Ludwig Wittgenstein - Philosophical Investigations For those five/eight members who have read this book or are reading it, here's a discussion thread. Aran and I are reading it now, and I know Whitefish has read it. I realize that this is longer than the other Philosophy Discussion Group works, but I figured it was best to structure the thread in this manner because it will cut down on spam significantly. To start: #65: "Instead of producing something common to all that we call language, I am saying that these phenomena have no one thing in common which makes us use the same word for all,—but that they are related to one another in many different ways. And it is because of this relationship, or these relationships, that we call them all "language". I will try to explain this." It strikes me that a reluctance to accept an answer like this is a huge source of philosophical confusion. Just briefly thinking about it, it seems to apply in practically all areas of aesthetics, species concepts in biology, and more. This is a great rejection of "essences", I feel. Wittgenstein specifically relates it to ethics and aesthetics in #77, where he compares trying to find a single, comprehensive definition of something like "the good" or "the beautiful" as akin to drawing a sharp picture corresponding to a blurred one: there's no one sharp picture that perfectly corresponds.
Just reposting our earlier discussion to this thread so everyone can see it: Yes, absolutely. It's easy to miss how radical that thought is, but really (I think) it's rejecting a general philosophical method that's been the heart of most philosophy since Socrates. The thing is, unless Wittgenstein can provide a good justification of that thought it's easy to dismiss it or miss the point of it. It's easy to think that because we understand a word or have a concept we must know some thing which is the essence/meaning of the word or concept, and philosophy is an attempt to discover essences. Hence, we think the questions, 'what is justice?', 'what is time?', 'what is truth?', 'what is a number?', 'what is pain?', 'what is beauty?', 'what is knowledge?', must have clear answers, and if we can't find any clear answers then the concept in question is empty. I think a vast amount of philosophical discussion therefore centres on suggesting different ways of answering these questions. On the one hand you have the realist answer: 'Pain' stands for pain, 'good' stands for good, 'beauty' stands for beauty. Hence, the thing the word stands for is a simple, indefinable element. That's the Platonic answer; I think Plato's theory of Forms is essentially a response to the problem of trying to explain what words like 'beauty' stand for. Socrates' attempts to define 'virtue', 'piety' and so on in the early Platonic dialogues fail, so Plato's thought is that either these words stand for nothing, or they must stand for virtue itself, piety itself, beauty itself. And the Forms are the things for which these words stand. I think the famous sections of Republic and Phaedo argue for that conclusion: But if you reject Plato's realist position then you're led to a reductivist one. Hence, we might say that 'good' stands for the greatest happiness of the greatest number, or that 'good' is a term uttered to express personal approval. I think Plato found the reductivist response dissatisfying so formulated forms. The other alternative is eliminativism; that is, you just eliminate the problematic concept. If you can't say what 'good' stands for then you just stop using 'good'. Those three positions - realism, reductivism and eliminativism - are positions in a huge number of philosophical debates. They're most prominently used in the philosophy of mind. What is pain? Is it a simple, indefinable mental element? That's a dualist position. Is it is a certain form of behaviour? That's a behaviourist position. Is it a certain brain state? That's an identity theorist's position. Is 'pain' a myth created by pre-scientific people? That's an eliminativist's position. Wittgenstein's insight is to say that all of these positions are mistaken because they all take the same mistaken view that philosophical questions are about essence. In the Investigations Wittgenstein was particularly concerned to discuss the philosophy of mind/psychology (which really gets going at section 244ish - might be remembering that wrong), and the philosophy of mathematics. But he doesn't discuss the philosophy of mathematics in the published version of PI. His original plan was to write the early sections on language and then show the consequences of that for a discussion of mathematics. Instead he retained the same early discussion of language but it flowed into a discussion of the philosophy of psychology instead. That was possible because his discussion of language underpins everything else. By replacing the false view of language with a new one he thinks he dissolves all those old philosophical problems. You can see that he thinks in that way because some remarks remain comparing certain reductivist positions in the philosophy of mind to reductivist positions in the philosophy of mathematics (somewhere around the early 300s there's such a remark I think). He says both conceptions make the mistake of denying something (what the realist posits), when they should actually deny that there is anything to deny. But to do that requires a radical move away from the standard picture of doing philosophy; discovering essences. But the really interesting parts of the Investigations, in my view, are his discussions of how language works that justify the claim that the search for essences is pointless and mistaken. I think sections 130ish to 240ish are the most interesting in that regard, although everything from 1 onward is important in painting the picture he wants to paint.
For orientation, here's what I think is a very simple picture of the 'argument' of the Philosophical Investigations. If you read or have read the book you'll know that this is a gross simplification of its content, and that Wittgenstein's discussions don't actually follow in this order, but I think it's helpful. 1. One view of how language works is that the meaning of a word determines how it ought to be used. This leads to a picture of language according to which written or spoken words correspond to meanings, which are somehow associated with the words, the words themselves being arbitrary written shapes or spoken sounds. Understanding a word means internalising or grasping the meaning with which it is associated. Once a person has done this internalising or grasping they are able to use the word correctly; the use of the word follows from the meaning they have grasped. 2. Wittgenstein suggests this picture of language doesn't work. He does this in several ways. a. Not all words are names - The idea that all words are associated with meanings brings naturally the idea that all words are names for things. 'Wittgenstein', 'Socrates' and 'Whitefish' are names for people; '1', '+' and '%' are names for concepts; 'beauty', 'justice' and 'goodness' are names for other types of concepts and so on. Throughout the Investigations, but particularly in the early part of it, Wittgenstein simply wants to show us how much of our language doesn't fit with the idea that words are names for things. (See, e.g., sections 11-14) b. It isn't so simple to 'associate' a word with a meaning - The picture of language use set out in (1) assumes that words are associated with meanings. But how does this association happen? It's tempting to think of things in the following terms: The word 'Socrates' is associated with its meaning - Socrates the actual person - because someone pointed at Socrates and said 'I call him 'Socrates'', and so the link was set up. Wittgenstein points out that pointing at something and uttering a word only makes sense within a language. Pointing at something, for example, is a piece of linguistic behaviour. Is this arrow, --->, pointing to the right or left? Is it even pointing at all? We only say that it's pointing to the right because we have a pre-existing understanding of the language. Pointing at something and saying a name (this is called 'ostensive definition') therefore only works from within a language. We can't begin the language by naming things in this way. A person who doesn't share our pre-existing language might not understand anything by pointing at Socrates and saying 'Socrates'. When I point at something and say 'look!', my cat just looks at the end of my finger and not the thing I'm indicating. So original associations between words and names can't be set up in this way. So we need another explanation of how words are associated with meanings. (See 28-31) c. We can perfectly well make sense of cases in which words do not correspond to anything, and yet they are still useful and, Wittgenstein would argue, meaningful. (41) d. Wittgenstein also spends a lot of time calling into question much of the view of language he held himself in his earlier work, the Tractatus. In that book he held a picture of language the same as that described in point (1), above. However, he worked out the consequences of that picture of language in extreme detail. He discusses flaws in his thought about those details in the opening sections of the Investigations (See, for example, the discussion of 'simple elements' or 'objects' at 44-64. He is here criticising Russell as well, and the famous theory of descriptions is called into question at this point.) e. Wittgenstein questions the picture of understanding associated with the original view of language - On that original view, understanding consisted in grasping the meaning associated with a word, and then being able to derive one's use of a word from that meaning. Grasping the meaning of a word gives one a rule from which one can derive the correct use of the word. In arguably the most important part of the Investigations, and certainly one of the most discussed, Wittgenstein sets out to argue that this view of how the understanding works is deeply flawed. He argues that there is nothing we could 'grasp' or 'internalise' that could show us how we ought to use a word in the future. The sections of the book that deal with this topic constitute, in my view, perhaps the most insightful and astounding philosophy ever written. Sections 82-88 hint at this argument, but the heart of the discussion is found in sections 138-201. Kripke's very famous book, Wittgenstein On Rules and Private Language provides an extremely useful introduction to this section of the Investigations. Kripke's interpretation of Wittgenstein is highly controversial, and I think it's significantly mistaken in several very important ways, but nevertheless I think it's still the best introduction to this part of the Investigations. It gives you a way in, and once you're in you can try to find your own way around. 3. Wittgenstein replaces the flawed picture of language with a new one that avoids the problems he has discussed. - The flawed picture of language is that meaning determines use. The new picture of language is that the use of a word determines the meaning of a word. This is first set out at section 43, but what the new picture really means only becomes clearer as you read on. Sections 195 - 242 are particularly helpful I think. 4. Wittgenstein discusses the consequences of replacing the old picture of language with the new one - This is where we come to the stuff Pnoom! and I discussed above. The old picture of language encourages the idea that there's a thing correspond to each word. It then becomes coherent to ask, 'what is beauty really?' We think there must be a thing, beauty, which is the meaning of 'beauty'. A huge number of philosophical problems emerge from this way of thinking, and arguably the majority of philosophers have attempted to answer questions that stem from this way of thinking. In sections 66-67 Wittgenstein suggests that there needn't be 'essences' that correspond to words. In sections 100-133, Wittgenstein presents us with the view that most philosophy stems from misunderstandings of how language works - i.e. stem from the idea that (1) is the correct picture of language. In sections 243 onwards, Wittgenstein discusses important other consequences of replacing the old picture of how language works with the new one. Most significantly, his new picture of language rules out the possibility of a 'private language'. The 'private language argument', which seems to become very focused at 258, asks us to revise our view of how words like 'pain' work. Much of the rest of the book is concerned with getting us to revise our idea of how various psychological concepts function, and explaining the misconceptions that arise from misunderstanding how they function. In the final remark of the Investigations, Wittgenstein notes that a reappraisal of mathematical concepts could be carried out along much the same lines as his reappraisal of psychological concepts. This is because both reappraisals utilise the new picture of language to expose the contradictions inherent in the old one. The mathematical investigation that Wittgenstein himself carried out has since been published as Remarks on the Foundations of Mathematics. So, to summarise, a huge oversimplification of the argument of the Investigations is as follows: 1. Wittgenstein criticises a false view of language. 2. Wittgenstein replaces the false view with what he believes to be the correct one. 3. Wittgenstein shows that doing so leads one to realise that many, or even most, philosophical problems are not actually problems.
I'm taking a Chinese Philosophy course this semester, and I just read three chapters of Herbert Fingarette's take on Confucius. I don't really know how well he's thought to have captured Confucius' intention, but a lot of it strikes me as very Wittgensteinian. For instance: "There is no inherent way or method of deciding. One simply decides... we could have made the same decision though our deliberations preceding it had taken a different route" Reminds me strongly of "a wheel that can be turned though nothing else moves with it, is not part of the mechanism" (PI 271). I think this really comes through in (Fingarette's interpretation of) Confucius' view of jen* as an outward way of action where your action displays a certain intention (NOT a specific mental process). *jen involves respecting others as worthy of equal dignity of yourself. I probably just horribly misrepresented it because I don't yet fully have a handle on how to talk about it.
#403. If I were to reserve the word "pain" solely for what I had hitherto called "my pain", and others "L.W.'s pain", I should do other people no injustice, so long as a notation were provided in which the loss of the word "pain" in other connexions were somehow supplied. Other people would still be pitied, treated by doctors and so on. It would, of course, be no objection to this mode of expression to say: "But look here, other people have just the same as you!" But what should I gain from this new kind of account? Nothing. But after all neither does the solipsist want any practical advantage when he advances his view. _______ Is Witty here saying that solipsism is just a new way of talking?
A new way of talking without important consequences. That's what I take it to mean. Whether or not other people are in pain is a matter of our responses to them and the way people treat them, etc. The solipsist changes none of that by refusing to say that other people 'have pains'. So the solipsist has introduced a new terminology, but that's apparently all he's done. I think it's an instance of Wittgenstein's general view that to understand the meaning of a word we should just look at what it does. Normal people (not solipsists) use the word 'pain' in such-and-such cases involving other people and so on. The solipsist is basically refusing to look at how the word is actually used and saying 'this is what I mean by pain' instead. Wittgenstein is saying, "Fair enough Mr Solipsist, but that doesn't change anything". At the same time Wittgenstein's admitting that he can't provide a knockdown response to the solipsist (that's the point of saying, 'It would, of course, be no objection to this mode of expression to say: "But look here, other people have just the same as you!"'). But we can see that the solipsist's view is pointless if we start to accept Wittgenstein's picture of how language works and how things are with philosophy. The solipsist, of course, isn't looking to gain anything practical through his position. His position is the end result of having the wrong picture. The solipsist has tied himself up in a knot, and he's interesting because many many philosophers and philosophically-minded people are tempted to get tied up in the same knot. But Wittgenstein's saying, 'if you follow my line of thought, the solipsism problem doesn't even emerge'. I don't really know if that's what W means there, but that's what I take it to mean at present.
480. Thus one could in fact take "grounds" for an opinion to mean only what a man had said to himself before he arrived at the opinion. The calculation that he has actually carried out. If it is now asked: But how can previous experience be a ground for assuming that such-and-such will occur later on?—the answer is: What general concept have of grounds fro this kind of assumption? This sort of statement about the past is simply what we call a ground for assuming that this will happen in the future.—And if you are surprised at our playing such a game I refer you to the effect of a past experience (to the fact that a burnt child fears the fire). This suggests that he would respond to Hume's problem of induction by saying that it's simply the way we behave.
Absolutely, which is also the same response Hume gives. The thing is, Hume doesn't really think he's discovered a 'problem of induction'. His point isn't, 'inductive reasoning is without grounds! Oh no! This is a problem!'. His point is rather, 'inductive reasoning is without grounds, and this tells us something interesting about human beings and human nature'. The idea that Hume discovered 'the problem of induction' is something that other philosophers have created. Lots of people dedicate their life to attempting to solve 'the problem of induction', but a clear reading of Hume shows that there's no problem at all. Hume was arguing against the idea that human beings are or have the capacity to be 'fundamentally rational', by which I mean the idea that all human activities could be grounded in reason. Hume is saying that at some level we can't justify anymore, and have to say: 'Well, this is just what people do'. In that sense Hume is pointing out that human beings are just like animals. Edward Craig describes Hume's aims as follows (not exact quote because I can't find it): 'Hume was concerned to show that man is not the least of the angels, but the greatest of the animals'. The point is that while human beings are very complicated animals, capable of hugely complex activities, we nevertheless do not transcend our animal natures. So while we can reason about the future in more detail than a cow can, that does not mean that our reasoning is fundamentally grounded in a way that the cow's is not. In making such an argument Hume was going directly against the mainstream in philosophy, and particularly against the Cartesian picture of man which is still very commonly accepted today. It seems to me that Wittgenstein was performing a very similar role to Hume, but with regard to the use of language. For even if we follow Hume in saying that humans are animals with regard to their inductive reasoning, it's still tempting to think that the human ability to use language implies a higher faculty. This particularly comes up with mathematics and logical reasoning. It's tempting to think that when we're doing mathematics we're doing something that is, in some way, fundamentally rational. So while I have an animal part, I also seem to have a rational part, and that rational part is what deals with language use and mathematics. Wittgenstein's arguments about rule-following (see sections 185-242 in particular) are, in my view, close to Hume's arguments on induction. Before Hume it was tempting to say that we decide how to act by reasoning inductively, and then acting. Hume showed that explanations of this sort come to an end somewhere, and only make sense if we presuppose that inductive reasoning works in the first place. Wittgenstein is attacking the view that I know how to use a word in the future/continue a sequence etc. because I can work out, through rational means, how to go on. But explanations of this sort break down very quickly. There's nothing we can point to that justifies me in going on in one way rather than another. So in the end we have to say, 'this is just what I do; this is just how I use the word'. So just as Hume argues that we can't be fundamentally rational in deciding how to act, so Wittgenstein argues that we can't be fundamentally rational in deciding how to use words. Indeed most of the time we don't decide at all. We just do things and say things. As I type these words I don't have them running through my head first; I just move my fingers on the keyboard and the words come out. Similarly, when I want to move my arm or run across the room, I don't think 'I must move my arm and move my legs so as to move across the room'; I just do those things. The crucial point that Wittgenstein and Hume are both making is that this 'lack of rationality' in our actions isn't a defect in human nature. We would only think that if we have an incorrect view of human nature. I.e. if we had the idea that human beings have a 'rational part' through which instructions are being processed and commands given to the body, the animal part, which then does things. I think this point, and the link between Hume and Wittgenstein, is brought out at PI 84: So just as we, at some point, are unable to justify our belief that a hole will not be behind the door, so we are unable to justify ourselves in using language/continuing a sequence in a certain way. But this does not create a problem for us. Because we just do things anyway, and we are generally very successful in what we do. This is Wittgenstein's general attitude toward scepticism (see below). I think Wittgenstein and Hume were both very much against the idea that human beings think first and then act, and instead in favour of the idea that, ultimately, human beings just act. 'Thinking' is a concept we learn later and apply in specific contexts. So what I'm trying to say is that while it's perfectly coherent to say that human beings think about what they do, it isn't coherent to say that human beings are ultimately thinking beings. Because thinking couldn't occur without the existence of complex language games which require a great deal of acting to get started. I would also argue that just as Hume is concerned to show that human beings are essentially animals, and do not transcend their animal natures, so is Wittgenstein. See, for example, PI 25: I think in this passage Wittgenstein is acknowledging the common idea that language requires an internal power of thought (in one's rational part), but then giving his response: "they simply do not talk". He then goes on to say that "Commanding, questioning, recounting, chatting", i.e. all language games, are 'as much a part of our natural history' as other activities that we share with animals. The point seems to be that insofar as we use complex languages we are different from other animals, but we are not fundamentally different. Humans do not build nests, but that doesn't mean that blackbirds are a fundamentally different sort of thing from us. If you're interested in these issues then Wittgenstein's On Certainty is certainly worth reading. It's basically using his philosophical method to discuss 'the possibility of knowledge'. He argues that when Cartesian sceptics get worried about whether or not we can have knowledge, they are misusing the term 'knowledge'. To see what 'knowledge' means we need only observe the use that's made of it. He goes on to argue that 'knowledge' is a precision concept; we can only use it in certain contexts. A sceptic demands that we must be able to everything, and if we don't then we know nothing. Wittgenstein says that we don't need to be able to know everything because it is incoherent to suggest that we could know everything since to do so would be to misuse 'knowledge', the precision tool. What's important on a fundamental level is not that we know things but that we are able to do things: In this sense we are exactly like other animals. Wittgenstein is explicit about this point: On Certainty is a wonderful wonderful book, and the best piece of philosophy on 'the problem of knowledge' that I've ever read.
I really need to read Hume. I started Treatise on Human Nature over the summer but got distracted and didn't finish. Everything I know about him (including what you say above) suggests that I would really like him.
Hume is brilliant but actually quite difficult I think. The Treatise is also off putting because it starts with a lot of terminological stuff (that he inherits from Locke) which isn't really argued for and isn't that interesting. He also seems to spend a lot of time arguing about whether or not space is infinitely divisible, or something, and it's hard to care either way (he's responding to the concerns of the day). But Hume gets really interesting when he starts talking about 'knowledge and probability', which is Part III of Book I. There you get the famous arguments about causation, which I think are beautifully argued, as well as the arguments about inductive reasoning. You also get a discussion of knowledge of the external world, and of personal identity. In my second year as an undergrad one of my essays was about Hume's views on personal identity. In the original version of the Treatise he confidently sets out his ideas, but in the Appendix to the book, added later, he says that his writings on personal identity have troubled him. He gives a cryptic reason for this which doesn't explain much, and that's it. I remember it being a really fun essay trying to come up with what Hume's problem was. The Dialogues Concerning Natural Religion are excellent as well. The arguments are generally quite familiar to anyone who's considered the arguments for and against the existence of God, but the book is so well composed that it's just really enjoyable to read.
673. The mental attitude doesn't 'accompany' what is said in the sense in which a gesture accompanies it. (As a man can travel alone, and yet be accompanied by my good wishes; or as a room can be empty, and yet full of light). Is he here (and elsewhere) trying to suggest that mental states only exist insofar as they are displayed by action (similar to Sartre's claim that an action determines the actor's values, rather than the other way around). I'm a bit confused by this general line of thought in the book.
I agree that this is a difficult line of thought and it's one I don't particularly well understand. What is clear is that Wittgenstein's arguing against the idea that the mental state is something hidden and private, and the external behaviour is just contingently related to it. We don't take someone's behaviour as evidence for a hidden mental state; their behaviour is in some sense constitutive of their having the mental state. That's why we can say, 'I can see that you're in pain', rather than, 'I can see that you're exhibiting behaviour that tends to be associated with pain, and that leads me to the conclusion that you're experiencing pain, even though I have no way to know for sure'. In that sense 'reading' other people's behaviour is like reading a page of text. When I read the words 'Please buy more milk' I don't take the written symbols as evidence for the message's communicating a certain meaning. There is no meaning hidden behind the written symbols. So I think the same is true for our responses to other people. When someone jabs himself with a pin and grimaces the fact that he's in pain is evident; his mental state is evident. It's not something separate from what I see that merely 'accompanies' it. So it's certainly clear that Wittgenstein is arguing for something like that. At the same time he certainly isn't taking the reductionist approach which says that to be in pain just is to exhibit certain forms of behaviour. There is a clear difference in our language between behaving as if one were in pain, and being in pain. But Wittgenstein argues that we don't need to explain that difference by reference to a hidden 'mental state', which is the mistake that dualists make. We would only make that mistake if we bought into the conception of language which says that all words stand for objects, and that's clearly a conception of language against which Wittgenstein is arguing. At around 300 Wittgenstein really focuses on behaviourism and gives his account of why he isn't a behaviorist. It's because he's rejecting the dichotomy between dualism and reductionism. Both dualist and reductionist say that 'pain' must stand for something. The dualist says it stands for something hidden and private and irreducible. The reductionist says it stands for a certain form of behaviour, or a certain brain state etc. Wittgenstein's saying that 'pain' doesn't need to stand for anything. The word 'pain' is meaningful if it has a clear use in the language game, and it does have a clear use in the language game. What's clear is that the use it does have is tightly bound up with observations of animal activity. This is why it's barely coherent to suggest that a mountain or a pencil could be in pain. So 'pain' certainly has a use in our language game, and the use of 'pain' occurs only in certain situations that resemble each other in certain ways. But that doesn't mean that we can say, 'pain just is...' and fill in the gap with a description of behaviour etc. So the basic idea is that what is said expresses a mental state and isn't just contingently related to it. But that doesn't mean that mental states are just expression. So I think your assessment - that mental states only exist insofar as they are displayed by action - isn't far off. At the same time, though, I'd be very wary of saying that Wittgenstein is arguing for or against the 'existence' of mental states. He's discussing the contexts within which terms like 'mental state' or 'pain' or 'understanding' or 'belief' can be used. He's not saying that mental states do or do not exist. If you think of 'understanding' as a thing that 'exists' then you're at risk of getting caught in the trap of thinking that 'understanding' is some sort of object, and then you get stuck in the dualist/reductionist dichotomy. I guess in general Wittgenstein wants us to avoid thinking of understanding and pain as things. Of course, there's still something confusing in all of this. I can very well sit in a room on my own and think to myself 'I love pancakes', and it seems perfectly coherent to say that I'm having a thought that no one else will ever know occured. So isn't that a private occurence? And then isn't my saying the words, 'I love pancakes' a completely inessential thing to my having the thought in the first place? I should be clear; I don't think that this thought refutes anything Wittgenstein is saying. I think the fact that I'm confused by this shows that I haven't yet understood Wittgenstein properly. After all, he spends a long time discussing such 'internal monologues'. So this is one of the areas of the Investigations that I find most confusing. It's also one of the most interesting.
In part II, number xi: "A concept forces itself on one. (This is what you must not forget.)" Is he saying here something akin to Heidegger's claim that the world comes to us in language? It seems similar, at the very least. Later in that same section: "The question whether the dreamer's memory deceives him when he reports the dream after waking cannot arise, unless indeed we introduce a completely new criterion for the report's 'agreeing' with the dream, a criterion which gives us a concept of 'truth' as distinct from 'truthfulness' here." This reminds me of the Rorty talking about how we generally take people's word for it when they report on what they feel/experience "internally," but I can't tell the extent to which Witty is arguing something similar.
I don't have my copy of PI to hand so I can't be sure of the context of that quotation, but I'm pretty sure it comes up in his discussion of aspect perception doesn't it? In which case, I assume that what he's saying there is that when you see the duck-rabbit as a duck, you don't simply see the shape and then interpret the picture as being a picture of a duck. You can't help but see it as a duck. On the other hand, if you see a grid of evenly sized squares, and someone says, 'imagine this is a street map of central Los Angeles', you don't see the grid as a picture of central LA, but rather you interpret it in that way. That's different from what happens where you see the duck-rabbit as a duck. So I take it in that quotation he's making the point that aspect perception can't simply be explained by saying that we interpret the shape differently, because interpretation is something different from seeing-as. I'm afraid I can't help with the comparison to Heidegger though, as I know next to nothing about him. Can you explain Heidegger's ideas to me? I guess Wittgenstein's saying that we have no criterion for deciding whether or not a person's account of their dream is accurate. We might have criteria for whether or not they're simply making it up on the spot though. Like, if someone hesitates or contradicts themselves in telling you about the dream, you might think they were just making it up. But if someone seemed to be sincerely recounting their dream we would have no way whatsoever of deciding whether or not their account was accurate. If we were going to be able to do that we'd have to introduce a new criterion. So the point is that we currently have no such criterion. I guess the point would seem to be that at present, the accuracy of someone's account of their dream isn't a consideration for us. When someone tells us about their dream we don't think about whether or not their account's accurate. We might say, 'well, what does this account tell us about your inner psyche' or something, and do some dream interpretation, but for that purpose the accuracy of the account is irrelevant. I don't know if that's really the point of this passage, but maybe it is. I don't know if that's the same as Rorty. Wittgenstein here seems to be saying that, for dreams, truthfulness of the account isn't relevant. So that's different from saying, 'we take your word for it that the account is truthful'. Wittgenstein's saying that truthfulness of dream accounts just isn't something we consider. I suppose in general Wittgenstein would agree that we tend to take people's reports of their mental states for granted, but that seems like a trivial thing to say. It's no different from saying that we tend to think that people aren't lying to us. Wittgenstein certainly argues that we are very good at working out whether or not people are accurately representing their mental states; it's normally very easy to tell whether or not someone's really in pain, for example. And that, for Wittgenstein, shows that 'pain' is not a private object going on in someone's mind. The difference between a person in pain and a person not in pain is something we can see.
So, for the first one, you see something ("interpretation" and all), and only then, when you talk about it, do you break it down into "seeing" and "interpreting". I don't claim to really understand Heidegger that well, but his claim that the world comes to us in language is that, as children, we are taught to call chalk "chalk". Once we learn this, "chalk" is a part of how we perceive chalk the object. This strikes me as similar, because again the "interpretation" is wrapped up in the seeing. I might ask my existentialism professor, since he has a mancrush on Wittgenstein. I'll have to mull over the second part more later because I read what you wrote but it just sort of slipped out of my grasp. I'll chalk that up to having just woken up.
I'm not quite sure what you mean by this. That does sound sort of like what I think Wittgenstein's saying. But I'm not sure he'd go as far as Heidegger. Wittgenstein discusses aspect perception in the context of images because he's interested in how an image can represent something to us other than simply a flat 2D shape on a page. And that's a very similar question to how some arbitrary shapes (letters and words) can come to have meaning. I think in both cases he says the meaning of the shapes (whether it be an image or a word) comes out in how people respond to them. That's an extremely difficult thought, and not one I fully understand. But he's not talking about all perception in section xi. He's not saying, 'how can this arbitrary selection of colours and shapes that I see when I look around represent a 3D world to me?' So he's not saying that we get a series of sense-data and then we interpret them or see them as something. I don't see have to see what I call 'a piece of chalk' as a piece of chalk. I just see a piece of chalk. In section xi he explicitly says that we don't encounter sense-data that must be interpreted when we look at things. I don't see a grey rectangle and a white rectangle above it: I see a computer. It's not natural to talk about things we see in that way. That's why learning to represent 3D objects in 2D is a skill that must be mastered. And we can see all this if we just pay attention to the way that 'seeing', 'seeing-as', 'interpreting', and so on, are used. The words all have different uses. We can't say 'all seeing is seeing-as', or 'all seeing is interpreting'. So I don't know if that contradicts Heidegger or if I'm just misunderstanding things. But Wittgenstein's particularly focused on how we see images, and why we see them as we do rather than as collections of lines on a page. He's not talking about perception in general. There's also a very interesting link in section xi to the Tractatus and to Wittgenstein's earlier 'picture theory of meaning'. So there, he says that words in a sentence picture reality just as an image does. And an image pictures reality by sharing the same elements as reality in the same pictorial form. In the earlier parts of PI Wittgenstein tears his picture theory of meaning apart. So it makes sense that he would also tear apart his idea of how pictures represent, upon which the picture theory of meaning was based. And section xi is his attempt to deconstruct that earlier thought of how images work, at least I think so. So section xi is very closely related to the discussion of section 1-242. I mean, all the extra stuff in Part II was supposed to slot into Part I somehow, it's just that W couldn't work out how to do it before he died. So it's very closely related. I think once you become aware of the close relation the significance of section xi starts to become clearer. Because on the surface it looks like a weird discussion of a strange perceptual phenomenon. But it's deeper than that. Still, I'm not entirely clear on how it all works.
I've started re-reading this taking copious notes. For 17-18, I'm looking at these in a very anti-essentialist light. It seems like his goal with these is to tear down the notion of an essentialist view of language. Our classifications of words depend upon the goals we have in classifying them (17), and the notion of a "complete" language is bogus, because language adapts to human experience, and there's not essential human experience that's somehow "complete" (18). In 19, he says "to imagine a language is to imagine a life-form." I'm reading this as saying that imagining a language is the same as imagining the uses to which language is put, which tells you about the activities and life of a life-form. What I'm wondering is if he would accept looking at something like frog song as a (primitive, obviously) language, because when we imagine frog song we are imagining the ways in which those frogs attract mates, etc—in short, we're imagining a life form. Or would Witty say that classifying it as a language depends on your goals in describing language (and your personal preference), same as for classifying types of words. (And maybe he would say that it's not a language because it creates confusion if we don't draw the line at communication between conscious beings.)
Yes, I think this is a very accurate assessment of those sections. This is one of those claims that seem obvious to people who don't have a grounding in the philosophy of language, but would seem very controversial to anyone who comes to the Investigations after reading Russell, Frege or the Tractatus. One of my favourite of Wittgenstein's metaphors is at the end of 18: 'Our language may be seen as an ancient city: a maze of little streets and squares, of old and new houses, and of houses with additions from various periods; and this surrounded by a multitude new boroughs with straight regular streets and uniform houses.' He's taking aim at people like Russell and Frege who believe that all language is underpinned by a precise and common logic. They believe that although the surface is confusing, beneath it all lies a perfectly clear system that we all grasp implicitly. Wittgenstein's saying, on the contrary, that Russell and Frege have created a very precise form of language (modern formal logic), but that this is simply an addition to the messy language we had in the first place. Hence, logical analysis of language doesn't reveal to us the 'true meaning' of the things we say. Logical analysis is just stipulation; it isn't discovery. To people who hadn't heard how Wittgenstein's ideas developed in the period between the Tractatus and his death, this would have been a very surprising thing to hear. After all, in the Tractatus Wittgenstein throws his full support behind Russell's theory of descriptions, which is perhaps the paradigm of taking a 'messy' element of common language and analysing it away to discover the 'true logical form' that 'lies beneath'. At 3.323-3.325 of the Tractatus you can see Wittgenstein following the Russellian-Fregean line completely: Confusions arise in philosophy because of the messiness of language. They can be solved by finding the logic that sits beneath the surface. So these passages of the Investigations (and the whole of the first 130 or so sections) are a pretty radical rejection of his old views. Remember that Wittgenstein would have no interest whatsoever in the question, 'is this really a language?' or 'what is language really?' I think we can see from PI 25 that Wittgenstein doesn't come down on either side for those questions. He says that animals 'do not use language - if we except the most primitive forms of language'. But your general assessment of his use of 'life-form' is, I think, spot on. A language makes sense only within a context of use, so in so far as bees, frogs, dolphins, wales, whatever put vocalisations and so on to use in patterns like that then we can say that they use language. He would say that the difference between human life-forms and animal life-forms is not a difference of type; our language is more complex, but it's still fundamentally the same in how it works. That leads to the important conclusion that the human ability to use language doesn't depend upon some supra-animal faculty that humans have. People often say, 'humans can use language because they can reason', and 'reason' has to be explained as something that transcends anything that animals can do. Wittgenstein, I think, would be far more inclined to say: 'We say that humans can reason on a higher level only because they use certain forms of language that other animals do not'. So language use is not grounded in reason; language use is altogether more primitive. And rational activities like doing mathematics just come out of our use of language. This all links to what is one of Wittgenstein's most fundamental insights. The old paradigm of how language works is: 'The meaning of a word determines the use of the word'. Wittgenstein's new paradigm is: 'The use of a word determines its meaning'. If use is prior to meaning, then the idea that we 'grasp the meaning' before we work out how to use a word is wrong. That idea over-intellectualises the phenomenon of human language use.
