The Given Theory

The Given Theory came up in our discussion of BonJour's argument against basic beliefs. According to this Given Theory, it is possible to have a "cognitive grasp" of some facts without having beliefs or evidence about those facts. One gets ahold of those facts in a way that is more immediate and direct than having beliefs or evidence about them. For example, suppose you have a toothache. When you have the toothache, you're aware that you have it. How should we explain your awareness of your toothache? According to the Given Theorist, it would be wrong to explain this in terms of evidence and supporting beliefs. You can just directly and immediately apprehend the toothache. The toothache is "given" to your mind--your mind "gets directly ahold of" it--in a way that doesn't require any evidence. This all happens before the issue of believing ever arises. (Perhaps that's why toothaches are so hard to ignore?) Your mind's direct apprehension of your toothache is not itself a belief; but it has the power to support or justify beliefs about your toothache.

Here's a picture which might help illustrate this view:

The Given Theory is relevant to our current discussion in two ways. First of all, this is a popular way to explain how we come to have justified beliefs about our own sensations and experiences. So if one wanted to defend a foundationalist view where the basic beliefs concerned one's own sense-data, for example, one might want to appeal to this kind of story. The Given Theory would be a story about what makes one's beliefs about sense-data justified. These beliefs are justified because you directly apprehend your own sense-data, and the beliefs are based on those direct apprehensions.

(The Given Theory is a popular way to explain how we come to have justified beliefs about our own sensations and experiences, but it is not the only way. There are also other ways one might try to explain our justification for beliefs about our own sense-data. Just to take an example: you might take a reliabilist line about these beliefs. You might say that our beliefs about our own sense-data are especially reliable, and this is why they are justified.)

Of course, you might wonder, if the direct apprehension is justifying the belief, what is justifying the direct apprehension? That's a good question. We'll come back to that in a moment.

The second way in which the Given Theory is relevant to our current discussion is as follows. BonJour argued that if you're to be justified in believing some belief B, you have to be "in cognitive possession"of some information about your belief: namely, the information that it has some feature F and that this makes B likely to be true. The Given Theorist can accept this move of BonJour's, but claim that the way we're "in cognitive possession" of that information is by directly apprehending it, not by believing it. Hence the fact that our belief B requires us to have that information doesn't preclude B from being a basic belief. To be basic, a belief just has to not rest on other beliefs. It's allowed to rest on direct apprehensions.

Of course, if the Given Theorist is going to make this move, then once again we're going to want to know: If these direct apprehensions are playing a role in justifying B, what is justifying the direct apprehensions? This is exactly the question that BonJour asks.

The Given Theorist's key idea is that direct apprehensions were a kind of cognitive state which couldn't ever be wrong or mistaken. They were infallible. If you directly apprehend some sense-datum, then the sense-datum has to be there, and it can't be other than you apprehend it. Hence, the question of justifying these states shouldn't arise. They can't ever be wrong, so they don't need to be justified.

The way the Given Theorists made direct apprehensions infallible is by denying that they have truth-conditions or propositional content. That is, unlike beliefs, which can be true or false, direct apprehensions aren't the sort of thing which can be true or false. Directly apprehending your sense-datum is a way of knowing that thing, the sense-datum, not a way of knowing some fact about the sense-datum. Contrast "knowing Bill" and "knowing that Bill is so-and-so." When you know Bill, this is a matter of standing in some special relation to Bill. It doesn't seem to be a matter of accepting certain propositions about Bill. There is a related difference between "seeing Bill," on the one hand, and "seeing that Bill is so-and-so," or "seeing that a certain person is Bill," on the other. Seeing Bill is standing in a certain relation to the person, Bill, not a matter of standing in a certain relation to propositions about Bill. Propositions can be true or false, but Bill himself cannot be true or false. Hence, if directly apprehending a toothache is a matter of knowing the thing, your toothache, rather than knowing any propositions about your toothache, then it will be clear why your direct apprehensions cannot be false.

Here is an example of how this view gets spelled out. Bertrand Russell held a version of the Given Theory. Instead of "directly apprehending" your toothache, he talked about "being acquainted with" your toothache. But the basic idea was the same. Russell wrote that:

and:

and:

The picture that emerges from these quotes is this. Some of your beliefs rest upon other beliefs, but ultimately, at the foundation, all your beliefs rest upon acquaintance. If you're acquainted with some fact, for example the fact that you have a toothache, this makes you justified in believing that you have a toothache. (As Russell puts it, the truth that you have a toothache is "self-evident" to you.) And acquaintance is not itself based on or mediated by any beliefs or evidence or knowledge. It's a way of directly becoming aware of your own sensations and experiences.

Now, like other Given Theorists, Russell thought that what made acquaintance such a secure foundation for our beliefs is that it was a way of being related to things--our own sensations and experiences--rather than to propositions or truths. If acquaintance is fundamentally a relation to things, rather than to propositions, then this will preclude you from ever having false states of acquaintance. Russell writes:

Let's add this into our picture of the Given Theory:

This is a funny view. Why should we think that direct apprehensions have no truth-conditions or propositional content? Some Given Theorists arrived at that view in the following way.

1. If a cognitive state has truth-conditions or propositional content, then there are situations in which it could be true and situations in which it could be false.

2. If there are situations in which the state's content could be false, then the state must not be infallible. When you are in the state, the possibility would always remain open that it is false.

3. So if the state is to be infallible, then it cannot have truth-conditions or propositional content.

A number of comments.

First, the step from "the state has truth-conditions" to "there are situations in which it could be false" is doubtful. The belief that "1+1=2" has truth-conditions. That doesn't show that there are situations in which it could be false.

Second, the step from "there are situations in which the state's content can be false" to "the state is fallible" is also doubtful. Consider the belief "I exist." There are situations in which that belief's content is false: that is, situations in which I don't exist. But at the same time, the belief is infallible: whenever I hold that belief, it has to be true. So it doesn't follow, from the fact that there are situations in which some state's content can be false, that whenever you are in the state, the possibility will remain open that it is false.

Most importantly, even if we accept that direct apprehensions have no truth-conditions, it's not clear how that would enable them to do the work that the Given Theorist wants them to do. True, the direct apprehensions couldn't ever be false or mistaken, if they have no truth-conditions. But that means they couldn't ever be true or correct, either. So how could they justify us in holding any beliefs, which do aim to be true or false? Why should a given direct apprehension justify a belief with any one propositional content over a belief with some other propositional content? How do we get from "knowing the toothache" to knowing something about the toothache?

This last complaint constitutes the central part of BonJour's argument against the Given Theory.

(Notice how, in the quotes we looked at above, Russell slips between talking about acquaintance as a kind of "knowledge of things"and talking about our being acquainted with "facts." This slippage can make it hard to tell whether acquaintance is supposed to have propositional content. After all, facts are a lot like propositions. But the official line, for Russell, is that acquaintance is a relation to things, not to propositions.)

BonJour's Argument

BonJour's argument takes the form of a dilemma. Either our direct apprehensions have propositional content, or they do not.

1. If they do have propositional content, then BonJour admits it's possible to see how they could provide justification for the beliefs we base on them. But BonJour argues that in this case, our direct apprehensions would themselves "require justification." He writes:

   [The intuition or direct apprehension] seems to be a cognitive state, perhaps somehow of a more rudimentary sort than a belief, which involves the thesis or assertion that-p. Now, if this is correct, it is easy enough to understand in a rough sort of way how an intuition can serve to justify a belief with this same assertive content. The problem is to understand why the intuition, involving as it does the cognitive thesis that-p, does not itself require justification. (p. 10)

   BonJour's complaint here is that, insofar as we think of "direct apprehension" as a contentful state, like belief, involving the "cognitive grasp" of some state of affairs, and the "thesis or assertion" that that state of affairs obtains, then it seems like "direct apprehensions" would require justification, just as beliefs do.

2. BonJour continues by turning to the other option, which says that our direct apprehensions don't have propositional content. In that case, then perhaps they can escape the demand that they be justified. But it becomes obscure how these direct apprehensions, without propositional content, could justify us in believing that certain specific propositions are true. Why those propositions rather than others? Why any propositions at all? BonJour writes:

   If, on the other hand, an intuition is not a cognitive state and thus involves no cognitive grasp of the state of affairs in question, then the need for a justification for the intuition is obviated, but at the serious cost of making it difficult to see how the intuition is supposed to justify the belief... How does the intuition give [the subject] a reason for thinking that his belief is true or likely to be true? (p. 10)

   What BonJour has in mind here is this. For "direct apprehensions" not to require justification, we have to think of them as unlike beliefs. We have to imagine that having a toothache "given" to you is like being in the same room as your friend Bill, or seeing Bill, and less like believing certain things about Bill. (It's not even like believing that Bill exists.) If we do that, it becomes understandable why your direct apprehension of your toothache, or having the toothache "given" to you, isn't the sort of thing that requires justifying. But it also becomes unclear how "direct apprehensions" could provide you with justification for any given proposition. You can't "read any proposition off" of the direct apprehension, since the direct apprehension itself doesn't have any propositional content.

So:

Either way, it seems a mystery how "direct apprehensions" can play the role the Doctrine of the Given assigns to them.

BonJour thinks that, without a more informative story about what "direct apprehensions" are and how they work, it's ad hoc to suppose they can have just those features in common with belief which enable them to provide justification, but not those features which make them in turn need justifying. (Indeed, it's not even clear that these are different features: BonJour thinks that it's one and the same feature of belief--its "assertive content"--which makes the belief able to provide justification to other states, and itself need justifying.)

Maybe BonJour's criticisms can be answered, if there are natural connections between relations to things and relations to certain propositions about those things.

For instance, if you directly apprehend a toothache, perhaps there is a natural connection between that state and a certain proposition, namely the proposition that you have a toothache, or that a toothache exists. We can defend this connection by saying that part of what it is to be a toothache is that whenever a subject apprehends a toothache, he or she will apprehend it as a toothache.

But notice: here we're no longer talking about apprehending things. Now we're talking about apprehending things as toothaches. And that is very close to apprehending truths or propositions.

Also, how does it help if the only way you can apprehend toothaches is by apprehending them as toothaches? By itself, that is compatible with its also being possible to apprehend other things (e.g. tickles) as toothaches, as well. If so, you would face an epistemological problem whenever you apprehended something as a toothache, of knowing whether or not it really was a toothache. The only way to avoid that problem would be to insist that the only things which can be apprehended as toothaches are toothaches. In other words, whenever your apprehension has the content that something is a toothache, that content must be true. This is what would really be crucial. It takes apprehensions of truths or propositions, to the effect that something is a toothache, to do the epistemological work. Apprehensions of things which happen to be toothaches is not enough. For the epistemologist, it's not what you're aware of, but how you're aware of it, that matters.

I have already expressed sympathy for 1/2 of BonJour's argument. I agree that it's obscure how direct apprehensions could justify us in holding certain beliefs, if those direct apprehensions don't have any propositional content. But the other half of BonJour's argument doesn't seem as convincing. If some state does have propositional content, why should it follow that the state "needs justifying"?

Perhaps BonJour is assuming that only beliefs, or belief-like states, which clearly do need justifying, are capable of having propositional content. But many philosophers nowadays say that our perceptual experiences have propositional content. I have an experience as of there being a table in front of me; this experience represents the world as being a certain way, and its propositional content is that the world really is that way. Yet no one would say that my experience needs justifying. I may need justification for believing I have the experience, but that's another matter. The experience itself is not the sort of thing which could be--and does not need to be--justified.

But perhaps I'm misunderstanding BonJour. When he says that these contentful states--here, the perceptual experiences--"need justifying," maybe he means something else. Not that the experiences themselves have to be justified. Perhaps what he means is that with contentful states like these, his demand for a metajustification is legitimate. It's legitimate to ask: Why should being based on these states be a feature that makes a belief justified? Why do we think that when we're in one of these propositional states, the proposition they represent is likely to be true?

Behind these questions is BonJour's conservative idea that if your experience as of a table makes you justified in believing there is a table, you also need to be justified in believing that there is in general a reliable connection between experiences as of tables and tables really being there. Liberals and externalists reject this conservative idea.

If we held a liberal foundationalist view which says that some of our perceptual beliefs about the external world are justified, then we might say that having an experience with a given content justifies you in holding a belief with the same content. (Of course, the justification your experience gives you has to be defeasible. If you get evidence that your experiences were produced by an evil neuroscientist, you would no longer be justified in believing that the world is the way your experience represents it as being.) I defend a view like this in a paper I wrote, called "The Skeptic and the Dogmatist." A copy of this should be on reserve in Robbins.

That would be occupying one of the horns that BonJour criticizes in his argument against the Given Theory. As I said, I share BonJour's misgivings about the other horn, which says that certain states without content can justify our beliefs.

But even if we accepted the liberal foundationalist view I just described, it would only give us a story about what justifies us in believing things like "There is a table." We haven't said what justifies us in beliefs about our own minds, beliefs about what sensations and experiences we're having. I think this is much harder. With the perceptual beliefs, we have an experience with a certain content, and that's supposed to justify us in holding a belief with the same content. This story won't work for beliefs about our own sensations and experiences. We don't seem to have any experiences of our own sensations and experiences. None of our mental states seem to represent our own sensations and experiences in the same way that our perceptual experiences represent the external world to us. So for beliefs about our own sensations and experiences, we need some other story about what makes them justified.

I don't know how that story should go. I agree with BonJour that the Given Theory seems problematic here. And I don't want to say that what makes our beliefs about our own sensations and experiences justified is just the fact that they're reliable. So I'm stumped. Do you have any good ideas?