a new home

I've decided that I like the look of wordpress better. So, I'm moving this infant into her new home.


Join me there! There's coffee brewing and biscotti on the table.

http://doingphilosophy.wordpress.com

Gettier's Challenge to the Traditional Analysis of Knowledge

This is one of the first philosophy blogs that I wrote. I wrote it about a year ago, during my first semester studying philosophy:

Knowledge is an interesting concept. What makes it interesting is that we use the word all the time, and presumably we know what we mean when we say it. But when pressed for a definition, we cannot provide one. We stumble around in the dark, fumbling with our words, usually using "know" to define itself! So what is this elusive concept? Do we know what knowledge is?

It turns out that we both do and don't. Philosophers identify three different types of knowledge. The first is knowledge by acquaintance. This is generally understood as a sort of familiarity with a person, place or really any sort of noun, perhaps even adjective or verb, where you have an interactive relationship with something. The second is know-how. This is reasonably self-explanatory, dealing with our knowledge of procedures for accomplishing tasks. The final is propositional knowledge. This is defined as knowing the truth value of statements, which are either true or false.

The interplay between these three facets of knowledge is quite interesting. Take for example Tokyo. I have never been to Tokyo. I am not acquainted with it at all. However, I do know some propositions about it. I know that it is the capital of Japan, that it is a densely populated city, that it is on the island of Honshu, etc. So it is clear that it is possible to know propositions about things and not have knowledge by acquaintance of those same things. Similarly, it is possible to know someone by acquaintance, but not know any propositional truths about them. Take for example a university. One could know a lot of propositional truths about a school, but have never been there. If asked, "Do you know Harvard?" the person might respond, "Yes," thinking of all the things he knows about Harvard. But when the follow-up question is, "Is the food good?" he would be unable to answer. Also, a person might live at Harvard and be very knowledgeable about it. But when asked what the enrollment is, he would be unable to answer. Similar examples can be made with regard to know-how and propositional knowledge.

Philosophers tend to focus on propositional knowledge (although I think that as Christian philosophers we should be mostly concerned with knowledge by acquaintance, both with God and others; the other types of knowledge flow out of that). But philosophy deals with propositional knowledge, so for now that's what I will focus on.

So then, what is knowledge? Well, there are several different ways of defining something. You can point to something that is an instance of the thing to be defined and say, "There; that's what I'm talking about." This is to give an ostensive definition. So I could point to instances of knowledge or give examples of things I know, and say, "This is knowledge." That's the way that we come to learn most definitions, starting from the time we were babies. (That claim is open to debate) Our parents would hold up a block and say, "Block." But what's to say that you understand block as the whole concept and that you don't confuse it with wood or cube or the color it is? So an ostensive definition never really gets us to the essence of being of the thing to be defined. I could give you hundreds of examples of where knowledge has occurred, but that doesn't get you much closer to actually understanding what constitutes an instance of knowledge. Perhaps a better way to define knowledge is to give a whole-part analysis. So to define a car, I would say that it has doors, a motor, wheels, windows, etc. I would point to parts of the car, give you as many as possible, and from that you could somewhat construct in your mind what I'm talking about. This seems to be a better way to define things than giving an ostensive definition because I am giving you individual things that, when put together, actually get you at the essence of what a car is. But that still doesn't seem to be enough.

The kind of definition that philosophers seek to give of a thing is called the intension. That is, what one can say about the things essence. This usually involves giving the necessary and sufficient conditions for it, such that if and only if those conditions obtain, then it counts as a genuine instance of the thing being defined.

So let's apply that sort of definition to knowledge. Philosophers have asserted for some time that knowledge is something akin to justified true belief. In fact, after centuries of counting knowledge as justified true belief, many philosophers came to think that perhaps knowledge just is justified true belief. That is, justification, truth and belief are each necessary and jointly sufficient to count as knowledge. If you believe something, you are justified in believing it and it is true, it is knowledge.

However, in 1963 a philosopher named Edmund Gettier published a paper called Is Knowledge Justified True Belief? He asked us to assume just two things. 1) It is possible to have a justified false belief. This happens all the time. We have good reason to believe something is true, so we believe it, and it turns out that we are wrong. 2) If a person is justified in believing a proposition p and justified in believing that p entails another proposition q, then the person is justified in believing that q. [ Jp & J(p->q) --> Jq ]That is classically accepted as true, referred to as the closure principle.

With just these two assumptions, Gettier provides two cases which challenge the sufficiency claim of the justified true belief (JTB) analysis of knowledge. I will present one of the cases to you, then another Gettier-like case which I believe to be stronger.

Gettier case: Suppose that Smith and Jones have applied for a certain job. And suppose that Smith has strong evidence for the following conjunctive proposition:

d. Jones is the man who will get the job, and Jones has ten coins in his pocket.

Smith's evidence for (d) might be that the president of the company assured him that Jones would in the end be selected, and that he, Smith, had counted the coins in Jones's pocket ten minutes ago. Proposition (d) entails:

e. The man who will get the job has ten coins in his pocket.

Let us suppose that Smith sees the entailment from (d) to (e), and accepts (e) on the grounds of (d), for which he has strong evidence. In this case, Smith is clearly justified in believing that (e) is true.

But imagine, further, that unknown to Smith, he himself, not Jones, will get the job. And, also, unknown to Smith, he himself has ten coins in his pocket. Proposition (e) is then true, though proposition (d), from which Smith inferred (e), is false. In our example, then, all of the following are true: 1) (e) is true, 2) Smith believes that (e), and 3) Smith is justified in believing that (e). But it is equally clear that Smith does not know that (e); for (e) is true in virtue of the number of coins in Smith's pocket, while Smith does not know how many coins are in Smith's pocket, and bases his belief that (e) on a count of the coins in Jones's pocket, whom he falsely believes to be the man who will get the job.

Gettier-like case: A man is driving down a country road. He sees what appear to be sheep and forms the belief that there are sheep in the field. In fact, what he saw were very hairy dogs. But, there are some sheep hiding behind some trees. So his belief that there are sheep in the field is true, but we would not say that he has knowledge.

As you can see, armed with the proper method, anyone can construct any number of Gettier cases. The formula seems to be constructing a case whereby what makes the belief true (in the sheep case, the actual sheep that are hiding) is not what justifies the belief (the dogs).

What makes Gettier-like cases work is that they appeal to our pre-theoretical concept of knowledge. We are trying to analyze and give necessary and sufficient conditions for something that we seem to intuitively understand. Gettier appeals to that intuition, saying (in effect), "Look at these cases. Under the TAK, these people have knowledge. But our intuition says they don't." An easy way out is just to assert that Gettier-like cases are instances of knowledge. But then we're changing knowledge from what most people intuitively think it to be. And the job of philosophy should be to define concepts as they are used in the pre-theoretical sense. It won't do to change knowledge from what we thought it to be in order to be able to provide necessary and sufficient conditions for it.

Another approach to the Gettier problem is to attempt to reduce the propositions in the Gettier cases to other propositions which themselves are either not justified or which do not provide inference to the conclusion. For example, (e) is probably not the belief that Smith holds. That proposition seems ambiguous (because of 'the man'). He probably simply believes (d), and (d) does not imply (e) without a change in the texture of the belief.

I think that Gettier-like cases (although not necessarily the Gettier cases themselves) do pose a problem for JTB. Perhaps the original Gettier cases can be reformulated in order to save the TAK, but I do not think that can be done for all Gettier-like cases. Gettier seems to be on to something: JTB is not sufficient for knowledge. There must be a fourth condition, and I think it must do something to connect the justification with the truth. That is, what makes the belief true (not 'truthmaker' used in the technical sense) must somehow be what justifies the belief.

There have obviously been attempts toward this. The three most promising are a causal theory of justification, a "no defeaters" condition and a "no false grounds" condition. The first claims that there must be some sort of causal chain between the truth of the belief and the justification for it. But it is not clear that beliefs must be causally related to each other in order to preserve inference. The "no defeaters" condition states there can no be no propositions which, if the "knower" was aware of them, would make his belief unjustified. But that doesn't seem to hold either. For example, in the sheep illustration, suppose there was a man who, if he encountered the "knower," would tell him that he kept his dogs in that field and they looked like sheep. In that case, the "knower" would not be justified. But suppose also that that man is a pathological liar. In that case, it seems like the "knower" would still be justified (still wrong, but justified), in believing that there were sheep in the field. However, if he didn't know that the man was a liar, he would have an unjustified true belief that there are no sheep in the field. So in order for the "no defeaters" theory to hold, we must either modify it to a "no true defeaters" theory or make some sort of chain from the defeaters all the way back to some basic grounding truth. Neither of those will hold. The "no false grounds" condition maintains that it is not an instance of knowledge if the justification is based on a false belief. But say that the man in the field also sees an actual sheep. So he believes that everything he sees is a sheep. There are five dogs that he thinks are sheep, so part of his justification is based on a false belief, and therefore according to "no false grounds" he would not have knowledge. But it seems to us like he would. So "no false grounds" will not work as stated.

Is there a fourth condition out there? Probably. I don't know what it is. Do you?

First post

This is my first post. In most blogs I visit, the first post is nothing substantial. Not wishing to buck the trend, this post will say little to nothing. Since I have already accomplished that, I have reached my goal and shall end.