Is physics turning into philosophy?

Physics is, by most people's--especially physicists'--accounts, the queen of the sciences. Philosophy is, according to some physicists (for example, Steven Weinberg, in his Dreams of a Final Theory), a useless, perhaps even dangerous, enterprise, because it can slow scientific progress. Yet, two essays published recently in Nature argue that physics is in danger of becoming "just" philosophy, or worse, indistinguishable from nonscientific notions such as the anthropic principle, the idea that the universe is fine-tuned to allow for the appearance of life.

George Ellis, reviewing Leonard Susskind's The Cosmic Landscape (Nature 438: 739-740), says that "heavyweight physicists" are claiming to have proven the existence of parallel universes "even though there is no chance of observing them." Michael Atiyah, commenting on Lawrence Krauss's Hiding in the Mirror (Nature 438: 1081-1082), observes that there is a danger of a "mathematical take-over" of physics, leading to speculations that, while mathematically elegant, are "far removed, or even alien to, physical reality."

What is going on here? According to Ellis, currently fashionable ideas in physics and cosmology, such as multi-verses or string theories, are characteristic of a discipline where "faith in theory tends to trump evidence." Take string theory, for example: it is a mathematically beautiful construct that has the potential of unifying general relativity and quantum mechanics. The problem is that--at the moment--it also is entirely untestable empirically, which makes it an interesting philosophical position but not science.

Of course, if one begins with the assumption that physics is not only the queen of the sciences but that science is in turn the sovereign of all intellectual disciplines, then it is an insufferable insult for string theory to be "downgraded" to mere philosophy. Then again, this is nothing new. Philosophy has often been the placeholder for areas of intellectual inquiry that have subsequently moved to the domain of science. Physics itself, up until Descartes, was part of natural philosophy. Indeed, Descartes considered himself a scientist, not a philosopher, and it is obvious only in hindsight that he lived at the time of separation between the two modes of inquiry. Psychology was once a branch of philosophy, and philosophy of mind is only now slowly turning into neurobiology and cognitive science. The role of philosophy in each of these cases was to explore the intellectual terrain and clear up the conceptual mud, until ideas and techniques were mature enough to permit a shift to empirically-based inquiry.

Of course, there is no guarantee that all areas of logical and philosophical analysis will eventually turn into science, regardless of how much time or technological advance one allows. Despite the fact that biology provides some insights into, for example, both morality and aesthetics, it is hard to imagine turning these two disciplines over to science departments, because they involve human values, not just matters of fact. Similarly, it may very well be the case that we will never find a way to test the idea that there are multiple universes out there. Just because we are curious animals, there is no guarantee that nature behaves in a way that allows us the possibility to get answers to any mystery we are intrigued by.

Another way to look at the conundrum presented by some theories currently being entertained by physicists is that there is a distinction between logical and empirical possibility, and, correspondingly, between logical and empirical truth. The realm of logical/mathematical possibility (mathematics being a branch of logic) is much more vast than that of empirical possibility. For example, it is logically possible for a substance made of two atoms of hydrogen and one of oxygen to behave very differently from water, if the laws of physics were just slightly different than they are. Yet, once the laws of our universe are established, certain facts become empirically impossible. There are other things that are both logically and empirically possible, but have not been realized--for example, the existence of unicorns. At the opposite extreme, if something is logically impossible (for example, for me to be and not be myself), then it is a fortiori empirically impossible.

Strings and multiple universes are certainly logical possibilities, since they are features of mathematical theories (assuming that the math doesn't contain mistakes). What we don't know is whether they are also empirically possible and, if so, whether they have in fact been realized. It is the search for an answer to the latter two questions that distinguishes science from logic/mathematics (and philosophy). Yet, neither string theory nor the possibility of multiverses have yet crossed the conceptual line separating logical from empirical possibility. If we pretend that such a line is irrelevant to scientific practice, we risk turning physics into an exercise in faith, not much better than intelligent design "theory" or its close relative, the anthropic principle.