Non-Perceptual Knowledge

Non-Perceptual Knowledge

Perceptual knowledge is quite sharply limited, though clearly a type of knowledge. But we don’t customarily stop there; we generally extend the concept of knowledge beyond this restricted domain, well beyond it. Thus, we recognize several kinds of inferential knowledge—knowledge about things we don’t and can’t perceive. Are there any reasons to doubt this extension, stopping short of invoking extreme skepticism. Is the concept of knowledge inherently resistant to such epistemic promiscuity? Given that this “secondary knowledge” is not perceptual, we cannot locate its vehicle in sensory experience, so we will need some other psychological vehicle. This has various names: belief, judgment, opinion, surmise, conjecture, supposition, postulation, hypothesis, etc. Then we will be saying that some knowledge is grounded in this kind of state: for example, some knowledge is a type of belief (though many beliefs are not knowledge). In this way we take in knowledge of other minds, the past, the future, remote parts of the universe, the fine structure of matter, and so on. We don’t perceive these things but we do have beliefs about them. This has the consequence, obviously, that knowledge and belief are compatible states; as Plato would put it, some “opinions” (not all) count as instances of knowledge. The state of mind of knowing is compatible with the state of mind of having an opinion or surmising or conjecturing. But Plato would have none of that; for him, knowledge and belief (etc.) exclude each other. It is impossible to believe and know the same thing at the same time. And ordinary language would appear to back him up: no one in his right mind would say “I know that p and I am also of the opinion that p”, since the second conjunct contradicts the first. The well-educated analytical philosopher has his riposte ready—“That’s just a conversational implicature, my friend!” End of discussion. But is it just an implicature? Isn’t there something right in the idea that knowing is not compatible with believing? Don’t I need to do more than conjecture in order to know? I have to be convinced. Suppose I form a belief irrationally, based on my whims and wishes; but the belief happens to be true and I have gathered evidence sufficient to support my belief. However, I don’t hold the belief because of the evidence but because of my wishes: do I then have knowledge? Surely, I am in the wrong state of mind. The trouble is that beliefs and opinions can be held for irrational reasons, so they are unsuitable as the psychological basis of knowledge (the same is not true of sense experiences). Beliefs are often irrational, easily manipulated, and prone to error; so, we don’t want to involve them in the serious business of possessing knowledge. They also allow people to go out on a limb, get fooled by conspiracy theories, make mistakes of reasoning. Beliefs are just not a good way to run your cognitive life. You would be better off without them. They are an epistemic trap.

Many serious philosophers have advocated just that. Stick to what you really know, what you have seen with your own eyes, don’t believe other people with all their lies and bullshit. Remain agnostic about matters you really know nothing directly about. Don’t believe. Don’t form wacky opinions. You can behave as if you believe; you can provisionally entertain and act upon certain propositions; you can hold that a proposition has not yet been falsified; you can pretend the proposition is true: but for heaven’s sake don’t believe things of which you have no direct experience! That way error lies, if not madness. Don’t believe what you don’t directly know, and if you do directly know it you needn’t believe it. Abolish belief altogether! That would appear to be Plato’s view, and Popper’s too. Things will be much clearer and cleaner that way, once you have adopted the zero-belief lifestyle (“Just say no to belief!”). It might be replied that this is too sweeping: surely some beliefs are better than others, some opinions better informed than others. That is clearly correct, but it doesn’t blunt the force of the Plato-Popper position. We can register the necessary distinctions without employing the concept of knowledge so loosely and irresponsibly; we can simply talk of degrees of justification, warrant, cognitive virtue. Nor need we recommend any linguistic revision in our ordinary use of “know”; we just need to recognize that it is not literally true, a figure of speech, a useful convenience. We should stop thinking that we really know in these cases, whether in the primary sense or a secondary sense. We should stop assimilating these cases to the basic cases provided by perceptual knowledge—the only true knowledge. In fact, it is not clear that anyone is fooled by our common use of “know”: people are aware that this is so much loose talk—like talk of the sun rising. When I ask people about this, they generally say that of course we don’t really know other minds no matter what words we utter. Some propositions have higher epistemic credentials than others, that’s all; not much is really known. The language game involving “know” is just that—a game, not to be taken too seriously and literally.

It might be asked how this game came into being—why did we start using “know” so promiscuously? My theory is that it came from religion and then migrated into science. Clearly, it is not possible to perceive God, but religion encourages us to believe in him; so, it needs a notion of knowledge compatible with imperceptibility. Thus, belief (“faith”) was introduced as the route to knowledge of God. Religion needs a belief-based epistemology conducive to (non-perceptual) knowledge. Science, especially astronomy, needs a similar epistemological structure: belief as the vehicle of scientific knowledge. We need to stretch the concept of knowledge if we are to make room for scientific knowledge, thus conferring on it the honorific status of knowledge proper. However, there is really no pressing need to stretch the concept this far, to the breaking point as it were; we can settle for something short of knowledge, such as pragmatic acceptance. Then we avoid introducing an untenable dualism into our epistemology—between perceptual knowledge (the real deal) and something far removed from it (and not clearly knowledge at all). We don’t need to mess with the concept of knowledge, stipulating knowledge where it does not belong. There is simply a lot we don’t know (primary!), even when the relevant science is as good as it can be (pretty damn good). It is pointlessly provocative to insist that we know Darwin’s theory to be true, say, which invites the response “Have you seen it in action?” Epistemic merit is not the same as knowledge strictly so-called. The former is possible without the latter. Then we can say that only perceptual knowledge is really knowledge, bearing in mind that perception (“seeing”) is not just a matter of the physical senses. Unless we stick to this strict and sharp distinction, we end up completely confused about what counts as knowledge and what doesn’t. Opinions can differ in their epistemic credentials, but opinions are never knowledge, just as Plato maintained. Knowledge can be conceptualized as seeing in all cases, with no weakening contemplated. Belief therefore has nothing to do with knowledge proper: not perceptual knowledge and not some kind of dilution of that. Knowledge and belief are mutually exclusive. So far from knowledge implying belief, it implies non-belief. Knowledge is never true justified belief, or some improvement in that formula. All knowledge is acquaintance knowledge. The natural kind knowledgeis a single unified category not a congeries of different concepts. I would recommend the attitude of pragmatic acceptance for any truth claims that go beyond perception. There is no such thing as “inferential knowledge” or “testimony knowledge”. This is the way things should be carved up epistemologically, not into different types or degrees of knowledge.[1]

[1] How revisionary is this position? It is certainly revisionary of philosophical and scientific custom (not counting Popperians), but it is not clear that it contradicts the ordinary man’s tacit epistemology. From my informal surveys, people are quite ready to accept that nothing is really known beyond direct perception. It is deemed folly to suppose otherwise. This doesn’t prevent them from preferring some theories to others, quite rationally. People are surprisingly strict about the word “know”; they don’t like to see it bandied about. It is so in Barbados anyway (fine people).