The Possibility of Practical Reason, 2nd Edition

14  A Note on Practical Knowledge

Peter Pan sprinkles fairy dust on Wendy, Michael, and John, saying, “Now you can fly, but only by believing you will.” The Darling children are now in a position to know that they are going to fly, since they can believe so truly and with reliable justification. (Where flying is concerned, Peter is a reliable authority.) They are also in position to know that they won’t fly, since a belief to that effect will likewise be true and reliably justified. They are therefore in a position to have practical knowledge—knowledge that causes its truth-maker.

Two philosophers object to the idea of practical knowledge.[1] One thinks that the Darling children will never get off the ground. The children know that they cannot fly unless they believe that they will; and they cannot believe that they will fly (this philosopher claims) so long as they know that they can't. Another philosopher thinks that the children can get off the ground but only by being epistemically irresponsible, since (this philosopher claims) they shouldn't believe that they will fly unless they have evidence that they can; whereas their evidence indicates only that they will be able to fly once they have the belief.

We, of course, know that Wendy, Michael, and John soar off to Neverland with their epistemic innocence intact. Where have the skeptics gone wrong?

The first skeptic thinks the children cannot believe something without evidence that it’s true and so they cannot believe something given only evidence that it will be true if they believe it. If they could form a belief on the latter evidence, he points out, then they would be able to believe things at will; in the present case, they could choose to believe either that they will fly or that they can’t, since they have evidence that either belief would be true if they formed it. And, as we know, says this philosopher, believing at will is impossible.

But why is believing at will impossible? And how do we know that it is? We certainly don't know it from experience with bad trips on fairy dust. It’s true that we are usually unable to believe arbitrarily in a proposition or its negation. But the reason why we are usually unable to do so doesn’t apply to beliefs that are known to be self-fulfilling, like the belief made available to the Darling children on Peter’s authority.

To believe something is to regard it as really true—to regard it as true, that is, in an attempt to represent reality. Ordinarily, we cannot try to represent reality by choosing arbitrarily whether to accept a proposition or its negation, because reality may not correspond to an arbitrarily chosen proposition. Thus, choosing to believe is not the sort of attempt that ordinarily yields belief. But in the case of a belief known to be self-fulfilling, arbitrary choice between p and not-p is compatible with attempting to represent reality, since reality can be expected to conform itself to whichever proposition we accept. So the ordinary obstacle to believing at will is absent.

The second philosopher thinks the rules of theoretical reasoning forbid the adoption of a belief without prior evidence of its truth. But why do the rules forbid it?

Theoretical reasoning is not a game defined by arbitrary rules; it is a project with a constitutive aim, namely, knowledge—reliably justified true belief—and its rules are subservient to that aim. Peter Pan invites the Darling children to form a belief that will be both reliably justified and true. Rules requiring them to decline his invitation would not be subservient to the constitutive aim of theoretical reasoning. Such rules cannot be the ones governing that project.

To be sure, the story is pure fantasy: it requires us to suspend disbelief. But all it requires us to suspend is our disbelief as to whether fairy dust can give the children the power to fly by believing they will. We do not disbelieve for a moment that given the power to fly by believing they will, the children will be able to fly.

What if Peter tells the children, “Now you can fly, but only if you suppose that you will”? There ought to be no obstacle to their flying in this case, even according to our skeptics. One can suppose empirical propositions that are obviously false, and one can be rational in supposing them.

Yet in this case, the children will be unable to stop at making a supposition. One cannot merely suppose what one already knows. (Try merely supposing that you can’t fly.) So if the Darling children come to know that they will fly, they won't be able to suppose so; and since they already know that they will fly as soon as they suppose so, they will no sooner suppose so than they will know so, whereupon they will no longer be able to maintain a mere supposition.

If Peter means “Now you can fly, but only if you merely suppose you will,” then the children may still be grounded, because they won’t be able to stop short at supposition. But he could mean “Now you can fly, but only if you at least suppose that you will,” where believing and knowing are conceived as greater forms of acceptance than supposing. Then the children can aim at mere supposition, although they will inevitably overshoot it and land in a reliably justified belief. To aim at supposition will amount to undue modesty about their cognitive and aeronautic powers. After a few overly modest attempts, they will learn that aiming to suppose in this case always ends in knowledge—hence, that they might as well aim at belief.

Suppose that Peter Pan says to the children, “Now you can fly, but only if you choose to.” Oddly, our no-fly philosophers are committed to rejecting this case as well. Our first philosopher will have to say that the children cannot choose to fly, because they know that they lack the ability to fly until they choose to, and they cannot choose to do something that they are aware of being as yet unable to do. Our second philosopher will have to say that although the children can choose to fly, that choice would be irrational, since they should not choose to do something that, as they well know, lies beyond their current abilities.

Yet even Peter cannot fly unless he chooses to. His ability to fly just consists in the fact that he will fly if he so chooses. No less than Peter, the children will fly if they choose to. So they already have the ability to fly, and the choice is both possible and rational for them.

Clearly, then, the skeptics’ reasoning cannot be extended from the case of belief to the case of choice. And what is the difference between the children’s choosing to fly in the latter scenario and their voluntarily believing that they will fly in the former?

The children’s choosing to fly settles the matter: they are going to fly, pursuant to their choice. Yet believing they will fly also settles the matter: they are going to fly pursuant to their belief. In the second scenario, they choose to fly because they want to fly, and they know they will fly if they so choose. Likewise, in the first scenario, it’s because they want to fly—and know they will if they so believe —that they believe they will fly. Once believing they will fly, they needn’t take a further step of choosing to, since they have already settled the matter, not just in their minds but in fact, and they’ve settled it as they desire, out of that desire. So there is nothing for a choice to do that hasn’t already been done by the self-fulfilling belief. Nor would a choice do any less than a self-fulfilling belief, since it would merely settle, notionally and actually, that they are going to fly.

Well, philosophers insist that there is a difference between “We’re going to fly” as an expression of choice and the same utterance as an expression of belief. There is indeed a difference between choice and ordinary belief, which does not cause its own truth-maker and must therefore be responsive to it or to evidence of it. But where belief is a springboard to action, it is no different from the springboard of choice: believing is choosing; choosing is believing. And since it’s reliably justified and true, it is knowledge—knowledge embodied in choice, rightly called practical knowledge.

Ultimately, what separates believers from skeptics on the question of practical knowledge is a difference of methodology—specifically, a difference over the philosophical use of intuitions. There is consensus among philosophers—going back as far as Aristotle—that philosophical reflection must start with intuitions or commonsense beliefs, for the simple reason that there is nowhere else to start. But some philosophers believe that intuitions are not just the starting point of philosophy but also the criterion of success at the finish. They believe that philosophical conclusions should not be counterintuitive, or that counterintuitiveness is a cost to be avoided (as if philosophical conclusions could be evaluated by cost-benefit analysis).

This methodology turns philosophy into a discipline of achieving reflective equilibrium among our intuitions—of systematizing common sense. And no matter how common sense is systematized, it will never accommodate practical knowledge. Philosophers who adhere to this methodology are therefore right to be skeptical.

But there are good reasons not to go in for the discipline of systematizing common sense. One reason is that a systematized common sense is unlikely to be true. Another reason is that even if it is true, it will be uninteresting. Time spent systematizing common sense would be better spent on discovering uncommon truths. As Bertrand Russell said, “The point of philosophy is to start with something so simple as not to seem worth stating, and to end with something so paradoxical that no one will believe it.”[2] What is so simple as to go without saying is that knowledge is reliably justified belief; what’s so paradoxical as to defy belief is that belief can be reliably justified without having been justified antecedently.