The Illusive Test for Knowledge
The traditional analysis of propositional knowledge holds that knowledge is justified true belief (“JTB”) (Crumley II 36). JTB is necessary and may be sufficient for knowledge[1]. Thus, first one must have a propositional belief, e.g., “The earth has the shape of an oblate spheroid.” Second, the belief must be adequately justified, e.g., there is an abundance of scientific knowledge which is taught in schools and available from reputable academic sources online and in print which asserts conclusively that the earth is an oblate spheroid. Third, the belief must be true, i.e., the actual world (independent of the believer’s perception) must be such that the earth has the shape of an oblate spheroid[2]. But it is only via human perception and reasoning that the claim about the shape of the earth is inferred. The limitation here is that there is no way for a human to bypass human perception and belief to ascertain the truth value of any such proposition.
One might be tempted to propose a knowledge-testing algorithm. A proposition could be run through the system and evaluated on each of the three traditional criteria. Is it a belief? Is it justified? Is it true? It is conceivable that the system could be programmed with objective criteria to ascertain belief and justification. But the problem is with truth. How can the truth requirement for knowledge be tested? The answer is: Although one can “test,” such a test cannot yield a definitive result because in principle truth could only be ascertained with a God’s eye view, independent of human perception. There can be strong justification for a truth claim, but there is no test for the certain truth value of a belief, and therefore no test as to whether such belief would qualify as certain knowledge.
Now, I am not a skeptic and I do not claim that knowledge is impossible. Many beliefs are justified, and many are true. Indeed, if the earth has the shape of an oblate spheroid, then one who believes that the earth has the shape of an oblate spheroid, and has adequate justification for doing so, has such knowledge. Indeed, science gives justification for such a belief — very strong justification. But justification, no matter how strong, is not identical to truth. If we try to use the JTB model to test for truth, we fail because all we can have are varying degrees of justification — from no justification to overwhelming justification. As Laurence Bonjour states:
We have no such immediate and unproblematic access to truth, and it is for this reason that justification comes into the picture. The basic role of justification is that of a means to truth, a more directly attainable mediating link between our subjective starting point and our objective goal. We cannot, in most cases at least, bring it about directly that our beliefs are true, but we can presumably bring it about directly (though perhaps only in the long run) that they are epistemically justified (176).
The correct way to think about JTB is that it is a conditional argument, not an empirical test. If P1 and P2 and P3, then C. But if the truth value of P1 and P2 can be ascertained, but not the truth value of P3, then we cannot determine the truth value of C. This is the problem we encounter when attempting to use JTB as a test for knowledge. We may be able to establish that a person has a belief, and even that he is justified in that belief, but unfortunately a human cannot establish whether, independent of his perception, a proposition is objectively true. Therefore, the truth value of the conclusion cannot be determined. We therefore could possibly have knowledge, but to claim we have knowledge would not be rationally justified if using the JTB model as a test.
Let me illustrate my claim with another example. Mr. and Mrs. Jones park their car in a restaurant parking lot and go inside to eat dinner. During dinner, Mr. Jones expresses the belief that his car is in the parking lot. His belief is justified by the fact that he parked it there, and he has the keys to the car in his pocket. But does Mr. Jones have knowledge that his car is in the lot? Two criteria for knowledge, i.e., belief and justification, seem to have been met. The third — truth — is met if and only if Mr. Jones’s car is indeed in the lot. But Mr. Jones does not know if this is the case. That is to say, he does not know whether or not he has knowledge. What he can rightly claim is that if his car is in the lot, then he has knowledge that it is in the lot. Still seeking knowledge, Mr. Jones asks Mrs. Jones to go outside and check for the car, which she does. She returns and reports that she saw the car in the lot. Now does Mr. Jones have knowledge that his car is in the lot? The answer remains the same: If, and only if, the car is in the lot, does he have such knowledge. But didn’t the test (sending Mrs. Jones to check) establish truth? It did not, because it remains possible that Mr. Jones is mistaken: Someone could have stolen the car between the time his wife saw their car and when she reported her finding to him. It may have rolled away. She may have mistaken another car for theirs. But, we can say that Mr. Jones’s justification for his belief that the car is in the lot has become considerably stronger. Relentless in his quest for truth, Mr. Jones goes to the lot and himself sees the car. He touches it. His key fits the door and the ignition. The engine starts. Now does Mr. Jones know that the car is in the lot? Certainly, the justification for believing so has become overwhelming. But it is nonetheless possible that someone replaced Mr. Jones’s car with a counterfeit. It is also possible that Mr. Jones is dreaming[3] or hallucinating, and doesn’t observe his car at all. These scenarios may seem extremely unlikely, but they are logically possible. As before, Mr. Jones has knowledge that his car is in the lot if and only if the car is actually there — and no person knows this for absolute certain — the most a person can have is some degree of justification for believing it is so. There is no way that a human being can escape his own mind to verify — one hundred percent — that the car is in the lot. The best he can hope to do is obtain overwhelming justification for believing that it is the case[4].
The problem can also be expressed in the following way: There is a distinction between justification, which has a subjective mode of existence[5], and truth, which has an objective mode of existence. That is to say, justification is constructed in the human agent’s mind, whereas truth value is a fact of the world independent of human perception. Because justification and truth are two distinct things, we must not conflate the two. If I put the dog in the yard and go back inside the house, I might think I am justified in believing the dog is still in the yard. This is a subjective belief. Whether the dog is actually in the yard is an objective fact about the world; my subjective belief is a separate thing. If the dog is in the yard, it will be so no matter what my belief; if the dog is not in the yard, my belief that it is cannot make it so. And it is not hard to think of a scenario where my belief would be wrong: The dog may have jumped the fence, or may have quietly snuck back into the house behind me. But what if I then peer through a window and see the dog in the yard? One might be tempted to say that I have now discovered an objective fact — the dog is in the yard — via my sense of sight. But, is my sense of sight infallible? If I hold a finger close to my face and see two fingers, can I trust my sense of sight? Indeed, in that case it has misled me. Likewise, when I see the dog in the yard, it is possible I am being misled. I am not seeing an objective fact about the world, I am merely acquiring sense-data evidence from which I infer that the dog is in the yard[6]. That is to say, I am justified in believing that the dog is in the yard, and this is a subjective belief. Whether the dog is actually in the yard is an objective question of fact about the world. If and only if the world is such that the dog is in the yard do I have knowledge that the dog is in the yard.
We can see that the JTB model for knowledge is properly understood as a conditional argument, not an empirical test. Knowledge exists when all three criteria, viz., justified, true, and belief, have been met. But, there is no test a human can perform on a belief that will establish its truth independent of the perception of the believer, because such a test is logically impossible in that it would require a person to escape his own beliefs and perceptions. What a human can hope to do is accumulate massive amounts of justification from which to infer the truth of his belief about the objective state of the world. It is possible that he has acquired knowledge, if and only if the world is such that his belief is true. But the catch is that she cannot know that she has knowledge, because the JTB model, as a test, at best proves inconclusive. The essential point is that according to the JTB analysis of knowledge, it is possible for a person to have propositional knowledge, but logically impossible for a person to have knowledge that she has such knowledge. I understand that this is a paradox and it may be hard for one to accept, i.e., that she might have knowledge but no way to know it. But at least she could be justified in believing she had knowledge.[7]
Works Cited
Bonjour, Laurence. “The Concept of Epistemic Justification.” Crumley II, Jack S. Readings in Epistemology. Mountain View: Mayfield Publishing Company, 1999. Print.
Crumley II, Jack S. An Introduction to Epistemology. Mountain View: mayfield Publishing Company, 1999. Print.
Searle, John R. Mind: A Brief Introduction. New York: Oxford University Press, 2004. Print.
[1]Some have suggested that additional criteria are necessary. Further, exactly what constitutes justification, and the definition of truth, are not universally recognized.
[2] Per Crumley, most believe that truth is a necessary condition for knowledge, even if they do not concur with the traditional analysis in its entirety (38). Further, Crumley adopts a nonepistemic, as well as a correspondence theory of truth, viz., the truth of a proposition depends on its correspondence to the state of the world, independently of human beliefs (40).
[3] Descartes, in his Meditations on First Philosophy, warned of trusting perceptions, because they may be a dream.
[4] This is called induction, which Hume claims is not rationally justifiable — precisely because it is never a certainty. Modern science, which relies upon induction, accepts this limitation.
[5] I am assuming an internalist theory of justification, which holds that the reasonability of a belief is judged relative to the agent’s cognitive perspective (Crumley II 159).
[6] I am invoking the theory of indirect realism, which holds that people have access to sense-data representations of objects, but not the objects in themselves (Searle 184–186).
[7] Many consider the JTB criteria as sufficient for knowledge; but a limitation is, as argued in this paper, that truth of a proposition cannot be tested such that the test yields certainty independent of human assessment.
