John Derbyshire

There's been a little debate going on at The Corner about reason, truth, science, and faith. John Derbyshire started the debate here. It continues here, here, here, here, here, here, here, and here, at least so far. Derbyshire's major point, I think, is that religion is unfalsifiable and therefore, lacking criteria for verification, necessarily relativistic. Why, for example, prefer Catholicism over Buddhism?

It's an important question for scientists to ask. Actually, I see three important questions. First, is it possible for there to be something which is true which is unfalsifiable? Second, is it possible for us to know these truths? Third, is it wrong for us to hold as true something which is unfalsifiable, even if we do not know that it is true?

I think that Derbyshire would admit that it is possible that something be true which is unfalsifiable. However, it is quite clear that he denies that we could know them, at least with kind of certainty that we know scientific truths, and he implies that therefore these things are not worth believing. (He was widely reported to have left the Episcopal faith within the last few years or so, and is the most vocal opponent of religion on The Corner.)

However, Derbyshire himself holds to beliefs which are unverifiable by definition. For example, he believes that all proponents of Intelligent Design are motivated by religious, and not scientific, reasons. He knows quite well enough the scientific arguments which ID people make, but refuses to engage them, instead asserting that these arguments are simply cover for anti-scientific creationistic beliefs. It is impossible for Derbyshire to be refuted on this point, because any evidence to the contrary can be explained away (paraphrasing him slightly, it's ridiculous to suppose on his account anything other that unadulterated religious fervor (by definition anti-scientific) to the ID proponents).

The scientific method, therefore, cannot be the standard by which we judge all our beliefs. There are many things we believe, and Derbyshire must admit this, I think, which we cannot prove, and this state of affairs is or at least can be good. For example, it is appropriate to believe in the fidelity of my wife, and in fact, it would be inappropriate to subject that fidelity to scientific scrutiny. A relationship has been destroyed already when the husband hires a private investigator to follow his wife around, even if she has been completely faithful.

Secondly, we can ask, granted that we hold beliefs which are unverifiable, can we know anything for certain which is not verifiable, or at least falsifiable? I think it is obvious that the answer to this question is yes. For example, consider Aristotle's definition of nature: "Nature is a principle and cause of motion and rest, in that in which it is, primarily, in virtue of itself and not accidentally" (Physics 2.1). This is not falsifiable. There is no experiment that can prove this wrong. It is, however, in contradiction to Newton's definition of matter ("The quantity of matter is the measure of the same, arising from its density and bulk conjunctly"), which itself is unverifiable. Either Aristotle is right (there are things with natures and forms which determine things' actions from within) or Newton is right (there are no natures or forms, and the only account we can give of the world is mathematical). The scientific method assumes the second. Why? Because it is successful. We have a cure for influenza and we have put people on the moon. But it is one thing to say that it is successful and its another thing to say that it is verifiable.

We can put things another way: Newton's theory can (with modifications) be incorporated into and Aristotelian framework. Aristotle's theory is not opposed to materialistic explanations -- after all, he distinguishes four causes, and modern science is explicitly concerned with two of them. However, Aristotle's understanding of nature can never be subsumed into Newton's account. Newton's science works by systematically accounting for quantifiable differences in material properties, a practice which has no room for non-quantifiable realities like "nature."

Now, per suppositum Derbyshire would reject Aristotle's theory of "nature" on the grounds that there are no criteria by which it could be verified. However, the notion that something should be rejected as being true if it cannot be verified, itself cannot be verified, as we have shown by distinguishing Newton and Aristotle. Therefore, the act of asserting it to be true enacts its falsity.

It is on the basis of such truths (aka, ones known to be true prior to all scientific investigations) that Aquinas proves the existence of God, and all philosophy of any sort proceeds. And, as we have noted, science itself depends upon those truths. However, there is a further question. Granting that we cannot know for certain everything a certain religion proposes as true, does it follow that we have no good reasons for believing them? For example, the Apostles' appealed to one fact and one fact only when converting the Jews and the pagans: the Resurrection. The Resurrection is not a verifiable fact, in the scientific sense, but then again, neither is the existence of Newton in the 17th century. But that does not mean that we have no good reason for believing in the Resurrection (or, for that matter, in Newton's existence).

But all that aside, I think Ramesh Ponnuru is exactly right about Derbyshire.