Spring 2016, NYU Abu Dhabi

Knowledge and Facts

Knowing that P

The kind of knowledge we'll be concerned with in this course is called factual or propositional knowledge. That's where you know something to be a fact. For instance, you know that Princeton is located in New Jersey.

There are some cases where we use the word "know" but we're not talking about factual knowledge. For instance:

I know my mother pretty well.

That's not the same--or at least, it's not obviously the same--as knowing any particular fact about my mother. Another example:

I know how to walk.

That too is not the same--or at least, it's not obviously the same--as knowing any fact. There might be this really smart baby who reads all the books about the physiology and mechanics of walking, so he knows all the relevant facts. But he might not yet know how to walk.

In this class, we're just going to focus on factual knowledge, the kind of knowledge that's involved when you know some fact, when you know that something is the case.

Knowing a Fact

Some verbs are what philosophers and linguists call factive verbs. For instance, consider the verb "recognizes that Georgie has talent." In order for you to recognize that Georgie has talent, it seems like Georgie does have to have talent. If George lacks talent, we're uncomfortable saying that anybody "recognizes" Georgie's talent. They can at best believe that Georgie has talent, expect him to do well, and so on.

Perhaps this is because "Tom recognizes Georgie's talent" entails "Georgie has talent." Or perhaps, more plausibly, it's because it presupposes that Georgie has talent, in the same way that "Tom has stopped beating his wife" presupposes that Tom used to beat his wife.

Similarly, if you discover that P, then it seems like P has to really be the case. Same goes for overlooking that P, forgetting that P, admitting that P, being happy or lucky that P, and regretting that P. If you regret that P, it seems like P really has to be true. If P isn't true, but you just think it's true, then the thought that P might make you feel bad. But we won't feel comfortable saying that you "regret that P." Nor will we feel comfortable saying that you "discovered that P." You can only seem to discover it, or think you discovered it.

Here's a subtle but important distinction: suppose you ignore the weather forecast and pack a picnic without taking any umbrella or tent along. As it turns out, the day is sunny and clear, contrary to what the weather services had predicted. In this case, we might naturally say you overlooked that it might rain, or that it was supposed to rain, even though (fortunately for you) it didn't rain. Should we conclude then that the verb "overlooked" isn't factive?

No, because it really was a fact that it might rain, and it really was a fact that it was supposed to rain. Those are the things we can rightly say you overlooked. Would we also naturally say that you overlooked that it will rain, or that it did rain---when it didn't? This doesn't sound like natural English to me. Hence, I think the verb "overlook" really is factive. If you overlook that P, P must really be true. It's just that sometimes P is a truth like "it might rain" instead of a truth like "it did rain."

It's an unsettled question in linguistics exactly what mechanism underlies these phenomena, and what verbs display it. And I could be wrong about how some of these examples work. But I think they'll help you get the feel of this common linguistic pattern.

Verbs like "predict" and "report," on the other hand, are not factive. If I predict that P is the case, it does not follow that P really is the case.

Now what about the verbs "believe" and "know"?

"Believe" seems not to be factive: I can believe that P but be wrong.

"Knows," on the other hand, seems to be another example of a factive verb. It doesn't make sense to say:

In fact Sue is guilty of stealing the car, but the jury knows that she is innocent.

If you said that, it would sound like you were contradicting yourself. It'd be okay to say:

In fact Sue is guilty of stealing the car, but the jury believes that she is innocent.


In fact Sue is guilty of stealing the car, but the jury thinks it knows that she is innocent.

But it seems like, for the jury to really know that Sue is innocent, it has to be the case that Sue is innocent. You can't know things which are false. You can only seem to know them, or falsely think that you know them.

So this is one important fact about knowledge. If you know that P, it follows that P is true.

Does this mean the same as saying that knowledge requires absolutely certain evidence? No. The claim that knowledge is factive does not entail that:

Knowledge has to be based on indefeasible, absolutely certain evidence.

As we said a moment ago, things like forgetting that P, or being lucky that P, also require that P be true, and forgetting that P or being lucky that P presumably don't require you to have absolutely certain evidence that P.

The fallibilist agrees that knowledge is factive. On his view, you can know P on the basis of fallible evidence, but only if P is also true. If there are other people who believe things on the basis of the same kinds of fallible evidence as you, but their beliefs are false, then their beliefs won't count as knowledge on anybody's view.

This point often confuses students, so make sure that you've thought it through and understood it.

The fallibilist says: to know P, you need to have good evidence for P, and in addition, P has to be true. (The evidence by itself usually won't guarantee that P is true.)

The Certainty Principle, on the other hand, says: to know P, your evidence has to be maximally good. It has to be so good that no one could have that evidence without P's being true.

It may help you better understand the relation between fallibilism and knowledge's being factive if you think about the difference between tests and definitions. Consider the question: "Was Bush lying when he said Saddam had weapons of mass destruction?" Defining what we mean by lying isn't so hard. There may be some subtleties to it, but it doesn't seem like it would be that hard. It's a matter of what was going on in Bush's head when he said that. On the other hand, trying to construct a way to test whether Bush was lying might be extremely hard.

In the same way, what it means for knowledge to be factive is that part of the definition of your knowing that P is that P is true. That's not the whole definition, but it's part of it. That doesn't mean you'll have any easy way to test whether you know something. Defining what knowledge is, and coming up with a test for whether you have some knowledge, aren't the same thing. They're related, but they're not the same thing.

The fallibilist agrees that part of the definition of your knowing that P is that P is true. He's not yet said anything about how we should test for whether someone has knowledge, or how hard it will be to apply such a test. He's only saying that it's sometimes possible to know that P on the basis of evidence that falls short of being maximally good.

Sometimes students think, "According to the fallibilist, if Mars really isn't made of coffee, then your fallible evidence will be good enough to give you knowledge that it isn't. But it won't be good enough for you to know that you know." I guess they're thinking that knowing that you know requires you to be absolutely certain that you know, and hence, absolutely certain that Mars isn't made of coffee. But it's not clear why that should be so. If we accept the fallibilist's view that you can know things about Mars on the basis of less-than-certain evidence, why shouldn't we also be able to know whether we have knowledge, on the basis of less-than-certain evidence?

We'll talk more about this later.

The Truth Is "Out There"

We said that in order for you to know that P, P has to be true. Nonetheless, there is an important difference between what's true, on the one hand, and what we know or can know to be the case, on the other.

Let's think about the Matrix. Is the world that Trinity and Cypher experience when they're inside the Matrix, and seem to interact with, "less real" than the world they "awaken to" when they take Morpheus' red pill?

The standard view is that "yes," it is less real. As Morpheus says, the Matrix is "a dream world." The characters are just experiencing a "neural interactive simulation" of eating steak, dodging bullets, and so on. In truth, they never have eaten steak, and never will. It just seems to them that they have. And this would be so even if no one ever discovered that it was so; even if no one ever figured out that the Matrix was just a "dream world."

Philosophers would express this view by saying that facts about whether you've ever eaten steak, whether your eyes have ever been open, and so on, are all objective facts: facts that are true (or false) independently of what anybody believes or knows about them, or has evidence for believing.

Most facts seem to be objective facts. For instance, Mt. Everest is 8850 m tall. That's how far it rises above sea level. And Mt. Everest doesn't care very much about whether we exist or what we know. Mt. Everest would still be 8850 m tall no matter what any of us knew or believed or had evidence for believing.

You might wonder: what if we changed our minds about where to draw the boundaries of Mt. Everest? Wouldn't that make it have a different height? It would not. What we'd have done in that case would be to change what object it is we're talking about when we use the word "Mt. Everest." The old object--the object we used to call "Mt. Everest"--would still have the same height. And that's all that's being claimed by the philosophers who say that the facts about Mt. Everest's height are objective. They're claiming that what height a certain object has is an objective fact, that does not depend in any way upon people's beliefs or knowledge about it. If people come to have different beliefs, then it might be that their words no longer refer to that object. But we're talking about the object itself, not our words.

As a matter of fact, we call one hunk of rock "Mt. Everest," but we didn't have to call it that. We might not have called it anything at all. And how high that hunk of rock is is independent of what we call it or believe or know about it.

Van Inwagen has a good discussion of this in his selection in the coursepack. (He also discusses a case where we change our minds about what to count as "a meter.")

Some people get uneasy when you talk about "objective facts." They say: "Well, what's true for you might be different than what's true for me." I discuss this kind of langauge in my article What's so bad about living in the Matrix? Some of the things people might mean when they say this are quite plausibly true; but this is a misleading way for them to express themselves and what they're saying doesn't conflict with the claim that there are lots of objective facts. There are views some philosophers hold which do conflict with the claim that there are many---or even any---objective facts. But as I explain in the article, these are radical philosophical views that need to be argued for. They should not be assumed as a starting point.

Another claim that sometimes gets thrown up in opposition to "objective facts" is that knowledge is just a social construction. Let's try to get a handle on what this means. To do that, we need to introduce some background ideas.

A realist about Xs is someone who believes that Xs really exist, that they aren't mere fictions. Realists about Xs also think that Xs aren't radically different from the sort of thing we thought they were all along. And they think that the facts about Xs are independent of what we believe or have evidence for believing.

For example, a realist about the external world is someone who believes that there really are trees and mountains and so on; he thinks these are real things "out there" in the world, and that they're not just ideas in our mind, or constructions out of our experiences.

Another example. A realist about wealth believes that some people really are rich, and other people really are poor. That is, he believes that people really do have those properties. He needn't think that people who are rich will always be rich, or that they deserve to be rich, or anything like that. A realist about wealth just says that some people really are, as a matter of fact, and at some times, rich, and others are poor.

A reductionist about certain things or properties is someone who thinks that facts about those things or properties can be reduced to, or explained away in terms of, facts about something else. For instance, most of us believe that biology reduces to chemistry. There are no distinctive, brute biological facts. It's really all just chemistry. Facts about wealth might be reduced to facts about people and the ways they interact with each other.

Sometimes reductionists count as realists, and sometimes they don't. It depends on what they're reducing things to. It's hard to come up with any general rule that tells you for all cases whether a given reductionist is a realist or not. If you think that facts about trees reduce to facts about atoms and electrons, then philosophers will still count you as a realist about trees. If you think that facts about trees reduce to facts about what ideas and experiences people are having, then you don't count as a realist. That's too different from what we ordinarily take trees to be.

Let's consider the property of being fashionable. It is plausible to expect we can reduce this property to certain facts about us: to some of our social conventions, and our beliefs and intentions about what will be fashionable. The fashions aren't "out there" waiting for us to discover them. Instead, we make certain clothing styles fashionable, by treating them as fashionable. We have authority about what the fashions will be. Because of that, there are limits to what kinds of mistakes we can make about what's fashionable. We can make isolated mistakes--and some of us make more mistakes than others--but we can't all be systematically wrong about what's in fashion. Rather, the fashions are partly determined by what we believe they are.

The claim that "knowledge is a social construction" means that knowledge is like fashions. It means that facts about who knows what can also be reduced to facts about our social conventions. Suppose there's this guy Curtis who lives in the East Village and he wears plaid pants from the 70s that look really cool, he has hipster glasses, and so on. He's very stylish. He's also this serious researcher about cell biology. So he knows a lot about cells. Now, when you say that fashions are a "social construction," that means that the rest of us could get together and agree that the plaid pants and all of that will no longer be fashionable. And we can thereby change him from being fashionable to being unfashionable, behind his back. It's plausible that fashions are like that. If you say that knowledge is a "social construction," that means that we could also get together and agree that he won't know things about cells anymore. Just because we decided he won't.

I hope you realize how stupid that is. Of course, we can decide to say that Curtis doesn't have knowledge any more. But that wouldn't make it so. We can also take away his books and computer, so he won't be able to gather any more evidence. We could even brainwash him so that he forgets everything he knows about cells. What we can't do is make him stop having knowledge, just by agreeing that he won't know things anymore. So knowledge isn't a "social construction," in the way that fashions are.

I expect that some people who say "knowledge is a social construction" aren't really expressing themselves properly. What they really mean is something like this:

"Social and political factors play a large role in determining what evidence is gathered and preserved, who has access to it, and how knowledge gets disseminated."

Well, that's all true. But it doesn't show that knowledge is literally a "social construction," in the way that fashions are.

Here's an interesting way in which what you know might depend on facts about your social community. Suppose I live in a group with lax epistemic standards, and you live in a group with stricter standards. And suppose we both believe that Mt. Everest is 8850 meters tall, on the basis of the same body of evidence. Might it be that I count as knowing that fact, but you don't, because the evidence we're relying on is good enough to meet my society's standards but not good enough to meet yours? This is an interesting question. We'll be talking about it more in coming weeks.

Even if the answer to that question is "yes," though, that still won't show that knowledge is a "social construction." For the views we'll discuss later don't say that knowledge only depends on the epistemic standards of your society. It also depends on other things, such as whether it's really true that Mt. Everest is 8850 meters tall. That fact is not a "social construction." So these views would not make facts about who knows what reducible solely to facts about our social conventions.

Objectivity and Knowledge

Some people are uneasy with talk about "objective facts" because they think we're saying we have some superior insight or knowledge than everybody else. But that thought is wrong. The claim that certain facts are objective has nothing to do with anyone's having a superior epistemic position. If we say that the facts about Mt. Everest's height are objective, we're saying that whatever height it has, its having that height is independent of us and our beliefs and epistemic powers. We needn't also claim that we know what Mt. Everest's height is, or that we are better placed to know its height than other people are. Those are further claims, and they require further argument and defense.

The only matters I'll be claiming any special authority over in this class are the thought-experiments I ask us to consider. When I construct a thought-experiment, it's my story. So what I say is happening in the story goes. But if we're talking about the world we really inhabit, I'm presumably no better placed than the rest of you. Maybe we know what the world we really inhabit is really like, maybe we don't. That remains to be seen.

If we think it's an objective matter whether there are mountains of a certain height, we can go on to say:

  1. there are mountains of that height, and we know that there are

But we can also go on to say:

  1. there aren't any mountains, but there could have been (and if there were any, that would be because of how things were in the world outside us, not because of our beliefs or evidence)

And we can even go on to say, as the skeptic does, that:

  1. we can't know whether or not there are any mountains.

All three of those views count the facts about mountains as objective. They think that what it is, or what it would be, for there to be mountains of such-and-such a height is a matter of how certain piles of rock are configured, and has nothing to do with us and what we can believe or know.

Sometimes philosophers use the expression "realism about X" to mean "the view that the facts about X are objective." On this use, the skeptic counts as a realist about the external world. He's a realist who thinks that our epistemic powers are very limited. Other times philosophers use "realism about Xs" to mean "the view that Xs really exist, and the facts about them are objective." That's the way I've been using the term "realism." On this use, view (i) counts as realist, but the skeptic's view (iii) does not, and neither does view (ii).

It's very confusing that "realism" is used in both these ways. van Inwagen proposes using "Realism" with a capital "R" for realism in the first sense, and "realism" with a small "r" for realism in the second sense. I think that's still confusing. Here's what I'll do. I'll use "realism" in the way I've been using it all along, namely, to mean "the view that Xs really exist, and the facts about Xs are objective." I'll introduce a new word, objectivism, to mean "the view that the facts about Xs are objective." (This is not a standard philosophical usage; I'm just making it up right now.)

So with this terminology, view (i) is a realist view, but views (ii) and (iii) are not. But all three views are objectivist views. The skeptic and the realist are both objectivists. They agree that whether there is a mountain of a certain height does not depend on us and our epistemic powers.

In the Matrix article, I talk about a view called verificationism. Verificationists think that whether there is a mountain of a given height does depend on whether we know or can know there to be such a mountain. They say that:

  1. if all the evidence says there's a mountain there, then there really is a mountain there. It's impossible for us to be systematically deluded about such things.


As I say in the article, some philosophers have defended verificationism. But...that's not saying very much. Philosophers have at various times defended all sorts of crazy views:

None of these views is very plausible, is it? Verificationism isn't very plausible either. That doesn't mean we shouldn't talk about it. It's OK to discuss and investigate what the arguments for an implausible view might be. Often that can be philosophically illuminating. But I hope that you will have your guard up. I hope that you'll want to see a really damn good argument for verificationism, before you'll be tempted to accept it. (And if you think you have a really good argument, I hope that you'll scrutinize it really closely, and anticipate that people are going to be suspicious.)

Many of the philosophers who adopt verificationism do so because they think that it's impossible for a realist to resist the skeptic's arguments. They think that, faced with a choice between accepting skepticism or giving up objectivism, it's better to give up the objectivism.

Well, that will be a central focus of this class, whether it really is impossible for a realist to resist the skeptic's arguments. I hope you will agree that it's important for us first to get a clear understanding of what a realist's theory of knowledge would look like, and what the realist's options are for resisting skepticism, before we give up on objectivism and decide to become verificationists. That's what we'll be doing in this class.

There are other arguments for verificationism, too, coming from the philosophy of language and philosophy of mind. Those arguments are much too complicated for us to explore here. You may deal with them in other philosophy classes.

In brief, those arguments go something like this: "If there were objective external facts, then the skeptic is right that we wouldn't be able to have knowledge of them. If we couldn't have knowledge of them, then we wouldn't even be able to think or talk about them. So the only facts we can think or talk about must be non-objective facts, like the ones the verificationist countenances."

As you can see, these arguments also rest on the assumption that it's impossible for an realist to resist the skeptic's arguments. So here too it's a good idea to first understand what are the best things a realist can say in reply to skepticism.

So far, I've emphasized how alien the verificationist's view of the world is to our common-sense picture of things. To be sure, that does not show that verificationism is incorrect. But it should put the burden of proof on the verificationist. Philosophical argument might eventually convince us that objectivism is incorrect. But objectivism should at least be our starting point. A sensible way for us to proceed is to assume that our ordinary, objectivist picture of the world is correct, until we encounter good arguments that persuade us otherwise.