Is Belief Necessary For Knowledge?
Is Belief Necessary for Knowledge?
It has always seemed that the stipulation that belief is a necessary condition for knowledge is a touch unrealistic. One wants to say, “I don’t just believe it, I know it”. Belief goes with opinion, uncertainty, faith—but knowledge is a matter of being indisputably right, in the know (as we say). I am not of the opinion that I am writing an essay about belief and knowledge, as I am of the opinion that the Beatles are better than the Stones; I damn well know it. Do I believe that I exist or do I simply know it? Could Descartes have said, “I think, therefore I believe that I exist”? I can say that I merely believe something but not that I merely know it. Genuine knowledge appears to preclude belief: if I know something I know it, I don’t just take myself to know it. To this objection there is a standard reply: it confuses implication and implicature. To say “I believe” when you really know conversationally implies that you don’t feel entitled to claim knowledge, but logically it is compatible with knowing. Knowledge logically implies belief, but we don’t say the weaker thing when we could truly say the stronger thing—as in saying “It seems to me there is apple there” when you are in plain sight of an apple and have no doubt that you are in the presence of an apple. Now, that reply may indeed be a theoretical option, but what if it really is false that the knower is also a believer? What if belief really is incompatible with knowledge? What if the believing state of mind just doesn’t exist in an ordinary case of knowledge? What if someone could be a knower and yet not have the corresponding belief?[1] Is that logically possible? Maybe in most actual cases belief and knowledge coexist, but the former is not a necessary condition of the latter—knowledge doesn’t rule out belief, but it doesn’t presuppose it either. Maybe belief can be replaced by knowledge once the knower’s epistemic situation is improved, so that it no longer exists in the new knowledge state. The connection may be loose not logically tight. It is certainly not a truism that knowledge logically implies belief, as it is a truism that knowledge logically implies truth.[2]
Here is a counterexample to the claim of necessity. A scrupulously rational person, Bertie, has been reading a lot about skepticism recently and is mightily impressed with it. He becomes a passionate skeptic, refusing to accept that he knows anything; he undertakes to suspend belief about matters that most people unhesitatingly take for granted—for example, that there is a table in front of him. He declines to believe it, convinced that he might be wrong (he could be a brain in a vat). However, he is well aware that his well-being depends on acting in a certain way in response to his subjective experience—he has to behave as if he knows there is a table there, or else he will receive sensory impressions of a barked shin etc. Inwardly he doesn’t believe in tables (nor does he disbelieve in them); outwardly he behaves as if he does believe in them. You wouldn’t know to look at Bertie that he is a skeptic with regard to the external world. But if he tells you of his attitude towards such matters, you will refrain from ascribing the usual beliefs to him—for Bertie is a very determined rational man. He doesn’t believe in tables, period. But does he know anything about tables? Wouldn’t you say that he knows there is a table in front of him, even though he doesn’t believe it? He isn’t like a blind man: his eyes are open and he clearly sees tables; he acts as if he knows the disposition of tables. He knows the table is there; he just doesn’t believe it. The reason this verdict seems correct is obvious: Bertie’s sensorimotor system is giving him the information that tables are all around—he just refuses to convert this information into belief. He is a non-believing knower. His senses and actions track the presence of tables, but his belief system is disengaged. We could say that part of his mind tracks tables but not the belief part. He mentally represents tables perceptually (and in his actions) but his beliefs don’t match these representations. This mental representation might well be a necessary condition of his knowing, but the corresponding beliefs are not essential. Thus, it is possible to know that p and not believe that p. Less scrupulous believers may well dive into belief irrespective of skepticism, but in their case also the real basis for an ascription of knowledge is their sensorimotor capacities not their state of belief. The classic analysis of knowledge has mistaken the contingent for the necessary, elevating belief into a central role it does not deserve. And the sensorimotor basis is not a species of opinion—it isn’t a type of uncertain judgment or speculation or conjecture or article of faith. It doesn’t belong to that part of the mind.
Here is another counterexample of a more science fiction type. A certain individual, Phineas, has had an injury to the head in which his ability to form beliefs has been damaged. Phineas finds that he can’t form opinions anymore (he used to be full of them). If you ask his opinion on any subject, he will report that he has none. His doctors declare him a victim of “doxastic paralysis” and publish learned articles about him. But suppose that he is otherwise undamaged—nothing wrong with his eyes or motor system. Doesn’t he still know things? He perceives his environment, has memories, conducts himself like a normal person—he knows what’s what. He just has no beliefs about any of this (he is a “belief zombie”). Perhaps he slides into simply acting as if he believes this or that—his life works out better if he does that. Zero belief, much knowledge. The lesson is that belief is not essential to knowledge; what is essential is some sort of tracking of the world by the organism—by the brain and body. Knowledge is less intellectual than belief, less a matter of judgment and deliberation, of opinion formation. It isn’t a type of belief at all—though beliefs do populate the knowing mind in normal cases. And, come to think of it, belief is unsuitable for knowledge in most cases, because it is far too friable, far too shaky. One might almost say that it is not me that knows but my body and brain—whereas I am a believer. The rational ego forms beliefs, but knowledge typically arises from more basic capacities, which don’t require the participation of the conscious rational self. When did I ever form the belief that I am surrounded by physical objects? Do I really believe this, as I believe in democracy and the rule of law and the superiority of the Beatles? I don’t believe in physical objects; I know it without having to undertake a process of belief formation. Do animals believe in such things, or do they know them without benefit of belief? They are set up to know; they don’t need a faculty of belief to get them there. Belief is a luxury they can ill afford; they need to know things without such time-consuming lucubration. Knowing doesn’t generally involve study, reflection, thought, meditation, judicious decision. Evolution made us knowers before believing ever came into the picture. Knowing dates back millions of years, but believing only hundreds of thousands. Believing is coeval with civilization, roughly, but knowing is primitive and instinctual—as it needs to be. Knowing is not a superior form of belief (the true and justified kind) but in some ways more animalistic (not in any pejorative sense); it is part of animal nature, or the animal part of human nature. Belief is as inessential to knowledge as it is to perception. You can see and know without ever going to the trouble of believing things.[3]
[1] A point that used to be made is that it is possible to be unsure about the answer to a question and just guess the answer—correctly, because the memory of the answer still lingers. Here it is natural to say that the person really knows but doesn’t have the confidence to fully (or even partially) believe. I will be pressing this kind of point a lot harder.
[2] Question: are there any undisputed truths of philosophy that are not truisms? It’s hard to come up with any.
[3] Believing is hard, effortful, time-consuming, anxious, fraught, and modern; but seeing and knowing are automatic, reflexive, easy, well-honed, and ancient. There is no need to worry about seeing and knowing, but believing is inherently burdensome (“I don’t knowwhat to believe!”). The whole picture of knowledge encouraged by the belief condition is false to the reality of it—a typical philosopher’s error of over-intellectualizing the phenomena. Knowledge is not generally like philosophical knowledge, which is mostly opinion. Plato was right sharply to distinguish knowledge from opinion. Knowledge is not high-class opinion; it isn’t opinion at all. Knowledge is more democratic, widely distributed, humble, biological. It is pretty much coextensive with consciousness (though often unconscious). Think of our knowledge of language: we don’t have opinions about grammatical rules.
Hello,
I have a question related to your first counterexample (it can also be asked of your second counterexample). Could this not be salvaged (in a belief-is-necessary-for-knowledge framework) by saying that Bertie behaves as though tables exist because he suspects (but doesn’t believe) that the apparent (but not necessarily existent) tables will trigger a negative experience of pain (not because he knows anything)? I think this is agnostic to the existence of the table and any other firm belief.
Thanks,
Greg
I don’t think that helps. First, it concedes that belief is not a necessary condition. Second, Bertie doesn’t even suspect there is a table there, because he is a skeptic. Third, he knows quite well that apparent tables can lead to pain, whether real or not.
1. I know it doesn’t invoke belief. The whole point of your counterexample was to create a scenario in which not belief, but knowledge. I’m claiming that it’s not belief, but also not knowledge.
2. I’m not claiming that he suspects the table is there. I’m claiming that he suspects bumping into this table (whether that is what’s actually happening or if it’s all a simulation) will cause pain.
3. I’d challenge that. He suspects from a probability razor that bumping into the table will cause pain, whether or not the table is real.
As I suspected, I didn’t understand your comment, so I tried to interpret it. I understand your new comment even less.
Ok, let me try again. I’m not a philosopher by a goddamned sight so please be patient if I’m saying stupid stuff (but I don’t think I am here).
Your claim here is that knowledge doesn’t require a belief. To this end, you construct a scenario in which knowledge is present and belief isn’t. I agree with you that belief is absent, but I’d also say that knowledge is absent.
Bertie behaves in the way he does because previous experience informs him that bumping into the table will cause pain (whether or not the table or his leg or anything else exists). That perceived event will entail that negative result.
If you are talking about going full skeptic, and denying that Bertie’s memories are real, it makes less sense, not more, I’d say. You are equating knowledge with inherent dispositional tendencies (in this case, avoiding collisions with the table). But if this is your standard, you’d have to conclude that human beings’ universal genetic behavior of worshipping supernatural things constitutes knowledge that supernatural things exist. Among other things.
Defining knowledge as JTB means that our only two pieces of knowledge are that conscious experience exists and that reality exists.
You appear to be thinking that there is no actual table in the case I describe: but there is, by stipulation. The truth condition is satisfied and so is the justification condition–Bertie is seeing an actual table (he just doesn’t believe he is). That’s why our intuition is that he knows there is a table there. It’s nothing like supernatural beings: here the truth condition is not satisfied.
Hello again,
I apologize for leaving this as a separate comment, for some reason I don’t see the option to attach it to your last message.
I didn’t catch that the truth value of the table’s existence was stipulated. My bad. I have thought about this quite a bit today, and I have one more question.
“Justified” and “true” are adjectives, “knowledge” is a noun. I would think those adjectives must describe a particular noun for that noun = knowledge to make sense. You’ve already said that that noun is not “belief.” Is that noun “behavior?”
Or is this some kind of epistemic analog of metaphysical generalism, where qualities (and only qualities) constitute the object of study (“knowledge”)?
Thanks again and please forgive my shortcomings
I guessed correctly. You are right that truth and justification need a bearer, but it doesn’t have to be belief. It could be behavior (verbal or non-verbal); it could be a perceptual state; it could be an informational or computational state. Belief is just one kind of mental representation.
Ok, I’m intrigued. I am currently struggling to see how the following things are intelligible/applicable, given that we’re still committed to the truth and justification criteria:
“True behavior” seems unintelligible to me
“Justified perceptual state” seems strange to me, especially given the noumenal/phenomenal distinction.
“True informational state” seems to presuppose Platonism
You’ve been very kind to respond up to now, thanks
Isn’t a statement a type of behavior (speech act)?
A justified perceptual state is one that has the right causal history.
Information has nothing to do with Platonism; the brain processes information.
As always, thank you. I feel like I’m getting philosophy training for free haha. If you’d entertain continuing the conversation:
1. It is true that you have to convey the statement by an action, but I’m not sure that that means that the statement is an action. We already agreed that belief (or a statement or equivalent) is a noun, which is not a verb. (I’m finding this syntactical framework helpful for precision and understanding here, I hope you don’t mind it).
2. Ok, that makes sense, but I happen to reject the existence of causality, so I guess I’d personally have to reject this one, but I do understand why this should be accepted by most philosophers. An interesting connection between metaphysics and epistemology there for sure.
3. (just assuming physicalism here, although I’m not a physicalist (although I am sympathetic to your Mysterianism)) The brain processes physical stimuli, information is a Platonic attachment/description. As a nominalist, I don’t think the information exists, and so I don’t see how I can accept that it constitutes knowledge, which I hold is real (in the cases of the existence of consciousness and of reality). To break this down simply, I can’t simultaneously hold these views: 1. Information doesn’t exist 2. Information constitutes knowledge 3. Knowledge exists (as a mind state)
This will be my last free tutorial. An utterance is an action and utterances can be true or false; they have meaning. The causality point is simply that external stimuli often cause perceptual events, which is hard to deny. If information doesn’t exist, there can be no meaning or belief or knowledge. Information is not Platonic in any recognized sense. I suggest you study some philosophy texts.