Wittgenstein vs. Wittgenstein

We are pivoting in the course to the Investigations now. In our first discussion we went too far too fast – that is, we only discussed the first nine propositions, but couldn’t contain our desire to figure out all in one go what the relevance of language-games to sense-making was, and what the differences between the two books were. Impatience led to perplexity. We will try again today.

The Investigations does not always interpret the Tractatus charitably, or even correctly on its own terms. This is partly because of a significant change in direction. What §114 says about “Es verhält sich so und so” is more or less what the Tractatus wanted us to think about it, I think – with the result that we weren’t to affirm it, couldn’t even think it – it’s just another way of writing p, itself a formal concept – more nonsense. (“We make statements.”) But then nonetheless in §115 – “a picture held us captive.” The Tractatus wants to protest: “What picture? Didn’t you read the book? ‘Es verhält sich so und so’ isn’t a picture of anything!” Fair enough; but the picture of language’s working on which just this particular sort of use, assertion, is at the very heart of what we do in communication is now being reconceived as just one piece of a larger whole. (In fact, more than one piece, as what we call ‘asserting’ is not just one linguistic practice, not just one language-game.) The picture in the Tractatus is one on which asserting, describing, contesting, denying are the primary things we do with language, and other uses ones that will “take care of themselves.” (So that a command, say, might understood as a command to p, which in turn is perhaps analyzed in terms of commitment to the truth of a statements like “S performs x at t” and “S’s performance of x at t causes p to be the case at t + Δt,” so that “when fully logically analyzed” commands are “just” imperative-operators on propositions of the ordinary type – Wittgenstein never said anything like this, but one encounters this kind of thing fairly often in philosophy produced over the last half-century or so.) With that picture in place, and also in speech-contexts, language-games, where we are in fact communicating primarily in this manner, much of what Wittgenstein says in the Tractatus is presumably still all right, as far as it goes. But that picture is thought by the Investigations to be a substantial distortion in other contexts.

Moving on to §134 – “surely no one is going to call the letter “p” the general form of propositions” – actually, that’s more or less how I read “Es verhält sich so und so,” and I am somewhat inclined to think that Wittgenstein could have agreed in his Tractatus days. It’s ridiculous because it doesn’t communicate anything – and that’s the point – neither does “Es verhält sich so und so.” In §134 to §136 Wittgenstein is denying the author of the Tractatus the peculiar sort of erasure that overwrites the whole book.

“He explained his position to me, said that this was how things were, and that therefore he needed an advance.” If we interpret this as a meaningful utterance relative to the Tractatus, “this was how things were” means – the actual situation or position that He is in. “He went there” can be a perfectly good Tractarian sentence – it can be a logical picture of the facts – if it’s understood to mean say “Chris went to the Old Main Building.” (If it means anything in the Tractatus, it means something like that. On the other hand, it’s not clear that the Tractatus can handle “Is this really happening now?” as currently under discussion at http://languageisthingswedo.blogspot.com/2013/08/is-this-really-happening-now.html. If we view that question as a request for the truth-value of a proposition, it makes no sense – we gesture at a fact and so in effect assert p in order to ask the question “Is it the case that p?,” but the answer is presupposed in the way we set up question, which makes it in some sense a non-question. So if you agree with Lars that it has a perfectly good sense, there has to be something missing in the Tractatus.)

Wittgenstein though says something about the sentence that is impossible in Tractatus-land – “it is employed as a propositional schema.” There is no such employment in Tractatus-land. A person who knows what his position is puts that position into the sentence as a proposed reason to need money, and a person who doesn’t doesn’t understand the sentence, although he would know what kind of details to ask for concerning “his position” and “this was how things were” in an effort to try to give a meaning to the statement. “x is a black and white dog” is nonsense in the Tractatus, and so is a complex “proposition” which includes propositional variables among its constituents. It’s only the applicable utterance which has sense.

It’s true that if you treat our spoken languages as calculi then something like “E-s- -v-e-r-h-ä-l-t- -s-i-c-h- -s-o- -u-n-d- -s-o” – the spaces being an artificial device to indicate the combination of letters in question, say – could be regarded as a propositional variable, much as pronouns can be regarded as object variables. But in use a pronoun always means a particular person and an expression like “Es verhält sich so und so” always means a particular fact – or it means nothing. The variable structure of the word “his”, say, shows itself in that over here “his” means “Chris’”, over there it means “Greg’s,” etc. – the pronoun is made to take the place of a noun – it is a sign which can serve as more than one name, as it were. But as a symbol it just is one of those names naming one of those objects.

In Tractatus-land.

In the Investigations though Wittgenstein isn’t having any of this, except perhaps as a limit case in certain contexts. Rather, the discussion of “Es verhält sich so und so” in §134 to §136 is meant to undermine the idea that there is any one stable thing we mean when we talk about propositions. Is “Es verhält sich so und so” a proposition? Yes, it’s a German proposition, says Wittgenstein. But what does that mean? Well, not that it agrees or disagrees with reality. We use it as a propositional variable. It sounds like a sentence and we can plug it into complex sentences. It follows the rules of sentence formation in German. Wittgenstein doesn’t go this way in the discussion, but we might entertain a notion of ‘indeterminate reference’ – we assume that because money was asked for that there was something about the person’s situation that led him to ask for it, and if we decide it’s important we can look into it and see if that was the case and what it was, etc. You put it together with one set of propositions, it belongs, another, it doesn’t belong. Say what you like, as long as you see what the facts are.

Wittgenstein’s attention to the use of signs in the Investigations is actually baffling from the Tractarian point of view. “Maybe the author of the Investigations is right: language really is an assemblage of different uses in different contexts that don’t all work together, or that work together in some contexts and not in others, and as a result our terms are ambiguous, we get confused, etc. – so what? That’s just a matter of how we use signs. What matters is that you understand, and understanding is being able to see the symbol in the sign and know what it would be to apply that symbol to reality, no matter where it came from. The history might be of epistemological significance, might help us understand where different uses originate, but it’s surely not relevant to what a particular utterance means.”

How does the author of the Investigations respond to this? In different ways, but here’s one that’s useful, which comes up right at the beginning of the book. In what does ‘applying a proposition to reality’ consist? In the Tractatus it mostly consists of verifying and falsifying – checking to see if the thought and the fact match up. We do do this. But it is not what we usually do and often when we do do it it is at the service of, has a role to play in the context of, some other activity. The more usual sorts of application to reality are – we do something. We build something, go down on our knees to pray, play a game, kiss someone, try to predict the result of continuing to do what we are doing now, etc. Understanding in ordinary contexts is more a matter of knowing what to do than assigning a truth-value.

The reversal of perspective that the Investigations attempts is to try to deal with language primarily on these terms – in terms of the role it plays in our lives – and to contextualize and understand assertion and description within language understood in its “lived reality”, I guess, if we have to use terms like that. (It is hard for someone who’s read Heidegger not to think here of his claimed reversal of the customary dependence-relations between the thing as Vorhanden and the thing as Zuhanden, however.) The later Wittgenstein’s supposed “anthropologism,” “behaviorism,” “vulgar semantic Marxism,” etc. really just come to this – why don’t we think about meaning primarily in terms of the way language is involved in the ordering of our lives and activities, instead of primarily in terms of a kind of logical litmus paper held up to the world? There is no question of reductionism here, just of a different (and perhaps in many ways more perspicuous) way of thinking about language, and more particularly about certain kinds of philosophical problems that sometimes arise.

What’s involved in what we do are particular signs used in particular ways. We understand these – but understanding isn’t one thing any more. Sometimes it’s Tractatus-understanding, sometimes it’s do something understanding, sometimes it’s take a different perspective on the same facts understanding, sometimes it’s use-different-words-to-say-that understanding, sometimes it’s perceive the motive behind the action understanding – lots of different possibilities.

The language-games can in part be understood as tools of disambiguation. Sure, we can use the same bit of wood as a king in checkers and chess, but you shouldn’t think that you can’t move the chess king directly ahead for this reason.

I find myself unable to bring this post to a conclusion, but there are some interesting bits here, so I’ll go ahead and put it up.

Two Quick Bits of ‘Wittgensteinian’ Theology

1. 3.03 says “We cannot think anything unlogical, for otherwise we would have to think unlogically.” This is commentary/exposition on 3, “The logical picture of the facts is the thought.”

Should 3 have been put differently? As I read W there aren’t actually any ‘logical pictures’ of anything in the pure sense – ‘logical picture’ in the Tractatus is another formal concept. Why wouldn’t he write, ‘the picture of the facts is the thought’? When I think ‘Winnie is on the Tunisian rug,” surely my thought includes the spatial picture, not to mention the canine one, as well as an understanding of how to apply that picture, which latter understanding makes it also a logical picture (cf. 2.182), which applies or does not apply, is true or false.

I don’t suppose this is so important, if every picture is also a logical picture.

The broader point is this. In the Tractatus we have no proposition, no thought, no picture of the facts unless we know how to apply it to reality. Arguendo, we don’t know how to apply “Scott kept a runcible at Abbotsford” to reality. So this sentence isn’t a proposition, doesn’t express a thought, isn’t a picture of the facts. Thus it is also not a logical picture of the facts. But it is not in any sense a limitation on our thought that we can’t say or think that Scott kept a runcible at Abbotsford, since there’s nothing there to be said or thought in the first place.

Having a picture is having a logical picture and if you don’t have a logical picture you don’t have a thought. So you can’t think something unlogical because thinking just is having a logical picture, whatever else it is in addition.

Now 3.031: “It used to be said that God could create everything, except what was contrary to the laws of logic. The truth is, we could not say of an ‘unlogical’ world how it would look.”

This is a problem of theology – how God can be all-powerful if he is limited by the laws of logic? Wittgenstein’s answer in the Tractatus: the laws of logic are not a limit in this sense. What is thinkable is what is sayable is what is possible; we can give no content to the phrase “contrary to the laws of logic” or “unlogical thought.” Again with the application to reality: if you can think something it means that what you think is a picture which is applicable to the world, and the world can either be that way or not be that way. An all-powerful God conceived in the theological way would determine which, and there would be nothing beyond that to determine. It’s not a limitation that you can’t make a bit of nonsense so, because there is nothing to make so expressed by a bit of nonsense.

Although Wittgenstein does not mention it again, material later in the Tractatus is also relevant to this problem in a slightly different way. Why can’t God make it the case that p and the case that ~p? If you think of the law of noncontradiction as something apart from the facts, of the logical constants as representing (4.0312), then this seems like a limitation. But that would be the wrong picture. God would instead create all the facts, and in having done so he would also have made all the propositions true or false. God can make it the case that p and in so doing makes it not the case that ~p – they are one and the same choice, as it were.

The logical constants “supervene” on the facts in the Tractatus. Is that word too philosophically loaded? If I have ten one-dollar bills in my pocket, I also have $10, but the $10 isn’t an additional thing I have over and above the ten one-dollar bills – that’s all I mean by supervention here. Similarly with every truth-function of the elementary propositions – having specified the elementary propositions I have specified all of their truth-functional combinations s well. And all logical constants (not, and, or, the Sheffer stroke, etc.) do is specify particular truth-functions of interest, which were in effect already there in the elementary propositions themselves – logic is just the combinatorics of truth-functionality, as it were.

So you can’t do and not do something – having done it you haven’t not done it – etc. Not different facts but the same facts presented differently.

2. From the Lecture on Ethics: “I can describe my feeling by the metaphor, that, if a man could write a book on Ethics which really was a book on Ethics, this book would, with an explosion, destroy all the other books in the world”

John 21:25: “And there are also many other things which Jesus did, the which, if they should be written every one, I suppose that even the world itself could not contain the books that should be written.”

Who is Jesus Christ, for John? The Word made flesh, who dwelt among us. But the Word is also the Bible, whose central character is God, and Jesus Christ is God made flesh as well – fully divine and fully human.  

So the Word, that is, the Bible, is made flesh in Christ, and the doings of that Word are so great, that even the entire world can not contain enough books to express them.

Nonetheless here, says the Christian, is a book that does.

The Sense of a Proposition

One of the perennial problems of philosophy is the problem of the ‘aboutness’ of thought. When I was a child someone told me that the Andromeda Galaxy is the nearest galaxy to our own. Having read some science books I pictured a sort of vague fuzz of stars that was nearer to our little spiral fuzz than some other fuzzes farther away.  Being that sort of child, I would have confidently asserted that the Andromeda Galaxy was the nearest galaxy to the Milky Way to anyone who cared to quiz me on the matter.

The question is: what made my thoughts, my mental pictures, my words ‘about’ the Andromeda Galaxy? Gorgias, as paraphrased by Sextus Empiricus, puts the problem marvelously: “It is evident that objects of thought are not things with being. For if objects of thought were things with being, then everything that one thinks of, however one thinks of them, would have being. But this is nonsensical. For it is not the case that if one thinks of a man flying or a chariot being driven in the sea, there immediately is a man flying or a chariot being driven in the sea. And so it is not the case that objects of thought are things with being.” (R. G. Bury, trans.) 

So since the mental representation is not identical with the object it represents, in virtue of what does the representation represent?

The magnitude of the difficulty can be appreciated by opening Book III of Aristotle’s De Anima, in which one of the most sober-minded thinkers in the history of philosophy reads like Hegel. One way of parsing Aristotle’s theory, loosely, is as follows. The mind is a sort of infinitely plastic – something – that can take on any form at all, and what constitutes the aboutness-relation for our thoughts is that the infinitely plastic something that is your mind takes on the same form as the thing out there in the world that you are thinking of. Further, since form is abstract, and since your mind and the thing you are thinking of have the same form, there is a sense in which Gorgias was simply wrong: your thought is in fact the thing you are thinking of, at least when you think of something real – not materially, but formally, just as two matched silver cups share the same formal cause. Thus there is a sense in which, for Aristotle, when I think of the Andromeda Galaxy, my plastic mindstuff has taken on (vaguely?) the form of said galaxy, and because my thought and the galaxy share this form, I am thinking of the galaxy.

This sort of resemblance theory of mental representation got some additional support two thousand years later, when the author of La Géométrie pointed out that owing to the perfect clarity and distinctness with which we can form the idea of an infinitely good, powerful, and knowing being, such a being must exist, and that the existence of such a being is incompatible with wide-scale general failure of aboutness of thought.

These were extremely intelligent people. We are talking about the inventors of formal logic and analytic geometry here. Try to solve the problem on their terms for yourself and see if you do any better.

In the Tractatus Wittgenstein solves the problem essentially by rejecting its parameters. In order to serve as a picture, a thought or proposition does, as in Aristotle, ‘share a structure’ with the fact it represents if it is true. In fact, we can even go farther with the parallel: just as in Aristotle, the true proposition and the fact it speaks of share a common structure. 

But it’s not: the fact over there, the proposition which picks it out over here, and some common structure possessed by both which ‘makes it the case’ that the thing over here means the thing over there.

Rather, “The essential nature of the propositional sign becomes very clear when we imagine it made up of spatial objects (such as tables, chairs, books) instead of written signs. The mutual spatial position of these things then expresses the sense of the proposition.” (3.1431)

That is, the way things actually are expresses the sense of what we say. This is an inversion of the usual formula, wherein the sense of what we say expresses the way things actually are, or tries to. The usual formula is also OK by W, I think, but the fact that it goes both ways is highly significant.

Even more clearly: “What the picture represents is its sense.” (2.221)

That is, whatever it is we do to understand “Winnie is on the Tunisian rug” is (one of) the same things we do when we look at the Tunisian rug and see whether Winnie is on it; and (one of) the same things we do when we look at a crayon drawing and see that it’s a picture of Winnie on the Tunisian rug; and so on. To understand the sentence or the drawing is to know how to apply it; to be able to look at the rug and see whether Winnie is on it is to be able to understand the proposition. Likewise if we say it in German, etc. (cf. 4.243).

This is the thing in Wittgenstein which gets turned into verificationism, but it’s not verificationism. Verificationism takes a similar insight and crossbreeds it with empiricism to arrive at the doctrine that (roughly) only operationally defined terms with a clear sensory interpretation are meaningful. I think that in the Tractatus W sometimes leans in a verificationist direction, but even then he can’t have been a verificationist, because this is in effect a doctrine of sense, which we can’t have: what it makes sense to say is primary. At the very most verificationism would be the sort of thing that ‘shows itself,’ not a doctrine that could be directly articulated.

It’s also not idealism. “From what you’re saying, Wittgenstein thinks that there’s nothing more to reality than our true thoughts of it!” In a way that’s true, but you could also just as well say that there’s nothing more to our thoughts then their being true or false of reality. The world consists of facts; facts are what true statements state; understanding a statement just is understanding how the world would be if it were true.

Perhaps another way to put it is that ‘thought’ and ‘reality’ are interdependent notions in the Tractatus. (One of the last classes I sat in on with Peter Winch was a seminar on Spinoza which touched on this theme; I think also here of Cora Diamond’s portrayal of Berkeley as a certain sort of “realist.”) We have no conception of reality beyond our thoughts of it, beyond the propositions that could be applied to it; that’s the Berkeleyan side. But we likewise have no conception of thought or propositions beyond our ability to apply them to the world; that’s the ‘direct realist’ side, if you like.

So it’s not, it’s never in Wittgenstein, the fact over here and the proposition over there. Rather, the true proposition is the fact; the proposition is the sense is the position in logical space. Reality makes our statements about it true or false, and is thus in a sense ‘independent of our say-so’, but we have no conception of reality other than the one we can state.

The nonsense has gotten pretty heavy now, but I think this is the view. “The proposition determines reality to this extent, that one only needs to say ‘Yes’ or ‘No’ to it to make it agree with reality. Reality therefore must be completely described by the proposition.” (4.023)

“Whether or not ‘Winnie’s on the Tunisian rug’ is true depends on whether or not Winnie’s on the Tunisian rug.” Yes, but only in the sense that whether the light is green depends on whether the light is green.

The question as to how our thoughts and statements can be about reality is on this view simply a bad question. It’s like asking how a mountain can be next to a valley. They’re not the same thing, but they are linked to one another at a fundamental level.

Gorgias’ argument is addressed by the intrinsic bipolarity of thought. To the degree that a proposition like “A man is flying” has a clear sense, we know how to apply it to reality; either it applies or it doesn’t. If it does, the “object of thought” is the “thing with being”; if it doesn’t, there isn’t any “object of thought” either – only a projection which doesn’t meet with success, as it were.

In such a framework the Cartesian problem arguably never even gets off the ground. Likewise, there is no gap between epistemic and metaphysical possibility.

(This is a post that would really benefit from some spirited interrogation and challenge, if any students, friends, or casual passersby want to get into it a bit.)

Ein Zeichen sind wir

Wittgenstein offers a remarkable solution to the problem of the truth-values of statements ascribing propositional attitudes:

5.54 In the general propositional form, propositions occur in a proposition only as bases of the truth-operations.

5.541 At first sight it appears as if there were also a different way in which one proposition could occur in another. Especially in certain propositional forms of psychology, like “A thinks, that p is the case”, or “A thinks p”, etc. Here it appears superficially as if the proposition p stood to the object A in a kind of relation. (And in modern epistemology (Russell, Moore, etc.) those propositions have been conceived in this way.)

5.542 But it is clear that “A believes that p”, “A thinks p”, “A says p”, are of the form “‘p’ says p”: and here we have no co-ordination of a fact and an object, but a co-ordination of facts by means of a co-ordination of their objects.

How is a proposition like “A believes that p” applied? 

I say “Dave believes that we’re all talking nonsense.” I might have just heard Dave say “You’re all talking nonsense.” Or I might have heard Dave construct a clever parody of what we were just saying. Or I might just be watching Dave squirm in his seat while I’m off in some aside about scientific realism and, remembering that Dave has some Carnapian and nominalist tendencies, I can see the symbol (which is p – “You’re all talking nonsense”) in the sign (his squirming).

There isn’t any device which allows us to read off beliefs from neurological states that I know of, and it’s not clear that such a thing could be done. But if there were that would be fine too. Likewise a telepathic sense, etc. Any method of belief ascription ought to do.

If I say “My dog Winnie is on the Tunisian rug” and you’re in the room with me you can look for the rug and see if there’s something doggish on it, or just look for Winnie if you know her. If we’re talking on the phone and I’m saying the same thing you can’t do that, but if you understand what I’m saying you’d know that that was the sort of thing you’d do to apply the statement to reality.

It’s exactly the same thing with “Dave believes that we’re all talking nonsense.” In essence, the things Dave says and does are signs, in which we can see, hear, or think various propositions.

3.12 The sign through which we express the thought I call the propositional sign. And the proposition is the propositional sign in its projective relation to the world.

p’ says p has given many commentators pause because they forget that for Wittgenstein “the propositional sign is a fact” (3.14) – “the propositional sign is articulate” (3.141). Consider also in this connection 4.022: “The proposition shows how things stand, if it is true. And it says, that they do so stand.” “‘p’ says p” is nonsense in exactly the same way as 4.022 is; the sign, ‘p’, says nothing by itself, qua object, these syntactic tokens or that set of behaviors; it is only the sign ‘p’ as understood, in its projective relation to the world, that says anything at all; but that just is the proposition.

So when we say “Dave believes that we are all talking nonsense,” we are saying that Dave himself is in effect a sign symbolizing “we are all talking nonsense,” which could be expressed in behavior or words or even privately by Dave to himself through a certain quiet aggravated shrugging of the shoulders. And Dave’s belief is that that Ein Zeichen sind wir, deutungslos, just as Hölderlin had it: we are talking, but our words lack meaning.

“A coordination of facts by means of a coordination of their objects”: let’s switch to Winnie on the rug for a moment for convenience. The words I type, my thought of her as I look on, Kira’s gesture with her head after Miriam asks “where’s the dog?” and so on – and, if p is true, Winnie herself sitting on that Tunisian rug – all share the same structure, which is shown by thought and reality alike. (The gramophone record, the musical thought, the score, the waves of sound – 4.014.)

So now we are ready to understand 5.5421:

“This shows that there is no such thing as the soul – the subject, etc. – as it is conceived in contemporary superficial psychology. A composite soul would not be a soul any longer.”

Although there are interesting points of contact between Wittgenstein and Hume, I don’t think this is one of them, as some commentators have suggested. We don’t need ‘the soul’ because we don’t need to posit any sort of subject to evaluate propositions of the form “A believes that p.” To believe that p is just to be a sign/symbol for p yourself – and this definition holds in the first person (as when you say or think p) just as it does in the third. p can be asserted in your own words or thoughts or behavior or someone else’s words or thoughts or behavior – and it is the same p in all of them.

So the problem of the ‘composite soul’ here is not that some weird hybrid entity including person-objects and proposition-facts is incoherent – it’s that the things that were supposed to be possessed by the souls/minds/subjects in question aren’t possessed by them at all. We don’t need the hypothesis in the first place. Rather we – whatever we are, plain old persons or souls in some other sense – just instantiate them, serve as signs for them, at various times in various ways. If “Dave believes that p” just means that Dave is being a sign symbolizing p just as someone else’s words or thoughts or behavior might, then you can’t call that belief any sort of possession of a mind our soul, because it would be the same belief in my mind and in yours and in the minds of anyone else who thought it. It’s a zusammengesetzte, an agglomeration of disparate things that doesn’t at all track the individuals who were supposed to be the possessors of this sort of soul in the first place.

Plausible as an interpretation?

Is the view defensible more generally? I suspect it might be the only possible one if the Tractatus’ discussion of sense and the relationship of propositions to reality holds up, so perhaps that is really the place that one would want to dig in, but we can raise the question independently here too.

At the beginning of the post I said that Wittgenstein was here dealing with “the propositional attitudes”, but we have only talked about belief. What about “Travis fears the tiger?” Well, if we see Travis running from a tiger, his fear-behavior works just fine to symbolize “Travis believes that there is a tiger chasing him” too.