In his most recent post on the Q&A section of his website, William Lane Craig responds to an objection to his version of the fine-tuning argument. Talking about the fine-tuning argument, Tyson said:

Most places in the universe will kill life instantly – instantly! People say, ‘Oh, the forces of nature are just right for life.’ Excuse me. Just look at the volume of the universe where you can’t live. You will die instantly.”

In response to Tyson, Craig calls this objection “silly” and concludes, “Tyson doesn’t understand the fine-tuning argument.”

I don’t know if Tyson understands ‘the’ fine-tuning argument or not, but it seems to me that the universe’s hostility to life does pose a powerful objection to cosmic fine-tuning arguments for God’s existence, including the version defended by William Lane Craig. Here is his version.

(1) The fine tuning of the universe is due to either physical necessity, chance, or design.

(2) It is not due to physical necessity or chance.

(3) Therefore, it is due to design.

By “fine tuning,” Craig explains that he means the fact that the universe is “life-permitting rather than life-prohibiting.”

Craig thinks Tyson’s objection is irrelevant because it denies the fact that the universe is life-permitting. The fact that the life-permitting parameters of the universe are merely necessary, rather than sufficient, conditions for life, is a “fallacious” reason for denying that the universe is life-permitting.

Well, yes, the universe’s hostility to life is compatible with the fact that the universe is life-permitting. I don’t know if Tyson intended to make the claim that the universe’s hostility is incompatible with its life-permitting parameters or not, but it really doesn’t matter one way or the other. For by construing the “hostility objection” as the claim that the universe’s hostility contradicts its life-permitting conditions, Craig ignores a much stronger version of the objection.

As I explained almost two years ago, Craig’s version of the fine-tuning argument is a textbook example of how the distinction between deductive and inductive arguments can mask uncertainty. His argument is a valid (deductive) argument: the conclusion has to be true if both of the premises are true. The fact that the conclusion follows from the premises  however, tells us nothing about the probability of the conclusion, which Craig wrongly ignores. He seems to think that all that matters is whether the premises are more probable or not. In his words:

This is a logically valid argument. That is to say, if the two premises are true, then the conclusion necessarily follows. The only question is: are those two premises more plausibly true than false?

But the expression, “the conclusion necessarily follows,” gives the illusion of certainty, which is not warranted if at least one premise is uncertain. For example, suppose we are certain that (1) is true–i.e., we believe the probability of (1) is 100%–but only 70% confident that (2) is true. In that case, it follows that our our degree of belief in (3) should also be 70%. That is all the hostility objection needs to get going.

Assume that we are 70% confident that the life-permitting conditions of our universe are due to design. By itself, the fact that we are 70% confident that the life-permitting conditions of our universe are due to design does not justify us in being 70% confident that our universe is due to design. That would be the case only if we did not know of any other factors about our universe, besides life-permitting conditions, which were relevant to the probability that our universe was designed. But that’s false. We know much more about our universe’s habitability than the fact that it is life-permitting, however. We also know that the vast majority of our universe is hostile to life. Given that our universe is life-permitting,[1] the fact that so much of our universe is hostile to life (hereafter, “the hostility data”) is more probable on the assumption that naturalism is true than on the assumption that theism is true.

Here’s why. If we assume that naturalism is true, the fact that our universe is life-permitting gives us no reason at all to predict that the hostility data is false. In fact, naturalism, combined with the life-permitting data, predicts the hostility data because naturalism entails that nature is blind and so indifferent to our universe’s habitability.

With theism, however, things look quite different. Theism does not entail that the hostility data is false; God could have designed a universe that is overwhelmingly hostile to life. It’s also possible, and no less likely, that God could have designed a universe that is mostly habitable. But when combined with the fine-tuning data, theism not only does not predict the hostility data is true, it provides at least some reason to predict the hostility data is false.

So the hostility data is more probable on the assumption that naturalism is true than on the assumption that theism is true, and hence at least some evidence against theism. When we combine the hostility data with other data, we can formalize this argument, which I shall call the “Coarse-Tuning Argument,” as follows.

B: The Relevant Background Information

1. Our universe exists.

2. The universe is intelligible.

3. Our universe is life-permitting.

E: The Evidence to be Explained

1. So much of the universe is highly hostile to life.

2. Our universe is not teeming with life, including life much more impressive than human life.

3. The only intelligent life we know of is human.

4. Intelligent life is the result of evolution.

Logical Structure

(1) E is known to be true, i.e., Pr(E) is close to 1.

(2) Theism is not intrinsically much more probable than naturalism, i.e., Pr(|T|) is not much greater than Pr(|N|).

(3) E is more probable on the assumption that naturalism is true than the assumption that theism is true, i.e., Pr(E| N & B) > Pr(E | T & B).

(4) Other evidence held equal, theism is probably false, i.e., Pr(T | B & E) < 0.5.

The upshot is this. Even if the life-permitting data is more probable on theism than on naturalism, the hostility data is more probable on naturalism than on theism. Once the evidence about our universe’s habitability is fully stated, it’s far from obvious that it favors theism.

Therefore, far from being “silly,” I think Tyson’s observation that the universe is hostile to life is relevant to the final probability of design. Even if Tyson did not present his objection in the way this article has, Craig has failed to display philosophical charity and consider his opponent’s objection in its strongest form. For once we consider the objection in its strongest form, it becomes clear that what is “silly” is Craig’s smug reply to Tyson.

Notes

[1] By “Given that our universe is life-permitting,” I mean, “If we include ‘Our universe is life-permitting” within our background knowledge.”