This is the weekly Q & A blog post by our Research Professor in Philosophy, Dr. William Lane Craig.


First of all I would like to say thanks for the great job you are doing and for the big influence you have upon people's lives both spiritually and intellectually.

My question isn't really mine, actually I found it in one of the reasonable faith forums, and I think it's a very good question that intrigues me since it was raised in your debate with Kevin Scharp. I would like to look at your take on the divine psychology objection proposed by Scharp more closely. Here's the question as it was presented in the forum:

“Dr. Craig recently debated Dr. Kevin Scharp on the Veritas Forum. One very interesting objection that Dr. Scharp raised to the fine tuning argument is that it appeals to divine psychology to support the premise that design is more probable than chance and necessity.

“The basic point was this: Dr. Craig in his refutation to the probabilistic version of the Problem of Evil objects that since we can't have any confidence on the probability that god might decide to permit evil, we're not warranted in stating that it is improbable that god would decide to permit evil. That is to say, that we can't make probability judgements on god's decision making process, i.e. his psychology. The problem of evil argument therefore fails, due to our inability to know with any confidence what god may or may not choose to do in any given situation. I.e. because of an appeal to divine psychology.

“But by the same token, the Fine Tuning Argument (FTA) claims that chance and necessity are less probable than design. But the hidden assumption is that there is not an improbable chance that god would decide to fine tune a universe, which, according to Dr. Craig we cannot have any confidence in asserting (because it's an appeal to divine psychology).

“I felt Dr. Craig failed to respond adequately to this in the debate. All he said was that god could have good reasons to design the universe and that it's the Atheists burden of proof to show that it's less probable than chance and necessity. But of course, it's the person making the claim which has the burden of proof, in this case the proponent of the FTA. Just because a god could have good reasons to design a universe does nothing to show that it's more probable than chance (after all, a god could also have NO reason to design the universe, making the probability less than chance).

“I've yet to hear a good refutation of this objection. Can anyone do a better job than Dr. Craig in this debate?"

Thank you for your time.



Dr. William Lane Craig’s Response

Dr. William Lane Craig

As I remarked at the time, the dialogue with Kevin Scharp is a rich trove of interesting topics for discussion. Be sure to consult the transcript, which has now been posted. Thanks for the opportunity to discuss one of these topics, his objection from divine psychology.

What is meant in this context by “divine psychology”? It refers to reasoned conjectures about what God would likely do in a certain situation. As such it is unobjectionable. We make such surmises all the time about human persons. For example, I think it’s pretty likely that if Hillary Clinton were elected President of the United States, she would appoint a Supreme Court justice that is far to the left of Antonin Scalia. I couldn’t say specifically whom she would appoint, but I think it’s a pretty safe bet that whomever she chose, he would be far more liberal than Scalia. Of course, the more we know about a person, the better we’re able to make such reasoned conjectures. I am less certain, for instance, what sort of justice Donald Trump would appoint if elected! Still, given what I know about current politics, I’d say it’s not improbable that he would appoint a conservative to the Court. Obviously, these sorts of conjectures do not assign specific probability values to the actions but rest content with vague approximations. The question will be whether we can make such reasoned conjectures about certain of God’s actions.

Now I trust that it is clear that, contrary to Kevin Scharp, divine psychology plays no role in most of the arguments for God’s existence that I have defended. But it does, as I acknowledge in our dialogue, play a role in the design argument from fine-tuning. That is because I claimed that given the incomprehensible improbability of fine-tuning on atheism, it’s a pretty safe bet that fine-tuning is more probable on theism (or design) than on atheism. Robin Collins, the finest exponent of the fine-tuning argument today, is fully aware of the role played by divine psychology in the argument and addresses it in his article in the Blackwell Companion for Natural Theology, which I commend to everyone interested in this issue.

Collins argues that, given the existence of embodied conscious agents (Life) and our background information k' (including the laws of nature), theism is more probable than atheism [P(Theism | Life & k') > P(Atheism | Life & k')]. Key to his argument is assessing the comparative likelihoods of Life relative to theism and the background information [P(Life | Theism & k')] as opposed to Life relative to atheism and the background information [P(Life | Atheism & k')].

The first point to notice about Collins’ formulation of the argument is that it does not require us to assign a hypothetical likelihood to God’s creating a finely-tuned universe [P(FT | Theism & k')]. Rather one is asked to assign a likelihood to the existence of embodied, conscious agents (Life) given God’s existence and the laws of nature described in k'. Thus, one need not worry about the probability that God would create a finely-tuned universe; that probability plays no role in Collins’ formulation of the argument.

The second point to note is that Collins’ argument does not require that it be at all probable that God would create a universe with Life. Rather Collins’ claim is that however low the probability assigned to Life relative to theism and the laws of nature, it’s not going to be as low as the probability of Life on atheism and the laws of nature. That is to say, P(Life | Theism & k') >> P(Life | Atheism & k'). That point seems to me eminently reasonable, if not obvious. The improbability of Life given atheism and nature’s laws is just inconceivably large; I think people sometimes become numb at the odds and forget that we are talking here of incomprehensible numbers. By contrast, if God exists, I can’t see any reason at all to think that it would be improbable that He should create Life, much less more improbable than Life on atheism and nature’s laws. As I said in the dialogue, God could well have good reasons for creating a universe with embodied, conscious agents—for example, that they might come to know the great good of a personal relationship with God. The ball is now in the atheist’s court: he needs to give us some reason to think that such a surmise is absurdly improbable. But he can’t do that.

Thus, the appeal to divine psychology in the fine-tuning argument is pretty minimalist. The argument doesn’t assert that it’s probable that God would create a fine-tuned universe or even that God would create a universe with Life. It just claims that the probability of Life, given that God exists, is not absurdly small. That claim is justified in light of the plausible reasons God might have for wanting a universe with Life. We needn’t claim that we know what those reasons actually are, but just that it’s not absurdly improbable that He have such reasons.

But if we allow this appeal to divine psychology in the fine-tuning argument, will that undermine my response to the probabilistic version of the problem of evil, as Kevin Scharp claimed? No, not all, for my response to the problem of evil is not based, as Scharp imagined, on scepticism about divine psychology, an inability on our part to say what God would do. Look at my discussion of the problem in Philosophical Foundations for a Christian Worldview. There I make three points:

1. Relative to the full scope of the evidence God's existence is probable.

2. We are not in a good position to assess with confidence the probability that God has no morally sufficient reasons for permitting the evils that occur.

3. Christian theism entails doctrines that increase the probability of the co-existence of God and evil.

None of these points rests on scepticism about divine psychology.

Point (1) is an appeal to the arguments of natural theology, which outweigh any improbability that evil might throw on God’s existence.

Point (2) is based, not upon our ignorance of divine psychology, but rather upon our cognitive limitations, such as our confinement in space and time, that prevent our discerning the ultimate outcome of specific events. As I wrote,

The brutal murder of an innocent man or a child's dying of leukemia could send a ripple effect through history so that God's morally sufficient reason for permitting it might not emerge until centuries later or perhaps in another country. . . . this is not to appeal to mystery, but rather to point to the inherent cognitive limitations that frustrate attempts to say that it is improbable that God has a morally sufficient reason for permitting some particular evil. Ironically, in other contexts non-believers recognize these cognitive limitations. One of the most damaging objections to utilitarian ethical theory, for example, is that it is quite simply impossible for us to estimate which action that we might perform will ultimately lead to the greatest amount of happiness or pleasure in the world. Because of our cognitive limitations, actions which appear disastrous in the short term may redound to the greatest good, while some short term boon may issue in untold misery. Once we contemplate God's providence over the whole of history, than it becomes evident how hopeless it is for limited observers to speculate on the probability of God's having morally sufficient reasons for the evils that we see. We are simply not in a good position to assess such probabilities with any confidence.

This appeal to our cognitive limitations has nothing more to do with divine psychology than does the critique of utilitarian ethics. Dr. Scharp has simply misunderstood the point.

Point (3) appropriates Marilyn Adams’ insight that the probability of evil may be greater relative to theism plus some auxiliary hypotheses than it is relative to theism alone [P(Evil | Theism + auxiliary hypotheses) > P(Evil | Theism)]. I offer four auxiliary hypotheses which are part of Christian theism, including God’s plan to bring about people’s knowledge of God. I even give empirical evidence that evil actually contributes to a greater knowledge of God. So evil turns out to be less improbable on Christian theism than on a sort of bare-boned theism. To be sure, the addition of these auxiliary hypotheses to theism will make Christian theism less probable than simple theism; but I offer a robust evidential case for the truth of Christian theism. So the atheist has to show either that my case for Christian theism fails or else that evil is improbable even given my auxiliary hypotheses, which, I claim, he cannot do.

I trust that you can see that my response to the probabilistic version of the problem of evil makes no appeal to our inability to “make probability judgements on god's decision making process, i.e. his psychology.”

So making reasoned conjectures about what God would do is a legitimate endeavor, however vague our conclusions might be. Such an endeavor plays a modest and unobjectionable role in the argument from fine-tuning, and my response to the probabilistic version of the problem of evil does not rest on a rejection of divine psychology.

This post and other resources are available on Dr. William Lane Craig's website:

Learn more about Dr. Craig’s book, A Reasonable Response.