Anything approaching a fully satisfactory explanation of the phenomena of knowledge requires the co-operative efforts of all those who believe that there is a world of real existence independent of human minds and that this real existence can be truly known as it really is.

After listening patiently, the malefactor says with a smirk, “But why would I have done any of that?” And unless the scene occurs at the very end of the story, the detective must admit that he doesn’t know. “I don’t know yet. But I’ll find out,” he vows.

I find these scenes a little annoying because of the insult to logic, and, hence, to the intellect. The fact of the matter is that once you have assembled the evidence that something is the case, the fact that you don’t know why it is the case does not imply that it is not the case.

This is not to deny that there is a satisfaction in being able to answer the “why” questions, but that is not the same thing. When a lawyer is attempting to put a criminal behind bars, the jury wants to know why the criminal did what he did. There seems to be some deep, human need not to conclude that a person has done something if it is not possible to say why he did it.

And surely we all know in what that need consists. It is the desire to understand. If we are talking about — to pick a popular example — the assassination of John F. Kennedy, it is not enough to know who killed him; we want to know why he was killed. Logic leads us to understand what is true; but real understanding, satisfying understanding, includes why it is true. (As an aside, this is why science is not enough for the intellect; we need philosophy as well.)

Let us suppose that I have argued that Lee Harvey Oswald did in fact kill John Kennedy. And let us suppose that a person with different beliefs says to me, “But why would he do that?” and I am unable to supply a motive for the shooting. At that point, what am I expected to answer? “Well, gee whiz, I guess all that evidence I just gave you doesn’t count for anything”?

We see this, perhaps more dramatically, in discussions of the existence of God, and in particular what is known as The Problem of Evil: If God is good and if he is omnipotent, why would he let such-and-such happen? I confess that what I really want to say in such exchanges, is, “I don’t know. Ask him.” (I say this knowing what kind of answer Job got when he raised the question.) But perhaps more generally, we should realize that if I have offered a sound proof of the existence of God or even just a lot of good reasons for believing in God, no shortcomings in my ability to explain something he may have done or not do, do not refute my arguments. I grant that they may make my interlocutor dissatisfied with the conclusion — the things I am unable to explain may be precisely those things that are of greatest concern and interest to him — but even he should see that they do not count as a refutation.

I think that people who find the “But why would ...” reply so compelling is that they are very poor introspectors. Surely they have had the experience of being at a loss to explain just why they said or did something. When that happens to me, I am never inclined to conclude that therefore I did not say or do it.

In other words, the failure to account for a person’s actions is not evidence that he didn’t perform them.

To be able to supply the motive is a rhetorical advantage, but in logic it has no value, either for or against the proof. To treat that ability (or inability) as anything but a rhetorical device — even a rhetorical device with undeniable psychological power — is to commit a logical error.

Moreover, it should be noted that the “But why would” question is just that: it’s a question. A question does not refute an argument. (I speak, of course, of genuine questions, not rhetorical questions, which in their substance and despite their form are more like propositions than like questions.) Only propositions can refute an argument, or, more precisely, only propositions in the form of an argument can refute an argument.

To be sure, a deftly asked question may seem to refute an argument, but that is often because certain implications or propositions may be implicit and apprehended both by the person asking and by the person asked. Similarly, a single proposition may suffice in an attempted refutation because there is an unspoken or implicit argument.

Related to this error is the claim, “There is no reason that ...” Whenever I hear or read someone arguing a case who insists that his point must be correct because “There is no reason” for things to be thought to be other than what he says they are, I cannot help thinking that I am hearing or reading a person with little imagination.

When talking about human behavior, this argument form — “Why do you say Barton did such-and-such? There is no reason ...” — I am tempted to reply, “How do you know? How do you know there is no reason?”

“There is no reason” is at the very least a confession of lack of imagination or information. It may be a confession of intellectual arrogance. But whatever it may be, it is not a refutation of anything, and should not be treated as such by anyone.