Brian63

He came to my campus last week and gave a fairly impressive lecture. I have not (yet) read his book, but I do have a slightly better grasp of string theory than I did before (which is to say, none).

One item that he mentioned has been troubling me for a couple days, and I asked a couple other science-minded people about it, and we were not able to come up with conclusive answers, so I'd like to see what the science nerds here think about it.

It's called superstring "theory" rather than superstring "hypothesis." At one point in the lecture however, he mentioned that there exists no evidence to support superstring theory. Given this, how can he call it a theory? By definition, isn't what makes a theory a theory the fact that there is considerable evidence in support of it?

He mentioned also that string theory predicts certain phenomena, and those phenomena have been observed. However, other competing theories also predict those same phenomena. So the best guess that has come out with a couple brainstorming sessions with my friends is that he considers that evidence not for string theory exclusively, but as evidence for a bundle of theories.

Is that probably what he meant, or something else?

Also, for my own curiosity:

In general, if there is evidence obtained for a certain collection of hypotheses, but not one of those hypotheses exclusively, could that be considered evidence for each of those hypotheses individually, or must evidence for a hypothesis be exclusive to that hypothesis?

Brian

Vorkosigan

Well, a theory is just a framework for organizing data. It suggests areas for research, provides explanations/models of reality, offers areas where it could be tested and confirmed/disconfirmed, and perhaps has ramifications for other fields of knowledge. String theory could satisfy all of the above. Isn't the evidence for it largely deductive and mathematical?

The steady-state theory, which had no evidence to support it, was also a theory. Cosmology seems rife with this kind of stuff.

Michael

tronvillain

No. A hypothesis does not change into a theory.

Brian63

From the library, courtesy of Richard Carrier:

Am I misinterpreting this somehow?

Brian

Bill

I have the benefit of having actually read Brian Greene's book (twice, actually ... I got a lot more out of it the second time through).

As for the confusion over the word "theory," let me offer the most pertinant dictionary definition as a jumping-off point: First, as Brian Greene makes clear in his book, there is not just one superstring theory. There is actually a set of five superstring theories, all of which are believed to be related through the all-encompassing theory known as M-Theory. It is sort-of like each of the five small theories are out on the points of a five-pointed star while M-Theory adds the middle area and links all of the points together into a single framework.

Second, while the various string theories are mathmatically describable, the various descriptions represent only the vaguest sort of approximations of the actual theory. In other words, we are not presently able to write out any of the equations for any superstring theory presently known. Instead, we can only somewhat vaguely approximate what they would be if it were possible to actually deal with the real equations.

It is well-recognized that both Quantum Mechanics (QM) and General Relativity (GR) are both well proven by scientific facts. It is also well-recognized that QM and GR are very much at odds with each other as to how their respective worldviews operate in a small-scale environment. The smaller the distance scale, the more QM tends towards unimaginable chaos. Most physicists rebel at the thought of this sort of "unimaginable chaos" having a real existence. So, they searched and they searched for some answer which would explain away the unimaginable chaos of QM at the small-scale distances.

Superstring theory achieves that particular goal. It unifies QM and GR in such a way as to define a "shortest possible length" for any "real" phenomena. That "shortest possible length" is approximately the Planck length. Within the world of superstrings, reducing the distance scale to less than the Planck length results in the phenomema progressing in such a way as if the distance scale were actually increasing. In other words, superstring theory turns the distance scale into a hyperbolic function which reaches its smallest scale at the Planck length.

So, on the one hand, there have been several thousand papers written about various aspects of superstring theory. And superstring theory has been correlated with QM and GR in such a way as to tie the whole thing up in a nice-looking overall "Theory of Everything." I'm reasonably certain that this is the "verification" which scientists felt was sufficient to raise the "hypothesis" up to the level of a "theory." After several thousand papers failing to disprove superstring theory, they must have felt that this was verification enough to rate the promotion to a "theory." (One must also realize that the most recent version called "superstring theory" or "M-Theory" is at least the third major version of string theory; string theory was killed and resurrected at least twice. So, once again, they probably feel that this qualifies it as "standing up to criticism," or whatever else is necessary to merit the "theory" designation.)

So, "superstring theory" does appear to be "a plausible or scientifically acceptable general principle or body of principles offered to explain" the prior disconnect between the worldviews of QM and GR. This qualifies it as a "theory," at least by my dictionary......

=====

On your final question, if evidence supports all five points of the star (as per the above description of "M-Theory"), then I believe that each "hypothesis" is, in fact, supported by that evidence. In other words, evidence need not be exclusive to any given hypothesis in order for it to be supportive of that hypothesis.

But of course, at some point, science will wish to understand the various competing theories (or hypotheses) well enough to be able to design an experiment which will falsify all but one of the multiple theories (or hypotheses) and leave (hopefully just) one remaining as the only possible "true" one.

=====

Finally, I think that tronvillain was wrong when he said that "A hypothesis does not change into a theory." In fact, that is exactly what happens. When a hypothesis become accepted by the scientific community, it becomes a theory.

However, I believe that the acceptance can come for reasons other than evidence, and that this is exactly what has happened with respect to superstring theory. After all, we still don't really know exactly what superstring theory is because we still can't even write down the relevant equations! So, how could we possibly have really confirmed a theory with clear evidence if we can't even write the theory down in mathmatical language?

== Bill

Bill

A bit more on this distinction:

tronvillain

*shrugs* I don't consider those synonyms at all, but I suppose they could be under certain definitions. They just aren't the definitions I use.

Oolon Colluphid

That's a new one on me too! You sure you're not confusing facts -- empirical data -- with the hypotheses and theories we use to explain them?

Ref strings, I guess there is enough evidence for it to gain theory status... provided deductive maths counts as evidence!

Cheers, Oolon

tronvillain

How I think of a theory:
Out of curiosity, can anyone think of an something initially called a hypothesis becoming a theory simply by virtue of confirming evidence?

bluefugue

Is not empirical verification still the ultimate goal, though? I mean, presumably Greene and his ilk would very much like to find a way to conduct experiments that would test string theory?

I recall in his book "The End of Science," John Horgan accuses folks like Greene and Hawking of practicing "ironic science" because their ponderings are increasingly divorced from direct experimental testing. I certainly don't agree with Horgan's claim (I don't know enough to agree or disagree with it, to be honest), but it's an opinion...