5.1 Introduction to Knowledge

So this week's topic is mainly knowledge, with a little bit more to say to round off skepticism. Our four luminaries there, we've got Freddy, AH, who was a professor at New College when I was an undergraduate. Edmund Gettier, famous for writing a three-page paper in the journal Analysis in 1963. I’m not sure whether he's published anything since, but it suffices to make him very famous. Then we've got Hilary Putnam, who we will come to at the very end, and Tim Williamson on the right, who is a professor here and a successor to AJA. Tim's work isn't actually going to feature in this lecture, or indeed in the readings you have, but he's one of the most prominent epistemologists in the world, so it's as well to know that he's around. If you find yourself giving a paper and you see him in the audience, beware. He's universally feared for the sharpness and precision of his questions. Okay, so we've seen some skeptical arguments, most famously those of Descartes. And those sorts of arguments rather suggest that if we put a threshold for knowledge very high, then we're quite likely to be driven to the conclusion that we don't know anything at all. Descartes' own answers don’t seem to work very well. Other answers are all very controversial. It's rather tempting to try to get round this problem by redefining the notion of knowledge to provide a useful distinction amongst the beliefs we have. Maybe none of them are only very, very few, like "I exist," will actually reach the highest threshold, but surely it makes sense to try to define a more moderate, more reasonable threshold. Because we do want to distinguish between things that we know in a perfectly ordinary sense, and things that we don't know. So we naturally get the question, what is knowledge? How should we understand the notion of knowledge? Now, questions of the form "What is X" feature quite prominently in philosophy. If you go back to Plato and look in his dialogues, you'll see that Socrates is always asking this sort of question. In fact, it used to be the case, I think, that people thought of this sort of thing as absolutely paradigmatic of what philosophers do. Philosophers search for essences by trying to define things. I don't think you'll find it's nearly as prominent these days, but such questions still come up quite a lot in topics like personal identity or freedom. What do we mean by freedom? What is freedom? Now, such questions, if you think about it, are rather puzzling because they could just be asking, when do we apply the word X, where X is freedom, knowledge, or whatever. But that sort of question seems to be just about our use of language. Don't we want to go deeper than that, to ask what is a genuine case of X? If that question isn't just about our use of language, then what is it? It seems rather peculiar. What could knowledge be other than what we refer to using the word "knowledge"? So let me give you an example. Take the discipline of geography. Suppose that the study of geography started out as the study of places in terms of their location, physical characteristics, mineral resources, the natural environment, that sort of thing. I'm not sure whether that was true, but let's suppose that it was. Then over time, people became interested in things like land use, economic considerations, maybe even culture. And if you studied geography now, you will find that culture is one of the things that get studied. Now, you can imagine someone saying, "That's all very well, you now study culture, part of geography, but is culture really part of geography? Does geography really include cultural things?" Well, if the word "geography" is now used to cover cultural matters amongst others, then sure, the discipline of geography includes culture. How could it not? So it might well look as though the kinds of questions we're asking when we ask "What is X?" what is knowledge, just come down to language. And in the 1950s and '60s, Oxford philosophy was famously identified with ordinary language philosophy, as though the purpose of philosophy was just getting clear about how we use ordinary language. If that were all there is to it, then it would be rather an uninteresting kind of question. But with most of the concepts that interest philosophers, there is something deeper at stake. Take the case of freedom, which we'll be looking at in a week or two. There, we're not just interested in how we use the word "freedom." We want to know what kinds of acts we should describe as free, because the notion of freedom is tied to moral responsibility. We think it matters whether somebody is free. It could turn out that we describe actions as free when really, from a God's eye point of view, they're not. Maybe we describe people as free in certain circumstances in ordinary life, but actually, if we knew about it, there's no moral responsibility there, no genuine freedom. So there is a deeper metaphysical question underlying the linguistic question. Now likewise, in the case of knowledge, the concept of knowledge has a normative aspect. When we say something's knowledge, we're not just categorizing it as something that is called "knowledge." We're saying that it's reliable, that it has a certain authority. So, it is possible to ask of a particular belief, "Well, everyone says they know this, but is it really knowledge? Do they really know it?" Again, a similar issue arises with Strawson's response to the problem of induction, on your induction reading list. He famously says that inductive methods just are what we mean by "reasonable." When we describe an inference about the world as reasonable, that just means it meets inductive standards. But a very well-known answer to Strawson is to say, "Hang on a minute, no. When we say that a method of inference is reasonable, we're not just saying that this is the kind of inference that everybody calls reasonable. We actually mean that it is reasonable, that it has normative force, that this kind of inference really does convey assurance to the conclusion." The kind of conceptual analysis that we're doing on the concept of knowledge provides a nice example of this, and it's a good reason for having it in this general philosophy course. From this example, you can get an idea of the kinds of things that typically pop up in these sorts of discussions. So one of the things that often comes up is an appeal to linguistic intuitions. Now when people talk about intuition, it's sometimes a bit sloppy, as though they're saying, "Oh well, this is just something I think, an intuition you’ve got to accept it." But actually, linguistic intuitions have a particular authority because if you're a native expert speaker of your language, then certain things do just come naturally to you to say. And that does carry some authority for the standard use of language. Now obviously, that doesn't necessarily tell you anything about philosophical truth, but it does keep you on the rails of using language correctly. Puzzle cases also feature a lot, as we'll see in this lecture. Some people call these intuition pumps. The idea of a puzzle case is that you sketch out some hypothetical scenario, and then you ask, "Well, what would you say about that?" And clearly, what you try to do is devise puzzle cases which steer your hearers' intuitions in the way you want to take. So often, you'll find in philosophical debate, each side is producing puzzle cases to favor their own particular point of view. You'll find in personal identity, for example, puzzle cases feature quite highly. Then obviously, we get argument and we get systematization. We try to pull all these intuitions and thoughts together to make sense of them all together. So let's now embark on a discussion of knowledge and its variants, trying to employ some of these methods to straighten out what we want to say about it. First of all, let's distinguish between three different kinds of knowledge: acquaintance, knowing how, and knowing that. What we're interested in here is propositional knowledge, knowledge that P, a strange phrase, but you'll find philosophers use it quite a lot. So P is a proposition, could be any old proposition. But notice, knowing that P is the case is quite different from having acquaintance with somebody or something or having practical knowledge. For example, I know how to ride a bike. I don't exactly know how I do it. I'm ignorant of all sorts of propositions that would explain how I managed to remain upright on a bike. But I have the practical knowledge, that is, I can actually do it. That's irrelevant here. We're talking about factual knowledge.