This post is part of a series on the fine-tuning of the universe. Here I will respond to the work of Dr. William Lane Craig. Craig is Research Professor of Philosophy at Talbot School of Theology. He is known for his defence of arguments for the existence of God, both in philosophical journals and public debates. Here, I will respond to a point that Craig has made in response to the multiverse (or many-worlds hypothesis; James Sinclair makes a similar point in his essay in “Contending with Christianity’s Critics”):
The error that that is made by the many worlds hypothesis is that it is basically an attempt to multiply your probabilistic resources without having any justification for doing so. It’s a way of saying that the improbable roll of the dice that we have come up with is rendered probable because there have been many throws. If you’re allowed to do that, then you could explain away anything. For example, imagine a couple of card players in a west Texas saloon. And every time one of them deals, he gets four aces, and wins the game. The other guy gets outraged and says, “Tex! You’re a dirty cheater!” And old Tex says, “Well, Slim, you shouldn’t really be surprised that every time I deals I gets four aces. After all, in this infinite universe of ours there’s an infinite number of poker games goin’ on somewhere. And so chances are in some of them I gets four aces every time I deals.”
Why is Tex wrong? It’s not merely that his appeal to countless other poker games is unjustified. Even if these other poker games existed, they would still not explain his suspiciously good luck.
Suppose that Tex deals 10 times, and gets four aces each time – call this “M”, the “magic deal”. The probability of M, assuming he is dealing fairly, is approximately p(M | fair-deal) = one chance in . Now suppose that there are other poker games going on elsewhere. Then the probability that someone somewhere is seeing the magic deal is almost one. So far, so good.
As in many areas in life, the key is asking the right question. If we ask: what is the probability that someone is observing M? Answer: almost one. But if we ask: what is the probability that I (a particular observer) am observing M? Answer: one chance in . These other poker games don’t make it any more probable that I observe M. For every player who sees the magic deal, there are about poker players out there who don’t. Even with all these other poker games, it is still incredibly unlikely that Tex deals the magic deal. Given these overwhelming odds, I am justified in concluding that the hypothesis “Tex is cheating” is quite likely.
Compare this with the multiverse hypothesis: there are a vast number of other universes out there, each with different physics. Suppose that the laws of nature need to be fine-tuned for life i.e. the probability that a universe, chosen at random from among the possible universe, is life-supporting is very small. Now let’s ask our questions. What is the probability that there is a life-supporting universe out there somewhere? Answer: almost one (provided there are plenty of universes, randomly selected). What is the probability that I am observing a life-supporting universe (given that I am observing at all)? One! It is impossible for me to observe that I am in a universe that doesn’t support life. This is simply a statement of the Weak Anthropic Principle (WAP).
Here’s the point: for every poker player who observes the magic deal, there are about poker players out there who don’t. But for every observer who observes that they are in a life-permitting universe, there are exactly zero who don’t. The difference between the two cases is the existence of a selection effect (WAP) for life, which isn’t there for poker players.
Here we see the true power of the anthropic principle. On its own, it is an effete tautology. Similarly, a plethora of universes, in and of itself, cannot explain anything – contrary to Craig’s claim that “you could explain away anything” with a multiverse. But the conjunction of the two, as sample space and selection effect, has real explanatory power … if these other universes are really there.
The last point is an important one. Hypothetical universes will not do. Craig is correct is his concern that “the many worlds hypothesis is … basically an attempt to multiply your probabilistic resources without having any justification for doing so.” I have so far only argued that if these other universes existed, then they could explain why our universe is fine-tuned for life. But to be more than an ad hoc hypothesis, we would need independent reasons for believing that these universes are really there, in sufficient numbers and with sufficiently varied properties to make the existence of a life-permitting universe probable.
Apart from the above example, Craig’s work on the fine-tuning of the universe for life isn’t too bad (my next blog post notwithstanding). In particular, the other points he makes against the multiverse hypothesis deserve our consideration. His best article is “Design and the anthropic fine-tuning of the universe” in the volume “God and design: the teleological argument and modern science”, which also contains Robin Collins’ excellent “Evidence for Fine Tuning”. Both are worth a close read.
More of my posts on fine-tuning are here.