In my former life as an academic, I studied and taught the philosophy of science. It was a seemingly useless pursuit, but it is suddenly topical in 2006.
There is a war going on about string theory, and it’s getting nasty.
Physicists are regarded as the high-priests of science. They deal with the kind of foundational issues that make even things like evolution and the Oedipus complex seem trivial and very, very earthly and material. And they have deserved this high regard. For several hundred years, they have had a constant stream of wondrous discovery. But now, in the last 25 years or so, things have dried up, and only the most technical of discoveries have been made.
In this period, string theory has minted more PhDs than anything else, and in the exclusive club of particle physics, it’s the only road to full professorship most places. But, its critics say in two recent powerful books, that it is not science at all, because it fails to produce and falsifiable predictions.
This is an allusion to the metascience of Karl Popper, who defined science thusly. Of course, science existed a long time before Popper. I think his definition is not useless, but not the last word. Consider for a second Fermat’s last theorem. This went unproven for hundreds of years. But at the same time, it was never falsified. In the meantime, you could have called it Fermat’s law. Epistemically, the failure to falsify Fermat’s theorem was not a proof. (To be fair, empirical matters are not subject to apriori affirmative proof this way–it’s just an illustration.)
In other words, science is not the set of those propositions that remained unfalsified, even if the test is possible. That universe of propositions is infinite, and much of it is tautological. Science exists on the edge of that universe, somewhere in the space of that universe. Given that these propositions, to the extent we encounter them, have some language component too.
Newton’s explanation of method remains, in my researches, the most subtle and effective in all of philosophy for creating knowledge that is epistemically reliable. Newton’s idea is that you build and test theories, and, once proven, you stick with them even in the face of countervailing evidence, until it can’t stand anymore.
So, by this method, adding epicycles to correct for the orbit of Mars in the Ptolemaic system is ok, but when the predictive power of that model completely falls apart, it’s time to move on. This is vastly oversimplified, but I think it’s enough to distinguish from Popper.
No one doubts the predictive power of quantum theory. It has been held together with very little tweaking for a long time. But, let’s face it. The standard model of quantum theory looks a lot like Ptolemy’s universe. The empirical constants that are required to make it work lack any explanatory value, and are almost epicycle like kludges. On top of that, we have to accept that these particles have certain essences and innate urges to do things, just like Aristotelian physics.
Ptolemy could tell you where Mars would be, but not why, at least not without invoking the supernatural. Quantum mechanics can tell you a lot, but it leaves you with “just is” a lot of the time.
String theory may not have produced any falsifiable propositions, but I don’t think it’s empty research. Even if it does end being a dead-end, a dead-end which might cause a rethinking of over 100 years of physics, that alone will provide value.
One empirical fact I might note as a former historian and philospher of science: scientists spend little time discussing method or meta-science when things are going well.