From “The naturalism of the sciences,” by Gregory W. Dawes and Tiddy Smith, writing in the journal Studies in History and Philosophy of Science Part A:

The sciences are characterized by what is sometimes called a “methodological naturalism,” which disregards talk of divine agency. In response to those who argue that this reflects a dogmatic materialism, a number of philosophers have offered a pragmatic defense. The naturalism of the sciences, they argue, is provisional and defeasible: it is justified by the fact that unsuccessful theistic explanations have been superseded by successful natural ones. But this defense is inconsistent with the history of the sciences. The sciences have always exhibited what we call a domain naturalism. They have never invoked divine agency, but have always focused on the causal structure of the natural world. It is not the case, therefore, that the sciences once employed theistic explanations and then abandoned them. The naturalism of the sciences is as old as science itself.

From a quick scan, this is an interesting article — but their historiography looks more than a tad tendentious. Dawes and Smith say they’re simply describing (as a “matter of fact”) the history of science. But they’ve also carefully built escape or exception clauses into their history, so that any counterexample does not count against their thesis. As they write on page 28, opening the gate so that the exceptions can wander away, leaving only the obedient sheep in the pen:

 The naturalism of the sciences is a norm of scientific inquiry and norms represent both how a community regularly behaves and how its members think one ought to behave (Pettit, 1990, p. 728). So the existence of a norm is consistent with its occasional violation. [Emphasis added.]

Well — how convenient, as the Church Lady on Saturday Night Live used to say.

I grabbed a 19th-century science textbook from my office shelves: James Dana’s Manual of Geology (1871). Dana was professor of geology at Yale and by any dispassionate description fully a “scientist.” Here is how Dana ends his discussion of the topic “The Progress of Life” (paleontological trends — a summary of the signal from the fossil record):

Geology appears to bring us directly before the Creator; and while opening to us the methods through which the forces of nature have accomplished His purpose, — while proving that there has been a plan glorious in its scheme and perfect in its system, progressing through unmeasured ages and looking ever towards Man and a spiritual end, — it leads to no other solution of the great problem of creation, whether of kinds of matter or of species of life, than this: — DEUS FECIT.  (p. 602)

Deus fecit — Latin for “God created.”

This was a widely used geology textbook: “science” by any description. But this counterexample (one of hundreds possible) won’t count, because it’s “an occasional violation” of an otherwise universal norm.  Universal generalizations sleep undisturbed when the contrary evidence isn’t allowed anywhere near the doorbell.

Moreover, the relentless late 19th-century campaign by T.H. Huxley and others against scientific explanation by divine action and for fully naturalistic or materialistic explanation should not have been necessary, if Dawes and Smith are correct in their history.

But — check the article, it’s open access — Dawes and Smith tip their hand in their concluding paragraph. Any flexing of the methodological naturalism (MN) rule will fracture science along religious lines, they say, and that’s bad. So the provisional atheism of science should continue, because that’s what science since the Greeks has always done…

…Except when it hasn’t — but we’re not counting the many exceptions.

Editor’s note: For more on MN, see Paul Nelson’s comments here, “Methodological Naturalism: A Rule That No One Needs or Obeys.”

Image: Dark finger reef crab (Etisus dentatus), an illustration from James Dana’s book Crustacea (1855), via Wikimedia Commons.