As I note in my “About” page, I sometimes deal with the topic of philosophy. In this sense, I usually mean moral philosophy (or perhaps also political philosophy, but I usually label that under politics). My fascination with philosophy began with my intro to reasoning (i.e. logic) class with Dr. Williford (no longer at SCSU) in my first year here at the university, and continued with my business ethics course with Dr. Shaffer in my third year here. It now continues with my course in epistemology (i.e. the philosophical study of knowledge) with Dr. Mirza. One of the central topics dealt with was what constituted knowledge. In this sense, the philosophical question is much different than the other philosophical questions that I have dealt with in the past on this blog. In any case, I find it fascinating (even if irrelevant), and the label of “philosophy” in my About page does not preclude it, so I’ll explore the topic below in further detail. If you are not interested in the philosophical question of knowledge, then this would be a good time to stop reading (and keep in mind this is not one my ordinary posts on this blog, as I usually deal with politics or economics; see the categories in the box on the left). If, on the other hand, you are interested in the epistemic question of knowledge, then please bear with me as I explore the concept of philosophical skepticism. πŸ™‚

For well over two millennia, philosophers have dealt with the question of what knowledge is. Plato gave, but did not necessarily commit to, the basic description as “justified true belief.” That is, knowledge is justified true belief. More formally, subject, S, knows proposition, p, if and only if p is true, S believes that p, and S is justified in believing that p. It is easier to read if formulated like this:

S knows that p, iff (if and only if):
1. p is true
2. S believes that p
3. S is justified in believing that p

Thus, the three necessary and sufficient conditions for knowledge are truth, belief, and justification. This has become known as the justified true belief analysis of knowledge. The purpose is to find the conditions for knowledge such that when these conditions are met, one can be said to have knowledge (sufficiency) and that every case of knowledge meets these conditions (necessity). Put differently, if I meet these three conditions I have knowledge; and if I have knowledge, I meet these three conditions.

Thus, the justified true belief analysis of knowledge was generally accepted as the correct analysis for knowledge since Plato crudely formulated it. And it seems correct on the surface. I cannot be said to know something that is false. I cannot “know” it is raining outside currently if it is, in fact, not currently raining outside. Further, I also cannot be said to know something if I don’t actually believe it. If it is raining outside, but I do not believe it to be raining outside, I cannot be said to know that it is raining outside. Moreover, it seems I cannot “know” something if I am not properly justified. I may have the true belief that I will win the lottery tonight. That is, I might actually win the lottery tonight and I believed I would. However, if I simply formulated my belief on conjecture, some supernatural explanation, etc., I am not ordinarily said to have knowledge. If I merely make a lucky guess about some proposition, it’s not really knowledge. Thus, I need some justification for my beliefs. If I am properly justified in my belief that is true, then I can be said to have knowledge. For example, I observe that it is raining outside and I therefore formulate the belief that it is raining outside. Since it is true that is it raining outside (and I believe it and am justified in believing it), I can be said to know it’s raining outside.

In 1963, one Edmund Gettier turned this two-thousand-year-old analysis on its head with his single three-page paper. In it, he describes cases in which a person has a justified true belief but cannot be said to have knowledge (i.e. the three conditions are not sufficient for knowledge). Such counterexamples to the traditional analysis have come to be known as “Gettier problems” or “Gettier counterexamples.” Let us explore such a counterexample. The following case is one similar to the original Gettier presented in his seminal paper. Suppose Smith is a close friend of Jones. Smith formulates the belief that Jones owns a Ford, for Smith has always seen Jones drive his Ford, has seen the documentation for Jones’ ownership of Ford, has no reason to suspect Jones does not in fact own the Ford he drives around, and so on. Smith is entirely justified in believing Jones owns a Ford. However, Smith has been deceived. The documentation Smith saw was authentic-looking but was in fact fraudulent; in reality, Jones did not own the Ford he was driving around and was merely deceiving Smith. Thus, Smith’s justified belief that Jones owns the Ford is false. Smith does not have knowledge because he failed to satisfy all three of the previous conditions for knowledge (namely, the condition that the proposition believed is true). However, Smith is in a room with Jones and another man whom he has never met before, Jake. At the same time, Smith deduces from his (untrue) belief about Jones’ ownership of Ford that “Someone in this room owns a Ford.” Smith believes this proposition based on his false belief that Jones owns a Ford. As it just happens, however, Jake does own a Ford (but Jones does not). As we see, then, Smith has the justified true belief that “Someone in this room owns a Ford.” Smith was justified in believing “Someone in this room owns a Ford” because he validly deduced it from the justified belief that Jones owned a Ford. However, Jones did not own a Ford, but it still remains true that “Someone in this room owns a Ford” because Jake, by sheer circumstance, does in fact own a Ford. Since Smith had no previous knowledge of Jake (much less what car he drove), then it seems Smith’s belief that “Someone in this room owns a Ford” is only made true by the lucky circumstance that someone he with whom he is unfamiliar owns a Ford. Intuitively, we do not say lucky circumstances amount to knowledge. Gettier showed the justified true belief of knowledge to be insufficient, because all three conditions were met but knowledge was not obtained.

I apologize if the foregoing case was not very clear. Gettier probably explains it much better than I can, and you can read his paper here. The actual car example comes from Keith Lehrer, in this paper. Of course, Wikipedia explains it pretty well. A much simpler version that I like is that of a broken clock. Let’s say I purchase a brand new clock, set it up, and observe it for some time and ascertain that it is telling time accurately. I am justified in believing the time it displays because clocks are generally reliable sources for telling time. Several days later, however, I come home and observe the clock displays the time of 6:07. Based on this, I formulate the justified belief that it is 6:07. Unbeknownst to me, the clock stopped working 12 hours earlier. By sheer luck, though, the actual time is 6:07. Thus, my justified belief is true. However, we would not ordinarily say I “know” that it is 6:07 because of my reliance on a broken clock. Had I looked at it at any other time, my belief about the time would be false. Therefore, I was lucky and luck does not count for knowledge.

There have been many attempts to solve this puzzle. Some have said we need to find a fourth condition that eliminates the possibility of Gettier problems (called “JTB+G”). Others have decided to throw out the justification condition and find other conditions in its stead. (It is almost universally agreed that truth and belief are necessary but insufficient conditions for knowledge. That “justification” is necessary for knowledge has been argued against, for example by Robert Nozick.) The literature on this is quite vast so I will not review it in this post. The Wikipedia article I linked to does a satisfactory job in explaining the various approaches, if you are interested, as does this site.

Suffice it to say that one response has been that the justification condition on which we rely is not strong enough. It is possible that even completely justified beliefs are in fact wrong. If the premises on which I rely are always (logically) possibly wrong, then any account that relies on “justification” will always be susceptible to a Gettier-like problem. This line of thinking is referred to as fallibilism. Kirkham has a good defense of fallibilism here.

Fallibilism rests on the idea of skepticism. Epistemic skepticism says that any belief based on empiricism (that is, empirical evidence or reliance on sensory perception) could logically possibly be false. The arguments for this are quite nuanced and detailed. That I am seeing a computer before me could conceivably be false. To wit, there could be a conceivable world where my senses are being deceived perhaps because I am hallucinating or because some evil genius scientists is controlling my thoughts and is merely giving me the false belief that I see a computer before me. Thus it is logically possible that I am wrong in my perceptual beliefs. In fact, claims the skeptic, it is logically possible that all beliefs based on empiricism are false (see, e.g. the brain in a vat theory). If it is logically possible that all our beliefs based on perception are false, then we cannot be said to know anything for sure. That is the argument of skepticism. Again, though, the arguments for and against skepticism are nuanced and will not be reviewed in this post. (If you are really interested, this book has a good overview.)

In conclusion, since it appears it is logically possible that any (or every) empirical belief we have about things could be wrong, the case for knowledge should be stronger. That is, it is only possible for us to “know” when it is not logically possible that our belief is fallible (that is, there is no conceivable or possible world where it is wrong). Again, see the Kirkham paper I linked to for a further exposition of this argument.

This is a powerful argument, and it is one I appeal to. Of course many people reject this argument on many grounds and I do not mean to imply that Dr. Mirza endorses such an argument. It is merely one I have decided is the most convincing of those provided. Keep in mind, please, that I have also just condensed an entire semester of lecture, discussion, argument, and thought into about 1,500 words. The issue is much more complex and nuanced than this. I merely offer a rough overview of this topic that I find fascinating. I admit I probably did a rather poor exposition here, as my purpose was simply to formulate my thoughts about this subject in some words and I apologize if it is not exactly coherent. I would be pleased to clarify if anyone is interested.

If you were at all put off by this post, I won’t take it against you, but you should not worry because these types of philosophical posts probably will not be the norm for this blog. Cheers. πŸ™‚