The Problem of Self-Knowledge

This is about the problem of self-knowledge. And one of the hardest things in this area is motivating the problem of self-knowledge, saying why it is a problem. Most of us are intimately acquainted with our thoughts and feelings, with our current states of mind. And they're known to us in a way in which they're known to no one else. So you might wonder why philosophers should find it problematic that we have self-knowledge. It's hard at first to see why this should create a philosophical problem.

The trouble is that we're inclined to be so immediately aware of our own thoughts and feelings that we imagine them to be accessible to us and to us alone. It's a natural temptation to suppose that, uh, my my thoughts and feelings are part of an inner world that are immediately accessible to me, and that therefore they're removed from or screened off from anybody else. That they occur in a in a separate realm in which I can study or reflect on my states and which no one else gains access to.

But that that intuition is countered when we think of other people. When we think of other people, we take it for granted that we know quite a lot about other people's minds. We know about their thoughts, their aspirations, their hopes, their fears, their worries, their intentions. And we do this on the basis of observing what they do and what they say. We find it very easy to attribute states of mind to other people. And therefore, it's as if common sense seems to provide us with conflicting intuitions: intuitions that our own states of mind are private and accessible, and intuitions that other people's minds can be known.

Now, if I'm sure that I know something about what my friends and acquaintances are thinking, what they feel or preoccupies them, what that look means, and what I'm meant to do about it, if I think I know this about other people, then it's only a comforting illusion that other people are not in a similar position to know this about me. So I must think about my own mental states that they too are revealed to others, perhaps much more so than I I appreciate. And when we understand that, then we understand something that brings us to the edge of the philosophical problem. The idea that the very same states could be known by you on the basis of behavior and observation, and could be known by me without needing any such evidence or inference. How can it be that one in the same state, one in the same metaphysical state, is, uh, accessible in two quite different ways?

Now, the idea that the idea that other people can know our states of mind doesn't in any way remove the obvious truism that we know ourselves best. We know ourselves better than others know us, and better than we know them. And we have to both capture the intuition that people will have a knowledge of our minds on the basis of behavior, on the basis of what we say and what we do, but that we will still have to account for the very special way in which we ourselves know what we're thinking, know what we intend, know what we desire. There has to be a way in which we are made specially aware of that, especially and differently aware of that, while not at the same time making it impossible to honor the claim that other people can know our thinking to. That's that's part of the setup here.

Now, um, it's part of the idea that other people can know our minds that other people can tell us something quite important about ourselves. They can point out some facts about our psychology, our makeup, our our way of behaving or thinking, which we had overlooked. And that's why we can gain insight about ourselves from others. If that weren't true, it's not obvious why we would spend so much time telling people our problems. We we hope to recite some catalogue of of incidents to gain some understanding back. So self-knowledge is not always arrived at by the self. It can also come from other people who are studying and observing us and making astute judgments, astute, uh, claims about our current states of mind.

But those judgments that others make, and and we might even say that they're in a better position to make them, are usually about long-term dispositional states. They're usually what we call character traits: whether somebody is anxious, whether they're irritable, whether somebody is reliable, whether somebody is thoughtful. These are usually properties that other people can work out better than we can by observing our behavior and generalizing across patterns in our behavior. It's not so obvious, it's not so obvious that subjects are very good at describing their own characters. If I set you the exercise now of writing on a sheet of paper what your character is, you probably will be uncomfortable and less good at it than someone who knows you very well. So dispositional states, general character traits, are not the sort of thing that I claim we have, uh, special and, and it seems, um, immediate knowledge of.

So I'm more interested in states of mind which are the current conscious states of thinking, what I what I currently want or currently think. So if I, if if you ask me, "Do you want a cup of coffee?" I just know. I know on the basis of an immediate response or answer to that question. And if you ask me, "What time do you think the seminar will finish?" I'll tell you. This is something not like a disposition, not like a trait, something which I think I'm in, in the best position to pronounce on. It's states of that sort that I'm interested in here.

Now, let's try to tease out something about what's special about self-knowledge. Because if it is special, it's special because it's not like the knowledge other people have of us, and it's not like other kinds of knowledge. It seems to to try and say, and I haven't said yet what the problem is, to try and say what the problem of self-knowledge is, we have to see why it's special and different from the kind of knowledge we have about the world around us, the kind of knowledge people can have about our states of. And usually people get at this by talking about a number of asymmetries. There are a number of asymmetries between knowledge I have of my own mind and knowledge you can have of my mind, or I can have of yours.

The asymmetries are thought of like this: As far as other people's minds and motives are concerned, my claims to knowledge always have to be based on evidence, observation, and inference. If I want to know what what you think about the current political situation, if I want to know what you want for Christmas, I have to do quite a bit of observing and studying of you. It's not something that I can pronounce on as easily as I can pronounce on my own states of mind. So there's a difference in, uh, in this sense, there's an asymmetry. I need evidence to make judgments about your mental states. I don't seem to need evidence to make judgments about my own state of mind.

If you ask me what I want for Christmas, I don't have to start observing how I behave, seeing which stores I'm looking in, seeing which of two bottles of wine I show a preference for. I don't have to do that. I know. I can tell you. I mean, you can, I can tell you later, but it's not, it's not as though I need to figure out, uh, on the basis of evidence, what I want. Similarly, if I think, "I wonder if you're in the mood for coffee," I might judge that by your, you know, whether or not we're heading towards a coffee shop, whether or not when I ask you if you want one, you you accept. In my own case, I don't have to. You don't say to me, "Do you want coffee?" and I say, "Well, let's see how I do. Let's take me to the cafe, see whether I order one, see whether I I drink it." It's not like that. I just know. So already we have that asymmetry in terms of evidence.

Another asymmetry, another important asymmetry, is the fact that I'm authoritative about what I'm currently thinking or what I want, but I'm not authoritative about what you want. If you ask, "Do you want coffee?" and I say, "Yes," I'm the authority about that judgment. I'm the one who knows best. I'm the one who should be in a position to say. But even if I'm pretty good at figuring out what it is you want, I'm not authoritative. In fact, I will defer to you to find out whether or not what I take you to be currently wanting is is right. I want to find out whether my judgment's correct. So authority is only had in the first person case. That seems another asymmetry, something we have to, um, something we have to account for.

There's another asymmetry in terms of, um, which goes with the idea of evidence. We we don't need a justification. We don't need a justification to say what it is that we are currently thinking or currently feeling. If you, if you ask me, "What time do you think the seminar finishes?" and I say, "Six o'clock," or "Seven o'clock," "Seven o'clock, I guess," "Half past seven," "Half past seven." If I say that, see, I didn't know my own mind. But if, if, if you ask me, "What what do you think?" I actually thought it was sent. So I, I think it's 7 o'clock, maybe false, that's what I what I was thinking. And then you say to me, "But how do you know that's what you're thinking?" Not, "How do you know it's seven o'clock?" but "How do you know that's what you're thinking?" It doesn't seem to be a reasonable question. If you ask me, "How do you know you want a cup of coffee?" it doesn't seem like a reasonable question. It seems as though there's no need for me to give you a reason or a justification for what I say in answer to a question about what I want or what I think.

But if I say, "Pierre thinks the seminar will end at seven o'clock," I do have to have reasons or justification for that to be a knowledge claim. Spouses get into terrible trouble, uh, I gather, when, um, they say, "I know you want such and such," and you say, "How do you know that?" "I just know." "I just know." This is usually a very bad thing to say. You should give some evidence for why you know that someone else is in a certain state of mind. "I know you don't want to see that film." "How do you know this?" "You say, 'I just know.'" "I just..." It's not good enough. There must be some evidence. There must be some reasons that they are used for ascribing that state of mind to you. So there's a difference in terms of justification. And the difference is striking: claims to self-knowledge about current, present mental states seem to be based on no justification at all. If the demand that I justify my claim to know what I'm thinking is unreasonable, then it seems as though there's no justification in the offing.

Now we're creeping up on a bit of a problem. Why should it be the case that claims made without any justification are more authoritative and generally reliable than claims we can make by providing evidence or providing some grounds? That seems to be worth asking as a general question about knowledge. Now, keep that, keep these sets of asymmetries in mind. Let's now go back to the one about authority and ask, uh, what, what that involves. What does it mean to say a subject is authoritative about her own mental states? Well, the authority she has in judging is not, it's not the authority that's granted to you by being in a position of administrative power. So someone, someone like Francis, has administrative power. If he says, "The seminar will finish now," the seminar will finish now. He has the authority to do that. He can make it true. But it's not, it's not, it's not that kind of authority we're talking about when we say that people are authoritative about their mental states.

In this case, we want to say people, uh, are authoritative because, by and large, what they think is usually right. What they think about their current mental states is usually right. They are the, they're in the best position to judge, and what they judge is usually correct. Typically, it's correct. That's what makes them authoritative. That's why we ask them what they're thinking. Okay. Now, such, such first-person authority doesn't amount to infallibility. It doesn't amount to being unable to be wrong about your current state of mind. Perhaps I'm wrong about what I feel about X. Perhaps I'm wrong about what I really want. Beware, beware declarations of what you want. You can be surprised. So maybe there are ways in which I'm self-deceived. Maybe I'm mistaken about my own wants and wishes. Maybe I'm deceived about what I'm thinking. Or maybe I'm simply unself-knowing. Maybe there's some state of mind that has eluded me, and which is important and operative in my psychology, makes a difference to how I behave and how I treat others, but about which I'm wrong.

So perhaps I go to a meeting and I react very, uh, I'm very irate. I throw my hands up. I start getting louder and louder, making a lot of noise. Someone says, "You were very annoyed with the, with the chairman." And I say, "No, no, I wasn't annoyed. I mean, I certainly, it didn't bother me at all." People could say to you, "No, no, look, on the basis of how you behave, just look at how you behave, look at the way you reacted, you were clearly annoyed." Now, when I take the third-person point of view, when I look at my own behavior as others see it, I can recognize, "Yes, that's a better explanation of my behavior than the one I give where I say I was perfectly calm."

I can't resist telling you this story, though. When I edited, uh, this book, um, we had a conference in St. Andrews with Crispin Wright and John McDowell. And at one point, in a very heated discussion, the two of them were speaking at the same time, and neither would give way. And any of you who know John McDowell and Crispin Wright, now they're both excessively polite people, but they were, they would not stop. It was a very bad-tempered exchange. And I came out and I said to Crispin, um, "I wonder, you know, what, why was that so heated?" And he said, "Yes, there seemed to be a certain animus on John's part. I didn't feel anything myself." I thought, "Well, this is a conference on self-knowledge. That's that's a little surprising."

Now, it, you know, people did have to point out rather forcefully, "Look, this wasn't, this wasn't a very, um, well-mannered exchange. This was a very heated exchange." So you can learn about your own mental states from others. You could have overlooked or or failed to look at them. But just because you're not infallible, because it doesn't mean that whenever whatever I say about my mind is right, that's not always true. It's not always true that whatever I think I'm feeling or thinking now, I'm right about it. Because I'm not infallible doesn't mean I'm not authoritative. I'm typically right. I'm usually right when I'm not self-deceived, when I'm paying attention, when I'm not in any, uh, abnormal, unusual cognitive state. My judgments about what I'm thinking are by and large true. And that's something we need to account for. Why that should be, on the basis of no grounds at all, on the basis of no evidence at all? That's what we need to explain.

Now, there's another condition we have to build into an account of self-knowledge that that does the work here. And that's, um, I've talked about asymmetries. I want to talk about symmetries now. So if you look at the handout at the bottom of page two, um, here's a, here's a kind of symmetry that we mustn't lose sight of. There's a, there's a truth-conditional symmetry between ascriptions of mental states between people. And the symmetry is this: If I say, "I want a cup of coffee," that's true, said by me now, if and only if you or he wants a cup of coffee, said by you about me now, is true. Those truth conditions should be the same. There's something wrong in an account that gets the truth conditions to come out differently. That that biconditional should hold. And in general, we have this truth-conditional asymmetry: symmetry. "A believes, desires, intends that P" is true, said by subject A at a time T, if and only if "B believes, desires, intends that P," said by B about A at T. That's got to be in place. If that's not there, we we haven't got an account that illuminates the fact that we know about our own states of mind in a way that others don't, but in knowing about them, we know about the very same states that other people could know about on the basis of inference and evidence. That's what we have to say.

Now, where's the problem? Well, any account of our mental states, and now I do the metaphysics, since since France has invited me to talk about metaphysics, any account of our mental states has to be an account that accommodates these facts. It has to accommodate the fact that there is first and third person access to the very same states of mind, but they're known in different ways, and that the truth-conditional symmetry has to be there while the epistemological asymmetry has to be there. Can there be states? Can we explain the nature of states that that will account for, illuminate how things turn out to be just like this? Truth-conditional symmetries and epistemological asymmetries. What's the nature of states that that afford this, in some sense? Well, the symmetry is one thing, but the knowledge of our own thoughts, sensations, beliefs, desires, intentions, meanings is characteristically different from other people's states of mind. It's typically immediate, authoritative, and indeed effortless.

Now, here's one way I put the problem before. It's on your handout on page three, which now we're getting nearer to the problem. The problem arises when we try to square this objective and outward aspect of the mental with the special way in which we know our own minds from the first-person perspective. How can states whose natures belong partly in the public sphere be so readily available to us from the first-person point of view? That's the difficulty. If, if our mental states can be known by others, then they're part of the world of others. In which case, how can that be? And at the same time, I know about them from the first-person point of view in a completely different way. How can that be?

Looking at the same problem from the first-person perspective, we're inclined to think that we're just automatically aware of what we're thinking because it's part of our our inner world. It's part of the fabric of our inner life. How could we not know it? How could we overlook it? And now you get the other way up. I put the problem again on the handout: How can the facts of mental life be part of the inner world of a subject if they're also objective facts, knowable by others on the basis of outwardly observable behavior? How can they seem to have these two roles? I know that that's the, that's the, the problem. That's the tough metaphysical problem. How can they serve both, both roles? I think that's the problem of self-knowledge as I want to understand it.

But what sort of problem is it? Well, we're required to say what kind of mental states would make it the case that thinkers have this immediate awareness of them, but leave room for others to know them, for others to be able to observe facts about our mental life. As Wittgenstein says, "You see people's minds in their face." There's something right about that. I mean, if you, if you really treated people as just bags of skin moving around about which you had to do very complicated hypotheses to see them as having beliefs and desires, you would not see them as people. It's part of your immediate response to certain creatures in your environment as people, distinctive that you, you attribute mental states, emotions, you respond to minute changes in their facial, uh, expressions and so on. That's got to be part of having, having the mind in view in the world. But how can it then be available from the first person?

Okay, so we need to provide a satisfactory metaphysics of mind that does justice to both first and third person epistemology. So any solution is going to have to reconcile the first and third person roots to the mental. Has to make, make room for them. Now, the involvement of both epistemology and metaphysics here qualifies as a problem. Uh, it's an instance of what Christopher Peacock has recently called the integration challenge. Something we all knew about, but Christopher Peacock gave it a name. So we might as well just use the name. It's very useful. So here's the integration challenge. The integration challenge, Peacock says, on your handout, in its general form, is that of reconciling our metaphysics in any given area with our epistemology for that same area. That's what you've got to do. You've got to have the metaphysics and epistemology straight. They've got to, they've got to combine in some way that that doesn't make it problematic how there can be such states and we can know them in these ways. Okay.

So we need, we need to find, uh, a metaphysics. Is it a substantial metaphysics? And one of the issues I want to explore in this talk, and it's about the status of self-knowledge, is whether so-called deflationary accounts of self-knowledge will satisfy us. Accounts in which there isn't really much to be said about standing in special relations or having special access to one's mental states. I, I want to, I want to look at deflationary accounts. They're very clever. They maybe try to satisfy our criteria, but I think in the end they fail. But it's interesting to see why.

Okay, so, um, a number of accounts, a number of accounts have been given which try to reconcile this notion of the inner and the outer. They try to make room for the idea that there's, uh, their inner and outer aspects of the mind, that the mind is available first and third personally. But how do they do that? Well, they're two famous attempts that wrestled with this and it seems go absolutely nowhere, but, but interestingly, in a way that philosophers can learn from, and that's Cartesianism and behaviorism. Now, I'm sitting here in Paris, and I say Cartesian, I don't mean Descartes, because I'm not enough of a Descartes scholar. And if I say this to real Descartes scholars, they'll go crazy. So I say Cartesian, it's our sort of parody of what Descartes thought. Let me just, let me just use that as a term. I'll say what I mean by it.

Now, the Cartesian view of the mind, the Cartesian view of the mind removes the mental from other people's scrutinizing gaze. It's not on show. It's not available to others. For the Cartesian, all we do see, all we do observe, are people's bodies. Facts about their minds remain unobservable and hidden. And facts about their minds, of course, are not contained in the body. The mind is not located in the body like a pilot in a ship. But the mind of an individual is nevertheless a sphere about which the possessor of that mind cannot be in doubt, cannot make mistakes. So the Cartesian, on this view, sees the mental landscape as so well lit that there are no dark corners into which we can't peer. It's all immediately available and illuminated. That's the, the parody of the Cartesian view I want to, I want to talk about.

Now, it it exempts us from skeptical doubt. We can't be mistaken about our own states of mind because we can't overlook them. They are possessed insofar as they are subjectively experienced and present to their possessor. But the facts about those mental states are also constituted and exhausted by the subjective experience of them. There's no more than the subjective experience of them. And this is going to mean that we can have no doubt about our mental states, only at a cost of losing objectivity in our claims about the mental, only in reducing the content we can actually have about our thoughts and feelings.

Now, there's something interesting here which isn't often commented upon, but it's kind of worth commenting on. The reason that, um, we, the reason the fact, remember one of the asymmetries I mentioned, the fact that we cannot and do not need to give a justification for claims about our current state of mind, that is what renders maybe our judgments about our states of mind immune to skeptical doubt because it leaves no gap between the grounds on which we make knowledge claims and the truth of those claims. And yet, that's exactly where the skeptic works. As far as our knowledge of the external world goes, the skeptic tries to say, "Imagine that things are experienced in exactly the way you experience them now, that this is all the evidence you have through your senses, and yet there not be a world of people, tables, and chairs, a speaker in front of you talking." The idea is meant to be consistent with all the evidence you have. The claim on which you base that evidence might be false. That's, that's how the skeptic operates, to exploit a gap between the evidence or the grounds or the justification and the truth of the claim.

But look, we've just, we've just closed off a door to the skeptic. If there is no justification offered, there are no grounds, there's no room for the skeptic to exploit that strategy. They can't say, "Well, everything could be just the same, that's when you're thinking about your mind, only you're not." No possibility, because in thinking about your mind, that's that's it. That's what it is to have self-knowledge, to be thinking, okay, to be an experiencing being, a conscious being. So there isn't any gap. There is no, nothing we're basing those claims on. That's, that's supposed to be an advantage of this view.

Nevertheless, um, nevertheless, this would render, um, these these groundless claims would render, um, the content of those claims, it rendered the content of those claims unfit for a domain of truth and knowledge. We wouldn't have a kind of objectivity attaching to those opinions. Why not? Because now, if merely holding an opinion was enough to make it true, we've lost sight of one of the primary conditions for objective judgment. To have objective judgments, there has to be a distinction, a substantial distinction, between holding an opinion on the one hand, and what makes that opinion true on the other. Even if they coincide, even if they coincide, even if having an opinion and what makes it true coincide, there better be room for a substantial distinction between holding the opinion and what makes it the case that the opinion is right. That's required for there being any notion of objectivity. There ought to be a distinction between how things seem to me when I'm thinking and how they really are. And even if those two are given at the same time, we ought to make sense of there being two things here, not one. There isn't just how things seem. There's got to be how they are, as given to me by the way they seem. How things seem to me when I'm thinking about my own state of mind is how they are. But there better be how they are as well as just how it seems, otherwise there's no subject matter, otherwise there's no objectivity to our judgment.

But if we fail to make room for such a distinction between how things seem and how they are, then we suppose that what we're thinking of is constituted and exhausted by seemings. There's no more than seemings. And of course, that was the Cartesian idea. We can't be sure whether how things appear or how they are, but we can at least be sure how they appear. So we retreat to the inner and we scan a domain of appearances about which we're infallible. Now we're supposed to constitute claims, the truth and knowledge, out of the appearances. But of course, famously, we can't get a distinction between merely judging, a merely things appearing, and are are being even right about the appearances, because there's no, there's no distinction to be had there. So there's no claim to truth. There's no claim to objectivity. Famously. And if you really force me to do it, I'll try giving you the private language argument afterwards by insight that the idea is meant to be, look, the Cartesian, the Cartesian tries to avoid falling into error in judgment about the world by at least assuring himself or herself that they know how things are in the inner realm. But the retreat to the inner, which gives up the world and leaves you with an unbridgeable gap between the mind and the world, that actually is no good. Because if the Wittgenstein argument, the private language argument, any good, there isn't even any resting place. You can't, there's no stability to the inner. We can't make judgments about it because we can get no distinction between how we judge things to be and how they are. We can't, we we cannot entertain a proper distinction. So there is no, there's no safer retreat. Okay, that's the parody of the, that's the parody of the Cartesian position.

Okay, so we need something. We need something about which we're getting our judgments right or wrong. And what would that be? Well, of course, remember, if we forget the Cartesian view and we we return to the idea that other people can know quite a lot about our our thoughts and feelings, our wants and wishes, our hopes and fears, they could, they can do ordinary psychology. They've got theory of mind capabilities and abilities. They can attribute mental states to one another very effortlessly and and often accurately, and predict and explain behavior in the light of those attributions. If we remember that, if we remember there is an external perspective on the mind, then we can say, well, it's got to be, it's got to be about states like that, states that are confirmed externally, that we're getting matters right. So that would give us an objective subject matter.

But maybe then, and this is the alternative to Cartesianism, where we veer away to the other side, maybe then we we end up in the position of Gilbert Ryle, who says, actually, all our knowledge about minds is knowledge of that sort. Actually, all our knowledge is knowledge on the basis of observing behavior, even knowledge of your own mind. Now, how then does Ryle make sense of the asymmetry? The fact that we know our own minds best? Well, nothing could be easier. Ryle says, you know your own mind best because you're around yourself more. You tend to be with yourself more than you're with other people. It's not surprising you know your own mind best. And Ryle actually went on to say that the person who knows you second best is your wife. This is a typical old Oxford example of the of the 50s, right? So a man knows himself best, and the person who knows him second best is his wife. And this is, this is, I mean, this counted as serious, you know, philosophy of a philosophy of mind in those days.

Now, Ryle did much more besides, but, but on on self-knowledge, this won't do. It won't do because although it's a a very interesting attempt to show how we can preserve the symmetry and the authority, it loses sight of the groundlessness. It loses sight of the fact that I don't have to observe my behavior to know whether I want a cup of coffee. Now, when you take me to, or when you take me to the bar afterwards and you say, "Would you like a glass of wine?" I don't say, "Well, you know, let's see. Pour it. I'll see what happens. Oh, yeah, I seem to do want it." I say, as I'm drinking, "No, it's not like that." So Ryle can't overcome that problem. So we can't have a real solution. We still need a first and a third person way of accessing the same mental states.

So there's a hopelessness, there's a hopelessness in the in the Rylean position. We've got to rule that out. One thing that I haven't said, and that some of you might be tempted by, is the idea, "Look, you're confusing not having grounds with the idea that our knowledge of our own minds is mediated by some further evidence." Maybe we do have knowledge of our own minds, but it's, uh, it's rather direct. It's direct and unmediated. In that case, I do have grounds. When you say to me, "How do you know what you want?" I say, "Because I'm directly aware of it." And what is this awareness? Now, some people will posit an inner observation model. The idea that there's a kind of inner perception of our states. This is a terrible idea. It's a terrible idea. And in fact, one could, one could waste the rest of the seminar telling you why this is a bad idea, but I don't want to do that. We could take it up in discussion.

But notice that if there are grounds, but the grounds are sort of direct, they're unmediated, the possibility of grounds allows the skeptic back in because there could still be errors and illusions. There could be cases where we're misperceiving our states of mind, and yet we're, we're in in a perfectly good working state, and yet we're not getting facts right. Also, phenomenologically, it doesn't seem right. It just doesn't seem right to say that whenever I'm asked to comment on what I want or what I think, I have to go in for some internal mental scanning. As Gareth Evans pointed out once, sometimes the best way to know whether you believe that P is to look outside, to look at the facts, to, to, to look outwards, as it were. If you say, uh, "Do you think it's, do you think it's cold outside?" I don't sort of think, "Now, wait a minute. Let me see. Do I think that?" And I do a little bit of internal scanning. No, I can't. I go and look. I go and find out. I open the window. I, I peer outside. I see whether people are covering themselves, protecting themselves from the wind and the chill. I observe the world and come to know whether I believe that P. I look, I look outwards. It's, it's an eye directed outwards, as Mike Martin once put it.

Now, that idea is quite nice. Evans talks about the transparency, uh, of of our mental states for this reason. If, if I want to know, if you want to know whether you believe something, you want to know whether you believe a particular, uh, fact or or situation obtains, you concentrate on the fact or situation. You direct your attention outwards, and then you know. That's that's interesting. But there's no explanation as to why looking at the world also has as a byproduct my not, my knowledge of what I think when I look at the world. It needs to be explained how it, how it as it were, generates also a claim to self-knowledge and not just a claim about the world. That's, that's part of what's missing from the account. So it's a nice datum, but it's not missing. Also, the datum is good, just to just to remind us why the perceptual model won't do.

But I want to say something which would be more congenial to the perceptual model and not actually, and not actually, um, in the spirit of Evans and in the spirit of transparency. It doesn't look like the right account of desires. I mean, sometimes when you say to me, you know, "Do you want to drink the, you know, do you want to drink the Lalande Pomerol?" I could sort of, I, you know, I could concentrate on it and then I could try and see whether or not it's desirable. But, but pretty much, I am reflecting on me. Do I want to drink that? No, I don't think I do. No, I don't think I do. I say, "Shall we open your incredibly nice, uh, Corton, which you've been keeping for years, instead?" You know? So I could, I do actually have to, um, observe myself. I don't, I don't think that it's about direction out for desire. There are lots of cases which don't work this way. So, um, the looking outside isn't going to do it.

Also, there's another nice case which is brought up here by and made a lot of by Richard Moran, which is the idea that sometimes self-knowledge isn't like reporting, it's rather like making a decision. So when you say, "Do you want to drink red or white?" I don't think, "What do I want?" I, I make up my mind. I decide. I go for one or the other. That's, that's bringing into existence a desire of mine, and also a reflection on what that desire is. Making up one's mind is sometimes the right way of getting self-knowledge. So again, the perceptual model doesn't, the perceptual model doesn't accommodate that in any satisfactory way. These are all things that seem to be very familiar parts of, uh, of mental life, especially parts of French life. I mean, all the time, all that lunch, dinner, you know, so I'm put in that position and I have to make up my mind. It's not always the case I'm just consulting an internal state. I'm actually choosing. That seems, that seems not perceptual. Good. So that's all I want to say about the flaws of the perceptual model. I don't want to say more about it.

So let's return to the issue of how we're to explain the special features of our self-knowledge, and especially authority, which seems like the hallmark of self-knowledge. That seems like, you know, one of the things we're not prepared to give up. Now, the lack of us of a skeptical problem, the idea that we don't doubt that we have knowledge of our states of mind, and the ease of acquiring it, in no way gives us reason to think that it's easy to explain how we have this knowledge. Far from it. In a way, the explanation seems to go missing. The very ease of having self-knowledge and the authority we have for it seem to, uh, make it more difficult rather than easier, seem to exclude the idea that it's easy to get at why we know our own minds and why, why it has this feature. It's not easy to say. So how do I know what I'm thinking? I think contains, um, something important to it, because how do I know what I'm thinking? I, I'm going to, by the end, place a lot of emphasis on the "what I'm thinking." It's not just that I'm thinking, it's what I'm thinking. How do I have the right characterization of my, uh, states of mind? How do I, how do I get them right?

Now, some of you might be tempted at this stage to say, "Well, Luke, you haven't mentioned consciousness. You haven't mentioned consciousness." Uh, I probably haven't, right? It's a lot of work going on in consciousness. A lot of interesting things to say about consciousness. Isn't that the place to look? Shouldn't we try and get an account in terms of consciousness? No. No. There are many creatures who are conscious, about whom it seems reasonable to suppose that they are creatures with minds, that they have conscious minds, but they don't necessarily know they have conscious minds. They don't necessarily have the capacity to know what they're thinking. Being in a state of mind is one thing. Knowing what state of mind you're in is another. These are two different things. We mustn't confuse just being in a state of mind with knowing what state of mind one's in. And also, creatures who have consciousness are not necessarily creatures capable of self-knowledge or reflection in general. I think there's a division. There's a very large division in the natural world between creatures who have minds and creatures who have minds and know they have minds. This seems a very important division. And it doesn't seem to me that creatures who have minds, as opposed to creatures who have minds and know their minds, doesn't seem that creatures who have minds have to be denied consciousness. It's not that consciousness provides the extra ingredient for self-knowledge.

The other thing is that consciousness, think of the private language argument again, for those of you who have it, read a later mind. The private language argument for Wittgenstein was not a question about consciousness. He didn't doubt the private diarist had consciousness. The question is whether the private linguist is able to decide whether two conscious experiences count as the same ones experienced again or different. It's a question of whether or not I'm able, entitled, to classify my conscious experiences such that I can say, "Ah, that's S. That's the experience I had when I wrote down S in a diary." Now, conscious states of mind come in all stripes and flavors. They're modulations of our sort of general awareness. But the point is that knowing what conscious state of mind you're in takes an extra turn of the mind. It requires an extra step. Being in a state of mind is one thing. Classifying your states of mind correctly is another. Was that sore or just uncomfortable? Was that a pain or an itch? Was that, do I feel mildly exhilarated or anxious? It's not always easy to tell. It's a superb feat of phenomenological taxonomy that we're able to classify our mental states as, uh, being states of of exhilaration, anxiety, joy, pain, irritation, and so on. It's an extra step. And given that we have to classify our mental states, we have to categorize them. That's where we could go wrong, as well as going right.

So I agree with anyone who says, "But you can't overlook your conscious states. You can't fail to notice consciousness. You can't, you know, consciousness rather grabs your attention. It's rather kind of present in there." Absolutely. It's sort of self-illuminating about its presence, but not about what is present to the mind. Being conscious, one thing. Knowing what one is conscious of, another. Self-knowledge, I think, is demanding to that extent.

And now, when we say self, when I say self-knowledge is demanding, I want to resist another line that people often take when they say self-knowledge is not a cognitive achievement. I mean, it's so easily won. I can know without having any grounds whether what I want and what I'm thinking now. So how could something I can, I can know without it actually demanding any thought or reasoning or justification for me, how can it be much of a cognitive achievement? And they try to tell you that there's no cognitive achievement. But there is. There is to the extent that classifying one's states of mind correctly actually requires some cognitive competence. So we mustn't lose sight of that.

Okay. Now I want to just rub in a point which, uh, again, often gets missed. And the point is this: It might appear as though the public side to the mind, the publicly recognizable side of of people's minds, the fact that we can attribute beliefs and desires, hopes and fears, wants and wishes, intentions to them so as to make sense of them in rational terms as as agents deliberating about their world, planning, thinking, strategizing, reasoning, and so on. The fact that the mind is is known by others in this way, and also notice, a lot of our concepts of mind, a lot of our concepts of what it is to have a mind, to be a believer, desire, thinker, reasoner, they don't mention consciousness. I don't even know if you're conscious. I don't even care about the question of whether or not you're a zombie, as long as you're behaving in the right way. To me, that's good enough for me. I, I do not, I do not have access to never will have access to your conscious experiences. So when you say, "Do you feel it too?" and I say, "Yes," I'm lying. It's not, I can't, I don't, right? I can judge how you are responding. I can judge your emotions, but I can't have your consciousness. So a lot of consciousness does not feature in the the notion of our common sense psychology, our classification of states by which we describe and explain one another's behavior. Consciousness in common sense psychology is usually whether you're conscious or unconscious. Oh dear, Barry's had rather a lot of wine, he's unconscious. Right, that's different. That's kind of folk psychology. We're interested in that distinction, but we're not really able to get at much more than that, okay, about other people.

But is it the case that the public side to mental life threatens the possibility of self-knowledge? No. Okay, first major moral that we've got to draw here is that the fact that the mind has a public side is what secures the possibility of self-knowledge being objective. It was, it's what gives us something to have knowledge about. This is a way of independently verifying that there are mental states that people are in, such that if they know about their mental states, they'll by and large know about a contingent fact about themselves. They'll know a contingent fact that's there in the world for the taking, for the discovering. Okay, that's good. That's good. That secures, unlike the Cartesian picture, it secures a kind of objectivity. So maybe it's the case that self-knowledge actually requires other people have knowledge of our minds. So maybe knowledge of knowledge of other minds and self-knowledge are actually connected in a rather deep and important way. You couldn't have self-knowledge unless.

There was, uh, something to be known about. And maybe the possibility of knowing about it is partly constituted by a practice of us describing and explaining people's mental life in psychological terms. Okay, so self-knowledge may require it. Publicity ensures there's a genuine subject matter to which self-knowledge lays claim. Okay, it credits, uh, my self-knowledge, my opinions about my states of mind with, with something to be right or wrong about. That's over and above my merely having that opinion. It guarantees that there's something to get right.

What's puzzling is just how having immediate opinions, not even basing them on anything, just having immediate opinions, what's puzzling is how should, how should that succeed in giving me objective knowledge of contingent facts? How could I come to know about a bit of the world, the bit of the world where I am, without actually having to go to any effort to do that? How could I know such an objective fact about me? That's, that's part of the puzzle. And why should I be objective and authoritative about it?

Now, um, our mental states, um, okay, our mental states in this sense, um, are something that, um, we know about from the first-person perspective in a way that's different from, but maybe requires there be such states and that they be knowable, not known, but knowable from the third-person point of view. There should be no reason why other people, operating using a, a proper common sense psychological scheme which genuinely does explain people's behavior by describing beliefs and desires, them to, to make sense of what they do and what they want. There should be such a scheme, even if it's not operated by anyone, to provide a subject matter about which I'm right or wrong.

Okay, now this would suggest now that if we're to, um, if we're to solve the problem of self-knowledge, and remember the problem of self-knowledge is how can states of mind be part of my inner world if they're also accessible to others on the basis of public behavior, that's, that's how I was framing the problem. If we're going to do that, we're going to suggest that you have to solve the problem of how there can be first and third personal knowledge of the same mind. So knowledge of the self and knowledge of other minds have to be, uh, two ways of knowing about the same thinker's states. And we have to solve the problem of how there can be such knowledge at once, simultaneously. That's the idea. We're concentrating on the self-knowledge part of it.

So look at seven on page five here. Two distinctions I want to connect up. I want to connect up these distinctions in the following way. There's a distinction between being in mental states and knowing what mental states one is in. I've just said that we've got that distinction. But remember for objectivity, there was this other distinction between, uh, how things seem to I think are psychologically and how they are. Okay. And now I, we need to try and connect these two distinctions together. We need to connect these two distinctions to each other. And we're going to connect them like this. Obviously, being in a psychological state will correspond to how things are, right? But now we need to understand why and under what conditions knowing what psychological state we're in will correspond to, uh, sorry, that should be how things seem. I need to change that.

Okay, now we need to say why does knowing what psychological state we're in just correspond to how things seem? Why is it enough that when things seem to us a certain way, we're usually right about them? Why should that be for the very same states that can be known in these two different ways? Well, we have to say a little bit more about what it is to know other people's minds from the third-person point of view. And, uh, here I'm just going to make a blatant appeal to, um, Donald Davidson's views on what it is to see other people as minded creatures, to ascribe them beliefs and desires and intentions, to explain their actions, to give reasons for their actions, and to explain them in rational terms.

Now, it's part of our check on what it is to attribute to someone, uh, states of mind, that we're able to give a justification for attributing any particular belief and desire on the basis of a vast set of previously attributed beliefs and desires that we ascribe to someone to make sense of their behavior in rational terms. So when I see you do something, perhaps I've known you for some time, and I see you do something or say something, I ascribe motivations and mental states to you as your reasons for doing or saying what you do and say. But those reasons I ascribe to you, they earn their keep, they earn their home within a much larger portrait of other beliefs and desires you have that, uh, provide a conceptual surround and make sense of you having these reasons and these other beliefs and desires. They must play their part in explaining what you do and say elsewhere and at other times. They have to be part of an overall portrait of, of a person's attitudes and actions that make sense of their total life and conduct in intentional terms. There's got to be an ambition to try to fill out the map, the network of mental states that we can, uh, attribute as a whole, to make best sense possible, to make most intelligible and rational the actions, sayings, and doings of a creature, given the totality of their behavioral conduct. That's the, that's the Davidsonian ambition.

Now, objectivity is guaranteed, uh, as I said, when there are correct such explanations of people's mental states. But what is it to be, what is it to be a belief or a desire or a hope or a fear for Davidson? Well, he says this. Here's the quote on page six. States like belief, desire, hope, and fear are, quote, "just those states whose contents can be discovered in well-known ways. If other people are creatures are in states not discoverable by these methods, it can be not because the methods fail us, but because these states are not correctly called states of mind. They are not beliefs, desires, wishes, or intentions."

This shows you the normative and constitutive nature of the principles that govern our attribution of mental states to people. The principles that govern our interpretation of others are not principles which we think are very good at getting right the facts about their mental states. Rather, what we count as the facts of their mental states are what conform to the principles. So these are relatively a priori principles, constitutive and normative. Constitutive means synthetic a priori, if you're an Oxford, you keep using the word constitutive in a kind of mysterious way. Nobody says what it means, but it means synthetically a priori. It's okay. I tried this out. Davidson, he says it means synthetic, right? That's fine. These, so these are principles which constitute and exhaust what it is to be a mind, the principles of interpretation. So roughly, when we say, what is it for you to have a particular belief? The answer is, it's for you to be apt to be ascribed that belief in the course of giving an interpretation of you that makes best sense possible of your total life and conduct in terms of a set of attitudes and actions. That's the official picture. That's all there is, ascribing.

Okay, now we've got to wonder, well, how does this leave room for self-knowledge? How could it make room for self-knowledge? Of what it is for me to have a belief is to be apt to be interpreted as having that belief by an interpreter who makes best sense possible of my total life and conduct. Where's the, where's the room for self-knowledge? Well, what we need to discover is whether a subject's way of ascribing such states to himself counts among what Davidson's calling here, the well-known ways. The well-known ways of ascribing mental states to people. Because after all, there are well-known ways of ascribing mental states to one another. But that might include well-known ways of ascribing mental states to ourselves, namely, just saying what we think, just saying what we know. Okay, if that's, if that's among the well-known ways, all well and good.

But now the question is, what does it take for it to, uh, be conformable to or fit into the scheme by which we interpret somebody's mental life? Well, the answer's got to be this. And this is where the objectivity comes in that I was hinting at. The objectivity is this: what a subject says about themself is correct if and only if what they say conforms to the interpretations that a fully informed interpreter could make of them from the third-person point of view, subscribing to the principles of rationality and charity, the normative principles that govern interpretation. So we could think of a, a theory of interpretation, very loose idea of what we mean by theory, because Davidson's, Davidson doesn't believe in prediction, doesn't believe in scientific laws, but there's a kind of common sense theoretical organization of the way we interpret other people to make sense of them. And the thought would be, what is it for what I say about myself to be correct? It's for it to conform to, to fit in with what you say about me when you're an interpreter who has access to the full range of behavior and uses the principles of interpretation.

But now we get a mystery. And the mystery is, why should what I say about myself, not based on interpretation and not theoretically generated, why should it be answerable to and conform to what the best theory says or interpretation says in making pronouncements about me? Why should judgments that are not theoretically generated seem to conform to what the pronouncements of a theory? That's, that's puzzling. That's another way in which the puzzle reappears here.

Okay, now this idea of answerability, answerability to the, the interpreter, that provides a kind of objectivity. But it doesn't make much room for authority. We don't see why, we don't see why subjects are authoritative about their mental states. Because after all, if what I say about myself is only to be rendered intelligible by what other people say about me from the third-person point of view, then I don't see how I, I come to be authoritative. And now I think what we see is there's a kind of tension. There's a kind of tension between authority and objectivity. What objectivity requires is there better be something else there, better be something else other than what the subject says about himself or herself, by reference to which the judgments are true. There has to be that. But at the same time, that, that guarantees objectivity. The same time that puts authority in question. And what authority takes seems to be exactly the opposite. We're authoritative when how things seem to us is how they are. We're authoritative when the gap between how things seem and how they are is narrowed, perhaps closed. I'm authoritative when my pronouncements count as correct. But for them to count as correct means I'm actually in the best position to get things right. And that means when I think I'm in a certain mental state, I am in a mental state. So authority seems to require there be no clear gap between how things seem and how they are. But objectivity seems to require there be a discernible gap. There's a tension between objectivity and authority. And that's actually again, getting at the heart of the problem of self-knowledge. That's, that's what I think is the, the issue that self-knowledge that we care about here.

So finding a solution to this, finding a solution to this, um, requires us to fight, to answer two key questions. And the two quick questions at the bottom of page six, not long ago, we're nearly there. Um, what is it for things? The first question we have to answer these two questions to have a satisfactory account that meets our requirements. What is it for things to seem a certain way to a person psychologically? What is it for them to take themselves to be in a mental state? And secondly, why is it that how things seem psychologically usually provides knowledge of how they are? We need, we need to say both. We say both, we're home and dry. See, very easy. We answer these two questions, we're done.

Okay, so now, what, what I, I noticed that they addressed different parts of the problem. The first question makes room for the subjective viewpoint of a thinker, there being a way it seems to the thinker, that gets the subjectivity in. The second point gets the objectivity and why does seeming usually coincide with how things are? If we can get both of those together, we're there.

Okay, can we do it? You think I'm going to provide a solution, but I'm not. But I'm going to age towards one, right? I mean, I don't want to, I don't want to, yeah, I don't want to mislead you, but I'm going to suggest the materials we need to do this. Okay, so let's just quickly end by, um, I'm going to make at the very end a very tentative proposal, but, but I want to just quickly suggest a couple of, um, runs through what people have said and why they, they don't satisfy the problem. Is it just consciousness? Is it just awareness capturing, uh, our attention or having awareness? No, that's not yet a claim to knowledge. As I said, because going back to our previous discussion, taking things to be a certain way is not just about being conscious. It's making a judgment about which state of mind you're in. And so it doesn't even meet the first, doesn't answer the first question. Do we have some special epistemic relation to our inner lives, like perception? No, and I said I wouldn't say any more about that, so I won't.

So now the interesting options that are left to us are deflationary treatments of our self-knowledge. Now, deflationary treatment is interesting because, um, here we account for the first-person point of view as best we can, merely by honoring one of the marks of, uh, self-knowledge, namely authority. If we can try and explain why there's authority, then that's supposed to help us say we have an account of self-knowledge. If we can have the objectivity and we can have authority, we're supposed to be there. Most of the deflationary accounts can't meet the first, can't answer my first question on page six. They can't answer that. They can't get the subjective part in there. They can give you a mark of authority, they can show you why judgments count as authoritative, but not why they came to subjects taking themselves to be in certain mental states.

Okay, we'll see this in just two minutes. Wittgenstein just gives up on this claim. He says, when, when I say, um, "I have a headache" or, um, "I want a cup of coffee," and so on, that's, that's something for other people. I'm not making reports about myself. I'm making something that you can make use of. You can now say, he wants a cup of coffee. But now we don't have the truth-conditional asymmetry. We actually just cut the link. The biconditional line sister done earlier is not there because we don't have a truth value for statements about the first person. We have a third personal claim, and that's it. So we don't have, um, we can't get the, the symmetry there.

Richard Rorty says something, uh, mischievous, as usual for Rorty, but it's kind of, um, playful and mysterious. And Rorty says, he says, look, the authority that peop, that standardly attaches to people's words, it's just a courtesy we grant people authority for what they say about themselves. It's not as though they are authoritative. It's just that we, we grant them that. It's a kind of politesse. So, you know, you say, um, "Oh, I'm hungry" or "I want to go to the movies," and I just say, "Yes, yes, of course, of course." I defer because, you know, this is, this is what we do. Now, of course, why that, why that's no good is because, why that's not good is again because, um, it wouldn't explain the mysterious coincidence with the fact that what other people do when they interpret us tends to land on the same results that about what people say about themselves. If, if people speaking about themselves were just granted as a courtesy that what they say is true, it wouldn't explain the coincidence in truth value between what we say when we know a lot about people and what they say about themselves. We usually take it that, you know, we've known someone long enough and we make a kind of estimate of what, what's going on in their mind, and they confirm it, and they say yes, and we take it as a confirmation. We don't take it that, that, uh, it's just, it's just them saying something and we should just agree with it. It wouldn't explain why we're so good third person getting at the same thing they say first person. Like, it wouldn't explain the coincidence. It also leaves no room for overturning people's mental, mental pronouncements. And yet there are cases, as I said early on, where you can say, "No, no, no, you were clearly irritated with the chairman." We actually have to overrule what people say. And Rorty leaves no room for that. That's just, that's just being impolite when you do that.

Okay, but that doesn't really make sense of the case. It's not good enough. Richard Moran, "Making Up One's Mind," more like a decision than a judgment. That, that's good for some cases, but it doesn't cover them all. It doesn't cover them all. Sometimes I believe something and I don't have to, uh, come up with a decision. I actually, I actually am trying to be faithful to something which I think has been an enduring part of my mind and which is, which is there. Now, for me, uh, in speaking, also decisions, even if it's different from judgments, decisions nevertheless have to be true. And again, it's interesting that other people could perhaps predict or explain what I think and what I'm calling a decision would actually coincide with that. So again, there's an interesting issue about the symmetry.

But the real, the real issue I want to end on is just looking at Crispin Wright's view. Crispin Wright says, and we can cut to the chase by looking at the, um, he thinks rule-following and the nature of our knowledge of our intentions goes the same way. That's why rule-following's here. He says, "There is no essential epistemology, inner epistemology of rule-following. In fact, there's no inner epistemology of knowing our own minds." For Wright's striking view, and now he says this, but he's, he's, he's addressing the issues we're interested in because he says this: "How can judgments of our own intentions lack a substantial epistemology in this way, and yet still be objective? Still have to answer to something distinct from our actual dispositions to judgment?" So somebody says, "I want to go to the movies." Okay, but there's no essential inner epistemology. They're not, as it were, attending to something in themselves when they say that. So how can what they say, when it lacks a substantial epistemology, how can it be nevertheless objective? What they say is correct, and it's answerable to something more than them just saying it. It's not correct because they said it. It's got to be correct for other reasons.

Well, the answer for Wright comes from this, comes from identifying the conditions of correctness for ascribing intentional states with their interpretive utility. What this means is roughly something like this: When people make statements about themselves, about their intentions, those statements about themselves are usually right because we find that those statements they make about themselves give best service in interpreting what they'll do and what they think than accounts of their psychology given externally that leave them out or overturn them. By and large, if you credit what people are saying about their own minds and give it special place in making sense of them, accommodating what they say about themselves, respecting what they say about themselves, gives better explanations of what they do, of what they're up to, of what they're capable of thinking and feeling and doing than explanations that reject them or overturn them.

Why should that be? Wright says two things, and I think this is where he's really in trouble. He says two things on the same page. He says, on, on at one point, it's an a priori constraint of interpretation. It's a, it's an a priori requirement in interpreting other people that we respect what they say about themselves and we treat what they say about themselves as largely true unless a better interpretation of them can be had by overruling and overturning what they say about themselves. So the claim is, when you're interpreting someone, the data you get is not just what they do and what they say, but what they say about themselves. And by and large, you should try to treat as special, you should try to credit what they say about themselves as default true. It should be true unless you have to overturn it by providing explanations that ascribe other mental states to them that are better, that are more illuminating, more insightful, more accurate at interpreting what they say.

Okay, now if that's an a priori constraint on interpretation, namely that so interpret someone that you give special place and treat as true, as far as you can, subject to the other principles, what they say about their own minds, then how can it be also the case, as Wright says, that it's a deep contingency? It's a deep and contingently true fact that when we do credit other people with being right about what they say, it does explain them better. This is, this is metaphysically puzzling. It's metaphysically puzzling that it should be an a priori constraint that we have to interpret people so as to make them largely right about what they say, and we should find out it's a contingency that when we do credit other people's self-descriptions with this, uh, importance and and see it as true, it gives better explanation of them than ones that leave it out. How can it both be an a priori principle and a deep contingency that we're getting things right? There's a metaphysical confusion there that's unresolved. And it looks as though we better hang on to the idea it's genuinely explanatory. It's not a priori. It's actually right that we get better explanations when we hold on to what makes, uh, what makes it true.

Here's another way in which, um, Wright tries to, he has another go at it, uh, in the quote you now have in the middle. He says, "Remember, there's no inner epistemology. We do not cognitively interact with states of affairs which confer truth on our opinions concerning our unintentional states. Rather, we are, as it were, inundated day by day with opinions." And when you ask how we come to be inundated with opinions which which are largely true, he says, "We are ceaselessly but sub-cognitively moved to produce opinions for which truth is the default value."

So here's the thing. He's now going to say, I don't interact with some state of affairs that can serve, confers truth on my opinion. Rather, I'm inundated day by day with opinions that I'm just urged to produce sub-cognitively concerning my own intentional states, for which truth is the default position. The account is true, provided we hold them, and no good purpose is served in another's quest to find us intelligible by rejecting them. Here's the funny portrait. Wright's view is this: I sit here now, and I just keep being restlessly but ceaselessly moved to utter opinions which you take to be true. "I'm hungry," "I want a drink," "I want coffee," "I believe the seminar should finish soon." These are just things that I'm restlessly and ceaselessly moved to produce. And you, as an interpreter, take them as being true. Truth is the default position, he says. And they count as true, provided that we hold them, and no good purpose is served in making us intelligible by rejecting them.

What that does, notice, is something very, very interesting and intriguing. It says that what a subject literally says is counts as true according to the interpreter, as long as the interpreter can't do better by ignoring or overturning it. That then makes it true that the subject is largely right about their mental states in this sense, that the pronouncements they make are largely true. He shows you why what subjects say about themselves is largely true, therefore they're authoritative. But it doesn't count as an explanation of their knowledge. There is no explaining from the first person point of view what it is to have knowledge here. Technically speaking, we're meeting an obligation to show why what subjects say about themselves is true, why they are largely correct by the interpreters' lights from the third person. But it doesn't tell you anything about what it is from the first person to know whether or not they're right. That's okay for him because there's no inner epistemology. But no inner epistemology suggests no epistemology at all here. There's just no account of what it is to know our own states of mind.

Now there's another confusing thing that Wright says, and then I'll say my one sentence about what we should say instead. The other confusing thing that Wright says is this: He says, "My judging, my saying, my pronouncing, my vowing, my saying out loud what I want or what I intend is deep is a default sufficient condition for my being in the state I ascribe to myself." So when I say, "I intend to go to the movies," that's a default sufficient condition for it being true that I intend to do that. It could be overruled, but it's default true. So in other words, merely saying, under the right conditions, what I believe or intend is default sufficient for making it the case that I do so intend, that I do so believe. Why is it only a sufficient condition? It's only a sufficient condition because, Wright says, other creatures and other animals could be in those states without making judgments about their own minds. And now we have something again, metaphysically very puzzling. The mental state could be there, requiring no help from our judging or avowing that we're in that state, or the mental state could be there because it's partly constituted by and determined by our saying we're in that state. How can one and the same target state be both there without requiring any constitutive help from us as judging subjects, and also be brought into existence and be there by judging? That won't do. That's metaphysical madness. There's another problem here for Wright.

So we've got a priori contingency issue, and we've also got states being constituted by judging that we're in them, and they're potentially being there requiring no help and no constitutive role of a judging subject. That seems wrong. So here's, here's the hint that higher solution should go. We should make it a, we should make it a necessary condition for being in a certain mental state that we are in a position to judge and classify we're in that state, using our concepts, probably using our language, probably using the language we have to articulate our mental states. But that doesn't mean we couldn't be in mental states independently of using the language to classify them. It's that the mental states we're in themselves would be changed and transformed by our so classifying them. If you're in a mental state, phenomenologically rich and conscious, you could be in a mental state without making a judgment about it. But to know you're in that mental state, to have the means of classifying and describing it, uses concepts which shape that conscious state and articulate it, give it the structure and the content that the language we use to classify it infuses it with. So it's not strictly true that we can be in the same state before and after we make a judgment about it. But it doesn't mean there's no state there before we make a judgment about it. There can be a conscious state that isn't articulated and isn't therefore transformed by our means of classifying it. This should remind you, some of you, of the debate about conceptual content. The idea that there being a state there is different when we think conceptually about it. It actually changes the state. It cuts up and classifies it in a certain way. There was a state there, and it did have a content, but it didn't have the same content after we've structured it and organized it conceptually. Something about using our conceptual apparatus to apply to our states is what it is that makes what we say about those states about a different state of mind that could not be there without the judgment.

But last word, why is it that what we say about our state of mind is answerable to what other people say? Because the concepts and categories we use to classify our mental states as beliefs, desires, intentions, and so on, are the ones we learn from others in the, in the course of learning of folk psychology and learning a psychological common sense scheme by which we interpret ourselves and others, and which was first used by others to classify and categorize our mental states. So of course they coincide. I'll stop there. Thanks, Francis.