Kevin McMahon

Philosophy 326

What is Knowledge?

What does it take to truly “know” something? There are many different concepts that we accept and take for granted but, of these, what can be considered knowledge? For instance, I believe that I am sitting here typing on my computer. This is something that I feel very secure in saying that I know. The most famous example of skepticism (doubting knowledge of anything) is the Cartesian demon example; this is where the skeptic says that I am not seeing a computer but a demon is clouding my mind and implanting the thought processes and stimuli equivalent to my actually experiencing a computer in front of me. Since it is conceivable that I could be deceived in such a manner, how do I know that I see a computer. Even if I am right, and there is a computer in front of me, but the demon is not letting me see the computer that is actually in front of me, just a recreation of said computer, do I still know that the computer is there. Do I have actual knowledge of it? In this paper we will compare two different philosophers’ ideas on how to sort out what is knowledge and what is not, we will look at the positions of Keith Lehrer in Theory of Knowledge and Robert Nozick in Philosophical Explanations. Then we will examine the problems with both, and where they disagree. Finally we will come back and try to sort out the problems and reach a new solution and or definition of knowledge.

In Theory of Knowledge Lehrer sets forth a system of justification as the doorway to knowledge. There are different types of justification that have to be combined to get us to knowledge. They are: personal justification, verifical justification and undefeatable justification. Personal justification is an internal thing, everyone has an individual acceptance system built in. It is what we weigh information against to see if it conforms with what we already believe. Lehrer refers to this system of comparisons as coherence. We have to maintain a coherence between what we want to know and what we already believe. According to Lehrer it should take the form that any concept P is said to be personally justified for that person if it coheres with their personal acceptance system. He deals with the demon hypothesis by saying that it is more reasonable for him to believe based on his personal belief system that he is, for example, seeing my computer, than it is to believe that a demon is influencing his mind. He says this because he has evidence to believe he should trust his eyes and that it fits with his acceptance system and he has no evidence to the contrary to prove that there is a demon. For example, it is part of my acceptance system that I write papers for school and I turn them in to receive grades; I also have evidence that I am actually writing it because it has happened before and right now I am feeling the same sensory stimulus as I normally do. Since I have no conclusive evidence against the theory that I am actually in front of my computer or even to suggest otherwise, and I have some evidence to support said theory, I am personally justified in believing that I am in fact in front of the computer. The skeptic though would still claim that since it is impossible to be one hundred percent sure that there is no demon, since we have no way of testing the hypotheses, then we cannot know either way.

The next step in the process to get to knowledge is verifical justification. Here we start with the same acceptance system used to gives us personal justification for the things that we believe and then we add another clause to it. Since all of us may accidentally accept things that are incorrect (for example it was once widely thought that the earth was the center of the universe and this was accepted as fact), we must make a clause that says that the claims used to “accept” a new idea as true must also all be true. This new clause involves removing all false beliefs from the acceptance system used. Therefore it would not be knowledge if one were to base his belief in Martians on the false belief that Orson Wells’ War of the Worlds radio play was actually true. A subsystem of the acceptance system is then created, Lehrer calls it the “verific system” (Lehrer, 1990; 134). Now, the new idea P must be both personally justified as well as verifically justified. Lehrer uses the supposition that there is some omniscient being which can make a list of all true and untrue things in your personal acceptance system (134). This would be the only way to truly remove all of the untrue assumptions in our personal acceptance system. Without someone telling you specifically that an idea you have is false, how do you know what is untrue and therefore that the idea you are accepting based on the falsity is also untrue? The skeptic would reply that since we all could be living in a “big lie” society where all of our television and print media are constantly running false things like the War of the Worlds example, it is not possible to sort out the true ideas from the false without Lehrer’s “omniscient being.”

The final move in Lehrer is to reach undefeated justification. To reach undefeated justification Lehrer says we have to replace every statement found to be false by the verific system with the opposite statement. For example, if I hold the belief that the world is flat but it is shown that this is false, then not only do I have to remove the belief “the world is flat” from my acceptance system, I have to replace it with “the world is not flat.” This forces the acceptance system to then drop any other fact that would have been related to one which was shown false. For example, assume I believe that the world is flat, and based on the idea that the world is flat, I also accept that I will die if I sail over the horizon. If it is later proven to me that the world is not flat, I have to remove the fact that the world is flat from my acceptance system; it is still plausible to believe that I will fall off the earth at the horizon based on other things since there is no statement of the nature of the earth. We have to add the negation and make it, “the world is NOT flat” and therefore it is not reasonable to believe you will fall off the earth at the horizon. Adding the negation is a much stronger claim than just removing something from your acceptance system, you might remove something for lack of evidence, but you negate it based on truth. This to Lehrer is what defines knowledge, a personally and verifically justified belief that cannot be defeated by any false statement. Since no false statements remain in our acceptance system thanks to the final move, all things that survive this three pronged test are knowledge.

Since for Lehrer this justification model is the only way to truly reach knowledge, he does not see the cause of a belief to be as important as Nozick does. Lehrer sees the relationship between a belief and its’ cause as interesting and insightful, but he does not see it as a solution to the problem of Knowledge. Nozick on the other hand believes precisely the opposite in Philosophical Explanations. The two basic camps we are studying in this paper are the externalists and the internalists. Externalism tends to define knowledge as relating to external causes for your beliefs, whereas internalism defines knowledge using a system such as Lehrer’s to internally justify why you belief a fact. Nozick is an externalist. He believes that the only way to understand what is knowledge and what is not is by studying the relationship between beliefs and their causes. The relation he sees as most important is the causal one.

Nozick sets forth four criteria by which to judge whether something is knowledge. The first two are by far the simplest notions. First off, assume that we are trying to determine whether we have actual knowledge of idea P. We must first assume that idea P is actually true because we cannot have knowledge of something that is false. The second tenet is that we must actually believe it to be true for it to be knowledge. These two are only the basis for the system, they leave much open for there is no real relation between them there must be something more.

That something more, to Nozick, is a causal relationship. The fact that the belief is held, must be born out of the fact that the belief is true. If the idea P were not true, one would not hold the belief of idea P. Nozick uses a bit of logical notation to show this, he says ~P®~(the belief that P) (172). This is to say that if someone were to believe that something was true, even if it was not true, then it would fail this test. Earlier I said that I believe I am sitting here typing on my computer, if I were not typing on my computer I would not have the belief that I am because it is born out of my actual typing. If the computer were to be removed from my line of sight then I would no longer have the belief that I was typing on it. Logically, ~(the computer in front of me)®~(the belief that my computer is in front of me). If I were to take a large hit of acid immediately prior to sitting down in front of my computer and start hallucinating to the point that even after the computer was stolen by thieves, I thought I was typing on it, then it would also fail this test. If I did have the belief that my computer was in front of me after it was removed, then my belief is not based causally on the fact that it is so, and therefore, I have no knowledge of it.

The fourth tenet of Nozick’s externalism is the idea that the truth of something must cause the belief of it. P®(the belief that P and not the belief that ~P); if it is true then it absolutely must produce the belief that it is true (178). This might sound redundant but as is illustrated in Nozick’s example, it is necessary. He uses the example of a man in a tank of water that can only get his perceptions and experiences from scientists putting them into his brain. He only “knows” he is in the tank of water because that is the belief that they feed into his mind. His mind is being tricked using modifications of its’ chemistry into thinking it senses the vat. If we can imagine a possible world, Nozick says, where the same man is in the tank and yet they do not give him the sense of it, then he would be in the tank but not have the belief of it. Not having the belief of P when P is true disqualifies it from being true knowledge since it did not come about in the proper causal manner. It is from this relationship that Nozick thinks we get our knowledge. He puts forth a simple and relatively easy to understand system for testing whether our knowledge is earned properly. To him, without the proper causal relationship between facts and our beliefs there is no knowledge.

Lehrer and Nozick are on opposite ends of the spectrum. Nozick feels that the only way to reach knowledge is through causal relationships, and Lehrer feels that it is through a complex system of internal coherence and justification. This could be referred to as the internalist versus the externalist. Lehrer wrote one of his chapters on externalism which was basically a reply to Nozick and others with similar beliefs. Nozick’s book came out long before Lehrer and so it could not deal directly with Lehrer.

Philosophical Explanations did deal with justification as a type of system for gaining knowledge. Unfortunately the way that Nozick simplified the idea of justification is the not exactly the same as Lehrer’s conception of it. Nozick characterized the idea of justification by saying it was the same thing as reliabilism. He did this by saying that a person would be justified in believing something if the method they used to reach their conclusion tends to have a correct outcome more often than not. Although Nozick showed how this idea did not work, it really did not address Lehrer because Lehrer used a system that was vastly different from the justification system that Nozick was trying to refute.

Although Nozick does not directly conflict with Lehrer when he deals with justification que reliabilism, Lehrer does not walk away completely unscathed. In Lehrer, we are relying on an omniscient critic to tell the questioning skeptic everything that is true, and everything that is false in the personal justification system of the individual that holds belief P (134). Since there is no third party to tell us which of our beliefs are acceptable, we may hold many false beliefs and use them to justify other beliefs. The absence of this know-it-all party makes it possible for someone to justify any belief with any number of other false beliefs; since they would be fully justified within their personal system though, they would accept it as knowledge. Nozick’s system relies on our own intuitions about the causal relationship for our beliefs. Take this modified example from Nozick that we briefly mentioned earlier. A man, person A, is suspended in a vat of water and he is not directly aware of his surroundings. His only knowledge of himself or the rest of the world comes from the scientists feeding thoughts/experiences/sense-data into his brain; his brain actually thinks that all of these things have happened, he has not just been told that he is in the vat, they have tricked his brain into believing it has sensed it. They have fed him an entire acceptance system identical to the acceptance system of some other person B that is also floating in a vat of water. The acceptance system and the experiences of person B have all come about in the usual sensory way; he actually senses himself being in the vat. Person A, then, has a belief of being in the vat that is justified based on his acceptance system, and is also true. Not only that, since his acceptance system is all true, it passes the verific test and the undefeated justification test. So, according to Lehrer’s system, it is knowledge that he is in the vat of water.

Let us use the same example within Nozick’s system. First off, it is true that he is in a vat of water, and it is true that he believes it. The problem comes when we get to condition three [~P®~(the belief that P)]. Since all his beliefs come from person B who is in the vat of water, if we slowly pull person A out of the water, and yet continue feeding him the thoughts that he is still in the water he will continue to believe he is. This fails test number three, since he no longer is in the water he should not have the belief that he is still there. Therefore, by Nozick’s system, it is not knowledge that Person A has. This is somewhat like the Cartesian demon, since he is not really perceiving the vat in the normal sensory way, he does not have knowledge of it.

Nozick seems to have won that round against Lehrer’s justification theory, but Lehrer points out a flaw in Nozick’s system also. The problem would be with what Lehrer refers to as the causal fallacy. This is a confusion between how a person came to have a certain belief and the reasons that person has for holding onto that belief. We will start with a slightly modified version of Lehrer’s story of the racist whom he calls, Mr. Raco. Raco is very prejudiced against black people. He feels that there is a certain disease that only affects blacks and does not affect people of any other ethnicity. Let’s say, for simplicity’s sake, the disease is sickle cell anemia. He comes to hold this belief based solely on his belief that he is better and that they are inferior to him, not from any medical knowledge whatsoever. It was purely coincidental that the disease he picked to focus his attention on, really does only affect black people. Later on in his life, Raco decides he wants to become a doctor. He goes to medical school and graduates, he now has his M.D. and is a licensed doctor. He now knows though through the medical knowledge that he has gained about all the evidence that points to the fact that sickle cell anemia only affects people blacks. His medical knowledge is now his reason for believing that only blacks are affected by this disease even though the original cause of his belief was due to his prejudice and racism.

Let us examine this first through Nozick’s analysis of knowledge. The first and second steps are both okay because P is true and Raco does believe that fact P is true. Again, the problem is found in the third step. Even if we assume that P is false, unless we present Raco with an incredible amount of evidence to the contrary, he will still continue to believe that it is true. Therefore since the form ~P®(~the belief of P) fails, it must fail to be knowledge to Nozick. Nozick had a reply for an example such as this but it does not quite stop Lehrer’s example from working.

“1) P is true.

2)S believes, via method or way of coming to believe M, that P.

3)If P weren’t true and S were to use M to arrive at a belief whether (or not) P, then S wouldn’t believe, via M, that P.

4)If P were true and S were to use M to arrive at a belief whether (or not) P, then S would believe, via M, that P.”(Nozick, 179)

This is how Nozick characterized an argument of this type. Let P mean “sickle cell anemia only affects blacks”, M be his original method of racism, and S be Raco. Substituting these in for the variables in Nozicks proof, the statement would then still lead to a problem, again occurring at step 3. Because of his excessive racism, even if P weren’t true, Raco would still believe that it was. This has the consequence of still not being considered knowledge under Nozick’s externalist system.

If we run this same example through Lehrer’s system we get a different result. Still, the first two steps are fine; it is both true and believed to be true that sickle cell anemia only affects blacks. According to the personal acceptance theory of Raco, blacks are the only victims of the disease. It is verifically justified also, because even if we do not allow him to use the false racist claims to prove it, he can stand on his medical expertise to get him through without using any falsities. Also it is undefeatable, because none of the falsities logically exclude anything else, so it survives all of Lehrer’s tests. Clearly and correctly it survives as knowledge.

It does seem that we can use Lehrer to correctly illustrate the knowledge, or lack thereof, in the racist example. With a few simple additions though, I think Nozick can survive too. Although he originally came to believe that only blacks are affected by sickle cell due to his racism, he was eventually able to justify this belief better thanks to his medical knowledge. If in the previous example given by Nozick himself, we substitute for M his medical knowledge it should pass the test. Now, if we take away the truth of it and say that sickle cell affects all people, his medical knowledge would no longer give him the belief that P. We would need to specify in Nozick’s response that when we are given multiple methods M of reaching a given belief, as long as one of them satisfies all of the requirements, then it is knowledge. That would solve the problem in this example since using the medical understanding clearly meets all of Nozick’s requirements even though the racist cause definitely does not.

Lehrer’s system, at first glance, felt more right; upon further examination though, it did not stand up as well as Nozick’s. Without the causal link in there it seems to easy to accept certain things as true. It seems as though Lehrer is making the age old blunder that knowledge is simply a justified true belief; sure he sweetened the argument a little but the basis is the same. He added the fact that the belief must be verifically justified, meaning that it cannot be proved wrong by some omniscient creature. If we had something to set up a verifical system to tell us which beliefs are true and can therefore be used to justify things, we could simply ask it for all knowledge and then we would know everything. This omniscient creature test is really begging the question of knowledge; you have to have knowledge to gain knowledge.

Nozick’s externalist system, with one minor addition, came through all tests put to it with flying colors. It sets up more than simply another argument for a justified true belief. His link to the external cause of a belief really helps us understand where knowledge comes from. His system does though, in a sense, still incorporate a little bit of justification because of the third step. If belief P were not true and you still had the belief of P, it would fail Nozick’s test of knowledge; your belief in P has to come from the fact that P is actually true and not from elsewhere. It would seem then that the cause of your belief is your justification for it.

Nozick, clearly then, had the right idea about what is knowledge. It does have to be true and justified, but it also has to have that something more. There must be a link between the belief and its’ cause. The fact that a belief is true must be that which brought about its’ belief in the first place. This is how we can understand our knowledge. The externalist system, presented by Nozick,wins out over Lehrer’s internalist model.

Home Outdoors Various School Projects Rat Babies Wacky Test Results Contact Kevin McMahon