Editor’s note: We are delighted to welcome Tom Gilson as a new contributor. Mr. Gilson is a senior editor at The Stream and has hosted ID the Future. He blogs at Thinking Christian.

The pseudo-scientific doctrine of methodological naturalism has caused no end of trouble. It’s time we called for a more reasonable replacement: How about methodological theism?

Actually, for anyone who’d make a suggestion like that it’s time to duck for cover instead. The scientific community would stone you for advancing an idea so loaded with metaphysical and theological presuppositions. 

Except that’s precisely the point. Even if methodological theism had nothing else to recommend it — though it does — it can at least highlight how methodological naturalism contains every bit as much non-scientific bias. 

Methodological Theism, the Straw Man Version

Methodological naturalism, the usual doctrine, tells us we should treat our pursuit of science as if nature is all there is; that we should assume there is no supernatural reality, or if there is, it is completely and forever divorced from our theories, methods, and conclusions. Methodological theism, in contrast, supposes that there is a God who acts in and through nature, as Judaism and Christianity have traditionally understood God to do.

Many will reject this out of hand. J.B.S. Haldane said it best, perhaps, in 1934: “My practice as a scientist is atheistic. That is to say, when I set up an experiment I assume that no god, angel, or devil is going to interfere with its course; and this assumption has been justified by such success as I have achieved in my professional career.” More recently, philosopher of science Robert Pennock wrote, “Without the binding assumption of uninterruptible natural law there would be absolute chaos in the scientific worldview.” Philosopher A.C. Grayling has voiced his concerns over “local suspension of the laws of nature for arbitrary reasons e.g. in answer to personal prayer, which makes a nonsense of the idea that the world or the deity is rationally comprehensible.”

Methodological Theism for Real

But in fact none of this has anything whatever to do with theism. It’s all a straw man, conveniently crafted, one suspects, to support a secular scientific view of reality. True theism actually provides stronger support than naturalism for expecting nature to behave in a regular fashion. And this is no conveniently crafted theory on the theist’s part, served up to explain God in a world of science. This theism goes back millennia.

It’s in the very earliest understandings of God as creator, to start with. Genesis tells us that God created humans to be morally significant agents. Moral agency requires regularity and predictability in nature. Suppose my hand on your shoulder gave you strong personal encouragement one day, gout the next, a winning lottery ticket the next, and bone cancer the day after. If that were so, my decision to touch you could have no moral content at all, either good or bad. I would have no way of knowing what I was doing to you!

Also from the very beginning theists have taken it that God intended humans to learn from experience. The law of reaping and sowing appears early in biblical literature — long before science took off. Humans learned, for example, that planting barley seeds yielded a barley crop, and that proper cultivation, water supply, and so on increased that crop. If experience were as irregular as Haldane supposed it would be under God, no such experience-based learning could be possible. His view of God is obviously wrong in that. And what is science, but learning from experience writ very large? 

Theism also takes it that God would want to communicate with humans through miracles from time to time. Yes, I admit it: God does step in and change things once in a great while. But it’s very, very rare. In fact there’s a communication engineering principle at play here: If there’s too much noise — meaningless irregularity — in any transmission, no signal can be discerned. It applies to God and nature as much as it does to telephone lines: If God were always jumping in, making nature behave differently all the time (as Haldane imagined), no one could tell when his intervention had significance as communication. 

But watch out: “Aha!” shouts the secularist. “You admit it! God interferes! How can we possibly do science if he gets in the way, even a little bit?!” Answer? Easy! Scientists know full well what to do with the occasional anomalous result: Cast out the data point as an outlier. Or re-run the experiment. Failing that, simply shrug your shoulders, scratch your head, say, “Hmmm, that was odd,” and go on. This happens in science. It’s not the least bit out of the ordinary. Science survives anyway. And remember: every theist believes true miracles are very, very rare. 

So theism provides for a reasoned expectation that nature will behave regularly virtually all the time; certainly with enough regularity to allow science to succeed. In this expectation, theism actually outdoes naturalism; for naturalism knows no reason why nature should behave regularly. It just does, and no one can say why. Taking this advantage into account, methodological theism actually provides science a better undergirding than methodological naturalism.

Metaphysical Biases Either Way

But we still can’t go there anyway. It’s off limits. There’s that awful metaphysical bias that goes with methodological theism, and we can’t have any of that, can we? Science simply has no business assuming there’s a God! That’s theology and philosophy, not science! 

Sure. But the same could be said from the other direction concerning naturalism’s metaphysical bias. Need proof? Read the rest of what Haldane said back in 1934. Let’s pick it up at the middle again:

I assume that no god, angel or devil is going to interfere with its course; and this assumption has been justified by such success as I have achieved in my professional career. I should therefore be intellectually dishonest if I were not also atheistic in the affairs of the world.

In his mind, methodological naturalism implied the actual truth of atheism. More recently again, Lawrence Krauss said (it’s the title of the article) “All Scientists Should Be Militant Atheists.” His reasoning (such as it is) is similar to Haldane’s: 

In my more than thirty years as a practicing physicist, I have never heard the word “God” mentioned in a scientific meeting. Belief or nonbelief in God is irrelevant to our understanding of the workings of nature — just as it’s irrelevant to the question of whether or not citizens are obligated to follow the law.

And there’s more than just a bit of metaphysical bias there. Saying God doesn’t matter is functionally equivalent, after all, to saying there is no God, which is the conclusion Krauss so emphatically commends to us all. We are well acquainted with methodological naturalism’s insistence that whatever happened in natural history, it had to have happened naturally. Nothing supernatural allowed. Ever. That’s equally biased, for it rules out a whole category of potential explanations, not because of evidence but because of prior metaphysical expectations.

So am I actually proposing everyone adopt methodological theism as their scientific operating guideline? Obviously not. It could be the stronger theory in all kinds of ways, and in fact I think it is. But in a pluralistic world it would certainly never fly — much less in a secular-slanted world like academia today. No, I’ve brought it up here for a completely different reason: to expose the fact that there’s no scientific reason to reject it; nothing but that thorny issue of metaphysical bias. At the same time, though, it highlights the reality that there’s no scientific reason to insist on methodological naturalism; only the same problem of bias. Neither principle is fully scientific. Both principles are saturated with philosophical assumptions.

But I need to say that one more time. With emphasis — emphasis, that is, on the principle that’s held sway all these years. Methodological naturalism is saturated with non-scientific, philosophical assumptions. Therefore it has no business masquerading as a necessary principle for doing science. There’s no rational cause to think science benefits from that kind of needless bias imported into it.

Dump the Metaphysics with Methodological Regularism

That doesn’t mean scientists can’t hold their opinions on these things, provided they hold them openly and honestly. Theists should be as free to talk about methodological theism as naturalists are to speak of methodological naturalism. Which only leaves the question, how do they talk with each other? And what about those who have no opinion on either theism or naturalism? Is there any bias-free principle on offer? One that we could all agree on? 

Sure; and it isn’t complicated. I’ve hinted at it several times already here. We all know that science depends on nature behaving regularly and predictably virtually all the time. I suggest we settle on mere “methodological regularism” as science’s operating principle. It says what needs saying, and it knows enough to say no more than that, especially nothing more of a non-scientific nature. It’s simple, and it works — in fact it works so well, it seems to me that having introduced the principle here, at a point when I should go on elucidating and campaigning for it, I would do just as well to let it speak for itself.

Science doesn’t need methodological naturalism. It doesn’t need methodological theism, either — although as a believer in God, I’d have to condition that with the thought that we couldn’t do anything without him in any endeavor! Science does, however, need a principle of predictable regularity. Methodological regularism provides that principle. It could be just that simple.

Photo: Sunrise from the International Space Station, by NASA [Public domain], via Wikimedia Commons.