I will not wax millenial. But we are nearing the end of a century. It is conceivable that its most important philosophical event will take place between now and December 31. But it is unlikely. So we are at a point where we can, not unreasonably, look back and assess the philosophical events of our century. Any philosophically respectable century has its moments; a point, or points, at which ways of thinking about problems, and about philosophy itself, change radically. Such a moment might fairly be called revolutionary, if one does not puff up that term unduly. Two revolutions in a century would be a lot. I will describe what I think is the crucial revolutionary moment in our century.

I need to acknowledge that the view I will present is personal, or at least far from received wisdom. Some of you may find it biased. First, I do not think that the revolution means that philosophy is, or should be, at an end, that we are somehow in a period of ‘post-philosophy’, or that the idea of a philosophical problem is somehow passé. The revolution I am going to describe left us with some genuine, and severe, philosophical problems. At the end of my talk I will describe them, and hint, but no more, at a way with them–a way I do not know to be adequate, since it has not yet been tried fully. Second, it will occur to some of you that my revolution is suspiciously Anglophone, so, perhaps, parochial. If ‘Anglophone’ includes enough of Vienna and Berlin, and enough work written in German, then my revolution is, indeed, Anglophone. Given the political events of our century, it is not surprising that that should be so. In any event, such is life. If I thought the action were elsewhere, I would be there.

The revolution I have in mind can be dated as occurring between 1930 and 1960 (interrupted by the war). That is a period that opened with the start of Wittgenstein’s lecturing at Cambridge (roughly coincident with his new approach to philosophy), and ending with J. L. Austin’s death–a period I am construing as broad enough to capture two other key revolutionary figures, Hilary Putnam and Noam Chomsky. (Stuart Hampshire, in his memoriam, said of Austin, "He could not have adopted a special tone of voice, or attitude of mind, for philosophical questions." That attitude encapsulates the revolution.)

A more serious worry for me than those mentioned above is that some of my colleagues may think that the real important revolution in the twentieth century happened somewhat earlier than that. And they would be apt to mention Russell, and Wittgenstein before his change of approach, as at its centre. Now, I agree that those figures were at the centre of an important revolution. I will say a few more words about it later. But, as I see things, that was a nineteenth century revolution–indeed, that century’s most important one–begun by Frege in 1879. My twentieth century revolution is, in large part, though not just, a reaction against just the features of this earlier one that most exercised Russell and early Wittgenstein. Some, of course, will find that reaction unfounded. That is a controversy into which I plan to enter in what follows.

1. Empiricism: The twentieth century revolution has two main parts. It is a rejection of empiricism; and it is a rejection of a certain form of platonism (as we shall see, engendered by the Fregean revolution mentioned above). I will be more concerned in this lecture with the anti-platonism, and its consequences. But, since it is crucial for philosophical method, I want at least to mention the anti-empiricism.

The core of empiricism, in the present sense, is a two pronged thesis: first, there is a privileged class of facts; second, there is a determinate, specifiable topic neutral set of knowledge-yielding means: principles, procedures or methods. A given empiricism claims to be able to identify those facts, and those means, in advance of inquiry in any given field (that is, topic-neutrally). The key idea is that any fact is answerable to the privileged ones via the specified means. And an empiricist will usually tell us that some significant domain of what we thought fact consists either of no real facts at all, or of not the facts we thought there were. (His principles, he takes it, bite.) (Empiricism is often started by a healthy suspicion of loose talk.)

Answerability is, au fond, the idea that for any genuine way things might be (and we might think of things as being) it must be possible for someone who does not know what that way is to come to know it–what it is for things to be that way–by applying specified means to given areas of privileged fact. So either what it is for something to be that way is for the privileged facts to be arranged in such and such way, or at least the only real grounds there could be for taking things to be the way in question is that the privileged facts are arranged in such and such ways. So, for example, there is such a thing as a person’s being happy only if someone who did not know what it was for someone to be happy could (in principle) come to know this by extrapolating from the privileged facts by the specified means.

To be slightly less abstract, privileged facts are usually meant to be those we actually confront in experience–those it is open to us just to observe to be so. And, to repeat, the empiricist thinks these are a definite class of fact, and he can say (in advance of confronting experience) which ones they are. He might tell us, for example, ‘all we can really observe about others is their behavior’, or ‘all we can really observe is the way things appear to us.’ What moves empiricist thought here is usually some version of what is now known as a ‘highest common factor argument’. The argument turns on this thought. Suppose things are a given way, and, moreover, look that way to you. As far as you can tell they are that way. Now suppose there is a possible situation such that, were you in it, everything would seem just the same to you–you cold not notice any deviation from the situation you in fact are in. Then all you can really observe, in either situation, is what is in common to both. So, for example, Pia is exuding happiness. You can see her beaming and fluttering with it. Or so you think. But, in principle, there could be a highly trained actress whose performance was indistinguishable from Pia’s joy, though produced utterly cynically and with a bitter heart. Her performance could look as much like Pia’s happiness as you like. So, the argument concludes, all you really observe about Pia are (at most) flutterings and facial expressions of various sorts. The rest is inference.

If you behaved badly last night, that consisted in your constantly contradicting everyone, or insulting, or trying to humiliate, Pia, or making overly suggestive remarks, or whatever it is you did. That is our ordinary concept of behavior. If we stick to it, then the idea that all you observe about others is their behavior will be harmless. It will have no power to threaten our intuitive conviction that we can see that Pia is happy. To threaten it we need a proprietary notion of behavior. That is what the highest common factor argument promises to supply.

Empiricism purports to place an a priori and substantive requirement on genuine facts, or fact stating: its requirement of answerability. In Philosophical Investigations §135, Wittgenstein rejects the idea that there can be any such requirement. That is a cornerstone of his later philosophy, so, too, of the twentieth century revolution. It is a foundation of Austin’s attacks on particular empiricisms. As that rejection works out in particular cases, the revolutionary will reject either the empiricist’s conception of the privileged facts (usually the very idea that there is any such class of facts) or empiricist limitations on our knowledge-yielding means and capacities, and its idea of what they must work on, or both. In the first case rejection of the highest common factor argument plays a crucial role. It is the second rejection that most clearly carries a seeming threat of idealism. But idealism is a topic yet to come.

----

2. Platonism: How do we manage to think or to speak, about the world at all–either truly or falsely? That is, how does our thinking manage to be about the world? How, that is, can we think–or say–what is so just in case things are such and such way–think, say, that the lawn is green, thus thinking what is so precisely when the lawn is green? A very ancient form of answer has it that this is accomplished through mediation by something external both to us, and to whatever it is that we think about. Somehow we get in touch with the right somethings. The story at this end will not concern us much here. But, as it may be, we discern them, or mean them, or intend for them to mediate. Then those somethings do their job quite independent of us: they sort possible arrangements of the world into those that are just what we thought, or said, and others. That general form of answer–that sort of reference to something external to us for setting the standards for things being as we thought, or said–is what I am here calling platonism.

To illustrate with an old term, we might call the external items universals. I call the lawn green. My words connect to a certain universal-the universal greenness. The lawn, for its part, relates in its own way to that universal. Just to name the relation in question, it might partake of, or participate in it. Or it might fail to. If it does partake, it is its way of being colored–its look–that makes it do so. As for the universal, it decides, just by being the universal it is, what looks will constitute partaking in it, and what ones will not, so, thereby, whether the lawn participates. Since it is connected to my words as it is, it decides how things must be for what I said to be true, or equally, when what I said would be true. And the way it does its work has nothing to do with us. (Someone might think of that as a way of breaking out of what might otherwise look like a vicious regress.)

The term ‘universal’ nowadays has a sort of mystical ring. Philosophers are not particularly fond of using it. But vocabulary is not important. There are all sorts of ways of thinking of external mediators between us and the world which–thanks only to the mediation–makes our particular thoughts about it true or false. To take a thoroughly modern version, for example, one might just think that an English predicate–something like ‘is green’–has a certain property which we can specify this way: it is true of something just in case that thing is green. It is no proper business of the semanticist how the predicate came by that property. The important thing is that it has it. And to think of the property in that way just is to think of it as one that sorts out possible arrangements of the world into those that would make a use of ‘is green’ of something true, and others–just the job universals were meant to do. As with universals, how this property does its work–what sorting out it does–has nothing to do with us.

Who cares whether we think of our thinking platonistically? This may seem one of those dry areas of philosophy of little concern to anyone but philosophers, who, by definition, have nothing better to do. But philosophers only worry about problems people would naturally worry about without them. It is just, or so we think, that philosophy is the best preparation for doing that. I have been giving short shrift, so far, to the issue of how we get connected in the right way to specific mediators. But when that issue comes out into the open, as it must eventually, it may come to seem that that is none too easy a thing for us to do. In fact, empiricism and platonism coincide in easily making objective discourse about the world–discourse objectively correct or incorrect simply in virtue of the world’s being as it is–seem an impossible achievement for us. As we know, many today are prepared to succumb to the idea that in fact it is an impossible achievement. (As for empiricism’s influence, one might think of Richard Rorty’s reliance on Quine in reaching his conclusion that such discourse is impossible.) But if the revolution I am describing is correct, what they are actually succumbing to is a mythological view of what objective discourse would have to be like. Given the state of much of the humanities today, I think it is at least fair to say that this is a philosophical issue that matters.

Frege began his revolution with the injunction "always to separate the psychological from the logical, the subjective from the objective." As a research strategy, the idea has undeniably borne quite a lot of fruit: mathematical logic. But it has also been an understandable impetus to a form of platonism. For, in Frege’s hands, it encourages the idea that thoughts may be studied, so far as they concern logic, in splendid isolation from thinkers; that such things as relations of entailment between thoughts, or facts, if you prefer, are determined by laws that do whatever they do entirely independent of the users of thoughts; that the domain of thoughts–that is, things there are to think–has an intrinsic physiognomy, its having which has nothing to do with us. Here what is external to us is an intrinsic structure linking items in a domain–thoughts–and, on the usual line, structures intrinsic to the thoughts themselves. What the external determines in the first instance is what follows from what. But, in doing that, it also determines how what we think may be, or fail to be, what is so.

Fregean platonism depends on a particular picture of how logic, or in his terms the laws of truth, applies to the particular things we think (and on taking very seriously the count-noun status of ‘thought’ where a thought is something there is to think). The antiplatonism in the twentieth century revolution is largely a reaction against this tendency in Fregean thought (where ‘Fregean’ covers much more here than the thought of Frege himself). Indeed, I think we can see the whole change in ways of thinking from the platonism that went along with the early development of logic to the antiplatonism of the later Wittgenstein in these two remarks by Wittgenstein, one as a central figure in the nineteenth century revolution (then spilled over into the twentieth), and the second a later remark on the first:

Logic fills the world: the limits of the world are its limits.
Thus we cannot say in logic: Such and such there is in the
world; that there is not.For that would apparently presup-
pose that we exclude certain possibilities, and this cannot
be the case, since otherwise logic would have to get out-
side the limits of the world; as if it could also consider these
limits form the other side.

The ideal, as we think of it, is unshakably fixed. You can never
get outside it. You must always turn back. There is no outside
at all; outside there is no air to breathe.–Where does this idea
come from? It is just like a pair of glasses on our nose, through
which we see whatever we look at. It never occurs to us to take
them off.

But to see how that encapsulates the revolution, we would have to understand both remarks–not an easy task. We can begin to get an inkling of the point of the second remark if we note that it occurs just after an extended attack on platonism occupying the first 92 paragraphs of the Investigations (where such famous notions as language game, and family resemblance, are first introduced). One main idea there is that we should not see facts of meaning, or of content–notably facts as to when things would be as we said–as governed by, and always deriving from, some definite set of rules. As he says in § 81,

In philosophy we often compare the use of words with games and
calculi which have fixed rules, but we cannot say that someone who
is using language must be playing such a game.

We can begin to appreciate the point of all that if we see just what the antiplatonist picture of content is as it is developed in those 92 paragraphs. I will try to convey that here by laying out the picture as I have developed it myself, since developing it has been a major part of my own work over the last twenty five years or so.

Suppose someone says, ‘Jones has a desk in her study.’ Jones is a poor student. In the box room she uses for a study she has a door lain atop stacked milk crates. Is what was said true? A natural answer might be: it is if, but only if, she had a desk. But the question is, is what she had a desk? A reasonable answer would be: it depends on what you count as a desk. Which is to say: there is an understanding of what it is for something to be a desk (henceforth an understanding of being a desk) on which what she had is a desk; and another on which it is not. Neither understanding is incompatible merely with what it is to be a desk. Rather, what being a desk is permits either understanding; either corresponds to a permissible way of looking at the question of something’s being a desk. A note: when I say there is an understanding of being a desk on which what she had is one, all I mean is that counting what she had as a desk just is a permissible way of understanding what being a desk is. I am not thinking of an understanding of being a desk as something else from which that result derives.

To get where we are going, we need one more piece. The notion of being a desk, as such, admits of understandings. But a particular deployment of that notion, say, to describe, on a particular occasion, the way things are with Jones, may require some particular way of understanding being a desk, and exclude other such understandings. Suppose we have just attached all of Jones’ furnishings in connection with a debt, and are sending the sheriff to collect. We are speculating as to what we might get for her furniture at the brocante market. We are pessimistic about the worth of her somewhat over-used futon. But, we are led to believe, there is still hope. As someone remarks, ‘Well, she has a desk in her study.’ But what is thus said is not so. For all she has in her study is a door and some milk crates; and we certainly will not get much for that.

Remember this second piece. I will return to it in a moment. But first a brief digression. I chose this first example because it illustrates one salient feature of the meanings of the words we use, or at least many of them. What drives the wheels here is that there are many strands in what we would take ‘desk’ to mean; what we expect from it, or from a description of something as a desk. (By ‘we’, I mean we who are competent in that area of language if anyone is.) If we were to reflect on what a desk is, it would be reasonable to take a number of things as central to that notion. One would normally expect of a desk that it is an artifact; that it is made for writing on (and related activities); that it is used for writing on (and etc.); that, given its place in a certain tradition of furniture manufacture, there is a certain look and form that it would have (though, of course, we expect no strict account of the similarities that make for the right look and form). What is in Jones’ study respects some of this strand. It is certainly used for writing on. In one sense, it was made for writing on. That is why Jones assembled those bits in those ways. But it certainly does not fit our standard image of crafting, or manufacturing, a piece of furniture, nor our standard image of the form and look of a desk. So we face these questions: what relative importance should we attach to these various strands; notably to the ones that are honored here as opposed to the ones that are not? Just how insistent should we be that our normal expectations should be satisfied, rather than chalking them up to mere normal accompaniments of being a desk? And where a strand is honored if you look at things in one way (if Jones’ assembly of bits counts as manufacturing), but not if you look at things in another, just how should we look at things? If we put the facts of meaning in this way, then the point becomes that there is no uniquely right way of answering all these questions.

Another example, though, will present another way of viewing the phenomenon. Suppose I point at the ink in a certain bottle and say, ‘This ink is blue.’ Now, suppose that that ink has been designed to behave, in a certain respect, like blood. It is blue in the absence of oxygen, but turns red instantly on contact with oxygen. And it is packaged in air-tight bottles. (The accountant’s secret weapon.) So it contrasts with ink that looks blue in the bottle and writes blue as well. Now is what I said true? Well, we can rehearse all the points made in the last case. There is an understanding of what it is (for ink) to be blue on which the ink is blue, and another on which it is not. Both are permissible as such. What I said may have been said on the one understanding (that is, I may have said what is so if the ink is blue on that understanding), or said on the other, or said on neither. So I might have stated truth, or falsehood, or neither. (We are simply ignorant of the circumstances of my speaking.)

Here another aspect of meaning is to the fore. We can see it if we not a distinction between the color blue, and being colored blue. The color blue contrasts with the color red, for most purposes with the color green, and so on. We may have a perfectly secure idea of how to draw those distinctions. If asked which color a given color is, we have various techniques, or capacities, that allow us to give an answer. Those ways of distinguishing between colors form, as one might think of it, a core of techniques that may be employed in classifying objects–as we would say, by the way they are colored. But that core of techniques fits into an indefinitely diverse array of ways of classifying objects. Each of these constitutes, inter alia, a different understanding of being colored a given color. The core–our ways of distinguishing between colors remains constant across all of these. It is what makes them all understandings of being blue, or green, or whatever. So we may think of meaning (as opposed to what is said in using words, or in words on a given use of them) as putting tools at our disposal, to be combined more or less ad lib with other tools for generating ways of describing the way things are; but doing nothing in particular purely on their own.

So much for the digression. (I thought I should say something about meaning as such.) Now back to the main theme. Suppose we think of our words as connecting to universals (re-named however you like); and we suppose, just for a moment, that we know how to name those universals: their names are found in the ordinary vocabulary of English, with some scholastic nominalizations–blueness, for example. As platonism has it, those universals mediate between us and the world we describe: with no help from us, they sort out cases where the world would be as we described from cases where it would not be. Well, we now see that that just is not so. As it may be, I said ‘Jones has a desk’, and spoke truth. But that does not follow from anything the universal deskness might accomplish on its own. For there are both understandings of being a desk on which Jones has one, and others on which she does not.

What platonism now needs to suppose, then, is that the universals with which our words connect are much more arcane than first seemed. Where I said ‘Jones has a desk’, there is, we must suppose, some particular universal–let us call it flurgness-such that what I said is true just in case some item in Jones’ room partakes of that; and flurgness really does sort out cases in a unique way into ones of the world being as described in descriptions that connect with it and ones that are otherwise.

But it is just here that we revolutionary anti-platonists–Wittgenstein, Austin, Putnam and I–get off the boat. With what right do we assume that there are such arcane mediators? Trivially, there would be if there are specifiable ways of understanding being blue, or being a desk, that do not themselves admit of understandings–if, for example, we can say what it is for something to be flurg in a way that leaves no two distinct and permissible understandings of what it would be for something to be flurg. But we have no reason to believe that that is so. Perhaps desks may be made out of milk crates and doors. But if I take a standard manufactured desk and suspend it upside down from the ceiling, it is still, for most purposes, a desk. What if I suspend two strings of milk crates from the ceiling and tie a door to the bottom of each string. Is that a desk? Here we find new varieties of understandings of being one. And it does now seem that we have started on a process of discovery that may continue ad inf.

It is at this point that, for us revolutionaries, human beings re-enter the picture. There is a long story as to how, and no time to tell it here. But here is the gist. Put us normal humans (with a tolerable degree of linguistic competence) in a normal situation, and a certain way of understanding a description–of understanding being a desk, say–will strike us as reasonable, others as entirely unreasonable. That is just what I illustrated in the first example, to make the second point. Now, when Max, say, described Jones as having a desk, he said that on a certain understanding of being a desk just insofar as that is, by our human lights, the reasonable way to understand that matter in the circumstances of his so describing her. The anti-platonist point now becomes: what we humans are prepared to recognize as to what is reasonable in particular situations (what we are prepared to see those situations as making reasonable, or unreasonable) is not reducible to the work that any specifiable universal might unequivocally do.

3. Idealism: Every revolution has its price. And the price of this one is at least the apparent threat of idealism. Idealism, in brief, is the view that we, or our minds, or the mind, make up the world, to at least some extent; our thinking about the world, or our capacity for it, makes the world we think about (in part) the way it is. And that is certainly the wrong way to think about things, say, when you are on the interstate and your tank is showing empty. Merely thinking about it differently won’t help. There are, in fact, several different threats here. Since I see myself as standing with the revolutionaries, I am committed to those threats being only apparent. I wish I could prove that in the remainder of this talk (or, in fact, anywhere). But I will try to hint at at least one main idea.

Anti-empiricism carries its own idealist menace. We speak, for example, of others’ thoughts and feelings. Our talk makes sense if, supposing that it does, it is clear enough when the descriptions we thus deal in are rightly given. If it does make sense, then in engaging in it we really do describe the way things are (or are not). So there are facts as to, for example, who is happy and who is not. It is a feature of that sort of discourse that sometimes one can just see that someone is happy. So, equipped the way we are, that is the sort of thing we are sometimes able to see. For all this to be so, talk about happiness (say) need not satisfy any other, external, requirement. Notably, the facts about happiness need not follow from other facts, equally available to a thinker who did not know what happiness was, by principles available to just any thinker at all. We can see some facts as to who is happy, because we are equipped to think of the world in terms of happiness. A rational Martian, for example, might not be so equipped. He would then just not see that feature in things (people) at all. That is the anti-empricist idea.

It is part of the anti-empiricist idea, then, that the way we are designed to think opens our eyes to certain domains of fact. The question naturally arises, suppose we had been designed differently? Well, our eyes might be closed to those areas of reality. We simply would not find any such facts. But then, would there still be such facts? Or is what the facts are fixed by what we are prepared to recognize? If the latter, then our minds do construct the world. So we arrive at idealism.

The argument as it stands is not compelling. Different organisms, differently designed, might be sensitive to different facts about the world. That a given organism may be unable to discern some facts that another can is no reason to think the other is not really discerning facts. Nor, one might think, does the fact that we might have been designed differently show that the facts we can in fact discern might not have been facts. Wittgenstein, while a member of the nineteenth century revolution, provided a fitting image for this idea;

Let us imagine a white surface with irregular black spots. We now
say: Whatever kind of picture these make I can always get as near
as I like to its description, if I cover the surface with a sufficiently
fine square network and now say of every square that it is white or
black. ... This form is arbitrary, because I could have applied with
equal success a net with a triangular or hexagonal mesh. ... To the
different networks correspond different systems of describing the
world.

The way we are designed to think places a certain net against the world–the network of concepts we are equipped to use. Another design would have placed a different net. But that the world fits a given net in a given way is not jeopardized by the possibility of placing other nets against it, which it would fit differently.

Still, one might feel unease. To take a pointed case consider ethics. Suppose it really is wrong (other things equal) to step on babies, or to throw them in front of moving trains. So there are moral facts. Our ways of thinking allow us, in principle, at least, to see what they are. (It is hard to assign sense to the idea that, say, tying your left shoe before your right one is really wrong, though no human being will ever be capable of seeing that it is.) It is hard to see how the moral facts could extend farther than, or differ from, what we are in the end, prepared to recognize. Well, then, suppose that, on careful reflection, we can find nothing compelling in the idea that it is best to keep your feet off babies. Could it then be wrong to tread on them? But now suppose we had been differently designed, so that we did find nothing compelling in that idea. Would it still have been, for all that, wrong to tread? Anyway, it is hard to see how two such forms of mindedness could simply place different nets against the moral facts, either of which is a way of capturing them. So if we find two such nets, what could determine which, if either, does capture the facts?

In any event, anti-platonism deprives us of this simple line of response to charges of idealism. The anti-platonist idea is that whether a given description one might give of things–say, a description of some blood as blue–would fit the way things are depends, first, on the occasion of, or for, giving it, and, second, irreducibly on the reactions it would be natural for reasonable people to have in that situation–on some judgements as that, in such and such circumstances (selling ink to students, say), you ought not to call ink blue unless it will write blue. Anti-platonism brings human beings-expressers and thinkers of thoughts–back ineliminably into the picture. The reverse side of that is that it cannot follow merely from the fact that Jones, say, described things truly, and Pia, say, did not that each gave a different description of things, or applied different concepts. Each might, for example, have described the same blood as blue.

Pia described some blood as blue. Given human ways of thinking, she thus described it truly. Had our ways of thinking been different, in intelligible ways, in giving that description she would not have described it truly. That is just what anti-platonism comes to. (It is no anodyne view.) Blood that counted as blue might not have done so. It is an easy slide from that to the view that what is blue and what not depends on our reactions. And that, of course, would be idealism.

On a given occasion, someone describes something as blue. He thus describes things, say, truly. That he did so is, on the antiplatonist idea, thanks, in part, to our way of thinking of describing, and to the perceptions we would have of that occasion. And that fact shows itself in the existence of other occasions where to describe that thing as blue would not be to describe things truly. The misreading of that would be: what is blue, and what not, depends on our ways of thinking, or perceptions, or reactions, or something of the sort. And that would be idealism. It helps to see this as a misreading to note the following. Where Max, say, described some blood as blue and thus spoke truly, and Pia described that same blood as not blue and thus spoke truly, what Pia said does not contradict what Max did. (Of course not, since contradictory pairs cannot both be true.) The reason for that is that Max described the blood as blue on a certain understanding of what it would be for blood to be blue. Pia did the same, but on a different understanding of blood’s being blue. On the antiplatonist view, there is no way of calling something blue other than on some understanding of its being so. And that is why an occasion-independent question as to what is blue and what is not simply cannot arise.

That is some help. Still, Max described something as blue–correctly, as it happens. But were we different sorts of thinkers than we are, that would not be true describing. Well, perhaps there are thinkers different enough from us that they would not find that describing truly. We can at least imagine them. What now seems in jeopardy is the objectivity of judgement. Why should our own perceptions of describing, and of situations, count for more than the perceptions of these other thinkers? How can our perceptions make a judgement correct, when by other possible perceptions it would not be? And idealism could just as well have been taken to be the view that judgement is not objective as the view that the way the world is depends on the way our minds are. The one just amounts to the other.

Here I want to tell a tale about Sextus Empricus. Sextus suggests that different animals have different perceptions of color. He then says,

[I]f the same objects appear dissimilar depending on the variations
among animals, then we shall be able to say what the existing object
is like as observed by us, but as to what it is like in its nature we shall
suspend judgement.

Further,

If, therefore, appearances become different depending on the variations
among animals, and it is impossible to decide between them, it is necessary
to suspend judgement about external existing objects.

The lawn looks green to us, but who knows how it looks to a cow? So it would be rash to judge that it is green.

One might respond to Sextus this way: ‘What are we talking about when we talk about something being green? What is it for something to be green? However you understand that question, not something such that whether something is green or not is at all a question of how it looks to cows. Being green leaves cows out of the picture, and we are free to do likewise.’ That, I think, is a fair response.

Now let us turn to human and martian thinkers. Max calls the ink blue, and that satisfies our standards of good description. For unfathomable reasons, it fails to satisfy martian standards. But Max wasn’t speaking to, or for, martians. He was doing what he purported to do–what he presented himself as doing–if his words fit into human life as true words ought–if they have the uses for human beings that humans might (reasonably, by our standards) expect of them. If they are true, then they are a guide to certain sorts of human conduct; one on which we humans can rely (which give us no cause for complaint). They have a useful place in our activities of check signing, for example. It would just be a mistake–a misunderstanding–to expect any more of them. Which is to say that martian perceptions have no role in deciding whether Max described correctly. And if he did, then things are the way he said (a point, we have learned, that cannot always be put by saying: ‘That ink (really) is blue.’

4. Coda: In Investigations §242 (for cognoscenti, the paragraph immediately before the one in which he introduces the idea of private language), Wittgenstein says,

Communicating in language presupposes agreement, not just in
definitions, but (as odd as this may sound) in judgements. This
appears to abolish logic, but does not.

Here, I think, Wittgenstein expresses sensitivity to the idealist fears his antiplatonism might awaken. Why should there be so much as a false appearance of abolishing logic, and why should the appearance be false? We can understand his remark, I think, by returning to the contrasting quotes I used to encapsulate the shift from the nineteenth century revolution to the twentieth century one.

In the first remark; Wittgenstein thinks of logic as mapping the limits of thought, and thus of the world. Thoughts–the things there are to think, suppose, and so on–are thus its subject matter–what it is a science of. These thoughts, it is supposed, form a totality with a definite structure: for each thought there is such a thing as the way it relates to all the others. Logic’s task is to describe that structure at a suitable level of generality. Against that background, antiplatonism will seem to make human ways of thinking–natural human reactions, as the later Wittgenstein might put it–a determinant of what the logical facts are. That would be idealism of the most profound and disturbing kind imaginable. For if when things would be as any given item has them is always, unavoidably, in part a matter of how we human thinkers would see things, then so too is the entailments there are between any one such item and others. And it was just those entailments that logic was meant to map.

The second quote, though, gives us another way of thinking of logic’s appointed task. By that idea, logic is not (directly) the science of thoughts. Rather, it puts at our disposal a set of powerful tools we can use, as needed, to organize, and to systematize, thought. The tools are rigid, simply because that is how they are designed. It is a feature of them that certain relations between the organizing forms they put at our disposal are not up for grabs. That those relations hold depends on nothing, in just this sense: to imagine them not holding (if such were possible at all) would be to step outside the system, to abandon that tool. But how these tools grip onto the thoughts we wish to organize, and where they can get purchase at all, and where not, depends on the phenomenon of our thinking–on the material there is for them to work on. And if given tools should prove unsatisfactory for given purposes, we might always abandon them and construct new ones.

All that is metaphor, and at least for that reason unsatisfactory. In fact, it is unsatisfactory for more reason than that. I would not fail to sympathize with someone who found these images themselves disturbing. So there is more work to do. And that is just the point I would like to leave you with. There is a reading of Wittgenstein on which his later philosophy does away with philosophical problems, and thus, in a sense, with philosophy itself. Such problems on this reading, are to be thought of as mere pseudo-problems, even if sometimes deep for all that; as problems to be dissolved rather than solved. If that is how the twentieth century panned out, then at the end of it, one might well think, we find ourselves in the time of ‘post-philosophy’, whatever that may be. Our century is the one in which traditional philosophy died. It may be that Wittgenstein himself encouraged this view of his later work. But by working through the changed view that his later work, and, more generally, the twentieth century revolution represents, we have arrived at a batch of problems which certainly seem genuine problems, which are deep, and which even have a somewhat tradtional cast. My conclusion is that philosophy is far from over yet.

Charles Travis
Marshall Weinberg Distinguished Visiting Professor
of Philosophy
November 18, 1999

NOTES:

1. Stuart Hampshire, "In memoriam J.L. Austin 1911-1960",
    Proceedings of the Aristotelian Society, N. S. Vol. 60, 1959-1960, pp. I-ii.

2. Gottlob Frege, The Foundations of Arithmetic, J.L. Austin,
    trans, Basil Blackwell, Oxford, 1950, Introduction, p. x.

3. Tractatus Logico-Philosophicus, 5.61.

4. Philosophical Investigations, §103.

5. Tractatus Logico-Philosophicus, 6.341.

6. Outlines of Pyhrrhonism, Book I, Chapter 14, §59 (trans.
   Julia Annas and Jonathan Barnes).

7. By Op. Cit., §61.