Wittgenstein's remarks on private language in PI were partly in relation to Frege's private language arguments, and part of the later Wittgenstein's attempt to reduce Frege's third realm of sense to an interaction between the psychological realm (Frege's second realm) and the physical realm (Frege's first realm). — sime

I’m curious how Wittgenstein can reduce the third realm of sense (presumably language) to an interaction between the second realm of psychology (presumably inner feelings) and the first realm of the physical (presumably the world)?

On the one hand, the second realm of inner feelings drops out of language (the analogy of the beetle) and on the other hand the first realm, any correspondence with the world, also drops out of language (as the meaning of a linguistic expression is its use in language).

Let the concrete principle be “don’t touch a hot stove” and the abstract concept be “touching a hot stove causes pain”. A logical structure can be thought of as synonymous with formal rules.

Where are the formal rules in the abstract concept that touching a hot stove causes pain? — RussellA

You're example does not make sense to me at all. If the concrete principle is “don’t touch a hot stove”, then the principal concept involved is "do not touch", and that itself looks to me like a formal rule. The "a hot stove" is less formal, because criteria is required. I don't see how you even draw a relation between this and "causes pain". this is a common problem with ethical rules, the lack of relation between is and ought.

Wittgenstein definitely didn't adhere to the dogmatic community view (social platonism) that considers meaning to be necessarily social - for "Wittgenstein's manometer" example makes it clear that a diarist's private use of "S" might be turn out to be correlated to rising blood-pressure - a hidden cause of the diarist's behaviour that might be unknown to both the diarist who feels the urge to write "S" and to his community. (Wittgenstein even calls the appearance of a mistake an illusion). Hence Wittgenstein does indeed hint at what i previously called "self-justifying" verbal behaviour - namely verbal behavior that a community considers to be "private" because 1) the behaviour doesn't follow a recognizable existing convention, and 2) the behaviour has no presently known causal explanation. — sime

Could you please explain to me how Frege's "third realm", "sense", is related to this? What would be the sense of Wittgenstein's "S" in this example? The following is how I would interpret this.

From what you've explained, I take "sense" to be a sort of meaning-giving context. If that's the case, then for the public observer, the sense of "S" would appear to be the rising monometer (notice I say manometer, not blood-pressure, because that is what is evident to the observer). For the diarist, the sense would appear to be the inner feeling. Is the true "third realm" then, the blood-pressure itself? For the diarist, the inner feeling is a representation of that presentation (rising blood pressure), and for the observer, the manometer is a representation of that presentation.

From this perspective, we have two opposing, or inverse approaches to "the appearance of a mistake". If both, the outward representation, the manometer, and the inward representation, the feeling, are representations of an intermediary presentation, then we need to allow for the possibility of mistake in either directions of representation.

Suppose we assume that "sense" is the intermediary between speaker and hearer, in this way. To avoid the possibility of mistake, we might designate the true intermediary as the spoken words themselves. This implies that the sense is the words themselves, and that would leave the blood pressure as irrelevant, no longer qualifying as the intermediary. Then the public compares the spoken "S", with the manometer But I think that this would misrepresent the intent of the diarist, who's use of "S" is not as a public presentation, rather it's a private record. Now there is an issue of whether the spoken words (or written) are intended to be an intermediary, or not, and this is an issue in relation to whether the words can actually be the "sense".

If the concrete principle is “don’t touch a hot stove”, then the principal concept involved is "do not touch", and that itself looks to me like a formal rule. — Metaphysician Undercover

I agree that some concepts can be rules, such as “do not touch”, but some concepts are not rules, such as freedom, tree, happiness, colour or more/less.
=========================================

I think concepts are logical structures with formal rules. — Metaphysician Undercover

There is a difference between what a concept is and what a concept does.

I agree that as regards what a concept does, it can be a rule or not be a rule, but as regards what a concept is, I don’t see that a concept is something with a logical structure or formal rules.

Tool 12: Language on holiday

One of Wittgenstein’s most useful tools is noticing when a word has stepped out of its ordinary work. We're using the word, but it’s no longer doing the job it normally does, and this is when language goes on holiday to Bermuda. :grin:

You can see it when a conversation starts producing deep puzzles that never cash out into anything you could actually use, check, learn, correct, or dispute in any concrete way. We keep using familiar words, meaning, rule, proof, evidence, doubt, true, but what give those words traction has disappeared. This results in a kind of philosophical spinning of the wheels. This happens with the majority of discussions in this and other forums.

A simple example from this thread is when someone (sorry @RussellA) takes PI 43, meaning is use, and turns it into a premise in a formal proof, then objects that it’s circular. That move treats meaning as if it must be a detachable item attached to a word, and treats Witt’s reminder as if it were an axiom. The words are still there, but they’re no longer doing their clarifying work, they’re playing a different game while pretending they’re not.

Another example is prove it aimed at hinges. In ordinary inquiry, prove it has a role, it requests reasons inside a system of justification. But when the demand is aimed at what the justification itself depends on, the demand has pulled prove, evidence, and doubt out of the setting where they function.

So, the tool is used when a dispute starts to feel metaphysically impossible, ask what work the key words are supposed to be doing. What would count as a correction, a check, a settled disagreement, a mistake? If there’s no answer, the language has likely gone on a Bermuda holiday, and the right move is to bring it back to where it actually works, viz., everyday use.

When I say everyday use, I don’t mean “whatever the man on the street happens to say.” That’s true in one sense, and false in another.

Witt wants us to start from the words as they live in our actual practices, buying, promising, apologizing, measuring, calculating, arguing, teaching, etc. Not from a philosopher’s purified theory driven version of the word. So yes, he’s pulling us back from abstraction to ordinary life.

But it’s false if you hear “everyday use” as “the average person’s current opinions or sloppy speech is the standard.” Witt isn’t taking a poll. Use includes the practice’s norms, how words are taught, corrected, and applied. Ordinary use includes skilled and technical language games too, medicine, law, mathematics, because those are also ordinary human practices with standards.

So “everyday use” means: the role the word actually plays in our forms of life, including the criteria that make right and wrong intelligible, not “whatever someone says on the street at 2pm.”

It is a grammatical fiction because:

the term 'pain' becomes meaningless if the pain is regarded as a fiction because we can't observe it.
— Fooloso4 — Fooloso4

It is not a grammatical fiction because you cannot observe it. It is grammatical fiction because we misleadingly talk about pain as if we are talking about apples, tree, or tables. It is grammatical fiction because we imagine a private world full of private experiences that can be identified and named that supposedly we can shared with a private language. It is about the limits of language and what we can express with it.

The term “pain” is not meaningless, it serves as a term of expression, or gives us a term to describe the behavior of humans in particular situations.

It seems obvious that you would need bedrock certainties in order to have any certainty. — Sam26

The term 'bedrock' as it is used at PI 217 is not about certainties. It is rather about the limits of justification when it comes to one's reasons for following a rule:

How am I able to follow a rule?” If this is not a question about causes, then it is about the justification for my acting in this way in complying with the rule.
Once I have exhausted the justifications, I have reached bedrock, and my spade is turned. Then I am inclined to say: “This is simply what I do.”

The spade is turned not because one has arrived at some kind of certainty but rather because he cannot dig further in order to uncover something more that will justify what one does in following the rule.

It may turn out that what one simply does is in some cases wrong. One is not following the rule.

A simple example from this thread is when someone (sorry RussellA) takes PI 43, meaning is use, and turns it into a premise in a formal proof, then objects that it’s circular. That move treats meaning as if it must be a detachable item attached to a word, and treats Witt’s reminder as if it were an axiom. — Sam26

Not a problem, I am appreciating the opportunity to learn more about Wittgenstein.

On the one hand is the Augustinian view of ostensive definition, whereby each word corresponds to an object that is its meaning and on the other hand is the Philosophical Investigation (PI) view, whereby the meaning of a word is its use in the language.

I don’t think either is sufficient in itself, although both are necessary.

The Augustinian view cannot cope with fiction and figures of speech.

The PI view cannot cope with an unavoidable circularity. For example, in the expression “this slab is heavy”, the meaning of slab is understood within the context of being heavy, and the meaning of heavy is understood within the context of being a slab. I don’t see how PI gets around this problem.

It seems to me that the Augustinian view is necessary for the meaning of certain core individual words and the PI view is necessary for the combinations of these core words into meaningful propositions.

I don’t see that a concept is something with a logical structure or formal rules. — RussellA

Concepts are logical structure and have formal rules. A human is not a cup. Consciousness is not unconsciousness. A fool is not wise. Socrates is mortal. etc.

It is not a grammatical fiction because you cannot observe it. — Richard B

Right.

The term “pain” is not meaningless — Richard B

It is rendered meaningless if one assumes, as the behaviorist at PI 307 does, that everything except human behavior is a fiction. If everything except human behavior is a fiction then pain is a fiction because pain is not a behavior. The behavior is an expression of the pain.

There is a difference between what a concept is and what a concept does. — RussellA

A concept is not just a word, but it has meanings. When the meaning is stated, it presents the formal rules and logical structure of the concept.

Concepts are logical structure and have formal rules. A human is not a cup. Consciousness is not unconsciousness. A fool is not wise. Socrates is mortal. etc. — Corvus

As there is a difference between what a rock is and what a rock does, there is a difference between what a concept is and what a concept does.

I agree that the concept “Socrates is mortal” has a logical structure, but this is what the concept does.

Another question is, does a concept, in the sense of what it is, have a logical structure.

As there is a difference between what a rock is and what a rock does, there is a difference between what a concept is and what a concept does. — RussellA

Concept doesn't do anything. Humans do things - use concepts in thoughts and statements. Stone is heavier than water. - The concept of stone has the inherent meaning what stone is, which implies and states the clear logic and formal rule.

But even though what a hammer is as a result of what a hammer does, once the hammer has been created, the hammer exists as it is independently of any use, of what it can do. — RussellA

Yes. One can actually use the hammer in various ways that are not what it was designed for. I just wanted to point out that sometimes what something does is intertwined with our idea of what it is.

That there is a rock lying on the floor is not proof that the rock caused the window to break. It may be evidence, but not conclusive evidence.
That someone grimaces is not proof that they are in pain. It may be evidence, but not conclusive evidence.
These seem quite equivalent. — RussellA

There are important ways in which they are quite different.
You said the rock was hidden, but, assuming that it did break the window, it can be revealed. There is nothing that would count as revealing the hidden pain.
There could be evidence that the rock broke the window and it may be beyond reasonable doubt, even if it is not conclusive.

As you say, the connection is not empirical, as is the connection between rain and rainbows. — RussellA

Applying the word "evidence" glosses over the fact that the evidence for the rock breaking the window is of a different kind from the grimace as evidence of pain. I can show you the splinters of glass and the rock beside each other. I cannot show you the grimace and the pain next to each other. On the contrary, showing you the grimace is showing you the pain. But I grant you that the grimace is defeasible.

The fact that “ouch” is in quotation marks shows that it is part of language. — RussellA

Lewis Carroll wrote "'Twas slithy and the mome raths outgrabe". Does the fact that they are in quotation marks show that they are part of language or does the fact that they are meaningless show that they are not? Even if you think that "ouch" is part of language, the fact that it is in quotation marks shows that it is mentioned, not used.

“2+2=4” refers to 2+2=4 — RussellA

What I actually asked is 'What does "plus" as in "2+2=4" refer to?'

“The present king of France” refers to the present king of France — RussellA

There is no king of France, so it refers to no-one - that is does not refer to anyone.

“Nothing” refers to nothing. — RussellA

Quite. So not all words refer.

This is because they do not have such "concepts" when they learn how to use those words. That is the point Wittgenstein made with "game", we all use the word without having any specific concept of game. I believe he takes this idea further in On Certainty. Knowing how to use a word doesn't indicate that the person using it has a concept of the word. — Metaphysician Undercover

Well, I had the impression that Wittgenstein's point about "game" was that there could not be a single definition (formal rule) that would be the basis of a concept. "Game" is applied to a very wide range of games, but he explains his meaning by means of the metaphor. There is no single thread that runs through the whole of a rope; its strength is made by a number of distinct threads which interweave and overlap. Better known, perhaps, is his metaphor of "family likenesses" which connect member of a family. Similarly, there is no single likeness that connects all games; but there are a number of different likenesses that interweave and overlap to connect them.

I think concepts are logical structures with formal rules. — Metaphysician Undercover

I think we understand that we use the word differently; there doesn't seem to be any point about that. I think, though, that people mostly assume that if you can use a word competently, you can articulate a definition of it - and vice versa. But those are different skills.

It seems obvious that you would need bedrock certainties in order to have any certainty.
— Sam26

The term 'bedrock' as it is used at PI 217 is not about certainties. It is rather about the limits of justification when it comes to one's reasons for following a rule:

How am I able to follow a rule?” If this is not a question about causes, then it is about the justification for my acting in this way in complying with the rule.
Once I have exhausted the justifications, I have reached bedrock, and my spade is turned. Then I am inclined to say: “This is simply what I do.”

The spade is turned not because one has arrived at some kind of certainty but rather because he cannot dig further in order to uncover something more that will justify what one does in following the rule.

It may turn out that what one simply does is in some cases wrong. One is not following the rule. — Fooloso4

You’re right about the PI 217 passage, bedrock there is about the end of justification, my spade is turned because justificatory digging ended, not because I’ve reached epistemic certainty or anything like absolute certainty.

But when I said “bedrock certainties,” I wasn’t claiming PI 217 is literally a passage about “certainties.” I was using my own terminology, I’m equating bedrock certainties with hinge certainty, viz., what stands fast and what makes doubt, checking, and justification possible. Structurally, PI 217 is marking the same kind of stopping point that On Certainty later treats as hinge certainty, even if the emphasis and vocabulary differ.

Of course it can turn out that “this is simply what I do” is wrong in the sense that I’m not following the intended rule. But this presupposes a practice with standards of correct and incorrect. My claim isn’t “bedrock guarantees I’m right.” It’s that some things have to stand fast in the practice for the distinction between seems right and is right to be intelligible, and that’s what I mean by bedrock certainties / hinge certainty.

One more clarification: I’m not always trying to give a perfectly “pure” Wittgenstein exegesis. I do aim to be faithful where it matters, but I’m also extending some of his insights into my own thinking on epistemology, especially with my four senses of certainty. If a term or connection doesn’t line up one to one with Wittgenstein’s phrasing, that’s sometimes deliberate. The question is whether the extension is illuminating and coherent, not whether it matches every textual contour.

Just to be clear I use certainty in the following 4 ways: Subjective certainty, epistemic certainty, hinge certainty, and absolute certainty.

A simple example from this thread is when someone (sorry RussellA) takes PI 43, meaning is use, and turns it into a premise in a formal proof, then objects that it’s circular. That move treats meaning as if it must be a detachable item attached to a word, and treats Witt’s reminder as if it were an axiom.
— Sam26

Not a problem, I am appreciating the opportunity to learn more about Wittgenstein.

On the one hand is the Augustinian view of ostensive definition, whereby each word corresponds to an object that is its meaning and on the other hand is the Philosophical Investigation (PI) view, whereby the meaning of a word is its use in the language.

I don’t think either is sufficient in itself, although both are necessary.

The Augustinian view cannot cope with fiction and figures of speech.

The PI view cannot cope with an unavoidable circularity. For example, in the expression “this slab is heavy”, the meaning of slab is understood within the context of being heavy, and the meaning of heavy is understood within the context of being a slab. I don’t see how PI gets around this problem.

It seems to me that the Augustinian view is necessary for the meaning of certain core individual words and the PI view is necessary for the combinations of these core words into meaningful propositions. — RussellA

I think we agree the Augustinian picture captures something, and Witt grants that. Words are sometimes taught by pointing. The mistake is to treat “word = object” as the general model, as if every word were a name and understanding were always just pairing a sound with a thing.

On your circularity example, I don’t think PI is stuck. “This slab is heavy” isn’t learned by defining slab through heavy and heavy through slab. It’s learned in practice, i.e., you’re shown slabs, you handle them, you hear slab, you learn “heavy” by contrasts (heavy/light), you’re corrected, and you go on. The words don’t get their meaning from mutual definition inside the sentence, they get it from the training and the roles they play.

I wouldn’t split it into “core words need Augustinian meaning, then PI handles propositions.” Ostension is one way of teaching, but it only works within a wider practice, and that wider use is what stabilizes meaning.

Tool 13: Forms of Life

“Form of life” is Witt’s name for the shared human background that makes language possible. It’s not a theory, and it’s not a spooky foundation. It’s the fact that we’re creatures who react, act, train, and correct in very similar ways; we learn by imitation and instruction, we respond to pain with concern, we count, measure, promise, doubt, argue, etc. Language isn’t built on private meanings; it grows inside these activities.

This helps with a recurring confusion in Wittgenstein discussions, viz., form of life is not “whatever the community votes for,” and it’s not a central authority that legislates meanings. It’s the lived practice that makes right and wrong use intelligible in the first place. The norms are not outside the practice, but they’re not some mere whim either. We observe them in training, correction, agreement in judgment, and in what counts as going on in the right way.

When Wittgenstein says that explanations come to an end, he isn’t saying, “anything goes.” He’s saying that at some point our justifications bottom out in what we do, how we’re trained, what we count as a correction, what stands fast. That is the form of life. And it’s why philosophical demands for a deeper foundation often misfire; they’re asking for a kind of justification that only makes sense once the background is in place.

If you want a simplification, forms of life are the shared human activities that give our language games their home, and make meaning, rule following, and correction possible.

I agree that some concepts can be rules, such as “do not touch”, but some concepts are not rules, such as freedom, tree, happiness, colour or more/less.
=========================================
I think concepts are logical structures with formal rules.
— Metaphysician Undercover

There is a difference between what a concept is and what a concept does.

I agree that as regards what a concept does, it can be a rule or not be a rule, but as regards what a concept is, I don’t see that a concept is something with a logical structure or formal rules. — RussellA

I still don't understand what you could mean by "concept". Sure we can all use words such as those of your examples, "freedom, tree, happiness, colour or more/less" but unless there are definite rules of usage, how can you assume that there is any concept involved with these words?

Take "freedom" for example. We can all use the word in a variety of contexts, each context having a different meaning. Why would you think that being able to use the word implies that there is such a thing as a concept of freedom. On the other hand, if we stipulated well defined rules of usage, like we do with mathematical concepts, then we'd have the basis for the claim that this constitutes a concept of freedom.

Well, I had the impression that Wittgenstein's point about "game" was that there could not be a single definition (formal rule) that would be the basis of a concept. "Game" is applied to a very wide range of games, but he explains his meaning by means of the metaphor. There is no single thread that runs through the whole of a rope; its strength is made by a number of distinct threads which interweave and overlap. Better known, perhaps, is his metaphor of "family likenesses" which connect member of a family. Similarly, there is no single likeness that connects all games; but there are a number of different likenesses that interweave and overlap to connect them. — Ludwig V

So doesn't that indicate to you that there is no concept of game? There is the single word, like the single rope, but that is composed of may different fibres, ways of usage. There is no single meaning therefore no single concept. Overlapping, distinct but similar uses, which are analogous to "family likenesses" does not constitute what we commonly understand as "a concept".

However, he then goes on to explain how one could dictate boundaries of usage, for a specific purpose.
This, applying rules to limit usage, boundaries, I conclude is the production of a concept. For example, the concept of "triangle" is the explicit rule of plane figure with three sides and angles.

I think we understand that we use the word differently; there doesn't seem to be any point about that. — Ludwig V

How could you say that there is a "concept" involved if we each use the word differently? Doesn't the very essence of what it means to be "a concept" indicate that the word must be used in the same way? If someone was calling a round plane figure a triangle, and someone else called a rectangular plane figure a triangle, how could we claim that there is a concept of "triangle"?

but unless there are definite rules of usage, how can you assume that there is any concept involved with these words? — Metaphysician Undercover

How would rules conjure a concept? It's probably that both rules and concepts are elements of post hoc analysis of language.