanation of the standard model. | |
| Jun 10, 2013 at 14:56 | comment | added | Samuel Monnier | See for instance this paper: arxiv.org/abs/1202.1757 "If the string vacuum you pick actually predicts something and it turns out to be wrong, that will not falsify string theory." No it won't. You have to study a theory thoroughly and completely before hoping to falsify it. This means in particular understanding the space of vacua, exactly like for a quantum field theory. Also, my answer to the question is as it is written above and not as you rephrase it. And I guess it's valid as well for Urs' comments... | |
| Jun 10, 2013 at 14:20 | comment | added | Peter Woit | I don't believe it's accurate to say that a string vacuum "reproducing convincingly the standard model" exists. If you know of one, please explain what it is, and what its predictions are that will test it. In any case, I don't see how you get a falsifiable prediction of string theory out of this. If the string vacuum you pick actually predicts something and it turns out to be wrong, that will not falsify string theory. Your answer to the question still seems to me to be the same as that of Urs Schreiber, that the theory is unfalsifiable. | |
| Jun 10, 2013 at 14:01 | comment | added | Samuel Monnier | ... vacua reproducing convincingly the standard model. These solutions of string theory cannot be ruled out with our current experimental knowledge, as the standard model is not yet ruled out. They predict a well-defined completion of the standard model that we will be able to test with experiments probing higher energies. | |
| Jun 10, 2013 at 13:57 | comment | added | Samuel Monnier | Again, what the theory predicts depends on which solution you're looking at. In order to find such an experiment, you have to fix such a solution, and compute the physics it predicts. If the solution is 10-dimensional Minkowski space, the experiment consists in the observation of the number of macroscopic dimensions of the spacetime we are living in. As this number is 4, this particular solution of string theory is ruled out as an explanation for the world we live in. Then go on with more realistic vacua. From what I know (but I'm no expert in string phenomenology), we now have found some... | |
| Jun 10, 2013 at 13:10 | comment | added | Peter Woit | In response to the addendum: You still have not addressed the question that was asked. You claim that string theory is falsifiable, but you do not give what was asked for ("one experiment that would test it"). In addition, you seem to be trying to make the same argument as Urs Schreiber, but giving the opposite answer to the question (he seems to be claiming string theory is unfalsifiable, but that this is true of just about all physical theories). | |
| Jun 10, 2013 at 12:35 | history | edited | Samuel Monnier | CC BY-SA 3.0 |
cosmetic changes
|
| Jun 10, 2013 at 12:06 | comment | added | Urs Schreiber | mbsq is right, a large number of solutions has never been a problem of any theory. Compare for instance general relativity, which certainly has a highly infinite dimensional moduli space of solutions (certainly not a finite set, and be it of size 10^500). A theory alone never makes any prediction, it is only a theory together with a choice of model in the theory that does (a choice of some of the paramters). E.g. general relativity does not predict that we live in an FRW universe. That is the model we have chosen because it fits observation. | |
| Jun 10, 2013 at 0:20 | comment | added | Monroe Eskew | Why is a large space of solutions a problem? Do we think we are going to "explain" why reality is the way it is? As mathematicians don't we know that this is inherently hopeless? There are infinitely many mutually incompatible consistent theories. The job of physics should be to describe the world in a systematic way, not search for "ultimate explanations" which is infinitely regressive and hopeless. | |
| Jun 9, 2013 at 23:41 | comment | added | Peter Woit | You didn't actually answer the question. Your argument seems to be that the theory is falsifiable, so you should provide what is asked for: a specific experimental prediction. Note that if your prediction is that "at high enough energy one will see the soft scattering amplitudes of perturbative string theory", you really haven't taken into account M-theory in general, so you are talking about falsifying only one corner of the theory. | |
| Jun 9, 2013 at 18:00 | comment | added | Woett | $n > 0$ seems like a good start. | |
| Jun 9, 2013 at 16:26 | comment | added | JHI | @Woett So how do you actually 'test' a theory if its true? When is it fully tested? Once you have performed 1,2,3,...,n experiments consistent with it, which n? | |
| Jun 9, 2013 at 16:09 | comment | added | Woett | While I definitely understand your first issue, I once read somewhere (might be Lee Smolin's book 'The Trouble With Physics') something along the lines of "If you have any physical theory, there are going to be practical ways to test it. Find them.'I must say I quite strongly believe in that statement. | |
| Jun 9, 2013 at 10:36 | history | answered | Samuel Monnier | CC BY-SA 3.0 |