The anthropic principle and why it's not science

The anthropic principle (AP) has a number of forms and variants, but in essence, it connects the overall form and evolution of the universe with the existence of life, the existence of intelligent life, and/or with the existence of us. Stated as an empirical fact, we are here, and all the preconditions for our existence obviously place a constraint on theories we might invent to explain the universe. This version of the AP seems like an empty truism, but, like all truisms, it is true. The critical point is that such a meaningful principle is a statement about theories or explanations, not about universes. Then at least, the principle is applied in a way that be verified or falsified. Falsifiability is critical: only if your theory can be proven wrong, can you really be sure that it's right.

Starting in the 1970s, a different and nonscientific form of the AP gained popularity in some quarters as a statement about possible universes, rather than theories. One form, vaguely theistic or at least teleological, states that the universe takes on the form it does so that we can be here. Another form, recently popular among string theorists unable to sort through the apparently very large or even infinite number of possible string theories, states that every string theory is realized in some (unobservable) universe, a small piece of a multiverse. Yet another form posits the existence of new universes evolving from old ones and "selected" by the presence of life.

A number of well-known scientists have tried to claim successfully using anthropic reasoning to arrive at nontrivial conclusions, but none of the examples presented really do that. They use some general property of the cosmos (the presence of carbon, the age of the universe, etc.) to arrive at these conclusions, and none of them require life, much less intelligent life, as a necessary assumption. The originators of the anthropic principle had more explicitly teleological or theological goals in mind, and they were at least honest about the role that religious faith or teleology played in their thinking. (Teleology might prove correct, after all - we don't know enough now to say.) The most recent version is driven by another faith, faith in string theory. The requirement that string theory "must" be right, combined with the failure to produce any complete, consistent, and predictive string theories, has led theoretical physicists to this impasse.

Some have even claimed that we must accept anthropic reasoning or be doomed to accepting Intelligent Design (ID) theory. This is strictly a tempest in a teapot: ID is a type of AP reasoning and, like all such anthropic thinking, has no place in science. We have no alternative universes to compare ours to, so the endeavor is futile.

Besides the lack of observability, a more subtle problem invalidates the AP. Lacking a complete theory that could specify and enumerate possible universes, we have no way of identifying and counting what these universes could be. As mathematicians say, we lack a "measure on the space" of possibilities. Our universe could be one of a large number of random possibilities. ("Random" means: all possibilities are equally likely.) But we don't know the possibilities - we have no non-repetitive and exhaustive list, and no way to assign probabilities. No one knows whether our universe is "likely," or even what that would mean.

It's a sign of how misguided AP advocates are when you consider that, in various forms, it's used by religious fundamentalists, teleological types, and secular academic physicists, all to rescue their respective pet theories from being "not even wrong." The fact of our existence tells you that the Universe has to have a certain age, a certain structure, and the presence of life on one planet that we know of. Any theory not compatible with these facts is ruled out - that's it. No multiverses, no alternate realities, and so on. Whether it proves the existence of G-d is not something rationally decidable. That's why religion involves faith.

(I'm not arguing here against religious faith - humans couldn't live without faith in something: faith in G-d, faith in the future, faith in reason, and so on. I'm just saying it's not definitive and precise knowledge based on experience, which is, after all, what science is about.)

In brief, appeal to the AP and multiverses is not science. We know for sure what we know about the world based on the evidence of the senses and the reasoning based on that evidence. Although scientists use technical and specialized concepts and instruments to pursue knowledge, at the end of the day, they're operating with the same framework as everyone else. If the anthropic principle is about possible universes, we've left the realm of science for speculative metaphyiscs, theology, or science fiction.

Why is this happening?

The current invocation of the anthropic principle is a desperate fad among string theorists to evade the fact that no unique string theory has turned up, and there are no mathematical demonstrations of a unique or even set of realistic solutions to these non-existent "theories." Of all possible theories, they say, all are real, "somewhere". This is nonsense. Any outsider would have to rub his eyes in astonishment to see respected and talented scientific figures playing such games.

POSTSCRIPT: The deadend failure of string theory, what most theorists have devoted the last generation of theoretical physics to, drives this latest outbreak of anthropic lunacy. Theorists want to continue believing in it, even though it's obviously going nowhere. A crisp and illuminating new book by Lee Smolin (himself a quantum gravity theorist) politely but decisively dismantles string theory claims and explains not only what's gone wrong in physics, but why classical scientific self-correction mechanisms have been short-circuited in the postmodern university. I'll discuss Smolin's book in an upcoming posting.