“Atheists have no ultimate explanation for how the universe is. They have to leave it as a random occurrence.”

There’s another common charge laid against the godless. And for naturalists, the accusation rings true. This need not, however, be a weakness.

The ultimate explanations of traditional theism do not work. They all come down to “God did it.” There is nothing intrinsically wrong with such a claim; if we had some independent idea of divine intentions or if we found a pattern of observations best explained by intelligent design, it could even help us make sense of the kind of universe we inhabit. But this is not how things stand. There are conservative ways of saying “God did it,” but they invariably make false claims. For example, assertions that God acted in history, or that living things show evidence for intelligent design rather than evolution. Sometimes “God did it” is a complete non-explanation, as with claims that the universe is fine-tuned for life and that this indicates intelligent design. It adds nothing new to our knowledge, saying only that a certain puzzle is solved by invoking The Solver Of All Puzzles. Then there are more liberal approaches, which usually translate as “God may well have done it, so we’ll take it as God did it.” These are annoyances, not arguments.

Still, naturalists also end our explanations somewhere, even if we dislike stopping at the Explainer Of All Things. Sometimes explanations stop when we say we have no idea, and don’t think anyone else has either. Sometimes we can speculate, or issue promissory notes about naturalistic solutions to a puzzle. Since the history of science has plenty of failed research programs as well as successes, some of these promissory notes will turn out to be less than what we first advertised. (Behaviorism comes to mind.) But then, all we can do is remind ourselves that we are fallible, and move on to the current best prospect for making cognitive progress, if there is any.

Sometimes we admit that things are confusing. The nature of dark energy, for example, is a head-scratcher. Knowing more about the universe—that there is such a thing as dark energy—has in some respects made cosmologists feel that they know less about the universe rather than more. If we include dark matter, we do not even know what over 95% of our universe is.

Sometimes we bite the bullet and say yes, it looks like certain things are indeed random. Modern physics is full of examples. And in such cases, we have good reasons (though never infallible reasons) to invoke randomness. I like to argue that today, a science-minded nonbeliever has to take randomness very seriously, making it a centerpiece of how they describe the world. Intelligent design proponents regularly accuse scientists of relying on a “chance of the gaps.” But there is a difference. Randomness is where pattern-finding comes to an end. A God of the gaps, in contrast, is an illegitimate extension of anthropomorphism. It brings in divine purposes without an adequate demonstration of a particular pattern that might be a signature of an intelligent agent.

There are lots of mysteries. Given our limitations, and especially how we will almost certainly always will be in a position of extrapolating from a finite amount of information, naturalists especially are aware of how we have to live with uncertainty. We are skeptical, however, of attempts to convert mystery into Mystery with a supernatural valence. Talk of capitalized Mystery, it seems, short-circuits what could be a sober acknowledgment of limitations. It claims knowledge, or perhaps a hint or a feeling of knowledge, where ignorance is a more accurate understanding of our situation.

But all this, as always, will be to the point only if we are concerned about achieving the most reliable broadly-scientific description of the world that we can. For most of us, an accurate understanding of nature is not a commanding interest—it is something that is of interest only so far as it serves our pursuit of other interests, often linked to biological and cultural reproduction.

In that case, perhaps the temptation to look for ultimate explanations and to be prematurely satisfied with “God did it” is not a mistake. After all, if religion for most people is about pragmatically coping with life, maybe short-circuiting certain kinds of inquiry is a good idea. After all, “God did it” is often associated with a kind of cosmic optimism, or at least a conviction of a humanly-meaningful purpose behind the seemingly mindless workings of the universe. If satisfaction with “God did it” prevents us from wasting time on questions with no immediate pragmatic significance in terms of the interests of everyday life, that makes it useful. If it is also associated with a kind of cosmic optimism, even better.

So, yes, the naturalistic variety of atheists do not have much in the way of ultimate explanations, at least not beyond those areas where we think we run up against fundamental physical randomness. And that, in a cognitive context, is a good thing.

But then perhaps the theistic accusation incorporates a legitimate worry. Perhaps someone stating that they are satisfied with “God did it” signals that they care about pragmatic questions rather than philosophical or scientific puzzles. They signal loyalty to a particular moral order. So maybe theists questioning nonbelievers about ultimate explanations, like so many other questions, translates into asking whether atheists can be trusted.