Big Think

Bruce Schechter on April 11, 2008, 1:50 PM

I’ve been working on a book about string theory and this question is something I’ve been wrestling with.
Taken at face value, the answer to your question is obviously that, if the ultimate theory of everything, or grand unified theory, or whatever you want to call it, is untestable, then we are stuck. If that’s the way things are then it limits what we can know scientifically. Which is annoying, but most scientists accept that there may be questions about the universe that can’t be settled by science.

The question (or one question) is, at what point do we decide that a theory is truly untestable? Until fairly recently it seemed impossible to talk about what happened before the big bang. But in the 80s the theory of inflation was developed that proposed a whole scenario, which I won’t go into here, for what happened before the big bang banged. Inflation seemed to clear up a lot of mysteries that surrounded the big bang and, better still, it made some very specific predictions that have been recently confirmed by astonishingly beautiful measurements of the so-called cosmic black body radiation. So, a question that seemed obviously beyond the reach of experiment, and therefore scientific inquiry, has proved to be experimentally accessible.

It is certainly conceivable (and people are working on this) that string theory makes predictions that can be tested by technologically conceivable experiments. Lisa Randall talks on this site about the search for extra dimensions. Finding them would buttress the ideas of string theory (though it would not prove them true. The problem of proving a theory true is rather sticky, and most people believe that it can’t be done). Other people have other ideas, and as the mathematical structure of string theory becomes better understood it might happen that it will make predictions of relationships between physical constants or new processes or whatever that can be observed. So, although it may be extremely difficult to test string theory, it is too soon to rule out the possibility of ever testing it.

But suppose that someday a brilliant theorist dreams up a test of string theory and an ingenious experimentalist figures out how to perform the test, and the prediction is verified to a hundred decimal places. Great. But string theory still makes predictions of phenomena that will probably forever be untestable. What do we say then? That string theory captures some aspects of physical reality and does so better than any other known theory? Testable predictions are considered by many to be the hallmark of a scientific theory, but what of a theory that makes both testable predictions and others that cannot be tested? What if every new test that theorists dream up is convincingly confirmed? It seems to me that it would be hard to deny that this theory captured some essential features of reality. But what to make of the untestable predictions? Would it be reasonable to believe in them, at least a little? I think so, though I’d be hard-pressed to defend them philosophically. It would be especially difficult to remain skeptical of these untestable predictions if the theory that made them, and which had also made astonishingly accurate, testable predictions, was all of a piece mathematically—no part could be tampered with without the entire edifice crumbling. If we could be sure of this (and I have no idea if we ever could be) wouldn’t testing any (unique) prediction of the theory add credence to the untestable aspects?
We are nowhere near being in this situation today, yet there are scientists that act as if we were. For example, one version of cosmic inflation predicts that our universe is just one of countless universes that float like bubbles in a vast and expanding sea. These other universes are too distant to ever be seen by observers in our universe; they are as isolated from us as the interior of a black hole. Some theorists have clutched onto this idea as an answer to a problem that string theorists have been wrestling with: the laws of physics depend upon how the extra dimensions of a 10-dimensional string curl up, and they can do so in more than 10^500 ways (nobody knows for sure, but the number seems to be getting bigger). This number is effectively infinite and is seen by some to be an embarrassment to string theory. But some physicists claim that this is instead a strength of the theory. They point to the observation that the physical constants of our universe—the strengths of the forces, the masses of the particles—seem to be sensitively tuned to make life possible. Vary any one of them by just a few percent and stars won’t form, or they’ll burn too brightly, or something else will happen that will make the existence of intelligent life unlikely. It’s hard to think of a reason why the constants should be so finely tuned for life—what does the subatomic world care about structures dozens of orders of magnitude beyond it? The answer, according to some scientists, is that we are living in just one among a vast number of universes and that the universe we occupy is as it is simply because it is the only universe beings like us could occupy. Things are as they are because how could they be any different still have parts of it asking this question?

Maybe this is how things are, but can we ever know it? There could be indirect tests but we will never actually see these other universes. This is unsatisfactory, but if it’s true, I suppose, we’ll have to live with it.

But if it’s true it is too soon to believe it. It is too soon, as Nobel laureate David Gross puts it, to throw in the towel, to give up. He points out that the history of physics contains many of examples of times when an anthropic argument could have been invoked, but would have proved wrong.

I’ve asked many physicists what could kill string theory. Experiments might. But it might also be that some day string theory will dry up on its own as theorists run out of new ideas and it is still well short of its goals. The strength of string theory and why it keeps attracting the brightest minds, is that this doesn’t seem to be happening. It continually throws off new ideas and insights into such fields as cosmology and black hole physics. If and when string theorists run out of ideas the field will wither.