Sometimes I hear people cite their belief in modern science as a reason to reject belief in God. One of the many reasons this perplexes me is that scientific findings, particularly in cosmology, give us strong evidence that there is a fine-tuning creator.

MIT’s Roger White summarizes these findings as follows:

“The inhabitability of our universe depends on the precise adjustment of what seem to be arbitrary, contingent features. Had the boundary conditions in the initial seconds of the big bang, and the values of various fundamental constants differed ever so slightly we would not have had anything like a stable universe in which life could evolve. In the space of possible outcomes of a big bang, only the tiniest region consists of universes capable of sustaining life. Most either last only a few seconds, or contain no stable elements or consist of nothing but black holes.”1

imagesWhite says that the situation is analogous to the following: nuclear bombs are connected to a high security combination lock, such that dozens of dials have to be adjusted with extreme precision to avoid detonating the bombs. Had any one dial differed ever so slightly from its actual position, the world would have been destroyed. In the absence of an explanation of why the dials were adjusted as they were (suppose they had been spun at random) we would find it astonishing that we were here to consider the matter.

Standard confirmation theory used by scientists and philosophers of probability says that evidence E confirms (i.e., gives evidence for) a hypothesis H (holding fixed all other background knowledge) if and only if P(H | E) > P(H).

Let LP = the universe is life-permitting.

Let T (for Theism) = there exists a God who fine-tuned the cosmological constants for the purpose of creating life.

Then Pr(LP | T) ≈ 1, and Pr(LP) is low, so Pr(LP | T)*Pr(T)/Pr(LP) > Pr(T).

By Bayes’ Rule, then, Pr(T | LP) > Pr(T). In other words, the fact that our cosmological constants are life-permitting gives evidence for the existence of a fine-tuning God.

Many people have tried to reject this conclusion by arguing that LP also gives evidence for some version of the Many Universe Hypothesis. This is supposed to help because if our universe is just one of very many universes, randomly varying in their initial conditions and fundamental constants, then it is to be expected that at least one of them is life-permitting, that is Pr(LP | M) is high. Add to this the fact that we could only find ourselves in a life-permitting universe and we seem to have satisfyingly accounted for what at first seemed amazing.

However, as White shows, the fact that our universe is fine-tuned does not, in itself, gives us any reason to suppose that there are universes other than ours. To argue that it does is akin to someone rolling double sixes and arguing that this makes it more likely that there exist 35 adjacent rooms in which other people all rolled other pairs of dice. Or someone concluding from being dealt a royal flush that there must have been millions of other hands dealt from that deck before she got there. This way of reasoning is clearly fallacious, and stems from a failure to distinguish between the occurrence of some unlikely event and a particular unlikely event. The existence of many universes might make it more likely that some universe is life-permitting, but it doesn’t make it more likely that ours is. That is, Pr(LP | M) is still low; as White says, Pr(LP | M) ≈ Pr(LP). It follows from Bayes’ Rule that Pr(M | LP) ≈ Pr(M). So the fact that our universe is life-permitting doesn’t give us evidence for the Many Universe Hypothesis. It does, however, give us evidence for a fine-tuning God.

Nathan Otey ’15 is a philosophy concentrator in Pforzheimer House.

  1. White, Roger (2000). Fine-Tuning and Multiple Universes. Noȗs 34:2, pp. 260–276.