Reading Wittgenstein's On Certainty as a Whole: An Interpretive Picture

Metaphysics & Epistemology
41

I can’t see your point Ludwig. My point was that certainty and doubt do not share the same grammar, and therefore cannot be contrasted in that way. Certainty is produced within the grammar of a single language-game, the requirement of consistency being essential. On the other hand, doubt is produced by words serving different, and inconsistent purposes (meaning) between one language-game and another. Certainty is the product of a single game, doubt the product of a multitude of games.

So to take your analogy, that “winning” is different in poker and racing, we could also say that “certainty” is different in science and ethics. However, just like “winning” in one sense is specific to one game, and “winning” in the other sense is specific to the other game, “certainty” in one sense is specific to that field, and “certainty” in the other sense is specific to the other field.

Now “certainty” as a general thing maintains the character I assigned, of being produced by adhering to the grammar of the field, regardless of the field. Likewise, “winning” in general is produced by following the rules of the specific game regardless of the game.

Doubt, as I described is produced from the reality of a multitude of different language-games. The skeptic says that if “winning” is different in poker and racing, then there is really no such thing as “winning”. It is not a well defined word, and skepticism about “winning” is warranted. Likewise, if “certainty” means something different in science from what it means in ethics, then there is no such thing as “certainty” itself, and doubt is justified.

Wittgenstein’s example is “existence”. What would justify me not doubting the existence of my hands? Clearly, even within the study of philosophy, “existence” has a multitude of meanings. Some say only ideas exist, some say only material objects exist, and there is a multitude of other claims about “existence”. So a clear definition of “existence” is required as a first step toward justifying not doubting the existence of my hands.

42

Would you make a distinction between linguistic propositions treated as formal statements of the form S is P, and a proposition treated as a move in a language game? I read Wittgenstein as arguing that the former are confused ways of thinking about how language works, and the latter reveal what words are actually doing as we use them. I would also avoid treating hinges as if they functioned outside the space where we give grounds. To me hinges are not outside the space of language games and languages games are not outside the space of moves within them.

We are tempted to treat hinges and the rules of language games as conditions of possibility which sit in the background imposing overarching schemes of meaningful organization on actions. But a fundamental feature of the application of a rule or our functioning within the purview of a hinge is that every action involves an element of creative improvisation.No rules or criteria can tell us how to go on. When we make a move within a language game the move subtly redefines the sense of the game, and subtly redefines the sense of the hinge which ties together a system of language games.

That’s why I think it’s better to understand rules and hinges as dimensions of each new action, verbal, gestural or otherwise, that we make, rather than as before, outside or encompassing of them. The only way we can know how a hinge functions is to discover freshly how it is used in a present action. The use of a word or gesture tells more about what the hinge is doing than the hinge can tell us about what the word is doing.

43

“Losing” is simply “not winning”. Winning is well defined in each game that it occurs in. Therefore Losing is also well defined in each game it occurs in. So scepticism about losing is not warranted.
There are many other terms that a like these two - they (and the related opposite) are well defined in each context that they occur in. The attempt to use them outside any context is empty.

You give the game away by suddenly introducing this new concept of “certainty itself”. What does that mean, exactly? By doing so in this way, without any explanation of what you mean, you leave your reader to supply a meaning - or not.

In fact, “winning” and “losing” may have different meanings in each game, but there is a common thread. Roughly winning is what players are expected to try for; if they win, they succeed. If they do not even try, they are not even playing the game. So there is a role in those games that have the concept and this role can be identified across games. It is the detail of what counts as winning that changes. Something similar can be said of “certainty” (objective sense) and “doubt”.

Your sense of doubt has the apparent strength of being proof against any answer. But the weakness of not saying anything concrete.

44

By this definition, there is a multitude of losers, second place, third place, they’re all losers. But this is not an accurate definition of losing, so your analogy fails.

That is the point, and why doubt is justified. Wittgenstein’s example is “existence”. Even within the discipline of philosophy “existence” has different meanings, depending on if you follow idealism, materialism, etc.. Therefore until we settle on a definition of “existence”, we’d be fools not to doubt the “existence” of our hands. So doubt is fundamental to philosophers.

Again, your analogy doesn’t work. On the one hand you are repeating what I said about “certainty”, it can be identified across games, as something sought produced, and defined within the context of a specific game, just like winning.

But you are not paying attention to what I said about doubt. It is not proper to any specific game, as winning is, but it is a feature of the fact that a multitude of different games use the same words in different way. Try this to help you understand. Suppose a word is a rule. Now imagine that a number of games use the very same rules, but expect you to interpret the very same rules in different ways depending on the game. That is cause for doubt.

So winning and certainty are similar in your analogy, but doubt is dissimilar. And doubt doesn’t even compare with losing because the skeptic is not trying to win, or even to play the game, but to question the required interpretation of the rules. Maybe the doubter could be portrayed as playing a meta-game.

I think that’s right, it seems to be exactly what doubt is. As a suspension of judgement, it may be applied against anything that one is supposed to judge, but at the same time it provides absolutely nothing concrete. That’s why it can’t be placed in the same category as certainty. Doubt doesn’t provide a weak foundation in contrast to certainty’s strong foundation, it doesn’t provide a foundation at all, the intent is to deconstruct.

45

No, it doesn’t. A given form of words with two or more different interpretations is not the same rule.

That presupposes that all those systems are equally legitimate. Someone might take that position and I would have some sympathy with it. Given that they all have different definition of “exist”, there’s a case for saying that on this view, they are simply talking past each other. But most philosophers think that they are not all equally legitimate.

Yes, it is possible to argue that the person who comes second or third in a race has not lost. But in many games there are only two players (or sides), so the definition works there. In other cases, we might explain that the binary win/lose is not appropriate and develop a more complex structure. That doesn’t change anything.

If idealism and materialism have different definitions of existence, they are simply talking past each other. Your doubt is not necessary, since what can be doubted is differently defined in different games.
Besides, part of the point of all this is the idea that philosophy gets itself into endless pointless debate by trying to think and talk in that way.You can’t assume that philsophical doubt is meaninful. You are begging the question.
Yes philosophy does change if you abandon Cartesian doubt. It changes for the better, since meaningless debates are identified and abandoned.

It is certainly possible to consider changing the rules of a game, or creating a new game by altering some of the rules of an existing game. But truth and falsity don’t play any part in that debate. The relevant criteria are not those of the game. Call that a meta-game if you like.

It depends what you mean by “deconstruct”. Once you have defined that, you will have developed a new game, and people will decide whether it is playable or interesting.
I’m wondering whether we would do better to think of the opposite of certainty as uncertainty and leave you with doubt as this vacuous kind of suspension of judgement.

46

I think that this little digression has only demonstrated that you and I do not agree on what “doubt” means in this context. I think that’s what Wittgenstein expected when he said at 24: “Hence, that we should first have to ask: what would such a doubt be like?, and don’t understand this straight off”.

We should probably not pretend to understand this straight off.

47

Post 11 OC 28-30

Text

OC 28: “What is ‘learning a rule’? — This. What is ‘making a mistake in applying it’? — This. And what is pointed to here is something indeterminate.”

OC 29: “Practice in the use of the rule also shews what is a mistake in its employment.”

OC 30: “When someone has made sure of something, he says: ‘Yes, the calculation is right’, but he did not infer that from his condition of certainty. One does not infer how things are from one’s own certainty. Certainty is as it were a tone of voice in which one declares how things are, but one does not infer from the tone of voice that one is justified.”

Remarks:

OC 28 and 29 should be read together because they complete the line of thought from OC 27. Witt asks what learning a rule is, and what making a mistake in applying it is, and his answer in both cases is simply “This,” pointing to something indeterminate. OC 29 then adds that practice in the use of the rule shows what counts as a mistake. These two remarks are doing something seemingly straightforward. Witt is saying that the concepts of rule-following and mistake-making can’t be fully defined in advance. You learn a rule by being trained in its use, and you learn what a mistake is by being corrected within that practice. There’s no prior definition that settles the matter independently of the practice itself. The word “indeterminate” is important. It doesn’t mean vague or arbitrary. It means that what counts as following a rule and what counts as a mistake can only be shown within the practice, not stated in a formula that stands outside it. This is continuous with OC 27’s point about normal circumstances. Just as we recognize normal circumstances without being able to precisely describe them, we recognize correct and incorrect applications of a rule without being able to ground that recognition in a further rule. The practice is where the buck stops. And for epistemology this means that our justificatory practices can’t be grounded in anything more fundamental than the practices themselves. The training, the correction, the shared recognition of what counts as getting it right, these are the bedrock. This is what I’ve been calling the hinge level, the foundation that makes epistemic moves possible without itself being an epistemic move. This seems to be a point of confusion for people.

OC 30 shifts the focus back to certainty and makes one of the most vivid points in the text. Witt says that when someone has made sure of something, he says “Yes, the calculation is right,” but he did not infer that from his condition of certainty (his mental state). Then something to remember, i.e., “Certainty is as it were a tone of voice in which one declares how things are, but one does not infer from the tone of voice [Moore’s announcement, “Here’s is a hand” comes to mind] that one is justified.” This is a direct strike at subjective certainty. The feeling of being certain, the tone of confidence and conviction, is real as a psychological phenomenon. Witt isn’t denying that it exists. But he’s showing that it does no epistemic work. When I check a calculation and say “yes, that’s right,” my justification comes from the checking, from the practice of calculation, not from how certain I feel about the result. The feeling of certainty accompanies the judgment but doesn’t ground it. Moore’s mistake is exactly this confusion even if he’s not aware of it. He takes his tone of voice, his psychological conviction that he can’t be wrong, and treats it as though it were doing justificatory work. But certainty as a tone of voice is just subjective certainty, and subjective certainty, no matter how forceful your tone of voice, doesn’t justify anything. Justification comes from the practice within the language game. This connects to the point I made in my JTB+U paper about understanding being internal to justification. A person can feel certain about a calculation without understanding why it’s correct, and that feeling adds nothing to the justificatory standing of the belief. What matters is whether the person has navigated the practice correctly, not whether they feel confident about the result. This use of I know is often conflated, i.e., the difference between expressing a conviction (subjective feeling), and expressing I know as an objective justificatory assertion.

48

Your opening definition of “non-linguistic” as “not normally functioning as a proposition inside the space where we give grounds, check, doubt, correct, and argue” is very close to what I mean by non-epistemic. So, there may be common ground here. But then you say you’d avoid treating hinges as if they functioned outside the space where we give grounds, and that’s where I think you part ways with the text. Witt’s point is precisely that some things do stand outside that space. That’s what makes them hinges.

OC 166 says, “The difficulty is to realize the groundlessness of our believing.”

OC 204 also “Giving grounds, however, justifying the evidence, comes to an end; but the end is not certain propositions’ striking us immediately as true, i.e. it is not a kind of seeing on our part; it is our acting, which lies at the bottom of the language-game.”

OC 205, “If the true is what is grounded, then the ground is not true, nor yet false.” OC 253, “At the foundation of well-founded belief lies belief that is not founded.”

OC 341-343, “The questions that we raise and our doubts depend on the fact that some propositions are exempt from doubt, are as it were like hinges on which those turn… If I want the door to turn, the hinges must stay put.”

OC 559, “You must bear in mind that the language-game is so to say something unpredictable. I mean: it is not based on grounds. It is not reasonable (or unreasonable). It is there — like our life.” The cumulative force of these passages is hard to resist. Witt is saying repeatedly that there is a level at which giving grounds comes to an end, where what we find is not more justified belief but our acting, our form of life. Hinges aren’t moves within the game. They’re what makes the game possible.

Your point about creative improvisation is interesting and I think it applies to moves within language-games. Every application of a rule involves judgment that can’t be fully captured by further rules, and OC 26-29 make exactly this point. But hinges are precisely what doesn’t get improvised, at least at the bedrock level. They’re the stable background that makes improvisation possible. If the ground shifted with every step, walking wouldn’t be creative. It would be impossible. You say hinges are dimensions of each new action and that their sense is subtly redefined by each move. But Witt says hinges stand fast. That’s their defining characteristic. OC 94 says our world-picture is “the inherited background against which I distinguish between true and false.” The riverbed can shift over time, yes, but at any given moment the bed holds and the water flows because it holds.

49

But he also says there is no categorical distinction between bedrock and flowing river, it’s a matter of degree.

It might be imagined that some propositions, of the form of empirical propositions, were
hardened and functioned as channels for such empirical propositions as were not hardened but fluid;
and that this relation altered with time, in that fluid propositions hardened, and hard ones became
fluid.

The mythology may change back into a state of flux, the river-bed of thoughts may shift. But I
distinguish between the movement of the waters on the river-bed and the shift of the bed itself;
though there is not a sharp division of the one from the other.

And the bank of that river consists partly of hard rock, subject to no alteration or only to an
imperceptible one, partly of sand, which now in one place now in another gets washed away, or deposited.”

If the alteration of the hinge is imperceptible, isnt it still
an alteration? I think this matters because it seems to me the only way hinges form and change is through our actual moment to moment, day to day use of words and gesture. If the hinge is the larger system of practices that participates in, informs and organizes our interactions, if each interaction provides an element of creative improvisation, and if we eventually find ourselves moving through a course of interactions which have the ability to significantly alter that hinge bedrock, mustn’t this potential responsiveness of hinges to certain life-altering varieties of experience be made possible by the tacit responsiveness of hinges to familiar rule-governed interactions?

I’m reminded of Piaget’s overarching epistemological schemes, which maintain their overall stability even as they accommodate themselves to the novel particularities of events. If hinges define the meaning of the theme, then variations within the theme don’t just occur within it but extend, enrich and redefine the meaning of the system as a whole.

If we were to insist that a hinge, as a reciprocally unified system of practices, could remain utterly unchanged by the routine events which
add a creative element to that system (i.e. moves within chess), then where do we locate the point of unresponsiveness in this unified reciprocal system of actions such that the superordinate dimension is cut off from changes in subordinate events? Could this direction of influence ever be just one way, from superordinate to subordinate?

50

Josh, this is a more nuanced than before and I think the riverbed metaphor already explains it. Witt says that hinges can shift over time. The riverbed generally isn’t permanent. But the shift is gradual and happens across the life of a practice, not in each individual interaction. When you ask whether imperceptible alteration is still alteration, the answer is yes, but that doesn’t make hinges responsive to individual moves in the way you seem to be suggesting. The erosion of a riverbed by water is real, but at any given moment the bed is what makes the flow of water possible. You can’t explain the flow by saying the bed is constantly being redefined by each drop. The asymmetry between bed and water is what makes the metaphor work, even though over time the bed shifts.

But it’s also important to recognize that not all hinges are equally susceptible to change. The distinction I’ve been drawing between non-linguistic and linguistic hinges maps directly onto this question. The deepest non-linguistic hinges, the continuity of experience, the persistence of objects, the existence of other beings, aren’t susceptible to change through practice at all. No amount of interaction is going to erode the certainty that the world holds together from moment to moment. These are conditions of any possible engagement with the world. Non-linguistic hinges at the level of bodily engagement, like the certainty that the ground will hold, are also extremely stable. Linguistic hinges are where your point has the most traction. “No one has been to the moon” functioned as a hinge in Witt’s time and no longer does. “The earth is flat” functioned as a hinge for centuries and then stopped. These hinges take propositional form within our language-games and are the most responsive to changes in practice, science, and experience. So your picture of hinges being gradually reshaped captures part of the story, but it applies to some hinges and not others, and it doesn’t apply at all to bedrock (I distinguish between bedrock hinges and general foundational hinges) ones.

Your Piaget comparison is actually helpful, though not in the way you intend. Piaget’s schemas accommodate new experience through assimilation and accommodation, but the schemas themselves maintain structural stability precisely so that accommodation can occur. If the schema dissolved with every new input, there would be nothing to accommodate to. The stability is what makes the responsiveness possible. That’s exactly the relationship between hinges and ordinary moves within language games. The hinge holds so that the moves can happen. If hinges were as fluid as you’re suggesting, they couldn’t play the organizing role that Piaget’s schemas play in his framework.

Your final question about where to locate the point of unresponsiveness assumes that if influence ever flows from subordinate to superordinate, it must flow constantly and in each interaction. But that doesn’t follow. A single chess game doesn’t redefine the rules of chess, even though over centuries the rules have changed. The fact that change eventually happens doesn’t mean each move is contributing to a redefinition of the rules. Most moves simply occur within a stable framework, and that stability is what makes them intelligible as moves. The direction of influence isn’t just one way, but the timescales are radically different. Hinges shape each interaction in the moment. Interactions reshape hinges only over the long life of a practice, and only the linguistic ones at that. Collapsing those timescales and those levels into one loses the asymmetry that makes Witt’s account work.

The points you’re making are important to the conversation.

51

Here I’d like to add, that there are two very distinct types of mistakes in the employment of a rule. One is when the rule is incorrectly followed, and the other is when the rule is incorrectly applied. The first type of mistake is exemplified by invalid logic. If the rule is not properly followed there is logical fallacy. The second type is exemplified by the unsound conclusion produced from a false premise. In this case, the rule of logic is properly followed, but the rule is incorrectly applied because of a false premise.

To relate this to mathematical calculations, here’s an example. If someone concludes that 2+2=5, this is an instance of not correctly following the rule. On the other hand, if someone takes a group of two things, and puts them together with a group of three things, and claims that this is represented as 2+2=4, therefore there is now four things, that would be a case of misapplication. The group of three things has been falsely represented as 2 in the application of the rule, and this produces an unsound conclusion.

The issue with the second type of mistake, “mistake in applying”, is that this type of mistake cannot be completely understood as a failure in following a rule. We may assume rules of application, but those rules would need rules of application, so this would produce infinite regress.

52

You subscribe to a more realist reading of Wittgenstein than I do.You see stable aspects of the world as grounding what you call the deepest non-linguistic hinges. According to my interpretation, there are no aspects of the world which can be extracted from the changing uses we make of them in our practices. There is no determinate scheme or context that can fix the content of utterances, and hence no way to get outside of language. How a theory or practice interprets the world is itself inescapably open to further interpretation, with no authority beyond what gets said by whom, when.

Piaget’s reciprocal approach to the relation between schemes and content is in line with my claim that assimilation and accommodation are not just minor adjustments to subordinate structures; they recursively reshape higher-level schemas as well. Piaget has written that each interaction with the environment, no matter how small, is incorporated into the organism’s total cognitive structure. Superordinate schemes are relatively stable because they are deeply integrated across multiple experiences, but their stability is always provisional. Every new experience subtly modifies them, otherwise, development would be impossible. The ‘hinge’ is never separate from action; it exists only in the ongoing structuring of experience.

53

I find this interesting.

You say there are no aspects of the world that can be extracted from the changing uses we make of them in our practices, and that there’s no way to get outside of language. That’s interesting, but I don’t think this position is coherent, and I don’t think it’s Witt’s position. The claim that there’s no getting outside language is itself made in language and claims to be saying something true about the relationship between language and the world. But if there’s no getting outside language, then the claim can’t be saying anything about the world as it is independent of language, which means it can’t establish what it claims to establish. It’s self-undermining in the same way that radical skepticism is self-undermining. The skeptic uses the framework of rational inquiry to argue that rational inquiry can’t reach the world, and the “no outside of language” position uses language to assert that language can’t be transcended. Both saw off the branch they’re sitting on.

And more to the point, the position can’t account for pre-linguistic creatures. A dog navigates the world with basic certainties that function very well without language. An infant acts with the certainty that objects persist before it has a single word. These aren’t interpretations. They’re conditions of engagement that exist prior to and independent of any linguistic activity. Either you deny that these creatures have any engagement with the world at all, which is absurd, or you admit that there is a level of engagement that operates prior to language, which concedes the point. When Witt says at OC 204 that what lies at the bottom of the language-game is “our acting,” he’s pointing to something that isn’t itself linguistic or interpretive. Acting isn’t interpreting. The child reaching for a toy isn’t applying a theory of object permanence. The child is reaching. The certainty is in the reaching, not in any interpretive act. Witt’s whole method in OC is descriptive, and what he describes includes a level that is prior to language and interpretation.

On Piaget (I had to look this up because I’m not familiar with Piaget), I think you’re misreading the source. Piaget does say every interaction modifies cognitive structures, but he insists on a distinction between assimilation, where new experience is fitted into existing structures, and accommodation, where structures change in response to experience that can’t be assimilated. Most interactions are assimilative. The structures hold and the new experience is absorbed. Accommodation happens when assimilation fails, and it’s the exception, not the rule. If every interaction equally modified superordinate structures, there would be no distinction between assimilation and accommodation, and Piaget’s entire framework collapses. You’re flattening a distinction that Piaget himself insists on. The claim that stability is “always provisional” also smuggles in an assumption that doesn’t hold across all levels of certainty. Provisions apply to linguistic hinges that can shift over time. “No one has been to the moon” was provisional and eventually shifted. But the persistence of objects, the continuity of experience, the existence of other beings, these aren’t provisionally stable. They’re conditions of there being a world in which stability and provisionality have any meaning at all. Calling them provisional puts them on the same level as linguistic hinges that do change, and that flattens exactly the distinction between levels that I’ve been arguing is essential to understanding OC.

Your reading of Witt pulls him closer to Rorty or Derrida than the text warrants. OC isn’t saying that the world is constituted by our interpretive practices. It’s saying that our basic certainties show themselves in how we act, and that some of those certainties are so deep that the question of interpretation doesn’t arise. They aren’t interpreted because they’re the ground on which interpreting becomes possible.

54

I respect the fact that you dont want to hand Wittgenstein over to the postmodernists. Wittgenstein scholars are strongly divided over whether to read him this way or as a realist. But my mention of Piaget was designed to show that one doesn’t have to go full-on relativist in order to interpret hinges as intrinsically responsive to events.

Piaget writes:

The subjective and the objective, integration and differentiation, affirmation and negation, structure and genesis, form and content, are implicit in the event of meaning; … every assimilatory scheme has to be accommodated to the elements it assimilates … " \ and thus, “from the beginning assimilation and accommodation are indissociable” •.
…assimilation can never be pure because by incorporating new elements into its earlier schemata the intelligence constantly modifies the latter in order to adjust them to new elements. Conversely, things are never known by themselves, since this work of accommodation is only possible as a function of the inverse process of assimilation. …We thus see how the very concept of the object is far from being innate and necessitates a construction which is simultaneously assimilatory and accommodating. “

55

Post 12 OC 31-35

Text

OC 31: “The propositions which one comes back to again and again as if bewitched — these I should like to expunge from philosophical language.”

OC 32: “It’s not a matter of Moore’s knowing that there’s a hand there, but rather we should not understand him if he were to say ‘Of course I may be wrong about this.’ We should ask ‘What is it like to make such a mistake as that?’ — e.g. what’s it like to discover that it was a mistake?”

OC 33: “Thus we expunge the sentences that don’t get us any further.”

OC 34: “If someone is taught to calculate, is he also taught that he can rely on a calculation of his teacher’s? But these explanations must after all sometime come to an end. Will he also be taught that he can trust his senses — since he is indeed told in many cases that in such and such a special case you cannot trust them? — Rule and exception.”

OC 35: “But can’t it be imagined that there should be no physical objects? I don’t know. And yet ‘There are physical objects’ is nonsense. Is it supposed to be an empirical proposition? — And is this an empirical proposition: ‘There seem to be physical objects’?”

Remarks

OC 31 and 33 tells us something about what Witt is doing. The propositions he wants to expunge are the ones philosophers keep returning to “as if bewitched.” Moore’s “I know that here is a hand” is the paradigm case. These propositions bewitch because they look like they’re doing epistemic work when they’re actually idle. They don’t get us anywhere because they’re attempting to operate in a space where the grammar of know has no epistemic power. Moore keeps coming back to them because he’s captivated by a picture of what knowing looks like.

OC 32 explains that the problem isn’t that Moore might be wrong about having a hand. The problem is that we wouldn’t understand him if he said he might be wrong. This is a grammatical observation, not an empirical one. It’s not that error is unlikely. It’s that we can’t give content to the idea of error here. “What is it like to discover that it was a mistake?” is a devastating question because there is no answer. There’s no procedure, no discovery, no experience that would count as finding out you were wrong about having a hand under ordinary circumstances. And if there’s no intelligible content to the possibility of error, then I know (In Moore’s sense) has no grip either, because knowing only has sense where doubt has sense. This reinforces OC 21, if I know is meant to mean “I can’t be wrong,” it collapses, because the impossibility of being wrong here isn’t epistemic. It’s a feature of the grammar.

OC 34 connects back to the regress problem from OC 25-27. When a child learns to calculate, she also learns to trust her teacher’s calculations. But this trust isn’t itself taught to the child. It’s part of learning to calculate. You don’t first learn arithmetic and then, in a separate step, learn to trust your teacher. The trust is embedded in the training. The same goes for trusting one’s senses. We aren’t taught as a general rule to trust our senses. We’re taught specific cases where they can and can’t be relied on, “Don’t judge the color in this light,” “Look more carefully.” Rule and exception. But the general reliance on the senses isn’t a rule we learn. It’s the background within which the rules about when to trust or distrust function. The child doesn’t justify trusting her teacher. The trust is the condition under which the teaching can proceed. This is hinges at work, something that must stand fast for the epistemic language games to function.

OC 35 asks whether it can be imagined that there should be no physical objects and then says “There are physical objects” is nonsense. Witt gets to the point. “There are physical objects” isn’t an empirical proposition because nothing could count as evidence for or against it. It doesn’t describe a discovery about the world. If it were empirical, there would have to be circumstances under which it could turn out to be false, and there aren’t. But it’s not a logical truth either. It has the form of an empirical proposition without the function of one. This is why it’s a hinge. It stands fast not because we’ve verified it but because it’s part of the framework within which justification has sense. The question “Is this an empirical proposition: ‘There seem to be physical objects’?” twists the knife further. Even the weakened version can’t be made into a proper empirical claim, because the entire language of appearance and reality presupposes the framework that physical objects provide. You can’t step outside that framework and report on how things seem from nowhere. The framework isn’t a conclusion we’ve reached. It’s the inherited background against which conclusions are reached.

56

Did you mean to say here that the entire language of appearance and reality presupposes the framework that the grammatical language game of physical objects provides?That wording would clarify that it’s not physical objects which ground language, as if there were an ontological scaffold of objects that makes discourse possible. Instead, it’s language-games which ground the sense of objects.

We don’t discover that physical objects form the necessary structure of reality; rather, we recognize that our language-game involving physical objects sets the conditions under which “empirical proposition,” “appearance,” and “reality” have their sense at all. This is the difference between “this is how our language works” and “this is how reality must be structured.”

57

I wasn’t making an ontological claim that physical objects ground language. So “the framework that the grammatical language game of physical objects provides” is better wording.

You present two options, “this is how our language works” versus "this is how reality must be structured, "and place Wittgenstein on the language side. I don’t think he’s on either side. At the hinge level, language and world aren’t two separate things that need connecting. The separation between them is something that only arises once we step back and start doing philosophy. “There are physical objects” isn’t a discovery about reality, but it’s also not a convention of our language game. It’s the ground on which both language and world already stand together before anyone thinks to ask which one comes first.

Saying language games ground the sense of objects just reverses the priority. Instead of objects grounding language, language grounds objects. But Witt refuses the priority question altogether. The hinge holds fast, not because it’s grounded in anything. That’s what makes some hinges bedrock.

And this connects to a familiar problem. If language games ground the sense of objects, then a pre-linguistic child reaching for a cup can’t have anything functioning as object-certainty. But the child does and so do animals. The certainty is in the act of reaching. It’s not yet propositional, not yet inside a language game, but it’s already functioning as a hinge. That’s the level Witt keeps pointing to when he talks about acting with basic certainty rather than asserting it.

58

You’re right to point out that we shouldn’t succumb to the temptation of replacing ’physical reality grounds language’ with a simple reversal in the direction of ‘language grounds physical reality’.
But what would it mean to think together language and world prior to any ontological split? Specifically, how does the sentence “there are physical objects” ground language-world? There is a risk of reading this as a pre-normative fusion of language and world, a kind of undifferentiated bedrock, as if there were a primordial unity prior to grammar, prior to standards, prior even to correctness. In other words, that “there are physical objects” points to a brute entanglement of organism and world.

But if we are serious about integrating language and world, then we have to extend the contingently relative normative function of language to pre-linguistic interactions as well. Before the child learns to speak, their entanglement with a world already pre-supposes the construction of normative, rule-governed perceptual practices. There is no layer of sheer entanglement beneath normativity. Even perception is rule-governed in a tacit way. The child does not first inhabit a brute world of perception and later acquire linguistic norms. The child is initiated into a normatively articulated world from the beginning.

If language games ground the sense of objects, then a pre-linguistic child reaching for a cup can’t have anything functioning as object-certainty. But the child does and so do animals. The certainty is in the act of reaching. It’s not yet propositional, not yet inside a language game, but it’s already functioning as a hinge. That’s the level Witt keeps pointing to when he talks about acting with basic certainty rather than asserting it.”

Whereas Wittgenstein doesn’t wade into this arena, phenomenologists like Husserl and Merleau-Ponty investigate the constructed nature of norms of pre-linguistic perception like object permanence. I see their work as consistent with and an extension of Wittgenstein’s analysis of the social construction of linguistic norms. For them, object permanence is a rule-bound normative achievement. The certainty comes from the ground provided by an idealizing construction. But the normative ground providing object permanence is itself contingent. It is a belief in the persistent self-identity of the object. This belief is stable but ultimately changeable. In sum, moving from the realm of linguistic social practices to pre-linguistic perceptual construction does not land us in a space prior to contingent norm-based rules.

When Merleau-Ponty states

“The world is inseparable from the subject, but from a subject which is nothing but a project of the world; and the subject is inseparable from the world, but from a world which it itself projects”

he means this to apply equally to the linguistic and pre-linguistic worlds.

59

I’m getting quite confused by the mixing of metaphors here. It seems very odd to call something that is described as a hinge as a foundation. My impression is that W talks about hinges in the context of debates and doubts. I understand them as propositions that are protected from question and debate, at least in a given context. Perhaps its not unduly mixing metaphors to say that an actual hinge enables the door to swing and also defines the space through which it can swing.
The idea of protecting a proposition is exemplified for me by the following. We may well notice that water that is heated will often begin to boil at a certain point. We can measure that point, and when we do, we find that it does not always boil at the same temperature. We could abandon our generalization, but we don’t. We look for additional variables and end up with a more complicated, but more accurate generalization.

Two more metaphors. One of which is very useful because it allows us to think about multiple levels of certainty (the water in the bed and the bed in the landscape) But can it be applied to hinge propositions. But multiple levels of hinges doesn’t make sense.

There’s a wide range of metaphors in the text.

There’s no reason to suppose that these are interchangeable. I don’t think we are likely to be able to build a taxonomy out of this.
Should we not expect that all these metaphors will give us slightly different pictures, which might be more appropriate in some cases rather than others. Treating them as interchangeable seems very confusing.

A taxonomy of hinges?
60

You’re right that I’m not pointing to a “brute entanglement” or some undifferentiated primordial unity. But extending normativity all the way down to pre-linguistic perception creates a different problem. For e.g., if a cat chasing a mouse is engaged in a rule-governed normative activity, what isn’t normative? The term does work when it marks a distinction, between activities subject to standards of correctness and activities that aren’t. If everything is normative, the concept stops distinguishing anything.

Object permanence in a kitten isn’t normative. The kitten doesn’t construct object permanence according to rules that could be otherwise. It develops object permanence because that’s how perception matures in creatures with its kind of nervous system. Calling that “rule-governed” stretches the term past its breaking point.

You acknowledge that Witt doesn’t wade into this arena, and I don’t think that’s accidental. When he describes certainty that shows itself in action, sitting in a chair, reaching for a cup, he’s not characterizing it as normatively constructed. When he says, “It is there, like our life” (OC 559), he’s pointing to something prior to the question of whether it’s constructed or natural, contingent or necessary. That’s not a gap in his analysis. It’s the point.

The Merleau-Ponty quote is actually closer to what I’m saying. “The world is inseparable from the subject, but from a subject which is nothing but a project of the world,” that’s mutual implication, not normative construction. The subject doesn’t construct the world according to contingent rules. Subject and world are already intertwined before the question of construction arises. That’s very close to what I mean by the hinge level being where language and world stand together before anyone asks which grounds which.