“Valid” derives from the Latin valere, “to be strong, powerful, healthy; to influence or prevail (“prevail” itself derives also from valere).”
When we reason validly using, e.g., modus ponens, the advantage to us is that we are able to generalize. Modus ponens relieves us of the necessity of performing a conclusively exhaustive experiment in order to verify the appropriateness of our behavior in respect to each and every instance of novelty we encounter. E.g., it allows us to infer reliably that Socrates will eventually die on account of the facts that all men are mortal and that he is a man, rather than being forced to try to kill him in order to find out for sure via empirical means. Likewise, it allows us to infer that he will fall to Earth if unsupported, sparing us of the necessity of dropping him in order to find out for sure. These inferences enable us to plan our behavior toward Socrates with enough efficiency that we are able to respond to him quickly enough to do us some practical good – in the event, say, that we were an enemy of Athens and he was rushing at us with the rest of the Athenian phalanx.
Valid reasoning, then, gives us great epistemological leverage. We take our premises, that have been somehow or other tested, and infer reliably to their conclusion, and thus to its general application. Because it vastly increases the economic efficiency of our cognitive operations, reliable generalization confers a great deal of biological advantage – or, to put it another way, is a really good thing, a really healthy, moral thing. Valid reasoning promotes our welfare.
So, valid reasoning is morally good. To reason validly from premisses is a virtuous act.
But validity is not virtuous merely on account of its beneficial consequences for our welfare. Indeed, it goes the other way: the beneficial consequences for our welfare of acts informed by valid reasoning turn out to be beneficial only on account of the fact that the valid reasoning that informed them is virtuous per se: is strong, is excellent in and of itself, and without regard to its practical consequences. This is easy to see when we reflect upon the virtue of valid reasoning about matters that have no practical consequences, such as our conclusion about the sum of two randomly selected numbers that, in themselves, don’t matter to us in any way. Even when a valid inference cannot touch our lives in any possible way, we apprehend nonetheless its strength and excellence.
Our acts – including our acts of ratiocination – are not good because they contribute to our survival and prosperity; they contribute to our survival and prosperity because they are good. The goodness of survival and prosperity, after all, consists in the goodness of the acts of the persons who, by performing them, therefore survive and prosper. The value of surviving consists in the value it enables us to realize.
An act that is informed by invalid reasoning or erroneous premisses may of course nevertheless turn out well; as may an act that is not informed at all, if such there may ever even be. But such misinformed or uninformed acts can turn out well only in the same way that a broken clock may “tell” the time accurately twice each day. The benefit such acts confer is wholly adventitious. Thus to the extent that the actual benefit of an act does not follow from the reasoning that informed it, that benefit has nothing to do with our reasoning.