The other day I came across an old clipping from the December 2005 issue of Physics World. It’s from an article called What will they think in 2105? looking forward from 2005 at likely developments in the next 100 years of physics, given the context of the centenary of Einstein’s “year of miracles” (1905) in which he came up with, among other things, Special Relativity which I start teaching today.
The article asks what present-day discoveries would be remembered in a hundred years. Many of those asked the question said string theory. My response was somewhat less enthusiastic:
I got quite a lot of stick at the time from senior physicists for this statement! My use of the phrase “dead again” was based on the observation that the popularity of string theory has waxed and waned several times over the years. It may not have died in 2015 as I predicted, but it does seem to me to be in a moribund state, in terms of its impact (or lack thereof) on physics.
I’m mindful of the fact that many mathematicians think string theory is great. I’ve had it pointed out to me that it has a really big influence on for example geometry, especially non-commutative geometry, and even some number theory research in the past few decades. It has even inspired work that has led to Fields medals. That’s all very well and good, but it’s not physics. It’s mathematics.
Of course physicists have long relied on mathematics for the formulation of theoretical ideas. Riemannian geometry was `just’ mathematics before its ideas began to be used in the formulation of the general theory of relativity, a theory that has since been subjected to numerous experimental tests. It may be the case that string theory will at some point provide us with predictions that enable it to be tested in the way that general relativity did. But it hasn’t done that yet and until it does it is not a scientifically valid physical theory.
I remember a quote from Alfred North Whitehead that I put in my PhD DPhil thesis many years ago. I wasn’t thinking of string theory at the time, but it seems relevant:
There is no more common error that to assume that, because prolonged and accurate mathematical calculations have been made, the application of the result to some fact of nature is absolutely certain.
My problem is not with string theory itself but with the fact that so many string theorists have become so attached to it that it has become a universe in its own right, with very little to do with the natural universe which is – or at least used to be – the subject of theoretical physics. I find it quite alarming, actually, that in the world outside academia you will find many people who think theoretical physics and string theory are more-or-less synonymous.
The most disturbing manifestation of this tendency is the lack of interest shown by some exponents of string theory in the issue of whether or not it is testable. By this I don’t mean whether we have the technology at the moment to test it (which we clearly don’t). After all, many predictions of the standard model of particle physics had to wait decades before accelerators got big enough to reach the required energies. The question is whether string theory can be testable in principle, and surely this is something any physicist worthy of the name should consider to be of fundamental importance?
In the Dark 