I think you may misunderstand how theoretical physics works. Theorists devise competing theories to explain existing evidence, while making some as-yet unproven predictions. Experimentalists then test those predictions using both controlled experiments (e.g. LHC) and observations (e.g. Hubble). Theories whose predictions were invalidated are then revised or discarded (e.g. Luminiferous Aether[0]).

That is the very definition of the scientific method, not at all "unprovable" as you claimed. The same process produced the current theory of evolution, and is just not as far along in the case of string theory. Seeing conflict in science and declaring it to be religious is like seeing a half-assembled car in a factory and declaring it to be undriveable. You are taking a snapshot and ignoring the process.

A note about the word "theory": it doesn't mean "untested"; it's the same sense as "music theory", that is, a comprehensive body of explanatory knowledge about a subject.

[0] http://en.wikipedia.org/wiki/Luminiferous_aether

>I think you may misunderstand how theoretical physics works. Theorists devise competing theories to explain existing evidence, while making some as-yet unproven predictions. Experimentalists then test those predictions using both controlled experiments...

I think you're misunderstanding what the OP is saying. He is saying that many physicists consider string theory to be "untestable", not just that it hasn't yet been tested. He's saying that given the limits of reality it is likely unfalsifiable, and therefore it could be argued that string theory is outside of the realm of science.

Any theory that is untestable is meaningless and will be relegated to math for math's sake (which is also an incredibly valuable endeavor) or abandoned. So if string theory turns out to be untestable, it will find the same end. In fact theorists even study theories they have contrived to be knowingly unrepresentative of reality, for the math.

How long until they are relegated to math for math's sake? Is there a finite period. Why are they even theories[1] when they are yet to be proven?

[1]: https://en.wikipedia.org/wiki/Theory

Science works by evidence-driven consensus. There is no central ruling body that decides what "science" is, or how long to keep working on it. Different scientists make proposals to different institutions, and those institutions decide what to fund. As long as there are tenured researchers or funding institutions that want to work on string theory, there will be people working on string theory. But if research into string theory stops yielding interesting results, those still researching it will get fewer publications and citations and they will change to a more promising line of research.

Also, scientific theories are never "proven", and the word "theory" doesn't have any sort of officialness to it, as is clear in the Wikipedia article you linked. To be a "theory", a set of ideas basically has to be comprehensive, logically sound, agree with existing evidence, and generally useful for something.

Experiments and consensus determine which theories are most promising for continued research and experimentation. In a given domain of science, the simplest theory (see Occam's razor[0]), with the best agreement with available evidence, with the most useful predictions, and with the greatest consensus, might be considered the "accepted" theory of the day. An accepted theory can be displaced if new evidence does not match predictions made by the theory and the theory cannot adapt to the new evidence, or if a new theory provides better agreement with evidence while being simpler or more predictive.

Science as a process transcends human lifetimes. Sometimes it takes a new generation of scientists to see past the blind spots of a previous generation, but that doesn't mean that the previous generation wasn't doing "science".