Since we're only getting loose direction in proceeding and folks have moved on a bit past what I commented on before, I'll do my comments on the next 10 sections, through 40.

31

What he's describing here, and in surrounding sections, is someone who already knows language to some extent who is asking for ostensive demonstrations of things they don't know or aren't sure of. That's certainly a possibility--that's often what's the case when we're dealing with someone who asks for or could use an ostensive definition of something. Say when we're dealing with other adults, school-aged children, etc.--as obviously they're already going to know some language as they enter school. This is what we usually are dealing with.

However, this is by no means an argument to the effect that it's not logically possible to learn a language solely by ostension.

My only other comment on this section is that I'm a bit baffled by this:"One can also imagine someone's having learnt the game without ever learning or formulating rules." If we're talking about chess a la anything like what conventionally counts as knowing chess, I don't think Wittgenstein's claim there makes any sense.

32

If Wittgenstein is only talking about the passage he quoted I don't think this comment is warranted, but I don't suppose Wittgenstein only has in mind the brief passage he quotes at the start of the book.

At any rate, on my view children definitely think prior to learning language. They just think non-linguistically prior to learning language.

33

The first part of that, re "It is not true that you must already be master of a language in order to understand an ostensive definition: all you need—of course!—is to know or guess what the person giving the explanation is pointing to. That is, whether for example to the shape of the object, or to its colour, or to its number, and so on." is what I say. Re his comment immediately following this, it's as if Wittgenstein just quoted (via the hypothetical) "know or guess" but then immediately ignored that he included the "or guess" part.

The way that guesses are refined is via subsequent ostensive demonstrations. The student refines their abstraction in response to that.

"Suppose someone points to a vase and says 'Look at that marvellous blue—the shape isn't the point.'—Or: 'Look at the marvellous shape—the colour doesn't matter.'" You're not going to say anything like that to a baby or toddler to whom you're doing your first ostensive demonstrations of language. That's way too complicated for what they can understand. You're not going to speak a bunch a words in a row to them like that, unless you're a horrible teacher. It seems like Wittgenstein is very drifty in these comments re just what sort of student he has in mind for this argument.

"But it isn't these things themselves that make us say someone is attending to the shape, the
colour, and so on." It depends on who we're talking about. Some people definitely would say someone is "attending to the shape" etc. due to those things themselves.

At any rate, this last section of 33 seems like it's going to matter more for following sections....

34

Yeah, no demonstration is going to be deterministic with respect to the way the student parses it. That's just how language works in general. It's a truism about language.

35

"For example, following the outline with one's finger or with one's eyes as one points" -- That still can't make anything epistemically deterministic re how the student is going to parse anything in their head. It's not possible to make that epistemically deterministic.

The end part of this section seems rambly to me. If we're talking about someone who already knows language a fair amount, sure we can clarify what we're pointing to in many different ways. What of it?

36

Not sure what in the world he's talking about in this section. It reads like something incoherent, but it's so brief and sketchy. Maybe he'll explain it better in a bit.

37

"What is the relation between name and thing named?"

There are a bunch of relations, both subjective and objective. The ones he gives as examples are fine.

38

"This" and "that" are conventionally basically verbal forms of pointing or gesturing towards something.

"Yet, strange to say, the word 'this' has been called the only genuine name; so that anything else we call a name was one only in an inexact, approximate sense."--no memory of who would have said that, but it's indeed a ridiculous thing to say.

"Can I say 'bububu' and mean 'If it doesn't rain I shall go for a
walk'? --Of course you can do that. You can use any word, phrase, etc. to mean whatever you like.

"t is only in a language that I can mean something by something." ---only in the sense that saying "bububu" and meaning "If it doesn't rain I shall go for a walk" thus is sufficient to count as language. In other words, we'd just tautologically be saying that meaning something by something, where that involves words, counts as a language.

"To mean" is like "to imagine" in that it's picking out something that individuals are doing mentally.

The pdf version of the book has a lot of glitches by the way. There's something else missing in this section (and there were a couple things missing or a bit garbled in the preceding sections that I didn't comment on), anyway, continuing ....

You can give ostensive explanations of "that" and "this," sure. Since they're more abstract, that would just take more work.

Ummm . . . not sure what the heck he's saying in the "queer connection" part, either.

39

The idea that the meaning of a term is its extension is simply a dumb theory that leads to a lot of errors (as he demonstrates out in this section).

The last two sentences of this section do not make much sense to me:"So the word 'Excalibur' must disappear when the sense is analysed and its place be taken by words which name simples. It will be reasonable to call these words the real names."

40

Just reiterating that saying that the meaning of a term is its extension is a dumb theory that leads to mistakes. I agree with that. (If that's indeed where he's going to continue to go with this.)

....okay, I'll leave it at that until other folks are moving on more.

My only other comment on this section is that I'm a bit baffled by this:"One can also imagine someone's having learnt the game without ever learning or formulating rules." If we're talking about chess a la anything like what conventionally counts as knowing chess, I don't think Wittgenstein's claim there makes any sense. — Terrapin Station

We just finished going over this, starting here: https://thephilosophyforum.com/discussion/comment/232532

'why and how does ostensive explanation pick out this rather than that?' — StreetlightX

What it picks out, for the learner, is whatever the learner takes it to pick out. Again, that can be refined via numerous examples, which help the learner refine an abstraction to fit all of the examples (if the learner can manage to do so, assuming the teacher is doing something coherent, etc.)

Yeah, I agree with your objections about that. Sam's "cup" example didn't cut it, because using "cup" in sentences that people don't have a problem with is nothing like knowing how to play chess, where we mean anything like the conventional sense of what it is to know how to play chess. There's just no way to have learned how to play chess without knowing the rules of playing chess, since each piece has such specific ways it can move, with all of those being different, not being intuitive, etc.

For example, if you don't know that a pawn can only move straight forward except when capturing an opponent piece, and only one space at a time forward except on the first move, then it wouldn't make sense to say, with any of the conventional connotations, that "you've learned chess," But if you know that sort of stuff, you've learned the rules.

To make sense of saying, "You can have learned how to play chess without having learned the rules," someone would need to give an example of how someone would play chess, where they wouldn't know the rules, but where we'd agree, "Yep, they've learned chess." They'd need to give some examples of how that person would proceed. Which chess pieces would they move on some of their turns and how would they move them?

Yeah, I agree with your objections about that. Sam's "cup" example didn't cut it, because using "cup" in sentences that people don't have a problem with is nothing like knowing how to play chess, where we mean anything like the conventional sense of what it is to know how to play chess. There's just no way to have learned how to play chess without knowing the rules of playing chess, since each piece has such specific ways it can move, with all of those being different, not being intuitive, etc. — Terrapin Station

What about something like the way that AI learns, just by observing? I don't think it will do anything outside of what has been observed, but it amasses, and processes such a huge volume of possibilities just from observing. Isn't that how those Go playing computers work? You can't say that it has learned the rules, yet it won't do anything outside the rules because it hasn't learned anything outside the rules, as a possible move.

For example, if you don't know that a pawn can only move straight forward except when capturing an opponent piece, and only one space at a time forward except on the first move, then it wouldn't make sense to say, with any of the conventional connotations, that "you've learned chess," But if you know that sort of stuff, you've learned the rules. — Terrapin Station

By observing, you would know that these are the only possible moves that a pawn could make, without having learned the rules. It's like inductive reasoning, by observing the same thing over and over again, you come to conclude that this is the only possibility, without first learning the rule.

you would know that these are the only possible moves that a pawn could make, without having learned the rules. — Metaphysician Undercover

But "these are the only possible moves the pawn can make" and so on are the rules.

I have no doubt that the rules of Go/chess are programmed into a Go/chess-learning AI. A specification of the rules is so tiny compared to the set of strategies the AI develops that the opportunity cost of specifying them is negligible, and the cost of not specifying them is wasting time in the search process in identifying rules that could have been known at the start.

No doubt an AI could learn most of the rules by observation (not all of them. There are some rules that are almost never used, such as changing a pawn into a bishop when it reaches the other end), but it would just waste processing time to make it do so rather than focusing on strategy and tactics.

Streetlight is doing a good job of staying on the topic at hand. My comments were meant as generalizations, but what Streetlight is doing is what should be done, especially if you want to follow the steps in Wittgenstein's thinking.

It's a matter of knowing possible moves, rather than a matter of knowing rules. The two are distinct. For instance, I know that these possible actions, walk, bike, drive, take the bus, or take the train, will get me to work. When I want to go to work I simply chose one of these actions, depending on the circumstances. I do not know any rule which states that if I want to get to work I must chose one of these actions, or any such thing. So when I choose a means of getting to work, I am simply choosing a means of getting to work, from the possibilities that I know. I am not following any rule.

Hey Sam26, I agree with you (fancy that), SLX is doing a fine job.

We agree, damn miracles will never cease, there is a God. :wink:

Either that or SLX is one hell of a mediator.

It doesn't necessarily mean that you have learned rules or that rules have been stipulated in some way prior to the learning of the game. — Sam26

Would it be too simple to observe that we check our grammar by trying out phrases and knowing they are wrong or right without checking on the basis of another authority?

It's much easier when there's a common point of reference to work between everyone with :)

§34

In §34, Witty continues to pose an objection to himself, one that goes one step further than the failed self-critique of §33. Recall that in §33, the appeal to 'attention' was criticised for being too equivocal a notion to answer the question of how ostensive explanations are individuated ('attention' can work in many different ways, and so cannot function to explain how ostensive explanation works to pick out one kind of thing and not another kind of thing). §34 tries to address this counter-objection by turning instead to 'characteristic experiences', which, unlike 'attention', are always the 'same' - that is, univocal, rather than equivocal.

This is meant to correct the 'defect' of the (self-critical) argument in §33, by having the same experience always undergird the individuation of ostensive explanation (rather than the different - and hence inadequate - modes of attention offered in §33). However, Witty notes that this too is not good enough, because even if the 'same experience' always accompanies an ostensive explanation, there is no guarantee that such an experience will itself always be interpreted in the same way:

§34: "Can’t his hearer still interpret the explanation differently, even though he sees the other’s eyes following the contour, and even though he feels what the other feels?".

Univocity of experience gives way once again to equivocity (diversity) of interpretation. In place of experience then, Wittgenstein once again emphasizes the use of words in an ostensive explanation - a use which cannot be reduced to the experiences which merely accompany 'the giving and hearing of an explanation'.

Historical note: the above contains the seed of the idea - elaborated subsequently by Wilfrid Sellars - regarding the critique of the 'myth of the given': roughly, the idea that sheer experience (or 'sensation') alone cannot dictate meaning in any straightforward manner, and that meaning is irreducibly normative (governed by certain standards - which Witty will conceptualize in terms of grammar or 'criteria'). This theme of an attack on 'the given' - not named as such - is found all through the PI, perhaps most crucially in the upcoming attack on 'private language'.

Ah, the "myth of the given," Fafner and I argued over this about 3 years ago in the other forum.

Trying to define the myth of the given in earshot of any semi-qualified philosophy grad student or professor is usually occasion for a frustrating, long, and pointed confrontation.

Anyway, I don't want to derail the conversation but I do want to push back a tad on the notion that Sellars inherits or continues Wittgenstein's legacy in any significant way. *push*

It's a matter of knowing possible moves, rather than a matter of knowing rules. The two are distinct. For instance, I know that these possible actions, walk, bike, drive, take the bus, or take the train, will get me to work. When I want to go to work I simply chose one of these actions, depending on the circumstances. I do not know any rule which states that if I want to get to work I must chose one of these actions, or any such thing. So when I choose a means of getting to work, I am simply choosing a means of getting to work, from the possibilities that I know. I am not following any rule. — Metaphysician Undercover

The analogy doesn't work because there aren't any "rules of getting to work" akin to the rules of chess, especially with respect to what people have in mind when they say that someone has learned chess.

Anyway, I don't want to derail the conversation but I do want to push back a tad on the notion that Sellars inherits or continues Wittgenstein's legacy in any significant way. *push* — John Doe

I will definitely agree with this.

32. Augustin's description was "as if the child could already think, only not yet speak".

Wittgenstein damming the notion of a private language.

The analogy doesn't work because there aren't any "rules of getting to work" akin to the rules of chess, especially with respect to what people have in mind when they say that someone has learned chess. — Terrapin Station

That's exactly the point, I can learn the possible moves without learning rules. It wasn't an analogy, I was just demonstrating that learning the possible moves is not the same thing as learning rules. You had suggested that it was. Now you seem to be in agreement that it is not the same thing.

32. Augustin's description was "as if the child could already think, only not yet speak".

Wittgenstein damming the notion of a private language. — Banno

What Wittgenstein is discussing here, is the fact that learning a language when one already knows a language is quite different from learning a first language. And, he interprets Augustine's description of learning a first language as similar to how one would learn a secondary language. So he says that Augustine's description of ostensive learning requires that the child already knows how to think in the sense of "talk to itself", i.e. think using words. Clearly this description of ostension is incorrect, because the child hasn't yet learned how to use words, so it cannot think using words, to figure out the meaning of the words being ostensively demonstrated.

That's exactly the point, I can learn the possible moves without learning rules — Metaphysician Undercover

The problem is that if you know the possible moves in a manner to prompt "He's learned chess," then you know the rules . Are you simply saying that you don't realize that you know the rules per se, even though you do?

I am saying that he would not know the rules, even though he learned how to play chess. He'd know possible moves, and knowing possible moves, and being able to choose from them, would allow him to play the game. And, as I demonstrated, knowing possible moves is not the same thing as knowing rules. Possible moves are particulars, rules are general. Knowing the particular possibilities is not the same thing as knowing the general "rule", because the latter requires an act of inductive reasoning. So you are creating a false representation, claiming a falsity, when you insist that knowing the particular possibilities is the same thing as knowing the rules.

Knowing the particular possibilities is not the same thing as knowing the general "rule", because the latter requires an act of inductive reasoning. — Metaphysician Undercover

What would the difference be there when we're talking about chess? How do chess rules require inductive reasoning where knowing the possible moves does not?

Wittgenstein damming the notion of a private language. — Banno

Consider that Wittgenstein has been arguing that one must already know some sort of language-games prior to being able to learn words ostensively. These language-games might just be private, so this is clearly not yet damning the notion of private language.

What would the difference be there when we're talking about chess? How do chess rules require inductive reasoning where knowing the possible moves does not? — Terrapin Station

I told you, rules are generalities, individual possibilities are particulars. So for instance the rule states that the bishop may only move diagonally. That's something general. The person who knows the possible moves, but doesn't know the rules, will look at the position of the bishop and know every possible square that the bishop can move to, without knowing that rule. This person might not even know what "diagonally" means. You of course will ask how could the person know these possibilities without knowing the rule, but we went trough this already, it is proposed that the person may learn this from observation.

So first, are we imagining people saying, "Joe has learned chess," as a response simply to watching Joe play chess?

Re this:

You of course will ask how could the person know these possibilities without knowing the rule, — Metaphysician Undercover

Actually, I'm trying to figure out how "Joe memorized every square the bishop can move to" (ignoring the ridiculous of them doing that without mentally forming an abstraction of it) is different, functionally, than "Joe has learned that the bishop can move only diagonally."

Actually, I'm trying to figure out how "Joe memorized every square the bishop can move to" (ignoring the ridiculous of them doing that without mentally forming an abstraction of it) is different, functionally, than "Joe has learned that the bishop can move only diagonally." — Terrapin Station

Functionally? Of course it's the same functionally. Joe can play the game. We're not discussing whether it's different functionally, we agree that it's the same functionally. It's two distinct ways of knowing, which may serve the same function.

So you're just saying that it's different in how Joe is thinking about it?

My only other comment on this section is that I'm a bit baffled by this:"One can also imagine someone's having learnt the game without ever learning or formulating rules." If we're talking about chess a la anything like what conventionally counts as knowing chess, I don't think Wittgenstein's claim there makes any sense. — Terrapin Station

He only suggests that one can imagine it, but even if you can't imagine it, it's a minor detail. The point of this section is in relation to use of the phrase "This is the king". This phrase might be used when explaining to someone who otherwise knows all the rules, or who otherwise knows how to play chess (e.g. only via mimicking the behaviours of others), which piece or shape represents the king. "This is the king" might also be used in the case of teaching someone how to play chess, but Wittgenstein stresses that this is possible only:

if the learner already 'knows what a piece in a game is'. That is, if he has already played other games, or has watched other people playing 'and understood'—and similar things. Further, only under these conditions will he be able to ask relevantly in the course of learning the game: "What do you call this?"—that is, this piece in a game.
We may say: only someone who already knows how to do something with it can significantly ask a name.

This section (31) relates back to his previous remarks on ostensive definition, and also leads into his brilliant summary at section 32 of the problem he has identified in Augustine's description of learning "human language": Augustine describes it "as if the child...already had a language only not this one". Wittgenstein spells out his position in section 33: "you must already be master of a language in order to understand an ostensive definition".

This leads him into a defence of this claim and an attack on the anticipated argument/assumption that ostension (i.e. pointing or "attending to") is only mental, intentional, or even spiritual.

I agree that it may be difficult to imagine someone knowing how to play chess without having learnt the rules of chess, but consider that children learn how to speak before they learn any rules of language, by mimicking the behaviours of others. Also, bear in mind that Wittgenstein is attacking certain prevalent philosophical assumptions of his time, including those of his Tractatus, that language is exclusively a private, mental phenomenon. He is trying to remind us that language is instead (or also?) a shared, public, cultural and behavioural phenomenon.

So you're just saying that it's different in how Joe is thinking about it? — Terrapin Station

Yes, how Joe is thinking about it is different, because how Joe learnt it is different. He didn't learn how to play the game by learning the rules of the game, he learnt how to play the game by observing the play of the game.