The Nature of Physical Proof: Is String Theory Proven?

[hyperds]

Edit: Here is the short (but more confusing and less rigorous) version of what I wrote below: For something to be considered a "law" it needs to make predictions which can then be tested, and if they are proven then it is considered law. However, scientists have been looking at the nature of physical proof from only one angle. They don't realize that logically speaking, a backwards proof is equivalent to a forwards proof. A "forwards proof" is where first a theory is created which creates predictions, and then the predictions are tested. It shows correlation between reality and the theory. A "backwards proof" is what Newton did when he came up with gravity. He took empirical observations and then synthesized the theory of gravity to explain the relationships he saw. This also shows correlation between reality and theory. String Theory does the same thing as Newton did, it takes observations (the fact that QM and GR exist in the same universe) and then develops the math to explain how this can be true. It shows correlation between reality and theory. If correlation between reality and theory is what makes something "law", then String Theory is law.

People keep talking about how string theory/m-theory needs to make "testable predictions" and people have even argued that string theory is nothing more than philosophy because of the lack of empirical evidence.

I think this is completely wrong, and stems from illogical thinking about what it means for a theory to be proven.

Take Newton's law of universal gravitation. Obviously mainstream science takes

as law. What is this based on? Newton is said to have done empirical observations, and then developed the mathematical relationship that explained the data. Further experiments showed that the mathematical relationship held, and gravity became a law.

Logically, the fact that a certain mathematical relationship explains the data perfectly does not imply that the mathematical relationship is in fact how the data is being created. It is conceivable that there could be some completely different math which led to the same exact observations. Therefore, for gravity to be absolutely proven, it would be necessary to do a mathematical proof which showed that Newton's formula is the only possible mathematical relationship which can explain the data that was seen.

For the purposes of argument, let's assume that this is the case--that it has been mathematically proven that Newton's equation is the only possible mathematical relationship which can explain the experimental data.

So this means that for something to be "law", all that is required is for a mathematical relationship which explains observed data to be developed, and then that math to be proven to be the only possible math which can explain the data.

Now let's look at string theory. String Theory takes the observation that GR and QM exist in the same universe, and tries to explain it. It comes up with the math to show how GR and QM can coexist.

In other words, it does exactly the same thing that Newton's law of gravity does-it takes observed data, and then proposes the mathematics which explains that data.

The only thing missing is the mathematical proof that String Theory is the only possible way of explaining the observed data. But then this proof is missing for Newton's Law of Gravity as well. Which means String Theory and gravity are equal. So if gravity is considered law, then so should String Theory.

And in fact, there is evidence that String Theory is probably the only possible way to reconsile GR and QM, because many physicists agree that reconciling GR and QM without extra dimensions is impossible. And then reconciling GR and QM using particles instead of strings is also impossible.

In contrast, Newton's law of gravity does have feasible alternative explanations, including General Relativity, and Jhttp://www.physorg.com/news85310822.html" . The point being that even Newton's "law" of gravity and his formula, actually has formulas which produce equivalent results, and thus clearly does not satisfy the requirement that the proposed math is the only possible math which can explain the data. Meanwhile, String Theory, may in fact be the only possible math that can explain the data, and yet it is still not considered "String Law" even though there is exactly the same reason to believe it is true, as there is to believe Gravity, or General Relativity is true.

 

[atyy]

We do already have a theory of quantum gravity for low energies. The aim of string theory is to provide a high-energy (UV) completion. It is unknown if string theory is the only UV completion possible.
http://relativity.livingreviews.org/Articles/lrr-2004-5/
http://golem.ph.utexas.edu/~distler/blog/archives/000639.html

 

[mitchell porter]

Since string theory is a theory of everything and not just a theory of gravity, to answer the question, you really have to judge it against the whole of the standard model. It would be no good if string theory got gravity right and something else wrong. Fortunately it appears to be capable of giving us everything else we need to describe reality. What remains to be demonstrated is that it can get the details right too - I mean the exact particle masses, and so on.

 

[member 11137]

Edit: Here is the short (but more confusing and less rigorous) version of what I wrote below: For something to be considered a "law" it needs to make predictions which can then be tested, and if they are proven then it is considered law. However, scientists have been looking at the nature of physical proof from only one angle. They don't realize that logically speaking, a backwards proof is equivalent to a forwards proof. A "forwards proof" is where first a theory is created which creates predictions, and then the predictions are tested. It shows correlation between reality and the theory. A "backwards proof" is what Newton did when he came up with gravity. He took empirical observations and then synthesized the theory of gravity to explain the relationships he saw. This also shows correlation between reality and theory. String Theory does the same thing as Newton did, it takes observations (the fact that QM and GR exist in the same universe) and then develops the math to explain how this can be true. It shows correlation between reality and theory. If correlation between reality and theory is what makes something "law", then String Theory is law.

People keep talking about how string theory/m-theory needs to make "testable predictions" and people have even argued that string theory is nothing more than philosophy because of the lack of empirical evidence.

I think this is completely wrong, and stems from illogical thinking about what it means for a theory to be proven.

Take Newton's law of universal gravitation. Obviously mainstream science takes

as law. What is this based on? Newton is said to have done empirical observations, and then developed the mathematical relationship that explained the data. Further experiments showed that the mathematical relationship held, and gravity became a law.

Logically, the fact that a certain mathematical relationship explains the data perfectly does not imply that the mathematical relationship is in fact how the data is being created. It is conceivable that there could be some completely different math which led to the same exact observations. Therefore, for gravity to be absolutely proven, it would be necessary to do a mathematical proof which showed that Newton's formula is the only possible mathematical relationship which can explain the data that was seen.

For the purposes of argument, let's assume that this is the case--that it has been mathematically proven that Newton's equation is the only possible mathematical relationship which can explain the experimental data.

So this means that for something to be "law", all that is required is for a mathematical relationship which explains observed data to be developed, and then that math to be proven to be the only possible math which can explain the data.

Now let's look at string theory. String Theory takes the observation that GR and QM exist in the same universe, and tries to explain it. It comes up with the math to show how GR and QM can coexist.

In other words, it does exactly the same thing that Newton's law of gravity does-it takes observed data, and then proposes the mathematics which explains that data.

The only thing missing is the mathematical proof that String Theory is the only possible way of explaining the observed data. But then this proof is missing for Newton's Law of Gravity as well. Which means String Theory and gravity are equal. So if gravity is considered law, then so should String Theory.

And in fact, there is evidence that String Theory is probably the only possible way to reconsile GR and QM, because many physicists agree that reconciling GR and QM without extra dimensions is impossible. And then reconciling GR and QM using particles instead of strings is also impossible.

In contrast, Newton's law of gravity does have feasible alternative explanations, including General Relativity, and Jhttp://www.physorg.com/news85310822.html" . The point being that even Newton's "law" of gravity and his formula, actually has formulas which produce equivalent results, and thus clearly does not satisfy the requirement that the proposed math is the only possible math which can explain the data. Meanwhile, String Theory, may in fact be the only possible math that can explain the data, and yet it is still not considered "String Law" even though there is exactly the same reason to believe it is true, as there is to believe Gravity, or General Relativity is true.

I really appreciate the forum because it gives a rare opportunity of discussion on complicated topics. I don't know so much on the details of String Theory. That's why I shall stay at logical level. You say that data can be explained by mathematical theories A, B or C (e.g. for Gravity). Let us say that String theory is actually trying to put A and the Quantum theory (QT) together. But was is the chance to get a real success if would have been more accurate to work Gravity with the B Theory? I mean: perhaps do we loose some precious time because we start with an incomplete mathematical tool?

But please understand me correctly: I respect the rules of the forum = I never have said that GR is false. Just thinking about the way to built a theory.

 

[Fra]

Your idea of the scientific process seems strange.

What you call "backwards proof", is a descriptive problem. In particular to represent the information in historical evidence, in a simpler way. A kind of data compression.

I think think such new descriptions or representations are never unique. The point with data compression is rather just to find a simpler representation. But dependending on context, the selection principle based on simplicity is ambigous.

Without law, all we have are loads of historical empirical data, that constitutes our knowledge. Once we realize that there are patterns in the data, in particular of the form where one can represent a whole ordered historical swequence of data by a single initial condition and then a dynamical law that is fixed for the entire set, we have manged to reduce the information capacity required to encode the same information state - clearly preferrable.

"Future proof", is more like a decision and learning problem. You act according to expectations, and see if your expectations are right. If they are right your confidence is strengtened, if they are wrong it's weakened. This isn't "proofing", it's maybe corroboration, but even better abstracted as a learning process. An in that learning process, it does not really matter so much wether your expectation are right, what is more interesting is how you act in the event of beeing wrong. Ie. exactly how to rationall revise your expectation of laws.

Becuase all we EVER have are EXPECTATIONS of what the laws are. These expectations clearly determine our actions in the process you call future proof. Because which theories to we test first? Yes the most promising ones of course. As it involves resources and risks to test theories.

I would argue, that for a theory that aspires to be a "theory of theory" the more important trait is exactly how the laws evolve. In my gaming example, the important things is how the expectations of law (which are the current EFFECTIVE LAWS) are to change, when proven wrong?

This is as I see it, one of the weaknesses of ST. There is some large space of theories, but the most important thing: the selection principle is missing. That's not to say it can't be found to complete it.

/Fredrik

 

[hyperds]

The way I see it, since ST has been shown to be the only way that has had success in uniting ALL of the forces into a consistent mathematical structure, it HAS to be true.

What would it take to "prove" it in the minds of skeptics? Would the discovery of the Higgs Boson be sufficient, or would the skeptics wait until we can harness the astronomical amounts of energy necessary to probe the strings themselves?

In either case, it is unnecessary. If you believe that the universe follows logic, and has a consistent mathematical structure, (i.e. you don't believe magic), and if string theory is the only way to unite the forces we can see, then it is PROVEN. Unless someone can come up with an alternative explanation, which is mathematically consistent, and doesn't require "magic" where the laws at one energy or size magically change into different laws with no explanation, then string theory should be considered proven.

Fortunately, I think we will be able to have our cake and eat it too. ST has already led to insights into real world physical problems which appeared intractable using the Standard Model, such as the use of ST in solving the problems at the RHIC, and there are many other things which are showing promise as well, such as this: http://www.sciencedaily.com/releases/2009/07/090706113702.htm where ST was applied to explain superconductors.

After days and nights of hard grind, it was a puzzle that fitted. "We hadn't expected it to work so well," says a delighted Zaanen. "The maths was a perfect fit; it was superb. When we saw the calculations, at first we could hardly believe it, but it was right."

 

[ZapperZ]

Er... I would try to tamper such outlandish claim, if I were you.

The "proof" of a valid theory has always been convincing experimental evidence. Currently, String theory has none. Furthermore, there are almost an unlimited versions of String/M-theory. So your claim that it has been shown to be the ONLY way is not valid.

And please note, String theory in that Zaanen article (citing Hartnoll et al. paper) does NOT explain superconductors. It tried to explain the Nernst effect seen in heat and charge transport in superconductors. This doesn't solve the mechanism of superconductivity in the cuprates/pnictides at all!

Zz.

 

[MTd2]

String Theory is probably the only possible way to reconsile GR and QM, because many physicists agree[...] .

If they agreed that Flying Spaghetti Monster exists, would you believe them?

 

[suprised]

If they agreed that Flying Spaghetti Monster exists, would you believe them?

This is a very unfair comment, since it puts string theory on a similar level of speculation as the FSM. What is routinely not said is that there are highly non-trivial computational results - one may call them theoretical hard facts; like the emergence of GR. So in contrast to some random idea, string theory is a well-motivated line of research.

 

[MTd2]

So in contrast to some random idea, string theory is a well-motivated line of research.

An analogue calculator can be built from systems with no common physical principles. I can integrate, differentiate, solve systems of differential functions, with use of only capacitors, resistances, inductors, wires and batteries. (This is not trivial: http://en.wikipedia.org/wiki/FERMIAC or even a quantum computer http://en.wikipedia.org/wiki/D-Wave_Systems) But to say that these elements are the basic fabric of the universe is similar to summon the Flying Spaghetti Monster as an explanation for anything.

 

[suprised]

You don't know what are you talking about, discusssion finished.

 

[MTd2]

The example above, which is not quite successful though, about explaining superconductivity using results of string theory, is about using it as an analogue calculator. Even QCD/CFT are analogue models. Why not to say that about the whole ADS/CFT research or even every kind of string theoretical kind of models?