Beyond the death of God - scientific challenges to atheism

National Review, May 6, 1996 by Patrick Glynn

The first line of defense has been logical hair-splitting. The Anthropic Principle has been said to be a tautology, since we could not expect to observe a universe that was not capable of producing us. This is the purest sophistry, since it pretends to ignore the surprise we register at stumbling upon such a multitude of coincidences. It evades the question rather than attempting to answer it. It is as if to say, We would not be observing the elephant we see standing in our living room if the elephant had not gotten there in the first place. True, but nonsense. This essentially tautological formulation sometimes has been called the "weak Anthropic Principle." It is "weak" in more ways than one.

Coincidences do not prove, but they suggest, sometimes powerfully. Indeed, our ability to detect and infer from coincidences plays a critical role in our basic capacity to interpret and find our way in the world around us. Without such ability, life would be for us a tale told by an idiot. If a detective investigating a crime, for example, stumbles on a series of mysterious coincidences, he will look for a human hand behind them. The hand may not be there, or may not be found. But the presumption will favor the existence of such human intervention, and a good detective will follow the trail of evidence until the supposition is disproved or proved. He will not take refuge in platitudes to the effect that "coincidences are a part of life."

The second line of defense has been a resort to imagination or fantasy. Our universe has been said to be simply one of billions of universes existing either in sequence or in parallel -- none of which, of course, save our own, we can detect. Given the supposed existence of these billions of universes, the fact that one (ours) happened to hit on the precise combination of values and relationships to produce life would not be surprising. This is actually a very old, pre-scientific argument, a traditional mainstay of the atheist case -- invoked by Diderot and Hume and dating back to the Epicureans of Roman times -- that, given an infinite duration, nature, acting randomly, would eventually assemble the order we see around us. A monkey at a word processor, over infinity, would eventually type the works of Shakespeare, or so it is supposed.

Each of these proposals is a grand balloon of speculation anchored to a tiny grain of scientific hypothesis. To be sure, scientists are not yet certain whether the present universe is a one-time event or whether, at a certain point, it collapses back in on itself, only to begin the cycle of expansion anew. That there is an oscillating sequence of universes remains at least theoretically possible, though the present evidence suggests otherwise. The somewhat different "parallel universes" idea rests on a highly theoretical proposal advanced by Hugh Everett in 1957 to solve the "problem of measurement" in quantum mechanics. Everett proposed that all the possibilities implicit in matter before it is actually observed -- before, for example, light in its fuzzy "wave" state is observed and collapses into a photon particle -- actually exist in reality. Everett imagined that, at each observation, reality was infinitely branching out. My eye would observe one photon with certain properties in this universe, while, in effect, copies of "me" would be observing photons with the other possible properties in a series of parallel universes ad infinitum. It is a powerful speculation, and a theoretically interesting way of posing the genuine paradoxes implicit in the quantum theory -- but it is just that, a speculation. There appears to be no way that such parallel universes could be detected, even in theory; certainly, no one has stumbled on them so far.