There has been a lot of debate lately about string theory and testability (see for instance Hanging on by a Thread) . It seems to be one of the only theories in the modern time that is taken seriously by physicists in spite of the fact that it cannot be empirically tested, and in fact it is hard to see how it could be so tested. This gets a lot of people upset as it looks like string theory can’t be falsified and so shouldn’t count as a scientific theory.
Ed Witten is famous for, among other things, arguing that string theory does make one prediction. It predicts gravity. It is concievable, he argues, that some alien society discovered string theory before they discovered gravity. In that world gravity is a prediction of string theory and not of quantuum mechanics. So it merely an historical accident that in our actual world we discovered gravity first and string theory second. So there is a sense in which string theory has already made one bery important ‘prediction’ albeit one that is already established.
Is this all the evidence for string theory that we can muster? I have been thinking lately that perhaps we can get some evidence for something that is at least more string-ish than particle physics if we think about special relativity….So perhaps the most famous result of the special theory of relativity is the equivelance of mass and energy captured by E=MC^2. There have been two philosophical interpretations of this equivelance (see the Stanford Encyclopedia entry linked to above). There are those who take it to show that the properties picked out by ‘energy’ and ‘mass’ turn out to be the same property, and those who take it to show that there is no ontological distinction between fields and matter, or in other words that there is just one kind of stuff out there. It seems to me that either way one interprets the equivelance of mass and energy it provides some reason for thinking that a theory like string thoery will turn out to be correct (as opposed to a theory like particle physics).
So, say that you take it to show that there is only one kind of fundamental stuff out there that we can describe as either energy or mass. This is evidence for something string-ish in that string theory posits only on fundamental entity, viz. the string. Particle physics, on the other hand, posits a zoo of particles that are all made from different stuff. Electrons are made from one kind of stuff quarks from another kind of stuff. Thus particle physics as standardly construed seems fundamentally at odds with the ‘one stuff’ interpretation of E=MC^2.
On the other hand, let’s say that you take the equivelance to show that energy and mass are the same property, though that property may be had by several different kinds of stuff. This is evidence for something string-ish in that string theory posits that the one property is actually string vibration and so can explain what the property is as well as how it will seem to us that one converts into the other (i.e. the pattern of vibration changes). What is the candidate property that particle physics offers? It’s only answer is ‘a property, we know not what’. Thus the two fundamental properties of physics are rendered completely mysterious.
Now, I don’t think this second argument is as decisive as the first, and I also think that the property interpretation is more likely to be true, so I tentatively conclude that special relativity does give us some reason to prefer a string-ish theory over particle physics…but I will have to think about it some more…
Richard Brown 
As for the testability issue, I think there might be an analogy in the economics idea of market efficiency (or perfect economies). There are a number of fundamental indicators that we look at when evaluating the worth of a company. Likewise, we tend to think we have some indicators about the worth of scientific theories – falsifiability and so on. In either case, we have little choice but to make our best predictions based on these indicators. But, as economic research shows, fundamental analysis ultimately fails, and very few if any can do it well enough to beat market averages. Of course, financial markets are at least implicitely structured to achieve just this very effect as predictable trading profits imply inefficient capital allocation.
Now look at theoretical physics as a competitive market of ideas. Falsifiability is one of those fundamental indicators. It’s no secret, and a very large crowd of physicists who are all smart as hell have apparently looked carefully at this fundamental and taken it into consideration when choosing their research projects. So I kind of see attempts to look at the structure of string theory and bet on its success or failure based on price/earnings ratios (falsifiability etc.) as an attempt to arbitrage the future of science.
I think our only real hope is that the market of ideas in physics remains as competitive as possible and the fundamentals of whatever theory pans out in the future will only be clear as day as the one that was true in retrospect.
Well, something will have to change the standard model. It’s a mess 🙂
I’m not convinced by this. The Standard Model does divide matter into many different kinds, but since it acknowledges that they all have mass or energy one can interpret it to be saying that there is one fundamental stuff that matter is made of, but it has different flavours (like saying everything is ice cream, only with different flavours). So I don’t think the SM is at odds with mass-energy ontological equivalence. If we want to show that all the different flavour stem from different conformations of strings, then we would be continuing a kind of pattern of advancing physical theories (that is, unification of ontologies), but no more than that.
In any case, even if this argument works, it wouldn’t favour string theory over rivals like loop quantum gravity.
Hi AG,
I think you are right that the ‘market of ideas’ is important and that there are several indicators that scientists look at when evaluating a theory…my point is just that special relativity seems to require a theory that is more like string theory than the standard model…
Hi Ponder,
Thanks for the comment!
Someone is not convinced by something they read on a a blog? What is the world comming to? 🙂
Seriously, though, I don’t think your ‘everything is ice cream’ strategy avoids the arguement that I am trying to develop…For, what is it that distinguishes the flavors of matter from each other? More importantly, though, what distinguishes the quark from the electron? And what else is the ‘pattern of unification of ontologies’ if not moving to the ‘one stuff’ interpretation that special relativity requires?
Finally, re the loop quantuum gravity stuff, as far as I know (which is not all that far) LQG is only a theory of space-time and NOT a theory of matter at all, so in so far as LQG relies on the standard quantuum field interpretation of matter then the argument, if it works, does indeed favor string theory over LQG.
Well, but the SM already has the ‘one stuff’ part that relativity requires. The pattern of unification of ontologies is a justification outside of special relativity. Perhaps it does justify string theory, I don’t know. But the point is that SR alone does not justify a unification of ontologies.
I know almost nothing about LQG at all. But from the little I’ve read about it, it seems that people have tried to derive different forms of matter from it. See, for example, this.
Thanks for that interesting link! I enjoyed reading the paper, though I am sure I did not fully comprehend everything in it!!
It does seems a bit odd to me that the particles (i.e. the braided graphs) have to be placed into the theory. They do no arise from LQG in a natural way (for instance, in the way that the do from string theory)…that seems to mean that LQG does not (nor could not) predict any new particles…it would have to wait until they were discovered and then they would have to hand placed into the theory…that seems odd to me…So maybe you are right that the argument based on SR that I want to give does not decide between LQG and string theory (though, it does look like LQG ends up saying that there really is ‘no stuff’ and so still might be in conflict with special relativity in that way…but I am still thinking about that). BUT I think that LQG is inconsistent with special relativity in another way. Namely it doesn’t seem to respect Lorentz invariance…though I gather that people are working on this issue…
Re the other thing, I’m not sure I am seeing the argument. You say that SM already has the ‘one stuff’ part, but as far as I know that is not a standard interpretation…though I agree that maybe it is a view that someone might try to defend (as, I take it, you are)…but why doesn’t SR ‘by itself’ justify a unification of ontologies? Maybe I am missing your point…
The Standard Model already has a definition for the property that is variously referred to as ‘mass’ or ‘energy’–namely, ‘the (locally) conserved quantity conjugate to time’. (Actually, in relativistic QM and General Relativity we have ‘energy-momentum’ which is ‘the locally conserved quantity conjugate to position in space and time’, but you get the idea.) (And for extended bodies rather than particles, we have the energy-momentum-stress tensor.) All of these are perfectly well-defined descriptions that are philosophically and mathematically related and pick out definite properties in a wide class of physical theories. They’re not as easy to explain to non-experts as ‘string vibration’, but they do the job just as well.
Hi xplat, thanks for then comment!!!
I am getting in a bit over my head here, but it looks like what you have offered is a description that doesn’t solve the problem I was trying to raise. So, all you have said is that energy is ‘whatever is conserved when time is held invariant’ how does that say anything more than ‘a property, we know not what’?